Electrically conductive textile coatings

with PEDOT:PSS

Maria Åkerfeldt

Copyright © Maria Åkerfeldt

Research School of Textiles and Fashion Faculty of Textiles, Engineering and BusinessUniversity of BoråsISBN 978-91-87525-39-1 (tryckt)ISBN 978-91-87525-40-7 (pdf)ISSN 0280-381X, Skrifter från Högskolan i Borås, nr. 56

Printed in Sweden by Ale Tryckteam, Bohus 2015

iii

ABSTRACT

In smart textiles, electrical conductivity is often required for several functions, especially contacting (electroding) and interconnecting. This thesis explores electrically conductive textile surfaces made by combining conventional textile coating methods with the intrinsically conductive polymer complex poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate) (PEDOT:PSS). PEDOT:PSS was used in textile coating formulations including polymer binder, ethylene glycol (EG) and rheology modifier. Shear viscometry was used to identify suitable viscosities of the formulations for each coating method. The coating methods were knife coating, pad coating and screen printing.

The first part of the work studied the influence of composition of the coating formulation, the amount of coating and the film formation process on the surface resistivity and the surface appearance of knife-coated textiles. The electrical resistivity was largely affected by the amount of PEDOT:PSS in the coating and indicated percolation behaviour within the system. Addition of a high-boiling solvent, i.e. EG, decreased the surface resistivity with more than four orders of magnitude. Studies of tear strength and bending rigidity showed that textiles coated with formulations containing larger amounts of PEDOT:PSS and EG were softer, more ductile and stronger than those coated with formulations containing more binder. The coated textiles were found to be durable to abrasion and cyclic strain, as well as quite resilient to the harsh treatment of shear flexing. Washing increased the surface resistivity, but the samples remained conductive after five wash cycles.

The second part of the work focused on using the coatings to transfer the voltage signal from piezoelectric textile fibres; the coatings were first applied using pad coating as the outer electrode on a woven sensor and then as screen-printed interconnections in a sensing glove based on stretchy, warp-knitted fabric. Sensor data from the glove was successfully used as input to a microcontroller running a robot gripper. These applications showed the viability of the concept and that the coatings could be made very flexible and integrated into the textile garment without substantial loss of the textile characteristics. The industrial feasibility of the approach was also verified through the variations of coating methods.

Keywords: Textile coating, conductive coating, conjugated polymers, ICP, PEDOT:PSS, textile properties, textile sensor, printed electronics, Smart textiles, poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate)

v

PREFACE

The work included in this thesis was carried out between the years 2010-2012 at The Swedish School of Textiles, University of Borås (Borås, Sweden) and between the years 2013-2015 at the Materials department, Swerea IVF AB (Mölndal, Sweden).

vi

ACKNOWLEDGEMENTS

Thanks to:

My supervisor Pernilla Walkenström, I wouldn’t have done any of this without you!

My co-supervisors during the years: Weronika Rehnby (Naturskyddsföreningen) and Nils-Krister Persson (Smart Textiles) for

initializing the project, Martin Strååt (Swerea IVF), for introducing me to the secrets of being a PhD student and Philip Gillgard (Swerea IVF), for your

support and interest in my work.

My present examiner, Vincent Nierstrasz (University of Borås), for your encouragement.

My former examiner, Mikael Rigdahl (Chalmers University of Technology), for your curiosity.

My mentors and colleagues: Anja Lund, for great inspiration, Veronica Malm, for (above all!) being a friend and Erik Nilsson, for fun collaboration.

The skilled technicians and specialists that have helped me throughout my work (I hope I haven’t forgotten too many of you on this list): Maria

Stawåsen, Catrin Tammjärv (University of Borås); Anders Kvist (Chalmers university of Technology); Ann Stare, Simonetta Granello, Desiré Rex,

Marie-Louise Helgee, Eva Carlbom, Karin Christansen, Lars Eklund, Bengt Hagström and Hans Grönquist, (Swerea IVF).

All of my great colleagues, both present and former, at Swerea IVF and The Swedish School of Textiles (University of Borås).

The best friends in the world and my supportive family.

And, for the financial support for this work: Sparbanksstiftelsen Sjuhärad, VINNOVA (through the Smart Textiles initiative) and University of Borås.

Avhandlingen tillägnas särskilt Syster Yster, Zäta och Anita Dahrén.

vii

LIST OF APPENDED PAPERS

I. Åkerfeldt, M., Strååt, M., & Walkenström, P. (2013). Electrically conductive textile coating with a PEDOT-PSS dispersion and a polyurethane binder. Textile Research Journal, 83(6), 618-627. doi: 10.1177/0040517512444330

II. Åkerfeldt, M., Strååt, M., & Walkenström, P. (2013). 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), 2164-2176. doi: 10.1177/0040517513487786

III. Åkerfeldt, M., Nilsson, E., Gillgard, P., & Walkenström, P. (2014). Textile piezoelectric sensors – melt spun bi-component poly(vinylidene fluoride) fibres with conductive cores and poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate) coating as the outer electrode. Fashion and Textiles, 1(1), 1-17. doi: 10.1186/s40691-014-0013-6

IV. Åkerfeldt, M., Lund, A., & Walkenström, P. Textile sensing glove with piezoelectric PVDF fibres and printed electrodes of PEDOT:PSS. Manuscript accepted for publication in Textile Research Journal, 2015.

Conference proceedings (not included in the thesis):

Åkerfeldt, M., & Strååt, M. (2011). A Rheological Study of a Textile Coating Paste Containing PEDOT:PSS. Paper/poster presented at the Nordic Rheology Conference, Helsinki, Finland.

Åkerfeldt, M. (2013). The influence of ethylene glycol on the properties of electrically conductive textile coatings obtained with PEDOT:PSS. Paper/poster presented at the 13th AUTEX World Textile Conference, Dresden, Germany.

Åkerfeldt, M. (2014). Towards sceen-printed electronics for smart textile applications with PEDOT:PSS. Paper/oral presentation presented at the 14th AUTEX World Textile Conference, Bursa, Turkey.

viii

CONTRIBUTION TO APPENDED PAPERS

I. The author planned the experiments together with co-authors, conducted the experimental work, wrote the first draft of the article and finalized it together with the co-authors.

II. The author planned and conducted the experimental work, wrote the first draft of the article and finalized it together with the co-authors.

III. The author planned and conducted most of the experimental work; except for the fibre spinning that was conducted by Erik Nilsson, wrote most of the first draft of the article and finalized it together with the co-authors.

IV. The author planned and conducted the experiments in collaboration with Anja Lund, wrote most of the first draft of the article and finalized it together with the co-authors.

ix

TABLE OF CONTENTS

Abstract ________________________________________________ iii

List of appended papers ___________________________________vii

Introduction _____________________________________________1Aim _______________________________________________________2

Background _____________________________________________3Textile coatings _____________________________________________3

Fabrics ________________________________________________________ 4Coating methods_________________________________________________ 5

Conductive materials ________________________________________9Metals ________________________________________________________ 12Carbons ______________________________________________________ 13Intrinsically conductive polymers (ICP) _____________________________ 13

Smart textiles ______________________________________________15Sensing _______________________________________________________ 16Interconnecting _________________________________________________ 17

Textile coatings with PEDOT:PSS __________________________19Materials _________________________________________________19

Fabrics _______________________________________________________ 19PEDOT:PSS ___________________________________________________ 19Conductivity enhancer ___________________________________________ 21Binders _______________________________________________________ 22Rheology modifier ______________________________________________ 23

The coatings _______________________________________________25Electrical resistance _____________________________________________ 25Surface appearance/Microstructure _________________________________ 27Textile mechanical properties _____________________________________ 29

x

Textile application _______________________________________34Active fibre sensors _________________________________________34

Electroding ____________________________________________________ 34Interconnection _________________________________________________ 36

Conclusions ____________________________________________38

Future work ____________________________________________40

References _____________________________________________41

1

Introduction

Most textiles consist of polymers with very low conductivity, their surface resistivity being typically ≥ 109 Ω/square1. This leads to unappreciated static cling and electrostatic discharge (ESD). To avoid these features, the conductivity of textiles are increased by adding metals and carbon compounds either as particles, fibres or yarns1, 2.

The interest in conductive textiles was renewed when the concept smart textiles emerged some fifteen years ago3. Smart textiles are defined as textiles that can sense and react to environmental conditions or stimuli4. Important functions to build up such a system include sensing, actuating, powering, communicating, data processing and interconnecting; functions that often require electro-activity5. Miniaturization of electrical components and new materials development enabled the incorporation of electro-activity with textiles3.

However, many electro-active components and materials do not meet the criteria for textiles regarding stability, comfort, washability, flexibility, safety for the environment, or even safety for a potential wearer if used in garments. To meet as many of these criteria at once is today the main challenge for material researchers within the field. Since textiles are polymeric, suitable materials may be intrinsically conductive polymers (ICP), a class of polymers that allow for electron mobility within the molecules.

Another problem with many smart textiles is their lack of compatibility with the processes used in the textile industry. This has made commercialization of the products difficult. The market success of smart textile products is largely dependent on the possibility of industrial-scale production.

Coating methods are industrial technologies used to adapt the properties of textiles to those required for technical and specialty applications. The variety of methods allows for good adaptation to end-use requirements. As a final processing (finishing) step, these adaptations do not need to be made prematurely, such as can be the case with fibre alterations. Printing, which is basically a patterned coating, also offers very precise control of the addition of functional substances and is therefore advantageous from a sustainability perspective. In combination with ICPs, textile coatings can offer a route to safe, flexible and more economical conductive textiles. A readily available ICP is poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate) (PEDOT:PSS),

2

which also comes in a dispersion compatible with many industrially used coating polymers.

AimThe aim of the work presented in this thesis was to obtain organic electrically conductive textile coatings, using industrial methods and PEDOT:PSS as the conductive material. PEDOT:PSS was chosen because of its availability, stability, easy processing and high conductivity compared to other ICPs.

The first part of the work aimed at optimizing the coating parameters with respect to the surface resistivity of the coated textiles. To achieve this, Papers I and II included comparative studies of knife-coated woven poly(ethylene terephthalate) (PET) textiles with coating formulations of different compositions, different amounts of coating and variations in the film formation process. Studies of shear viscometry, microstructure/surface appearance and textile mechanical properties were also included in the optimization.

The purpose of Papers III and IV was to investigate possible applications of the PEDOT:PSS coatings, in particular as outer electrodes and interconnections for fibre sensors made from melt-spun piezoelectric poly(vinylidene fluoride) (PVDF). For these applications, pad coating and screen printing were better suited coating methods than the previous knife coating. Another aim of these papers was therefore to study the influence of these coating methods on the coated textiles, using similar characterization methods as previously.

The work in Paper IV specifically addressed the possibility to obtain a comfortable and washable smart garment, using only non-metallic electroactive materials and industrial textile processes. Thus, the paper included a proof-of-concept study using the piezoelectric fibres as sensors and the PEDOT:PSS coating as printed interconnections in a glove where finger motions were used to control a robot gripper.

3

Background

Electrical conductivity in textiles can be achieved in many ways, depending on the desired level of conductivity or specific geometrical, durability and process demands. Textile fibres can be spun with conductive particles throughout the fibres or only in the cores6, 7 and metallic yarns inserted in woven or knitted constructions 8. The focus of the work presented here was to make electrically conductive textile surfaces and this chapter aims at describing the prerequisites for this. Since the area is multi-disciplinary it requires a basic understanding of concepts such as textile coating, conductive materials and smart textiles.

