Smart textiles for in situ monitoring of composites 9780081023099, 008102309X

Citation preview

Smart Textiles for In Situ Monitoring of Composites

The Textile Institute Book Series Incorporated by Royal Charter in 1925, The Textile Institute was established as the professional body for the textile industry to provide support to businesses, practitioners and academics involved with textiles and to provide routes to professional qualifications through which Institute Members can demonstrate their professional competence. The Institute’s aim is to encourage learning, recognise achievement, reward excellence and disseminate information about the textiles, clothing and footwear industries and the associated science, design and technology; it has a global reach with individual and corporate members in over 80 countries. The Textile Institute Book Series supersedes the former ‘Woodhead Publishing Series in Textiles’, and represents a collaboration between The Textile Institute and Elsevier aimed at ensuring that Institute Members and the textile industry continue to have access to high calibre titles on textile science and technology. Books published in The Textile Institute Book Series are offered on the Elsevier web site at: www.elsevier.com/books-and-journals and are available to Textile Institute Members at a substantial discount. Textile Institute books still in print are also available directly from the Institute’s web site at: www.textileinstitute.org To place an order, or if you are interested in writing a book for this series, please contact Matthew Deans, Senior Publisher: [email protected]

Recently Published and Upcoming Titles in The Textile Institute Book Series Handbook of Technical Textiles, Volume 1, 2nd Edition, A. Richard Horrocks and Subhash C. Anand, 9781782424581 Handbook of Technical Textiles, Volume 2, 2nd Edition, A. Richard Horrocks and Subhash C. Anand, 9781782424659 Geotextiles, Robert Koerner, 9780081002216 Advances in Braiding Technology, Yordan Kyosev, 9780081009260 Antimicrobial Textiles, Gang Sun, 9780081005767 Active Coatings for Smart Textiles, Jinlian Hu, 9780081002636 Advances in Women’s Intimate Apparel Technology, Winnie Yu, 9781782423690 Smart Textiles and Their Applications, Vladan Koncar, 9780081005743 Advances in Technical Nonwovens, George Kellie, 9780081005750 Activated Carbon Fiber and Textiles, Jonathan Chen, 9780081006603 Performance Testing of Textiles, Lijing Wang, 9780081005705 Colour Design, Janet Best, 9780081012703 Forensic Textile Science, Debra Carr, 9780081018729 Principles of Textile Finishing, Asim Kumar Roy Choudhury, 9780081006467 High-Performance Apparel, John McLoughlin and Tasneem Sabir, 9780081009048 Handbook of Properties of Textile and Technical Fibres, 2nd Edition, Bunsell, 9780081012727

Related Titles Smart Textiles and their Applications, 978-0-08-100574-3 Sensor Technologies for Civil Infrastructures, Volume 2: Applications in Structural Health Monitoring, 978-1-78-242242-6 Structural Health Monitoring (SHM) in Aerospace Structures, 978-0-08-100148-6 Structural Health Monitoring of Aerospace Composites, 978-0-12-409605-9

The Textile Institute Book Series

Smart Textiles for In Situ Monitoring of Composites

Vladan Koncar

Woodhead Publishing is an imprint of Elsevier The Officers’ Mess Business Centre, Royston Road, Duxford, CB22 4QH, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, OX5 1GB, United Kingdom Copyright © 2019 Elsevier Ltd. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN (Print): 978-0-08-102308-2 ISBN (Online): 978-0-08-102309-9 For information on all Woodhead publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Matthew Deans Acquisition Editor: Brian Guerin Editorial Project Manager: Charlotte Rowley Production Project Manager: Sojan P. Pazhayattil Cover Designer: Vicky Pearson Esser Typeset by TNQ Technologies

Contents

General introduction

vii

1

Smart textiles for monitoring and measurement applications 1.1 Introduction 1.2 Smart textiles 1.3 Sensorsddefinitions and classifications 1.4 Connectors 1.5 Conductive polymers, fibers, and structures 1.6 Materials and sensors for glass fibers based composites monitoring References Further reading

1 1 1 2 27 49 95 142 146

2

Composites and hybrid structures 2.1 Compositesdterms and definitions 2.2 Textile reinforced composites 2.3 Outlookdcomposite structures 2.4 Reinforcing fibers 2.5 Matrices 2.6 Failure mechanisms in composites 2.7 Hybrid structures, production methodology and principles, state of the art 2.8 Hybrid structuresdbonding issuesdinnovative joining techniques 2.9 Conclusion References Further reading

153 153 159 184 185 194 196

Structural health monitoring of composite structures 3.1 Health monitoring definitions 3.2 State of the art of monitoring techniques 3.3 Characterization of textile sensors before insertion in textile preforms 3.4 Characterization of textile sensors after insertion in textile preforms

217 217 217

3

200 201 205 205 210

222 229

vi

4

5

Contents

250

3.5 Results and discussiondinterface phenomena 3.6 Results and discussiondtomography analysis of textile reinforced 2D thermoplastic composites with integrated textile sensors 3.7 Results and discussiondelectrical resistance dependence of textile sensors on climatic conditions 3.8 ResultsdSEM and EDS analysis of yarns 3.9 Results and discussiondthermal properties of yarns and textile reinforced 2D thermoplastic composites with integrated sensor yarns 3.10 Toward wireless structural health monitoring 3.11 Predictive maintenance concept 3.12 Conclusion References Further reading

270 277 281 283 283 284

Structural health monitoring of processes related to composite manufacturing 4.1 Study case 1, interlock weaving process monitoring 4.2 Study case 2, Stamping process supervision 4.3 Study case 3, Infusion process supervision References Further reading

295 295 348 360 371 372

General conclusion Acknowledgments

383 385

Index

260 262 264

387

General introduction Smart textiles Smart textiles, encompassing electronics combined with textiles also called textronics, have a very promising realm in science and technology nowadays because of commercial viability and public interests [1,2]. Smart textiles play a significant role as well in the European textile sector and assist the textile industry in its transformation into a competitive knowledge driven industry. These kinds of textiles combine knowledge from many disciplines with the specific requirements of textile. Numerous materials and systems are available together with devices for sensing and actuation, but they are not compatible with a textile or with the textile production processes. They could be transformed into a textile compatible structure or even in a full textile structure. Smart textiles can be defined as textiles that are able to sense and respond to changes in their environment. They may be divided into two classes: passive and active smart textiles. Passive smart textiles have the ability to change their properties according to an environmental stimulation. Shape memory materials, hydrophobic or hydrophilic textiles, etc., make part of this category. Active smart textiles are fitted with sensors and actuators to connect internal parameters to the transmitted message. They are able to detect different signals from the environment (temperature, light intensity, pollution, etc.), to decide how to react, and finally to act using various textile based, flexible or miniaturized actuators (textile displays, microvibrating devices, LED, OLED, etc.). The “decision” can be taken locally in case of embedded electronic devices (textile electronics) to smart textile structures or remotely in case the smart textile is wirelessly connected to clouds containing data base, servers with artificial intelligence software, etc., and may be a part of Internet of Things (IoT) concept. The notion Smart Material was for the first time defined in Japan in 1989. The first textile material that, in retroaction, was labeled as a “smart textile” was silk thread having a shape memory. The discovery of shape memory materials in the 1960s and intelligent polymeric gels in the 1970s was, however, generally accepted as the birth of real smart materials. It was not before the late 1990s that intelligent materials were introduced in textiles. First researches related to communicative textiles have been done in different laboratories in late 1990s and first textile electronic semiconductive components have been realized in early 2000s.

viii

General introduction

Today’s dynamic market segments such as digital, health, transportation, energy, or security have an important added value that may be captured by textiles as supports. Currently, the value of 1 kg of technical textiles is estimated to be 5.3 US dollar comparing to 3.4 US dollar for nonwoven textiles and 10.5 for textile composites. The smart textile market is also rapidly growing. On the other side, the market of connected objects also called “Internet of Things” or “Internet of staff” will possibly contain 30 billion devices connected in 2020, and 10% of those things will be clothes. If these figures are realistic, it means that in 2020, about 3 billion of clothes will be connected and will use some aspects of smart textiles such as sensors, or actuators together with emitters, receivers, and units of computation. The market share of smart textiles in that case will be even larger than statistics have forecasted. These figures are very encouraging for the textile industry globally and indicate major changes in our way of life. Social networks have introduced a completely different approach to interactions among human beings. New services, such as Uber, Blablacar, AirBandB, crowd funding, etc., radically modified the economy and the cost of transportation, accommodation, etc. Textile manufacturers have brought sensor-based smart textiles products to the market, mainly for the collection of biodata (e.g., heart-rate, body temperature, etc.) and in workplace safety. The following table shows an example of individual products in the market that serve a limited scope of providing the wearable sensing device mostly based on a bespoke and proprietary hardware (Table 1). However, these early stage products are yet far from being integrated in a comprehensive IoT infrastructure, with standard protocols for secure data access from/to cloud and, as platforms for smart textiles, lack of the possibility for both the users and development communities to build innovative solutions by integrating them with new functionalities, still guaranteeing the elevated level of security and safety that is essential to secure user acceptance. From that perspective, there is still room to build an entirely new industry of smart e-textiles that will constitute the future IoT nodes for wearable electronics, smart and communicating composite parts and integrate fully with existing IoT infrastructures standardized interfaces and open tools for applications developers. Solving this dilemma and unlocking the IoT of smart wearable e-textiles and smart technical textiles is exactly the vision for the future. Once this is done, and validated in the use-case scenarios that may be envisaged, the market opportunity to look at is enormous. Until around 5 years ago, the healthcare sector was the main consumer of a relatively small market for smart textiles and wearable electronics and sensors market. Although, since then the market has progressed considerably, with applications in personal health and fitness monitoring, immersive gaming and, wireless communications. This has been supported by the reduction in the costs of sensors and wireless connectivity with smart devices. Besides bringing low-cost monitoring and MEMS sensor fusion to devices such as smart watches, the growth in the consumer wearable’s market

General introduction

ix

Table 1 Smart textiles generations Generation

Products

First generation

Sensoria fitness garments Adidas MiCoach Under Armor Healtbox

Second generation

Ralph Lauren Polo Tech Jabit Circuit Hearth Monitoring Hexoskin performance management Asensei personal trainer OM signal smart clothing Stretch sense sensor Nu metrix heart rate monitor

Third generation

Advan Pro pressure sensing shoes AIQ Smart clothing stainless steel yarns CLIM8 GEAR heated textiles NTT data and Toray hitoe smart shirt Bioserenity cardio skin @Health Sense box

has resulted in increased investment in new types of smart textiles and wearables and their associated manufacturing technologies. There is now increasing demand from consumer electronics, sports, and gaming industries. On the other side, technical textiles that are higher added-value products, will also become smarter in the future having functions of real time in situ structural health monitoring (SHM) to achieve the concept of predictive maintenance. Printed electronics and energy harvesting technologies are evolving to meet the demands of new, wearable formats. Next-generation wearables and smart technical textiles will rely on active fabrics made by weaving conductor, insulator, and semiconductor fibers sparsely into textile yarn. Fabrics woven from such yarns will allow electronic functions that seamlessly integrate into every day, comfortable, lightweight clothing, and also smart and connected composite structural parts of trains, planes, and cars. Sensor tattoos and wearable motion charging devices are now in early commercial stages. The following graph represents a forecast of the possible growth in smart textiles market (Fig. 1). Materials such as metals, optical fibers, and conductive polymers may be directly integrated into the 2D or 3D textile structures, thus supplying electrical conductivity, sensing capability, and data transmission capability to the material [1,2]. Electronic components, such as electrodes, antennas, transmission lines, or heating elements require electroconductive yarns to increase textile compatibility [3]. In recent years, conductive and semiconductive polymers and textile products have been considered

x

General introduction

Million US$

Market demand in million US$ for smart textiles

3500,00 2900,00

3000,00 2500,00 2100,00

2000,00 1500,00

1300,00

1000,00

750,00 549,00

500,00 79,00

0,00

2014

154,00

2015

260,00

2016

2017 2018 Year

2019

2020

2021

Figure 1 Evolution of smart textiles market (prevision).

for electromagnetic shielding and antielectrostatic purposes, heating, transport of electrical signals, transistors, etc., in various applications for the electronic and defense industries. This is mainly due to their desirable properties in terms of electrostatic discharge, radio frequency interference protection, thermal expansion matching, and weight [4e7]. Electroconductive yarns must satisfy the electric characteristics requirements of textile (resistance, capacitance, inductance) and physical characteristics (tensile strength, extensibility) [8]. The conductive fibers may be transformed in textile structures by weaving, knitting, or other manufacturing processes. Textile fibers are typical examples of electrical insulators. They do not permit the flow of electric current through them. However, for certain technical applications it is essential to use fibers with a considerable electrical conductivity. For textile-reinforced composites, one possible solution is to use intelligent textile materials and structures, which provide real possibility for online and in situ monitoring of structural integrity. Such intelligent materials are made by coating or treating textile yarns, filaments, or fabrics with nanoparticles or conductive and semiconductive polymers, giving them specified performance [8]. In this study novel textile sensors are introduced aiming at: l

l

l

monitoring of composite structures in real time in situ; composite manufacturing processes monitoring, such as stamping and weaving of reinforcements (multi layer, 3D, etc.); monitoring of infusion process for thermoset composites and thermo consolidation process for thermoplastic composite structures.

General introduction

xi

An example of textile fibrous sensors are E-glass/polypropylene (GF/PP) commingled yarns, E-glass/poly(N,N0 -hexamethylene adipamide) (GF/PA66) commingled yarns, and E-glass (GF) yarns that have been coated with an aqueous dispersion of polymer complex poly[3,4-(ethylenedioxy)thiophene]-compl-poly(4-vinylbenzenesulfonic acid) (PEDOT-compl-PSS). Textile sensors have been integrated during the weaving process of 2D/3D textile preforms. These sensors mostly demonstrate resistance to high pressure and temperature after thermal consolidation of 2D textile preforms with integrated textile sensors and support in situ SHM of textile-reinforced thermoplastic composites. Developed textile sensors give possibility for their integration in advanced 3D textile products by using industrial machines. Characterization of new textile sensors, preforms, and textile-reinforced thermoplastic composites included determination of their electrical, electromechanical, mechanical, thermal, and interface phenomena properties and morphological characteristics of the surface and cross section of textile sensors and textile-reinforced thermoplastic composites. In addition, the electrical conductivity of textile sensors, depending on temperature and humidity conditions, was investigated as well.

Composite structures A material composed of two or more distinct phases (matrix phase and dispersed or reinforcing phase) and having bulk properties significantly different from those of any of the constituents is defined as a composite material. Composite materials can be adapted to the intended application area and conceived according to the desired mechanical performance of the structure. Because of this fact, composite materials find broad potential in structural applications. Structural composite materials are being developed in many different domains such as: aeronautics, railway, marine, automotive, civil engineering, and all the others where specific mechanical properties related to weight are important. Currently, fiber-reinforced composites (i.e., composites having fibers as the reinforcing phase) are being widely used in high performance structural applications. Various textile-related techniques can be used to manufacture these fibrous reinforcements. Weaving, knitting, stitching, and braiding manufacturing techniques that are traditional textile technologies are used to manufacture interlaced fibrous assemblies to serve as reinforcements for composite materials. Hybridization of those technologies is also possible to realize more complex textile reinforcement structures with almost endless possibilities. A suitable design of the reinforcement coupled with the use of high performance fibers and a compatible matrix are essential for high-performance composite structural part. For instance, woven interlock reinforcements are unique as they can be manufactured on a conventional adapted loom. In this way, a single complex net-shaped preform

xii

General introduction

having through the thickness reinforcing fibers, to minimize delamination can be produced in a single manufacturing step. Because a composite owes its mechanical strength to the reinforcement properties, it is important to design reinforcement according to the target application of the composite. Complete understanding of the reinforcement properties, its manufacturing process, and the nature of matrix are critical. Moreover, woven structures are hierarchically organized. Three hierarchy levels are usually associated with textiles namely macro (structural parts ranging from a cm to several meters), meso (yarns and tows in mm), and micro (fibers in mm). At each level, different degrees of approximations are needed to describe the textile structure. Geometrical modeling at meso structural level is advantageous as it can be used to describe all the essential parameters of a fabric while avoiding unrealistic approximations of the macro scale and unnecessary complications at the micro scale. It is important that the woven reinforcement geometry be modeled keeping in view the technological constraints, so that the resultant geometrical description becomes realistic. Weaving parameters should be coupled with the geometry at meso structural level to obtain a sufficiently accurate and user friendly model [9,10]. Manufacturing process not only influences the geometry of tows inside the reinforcement but also their mechanical properties. Because of mechanical stresses and deformations induced during manufacturing process, tow-to-tow friction and abrasion between tows and the loom parts, the tows lose their properties significantly. Moreover, tow strength changes significantly along the complete production line of a woven composite, i.e., from tow to its integration in warp and weft to form a reinforcement and finally to the composite that is formed by resin impregnation. It is important to study this evolution and the influence of weaving parameters to understand and correctly asses the final reinforcement and composite properties. In addition to the understanding of mechanical behavior of a composite in laboratory, it is important to set up some sort of system to monitor the state or “health” of the structure during service life of the structural part or component. This scheme is termed as SHM [11]. Therefore, having studied the global tensile behavior of composites, the next step is introducing sensing mechanism in the composite to measure in-situ local deformations in real time. In the context of textile materials, these sensors should be compatible with the reinforcement and its manufacturing process. Like this, the cost of integration of sensors can be reduced considerably as it is one step insertion during manufacturing. Moreover, any sensor that follows the geometry of the reinforcement is expected to follow the deformation pattern when the composite is subjected to stresses. Strategically located in situ sensor can also give useful information about the deformation pattern of the reinforcement inside the composite during loading.

General introduction

xiii

Main contributions of this manuscript, in reference to the scope of the researches, are quoted below: l

l

l

l

l

Introduction to the development of fibrous sensors based on smart textile materials adapted to SHM in real time in situ of composite structures to achieve the concept of predictive maintenance; Development of specific textile sensors adapted to the monitoring of composite manufacturing and set up processes such as textile reinforcement production, stamping process, and infusion process; Introduction and classification of composite structures using reinforcement realized by textile manufacturing processes such as weaving, knitting, braiding, and hybrid processes; SHM of composites, including health monitoring definitions, state of the art of traditional monitoring techniques, characterization of textile sensors before insertion in textile preforms, mechanical stress response of textile sensors, interface phenomena of textile sensors, and related textile-reinforced thermoplastic composites, etc. SHM of processes related to composite manufacturing, weaving, braiding, infusion process, etc. This section contains three study cases to better explain our approach on specific applications.

The manuscript is organized in four distinct chapters to facilitate the reading and comprehension of its main contributions. The first chapter is focused on smart textiles for monitoring and measurement applications. The second chapter introduces the main concepts of composites and hybrid structures. In the third chapter, the SHM of composite structures is presented, and the fourth chapter is dedicated to SHM of processes related to composite manufacturing including three study cases. Finally, the conclusion and the discussions are given at the end of the manuscript. Finally, the PhD students under my scientific supervision and also coauthors of numerous scientific articles with me, Ivona Jerkovic, Ahmad Labanieh, Saad Nauman, and Nicolas Trifigny deserve a very special thanks for their hard work, outstanding results, and contributions to this manuscript.

References [1] V. Koncar (Ed.), Smart Textiles and Their Applicatons, Elsevier Science Ltd., London, 2016. [2] X. Tao, V. Koncar (Eds.), Handbook on Smart Textiles, Springer, Hong Kong, 2015. [3] A. Schwarz, I. Kazani, L. Cuny, Electro-conductive and elastic hybrid yarns e the effects of stretching, cyclic straining and washing on their electro-conductive properties, Materials and Design 32 (1) (2011) 4247e4256. [4] B. Kim, V. Koncar, E. Devaux, C. Dufour, P. Viallier, Electrical and morphological properties of PP and PET conductive polymer fibers, Synthetic Metals 146 (2) (2004) 167e174. [5] B. Kim, V. Koncar, C. Dufour, Polyaniline (PANI) coated PET conductive yarns: study of electrical, mechanical and electro-mechanical properties, Journal of Applied Polymer Science 101 (3) (2006) 1252e1256.

xiv

General introduction

[6] X. Tao, I. Fsaifes, V. Koncar, C. Dufour, C. Lepers, L. Hay, B. Capoen, M. Bouazaoui, CO2 laser-induced crystallization of solegel-derived indium tin oxide films, Applied Physics A 96 (3) (2009) 741e749. [7] X. Tao, V. Koncar, C. Dufour, Novel geometry pattern for the wire organic electrochemical textile transistor, Journal of the Electrochemical Society 158 (5) (2011) 572e577. [8] C. Cochrane, V. Koncar, M. Lewandowski, C. Dufour, Design and development of a flexible strain sensor for textile structures based on a conductive polymer composite, Sensors 7 (1) (2007) 473e492. [9] S. Nauman, I. Cristian, V. Koncar, Simultaneous application of fibrous piezoresistive sensors for compression and traction detection in glass laminate composites, Sensors 11 (10) (2011) 9478e9498. [10] I. Cristian, S. Nauman, F. Boussu, V. Koncar, A study of strength transfer from tow to textile composite using different reinforcement architectures, Applied Composite Materials 19 (3) (2012) 427e442. [11] D. Coutelier (Ed.), Advanced Material Textiles for Reinforced Structures for Complex Lightweight Applications, ENSIAME, Valenciennes, 2016.

Further reading
[1] A. Lobnik, S.K.U.M. Turel, Optical chemical sensors: design and applications, in: Advances in Chemical Sensors, Intech, Maribor, 2012, pp. 3e28. [2] S.A. Dyer, Wiley Survey of Instrumentation and Measurement, Wiley, Kansas city, 2001, pp. 84e98. [3] C. Cochrane, Développement d’un systeme de mesure d’allongement pour voilure de parachute, L’universite des Sciences et Technologies de Lille, Villeneuve d’Ascq, 2007. [4] Peratech, QTCTM Material Science the Physics of QTCTM Material Operation, Peratech, August 27, 2013 [En ligne]. Available: http://www.peratech.com/qtc-science.html. [5] B. Hafner, Energy Dispersive Spectroscopy on the SEM: A Primer, Characterization Facility, University of Minnesota, Twin Cities, 2007. [6] A. Kaw, Mechanics of Composite Material, Taylor & Francis Group, New York, 2006. [7] S.T. Peters, Handbook of Composites, Chapman & Hall, UK, Mountain View, 1982. [8] D. Gay, Matériaux Composites, Hermes, Paris, 1997. [9] M.J. Robert, Mechanics of Composite Materials, second ed., Taylor & Francis Group, , New York, 1999. [10] A. Elschner, S. Kirchmeyer, W. L€ovenich, PEDOT: Principles and Applications of an Intrinsically Conductive Polymer, CRC Press Taylor & Francis Group, Boca Raton, 2010. [11] G. Asch, Les capteurs en instrumentation industrielle, 7eme édition, Dunod, Paris, 2010. [12] C. Cochrane, V. Koncar, M. Lewandowski, Development of a flexible strain sensor for textile structures, in: 2nd Ambience 08 International Scientific Conference Proceedings, Smart Textiles e Technology and Design, Borås, 2008. [13] J.M. Cole, J. Berthelot, Composite Materials: Mechanical Behavior and Structural Analysis, Springer-Verlag, New York, 1999. [14] D. Yang, X. Tao, A. Zhang, Optical responses of FBG sensors under deformations, in: Smart Fibres, Fabrics & Clothing, Woodhead Publishing Limited, Hong Kong, 2001, p. 150.

General introduction

xv

[15] F.K. Ko, T.W. Chou, Composite Materials Series e Textile Structural Composites, Elsevier Science Ltd., Amsterdam, 1989. [16] G.C. Sih, A.M. Skudra, Failure Mechanics of Composites, Elsevier Science Publishers BV, New York, 1986. [17] H.P. Konka, M.A. Wahab, K. Lian, Piezoelectric fiber composite transducers for health monitoring in composite structures, Sensors and Actuators A: Physical 194 (2013) 84e94. [18] I. Cristian, S. Nauman, C. Cochrane, V. Koncar, Electro-conductive sensors and heating elements based on conductive polymer composites in woven structures, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabrics Technology, Athens, 2011, pp. 3e22. [19] L. Tong, A.P. Mouritz, M.K. Bannister, 3D Fibre Reinforced Composites, Elsevier Science Ltd., London, 2002. [20] P. Lucas, G. Zanella, Mise en oeuvre des composites TP, Compounds a fibres courtes, longues, Techniques de l’ingénieur, Paris, 2007. [21] R. Bartalesi, N. Carbonaro, F. Lorussi, M. Tesconi, A. Tognetti, G. Zupone, DeRossi, Smart textiles: toward a wearable motion capture system, in: Ninth International Symposium on the 3D Analysis of Human Movement, Valenciennes, 2006. [22] V.V. Vasiliev, E.V. Morozov, Mechanics and Analysis of Composite Materials, Elsevier Science Ltd., Oxford, 2001. [23] B. Adhikari, S. Majumdar, Polymers in sensor applications, Progress in Polymer Science 29 (7) (2004) 699e766. [24] M. Akerfeldt, M. Straat, P. Walkenstrom, Influence of coating parameters on textile and electrical properties of a poly(3,4-ethylene dioxythiophene):poly(styrene sulfonate)/ polyurethane-coated textile, Textile Research Journal 83 (20) (2013) 2164e2176. [25] J.N. Aneli, G.E. Zaikov, L.M. Khananashvili, Effects of mechanical deformations on the structurization and electric conductivity of electric conducting polymer, Black-Filled Rubbers Under, Journal of Applied Polymer Science 74 (1) (1999) 601e621. [26] E.R. Arakelova, A.M. Khachatryan, K.E. Avjyan, et al., Deposition of high-ohmic oriented ZnO films on glass, Si, and PEDOT-PSS, PEDOT-PSS(PVA) substrates at low temperatures by DC-magnetron sputtering, Journal of Contemporary Physics (Armenian Academy of Sciences) 46 (6) (2011) 293e299. [27] F.J. Arregui, I.R. Matıas, M. Lopez-amo, Optical fiber strain gauge based on a tapered single-mode fiber, Sensors and Actuators 79 (1) (2000) 90e96. [28] P.T. Asada, H.H. Gibbs, Wearable conductive fiber sensors for multi-axis human joint angle measurements, Journal of Neuroengineering and Rehabilitation 2 (1) (2005) p. on line. [29] T. Bashir, M. Ali, S. Cho, N. Persson, M. Skrifvars, OCVD polymerization of PEDOT: effect of pre-treatment steps on PEDOT-coated conductive fibers and a morphological study of PEDOT distribution on textile yarns, Polymers for Advanced Technologies 24 (2) (2013) 210e219. [30] S. Bhadra, D. Khastgir, N.K. Singha, Progress in preparation, processing and applications of polyaniline, Progress in Polymer Science 34 (8) (2009) 783e810. [31] A.B. Bradley, P.D. Render, The Development of a Parachute Strain Measurement Technique, vol. 1, American Institute of Aeronautics and Astronautics, 1986, pp. 194e202, 1. [32] B. Carter, J. Shannon, J. Forshaw, Measurement of displacement and strain by capacitance methods, Proceedings of the Institution of Mechanical Engineers 5 (1945) 215e221. [33] A. Chen, C.H. Sawhney, Soft-structured sensors and connectors by inkjet printing, AATCC Review 7 (6) (2007) 1e10.

xvi

General introduction

[34] C. Chen, J.C. Larue, R.D. Nelson, L. Kulinsky, M.J. Madou, Electrical conductivity of polymer blends of poly(3,4-,ethylenedioxythiophene): poly(styrenesulfonate): N-Methyl2-pyrrolidinone and polyvinyl alcohol, Journal of Applied Polymer Science 1 (2011) 1e8. [35] C. Chen, A. Torrents, L. Kulinsky, R. Nelson, M. Madou, L. Valdevit, J. LaRue, Mechanical characterizations of cast poly(3,4-ethylenedioxythiophene): poly (styrenesulfonate)/polyvinyl alcohol thin films, Synthetic Metals 161 (2011) 21e22. [36] O.P. Dimitriev, D.A. Grinko, Y.V. Noskov, et al., PEDOT: PSS films-effect of organic solvent additives and annealing on the film conductivity, Synthetic Metals 159 (21) (2009) 2237e2239. [37] G. Du, P. Popper, Analysis of a circular braiding process for complex shapes, Journal of the Textile Institute 85 (3) (1994) 316e337. [38] L. Flandin, A. Chang, S. Nazarenko, A. Hiltner, E. Baer, Effect of strain on the properties of an ethylene-octene elastomer with conductive carbon fillers, Journal of Applied Polymer Science 76 (6) (2000) 894e905. [39] L. Flandin, A. Hiltner, E. Baer, Interrelationships between electrical and mechanical properties of a carbon black-filled ethyleneeoctene elastomer, Polymer 42 (2) (2001) 827e838. [40] B. Friedel, T.J.K. Brenner, C.R. McNeill, et al., Influence of solution heating on the properties of PEDOT: PSS colloidal solutions and impact on the device performance of polymer solar cells, Organic Electronics 12 (10) (2011) 1736e1745. [41] L. Guo, L. Berglin, M. Heikki, Textile strain sensors characterization e sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 2 (1) (2010) 51e63. [42] L. Guo, L. Berglin, M. Heikki, Textile strain sensors characterization e sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 1 (2) (2010) 51e56. [43] J. Han, M. Shen, W. Cao, Hopping conduction in Mn-doped ZnO, Applied Physics Letters 82 (1) (2003) 67e69. [44] C.T. Huang, C.L. Shen, C.F. Tang, S.H. Chang, A wearable yarn-based piezo-resistive sensor, Sensors and Actuators A: Physical 141 (2) (2008) 396e403. [45] C.T. Huang, C.F. Tang, M.C. Lee, S.H. Chang, Parametric design of yarn-based piezoresistive sensors for smart textiles, Sensors and Actuators A: Physical 148 (1) (2008) 10e15. [46] J. Hwang, F. Amy, A. Kahn, Spectroscopic study on sputtered PEDOT PSS: role of surface PSS layer, Organic Electronics 7 (5) (2006) 387e396. [47] B. Kim, V. Koncar, E. Devaux, Electrical properties of conductive polymers: PET e nanocomposites’ fibres, AUTEX Research Journal 4 (1) (2004) 1e13. [48] G. Latessa, F. Brunetti, A. Reale, G. Saggio, A. Di Carlo, Piezoresistive behaviour of flexible PEDOT: PSS based sensors, Sensors and Actuators B: Chemical 139 (2) (2009) 304e309. [49] Z.S. Levin, C. Robert, J.F. Feller, M. Castro, J. Grunlan, Flexible latex d polyaniline segregated network composite coating capable of measuring large strain on epoxy, Smart Materials and Structures 22 (1) (2013) p. on line. [50] J. Li, J.P. Longtin, S. Tankiewicz, A. Gouldstone, S. Sampath, Interdigital capacitive strain gauges fabricated by direct-write thermal spray and ultrafast laser micromachining, Sensors and Actuators A: Physical 133 (1) (2007) 1e8. [51] L. Li, S. Liu, F. Ding, T. Hua, W. Man Au, K. Wong, Electromechanical analysis of length-related resistance and contact resistance of conductive knitted fabrics, Textile Research Journal 82 (20) (2012) 2062e2070. l

General introduction

xvii

[52] X. Lin, K. Zhou, X. Zhang, D. Zhang, Development, modeling and application of piezoelectric fiber composites, Transactions of Nonferrous Metals Society of China 23 (1) (2013) 98e107. [53] F. Lorussi, W. Rocchia, E.P. Scilingo, A. Tognetti, D. DeRossi, Wearable, redundant fabric-based sensor arrays for reconstruction of body segment posture, IEEE Sensors Journal 4 (6) (2004) 807e818. [54] J. Raush, E. M€ader, Health monitoring in continuous glass fibre reinforced thermoplastics: tailored sensitivity and cyclic loading of CNT-based interphase sensors, Composites Science and Technology 70 (13) (2010) 2023e2030. [55] J. Nevrela, M. Micjan, M. Novota, et al., Secondary doping in poly(3,4ethylenedioxythiophene):poly(4-styrenesulfonate) thin films, Journal of Polymer Science Part B: Polymer Physics 53 (16) (2015) 1139e1146. [56] D. Orata, M. Mukabi, H. Njenga, Electro-characterization of polypyrolle electrosynthesized on a montmorillonite host-matrix, in aqueous media containing sulphuric acid as supporting electrolyte, IOSR Journal of Applied Chemistry 7 (5) (2014) 59e72. [57] A. Ramanavicius, A. Malinauskas, Electrochemical sensors based on conducting polymer e polypyrrole, Electrochimica Acta 51 (27) (2006) 6025e6037. [58] J.V. Risicato, F. Kelly, D. Soulat, X. Legrand, W. Tr€ umper, C. Cochrane, V. Koncar, A complex shaped reinforced thermoplastic composite part made of commingled yarns with integrated sensor, Applied Composite Materials 22 (1) (2014) 81e98. [59] M. Rothmaier, M.P. Luong, F. Clemens, Textile pressure sensor made of flexible plastic optical fibers, Sensors 8 (7) (2008) 4318e4329. [60] E.P. Scilingo, F. Lorussi, A. Mazzoldi, D. De Rossi, Strain-sensing fabrics for wearable kinaesthetic-like systems, IEEE Sensors Journal 3 (4) (2003) 460e467. [61] A. Soroudi, M. Skrifvars, Melt blending of carbon nanotubes/polyaniline/polypropylene compounds and their melt spinning to conductive fibres, Synthetic Metals 160 (11) (2010) 1143e1147. [62] X. Tao, V. Koncar, T.H. Huang, C.L. Shen, Y.C. Ko, G.T. Jou, How to make reliable, washable, and wearable textronic devices, Sensors 17 (4) (2017) 673. [63] D.R. Thevenot, K. Toth, R.A. Durst, G.S. Wilson, Electrochemical biosensors: recommended definitions and classification, Pure and Applied Chemistry 71 (12) (2009) p. on line. [64] R. Matsuzaki, A. Todoroki, Wireless flexible capacitive sensor based on ultra-flexible epoxy resin for strain measurement of automobile tires, Sensors and Actuators A: Physical 140 (1) (2007) 32e42. [65] J.P. Wang, P. Xue, X.M. Tao, Strain sensing behavior of electrically conductive fibers under large deformation, Materials Science and Engineering A 528 (6) (2011) 2863e2869. [66] A. Weder, S. Geller, A. Heinig, T. Tyczynski, W. Hufenbach, W. Fischer, A novel technology for the high-volume production of intelligent composite structures with integrated piezoceramic sensors and electronic components, Sensors and Actuators A: Physical 1 (2013) 1e5. [67] S. Xiong, L. Zhang, X. Lu, Conductivities enhancement of poly(3,4,hylenedioxythiophene)/poly(styrene sulfonate) transparent electrodes with diol additives, Polymer Bulletin 70 (2013) 237e247. [68] P. Xue, K.H. Park, X.M. Tao, Electrically conductive yarns based on PVA/carbon nanotubes, Composite Structures 78 (2) (2007) 271e277.

xviii

General introduction

[69] P. Xue, X. Tao, H.Y. Tsang, In situ SEM studies on strain sensing mechanisms of PPy-coated electrically conducting fabrics, Applied Surface Science 253 (7) (2007) 3387e3392. [70] R. Zeiser, T. Fellner, J. Wilde, Academia Day, Development and Testing of Capacitive Strain Gauges, vol. 1, 2011, pp. 1e6. [71] R. Zhang, H. Deng, R. Valenca, J. Jin, Q. Fu, E. Bilotti, T. Peijs, Carbon nanotube polymer coatings for textile yarns with good strain sensing capability, Sensors and Actuators A: Physical 179 (2012) 83e91. [72] J. Zhou, G. Tzamalis, N.A. Zaidi, Electrically conductive PANI multifilaments spun by a wet spinning process, Journal of Materials Science 36 (13) (2001) 3089e3095. [73] J. Zieba, M. Frydrysiak, Textronics e electrical and electronic textiles, Sensors for Breathing Frequency Measurement, Fibres and Textiles in Eastern Europe 14 (5) (2006) 43e48. [74] K. Kim, C.S. Park, D.W. Lee, S.M. Cho, C.R. Han, Measurement of the gauge factor of carbon fiber and its application to sensors, Microelectronic Engineering 85 (5) (2008) 787e791. [75] U. Lang, N. Naujoks, J. Dual, Mechanical characterization of PEDOT: PSS thin films, Synthetic Metals 159 (5) (2009) 473e479. [76] N. Trifigny, F.M. Kelly, C. Cochrane, V. Koncar, et al., In-situ measurements of strain and stress of e-glass warp interlock fabric using innovative sensors, in: CompositeWeek@ Leuven(TexComp-11): Proceedings, Leuven, 2013. [77] S. Sha, X. Tao, F. Boussu, V. Koncar, Mechanical properties of high modulus UHMW yarn measured by intelligent sensors, in: 13th AUTEX World Textile Conference Book of Proceedings, Dresden, 2013. [78] Co Heraeus Precious Metals GmbH, Coating Guide CleviosÔ P Formulations, Heraeus, Hanau, 2012. [79] S.L. Ogin, Textile-reinforced composite materials, in: A.R. Horrocks, S.C. Anand (Eds.), Handbook of Technical Textile, Woodhead Publishing Limited, London, 2000, pp. 264e279. [80] J. M€ader, E. Rausch, Health monitoring in continuous glass fibre reinforced thermoplastics: manufacturing and application of interphase sensors based on carbon nanotubes, Composites Science and Technology 70 (11) (2010) 1589e1596. [81] R. Kamiya, et al., Some recent advances in the fabrication and design of three-dimensional textile preforms: a review, Knowledge: Creation, Diffusion, Utilization 60 (2000) 33e47. [82] B. Cox, M. Dadkhah, The macroscopic elasticity of 3D woven composites, Journal of Composite Materials 29 (6) (1995) 785e819. [83] B.K. Behera, P.K. Hari, Woven Textile Structure: Theory and Application, Woodhead Publishing Limited, Cambridge, 2010. [84] P.T. Curtis, S.M. Bishop, An assessment of the potential of woven carbon fibre-reinforced plastics for high performance applications, Composites 15 (4) (1984) 259e265. [85] The Hong Hong Cotton Spinncer Association, Textile Handbook, Hong Kong Productivity Council, Hong Kong, 2001. [86] M. Rao, B. Sankar, G. Subhash, Effect of Z-yarns on the stiffness and strength of threedimensional woven composites, Composites Part B: Engineering 40 (6) (2009) 540e551. [87] Y. Mahadik, K. Brown, A.S. Hallett, Characterisation of 3D woven composite internal architecture and effect of compaction, Composites Part A, Applied Science Manufacture 41 (7) (2010) 872e880.

General introduction

xix

[88] M. Ansar, W. Xinwei, Z. Chouwei, Modeling strategies of 3D woven composites: a review, Composite Structures 93 (8) (2011) 1947e1963. [89] S. Buchanan, A. Grigorash, E. Archer, A. McIlhagger, J. Quinn, G. Stewart, Analytical elastic stiffness model for 3D woven orthogonal interlock composites, Science and Technology 70 (2010) 1597e1604. [90] X. Chen, L. Taylor, L. Tsai, An overview on fabrication of three-dimensional woven textile preforms for composites, Textile Research Journal 81 (9) (2011) 932e944. [91] S. Nehme, A. Hallal, F. Fardoun, R. Younes, B. Hagege, Z. e. a. Aboura, Numerical/ analytical methods to evaluate the mechanical behavior of interlock composites, Journal of Composite Materials 45 (16) (2011) 1699e1716. [92] F. Boussu, The use of warp interlock fabric inside textile composite protection against ballistic impact, Textile Research Journal 81 (4) (2011) 344e354. [93] S. Lomov, I. Verpoest, F. Robitalle, Manufacturing and internal geometry of textiles, in: Design and Manufacture of Textile Composites, Woodhead Publishing Limited, London, 2005, pp. 1e61. [94] K. Leong, B. Lee, I. Herszberg, M. Bannister, The effect of binder path on the tensile properties and failure of multilayer woven CFRP composites, Composites Science and Technology 60 (2000) 149e156. [95] J. Solden, B. Hill, Conventional weaving of shaped preforms for engineering composites, Composites Part A 29 (1998) 757e762. [96] S. Rudov-Clark, A.P. Mouritz, L.L.M. Bannister, Fiber damage in the manufacture of advanced three-dimensional woven composites, Composites Part A, Applied Science Manufacture 34 (10) (2003) 963e970. [97] V. Guenon, T. Chou, J. Gillespie, Toughness properties of a three-dimensional carbonepoxy composite, Journal of Materials Science 24 (1989) 4168e4175. [98] B.N. Cox, M.S. Dadkhah, W.L. Morris, J.G. Flintoff, Failure mechanisms of 3D woven composites in tension, compression, and bending, Acta Materialla Material 42 (12) (1994) 3967e3984. [99] S.V. Lomov, D. Ivanov, G. Perie, I. Verpoest, Modelling 3D fabrics and 3D reinforced composites: challenges and solutions, in: 1st World Conference on 3D Fabrics, Manchester, 2011. [100] M. Lefebvre, F. Boussu, D. Coutelier, Influence of high-performance yarns degradation inside three-dimensional warp interlock fabric, Journal of Industrial Textiles 42 (2013) 475e488. [101] J. Ruzand, G. Guenot, Multiaxial Three-Dimensional Fabric and Process for its Manufacture, 1994. International Patent Brevet WO: 94/20658. [102] G. Mood, Multi-Axial Yarn Structure and Weaving Method, 1996. United States Brevet 5-540-260. [103] L. Bryn, M.A. Islam, W.L. Lowery, H.D. Harries, Three Dimensional Woven Forms with Integral Bias Fibers and Bias Weaving Loom, 2004. United States Brevet 6-742-547. [104] S. Nayfeh, J. Rohrs, O. Rifni, et al., Bias Weaving Machine, 2006. United States Brevet 7-077-167. [105] M. Lima, R. Fangueiro, A. Costa, C. Rosiepen, V. Rocha, Multiweave, a prototype weaving machine for multiaxial technical fabrics, Indian Journal of Fibre and Textile Research 34 (2009) 59e63. [106] G.L. Farley, Method and Apparatus for Weaving a Woven Angle Ply Fabric, 1993. United States Brevet 5,224,519.

xx

General introduction

[107] M. Anahara, Y. Yasui, H. Omori, Three-Dimensional Textile and Method of Producing the Three Dimensional Textile, 1991. Europe Brevet 0-426-878-A1. [108] Y. Yasui, M. Anahara, F. Hori, Y. Mita, Production of Three-Dimensional Woven Fabric, 1993. Japan Brevet 05-106-140. [109] H. Uchida, T. Yamamoto, H. Takashima, T. Yamamoto, S. Nishlyama, M. Shinya, Three-Dimensional Weaving Machine, 1999. United States Brevet 6-003-563. [110] A.K. Bilisik, Multiaxis three dimensional woven fabric, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabric Technology, Intech, Athens, 2011 p. on line. [111] A.K. Bilisk, M.H. Mohamed, Multiaxis three-dimensional (3D) flat woven preform-tube Carrier weaving, Textile Research Journal 80 (8) (2010) 696e711. [112] A.K. Bilisik, Multiaxial Three Dimensional (3D) Woven Carbon Preform, 2000. United States Brevet 6-129-122. [113] K. Bilisik, Three-dimensional braiding for composites: a review, Textile Research Journal 83 (13) (2012) 1414e1436. [114] P. Stickler, M. Ramulu, P. Johnson, Experimental and numerical analysis of transverse stitched T-joints in bending, Composite Structures 50 (1) (2000) 17e27. [115] A.P. Mouritz, B. Cox, A mechanistic approach to the properties of stitched laminates, Composites Part A 31 (1) (2000) 1e27. [116] A.P. Mouritz, C. Baini, I. Herszberg, Mode I interlaminar fracture toughness properties of advanced textile fiber glass composites, Composites Part A, 30 (7) (1999) 859e870. [117] F. Edgren, C. Soutis, L. Asp, Damage tolerance analysis of NCF composite sandwich panels, Composites Science and Technology 68 (2008) 2635e2645. [118] G. Bibo, I.K.M. Hogg, Mechanical characterization of glass- and composites made with non-crimp fabrics, Composites Science and Technology 57 (1997) 1221e1241. [119] P. Chang, A.P. Mouritz, B.N. Cox, Elevated temperature properties of pinned composite lap joints, Journal of Composite Materials 42 (8) (2008) 741e769. [120] A.P. Mouritz, Compression properties of z-pinned composite laminates, Composites Science and Technology 67 (15) (2007) 3110e3120. [121] M. Niggemann, M. Pellmann, P. Bolsmann, J. Reinbold, K. Cammann, Remote sensing of tetrachloroethene with a micro-fibre optical gas sensor based on surface plasmon resonance spectroscopy, Sensors and Actuators B: Chemical 34 (1) (1996) 328e333. [122] E.D. Witten, T. Kraus, M. Kuhnel, Composite Market Report 2017, November 28, 2016 [En ligne]. Available: http://www.eucia.eu/userfiles/files/20161128_market_report_ 2016_english.pdf. [123] E.D. Witten, T. Kraus, M. Kuhnel, Composite Market Report 2017, November 15, 2016 [En ligne]. Available: http://www.eucia.eu/userfiles/files/20161128_market_report_ 2016_english.pdf. [124] C. Warren, Lower Cost High Performance Carbon Fibre, U.S. Department of Enenrgy, Oak Ridge, 2011. [125] F. Composites, Aramid Fiber Stronger and Safer with Aramid Fiber, Fibermax Composites, 2017 [En ligne]. Available: http://www.aramid.eu. [126] R. i. China, Global and China Aramid Fiber Industry Report, 2015e2018, Research in China, Beijing, 2015. https://www.reportlinker.com/p02515745/Global-and-ChinaAramid-Fiber-Industry-Report.html. [127] Marketsandmarkets.com, Aramid Fiber Market by Type (Para-aramid, Meta-aramid, and Others), and Application (Security & Protection, Frictional Materials, Optical Fiber, Tire Reinforcement, Rubber Reinforcement, Electrical Insulation, Aerospace, and Others) e Global Trends & Forecasts to 2021, Marketsandmarkets, London, 2016.

General introduction

xxi

[128] R. linker, Global Natural Fiber Composite Market 2015e2020: Trends, Forecast, and Opportunity Analysis, December 2015, March 30, 2016 [En ligne]. Available, http:// www.prnewswire.com/news-releases/global-natural-fiber-composite-market-2015-2020trends-forecast-and-opportunity-analysis-december-2015-300243712.html. [129] N. Institute, WPC/NFC Market Study, NOVA Institute, H€ urth, 2014. [130] P. Irving, C. Soutis, Polymer Composites in the Aerospace Industry, Woodhead Publishing Limited, London, 2014. [131] P. Pasbakhsh, A. Haghi, E. Zaikov, Applied Research on Polymer Composites, AAP Research Notes on Polymer Engineering, Science and Technology, 2015. [132] G. Koronis, A. Silva, M. Fontul, Green composites, a review of adequate materials for automotive applications, Composites Part B 44 (2013) 120e127. [133] M. Reddy, S. Vivekanandhana, M. Misraa, S. Bhatiac, A. Mohantya, Biobased plastics and bionanocomposites: current status and future opportunities, Progress in Polymer Science 38 (2013) 1653e1689. [134] C.R. Farrar, K. Worden, An introduction to structural health monitoring, Philosophical Transactions of the Royal Society A 365 (2007) 303e315. [135] A.E. Aktan, F. Catbas, K. Grimmelsman, M. Pervizpour, Development of a Model Health Monitoring Guide for Major Bridges, Drexel Intelligent Infrastructure and Transportation Safety Institute, Drexel, 2003. [136] O. Buyukozturk, T.Y. Yu, Structural health monitoring and seismic impact assessment, in: Proc. 4th Nat. Conf. Earthquake Eng., Istanbul, 2003. [137] S. Nassar, X. Yang, Fastening and joining of composite materials, in: E. Patterson, D. Backman, G. Cloud (Eds.), Composite Materials and Joining Technologies for Composites, Proceedings of the 2012 Annual Conference on Experimental and Applied Mechanics, vol. 7, Springer New York, New York, 2012, pp. 5e23. [138] M. Ueda, S. Miyake, H. Hasegawa, Y. Hirano, Instantaneous mechanical fastening of quasi-isotropic CFRP laminates by a self-piercing rivet, Composite Structures 94 (2012) 3388e3393. [139] V. Virupaksha, S. Nassar, Effect of washers and bolt tension on the behavior of doublelap S2-glass fabric-epoxy composite joints, in: International Conference on Pressure Vessels and Piping (PVP 2008), Chicago, IL, USA, July 27e31, 2008. [140] D. Liu, Y. Tang, W. Cong, "A review of mechanical drilling for composite laminates, Composite Structures 94 (2012) 1265e1279. [141] J. Oh, Y. Kim, D. Lee, Optimum bolted joints for hybrid composite materials, Composite Structures 38 (1e4) (1997) 329e341. [142] J. Gebhardt, J. Fleischer, Experimental investigation and performance enhancement of inserts in composite parts, in: 5th CATS (Conference on Assembly Systems and Technologies) 2014-CIRP, Procedia CIRP 23, Dresden, Germany, November 12e14, 2014. [143] C. Dufour, F. Boussu, P. Wang, D. Soulat, K. Kalnins, E. Labans, P. Lefort, M. Wallenius Henrinksson, 3D warp interlock fabric structure as a solution of composite joining with metallic part, in: Proceedings of the 6th World Conference on 3D Fabrics and Their Applications, Raleigh, NC e USA, 2015. [144] C. Ageorges, L. Ye, M. Hou, Advances in fusion bonding techniques for joining thermoplastic matrix composites: a review, Composites: Part A 32 (2001) 839e857. [145] J. Esteves, S. Goushegir, J. dos Santos, L. Canto, E. Hage Jr., S. Amancio-Filho, Friction spot joining of aluminum AA6181-T4 and carbon fiber-reinforced poly(phenylene sulfide): effects of process parameters on the microstructure and mechanical strength, Materials and Design 66 (2015) 437e445.

xxii

General introduction

[146] S. Ucsnik, M. Scheerer, S. Zaremba, D. Pahr, Experimental investigation of a novel hybrid metalecomposite joining technology, Composites: Part A 41 (2010) 369e374. [147] A. Bogdanovich, M. Dannemann, J. D€oll, T. Leschik, J. Singletary, W. Hufenbach, Experimental study of joining thick composites reinforced with non-crimp 3D orthogonal woven E-glass fabrics, Composites: Part A 42 (2011) 896e905. [148] Z. Huang, S. Sugiyama, J. Yanagimoto, Applicability of adhesiveeembossing hybrid joining process to glass-fiber-reinforced plastic and metallic thin sheets, Journal of Materials Processing Technology 214 (2014) 2018e2028. [149] Z. Huang, S. Sugiyama, J. Yanagimoto, Adhesiveeembossing hybrid joining process to fiber-reinforced thermosetting plastic and metallic thin sheets, in: 11th International Conference on Technology of Plasticity, ICTP 2014, Procedia Engineering, Nagoya, Japan, 19e24 October, 2014. [150] M. Gude, W. Hufenbach, R. Kupfer, A. Freund, C. Vogel, Development of novel formlocked joints for textile reinforced thermoplastics and metallic components, Journal of Materials Processing Technology 216 (2015) 140e145. [151] B. Cosson, M. Deléglise, W. Knapp, Numerical analysis of thermoplastic composites laser welding using ray tracing method, Composites: Part B 68 (2015) 85e91. [152] E. Rodríguez-Vidal, J. Lambarri, C. Soriano, C. Sanz, G. Verhaeghe, A combined experimental and numerical approach to the laser joining of hybrid polymer e metal parts, in: 8th International Conference on Photonic Technologies LANE 2014, Physics Procedia, Furth, Germany, 8e11 September, 2014. [153] P. Cognard, Assemblage des composites e Les points forts du collage, Techniques de l’Ingénieur, traité Génie mécanique, Juillet 2003, pp. 1e4, n %1BM 7 624. [154] P. Cognard, Collage des composites e Secteurs routier, ferroviaire, Techniques de l’Ingénieur, traité Génie mécanique, October 2003, pp. 1e7, n %1BM 7 627. [155] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Procédé pour l’assemblage entre une piece en matériau métallique, une piece en matériau composite a matrice organique, December 19, 2014. France Brevet FR3 030329 A1. [156] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Assembly Method Between a Part Made of Metal Material and a Part Made of Organic Matrix Composite Material; Corresponding Parts Made of Organic Matrix Composite Material and Assembly, December 23, 2016. France Patent WO 2016/097656 A1. [157] C.Q. Yang, Q. He, R.E. Lyon, et al., Investigation of the flammability of different textile fabrics using micro-scale combustion calorimetry, Polymer Degradation and Stability 95 (2) (2010) 108e115. [158] R. Gupta, V. Kumar, P.K. Goyal, et al., Thermogravimetric analysis of heat treated CR-39 polymer, Journal of Chemical and Pharmaceutical Research 2 (4) (2010) 629e634. [159] N. Amir, F. Ahmad, P.S.M. Megat-Yusoff, Study on the fibre reinforced epoxy-based intumescent coating formulations and their char characteristics, Journal of Applied Sciences 11 (10) (2011) 1678e1687. [160] R.E. Lyon, R. Walters, Flammability of Automotive Plastics, Federal Aviation Administration William J. Hughes Technical Center, New Jersey, 2005. [161] F. Laoutid, L. Bonnaud, M. Alexandre, New Prospects in Flame Retardant Polymer Materials: From Fundamentals to Nanocomposites, Materials Science and Engineering: R: Reports, 2009. [162] A.M. Grancaric, A. Tarbuk, E. Chibowski, Surface free energy of textiles, Tekstil 57 (1) (2008) 29e39.