Textile coatingsA textile coating is essentially a material layer adhered to a textile structure. In its most archetypical sense, polymers are applied in the form of a thickened dispersion, a paste, through a spreading technique onto a woven fabric, rather similar to the buttering of bread. In its wider sense, textile coating methods may include any method functionalizing textile surfaces, including anything from basic textile finishing to plasma, dyeing/printing methods to lamination and nonwoven bonding techniques. In this case, textile surfaces refer to any material intended for textile use, including loose staple fibres, filaments, yarns, braids, nonwovens, woven, warp- and weft-knitted fabrics of both organic and inorganic origin.

Textile coatings are used to add or alter the functionality of the textiles. The functionality is often the property of a substance that does not form covalent bonds or adhere to the fabric on its own. Polymeric binders are thus required to form enough secondary bonds for adhesion. Polymers may also be the functional substances for some properties, e.g. hydrophobicity or increased rigidity. The so-called binder polymers generally come in the form of waterborne dispersions, into which functional materials and auxiliaries are added. The coating binder can be any polymer enabling enough adhesion; the only prerequisites being compatibility with the coating method and the end-use purpose. Common polymer groups used include rubbers, polyvinyl chloride, silicones, acrylates and polyurethanes (PU)9.

4

Fabrics

The variety of coating methods allows for almost any textile to be coated, whether fibres or yarns, knitted or woven, of synthetic or natural origin10. Conventional, continuous coating methods will subject the fabrics to substantial tension, basically requiring woven textiles as these are comparatively dimensionally stable. Fabrics made from staple fibres, especially cotton, were the most commonly used for coating until the 60s, but now fabrics made from synthetic filaments are predominating the industry due to their strength and easy processing11.

Three basic, or plain, types of textile constructions are illustrated in Figure 1: a woven, a weft-knitted and a warp-knitted fabric. The fabrics used during the course of this project were both woven and knitted, but the appended Papers include only studies of plain woven and warp-knitted fabrics.

A plain-woven fabric is constructed so that each weft yarn goes between every warp yarn, where every other weft insertion takes the opposite sides of the warp yarns. Since this locks every yarn in place this is also the most dimensionally stable woven construction. The density of the fabric depends on the thickness (e.g. dtex) of the yarns and how tightly the yarns are packed. This also influences the rigidity of the fabric.

Weft-knitted fabrics can be made by flat- and circular knitting techniques. In principal, a single yarn bobbin can be used because it is made to form loops (held by needles) into which new loops of the same yarn can be formed. The yarn is led from side to side in flat-knitting and circularly upwards in circular knitting. Knitted fabrics and plain weft-knits especially, tend to be easily deformed. This makes the fabrics easy to fit onto a human body for example, but may be a drawback when stability and rigidity is desired.

Warp knitting can be described as an intermediate between weaving and weft knitting, but is probably the most versatile construction technique of textile fabrics; the fabrics can be anything from the finest lace to the most robust automotive interior. Yarns are vertically placed side-by-side, similar to the warp yarns in weaving, and placed in guiders that make them interlock with each other horizontally. The fabrics often mechanically outshine woven and weft-knitted fabrics. Synthetic filaments are basically required for this construction in order to avoid entanglements and a high degree of elasthane can be incorporated. This makes them ideal for many high-tech applications, such as sports, automotive interior, personal protective equipment (PPE), geo- and greenhouse fabrics etc., although not always the most comfortable choice for everyday wear.

5

Figure 1. Illustration of plain textile constructions: a) woven, b) weft- and c) warp-knitted.

Coating methods

The coating methods used in the work included in this thesis were: knife coating, pad coating (impregnation) and screen printing; these are also the focus of this section. Laboratory versions of the coating methods were used for the experimental work; up-scaling would require adaptation of the parameters11.

Knife coating

Knife coating, or direct coating, is probably the most widely used technique within the textile coating industry. The principle is simple, with a blade smearing out a thickened polymeric formulation across a moving textile substrate11. If the substrate rests on a roll beneath the blade the technique is called knife-over-roll, see illustration in Figure 2. Other set-ups are e.g. knife-over-air (or floating knife) and knife-over-blanket. The different set-ups are mainly of consequence for the penetration of the coating into the textile, knife-over-roll yielding the highest degree of penetration and floating knife the lowest12. Blade angle, gap height, coating speed and the shear viscosity of the formulation are also important for the degree of penetration which, in turn, will affect the adhesion of the coating to the substrate. These are in turn affected by tension, air flow, pressure, temperatures, time, relative humidity etc.12.

6

Figure 2. Schematic illustration of knife-over-roll coating.

Pad coating

Pad coating, also known as impregnation, pad-mangle, dip coating and saturation, is basically a dipping procedure where excess formulation is squeezed out of the fabric by rollers. A simple version of the arrangement is illustrated in Figure 3. The method allows for the coating to penetrate into the interstices in the fabric and between the fibres in the yarn. The coating will thus be within and on the entire fabric, whereas most other coating methods lead to a more one-sided coating. The amount of applied material to the fabric will depend on coating viscosity as well as the speed by which the fabric moves through the coating liquid; the time the fabric is in the liquid is referred to as dwell time9.

From a fluid-mechanical point of view, the pad-mangle coating process can be viewed as a combination of dip coating and a forward two-roll coating process, see Figure 3. The dip coating is mainly governed by substrate speed, surface tension and viscosity of the formulation (with water as solvent, the effect of drying during the process is negligible); the roll coating will be influenced by the nip pressure, roll speed and coating viscosity. It is difficult to determine the shear forces applicable during the entire process due to the great number of process parameters, but theoretically they have been appreciated to reach instantaneously extremely high values, of 103 s-1. This is mainly during pulling the substrate from the bath and in the turbulent areas surrounding the nip of the rollers, in the rolling bank of the coating formulation just before the nip and in the film-splitting meniscus region, i.e. where the rollers part from each other13.

7

Figure 3. Schematic illustration of pad-mangle coating.

Screen printing/coating

Rotary screen printing is one of the most common printing methods in the textile industry2. It is efficient, fast and versatile, especially in comparison to older methods, e.g. engraved roller printing and flat-bed screen printing.

The basic principle behind both rotary and flat-bed screen printing is that the printing formulation is pressed through a screen onto the substrate, illustrated in Figure 4. Normally, the screen is printed with the inverse of the desired pattern with an UV-curable ink, but the method is also used for coating, then with an open screen mesh. As can be foretold by their names, rotary screen printing entails a cylindrical screen, often made of perforated nickel, whereas the screen is flat in flat-bed screen printing, and often made out of polyester mesh9.

For industrial rotary screen printing, the fabric rests on a rubber mat that moves the fabric forward under the rotating screen. Since the friction between the fabric and screen is low, the method is suitable even for delicate, stretchable, lightweight textiles or fabrics with uneven surfaces9. The print is squeezed out through the cylinder by a squeegee or a rod, the latter if the table beneath the substrate is magnetic11. With the rotary screen printing method the printing can be performed continuously without having to move the screen exactly into a new place, as with flat-bed screen printing. The flat-bed technique is however still commonly used industrially for individual printing of textile products, e.g.

8

t-shirts and tablecloths; it is also favourable for laboratory studies of screen coating.

Figure 4. Schematic illustration of manual flat-bed screen printing.

The amount of coating deposited on the fabric, as well as the quality of the print, is determined by the viscosity of the formulation, the thread density of the screen (or screen mesh number), the pressure from the rod or squeegee, the speed of the process and if there is an angle of the squeegee12. The formulation will be deposited onto the fabric as dots that should merge together as the water (or solvent) evaporates. For some formulations, the attachment of a whisper blade may be necessary to aid this process11.

Film formation

The film formation is the drying and annealing of the coating formulation after deposition. This process is sometimes divided into four stages, as illustrated in Figure 5: first, the particles are randomly distributed in the medium; secondly, water evaporates from the formulation and the polymer particles are packed together; thirdly, the polymer particles deform for a more thermodynamically favoured structure and in the fourth phase the polymers interdiffuse across particle boundaries and sometimes even chemically crosslink. Stages two and three is often referred to as drying in an industrial context and the last stage may thus be called curing or annealing, although these terms are highly dependent on where and by whom they are used. Drying is normally performed at a lower temperature to ensure that the evaporation of solvent (e.g. water) does not cause irregularities in the coating, whereas annealing is performed in higher temperatures to boost the interdiffusion or, when possible, catalyst the chemical reaction necessary for crosslinking.

For waterborne coatings, the film formation process is often the most energy-demanding step in the production.

9

Figure 5. Illustration of four stages of film formation.

Conductive materialsElectrical conductivity, i.e. electron mobility, is a gradual property. Figure 6 illustrates a typical conductivity/resistivity scale with corresponding materials and applications for different levels.

10

Figure 6. Electrical resistivity scale

Basically, only three classes of materials are used on textiles for conductive purposes: metals, carbon allotropes and intrinsically conductive polymers (ICP). For semi-conductive and antistatic applications, ionomers and silicones may be used as well14, 15.

All mixtures of insulating materials with conductive materials, including those where conductive particles are added directly to the textile fibre, can be viewed as conductive composites. An often encountered feature of conductive composites is the electrical percolation behaviour 16, 17. The percolation behaviour means that the conductivity, or inverse resistance, will increase drastically at a certain concentration of conductive material. The concentration where the conductivity increases the most is called the percolation threshold. The particles may have clustered at a lower concentration than the threshold concentration, but the clusters were not sufficiently close to form conductive pathways, i.e. the electrons were not able to jump between particles throughout the material. At the threshold concentration the particles, or clusters, are close enough for the electrons to find pathways throughout the insulating matrix. Above the threshold concentration, more potential pathways may

11

appear, but since the amount of excited electrons is limited to the material properties, the conductivity will not increase substantially. Figure 7 illustrates the corresponding percolation behaviour on the electrical resistance of a typical composite. The shape and size of the conductive material will strongly influence at what concentration the percolation occurs18.

Figure 7. Illustration of a conductive polymer composite: a) below the percolation threshold and b) above the percolation threshold; c) a schematic diagram showing electrical percolation behaviour on the measured resistance.

12

Metals

Originally, thin wires and flakes of metals were added to textiles for decorative purposes, but it was not until the development of vapour deposition in the 1960s that their electrical conductivity was utilized and metals found more widespread use in the textile industry. Silver and copper are amongst the most conductive elements, but aluminium, nickel, tin, steel, gold etc. are also used for conductivity. Although metals have some drawbacks, such as their lack of flexibility, their price and their environmental effects, they are still the only viable alternatives for applications where high conductivity is required.

The application methods for metals can be roughly divided into four areas: binder coating, vacuum deposition, sputtering and electroless plating9.

The binder coatings are similar to conventional textile coatings, using polymers carriers and incorporating metal particles. Since the insulating polymer matrix will counteract the conductivity, several routes to facilitate the contact between the conductive particles, without increasing the mass of metals added, are being explored. Most include increasing the surface area and playing with different shapes and sizes of the metal particles: oriented fibres and flakes19, metal-coated mica particles and smaller (e.g. nano-) particles20. Other properties of the coatings will largely depend on the choice of binder polymer.