General introduction

xxiii

[163] K.E. Jackson, L. Edwin, R. Fasanella, L. Boitnott, K. Lyle, Full scale crash test and Finite Element Simulation of a composite prototype helicopter, Aircraft Design, Testing and Performance 1 (2003) 121e129, n %11. [164] J. Obradovic, S. Boriab, G. Belingardia, Lightweight design and crash analysis of composite front impact energy absorbing structures, Composite Structures 94 (2012) 423e430, n %12. [165] E. Mehdi, A. Mokhtar, D. Magid, F. Ahmadun, A. Prithvi Raj, S. Taher, A new composite energy absorbing system for aircraft and helicopter, Composite Structures 75 (2006) 14e23, n %11-4. [166] A.M. Grancaric, A. Tarbuk, I. Jerkovic, et al., Surface free energy of multilayered weftknitted fabrics and related composite plates, in: CompositeWeek@Leuven (TexComp11): Proceedings, Leuven, 2013. [167] A.M. Grancaric, J.V. Risicato, I. Jerkovic, et al., Interface phenomena of PP/glass braided structures, in: XIIIth International Izmir Textile and Apparel Symposium Book of Proceedings, Izmir, 2014. [168] A.M. Grancaric, V. Kova
cevic, M. Leskovac, et al., Interface phenomena of hydrolyzed polyester fabric, in: 2nd International Textile, Clothing & Design Conference e Magic World of Textiles Book of Proceedings, Dubrovnik, 2004. [169] E. Chibowski, L. Holysz, G.A.M. Kip, et al., Surface free energy components of glass from ellipsometry and zeta potential measurements, Journal of Colloid and Interface Science 132 (1) (1989) 54e61. [170] C. Faulkner, What is NFC? Everything You Need to Know, Techreader, May 9, 2017 [En ligne]. Available: http://www.techradar.com/news/what-is-nfc. [171] N. Pelly, J. Hamilton, How to NFC, May 10, 2011 [En ligne]. Available: https://www. youtube.com/watch?v¼49L7z3rxz4Q. [172] I. p. NFC.org a Square, Near Field Communication Versus Bluetooth, November 28, 2012 [En ligne]. Available: http://nearfieldcommunication.org/bluetooth.html. [173] S. Nauman, P. Lapeyronnie, I. Cristian, F. Boussu, V. Koncar, On line measurement of structural deformations in composites, IEEE Sensors 11 (6) (2010) 1329e1336. [174] I. Jerkovic, V. Koncar, A.M. Grancaric, New textile sensor for in situ structural health monitoring of textile reinforced thermoplastic composites based on the conductive poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) polymer complex, Sensors 17 (2017) 1e21, n %12297.

This page intentionally left blank

Smart textiles for monitoring and measurement applications 1.1

1

Introduction

The chapter is dedicated to the presentation of smart textile concepts in general, with the focus on monitoring and measurement applications. An overview of sensory technologies is given together with basic definitions and classifications. More specifically, fibrous textile sensors adapted to the integration to composite structures for monitoring purposes are developed, characterized, and described.

1.2

Smart textiles

Smart textiles are able to sense different stimuli (signals), to respond and adapt their behavior to them in an intelligent, or at list logical way [1,2]. The stimuli can be thermal, mechanical, chemical, electrical, magnetic, optical, etc. The functional activity of these materials is an important aspect. However, there is no agreement about the definition of “smart.” Among other better-known terms for the same purposes are adaptive, intelligent, interactive, responsive, connected, and multifunctional. Some of the first usages of smart textiles were in military and medical applications [1]. There is an increasing trend of integrating intelligence in our daily environment. These systems can provide customers with information needed in the current situations and help to manage everyday life more efficiently. Smart textiles present a challenge in several fields such as sport, artistic communities, medical, security and safety, railway, automotive, and aerospace. The term “smart textile” refers to a broad field of studies and products that extend the functionality and usefulness of common fabrics. Smart textiles are defined also as textile products such as fibers and filaments, yarns together with woven, knitted or nonwoven structures, which can interact with the environment/user. The convergence of textiles and electronics (e-textiles or textronics) can be relevant for the development of smart materials that are capable of accomplishing a wide spectrum of functions, found in rigid and nonflexible electronic products nowadays. Fundamental components in smart textile include sensors, actuators, data processors, communication units, and energy supply. Besides, smart textiles have some benefits such as noninvasive and continuous monitoring; besides the most important requirements for smart textiles include soft and flexible sensors production. Advanced functionalized materials, such as breathing, fire-resistant or ultra strong fabrics are not considered as intelligent, no matter how high-technological they might be.

Smart Textiles for In Situ Monitoring of Composites. https://doi.org/10.1016/B978-0-08-102308-2.00001-2 Copyright © 2019 Elsevier Ltd. All rights reserved.

2

Smart Textiles for In Situ Monitoring of Composites

According to the applications developed during the last decades, smart textiles can be classified as: 1. Passive Smart Textilesdthe first generation of smart textiles found in clothing, which can detect environmental or body conditions; 2. Active Smart Textilesdthe second generation has both actuators and sensors with the capacity to sense (detect) and to react (chameleonic, water-repellent, thermoregulated, etc.); 3. Very Smart Textilesdthe third generation of smart textiles, which can sense, react, and adopt themselves to the given circumstances (stimuli).

The first generation of smart textiles uses conventional materials and tries to adapt the textile design to fit in the external elements. They can be considered as e-apparel, where electronics are added to the textile. The first successful step toward wearability was the ICDþ line at the end of the 1990s of 20th century, resulting from a cooperation between Levi’s and Philips. This line’s coat architecture was adapted in such a way that existing apparatus could be put away in the coat: microphone, earphone, remote control, mobile phone, and MP3 player. All these components were carefully removed from the coat before it went into washing machine.

1.3

Sensorsddefinitions and classifications

Sensor is defined as device providing information mostly in the form of an electrical signal. It can sense the measured object or medium and emit a signal related to the variations of the measured quantity [3]. However, for our application a major problem is to maintain sensors’ reliability and stability to cleaning, washing, etc. [4]. Chemical sensors are miniaturized analytical devices that can deliver real time and online information on the presence of specific compounds or ions in complex samples [5]. Biosensors are self-contained integrated devices [6] that are capable of providing specific quantitative or semiquantitative analytical information using a biological recognition element (biochemical receptor) that is in direct spatial contact with transducer element. The selectivity and sensitivity provided by nature have been utilized in such sensors, frequently by immobilizing the biologically active compounds, such as enzymes and immunoglobulins, within a receptor part of the sensor. An effective way of obtaining the biological selectivity is a combination of cell cultures, tissue slices, organs, and sometimes of whole living organisms with the transducer. The whole effect of a biosensor is to transform a biological event into an electrical signal. Biosensors found extensive applications in medical diagnostics, environmental pollution control for measuring toxic gases in the atmosphere, and toxic soluble compounds in river water. Biosensors that include transducers based on integrated circuit microchips are known as biochips. Optical sensors, or opt(r)odes, represent a group of chemical sensors in which electromagnetic (EM) radiation is used to generate the analytical signal in a transduction element. The interaction of this radiation with the sample is evaluated from the change of a particular optical parameter and is related to the concentration of the analyte. Typically, an optical chemical sensor consists of a chemical recognition phase (sensing element or receptor) coupled with a transduction element (Fig. 1.1). The receptor

Smart textiles for monitoring and measurement applications

3

Interferent Membrane Optical transducer

A

Signal processing

A A

A

Light

Measurable signal

A (analyte)

Sample

Receptor chemical recognition phase

Figure 1.1 Schematic representation of the composition and function of an optical chemical sensor.

identifies a parameter, e.g., the concentration of a given compound, pH, and provides an optical signal proportional to the magnitude of this parameter. The function of the receptor is fulfilled in many cases by a thin layer that can interact with the analyte molecules, catalyse a reaction selectively, or participate in a chemical equilibrium together with the analyte. The transducer translates the optical signal produced by the receptor into a measurable signal that is suitable for processing by amplification, filtering, recording, display, etc. [7,8,9]. Optical sensors can be based on various optical principles (absorbance, reflectance, luminescence, fluorescence), covering different regions of the spectra (UV, Visible, IR, NIR) and allowing the measurement not only of the intensity of light, but also of other related properties, such as lifetime, refractive index, scattering, diffraction, and polarization. These sensors have numerous advantages over conventional electricity-based sensors, such as selectivity, immunity to EM interference, and safety while working with flammable and explosive compounds. They are also sensitive, inexpensive, nondestructive, and have many capabilities. Optical sensors also exhibit disadvantages: ambient light can interfere with their operation, the long-term stability is limited due to indicator leaching or photobleaching, there may be a limited dynamic range, selectivity may be poor, and a mass transfer of the analyte from the sample into the indicator phase is necessary to obtain an analytical signal. Fiber-optic chemical sensors represent a subclass of chemical sensors in which an optical fiber is commonly employed to transmit the EM radiation to and from a sensing region that is in direct contact with the sample. The spectroscopically detectable optical property can be measured through the fiber optic arrangement, which enables remote sensing. In addition to advantages in terms of cheapness, ease of

4

Smart Textiles for In Situ Monitoring of Composites

Figure 1.2 Commonly used textile sensors: (a) textile composite samples containing piezoresistive sensors, (b) pressure sensor used as an on/off switch for a light emitting diode [12].

miniaturization, obtaining safe, small, lightweight, compact and inexpensive sensing systems, and a wide variety of sensor designs are possible [10]. The receptor in a chemical sensor usually contains a thin layer where the analyte interacts with the sensor. The receptor part may be based on various principles: physical, chemical, and biochemical [11]. There are lot of common sensors existing for the application in smart textiles [8], but piezoresistive and pressure sensors (Fig. 1.2) are the most usually used. Piezoresistive sensors can indicate the stress or strain change during time by recording resistance change of the material [10]. Pressure sensors are commonly used either as switches and interfaces with electronic devices or to monitor vital signs of the user. As most signals transmitted by the sensors are in electrical form, the most effective way to create a textile sensor is therefore by using electroconductive materials [12]. Sensors incorporation into textile is divided in two specific groups as follows (Fig. 1.3): 1. Removable sensors developed mostly for clothing applications: usage of velcom tape, zipper, sewing; 2. Fixed sensors for technical textile applications: weaving, knitting, laminating, sensor nanofibers, dyeing, printing, coating [11].

The modification of the fibers can give the required electrical characteristics to serve the heating, protecting, and signal-transmitting operations. The conductive character of the fibers supports the multifunctionality of the end products. For a very long historical period the textile fibers were used almost exclusively for clothing applications. Step by step, the technical uses started to be more important and electrically conductive textile fibers are one of the most recent materials for technical applications.

Smart textiles for monitoring and measurement applications

5

Figure 1.3 Sensors incorporation into textile: (a) removable sensor, (b) fixed sensor [11].

Conductive fibers can be classified into two broad categories: 1. naturally conductive, 2. specially treated or functionalized to create conductivity.

Metallic fibers as naturally conductive fibers are developed from electrically conductive materials such as ferrous alloys, nickel, stainless steel, titanium, aluminum, copper, etc. Other naturally conductive fibers are those made of inherently conducting polymers and carbon fibers. Specially treated fibers represent metal-coated fibers, fibers filled or loaded with carbon of metallic salts, bicomponent fibers, etc.

1.3.1 1.3.1.1

Mechanical sensorsdgeneral definitions Strain gauges

The classical system for strain (and deformation) measurements is the strain gauge. Basically, it is made of thin metallic foil and the electrical insulation thin backing substrate. The gauge is attached to the object by a suitable adhesive, such as cyanoacrylate. A metallic sheet is cut (by lithography or using the acid) to form extremely tight conductive paths in parallel making very long overall conductive path sensitive to elongations in its direction. This geometry (Fig. 1.4) maximizes the length of electrical circuit connecting two electrodes on a reduced surface. When the gauge is elongated in the direction of conductive path, it is deformed (lengthened) and the electrical resistance increases. The lengthening of the conductive circuit increases the value of its electrical resistance. The variation of the electrical resistance in function of the geometrical deformation outlines the gauge factor K [13] defines as follows: DR R K¼ 0 ε

6

Smart Textiles for In Situ Monitoring of Composites

Stress direction

Figure 1.4 Classical strain gauge.

Equation 1.1: Gauge factor.where DR is the variation of electrical resistance between the moment t and the initial resistance defined by DR ¼ Rt  R0, and ε is the geometrical deformation relative to the difference between the initial length L0 and the length at the moment t, Lt divided by the initial length defined as follows: ε¼

L t  L0 L0

Equation 1.2: Elongation expression. In case of piezoelectrical materials, the gauge factor is directly related to mechanical properties of the material within its Poisson coefficient n. It may be expressed by the following equation: Kl ¼ 1 þ 2 n Equation 1.3: Gauge factor in longitudinal direction.

1.3.2

Capacitive sensors

An alternative solution to classical strain gauges and a measuring of their resistance changes is a capacitive sensor. Its electrical capacitance varies in function of elongation. Therefore, capacitive sensors measure the electrical capacitance between two conductors deposited on the insulation substrate. The advantage of those sensors is a very high gauge factor (K) going up to 15 to 30 and in the same time the capacitive sensors are less sensitive to an electronic noise and to the temperature changes than resistive sensors. A development of the technology used for capacitive sensors started in 1940s [15]. Nowadays, it is possible to find on the market distinct types of capacitive gauges, even

Smart textiles for monitoring and measurement applications

7

Figure 1.5 Capacitive flexible strain gauge on polyamide, Zeiser et al. [14].

flexible ones (Fig. 1.5) using technologies based on thin films. Therefore, the possible fields of applications are extended [15,16,17]. Research works of R. Zeiser et al. were focused on this type of flexible capacitive strain gauge where two polymeric substrates have been used as a base for capacitive flexible strain gauge. The first one was a polyamide and the second one the liquid crystal polymer. To form the sensor, a thin layer of chromeecopper alloy is deposited on the polymer film by lithography. The geometry of this metallic track plays a key role in the sensor properties. In this study as well as in many recent studies the interdigital geometry is used (Fig. 1.6). The spacing among electrodes is 15 mm and their width 45 mm. This geometry is supposed to optimize the sensor sensitivity. The lengthening provokes a decrease of the sensor capacitance proportionally. These sensors are adapted to the measure of small deformations ranging from 0.1% to 0.5% with the gauge factor of 1.38 (capacitance decreases with the elongation). This study also shows that the sensors performances, their sensitivities, and noise rejection depend strongly on the substrate used.

1.3.3

Piezoelectric sensors

Piezoelectric materials generate an electrical field when they are mechanically excited and inversely they deform when they undergo an electrical current. This phenomenon may be explained by their crystalline structure. A large number of materials exhibit this property, and they could be classified into different families: ferroelectric oxides, quartz similar compounds, semiconductors, polymers, and ceramics. They can be found in different forms (independently on the chemical family), composites, or thin layers. The ceramic family is the most represented, and the most used material currently is PZT Led titanium zircon, for its high coefficient of piezoelectricity and its thermal stability. Those materials are able to play a role of mechanical actuators or strain gauges. Render et al. [18] used a flexible PVDF film (25 mm) that has been coated by

8

Smart Textiles for In Situ Monitoring of Composites

(a)

Strain direction x

Attachment area Copper electrode Flexible resin

ws

ls

(b)

ws

ls

Figure 1.6 Schematic representation of capacitive gauges with interdigital, Matsuzaki and Todoroki [17].

aluminum on both sides for electrical contacts. It has been directly glued to a parachute canopy. Tests in laboratory conditions have shown a quasi linear response up to 7% of elongation. The previously developed sensory technology has been transferred to fibers and threads to make it adapted and compatible with composites monitoring in situ [19,20,21]. Lin et al. has developed in 2013 [19] and applied piezoelectric strain gauges composed of fibers exhibiting better properties than those of mono filament piezoelectric material. The innovative structure that has been proposed is composed of piezoceramic and polymeric fibers guarantying optimal geometrical adaptability and bringing more longitudinal resistance to mechanical stresses comparing to simple polymeric thread. This “composite” sensory structure is generally composed of PZT piezoelectric fibers aligned and surrounded by polymeric matrix. This sensor is then located between two polyimide films with previously printed conductive interdigitally disposed capacitive electrodes (Fig. 1.7). Electrodes are used to measure the electrical

Smart textiles for monitoring and measurement applications

9

0.34 mm

Electrode 0.25 mm

PZT fibres Matrix

Figure 1.7 Cross section of an active fibers composite (Lin et al. [19]).

field generated by the “composite” sensor, the whole structure is commonly called Active Fibers Composite (AFC). However, many drawbacks could be observed related to AFC structure, mainly correlated to a relative fragility of fibers that break easily and are very difficult to manipulate. Their size is very small between 100 and 250 mm in diameter. Also, a polymeric matrix is not easy to set up, particularly due to air bulbs among fibers and matrix. They also increase a risk of electrical reliability. Moreover, the polymer matrix needs high voltages necessary to electric field to be recorded between electrodes and fibers limiting consecutively the possibilities of utilization of those sensors for composites structural health monitoring in situ. Those difficulties have partially been overwhelmed by the development of Macro Fibers Composites, where the cross section of fibers is rectangular rather than circular. This technique improves the filling rate of fibers inside the matrix and to increase the contact surface between the yarn composed of fibers and the film of electrodes. Mechanical properties are also better, and sensing or actuating efficiency increases up to 50%.

1.3.4

Optical fibers based sensors

An optical fiber is composed of two concentric cylindrical wave guides, with different refraction indexes. The internal wave guide has its refraction index higher than the external one, making the light beam remain inside it. Optical fibers are mainly used for information transfer, but they can also be used as strain gauges. In that case, optical fibers structure has to be modified to be sensitive to the elongation. Various structures exist as stated below: • • •

Bragg network; Multimodal distribution; Biconically tapered single-mode fiber.

Bragg network is a periodical modification of the refraction index of the core (internal cylinder) of the optical fiber [22]. Its periodical modification acts as a mirror inside the fiber, but only for a defined wavelength light beam, which is associated to the period of refraction index alterations (Fig. 1.8). When the optical fiber sensor is elongated, the step of the Bragg net is modified (series of alterations) and consecutively a light beam with another wavelength is reflected.

10

Smart Textiles for In Situ Monitoring of Composites

Input signal

Optical fiber

Bragg network

Reflected signal

Transmitted signal

Figure 1.8 Bragg network functioning.

Main application of those sensors concerns composite structures health monitoring and control in maritime and aerospace fields. This may be explained by relative facility of their integration to a composite part during the resin infusion process. The interface between the monitored part and the sensors is almost perfect. They may however weaken the overall structure by their presence, in the case of an important difference between the reinforcement tows and optical fibers mechanical properties. Bragg effectebased sensors may also be used to develop planar strain gauges. The fiber is twisted to make a triangular rosette, where each side represents a particular Bragg network. Some multimodal distribution optical fiber sensors relay on the measuring of light intensity variations due to the geometrical modification of the optical fiber when stressed. It may also relay on the refraction index variations of the multimode light wave guide in case of external mechanical disturbance. Biconically tapered single-mode fiber sensor, shown in Fig. 1.9, and functioning principle for different applications are based on the research results of Arregui et al. [23]. X

ρcore (z)

Z

ρcladding (z) ρ (z)

Y Zone I

Zone II

Zone III

Figure 1.9 Scheme of the biconically tapered single-mode fiber [23].

Smart textiles for monitoring and measurement applications

11

0

Output optical power (dB)

–5

40 mm

–10

30 mm

–15 –20

22 mm

–25 –30 –35 0

100

200 300 Displacement (micro m)

400

500

Figure 1.10 Optical output power (for a wavelength of l ¼ 1550 nm) in function of the displacement for diameters of 40, 33, and 22 mm [23].

The range of possible elongations is rather limited. The authors mentioned maximal deformation of 87,500 mε (microstrains) for a variation of 19 dB (output optical power). The elongation sensor output is not linear and depends on the diameter of the reduced zone (Fig. 1.9). The sensor dimensions, particularly a diameter of the reduced zone, should be defined in function of the measuring range. More the diameter reduction is small, more the sensor will be able to measure small elongations. For instance, a sensor with a diameter of 22 mm exhibits quasi linear behavior (Fig. 1.10) and corrects gauge factor for elongations ranging from 50 to 100 mm. The light wavelength that is used influences the output concerning the graphs shift, but the overall behavior remains comparable. In conclusion, optical fiber sensors are quite interesting, specifically because of their immunity to an EM noise and a possibility to avoid the electrical shielding.

1.3.5

Textile strain gauges with mobile electrodes

In 2005, Gibbs and Asada have published their results on a new textile sensor able to measure the bending angle of human articulated limbs [24]. The principle that has been used is not related on the physical properties modifications such as electrical conductivity, optical properties, capacitance etc., when deformed, but simply on the principle of the variable resistor or potentiometer. Indeed, a distance between two electrodes deposited on the conductive core is varying. Electrodes are fixed on the substrate. The measured electrical resistance is related to the length of the conductive yarn. This length is then used to compute, using equations, the angle formed by the articulation between limbs. In Fig. 1.11 it is possible to notice the aforementioned textile sensor and to understand its functioning. A and B points are used as electrodes for the measurement of a

12

Smart Textiles for In Situ Monitoring of Composites

Conductive fiber A

W

B C Elastic cord

Wire contact point D

Figure 1.11 Scheme of the Gibbs and Asada strain gauge [24].

conductive thread resistance. They are fixed on the elastic textile fabric (blue) that follows the movements of the articulation. The conductive thread is stitched in the manner to be fixed on A electrode and to be mobile on B electrode. B electrode is fixed at the level of point C to an elastic ribbon fixed to the point D. When the angle of the articulation changes the fabric elongates and the distance AB increases because the B electrode slides down on the conductive thread. At the same time, the conductive thread does not elongate because the elastic ribbon does. Conductive threads used to make this sensor are silver-plated nylon yarns. Their linear resistance is approximately 3.6 U/cm. There is no further information on the nature of A and B electrodes. It is possible to notice that the sensor, in spite of its functioning that does not imply any deformation of conductive thread, shows an important hysteresis. The maximal frequency of detection for this sensor is about 2 Hz. The sensor lifetime has not been studied. It is stated in the study that its recalibration has to be done frequently to guarantee an optimal precision of measurements. This sensor is interesting for the measurement of large amplitude deformations and for low electrical resistance output. Nevertheless, it is not adapted to small deformation measurements often met with composite structures. The bandwidth is also too narrow for the monitoring of composites, especially in case of high-frequency vibrations. Finally, it is not possible to realize a mobile electrode inside composite structures without their severe weakening (Fig. 1.12).

1.3.6

Piezoresistive textile sensorsdconductive polymer composites based

Conductive polymer composites (CPCs) (as composite materials) are by definition made of two components: an insulating matrix and conductive charges. The matrix

Smart textiles for monitoring and measurement applications

13

30

Sensor output (Ohms)

25 20 15 10 5 0

0

20

40 60 80 Joint angle (deg)

100

120

Figure 1.12 Electrical output (in U) of Gibbs and Asad sensor for angular movements measurement of articulation between two limbs.

is often a polymer with electrical insulating properties and the “reinforcement” is composed of some electrically conductive charges under various forms. There are three main classes of conductive charges used in CPCs described below. Two of them are well known and characterized while the third appeared in last decades and is less known and used. • • •

Metallic (copper, silver, aluminum, nickel, etc.). They are excellent electrical and heat conductors. On the other side their surface energy is very high that leads to fast oxidation. Carbon (graphite, carbon black [CB], carbon nanotubesdCNT, graphene, etc.). They exhibit excellent stability and relatively low cost expecting CNT and graphene. Conductive polymers (polypyrrole [PPy], polyaniline, PEDOT, etc.). They appeared recently and will be presented in detail in following sections.

Concerning polymer matrixes, there are also two possible approaches of manufacturing: thermoset and thermoplastic. Finally, there are three different methods of CPC processing to realize conductive fibers, which may be selected in function of the target application and materials that are used: (1) in situ polymerization followed by the vulcanization, (2) melt spinning, and (3) solvent spinning.

1.3.6.1

Electrical properties and percolation phenomenon

In CPCs, when the concentration of conductive charges increases relatively to the polymer matrix content, the electrical conductivity also increases. This statement is true only if the conductive charges are well dispersed inside the polymer matrix, implying their homogeneous space concentration. The increase is not linear as shown in Fig. 1.13. It is possible to identify three main zones of conductivity characteristic in function of conductive charges concentration. Those three zones depend on the three different levels of particules arrangements serving as paths for conductive particules inside polymer matrix (see Fig. 1.13, graphs I, II, III).

14

Smart Textiles for In Situ Monitoring of Composites

Log ρ I

ρm

II

III

φ 1 φc

φ2

φ

Figure 1.13 Log of the resistivity of a conductive polymer composite versus conductive particles concentration (Graphs I, II and III). Schematic representation of the conductive particles arrangements inside polymer matrix [25].

First, a slow decrease of the resistivity is observed forming a superior asymptote. This zone correspond to the state when the particles are rather rare inside polymer matrix and unable to form electrical conductive paths. The CPC is somewhat insulator, and its resistivity is close to those of the polymer matrix rm. The following zone is characterized by fast decrease of electrical resistivity matching to relatively small variation of conductive particles concentration. The conductive particles are sufficiently numerous to form conductive paths by connecting among them. The conductivity is realized either by direct contacts or by tunneling effect or by combination of two phenomena. This zone representing a discontinuity in a conductivity (or resistivity) is also called the percolation threshold. Its exact value in terms of concentration noted 4c is established on the inflexion point of the curve. Finally, the last zone of the curve forms almost horizontal asymptote that approaches to the conductivity (resistivity) value of conductive particles (charges) only. Inside the CPC the particles are present in a substantial number. In reality, there is a maximal possible concentration of charges inside polymer matrix depending on the nature and shape of particles as well as on the processing method. Conductive paths scheme among particles inside the CPC are given below: Several different conductive phenomena may be met in CPCs, they are designated below: • •

Metallic conductivity (direct contacts among particles Fig. 1.14). This is possible only when particles are in direct contact. Electrons move freely from one particle to another similarly as inside metal conductors. Hopping (Bdwithout direct contact among particles Fig. 1.14). In this case an electron coming from one particle has to “jump” to another particle and cross the gap inside polymer matrix using its kinetic energy that has to be higher than the potential energy of the barrier

Smart textiles for monitoring and measurement applications

15

Coating

Direct contact

A

No direct contact

B

Figure 1.14 Graphic representation of the conductive particles location inside polymer matrix with and without contacts among them.

•

(gap) to be crossed. The conductivity is established between particles that have sufficient energy levels not only between the closest ones. Tunneling effect (Adwithout direct contacts among particles Fig. 1.14). This conductivity principle could be explained by quantum mechanics. There is also a “jump” of an electron from one particle to another by crossing the matrix barrier. The difference between tunneling and hopping effect is that even if the electron does not have a sufficient kinetic energy to cross the gap, the probability that this phenomenon occurs is not equal to zero. The electron is assimilated to a wave (see Fig. 1.15) moving inside the particle. When the wave hits a barrier, it does not disappear immediately but decreases exponentially during this crossing of the barrier. If the barrier is not too “high,” the wave magnitude is not null when it reaches the other side of the Electric field

v+

Incoming ψ (particle wavefunction)

Metal particle A

0v

Narrow barrier width

Polymer

ψ exits with lower reduction in energy

Metal particle B

Tuneling barrier > low ψ decay within barrier.

Figure 1.15 Schematic representation of the tunneling effect, between two particles, inside polymer matrix [26].

16

Smart Textiles for In Situ Monitoring of Composites

barrier, consecutively there is a probability that is not equal to zero that the electron will appear in another particle. The barrier “thickness” plays an important role in the tunneling effect.

1.3.6.2

Conductive polymer composite behavior in the presence of deformations (elongation and pressure)

When a CPC undergoes deformation, its electrical resistance changes similarly, as for other conductive materials. This resistance variation is related to two phenomena explained below: •

•

The geometrical properties modifications The electrical resistance of a conductor is defined in function of its resistivity times its length on its cross section (see the Eq. 1.4). As for a traditional strain gauge, when the conductor’s length increases, its cross section decreases (depending on material’s Poisson coefficient). Inversely, in case of pressure on the conductor, its length decreases and its section increases. Following the Eq. 1.4, if the material undergoes elongation (length increase and section decrease), its electrical resistance increases. This phenomenon is true for any possible concentration of conductive particles inside the polymer matrix, but it is predominant only if the concentration is significantly higher than the percolation threshold. The material resistivity changes As shown in the Eq. 1.4, the electrical resistance of the conductor is directly depending on the material resistivity. When the concentration of conductive particles is high in the CPC, many conductive paths are formed and the resistivity does not change when the conductor undergoes geometrical deformations. On the other side, when the conductive particles concentration is close to the percolation threshold, the elongation or the compression of the conductor makes particles to be closer to each other or further (see II, Figs. 1.14 and 1.15). In that case the number of conductive paths changes strongly implying important changes of material electrical resistivity. Simple 3D geometrical analysis shows that when the conductor is elongated, numerous conductive paths break implying an important increase of the electrical resistivity of the CPC material. R¼ r 

l s

Equation 1.4: Conductor’s electrical resistance with r electrical resistivity (Um), l length (m), and s cross section (m2).

Flandin et al. published in 2000 [27,28] research results stating that the modification of a CPC resistance provoked by its geometrical deformation could be more complex than a simple linear increase. They have studied several CPC composed of polymer matrixes such as elastomers, ethyleneeoctene (EO) with volume mass 0.863 g/cm3, containing conductive particles given below and defined in Table 1.1: • • •

Carbone fibers with the diameter of 10 mm and average length 200 mm; Carbone black weakly structured (NCFS), MS-ST from Columbian Chemical co.; Two highly structured CBs (NCHS), Printex 30 from Degussa, and Conductex 975 Ultra from Columbian Chemical co.

Measures of resistances with different concentrations of conductive particles demonstrate that the percolation threshold depends on the type of particles.

Smart textiles for monitoring and measurement applications

17

Table 1.1 Properties of conductive particles used in the study realized by Flandin et al. Particle diameter

Structure (DBPA number)

Surface area (m2/g)

Volatiles (%)

Conductex 975U

21

170

242

1

Printex 30

27

102

80

1

MS-TS

300

33

8

1

Material

1014 LSCB HSCB printex HSCB 975 U CF

Resistivity (W cm)

1012 1010 108 106 104 102 100 10–2 0

10

20

30

40

50

60

Filler content (% v/v)

Figure 1.16 Conductive polymer composite resistivity in function of conductive particles concentrations Flandin et al. [27].

In Fig. 1.16, significant differences among percolation thresholds and slopes of curves at percolation threshold concentrations may be perceived. Mathematical model based on the power law (Eq. 1.5) applied to the measured results gives the approximate values of different percolation thresholds and critical power values t that are summarized in Table 1.2. The coefficient t value designates the slope of the percolation curve characteristics. It may also be described as an intensity of the conductivity (resistivity) changes at percolation threshold. Usually t is equal to 2 for this kind of CPC systems. 1 ¼ r


 V  Vc t 1  Vc

Equation 1.5: Modeling of a CPC resistivity in function of the conductive particles concentration in function of the nature of those particlesdV is the conductive particles concentration, Vc is the concentration corresponding to the percolation threshold, t is the critical power parameter.

18

Smart Textiles for In Situ Monitoring of Composites

Table 1.2 Computed values of the coefficient and percolation thresholds for different conductive fillers, Flandin et al. Filler

t

Vc (vol.%)

HSCB (975U)

2.05

12

HSCB (Printex 30)

2.55

14.5

LSCB (MT-ST)

3.8

39.5

CF (DKDX)

2.04

9.5

The obtained results prove that the spherical structure and the reduced specific surface of LSCB (Low Surface Carbon Black) compared to HSCB (High Surface Carbon Black) increase the value of the percolation threshold more than two times. Even better results are obtained using Carbon Fibers (CFs) as nanofillers. Moreover, many studies on the sensors development based on CPCs with carbon nanotubes (CNT) as conductive filler [29] have shown that the percolation threshold can be as low as 1% of mass concentration. This may be explained by CNTs extremely high specific surface. The study of Flandin et al. continues by uniaxial traction tests of different CPCs. The electrical resistance is computed for all the samples and the normalized resistivity is then calculated. For CPCs with CFs filler with 20% concentration (volume), the resistance increases linearly with the elongation until approximately 30%. After that elongation, CPC may be considered as an insulator. When the cycles of deformations are repeated, it is possible to observe a decrease of the resistivity. Similar behavior is witnessed for CPCs with other High Structure Nanocarbon (HSNC) fillers. The authors explain this decrease of the resistivity by the alignment of conductive particles during the lengthening of sensor and accordingly forming and reinforcing conductive paths that would then remain, even after the releases. On the other side, the elongation tests on the CPCs with HSNC fillers show a significantly different behavior. In Fig. 1.17, it is shown that the resistance increases

Figure 1.17 Parallel and perpendicular resistance and resistivity of a conductive polymer composite with 20% (volume) of high structure nanocarbon filler [28].

Smart textiles for monitoring and measurement applications

19

slightly (about 50%) until 2% of elongation, and then it decreases below its initial value until approximately 20% of lengthening. After that, the resistance increases again very slowly. The resistivity (parallel and perpendicular) exhibits the same behavior. In the case of repeated cycles, the resistivity regains its initial values after cycles, showing a decrease in elongation. With other concentrations of conductive particles inside CPCs, the same behavior may be observed. This concentration influences the initial resistivity of the CPC with HSNC fillers. More the concentration is important, more the initial resistance is low. The concentration of filler also influences the slope of the resistance increase toward the final resistance value. Also, if the filler concentration is close to the percolation threshold, the increase speed of the resistivity is more important. The resistivity evolution versus elongation may be decomposed in four zones (Figs. 1.18 and 1.19). In the first zone, the fast increase of the resistivity may be the steady behavior of CPC with microcracks apparition inside the network structure of CB conductive paths. The second zone is characterized by the decrease of the resistivity between 3% and 20% of elongation. Finally, the last increase of the resistivity between 20% and 1000% of elongation is divided into two zones: the zone III where the resistivity increase is moderate until 600% of elongation and the zone IV where the increase is fast until 1000% of elongation. The zone III is related to ruptures inside CPC, but without irreversible alteration of mechanical and electrical properties of the CPC. The zone IV is characterized by the depercolation phenomenon, meaning that conductive paths formed from conductive particles fillers are irreversibly broken, provoking a fast increase of the resistivity. Even if it is not a standard, the behavior presented in Fig. 1.19 of a CPC-containing EO and HSNC-conductive particles has also been observed for several other CPCs

Reversible I

Recoverable damage

II

III

Resistivity (W cm)

20

'Depercolation' 108 IV 107 r μ exp(be) 106 105

15

b ~ 0.043 10

104 103 102 101

b ~ 0.0075

5 0

5

10 15 20

200

400

600

Resistivity (W cm)

Initiation

800

100 1000

Strain (%)

Figure 1.18 Four zones of the resistivity evolution in function of the Flandin et al. [28], for a conductive polymer composite of EthyleneeOctene and high structure nanocarbon with the concentration 20 (v/v)%.

20

Smart Textiles for In Situ Monitoring of Composites

(c) (b) (a)

Initial structure

Regime II

Regime III

Figure 1.19 Rearrangement of conductive paths during elongationdFlandin et al. [28].

containing highly structured conductive fillers, Aneli et al. [30]. In that study, the polymer matrix used was Polysiloxane vulcanized. The phenomenon of the positive and negative slopes for the CPC resistivity has been noticed for two different types of NCHS (P357E et P803) used as fillers in matrixes of Polysiloxane, polymerized with Toluen, but only for low fillers concentrations. The theoretical explanation is similar to those proposed by Flandin (Fig. 1.19). This study also shows the importance of the CPC manufacturing process, particularly on the good dispersion of conductive fillers, to obtain the optimal final properties. Therefore, a CPC with the same components (polymer matrix and filler) and identical concentration of conductive particles, but vulcanized with peroxide, exhibits quite different properties. Particularly, for low filler’s concentrations, CPC resistivity increases linearly with the elongation. In case of uniaxial CPC compression, Aneli et al. have noticed a decrease of the resistivity with the pressure increase for different CPCs with different fillers’ concentrations. Moreover, the filler’s concentration influences the magnitude of the resistivity increase. More the concentration is close to the percolation threshold (but remaining superior), more the resistivity will be high at maximal pressure. This magnitude decreases strongly with the concentration increase above the percolation threshold. For very high filler’s concentration, this increase is almost invisible (Table 1.3). This behavior of a CPC submitted to a pressure may seem surprising because in CPCs containing metallic fillers it is opposite. This may be justified, by the limited number of conductive paths that are also fragile. The conductive paths are destroyed by disalignment, when a CPC is compressed uniaxially. When the fillers’ concentration is higher, a CPC is saturated in conductive particles, which have subsequently very limited possibilities of displacements that may lead to breakage of contacts among them.

Smart textiles for monitoring and measurement applications

21

Table 1.3 Resistivity measurements for different conductive polymer composites with different filler’s concentrations and different pressures applied within uniaxial compressiondAneli et al. [30] r (U$m) at compression pressure (MPa) Elastomer

Filler content (%)

Atmosphere pressure

0.1

0.5

1.5

SCI-3

30

2.1

47.8

512

1250

50

0.18

0.6

2.2

3.8

70

0.06

0.12

0.31

0.32

100

0.04

0.05

0.08

0.07

30

3.2

26.3

360

423

50

0.2

0.5

0.8

1.2

70

0.07

0.09

0.13

0.13

100

0.04

0.04

0.05

0.05

30

9.7

152

1210

5200

50

0.4

1

6.6

16.6

70

0.09

0.16

0.56

0.67

100

0.04

0.06

0.07

0.06

30

10.8

203.8

2010

6250

50

0.5

2.8

7.2

12.3

70

0.1

0.2

0.6

0.6

100

0.03

0.05

0.07

0.06

SCD

SCDd26M

Nairit A

The coating with CPCs, developed by Bilotti et al. in 2012, using a blend of thermoplastic polyurethane (TPU) and carbon nanotubes (NTC) exhibits similar behavior, when deposited on a Spandex yarn to realize a fibrous strain gauge. Fig. 1.20 shows the behavior of the sensor submitted to cycles of elongations and relaxations, where its resistance increases with the first elongation but decreases only partially (graph a) and even increases (graph b), corresponding to more important elongation. After several cycles, the resistance decreases with the elongation of the sensor and increases when the yarn returns to its initial length. This behavior is explained by two phenomena that are in competition inside the coating layer when the yarn is elongated. • •

Destruction of conductive paths implying the resistance increase; Realignment of conductive particles implying reforming of conductive paths and decreases the resistance.

These observations confirm the conclusions proposed by Flandin and Aneli in their researches on CPCs made of amorphous polymers and highly structures conductive particles.

22

Smart Textiles for In Situ Monitoring of Composites

(a) R/R0

1.5

Strain (%)

1.0

10

0 100

0

200 t (s)

300

(b) 5 R/R0

4 3 2 1

Strain (%)

30 20 10 0 0

100

200

300

400 t (s)

500

600

700

Figure 1.20 Elongation imposed and relative resistance variations in function of time for sensors realized by Bilotti [29]. Coating with the concentration of 2% (mass) of CNT fillers inside the conductive polymer composite. (a) 15% maximal elongation; (b) 30% maximal elongation.

1.3.7

Mechanical properties of conductive polymer composites

The addition of conductive particles to a polymer matrix, to form a CPC, modifies its electrical and also mechanical properties. In general, the addition of particles increases the Young modulus of the material (comparing to the Young modulus of polymer matrix alone). In case of vulcanized elastomers, it is possible to notice an increase of the elastic limit. These modifications of mechanical properties depend mainly on the quality of interactions between the matrix and conductive particles. The efforts applied to the CPC are focused on particles that are more resistant, but the ruptures appear at the interactions between particles and matrixes. Therefore,

Smart textiles for monitoring and measurement applications

23

1200 CF HSCB 975U HSCB printex LSCB

Strain at break (%)

1000 800 600 400 200 0 0

10

20

30

40

50

Filler content (% v/v)

Figure 1.21 Fracture stresses for conductive polymer composites composed of different fillers at different concentrations.

the parameters that influence mainly mechanical properties of CPCs are (also see Fig. 1.21): • • •

The filler’s (conductive particles) concentration; Specific surface of conductive particles; Conductive particles structure.

1.3.7.1

Microruptures phenomenon of piezoresistive coatings

Xue et al. published in their study [31] an analysis of the structure of CPC-coated yarns during elongation cycles. The main objective is focused on phenomena contributing to the changes of resistivity when sensor is deformed geometrically. To better understand those phenomena, a Scanning Electron Microscopy (SEM) has been used. The sensor that has been studied was formed of knitted structure made with Nylon (Polyamide 6, 6; Pa66) threads and Polyuretan (Pu) coated by PPy. Identical yarns as those used in knitted structure have been coated with the same material. The electrical resistance has been measured using four-point methods with Keithley 2010. The samples have been tested during 10 cycles of elongationerelaxation, with the maximal elongation of 50%. The speed of deformations has been set up low (0.005 mm/min) to be able to follow the morphology of the coated structure in situ, with an MEB and to stop the movement to realize micrographs. The gauge factor observed for this sensor is very high going up to 400, for 50% of elongation. In the next paragraphs, an analysis of the reasons of this high sensitivity has been carried out, and the role of each component of the sensor has been investigated. The initial resistances of sample sensory threads Pa66 coated by PPy and Pu coated by PPy ranged from 100 to 200 kU for the first one and 200 to 300 kU for the second. Both sensory threads had similar behavior when exposed to elongation

24

Smart Textiles for In Situ Monitoring of Composites

2.5 2.0

(R-R0)/R0

1.5 1.0 0.5 0.0

PA6 yarn PU yarn

–0.5 0.0

0.2

0.4 0.6 0.8 Relative elongation

1.0

1.2

Figure 1.22 Relative resistance versus relative elongation for Pa6- and Pu-based sensors coated by PPy [31].

with approximately equal gauge factors until 60% of elongation (Fig. 1.22). Above this limit elongation, Pa66-based sensor started to break gradually and Pu-based sensor resistivity increased quickly with the elongation. Knitted structure containing two distinct types of sensors (Pu and Pa6 base) has been realized on a knitted machine with two different feeding devices having different tensions. The tension imposed to Pu thread was higher, implying a preelongation relatively important. This may explain the important gauge factor of the knitted structure sensor because it is directly related to the gauge factor of Pu-based sensor for elongations above 60%. On the images obtained from MEB observations, it is possible to notice microruptures on the surface of Pu-based sensors starting at 9.4%. Above this value of elongation, the number and the size of microruptures increase. Until 110% of elongation, conductive paths are destroyed progressively and the resistivity increases. Above 110%, the number and the size of microruptures become too important and the sensor does not conduct the electricity any more. After relaxation, the microruptures disappear (they close) and the sensor becomes again conductive. After several cycles, similar phenomena appear at different levels of elongation. On the other hand, microscopic observation reveals that behavior of Pa66-based sensors is completely different. The coated structure remains homogeneous until the maximal accepted elongation of 60%, the geometrical variations of length versus cross section contributes to the variations of electrical resistance. After this preliminary study and based on conclusions on the behavior during the elongation and microruptures, particularly on Pu-based sensory yarns, a new study has been published by Xue in 2011 [32]. This study focused on electromechanical characterization of Lycra (Pu) coated by PPy for important elongations going up to 110%. The same technique has been used, sensory yarns have been fixed in a tensile testing machine (low speed), and a MEB has been coupled to the experience to obtain

Smart textiles for monitoring and measurement applications

25

Image treatment

Elementary cell

Lb Ra

Rb

Equivalent circuit

La

Figure 1.23 Microruptures analysis using automated procedure for image treatment [32].

images of the coated yarns in real time during the elongation and to analyze microruptures. The 4-points resistance measuring device has been coupled also to the experiment to get the resistance in real time. Numerical treatment has been applied using Matlab software and results are given in Fig. 1.23. It is possible to notice an evolution of the electrical resistance of yarns versus elongation. From 50% deformation, the slope of relative resistance variation becames linear and steep. This confirms that the apparition of microruptures that are perpendicular to the elongation is due to the significant difference of Young modulus between two materials (Pu and PPy). In fact, the PPy is considerably less elastic than the Pu. Therefore, the microruptures characteristics may be defined by three parameters that are: their number, their lengths, and widths. These parameters are obtained from the numerical analysis of the binarized image of the area with microruptures. R¼ r 

L A

The number and the width of ruptures increase with elongation of the sensory yarn; their length seems to remain constant. Two polynomial models have been set up to follow the evolution of a number and width of ruptures in function of deformation. The electrical resistance is then computed: from the Eq. 1.6 and from considerations on the geometry of the system composed of the yarn with the coating and its mechanical properties based on the elementary cell of the system. The polynomial models are then applied to the elementary cell (with and without microruptures), the equations are

26

Smart Textiles for In Situ Monitoring of Composites

obtained from each one of elementary cells. When those equations are combined with the number of ruptures existing on the yarn, the expression of the coated yarn resistance for a given deformation is obtained. Finally, taking into account the initial resistance of the coated yarn based on first geometrical expression and the resistance law, the model is defined by the Equation 1.6. Describing the evolution R/R0 in function of the deformation ε, of the Poisson coefficient n of the system yarn/coated layer (approximately 0,4), of the initial dimensions of the yarn, L0 its length, d0 its diameter, and of the previously defined parameters which are length L0 of microruptures, their width W(ε), and their number N(ε). R ¼ R0



L0 ð1 þ εÞ  NðεÞ$WðεÞ NðεÞ$WðεÞ þ pd0 ð1  nεÞ pd0 ð1  nεÞ  2L0

 

pd0 L0 ð1  nεÞ

Equation 1.6: Numerical model defined by Wang et al., of the relative electrical resistance evolution of the Pu yarn coated by PPy in function of its relative elongation [33]. The simulation of the relative electrical resistance evolution (Fig. 1.24) shows small difference compared to measured values, but the general behavior remains correct. The author proposed two explanations: the samples preparation that is difficult to be sure that the yarn has not been already preelongated at the beginning of measuring process and the method of measuring that is static, meaning that before each measuring the elongation has to be stopped implying some errors. The most influent parameters acting on the electromechanical properties of sensory yarns have been identified. Three parameters related to microruptures that are their length, width, and number have been tested independently. The most important are the length and the width. If the length is equal to the circumference of the yarn, or if the width is more important that the threshold value, the conductive paths are definitely broken and the electrical resistance becomes infinite.

(R-R0)/R0 (Ω/Ω)

30

20

Experimental results

10

0

Modeling results 0

0.2

0.4 0.6 Deformation (mm/mm)

0.8

1

Figure 1.24 Modeling and experimental results in function of the sensory yarn relative electrical resistance evolution in function of the Pu yarn coated by PPy deformation [32].

Smart textiles for monitoring and measurement applications

1.4 1.4.1

27

Connectors Basic definitions

Conductive wires, cables, and connectors are common physical materials used in the electronics to connect electronic components such as resistor, capacitors, inductors, transformers, chips, sensors, actuators, etc., among them on single or multilayer rigid printed circuit boards (PCBs). The common property of aforementioned electronic circuits is that they are rigid. In the field of textiles and flexible materials, the interconnection, power supply, or communication among different components can be realized by electroconductive yarns integrated into fabrics to form a bus structure. Several types of electroconductive yarns are already on the market. As new electroconductive flexible materials are available, fiber and yarn manufacturers find ways of converting them into fibers and yarns compatible with textile structures. There are already many competitors on this market and some examples are given in Fig. 1.25.

1.4.2

Washability and reliability of connecting devices

The washability and reliability of smart textile devices using flexible connectors such as conductive threads etc. has been and still is extensively studied [4]. Two different approaches aiming at designing, producing, and testing robust washable and reliable smart textile systems are presented in this section. The common point of two approaches is the use of flexible conductive PCB to interface the miniaturized rigid (traditional) electronic devices to conductive threads and tracks within the textile flexible fabric and to connect them to antenna, textile electrodes, sensors, etc. The first approach is focused on the protection of the whole system composed of rigid electronic

Figure 1.25 Electroconductive yarn: (a) stainless steel yarn (Bekaert), (b) silver-coated nylon yarn (Statex).

28

Smart Textiles for In Situ Monitoring of Composites

device, flexible PCB, and textile substrate by the barrier made of latex. The second approach consists in the use of TPU films that are deposited by the press under controlled temperature and pressure parameters to protect the electrical contacts. Several prototypes were realized and washed. Their reliabilities are studied. The washability and more generally the resistance to the humidity and water issues are always an obstacle in terms of application reducing the reliability of smart textile devices and making them not robust enough and therefore not ready for the market. Many of experimental wearable textile devices cannot be used in the real life because of the washability problem. Because of the capillary effect, even the hydrophobic textile substrate can still absorb the water in the textile bulk and make electronic devices fail. Besides, the mechanical stresses provoked by the washing process may destroy the electrical contacts between the conductive thread and the electronic wearable device. As a result, the electric impedance becomes uncontrollable after several cycles of washing process and the wearable device becomes unstable and in some cases, stops to function. In this section two different approaches aiming at designing, producing, and testing robust washable and reliable smart textile systems are presented. The common point of both approaches is the novel method developed in our laboratory consisting in the use of flexible conductive PCB to interface the miniaturized rigid (traditional) electronic devices to conductive threads and tracks within the textile flexible fabric and to connect them to antenna, textile electrodes, sensors, etc. The first approach is focused on the protection of the whole system composed of rigid electronic device, flexible PCB, and textile substrate by the barrier made of latex. Several prototypes have been designed, realized, and tested to verify the washability and reliability of the protected systems. The second approach consists of the use of TPU film that is deposited by the press under controlled temperature and pressure parameters to protect the electrical contacts against water and mechanical stresses. Finally, the third hybrid approach using latex and TPU film protection may also be foreseen to realize the optimal protective barrier. For example, the conductive thread (Shieldtex 234/34-2 ply HC) is silver-coated polyamide, which is used as transmission wire. The first number expresses the fineness of the thread or more professionally expressed the yarn count and is measured in dtex, i.e., 10,000 m of a thread with 234 dtex weighs 234 g. 34 represents the number of fibers in a threadddealing with a twine means there are 68 fibers in total. The linear resistance is less than 100 U/m, which is acceptable for sensing the ECG signal in lowenergy consumption system. This 2-ply structure allows the thread to be adaptable in the sewing machine or embroidery machine. In Fig. 1.26, first prototypes using flexible PCB fixed by stitching to knitted fabric with the LED welded and glued to perform washing tests are shown. To make it compatible with the textile structure, the flexible PCB, Pyralux LF9120R, is employed for electrode parts to make the interconnection with the conducting yarn. In our example, the woven structure with embedded conductive tracks for LEDs application has also been developed and analyzed to verify the reliability and washability of connecting conductive threads. The nickel-plated copper wire (12 nickel

Smart textiles for monitoring and measurement applications

29

Figure 1.26 First prototype using flexible PCB fixed by stitching to knitted fabric with the LED welded and glued to perform washing tests [4].

plated linesþ100D inelastic yarn, Maeden Innovation Ltd) and the silver-plated silver copper tinsel (Tinsel AS32JTE20SZ, Maeden Innovation Ltd) are chosen as conductive threads. Their resistances are less than 5 U/m. The nickel-plated copper wire is used in the small LED sample and the silver-plated silver copper tinsel is used in the LED array samples. The interconnection technology is realized by using professional sewing machine (Mitsubishi LS2-190). The conductive thread is used as needle and as spool thread. The distance between each stitch is 2.5 mm. In this case, the lockstitch type is realized by two parallel conducive thread and results of the low resistance. The electric resistance model of interconnect is shown in Fig. 1.27. The electric impedance between the conductive thread and PCB is composed of three parts: thread resistance Rt, contact resistance Rc, and the PCB resistance RPCB . The thread electrical resistance depends on the material and structure of conductive thread, which is in the range of 10e100 U/m. The PCB electric resistance is as low as negligible compared with the thread electric resistance. According to the contact resistance theory [7], the factors that determine the contact resistance are show in equation: r Rc ¼ 2

rffiffiffiffiffiffiffi pH nP Rt

Rc

RPCB

Needle thread Bobbin thread Flexible PCB Textile structure

Figure 1.27 Illustration of electric resistance model between the conductive thread and PCB on the textile structure [4].