Vacuum deposition requires that the metal source (powder/wire) is placed in vacuum and heated until it starts to evaporate (≥ 1000°C) and then allowed to condense onto the cool textile substrate.

Sputtering also involves a vacuum chamber, but one that contains argon (or another inert gas), an anode with the function of substrate holder and a cathode working as source of metal particles. An electrical potential induces ionization of the argon gas; the ions bombard the metal on the cathode and metal ions and atoms are thus sputtered onto the substrate where they condense.

Electroless plating uses a chemical redox reaction to adhere a metallic film to the substrate. The dielectric textile substrate is immersed in a solution containing a reducing agent and metal ions, resulting in a homogenous layer of metal deposited onto the textile21.

Metallization techniques, such as plating and sputtering, both have the advantage and the disadvantage of yielding entirely metalized fabrics, meaning excellent conductivity but also stiffer fabrics with reflective surfaces22, 23. Metal particles (e.g. Ag and Cu) may also leak out into the environment and cause damage, especially if these materials are used in applications where washing or direct wear (abrasion) is part of the life cycle24, 25.

13

Carbons

Materials that only consist of carbon atoms can have a wide range of conductivities, from the insulator diamond to conductors such as carbon black (CB), graphene and carbon nanotubes (CNT). The level of conductivity will depend on the degree of delocalized electrons; thus making graphitization and purity of the carbon compounds important factors.

For textile surface deposition, the most common method of application is that of CPC coating, a mixture of conductive carbon particles, often CB or CNT, and thermoplastic or thermoset polymer binder. Although sputtering with carbon is also a possibility, this has mainly found its application as a preparatory finishing before imaging with scanning electron microscopy.

CPC coating offers moderate to good conductivity, depending on type of carbon particle, degree of loading and degree of dispersion. Considering geometry, CPCs based on CNTs should theoretically have lower percolation thresholds than those based on CBs, but their often-encountered fragility and processing difficulties does not always make this the case in reality. CBs are on the other hand always more or less agglomerated which can be an advantage for their formation of conductive networks.

The colours of carbon coatings are different shades of black and for a stable conductivity, the carbon fillers require a rigid coating in order to not drift apart when used. On the other hand, with a less rigid coating, this change in conductivity can be used as a resistive strain sensor26. Mostly however, carbon coatings are deposited in thick layers that will alter the mechanical properties of the textile to a great extent.

Intrinsically conductive polymers (ICP)

Polymers are normally insulating, but during the 1970s Shirakawa et al. discovered that polyacetylene could be made conductive by extensive iodine doping27. It was subsequently realized that several other conjugated polymers, such as polyaniline, polypyrrole and polythiophene, could be doped to conductivity as well (chemical structures of the most common ICPs are included in Figure 8). However, the instability and processing difficulties, as well as the often dark or intense colour, of the ICPs almost shattered the newly arisen dreams of all-plastic electronics. Decades of intense research on synthesis routes, substitutions and regioregularity, i.e. the position of the double bonds and specific groups, resulted in better combinations of properties and commercial applications started to arise.

14

Figure 8. Chemical structures of common ICPs.

For the textile sector, the method of polymerizing polypyrrole, polyaniline and poly(ethylene dioxythiophene)28, 29 in situ on textile substrates has been the most relevant. The first patent is from 199030 and products based on this technology are now commercially available, such as EeonTex™ from Eeonyx31. However, the in situ polymerization method requires stepwise addition of polymer and dopant under controlled circumstances. This can be achieved in a process similar to jig-dyeing of textiles32, but under strongly acidic conditions and with the drawbacks of excessive washing as necessary post-treatment. The problem of atmospheric instability also remains for these composites29, 33.

Other commercially available conjugated polymer systems include dispersions or solutions of fully polymerized polyaniline34, polypyrrole35 and PEDOT:PSS36, 37. ICP-coated latex particles are also possible alternatives with potential benefits of offering a favourable microstructure for the conductive network. These water-based systems are obviously compatible with the water-based systems used in industrial textile coating; this was also the starting-point for the work included in this thesis.

Smart textilesThe interest in electro-active textiles took on about fifteen years ago with the dawn of smart textiles. Initially, smart textiles were primarily design projects,

15

achieved by simple addition of electrical components to textile products. The bulkiness of the electrical components and their incompatibility with textile use (e.g. washability) soon evolved the research towards the integration of smart functions to the textiles themselves.

But what are smart textiles? The most commonly encountered definition is based on that made by Tao regarding smart materials in 20014:

“Smart materials and structures can be defined as the materials and structures that sense and react to environmental conditions or stimuli, such as those from mechanical, thermal, chemical, electrical, magnetic or other sources. According to the manner of reaction, they can be divided into passive smart, active smart and very smart materials. Passive smart materials can only sense the environmental conditions or stimuli; active smart materials will sense and react to the conditions or stimuli; very smart materials can sense, react and adapt themselves accordingly. An even higher level of intelligence can be achieved from those intelligent materials and structures capable of responding or activated to perform a function in a manual or pre-programmed manner.”

Applying these definitions on textiles, a passive smart textile is thus a sensor, an active smart textile an actuator and a very smart textile is one that can adapt its response to the environment in a nonlinear fashion. Only the last definition corresponds to Kirstein’s definition of a truly smart textile38. In some circumstances, the term smart textiles is used in a much more general sense; including advanced, functional and technical textiles39.

In order to achieve a very smart textile product/garment, a whole system of components offering required partial functions may have to be incorporated. The partial functions have been appreciated by Schwarz et al. as: sensing, actuating, powering/generating/storing, communication, data processing and interconnection5. Out of these functions it is only the data processing that has yet to be integrated into the textile product; the other functions have been integrated with varying degrees of success. As a further simplification of a typical smart textile system, the functions of powering/generating/storing, communication and data processing may be collected under the term of adaption, since these will collectively answer for the response the user experiences from the product. With this, the system can be divided into: sensing, actuating, adaption and, of course, interconnections between the three, as illustrated in Figure 9.

16

Figure 9. Illustration of a smart textile system/garment including essential parts of such a system.

Sensing

A large and important sub-section to smart textile research is the area of textile sensors. This is partly because the sensing ability is the first and most essential part of any smart textile system, but also because textile sensors can find a widespread range of applications on their own. This includes the medical, sports, PPE, geo-protection, military and aerospace sectors, where sensing and monitoring are already important and would only be made more efficient if integrated to the materials themselves.

Different types of textile sensors have been developed and can be found in the literature40. Sensors are divided into two categories according to their response: passive, which consumes energy and active, which produces energy. Passive sensors respond to the environment by change of a property, such as a change in resistance for resistive sensors. In order to be used, the changing property requires recording or connection to an actuator. Active sensors on the other hand, will yield a physical signal that can be used at once.

17

Sensors are also classified according to what triggers them: optical, chemical, temperature, humidity sensors as well as mechanical sensors that respond to force, pressure and strain. Strain sensors can be used for physiological monitoring such as measurement of breathing, heart rate or frequency of a movement.

A typical passive strain sensor is that of a resistive sensor, where the resistance of the material is changed upon strain26. This is also the most common textile sensor in the literature. The change in resistance can occur due to a geometrical change of the device, filler disconnection or a change in the contacting of the conductive material. Since it is the change in resistance, and not high conductivity, which is required for this application, textile resistive sensors are often carbon- or ICP-based CPCs.

An active strain sensor requires a rather special material, such as one that can produce a voltage upon mechanical deformation. This property is called piezoelectricity and is seen in certain ceramics, but also in fluoropolymers, such as poly(tetrafluoroethylene) (PTFE) and poly(vinylidene fluoride) (PVDF), as discovered by Kawai in 196941. PVDF is today being studied as a fibre material, both with electro- and melt spinning, in order to achieve inherently active textile fibre sensors. The contacting of such fibre sensors was one of the topics studied in the work included in this thesis.

Interconnecting

Regardless of how the sensing occurs, the signal needs to be transported to a processing unit, or from a processing unit to an adapting function, depending on how elaborate the system is. This requires stable and reliable conductivity at the very least, perhaps even high conductivity depending on the strength of the signal and the length it is required to travel. Metal wires, or even cables, are therefore often turned to, but are seldom stretchable, or comfortable, enough for textile applications. The material development during recent years has opened up for fibres, yarns and printed tracks that offer more textile solutions to this problem.

Three types of solutions were distinguished by Zeng et al. in a recent review42: One approach was to incorporate conductive metal wires or metal-plated yarns into knitted structures, where the yarns are coiled. The results of these materials seem promising, but processing and environmental/health concerns still remain. Metals can also be added in a wrinkled form, encapsulated in a thick layer (1 mm) of poly(dimethyl siloxane) to textile surfaces, but this approach suffers from problems with flexibility and delamination. A third solution is to add

18

stretchable conductive materials, such as CPC with carbon nanotubes, metal films or particles in elastomeric membranes or ICPs in stretchable composites. The review stated that the conductivity of this last group generally is too low for integrated circuits and that repeated strain may lead to a further, substantial decrease in conductivity.

19

Textile coatings with PEDOT:PSS

The previous chapter described the background and prerequisites of the work included in this thesis. This chapter aims at describing the coated textiles of this work: the materials, methods and the results of the studied properties.

Materials

Fabrics

Plain-woven fabrics were used for Paper I-III and a highly stretchable, plain warp-knitted fabric was used for Paper IV. As described in the previous chapter, a plain-woven fabric is very dimensionally stable and thus suitable for basically any coating method; the warp-knitted fabric however, required a coating method with little or no tension and an increased flexibility of the conductive coatings themselves.

Recently developed piezoelectric PVDF fibres43 were used as weft insertions, with a polyethylene (PE) monofilament warp yarn, in the plain woven fabric used in Paper III. The piezoelectric fibres were also twisted into a yarn and embroidered onto the knitted fabric used in Paper IV. The low surface energy and wettability of PVDF was expected to pose a problem for the adhesion, especially since the coating was water-based.

PEDOT:PSS

The polythiophene derivative poly(3,4-ethylene dioxythiophene) (PEDOT) was developed in the late 1980s by Bayer AG and found to have the then rare combination of high conductivity, transparency and environmental stability44. Its insolubility still posed a problem but was later circumvented by pairing it with poly(styrene sulfonate) (PSS), a water-soluble polyelectrolyte that also functions as a charge-balancing dopant. In Figure 10, the chemical structure

20

of the PEDOT:PSS complex is shown to the left and its secondary structure indicated to the right. In the chemical structure, the upper structure is the PEDOT and it is interesting to notice the positions of the dioxy-rings and how they uphold the regioregularity of the thiophene. It is also worth noticing the oligomeric, rather than polymeric, lengths of the PEDOT molecules from the secondary structure.

Figure 10. Structure of PEDOT:PSS; primary and secondary structure.

The addition of PSS to the structure made it possible to commercially offer PEDOT:PSS dispersions, but also resulted in a high amount of water, ca 90-95%, adsorbed to the PSS in the products; in spite of this, dried PEDOT:PSS films have exceptional stability to humidity and pH. PEDOT:PSS is inherently blue and films made from it tend to have a bluish tint. If however, they are thin enough, corresponding to ≤ 200 nm pure PEDOT:PSS, the visible light transmission through the film will be at least 85%45, 46. The coatings in this work were all in different shades of blue.

The conjugated polymer system PEDOT:PSS was first used for antistatic films in the photographic industry but soon found other applications as PEDOT:PSS dispersions became commercially available. Mixtures with latex (or elastomers)47, 48 and textile applications49-51 are also emerging within the literature.

Throughout this work, the PEDOT:PSS dispersion used was Clevios PH1000, from Heraeus Clevios GmbH, Germany. This grade is supposed to offer the highest conductivity (≤ 1000 S/cm) of the range of ready-made Clevios P. Some properties of Clevios PH1000 are listed in Table 1.

21

Table 1. Properties of the PEDOT:PSS dispersion Clevios PH1000 according to supplier.

Solids content 1.1 wt-%Viscosity (20°C) 33 mPa∙sAverage particle diameter 30 nmPEDOT:PSS ratio 1 : 2.5

Conductivity enhancer

The conductivity of PEDOT:PSS films has been found to be enhanced by the addition of so-called “secondary dopants”, i.e. certain high-boiling solvents such as glycerol52, sorbitol53, dimethyl sulfoxide (DMSO), N-methylpyrrolidone (NMP) and ethylene glycol (EG)54, 55. The increase in conductivity is attributed to swelling of the PEDOT:PSS grains, facilitating contact and electron transport between the conductive clusters, but it has also been suggested that it leads to a change in the conformation to more extended PEDOT-polymer chains. Figure 11 illustrates how the high-boiling solvent EG, used in this project, is thought to affect the clusters.

Figure 11. Illustration of the swelling of PEDOT:PSS clusters, tertiary structure, upon the addition of EG.

22

The chemical structure of EG is included in Figure 12. Compared to the polymers otherwise present in the coating formulation the EG molecules are very small, with a molar mass of only 62.1 g/mol, and are also water-soluble. The boiling point for EG is 198°C and it was therefore assumed that it would not evaporate during the drying and annealing but instead, to some extent, remain in the finished coatings.

The viscosity of EG (at 20°C) is 16.1 mPa∙s. Simpson et al.45 suggested adding about 5 % of the amount PEDOT:PSS dispersion, thus for a formulation containing 50 % PEDOT:PSS dispersion, 2.5 % solvent would be added. For the textile coating formulation it was found that an amount of 10 % of the amount PEDOT:PSS dispersion was needed for the best conductivity and that this amount decreased the viscosity of the formulation.

Figure 12. Chemical structure of ethylene glycol, C2H4(OH)2 .

Binders

According to Simpson et al.45, as well as Heraeus own coating guides56, polyurethanes generally show good compatibility with PEDOT:PSS. Initial studies showed that the commercial PU-based textile coating formulation Performax™ 16297G from Lubrizol Advanced Coatings (Belgium) gave good results in terms of low surface resistivity (high conductivity) when combined with the PEDOT:PSS and EG. As a ready-made formulation used in the textile coating industry, the binder also served as a reference for the viscosity as well as for the properties of the finished coating. Performax 16297G is an aqueous dispersion of aliphatic polyester-polyurethane thickened with hydroxyethyl cellulose. The solids content is 32 weight-% according to the supplier.

For Paper IV, another commercial binder system was used, Performax™ XPE 1210 (Lubrizol). This system was based on self cross-linking acrylates and chosen because of its combination of flexibility and durability.

23

Rheology modifier

The low viscosity of the PEDOT:PSS dispersion and EG, as well as the low solids content of the former, indicates that they will have a diluting effect on the formulation. In many coating processes, the viscosity of the formulation should be low enough for the formulation to flow and level out on the rough surface, but it also needs to be sufficiently high in order to ensure that the formulation does not pass through the pores (“bleed through”) and penetrates the fabric entirely during the film formation57. This corresponds to a shear-thinning (pseudoplastic) behaviour of the formulation, where the viscosity is high at low shear rates and lower at high shear rates. A typical, illustrative curve showing such behaviour of viscosity versus shear rate is included in Figure 13.

Figure 13. Typical shear viscometry curve showing a shear-thinning (pseudoplastic) behaviour.

24

There are a variety of rheology modifiers and thickeners available on the market58. Natural thickeners, such as hydroxyethyl cellulose, are mainly active in the aqueous phase of the formulation by adsorption of water and chain entanglement. The result is a pronounced pseudoplastic character and it can be difficult to obtain proper control of this behaviour. Amongst the synthetic thickeners, alkali swellable acrylic emulsions (ASE), hydrophobically modified ASE (HASE) and hydrophobically modified ethoxylated polyurethanes (HEUR) are the most commonly used classes. As their names indicate, the ASE and HASE types require alkali conditions to perform their function and since the PEDOT:PSS dispersion has a pH as low as 2, they are not realistic alternatives for the formulation.

HEUR-type thickeners consist of low-molecular weight polymers of hydrophilic poly(ethylene glycol) chains with hydrophobic end-groups. Their composition enables hydrophobic interactions with the PU-particles in the formulation and the chains to reside in the water phase between the particles, in this way forming an associative network of flower-like micelles, an illustration of this behaviour is found in Figure 14. The rheological properties of the formulation will consequently depend on the hydrophilic chain length and the degree of hydrophobicity of the end-groups of the HEUR59, 60.

For the work presented here, the HEUR used was Borchi® Gel L75N (BGL75N) which has been extensively studied by Orgilés-Calpena et al61, 62. The solids content is 48 weight-%, according to the technical data sheet.

Figure 14. Illustration of HEUR-molecules forming flower-like interactions between polymer particles in dispersion.

25

Shear viscometry of the formulations were included in Paper I. The amounts of PEDOT:PSS dispersion was between 20 and 80 weight-% of the formulations, and the amounts of EG and BG L75N were 10 % of the amount of PEDOT:PSS dispersion, respectively. The viscosity as function of shear rate was correlated to the physical appearance of the samples coated with the various formulations, where the larger amounts of PEDOT:PSS dispersion also resulted in more impregnated, instead of coated, fabrics.

Further viscometry was performed continuously throughout the project in order to verify the suitability of the coating composition to each deposition method.

The coatings

Electrical resistance

Electrical conductivity is, as mentioned, a measure of the electron mobility in a material, with the units of Siemens per metre (S/m). In reality it is calculated from its inverse: the electrical resistivity (Ωm), because it is easier to measure the proportion of voltage to current, i.e. the resistance of the material. Electrical resistivity is an inherent material property but can only be determined for an unknown material if its dimensions are known. When measuring plane materials of undetermined thickness, such as coatings on textiles, it is preferred to approximate it in terms of surface resistivity. This is measured through contacting only the surface of the material, neglecting the thickness entirely, yielding the resistance of a square of any size, Ω/square. The mobile electrons in the material will probably move throughout the material, and not only in its uppermost layer. A thicker layer of material will offer more possible routes for the electrons and thus, if the material is a conductive polymer composite at around the percolation threshold, the thickness can have a significant effect.

26

Figure 15. Schematic illustration of a) the two-point ring probe (Paper I), b) the four-point probe (Papers II-IV). All measurements are in millimeters.

In Paper I, a concentric, two-point ring probe (Warmbier model 880), illustrated in Figure 15, connected to a multimeter (Agilent 24405A) was used for the measurements of electrical resistivity. The samples were placed on an insulating material and a 2.2 Kg weight was placed on top. In accordance with standard CEI/IEC 93:1980, initial fluctuations of the measured values were circumvented by waiting one minute before recording the resulting resistivity.

Paper II-IV continued the electrical measurements with an in-house made four-point measurement probe, specifically designed and hand-crafted for the coated textiles of this project, see illustration in Figure 15. This probe was also connected to a multimeter (Fluke 8846A, USA), in a four-wire resistance mode, but otherwise in a similar measurement set-up as previously. Since it has been suggested that the resistance of PEDOT:PSS does not follow a linear I-V characteristic63, the ohmic linearity of certain samples was accounted for in Paper II. Linearity was obtained at the voltages tested.

Paper I investigated the influence of the amount PEDOT:PSS in the coating, the amount of coating on the substrate after drying and annealing as well as the influence of drying procedure before and after a subsequent annealing step on the surface resistivity. The concentration of PEDOT:PSS was found to be the most influential of these parameters, decreasing the resistivity with five orders of magnitude, and also showed an indication of percolation behaviour, with an approximated threshold concentration at between 1 and 3 weight-%

27

PEDOT:PSS in the coating. The amount of coating followed a similar pattern, but the decrease in resistivity was in total only about one order of magnitude. Any significant influence of drying process could not be ascertained from the results, especially not after the samples had been subjected to annealing.

With the four-point measurements in Paper II the samples were found to have a surface resistivity of up to two orders of magnitude lower than what was determined previously with the two-point measurements. This was accounted for by the inherent contact resistance in the two-point measurement system. Contact resistance is frequently encountered in two-point systems when measuring on conductive materials. All of the samples used in Paper I were re-measured with the four-point system and the trends were found to be the same as those concluded in Paper I, although some became more pronounced with the lower resistivity of the more conductive samples.

Thus, Paper II included results of a very low surface resistivity, of 10-20 Ω/square, for the samples with the largest amount of PEDOT:PSS. The conductivity-enhancing effect of adding EG to the formulation was also confirmed; the difference was more than three orders of magnitude, with a mean value of 130 Ω/square for the samples containing 3.4 weight-% PEDOT:PSS and EG in the coating and 650 000 Ω/square for samples with the same amount of PEDOT:PSS but without EG.

To a large extent the same samples were used in Papers I and II, and although almost a year had passed in between, their conductivity and appearance were unimpaired. Instability is, as previously mentioned, often mentioned as a problem with conjugated polymers, but it is here confirmed that PEDOT:PSS complexes embedded in a PU-matrix do not suffer from this. New samples were also produced for Paper II as control and proved to exhibit the same properties as those made more than a year earlier.

In terms of surface resistivity, the levels achieved in Paper II were sustained throughout the rest of the work. For the study in Paper III, the lowest surface resistivity was 12 Ω/square (without binder) and in Paper IV, which did not concern increased amounts of neither coating nor PEDOT:PSS, the mean value was 57 Ω/square. This showed not only that the coating system itself was reliable, but also that it survived the transition between the different coating methods very well.

Surface appearance/Microstructure

The appearance of the coated textiles was used as a quality indicator and to investigate the detailed deposition of the coatings. Ocular inspection and light

28

microscopy were used for initial appreciations, whereas scanning electron microscopy (SEM) was used for further analysis in Papers I-III.

Three different SEM were used during the course of this work. For Paper I, the SEM was FEI Quanta 200 FEG ESEM used in a low vacuum mode and thereby not requiring additional sputtering. In Paper II, and for some of the preparatory work for Paper III, the low vacuum SEM was JEOL JSM-6610 LV. Paper III included studies with a field emission scanning electron microscopy (FE-SEM) (JEOL JSM-7800 F, Japan), equipped with energy dispersive spectroscopy (EDS) (Quantax X-ray mapping system, Bruker Nano GmbH, Germany), allowing elemental analysis of the samples.

Papers I and II included studies of the surface appearance of the coated fabrics with respect to variations in the coating formulation, the amount of added coating and the film formation. From the SEM-images in these Papers it was derived that the amount solids in the formulation and the amount of coating had a great impact on the surface appearance. The film formation however, showed no visible effects on the appearance of the samples.