30

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

V

A V2

V1

A

Figure 1.28 Four-point method for electric resistance measurement on TPU-protected samples. (a) conductive thread; (b) sewn PCB by the conductive thread [4].

Equation 1.7: Contact resistance.where r (U.m), H (N/m2), n, and P (N) are electrical resistivity, material hardness, number of contact points, and contact pressure between the metal material and the conductive thread interface, respectively. The contact resistance is therefore inversely proportional to the number of contact points and the contact pressure [34]. The conductive thread and interconnection part are protected with the thin film of TPU by the heat press transfer molding machine ADKINS BETA. The thickness of TPU is 1 mm. The pressure is 3 bars, the temperature is 180 C, and the pressure time is 20 s. Both sides of conductive thread are protected by TPU film. To measure the electric resistance by the four-point method, the conductive thread is not totally covered by the TPU on the topside for the sewing conducting samples (Fig. 1.28(a)). For the sewn flexible PCB samples, only the PCB is covered by TPU (Fig. 1.28(b)). Fig. 1.28 shows the schema of four-point method for the electric resistance measurement. The contact resistance is given by the half value of resistance difference between two PCB electrodes and the uncovered conductive thread. The thread resistance is obtained from the following equation: Rt ¼

V A  number of loop

Equation 1.8: Thread electrical resistance. The contact resistance is obtained from the following equation.
 1 V2 V1  Rc ¼  2 A A Equation 1.9: Measured contact electrical resistance.

Smart textiles for monitoring and measurement applications

1.4.2.1

31

Washing test

The washing test is the one of the widely used as reliability tests for electronics textiles. For this purpose ISO 6330, a standard test for domestic washing and drying procedures for textiles [9], has become a popular test for electronics-in-textiles as well. It was used by many others for all kinds of components of electronics-in-textiles. The standard offers a variety of water temperature levels for different applications from 30 to 92 C. In this study, the temperature of 30 C is used and 50 wash cycles are run. Every wash cycle is programmed for 30 min. The rotation speed is 30 rpm. Between cyclesdbut not necessarily between every washing cycledtest vehicles are drip dried and their functionality is tested. The washing test machine is Datacolor AHIBA IR in this study.

1.4.2.2

Washability of silver conductive thread

Twenty samples are tested after every vehicle of washing tests. Fig. 1.29(a) shows the evolution of resistance during first 10 cycles of washing tests. All samples possess the

(a)

R'/R

7 6

Min Avg

5

Max

Black: thread without TPU Red: thread with TPU

4 3 2 1 0

2

4 6 Wash cycle

8

10

(b) Min Avg

100

Black: thread without TPU Red: thread with TPU

R'/R

Max

10

1 0

10

20 30 Wash cycle

40

50

Figure 1.29 Evolution of electric resistance of sewn conductive threads (a) from 0 to 10 wash cycles and (b) from 0 to 50 wash cycles.

32

Smart Textiles for In Situ Monitoring of Composites

Table 1.4 Resistance per loop during washing tests for the conductive silver thread Resistance (U/loop) Without TPU

With TPU

Wash cycle

Min

Avg

Max

Std dev

Min

Avg

Max

Std dev

0

0.16

0.19

0.27

0.03

0.09

0.13

0.24

0.05

1

0.20

0.28

0.42

0.06

0.12

0.15

0.28

0.05

2

0.28

0.39

0.60

0.08

0.16

0.20

0.35

0.06

3

0.32

0.52

0.80

0.13

0.20

0.25

0.42

0.07

5

0.40

0.67

0.92

0.12

0.23

0.31

0.54

0.10

10

0.57

0.91

1.26

0.23

0.29

0.42

0.73

0.14

20

0.77

1.96

3.15

0.69

0.30

0.45

0.69

0.13

30

1.23

3.22

4.54

1.18

0.42

0.59

1.01

0.20

40

2.72

4.78

6.67

1.21

0.45

0.72

1.31

0.28

50

2.72

13.27

25.30

6.93

0.52

0.77

1.30

0.27

linear increase behavior, even for the samples without TPU protection. After 10 wash cycles, the resistances of conductive threads without protection of TPU film exponentially increase with the cycles of washing tests. Meanwhile, for the samples with TPU protection, their electric resistances linearly increase with the cycles of washing tests. During all washing tests, the resistances of the samples without TPU are always higher than the samples with TPU protection. From Fig. 1.29, we can observe that the resistances for samples with TPU protection arrive at a stable stage and increase lightly; meanwhile the resistances of samples without TPU protection continue to increase. The absolute values of resistance per loop are shown in Table 1.4. After 50 wash cycles, the average resistance increases to more than 70 times to the initial value. The deviation of resistance for samples without TPU is as high as 52% after 50 cycles of wash. As for the samples with TPU protection, the resistance is in the acceptable range from 0.13 to 0.77 U/loop. The deviation is 35% after 50 wash cycles. The lower initial resistance for the sample with TPU can be explained by the increase of contact surface between filaments inside the thread under the pressure of TPU film. The effective numbers of samples are shown in Fig. 1.30. After 3 washing cycles, resistances of some samples without TPU cannot be measured. After 40 cycles, 30% of samples were failed. For the samples with TPU protection, all the samples are measurable till 40 cycles, and this result indicates that the samples are better protected by the TPU film. From the previous studies, it is possible to notice that the conductive thread is not affected if the washing temperature is 30 C. To understand the strong increase of our threads resistance, the scanning electron microscopy (SEM) measurement has been realized. As shown in Fig. 1.31(a), a few peel-off of silver-coating areas can be

Smart textiles for monitoring and measurement applications

33

Effective number of samples (%)

100 90 80 70 With TPU Without TPU

60 50 0

10

20 30 Wash cycle

40

50

Figure 1.30 Evolution of effective number of conductive thread samples during washing tests.

Figure 1.31 SEM images for sewn yarn without TPU protection during wash tests. (a) Before wash test; (b) 5 cycles of wash test; (c) 10 cycles of wash test; (d) 20 cycles of wash test; (e) 40 cycles of wash test; (f) 50 cycles of wash test.

34

Smart Textiles for In Situ Monitoring of Composites

observed even if the thread is not washed. This failure of coating derives from the mechanical movement during the sewing process. With the increase of the number of washing cycles, more and more plated silvers are destroyed by the mechanical movement during the washing (Fig. 1.31(b)e(f). After 50 washing cycles, the silver material can be rarely observed on the outside surface of thread. This may explain the bad conductivity results of washed unprotected threads and the improvements obtained when TPU is used. It should be mentioned that even when the high temperature of 180 C was applied during the heat press transfer process, the plated silver on the filament surface has not been damaged because of the short operation time. The operation temperature and time were adjusted according to the material and thickness of thermoplastic film.

1.4.2.3

Washability of nickel-plated copper wire

Fig. 1.32 shows the evolution of resistance of the sewn nickel-plated copper wire without TPU and with TPU during washing tests. All samples possess the linear increase behavior, even for the samples without TPU protection. After 5 wash cycles, the resistances of nickel-plated copper wire without TPU exponentially increase with the cycles of washing tests and all of them cannot be measured for any resistance after 20 washing cycles. Meanwhile, for the samples with TPU protection, their electric resistances also increase with the cycles of washing tests but the average resistance (3.98 U/loop) is still low. The absolute values of resistance per loop are shown in Table 1.5. After 50 washing cycles, the average resistance increases to more than 50 times to the initial value. The deviation of resistance became 17.36 U/loop for the samples without TPU. As for the samples with TPU protection, the resistance is in the acceptable range from 0.34 to 3.98 U/loop. The deviation became 2.37 U/loop after 50 wash cycles. The effective numbers of samples of the nickel-plated copper wire samples are shown in Fig. 1.33. After 5 washing cycles, resistances of some samples without TPU cannot be measured. After 30 cycles, 100% of samples failed. For the samples with TPU protection, all the samples are measurable after 50 cycles and this result indicates that the samples are well protected by the TPU film. The detailed image for the nickel-plated copper wire samples during wash testing is shown in Fig. 1.34(a). For the samples without TPU, the number of broken nickelplated copper lines increases with the washing cycles (Fig. 1.34(b)e(d)). After 50 washing cycles, the silver material can be rarely observed on the outside surface of thread. The break of copper lines results from the mechanical movement during the test. This may explain the bad conductivity results of washed unprotected threads and the improvements obtained when the TPU is used. Meanwhile, for the samples with TPU protection, the nickel-plated copper lines are intact. However, there was some water inside the TPU showing the connection of TPU and textile becoming loose.

1.4.2.4

Washability of silver-plated silver copper tinsel

Fig. 1.35(a) shows the evolution of resistance of the sewn silver-plated silver copper tinsel without TPU and with TPU during washing tests. The resistances of all samples

Smart textiles for monitoring and measurement applications

(a)

35

1000 900 800

Avg Max

600 R'/R

Black: thread without TPU

Min

700

500 400 300 200 100 0 0

5

10 Wash cycle

15

20

(b) 12 10 Min Avg

8 R'/R

Red: thread with TPU

Max 6 4 2 0

0

10

20 30 Wash cycle

40

50

Figure 1.32 Evolution of electric resistance of sewn nickel-plated copper wire (a) without TPU (b) with TPU.

are almost constant, even for the samples without TPU protection. The effective number without TPU and with TPU is 100% after 50 washing tests, shown as Fig. 1.35(b). The average resistances are from 0.41 to 0.45 U/loop for the samples without TPU and from 0.42 to 0.46 U/loop for the samples with TPU, shown as in Table 1.6. The result shows the structure of silver-plated silver copper tinsel is strong enough to avoid the destruction provoked by the mechanical stresses during the washing cycles inside washing machine. Therefore, the resistances are not significantly different between those of samples without TPU and with TPU. Fig. 1.36 shows detailed images for the washing testing. No matter with or without TPU, the silver-plated silver copper tinsels are all intact in all washing test cycles. However, it is possible to observe the silver-plated silver copper tinsel is rusty for the samples without TPU after 50 washing

36

Smart Textiles for In Situ Monitoring of Composites

Table 1.5 Resistance per loop during washing tests for the nickel-plated copper wire Resistance (U/loop) Without TPU

With TPU

Wash cycle

Min

Avg

Max

Std dev

Min

Avg

Max

0

0.32

0.34

0.36

0.01

0.32

0.34

0.36

0.01

1

0.48

0.73

1.56

0.24

0.36

0.43

0.52

0.05

2

0.55

1.20

4.65

0.98

0.37

0.51

0.94

0.13

3

0.93

2.15

5.80

1.20

0.41

0.54

0.88

0.13

5

1.36

4.78

13.85

4.15

0.39

0.50

0.83

0.10

10

2.8

17.33

64.18

17.36

0.40

0.53

0.87

0.12

20

120.1

352.14

961.1

279.41

0.47

0.67

0.90

0.14

30

0.59

1.15

2.29

0.53

40

1.35

2.58

4.76

0.79

50

1.28

3.98

10.3

Std dev

2.37

Effective number of samples (%)

100 90 80 70 60 50 40 30 With TPU Without TPU

20 10 0

0

10

20 30 Wash cycle

40

50

Figure 1.33 Evolution of effective number of the nickel-plated copper wire samples during washing tests.

cycles, as shown in Fig. 1.36(d). Therefore, the TPU is still useful for prevention of the rusting. Comparing to the silver conductive yarn and nickel-plated copper wire, the silver-plated silver copper tinsel has the best resistance performance. Even when there is not any protection, the resistance is still almost the same as the initial value. But it is also less flexible than the other two, it is not suitable for the knitted structure, but it can be used in the woven structure for high current applications.

Smart textiles for monitoring and measurement applications

Figure 1.34 Images for sewn the nickel-plated copper wire without TPU. (a) Before wash test (b) 5 cycles of wash test; (c) 20 cycles of wash test; (d) 50 cycles of wash test with TPU; (e) before wash test; (f) 5 cycles of wash test; (g) 20 cycles of wash test; (h) 50 cycles.

37

38

Smart Textiles for In Situ Monitoring of Composites

(a) 0.58 0.56

Min

0.54

Avg Max

Black: thread without TPU Red: thread with TPU

R'/R

0.52 0.5 0.48 0.46 0.44 0.42 0.4

0

10

20 30 Wash cycle

40

50

Effective number of samples (%)

(b) 100 90 80 70 60 50 40 30 With TPU Without TPU

20 10 0

0

10

20 30 Wash cycle

40

50

Figure 1.35 (a) Evolution of electric resistance of sewn silver-plated silver copper tinsel; (b) Evolution of effective number of the silver plated silver copper tinsel samples during washing tests.

1.4.2.5

Washability of interconnections

Fig. 1.37 shows the evolution of contact resistance throughout washing tests. The contact points and contact pressure between the conductive thread and PCB are reduced for both kinds of samples because of the mechanical stresses during the washing process. For the first 10 washing cycles, there is slight difference between samples with and without TPU protection. Their contact resistances linearly increase with the number of washing cycles (Fig. 1.37(a)). The increase of contact resistances of sample with TPU is lower than the samples without TPU. However, after 10 washing cycles, the samples without TPU are damaged. After 20 washing cycles, all the samples without TPU are destroyed and the contact resistance cannot be measured (Fig. 1.37(b)).

Smart textiles for monitoring and measurement applications

39

Table 1.6 Resistance per loop during washing tests for the silver-plated silver copper tinsel Resistance (U/loop) Without TPU

With TPU

Wash cycle

Min

Avg

Max

Std dev

Min

Avg

Max

Std dev

0

0.40

0.41

0.42

0.06

0.40

0.42

0.43

0.009

1

0.42

0.44

0.46

0.011

0.40

0.44

0.46

0.013

2

0.41

0.43

0.45

0.012

0.41

0.44

0.45

0.01

3

0.40

0.42

0.44

0.012

0.41

0.43

0.45

0.012

5

0.43

0.44

0.48

0.013

0.43

0.44

0.46

0.008

10

0.41

0.43

0.50

0.025

0.41

0.43

0.46

0.014

20

0.41

0.44

0.49

0.019

0.43

0.47

0.51

0.025

30

0.42

0.45

0.50

0.020

0.41

0.45

0.49

0.021

40

0.42

0.45

0.49

0.019

0.42

0.46

0.49

0.022

50

0.41

0.45

0.50

0.024

0.42

0.46

0.55

0.031

The absolute values of contact resistance are shown in Table 1.7. The average values of contact resistance of samples without TPU are almost three times higher than those of samples with TPU. The average value of contact resistance of samples with TPU after 50 washing cycles is less than that of samples without TPU after 5 washing cycles. This result indicates that the TPU protection improves the contact points and contact pressure between the conductive thread and PCB. As a result, the heat press molding process reduces the contact resistance. The number of measurable contact resistances is shown in Fig. 1.38. The number of measurable contact resistances is tremendously reduced after 2 washing cycles. After 30 washing cycles, all the samples without TPU failed. The failure of electric contact for the samples without TPU comes from the fact that the physical contact between the conductive threads and the PCB is loose. This phenomenon can be easily obtained by visual observation. The thread is suspended over the PCB. Meanwhile, the samples with TPU maintain the physical contact between the thread and the PCB. The failure of samples with TPU after 50 wash cycles may be explained by the shedding of plated silver from the surface of thread because the resistance between two PCBs can still be measured even after 50 wash cycles. Fig. 1.38 shows the evolution of contact resistance during washing tests. The contact points and contact pressure between the conductive thread and PCB are reduced for both kinds of samples because of the mechanical movements during the washing process. For the first 10 washing cycles, there is slight difference between samples with and without TPU protection. Their contact resistances linearly increase with the number of washing cycles. The increase of contact resistances of sample with TPU is lower than the samples without TPU. However, after 10

40 Smart Textiles for In Situ Monitoring of Composites

Figure 1.36 Images for the sewn silver-plated silver copper tinsel without TPU (a) before wash test, (b) 5 cycles of wash test, (c) 20 cycles of wash test, (d) 50 cycles of wash test; with TPU (e) before wash test, (f) 5 cycles of wash test, (g) 20 cycles of wash test, (h) 50 cycles of wash test.

Smart textiles for monitoring and measurement applications

41

Evolution of contact resistance

(a) 12 Min Avg Max

10

Black: sample without TPU Red: sample with TPU

8 6 4 2 0

0

5

10

Wash cycle

Evolution of contact resistance

(b)

25 Min Avg Max

20

Black: sample without TPU Red: sample with TPU

15

10

5

0 0

5

10

15

20

25

30

35

40

45

50

Wash cycle

Figure 1.37 Evolution of electric contact resistance between conductive thread and PCB. (a) First 10 wash cycles; (b) all wash cycles.

washing cycles, the samples without TPU are damaged. After 20 washing cycles all the samples without TPU are destroyed and the contact resistance cannot be measured.

1.4.3

Samples for LATEX-based barrier

To asses in depth the flexible protective barriers, three categories of textile samples have been set up and tested: samples with three LEDs, textiles with LED array, and textiles with ECG electronic devices and battery.

42

Smart Textiles for In Situ Monitoring of Composites

Table 1.7 Resistance per loop during washing tests for the conductive thread Contact resistance (U) Without TPU

With TPU

Wash cycle

Min

Avg

Max

Std dev

Min

Avg

Max

Std dev

0

0.52

0.90

1.37

0.20

0.19

0.30

0.42

0.07

1

0.92

1.57

2.38

0.34

0.25

0.40

0.58

0.10

2

1.21

2.08

3.03

0.47

0.35

0.68

1.15

0.23

3

1.99

3.17

4.93

0.62

0.29

0.86

1.62

0.36

5

1.99

4.37

5.83

0.94

0.70

1.25

2.18

0.38

10

2.79

5.10

7.17

1.23

0.90

1.63

2.46

0.55

20

7.42

8.76

10.05

0.89

1.05

1.67

3.15

0.55

30

1.17

2.52

5.00

1.09

40

1.09

2.64

5.71

1.23

50

1.63

3.44

6.18

1.33

Effective number of samples (%)

100 80 60

Without TPU With TPU

40 20 0 0

10

20 30 Wash cycle

40

50

Figure 1.38 Evolution of effective number of contact resistance samples during washing tests.

All the samples have been protected using latex-based barrier for smart textile systems to make them washable.

1.4.3.1

Textiles with three LEDs

For the washing test concerning textiles with three LEDs, there are totally 12 samples (2 textile structures  2 connections  3 encapsulation conditions). They are made of two different types of textile structures (knitted and woven), two different types of

Smart textiles for monitoring and measurement applications

(a)

Cooper film

43

Zig-zag sewing + Solder or glue LED

VS –

Knit textile

(b) Solder or glue LED

+ VS –

Conductive fiber Woven textile

Figure 1.39 The two kinds of bases for the textiles with three LEDs. (a) The samples of the textiles with three LEDs by PCB board; (b) The samples of the textiles with three LEDs by conductive fiber.

connection between LEDs and textiles (soldering with tin and gluing with silver adhesive), and three types of encapsulation conditions (silicon 1, silicon 2, and nonencapsulated). The conductive thread is sewn in zig-zag to link the flexible PCB with textile structures (Fig. 1.39(a)) or woven inside the textile structure without flexible PCB (Fig. 1.39(b)). The LEDs are connected with the flexible PCB or conductive line by tin soldering gun or silver adhesive gluing (Conductive Silver Adhesive #051908-1R, Polychem UV/EB International Corp.), shown in Fig. 1.40. The process of encapsulation is shown in Fig. 1.41. The painter’s tape was employed to make a mask around the LEDs. Then, the silicon (Super XG-NO.777 or Rubson Silicone) was spread on the LED and wiped to make a smooth surface by a chemistry spatula. The samples are dried in an ambient temperature over one night. The final samples are shown in Fig. 1.42. Eight different types of samples are realized with two different silicones, two different conductive thread integration methods, and two different LED integration methods. In addition, four types of samples without silicon encapsulation are made to make a comparative study for the washing tests.

1.4.3.2

Textiles with LEDs array

A large-scale sample with 16  23 LED array was based on a woven structure with silver-plated silver copper tinsel (Tinsel AS32JTE20SZ, Maeden Innovation Ltd).

44

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

Figure 1.40 Connection methods between LEDs and textiles. (a) Solder the LED on the inside conductive thread; (b) Glue the LED by silver adhesive on the flexible PCB.

Silico

n

+ VS –

Tape

Figure 1.41 Encapsulation for the LEDs on the textiles.

The LEDs were connected with the warp conductive line, and the power lines were connected with the four-weft conductive line. The tin solder paste (LOCTITE GC 10, Henkel Taiwan Ltd) were used to glue the LEDs and warp conductive lines and the textile with LEDs array were put into the 230 C oven about 20 s for soldering (Fig. 1.44(a)). After the soldering, the silicon (Super XG-NO.777) is used to encapsulate the LEDs and the soldered dots (Figs. 1.43 and 1.44(b)).

1.4.4

Washing tests

For the washing tests on the textiles with three LEDs and ECG device, the temperature of 40 C and the duration of 30 min were set up from 1 to 10 wash cycles and the temperature of 60 C and the duration of 45 min were set up from 11 to 18 washing cycles. The rotation speed was 30 rpm. After every vehicle of washing tests, the samples were conducted respectively after 1 h of drying by a dryer. As for the LEDs array, the washing tests were done till 30 cycles.

Smart textiles for monitoring and measurement applications

45

Control group (nonencapsulation)

Experimental group (encapsulation with silicon)

Conductive yarn

Flexible PCB

Silicon1

Silicon2

Silicon1

Silicon2

#1 solder, silicon1

#2 solder, silicon2

#3 solder, silicon1

#4 solder, silicon2

#5 glue, silicon1

#6 glue, silicon2

#7 glue, silicon1

#8 glue, silicon2

#9 solder

#10 glue

#11 solder

#12 glue

Figure 1.42 Twelve samples of the textile structures with three LEDs for the washing test. Silicon 1 (Super XG-NO.777, Cemedine Co., Ltd.) and silicon 2 (Rubson Silicone, Transparent 280 mL, Rubson). Warp conductive line

Weft conductive line

+ VS – LED Solder

Figure 1.43 The sample for the textile with LEDs array.

1.4.4.1

Washability of the textiles with three LEDs

Table 1.8 shows the number of LEDs still functioning after the washing tests and exhibits the result after 18 wash cycles. There are failures for samples with LEDs soldered directly on conductive thread from 1 to 18 wash cycles. Sample #1 shows one broken LED, sample #2 shows two broken LEDs, and sample #3 shows three broken LEDs after 18 washing cycles. It reveals soldering LEDs with the conductive line was easy to get broken, but the encapsulation still had the effect to reduce the quantity of broken LEDs. It can be seen that the best result has been obtained for the sample S1

46

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

Figure 1.44 The connection and encapsulation of the textile with LEDs array. (a) Solder with Tin solder paste; (b) Encapsulation with Silicon (Super XG-NO.777).

Table 1.8 Lighting test during washing tests for the textiles with three LEDs. Silicon 1 (Super XG-NO.777, Cemedine Co., Ltd.) and silicon 2 (Rubson Silicone, Transparent 280 mL, Rubson) Lighting quantity after wash cycles Sample

Structure

Connect

Encapsulation

1

5

10

15

18

S1

Conduct line þ Woven textile

Solder

Silicon 1

3

2

2

2

2

Silcon 2

2

2

1

2

1

None

3

3

1

0

0

S2 S9 S5

Silicon 1

3

3

3

3

3

S6

Silcon 2

3

3

3

3

3

S10

None

3

3

3

3

3

S3

Glue

Flexible PCB þ Knit textile

Silicon 1

3

3

3

3

3

S4

Silcon 2

3

3

3

3

3

S11

None

3

3

3

3

3

Silicon 1

3

3

3

3

3

S8

Silcon 2

3

3

3

3

3

S12

None

3

3

3

3

3

S7

Solder

Glue

Smart textiles for monitoring and measurement applications

47

Experimental group (encapsulation with silicon)

Conductive yarn

Flexible PCB

Silicon1

Silicon2

Silicon1

Silicon2

#1 solder, silicon1

#2 solder, silicon2

#3 solder, silicon1

#4 solder, silicon2

#5 glue, silicon1

#6 glue, silicon2

#7 glue, silicon1

#8 glue, silicon2

#9 solder

#10 glue

#11 solder

#12 glue

Figure 1.45 The washing test results after the 18 wash cycles, failed are in red color.

(solder þ silicon 1), S2, and S9. Compared to the samples S5, S6, and S10, glued LEDs with conductive lines appear to have the better performance than the soldering. It is because the conductive thread is nickel-plated copper tinsel and the solder joint is brittle for nickel. Regardless of the woven textile with conductive line or flexible PCB, all of them had zero broken LED. This suggests that the flexible PCB combined with the knitted textile by zig-zag sewing is more stable and it could be used to embed the flexible circuit into the textile. All the samples after 18 wash cycles are shown in Fig. 1.45.

1.4.4.2

Washability of the textiles with LEDs array

Fig. 1.46 shows the lighting results after washing test and Fig. 1.47 demonstrates the effective percentage and the broken quantity of LEDs. It reveals that there were four broken LEDs being observed after 6 wash cycles and there were only two broken LEDs after 30 wash cycles. It might be because the sample was hand-made and some of the LEDs were not soldered well, leading to an early breakage in the washing test. It was thought that the problem can be improved by using the automatic machine for manufacturing. Although there were still five broken LEDs found after 30 washing cycles, the encapsulation on the LEDs and LED dots were still feasible and the effective percentage of LEDs can be achieved up to 98.6%. In the future, the automatic tin paste extrusion and LEDs placement on the textile should be investigated for the mass manufacturing and better production quality.

48

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

5 washing cycles

(c)

10 washing cycles

(d)

15 washing cycles

(e)

20 washing cycles

(f)

25 washing cycles

30 washing cycles

Figure 1.46 The samples of the textile with LEDs array after 5, 10, 15, 20, 25, and 30 washing cycles. 5

Effective percentage of LEDs Broken quantity of LEDs

99.7

4

99.5

3

99.2

2

98.9

1

98.6

0

5

10

15 Wash cycle

20

25

Broken quantity of LEDs

Effective percentage of LEDs (%)

100

0 30

Figure 1.47 The effective percentage of LEDs from 1 to 30 washing cycles results for the textile LEDs array.

1.4.5

Conclusion

In this section, the main problem and the most important barrier to be ready for the market of smart and connected textiles related to their washability has been investigated. The lack of reliability is often due to bad contacts or to an important increase of contact and thread resistance. For the conductive thread, the shedding of conducting

Smart textiles for monitoring and measurement applications

49

material occurs because of the mechanical movements during the washing process. Similarly, an important increase of the contact resistance between the conductive thread and the flexible PCB is provoked by the shedding of conductive material on conductive threads. With the thermoplastic polymer film, the conductive threads and their contacts with the PCB can be protected against the mechanical stresses during the washing process. After 50 washing cycles, most of samples were still measurable. Their conductivity property was improved compared to the samples without film protection. The flexible PCB with the thermoplastic protection is a promising approach to the electronic-in-textile research, which provides a reliable method to integrate the electronic component into textile structures. As for the latex-based barrier in smart textile application, the encapsulation method for LEDs was found to perform well, which is due to a reliable method to embed the LEDs, flexible PCB, and hardware circuit into woven and knitted structures. For the e-textile applications in the daily life, two approaches (the thermoplastic polymer for the conductive thread and flexible PCB and the latex-based barrier for the rigid PCB or electronic components) can also be combined together to have better washability.

1.5

Conductive polymers, fibers, and structures

The demand for electrically conductive fibers, yarns, and threads and textiles structures used as industrial materials, such as sensors, electrostatic discharge, welding of plastics, EM shielding, etc., is growing up [35,36]. Conductive fibers can be made from conductive polymers by melt spinning, wet spinning, or coating fibers with electrically conductive materials. Modification of fibers based on conductive polymers seems to be an interesting approach enabling new functionalities. Conductive polymers comprise an interesting class of materials combining some of the mechanical features of plastics with the electrical properties typical of metals. Conductive polymers are specified by the presence of alternating single and double bonds between carbon atoms along the polymer main chains: due to such bond conjugation, broad valence, and conduction bands are generated. Their main disadvantages are that they are insoluble in common organic solvents and they have rather weak mechanical properties and poor processability. The conductivity in these polymers is influenced by a variety of factors including polar on length, the conjugation length, and the overall chain length and by the charge transfer to adjacent molecules. Researches in the field of conductive polymers have attracted considerable attention for more than 20 years [37]. Conductive polymers, due to their lightweight, processability, relatively high conductivity, stability, and flexibility, are well suited for the production of conductive fibers and textiles. Conductivity borderlines between electrically isolating, semiconducting, and conductive materials are fluent and not precisely defined. An overview with typical, widely accepted ranges of conductivity for these classes is shown in Fig. 1.48 [38].

50

Smart Textiles for In Situ Monitoring of Composites

Conductivity (S/cm) 106

Metallic conductors

104 102

Indium/antimony

100

Gallium/arsenic Germanium

10–2

Semiconductors

Materials Copper Iron Graphite Bismuth

10–4

Conjugated polymers

Silicon

10–6 10–8 10–10

Glass

10–12

Isolators

10–14 10

Diamond Sulfur Polyethylene Polystyrene Teflon Quartz

–16

10–18 10–20

Figure 1.48 Illustration of the range of electronic conductivities of conductive polymers in comparison with other materials.

The Nobel prize in Chemistry for 2000 obtained by Alan Heeger, Alan Graham MacDiarmid, and Hideki Shirakawa for the discovery and development of conductive polymers has highlighted the significance of these materials. The conductivity experimental data of conjugated polymers can be explained by hopping mechanisms. A large number of hopping mechanisms are proposed such as nearest neighbor hopping, variable range hopping, fluctuation-induced tunneling, etc. Hopping means phonon-assisted quantum mechanical tunneling and it is an anthropomorphic term [39]. Hopping conduction (Fig. 1.49) is defined as electric conduction Conduction band

Ehop (1)

(2)

(3)

Valence band

Figure 1.49 Electron transport between localized states: (1) thermal-assisted tunneling, (2) tunneling, (3) tunneling with the emission of phonon(s).

Smart textiles for monitoring and measurement applications

51

where electron hops from one localized state to another. Thermally assisted tunneling is a process where electron hops from one state to another one that has a higher energy depending on the temperature. As result, the energy hopping difference appears (Ehop). Tunneling processes can explain electrons hopping from a state to another state that has equal energy and does not depend on temperature. When electron hops to a state that has a lower energy, tunneling process with the emission of a phonon(s) occurs [40]. Two subclasses of conductive polymers for textile sensors development and integration into textile structures to measure different physical values are: • •

intrinsically conductive polymers (ICPs), extrinsically conductive polymers (ECPs or CPCs).

ECPs owe their conductivity due to the presence of externally added ingredients in them such a conductive particles based on CB, carbon nanotubes, etc. Textile materials are very flexible and easily deformable in all directions, and the sensors used should be able to support, often all at the same time, tensile, shear, bending, and even compression deformations. The sensors should be integrated in the textile structure to track all these mechanical deformations, without affecting the original textile characteristics such as softness, feel, etc. The sensors must be sensitive enough to measure in situ strains inside the composite parts. Sensitivity is important as the targeted application usually undergoes very low strains (S12).

Optimum adhesion results are related to the formation of homogeneous morphology and improvement in mechanical and thermal properties of the selected system. The interfacial energy between the two phases in contact, g12, can be calculated using the harmonic mean model (Eq. 1.25): g12ðSLÞ ¼ g1ðSÞ þ g2ðLÞ 

4gd1ðSÞ gd2ðLÞ gd1ðSÞ þ gd2ðLÞ



4gp1ðSÞ gp2ðLÞ gp1ðSÞ þ gp2ðLÞ

Equation 1.25. Thermodynamic work of adhesion represents the work (energy), Wa, required to separate the two surfaces in contact. The work of adhesion can be calculated using the equation (Eq. 1.26): W12ðAÞ ¼ g1ðSÞ þ g2ðLÞ  g12ðSLÞ Equation 1.26. Wetting coefficient, S, represents a measure of the degree of wetting one substance through the other one, S12, or the second phase over the first one, S21, using the equation (Eqs. 1.27 and 1.28): S12 ¼ g2ðLÞ  g1ðSÞ  g12ðSLÞ Equation 1.27.

142

Smart Textiles for In Situ Monitoring of Composites

S21 ¼ g1ðSÞ  g2ðLÞ  g12ðSLÞ Equation 1.28. Wetting coefficient can be calculated as well from the difference between the thermodynamic work of adhesion, W12(A), and the thermodynamic work of cohesion, Wc: (Eq. 1.29): Wc ¼ 2gi Equation 1.29. Wetting coefficient has significant importance for the problem of wetting in dispersions, during cleaning with detergents, and adhesion to different substrate.

References [1] V. Koncar (Ed.), Smart Textiles and Their Applicatons, Elsevier Science Ltd., London, 2016. [2] X. Tao, V. Koncar (Eds.), Handbook on Smart Textiles, Springer, Hong Kong, 2015. [3] G. Asch, Les capteurs en instrumentation industrielle, 7eme édition, Dunod, Paris, 2010. [4] X. Tao, V. Koncar, T.H. Huang, C.L. Shen, Y.C. Ko, G.T. Jou, How to make reliable, washable, and wearable textronic devices, Sensors 17 (4) (2017) 673. [5] M. Niggemann, M. Pellmann, P. Bolsmann, J. Reinbold, K. Cammann, Remote sensing of tetrachloroethene with a micro-fibre optical gas sensor based on surface plasmon resonance spectroscopy, Sensors and Actuators B: Chemical 34 (1) (1996) 328e333. [6] D.R. Thevenot, K. Toth, R.A. Durst, G.S. Wilson, Electrochemical biosensors: recommended definitions and classification, Pure and Applied Chemistry 71 (12) (2009) p. on line. [7] L. Guo, L. Berglin, M. Heikki, Textile strain sensors characterization - sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 2 (1) (2010) 51e63. [8] P. Xue, K.H. Park, X.M. Tao, Electrically conductive yarns based on PVA/carbon nanotubes, Composite Structures 78 (2) (2007) 271e277.  [9] A. Lobnik, S.K.U.M. Turel, Optical chemical sensors: design and applications, in: Advances in Chemical Sensors, Intech, Maribor, 2012, pp. 3e28. [10] J. Zhou, G. Tzamalis, N.A. Zaidi, Electrically conductive PANI multifilaments spun by a wet spinning process, Journal of Materials Science 36 (13) (2001) 3089e3095. [11] M. Rothmaier, M.P. Luong, F. Clemens, Textile pressure sensor made of flexible plastic optical fibers, Sensors 8 (7) (2008) 4318e4329. [12] B. Adhikari, S. Majumdar, Polymers in sensor applications, Progress in Polymer Science 29 (7) (2004) 699e766. [13] S.A. Dyer, Wiley Survey of Instrumentation and Measurement, Wiley, Kansas city, 2001, pp. 84e98. [14] R. Zeiser, T. Fellner, J. Wilde, Academia day, Development and Testing of Capacitive Strain Gauges, vol. 1, 2011, pp. 1e6. [15] B. Carter, J. Shannon, J. Forshaw, Measurement of displacement and strain by capacitance methods, Proceeding of the Institution of Mechanical Engineers 5 (1945) 215e221.

Smart textiles for monitoring and measurement applications

143

[16] J. Li, J.P. Longtin, S. Tankiewicz, A. Gouldstone, S. Sampath, Interdigital capacitive strain gauges fabricated by direct-write thermal spray and ultrafast laser micromachining, Sensors and Actuators A: Physical 133 (1) (2007) 1e8. [17] R. Matsuzaki, A. Todoroki, Wireless flexible capacitive sensor based on ultra-flexible epoxy resin for strain measurement of automobile tires, Sensors and Actuators A: Physical 140 (1) (2007) 32e42. [18] A.B. Bradley, P.D. Render, The Development of a Parachute Strain Measurement Technique, vol. 1, American Institute of Aeronautics and Astronautics, 1986, pp. 194e202, 1. [19] X. Lin, K. Zhou, X. Zhang, D. Zhang, Development, modeling and application of piezoelectric fiber composites, Transactions of Nonferrous Metals Society of China 23 (1) (2013) 98e107. [20] H.P. Konka, M.A. Wahab, K. Lian, Piezoelectric fiber composite transducers for health monitoring in composite structures, Sensors and Actuators A: Physical 194 (2013) 84e94. [21] A. Weder, S. Geller, A. Heinig, T. Tyczynski, W. Hufenbach, W. Fischer, A novel technology for the high-volume production of intelligent composite structures with integrated piezoceramic sensors and electronic components, Sensors and Actuators A: Physical 1 (2013) 1e5. [22] D. Yang, X. Tao, A. Zhang, Optical responses of FBG sensors under deformations, in: Smart Fibres, Fabrics & Clothing, Woodhead Publishing Limited, Hong Kong, 2001, p. 150. [23] F.J. Arregui, I.R. Matıas, M. Lopez-amo, Optical fiber strain gauge based on a tapered single-mode fiber, Sensors and Actuators 79 (1) (2000) 90e96. [24] P.T. Asada, H.H. Gibbs, Wearable conductive fiber sensors for multi-Axis human joint angle measurements, Journal of Neuroengineering and Rehabilitation 2 (1) (2005) p. on line. [25] C. Cochrane, Développement d’un systeme de mesure d’allongement pour voilure de parachute, L’Universite des Sciences et Technologies de Lille, Villeneuve d’Ascq, 2007. [26] Peratech, QTCTM Material Science the Physics of QTCTM Material Operation, Peratech, August 27, 2013 [En ligne]. Available: http://www.peratech.com/qtc-science.html. [27] L. Flandin, A. Chang, S. Nazarenko, A. Hiltner, E. Baer, Effect of strain on the properties of an ethylene-octene elastomer with conductive carbon fillers, Journal of Applied Polymer Science 76 (6) (2000) 894e905. [28] L. Flandin, A. Hiltner, E. Baer, Interrelationships between electrical and mechanical properties of a carbon black-filled ethyleneeoctene elastomer, Polymer 42 (2) (2001) 827e838. [29] R. Zhang, H. Deng, R. Valenca, J. Jin, Q. Fu, E. Bilotti, T. Peijs, Carbon nanotube polymer coatings for textile yarns with good strain sensing capability, Sensors and Actuators A: Physical 179 (2012) 83e91. [30] J.N. Aneli, G.E. Zaikov, L.M. Khananashvili, Effects of mechanical deformations on the structurization and electric conductivity of electric conducting polymer, Black-Filled Rubbers Under, Journal of Applied Polymer Science 74 (1) (1999) 601e621. [31] P. Xue, X. Tao, H.Y. Tsang, In situ SEM studies on strain sensing mechanisms of PPy-coated electrically conducting fabrics, Applied Surface Science 253 (7) (2007) 3387e3392. [32] J.P. Wang, P. Xue, X.M. Tao, Strain sensing behavior of electrically conductive fibers under large deformation, Materials Science and Engineering A 528 (6) (2011) 2863e2869. [33] A. Soroudi, M. Skrifvars, Melt blending of carbon nanotubes/polyaniline/polypropylene compounds and their melt spinning to conductive fibres, Synthetic Metals 160 (11) (2010) 1143e1147.

144

Smart Textiles for In Situ Monitoring of Composites

[34] L. Li, S. Liu, F. Ding, T. Hua, W. Man Au, K. Wong, Electromechanical analysis of lengthrelated resistance and contact resistance of conductive knitted fabrics, Textile Research Journal 82 (20) (2012) 2062e2070. [35] B. Kim, V. Koncar, E. Devaux, C. Dufour, P. Viallier, Electrical and morphological properties of PP and PET conductive polymer fibers, Synthetic Metals 146 (2) (2004) 167e174. [36] I. Cristian, S. Nauman, C. Cochrane, V. Koncar, Electro-conductive sensors and heating elements based on conductive polymer composites in woven structures, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabrics Technology, Athens, 2011, pp. 3e22. [37] B. Kim, V. Koncar, E. Devaux, Electrical properties of conductive polymers: PET e nanocomposites’ fibres, AUTEX Research Journal 4 (1) (2004) 1e13. [38] A. Elschner, S. Kirchmeyer, W. L€ovenich, PEDOT: Principles and Applications of an Intrinsically Conductive Polymer, CRC Press Taylor & Francis Group, Boca Raton, 2010. [39] J. Han, M. Shen, W. Cao, Hopping conduction in Mn-doped ZnO, Applied Physics Letters 82 (1) (2003) 67e69. [40] G. Latessa, F. Brunetti, A. Reale, G. Saggio, A. Di Carlo, Piezoresistive behaviour of flexible PEDOT: PSS based sensors, Sensors and Actuators B: Chemical 139 (2) (2009) 304e309. [41] S. Bhadra, D. Khastgir, N.K. Singha, Progress in preparation, processing and applications of polyaniline, Progress in Polymer Science 34 (8) (2009) 783e810. [42] A. Ramanavicius, A. Malinauskas, Electrochemical sensors based on conducting polymer e polypyrrole, Electrochimica Acta 51 (27) (2006) 6025e6037. [43] D. Orata, M. Mukabi, H. Njenga, Electro-characterization of polypyrolle electrosynthesized on a montmorillonite host-matrix, in aqueous media containing sulphuric acid as supporting electrolyte, IOSR Journal of Applied Chemistry 7 (5) (2014) 59e72. [44] J. Hwang, F. Amy, A. Kahn, Spectroscopic study on sputtered PEDOT•PSS: role of surface PSS layer, Organic Electronics 7 (5) (2006) 387e396. [45] B. Friedel, T.J.K. Brenner, C.R. McNeill, et al., Influence of solution heating on the properties of PEDOT: PSS colloidal solutions and impact on the device performance of polymer solar cells, Organic Electronics 12 (10) (2011) 1736e1745. [46] B. Kim, V. Koncar, C. Dufour, Polyaniline (PANI) coated PET conductive yarns: study of electrical, mechanical and electro-mechanical properties, Journal of Applied Polymer Science 101 (3) (2006) 1252e1256. [47] O.P. Dimitriev, D.A. Grinko, Y.V. Noskov, et al., PEDOT: PSS films-effect of organic solvent additives and annealing on the film conductivity, Synthetic Metals, 159 (21) (2009) 2237e2239. [48] S. Xiong, L. Zhang, X. Lu, Conductivities enhancement of poly(3,4 ethylenedioxythiophene)/poly(styrene sulfonate) transparent electrodes with diol additives, Polymer Bulletin 70 (2013) 237e247. [49] J. Nevrela, M. Micjan, M. Novota, et al., Secondary doping in poly(3,4-ethylenedioxythiophene): poly(4-styrenesulfonate) thin films, Journal of Polymer Science Part B: Polymer Physics 53 (16) (2015) 1139e1146. [50] E.R. Arakelova, A.M. Khachatryan, K.E. Avjyan, et al., Deposition of high-ohmic oriented ZnO films on glass, Si, and PEDOT-PSS, PEDOT-PSS(PVA) substrates at low temperatures by DC-magnetron sputtering, Journal of Contemporary Physics (Armenian Academy of Sciences) 46 (6) (2011) 293e299. [51] E.P. Scilingo, F. Lorussi, A. Mazzoldi, D. De Rossi, Strain-sensing fabrics for wearable kinaesthetic-like systems, IEEE Sensors Journal 3 (4) (2003) 460e467.

Smart textiles for monitoring and measurement applications

145

[52] F. Lorussi, W. Rocchia, E.P. Scilingo, A. Tognetti, D. De Rossi, Wearable, redundant fabric-based sensor arrays for reconstruction of body segment posture, IEEE Sensors Journal 4 (6) (2004) 807e818. [53] R. Bartalesi, N. Carbonaro, F. Lorussi, M. Tesconi, A. Tognetti, G. Zupone, D. DeRossi, Smart textiles: toward a wearable motion capture system, in: Ninth International Symposium on the 3D Analysis of Human Movement, Valenciennes, 2006. [54] Z.S. Levin, C. Robert, J.F. Feller, M. Castro, J. Grunlan, Flexible latex d polyaniline segregated network composite coating capable of measuring large strain on epoxy, Smart Materials and Structures 22 (1) (2013) p. on line. [55] U. Lang, N. Naujoks, J. Dual, Mechanical characterization of PEDOT: PSS thin films, Synthetic Metals 159 (5) (2009) 473e479. [56] M. Akerfeldt, M. Straat, P. Walkenstrom, Influence of coating parameters on textile and electrical properties of a poly(3,4-ethylene dioxythiophene):poly(styrene sulfonate)/ polyurethane-coated textile, Textile Research Journal 83 (20) (2013) 2164e2176. [57] C. Chen, J.C. Larue, R.D. Nelson, L. Kulinsky, M.J. Madou, Electrical conductivity of polymer blends of poly (3,4-ethylenedioxythiophene) : poly (styrenesulfonate) : N-methyl2-pyrrolidinone and polyvinyl alcohol, Journal of Applied Polymer Science 1 (2011) 1e8. [58] C. Chen, A. Torrents, L. Kulinsky, R. Nelson, M. Madou, L. Valdevit, J. LaRue, Mechanical characterizations of cast poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate)/ polyvinyl alcohol thin films, Synthetic Metals 161 (2011) 21e22. [59] J. Zieba, M. Frydrysiak, Textronics e Electrical and Electronic Textiles. Sensors for Breathing Frequency Measurement, Fibres and Textiles in Eastern Europe, vol. 14, 2006, pp. 43e48, 5. [60] L. Guo, L. Berglin, M. Heiki, Textile strain sensors characterization e sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 1 (2) (2010) 51e56. [61] A. Chen, C.H. Sawhney, Soft-structured sensors and connectors by inkjet printing, AATCC Review 7 (6) (2007) 1e10. [62] T. Bashir, M. Ali, S. Cho, N. Persson, M. Skrifvars, OCVD polymerization of PEDOT: effect of pre-treatment steps on PEDOT-coated conductive fibers and a morphological study of PEDOT distribution on textile yarns, Polymers for Advanced Technologies 24 (2) (2013) 210e219. [63] C. Cochrane, V. Koncar, M. Lewandowski, C. Dufour, Design and development of a flexible strain sensor for textile structures based on a conductive polymer composite, Sensors 7 (1) (2007) 473e492. [64] S. Nauman, P. Lapeyronnie, I. Cristian, F. Boussu, V. Koncar, On line measurement of structural deformations in composites, IEEE Sensors 11 (6) (2010) 1329e1336. [65] J. Rausch, E. M€ader, Health monitoring in continuous glass fibre reinforced thermoplastics: tailored sensitivity and cyclic loading of CNT-based interphase sensors, Composites Science and Technology 70 (13) (2010) 2023e2030. [66] J. Rausch, E. M€ader, Health monitoring in continuous glass fibre reinforced thermoplastics : manufacturing and application of interphase sensors based on carbon nanotubes, Composites Science and Technology vol. 70 (11) (2010) 1589e1596. [67] C.T. Huang, C.F. Tang, M.C. Lee, S.H. Chang, Parametric design of yarn-based piezoresistive sensors for smart textiles, Sensors and Actuators A: Physical 148 (1) (2008) 10e15. [68] I. Jerkovic, V. Koncar, A.M. Grancaric, New textile sensor for in situ structural health monitoring of textile reinforced thermoplastic composites based on the conductive poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) polymer complex, Sensors 17 (2017) 1e21, pii: E2297.

146

Smart Textiles for In Situ Monitoring of Composites

[69] C.T. Huang, C.L. Shen, C.F. Tang, S.H. Chang, A wearable yarn-based piezo-resistive sensor, Sensors and Actuators A: Physical 141 (2) (2008) 396e403. [70] C. Cochrane, V. Koncar, M. Lewandowski, Development of a flexible strain sensor for textile structures, in: 2nd Ambience 08 International Scientific Conference Proceedings, Smart Textiles e Technology and Design, Borås, 2008. [71] N. Trifigny, F.M. Kelly, C. Cochrane, V. Koncar, et al., In-situ measurements of strain and stress of e-glass warp interlock fabric using innovative sensors, in: CompositeWeek@ Leuven(TexComp-11): Proceedings, Leuven, 2013. [72] S. Sha, X. Tao, F. Boussu, V. Koncar, Mechanical properties of high modulus UHMW yarn measured by intelligent sensors, in: 13th AUTEX World Textile Conference Book of Proceedings, Dresden, 2013. [73] Co Heraeus Precious Metals GmbH &, Coating Guide Clevios™ P Formulations, Heraeus, Hanau, 2012. [74] B. Hafner, Energy Dispersive Spectroscopy on the SEM: A Primer, Characterization Facility, University of Minnesota, Twin Cities, 2007. [75] C.Q. Yang, Q. He, R.E. Lyon, et al., Investigation of the flammability of different textile fabrics using micro-scale combustion calorimetry, Polymer Degradation and Stability 95 (2) (2010) 108e115. [76] R. Gupta, V. Kumar, P.K. Goyal, et al., Thermogravimetric analysis of heat treated CR-39 polymer, Journal of Chemical and Pharmaceutical Research 2 (4) (2010) 629e634. [77] N. Amir, F. Ahmad, P.S.M. Megat-Yusoff, Study on the fibre reinforced epoxy-based intumescent coating formulations and their char characteristics, Journal of Applied Sciences 11 (10) (2011) 1678e1687. [78] R.E. Lyon, R. Walters, Flammability of Automotive Plastics, Federal Aviation Administration William J. Hughes Technical Center, New Jersey, 2005. [79] F. Laoutid, L. Bonnaud, M. Alexandre, New prospects in flame retardant polymer materials: from fundamentals to nanocomposites, Materials Science and Engineering: R: Reports 63 (3) (2009) 100e125. [80] A.M. Grancaric, A. Tarbuk, E. Chibowski, Surface free energy of textiles, Tekstil 57 (1) (2008) 29e39.

Further reading [1] [2] [3] [4] [5] [6] [7] [8] [9]

A. Kaw, Mechanics of Composite Material, Taylor & Francis Group, New York, 2006. S.T. Peters, Handbook of Composites, Chapman & Hall, UK, Mountain View, 1982. D. Gay, Matériaux Composites, Hermes, Paris, 1997. M.J. Robert, Mechanics of Composite Materials, second ed., Taylor & Francis Group, New York, 1999. J.M. Cole, J. Berthelot, Composite Materials: Mechanical Behavior and Structural Analysis, Springer-Verlag, New York, 1999. F.K. Ko, T.W. Chou, Composite Materials Series e Textile Structural Composites, Elsevier Science Ltd., Amsterdam, 1989. G.C. Sih, A.M. Skudra, Failure Mechanics of Composites, Elsevier Science Publishers BV, New York, 1986. L. Tong, A.P. Mouritz, M.K. Bannister, 3D Fibre Reinforced Composites, Elsevier Science Ltd., London, 2002. P. Lucas, G. Zanella, Mise en oeuvre des composites TP, Compounds a fibres courtes et longues, Techniques de l’ingénieur, Paris, 2007.