In addition to the imaging of the surfaces of the knife-coated samples in Papers I-II, attempts at cross-sectional analysis of them were also performed in order to evaluate the degree of penetration of the coating. Figure 16 shows two images that reveal a large difference between the coatings: sample b) has a distinct coating layer at the top surface, in sharp contrast to sample a) where most of the coating has penetrated into the substrate.

Figure 16. Cross-sectional SEM-images of samples a) coated with a formulation of 60 wt-% PEDOT:PSS dispersion, 28 wt-% binder, 6 wt-% EG and 6 wt % of BG L75N and the other b) coated with a formulation of 20 wt-% PEDOT:PSS dispersion, 76 wt-% binder, 2 wt-% EG and 2 wt-% BG L75N. The solids content of the formulations (including EG) was for a) 19 wt-% and for b) 30 wt-%.

29

Paper III studied the deposition of the coatings following pad-coating of PVDF fibres with PEDOT:PSS coatings with and without the polyurethane binder. The Paper included studies of several important aspects of the distributions and microstructures of the coatings: EDS-mapping of sulphur indicated the influence of shear viscosity on the distributions; the interrelation between the PEDOT:PSS and polyurethane phases was studied at microscale and the influence of harsh mechanical treatment (shear flexing) illustrated.

Textile mechanical properties

The mechanical properties of coated fabrics is difficult to predict because of the complicated interactions between the textile and the coating materials64. Modelling and simulations have been performed for tear65 and shear64 behaviour and good correlations to actual testing results were reported. Rather recent publications have also simply tested textile/composite properties, such as tensile strength and elongation, bursting strength, tear strength, bending rigidity and abrasion resistance, of traditionally coated66, 67, metal-plated23, 68,

69 or in situ polymerized conductive textiles32, 70. The tensile strength of coated fabrics depends more on the strength of the fabric than on its interactions with the coating. Such measurements are thus generally superfluous for comparative studies of textile coatings; this said, tensile testing was performed on the coated plain woven substrates used in Papers I and II, but no significant differences between the samples were found.

Tear strength

The tear strength of coated textiles is strongly influenced by various coating parameters and is also an important property in many of their applications65. Plain-woven fabrics have lower tear strength than baskets and twills of the same material composition, but generally the tear strength is further deteriorated by a coating because the structure becomes less flexible71. However, in a study by Yesilalan et al.67 better tear strength was achieved with coating than without and it was suggested that the coating helped to share the tear load between the yarns. Bulut et al.66 showed that the choice of coating material and its degree of penetration into the substrate were the most influencing factors with regard to the tear strength66.

In Paper II, the tear strength of the coated textiles was tested with the dynamic pendulum (Elmendorf) method, EN ISO 4674-2; for Paper III, the tear strength was evaluated on trouser-shaped specimens in a tensile tester (Instron 4502, UK) according to EN ISO 4674-1B. Witkowska and Frydrych72 studied

30

the correlation between different tear strength methods and found a strong correlation between the results of these two in particular. Figure 17 shows representative examples of torn and untorn trouser-shaped test specimens.

In Paper II, the influence of the coating formulation and the amount of coating on the tear strength was studied. It was found that for two samples, those with the largest amount of binder, the tear strength was lower, as expected. However, for the two other samples tested, with coatings having rather low or low binder content, the tear strength was actually substantially increased. It was concluded that these results could possibly be related to increased ductility and softness following the coating of these samples, since brittleness is the cause for lower tear strength.

Figure 17. Trouser-shaped specimens for tear strength tests, untorn above and torn below. (The specimens in the picture were prepared with different coatings.)

Although the fabric was of a completely different material composition (PE and PVDF) in Paper III, the coating with binder resulted in coated fabric samples with lower tear strength compared to the samples coated without binder, in agreement with the previous results from Paper II.

31

The results of the tear strength testing indicated differences in the sensational properties between the samples. Bending (or flexural) rigidity is a common way to appreciate the stiffness of fabrics, especially with regard to how it feels and is perceived of a person holding it in her hand73. Coating normally imparts increased stiffness to the fabric, simply because the polymers will restrict the fibre and yarn mobility67. The softness of the binder can be appreciated from its glass transition temperature, but this would not be sufficient to describe the sensation of the fabric when coated with the same binder11.

Bending rigidity

Paper II tried to measure the perceived stiffness of the coated fabrics in a systematic way. Textiles are often subjectively judged by their fabric hand and although attempts at objective quantification of this sensation were made from 1930 and onwards, it was not until Kawabata developed an intricate measurement system in the 70s that they became widely used in textile characterization74. An important parameter for this sensation is the bending rigidity, which was shown to greatly influence the end-use performance73.

The bending rigidity was studied in Paper II with the Kawabata evaluation system for fabrics, KES-F-2, pure bending testing. The results showed that the sample with the least amount of binder exhibited extremely low rigidity in comparison to the other samples. Since the rigidity of other samples with low binder content was not nearly as low, this meant that it was not only the binder polymers that influenced the increase in tear strength and the great loss of rigidity. Instead, the change in mechanical properties was probably a synergistic effect between the binder, the PEDOT:PSS and EG.

Durability: abrasion, flexing, cyclic strain and washability

The durability of the coated textiles is of paramount importance for any application, but is evaluated in different ways depending on what the textile should endure. The most basic, scientific approach is to measure the force needed to separate the coating from the textile, i.e. the adhesion. This requires that the coating has formed a strong enough film so that it does not merely break up internally when separation is attempted. The PEDOT:PSS coatings were too thin and internally weak for most conventional adhesion tests and their durability was therefore tested with ISO-standardized textile test methods, also more relevant to their textile character.

Abrasion resistance of fabrics is often tested as a means to predict their ability to withstand wear. Coatings generally increase the abrasion resistance of fabrics because they smoothen and protect the surface structure. With electrically

32

conductive coatings, it is especially vital that this function is not immediately worn off since it would limit their application and purpose significantly.

The study of abraded samples in Paper II found no differences between the obtained loss of conductivity or in changed surface appearances between samples with varying coating formulations and amounts. The differences between their surface resistivities remained at the same relative levels as before abrasion. Although the surface resistivity increased for all samples, it was still lower than expected. The analysis of the surface appearances also confirmed coating remained on the substrate, albeit distorted from the circular motions of the abrasion.

The influence of film formation conditions on abraded samples was also investigated. The result for the samples obtained with a faster film formation process stood out in its pattern regarding surface resistivity and increased significantly faster than the others. An attempt to explain this is that the faster kinetics during the drying of these samples was perhaps not sufficient for full entanglement of the binder polymers, thus leaving more open paths between the PEDOT:PSS complexes. During the abrasion process enough frictional may have developed to finalize the film formation process and closing these paths, thereby decreasing the electron mobility within the coating.

Due to the low influence of abrasion on the samples, a method to further challenge the durability of the coatings was used in the work with Paper III. The test method Determination of resistance to shear flexing and rubbing (EN-ISO 5981:2007) (method B, without pressure foot) was used as this method is much harsher than that of abrasion resistance. In terms of durability, the coating with binder performed significantly better than the coating without binder, resisting both abrasion and shear flexing to a much greater extent. Figure 18 contains micrographs illustrating the difference between the influence of abrasion and shear flexing. Aside from the different fabrics, the abraded sample shows a distorted surface structure, whereas the coating in the flexed sample is distorted throughout. This is also reflected in the effect on surface resistivity of respective test method.

33

Figure 18. Micrographs showing a) the influence of 50 000 martindale abrasion cycles on the coated plain-woven fabric; b) influence of 1 000 shear flexing cycles on PVDF/PE woven fabric. Images adapted from Paper II and III.

Paper IV presented a strain sensing glove, developed as a proof-of-concept and as a model for a smart textile application. Since the glove was made of a highly elastic warp-knitted fabric and the strain sensing function, the durability of the here screen-printed PEDOT:PSS coating was studied in terms of surface resistivity after repeated strain. It was also discussed that a textile garment meant for skin-close wear should require basic durability to washing of all non-removable components. The surface resistivity of the PEDOT:PSS prints was measured after up to 15 wash cycles and showed that they remained within a conductive range, although in order to retain the surface resistivity well below 1000 Ω/square it was necessary to stay within five wash cycles.

34

Textile application

In the latter part of the work included in this thesis (Papers III and IV), the textile coatings were studied from an application perspective. In particular, it was relevant to investigate what functions these textile coatings could fill in smart textile applications.

Active fibre sensorsFor the integration of active fibre sensors in textile applications, such as the piezoelectric fibres based on PVDF (described in the Sensing section), the voltage signals require a conductive medium for electroding and transference to a data collection device or similar. This can be achieved with conductive fibres and yarns, but with respect to contacting, coatings or prints would be preferable. The possibility of using the PEDOT:PSS coatings as that conductive medium was studied in Paper III and IV, with the specific aim to obtain fully organic, flexible and comfortable, active textile sensors. The melt spinning of PVDF into piezoelectric textile fibres was explored by several research groups75-77, including by colleagues at University of Borås78 and Swerea IVF43. The necessary draw ratio and polarization conditions to render the fibres piezoelectric were thoroughly studied in these works. The focus of the papers included in this thesis was the electroding and interconnecting function of the PEDOT:PSS coatings for the fibre sensors.

Electroding

For the piezoelectric PVDF fibres, one electrode can run as a continuous core within the fibres and the other electrode continuously on the outside of the fibres; see the schematic illustration of fibre cross-section in Figure 19. The former was achieved with melt spinning of bi-component multifilament fibres, where PVDF was used as sheath material and the core material was a CPC consisting of HDPE/CB. When spun as multifilaments, these fibres require very little post-treatment to be used as regular textile yarns in actual textile processing, such as weaving, knitting and sewing43.

35

Figure 19. Schematic illustration of bi-component PVDF fibre: a) side-view; b) cross-sectional view, darker areas represents conductive material, i.e. the electrodes.

In order to achieve the outer electrode, many methods are possible: • the yarn can be applied in a woven construction with insertions of

conductive yarn; • the yarns can be sandwiched in a thermoset CPC43, but this will have a

negative effect on the flexibility (textile sensation may be lost);• tri-component fibre spinning, so that both electrodes are inherent parts

of the individual fibres was recently achieved by Martins et al. with monofilaments76;

• coating and printing the yarns, or parts of the yarn, with conductive material.

The work included in this thesis, Papers III and IV, focused on the last suggestion by using the PEDOT:PSS coating as the outer electrode for piezoelectric PVDF yarns.

Figure 20. Picture of the woven PVDF fabric used in Paper III, the grey insertions are the piezoelectric bi-component fibres and the white insertions only PVDF.

36

For Paper III, the PVDF yarn was used as insertions in a plain-woven fabric (Figure 20) that was coated with PEDOT:PSS coatings via impregnation. Two coating formulations were prepared, one with and one without PU-binder, and their performance during tearing, shear flexing and abrasion were compared as discussed in the previous chapter. The impregnation method was chosen in order to maximize the contacted surface of the fibres. The distribution of the coatings within the fibre bundles and their microstructure on individual fibres was also studied with SEM, as discussed in the previous chapter. Piezoelectric characterization showed that both coatings could be used as the outer electrodes, but the textile characterization showed important differences in their mechanical properties. Keeping in mind the low surface energy of PVDF, the direct adhesion was not the problem it was expected to be.

Interconnection

The focus of Paper IV was to apply both the piezoelectric yarn and the conductive coating in a smart textile application, where their inherent flexibility and advantages in textile compatibility could be fully exploited. The concept was proven with a sensing glove, in which the piezoelectric yarn was embroidered into sensors for finger movement and the coating was applied by screen printing to perform the function of both electrode and interconnection. The glove was connected to an external acquisition board and a robot gripper that could follow the finger movements of a test subject wearing the glove, illustrated in Figure 21.

37

Figure 21. Picture showing the prototype sensing glove controlling a robot gripper.

Both papers showed the feasibility of the coatings as part of smart textile systems, but required incorporation of a separate sensing unit. According to the discussion of sensors in the section Smart Textiles, the coatings should, in theory, be able to function as resistive sensors on their own, if an appropriate system for the signal uptake was used. However, with the PVDF fibres, the coatings can be directly integrated to a smart textile system entirely based on organic materials.

38

Conclusions

This thesis presented routes to achieving organic and electrically conductive textile coatings, using the conjugated polymer system PEDOT:PSS. The conductive polymer was used with binder polymers of polyurethane and polyacrylate type in order to obtain adhesion, but also to obtain a system flexible enough for textiles. EG was added to boost the conductivity by swelling the PEDOT:PSS clusters. The viscosity of the coating formulation was continuously adapted to the different coating methods by using a rheology modifier of HEUR-type.

The PEDOT:PSS/EG/textile coating blend showed a clear tendency to electrical percolation behaviour, with a drastic decrease in surface resistivity of the coated textiles at a specific amount of PEDOT:PSS. The addition of EG decreased the surface resistivity with more than four orders of magnitude. Surface resistivity down to 12 Ω/square was achieved with the textile coatings in this work; this is considered a very low resistivity for non-metallic coatings. Tear strength and bending rigidity measurements showed that increased amounts of PEDOT:PSS/EG in the coatings resulted in softer, more ductile and stronger coated textiles.

Influence of coating amount and drying conditions on electrical resistivity and textile mechanical properties were also studied and showed that while the amount of coating was highly influential, especially on the textile mechanical properties, variations in the drying procedure did not have any significant effects. The impact of coating amount was attributed to the coverage of the coating on the fabric.

The coated textiles were found to be durable to abrasion and cyclic strain, as well as quite resilient to the harsh treatment of shear flexing. Regarding washability, the surface resistivity of the coated textiles increased with one order of magnitude after the first wash cycle, but this was without a protective coating, and the surface resistivity remained well below 1000 Ω/square even after five wash cycles.

The conductive coatings were used in strain sensing systems together with bi-component piezoelectric PVDF yarns. The coatings were first applied as the outer electrode on a woven sensor and then also as interconnections in a sensing glove based on highly elastic, warp-knitted fabric. These applications showed the viability of the concept and that the coatings could be made very

39

flexible and integrated into the textile garment without substantial loss of the textile characteristics. The industrial feasibility of the approach was also verified through the variety of coating methods used during the course of this work.

40

Future work

The coatings’ potential use in more applications, e.g. biopotential monitoring electrodes, resistive sensors, heating elements etc., is probably the most interesting area for future studies. This aspect of research could also be used to ensure that the following proposals of study would be performed with a relevant focus.

Although the coated textiles were found to be quite durable with the methods tested, it is preferable that their surface resistivity does not increase at all, or at least that the increase is kept at a minimum. For most actual applications, the prints/coatings would require a protective coating to provide insulation and prevent short-circuits if the coated areas come into contact. This protective coating could also further protect the PEDOT:PSS coating from wear, tear and washing and, used as a base coat, provide increased adhesion. These multi-layered coatings require specific study for each application. Another possible alternative is to study coating binders that could provide one or several of these functions within one layer of coating.

From a sustainability perspective, two aspects of the coatings are especially imperative for their success: more material-efficient deposition methods and more energy-efficient drying. Deposition of the coatings should be performed with little or preferably no waste; stencil and ink-jet printing may therefore be prosperous alternatives. The material deposition efficiency would also lead to better energy efficiency in the drying process. This aspect could be further optimized with more advanced drying systems, such as UV or IR-radiation, but also with PEDOT:PSS systems containing less water.

41

References

1. Wang L, Wang X and Lin T. Conductive coatings for textiles. Smart textile coatings and laminates. Cambridge, UK: Woodhead Publishing Ltd., 2010, p. 155-88.

2. Hatch KL. Textile Science. West Pub., 1993.

3. Smith WC. Overview of textile coating and lamination. Smart textile coatings and laminates. Cambridge, U.K. : Woodhead Publishing Series in Textiles, 2010, p. 3-9.

4. Tao X. Chapter 1: Smart technology for textiles and clothing - introduction and overview. Smart Fibres, Fabrics & Clothing. Boca Raton: Woodhead Publishing Limited, 2001, p. 1-6.

5. Schwarz A, Van Langenhove L, Guermonprez P and Deguillemont D. A roadmap on smart textiles. Textile Progress. 2010; 42: 99-180.

6. Strååt M, Rigdahl M and Hagström B. Conducting bicomponent fibers obtained by melt spinning of PA6 and polyolefins containing high amounts of carbonaceous fillers. Journal of Applied Polymer Science. 2012; 123: 936-43.

7. Strååt M, Toll S, Boldizar A, Rigdahl M and Hagström B. Melt spinning of conducting polymeric composites containing carbonaceous fillers. Journal of Applied Polymer Science. 2011; 119: 3264-72.

8. Cömert A, Honkala M, Puurtinen M and Perhonen M. The Suitability of Silver Yarn Electrodes for Mobile EKG Monitoring. In: Katashev A, Dekhtyar Y and Spigulis J, (eds.). 14th Nordic-Baltic Conference on Biomedical Engineering and Medical Physics. Springer Berlin Heidelberg, 2008, p. 198-201.

9. Sen AK. Coated textiles : principles and applications. 2nd ed. Boca Raton, FL: CRC Press, 2008, p.xxi, 236 p.

10. Farboodmanesh S, Chen J, Tao Z, Mead J and Zhang H. Base fabrics and their interaction in coated fabrics. In: Smith WC, (ed.). Smart textile coatings and laminates. Woodhead Publishing Series in Textiles, 2010, p. 42-94.

11. Fung W. Coated and Laminated Textiles. Cambridge, U.K. : Woodhead Publishing, 2002, p.413.

12. Shim E. Coating and laminating processes and techniques for textiles. In: Smith WC, (ed.). Smart textile coatings and laminates. Woodhead Publishing Series in Textiles, 2010, p. 10-41.

42

13. Kistler SF and Schweizer PM. Liquid Film Coating: Scientific principles and their technological implications. SpringerLink, Archive ed. Dordrecht: Springer Netherlands, 1997.

14. Schwarz A and Van Langenhove L. Types and processing of electro-conductive and semiconducting materials for smart textiles. In: Kirstein T, (ed.). Multidisciplinary know-how for smart-textiles developers. Woodhead Publishing Ltd., 2013, p. 29-69.

15. Harlin A and Ferenets M. Chapter 13 Introduction to conductive materials. In: Mattila HR and Institute T, (eds.). Intelligent Textiles and Clothing. Boca Raton: Woodhead Publishing, 2006.

16. Stauffer D and Aharony A. Introduction To Percolation Theory. Taylor & Francis, 1992.

17. Wypych G, Knovel and Plastics Design L. Handbook of fillers. Norwich, N.Y; Toronto, Ont: ChemTec, 2000.

18. Lux F. Models proposed to explain the electrical conductivity of mixtures made of conductive and insulating materials. J Mater Sci. 1993; 28: 285-301.

19. Yang K, Torah R, Wei Y, Beeby S and Tudor J. Waterproof and durable screen printed silver conductive tracks on textiles. Textile Research Journal. 2013; 83: 2023-31.

20. Osório I, Igreja R, Franco R and Cortez J. Incorporation of silver nanoparticles on textile materials by an aqueous procedure. Materials Letters. 2012; 75: 200-3.

21. Jiang SQ and Guo RH. Modification of textile surfaces using electroless deposition. In: Wei Q, (ed.). Surface modification of textiles. Cambridge, U.K. : Woodhead Publishing Ltd., 2009, p. 108-25.

22. Depla D, Segers S, Leroy W, Van Hove T and Van Parys M. Smart textiles: an explorative study of the use of magnetron sputter deposition. Textile Research Journal. 2011; 81: 1808-17.

23. Zhang H, Shen L and Chang J. Comparative Study of Electroless Ni-P, Cu, Ag, and Cu-Ag Plating on Polyamide Fabrics. Journal of Industrial Textiles. 2011; 41: 25-40.

24. VandeVoort AR and Arai Y. Chapter two - Environmental Chemistry of Silver in Soils: Current and Historic Perspective. In: Sparks DL, (ed.). Advances in Agronomy. Amsterdam, The Netherlands. : Academic Press, 2012, p. 59-90.

25. Rezić I. Chapter 7 - Engineered Nanoparticles in Textiles and Textile Wastewaters. In: Farré M and Barceló D, (eds.). Comprehensive Analytical Chemistry. Amsterdam, The Netherlands. : Elsevier, 2012, p. 235-64.

43

26. Cochrane C and Cayla A. Polymer-based resistive sensors for smart textiles. In: Kirstein T, (ed.). Multidisciplinary know-how for smart-textiles developers. The Textile Institute and Woodhead Publishing, 2013, p. 129-53.

27. Shirakawa H. The discovery of polyacetylene film - the dawning of an era of conducting polymers. Current Applied Physics. 2001; 1: 281-6.

28. Gregory RV, Kimbrell WC and Kuhn HH. Conductive textiles. Synthetic Metals. 1989; 28: 823-35.

29. Knittel D and Schollmeyer E. Electrically high-conductive textiles. Synthetic Metals. 2009; 159: 1433-7.

30. Kuhn HH, Kimbrell JR and William C. Electrically conductive textile materials and method for making same. US patent 4975317; 1990.

31. http://www.eeonyx.com/eeontex.php.

32. Patil AJ and Deogaonkar SC. A novel method of in-situ chemical polymerization of polyaniline for synthesis of electrically conductive cotton fabrics. Textile Research Journal. 2012.

33. Patil AJ and Deogaonkar SC. Conductivity and atmospheric aging studies of polypyrrole-coated cotton fabrics. Journal of Applied Polymer Science. 2012; 125: 844-51.

34. http://www.sigmaaldrich.com/catalog/product/aldrich/650013. 2012-10-21.

35. http://www.sigmaaldrich.com/catalog/product/aldrich/482552. 2012-10-21.

36. http://www.sigmaaldrich.com/catalog/product/aldrich/483095. 2012-10-21.

37. http://clevios.com/en/conductivepolymers/pedot-pss-conductive-polymers.aspx. 2012-10-21.

38. Kirstein T. The future of smart-textiles development: new enabling technologies, commercialization and market trends. In: Kirstein T, (ed.). Multidisciplinary know-how for smart-textiles developers. Woodhead Publishing Ltd., 2013, p. 1-26.

39. http://smarttextiles.se/en/new-idea/business-innovation/ongoing-projects/. Borås.

40. Castano LM and Flatau AB. Smart fabric sensors and e-textile technologies: a review. Smart Materials and Structures. 2014; 23: 053001.