Smart textiles for monitoring and measurement applications

147

[10] V.V. Vasiliev, E.V. Morozov, Mechanics and Analysis of Composite Materials, Elsevier Science Ltd., Oxford, 2001. [11] D. Coutelier (Ed.), Advanced Material Textiles for Reinforced Structures for Complex Lightweight Applications, ENSIAME, Valenciennes, 2016. [12] G. Du, P. Popper, Analysis of a circular braiding process for complex shapes, Journal of the Textile Institute 85 (3) (1994) 316e337. [13] S. Nauman, I. Cristian, V. Koncar, Simultaneous application of fibrous piezoresistive sensors for compression and traction detection in glass laminate composites, Sensors, 11 (10) (2011) 9478e9498. [14] J.V. Risicato, F. Kelly, D. Soulat, X. Legrand, W. Tr€ umper, C. Cochrane, V. Koncar, A complex shaped reinforced thermoplastic composite part made of commingled yarns with integrated sensor, Applied Composite Materials 22 (1) (2014) 81e98. [15] A. Schwarz, I. Kazani, L. Cuny, Electro-conductive and elastic hybrid yarns e the effects of stretching, cyclic straining and washing on their electro-conductive properties, Materials and Design 32 (1) (2011) 4247e4256. [16] X. Tao, I. Fsaifes, V. Koncar, C. Dufour, C. Lepers, L. Hay, B. Capoen, M. Bouazaoui, CO2 laser-induced crystallization of solegel-derived indium tin oxide films, Applied Physics A 96 (3) (2009) 741e749. [17] X. Tao, V. Koncar, C. Dufour, Novel geometry pattern for the wire organic electrochemical textile transistor, Journal of the Electrochemical Society 158 (5) (2011) 572e577. [18] K. Kim, C.S. Park, D.W. Lee, S.M. Cho, C.R. Han, Measurement of the gauge factor of carbon fiber and its application to sensors, Microelectronic Engineering 85 (5) (2008) 787e791. [19] I. Cristian, S. Nauman, F. Boussu, V. Koncar, A study of strength transfer from tow to textile composite using different reinforcement architectures, Applied Composite Materials 19 (3) (2012) 427e442. [20] S.L. Ogin, Textile-reinforced composite materials, in: A.R. Horrocks, S.C. Anand (Eds.), Handbook of Technical Textile, Woodhead Publishing Limited, London, 2000, pp. 264e279. [21] R. Kamiya, et al., Some recent advances in the fabrication and design of threedimensional textile preforms : a review, Knowledge Creation Diffusion Utilization 60 (2000) 33e47. [22] B. Cox, M. Dadkhah, The macroscopic Elasticity of 3D woven composites, Journal of Composite Materials 29 (6) (1995) 785e819. [23] B.K. Behera, B.K. Hari, Woven Textile Structure: Theory and Application, Woodhead Publishing Limited, Cambridge, 2010. [24] P.T. Curtis, S.M. Bishop, An assessment of the potential of woven carbon fibre-reinforced plastics for high performance applications, Composites 15 (4) (1984) 259e265. [25] The Hong Hong cotton spinncer association, Textile Handbook, Hong Kong Productivity Council, Hong Kong, 2001. [26] M. Rao, B. Sankar, G. Subhash, Effect of Z-yarns on the stiffness and strength of threedimensional woven composites, Composites Part B: Engineering 40 (6) (2009) 540e551. [27] Y. Mahadik, K. Brown, A.S. Hallett, Characterisation of 3D woven composite internal architecture and effect of compaction, Composite Part A, Applied Science Manufacture 41 (7) (2010) 872e880. [28] M. Ansar, W. Xinwei, Z. Chouwei, Modeling strategies of 3D woven composites: a review, Composite Structures 93 (8) (2011) 1947e1963. [29] S. Buchanan, A. Grigorash, E. Archer, A. McIlhagger, J. Quinn, G. Stewart, Analytical elastic stiffness model for 3D woven orthogonal interlock composites, Science and Technology 70 (2010) 1597e1604.

148

Smart Textiles for In Situ Monitoring of Composites

[30] X. Chen, L. Taylor, L. Tsai, An overview on fabrication of three-dimensional woven textile preforms for composites, Textile Research Journal 81 (9) (2011) 932e944. [31] S. Nehme, A. Hallal, F. Fardoun, R. Younes, B. Hagege, Z. Aboura, et al., Numerical/ analytical methods to evaluate the mechanical behavior of interlock composites, Journal of Composite Material 45 (16) (2011) 1699e1716. [32] F. Boussu, The use of warp interlock fabric inside textile composite protection against ballistic impact, Textile Research Journal 81 (4) (2011) 344e354. [33] S. Lomov, I. Verpoest, F. Robitalle, Manufacturing and internal geometry of textiles, in: Design and Manufacture of Textile Composites, Woodhead Publishing Limited, London, 2005, pp. 1e61. [34] K. Leong, B. Lee, I. Herszberg, M. Bannister, The effect of binder path on the tensile properties and failure of multilayer woven CFRP composites, Composites Science and Technology 60 (2000) 149e156. [35] J. Solden, B. Hill, Conventional weaving of shaped preforms for engineering composites, Composite Part A 29 (1998) 757e762. [36] S. Rudov-Clark, A.P. Mouritz, L. Lee, M. Bannister, Fiber damage in the manufacture of advanced three-dimensional woven composites, Composite Part A, Applied Science Manufacture 34 (10) (2003) 963e970. [37] V. Guenon, T. Chou, J. Gillespie, Toughness properties of a three-dimensional carbonepoxy composite, Journal of Material Science 24 (1989) 4168e4175. [38] B.N. Cox, M.S. Dadkhah, W.L. Morris, J.G. Flintoff, Failure mechanisms of 3D woven composites in tension, compression, and bending, Acta Materialla Material 42 (12) (1994) 3967e3984. [39] S.V. Lomov, D. Ivanov, G. Perie, I. Verpoest, Modelling 3D fabrics and 3D reinforced composites: challenges and solutions, in: 1st World Conference on 3D Fabrics, Manchester, 2011. [40] M. Lefebvre, F. Boussu, D. Coutelier, Influence of high-performance yarns degradation inside three-dimensional warp interlock fabric, Journal of Industrial Textiles 42 (2013) 475e488. [41] J. Ruzand, G. Guenot, Multiaxial Three-Dimensional Fabric and Process for its Manufacture, International Patent Brevet WO: 94/20658, 1994. [42] G. Mood, Multi-axial Yarn Structure and Weaving Method, United States Brevet 5-540260, 1996. [43] L. Bryn, M.A. Islam, W.L. Lowery, H.D. Harries, Three Dimensional Woven Forms with Integral Bias Fibers and Bias Weaving Loom, United States Brevet 6-742-547, 2004. [44] S. Nayfeh, J. Rohrs, O. Rifni, et al., Bias weaving machine, United States Brevet 7-077167, 2006. [45] M. Lima, R. Fangueiro, A. Costa, C. Rosiepen, V. Rocha, Multiweave, a prototype weaving machine for multiaxial technical fabrics, Indian Journal of Fibre and Textile Research 34 (2009) 59e63. [46] G.L. Farley, Method and Apparatus for Weaving a Woven Angle Ply Fabric, United States Brevet 5,224,519, 1993. [47] M. Anahara, Y. Yasui, H. Omori, Three-Dimensional Textile and Method of Producing the Three Dimensional Textile, Europe Brevet 0-426-878-A1, 1991. [48] Y. Yasui, M. Anahara, F. Hori, Y. Mita, Production of Three-Dimensional Woven Fabric, Japan Brevet 05-106-140, 1993. [49] H. Uchida, T. Yamamoto, H. Takashima, T. Yamamoto, S. Nishlyama, M. Shinya, ThreeDimensional Weaving Machine, United States Brevet 6-003-563, 1999.

Smart textiles for monitoring and measurement applications

149

[50] A.K. Bilisik, Multiaxis three dimensional woven fabric, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabric Technology, INTECH, Athens, 2011 (p. on line). [51] A.K. Bilisk, M.H. Mohamed, Multiaxis three-dimensional (3D) flat woven preform-tube Carrier weaving, Textile Research Journal 80 (8) (2010) 696e711. [52] A.K. Bilisik, Multiaxial Three Dimensional (3D) Woven Carbon Preform, United States Brevet 6-129-122, 2000. [53] K. Bilisik, Three-dimensional braiding for composites: a review, Textile Research Journal 83 (13) (2012) 1414e1436. [54] P. Stickler, M. Ramulu, P. Johnson, Experimental and numerical analysis of transverse stitched T-joints in bending, Composite Structures 50 (1) (2000) 17e27. [55] A.P. Mouritz, B. Cox, A mechanistic approach to the properties of stitched laminates, Composite Part A 31 (1) (2000) 1e27. [56] A.P. Mouritz, C. Baini, I. Herszberg, Mode I interlaminar fracture toughness properties of advanced textile fiber glass composites, Composite Part A 30 (7) (1999) 859e870. [57] F. Edgren, C. Soutis, L. Asp, Damage tolerance analysis of NCF composite sandwich panels, Composites Science and Technology 68 (2008) 2635e2645. [58] G. Bibo, I.K.M. Hogg, Mechanical characterisation of glass- and carbon-fibre-reinforced composites made with non- crimp fabrics, Composites Science and Technology 57 (1997) 1221e1241. [59] P. Chang, A. Mouritz, B.N. Cox, Elevated temperature properties of pinned composite lap joints, Journal of Composite Material 42 (8) (2008) 741e769. [60] A.P. Mouritz, Compression properties of z-pinned composite laminates, Composite Science Technology 67 (15) (2007) 3110e3120. [61] E.D. Witten, T. Kraus, M. Kuhnel, Composite Market Report 2017, November 28, 2016 [En ligne]. Available: http://www.eucia.eu/userfiles/files/20161128_market_report_20 16_english.pdf. [62] C. Warren, Lower Cost High Performance Carbon Fibre, U.S. Department of Energy, Oak Ridge, 2011. [63] F. Composites, Aramid Fiber Stronger and Safer with Aramid Fiber, Fibermax Composites, 2017 [En ligne]. Available: http://www.aramid.eu. [64] R. i. China, Global and China Aramid Fiber Industry Report, 2015e2018, Research in China, Beijing, 2015. https://www.reportlinker.com/p02515745/Global-and-ChinaAramid-Fiber-Industry-Report.html. [65] marketsandmarketscom, Aramid Fiber Market by Type (Para-aramid, Meta-aramid, and Others), and Application (Security & Protection, Frictional Materials, Optical Fiber, Tire Reinforcement, Rubber Reinforcement, Electrical Insulation, Aerospace, and Others) e Global Trends & Forecasts to 2021, Marketsandmarkets, London, 2016. [66] R. Linker, Global Natural Fiber Composite Market 2015-2020: Trends, Forecast, and Opportunity Analysis, December 2015, March 30, 2016 [En ligne]. Available: http:// www.prnewswire.com/news-releases/global-natural-fiber-composite-market-2015-2020trends-forecast-and-opportunity-analysis-december-2015-300243712.html. [67] N. Institute, WPC/NFC Market Study, NOVA Institute, H€ urth, 2014. [68] P. Irving, C. Soutis, Polymer Composites in the Aerospace Industry, Woodhead Publishing Limited, London, 2014. [69] P. Pasbakhsh, A. Haghi, E. Zaikov, Applied Research on Polymer Composites, AAP Research Notes on Polymer Engineering, Science and Technology, 2015. [70] G. Koronis, A. Silva, M. Fontul, Green composites, a review of adequate materials for automotive applications, Composites Part B 44 (2013) 120e127.

150

Smart Textiles for In Situ Monitoring of Composites

[71] M. Reddy, S. Vivekanandhana, M. Misraa, S. Bhatiac, A. Mohantya, Biobased plastics and bionanocomposites: current status and future opportunities, Progress in Polymer Science 38 (2013) 1653e1689. [72] C.R. Farrar, K. Worden, An introduction to structural health monitoring, Philosophical Transactions of the Royal Society A 365 (2007) 303e315. [73] A.E. Aktan, F. Catbas, K. Grimmelsman, M. Pervizpour, Development of a Model Health Monitoring Guide for Major Bridges, Drexel Intelligent Infrastructure and Transportation Safety Institute, Drexel, 2003. [74] O. Buyukozturk, T.Y. Yu, Structural health monitoring and seismic impact assessment, in: Proc. 4th Nat. Conf. Earthquake Eng., Istanbul, 2003. [75] S. Nassar, X. Yang, Fastening and joining of composite materials, in: E. Patterson, D. Backman, G. Cloud (Eds.), Composite Materials and Joining Technologies for Composites, Proceedings of the 2012 Annual Conference on Experimental and Applied Mechanics, vol. 7, Springer New York, New York, 2012, pp. 5e23. [76] M. Ueda, S. Miyake, H. Hasegawa, Y. Hirano, Instantaneous mechanical fastening of quasi-isotropic CFRP laminates by a self-piercing rivet, Composite Structures 94 (2012) 3388e3393. [77] V. Virupaksha, S. Nassar, Effect of washers and bolt tension on the behavior of doublelap S2-glass fabric-epoxy composite joints, in: International Conference on Pressure Vessels and Piping (PVP 2008), Chicago, IL, USA, July 2008, pp. 27e31. [78] D. Liu, Y. Tang, W. Cong, A review of mechanical drilling for composite laminates, Composite Structures 94 (2012) 1265e1279. [79] J. Oh, Y. Kim, D. Lee, Optimum bolted joints for hybrid composite materials, Composite Structures 38 (1e4) (1997) 329e341. [80] J. Gebhardt, J. Fleischer, Experimental investigation and performance enhancement of inserts in composite parts, in: 5th CATS (Conference on Assembly Systems and Technologies) 2014-CIRP, Procedia CIRP 23, Dresden, Germany, November 12e14, 2014. [81] C. Dufour, F. Boussu, P. Wang, D. Soulat, K. Kalnins, E. Labans, P. Lefort, M. Wallenius Henrinksson, 3D warp interlock fabric structure as a solution of composite joining with metallic part, in: Proceedings of the 6th World Conference on 3D Fabrics and Their Applications, Raleigh, NC e USA, 2015. [82] C. Ageorges, L. Ye, M. Hou, Advances in fusion bonding techniques for joining thermoplastic matrix composites: a review, Composites: Part A 32 (2001) 839e857. [83] J. Esteves, S. Goushegir, J. dos Santos, L. Canto, E. Hage Jr., S. Amancio-Filho, Friction spot joining of aluminum AA6181-T4 and carbon fiber-reinforced poly(phenylene sulfide): effects of process parameters on the microstructure and mechanical strength, Materials and Design 66 (2015) 437e445. [84] S. Ucsnik, M. Scheerer, S. Zaremba, D. Pahr, Experimental investigation of a novel hybrid metalecomposite joining technology, Composites: Part A 41 (2010) 369e374. [85] A. Bogdanovich, M. Dannemann, J. D€oll, T. Leschik, J. Singletary, W. Hufenbach, Experimental study of joining thick composites reinforced with non-crimp 3D orthogonal woven E-glass fabrics, Composites: Part A 42 (2011) 896e905. [86] Z. Huang, S. Sugiyama, J. Yanagimoto, Applicability of adhesiveeembossing hybrid joining process to glass-fiber-reinforced plastic and metallic thin sheets, Journal of Materials Processing Technology 214 (2014) 2018e2028. [87] Z. Huang, S. Sugiyama, J. Yanagimoto, Adhesiveeembossing hybrid joining process to fiber-reinforced thermosetting plastic and metallic thin sheets, in: 11th International Conference on Technology of Plasticity, ICTP 2014, Procedia Engineering, Nagoya, Japan, October 19e24, 2014.

Smart textiles for monitoring and measurement applications

151

[88] M. Gude, W. Hufenbach, R. Kupfer, A. Freund, C. Vogel, Development of novel formlocked joints for textile reinforced thermoplastics and metallic components, Journal of Materials Processing Technology 216 (2015) 140e145. [89] B. Cosson, M. Deléglise, W. Knapp, Numerical analysis of thermoplastic composites laser welding using ray tracing method, Composites: Part B 68 (2015) 85e91. [90] E. Rodríguez-Vidal, J. Lambarri, C. Soriano, C. Sanz, G. Verhaeghe, A combined experimental and numerical approach to the laser joining of hybrid Polymer e metal parts, in: In 8th International Conference on Photonic Technologies LANE 2014, Physics Procedia, Furth, Germany, September 8e11, 2014. [91] P. Cognard, Assemblage des composites e Les points forts du collage, Techniques de l’Ingénieur, Traité Génie mécanique (Juillet 2003) 1e4, n %1BM 7 624. [92] P. Cognard, Collage des composites e Secteurs routier et ferroviaire, Techniques de l’Ingénieur, Traité Génie mécanique (October 2003) 1e7, n %1BM 7 627. [93] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Procédé pour l’assemblage entre une piece en matériau métallique et une piece en matériau composite a matrice organique, France Brevet FR3 030329 A1, December 19, 2014. [94] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Assembly Method between a Part Made of Metal Material and a Part Made of Organic Matrix Composite Material; Corresponding Parts Made of Organic Matrix Composite Material and Assembly, France Patent WO 2016/097656 A1, December 23, 2016. [95] K.E. Jackson, L. Edwin, R. Fasanella, L. Boitnott, K. Lyle, Full scale crash test and Finite Element Simulation of a composite prototype helicopter, Aircraft Design, Testing and Performance 1 (2003) 121e129, n %11. [96] J. Obradovic, S. Boriab, G. Belingardia, Lightweight design and crash analysis of composite front impact energy absorbing structures, Composite Structures 94 (2012) 423e430, n %12. [97] E. Mehdi, A. Mokhtar, D. Magid, F. Ahmadun, A. Prithvi Raj, S. Taher, A new composite energy absorbing system for aircraft and helicopter, Composite Structures 75 (2006) 14e23, n %11-4. [98] A.M. Grancaric, A. Tarbuk, I. Jerkovic, et al., Surface free energy of multilayered weft-knitted fabrics and related composite plates, in: CompositeWeek@Leuven (TexComp-11): Proceedings, Leuven, 2013. [99] A.M. Grancaric, J.V. Risicato, I. Jerkovic, et al., Interface phenomena of PP/glass braided structures, in: XIIIth International Izmir Textile and Apparel Symposium Book of Proceedings, Izmir, 2014. [100] A.M. Grancaric, V. Kovacevic, M. Leskovac, et al., Interface phenomena of hydrolyzed polyester fabric, in: 2nd International Textile, Clothing & Design Conference e Magic World of Textiles Book of Proceedings, Dubrovnik, 2004. [101] E. Chibowski, L. Holysz, G.A.M. Kip, et al., Surface free energy components of glass from ellipsometry and zeta potential measurements, Journal of Colloid and Interface Science 132 (1) (1989) 54e61. [102] C. Faulkner, What Is NFC? Everything You Need to Know, Techreader, May 9, 2017 [En ligne]. Available: http://www.techradar.com/news/what-is-nfc. [103] N. Pelly, J. Hamilton, How to NFC, May 10, 2011 [En ligne]. Available: https://www. youtube.com/watch?v¼49L7z3rxz4Q. [104] I. p. NFC.org a Square, Near Field Communication Versus Bluetooth, November 28, 2012 [En ligne]. Available: http://nearfieldcommunication.org/bluetooth.html.

This page intentionally left blank

Composites and hybrid structures 2.1 2.1.1

2

Compositesdterms and definitions Introduction

The reinforced composite materials, which are also called “Advanced composite materials,” are used increasingly after the Second World War as structural engineering materials because of their light weight, high specific stiffness, high specific strength, excellent corrosion resistance, fatigue resistance, and impact resistance compared to common metallic alloys. Furthermore, their ability to be manufactured in complex shapes reduces the number of parts in an assembly. All those reasons explain the interest, despite high fabrication cost, compared to conventional metal alloys [1]. The reinforced composite material is a subgroup of the composite material class, which is one of the four engineering material classes (metals, polymers, ceramics, and composites) [2]. The composite material is defined as multiphase material composed of two or more constituents, with distinct boundary and different properties, combined together to optimize one or more specific properties [1,3e7]. The reinforced composite material is composed of reinforcing phase embedded in continuous matrix phase. The matrix fixes the reinforcement in position, transfers and distributes the loads between the reinforcing components by the shear adhesion forces, and preserves the reinforcement from the external environment conditions. The typical matrix materials are thermoset polymer, thermoplastic polymer, metal, carbon, and ceramic. The reinforcement plays the important role to control the stiffness and strength properties of the composite structure. It exists in three forms (Fig. 2.1), particles, flakes, and fibers, which could be made of short (discontinuous) fibers or long (continuous) fibers [1]. This chapter deals with the composite reinforced with continuous fibers, made of different materials. The typical used fiber material is carbon, glass, quartz, metal, boron, and organic (natural and synthetic) fibers. Furthermore, the continuous fiber reinforcement could be tailor-made in various architectures, according to the final product application conditions (type and orientation of the loads that the product will support). The architecture of the reinforcement refers to the position of the fibers in the reinforcement relative to a defined coordinate system. Different classifications are introduced in the literature for the architecture of the textile reinforcement depending on the fabrication technology; the alignment plans of the constitutive fibers and the orientation of the constitutive fibers. Regarding the fabrication technology, the conventional textile technologies are used to design and manufacture fiber-reinforcements such as weaving, knitting, stitching, and braiding, Fig. 2.2. Furthermore, several novel technologies were also developed to produce special-shaped composite [9].

Smart Textiles for In Situ Monitoring of Composites. https://doi.org/10.1016/B978-0-08-102308-2.00002-4 Copyright © 2019 Elsevier Ltd. All rights reserved.

154

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

(c)

Figure 2.1 Three forms of the reinforcement; (a) particle, (b) flake, (c) fiber (long fiber).

In terms of the alignment plans of the constitutive fibers, the fiber-reinforced composites are classified into three groups, as illustrated in Fig. 2.3. 1D fiber-reinforced composites: in which the fibers are parallel and aligned in one direction. 2D fiber-reinforced composites: in which the fibers are aligned in one plane of the structure (XY plane). 3D fiber-reinforced composites: in which, in addition to the aligned fibers on XY plane of the structure, another set of fiber is aligned along the orthogonal axis (Z axis) corresponding to through-thickness axis of the structure denoted through-thickness fiber reinforcement or Z fiber.

Regarding the orientation of the main axis of constitutive fibers, the fiber-reinforced composites are classified into nonaxial, monoaxial, biaxial, triaxial, and multiaxial structure, as presented in Fig. 2.4. In the present section, the preform expression is used to mention the dry state of the fiber-reinforcement before being impregnated into the matrix.

Composites and hybrid structures

155

Biaxial woven

High modulus woven

Multilayer woven

Triaxial woven

Tubular braid

Tubular braid laid in warp

Weft knit

Weft knit laid in weft

Weft knit laid in warp

Weft knit laid in weft laid in warp

Square braid

Square braid laid in warp

Warp knit

Warp knit laid in warp

Weft inserted Weft inserted warp knit warp knit laid in warp

XD

Stitchbonded laid in warp

Biaxial bonded

XYZ laid in system

Flat braid

Flat braid laid in warp

3-D braid

3-D braid laid in warp

Figure 2.2 Textile fiber reinforcements [8].

In following paragraphs, architectures of the 2D/3D fiber-reinforced composites produced by different textile technologies, in addition to their advantages and disadvantages, are briefly presented. Then, the necessity to elaborate novel 3D multiaxis woven reinforcement architectures, which are the focus of this work, are presented.

2.1.2

Laminate fiber reinforced composites

The laminate fiber reinforced composites are 2D structures, made by stacking various laminas one upon the other in specific order and specific relative orientation, as illustrated in Fig. 2.5 [10]. Each ply can be a unidirectional layer of juxtaposed long continuous fibers (filaments) aligned on one plane, or it could be a layer of conventional 2D woven fabrics, Fig. 2.6. The 2D woven fabric will be presented in the next section. The orientation of the lamina is expressed by the orientation of the longitudinal axis of the fibers in the case of unidirectional lamina or by the orientation

156

Smart Textiles for In Situ Monitoring of Composites 3-D

Planar (2-D)

Linear

Discon- Continuous tinuous (filament) (spun)

Woven

Knit

Braided Woven Knit

Braided Nonwoven 3-D

Angel interlock Triaxial Biaxial Triaxial Weft

2 step

Core

Biaxial Triaxial

XYZ

MWK

Warp

Fully fashioned weft knit

Multifilament Monofilament 6-ply 4-ply impaled impaled Flat

Textured Twisted

Figure 2.3 Classification of the fiber-reinforced composites in terms of the alignment plans of the constitutive fibers [8].

Axis Dimension

Nonaxial

Monoaxial

1D

2D Linear element

Biaxial

Triaxial

Multiaxial

Triaxial weave

Multiaxial

Roving yarn Preimpregnation sheet

Chopped

Plane weave

Strandmat z

Z

X

Y 3D braid

X 3D

Multi-ply weave

Y

Triaxial 3D weave

Multiaxial 3D weave

z Plane element

Y X

Laminate type

H or I beam

Honeycomb type

Figure 2.4 Classification of the fiber-reinforced composites in terms of the orientation of the constitutive fibers [8].

Composites and hybrid structures

157

Warp yarns

+45° yarns

Weft yarns

–45° yarns

Warp yarns

Figure 2.5 Laminate made by stacking laminas in different orientation [10]. Z

X

Y

Warp yarns Weft yarns

Figure 2.6 Lamina fiber reinforced composite: on left lamina with unidirectional fiber, on right lamina with 2D woven fabric [10].

of the longitudinal axis of the warp yarns in the case of woven fabric relative to referencing coordinate system. The orientation of stacked laminas could vary between
90 and þ90 degrees. The filaments of the unidirectional lamina are often saturated with resinous material, which will be used subsequently as matrix, without consolidation to maintain the aligned filaments in parallel and to be handled easily in the next lay-up process. Such laminas are called “prepregs.” The two basic steps to fabricate the laminate fiber reinforced composite materials are lay-up and curing [10]. The lay-up process consists of laying down the filaments

158

Smart Textiles for In Situ Monitoring of Composites

in specific orientation to form the laminas then stacking them to form the laminates, whereas the curing process consists of solidification of the matrix material to get the final rigid desired shape. The principal lay-up processes are winding, laying, and molding. However, the lay-up process implies high amount of skilled labor. Furthermore, the cost of fabrication and the number of parts in the assembly increase when forming a complex part shape by the laminate [11]. The laminate composite materials are used in various applications from sport equipment to the spacecraft as result of their favorable properties resulting from the combination of the properties of advanced fibers with that of the matrix. These advantages could be summarized in: 1. High in-plane specific strength (strength to density) and high in-plane specific stiffness (stiffness to density) for the laminate composite comparing to metallic materials, as illustrated in Fig. 2.7 where the specific strength and specific modulus of graphite fiber, unidirectional graphite/epoxy lamina, and cross-ply graphite/epoxy laminate are compared to that of steel and aluminum [1]. 2. Low amounts of defects (fiber breakage, fiber misalignment, resin pocket) are caused. Such defects, which are usually associated to textile process (weaving, stitching, braiding, etc.) used usually to produce fiber reinforcements, contribute to pull down the mechanical properties of final composite products. 3. Low cost in comparison with the metallic material, where the initial cost of the laminate composite materials (cost of raw material, design, fabrication, and assembly) and the operating cost is lower [10]. 4. Saving weight resulting from high specific strength and specific stiffness leading to lower operating cost such as the fuel cost for transport applications. 5. Inversely, this structure has, on the other side, many drawbacks such as being a directional material, its mechanical properties are not identical in all direction and they depend on the fibers’ directions. Further, as it was mentioned earlier that the cost increases when forming

3

Specific strength (Ksi – 36.13 cm /kg)

5000 Graphite fiber 4000 Unidirectional graphite/epoxy 3000

2000

1000

Quasi-isotropic Cross-ply graphite/epoxy graphite/epoxy Aluminum Steel 0 0 300 500 100 200 400 3 Specific modulus (Msi – 36.13 cm /kg)

600

Figure 2.7 Specific strength in function of specific modulus of fiber, lamina, laminate, and metals [1].

Composites and hybrid structures

159

a complex part shape using the laminas because of increasing the parts count for the assembly and hence the labor cost [10]. Moreover, it has high thermal and moisture expansion coefficient and low operating temperature when use polymer matrix [1]. However, the principal drawback relating to the reinforcement architecture is the absence of the through-thickness fiber reinforcement, which is the origin of the poor strength and strength properties in the through-thickness direction. This is strongly influencing poor delamination resistance leading to low impact resistance and low post impact properties [8,12]. As presented by Tong [11], the strength and stiffness properties, in the through-thickness direction of the laminate, are often less about 10% of the in-plane properties, Fig. 2.8. Tong also mentioned poor post impact mechanical properties of the laminates by showing the degradation of the in-plane tensile and compressive strength after impact, as illustrated in Fig. 2.9. It was proposed, as a recommendation in view of the poor post impact properties, to increase the ply count but that, on other side, increases labor and cost. Otherwise, the 3D-reinforced structures can be considered as alternative materials, as they are delamination-free and damage tolerant [13e15].

It is also important to introduce proper definitions of Compressive and Tensile strengths: • •

Compressive strength is a limit state of compressive stress that leads to failure in a material in the manner of ductile failure (infinite theoretical yield) or brittle failure (rupture as the result of crack propagation, or sliding along a weak planedsee shear strength). Tensile strength or ultimate tensile strength is a limit state of tensile stress that leads to tensile failure in the manner of ductile failure (yield as the first stage of that failure, some hardening in the second stage and breakage after a possible “neck” formation) or brittle failure (sudden breaking in two or more pieces at a low stress state). Tensile strength can be quoted as either true stress or engineering stress, but engineering stress is the most commonly used.

2.2

Textile reinforced composites

2.2.1 2.2.1.1

Woven fiber reinforced composites 2D woven fabric

The 2D woven fabric is made by conventional weaving process consisting of interlacing two orthogonal sets of yarns, warp 0 degree and weft 90 degrees [16], although it is not possible to have yarns in other in-plane orientation on the conventional 2D weaving loom, Fig. 2.10. The weaving loom technology is well developed providing high productivity and reducing the manufacturing cost of composites. The interlacement of yarns attributes to the lamina stability when handling and conformability to form complex shape with no gaps. Therefore, it is not necessary to preimpregnate preforms. Thereby one manufacturing step is bypassed in comparison to prepreg. However, bending and sliding against loom machineries causes abrasion and fiber breakage; further, the interlacement of yarns induces to crimping them. That reduces the yarns strength and leads to reduction of the lamina strength and toughness in comparison with equivalent crossply (0e90 degrees) unidirectional laminate, as shown by Curtis and Bishop [17].

160

Smart Textiles for In Situ Monitoring of Composites

Compressive strength (MPa) 1400 1200 1000 800 600 400 200 0 Carbon/epoxy

E glass/epoxy In-plane property

Kevlar/epoxy

Through-thickness property

Tensile strength (MPa) 1400

1240

1240

1200

1020

1000 800 600 400 200

40

41

30

0 Carbon/epoxy

E glass/epoxy In-plane property

Kevlar/epoxy

Through-thickness property

Compressive strength (MPa) 1400

1240

1200 1000 800

620

600 280

400 200

170

140

140

Carbon/epoxy

E glass/epoxy

Kevlar/epoxy

0

In-plane property

Through-thickness property

Figure 2.8 Comparison of the in-plane tensile modulus, tensile strength, and compressive strength to that in the through-thickness properties for laminate composite materials [11].

Composites and hybrid structures

161

Post impact tensile strength (N) 1.2 1 0.8 0.6 0.4 0.2 0 0

2

4

6

8

10

12

14

Impact energy (J) Post impact compressive strength (N) 1.2 1 0.8 0.6 0.4 0.2 0 0

1

2

3

4

5

Impact energy (J) Figure 2.9 Degradation of the in-plane tensile and compressive strength after impact normalized to the strength before impact in function of the impact energy [11].

2.2.1.2

2D weaving process

On the conventional 2D weaving loom, Fig. 2.11, a fabric is made by interlacing two sets of yarns (warp and filler), which are perpendicular, with specific weave pattern by a series of weaving operations in a following order [18]: 1. Warp left-off: Warp yarns are wound in parallel with even tension on warp beam. The beam is turned a specific step for each weaving cycle, corresponding to demanded filler yarns count per unit length, to feed warp yarns into weaving zone under controlled tension with the aid of tensioning devices.

162

Smart Textiles for In Situ Monitoring of Composites Weaving zone

Warp yarns

Heddle

Fabric formation line

Shed

Fabric

Shuttle Warp beam

Closed reed

Lifting shafts

Blade Warp yarn

Fabric beam

Figure 2.10 Schematics of the conventional 2D weaving loom.

Filler yarn 90 degrees

Shed

Warp yarn 0 degree

Figure 2.11 Schematic representing formation of the shed on the 2D weaving loom. 2. Shedding: Each warp yarn passes through a heddle eye (one heddle for one warp). These heddles are attached to lift shafts. Warp yarns, which have the same state in weave pattern (that means same sequence of interlacement up/down with filler yarns), have their heddles attached to the same shafts. For each weaving cycle, a part of lift shaft moves up while the others move down to separate the warp yarns into two layers creating appropriate space between them across the loom, called shed, Fig. 2.11. The shed allows insertion of filler yarn across loom width. This proper space is essentially required to avoid crossing and rubbing warp yarns while insertion of the filler yarn. The distribution of the warp yarns on lift shafts and the sequence of lifting the shafts (up/down) across weaving carry out the demanded weave pattern for fabric.

Composites and hybrid structures

163

3. Filler insertion: For each weaving cycle, a filler yarn is inserted through created shed and it passes across the loom width. Different apparatuses are used to carry filler yarn such as shuttle, rapier, projectile, and air-jet. 4. Beating: The filler yarn, after being inserted, is packed tightly to the fabric by means of reed that is closed combs through the gaps created between its blades the warp yarns are passed and the number of warp yarns in one space in addition to the count of blades per unit length defining the count of warp yarns per unit length of fabric on loom. The reed has alternative movement; it moves back away from the fabric formation line to allow creating the shedding and insertion of the filler yarn, then it moves forward toward the fabric formation line to pack the inserted filler yarn. 5. Fabric take-up: The new packed filler yarn is pulled from the weaving formation zone by rolling up the fabric on fabric beam. The take-up roller and the fabric beam turn one step corresponding to the filler yarns count per one-unit length of fabric.

2.2.1.3

Multilayered (or 3D woven) fabric

The multilayered fabric is composed of several in-plane woven layers linked together by yarns passing in the through-thickness direction of the fabric called binder yarns or weaver yarns. The aim of developing this woven architecture was to overcome the disadvantage of laminate by combining the different layers of structure via through-thickness fiber reinforcement [19e24]. The composite made of this preform is classified as 3D woven fiber reinforced composites. Generally, the in-plane yarns of the 3D woven preform are oriented in 0 and 90 degrees (warp and filler yarns), such as the 2D woven preform. However, several technologies are developed recently to obtain 3D multiaxis woven preform containing in-plane yarns oriented in a direction other than 0 and 90 degrees. This architecture and the used technologies are the axis of the next chapter (Fig. 2.12). Lifting shafts

Shed

Warp yarns

Fabric and take-up rollers

Warp beam

Heddles

Reed

Shuttle with weft yarn

Figure 2.12 Schematic representing of conventional 2D weaving loom.

164

Smart Textiles for In Situ Monitoring of Composites

Figure 2.13 Typical 3D woven architectures; (a) layer-to-layer angle interlock, (b) throughthickness angle interlock, (c) orthogonal.

Different weave patterns for the binder yarns are used to link the fabric layers [25], it could be classified into three weaving architectures categories: • • •

Layer-to-layer angle interlock: when the binder yarn passes from layer to adjacent one, then it returns to the first layer thus it links just two adjacent layers, Fig. 2.13(a). Through-the-thickness angle interlock: when binder yarn passes through the whole thickness of the fabric across more than two columns of weft yarns, Fig. 2.13(b). Orthogonal: when binder yarn passes through the whole thickness of the fabric for each column of weft yarns, Fig. 2.13(c). Here, there is no interlacement between warp and weft yarns where the alternate layers of virtually uncrimped in-plane yarns (warp and weft) are combined together just by the binder yarns [26].

The multilayered woven preform could be produced on the 2D weaving loom even with one weft insertion by adequate weave pattern. The multilayer weaving requires higher number of fed warp yarns, which raise the manufacturing cost and multiple weaving cycles. Certain 3D weaving looms are developed to reduce the abrasion incurred on warp and binder yarns while shedding aims at producing the orthogonal architecture, where just alternative lifting for binder yarns is required. Contrary to 2D weaving, in case of 3D weaving loom, multiple weft yarns are simultaneously inserted for each weaving cycle, Fig. 2.14. In addition to the flat woven preforms either 2D or 3D, more complex threedimensional shapes, such as I-beam, could be produced on the conventional 2D weaving loom by means of programmable shedding apparatus executing adequate weave pattern [27]. The manufacture cost of the 3D woven composite is higher than for laminates for many applications, where the woven composite is not produced in large commercial quantities [11].

Composites and hybrid structures

165

Filler yarn 90 degrees

Warp yarn 0 degree Binder yarn (Z)

Figure 2.14 Schematic representation of a 3D orthogonal weaving process with multiple filler insertion.

As it was mentioned for 2D woven preform, the weaving process causes strength reduction for the warp, weft, and binder yarns especially for brittle yarns, because of abrasion and breakage of fibers resulting from bending and sliding the yarns against the loom machineries. Rudov-Clark et al. [28] analyzed the damage on warp yarn and binder yarn of glass fiber arisen while weaving 3D preform on 2D jacquard loom. They noticed a strength reduction of about 30% and 50% for warp yarn and binder yarn respectively with a loss of 5%e7% in the young’s modulus, Fig. 2.15. The main reason of this reduction is attributed to the abrasion; therefore, it was suggested to coat the loom apparatus with wear resistant material having low coefficient of friction. As-received (E = 26.4 + – 0.61 GPa) Warping stage (E = 25.7 + – 1.01 GPa) Tensioning stage (E = 26.4 + – 0.62 GPa) Take-up stage (E = 24.8 + – 0.81 GPa)

100 80 60 40 20 0

20

22

24 26 28 Young’s modulus (GPa)

30

As-received (= 1190 + – 120 MPa) Warping stage (= 1010 + – 110 MPa) Tensioning stage (= 930 + – 120 MPa) After weaving stage (= 842 + – 63 MPa)

(b) 100

Cumulative probability (%)

Cumulative probability (%)

(a)

80 60 40 20 0 600

800

1000 1200 1400 Tensile strength (MPa)

1600

Figure 2.15 Cumulative probability distribution of young’s modulus (a) and tensile strength (b) of warp yarns at different weaving stages [28].

166

Smart Textiles for In Situ Monitoring of Composites

The waviness in through-the-thickness direction of the yarns, resulting from the interlacement of yarns when weaving, affects strength, ductility, and fatigue life of woven composite. The waviness magnitude varies between yarns sets depending on their tension while weaving. Generally, filler yarns are suffering from crimp higher than warp yarns. Furthermore, the waviness magnitude depends on the yarn path associated with the weave pattern. Therefore, the crimp is relatively high in layer-to-layer architecture of 3D woven fabric, whereas it is less in the through-thickness angle interlock architecture and least in orthogonal architecture [14]. After consolidation process, two types of regions could be found in the 3D woven composite: the first is regions of high-fiber content making infiltration of viscous resin difficult resulting in formation of porosity and the second is regions rich in resin (low-fiber content) resulting from crimp and push aside of the in-plane yarns by the binder yarns [26]. These microstructure damages and defects caused by weaving process influence the in-planes mechanical properties of composite. Tong et al. [11] analyzed the experimental results provided from researches about characterization the mechanical properties of 3D woven composites and he found that the young’s modulus of 3D woven composite is within 20% (improved or degraded) of that of equivalent 2D laminates. Further, either the content of the binder yarns or the 3D weave pattern (angle interlock or orthogonal) has no significant effects on the Young’s modulus of 3D woven composite. The same results were revealed for compressive strength of 3D woven composite in comparison with 2D laminates, whereas the tensile strength is mostly lower, but no more than 20%. It has also been found that the plastic straightening of heavily crimped in-plane yarns at low strain is the origin of the beginning of softening the stiffness of 3D woven composite, which continues with straightening all yarns while raising the strain. Likewise, the plastic deformation of the resin within the crimped yarn is the cause of the formation of kink band responsible of failure compression. Contrary to the laminate composites, the 3D woven composites have high fracture toughness and high delamination resistance due to the through-thickness fiber reinforcement (binder). Guénon et al. [29] showed that even for low binder yarns content about 1% in 3D carbon/epoxy composite, the delamination toughness for mode I is about 14% higher than for 2D carbon/epoxy prepreg laminates. In the same manner, it was found that the 1% content of binder yarns leads to delamination toughness for mode II two or three times higher. Lomov et al. [30], compared noncrimp 3D orthogonal woven composites and four-ply laminates of plain weave. He noted a higher strength, failure strain, and damage initiation thresholds for 3D woven composites. Further, he mentioned that the delamination between plain weave plies of laminates initiates at applied strain between 1.5% and 2%, whereas it never arises in the 3D woven composites. The high delamination resistance gives better damage resistance to 3D woven composites. That results in higher post impact mechanical properties for 3D woven composite than 2D laminates.

Composites and hybrid structures

2.2.1.4

167

3D weaving process

Contrary to the 2D woven fabric, the 3D woven preform is characterized by multilayer of in-plane yarns through the thickness of the preform linked together by binder yarns (Z-yarns) inserted via weaving process [23,25]. The 3D woven structure could be produced on the conventional 2D weaving loom to profit from its high productivity [27,31]. However, some adjustments are required [11]. Firstly, it is necessary to equip the loom with multiwarp beams, one beam for each warp yarns layer because of the difference in required tension between warp layers while weaving cycles. Binder yarns are also fed from separate beams because of the difference in its path in the structure comparing to warp yarns so dissimilar tension is required. Secondly, because on the 2D loom only one shed could be created enabling one insertion of filler yarn, a succession of shed formations and filler insertions are required to make one column of filler yarns in the through the thickness of the preform corresponding to one weaving cycle. The number of filler insertion per weaving cycle corresponding to the number of rows in one column of fabric is equal to the filler layers’ number. The formed fabric is pulled out of weaving zone for one step at the end of each weaving cycle and not at the end of each filler insertion operation. The multilayered woven fabric is characterized by important thickness regarding to the 2D woven fabric. Consequently, the use of the conventional take-up device consists in winding the formed fabric on fabric beam that causes distortion of the formed layers. Therefore, the linear take-up device is preferred here. However, the formation of multiple successive sheds for one weaving cycle increases the microstructure damages incurred on warp yarns by rubbing this yarns with the heddle’s eye and by abrasion to the other loom machineries. That reduces the mechanical properties of warp yarns [28,31], as explained in the previous chapter. Furthermore, the higher number of fed warp yarns and of the multiple filler insertion for one weaving cycle decreases the loom productivity and raises the manufacturing cost. Therefore, some researchers worked on construction of a special 3D weaving loom enabling formation of multiple sheds allowing simultaneous insertions of multiple filler yarns, meaning one column by one stage. On a 3D loom used to produce special 3D architectures such as 3D orthogonal woven preform, where the in-plane yarns are not interlaced and are bonded together only by the binder yarns, the sheds are fixed (no change warp level), which minimizes warp yarns rubbing by heddles.

2.2.1.5

Multiaxis weaving process

On 2D/3D loom, warp yarns are fed parallel to loom axis (0 degree orientation) and they advance, while weaving process with keeping this orientation. On the contrary, the bias yarns have to be displaced for each weaving cycle one step transversally to loom main axis to get q degrees orientation within the fabric relative to warp yarns axes. That is the essential difference of multiaxis weaving process comparing to conventional 2D and 3D

168

Smart Textiles for In Situ Monitoring of Composites

weaving process. To achieve this insertion, special mechanisms dedicated to feed bias yarns and to control their position in the weaving zone are required in addition to adapt the other conventional weaving operations to these requirements.

2.2.1.6

Two dimensional multiaxis weaving

Ruzand and Guenot [32] proposed lappet-weaving apparatus, by modifying standard 2D weaving machine. That enables forming a pair of symmetric bias yarns layer on the top and/or bottom faces of woven fabric. Heddle hooks (one for each bias yarn) are disposed on segmented transversal bar called lappet bar, Fig. 2.16. The heddle hook has also two movements; first, the vertical one allowing insertion of filler yarns and it is driven away to enable beating up by reed, and the second, horizontal movement to put the bias yarns in right orientation. However, the bias yarns are not interlocked to filler and warp yarns where they are maintained in position within the fabric by the interlacement between filler and warp yarns. The two lappet bars on each side of the fabric move oppositely in transversal direction. The length of the lappet bar is greater than the fabric width. Once the yarns of one segment of lappet bar reach the fabric edge, they are gripped between the selvedge and the guide and they are cut; then this segment is transferred to the opposite lappet bar and it is reattached. The bias layers cover the entire width of the fabric and the boundary bias yarns are folded to next layer. Consequently, edge-to-edge uniform bias yarns layers is formed. Moreover, disposing pair of opposing bias yarns layer (one at 45 degrees and the other on
45 degrees) creates balanced fabric. But it is not possible by this technique to arrange the bias layers other than the two outer faces of the fabric. As well, this technology solution is limited to fabricate layer-to-layer interlock, as reported by Kamiya [13]. Mood [33] worked also on modifying the 2D weaving loom to produce four layers multiaxis woven fabric (warp, filler, þ and
bias) by means of special split reed and

Warp yarns Bias yarns

Bias yarns Filler yarns

Figure 2.16 Schematic of the modified lappet multiaxis weaving by Ruzand and Guenot [32] with produced architecture.

Composites and hybrid structures

169

(a)

(b) 1

2

3

4 8 5

6

7

Figure 2.17 Multiaxis four layers woven fabric (a), one case of distribution the bias yarns by jacquard mechanism into upper movable reed and lower fixed reed (b), Mood [33].

jacquard shedding mechanism, Fig. 2.17. The reed is split into open upper reed that is able to move transversally to loom main axis and open lower one that is fix. The bias yarn passes through the eye of one especial jacquard heddle. To shift the bias yarns transversally, they are selected by jacquard mechanism and placed in the upper reed or in the lower reed. The jacquard mechanism also creates required shed for insertion filler yarn to be interlaced with warp yarns; then the filler yarns is beaten to the fabric by other reed. Thereby, the þ and
bias yarns are locked in position between warp and filler yarns. However, the large shedding motion could cause higher fiber damage as reported by Kamiya [13]. Furthermore, using this technique to fabricate multilayer multiaxis fabric could be too complicated for higher production rate. Bryn et al. [34] also worked on producing four layers multiaxis fabric in various cross-section shapes, Fig. 2.18, based on multilayer narrow weaving principle. Individual hook is used to control the position of one bias yarns. When bias yarn reaches the fabric edge, it will be folded to next opposite bias yarns layer. Comparing to the four layers multiaxis fabric fabricated by Mood [33], on Bryn’s loom the þ and
bias yarns are also interlaced with warp and filler yarns. The length of produced fabric is restricted by the length of bias yarns so the continuous manufacturing of fabric is not available on this loom [15]. Nayfeh et al. [35] used the braider carrier to feed bias yarns on 2D multiaxis weaving loom, Fig. 2.19. The carriers are equipped with controlling tension device able to maintain the tension of bias yarn during the weaving process. They move along predetermined path according to desired fabric cross section. A shuttle into formed shed, to be interlaced with warp yarns, inserts a filler yarn. Then, an open reed is used to pack the inserted filler yarn into the fabric.

170

Smart Textiles for In Situ Monitoring of Composites

Figure 2.18 Multiaxis four layers woven fabric (a) and the developed loom by Bryn [34] (b), which is able to produce fabric in various cross-section shape (c).

Figure 2.19 Multiaxis weaving machine developed by Nayfeh [35].

Lima et al. [36] developed a 2D multiaxis weaving loom, Fig. 2.20(a). The manufactured fabric is composed of four yarn sets: warp, filler, þ and
bias yarns. The bias yarns are fed from two rotating beams, and then they pass through special tension compensation device. By stepwise transversal movement for bias yarns, they form pair of two opposite parallel uniform layers. For each weaving cycle a layer of warp yarns is raised and form with bias yarns layers a shed to insert filler yarns. However, the produced fabric is in open weave architecture, Fig. 2.20(b), resulting in low fiber volume fraction obtained when it is impregnated. Furthermore, this machine is limited to 2D woven fabrics.

2.2.1.7

Multilayer multiaxis weaving

Farley [37] presented a lappet weaving apparatus enabling insertion of bias yarns during the weaving process. This apparatus consists of needles disposed on transversal rack holding also bias yarns supply, where one bias yarn passes through needle’s eye

Composites and hybrid structures

171

Figure 2.20 (a) Scheme of multiaxis weaving machine developed by Lima et al. [36], (b) architecture of produced multiaxis 2D woven fabric. 5

4

3

2

1

Figure 2.21 Scheme of lappet multiaxis weaving apparatus developed by Farley [37]. 1dfiller yarns, 2dbias yarns, 3dneedle’s eye, 4dneedles, 5dwarp yarns.

as illustrated in Fig. 2.21. The holding rack is placed in front of the shedding device and it has two movements: firstly, it moves up and down forming a shed to pass filler yarns; secondly, the bias moves transversally relative to loom axis. Thereby, bias yarns are interlaced with filler yarns so an interlocked bias ply structure could be produced without limitation concerning the number of layers and bias yarns ply could be placed in any layer through the thickness. Further, this apparatus produces edge to edge bias yarns ply, but there are no details about the uniformity of produced bias layers. However, there is no possibility, by the proposed apparatus, to produce multilayer woven architecture. Otherwise, high crimp is noted for bias yarns. Anahara et al. [38] worked on developing a multilayer multiaxis weaving machine using screw shafts and rapier needles apparatus, Fig. 2.21. The screw shafts displace

172

(a)

Smart Textiles for In Situ Monitoring of Composites

(b)

(c) Weft Warp

Figure 2.22 (a) Schematic of developed multiaxis multilayer weaving loom by Anahara et al. [38] using screw shaft and rapier needle, (b) produced architecture, (c) path of a pair bias yarns layers during weaving process by screw shafts.

the bias yarns transversally across the entire width of fabric forming edge-to-edge layer. However, as illustrated in Fig. 2.22(c), the path of bias yarns resulting from rotation of the screws, a pair of nonuniform bias layers in opposite orientation is formed. The screw shafts for each weaving cycle dispose the bias yarns at right position forming proper gaps to insert rapier needle holding binder (Z) yarns in the through-thickness direction of fabric. Thereby, the four in-plane yarn sets (warp, filler, þ and
bias) are locked in position within the fabric by the binder yarns. So, five-axial woven structure is manufactured. Multiple screw shafts could be used on the loom to get multiaxis multilayer woven fabric. Yasui et al. [39] also developed 3D multiaxis weaving machine, but they used movable guide blocks to displace the bias yarns in the weaving zone instead of screw shafts, Fig. 2.22. The guide blocks are actuated to move in closed cycle, Fig. 2.22(b), resulting in formation of pair of uniform bias yarns layers in opposite orientation. However, that compels to rotate the bias yarns supplying apparatus for continuous fabric production without intertwining the bias yarns. The bias yarns extend the entire fabric width that allows producing edge-to-edge layers. The rapier needles are also used to insert binder yarns in the through-thickness direction to fabricate orthogonal multilayer fabric. According to Kamiya et al. [13] using this technique to insert through the thickness reinforcement causes considerable damage to in-plane yarns. Uchida et al. [40] used as well the guide blocks to control the position of bias yarns in the weaving zone. Different method, Fig. 2.23, is used to actuate the guide blocks where special means are used to displace guide blocks sideways (transversally) and shift them up or down to fold the bias yarns once they reach the fabric edge so alternating its orientation from (
q/þq) to (þq/
q) but bias yarn still in the same layer. As a result, a pair of edge-to-edge nonuniform bias yarns layer is formed. Thereby, there is no need for circularly moving of bias yarns supply mechanism and the bias are fed

Composites and hybrid structures

173

Figure 2.23 Scheme of multiaxis 3D weaving machine developed by Uchida et al. [40] using guide blocks to control the position of bias yarns.

from beam. To control the variation of their tension during the weaving process, a tension device is developed. A pair of chain conveyors is used at the interval of fabric width to retention woven fabric to stabilize its width after weaving zone and to prevent high shrinkage caused by folded bias yarns (Fig. 2.24). Bilisik and Mohamed [41] developed a tube rapier weaving method to produce muliaxis three-dimensional fabric. Both warp and bias yarns pass through tube rapiers, Fig. 2.25(a), which are arranged in a matrix corresponding to a cross-section shape of

Warp guide

Guide block

Figure 2.24 Scheme of multiaxis weaving machine developed by Yasui et al. [39] using guide blocks to control the position of bias yarns with circularly moving for feeding mechanism.

Insertion

(c)

Three dimensional multiaxial woven structure

Z-yarn

±

Z-yarn needle

174

Weaving direction

(a)

Guide cylinders

Z-yarn Bias yarn

Warp (axial) yarn Filling yarn ±

Filling insertion needle

Bias yarn

Latch - needle

±

Bias yarn Tube-rapier for

bias yarn

±

Tube-rapier for axial yarn

Unit cell length

Tube-rapier for + bias yarn Filling yarn

Selvage yarn

Unit cell thickness

Selvage needle

Warp (axial) yarn

Z

Bobbins for axial yarn

Perforated table

Y

Unit cell width X

(b)

±

Bobbins for

bias yarn

Tube rapier for bias

Tube rapier for warp

Figure 2.25 Tube rapier weaving method; (a) scheme of developed multiaxis 3D weaving machine, (b) representation of transversal movement of each assembly of tube rapier for one bias yarns layer, (c) a unit cell of produced fabric illustrating of zig-zag path of bias yarns across fabric width [41].

Smart Textiles for In Situ Monitoring of Composites

Z-yarn insertion needle

Composites and hybrid structures

175

demanded preform. Filler and binder yarns are inserted by needles passing between the tube rapiers of bias and warp yarns. That reduces the damage incurred by insertion of the needles to warp and bias yarns met in other technologies. Tube rapiers of bias yarns displaces only transversally with respect to the loom main axis, Fig. 2.25(b). The yarns are folded when it reaches the fabric edge, but they are still in the same layer resulting in zig-zag nonuniform bias yarns layer, Fig. 2.25(c). On this loom, the bias yarns layer could be placed in any layer in through the thickness of the fabric allowing producing a 3D multiaxis woven fabric. To get pair of opposing uniform edge to edge bias yarns layer within multiaxis 3D woven fabric, Bilisik and Mohamed [41] developed another weaving method using a tube carrier. So each bias yarn is fed from individual bobbin and then it passes through one tube. The tubes move transversally inside a box and when one tube reaches fabric edge, it is shifted up or down to the next opposite bias yarns layer so it changes its orientation form (
q/þq) to (þq/
q) Fig. 2.26(b). Thereby, the tubes are in a circular movement requiring circular moving for feeding bobbins. The filler yarns and the binder yarns are inserted in the same manner as tube rapier weaving method by means of rapier needles. The applied tension on folded bias yarns while weaving process leads to high shrinkage for fabric width and trapezoidal weaving zone as illustrated by Fig. 2.27; the difference between the fabric width and weaving zone width. As consequence, the standard reed, having fix gap width between its blades, could not be used because of high friction between the inclined bias yarns and reed blades. Therefore, a special reed is used to beat the filler yarn into fabric formation line. In this reed, the distance between its blades decreases as it advances toward the fabric. However, this type of reed still causes important friction with in-plane yarns [41] and rotation of blades around its axis is also required. Otherwise, the needles rapiers damage also the in-planes yarns.

2.2.1.8

Polar multilayer multiaxis weaving

Yasui et al. [39] developed a polar multiaxis 3D weaving loom, Fig. 2.28(a). This machine is capable of fabricating preforms made of 24 layers (circumferential filler yarns, bias yarns, and axial warp yarns) with through the thickness fiber reinforcement (radial yarns). That is carried out by 144 yarn bobbins exchanged between two disks that are able to rotate and translate. Nevertheless, the large shedding motion resulting from exchanging the bobbins between the two disks could cause fibers damages. Further, concerning this method, it seems to be complex for high production rates. Bilisik also developed in 2000 [43] a multiaxis 3D circular weaving loom based on the 3D braiding principle, Fig. 2.29. The axial yarns form warp yarns and circumferential yarns form filler yarns while the bias yarns are fed by braider bobbins and placed only on the outside and inside surfaces of cylinder. The radial yarns form through the thickness yarn linking the different layers and they are inserted by means of special carrier units. This machine could produce thick and thin cylinders, and the presence of bias yarns improves its torsion properties, as reported by Bilisik in 2000 [43]. The stitched composites are 3D fiber reinforced composites produced by sewing high tensile yarns through stacked 2D conventional laminas as illustrated in Fig. 2.30 [11].

176

(a)

Warp tubes

Warp creel

± Bias yarn assembly Warp yarn

Z-needles-I Z-yarn Multiaxis 3D woven preform

Take-up

(c)

Z-yarn ± Bias yarns

Warp(axial) yarn ± Bias tube carrier boxes

Tension unit Filling

Beat-up Selvage bar Filling needles Z-needles-II

Unit cell length

Unit cell thickness

Unit cell width

Figure 2.26 Tube carrier weaving method; (a) scheme of developed multiaxis 3D weaving machine, (b) representation of circular moving of each assembly of tube carrier for one bias yarns layer, (c) a unit cell of produced fabric illustration of edge-to-edge path of bias [41].

Smart Textiles for In Situ Monitoring of Composites

(b)

Filling yarn ± Bias yarns

Composites and hybrid structures

177 Weaving zone width

Fabric width

Figure 2.27 Weaving zone on a developed multiaxis 3D weaving machine by Bilisik [42] using tube carrier weaving method.

Figure 2.28 (a) Polar multiaxis weaving machine developed by Yasui et al. [39], (b) architecture of fabricated woven preform.

178

(a)

(b) + Bias yarn – Bias yarn

Circumferential yarn Radial yarns

Multiaxis 3D cylindrical aramid preform

(c) Radial – Bias

Axial yarns Axial + Bias Circumferential

Circumferential yarns

Radial yarn – Bias yarn Circumferential yarn + Bias yarn

Figure 2.29 Cylindrical multiaxis 3D woven preform architecture (a), preform and weaving zone on polar loom (b), polar multiaxis 3D weaving loom, Bilisik [43].

Smart Textiles for In Situ Monitoring of Composites

Multiaxis 3D conical aramid preform

Composites and hybrid structures

179

Needle thread Layers in 2D

Bobbin thread

Figure 2.30 Stitched composite; stacked 2D laminas sewed with modified lock stitch [11].

The stitching technology is used to manufacture special-shaped composite structure such as T-joints [44], developed by Stickler 2000. The typical reinforcing yarns used for stitching are of carbon, glass, or Kevlar. Different configuration of stitch could be sewed such as lock stitch, modified lock stitch, and chain stitch, but the most used is modified lock stitch. Two other parameters of sewing process having important effects on resultant products are stitches density per unit area and stitching yarn diameters. The insertion of the stitching needle and yarn into through the thickness of stacked layers causes microstructure damages to stacked laminas. Mouritz and Cox 2000 [45] and Tong et al. [11] have summarized these damages in: (1) breakage of the in-plane fibers, crimping of the in-plane yarns at the surface into the through thickness, whereas insertion and due to stitching yarn tension, (2) misalignment of yarns around the stitch. Therefore, resin-rich region (resin pocket) associated to misalignment and crimping provokes stitch distortion in the case of heavy compaction while curing, microcracking resulting from the mismatch in the coefficients of thermal expansion of stitches and surrounding materials and compaction of the laminate, Fig. 2.31. Furthermore, bending and sliding of stitching yarns against sewing machineries while process degrades its mechanical properties. As a result of the microstructure damages induced by sewing process and stitches, Mouritz and Cox [45] revealed that it is common to reduce the in-plane strength and fatigue life of stitched laminate in comparison with equivalent unstitched laminate by up to 20%. Instead, the influence on in-planes Young’s modulus, compressive strength, and flexure is within 20% (improved or degraded). There is no or slightly noticed dependence of the variation of mechanical properties for stitched laminated on stitches density, whereas post impact mechanical properties are improved with large stitches density. The interlaminar shear strength of the laminates can be enhanced or degraded by stitches in a range of 15%e20%, according to Mouritz and Cox [45]. Preventing the stitch growing up the delamination crack induces the enhancement, whereas the initiation of the failure crack around the most distorted yarn surrounding a stitch could be the raison of the reduction in the interlaminar shear strength. However, high mode I delamination resistance is prompted to stitched laminate compared to unstitched one by bridging action of stitches. Likewise, the mode II delamination resistance of stitched laminate is promoted, but the improvement rate is less important.

180

Smart Textiles for In Situ Monitoring of Composites

Surface segment of stitch

(a)

x3 x2

(b)

Through-thickness segment of stitch

Resin pocket

x2 x1

Through-thickness segment of stitch

(c)

Laminate

x3 x1

(d)

Resin pocket between strands of stitch

Laminate x3

Through-thickness segment of stich x1

Figure 2.31 Schematics for the damages to laminates caused by stitching; (a) crimping of the in-plane yarns, (b) misalignment of in-plane yarns around stitch and formation of resin pocket, (c) distortion of stitch due to heave compaction while curing, (d) formation of resin pocket between the two strands of stitch heavily stretched. x1, x2 in-plane laminate direction, and x3 through-thickness direction [45].

Composites and hybrid structures

(b)

Line

(a)

181

Row

Figure 2.32 Schematic of (a) weft knitted fabric, (b) warp knitted fabric [2].

2.2.2

Knitted composites

Two basic types of knitting processes are warp knitting and weft knitting, Fig. 2.32. In warp knitting process multiple yarns are fed into the longitudinal direction of the machine and each yarn forms line of knit loops in the fabric direction. While in the weft knitting process, single yarn is fed into the transversal direction of the machine and it forms a row of knit loop [11]. The loop formed gives the structure with high conformability to cover complex shape surface without wrinkles or need to cut and overlap sections. Further, the knitting process enables to produce net shape or near net shape performs. From the experimental study investigating the in-plane properties of weft knit preforms, it is seen a similar performance between weft knitted composite and random mat composite, further its in-planes performance is much lower than that of conventional 2D woven composite. However, knitted composite exhibits fracture toughness greater than conventional 2D woven, unidirectional, or random mat composite [46]. Similarly, knitted composite showed higher damage tolerance and impact energy absorption.

2.2.2.1

Noncrimp fabrics

The noncrimp fabric (NCF) technology is a subgroup of knitting technology. This process consists of aligning technical yarns in specific orientation and in successive plies without inducing crimp to yarns; then the warp knitting technique is used to link these plies using sharp-head needles, as illustrated in Fig. 2.33; schematic for the principle of this technology elaborated by LIBA manufactory [13]. By the NCF machine of LIBA, the maximum number of plies, could be obtained, is 8 with orientation of the last ply in 0 degree. The warp knit, linking the fiber unidirectional plies, makes manipulating and handling the preform in the next step of manufacturing the composite materials easier

182

Smart Textiles for In Situ Monitoring of Composites

Knitting yarns

0°

1

Warp inlay yarns –0 ,–45°

2

Knitted weft yarn layers

+0

0 – 45°

90° 3

7

90° 4

0 – 45° 5

90°

6

Nonwoven material

Figure 2.33 Principle of the warp-knit noncrimp fabric technology developed by LEBA [13].

and reduces the amount of labor required to hand lay-up of laminas. The NCF reinforcement overcomes the disadvantage of the applying the 2D woven fabric lamina in terms of reduction of in-plane properties resulting from the crimp associated of woven fabric also the NCF fit better the complex shape without wrinkling than the standard woven fabric. However, insertion of the needles into through the thickness of the plies causes damage to in-plane fiber, misalignment of in-plane yarns around the through thickness part of loop knit and crimp of in-plane yarns into the through thickness of the preform. Therefore, polyester knitting yarns with low linear density are used to minimize these damages. The noncrimp composites have inferior tension, compression, and flexure properties than equivalent laminate composites made from unidirectional laminas, because of the damage incurred by warp knitting process to in-plane yarns [47]. Bibo et al. [48] observed similar fracture mechanism for noncrimp composite to equivalent laminate composite with improved resistance to interply failure and separation due to the knitting yarns.

2.2.3

Braided composites

The standard 2D braiding process consists of intertwining two sets of yarns in q degrees, in addition to the possibility to introduce axial yarns 0 degree [2], Fig. 2.34. Thereby, unlike woven fabric, the braided fabric contains q and 0 degree yarns but no 90 degrees yarns could be introduced by standard braiding process. As well, braiding process is not suitable to fabricate wide fabric as weaving process, and it is suitable to fabricate narrow width flat or tubular fabric [11]. Four-step 3D braiding, two-step 3D braiding, and multilayer interlock braiding are developed to produce 3D braided preform also to form a quite complex shape [49]. The mechanical properties of the braided composite depend largely on the amount of axial yarns, the angle of braided yarns, and their braiding pattern. The presence of axial yarns improves the axial tensile compressive and flexure properties. Moreover, decreasing the braiding angle improves the axial tensile and compressive properties, but at the sacrifice of the transverse one. The 3D braid shows a less tensile strength in both directions and less transverse tensile modules, whereas the longitudinal

Composites and hybrid structures

183

Axial yarns

0° direction –θ

+θ Braid angle

Braider yarns

Axial loading direction

Transverse loading direction

Figure 2.34 Schematic of braided fabric [2].

compressive properties and tensile modulus were better than laminate. Crimp of braid yarns reduces the performance of braided composite relative to unidirectional lamina and it contributes to degrade the fatigue performance of the 3D braided composite compared to laminate, but the presence of axial yarns improves the resistance to shattering fatigue damage. The 3D braided composite exhibits greater damage resistance and tolerance than laminate composite.

2.2.4

Z-pinned composites

In the Z-pinned composite, thin pins are inserted in the through-thickness direction of laminate using manual or automated pinning process. The Z-pinning process is presented in Fig. 2.35. The diameter of the pin is between 0.1 and 1 mm, whereas the volume content of the laminate is between 0.5% and 5%. And the pin is made of high stiffness and high strength material, such as fibrous carbon composites or titanium alloys [51]. The pins increase the delamination toughness, damage tolerance, and the throughthickness properties of the laminate [51]. However, insertion of the pins in the through-thickness of the laminate causes microstructure damages: in-plane waviness, out-of-plane crimping and breakage of fibers and resin-rich regions, and swelling of the laminate (to accommodate the pins), Fig. 2.36. That decreases the fiber volume fraction in addition to preventing the compaction of laminate while curing by the stiff pins. These microstructure damages lead to important reduction of the in-plane strength, stiffness, and fatigue life of composite in comparison with equivalent laminate.

184

Smart Textiles for In Situ Monitoring of Composites

STEP 1

STEP 2

Preform on laminate ready for ultrasonically assisted insertion

Low-density foam (upper half) of preform collapses allowing Z-pins to penetrate the lamintate

STEP 3

STEP 4

Excess Z-pin remaining in preform is sheard away

Z-pinned laminate

Figure 2.35 Schematic of the Z-pinning process of laminate composite [50].

Figure 2.36 Micrographs of laminate reinforced by pins illustrate caused microstructure damages; (a) shows waviness of in-planes yarns around pins and the formation of resin riche region, (b) shows crimp of in-plane yarns near pins [51].

Moreover, this reduction increases with the volume content increase and diameter of pines increase [50].

2.3

Outlookdcomposite structures

The low delamination resistance, poor through the thickness properties and poor toughness of laminates composite material is a result of the absence of the fiber reinforcement in through the thickness direction of composite structure. Different textile technologies were used to produce preforms with through the thickness fiber reinforcements. Among these technology, braiding and weft knitting process are applied to produce 3D complex shape composite products. The weaving and stitching processes

Composites and hybrid structures

185

are suitable to fabricate flat (planar) 3D composite structure. Although the improvement of through the thickness properties and toughness, an important degradation of the in-plane properties is observed as a result of the microstructure damages induced by textile machinery and the geometry of yarns attributed by textile technology. The woven composites show poor plane off axis properties and poor in-plane shear properties because of absence of in-plane biases yarns. Further, crimp of in-plane yarns into the through thickness of preform contributes to decreased in-plane properties. On the other side, although, stitching technology has no restriction concerning orientation of in-plane yarns, it causes important microstructure damages to in-plane yarns. The 3D multiaxis weaving technology is developed to overcome the drawbacks of standard weaving process by enabling aligning in-plane yarns in biases direction in addition to warp and filler yarns with minimization the effect of weaving machinery on fibers. However, this technology is still under development and few works have been realized to characterize the geometry and the mechanical performance of produced preform.

2.4

Reinforcing fibers

This section focuses on the industrial use and development of two largely used synthetic fibers that are glass and carbon fibers. The environmental regulation REACH (Registration, Evaluation, Authorization and restriction of Chemicals) application reduced strongly the use of aramid fibers even if they are still principally used for some military and security applications such as armors and helmets. Therefore, aramid fibers are only shortly mentioned and presented in following paragraphs. On the other side, this section ends by the presentation of natural fibers that are attracting more and more attention from automotive sports and leisure and building and construction sectors [52].

2.4.1

Glass fibers

Glass fibers present an excellent compromise between their mechanical properties and their rather low cost resulting by the market share close to 95% in composite material industry. They exhibit an excellent strength, but relatively low modulus and important density comparing to carbon fibers. This implies moderate weight gain of composite structures containing glass fibers as reinforcement. The market share of glass fibers within composite structures is rapidly growing in developing countries such as Brazil, Russia, India, China, South Africa (BRICS), Eastern European countries and Turkey. This may be explained by rapid construction of a large amount of new infrastructures under construction and where the labor cost is still comparatively low. The most popular fields of application of glass fibers are: construction and transportation followed by sports goods and electronic devices. The overall production of composites containing glass fibers is stagnating in France and Spain, while it is increasing in the United Kingdom and Germany particularly in the areas of automotive and construction industry.

186

Smart Textiles for In Situ Monitoring of Composites

Table 2.1 Differences between glass and carbon fibers Properties

E glass fibres

S glass fibres

Carbon fibres

Modulus (GPa)

72e78

88e91

230e600

Strength (MPa)

2600e3800

4380e4590

3500e6000

Elongation (%)

4.5e4.9

5.4e5.8

1.5e2

Diameter (microns)

6e21

6e21

5e15

Density (g/cm )

2,54e2,55

2,48e2,49

1,79

Cost (US$/kg)

1,1

1,5

20e30

3

Table 2.1 shows the differences between glass and carbon fibers in terms of mechanical and morphological properties. The total world glass fibers based composites in 2016 was approximately 2.2 million tons, comprising 1.2 million tons of discontinued fiber reinforced plastics mostly due to their low cost and the possibility to easily use conventional polymer processing methods such as injection molding and extrusion. Fig. 2.37 summarizes the glass fibersebased composite structures in Europe from 2000 to 2016. Table 2.2 gives the GRP production volumes in Europe according to processes/ components.

2.4.1.1

Sheet molding compound/bulk molding compound

Large-scale series manufacturing processes for composites are currently a hot topic in the media. Yet these have been a reality for SMC (sheet molding compound) and BMC 1400 1200 1000

986 984 992

1065 1009 1041

1131

1195 1058

1015 1049 1010 1020 1043

1069 1096

815

800 600 400 200

02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 *

01

20

20

20

00

0

Figure 2.37 Glass reinforced composites (GRC) production in Europe since 2000 (in thousands of tons). From E.D. Witten, T. Kraus, M. Kuhnel, et al., Composite Market Report 2017, November 28, 2016. (En ligne). Available: http://www.eucia.eu/userfiles/files/20161128_market_report_ 2016_english.pdf.

Composites and hybrid structures

187

Table 2.2 GRP production volumes in Europe according to processes/components [53]

a

2013

2014

2015

2016a

Kt

Kt

Kt

Kt

SMC

184

190

191

198

BMC

71

74

74

76

S SMC/BMC

255

264

265

274

Hand lay-up

142

138

139

140

Spray-up

90

94

96

97

S Open mold

232

232

235

237

RTM

126

132

137

141

Sheets

84

84

86

89

Pultrusion

47

48

49

50

S Continuous processing

131

132

135

139

Filament winding

78

79

80

80

Centrifugal casting

66

66

68

68

S Pipes and Tanks

144

145

148

148

GMT/LFT

114

121

132

140

Others

18

17

17

17

Sum:

1020

1043

1069

1096

estimation.

(bulk molding compound) components for many yearsdwith production of some components even exceeding 100,000 pieces per year [53]. SMC and BMC semifinished products are turned into components using pressing (SMC) and injection molding (BMC) processes. They are primarily used in the electro/electronic sectors and the automotive industry. Although the SMC/BMC segment had the weakest GRP production growth last year, in 2016 it has outperformed all other thermosetting materials with growth of over 3%. Over one-quarter of all GRP produced in Europe is manufactured from SMC or BMCda total of 274,000 tons. The sector is dominated by SMC production (198,000 tons), which not only accounts for the lion’s share of volume but is also growing somewhat faster than the smaller BMC sector compared to last year. Mass-produced car headlamp reflectors are typical components made from BMC. SMC is used in the automotive sector, e.g., for tailgates, interior paneling and cabin components, oil sumps, or covers. In the construction industry it is used, e.g., for light shafts, cable ducts, and shaft covers, in the electro/electronic sector for switches, control cabinets, and home junction boxes [53].

188

2.4.1.2

Smart Textiles for In Situ Monitoring of Composites

Open mold/open processes

“Open processes”dhand lay-up and spray-updare a segment that places greater emphasis on manual skills and craftsmanship. It is the second largest segment in the European GRP market with total production of 237,000 tons. The segment’s trend of comparatively weak growth over recent years continued and is under 1% in 2016. The business is characterized by a large number of small companies with few employees and often individual orders. It has a relatively low level of automation [53].

2.4.1.3

Resin transfer molding

RTM (resin transfer molding) components have continued their trend of slightly stronger than average growth (nearly 3%), which was also observed last year. European production in the segment totals 141,000 tons [53].

2.4.1.4

Continuous processing

The consistently strong trend of recent years continues in the continuous processing segment. In 2016, European production has risen by 3% to 139,000 tons. Panels have primarily been used in vehicles for many years, e.g., truck side panels, caravan superstructures, or the conversion of commercial vehicles. These are supplemented by applications in the facade industry. However, innovations are also an important driver in the segment. These include antiseptic laminates for paneling in operating theaters and sports equipment such as skis, wakeboards, or longboards [53].

2.4.1.5

Glass mat thermoplastic/long fibers thermoplastic

GMT (glass mat thermoplastic)/LFT (long fibers thermoplastic) composites production is steadily increasing with main applications for automotive parts where lightweight is required. GRP production by country from 2013 to 2016 is presented in Table 2.3. To better expand the market for glass fibers composites two technical challenges will have to be addressed: 1. The utilization of thermoplastic matrices Thermoplastic matrices have a numerous advantages comparing to thermoset ones that are: facility of storage, higher impact resistance, and a possibility to recycle. In the automotive sector SMC/BMC compound parts are being replaced by GMT/LFT, whose market is gradually increasing at the rate of 6% a year, due to their recyclability and mechanical strength properties. 2. New process development and automation Even if the raw material (glass fibers) and matrices are quite cheap, the manufacturing processes are still very expensive. This is mainly due to long process cycling times and expensive semi-produces such as prepregs. Also, the manufacturing processes are laborintensive and more automation is necessary to decrease the production cost of GRP composites.

Composites and hybrid structures

189

Table 2.3 GRP production by country from 2013 to 2016 in Europe and Turkey [53] 2013 Kt

2014 Kt

2015 Kt

2016a Kt

UK/Ireland

140

146

150

152

Belgium/Netherlands/Luxembourg

42

43

44

45

Finland/Norway/Sweden/Denmark

44

42

39

40

Spain/Portugal

152

154

156

158

Italy

146

148

150

154

France

112

108

108

110

Germany

192

200

212

220

17

18

18

18

175

184

192

199

1020

1043

1043

1096

214

245

245

257

Austria/Switzerland Eastern Europe

b

Sum: b

Turkey a

estimated/predicted data at that period. estimations for Eastern Europe and Turkey.

b

2.4.2

Carbon fibers

In spite of their excellent mechanical properties and light weight (low density), carbon fibers are mostly dedicated to high-end products in aero space, military, and some medical and sports applications. This may be explained by their relatively high cost. However, with novel environmental strict regulations, the principal solution to reduce gas emission in the field of transportation is to decrease the weight of cars, trucks, planes, and even trains. The challenges are to replace metallic parts with composite ones and to realize hybrid metallicecomposite parts. However, there are lot of issues related to bonding compositeemetal, life time of composite parts, etc. Therefore, the structural health monitoring (SHM) in real time in situ seems as a very promising solution to those problems. Globally, the demand for carbon fiber has shown steady growth since the general economic recession of 2009 (see Fig. 2.38). The initial strong annual growth rates of over 20% seen after 2009 have reduced over the following years to a normal growth rate of 6.9% for 2013. From 2013 to 2014 it is possible to notice again a marked increase in the annual growth rate of 14%. With 9.4% for 2015 we see growth rates leveling out to a stable level. It can be assumed that further annual growth rates (AGR) of double figures will be seen, which should swing between 10% and 13%, so that we could break through the 100,000 tons of carbon fibers demand mark already in 2020 [53]. Behind these impressive figures (Fig. 2.38) is the wide utilization of carbon fibers in the transportation sectors. For instance, electric vehicles such as BMW I3 and I8, Lamborghinis, Bentleys, Tesla Roadster, etc. have adopted carbon

190

Smart Textiles for In Situ Monitoring of Composites

140 120.0 120.0

120 100.5 100.5

100 80 60 40

33.0 33.0

43.5 38.5 43.5 38.5

53.0 53.0

46.5 46.5

72.0 72.0

64.0 64.0

58.0 58.0

20 0 2010 2011 2012 2013 2014 2015 2016* 2017*

2020*

2022*

Figure 2.38 Global demand for carbon fibers in thousands of tons from 2010 to 2022 (projection) [53].

reinforced composites. Also many car makers are in close relations with carbon fibers reinforced composites (BMW-SGL, Ford-DOW AKAS, etc.) to reduce the production cost and to guarantee their supplies. Boeing 787 Dreamliner reached 50% share in the weight of carbon composite parts and Airbus A350 53%. In Fig. 2.39, it is possible to appreciate the annual theoretical production of carbon fibers in thousands of tons for major CF producers. The United States and the Western Europe are the leading carbon fibers reinforced composite markets followed by Japan. In the United States the main markets are related to aeronautics and defense industries. In Western European countries there are three main markets that are aeronautics and defense, the automotive, and the 45

1.9

40 35

15.5 Yearly CF capacity 2014

Zoltek

Growth 2015/2016

30 25 20 15

27.1

0.9

3.0 1.0

13.6

3.0

Fi

e tig

su

ns So

at

ng

.

2.0

yo

en Sh u-

gf on Zh

M

yi

ec yt (c

ay

3.0

sa

ng

)

l ce ex H lv

.p om Fr

So

la st

M

.c

R

or .

C

o To h

SG L

k lte Zo + ay To r

3.6

H

4.0

0.5

b.

4.0

0

Ak

7.3

n

8.8

he

10.1

gs

11.5

en

12.0

5

H

10

Figure 2.39 Theoretical, annual carbon fibers production in thousands of tons according to manufacturers (September 2016) [53].

Composites and hybrid structures

191

wind mills industries. Japan is the main carbon fibers producer (raw materials) and third carbon fiberebased composites market. On the other side, China is a very fast growing market due to the development of the wind energy and aerospace and military applications. Until now, the carbon fibers composites market has been driven by civil aerospace industry, whereas the military and defense market is rather limited. In the future, automotive and railway rolling stocks markets seem to be promising and able to increase volumes of carbon fibers composites. The main problem is the cost, that will have to be under 10 US$ per kilogram of carbon composites to be able to replace conventional metallic parts. Carbon fibers producers, mainly in Japan and Germany make important efforts to increase the production and reduce the price of carbon composites under 10 US$ per kilogram. Currently, their price for the automotive industry is approximately 12e14 US$ per kg. Possible solutions supposed to be able to increase the production and decrease the price could be: 1. Development of the supply chain for low-cost carbon fibers good enough for automotive applications having 1700-MPA strength and 1.7-GPa modulus; 2. Organization of strategic alliances and joint ventures among car manufacturers and carbon fibers producers; 3. Set up novel low cost precursors such as lignin-based precursor developed by Zoltek and ORLN [54].

2.4.3

Aramid fibers

The Fiber structure of aramid fibers is composed of a series of synthetic polymers in which repeating units containing large phenyl rings are linked together by amide groups. Amide groups (CO-NH) form strong bonds that are resistant to solvents and heat. Phenyl rings (or aromatic rings) are bulky six-sided groups of carbon and hydrogen atoms that prevent polymer chains from rotating and twisting around their chemical bonds [55]. Aramid fibers are characterized by medium to ultra-high strength, medium to low elongation, and moderately high to ultra-high modulus with the densities ranging from 1.38 to 1.47 g/cm3. Heat-resistant and flame-resistant aramid fibers contain high-proportion or meta-oriented phenylene rings. Ultra-high strength high-modulus fibers contain mainly para-oriented phenylene rings. All aramids contain amide links that are hydrophilic. However, not all aramid products absorb moisture in the same way. The PPD-T (poly-phenylene terephthalamide) fiber has very good resistance to many organic solvents and salt, but strong acids can cause substantial loss of strength. Aramid fibers are difficult to dye due to their high Tg (glass transition temperature). Also, the aromatic nature of para-aramid is responsible for oxidative reactions when exposed to UV light, which leads to a change in color and loss of some strength. Aramid fibers do not melt in the conventional sense but decompose simultaneously. They burn only with difficulty because of Limited Oxygen Index values. It should be mentioned that at 300 C some aramid types can still retain about 50% of their strength. Aramids show high crystallinity, which results in negligible shrinkage at high temperature.

192

Smart Textiles for In Situ Monitoring of Composites

Aramid yarn has a breaking tenacity of 3045 MPa, in other words more than five times than this of steel (under water, aramid is four times stronger) and twice than this of glass fiber or nylon. High strength is a result of its aromatic and amide group and high crystallinity. Aramid retains strength and modulus at temperatures as high as 300 C. It behaves elastically under tension. When it comes to severe bending, it shows nonlinear plastic deformation. With tension fatigue, no failure is observed even at impressively high loads and cycle times. Creep strain for aramid is only 0.3%. Aramid fiber applications are separated into two categories: (1) Reinforcement in composites such as sport goods, aircraft, military vehicles, and many others; (2) Fabrics in clothing such as fire protection clothes or bullet proof vests. More sophisticated usages of aramid are: Protective gloves, helmets, body armor; Filament wound pressure vessels; Flame and cut resistant clothing; Asbestos replacement; Ropes and cables; Jet engine enclosures; Tennis strings and hockey sticks; Wind instrument reeds; Reinforcement for tires and rubber goods and Circuit board reinforcement. Two main kinds of aramid fibers are produced globally: Meta-aramid and Para-aramid fibers.

2.4.3.1

Metaaramid fiber

In 2013, global metaaramid fiber capacity totaled 42,000 tons, mostly distributed in the United States, Europe, and Asia, 27.4% of which was swept by China. Confronted with fierce competition in domestic market, some metaaramid fiber products in China are exported overseas. In 2014, the output of metaaramid fiber in China will hit 11,000 tons or so and net export volume of the product was more than 800 tons, as it is predicted. Metaaramid fiber in China is mainly applied in such fields as high temperature resistant filtration materials, safety protection, insulation paper, etc., among which high temperature resistant filtration materials accounted for over 50% of the total demand for metaaramid fiber.

2.4.3.2

Para-aramid fiber

In 2013, around 80,000 tons of para-aramid fiber was manufactured worldwide, involving major producers DuPont and Teijin Limited. Para-aramid fiber capacity of the aforementioned two giants approximated 80% of the world’s total para-aramid fiber capacity. Para-aramid fiber industry in China developed late and progressively achieved industrialization in 2010. During 2010e14, with multiple para-aramid fiber projects being put into production, para-aramid fiber output in China witnessed soaring growth and hit about 1.3 ktons in 2014 as it is predicted, lifting the country’s self-sufficiency rate of para-aramid fiber market to over 50%. The aramid fiber market is estimated to have been USD 2.99 billion in 2015 and is projected to reach USD 5.07 billion by 2021 [56], registering a CAGR (Compound Annual Growth Rate) of 8.9% between 2016 and 2021. In this study, 2014 has been considered as the historical year, 2015 as the base year, and 2016e21 as forecast period for estimating market size of aramid fibers. The global aramid fiber market

Composites and hybrid structures

193

has been segmented based on type, application, and region. Para-aramid is expected to play a key role in feeling the growth of the overall aramid fiber market owing to its unique properties, which makes it suitable for use in a wide range of applications. The demand for aramid fiber in security and protection application is increasing rapidly, thereby driving the aramid fiber market.

2.4.4

Natural fibers

Bio composites, having bio-based fibers reinforcements and/or bio-based matrix may be seen as an effective solution for achieving carbon neutral economy. The market of biocomposites is growing fast, particularly in Europe. They can be classified in two main categories: Wood Plastic Composites (WPC) and Natural Fiber Composites (NFC). The North America with the United States and Canada is the main market for WPCs with principal applications in construction industry with decking, siding, and fencing. On the contrary NFCs are dominant in European market for automotive applications. According to various sources, the future of the NFCs market looks attractive with opportunities in the automotive and building & construction industries. The global NFCs market is forecasted to grow at a CAGR of 8.2% from 2015 to 2020 [57]. The major driver for the growth of this market is the rise in demand for lightweight and environmentally sustainable composite materials in various applications, such as automotive, building and construction, and others. Within the NFCs market, the automotive segment is expected to remain the largest application by both value and volume. Increasing concern for passenger safety, government mandates for better fuel economy, and end-of-life vehicles directive are the major driving forces that spur growth for this segment over the forecast period. Europe is expected to remain the largest market during the forecast period due to higher acceptance level of environmentally sustainable composite materials by automotive OEMs, government directives, and growth in end-use industries. The market size of NFCs is projected to reach USD 6.50 billion by 2021, at a CAGR of 11.68%, between 2016 and 2021. It is also important to outline that the current demand for bio composites in automotive industry is not driven by a technical demand, but by the eco-marketing and governmental regulations. According to NOVA institute [58], the market of NFC is expecting to reach 350,000 tons with strong governmental incentives whereas it would be only 120,000 tons without those incentives. Among natural fibers, flax, hemp, and kenaf natural fibers are considered as potential composites reinforcement materials instead of glass fibers by dint of their good mechanical properties such as specific modulus and strength. Many scientific efforts, within large-scale research projects funded by national governments in Europe or by European Commission within Framework Programs, dedicated to find solutions able to improve inherent variability of natural fibers quality and their poor compatibility with matrix polymers, have been undertaken. However, the poor lifetime with the hygrothermal aging and rather low thermal stability are still important issues that will have to be addressed in the future concerning the utilization of natural fibers in composite structures.

194

2.5

Smart Textiles for In Situ Monitoring of Composites

Matrices

Composites may be classified in function of their matrix to three distinct classes: (1) Metal Matrix Composites (MMC), (2) Ceramic Matrix Composites (CMC), and (3) Plastic Matrix Composites (PMC). Relatively large diffusion of PMCs comparing to MMC and CMC can be explained with an excellent ratio of cost/performances, they exhibit relatively low cost of raw material, simple manufacturing methods, and excellent performances/cost ratio. The present book focuses on PMCs and their monitoring in real time in situ, using smart textile methodology applied to composites’ reinforcements. When heated, polymer matrices undergo melting and softening enabling their mixing with reinforcement structure and forming the final shape of composite part. Two different kinds of matrices, thermoset, and thermoplastics are presented in following sections. Moreover, thermoset undergoes irreversible chemical reactions (cross-linking, curing) during the thermo consolidation process.

2.5.1 2.5.1.1

Thermosetting matrices Unsaturated polyester resins

Those resins are largely used in numerous mass production products. They can be cured at room temperature and may be used in a broad range of manufacturing techniques such as open mould spray up, compression molding, RTM, and casting. They also have lower mechanical properties and larger curing volumetric shrinkage compared to phenolic and epoxy resins, but their utilization is simple and they are rather cheap [59].

2.5.1.2

Phenolic resins

Phenolic resins are high cross-linked aromatic structures exhibiting rigidity, strength, chemical resistance, and good electrical properties. Their solidification process can be conducted under high pressures; they undergo relatively large curing volumetric shrinkage. A drawback of phenolic resins is their brittleness. The good fire resistance led phenolic resins to applications in the field of aeronautics (interior), off shore gas and oil platforms, and mass transit and electronic devices [59].

2.5.1.3

Epoxy resins

Epoxy resins exhibit excellent mechanical behavior, good chemical resistance, and optimal dielectric properties. Their performances are good up to relatively high temperature of 121 C. They are also easy to cure and have low shrinkage. Their adhesion with the reinforcing fibers is good and they exhibit excellent dimensional stability and durability [59].

Composites and hybrid structures

2.5.2 2.5.2.1

195

Thermoplastic matrices Polyolefin

Polyolefin composites (polypropylene and polyethylene) are very popular and widely used in various applications such as automotive, construction, and consumer products. This may be explained by their excellent compromise in cost performances. They are often used in combination with glass fibers as reinforcements [59].

2.5.2.2

Polyketone resins

Polyketone resins exhibit outstanding mechanical properties including radiation and flame resistance, toughness, and strength. The best polyketone resin from the commercial point of view is Polyether ether ketone well known under the name PEEK [59].

2.5.2.3

Polyether imide

Polyether imide has excellent mechanical properties and optimal heat resistance. It overwhelms the PEEK performances at high temperatures, it has very high glass transition temperature, and also high utilization and low molding temperature.

2.5.2.4

Polyarylene sulfide resins

PolyPhenylene Sulfide exhibit excellent mechanical properties and very good stability. The macromolecules have also very low flammability. Those resins are often reinforced with short fibers rather than with continuous ones.

2.5.2.5

Bio-based resins

Bio-based polymers also called bio-based resins are obtained from renewable resources (algae, bacteria, microorganisms, plants, etc.). They can be synthetized either directly or through the monomers synthesis that have to be followed by the polymerization. There are many different market available bio-based polymers such as polylactic acid, the poly L lactide, polyhydroxybuturate, polyhydroxyalkalonates (PHAs), polyamide, polypropylene (PP) obtained from bio-based ethylene obtained by converting ethanol, polyethylene terephthalate and all other thermoplastic materials. There are also other bio-based resins partially obtained (polyurethanes with bio-based dyols and bio based epoxy epichlorohydrine from glycerol) that are thermosetting [60e62]. Bio-based polymers are mostly used in packaging applications; their use in more demanding applications is still a great challenge. This can be explained by following evidences: • • • •

Poor durability of bio-based matrices in long life applications such as transportation and construction; Important modifications of existing processes are necessary; Higher cost of bio-based polymers comparing to oil-based one; and Significantly lower performances of bio-based comparing to oil-based matrices.

196

Smart Textiles for In Situ Monitoring of Composites

2.6

Failure mechanisms in composites

Fundamental definitions related to failure mechanisms in composite structures are presented in this section to better define the targets of SHM using smart textile configurations such as fibrous strain gauges.

2.6.1

Damage

Damage can be defined as changes introduced into a system that adversely affects its current or future performance. Implicit in this definition is the concept that damage is not meaningful without a comparison between two different states of the system, one of which is assumed to represent the initial, and often undamaged, state. In structural and mechanical systems, the definition of damage will be limited to changes to the material and/or geometric properties of these systems, including changes to the boundary conditions and system connectivity, which adversely affect the current or future performance of these systems [63].

2.6.2

Defect/flaw

In terms of length scales, all damage begins at the material level. Although not necessarily a universally accepted terminology, such damage is referred to as a defect or flaw and is present to some degree in all materials. Under appropriate loading scenarios, the defects or flaws grow and coalesce at various rates to cause component and then system-level damage [63].

2.6.3

Failure

As the damage grows, it will reach a point where it affects the system operation to a point that is no longer acceptable to the user. This point is referred to as failure. In terms of time scales, damage can accumulate incrementally over long periods of time such as that associated with fatigue or corrosion damage accumulation. On relatively shorter time scales, damage can also result from scheduled discrete events such as aircraft landings and from unscheduled discrete events such as enemy fire on a military vehicle or natural phenomena hazards such as earthquakes [63].

2.6.4

Performance

Health monitoring is a concept that requires a comprehensive and quantitative description of performance covering a large spectrum of limit states that may govern throughout the life cycle of a composite material. There are four principle limit states defined as: • • • •

Utility and functionality; Serviceability and Durability; Safety and stability at failure; Safety at conditional limit states.

Composites and hybrid structures

197

Each limit state incorporates a number of limit events that should be considered in design and evaluation. The proper subjective as well as objective and quantitative indices for limit events determine the attainment of a limit state. Most limit events are governed by defects in design, materials, or fabrication, or by various loading events associated with some probabilistic model. The critical remaining issue related to the definition of performance is the formulation of objective, quantitative criteria for each limit event, and then find appropriate tests, measurements, and simulation methods for assurance that the desired performance limits will not be exceeded in the limit events expected during the life cycle of the composite part. The concept of health monitoring promises to provide the data and information for various types of composites to serve for formulating objective performance indices. The importance of unambiguous and quantitative indices for ensuring performance is quite evident; performance limit states and the corresponding limit events serve as critical foundations for performance-based design and evaluation [64].

2.6.5

Health

Engineers use various indices for defining health depending on purpose. Although it is pragmatic to use deterministic indices that are mainly related to “structural safety,” most engineers recognize the need for a broader definition that relates to the definition of performance discussed earlier. After all, the main purpose for evaluation of health is the prognosis of future performance. It is thus desirable to define the health of a composite part as its system reliability to possess adequate capacity against any probable demands that may be imposed on it in conjunction with the limit states defined previously. It is important to emphasize that system reliability should cover the entire spectrum of limit states and limit events and not just “structural safety” [64].

2.6.6

Health monitoring

Health monitoring is the tracking of any aspect of a structure’s health by reliably measured data and analytical simulations in conjunction with heuristic experience so that the current and expected future performance of the composite part for at least the most critical limit events, can be described in a proactive manner. The single most important distinction of health monitoring from a typical in-depth composite part evaluation and testing practice is the minimum standards that are required for analytical modeling for reliable computer simulations, and how the measurements, loads, and tests are designed and implemented in conjunction with the analytical simulations [64,65].

2.6.7

Structural identification

Structural identification is defined as the development of an analytical conceptualization leading to an analytical model of a composite structure that is quantified, tested,

198

Smart Textiles for In Situ Monitoring of Composites

and validated by correlating model simulations with corresponding measurements from the structure. It serves as both the starting point and the core of health monitoring. Applications of the structural identification principle provide the most reliable manner of characterizing a structure for analysis and decision-making as it goes through its life cycle. The structural identification principle is what should guide engineers in determining the minimum number of measurements needed to accurately and completely characterize a structure so that its health may be reliably evaluated at the serviceability and safety limit states [64,65]. Structural identification provides the most objective and reliable manner of documenting the global mechanical properties of a structural system and assessing its physical condition, load-rating, vulnerability, maintenance and retrofit needs, etc. The “characterization” is also used by the engineers to describe the concept. It offers an understanding of all the critical mechanisms of flexibility, energy dissipation, and inertia; the three-dimensional response kinematics; the resistance mechanisms; the critical structure regions; the mechanisms of resistance at the critical regions, the localized force, deformation and accumulated damage states at the critical regions. This information is essential to reliably assess and identify the available capacities of stiffness, strength stability, deformability, hardening, and energy dissipation in the structure. The following steps are essential for structural identification of a structure: Collecting information, conceptualizing and a-priori modeling to best represent the existing knowledge about the structure. 1. Experiment design based on analytical and preliminary experimental studies. 2. Full scale modal and controlled load tests are conducted to identify, verify, and evaluate global and local behavior. These tests are conducted in conjunction with in-depth visual inspections of the composite structure, local Nondestructive Evaluation (NDE) techniques, and in many cases material testing to identify and assess the extent of any existing deterioration and damage. 3. Processing and conditioning of the resulting experimental data to mitigate any errors and assure its quality. 4. Progressive calibration of analytical models by adjusting mechanical properties and boundary and continuity conditions to reflect the physical parameters measured from the full-scale states. 5. Utilization of the calibrated model as a basis for management decisions.

2.6.8

Structural health monitoring

The process of implementing a damage detection and characterization strategy for engineering structures is referred to as SHM. Here, damage is defined as changes to the material and/or geometric properties of a structural system, including changes to the boundary conditions and system connectivity, which adversely affect the system’s performance. The SHM process involves the observation of a system over time using periodically sampled dynamic response measurements from an array of sensors, the extraction of damage-sensitive features from these measurements, and the statistical analysis of these features to determine the

Composites and hybrid structures

199

current state of system health. For long-term SHM, the output of this process is periodically updated information regarding the ability of the structure to perform its intended function in light of the inevitable aging and degradation resulting from operational environments. After extreme events, such as earthquakes or blast loading, SHM is used for rapid condition screening and aims to provide, in near real time, reliable information regarding the integrity of the structure [63e65]. The classical methods for periodical maintenance use many NDE techniques that require extensive human involvement and expensive procedure. Moreover, this kind of periodical inspection cannot give any information concerning accidents and failures occurring between two successive overhauls. To overcome such shortcomings, it is now possible to use “sensitive” material. A material or a structure is said to be “sensitive” when it includes sensors providing realtime information about the material itself, or its environment. Continuous health monitoring of materials would result in improved durability and safety of structures. The basic idea is to build a system similar to the human nervous system, with a network of sensors placed in critical areas where structural integrity must be maintained. Mathematical algorithms based on a “neural network” then can be trained to recognize patterns of electrical signals that represent damage, such as fiber strains or breakage and matrix cracks. In broadest terms, therefore, SHM comprises a distributed network of sensors that are deployed as integral parts of a structure and capable of collecting and sending information to an interrogator or electronic monitoring device. Damage identification is carried out in conjunction with five closely related disciplines that include [63e65]: • • • • •

Structural Health Monitoring; Condition Monitoring (CM); Nondestructive Evaluation; Statistical Process Control (SPC); and Damage Prognosis (DP).

Typically, SHM is associated with onlineeglobal damage identification in structural systems such as aircrafts and buildings. CM is analogous to SHM, but addresses damage identification in rotating and reciprocating machinery, such as those used in manufacturing and power generation. NDE is usually carried out offline in a local manner after the damage has been located. There are exceptions to this rule, as NDE is also used as a monitoring tool for in situ structures such as pressure vessels and rails. NDE is therefore primarily used for damage characterization and as a severity check when there is
a priori knowledge of the damage location. SPC is process-based rather than structure-based and uses a variety of sensors to monitor changes in a process, one cause of which can be structural damage. Once damage has been detected, DP is used to predict the remaining useful life of a system.

200

2.7

Smart Textiles for In Situ Monitoring of Composites

Hybrid structures, production methodology and principles, state of the art

Assembling composite material with metallic part is still a challenge. Differences of material properties, due to anisotropic behavior of the fibrous reinforcement within the composite material and the isotropic behavior of the metallic part, generate difficulties related to joining. Various techniques are available. In recent overview, Nassar and Yang [66] have summarized the latest advances in fastening and joining of composites. The right choice of joining technique is essential to provide a safe load-transfer. Hybrid high-strength interfaces between metal and composite parts are a major issue for joining technologies. In the research study of Ueda et al. [67], an instantaneous and low cost process of self-piercing rivet to mechanically fasten carbon fiber reinforced plastic (CFRP) laminates seems to be an efficient solution, slightly higher in tensile strength resistance compared to bolted joint. In the research work of Virupaksha and Nassar [68], experimental characterizations of thick composite bolted joints have been done to check the effect of washer size and bolt preload on bearing properties. Joining metalecomposite parts by bolting imposes to drill the composite material, which tends to promote delamination inside the composite laminates, as reported in the recent literature review done by Liu et al. [69]. Research works have also been done to reveal the effect of geometry and laminate properties on the bolt-bearing behavior. Oh et al. [70] have worked on bolted joints for hybrid composites made of glasseepoxy and carboneepoxy combinations under tensile loading. They have investigated the influence of different design parameters as: laminate ply angle, stacking sequence, the ratio of glass-epoxy to carbon epoxy, the outer diameter of the washer and clamping pressure. Their results have shown that the peak load occurred before the maximum failure load due to the delamination of the laminate under the washer. In the research study of Gebhardt and Fleischer [71], metallic inserts in CFRP parts with a bead pattern have revealed a significant increase of the bending strength. Tested surface treatments and coatings also help to raise the strength of the co-cured bonding between the inserts and the laminate. Recently, Dufour et al. [72] have experimented metallic inserts inside 3D fabric architecture made with thermoplastic yarns, which tend to absorb more energy during dynamic pull-out tests in the axial direction. Ageorges et al. [73] published a complete review with the pros and cons of the fusion bonding techniques for joining thermoplastic matrix composites. They highlighted a huge potential for volume intensive applications in which short processing cycles are necessary. In the research study of Esteves et al. [74], the process parameters of friction spot joining have been statistically studied to better understand the mechanical performance of metalecomposite overlap joints. Hybrid technology of metalecomposite joining proposed by Ucsnik et al. [75], based on interaction between fibrous reinforcement of composite material and small spikes welded onto metal surface, has led to better performance on strength, local strains, and energy absorption capacity, particularly adapted in crash applications.

Composites and hybrid structures

201

In the research study of Bogdanovich et al. [76], several tests on a variety of single lap and double butt-strap coinfused and adhesively bonded joints of 3D woven NCF composites have revealed the lack of delamination within any unitary 3D woven composite adherent or strap of the studied joints. In the study of Huang et al. [77,78], adhesiveeembossing hybrid joining process has been applied to metallic part and three types of glassefiber reinforced plastic thin sheets. They conclude that the optimization of the internal structures of composites and development of forming techniques to improve the formability of the adherent composite would increase the potential of the adhesiveeembossing hybrid joining process for realizing more ultra-lightweight thermosetting FRPemetallic hybrid components. Gude et al. [79] have proposed another solution of joining metal and composite parts by the adaptation of a thermo-clinching process adapted for thermoplastic composites allowing the production of hybrid structures with continuous fiber reinforced thermoplastics and metallic components. This new thermo-clinching joining method provides joining element free and form-locked joints with thermoplastic composites and metallic joining parts by means of plastic deformation. In the research study of Cosson et al. [80], an innovative numerical analysis helps to evaluate changes in local energy diffusion directly linked with local fiber arrangements using laser welding technology for continuous thermoplastics composites structures. In the research paper of Rodríguez-Vidal et al. [81], two-step technique of laser welding to join metallic and composite part is presented, highlighting the influence of the appropriate microstructure of the metallic part leading to high breaking forces for glass reinforced polyamideesteel joints. In the technical report of Cognard [82], the joining by metallic inserts introduced inside the fibrous structure of the composite material before matrix fixation is considered as a compromise between gluing and mechanical assembling, without altering the global strength. In another technical report of Cognard [83], the Mosaic project, in 1996 leaded by Renault company, mentioned the use of glue technique to assembly metalecomposite due to the heterogeneity of these different materials. Faurecia and the Institute of Civil Engineering and Mechanics at Ecole Centrale de Nantes proposed a new technology for composite/metal assembling based on Magnetic Pulse Welding (MPW). This technology is very fast, versatile. It is similar to electric spot welding [84,85]. A global overview of the different main types of joining technologies can be summarized by highlighting respectively their main advantages and disadvantages (Table 2.4).

2.8

Hybrid structuresdbonding issuesdinnovative joining techniques

A number of innovative joining techniques, between composite and metallic parts, exist currently based on metallic inserts placed inside composite structures. They are presented in following paragraphs.

202

Smart Textiles for In Situ Monitoring of Composites

Table 2.4 Overview of advantages and disadvantages of joining technologies Technology

Advantages

Disadvantages

Mechanical Fastening

• Joining Dissimilar materials possibilities • Well-known prediction methods and analysis • Easy repair and replacement • Ease of disassembling when required

• Stress concentrations caused by holes and nonuniform distribution of local peak bearing stress • Delamination and exposure of the fibers due to drilling and machining in the composites • Changes in clamping load due to temperature changes • Galvanic corrosion due to reaction between metallic fasteners and composite fibers • Additional Steps in production cycle • Additional mass due to fasteners

Clinching Techniques

• Economic efficiency due to saved costs of additional parts • Joining of more than two parts in one step is possible

• Great delamination in the composite sheet

Adhesive Bonding

• Possibility of joining any combination of similar and dissimilar materials • Minimizes or prevents galvanic corrosion in dissimilarities cases • Reduce weight of the assembly

• Good mechanical performance is limited to shear resistance • Extreme surface preparations • Long cure times for some adhesives • Hazardous chemicals and solvents handling present a deceleration for the production cycle time and increase in costs • Uncertainty in bond’s failure prediction • Temperature sensitivity • Volatile organic compounds (VOC) emission

Composites and hybrid structures

203

Table 2.4 Continued Technology

Advantages

Disadvantages

Hybrid combined joints (combination between mechanical fastening and adhesives techniques)

• High static strength • Better resistance to fatigue and peeling • Better resistance to corrosion • Two stage cracking process that makes easier and safer to detect before the joint breaks

• Fiber damage and delamination • Increased manufacturing time due to curing time of the adhesive and the mechanical fastening process additional time • Thermal damage that can be caused to the composite when curing requires elevated temperature

Thermal-Based Techniques (Ultrasonic spot welding, laser direct joining, friction spot welding, friction lap welding, etc.)

• Strong joints can be achieved • Short joining cycles • Absence of emissions

• Great thermal damage caused to the composite • Applicable only for thermoplastic composites matrices

2.8.1

Continuous laser welding

Laser welding consists in a high energy, low contact, low heat input, and high performance bonding process, which can be used for deep penetration, high impact ratio, and low distortion welding. Because of recent advances in the laser sources technology, based on solid state components, the operating costs have been reduced and the global share of laser welding equipment is steadily increasing 6%e8% per year. It reached more than $2 billion in 2016. The metal inserts in composites could be bonded to metallic structures such as sheets by laser welding without affecting the integrity of metal inserts in composite joint as well as laminates of the composite itself.

2.8.2

Friction welding

Friction Stir Welding (FSW) of steel structures has been in development for over decades, but only in recent years the strength and high temperature wear resistance of tool materials have reached optimal values. This technique could also be used to assemble metallic and composite parts containing metallic inserts.

2.8.3

Magnetic pulse welding

The setup system for MPW and Magnetic Pulse Crimping (MPC) consists of the pulsed power generator, the tool (a coil also called inductor including a field shaper

204

Smart Textiles for In Situ Monitoring of Composites

if applicable), the joining partner that is deformed during the process (also called a flyer), and the static joining partner. Joining tubular structures has been the most common application of this technology; however, MPW and MPC of sheets receive more and more attention currently. The impact velocity and the progressive collision contributes to the forming of the jet that which eliminates the oxides guaranteeing two clean surfaces and generating a very short, but intensive local heat at the interface creating a metallurgical bonding having in general a wave pattern. To guarantee a required standoff distance, MPW of metal sheets uses in general a specific geometrical configuration. To overcome overlapping problems for potential robotized industrial applications, Magnetic Pulse Spot Welding (MPSW) has been introduced where a prior local stamping in the flyer plate is introduced creating a hump in the intended welding region. Therefore, the gap is not needed between two plates because the required standoff distance is insured due to the hump and after the impact a spot weld is obtained. Although the main idea of MPW, MPC, and MPSW of tubes and metal sheets is known since a while, the process is not yet fully understood and technologically applicable for composite containing metal inserts and metal parts joining.

2.8.4

Electromagnetic driven self-piercing riveting

Electromagnetic driven self-piercing riveting is a state of the art riveting process if it is executed at standard rivet setting velocity. The main advantage of this method is that additional manufacturing steps such as drilling of holes or extra cleaning are not necessary. Also, the die side sheet does not have to be cut, remaining intact and tight surface of it. However, applying this riveting process to CFRP sheets is very challenging for the reason of uncontrolled damaging of the reinforcement fibers by the piercing rivet. This implies difficulties in achieving optimal usage strength of the assembled hybrid metal composite parts. The use of electromagnetic drive to accelerate up to 50 m/s the punching velocity enables a generation of high-quality joining if an additional riveting aid is used. The riveting aid is an additional metal sheet or foil, as thin as possible but thick enough for the joining strength.

2.8.5

Electron beam welding

Conventional in chamber electron beam welding presents the challenge and limitation that the entire assembly for welding must fit to the vacuum chamber, providing the required environment and radiation protection. The cost of this kind of welding systems is often prohibitive for large joining and limits the process to modest size welding set ups. For too large applications to be constrained by vacuum chamber there is a possibility to perform out-of-vacuum-chamber local vacuum welding. Although out-of-vacuum-chamber local welding has been carried out on very large structures such as pressure vessels or offshore wind mils circular foundation structures it has not yet been developed for complex geometrical shapes.

Composites and hybrid structures

2.9

205

Conclusion

As stated in previous sections, composites are very complex structures with mechanical properties and aging depending on various parameters termed below: • • • • •

nature of reinforcing fibers (carbon, glass, bio sourced, etc.), nature of matrix, geometrical structure of reinforcement, process of shapes manufacturing (stamping, 3D weaving, knitting, etc.), process of thermo consolidation (thermoset, thermoplastic).

A large variety of possible combinations of aforementioned parameters and applications, in which composite structures are used, also complicate even more their use and set up. Furthermore, hybrid materials structures are more and more used and necessary for complex applications in aerospace, automotive, railway, construction, military, and other sectors. Bonding issues and joining techniques, between composite and metallic parts, exist currently and some of them are based on metallic inserts placed inside composite structures. As the reliability and durability of complex composite structures within various applications are extremely important issues, it is of utmost significance to guarantee monitoring of composite manufacturing processes and also of composite mechanical characteristics all along their lifetime. Flowing parts 3 and 4 describe the latest achievements related to smart textile-based sensors used for SHM of composite structures and processes related to composite manufacturing. Moreover, the fourth part contains three case studies on: • • •

Interlock weaving process monitoring; Stamping process supervision; and Infusion process supervision.

References [1] A. Kaw, Mechanics of Composite Material, Taylor & Francis Group, New York, 2006. [2] S.L. Ogin, Textile-reinforced composite materials, in: A.R. Horrocks, S.C. Anand (Eds.), Handbook of Technical Textile, Woodhead publishing Limited, London, 2000, pp. 264e279. [3] S.T. Peters, Handbook of Composites, Chapman & Hall, UK, Mountain View, 1982. [4] D. Gay, Matériaux Composites, Hermes, Paris, 1997. [5] J.M. Cole, J. Berthelot, Composite Materials: Mechanical Behavior and Structural Analysis, Springer-Verlag, New York, 1999. [6] V.V. Vasiliev, E.V. Morozov, Mechanics and Analysis of Composite Materials, Elsevier Science Ltd, Oxford, 2001. [7] P. Lucas, G. Zanella, Mise en oeuvre des composites TP, Compounds
a fibres courtes et longues, Techniques de l’ingénieur, Paris, 2007.

206

Smart Textiles for In Situ Monitoring of Composites

[8] F.K. Ko, T.W. Chou, Composite Materials Series e Textile Structural Composites, Elsevier Science, Amsterdam, 1989. [9] J.V. Risicato, F. Kelly, D. Soulat, X. Legrand, W. Tr€ umper, C. Cochrane, V. Koncar, A complex shaped reinforced thermoplastic composite part made of commingled yarns with integrated sensor, Applied Composite Materials 22 (1) (2014) 81e98. [10] M.J. Robert, Mechanics of Composite Materials, second ed., Taylor & Francis Group, New York, 1999. [11] L. Tong, A.P. Mouritz, M.K. Bannister, 3D Fibre Reinforced Composites, Elsevier Science Ltd., London, 2002. [12] G.C. Sih, A.M. Skudra, Failure Mechanics of Composites, Elsevier Science Publishers BV, New York, 1986. [13] R. Kamiya, et al., Some recent advances in the fabrication and design of three-dimensional textile preforms: a review, Composites Science and Technology 60 (2000) 33e47. [14] B. Cox, M. Dadkhah, The macroscopic elasticity of 3D woven composites, Journal of Composite Materials 29 (6) (1995) 785e819. [15] A.K. Bilisik, Multiaxis three dimensional woven fabric, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabric Technology, Athens, INTECH, 2011 p. on line. [16] B.K. Behera, P.K. Hari, Woven Textile Structure: Theory and Application, Woodhead Pub, Cambridge, 2010. [17] P. Curtis, S.M. Bishop, An assessment of the potential of woven carbon fibre-reinforced plastics for high performance applications, Composites 15 (4) (1984) 259e265. [18] The Hong Hong Cotton Spinncer Association, Textile Handbook, Hong kong productivity council, Hong Kong, 2001. [19] M. Rao, B. Sankar, G. Subhash, Effect of Z-yarns on the stiffness and strength of threedimensional woven composites, Composites Part B: Engineering 40 (6) (2009) 540e551. [20] Y. Mahadik, K. Brown, A.S. Hallett, Characterisation of 3D woven composite internal architecture and effect of compaction, Composite Part A: Applied Science and Manufacturing 41 (7) (2010) 872e880. [21] M. Ansar, W. Xinwei, Z. Chouwei, Modeling strategies of 3D woven composites: a review, Composite Structures 93 (8) (2011) 1947e1963. [22] S. Buchanan, A. Grigorash, E. Archer, A. McIlhagger, J. Quinn, G. Stewart, Analytical elastic stiffness model for 3D woven orthogonal interlock composites, Science and Technology 70 (2010) 1597e1604. [23] X. Chen, L. Taylor, L. Tsai, An overview on fabrication of three-dimensional woven textile preforms for composites, Textile Research Journal 81 (9) (2011) 932e944. [24] S. Nehme, A. Hallal, F. Fardoun, R. Younes, B. Hagege, Z. Aboura, et al., Numerical/ analytical methods to evaluate the mechanical behavior of interlock composites, Journal of Composite Material 45 (16) (2011) 1699e1716. [25] S. Lomov, I. Verpoest, F. Robitalle, Manufacturing and internal geometry of textiles, in: Design and Manufacture of Textile Composites, Woodhead publishing Limited, London, 2005, pp. 1e61. [26] K. Leong, B. Lee, I. Herszberg, M. Bannister, The effect of binder path on the tensile properties and failure of multilayer woven CFRP composites, Composites Science and Technology 60 (2000) 149e156. [27] J. Solden, B. Hill, Conventional weaving of shaped preforms for engineering composites, Composite Part A: Applied Science and Manufacturing 29 (1998) 757e762. [28] S. Rudov-Clark, A.P. Mouritz, L. Lee, M. Bannister, Fiber damage in the manufacture of advanced three-dimensional woven composites, Composite Part A: Applied Science and Manufacturing 34 (10) (2003) 963e970.

Composites and hybrid structures

207

[29] V. Guenon, T. Chou, J. Gillespie, Toughness properties of a three-dimensional carbonepoxy composite, Journal of Material Science 24 (1989) 4168e4175. [30] S.V. Lomov, D. Ivanov, G. Perie, I. Verpoest, Modelling 3D fabrics and 3D reinforced composites: challenges and solutions, in: 1st World Conference on 3D Fabrics, Manchester, 2008. [31] M. Lefebvre, F. Boussu, D. Coutelier, Influence of high-performance yarns degradation inside three-dimensional warp interlock fabric, Journal of Industrial Textiles 42 (2013) 475e488. [32] J. Ruzand, G. Guenot, Multiaxial Three-dimensional Fabric and Process for its Manufacture, International patent Brevet WO : 94/20658, 1994. [33] G. Mood, Multi-axial Yarn Structure and Weaving Method, United States Brevet 5-540260, 1996. [34] L. Bryn, M.A. Islam, W.L. Lowery, H.D. Harries, Three Dimensional Woven Forms with Integral Bias Fibers and Bias Weaving Loom, United States Brevet 6-742-547, 2004. [35] S. Nayfeh, J. Rohrs, O. Rifni, et al., Bias Weaving Machine, United States Brevet 7-077167, 2006. [36] M. Lima, R. Fangueiro, A. Costa, C. Rosiepen, V. Rocha, Multiweave, a prototype weaving machine for multiaxial technical fabrics, Indian Journal of Fibre and Textile Research 34 (2009) 59e63. [37] G.L. Farley, Method and Apparatus for Weaving a Woven Angle Ply Fabric, United States Brevet 5,224,519, 1993. [38] M. Anahara, Y. Yasui, H. Omori, Three-Dimensional Textile and Method of Producing the Three Dimensional Textile, Europe Brevet 0-426-878-A1, 1991. [39] Y. Yasui, M. Anahara, F. Hori, Y. Mita, Production of Three-Dimensional Woven Fabric, Japan Brevet 05-106-140, 1993. [40] H. Uchida, T. Yamamoto, H. Takashima, T. Yamamoto, S. Nishlyama, M. Shinya, ThreeDimensional Weaving Machine, United States Brevet 6-003-563, 1999. [41] A.K. Bilisk, M.H. Mohamed, Multiaxis three-dimensional (3D) flat woven preform-tube carrier weaving, Textile Research Journal 80 (8) (2010) 696e711. [42] K. Bilisik, Multiaxis 3D weaving: comparison of developed two weaving methods (tuberapier weaving versus tube-carrier weaving) and effects of bias yarn path to preform properties, Fibers and Polymers 11 (2010) 104e114, n %11. [43] A.K. Bilisik, Multiaxial Three Dimensional (3D) Woven Carbon Preform, United States Brevet 6-129-122, 2000, 2 p. [44] P. Stickler, M. Ramulu, P. Johnson, Experimental and numerical analysis of transverse stitched T-joints in bending, Composite Structures 50 (1) (2000) 17e27. [45] A.P. Mouritz, B. Cox, A mechanistic approach to the properties of stitched laminates, Composite Part A: Applied Science and Manufacturing 31 (1) (2000) 1e27. [46] A.P. Mouritz, C. Baini, I. Herszberg, Mode I interlaminar fracture toughness properties of advanced textile fiber glass composites, Composite Part A: Applied Science and Manufacturing 30 (7) (1999) 859e870. [47] F. Edgren, C. Soutis, L. Asp, Damage tolerance analysis of NCF composite sandwich panels, Composites Science and Technology 68 (2008) 2635e2645. [48] G. Bibo, I.K.M. Hogg, Mechanical characterization of glass- and composites made with non-crimp fabrics, Composites Science and Technology 57 (1997) 1221e1241. [49] K. Bilisik, Three-dimensional braiding for composites: a review, Textile Research Journal 83 (13) (2012) 1414e1436. [50] P. Chang, A.P. Mouritz, B.N. Cox, Elevated temperature properties of pinned composite lap joints, Journal of Composite Material 42 (8) (2008) 741e769.

208

Smart Textiles for In Situ Monitoring of Composites

[51] A.P. Mouritz, Compression properties of z-pinned composite laminates, Composite Science Technology 67 (15) (2007) 3110e3120. [52] D. Coutelier (Ed.), Advanced Material Textiles for Reinforced Structures for Complex Lightweight Applications, ENSIAME, Valenciennes, 2016. [53] E.D. Witten, T. Kraus, M. Kuhnel, Composite Market Report 2017, November 28, 2016 (En ligne). Available: http://www.eucia.eu/userfiles/files/20161128_market_report_2016_ english.pdf. [54] C. Warren, Lower Cost High Performance Carbon Fibre, U.S. Department of Energy, Oak Ridge, 2011. [55] Fibermax Composites, Aramid Fiber Stronger and Safer with Aramid Fiber, Fibermax Composites, 2017 (En ligne). Available: http://www.aramid.eu. [56] marketsandmarkets.com, Aramid Fiber Market by Type (Para-aramid, Meta-aramid, and Others), and Application (Security & Protection, Frictional Materials, Optical Fiber, Tire Reinforcement, Rubber Reinforcement, Electrical Insulation, Aerospace, and Others) Global Trends & Forecasts to 2021, Marketsandmarkets, London, 2016. [57] R. linker, Global Natural Fiber Composite Market 2015-2020: Trends, Forecast, and Opportunity Analysis, December 2015, March 30, 2016 (En ligne). Available: http://www. prnewswire.com/news-releases/global-natural-fiber-composite-market-2015-2020-trendsforecast-and-opportunity-analysis-december-2015-300243712.html. [58] N. Institute, WPC/NFC Market Study, NOVA Institute, H€ urth, 2014. [59] P. Irving, C. Soutis, Polymer Composites in the Aerospace Industry, Woodhead, London, 2014. [60] P. Pasbakhsh, A. Haghi, E. Zaikov, Applied research on polymer composites, in: AAP Research Notes on Polymer Engineering, Science and Technology, 2015. [61] G. Koronis, A. Silva, M. Fontul, Green composites, a review of adequate materials for automotive applications, Composites Part B: Engineering 44 (2013) 120e127. [62] M. Reddy, S. Vivekanandhana, M. Misraa, S. Bhatiac, A. Mohantya, Biobased plastics and bionanocomposites: current status and future opportunities, Progress in Polymer Science 38 (2013) 1653e1689. [63] C.R. Farrar, K. Worden, An introduction to structural health monitoring, Philosophical Transactions of the Royal Society A 365 (2007) 303e315. [64] A.E. Aktan, F. Catbas, K. Grimmelsman, M. Pervizpour, Development of a Model Health Monitoring Guide for Major Bridges, Drexel Intelligent Infrastructure and Transportation Safety Institute, Drexel, 2003. [65] O. Buyukozturk, T.Y. Yu, Structural health monitoring and seismic impact assessment, in: Proc. 4th Nat. Conf. Earthquake Eng., Istanbul, 2003. [66] S. Nassar, X. Yang, Fastening and joining of composite materials, in: E. Patterson, D. Backman, G. Cloud (Eds.), Composite Materials and Joining Technologies for Composites, Proceedings of the 2012 Annual Conference on Experimental and Applied Mechanics, vol. 7, Springer, New York, New York, 2012, pp. 5e23. [67] M. Ueda, S. Miyake, H. Hasegawa, Y. Hirano, Instantaneous mechanical fastening of quasi-isotropic CFRP laminates by a self-piercing rivet, Composite Structures 94 (2012) 3388e3393. [68] V. Virupaksha, S. Nassar, Effect of washers and bolt tension on the behavior of double-lap S2-glass fabric-epoxy composite joints, in: International Conference on Pressure Vessels and Piping (PVP 2008), Chicago, IL, USA, 27e31 July 2008, 2008. [69] D. Liu, Y. Tang, W. Cong, A review of mechanical drilling for composite laminates, Composite Structures 94 (2012) 1265e1279.

Composites and hybrid structures

209

[70] J. Oh, Y. Kim, D. Lee, Optimum bolted joints for hybrid composite materials, Composite Structures 38 (1e4) (1997) 329e341. [71] J. Gebhardt, J. Fleischer, Experimental investigation and performance enhancement of inserts in composite parts, in: 5th CATS (Conference on Assembly Systems and Technologies) 2014-CIRP, Procedia CIRP 23, Dresden, Germany, November 12 to 14, 2014, 2014. [72] C. Dufour, F. Boussu, P. Wang, D. Soulat, K. Kalnins, E. Labans, P. Lefort, M. Wallenius Henrinksson, 3D warp interlock fabric structure as a solution of composite joining with metallic part, in: Proceedings of the 6th World Conference on 3D Fabrics and Their Applications, Raleigh, NC - USA, 2015. [73] C. Ageorges, L. Ye, M. Hou, Advances in fusion bonding techniques for joining thermoplastic matrix composites: a review, Composites: Part A 32 (2001) 839e857. [74] J. Esteves, S. Goushegir, J. dos Santos, L. Canto, E. Hage Jr., S. Amancio-Filho, Friction spot joining of aluminum AA6181-T4 and carbon fiber-reinforced poly(phenylene sulfide): effects of process parameters on the microstructure and mechanical strength, Materials and Design 66 (2015) 437e445. [75] S. Ucsnik, M. Scheerer, S. Zaremba, D. Pahr, Experimental investigation of a novel hybrid metalecomposite joining technology, Composites: Part A 41 (2010) 369e374. [76] A. Bogdanovich, M. Dannemann, J. D€oll, T. Leschik, J. Singletary, W. Hufenbach, Experimental study of joining thick composites reinforced with non-crimp 3D orthogonal woven E-glass fabrics, Composites: Part A 42 (2011) 896e905. [77] Z. Huang, S. Sugiyama, J. Yanagimoto, Applicability of adhesiveeembossing hybrid joining process to glass-fiber-reinforced plastic and metallic thin sheets, Journal of Materials Processing Technology 214 (2014) 2018e2028. [78] Z. Huang, S. Sugiyama, J. Yanagimoto, Adhesiveeembossing hybrid joining process to fiber-reinforced thermosetting plastic and metallic thin sheets, in: 11th International Conference on Technology of Plasticity, ICTP 2014, Procedia Engineering, Nagoya, Japan, 19e24 October 2014, 2014. [79] M. Gude, W. Hufenbach, R. Kupfer, A. Freund, C. Vogel, Development of novel formlocked joints for textile reinforced thermoplastics and metallic components, Journal of Materials Processing Technology 216 (2015) 140e145. [80] B. Cosson, M. Deléglise, W. Knapp, Numerical analysis of thermoplastic composites laser welding using ray tracing method, Composites: Part B 68 (2015) 85e91. [81] E. Rodríguez-Vidal, J. Lambarri, C. Soriano, C. Sanz, G. Verhaeghe, A combined experimental and numerical approach to the laser joining of hybrid polymer e metal parts, in: 8th International Conference on Photonic Technologies LANE 2014, Physics Procedia, Furth, Germany, 8 e 11 September 2014, 2014. [82] P. Cognard, Assemblage des composites - Les points forts du collage, Techniques de l’Ingénieur, traité Génie mécanique, n %1BM 7 624, Juillet 2003, pp. 1e4. [83] P. Cognard, Collage des composites - Secteurs routier et ferroviaire, Techniques de l’Ingénieur, traité Génie mécanique, n %1BM 7 627, October 2003, pp. 1e7. [84] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Procédé pour l’assemblage entre une pi
ece en matériau métallique et une pi
ece en matériau composite
a matrice organique, France Brevet FR3 030329 A1, Devember 19, 2014. [85] G. Racineux, D. Priem, J.-M. Lebrun, Y. Amosse, C. Khalil, Assembly Method between a Part Made of Metal Material and a Part Made of Organic Matrix Composite Material; Corresponding Parts Made of Organic Matrix Composite Material and Assembly, France Patent WO 2016/097656 A1, Devember 23, 2016.

210

Smart Textiles for In Situ Monitoring of Composites

Further Reading  [1] A. Lobnik, S.K.U.M. Turel, Optical chemical sensors: design and applications, in: Advances in Chemical Sensors, Intech, Maribor, 2012, pp. 3e28. [2] S.A. Dyer, Wiley Survey of Instrumentation and Measurement, Wiley, Kansas city, 2001, pp. 84e98. [3] C. Cochrane, Développement d’un syst
eme de mesure d’allongement pour voilure de parachute, L’Universite des Sciences et Technologies de Lille, Villeneuve d’Ascq, 2007. [4] Peratech, QTCTM Material Science the Physics of QTCTM Material Operation, Peratech, August 27, 2013 (En ligne). Available: http://www.peratech.com/qtc-science.html. [5] B. Hafner, Energy Dispersive Spectroscopy on the SEM: A Primer, Characterization Facility, University of Minnesota, Twin Cities, 2007. [6] A. Elschner, S. Kirchmeyer, W. L€ovenich, PEDOT: Principles and Applications of an Intrinsically Conductive Polymer, CRC Press Taylor & Francis Group, Boca Raton, 2010. [7] G. Asch, Les capteurs en instrumentation industrielle - 7
eme édition, Dunod, Paris, 2010. [8] C. Cochrane, V. Koncar, M. Lewandowski, Development of a flexible strain sensor for textile structures, in: 2nd Ambience 08 International Scientific Conference Proceedings, Smart Textiles - Technology and Design, Borås, 2008. [9] C. Cochrane, V. Koncar, M. Lewandowski, C. Dufour, Design and development of a flexible strain sensor for textile structures based on a conductive polymer composite, Sensors 7 (1) (2007) 473e492. [10] D. Yang, X. Tao, A. Zhang, Optical responses of FBG sensors under deformations, in: Smart Fibres, Fabrics & Clothing, Woodhead, Hong Kong, 2001, p. 150. [11] H.P. Konka, M.A. Wahab, K. Lian, Piezoelectric fiber composite transducers for health monitoring in composite structures, Sensors and Actuators A: Physical 194 (2013) 84e94. [12] I. Cristian, S. Nauman, C. Cochrane, V. Koncar, Electro-conductive sensors and heating elements based on conductive polymer composites in woven structures, in: S. Vassiliadis (Ed.), Advances in Modern Woven Fabrics Technology, Athens, 2011, pp. 3e22. [13] R. Bartalesi, N. Carbonaro, F. Lorussi, M. Tesconi, A. Tognetti, G. Zupone, DeRossi, Smart textiles: toward a wearable motion capture system, in: Ninth International Symposium on the 3D Analysis of Human Movement, Valenciennes, 2006. [14] V. Koncar (Ed.), Smart Textiles and Their Applicatons, Elsevier, London, 2016. [15] X. Tao, V. Koncar (Eds.), Handbook on Smart Textiles, Springer, Hong Kong, 2015. [16] B. Adhikari, S. Majumdar, Polymers in sensor applications, Progress in Polymer Science 29 (7) (2004) 699e766. [17] M. Akerfeldt, M. Straat, P. Walkenstrom, Influence of coating parameters on textile and electrical properties of a poly(3,4-ethylene dioxythiophene):poly(styrene sulfonate)/polyurethane-coated textile, Textile Research Journal 83 (20) (2013) 2164e2176. [18] J.N. Aneli, G.E. Zaikov, L.M. Khananashvili, Effects of mechanical deformations on the structurization and electric conductivity of electric conducting polymer black-filled rubbers under, Journal of Applied Polymer Science 74 (1) (1999) 601e621. [19] E.R. Arakelova, A.M. Khachatryan, K.E. Avjyan, et al., Deposition of high-ohmic oriented ZnO films on glass, Si, and PEDOT-PSS, PEDOT-PSS(PVA) substrates at low temperatures by DC-magnetron sputtering, Journal of Contemporary Physics (Armenian Academy of Sciences) 46 (6) (2011) 293e299.

Composites and hybrid structures

211

[20] F.J. Arregui, I.R. Matıas, M. Lopez-amo, Optical fiber strain gauge based on a tapered single-mode fiber, Sensors and Actuators 79 (1) (2000) 90e96. [21] P.T. Asada, H.H. Gibbs, Wearable conductive fiber sensors for multi-axis human joint angle measurements, Journal of Neuroengineering and Rehabilitation 2 (2005) 1, p. on line. [22] T. Bashir, M. Ali, S. Cho, N. Persson, M. Skrifvars, OCVD polymerization of PEDOT: effect of pre-treatment steps on PEDOT-coated conductive fibers and a morphological study of PEDOT distribution on textile yarns, Polymers for Advanced Technologies 24 (2) (2013) 210e219. [23] S. Bhadra, D. Khastgir, N.K. Singha, Progress in preparation, processing and applications of polyaniline, Progress in Polymer Science 34 (8) (2009) 783e810. [24] A.B. Bradley, P.D. Render, The development of a parachute strain measurement technique, American Institute of Aeronautics and Astronautics 1 (1) (1986) 194e202. [25] B. Carter, J. Shannon, J. Forshaw, Measurement of displacement and strain by capacitance methods, Proceedings - Institution of Mechanical Engineers 5 (1945) 215e221. [26] A. Chen, C.H. Sawhney, Soft-structured sensors and connectors by inkjet printing, AATCC Review 7 (6) (2007) 1e10. [27] C. Chen, J.C. Larue, R.D. Nelson, L. Kulinsky, M.J. Madou, Electrical conductivity of polymer blends of poly (3, 4-ethylenedioxythiophene): poly (styrenesulfonate): N-methyl-2-pyrrolidinone and polyvinyl alcohol, Journal of Applied Polymer Science 1 (2011) 1e8. [28] C. Chen, A. Torrents, L. Kulinsky, R. Nelson, M. Madou, L. Valdevit, J. LaRue, Mechanical characterizations of cast poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate)/polyvinyl alcohol thin films, Synthetic Metals 161 (2011) 21e22. [29] O.P. Dimitriev, D.A. Grinko, Y.V. Noskov, et al., PEDOT: PSS films-effect of organic solvent additives and annealing on the film conductivity, Synthetic Metals 159 (21) (2009) 2237e2239. [30] G. Du, P. Popper, Analysis of a circular braiding process for complex shapes, Journal of the Textile Institute 85 (3) (1994) 316e337. [31] L. Flandin, A. Chang, S. Nazarenko, A. Hiltner, E. Baer, Effect of strain on the properties of an ethylene-octene elastomer with conductive carbon fillers, Journal of Applied Polymer Science 76 (6) (2000) 894e905. [32] L. Flandin, A. Hiltner, E. Baer, Interrelationships between electrical and mechanical properties of a carbon black-filled ethyleneeoctene elastomer, Polymer 42 (2) (2001) 827e838. [33] B. Friedel, T.J.K. Brenner, C.R. McNeill, et al., Influence of solution heating on the properties of PEDOT: PSS colloidal solutions and impact on the device performance of polymer solar cells, Organic Electronics 12 (10) (2011) 1736e1745. [34] L. Guo, L. Berglin, M. Heikki, Textile strain sensors characterization - sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 2 (1) (2010) 51e63. [35] L. Guo, L. Berglin, M. Heiki, Textile strain sensors characterization - sensitivity, linearity, stability and hysteresis, Nordic Textile Journal 1 (2) (2010) 51e56. [36] J. Han, M. Shen, W. Cao, Hopping conduction in Mn-doped ZnO, Applied Physics Letters 82 (1) (2003) 67e69. [37] C.T. Huang, C.L. Shen, C.F. Tang, S.H. Chang, A wearable yarn-based piezo-resistive sensor, Sensors and Actuators A: Physical 141 (2) (2008) 396e403. [38] C.T. Huang, C.F. Tang, M.C. Lee, S.H. Chang, Parametric design of yarn-based piezoresistive sensors for smart textiles, Sensors and Actuators A: Physical 148 (1) (2008) 10e15.

212

Smart Textiles for In Situ Monitoring of Composites

[39] J. Hwang, F. Amy, A. Kahn, Spectroscopic study on sputtered PEDOT$PSS: role of surface PSS layer, Organic Electronics 7 (5) (2006) 387e396. [40] B. Kim, V. Koncar, E. Devaux, Electrical properties of conductive polymers: PET nanocomposites’ fibres, AUTEX Research Journal 4 (1) (2004) 1e13. [41] B. Kim, V. Koncar, C. Dufour, Polyaniline (PANI) coated PET conductive yarns: study of electrical, mechanical and electro-mechanical properties, Journal of Applied Polymer Science 101 (3) (2006) 1252e1256. [42] B. Kim, V. Koncar, E. Devaux, C. Dufour, P. Viallier, Electrical and Morphological properties of PP and PET conductive polymer fibers, Synthetic Metals 146 (2) (2004) 167e174. [43] G. Latessa, F. Brunetti, A. Reale, G. Saggio, A. Di Carlo, Piezoresistive behaviour of flexible PEDOT: PSS based sensors, Sensors and Actuators B: Chemical 139 (2) (2009) 304e309. [44] Z.S. Levin, C. Robert, J.F. Feller, M. Castro, J. Grunlan, Flexible latex d polyaniline segregated network composite coating capable of measuring large strain on epoxy, Smart Materials and Structures 22 (1) (2013) p. on line. [45] J. Li, J.P. Longtin, S. Tankiewicz, A. Gouldstone, S. Sampath, Interdigital capacitive strain gauges fabricated by direct-write thermal spray and ultrafast laser micromachining, Sensors and Actuators A: Physical 133 (1) (2007) 1e8. [46] L. Li, S. Liu, F. Ding, T. Hua, W. Man Au, K. Wong, Electromechanical analysis of length-related resistance and contact resistance of conductive knitted fabrics, Textile Research Journal 82 (20) (2012) 2062e2070. [47] X. Lin, K. Zhou, X. Zhang, D. Zhang, Development, modeling and application of piezoelectric fiber composites, Transactions of Nonferrous Metals Society of China 23 (1) (2013) 98e107. [48] F. Lorussi, W. Rocchia, E.P. Scilingo, A. Tognetti, D. DeRossi, Wearable, redundant fabric-based sensor arrays for reconstruction of body segment posture, IEEE Sensors Journal 4 (6) (2004) 807e818. [49] J. Rausch, E. M€ader, Health monitoring in continuous glass fibre reinforced thermoplastics: tailored sensitivity and cyclic loading of CNT-based interphase sensors, Composites Science and Technology 70 (13) (2010) 2023e2030. [50] S. Nauman, I. Cristian, V. Koncar, Simultaneous application of fibrous piezoresistive sensors for compression and traction detection in glass laminate composites, Sensors 11 (10) (2011) 9478e9498. [51] J. Nevrela, M. Micjan, M. Novota, et al., Secondary doping in poly(3,4ethylenedioxythiophene):poly(4-styrenesulfonate) thin films, Journal of Polymer Science Part B: Polymer Physics 53 (16) (2015) 1139e1146. [52] D. Orata, M. Mukabi, H. Njenga, Electro-characterization of polypyrolle electrosynthesized on a montmorillonite host-matrix, in aqueous media containing sulphuric acid as supporting electrolyte, IOSR Journal of Applied Chemistry 7 (5) (2014) 59e72. [53] A. Ramanavicius, A. Malinauskas, Electrochemical sensors based on conducting polymer polypyrrole, Electrochimica Acta 51 (27) (2006) 6025e6037. [54] M. Rothmaier, M.P. Luong, F. Clemens, Textile pressure sensor made of flexible plastic optical fibers, Sensors 8 (7) (2008) 4318e4329. [55] A. Schwarz, I. Kazani, L. Cuny, Electro-conductive and elastic hybrid yarns - the effects of stretching, cyclic straining and washing on their electro-conductive properties, Materials and Design 32 (1) (2011) 4247e4256. [56] E.P. Scilingo, F. Lorussi, A. Mazzoldi, D. De Rossi, Strain-sensing fabrics for wearable kinaesthetic-like systems, IEEE Sensors Journal 3 (4) (2003) 460e467.

Composites and hybrid structures

213

[57] A. Soroudi, M. Skrifvars, Melt blending of carbon nanotubes/polyaniline/polypropylene compounds and their melt spinning to conductive fibres, Synthetic Metals 160 (11) (2010) 1143e1147. [58] X. Tao, I. Fsaifes, V. Koncar, C. Dufour, C. Lepers, L. Hay, B. Capoen, M. Bouazaoui, CO2 laser-induced crystallization of solegel-derived indium tin oxide films, Applied Physics A 96 (3) (2009) 741e749. [59] X. Tao, V. Koncar, C. Dufour, Novel geometry pattern for the wire organic electrochemical textile transistor, Journal of the Electrochemical Society 158 (5) (2011) 572e577. [60] X. Tao, V. Koncar, T.H. Huang, C.L. Shen, Y.C. Ko, G.T. Jou, How to make reliable, washable, and wearable textronic devices, Sensors 17 (4) (2017) 673. [61] D.R. Thevenot, K. Toth, R.A. Durst, G.S. Wilson, Electrochemical biosensors: recommended definitions and classification, Pure and Applied Chemistry 71 (12) (2009) p. on line. [62] R. Matsuzaki, A. Todoroki, Wireless flexible capacitive sensor based on ultra-flexible epoxy resin for strain measurement of automobile tires, Sensors and Actuators A: Physical 140 (1) (2007) 32e42. [63] J.P. Wang, P. Xue, X.M. Tao, Strain sensing behavior of electrically conductive fibers under large deformation, Materials Science and Engineering A 528 (6) (2011) 2863e2869. [64] A. Weder, S. Geller, A. Heinig, T. Tyczynski, W. Hufenbach, W. Fischer, A novel technology for the high-volume production of intelligent composite structures with integrated piezoceramic sensors and electronic components, Sensors and Actuators A: Physical 1 (2013) 1e5. [65] S. Xiong, L. Zhang, X. Lu, Conductivities enhancement of poly(3,4 ethylenedioxythiophene)/poly(styrene sulfonate) transparent electrodes with diol additives, Polymer Bulletin 70 (2013) 237e247. [66] P. Xue, K.H. Park, X.M. Tao, Electrically conductive yarns based on PVA/carbon nanotubes, Composite Structures 78 (2) (2007) 271e277. [67] P. Xue, X. Tao, H.Y. Tsang, In situ SEM studies on strain sensing mechanisms of PPycoated electrically conducting fabrics, Applied Surface Science 253 (7) (2007) 3387e3392. [68] R. Zeiser, T. Fellner, J. Wilde, Academia day, Development and Testing of Capacitive Strain Gauges, vol. 1, 2011, pp. 1e6. [69] R. Zhang, H. Deng, R. Valenca, J. Jin, Q. Fu, E. Bilotti, T. Peijs, Carbon nanotube polymer coatings for textile yarns with good strain sensing capability, Sensors and Actuators A: Physical 179 (2012) 83e91. [70] J. Zhou, G. Tzamalis, N.A. Zaidi, Electrically conductive PANI multifilaments spun by a wet spinning process, Journal of Materials Science 36 (13) (2001) 3089e3095. [71] J. Zieba, M. Frydrysiak, Textronics e electrical and electronic textiles. Sensors for breathing frequency measurement, Fibres and Textiles in Eastern Europe 14 (5) (2006) 43e48. [72] K. Kim, C.S. Park, D.W. Lee, S.M. Cho, C.R. Han, Measurement of the gauge factor of carbon fiber and its application to sensors, Microelectronic Engineering 85 (5) (2008) 787e791. [73] I. Cristian, S. Nauman, F. Boussu, V. Koncar, A study of strength transfer from tow to textile composite using different reinforcement architectures, Applied Composite Materials 19 (3) (2012) 427e442.

214

Smart Textiles for In Situ Monitoring of Composites

[74] U. Lang, N. Naujoks, J. Dual, Mechanical characterization of PEDOT: PSS thin films, Synthetic Metals 159 (5) (2009) 473e479. [75] N. Trifigny, F.M. Kelly, C. Cochrane, et al., In-situ measurements of strain and stress of e-glass warp interlock fabric using innovative sensors, in: CompositeWeek@ Leuven(TexComp-11): Proceedings, Leuven, 2013. [76] S. Sha, X. Tao, F. Boussu, V. Koncar, Mechanical properties of high modulus UHMW yarn measured by intelligent sensors, in: 13th AUTEX World Textile Conference Book of Proceedings, Dresden, 2013. [77] Co Heraeus Precious Metals GmbH & Co., Coating Guide Clevios™ P Formulations, Heraeus, Hanau, 2012. [78] J. M€ader, E. Rausch, Health monitoring in continuous glass fibre reinforced thermoplastics : manufacturing and application of interphase sensors based on carbon nanotubes, Composites Science and Technology 70 (11) (2010) 1589e1596. [79] F. Boussu, The use of warp interlock fabric inside textile composite protection against ballistic impact, Textile Research Journal 81 (4) (2011) 344e354. [80] B.N. Cox, M.S. Dadkhah, W.L. Morris, J.G. Flintoff, Failure mechanisms of 3D woven composites in tension, compression, and bending, Acta Metallurgica et Materialia 42 (12) (1994) 3967e3984. [81] M. Niggemann, M. Pellmann, P. Bolsmann, J. Reinbold, K. Cammann, Remote sensing of tetrachloroethene with a micro-fibre optical gas sensor based on surface plasmon resonance spectroscopy, Sensors and Actuators B: Chemical 34 (1) (1996) 328e333. [82] E.D. Witten, T. Kraus, M. Kuhnel, Composite Market Report 2017, November 15, 2016 (En ligne). Available: http://www.eucia.eu/userfiles/files/20161128_market_report_ 2016_english.pdf. [83] R. i. China, Global and China Aramid Fiber Industry Report, 2015-2018, Research in China, Beijing, 2015. https://www.reportlinker.com/p02515745/Global-and-ChinaAramid-Fiber-Industry-Report.html. [84] C.Q. Yang, Q. He, R.E. Lyon, et al., Investigation of the flammability of different textile fabrics using micro-scale combustion calorimetry, Polymer Degradation and Stability 95 (2) (2010) 108e115. [85] R. Gupta, V. Kumar, P.K. Goyal, et al., Thermogravimetric analysis of heat treated CR-39 polymer, Journal of Chemical and Pharmaceutical Research 2 (4) (2010) 629e634. [86] N. Amir, F. Ahmad, P.S.M. Megat-Yusoff, Study on the fibre reinforced epoxy-based intumescent coating formulations and their char characteristics, Journal of Applied Sciences 11 (10) (2011) 1678e1687. [87] R.E. Lyon, R. Walters, Flammability of Automotive Plastics, Federal Aviation Administration William J. Hughes Technical Center, New Jersey, 2005. [88] F. Laoutid, L. Bonnaud, M. Alexandre, New prospects in flame retardant polymer materials: from fundamentals to nanocomposites, Materials Science and Engineering: R: Reports 63 (3) (2009) 100e125. [89] A.M. Grancaric, A. Tarbuk, E. Chibowski, Surface free energy of textiles, Tekstil 57 (1) (2008) 29e39. [90] K.E. Jackson, L. Edwin, R. Fasanella, L. Boitnott, K. Lyle, Full scale crash test and Finite Element Simulation of a composite prototype helicopter, aircraft design, Testing and Performance 1 (2003) 121e129, n %11. [91] J. Obradovic, S. Boriab, G. Belingardia, Lightweight design and crash analysis of composite front impact energy absorbing structures, Composite Structures 94 (2012) 423e430, n %12.

Composites and hybrid structures

215

[92] E. Mehdi, A. Mokhtar, D. Magid, F. Ahmadun, A. Prithvi Raj, S. Taher, A new composite energy absorbing system for aircraft and helicopter, Composite Structures 75 (2006) 14e23, n %11-4. [93] A.M. Grancaric, A. Tarbuk, I. Jerkovic, et al., Surface free energy of multilayered weftknitted fabrics and related composite plates, in: CompositeWeek@Leuven (TexComp11): Proceedings, Leuven, 2013. [94] A.M. Grancaric, J.V. Risicato, I. Jerkovic, et al., Interface phenomena of PP/glass braided structures, in: XIIIth International Izmir Textile and Apparel Symposium Book of Proceedings, Izmir, 2014. [95] A.M. Grancaric, V. Kovacevic, M. Leskovac, et al., Interface phenomena of hydrolyzed polyester fabric, in: 2nd International Textile, Clothing & Design Conference - Magic World of Textiles Book of Proceedings, Dubrovnik, 2004. [96] E. Chibowski, L. Holysz, G.A.M. Kip, et al., Surface free energy components of glass from ellipsometry and zeta potential measurements, Journal of Colloid and Interface Science 132 (1) (1989) 54e61. [97] C. Faulkner, What Is NFC? Everything You Need to Know, Techreader, May 9, 2017 (En ligne). Available: http://www.techradar.com/news/what-is-nfc. [98] N. Pelly, J. Hamilton, How to NFC, May 10, 2011 (En ligne). Available: https://www. youtube.com/watch?v¼49L7z3rxz4Q. [99] I. p. NFC.org a Square, Near Field Communication Versus Bluetooth, November 28, 2012 (En ligne). Available: http://nearfieldcommunication.org/bluetooth.html. [100] S. Nauman, P. Lapeyronnie, I. Cristian, F. Boussu, V. Koncar, On line measurement of structural deformations in composites, IEEE Sensors 11 (6) (2010) 1329e1336. [101] I. Jerkovic, V. Koncar, A.M. Grancaric, New textile sensor for in situ structural health monitoring of textile reinforced thermoplastic composites based on the conductive poly(3,4-ethylenedioxythiohene)-poly(styrenesulfonate) polymer complex, Sensors 17 (2017) 1e21, n %12297.

This page intentionally left blank

Structural health monitoring of composite structures 3.1

3

Health monitoring definitions

As stated in Section 2.6.6, health monitoring is defined as the tracking of any aspect of a structure’s health by reliably measured data and analytical simulations in conjunction with heuristic experience, so that the current and expected future performance of the composite part for at least the most critical limit events, can be described in a proactive manner. The single most important distinction of health monitoring from a typical in-depth composite part evaluation and testing practice is the minimum standards that are required for analytical modeling for trustworthy computer simulations, and how the measurements, loads, and tests are designed and implemented in conjunction with the analytical simulations. The process of implementing a damage detection and characterization strategy for engineering structures is referred to as Structural Health Monitoring (SHM). Here, damage is defined as changes of the material and/or geometric properties of a structural system, including changes to the boundary conditions and system connectivity, which adversely affect the system’s performance. The SHM process involves the observation of a system over time using periodically sampled dynamic response measurements from an array of sensors, the extraction of damage-sensitive features from these measurements, and their statistical analysis to determine the current state of system health. For long-term SHM, the output of the process is periodically updated with information regarding the ability of the structure to perform its intended function in light of the inevitable aging and degradation resulting from operational environments. Continuous health monitoring of materials would result in improved durability and safety of structures. The basic idea is to build a system similar to the human nervous system, with a network of sensors placed in critical areas where structural integrity must be maintained. Mathematical algorithms based on a “neural network” then can be trained to recognize patterns of electrical signals that represent damage, such as fiber strains or breakage and matrix cracks. In broadest terms, therefore, SHM comprises a distributed network of sensors that are deployed as integral parts of a structure and that are capable of collecting and sending information to an interrogator or electronic monitoring device.

3.2

State of the art of monitoring techniques

Testing is an area of utmost importance for composite structures playing also an important role in composites design and manufacturing and providing fundamental information important for their initial and final assessments. Smart Textiles for In Situ Monitoring of Composites. https://doi.org/10.1016/B978-0-08-102308-2.00003-6 Copyright © 2019 Elsevier Ltd. All rights reserved.

218

Smart Textiles for In Situ Monitoring of Composites

Currently, several norms such as ASTM, EN, ISO, etc., are available and used to determine intrinsic properties of composite materials. However, these norms define tests on relatively simple samples and under basic loading conditions. Consequently, it is fairly impossible to generalize the characterization of complex-shaped parts under realistic loading conditions. The following chapters focus on conventional and some nonconventional testing methods with their advantages and drawbacks.

3.2.1

Drapability assessment of composite preforms

When dealing with technical textiles, the drapability is defined as the ability of textile preforms to conform to the surface of molds. It has a technical meaning and is referred to mold surface, used during the consolidation of composites, and not to an ability of textile fabric to conform to a human body. There is only one drapability test defined by a standard norm, even if various devices able to assess the fabric drapability exist on the market and in research laboratories including similarly vision systems and software adapted to drapability measurement. The drapability test BS 5058 73, defined in 2012, similar to the AFNOR G 07e109 norm assesses the capability of the fabric to modify its shape under the sole effect of its weight, from flat to undulated, against an edge with a low curvature laying on a horizontal plane. This standard is fully satisfactory as far as it is applied to apparel and upholstery, or to nonwoven fabrics. It is however not well adapted to composites preforms. Another type of test called “jutting strip method” or “cantilever method” has been developed by Pierce in 1930 and included in the FAST system standing for Fabric Assurance by Simply Testing that is an objective method for the determination of fabric samples mechanical properties. The bending property, making a part of overall mechanical characterization may easily be correlated to a drapability. As existing testing methods have not been well adapted to the applications involving composites preforms and their drapability capabilities, new nonconventional methods have been introduced [1]. A preform fabric for thermoplastic or thermoset composites has adapted drapability if it is able to conform to a particular shape without making wrinkles that could generate an abnormal material distribution resulting in unnecessary thickening of the fabric at some points at in the worse case to structural weak areas. Accordingly, nonstandard tests have been developed for woven and knitted fabric preforms. The thermoforming process requires good knowledge of the fabric behavior as it has to adapt its shape to a mold containing different radius curvatures in different directions and under pressure. Therefore, it is necessary to asses the property of the fabric to let yarns slip under shear tensions while adapting to a shape. An important issue is also the evaluation of the resulting area density of the reinforcement fabric through local measurement particularly in the areas supposed to undergo important deformations. The effect of the pressure has to be taken into consideration as the force applied to the mold may produce an important stress. When the fabric is not well adapted to a complex mold shape, wrinkles may appear, which could under the

Structural health monitoring of composite structures

219

Figure 3.1 Pretensioned fabric during the test of conformation.

mold pressure damage the fibers up to their breakage. The wrinkles may be avoided by the fabric pretension, which diminishes the risk to have extremely dense areas at critical zones of the mold. Various thermoforming processes exist, such as heating of the preformed fabric before its insertion in the mold, or using the fabric containing reinforcement yarns together with thermoplastic fibers or coupling the reinforcement fabric with thermoplastic films. Anyway, regardless of the selected thermoforming process, the capability of the reinforcing fabric to adapt to a complex mold shape is highly critical for thermo consolidation process (Fig. 3.1).

3.2.2

Biaxial tensile testing of flat structures

Traditional dynamometer tests performed on composite structures are typically unidirectional. For anisotropic materials, within various applications, biaxial tensile tests along two orthogonal axes become necessary for flat structures. Several devices able to perform biaxial tests have been recently developed and put on the market. They are able to generate a large amount of data during the testing procedure giving reliable and repeatable information on the flat structures behavior. The base structure of the biaxial testing machine is constituted of steel profiles mounted in a prestress condition, supported by four columns putting the working horizontal plane at convenient height. Four actuators are placed in this plane in two opposite orthogonal directions, sized in function of the testing samples dimensions. The testing system is able to apply a wide range of stresses and strains on samples. The adapted and versatile clamping system is installed suiting various applications. The tensile actuators are controlled regarding their displacements. The consolidations standards for uniaxial testing are taken as a reference due to the lack of standardized procedures for tests in two dimensions. Displacement speeds may vary and are controlled too; therefore, the system allows the detection of anisotropies, load strength of the material, and its deformation limits. The forces applied to samples in two directions are detected by eight load cells mounted in couples at each clamp able also to detect each nonsymmetrical load due to bad positioning of the specimen or existing defects.

220

Smart Textiles for In Situ Monitoring of Composites

Specific procedures of measurement have been set up. The first one couples the vision system that is added to the measuring device. Reference marks are positioned on the top of the specimen and the vision system is able to detect their location in real time (trajectories) during the biaxial test. The maximal resolution of this method is 1 mm, which in some cases is not sufficient to determine strain values. The second method detects the overall deformation along all the axes using the clamps displacements and strain gauges installed on them. The biaxial-measuring device with the adapted software is flexible and many parameters are controllable by the user. Measuring performances are reliable and repeatable. Concerns of potential applications are that various and data analysis may give a lot of useful information of flat structures undergoing complex deformations and loads. Technical specifications of typical biaxial measuring device are given below: • • • • • • • • • • •

Number of independently controlled axes: 4; Number of load cells par axis: 2; Maximum load per axis: 12,000 N
1%; Strain rat minemax: 10e500 mm/min; Max displacement per axis: 50 mm; Sampling rate (measurement): 1 Hz; Clamp width: 200 mm; Min distance between clamps: 150 mm; Deformation control via cameraeframe rate: 1 Hz; Resolution: 0.3 mm; Sample size (max): 200  200 mm2.

3.2.3

Crash tests

The cost of crash tests may be extremely important as they require full scale prototypes, complex infrastructures, accelerometers, and high-speed cameras [2e4]. Crash tests are often used to test inexpensive and relatively small composite parts such as helmets, on the other side vehicle parts, and crash tests aiming at the selection and validation of new composite parts are made on small-scale prototypes to reduce testing costs. There are also nonconventional crash tests with the instrumentation able to deliver quickly a maximal number of information placed so as to avoid any risk of destruction. The impact energy is produced by accurately placing an impact object of suitable mass hardness and shape. The air friction is constant for equal object size and similar speed of impact. In some cases, the first crash tests are performed only with high speed cameras to better determine the placement of accelerometers and avoid their quick destruction.

Structural health monitoring of composite structures

221

Also, virtual crash tests are utilized massively in application that have been properly modeled and whose behavior may be simulated with minimal errors.

3.2.4

Split Hopkinson bar testdcharacterization under dynamic conditions

In many cases the stressestrain characteristic of materials is obtained at relatively slow strain rate. The proper design of composite parts, particularly structural ones, requires the knowledge of materials behavior at different strain rates including at very high speed such as in case of impact (crash or ballistic tests) and comprising shock waves. Therefore, the Split Hopkinson bar test is the most often used to determine the material behavior at high-speed strain rates. It allows the material testing at frequencies ranging from 100 Hz to 10 000 Hz. The main elements of a Split Hopkinson bar are three long aligned pressured bars: • • •

The striker bar (red); The incident bar (blue); The transmission bar (blue).

The tested sample is mounted and fixed between incident and transmission bars. The striker bar generates the stress pulse by the impact and therefore the tensile wave propagation through the device. The incident wave is partially reflected by the sample at the incident bar/sample interface and partially transmitted through the sample to the transmission bar. The dynamic strain response along the incident and the transmission bar is then compared to determine the constants of the material such as Young’s modulus or mechanical stress under dynamic conditions. It is also important to stress that the minimal frequency response of all components in the data acquisition system should be at least 100 KHz (Fig. 3.2). Nowadays, a number of Split Hokinson’s bar setups have been developed to test dynamically composite structures. They are able to generate reliable data on high strain rate behavior of composites particularly for lightweight polymeric composite parts used, more and more in transportation applications. All aforementioned characterization methods are however not adapted for in situ real time monitoring of composites that could be achieved by the utilization of fibrous sensors based on the concept of smart textiles developed in this manuscript. Pulse Striker bar

Incident bar

Figure 3.2 Scheme Split Hopkinson bar.

Sample

Transmission bar

222

3.3

Smart Textiles for In Situ Monitoring of Composites

Characterization of textile sensors before insertion in textile preforms

3.3.1

Textile sensors production according to percolation threshold final study

The results of electrical resistance measurements of developed textile sensors made from Glass Fiber (GF)/PP commingled yarn according to percolation threshold prestudy are presented in Table 3.1.

3.3.2

Resultsdviscosity determination of final conductive dispersion used

Viscosity of conductive dispersion used for finally textile sensors production recorded at 20 C and 100 s1 is n ¼ 375 mPa s. Table 3.1 Electrical resistance measurements of developed textile sensors according to prestudy Sample label

Sensor electrical resistance R2

GF/PP Sy-PP15-1

1000 kU

GF/PP Sy-PP15-2

930 kU

GF/PP Sy-PP15-3

970 kU

GF/PP Sy-PP15-4

1600 kU

GF/PP Sy-PP15-5

430 kU

Average

986 kU

Standard deviation

415 kU

GF/PP Sy-PP08-1

930 U

GF/PP Sy-PP08-2

950 U

GF/PP Sy-PP08-3

920 U

GF/PP Sy-PP08-4

1110 U

GF/PP Sy-PP8-5

1360 U

Average

1054 U

Standard deviation

188 U

GF/PP Sy-sp-1

1100 U

GF/PP Sy-sp-2

1760 U

GF/PP Sy-sp-3

960 U

Average

1273 U

Standard deviation

427 U

Structural health monitoring of composite structures

3.3.3

223

Results and discussiondtensile properties of yarns

Textile fibrous sensory yarns have been manufactured following definitions presented in previous chapters. Those sensors have been then characterized to determine, in the first time, the mechanical properties and their eventual modifications or alteration comparing to nonfunctionalized neat yarns. Tensile tests have been performed and presented in following figures. Ten base yarns made of: • • •

Comingled GFs and thermo fusible polypropylene fibers (PP), Comingled GFs and thermo fusible polyamide fibers (PA), and Glass Fibers

have served as the base (neat yarn) that has been later functionalized to become sensors. In all the cases GFs are used as reinforcement fibers and PP or PA fibers as thermo fusible ones. To determine the influence of diverse coated layers GF/PP yarns have been tested first without any coating (neat yarns), and then they have been coated with one layer of conductive compound and tested. The same neat yarns have also been coated with the protective layer that is used to protect the conductive layer and may modify yarns’ mechanical properties and finally the same neat yarns have been coated with two conductive layers giving better electromechanical properties and lower initial resistance and protected by a protective layer before testing. Tensile test of pure, one-layer conductive coated, one-layer protective coated, and two-layers conductive coated yarn between protective coating applied (sensor yarn) were performed with MTS tensile testing device by using speed of 150 mm/min and preload of 0.5 N with clamp distance of 150 mm according to ISO 2062:2009. The goal was to analyze the quality of pure GF/PP, GF/PA66, and GF yarns and related coated and finally developed sensor yarns from the mechanical point of view. All the yarns have been produced in 10 samples that have been tested in strictly identical conditions to determine the convergence of measuring results. The tensile results of neat and coated yarns are shown in Figs. 3.3e3.5. Additional tensile results are presented in Table 3.2. According to tensile results of yarn (Table 3.2), the neat GF yarn shows the highest stress at break, 769.64 GPa, whereas GF/PP yarn the lowest one, 585.21 MPa. Elongation at break are similar for all neat yarns, w3.20%. Improvement of the mechanical properties (Figs. 3.3e3.5) (higher stress at break and elongation at break) of one-layer conductive, one-layer protective coated yarns and sensor yarn is evident and is increasing with number of coating layers applied. GF/PP and GF/PA66 sensor yarns demonstrate similar stress at break, 895.84 and 908.36 GPa, whereas GF sensor yarn the higher one, 1063.91 GPa.

3.3.4

Results and discussiondelectromechanical properties of textile sensors

To carry out electromechanical tests (electrical resistance variation, DR/R0, during tensile testing) on textile sensors developed the MTS device was used. Samples were submitted

224

Smart Textiles for In Situ Monitoring of Composites

(a)

(b) 800 800

GF/PP-1

GF/PP-C-1

GF/PP-2

GF/PP-C-2

GF/PP-3 GF/PP-4 GF/PP-5

400

GF/PP-6 GF/PP-7 GF/PP-8

200

600 Stress, s (GPa)

Stress, s (GPa)

600

GF/PP-C-3 GF/PP-C-4 GF/PP-C-5

400

GF/PP-C-6 GF/PP-C-7

200

GF/PP-C-9

GF/PP-9

GF/PP-C-10

GF/PP-10 5.80%

5.30%

4.80%

4.20%

3.70%

3.30%

2.80%

2.30%

1.70%

1.20%

0.30%

4.20%

3.90%

3.50%

3.20%

2.80%

2.50%

2.20%

1.90%

1.50%

1.20%

0.80%

0.50%

0.20%

4.40%

Strain, e (%)

Strain, e (%)

(c)

(d)

1000

GF/PP-LApp96100-1 GF/PP-LApp96100-2

800

1200

GF/PP-Sy08-1 GF/PP-Sy08-2

1000

GF/PP-Sy08-3

GF/PP-LApp96100-3 GF/PP-LApp96100-4

600

GF/PP-LApp96100-5 GF/PP-LApp96100-6

400

GF/PP-LApp96100-7 GF/PP-LApp96100-8

200

Stress, s (GPa)

Stress, s (GPa)

0.80%

0

0

800

GF/PP-Sy08-4 GF/PP-Sy08-5

600

GF/PP-Sy08-6 GF/PP-Sy08-7

400

GF/PP-Sy08-8 200

GF/PP-Sy08-9

GF/PP-LApp96100-9

GF/PP-Sy08-10

0 0.10% 0.70% 1.20% 1.60% 2.10% 2.60% 3.20% 3.70% 4.10% 4.60% 5.10% 5.70% 6.20% 6.60% 7.10% 7.60%

GF/PP-LApp96100-10 0.20% 0.70% 1.20% 1.70% 2.20% 2.70% 3.20% 3.70% 4.20% 4.70% 5.20% 5.70% 6.20% 6.70% 7.20%

0

Strain, e (%)

Strain, e (%)

Figure 3.3 Tensile results of GF/PP yarn: (a) neat yarn, (b) one-layer conductive coated yarn, (c) one-layer protective coated yarn, (d) two-layers conductive coated yarn between protective coating applied (sensor yarn).

to quasi-static tensile loading at a constant test speed of 150 mm/min with a preload of 0.5 N. The distance between the clamps was 150 mm. The electrical resistance measurements have been done using a Keithley KUSB data acquisition digital I/O counter/timer and simple resistance box connected to a computer (QuickDAQ software). The results of electromechanical tests of textile sensors developed are presented in Figs. 3.7 and 3.8. Additional results of electromechanical tests of textile sensors are shown in Table 3.3. Textile sensors were prepared according to development steps previously described (usage of 842 tex GF/PP yarn, 957 tex GF/PA66 yarn, 831 tex GF yarn). For these tests, GF/PP and GF sensors with 5 þ 2 twists of copper twisted wires were produced (Fig. 3.6), whereas GF/PA66 sensors with 3 twists of copper twisted wires according to composite development prestudies. When the textile sensor is stretched, two phenomena occur, the first one is related to the geometrical properties of the fibrous sensor; the cross-sectional area is decreasing; the length is increasing; the sensor electrical resistance is increasing. The second phenomenon is related to the conductive layer made of polymer complex PEDOT-compl-PSS and its electrical properties. As the

Structural health monitoring of composite structures

225

(a)

(b)

800

1000

GF/PA66-1

GF/PA-C-1

GF/PA66-2

GF/PA-C-2

GF/PA66-3 GF/PA66-4 GF/PA66-5

400

GF/PA66-6 GF/PA66-7

200

Stress, s (GPa)

Stress, s (GPa)

600

800

GF/PA66-8

GF/PA-C-5 600

GF/PA-C-6 GF/PA-C-7

400

GF/PA-C-8 200

GF/PA-C-9

GF/PA66-9

GF/PA66-LApp96100-2 GF/PA66-LApp96100-3 GF/PA66-LApp96100-4

600

GF/PA66-LApp96100-5

400

GF/PA66-LApp96100-6

1200 GF/PA66-Sy08-1 1000 Stress, s (GPa)

GF/PA66-LApp96100-1

800

0.10% 0.50% 0.80% 1.20% 1.50% 1.80% 2.10% 2.50% 2.80% 3.20% 3.50% 3.80% 4.20% 4.50% 4.80% 5.10% 5.50% 5.80% 6.20% 6.50%

(d)

1000

Stress, s (GPa)

0

Strain, e (%)

Strain, e (%)

(c)

GF/PA-C-10

GF/PA66-10 0.00% 0.20% 0.60% 0.90% 1.30% 1.60% 1.90% 2.20% 2.60% 2.90% 3.30% 3.60% 3.90% 4.20%

0

GF/PA66-Sy08-2 GF/PA66-Sy08-5

800

GF/PA66-Sy08-6 600

GF/PA66-Sy08-7

400

GF/PA66-Sy08-8

GF/PA66-LApp96100-7

200

GF/PA66-LApp96100-9 GF/PA66-LApp96100-10

GF/PA66-Sy08-10 0.20% 0.70% 1.20% 1.70% 2.20% 2.70% 3.20% 3.70% 4.20% 4.70% 5.20% 5.70% 6.20% 6.70% 7.20% 7.70% 8.20%

0

0.20% 0.70% 1.20% 1.70% 2.20% 2.70% 3.20% 3.70% 4.20% 4.70% 5.20% 5.70% 6.20% 6.70%

0

GF/PA66-Sy08-9

200

Strain, e (%)

Strain, e (%)

Figure 3.4 Tensile results of GF/PA66 yarn: (a) neat yarn, (b) one-layer conductive coated yarn, (c) one-layer protective coated yarn, (d) two-layers conductive coated yarn between protective coating applied (sensor yarn).

concentration of the PEDOT-compl-PSS is defined to be at the percolation threshold, the electrical conductivity is strongly decreasing when this layer is stretched, because a number of conductive paths inside the conductive material are broken. Therefore, this electrical conductivity is decreasing, or the electrical resistivity is increasing contributing to the increasing of the sensor electrical resistance together with “geometrical” increase of its resistance. Primarily, GF/PP sensors with 8% PEDOT-compl-PSS FET-LApp96100 conductive drops (cd) and silver drops (sp) applied after copper wire insertion (5 þ 2 twists of copper twisted wires) were compared. According to results, GF/PP sensors produced with 8% PEDOT-compl-PSS FET-LApp96100 conductive drops display slightly higher stress at break 912.68 MPa, and slightly lower elongation at break, 6.57%, but notable higher gauge factor, 3.9373, than GF/PP sensors made with silver drops with stress at break of 722.15 MPa, elongation at break 6.93% and gauge factor 2.8154. Therefore, new textile sensors were made with silver drops addition in several series for their integration during weaving of 2D/3D fabrics. Besides, more secure connection between copper wire and sensor yarn was realized with silver drops. The stress versus elongation curve has similar path as the electrical resistance variation

226

Smart Textiles for In Situ Monitoring of Composites

(a)

(b) 1200

1000

GF-4

GF-C-1

800

GF-C-2

GF-C-4 400 GF-C-5

200

Strain, e (%)

4.70%

4.20%

3.70%

3.20%

2.70%

2.20%

1.70%

0

3.50%

3.10%

2.80%

2.40%

2.10%

1.80%

1.40%

1.10%

0.80%

0.50%

0.10%

0

1.20%

200

GF-5

Strain, e (%)

(c)

(d)

1000

1600 GF-Sy08-1

GF-LApp96100-1

800

1200 GF-LApp96100-2

600 GF-LApp96100-3 400

GF-LApp96100-4

200

Stress,s (GPa)

Stress,s (GPa)

GF-C-3

600

0.70%

GF-3

400

1000

0.20%

Stress,s (GPa)

GF-2

600

Stress,s (GPa)

GF-1

800

GF-Sy08-2 800

GF-Sy08-3 GF-Sy08-4

400

GF-LApp96100-5 GF-Sy08-5

0.10% 0.40% 0.70% 1.00% 1.40% 1.70% 2.10% 2.40% 2.70% 3.00% 3.40% 3.70%

0.20% 0.70% 1.20% 1.70% 2.20% 2.70% 3.20% 3.70% 4.20% 4.70% 5.20% 5.70% 6.20% 6.70% 7.20%

0 0

Strain, e (%)

Strain, e (%)

Figure 3.5 Tensile results of GF yarn: (a) neat yarn, (b) one-layer conductive coated yarn, (c) one-layer protective coated yarn, (d) two-layers conductive coated yarn between protective coating applied (sensor yarn).

(DR/R0) versus elongation curve. In general, the coatings obtained on neat yarn are homogenous and the results for different coated yarns do not vary notable in their response to tensile loading. In comparison with GF/PP sensors made with silver drops, GF sensors show lower electrical resistance, w850 U, and elongation at break, 5.45%, higher stress at break, 1318.25 GPa, and higher gauge factor 3.5939. GF/PA66 sensors show higher elongation at break, 7.20% and lower gauge factor, 1.3412, than other textile sensors although number of copper wire turns applied during their production has to be taken into account. In addition, electrical resistance of GF/PA66 sensor is slightly lower compared to other textile sensors, with lower dispersion of results. Higher quality of production of neat GF/PA66 yarn in comparison with GF/PP gives contribution to obtained results as well. Greater deviations in electrical resistance variations could not be observed at higher elongation what confirms uniformity achieved of coating neat yarn. Nonuniform

Structural health monitoring of composite structures

227

Table 3.2 Tensile results of neat and coated yarns Sample label GF/PP

Average Standard deviation

GF/PP-LApp96100

Average Standard deviation

GF/PP-C

Average Standard deviation

GF/PP-Sy08

Average Standard deviation

GF/PA66

Average Standard deviation

GF/PA66-LApp96100

Average Standard deviation

GF/PA66-C

Average Standard deviation

GF/PA66-Sy08

Average Standard deviation

GF

Average Standard deviation

GF-LApp96100

Average Standard deviation

GF-C

Average Standard deviation

GF-Sy08

Average Standard deviation

Force (N)

Stress at break (GPa)

Strain (%)

183.85

585.21

3.28

26.95

85.80

0.24

193.34

615.41

4.53

34.12

108.62

1.00

190.29

605.71

4.92

27.61

87.88

0.74

281.44

895.84

6.33

23.99

76.35

0.73

201.29

640.71

3.20

14.67

46.71

0.32

250.53

797.47

5.71

22.74

72.40

0.46

265.13

843.94

4.91

44.50

141.63

0.53

285.37

908.36

6.75

25.85

82.28

0.76

241.79

769.64

2.84

13.79

43.91

0.40

268.85

855.79

3.30

28.11

89.49

0.34

296.52

943.86

3.90

34.50

109.83

0.62

334.24

1063.91

5.24

38.52

122.62

0.73

coatings tend to crack where there is a thin deposited layer. This causes a marked increase in electrical resistance whenever a conductive track breaks up. Finally, electrical resistance measurements during tensile loading of these textile sensors show promising results for their usage inside composite structural parts for in situ SHM during mechanical (tensile) loading.

Figure 3.6 Textile sensors: (a) GF/PP sensor, (b) GF sensor.

(a)

Electromehanical test of GF/PP sensors tensile results

350

GF/PP Sy-cd-1

GF/PP Sy-cd-2

GF/PP Sy-cd-3

300

Force (N)

250

200

150

100

0

1.0% 1.2% 1.4% 1.5% 1.7% 1.9% 2.1% 2.2% 2.4% 2.5% 2.7% 2.9% 3.0% 3.2% 3.4% 3.5% 3.7% 3.9% 4.0% 4.2% 4.4% 4.6% 4.7% 4.9% 5.0% 5.2% 5.4% 5.5% 5.7% 5.9% 6.0% 6.2% 6.4% 6.5% 6.7% 6.9% 7.1% 7.2% 7.4% 7.5%

50

Elongation (%)

(b)

Electromehanical test of GF/PP sensors DR/R (%) versus DL/L (%)

30%

GF/PP Sy-cd-1

GF/PP Sy-cd-2

GF/PP Sy-cd-3

25%

y = 3.5767x – 0.0162 2

R = 0.9455

DR/R (%)

20%

y = 4.7615x – 0.023 2 R = 0.8991

15%

y = 3.4738x – 0.0693 2

R = 0.9486 10%

5%

0% 0% –5%

1%

2%

3%

4%

5%

6%

7%

8%

DL/L (%)

Figure 3.7 Electromechanical tests of textile sensors developed: (a and b) GF/PP sensors (GF/PP Sy-cd-1: blue, GF/PP Sy-cd-2: red and GF/PP Sy-cd-3: green) with conductive 8% PEDOT-compl-PSS CPP-Lapp96100 drops added, (c and d) GF/PP sensors (GF/PPSy08sp07-1: blue, GF/PP-Sy08sp07-2: red and GF/PP-Sy08sp07-3: green) with silver points added after copper wires insertion.

Structural health monitoring of composite structures

229

Electrochemical test of GF/PP sensors tensile results

(c) 300

GF/PP Sy-sp-1

GF/PP Sy-sp-2

GF/PP Sy-sp-3

250

Force (N)

200

150

100

0

1.70% 2.70% 3.70% 4.80% 5.70% 6.70% 7.70% 8.70% 9.70% 10.70% 11.70% 12.70% 13.70% 14.70% 15.70% 16.70% 17.80% 18.70% 19.70% 20.70% 21.70% 22.80% 23.70% 24.70% 25.70% 26.70% 27.70% 28.70% 29.70% 30.80% 31.70% 32.70% 33.70% 34.80% 35.80% 36.70% 37.80% 38.80% 39.80% 40.70% 41.70%

50

Elongation (%)

Electromehanical test of GF/PP sensors DR/R (%) versus DL/L (%)

(d) 30%

GF/PP Sy-sp-1

GF/PP Sy-sp-2

GF/PP Sy-sp-3

25%

y = 3.8872x + 0.0341 2

R = 0.9179

20%

y = 2.5027x – 0.0518

DR/R (%)

2

R = 0.9603

15%

10%

y = 2.0564x – 0.0503 2 R = 0.8864

5%

0% 0% –5%

1%

2%

3%

4%

5%

6%

7%

8%

9%

DL/L (%)

Figure 3.7 cont’d.

3.4 3.4.1

Characterization of textile sensors after insertion in textile preforms Textile sensors integration during weaving of 2D fabric, consolidation pretest analysis

Preliminary experiences have been done to better apprehend the integration of developed textile sensors to textile structures used for the production of thermoplastic composites. Three cases of GF/PP sensors integration during weaving of 2D fabric and 2D textile preforms preparation for consolidation step are shown in Fig. 3.9. From the

230

Smart Textiles for In Situ Monitoring of Composites Electromehanical test of GF sensors tensile results

(a) 600

GF Sy-sp-1

GF Sy-sp-2

500

Force (N)

400

300

200

13.20% 13.70%

12.20% 12.70%

11.20% 11.70%

10.20% 10.70%

9.20% 9.70%

8.20% 8.70%

7.20% 7.70%

6.20% 6.70%

5.20% 5.70%

4.20% 4.70%

3.20% 3.70%

2.20% 2.70%

1.20% 1.70%

0

0.20% 0.70%

100

Elongation (%)

Electromehanical test of GF sensors ΔR/R (%) versus ΔL/L (%)

(b) 20%

GF Sy-sp-1

ΔR/R (%)

15%

GF Sy-sp-2

y = 3.9768x – 0.0114 2 R = 0.9413

10%

y = 3.211x – 0.0332 2

R = 0.9007 5%

0% 0%

–5%

1%

2%

3%

4%

5%

6%

7%

ΔL/L (%)

Figure 3.8 Electromechanical tests of textile sensors developed: (a, b) GF sensors (GF-Sy08sp03-1: yellow (a) and red (b), GF-Sy08sp03-2 green), (c, d) GF/PA66 sensors (GF/PA66-Sy08sp07-1: blue, GF/PA66-Sy08sp07-2: red and GF/PA66-Sy08sp07-3: green) with silver points added after copper wires insertion.

GF/PP_Sy series four samples of sensors with 5þ2 copper wires turns for connection (case I) were taken for textile sensors integration during weaving of 2D fabrics (two textile sensors per two coupons). Furthermore, four textile sensors were made only with 3 copper wires turns (case II) and integrated during a weaving step (two textile sensors per two coupons), while four textile sensors were integrated without copper wires (case III). Copper wires have been inserted after 2D fabrics thermal consolidation to connect those sensors. Three cases of textile sensors integrated in 2D fabrics after two different thermal consolidation conditions are shown in Fig. 3.10. Related electrical resistance measurements, when available, of GF/PP sensors are presented in Table 3.4.

Structural health monitoring of composite structures

231

Electrochemical test of GF/PA66 sensors tensile results

(c) 350

GF/PA66 Sy-sp-1

GF/PA66 Sy-sp-2

GF/PA66 Sy-sp-3

300

Force (N)

250 200 150 100

0

0.10% 0.70% 1.20% 1.60% 2.10% 2.60% 3.20% 3.70% 4.10% 4.60% 5.10% 5.70% 6.20% 6.60% 7.10% 7.60% 8.20% 8.70% 9.10% 9.60% 10.10% 10.70% 11.20% 11.70% 12.10% 12.60% 13.20% 13.70% 14.20% 14.60% 15.10% 15.70% 16.20% 16.70% 17.10% 17.60% 18.10%

50

Elongation (%)

(d)

Electromehanical test of GF/PA66 sensors ΔR/R (%) versus ΔL/L (%)

14.00%

GF/PA66 Sy-sp-1 12.00%

GF/PA66 Sy-sp-2

GF/PA66 Sy-sp-3

y = 1.4665x + 0.0092 2

R = 0.7816

10.00%

y = 1.4136x – 0.0213 2 R = 0.9397

ΔR/R (%)

8.00%

6.00%

y = 1.143x – 0.0089 2

R = 0.8756

4.00% 2.00% 0.00% 0.00% –2.00%

0.00%

4.00%

6.00%

8.00%

10.00%

12.00%

ΔL/L (%)

Figure 3.8 cont’d

These measurements show that the sensors’ initial resistance increases significantly after the thermo consolidation process, but it remains still acceptable for in situ measurements of composites’ deformations, delamination, cracks, and other possible internal damages. In following chapters different previously developed fibrous sensors (strain gauges) have been integrated to composite preforms, then they have been thermo consolidated and finally electromechanical properties of sensors have been determined to test their sensitivity and capacity to measure composites deformations in situ in real time. According to sensors electrical resistance after consolidation step of 2D fabrics (Table 3.5) under pressure of 2e3 MPa and temperature of 185 C, case I shows the most acceptable results. Taking into account that higher pressure has to be used, case II gives positive sensors electrical resistance responses under pressure of 4e5 MPa and temperature of 185  C during 5 min. This case was studied deeply through larger number of coupons.

Table 3.3 Tensile results and gauge (“k”) factor of developed textile sensors 232

Sample label

Sensor electrical resistance before test (U)

Sensor electrical resistance during test (U)

Force at break (N)

Elongation at break (%)

Stress at break (Pa)

Gauge factor

930

2560

264.24

7.40

841.10

3.4738

GF/PP-Sy08cd07-2

950

2450

305.15

6.10

971.32

4.7615

GF/PP-Sy08cd07-3

920

2510

290.79

6.20

925.63

3.5767

Average

933

2507

286.73

6.57

912.68

3.9373

Standard deviation

15

55

20.76

0.72

66.07

0.7156

GF/PP-Sy08sp07-1

880

980

269.03

7.90

856.34

2.5027

GF/PP-Sy08sp07-2

700

800

140.80

7.20

448.17

2.0564

GF/PP-Sy08sp07-3

1540

1460

270.79

5.70

861.94

3.8872

Average

1040

1080

226.87

6.93

722.15

2.8154

442

341

74.55

1.12

237.29

0.9546

GF-Sy08sp03-1

1240

1200

337.69

4.50

1074.88

3.9768

GF-Sy08sp03-2

500

500

490.60

6.40

1561.63

3.2110

Average

870

850

414.14

5.45

1318.25

3.5939

Standard deviation

523

495

108.13

1.34

344.188

0.5415

GF/PA66-Sy08sp07-1

990

1150

295.41

7.30

940.33

1.1434

GF/PA66-Sy08sp07-2

810

930

271.85

8.00

865.311

1.4136

GF/PA66-Sy08sp07-3

690

820

214.93

6.30

684.13

1.4665

Average

830

967

260.73

7.20

829.93

1.3412

Standard deviation

151

168

41.38

0.85

131.71

0.1733

Standard deviation

Smart Textiles for In Situ Monitoring of Composites

GF/PP-Sy08cd07-1

Figure 3.9 Textile sensors integration during weaving of 2D fabric (a) and 2D textile preforms preparation for thermal consolidation: case I and II (b) and case III (c).

Figure 3.10 Textile reinforced 2D thermoplastic composites with integrated textile sensors: (a) 2D fabric consolidated under pressure of 2e3 MPadcase I, (b) 2D fabric consolidated under pressure of 4e5 MPadcase II, (c) 2D fabric consolidated under pressure of 2e3 MPadcase III, (d) 2D fabric consolidated under pressure of 4e5 MPadcase III.

Sample label

Sensor electrical resistance R2 (U)

I

GF/PP_Sy-sp07-1

570

600

11.5 kU

e

GF/PP_Sy-sp07-2

770

840

36 kU

e

GF/PP_Sy-sp07-3

720

1080

e

10.1 MU

GF/PP_Sy-sp07-4

880

930

e

10 MU

Average

610.00

862.50

23.75 kU

10.05 MU

Standard deviation

175.31

201.06

17.32 kU

0.07 MU

GF/PP_Sy-sp03-1

740

850

12.9 kU

e

GF/PP_Sy-sp03-2

570

620

42.1 kU

e

GF/PP_Sy-sp03-3

540

640

e

160 kU

GF/PP_Sy-sp03-4

780

450

e

31.1 kU

Average

657.50

640.00

27.50 kU

95.55 kU

Standard deviation

120.10

163.91

20.65 kU

91.15 kU

GF/PP_Sy-sp01-1

e

e

75 kU

e

GF/PP_Sy-sp01-2

e

e

9.5 kU

e

GF/PP_Sy-sp01-3

e

e

e

90 kU

GF/PP_Sy-sp01-4

e

e

e

550 kU

Average

e

e

42.25 kU

320.00 kU

Standard deviation

e

e

46.32 kU

325.27 kU

II

III

Sensor electrical resistance after consolidation of 2D fabric R2 1858C, 2e3 MPa

1858C, 4e5 MPa

Smart Textiles for In Situ Monitoring of Composites

Case

Sensor electrical resistance after integration in 2D fabric R2 (U)

234

Table 3.4 Electrical resistance measurements of developed GF/PP sensors for three cases of integration in 2D fabric

Structural health monitoring of composite structures

235

Table 3.5 GF/PP sensors electrical resistance after production and integration in 2D fabric Sample label

Sensor electrical resistance after production R2 (U)

Sensor electrical resistance after insertion in 2D fabric R2 (U)

GF/PP-Sy08sp03-1

490

550

GF/PP-Sy08sp03-2

500

550

GF/PP-Sy08sp03-3

370

410

GF/PP-Sy08sp03-4

350

390

GF/PP-Sy08sp03-5

440

510

GF/PP-Sy08sp03-6

390

450

Average

423.33

476.67

Standard deviation

63.14

70.05

3.4.2

ResultsdGF/PP composites with integrated GF/PP sensors

In this section GF/PP sensors have been integrated to GF/PP composites and their electrical resistances after the production and integration in 2D fabric, but before a thermo consolidation are presented in Table 3.6. GF/PP composites with integrated GF/PP sensors are presented before electromechanical tests in Fig. 3.11 and after tests in Fig. 3.12. Typical stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) or time (t) curves are presented in Fig. 3.13. Tensile results of GF/PP composites with integrated GF/PP sensors are presented in Table 3.6. GF/PP sensors electrical resistance values are compared after 2D fabric thermal consolidation and during electromechanical test of developed composites (Fig. 3.14). Composites regular breakage can be seen after electromechanical tests between clamps for the first sample shown in Fig. 3.16, which is not in case for the last two samples. Gauge factor was determined from a polynomial equation of degree 2. The first composite sample shows similar electrical resistance-elongation curves of GF/PP sensors, whereas other two tested samples display different curves. Earlier fractures inside the composite structure were detected for other composites. One GF/PP sensor breakage occurred at the beginning of the electromechanical test of the last sample. In almost all cases, the textile sensors are able to follow the tensile loading and damage detection in the composite. Maximum peaks of textile sensors curves appeared before the maximum stresses achieved for the composite specimens. Fractures were accompanied by cracking sounds and slight changes in the electrical resistance variations versus elongation curves. Composite specimens tensile loading until break were performed approximately after 152 s, whereas notable changes in textile sensors outputs occurred after 85 s (2.5% elongation).

236

Table 3.6 Tensile results of GF/PP composites with integrated GF/PP sensors GF/PP composites with integrated GF/PP sensors

GF/PP sensors Time (s)

Force at break (N)

Stress at break (MPa)

Elongation at break (%)

153.6

27022.07

217.33

4.45

151.6

24632.21

206.99

4.40

152.6

25629.70

207.74

4.42

Average

152.6

25761.33

210.69

4.23

Standard deviation

1.00

1200.35

5.76

0.026

Sample label GF/PPcmp-s45P185T-GF/PPSy08sp03-1

GF/PP-Sy08sp03-1

GF/PPcmp-s45P185T-GF/PPSy08sp03-2

GF/PP-Sy08sp03-3

GF/PPcmp-s45P185T-GF/PPSy08sp03-3

GF/PP-Sy08sp03-5

GF/PP-Sy08sp03-2

GF/PP-Sy08sp03-6

Smart Textiles for In Situ Monitoring of Composites

GF/PP-Sy08sp03-4

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-1

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-2

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-3

Figure 3.11 GF/PP composites with integrated GF/PP sensors before mechanical test: front side.

(a)

(b)

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-1

(c)

(d)

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-2

(e)

(f)

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-3

Figure 3.12 GF/PP composites with integrated GF/PP sensors after mechanical test: (a, c, e) front side and (b, d, f) back side.

238

Smart Textiles for In Situ Monitoring of Composites

Stress, s (MPa)

200

70.00%

Stress (MPa) GF/PP-Sy08sp03-1 GF/PP-Sy08sp03-2

60.00%

2 y = 233.35x – 0.3484x – 0.0015 2 R = 0.9967

150 2 y = 2E–07x – 9E–05x + 0.0056 R2 = 0.9

100

50.00% 40.00% 30.00% 20.00%

50 10.00%

0.00% 0.17% 0.35% 0.52% 0.70% 0.87% 1.04% 1.22% 1.39% 1.57% 1.74% 1.91% 2.09% 2.26% 2.44% 2.61% 2.78% 2.96% 3.13% 3.31% 3.48% 3.65% 3.83% 4.00% 4.17% 4.35%

0

Electrical resistance variation, DR/R0 (%)

(a) 250

0.00%

Elongation, DL/L0 (%)

(b) 70.00% Stress (MPa) GF/PP-Sy08sp03-1 200

60.00%

GF/PP-Sy08sp03-2

Stress, s (MPa)

50.00% 150

40.00% 30.00%

100

20.00% 50

0

0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150

10.00%

Electrical resistance variation, DR/R0 (%)

250

0.00%

Time, t (s)

Figure 3.13 Electromechanical test of GF/PP composites with integrated GF/PP sensors: (a, c, e) stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) curves, (b, d, f) stress (s) and electrical resistance variation (DR/R0) versus time (t) curves.

Stress at break (Table 3.7) of composites developed is 210.69 MPa, whereas elongation at break is 4.42% with low dispersion of results. Electrical resistance of these sensors after insertion in 2D fabric (Table 3.6) was w477 U, slightly higher compared to sensor electrical resistance before their integration during weaving of 2D fabric, 423 U. Sensor electrical resistance after 2D GF/PP textile preforms with integrated GF/PP sensors thermal consolidation and at the beginning of electromechanical tests (Fig. 3.15) are similar and in range between 15 and 240 kU depending on the tested sample. Higher variation between two sensors, GF/PP-Sy08sp03e1 and GF/PP-Sy08sp03-2, integrated in composite,

Structural health monitoring of composite structures

Stress (MPa) GF/PP-Sy08sp03-3 GF/PP-Sy08sp03-4

90.00% 80.00%

200 2

Stress, s (MPa)

100.00%

y = –8E–08x + 0.0005x – 0.0432 R2 = 0.826

150

y = 536.69x2 – 8.544x + 0.0231 R2 = 0.9309 100

70.00% 60.00% 50.00% 40.00% 30.00%

50

20.00%

Electrical resistance variation, DR/R0 (%)

(c) 250

239

0

0.00% 0.16% 0.31% 0.46% 0.62% 0.77% 0.92% 1.08% 1.23% 1.39% 1.54% 1.69% 1.85% 2.00% 2.15% 2.31% 2.46% 2.61% 2.77% 2.92% 3.07% 3.23% 3.38% 3.54% 3.69% 3.84% 4.00% 4.15% 4.30%

10.00% 0.00%

Elongation, DL/L0 (%) 250

100.00%

Stress (MPa) GF/PP-Sy08sp03-3

90.00%

GF/PP-Sy08sp03-4 80.00%

200

Stress, s (MPa)

70.00% 150

60.00% 50.00%

100

40.00% 30.00%

50

20.00%

Electrical resistance variation, DR/R0 (%)

(d)

0 6 11 17 23 29 34 40 46 51 57 63 68 74 80 86 91 97 103 108 114 120 125 131 137 143 148

10.00% 0

0.00%

Time, t (s)

Figure 3.13 cont’d.

GF/PPcmp-s45P185T-GF/PP-Sy08sp03-1, can be seen only for the first composite sample. Textile sensors with low electrical resistance were detected for second composite sample (GF/PPcmp-s45P185T-GF/PP-Sy08sp03-2). For this composite, textile sensors plot of electrical resistance variation (DR/R0) versus elongation (DL/L0) were not totally regular with several interruptions in the second part of the plots due to beginning of the delamination and cracks inside the composite. Fig. 3.16 demonstrates irregular composite line breakage after test performed as well.

Smart Textiles for In Situ Monitoring of Composites

(e)

250

Stress (MPa) GF/PP-Sy08sp03-5 GF/PP-Sy08sp03-6

140.00% 120.00%

200

Stress, σ (MPa)

100.00% 150

80.00% y = 18.583x + 0.7634 R2 = 0.5882

100

60.00% 40.00%

50 20.00% y = 263.47x2 + 2.174x + 0.0069 R2 = 0.9828

0.00%

0.00% 0.23% 0.46% 0.69% 0.92% 1.15% 1.37% 1.60% 1.83% 2.06% 2.29% 2.52% 2.75% 2.98% 3.21% 3.44% 3.66% 3.89% 4.12% 4.35% 4.51% 4.74%

0

Electrical resistance variation, ΔR/R0 (%)

240

Elongation, ΔL/L0 (%)

(f)

200.00

140.00%

Stress (MPa) L-23

120.00%

L-24

Stress, σ (MPa)

100.00% 150.00

80.00% 60.00%

100.00

40.00% 50.00 20.00%

0.0 5.5 11.0 16.5 22.0 27.5 33.0 38.5 44.0 49.5 55.0 60.5 66.0 71.5 77.0 82.5 88.0 93.5 99.0 104.5 110.0 115.5 115.5 126.5 132.0 137.5 143.0 148.5 154.0 157.0 162.5 168.0

0.00

Electrical resistance variation, ΔR/R0 (%)

250.00

0.00%

Time, t (s)

Figure 3.13 cont’d.

3.4.3

GF/PP composites with integrated GF sensors

GF sensors electrical resistances after their production and integration in 2D GF/PP fabric, but before the thermo consolidation, are presented in Table 3.7. GF/PP composites after the thermo consolidation with integrated GF sensors are presented before electromechanical tests in Fig. 3.15, while after electromechanical tests in Fig. 3.16. Typical stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) or time (t) curves can be seen in Fig. 3.17. Tensile results of GF/PP composites with integrated GF sensors are presented in Table 3.8.

Structural health monitoring of composite structures

241

240.00

250

Sensor electrical resistance after consolidation (kW) Sensor electrical resistance in elemechanical test (kW) Average Standard deviation

Sensor electrical resistance (kW)

190.00 215.00 200

150

117.00 130.00 123.50 92.00 89.50

100

50

87.00

26.00 29.50 27.75

35.36 15.70 18.70 17.20 2.12

2.47

19.50 18.55 17.60 1.34

9.19

3.54

0 GF/PP-Sy08sp03-1 GF/PP-Sy08sp03-2 GF/PP-Sy08sp03-3 GF/PP-Sy08sp03-4 GF/PP-Sy08sp03-5 GF/PP-Sy08sp03-6 Sample label

Figure 3.14 GF/PP sensors electrical resistance after 2D GF/PP fabric consolidation and during electromechanical test. Table 3.7 GF sensors electrical resistance after production and integration in 2D fabric Sensor electrical resistance after insertion in 2D fabric R2 (U)

Sample label

Sensor electrical resistance after production R2 (U)

GF-Sy08sp03-1

790

920

GF-Sy08sp03-2

730

870

GF-Sy08sp03-3

550

700

GF-Sy08sp03-4

580

740

GF-Sy08sp03-5

330

380

GF-Sy08sp03-6

880

1000

Average

643

768

Standard deviation

198

221

GF sensor electrical resistance values are compared after 2D fabric thermal consolidation and during electromechanical test performed (Fig. 3.18). GF/PP composites with integrated GF sensors do not display totally regular line breakage between clamps after electromechanical tests (only the last sample) (Fig. 3.16). Gauge factor was determined from a polynomial equation of degree 2. The first composite sample shows similar electrical resistance-elongation curves of GF sensors, what is not in case of the last two composites, where for each specimen breakage of one GF sensor occurred at the beginning of the electromechanical test. Maximum peaks of textile sensors curves appeared before or at the maximum stresses achieved for the composite specimens. Fractures were accompanied by cracking sounds as well. Composite specimens tensile loading until break were performed after

GF/PPcmp-s45P185T-GF-Sy08sp03-1

GF/PPcmp-s45P185T-GF-Sy08sp03-2

GF/PPcmp-s45P185T-GF-Sy08sp03-3

Figure 3.15 GF/PP composites with integrated GF sensors before mechanical test: front side.

(a)

(b)

GF/PPcmp-s45P185T-GF-Sy08sp03-1

(c)

(d)

GF/PPcmp-s45P185T-GF-Sy08sp03-2

(e)

(f)

GF/PPcmp-s45P185T-GF-Sy08sp03-3

Figure 3.16 GF/PP composites with integrated GF sensors after mechanical test: (a, c, e) front side and (b, d, f) back side.

Structural health monitoring of composite structures

243

(a) 250

100.00%

y = 995.91x – 11.225x – 0.0448 R2 = 0.9888

60.00% 100 2 y = 840.95x – 6.1284x + 0.0196 R2 = 0.9998

50

40.00%

20.00%

0.05% 0.22% 0.40% 0.58% 0.75% 0.93% 1.11% 1.29% 1.46% 1.64% 1.82% 1.99% 2.17% 2.35% 2.52% 2.70% 2.88% 3.05% 3.23% 3.41% 3.58% 3.76% 3.94% 4.11% 4.29% 4.47% 4.64% 4.82% 4.98%

0

0.00%

Elongation, DL/L0 (%)

(b) 250

120.00%

Stress (MPa) GF-Sy08sp03-1 GF-Sy08sp03-2

100.00%

200

Stress, s (MPa)

80.00%

150

80.00% 150 60.00% 100 40.00% 50

0.00 5.90 11.80 17.70 23.60 29.50 35.40 41.30 47.20 53.10 59.00 64.90 70.80 76.70 82.60 88.50 94.40 100.30 106.20 112.10 118.00 123.90 129.80 135.70 141.60 147.50 153.40 159.30 165.20 170.13

0

20.00%

Electrical resistance variation, DR/R0 (%)

Stress, s (MPa)

2

Electrical resistance variation, DR/R0 (%)

200

120.00%

Stress (MPa) GF-Sy08sp03-1 GF-Sy08sp03-2

0.00%

Time, t (s)

Figure 3.17 Electromechanical test of GF/PP composites with integrated GF sensors: (a, c, e) stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) curves, (b, d, f) stress (s) and electrical resistance variation (DR/R0) versus time (t) curves.

177 s, whereas notable changes in textile sensors curves occurred after 85 s (2.5% elongation). Stress at break of composites developed (Table 3.8) is 257 MPa, whereas elongation at break is 5,16% with low dispersion. Electrical resistance of these sensors after insertion in 2D fabric was approximately 768 U with acceptable standard deviation, slightly higher compared to sensor electrical resistance before their integration during weaving of 2D fabric, 643 U. Sensor electrical resistance after 2D GF/PP textile preforms with integrated GF sensors after thermal consolidation and at the beginning of electromechanical tests (Fig. 3.15) are similar and in range between 3.5 and 56 kU depending on the tested sample. Higher variation between two sensors, GF-Sy08sp03-5 and GF-Sy08sp03-5, integrated in composite, GF/PPcmp-s45P185T-GF-Sy08sp03-3, can be seen only for the third composite sample.

244

Smart Textiles for In Situ Monitoring of Composites

Stress, s (MPa)

250

200

160.00%

Stress (MPa) GF-Sy08sp03-3 GF-Sy08sp03-4

120.00%

y = 380.59x2 + 7.8307x – 0.0464 R2 = 0.9978

80.00% 40.00%

150 0.00% 100

–40.00%

50

0.03% 0.24% 0.44% 0.65% 0.86% 1.07% 1.28% 1.49% 1.70% 1.91% 2.11% 2.32% 2.53% 2.74% 2.95% 3.16% 3.37% 3.58% 3.78% 3.99% 4.20% 4.41% 4.62% 4.83% 5.04% 5.16% 5.36% 5.56% 5.77% 5.98%

0

–80.00%

Electrical resistance variation, DR/R0 (%)

(c) 300

–120.00%

Elongation, DL/L0 (%) 160.00%

Stress (MPa) GF-Sy08sp03-3 GF-Sy08sp03-4

250

120.00%

Stress, s (MPa)

80.00% 200 40.00% 150 0.00% 100

–40.00%

50

0.00 7.40 14.80 22.20 29.60 37.00 44.40 51.80 59.20 66.60 74.00 81.40 88.80 96.20 103.60 111.00 118.40 125.80 133.20 140.60 148.00 155.40 162.80 170.20 176.94 181.65 189.05 196.45 203.85

0

–80.00%

Electrical resistance variation, DR/R0 (%)

(d) 300

–120.00%

Time, t (s)

Figure 3.17 cont’d.

Textile sensors with low electrical resistance were detected for the second composite sample (GF/PPcmp-s45P185T-GF-Sy08sp03-2). Breakage of one sensor occurred at the beginning of the electromechanical test, whereas regular curve of electrical resistance variation (DR/R0) versus elongation (DL/L0) can be seen for another one as shown in Fig. 3.17.

3.4.4

GF/PA66 composites with integrated GF/PA66 or GF sensors

GF/PA66 three layered consolidated under pressure of 4e5 MPa and diverse temperature are presented in Fig. 3.19. GF sensors electrical resistance after their production and integration in 2D fabric can be seen in Table 3.9.

Structural health monitoring of composite structures

Stress (MPa) GF-Sy08sp03-5 GF-Sy08sp03-6

Stress, s (MPa)

250

100.00%

2 y = 486.42x – 8.8974x + 0.0376 R2 = 0.9918

200

150

50.00%

0.00%

100 –50.00%

0

0.03% 0.25% 0.47% 0.69% 0.90% 1.12% 1.34% 1.56% 1.77% 1.99% 2.21% 2.43% 2.64% 2.86% 3.08% 3.29% 3.51% 3.73% 3.95% 4.16% 4.38% 4.60% 4.82% 5.03% 5.25% 5.36% 5.58% 5.79%

50

Electrical resistance variation, DR/R0 (%)

(e) 300

245

–100.00%

Elongation, DL/L0 (%) 300

Stress (MPa) GF-Sy08sp03-5 GF-Sy08sp03-6

Stress, s (MPa)

250

100.00%

50.00%

200

150

0.00%

100 –50.00%

0

0.00 7.80 15.60 23.40 31.20 39.00 46.80 54.60 62.40 70.20 78.00 85.80 93.60 101.40 109.20 117.00 124.80 132.60 140.40 148.20 156.00 163.80 171.60 179.40 183.46 191.26 199.06

50

Electrical resistance variation, DR/R0 (%)

(f)

–100.00%

Time, t (s)

Figure 3.17 cont’d.

GF/PA66 composite with integrated GF/PA66 sensor prepared according to case II with two diverse weaving patterns are presented in Fig. 3.20. GF/PA66 2D textile preform with integrated GF/PA66 sensors, GF-Sy08sp01, prepared according to case III, 4-end satin weaving pattern, is presented in Fig. 3.21. GF/PA66 composites with integrated GF/PA66 sensors prepared according to case III, 4-end satin weaving pattern, under diverse consolidation conditions are presented in Fig. 3.22. GF/PA66 composite with integrated GF sensors, 4-end satin weaving pattern, GF/ PA66cmp-s45P230T-GF-Sy08sp03-1, is presented before electromechanical tests in Fig. 3.23(a), while after tests in Fig. 3.23(b and c). Sensors electrical resistances are in range between 750 and 800 U after its production and insertion during weaving

246

Table 3.8 Tensile results of GF/PP composites with integrated GF sensors GF/PP composites with integrated GF sensors

GF sensors

Sample label GF/PPcmp-s45P185T-GF-Sy08sp03-1

GF-Sy08sp03-1

Time (s)

Force at break (N)

Stress at break (MPa)

Elongation at break (%)

167.5

27760.48

235.35

4.98

175.7

30885.40

269.14

5.15

183.5

29790.31

266.25

5.34

176.57

29478.73

256.91

5.16

5.52

1585.59

18.73

0.13

GF-Sy08sp03-2 GF-Sy08sp03-3 GF-Sy08sp03-4 GF/PPcmp-s45P185T-GF-Sy08sp03-3

GF-Sy08sp03-5 GF-Sy08sp03-6

Average Standard deviation

Smart Textiles for In Situ Monitoring of Composites

GF/PPcmp-s45P185T-GF-Sy08sp03-2

Structural health monitoring of composite structures

247

Sensor electrical resistance (kW)

250 Sensor electrical resistance after consolidation (kW) Sensor electrical resistance in elemechanical test (kW) Average Standard deviation

200

150

100

67

59 50

16.60 19 14.2 3.39

38.45 17.9 29.06

48

42.75 37.5 7.42

0

46 27

24.25 21.5 3.89

3.2

3.5 3.35 0.21

56.50

14.85

GF/PP-Sy08sp03-1 GF/PP-Sy08sp03-2 GF/PP-Sy08sp03-3 GF/PP-Sy08sp03-4 GF/PP-Sy08sp03-5 GF/PP-Sy08sp03-6 Sample label

Figure 3.18 GF sensors electrical resistance after 2D GF/PP textile preform consolidation and during electromechanical test of developed composites.

Figure 3.19 GF/PA66 three layered consolidated under pressure of 4e5 MPa and temperature: (a) 210 C, (b) 220 C, (c) 230 C, (d) 240 C.

of 2D fabric, whereas 10 MU after thermal consolidation of 2D textile preform in related composites. Stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) or time (t) curves can be seen in Fig. 3.24. Tensile results of GF/PA66 composites with integrated GF sensors are presented in Table 3.10.

248

Smart Textiles for In Situ Monitoring of Composites

Table 3.9 GF/PA66 sensors electrical resistance after production and integration in 2D fabric

Sample label

Sensor electrical resistance after production R2 (U)

Sensor electrical resistance after insertion in 2D fabric R2 (U)

2D fabric

GF/PA66-Sy08sp03-1

880

910

plain weave

GF/PA66-Sy08sp03-2

690

750

Average

785

830

Standard deviation

134

113

GF/PA66-Sy08sp03-3

570

600

GF/PA66-Sy08sp03-4

610

670

Average

590

635

28

49

Standard deviation

4-end satin

Figure 3.20 GF/PA66 composite with integrated GF/PA66 sensorsdcase II: (a) GF/PA66cmps45P230T-GF-Sy08sp03, (b) GF/PA66cmp-p45P230T-GF-Sy08sp03.

GF/PA66 three-layered consolidated at 230 C under pressure of 4e5 MPa during 5 min gives the most acceptable appearance, structure consolidation, and unique PA66 polymer distribution (Fig. 3.19). GF/PA66 sensors resistances after 2D fabrics consolidations according to case II and II, were too high, w10 MU, and confirmed that weaving patterns, weave plain or 4-end satin, and applied pressure do not affect

Structural health monitoring of composite structures

249

Figure 3.21 2D GF/PA66 textile preform with integrated GF/PA66 sensorsdpreparation for consolidation stepdcase III.

Figure 3.22 GF/PA66 composite with integrated GF/PA66 sensorsdcase III: (a) GF/PA66cmps23P230T-GF/PA66-Sy08sp01-1, (b) GF/PA66cmp-s45P230T-GF/PA66-Sy08sp01-1.

notable to high sensors resistance values. Hence, these parameters are not crucial to obtain positive sensor electrical responses. Consolidation temperature of 230 C has the most important impact to high electrical resistances due to PA polymer melting and mixing with protective and conductive coating layers. It was interesting to perform electromechanical test of the composite sample prepared according to case II to make comparison with previously tested GF/PP composites. GF/PA66 composite with integrated GF sensors displays regular line breakage between clamps, but without adequate response of textile sensors. One of the reasons could be mixture of thermoplastic polymer PA66 with protective and conductive coating during thermal consolidation step of 2D fabric. For the future work it will be interesting to find new protective solution for newly developed textile sensors.

250

Smart Textiles for In Situ Monitoring of Composites

Figure 3.23 GF/PA66 composite with integrated GF sensorsdcase II: (a) before mechanical testdfront side, (b) after mechanical testdfront side, (c) after mechanical testdback side.

Composite specimen tensile loading until break was performed approximately after 147 s. This composite shows the lowest stress at break 188.88 MPa, in comparison with previously tested composites (Table 3.11), with lower elongation at break, 4.209%.

3.5

Results and discussiondinterface phenomena

Contact angle measurements for applied two polar liquids and one nonpolar liquid of neat yarns, conductive dry films, coated and sensor yarns, and related textile reinforced 2D thermoplastic composites with integrated sensor yarns are presented in Tables 3.12e3.14 needed to determine their Surface Free Energy (SFE). SFE results are calculated and presented in Tables 3.15 and 3.16 for tablets of commingled yarns and their components. SFE results of conductive dry films are calculated and presented in Tables 3.17 and 3.18. SFE results for coated and finally developed sensor yarns are calculated and presented in Tables 3.19 and 3.20. For textile reinforced 2D thermoplastic composites with integrated sensor yarns, SFE results are shown in Tables 3.21 and 3.22. Adhesion parameters at the interface of the tablet-conductive dry film systems are presented in Table 3.23. Having determined the contact angle for applied two polar liquids and one nonpolar liquid of neat yarns (in the form of tablets), dry conductive and protective films, coated and sensor yarns, and related textile reinforced 2D thermoplastic composites with integrated sensor yarns it is possible to determine their SFE [5e8].

Structural health monitoring of composite structures

251

(a)

180

18.00%

GF-Sy08sp03-1

16.00%

GF-Sy08sp03-2

140

14.00%

120

12.00%

100

10.00%

80

8.00%

60

6.00%

40

4.00%

20

2.00%

0

0.00% 0.02% 0.16% 0.30% 0.44% 0.58% 0.72% 0.85% 0.99% 1.13% 1.27% 1.14% 1.55% 1.69% 1.83% 1.97% 2.11% 2.25% 2.39% 2.52% 2.66% 2.80% 2.94% 3.08% 3.22% 3.36% 3.50% 3.64% 3.78% 3.92% 4.05% 4.19% 4.30% 4.38% 4.52% 4.66%

Stress, σ (MPa)

160

Stress (MPa)

Electrical resistance variation, ΔR/R0 (%)

20.00%

200

Elongation, ΔL/L0 (%)

(b) 180

Stress (MPa)

18.00%

GF-Sy08sp03-1

16.00%

GF-Sy08sp03-2

140

14.00%

120

12.00%

100

10.00%

80

8.00%

60

6.00%

40

4.00%

20

2.00%

0

0.00% 0.0 4.6 9.2 13.8 18.4 23.0 27.6 32.2 36.8 41.4 46.0 50.6 55.2 59.8 64.4 69.0 73.6 78.2 82.8 87.4 92.0 96.6 102.2 105.8 110.4 115.0 119.6 124.2 128.8 133.4 138.0 142.6 147.2 151.8 156.4

Stress, σ (MPa)

160

Electrical resistance variation, ΔR/R0 (%)

20.00%

200

Time, t (s)

Figure 3.24 Electromechanical test of GF/PA66 composite with integrated GF sensors: (a) stress (s) and electrical resistance variation (DR/R0) versus elongation (DL/L0) curves, (b) stress (s) and electrical resistance variation (DR/R0) versus time (t) curves. Table 3.10 Tensile results of GF/PA66 composite with integrated GF sensors GF/PA66 composites with integrated GF sensors

GF sensors

Sample label GF/PA66cmps45P230T-GFSy08sp03-1

GF-Sy08sp03-1 GF-Sy08sp03-2

Time (s)

Force at break (N)

Stress at break (MPa)

Elongation at break (%)

147.4

25609.38

188.88

4.29

252

Smart Textiles for In Situ Monitoring of Composites

Table 3.11 Contact angle of used tablets Tablet label

Liquid

Contact angle (8)

GF

Water

0.0

e

Formamide

0.0

e

41.7

e

145.0

2.2

Formamide

92.2

0.8

Diiodmethane

86.6

0.8

100.2

1.6

94.3

2.0

0.0

e

Water

56.7

1.4

Formamide

61.0

0.7

0.0

e

66.2

1.7

Formamide

0.0

e

Diiodmethane

0.0

e

Diiodmethane PP

PA66

Water

Water Formamide Diiodmethane

GF/PP

Diiodmethane GF/PA66

Water

Standard deviation (8)

Table 3.12 Contact angle of conductive dry films Sample label 8% PEDOT-compl-PSS FET-LApp96100DF

15% PEDOT-compl-PSS CPP-LApp96100DF

Thickness (mm)

Liquid

Contact angle (8)

Standard deviation (8)

139.5

Water

78.6

0.8

Formamide

77.7

0.8

Diiodmethane

56.8

0.8

Water

75.9

0.9

Formamide

70.9

0.3

Diiodmethane

60.1

0.5

23.66

During contact angle measurements, it was difficult to capture water and formamide drops at GF tablets surface (one tablet per each liquid) what confirms that polar liquids wet polar surface (solid). Contact angle determined with diiodmethane as nonpolar liquid was w42 degrees. PP shows hydrophobicity for all liquids used, whereas contact angle for PA66 determined with diiodmethane was 0 degrees and has significant role in its SFE calculation what confirms the fact that liquids with high surface tension (water, formamide) cannot wet solids of lower SFE.

Structural health monitoring of composite structures

253

Table 3.13 Contact angle of coated and sensor yarns Sample label

Liquid

Contact angle (8)

Standard deviation (8)

GF/PP-LAp96100

Water

60.6

2.3

Formamide

60.7

2.1

Diiodmethane

29.9

1.6

Water

64.7

0.8

Formamide

63.4

1.8

Diiodmethane

43.5

0.2

Water

63.1

1.2

Formamide

44.8

1.8

Diiodmethane

35.5

1.5

Water

63.0

1.2

Formamide

60.8

2.0

Diiodmethane

32.5

1.3

Water

74.7

0.9

Formamide

76.6

1.2

Diiodmethane

55.7

0.9

Water

81.9

0.9

Formamide

54.6

1.6

Diiodmethane

59.2

1.4

GF/PP-LApp96100-2C

GF/PA66-LApp96100-2C

GF/PP-Sy08

GF/PA66-Sy08

GF-Sy08

Table 3.14 Contact angle of textile reinforced 2D thermoplastic composites with integrated sensor yarns Sample label

Liquid

Contact angle (8)

Standard deviation (8)

GF/PPcmp-s45P185T-GF/PPSy08

Water

107.5

3.0

Formamide

90.2

0.7

Diiodmethane

71.7

1.1

Water

85.4

2.5

Formamide

95.9

1.8

Diiodmethane

60.0

2.3

Water

67.9

1.5

Formamide

66.4

1.4

Diiodmethane

47.8

1.7

GF/PPcmp-s45P185T-GF-Sy08

GF/PA66cmp-s45P230T-GF/ PA66-Sy08

254

Smart Textiles for In Situ Monitoring of Composites

Table 3.15 OwenseWendt and Wu theorydSFE and its components of tablets Applied theory

Sample label

SFE (total) (mJmL2)

Disperse (mJmL2)

Polar (mJmL2)

OwenseWendt

GF

71.56

33.50

38.06

PP

21.47

14.40

7.07

PA

38.83

38.70

0.13

GF/PP

49.80

38.77

11.11

GF/PA66

59.12

52.99

6.13

GF

74.35

34.98

39.37

PP

19.79

19.79

0.00

PA

47.61

47.61

0.00

GF/PP

52.88

38.83

14.05

GF/PA66

62.42

49.80

12.62

Wu

Table 3.16 Acid-Base theorydSFE and its components of tablets Applied theory

Sample label

SFE (total) (mJmL2)

Disperse (mJmL2)

Acid (mJmL2)

Base (mJmL2)

AcideBase

GF

38.57

38.57

0.00

59.56

PP

32.87

32.87

42.41

0.00

PA66

50.56

50.56

0.00

8.73

GF/PP

49.27

49.27

0.00

42.96

GF/PA66

57.00

50.80

2.92

3.29

Table 3.17 OwenseWendt and Wu theorydSFE and its components of conductive dry films Applied theory OwenseWendt

Wu

SFE (total) (mJmL2)

Disperse (mJmL2)

8% PEDOT-compl-PSS FET-LApp96100DF

29.28

23.56

5.72

15% PEDOT-compl-PSS CPP-LApp96100DF

30.12

21.42

8.70

8% PEDOT-compl-PSS FET-LApp96100DF

35.10

26.86

8.24

15% PEDOT-compl-PSS CPP-LApp96100DF

36.81

26.21

10.60

Sample label

Polar (mJmL2)

Structural health monitoring of composite structures

255

Table 3.18 Acid-Base theorydSFE and its components of conductive dry films Applied theory AcideBase

SFE (total) (mJmL2)

Disperse (mJmL2)

Acid (mJmL2)

Base (mJmL2)

8% PEDOTcompl-PSS FETLApp96100DF

20.55

20.55

0.00

21.10

15% PEDOTcompl-PSS CPPLApp96100DF

19.72

19.72

0.00

17.89

Sample label

Table 3.19 OwenseWendt and Wu theorydSFE and its components of coated yarns and developed sensor yarns Applied theory

Sample label

SFE (total) (mJmL2)

Disperse (mJmL2)

OwenseWendt

GF/PP-LApp96100

31.12

28.37

2.76

GF/PP-LApp96100-2C

31.34

29.10

2.23

GF/PA66-LApp96100-2C

48.66

38.38

10.28

GF/PP-Sy08

34.29

32.59

1.69

GF/PA66-Sy08

24.51

22.86

1.65

GF-Sy08

27.95

27.76

0.19

GF/PP-LApp96100

49.37

35.48

13.90

GF/PP-LApp96100-2C

45.09

31.54

13.55

GF/PA66-LApp96100-2C

49.90

35.67

14.24

GF/PP-Sy08

47.55

35.98

11.57

GF/PA66-Sy08

35.65

26.69

8.97

GF-Sy08

37.28

26.62

10.66

Wu

Polar (mJmL2)

GF/PP tablets show similar contact angles by using polar liquids. It was not possible to catch the drops of diiodomethane on the tablet surface. In case of GF/PA66 tablets, contact angle was determined only by using drops of distilled water. From the electrical conductivity point of view, 8% PEDOT-compl-PSS FET-LApp96100DF dry conductive film shows notable higher levels of electrical conductivity compared to 15% PEDOT-compl-PSS CPP-LApp96100DF conductive dry film. This dry film shows a slightly greater hydrophobicity on polar solvents (liquids), water (78.6 degrees), and formamide (77.7 degrees). Contact angle determined with diiodomethane is lower (56.8 degrees). In addition, this conductive film is more hydrophilic to nonpolar solvents (liquids).

256

Smart Textiles for In Situ Monitoring of Composites

Table 3.20 Acid-Base theorydSFE and its components of coated yarns and developed sensor yarns Applied theory AcideBase

SFE (total) (mJmL2)

Disperse (mJmL2)

Acid (mJmL2)

Base (mJmL2)

GF/PPLApp96100

34.58

34.58

0.00

14.42

GF/PPLApp961002C

24.43

24.43

0.00

31.49

GF/PA66LApp961002C

34.18

33.66

0.01

11.21

GF/PP-Sy08

29.14

29.14

0.00

30.16

GF/PA66-Sy08

19.39

19.39

0.00

29.32

GF-Sy08

30.40

30.40

6.58

0.01

Sample label

Table 3.21 OwenseWendt and Wu theorydSFE and its components of textile reinforced 2D thermoplastic composites with integrated sensor yarns Applied theory OwenseWendt

Wu

SFE (total) (mJmL2)

Disperse (mJmL2)

GF/PPcmp-s45P185TGF/PP-Sy08

23.09

22.95

0.10

GF/PPcmp-s45P185TGF-Sy08

24.29

21.91

2.37

GF/PA66cmp-s45P230TGF/PA66-Sy08

29.33

27.31

2.02

GF/PPcmp-s45P185TGF/PP-Sy08

24.60

23.73

0.87

GF/PPcmp-s45P185TGF-Sy08

28.77

24.72

4.05

GF/PA66cmp-s45P230TGF/PA66-Sy08

42.57

30.07

12.50

Sample label

Polar (mJmL2)

From the results of the contact angle measurements, it is obviously that coated yarn and finally developed sensor yarns are more hydrophillic compared with textile reinforced 2D thermoplastic composites with integrated sensor yarns. For this observation, mass content of GF/PP (71:29) and GF/PA66 (65:35) commingled yarns has to be taken into account and different coating layers applied (protective and conductive). GF presence has crucial role in obtained contact

Structural health monitoring of composite structures

257

Table 3.22 AcideBase theorydSFE and its components of textile reinforced 2D thermoplastic composites with integrated sensor yarns Applied theory AcideBase

SFE (total) (mJmL2)

Disperse (mJmL2)

Acid (mJmL2)

GF/PPcmp-s45P185TGF/PP-Sy08

24.19

24.19

0.00

0.92

GF/PPcmp-s45P185TGF-Sy08

16.32

16.32

0.00

35.84

GF/PA66cmps45P230T-GF/ PA66-Sy08

23.09

23.09

0.00

28.88

Sample label

Base (mJmL2)

Table 3.23 Adhesion parameters at the interface of the tablet-conductive dry film systems Interfacial energy (mJmL2)

Thermodynamic work of adhesion (mJmL2)

Wetting coefficient (mJmL2)

21.42

88.03

17.83

GF/PP/8% PEDOT-compl-PSS FET-LApp96100DF

3.70

84.28

14.08

GF/PA66/8% PEDOT-complPSS FET-LApp96100DF

7.78

89.74

19.54

GF/15% PEDOT-compl-PSS CPP-LApp96100DF

17.82

93.34

19.72

GF/PP/15% PEDOT-compl-PSS CPP-LApp96100DF

2.93

86.76

13.14

GF/PA66/15% PEDOT-complPSS CPP-LApp96100DF

7.50

91.73

18.11

Sample label GF/8% PEDOT-compl-PSS FETLApp96100DF

angles results. Contact angles determined with diiodmethane are lower in comparison with polar liquids, water, and formamide. Protective and conductive layers applied increase yarns hydrophobicity. Application of PEDOT-compl-PSS conductive layers according to new formulation described in this study after first protective coating of GF/PP yarn decreases its hydrophobicity compared with only one-layer protective coated GF/PP yarn, whereas the last protective coating presently increases it once again. GF/PA66 yarn coated with one protective and two conductive coatings is more hydrophilic than GF/PP-coated yarn. Besides, GF/PA66 sensor yarn is more hydrophobic

258

Smart Textiles for In Situ Monitoring of Composites

than GF/PA66 yarn with one protective and two conductive coating what can be explained with better adhesion between all coating layers applied. GF sensor yarn, GF-Sy08, shows higher hydrophobicity compared to GF/PP and GF/PA66 sensor yarns, GF/PP-Sy08 and GF/PA66-Sy08, on water (gL ¼ 72.8 mJm2, h ¼ 0.01,000 P), whereas lower hydrophobicity on formamide (gL ¼ 58.0 mJm2, h ¼ 0.04550 P). Higher dispersion in results can be seen for obtained contact angle using diiodomethane (59.2 degrees) compared with GF/PP (32.5 degrees) and GF/PA66 (55.7 degrees) sensor yarns. GF/PA66 composites with integrated GF/PA66 sensor yarns are more hydrophilic than GF/PP composites with integrated GF/PP or GF sensor yarns. Thermal consolidation step of 2D GF/PA66 fabric and different conditions contribute to these results. Finally, SFE for tablets of commingled yarns and their components (Tables 3.15 and 3.16), dry films (Tables 3.17 and 3.18), coated and finally developed sensor yarns (Tables 3.19 and 3.20), and textile reinforced 2D thermoplastic composites with integrated sensor yarns (Tables 3.21 and 3.22) were calculated. According to the literature, SFE of PP is around 30 mJm2, SFE of PA66 is 46.5 mJm2, whereas SFE of GF is 100 mJm2 [9]. In this study, according to Wu and OwenseWendt theories, SFE of GF is w74 mJm2, whereas AcideBase theory does not give acceptable results. SFE of PP is 33 mJm2 according to AcideBase theory, while according to other theories slightly lower, 20 mJm2. SFE for PA66 is in range between 40 and 50 mJm-2 depending on the theory, whereas AcideBase theory shows Base component effect, 8.73 mJm2. Disperse component has major role in Total SFE for mostly all analyzed samples. Only in case of GF, polar component, 38.06 mJm2, according to Owense Wendt, or 39.37 mJm2, according to Wu, has higher influence to the Total SFE, but not too high as in literature [9]. Slightly lower SFE was shown by 8% PEDOT-compl-PSS FET-LApp96100DF dry film compared to 15% PEDOT-compl-PSS CPP-LApp96100DF film according to OwenseWendt and Wu theory and slightly higher effect of dispersed component. Besides, this film displays lower SFE of polar components. According to AcideBase theory, developed dry films have similar total SFE, w 20 mJm2, and higher influence of Base components (compared to SFE of Acid component, 0.00 mJm2). However, 8% PEDOT-compl-PSS FET-LApp96100DF conductive dry film has a slightly higher total SFE and SFE of other components, except Acid component. This dry film shows also slightly lower SFE compared to 15% PEDOT-compl-PSS CPP-LApp96100DF film according to OwenseWendt and Wu theory and slightly higher effect of dispersed component. Moreover, this film displays lower SFE of polar components. According to AcideBase theory, developed dry films have similar total SFE, w20 mJm2, and higher influence of Base components (compared to SFE of Acid component, 0.00 mJm2).

Structural health monitoring of composite structures

259

According to OwenseWendt and Wu, the highest SFE was shown by GF/PP sensor yarn, GF/PP-Sy08. Lower SFE of GF sensor yarn, GF-Sy08, than expected, can be explained with better coating process, yarn and coating adhesion, which in this case decreases SFE. In case of GF/PA66 sensor yarn, GF/PA66-Sy08, higher mass content of PA66, than PP mass content in GF/PP yarn, affects to its SFE. GF/PP yarn with one protective coating LApp96100 and two conductive coatings 8% PEDOT-compl-PSS FET-LApp96100, GF/PP-LApp96100-2C, shows lower or equal SFE compared with GF/PP yarn coated with one protective coating LApp96100, GF/PP-LApp96100. Finally, GF/PP sensor yarn, GF/PP-Sy08, has higher SFE than GF/PP-LApp96100-2C, but lower than GF/PP-LApp96100. In contrast, GF/PA66 sensor yarn, GF/PA66-Sy08, has lower SFE than GF/PA66 yarn with one protective coating LApp96100 and two conductive coatings 8% PEDOT-compl-PSS FET-LApp96100, GF/PA66-LApp96100-2C. According to AcideBase theory, the same conclusion can be found for SFE of GF/PP and GF/PA66 coated with one protective coating LApp96100 and two conductive coatings 8% PEDOT-compl-PSS FET-LApp96100 compared to SFE of sensor yarns. In general, results are lower according to this theory. GF and GF/PP sensor yarns, GF/PP-Sy08 and GF-Sy08, have similar SFE, w 30 mJm2, whereas GF/PA66 sensor yarn, GF/PA66-Sy08, has notable lower SFE. Base component can be seen for GF/PA66 and GF/PP sensor yarns (w30 mJm2), although it does not have influence on final SFE calculation. For textile-reinforced 2D thermoplastic composites with integrated sensor yarns analyzed, according to OwenseWendt and Wu theories, SFE is under influence mostly of PP or PA66 due to thermal consolidation effect of 2D fabrics (GF/PP or GF/PA66). Wu theory seems to be the most appropriate for SFE calculation. AcideBase theory gives notable lower SFE results, and it is not acceptable for SFE composites analysis, although in case of GF/PP composites with integrated GF/PP sensor yarns SFE is mostly the same for all the theories. Tablet-conductive dry film systems show low interfacial energy (Table 3.23), which is close to the optimum thermodynamic conditions when the interfacial energy is minimal and tends to zero. Conductive dry films of two key polymer complex PEDOT-compl-PSS formulations determined by percolation threshold do not show notable different total SFE. Slightly lower interfacial energy and higher work of adhesion for GF, GF/PP and GF/PA66 tablets in interactions with the conductive film 15% PEDOT-compl-PSS CPP-LApp96100DF, with wetting coefficient higher only for GF tablet, compared to tablets interactions with 8% PEDOT-compl-PSS FET-LApp96100DF, can be seen. Finally, in real systems of developed textile reinforced 2D thermoplastic composites, it is important to outline the impact of the first protective coating of acrylic esters (synthetic latex) of sensor yarns and its interaction with PP and/or PA 66 and PEDOT-compl-PSS polymer complex during thermal consolidation of 2D textile preforms in a hot press under certain conditions previously described.

260

3.6

Smart Textiles for In Situ Monitoring of Composites

Results and discussiondtomography analysis of textile reinforced 2D thermoplastic composites with integrated textile sensors

Tomography images performed by using Easy Tom machine (RX Solution, 3D X-ray CT scan, resolution 10e15 mm) were taken from three sides (x, y, z coordinates) of samples analyzed according to Fig. 3.25 in determined scan area previously mentioned. Images of composites with textile sensors integrated, before and after electromechanical test, can be seen in Figs. 3.26 and 3.27.

(a)

Front Right Top

Front Right Top

(b)

Front Right Top

Front Right Top

Figure 3.25 View of the samples by camera in MyVGL 2.2: (a) before electromechanical test, (b) after electromechanical test.

Structural health monitoring of composite structures

261

Figure 3.26 Tomography analysisdGF/PP composite with integrated GF/PP sensor before electromechanical test: (a) front side, (b) right side, (c) top side.

Figure 3.27 GF/PP composite with integrated GF/PP sensor after electromechanical test: (a) front side, (b) right side, (c) top side.

Because of consolidation step, applied temperature and pressure, destroyed fabric area can be seen around copper wires where silver drops have been applied (light part) before electromechanical test. GF/PP sensor cannot be detected clearly after electromechanical test, although GF yarn as a part of it was not destroyed. Composite breakage occurred at the surface. There are little cracks and fiber breakage inside the composite structure.

262

3.7

Smart Textiles for In Situ Monitoring of Composites

Results and discussiondelectrical resistance dependence of textile sensors on climatic conditions

The dependence of the sensors electrical resistance on climatic conditions is investigated in climatic chamber. Three samples of GF/PP sensors were mounted on the Teflon plate placed at the center of the chamber (Fig. 3.28). Textile sensors were connected to the Keithley KUSB data acquisition digital I/O counter/time and resistance boxes placed outside the chamber and interfaced with a computer (QuickDAQ software) to record the electrical resistance during the experiments. Isohumidity cycle was applied where RH of 65% is kept constant and temperature is varied from 20 C till 160 C (limited temperature of the climatic chamber) with increasing step of 20 C. The electrical resistance is expressed as a normalized relative electrical resistance Rrc, based on a reference value of electrical resistance, R0 (Eq. 3.1): Rrc ¼

ðR  R0 Þ R0

(3.1)

R is the sensor electrical resistance during measurements, whereas R0 is the sensor electrical resistance at a temperature of 20 C and a relative humidity of 65% in climatic chamber. The results of electrical resistance dependence of textile sensors on climatic conditions are presented in Table 3.24 and Fig. 3.29. The results obtained in the isohumidity experiment of the GF/PP sensors shown in Fig. 3.29 where normalized relative electrical resistance points corresponding to the rising temperature are presented as well. According to results, there are no notable changes of sensor electrical resistance (Table 3.24, Fig. 3.29) during measurements in temperature range between 40 and 160  C in comparison with reference electrical resistance value, R0. Till 100 C, R is decreased for 30e50 U, and after 100 C, R is increased for 20e40 U.

Figure 3.28 GF/PP sensors in climatic chamber.

Temperature (8C)

Textile sensor

Sample label

Sensor electrical resistance after production

GF/PP_Sy

L-1

750

810

770

740

690

670

700

740

760

L-2

580

630

590

490

450

450

470

500

530

L-3

520

520

510

490

460

450

470

500

530

R0

40

60

80

100

120

140

160

Sensor electrical resistance (U)

Structural health monitoring of composite structures

Table 3.24 Sensors electrical resistance studied in climatic chamber

263

Smart Textiles for In Situ Monitoring of Composites

Normalized relative electrical resistance, Rrc (W)

264 0.10 0.00 –0.10

GF/PP-Sy08sp07-1

GF/PP-Sy08sp07-2

GF/PP-Sy08sp07-3

–0.02 40°C –0.06

0.02 60°C

80°C

100°C

120°C

–0.06 –0.12

–0.05

–0.20

140°C –0.04

–0.16

–0.10 –0.13 –0.21

–0.22

–0.30 –0.40

–0.70

–0.06

–0.25 –0.09

–0.29

–0.09

–0.29

–0.50 –0.60

160°C

–0.14 –0.15 –0.17

Temperature (°C)

Figure 3.29 Normalized relative electrical resistance versus temperature in isohumidity cycle of GF/PP sensors.

It can be noted that the highest decrease in sensor electrical resistance for all GF/PP sensors is at 100 C. This behavior is due to an increase in the number of conducting chains and an increasing of the tunnel effect. The PP polymer undergoes an expansion following a temperature rise from 40 C till 100 C, and this may cause the higher number of electrical contacts between the conductive particles, and increase the effective section of electrical conductivity. Furthermore, the general change in the geometry of the GF/PP sensors will tend to decrease the electrical resistance. In case of disjoined particles, a temperature rise will increase the number of charge carriers, and hence favor conduction by tunnel effect. After 100 C, increase of sensor electrical resistance starts again, but because of climatic chamber temperature range limitation (160 C), there is restriction for sensors electrical resistance study at higher temperature (>160 C).

3.8

ResultsdSEM and EDS analysis of yarns

Scanning Electron Microscopy (SEM) with Energy Dispersive Spectroscopy (EDS) was used (Tescan, MIRA/LMU, Czech Republic) for microscopic analysis of GF, GF/PP, and GF/PA66 pure yarns. Scanning electron micrographs of GF, GF/PP, and GF/PA66 pure yarns are presented in Figs. 3.30e3.32. Scanning electron microscope monitoring shows GF fibers of approximately 15 mm diameter, PP of 40 mm diameter, and PA66 of 33 mm diameter. Scanning electron micrographs of electroconductive dry films, 15% PEDOTcompl-PSS CLEVIOS P FORM, CPP105D/Latex Appretan 96100 (thickness 94.90 mm), and 8% PEDOT-compl-PSS CLEVIOS F ET/Latex Appretan 96100 (thickness 135.67 mm) are presented in Fig. 3.33. Both electroconductive dry films have fibrous morphology, whereas second dry film shows more uniform surface, which is an important parameter for achieving its greater electro conductivity.

Structural health monitoring of composite structures

(a)

SEM HV: 10.00 kV SEM MAG: 1.00 kx Name: A_12

(b)

WD: 32.21 mm Det: SE

MIRA\\ TESCAN 50 mm Performance in nanospace

(c)

SEM HV: 10.00 kV SEM MAG: 1.00 kx Name: B_16

265

SEM HV: 10.00 kV SEM MAG: 3.00 kx Name: A_13

WD: 32.16 mm Det: SE

MIRA\\ TESCAN 20 mm Performance in nanospace

(d)

WD: 32.13 mm Det: SE

MIRA\\ TESCAN 50 mm Performance in nanospace

SEM HV: 10.00 kV SEM MAG: 3.00 kx Name: B_17

WD: 32.15 mm Det: SE

MIRA\\ TESCAN 20 mm Performance in nanospace

Figure 3.30 Scanning electron micrographs of yarn surface: (a) GF fibers (SEM mag 1.00k), (b) GF fibers (SEM mag 3.00k), (c) GF/PP fibers (SEM mag 1.00k), (d) GF/PP fibers (SEM mag 3.00k).

Results of EDS analysis of GF, PP and PA66 fibers are shown in Fig. 3.34 while for GF/PP sensor yarn without last protective coating (yarn coated with one protective and two conductive coating) in Fig. 3.35. The spectrum of GF fiber shows significant presence of oxygen (44.97 norm. wt.%), silicon (25.51 norm. wt.%), calcium (18.99 norm. wt.%), and aluminum (7.92 norm. wt.%); low presence of titanium (1.45 wt.%) and potassium (1.00 norm. wt.%) elements; and while magnesium in traces (0.16 wt.%).

266

Smart Textiles for In Situ Monitoring of Composites

(a)

SEM HV: 3.60 kV SEM MAG: 2.00 kx

(b)

WD: 12.24 mm Det: SE 20 mm Date(m/d/y): 12/09/15

MIRA\\ TESCAN Performance in nanospace

SEM HV: 3.60 kV SEM MAG: 350 x

WD: 12.52 mm Det: SE 100 mm Date(m/d/y): 12/09/15

MIRA\\ TESCAN Performance in nanospace

(c)

SEM HV: 3.60 kV SEM MAG: 1.01 kx

WD: 12.58 mm Det: SE 50 mm Date(m/d/y): 12/09/15

MIRA\\ TESCAN Performance in nanospace

Figure 3.31 Scanning electron micrographs of fiber surface: (a) GF (SEM mag 2.00k), (b) PP (SEM mag 350), (c) PA66 (SEM mag 1.01k).

In case, of thermoplastic polymers, the spectrum of PP displays significant presence of carbon (97.46 wt.%), with low presence of oxygen (2.54 wt.%) element. EDS analysis of PA66 fiber shows crucial presence of carbon (79.96 wt.%). Nitrogen element (14.79 wt.%) was detected as well with low presence of oxygen (5.24 wt.%). The spectrum of the surface of GF/PP sensor yarn without last protective coating (yarn coated with one protective and two conductive coating) shows significant presence of carbon (67.99 wt.%) and oxygen (26.10 wt.%) elements with low presence

Structural health monitoring of composite structures

(a)

SEM HV: 10.00 kV SEM MAG: 1.00 kx Name: A_4

(b)

WD: 26.65 mm Det: SE

50 mm

MIRA\\ TESCAN Performance in nanospace

(c)

SEM HV: 10.00 kV SEM MAG: 3.00 kx Name: B_9

267

SEM HV: 10.00 kV SEM MAG: 3.00 kx

WD: 26.72 mm Det: SE

20 mm

MIRA\\ TESCAN

(d)

WD: 26.43 mm Det: SE

20 mm

MIRA\\ TESCAN Performance in nanospace

SEM HV: 10.00 kV SEM MAG: 3.00 kx

WD: 26.42 mm Det: SE

20 mm

MIRA\\ TESCAN

Figure 3.32 Scanning electron micrographs of yarn cross-section: (a) GF fibers (SEM mag 1.00k), (b) GF fibers (SEM mag 3.00k), (c) GF/PP fibers (SEM mag 3.00k), (d) GF/PP fibers (SEM mag 3.00k).

of silicon (2.28 wt.%). Sulfur element (3.63 wt.%) confirms sulfonate groups of very thin conductive layers of PEDOT-compl-PPS. The spectrum of the cross-section of this sensor yarn shows similar chemical map in comparison with its surface observation, with low presence of calcium (6.65 wt.%) and aluminum (2.83 wt.%). It can be concluded that GF component has important role in chemical map detection of cross section of this yarn due to significant presence of

268

Smart Textiles for In Situ Monitoring of Composites

(a)

(b)

SEM HV: 3.60 kV

WD: 9.057 mm

SEM MAG: 1.00 kx

Det: SE

MIRA\\ TESCAN 50 μm

Date(m/d/y): 05/05/16

Performance in nanospace

(c)

SEM HV: 3.60 kV

WD: 17.45 mm

SEM MAG: 2.00 kx

Det: SE

20 μm

Date(m/d/y): 05/05/16

MIRA\\ TESCAN Performance in nanospace

(d)

SEM HV: 3.60 kV

WD: 9.050 mm

SEM MAG: 1.00 kx

Det: SE

50 μm

Date(m/d/y): 05/05/16

MIRA\\ TESCAN Performance in nanospace

SEM HV: 3.60 kV

WD: 17.43 mm

SEM MAG: 2.00 kx

Det: SE Date(m/d/y): 05/05/16

MIRA\\ TESCAN 20 μm Performance in nanospace

Figure 3.33 Scanning electron micrographs of electroconductive dry films: (a and b) 15% PEDOT-compl-PSS CLEVIOS P FORM. CPP105D/Latex Appretan 96100 and (c and d) 8% PEDOT-compl-PSS CLEVIOS F ET/Latex Appretan 96100.

oxygen (26.49 wt.%) and silicon element (10.22 wt.%). Cross-section analysis shows that aqueous conductive dispersion of polymer complex PEDOT-compl-PSS did not penetrate deeply in pure GF/PP yarn during roll to roll coating procedure. Sulfur element was detected in traces (0.42 wt.%).

Structural health monitoring of composite structures

269

(a) cps/eV

5

4 Ti K CO

3

Al Mg Si

Ti

K Ca

Ca

2

1

0

0

2

4

Element Oxygen Silicon Calcium Aluminium Titanium Potassium Magnesium

(b)

AN 8 14 20 13 22 19 12

6

8

keV

Series (wt.%) K-series 31.43825252 K-series 17.83264525 K-series 13.27754934 K-series 5.53835786 K-series 1.01101658 K-series 0.70229652 K-series 0.11373688 Sum: 69.9138549

10

12

14

16

18

(norm. wt.%) 44.96712782 25.50659703 18.99129915 7.92168858 1.44608902 1.00451694 0.16268145 100

(norm. at.%) 61.78704604 19.96532302 10.41728663 6.45441687 0.66396757 0.56481390 0.14714598 100

Error in wt.% 3.996417551 0.788986534 0.422983034 0.294667933 0.065423938 0.053929554 0.034887672

2.0

2.5

3.0

(norm. wt.%) 97.46321932 2.53678068 100

(norm. at.%) 98.08347935 1.91652065 100

cps/eV

100 80 60

C

O

40 20 0 0.0

0.5

Element Carbon Oxygen

AN 6 8

1.0

1.5 keV

Series (wt.%) K-series 97.46321932 K-series 2.53678068 100 Sum:

Figure 3.34 EDS analysis of fiber surface: (a) GF, (b) PP, (c) PA66.

Error in wt.% 10.66510459 0.532448011

270

Smart Textiles for In Situ Monitoring of Composites

(c) cps/eV

60 50 40 C

N

O

30 20 10 0 0.0

0.5

Element Carbon Nitrogen Oxygen

AN 6 7 8

1.0

Series K-series K-series K-series Sum:

1.5 keV

(wt.%) 79.96334142 14.79466327 5.24199531 100

2.0

2.5

(norm. wt.%) 79.96334142 14.79466327 5.24199531 100

(norm. at.%) 82.79039996 13.13522366 4.074376382 100

3.0

Error in wt.% 8.819520391 2.479144081 0.898035189

Figure 3.34 cont’d.

3.9

3.9.1

Results and discussiondthermal properties of yarns and textile reinforced 2D thermoplastic composites with integrated sensor yarns Thermogravimetric analysis

Commercial thermogravimetric analyser was used to thermally decompose milligram samples under controlled heating and environmental conditions. Thermogravimetric analysis (TGA) of dry films, pure yarns, and sensor yarns (5 mg test samples) have been done (TGA Q50, TA Instruments) in oxidizing (air) and inert (nitrogen) atmospheres under following conditions: flow rate of 50 mL/min and a heating rate of 10 C/min over a temperature range of 50e600 C (Figs. 3.36e3.38). The samples were placed in open alumina pans. From TG curves, the parameters Tonset (the temperature at which the sample lost 5 wt.% of its original mass), Tendset (decomposition temperature at which the final weight became stable), and pyrolysis residue at 600 C were detected and results are shown in Table 3.25. TGA shows that conductive dry film starts to decompose earlier than protective dry film while its final decomposition temperature is higher in both atmospheres. In air atmosphere its pyrolysis residue is lower, 1.0% for conductive dry film and 3.4% for protective dry film. Besides conductive dry film degrades in two degradation steps, whereas protective dry film only in one. In nitrogen, conductive dry film starts and ends degradation earlier than protective dry film. For both films, degradation occurs in one step, and conductive dry film has significantly higher pyrolysis residue, 7.4%.

Structural health monitoring of composite structures

271

C (a) Counts

O

Si 1.00 Element Carbon Oxygen Silicon Sulphur

S

2.00 Series K-series K-series K-series K-series Sum:

3.00 (wt.%) 67.99 26.10 2.28 3.63 100

4.00

5.00

(At.%) 75.62 21.79 1.08 1.51 100

6.00 K-ratio 0.4320 0.1004 0.0204 0.0331

7.00 Z 1.0107 0.9882 0.9285 0.9160

8.00 A 0.6286 0.3892 0.9649 0.9934

9.00

keV

F 1.0002 1.0000 1.0017 1.0000

(b)

Element Carbon Oxygen Aluminium Silicon Sulphur Calcium

Series K-series K-series K-series K-series K-series K-series Sum:

(wt.%) 53.38 26.49 2.83 10.22 0.42 6.65 100

(At.%) 65.85 24.54 1.56 5.39 0.2 2.46 100

K-ratio 0.2738 0.1009 0.0240 0.0914 0.0038 0.0600

Z 1.0242 1.0014 0.9164 0.9416 0.9288 0.9003

A 0.5008 0.3802 0.9200 0.9482 0.9656 1.0011

F 1.0002 1.0001 1.0042 1.0008 1.0023 1.0000

CK

OK

AIK

SiK

SK

CaK

Figure 3.35 EDS analysis of GF/PP sensor yarn without last protective coating: (a) surface, (b) cross-section.

272

Smart Textiles for In Situ Monitoring of Composites

(a)

TGA curves of dry films in air atmosphere

120%

8% PEDOT-compl-PSS C F ET/L App 96100 L App 96100

100%

Weight (%)

80% 60% 40% 20% 0% 50

(b)

100 150 200 250 300 350 400 450 500 550 600 Temperature (°C) TGA curves of dry films in nitrogen atmosphere

120% 8% PEDOT-compl-PSS C F ET/L App 96100 L App 96100

100%

Weight (%)

80% 60% 40% 20% 0% 50

150

250 350 Temperature (°C)

450

550

Figure 3.36 TGA curves of dry films, 8% PEDOT-compl-PSS CLEVIOS F ET/Latex Appretan 96100 and Latex Appretan 96100: (a) air atmosphere, (b) inert nitrogen atmosphere.

GF/PP starts to decompose earlier and degradation ends earlier also compared to GF/PA66 in both atmospheres. This can be explained with lower decomposition step of PP than PA66. GF fibers as an inorganic material do not burn or support combustion, but melts at 500 C and above.

Structural health monitoring of composite structures

(a)

273

TGA curves of pure yarns in air atmosphere

120% GF/PP

GF/PA66

GF

100%

Weight (%)

80% 60% 40% 20% 0% 50

(b)

150

250 350 Temperature (°C)

450

550

TGA curves of pure yarns in nitrogen atmosphere

120% GF/PP

GF/PA66

GF

100%

Weight (%)

80% 60% 40% 20% 0% 50

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

Figure 3.37 TGA curves of pure yarns: (a) air atmosphere, (b) inert nitrogen atmosphere.

Sensor yarns degrade in two degradation step in air atmosphere. It is interesting to see that GF/PP sensor yarn starts to decompose at higher temperature than GF/PA66 sensor yarn, which is not in case of neat yarn previously discussed. One of the reasons could be greater thicknesses of coating applied although the same number of coating during the textile sensors preparation was done. GF sensor yarn starts to decompose at higher temperature and has notable higher pyrolysis residue in both atmospheres.

274

Smart Textiles for In Situ Monitoring of Composites

(a) 120%

TGA curves of sensor yarns in air atmosphere GF/PP_Sy

GF/PA66_Sy

GF_Sy

100%

Weight (%)

80% 60% 40% 20% 0% 50

(b)

150

250 350 Temperature (°C)

450

550

TGA curves of sensor yarns in nitrogen atmosphere

120% GF/PP_Sy

GF/PA66_Sy

GF_Sy

250 350 Temperature (°C)

450

100%

Weight (%)

80% 60% 40% 20% 0% 50

150

550

Figure 3.38 TGA curves of sensor yarns: (a) air atmosphere, (b) inert nitrogen atmosphere.

3.9.2

Results and discussiondmicroscale combustion calorimetry analysis

The microscale combustion calorimetry (MCC) tests of conductive dry film, protective dry film, pure yarns, and related sensor yarns have been carried out according to ASTM: D7309 on a Govmark MCC-2 Microscale Combustion Calorimeter. Each sample was prepared as a powder and placed within the sample cup. Test was performed under following conditions: pyrolyzer operating temperature range of

Structural health monitoring of composite structures

275

Table 3.25 Thermogravimetrical data of dry films and pure and sensor yarns Pyrolysis residue at 6008C

Pyrolysis residue at 6008C

Tonset (8C)

Tendset (8C)

1.0%

233

480

7.4%

408

3.4%

328

447

3.5%

238

334

68.40%

344

447

71.7%

GF/PA66

318

555

65.0%

365

482

66.6%

GF

w600

>600

98.0%

w600

>600

98.8%

GF/PP_Sy

306

506

26.6%

345

483

27.8%

GF/PA66 Sy

289

532

20.5%

316

470

22.6%

GF Sy

316

518

42.6%

344

451

42.7%

Tonset (8C)

Tendset (8C)

8% PEDOT-complPSS C F ET/L App 96100

230

530

L App 96100

288

GF/PP

Sample label

75e750 C, detection sensitivity limit of 5 mW, and repeatability of
2% (5 mg test sample). The mean values of the MCC measurements were calculated as an average of three results of the MCC tests. Fig. 3.39 shows the peak Heat Release Rate, H.R.R. (W/g) results in correlation with the maximum temperature, Tmax ( C). The results are collected in Table 3.26. The higher the char yield, the more carbon/inorganic material left behind, decreased heat release rate, decreased amount of combustible volatile release, resulting in lower flammability (lower H.R.R. temperature and H.R.R. values). MCC data in this dissertation confirm this fact as well. Variation is higher only in case of GF/PP. The H.R.R peak (w340 W/g) of GF/PP and GF/PA66 sensor yarns (w350 W/g) occur at similar Heat Release Temperature, Tmax (w430 C), although there is additional peak detected for GF/PP sensor yarn at 493 C. GF sensor yarn displays lower H.R.R peak (253 W/g) at similar temperature. MCC data show that high pyrolysis residue indicates high GF content for pure yarns and related sensor yarns. In general, higher H.R.R peaks for sensor yarns occur at slightly lower temperature than of pure yarns. Pyrolysis residues are significantly lower than of pure yarns and have to be taken into account as in TG analysis. In addition, the Heat Release Capacity values obtained by dividing the Maximum value of the Specific Heat Release Rate (Qmax) with the heating rate in the test are slightly different as well between dry films, pure yarns, and sensor yarns. The heated sample will ignite when it reaches a particular temperature, Tign(onset), in this case at 422 C for GF/PP yarn, 371 C for GF/PA66 yarn, and at lower temperature for sensor yarns due to coating applied. Besides, MCC pyrolysis residues are mostly coherent to obtained TGA residue values.

276

Smart Textiles for In Situ Monitoring of Composites

(a)

Heat release rate of dry films

500 8% PEDOT-compl-PSS C F ET/L App 96100

450

L App 96100

400

HRR (W/g)

350 300 250 200 150 100 50 0

0

100

200

300 400 500 Temperature (°C)

600

700

800

700

800

Heat release rate of pure yarns

(b) GF/PP

309

GF/PA66

GF

257

HRR (W/g)

205 153 101 49 –3

0

100

200

300 400 500 Temperature (°C)

600

Figure 3.39 MCC curves: (a) conductive and protective dry films, (b) pure yarns, (c) sensor yarns.

3.9.3

Results and discussiondlimiting oxygen index

Limiting oxygen index (LOI) measurements of developed structures have been done by Dynisco LOI instrument (Table 3.27). Three measurements per sample have been performed with specific sample dimensions for textile reinforced 2D thermoplastic composites with integrated sensor yarns (10  140 mm2). It must be taken into consideration that LOI method is usually difficult to apply to textile-reinforced thermoplastic composites due to sample melting. For all the samples, during the heating of thermoplastics, the polymer softens, melts, and drips.

Structural health monitoring of composite structures

(c)

277

Heat release rate of sensor yarns

400 GF/PP_Sy

GF/PA66_Sy

GF_Sy

350

HRR (W/g)

300 250 200 150 100 50 0

0

100

200

300 400 500 Temperature (°C)

600

700

800

Figure 3.39 cont’d.

It must be taken into consideration that LOI method is usually difficult to apply to textile-reinforced thermoplastic composites due to sample melting. For all the samples, during the heating of thermoplastics, the polymer softens, melts, and drips. As mentioned earlier, GF fibers as an inorganic material do not burn but in this work support combustion, melts at 500 C and above. GF/PP composites with integrated GF/PP or GF sensor yarns have poor resistance to fire. PP has tendency to drip; hence, it may aid in the propagation of fire. According to results (Table 3.27), its LOI is 22. Less time is needed for LOI test of GF/PP composites with GF sensor yarns due to lower content of PP. GF/PA66 composites with integrated GF/PA66 sensor yarns show higher LOI due to different thermal consolidation conditions of 2D fabrics and higher decomposition temperature of PA66 observed as well during MCC and TG testing.

3.10

Toward wireless structural health monitoring

As mentioned in previous sections, the main effort should be focused on the development of Smart textiles, electronics combined with textiles also called textronics, as this technology will be able to monitoring the composite part during their production and through all it service-life. This requires the necessary insertion of sensors and actuators during the fabrication process of the composite. Table 3.28 summarizes the types of sensors and their location on or within composite parts together with the measured physical parameters. Often, it is quite difficult to connect embedded sensors to measuring devices that generate data and send them to data basis also called clouds. The data may be processed in real time, or afterward, to determine the composites health and decide on their eventual replacement.

278

Table 3.26 MCC data of dry films and pure and sensor yarns Maximum specific heat release rate or peak H.R.R., Qmax (W/g)

Heat release temperature, Tmax (8C)

Specific heat release or heat of combustion of sample, hc (kJ/g)

Specific heat of combustion of fuel gases, hc, gas (kJ/g)

Yield of pyrolysis residue, Yp (g/g)

Tign (8C)

Tends (8C)

8% PEDOTcompl-PSS C F ET/L App 96100

358.00

361.70

426.60

20.30

22.16

0.08

315

472

L App 96100

476.00

483.00

430

23.90

24.35

0.02

328

484

GF/PP

302.00

303.03

484.93

9.87

36.15

0.73

422

510

GF/PA66

233.00

216.45

446.85

11.75

30.17

0.61

371

478

GF

e

e

e

e

e

e

e

e

GF/PP Sy

337.00

341.10

434.10/493.20

18.60

24.76

0.25

334

511

GF/PA66 Sy

344

347.90

429.60

19.20

24.37

0.21

320

497

GF Sy

250

253.00

429.30

12.90

27.32

0.53

337

472

Sample label

Smart Textiles for In Situ Monitoring of Composites

Heat release capacity, hc (J/g-K)

Structural health monitoring of composite structures

279

Table 3.27 LOI of developed textile reinforced 2D thermoplastic composites Textile reinforced 2D thermoplastic composites/integrated sensor yarns

Sample label

Time t (s)

LOI

GF/PP composite/GF/PP sensor yarns

GF/PPcmp-GF/PP_Sy

421

22

470

22

GF/PP composite/GF sensor yarns

GF/PPcmp-GF_Sy

372

22

390

22

GF/PA66 composite/GF/PA66 sensor yarns

GF/PA66cmp-GF/PA66

242

23

315

26

Table 3.28 Composite sensors’ locations Physical parameter measurement

Category

Configuration

Polymer Optical sensor

Embedded

Strain and vibration measurements

Flexible piezoelectric sensors

Surface mounted and Embedded

Vibration measurements (active or passive)

Piezoelectric fibers

Embedded

Vibration and acoustic emission measurements

Vacuum Monitoring sensors

Surface mounted and Embedded

Pressure measurement

Carbon nanotubes

Embedded

Strain measurements

Flexible Piezoresistive fibrous sensors

Embedded

Strain measurements

The problem of connections with measuring devices could be solved by using the wireless connections for communication among them. To achieve this objective, several additional devices have to be added such as: • • •

embedded power supplies or wireless powering of sensors and communication devices; miniaturized communication devices (emitters) integrated to composite part together with sensors; antennas able to enhance the range of communication.

It is also very important not to weaken the composite part’s structural mechanical properties with the integration of additional devices in charge of wireless communication.

280

Smart Textiles for In Situ Monitoring of Composites

Currently, the Near-Field Communication (NFC) seems promising and could be used to develop the wireless SHM concept. NFC is a set of communication protocols that enable two electronic devices, one of which is usually a portable device such as a smartphone, to establish communication by bringing them within 4 cm (1.6 in) of each other [10]. NFC devices are used in contactless payment systems, similar to those used in credit cards and electronic ticket smartcards and allow mobile payment to replace/supplement these systems. This is sometimes referred to as NFC/CTLS (Contactless) or CTLS NFC. NFC is used for social networking, for sharing contacts, photos, videos, or files [11], NFC-enabled devices can act as electronic identity documents and keycards, NFC offers a low-speed connection with simple setup that can be used to bootstrap more capable wireless connections. Like other “proximity card” technologies, NFC employs electromagnetic induction between two loop antennas when NFC-enabled devicesdfor example a smartphone and a printerdexchange information, operating within the globally available unlicensed radio frequency ISM band of 13.56 MHz on ISO/IEC 18000-3 air interface at rates ranging from 106 to 424 Kbit/s. Each full NFC device can work in three modes: • • •

NFC card emulationdenables NFC-enabled devices such as smartphones to act like smartcards, allowing users to perform transactions such as payment or ticketing; NFC reader/writerdenables NFC-enabled devices to read information stored on inexpensive NFC tags embedded in labels or smart posters; NFC peer-to-peerdenables two NFC-enabled devices to communicate with each other to exchange information in an ad hoc fashion.

NFC tags are passive data stores that can be read, and under some circumstances written to, by an NFC device. They typically contain data (as of 2015 between 96 and 8192 bytes) and are read-only in normal use, but may be rewritable. Applications include secure personal data storage (e.g., debit or credit card information, loyalty program data, personal identification numbers, contacts). NFC tags can be customencoded by their manufacturers or use the industry specifications. Therefore, the embedded sensors should be equipped with the inexpensive rewritable NFC tags and the microcontroller card able to write the sensor’s output to the NFC tag. All those devices have to be integrated to a composite part together with the sensor. An NFC reader/writer (on smartphone typically) able to read the information provided by embedded NFC tag will then be used to collect the information generated by the sensor in real time and to send them further to a cloud for data treatment. Bluetooth communication devices could be another possible solution for wireless SHM. NFC and Bluetooth are both relatively short-range communication technologies available on smartphones. NFC operates at slower speeds than Bluetooth and has a much shorter range, but it consumes far less power and does not require pairing (see comparison in Table 3.29) [12]. It is obvious from the previous table that the main advantage of NFC tags is that it does not require the power source. However, the embedded sensor and the added microcontroller device do require the energy source and they should be minimized to not weaken the structure of the composite part.

Structural health monitoring of composite structures

281

Table 3.29 NFC versus bluetooth and bluetooth low energy Aspect

NFC

Bluetooth

Bluetooth low energy

Tag requires power

No

Yes

Yes

Cost of tag

US$0.10

US$5.00

US$5.00

RFID compatible

ISO 18000-3

Active

Active

Standardization body

ISO/IEC

Bluetooth SIG

Bluetooth SIG

Network standard

ISO 13157 etc.

IEEE 802.15.1 (no longer maintained)

IEEE 802.15.1 (no longer maintained)

Network type

Point-topoint

WPAN

WPAN

Cryptography

Not with RFID

Available

Available

Range