41. Kawai H. The Piezoelectricity of Poly (vinylidene Flouride). Japan J Appl Phys. 1969; 8: 975-6.

42. Zeng W, Shu L, Li Q, Chen S, Wang F and Tao X-M. Fiber-Based Wearable Electronics: A Review of Materials, Fabrication, Devices, and Applications. Advanced Materials. 2014; 26: 5310-36.

44

43. Nilsson E, Lund A, Jonasson C, Johansson C and Hagström B. Poling and characterization of piezoelectric polymer fibers for use in textile sensors. Sensors and Actuators A: Physical. 2013; 201: 477-86.

44. Groenendaal L, Jonas F, Freitag D, Pielartzik H and Reynolds JR. Poly(3,4-ethylenedioxythiophene) and Its Derivatives: Past, Present, and Future. Advanced Materials. 2000; 12: 481-94.

45. Simpson J, Kirchmeyer S and Reuter K. Advances and applications of inherently conductive polymer technologies based on poly(3,4-ethylenedioxythiophene). 2005 AIMCAL Fall Technical Conference and 19th International Vacuum Web Coating Conference. Myrtle Beach, South Carolina.

46. Reuter K, Kirchmeyer S and Elschner A. PEDOT–Properties and Technical Relevance. Handbook of Thiophene-Based Materials. John Wiley & Sons, Ltd, 2009, p. 549-76.

47. Sun J, Gerberich WW and Francis LF. Transparent, conductive polymer blend coatings from latex-based dispersions. Progress in Organic Coatings. 2007; 59: 115-21.

48. Hansen TS, West K, Hassager O and Larsen NB. Highly Stretchable and Conductive Polymer Material Made from Poly(3,4-ethylenedioxythiophene) and Polyurethane Elastomers. Advanced Functional Materials. 2007; 17: 3069-73.

49. Yamashita T, Takamatsu S, Miyake K and Itoh T. Fabrication and evaluation of a conductive polymer coated elastomer contact structure for woven electronic textile. Sensors and Actuators A: Physical. 2013; 195: 213-8.

50. Irwin M, Roberson D, Olivas R, Wicker R and MacDonald E. Conductive polymer-coated threads as electrical interconnects in e-textiles. Fibers and Polymers. 2011; 12: 904-10.

51. Odhiambo SA, De Mey G, Hertleer C, Schwarz A and Van Langenhove L. Discharge characteristics of poly(3,4-ethylene dioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) textile batteries; comparison of silver coated yarn electrode devices and pure stainless steel filament yarn electrode devices. Textile Research Journal. 2014; 84: 347-54.

52. Martin BD, Nikolov N, Pollack SK, et al. Hydroxylated secondary dopants for surface resistance enhancement in transparent poly(3,4-ethylenedioxythiophene)–poly(styrenesulfonate) thin films. Synthetic Metals. 2004; 142: 187-93.

53. Onorato A, Invernale MA, Berghorn ID, Pavlik C, Sotzing GA and Smith MB. Enhanced conductivity in sorbitol-treated PEDOT–PSS. Observation of an in situ cyclodehydration reaction. Synthetic Metals. 2010; 160: 2284-9.

45

54. Jönsson SKM, Birgerson J, Crispin X, et al. The effects of solvents on the morphology and sheet resistance in poly(3,4-ethylenedioxythiophene)–polystyrenesulfonic acid (PEDOT–PSS) films. Synthetic Metals. 2003; 139: 1-10.

55. Dimitriev OP, Grinko DA, Noskov YV, Ogurtsov NA and Pud AA. PEDOT:PSS films--Effect of organic solvent additives and annealing on the film conductivity. Synthetic Metals. 2009; 159: 2237-9.

56. Coating Guide: Clevios™ P Formulations. Heraeus Clevios GmbH, Germany, Issue 03/2012.

57. Sen AK. Coated textiles : principles and applications. Lancaster: Technomic Pub. Co., 2001, p.xv, 230 p.

58. Kästner U. The impact of rheological modifiers on water-borne coatings. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 2001; 183-185: 805-21.

59. Barmar M. Study of the effect of PEG length in Uni-HEUR thickener behavior. Journal of Applied Polymer Science. 2009; 111: 1751-4.

60. Orgilés-Calpena E, Arán-Aís F, Torró-Palau AM, Orgilés-Barceló C and Martín-Martínez JM. Influence of the Chemical Structure of Urethane-Based Thickeners on the Properties of Waterborne Polyurethane Adhesives. The Journal of Adhesion. 2009; 85: 665-89.

61. Orgilés-Calpena E, Arán-Aís F, Torró-Palau AM, Orgilés-Barceló C and Martín-Martínez JM. Addition of different amounts of a urethane-based thickener to waterborne polyurethane adhesive. International Journal of Adhesion and Adhesives. 2009; 29: 309-18.

62. Orgilés-Calpena E, Arán-Aís F, Torró-Palau AM, Orgilés-Barceló C and Martín-Martínez JM. Effect of annealing on the properties of waterborne polyurethane adhesive containing urethane-based thickener. International Journal of Adhesion and Adhesives. 2009; 29: 774-80.

63. Pretl S, Hamacek A, Reboun J, Cengery J, Dzugan T and Kroupa M. Electrical characterization of PEDOT:PSS. Electronic System-Integration Technology Conference (ESTC), 2010 3rd. 2010, p. 1-4.

64. Farboodmanesh S, Chen J, Mead JL, et al. Effect of Coating Thickness and Penetration on Shear Behavior of Coated Fabrics. Journal of Elastomers and Plastics. 2005; 37: 197-227.

65. Zhong W, Pan N and Lukas D. Stochastic modelling of tear behaviour of coated fabrics. Modelling and Simulation in Materials Science and Engineering. 2004; 12: 293.

46

PAPER I

ELECTRICALLY CONDUCTIVE TEXTILE COATING WITH A PEDOT-PSS DISPERSION AND A POLYURETHANE BINDER

ÅKERFELDT, M., STRÅÅT, M., & WALKENSTRÖM, P. TEXTILE RESEARCH JOURNAL, 83(6), 618-627 (2013)

DOI: 10.1177/0040517512444330

66. Bulut Y and Sülar V. Effects of process parameters on mechanical properties of coated fabrics. International Journal of Clothing Science and Technology. 2011; 23: 205-21.

67. Yesilalan HE, Warner SB and Laoulache R. Penetration of Blade-Applied Viscous Coatings into Yarns in a Woven Fabric. Textile Research Journal. 2010; 80: 1930-41.

68. Jiang S and Guo R. Effect of polyester fabric through electroless Ni-P plating. Fibers and Polymers. 2008; 9: 755-60.

69. Afzali A, Mottaghitalab V, Motlagh M and Haghi A. The electroless plating of Cu-Ni-P alloy onto cotton fabrics. Korean Journal of Chemical Engineering. 2010; 27: 1145-9.

70. Li R, Liu G, Gu F, Wang Z, Song Y and Wang J. In situ polymerization of aniline on acrylamide grafted cotton. Journal of Applied Polymer Science. 2011; 120: 1126-32.

71. Abbott NJ, Lannefeld TE, Barish L and Bpysson RJ. A Study of Tearing in Coated Cotton Fabrics. Journal of Industrial Textiles. 1971; 1: 4-17.

72. Witkowska B and Frydrych I. Protective clothing – test methods and criteria of tear resistance assessment. International Journal of Clothing Science and Technology. 2005; 17: 242-52.

73. Yu W and Du Z. Determination of the Bending Characteristic Parameters of the Bending Evaluation System of Fabric and Yarn. Textile Research Journal. 2006; 76: 702-11.

74. Behery HM. Effect of mechanical and physical properties on fabric hand Woodhead Publishing Ltd., 2005.

75. Steinmann W, Walter S, Seide G, Gries T, Roth G and Schubnell M. Structure, properties, and phase transitions of melt-spun poly(vinylidene fluoride) fibers. Journal of Applied Polymer Science. 2011; 120: 21-35.

76. Martins RS, Gonçalves R, Azevedo T, et al. Piezoelectric coaxial filaments produced by coextrusion of poly(vinylidene fluoride) and electrically conductive inner and outer layers. Journal of Applied Polymer Science. 2014; 131: n/a-n/a.

77. Hadimani RL, Bayramol DV, Sion N, et al. Continuous production of piezoelectric PVDF fibre for e-textile applications. Smart Materials and Structures. 2013; 22: 075017.

78. Lund A, Jonasson C, Johansson C, Haagensen D and Hagström B. Piezoelectric polymeric bicomponent fibers produced by melt spinning. Journal of Applied Polymer Science. 2012; 126: 490-500.

Due to copyright reason the articles included in this PhD thesis by publication are not available in the digital version of the thesis. Complete references to the articles can be foundon the following pages.

PaPer I

PAPER I

ELECTRICALLY CONDUCTIVE TEXTILE COATING WITH A PEDOT-PSS DISPERSION AND A POLYURETHANE BINDER

ÅKERFELDT, M., STRÅÅT, M., & WALKENSTRÖM, P. TEXTILE RESEARCH JOURNAL, 83(6), 618-627 (2013)

DOI: 10.1177/0040517512444330

Original article

Electrically conductive textile coatingwith a PEDOT-PSS dispersion and apolyurethane binder

Maria Akerfeldt1, Martin Straat2 and Pernilla Walkenstrom2

Abstract

Electrically conductive textile coatings have been prepared by the addition of a dispersion of poly(3,4-ethylenedioxy

thiophene)-polystyrene sulfonate (PEDOT-PSS) and ethylene glycol to a polyurethane-based coating formulation. The

formulations were designed to have similar viscosities, measured with a rheometer using a cone-and-plate set-up. The

formulations were applied to woven poly(ethylene) terephthalate substrates using a direct coating method. The con-

centration PEDOT-PSS in the finished coatings varied between 0.7 and 6.2wt%, the coating deposit between 19 and

155 g/m2 and the drying procedure between 4 hours at 20C and 10 minutes at 150C. Surface resistivity was measured

with a ring probe and surface topology was addressed with scanning electron microscopy (SEM). The PEDOT-PSS

concentration had a large effect on the resistivity, which dropped by five orders of magnitude with an increased con-

centration. The steepest decrease occurred between 1 and 3wt% PEDOT-PSS, indicating a percolation threshold. An

increased coating deposit resulted in a resistivity drop by a factor 10, but no significant effect on the resistivity of the

samples could be ascertained by variation of the drying conditions when samples had been subjected to subsequent

annealing.

Keywords

Coatings, composites, incorporation, intrinsically conductive polymers, measurement, properties

Introduction

During the past decade, the concept of ‘smart textiles’has developed into more than just an expression.1,2

Technical research has opened the way both to newmaterials and to new ways of integrating them withother materials, thereby making possible a whole newrange of applications. Textiles, being omnipresent inhuman life and appreciated for their often high flexibil-ity and wear resistance compared to competing mater-ials such as paper and plastics, are prime candidates forthe incorporation of ‘smart’ technology. Applications,such as textile sensors3 and actuators,4 electrostatic dis-charge (ESD) protection,5 electromagnetic interference(EMI) shielding6 and heating or cooling elements,7,8

have in common that they require electrical conductiv-ity for their function. Within the area of textile coatingand finishing, electrical conductivity may be achievedby methods including plating or sputtering of metalsonto textile substrates,9 direct coating or printing of a

conductive polymer composite (CPC)10 or by the in situpolymerization of conjugated polymer systems (e.g.polypyrrole and dopant) onto the textile fibre surface.11

Direct coating has for a long time been a commonway of adding value to a textile substrate.12 In general,a thickened polymeric formulation is smeared onto atextile substrate with a knife and thereafter subjected todrying and annealing for complete film formation. Inthe drying phase, water (or another solvent) evaporatesfrom the formulation and the polymer particles stacktogether and in the annealing phase the polymers

1The Swedish School of Textiles, University of Boras, Sweden2Swerea IVF, Sweden

Corresponding author:

Maria Akerfeldt, The Swedish School of Textiles, University of Boras,

Boras 50190, Sweden.

Email: maria.akerfeldt@hb.se

Textile Research Journal

83(6) 618–627

! The Author(s) 2013

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0040517512444330

trj.sagepub.com

PaPer II

PAPER II

INFLUENCE OF COATING PARAMETERS ON TEXTILE AND ELECTRICAL PROPERTIES OF A POLY(3,4-ETHYLENE DIOXYTHIOPHENE):POLY(STYRENE SULFONATE)/POLYURETHANE-

COATED TEXTILE ÅKERFELDT, M., STRÅÅT, M., & WALKENSTRÖM, P.

TEXTILE RESEARCH JOURNAL, 83(20), 2164-2176 (2013) DOI: 10.1177/0040517513487786

applications. Ambience 08. Boras, Sweden: University of

Boras, 2008.

21. Weiser M, Pohlers S, Neudeck A, et al. Application of

intrinsic conductive polymers (ICP) on textile materials.

Melliand Int 2010; 16: 226–228.22. Akerfeldt M and Straat M. A rheological study of a tex-

tile coating paste containing PEDOT:PSS. In: Sjoqvist M

(ed.) Nordic Rheology Conference. Helsinki, Finland:

Annual transactions of the Nordic Rheology Society,

2011.23. Orgiles-Calpena E, Aran-Aıs F, Torro-Palau AM, et al.

Addition of different amounts of a urethane-based

thickener to waterborne polyurethane adhesive. Int JAdhesion Adhesives 2009; 29: 309–318.

24. Methods of test for volume resistivity and surface resist-

ivity of solid electrical insulating materials, CEI/IEC 93:1980. International Electrotechnical Commision.

25. Maryniak WA, Uehara T and Noras MA. Surface resist-ivity and surface resistance measurements. Trek

Application Note. Trek, Inc., 2003, pp.1–4.26. Kastner U. The impact of rheological modifiers on water-

borne coatings. Colloids Surfaces A: Physicochem Eng

Aspects 2001; 183–185: 805–821.27. Stauffer D and Aharony A. Introduction to percolation

theory. London: Taylor & Francis, 1994.

Akerfeldt et al. 627

Original article

Influence of coating parameterson textile and electrical propertiesof a poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate)/polyurethane-coated textile

Maria Akerfeldt1, Martin Straat2 and Pernilla Walkenstrom2

Abstract

Textile coatings with electrical conductivity were obtained by the addition of poly(3,4-ethylene dioxythiophene)-

poly(styrene sulfonate) (PEDOT:PSS) and ethylene glycol (EG) to a polyurethane (PU)-based coating formulation.

Variations of the coating formulation, the coating amount and the drying conditions, as well as the absence of an

annealing step, were investigated. The coated fabrics were evaluated for tear strength and bending rigidity as well as

surface resistivity and appearance before and after Martindale abrasion. A high proportion of PEDOT:PSS dispersion in

the formulation and the presence of EG provided low surface resistivity. This composition resulted in softer samples with

higher tear strength than those containing more PU-binder. All coatings proved to withstand abrasion to a similar extent.

The surface resistivity increased gradually with the abrasion, about one half order of magnitude, except for those

coatings that had been subjected to a faster drying process, where the surface resistivity increased somewhat faster.

Keywords

Coatings, Chemistry, composites, materials, structure-properties, intrinsically conductive polymers

Electrical conductivity is becoming an increasinglydesired quality of textile structures.1 With this function-ality, textile applications, such as electrostatic discharge(ESD) protection,2 electromagnetic interference (EMI)shielding3 and even wearable sensors,4 can be realizedand ‘‘smart textiles’’ become more feasible. For manyapplications it is often more effective, economicallyand/or production-wise, to confine the conductivematerials to the surfaces of the textile substrates thanto disperse them throughout the material, such as byadding them to the spinning melt. Plating5 or sputter-ing6 of metals, coatings of conductive polymer compos-ites (CPCs) containing sp2-hybridized carbons7 or insitu polymerization of intrinsically conductive polymers(ICPs)8 are popular routes to achieve surface conduct-ivity in textiles.

Metal coatings have high and reliable conductivitybut also deteriorate the tear strength, and the adhesionof the environmentally hazardous, metallic materials isgenerally poor.9–11 Carbons in the nano-scale, the most

prevalent conductive materials for CPC coatings, arebecoming increasingly scrutinized in terms of healthaspects.12 ICPs, that is, conjugated polymer systems,have so forth become relevant conductive materialsfor textile applications, especially through the in situpolymerization method.13 Most common is the add-ition of pyrrole and dopant to yield polypyrrole com-posites,14,15 but similar composites have been achievedwith polyaniline16 and poly(3,4-ethylene dioxythio-phene) (PEDOT).17 Excessive subsequent washing toeliminate by-products is, however, a requirement for

Textile Research Journal

83(20) 2164–2176

! The Author(s) 2013

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0040517513487786

trj.sagepub.com

1The Swedish School of Textiles, University of Boras, Sweden2Department of Textiles and Plastics, Swerea IVF, Sweden

Corresponding author:

Maria Akerfeldt, The Swedish School of Textiles, University of Boras, attn

MAAK, Boras 50190, Sweden.

Email: maria.akerfeldt@swerea.se

PaPer III

PAPER III

TEXTILE PIEZOELECTRIC SENSORS – MELT SPUN BI-COMPONENT POLY(VINYLIDENE FLUORIDE) FIBRES WITH CONDUCTIVE CORES AND POLY(3,4-ETHYLENE

DIOXYTHIOPHENE)-POLY(STYRENE SULFONATE) COATING AS THE OUTER ELECTRODE ÅKERFELDT, M., NILSSON, E., GILLGARD, P., & WALKENSTRÖM, P.

FASHION AND TEXTILES, 1(1), 1-17 (2014) DOI: 10.1186/S40691-014-0013-6

RESEARCH Open Access

Textile piezoelectric sensors – melt spunbi-component poly(vinylidene fluoride) fibreswith conductive cores and poly(3,4-ethylenedioxythiophene)-poly(styrene sulfonate) coatingas the outer electrodeMaria Åkerfeldt1,2*, Erik Nilsson1,3, Philip Gillgard1 and Pernilla Walkenström1

* Correspondence:maria.akerfeldt@swerea.se1Swerea IVFAB, Textiles and PlasticsDepartment, Box 104, SE-43122Mölndal, Sweden2University of Borås, The SwedishSchool of Textiles, SE-50190 Borås,SwedenFull list of author information isavailable at the end of the article

Abstract

The work presented here addresses the outer electroding of a fully textilepiezoelectric strain sensor, consisting of bi-component fibre yarns of β-crystallinepoly(vinylidene fluoride) (PVDF) sheath and conductive high density polyethylene(HDPE)/carbon black (CB) core as insertions in a woven textile, with conductivepoly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate) (PEDOT:PSS) coatingsdeveloped for textile applications. Two coatings, one with a polyurethane binder andone without, were compared for the application and evaluated as electrode materialin piezoelectric testing, as well as tested for surface resistivity, tear strength, abrasionresistance and shear flexing. Both coatings served their function as the outerelectrodes in the system and no difference in this regard was detected betweenthem. Omission of the binder resulted in a surface resistivity one order of magnitudeless, of 12.3Ω/square, but the surface resistivity of these samples increased moreupon abrasion than the samples coated with binder. The tear strength of the textilecoated with binder decreased with one third compared to the uncoated substrate,whereas the tear strength of the coated textile without binder increased with thesame amount. Surface resistivity measurements and scanning electron microscopy(SEM) images of the samples subjected to shear flexing showed that the coatingswithout the binder did not withstand this treatment, and that the samples withthe binder managed this to a greater extent. In summary, both of the PEDOT:PSScoatings could be used as outer electrodes of the piezoelectric fibres, but inclusionof binder was found necessary for the durability of the coating.

Keywords: Piezoelectric sensor; PVDF; PEDOT:PSS; Textile coating

IntroductionOne of the most critical issues to address for the Smart Textiles concept is the integra-

tion of the smart components into textile structures (Kirstein 2013). Electroactive

components generally lead to perceived bulkiness and loss of flexibility in the textiles.

Furthermore, the lack of refined integration methods has been a serious pitfall for the

possibilities for industrialization and the following commercialization of products

© 2014 Åkerfeldt et al.; licensee springer. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in anymedium, provided the original work is properly credited.

Åkerfeldt et al. Fashion and Textiles 2014, 1:13http://link.springer.com/article/10.1186/s40691-014-0013-6

PaPer IV

PAPER IV

TEXTILE SENSING GLOVE WITH PIEZOELECTRIC PVDF FIBRES AND PRINTED ELECTRODES OF PEDOT:PSS

ÅKERFELDT, M., LUND, A., & WALKENSTRÖM, P. MANUSCRIPT ACCEPTED FOR PUBLICATION IN TEXTILE RESEARCH JOURNAL (2015)

1

TEXTILE SENSING GLOVE WITH PIEZOELECTRIC PVDF FIBRES AND PRINTED ELECTRODES OF PEDOT:PSS

MARIA ÅKERFELDT 1, 2, ANJA LUND 2 AND PERNILLA WALKENSTRÖM 1

1 Swerea IVF AB, Materials Department, Box 104, 431 22 Mölndal, Sweden 2 University of Borås, The Swedish School of Textiles, 501 90 Borås, Sweden

ABSTRACT

This paper describes the development of an entirely polymer-based motion sensing glove with

possible applications for example in physical rehabilitation. The importance of comfort for the

wearer, and the possibility to clean the glove in normal laundering processes, were important

aspects in the development. The glove is all textile, and manufactured using materials and

methods suitable for standard textile industry processes. For the first time, melt-spun

piezoelectric poly(vinylidene fluoride) (PVDF) fibres with conductive cores were machine

embroidered onto a textile glove to function as a sensor element. Electrodes and electrical

interconnections were constituted by a screen printed conductive poly(3,4-ethylene

dioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) formulation. Screen printing

interconnections was shown to be a reliable method for reproducible material deposition,

resulting in an average surface resistivity value of 57 Ω/square. A repeated strain of 10 %

only influenced the resistance of the interconnections initially and to a very limited extent.

The influence of washing on the electrical resistance of the printed interconnections was also

studied; after 15 wash cycles the average surface resistivity was still below 500 Ω/square,

which was deemed sufficient for the polymeric sensor system to remain functional during

long-term use. Sensor data from the glove was also successfully used as input to a

microcontroller running a robot gripper, in order to demonstrate its potential applications.

Keywords: wearable sensor , textile sensor , piezoelectric fibre or fiber , printed electronics , poly(vinylidene fluoride) , poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate)