Polymer Electronics

Polymer Electronics – A Flexible Technology

Frances Gardiner, Eleanor C

arter


Editors: Frances Gardiner Eleanor Carter


Editors:

Frances Gardiner

Eleanor Carter

iSmithers – A Smithers Group Company

Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118

http://www.rapra.net

First Published in 2009 by

Typeset by Argil Services Indexed by Argil Services

Printed and bound by Lightning Source

ISBN (Hard-backed): 978-1-84735-421-1

ISBN (Soft-backed): 978-1-84735-422-8 e-book: 978-1-84735-423-5

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

any have been overlooked.

A catalogue record for this book is available from the British Library.

All rights reserved. Except as permitted under current legislation no part of this publication may be photocopied, reproduced or distributed in any form or by any means or stored in a database or retrieval system, without

the prior permission from the copyright holder.

©2009, Smithers Rapra

iSmithersShawbury, Shrewsbury, Shropshire, SY4 4NR, UK

iii

1 Roadmap for Organic and Printed Electronics ................................................1

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

1.2 Applications ..........................................................................................2

1.3 Organic Electronics Association (OE-A) Roadmap for Organic Electronics Applications ........................................................................3

1.3.1 Technology................................................................................4

1.3.2 Materials ...................................................................................4

1.3.3 rinting and Patterning Techniques .............................................6

1.3.4 Devices ......................................................................................7

1.4 Principle Challenges/Red Brick Walls .....................................................8

1.5 Summary and Outlook ........................................................................10

1.6 Global Network for an Emerging Industry ..........................................10

2 Technical Issues in Printed Electrodes for All-Printed Thin-Film Transistor Applications 13

2.1 Introduction ........................................................................................13

2.2 Surface Roughness of Printed Electrodes .............................................14

2.3 Edge Waviness in Printed Electrodes ....................................................15

2.4 Solution-Process Organic TFT .............................................................17

2.5 Conclusion ..........................................................................................18

2.6 Acknowledgement ...............................................................................18

Contents

iv


3 All-Printed Flexible Organic Light-emitting Diodes .......................................21

Introduction 21

Roll-to-Roll Printing ......................................................................................21

Gravure Printing of Poly(3,4-ethylenediocythiophene):poly(styrene sulfonate) and Pentafluorobenzenethiol .........................................................................24

Screen Printing of Aluminium Cathode .........................................................25

Characteristics of All-Printed OLEDS ............................................................26

Roll-to-Roll Printed OLED Demonstrators ...................................................27

Summary 28

Acknowledgement .........................................................................................28

4 Roadmap for Organic and Printed Electronics ..............................................31

Introduction 31

4.1 Process Modelling and Simulation of Ink Jet Printing ..........................32

4.2 Characterisation of PEDOT-PSS Inks and Products .............................36

4.3 Characterisation of Carbon Nanotube - PEDOT:PSS Nanocomposite Products .....................................................................37

4.4 Electrospinning of Carbon Nanotube Inks ...........................................40

4.5 Conclusions .........................................................................................41

5 Highly Conductive Plastics – Custom-formulated Functional Materials for Injection Mouldable Electronic Applications ...........................................43

5.1 Introduction ........................................................................................43

5.2 Characteristics of the Hybrid Compound Within the Injection Moulding Process ................................................................................45

5.2.1 The Cavity Filling ...................................................................45

5.2.2 The Morphological Structure ..................................................46

5.2.3 Dependence of the Conductivity on Geometry, Material and Process Condition ...................................................................47

v

Contents

5.2.4 Temperature Dependency on the Conductivity ........................50

5.2.5 Electromagnetic Shielding Effectiveness ..................................51

5.2.6 Injection Moulding of Conductor Paths ..................................52

5.3 Conclusions and Outlook ....................................................................54

Acknowledgements ........................................................................................55

6 Additives in Polymer Electronics ...................................................................57

6.1 Introduction ........................................................................................57

6.2 Degradation and Stabilisation of Polymer ............................................57

6.3 Stabilisers for Polymer Electronics .......................................................60

6.4 Functional Additives for Polymer Electronics ......................................63

7 A Facile Route to Organic Nanocomposite Dispersions of Polyaniline – single Wall Carbon Nanotubes ................................................................................67

7.1 Introduction ........................................................................................67

7.2 Materials .............................................................................................68

7.3 Experimental Section ...........................................................................68

7.3.1 Scheme - 1 Polyaniline Synthesis [17] ......................................68

7.3.2 Scheme - 2 Organic Dispersions of PANI ................................68

7.3.3 Scheme - 3 Exfoliation of Single Wall Carbon Nanotubes .......69

7.3.4 Scheme - 4 Organic ‘Nanocomposite’ Dispersions of PANI-SWNT 69

7.4 Characterisation ..................................................................................70

7.5 Results and Discussions .......................................................................70

7.5.1 Phase Inversion Phenomenon ..................................................70

7.5.2 FT-IR Spectroscopy .................................................................70

7.5.3 UV-Vis Spectroscopy ...............................................................72

7.5.4 Four-probe Conductivity .........................................................73

7.6 Conclusions .........................................................................................73

vi


7.7 Acknowledgements ..............................................................................74

8 Preparation and Characterisation of Novel Electrical Conductive Rubber Blends 77

8.1 Introduction ........................................................................................77

8.2 Experimental .......................................................................................77

8.2.1 Chemicals and Raw Materials .................................................77

8.2.2 Synthesis of PANI.DBSA and Blends Preparation ....................77

8.3 Characterisation of NBR-PAni.DBSA Blends .......................................78

8.3.1 Methods of Characterisation ...................................................78

8.3.2 Calculation of Solubility Parameter Values ..............................78

8.3.3 Morphological Studies (Optical Microscopy and TEM) ..........78

8.3.4 FT-IR Spectroscopy .................................................................80

8.3.5 DSC Thermal Analysis ............................................................81

8.3.6 Electrical Conductivity Determination ....................................81

8.4 Conclusion ..........................................................................................85

Acknowledgements ........................................................................................85

9 Solar Textiles 87

9.1 Introduction ........................................................................................87

9.2 Why Textiles? ......................................................................................88

9.3 Solar Cells ...........................................................................................88

9.4 Technological Specifications .................................................................89

9.5 Suitable Textile Constructions .............................................................90

9.5.1 Fibres ......................................................................................90

9.5.2 Fabrics ....................................................................................91

9.6 Electrical Conductivity ........................................................................91

9.7 The Future ...........................................................................................93

vii

Contents

10 Flexible Sensor Array for a Robotic Fingertip Using Organic Thin Film Tran-sistors (OTFT) with Minimum Interconnects and Improved Noise Tolerance 95

10.1 Introduction ........................................................................................95

10.2 General Description of the System .......................................................96

10.3 Operation Mechanism .........................................................................96

10.4 Robotic Finger Tip Specifications .........................................................97

10.5 System Design ......................................................................................97

10.5.1 OTFT Devices .........................................................................97

10.5.2 Sensor Array ...........................................................................98

10.5.3 Active Matrix Addressing Electronics .....................................98

10.5.4 Astable Multivibrator .............................................................99

10.6 Important Design Features .................................................................101

Conclusion 102

11 An Organic Thin Film Transistor Pixel Circuit for Active-Matrix Organic Light Emitting Diode Flat Panel Display ...............................................................105

11.1 Introduction ......................................................................................105

11.2 Pixel Circuits Designs ........................................................................106

11.3 The Voltage Programming Pixel Circuit .............................................107

11.4 Simulation Results .............................................................................108

11.4.1 OTFT model .........................................................................110

11.4.2 OLED model .........................................................................110

11.4.3 The results of the voltage programming circuit .....................110

11.5 Conclusion ........................................................................................113

12 Intelligent Packaging for the Food Industry .................................................117

12.1 Introduction ......................................................................................117

12.2 RFID and Packaging Connection - The Age of Connectivity ..............119

viii


12.3 TTI and Other Solutions - The Viewpoint of Food Industries ............123

12.4 New Proposals - The Consumer’s Viewpoint .....................................127

12.5 Food Companies Under Attack - Sabotage and Class Actions ............130

12.6 Conclusions .......................................................................................130

Abbreviations .....................................................................................................133

Index ..................................................................................................................137

ix

Contributors

Kok Chong Yong

Advanced Rubber Technology Unit, Rubber Research Institute of Malaysia (RRIM), Malaysian Rubber Board, 47000 Sungai Buloh, Selangor, Malaysia

Seungjun Chung

Seoul National University, Department of Electrical Engineering and Computer Science

Kwanak-Ro 599 Shillim-9-Dong, Kwanak-Gu, Seoul 151-744, Korea

Jan Fragner

Institute of Plastics Processing at RWTH Aachen University (IKV), Pontstrasse 49, Aachen, D-52062, Germany

Daisuke Fujiki

Ciba Inc, R1038.4.07, Basel, CH-4002, Switzerland

Klaus Hecker

PolyIC GmbH & Co. KG, Tucherstrasse 2, 90763 Fürth, Germany

Yongtaek Hong



Jaewook Jeong


x



Jinwoo Kim



Darek Korzec

The German University in Cairo, 15th Moustafa el Maghraby-el, nozha Heliopolise, Eygpt

Constantina Lekakou

Faculty of Engineering and Physical Sciences, University of Surrey, Guildford, Surrey GU2 7XH, UK

Arto Maaninen

VTT Technical Research Centre of Finland, Kaitoväylä 1, FI-90571 Oulu, Finland

Tiina Maaninen


Ahmed Maitouk

Faculty of Electrical and Electronics Engineering, German University in Cairo

7, Rabaa al Istethnary buildings, Nozha St – Nasr, Egypt

Robert Mather

Power Textiles Limited, Upland House, Ettrick Road, Selkirk, EH14 4AS, UK

Walter Michaeli


Wolfgang Mildner

PolyIC GmbH & Co. KG, Tucherstrasse 2, 90763 Fürth, Germany

xi

Contributors

Mohamed Montasser

Electrical and Electronics Engineering Department, The German University in Cairo, 15th Moustafa el Maghraby-el, nozha Heliopolise, Eygpt

Pradipta Nayak



Salvatore Parisi

Department of Hygiene, Preventive Medicine and Public Health ‘R. DeBlasi’,

University of Messina, Via Roccazzo, 25-90135, Palermo, Sicily, Italy

Tobias Pfefferkorn


Sanjay Rastogi

Department of Materials, University of Loughborough, Loughborough, LE11 3TU, UK

Dimitris Schoinas


Markus Tuomikoski


Sainath Vaidya

Department of Materials, University of Loughborough, Loughborough, LE11 3TU, UK

Marja Välimäki


xii


John Watts


John Wilson

Power Textiles Limited, Upland House, Ettrick Road, Selkirk, EH14 4AS, UK

Peter Wilson


xiii

Preface

Preface

The worldwide market for polymer electronic products has been estimated to be worth up to £15 billion by 2015 and the opportunity for new markets could be as high as £125 billion by 2025.”

The rapid development of polymer electronics has revealed the possibility for transforming the electronics market by offering lighter, flexible and more cost effective alternatives to conventional materials and products. With applications ranging from printed, flexible conductors and novel semiconductor components to intelligent labels, large area displays and solar panels, products that were previously unimaginable are now beginning to be commercialised.

Smithers Rapra is pleased to announce its first Polymer Electronics conference, designed to inform researchers, material suppliers, component fabricators and electronics manufacturers of the latest research and developments in this dynamic and rapidly evolving field.

With a prestigious panel of speakers presenting on a wide range of polymer electronics and related technology topics, this conference will bring together senior delegates from across the world to discuss the technologies and opportunities at the heart of polymer electronics.

The conference offers two full days of application focussed technical papers detailing current and future developments in the fast-moving polymer electronics industry: it is a must for forward thinking companies wishing to participate in this predicted £125 billion market.

Who Should Attend:

The conference will particularly appeal to those with the following roles:

xiv


The conference will also be of use to companies involved in the following sectors:

1

1 Roadmap for Organic and Printed Electronics

Wolfgang Mildner and Klaus Hecker

1.1 Introduction

The key trends that will drive modern societies in the next decades are:

2


1.2 Applications

3

Roadmap for Organic and Printed Electronics

1.3 Organic Electronics Association (OE-A) Roadmap for Organic Electronics Applications

Figure 1.1

Figure 1.1

Source: OE-A

4


1.3.1 Technology

1.3.2 Materials

Figure 1.2

Figure 1.3


Figure 1.2

Figure 1.3


1.3.3 Printing and Patterning Techniques

Table 1.1


1.3.4 Devices

Figure 1.4

Figure 1.5

Table 1.1 Key parameters for several printing technologies

Printing Method

Speed (m2/s) Resolution (μm)

Film thickness (μm)

Viscosity (Pa-s)

Source: OE-A; Printed Electronics Consulting, in ‘Organic Electronics Technology’, VDMA, 2006; Fraunhofer IZM, ‘Abschlussbericht zum Projekt PropolyTec’, 2006.

8


1.4 Principle Challenges/Red Brick Walls

Figure 1.4

Figure 1.5


2

2


1.5 Summary and Outlook

1.6 Global Network for an Emerging Industry

11


www.oe-a.org

12


13

2 Technical Issues in Printed Electrodes for All-Printed Thin-Film Transistor Applications

Yongtaek Hong, Jaewook Jeong, Seungjun Chung, Jinwoo Kim, and Pradipta K Nayak

2.1 Introduction

Direct-printing of functional layers of solution-process thin-film devices has recently been a hot topic in rugged, disposable, large-area, printed, and flexible electronics applications. Several direct printing methods such as screen printing, microcontact printing, gravure printing, imprinting, laser assisted patterning, and inkjet printing have been considered as the next-generation patterning methods because they can provide high-throughput, low-cost, low-temperature manufacturing capabilities [1-5].

For all-printed thin film transistors (TFT), various organic and inorganic metal electrode materials, such as conductive polymer, carbon nanotube (CNT), organic metal compound, or metal nano-particles, have been used as gate and source/drain electrodes [6-11] in a combination with inkjet- and laser-based printing methods. One of the immediate applications for all-printed TFT would be flexible or rugged display backplane and disposable radio frequency identification (RFID) tags. In addition, printed metal electrodes and TFT have also been used to fabricate passive circuit components, power transmission sheets and sensors for ambient electronics and electronic skin [12-13].

When a narrow line is patterned by printing-based methods, rough surfaces, coffee rings, and/or the creation of wavy edges along the printed lines are frequently observed. Rough surface and coffee ring effects of printed gate electrodes are directly related to gate leakage current, thus current on/off ratio, of the fabricated TFT. If edge waviness of the printed electrodes exists in source/drain (S/D) electrodes, current flow variation and possible changes in operation stability are to be expected. Ink and printing process conditions, especially for inkjet printing methods, must be well controlled to suppress these undesirable structural effects due to the relatively large ink drop size for narrow line patterning. Even in the laser-printing method where a few micron-size beam spots are used, edge waviness can be obtained if the overlap between laser

14


beam spots is not well optimised [14]. Therefore, in this paper, technical issues of the printed electrodes for all-printed TFT applications, including surface roughness and edge waviness, are addressed.

2.2 Surface Roughness of Printed Electrodes

An inkjet printer from DIMATIX Corporation (DMP-2800 series) and silver ink from INKTEC Corporation (IJ-TEC-010) have been used to print silver electrodes on various substrates. When the width of the printed electrodes was in the range of several hundred micrometers, it was possible to suppress the surface roughness of the printed electrodes to a level similar to that of comparable thermally evaporated electrodes. A surface roughness of about 2.2-2.4 nm in root-mean-square (RMS) and 19-20 nm in peak-to-valley values was obtained from both printed and evaporated silver electrodes. If the line width of the printed silver electrode was reduced to some tens of micrometers, it was found that the surface roughness increased to approximately 5 nm in RMS and 29 nm in peak-to-valley values, respectively. The atomic force microscope (AFM) images are shown in Figure 2.1. The AFM images were measured by using the non-contact mode at a rate of 1 Hz for a scan area of 10 10 μm2. Although surface roughness increased, if an organic dielectric layer was used on the printed electrodes, surface roughness was further reduced due to the planarisation capability of the organic dielectric layers. It was observed that, for wide silver electrodes, surface roughness was reduced to about 1.5 nm in RMS and 10 nm in peak-to-valley values when a poly(4-vinylphenol) (PVP) layer was spin-coated onto the electrodes.

Figure 2.1 Microscope and atomic force microscopy images of the printed silver electrodes

15

Technical Issues in Printed Electrodes for All-Printed Thin-Film Transistor Applications

Sheet resistance of the printed silver electrode also increased from 0.2-0.4 ohm/square to 0.6-0.8 ohm/square when the line width was reduced, which correspond to approximately 4-5 10-6 ohm-cm and 16-20 10-6 ohm-cm, respectively. The 20 m diameter nozzle used in this investigation gave a typical ink drop size on substrates of approximately 50 m. This gave a smaller potential line width than is obtained from current processes suggesting that if the nozzle size can be reduced, narrow line widths are achievable. However, for inkjet printing, it can be difficult to obtain sufficiently narrow lines and to ensure a high enough aspect ratio at the same time. If the aspect ratio is not high enough, resistance of the printed line will be high and current flow capacity through the lines will be significantly reduced.

2.3 Edge Waviness in Printed Electrodes

Edges of the printed narrow lines frequently show irregular wavy patterns depending on the material and printing process conditions. Extra care should be taken during the inkjet-printing process, since the properties of inks, such as concentration and viscosity, change with time. Figure 2.2 shows examples of printed silver electrodes with wavy edges. Edge waviness can be modeled and numerically analyzed [15]. If S/D electrodes of TFT have wavy edges, TFT performance and operation stability can be significantly affected. The ATLAS 3D device simulator (Silvaco Inc.) was used to simulate this S/D edge waviness effect. Peak-to-peak magnitude of the wavy patterns in S/D electrodes had the most significant effects on the TFT performance variations. In addition to performance variation, the effects of the overlap between gate and S/D electrodes on TFT performances were also analysed. It was assumed that gate and S/D electrodes had flat and wavy edges, respectively. Due to highly localised electric field distribution along the edges of the wavy patterns, as shown

Figure 2.2 Examples of edge waviness in printed silver electrodes

16


in Figure 2.3, overlap smaller than peak-to-peak magnitude of the wavy patterns provided sufficiently low contact resistance as long as the overlap was larger than characteristic lengths of TFT. The characteristic lengths of TFT are typically determined by current spreading under the S/D electrodes, which is a function of channel resistance and parasitic resistance of the TFT. Therefore, there exits an optimised gate overlap for given TFT structure from the viewpoint of reducing parasitic resistance. Although the characteristic length concept was used for typical TFT with flat edges in the S/D electrodes, similar analysis can be applied to the TFT with wavy S/D electrodes. However, in this case, due to the current localisation effect, current spreading occurs along the peak region of the wavy patterns, leading to variation in characteristic length according to the waviness parameters, such as peak-to-peak magnitude. The variation of characteristic length can be obtained from a transmission line method [15] and can be used for optimisation of gate overlap in the TFT with wavy S/D electrodes.

Hydrogenated amorphous silicon (a-Si:H) TFT were used as test vehicles for the simulation results. Wavy edges were photolithographically patterned in the S/D electrodes and TFT performance and operation stability were tested [16]. Stress bias-dependent operation instability was shown by a-Si:H TFT. Threshold voltage

Figure 2.3 Current density distribution of TFT with flat gate and wavy S/D electrodes in (a) lateral and (a) vertical direction with respect to channel. Dotted

white and red lines show defined effective channel length and flat edge of the gate electrodes. GO means gate overlap

17


shift was relatively independent of peak-to-peak magnitude of the wavy edges. It was also observed that when TFT were stressed in a constant current mode, the threshold voltage shift demonstrated a large dependency on the peak-to-peak magnitude variations. This shows that edge waviness will have different effects for switching and current-driving TFT in terms of the operation stability. The same analysis can be applied to organic or metal-oxide TFT if their source/drain electrodes are printed.

2.4 Solution-Process Organic TFT

Inkjet-printed silver electrodes, spin-coated PVP, and spin-coated 6,13-bis (triisopropylsilylethynyl) pentacene (TIPS-pentacene) were used as gate and S/D electrodes, gate dielectric layer, and semiconductor layer in solution-process organic TFT as shown in Figure 2.4. PVP layer (700-800 nm thick) showed a leakage current of 8-10 nA/cm2 at 0.5 MV/cm and a capacitance of 5-6 nF/cm2 at 1 kHz. Details of the fabrication process and properties of the printed silver electrodes and spin-coated PVP insulator have been reported elsewhere [17]. Organic thin film transistor (OTFT) showed mobility of about 0.07 cm2/Vs, a threshold voltage of 2.74 V, and on/off current ratio of >103 as shown in Figure 2.5, which are comparable to other previously reported results [18-21]. When the printing process and gate dielectric interface are further optimised, better performance is expected. In addition, printed-version of TIPS-pentacene based TFT showed similar TFT performances.

Figure 2.4 Schematic description of the fabricated OTFT

18


2.5 Conclusion

Technical issues in printed electrodes were briefly reviewed for all-printed TFT applications. Surface morphology and edge waviness of the printed electrode should be well controlled to produce uniform and stable TFT behavior and consistent thin-film device performances. This investigation fabricated solution-process TIPS-pentacene based TFT with the printed silver electrodes. Solution-process materials can be readily combined with a low-cost printing process, which can significantly reduce complexity in the fabrication and manufacturing process. In addition, these types of solution-process TFT can be fabricated at low temperatures and they can be also readily implemented on plastic substrates for flexible electronics applications.

2.6 Acknowledgement

This work was supported by the ‘SystemIC 2010’ project of the Korea Ministry of Knowledge Economy and the Seoul R&BD Program (CRO70048). Nayak would like to thank the BK21 project of the Korean Ministry of Education, Science and Technology (SNU BK21 Research Division for Information Technology). The authors would like to thank Professor Soon-Ki Kwon at Gyeongsang National University for providing the TIPS-pentacene material. The authors would also like to thank Professor Changhee Lee and the OLED Center at Seoul National University for valuable technical discussions and process facility support, respectively.

Figure 2.5 Transfer characteristics of the fabricated OTFT. High gate leakage current reduced current of/off ratio. (W/L = 1000/40-60 m/ m)

19


References

1. D.A. Pardo, G.E Jabbour and N. Peyghambarian, Advanced Materials, 2000, 12, 17, 1249.

2. D.B. Wolfe, J.C. Love, L.E. Paul, M.L. Chabinyc and G.M. Whitesides, Applied Physics Letters, 2002, 80, 12, 2222.

3. J. Puetz and M.A. Aegerter, Thin Solid Films, 2008, 516, 14, 4495.

4. J. Koch, E. Fadeeva, M. Engelbrecht, C. Ruffert, H.H. Gatzen, A. Ostendorf and B.N. Chichkov, Applied Physics A, 2006, 82, 1, 23.

5. T.H. J. van Osch, J. Perelaer, A.W.M. de Laat and U.S. Schubert, Advanced Materials, 2008, 20, 2, 343.

6. H. Sirringhaus, T. Kawase, R.H. Friend, T. Shimoda, M. Inbasekaran, W. Wu and E.P. Woo, Science, 2000, 290, 5499, 2123.

7. M-F. Chen, Y-P. Chen, W-T. Hsiao and Z-P. Gu, Thin Solid Films, 2007, 515, 24, 8515.

8. Y-Y. Noh, N. Zhao, M. Caironi and H. Sirringhaus, Nature Nanotechnology, 2007, 2, 12, 784.

9. W.R. Small and M. in het Panhuis, Small, 2007, 3, 9, 1500.

10. S.P. Speakman, G.G. Rozenberg, K.J. Clay, W.I. Milne, A. Ille, I.A. Gardner, E. Bresler and J.H.G. Steinke, Organic Electronics, 2001, 2, 2, 65.

11. T. Kawase, S. Moriya, C.J. Newsome and T. Shimoda, Japanese Journal of Applied Physics, 2005, 44, 1, 6A, 3649.

12. T. Someya, Y. Kato, T. Sekitani, S. Iba, Y. Nguchi, Y. Murase, H. Kawaguchi and T. Sakurai, Proceedings of the National Academy of Sciences, 2005, 102, 35, 12321.

13. T. Sekitani, M. Takamiya, Y. Noguchi, S. Nakano, Y. Kato, T. Sakurai and T. Someya, Nature Materials, 2007, 6, 6, 413.

14. M-F. Chen, Y-P. Chen, W-T. Hsiao and Z-P. Gu, Thin Solid Films, 2007, 515, 24, 8515.

15. J. Jeong, Y. Hong, S.H. Baek, L. Tutt and M. Burburry, Electronics Letters, 2008, 44, 10, 616.

20


16. J. Jeong, S. Chung, Y. Hong, S.H. Baek, L. Tutt and M. Burburry, Journal of Korean Physical Society, 2009, 54, 1.

17. J. Kim, J. Cho, S. Chung, J. Kwak, C. Lee, J-J. Kim and Y. Hong, Journal of Korean Physical Society, 2009, 54, 1.

18. J.G. Park, R. Vasic, J.S. Brooks and J.E. Anthony, Journal of Applied Physics, 2006, 100, 4, 44511.

19. Y-H. Kim, Y.U. Lee, J-I. Han, S-M. Han and M-K. Han, Journal of the Electrochemical Society, 2007, 154, 12, H995.

20. J-H. Kwon, J-H. Seo, S-I. Shin, K-H. Kim, D.H. Choi, I.B. Kang, H. Kang and B-K. Ju, IEEE Transactions on Electron Devices, 2008, 55, 2, 500.

21. S-I. Shin, J-H. Kwon, H. Kang and B-K. Ju, Semiconductor Science and Technology, 2008, 23, 8, 85009.

21

3 All-Printed Flexible Organic Light-emitting Diodes

Arto Maaninen, Markus Tuomikoski, Marja Välimäki and Tiina Maaninen

Introduction

Organic electronics has grown into a promising candidate to replace several low end silicon applications. Printed and organic electronics are expected to become worth between 250 – 300 billion USD within the next 20 years [1]. To date, fabrication has been carried out using traditional and expensive batch processes such as the vacuum evaporation technique. However, the potential application of roll-to-roll printing represents a new and interesting method for the manufacture of flexible organic devices [2] that would provide high throughput capability with decreased fabrication costs.

Printing techniques for roll-to-roll production of organic light-emitting diodes (OLED) were investigated by the EU funded project; roll-to-roll manufacturing technology for flexible OLED devices and arbitrary size and shape displays (ROLLED) [3]) under the Sixth EU-Framework Programme. The project objective was to fabricate an entire OLED structure by using roll-to-roll manufacturing methods and examine how commercial production could be set up and integrated into an existing printing process. In order to attain roll-to-roll compatibility, it was essential for all materials, inks and devices to be suitable for printing.

Roll-to-Roll Printing

In this study, all the printing experiments were carried out using a roll-to-roll pilot-line production machine, which contains four exchangeable printing units for reverse and forward gravure, flexography and rotary screen printing techniques. The printing machine also provided thermal heating units and sensors for controlling the registration and tension of the web. A photograph of the pilot-line printing machine is shown in Figure 3.1.

22


The procedure for OLED printing development was carried out in the following way:

1) Developing ink suitable for the printing technique;

2) Printing experiments using lab scale printers with variable printing parameters;

3) Re-formulation of printing inks to optimise film quality and performance;

4) Scaling-up printing inks and fabrication of printing tools for roll-to-roll printing experiments.

Figure 3.2(a) outlines the gravure printing process in which ink picked up from a reservoir fills the engraved cells of a rotating gravure cylinder. A doctoring blade removes any excess ink so that only ink contained within these cells is transferred onto the surface of the impression roll, which is covered with a flexible substrate. The thickness and quality of the printed films are determined by parameters such as the load of the doctoring blade, ink properties, the characteristics of the gravure cylinder, the pressure between the cylinder and the impression roll, and printing speed. In this study, the gravure printing technique was used for thin film printing of a hole injection and green and red light-emitting polymers.

Figure 3.1 The photograph of roll-to-roll pilot-line printing machine used for manufacturing of OLED [3]

23

All-Printed Flexible Organic Light-emitting Diodes

Screen printing is one of the simplest printing processes and can be used for a number of applications such as printing textiles, ceramics and electronics. However, print quality can be affected by factors such as the properties of the ink and substrate, the type of material used for the stencil and its thickness, the number of mesh openings of the screen fabric, the properties of the squeegee and printing speed. [5] Rotary screen printing is carried out at a higher speed than flat screen printing. The screen has a cylinder shape and the ink is placed inside the cylinder as shown in Figure 3.2(b). The squeegee, which is also within the cylinder, forces the ink through the openings in the screen and onto the substrate [6]. In the study, the rotary screen printing technique was used for printing an insulator, a cathode and electric contacts.

Figure 3.2 The schemes of (a) gravure printing and (b) rotary screen printing processes used in the roll-to-roll OLED processing [4]

24


Gravure Printing of Poly(3,4-ethylenediocythiophene):poly(styrenesulfonate) and Pentafluorobenzenethiol

The printability of standard poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) polymer was examined on the top of indium tin oxide (ITO) coated plastic foil with various gravure printing parameters and ink formulations. The experiments showed that a modified aqueous PEDOT:PSS solution was necessary for the gravure printing process. According to printability tests, the addition of the surfactant Tween®80 enhanced the wetting properties of the ink, as did the addition of the solvent: isopropyl alcohol. Isopropyl alcohol also influenced the morphology of the printed structure by altering the fluid, flow and drying properties, of the ink.

A 70 nm thick, smooth and pinhole-free film was successfully printed using a 100 lines/cm and 37 μm cell depth gravure cylinder and modified ink containing 65 wt% of PEDOT:PSS, 1 wt% Tween®80, and 34 wt% isopropyl alcohol. The PEDOT:PSS layer was subsequently dried at 110 °C with a web speed of 8 m/min.

The printability of the yellow light-emitting polymer pentafluorobenzenethiol (PFBT) was examined on the top of printed PEDOT:PSS. The 240 nm thick, smooth and pinhole-free films were roll-to-roll gravure printed using the mixture of 2.9 wt% of PFBT in p-xylene without additives. Waviness or interfacial mixing was not observed after printing and drying at 110 oC with the web speed of 2 m/min. Figure 3.3 (a) and (b) present the white light interferometer (WLI) 3D surface profiles from gravure printed PEDOT:PSS and PFBT films on ITO foil.

Figure 3.3 WLI 3D surface profiles of (a) PEDOT: PSS (RMS 16 nm) and (b) PFBT (RMS 39 nm) polymers

25


Screen Printing of Aluminium Cathode

In the OLED structure, the cathode layer normally consists of vacuum evaporated low and high work function metals such as calcium and silver, respectively. In this study, the target was to process a highly conductive cathode from printable metal ink using a ball milling process [7]. Aluminium was selected for the metal ink preparation as it has a lower work function (Al 4.2 eV) than silver (Ag 4.6 eV), which is the more common metal used in conductive pastes for other electronic applications. After preparation of aluminium ink, the screen printing technique was selected in order to achieve a thick and opaque layer on top of the OLED structure. An aluminium layer of around 10 μm thick, and sheet resistance of 0.06 /sq was successfully screen printed using a 200 lines/cm mesh in an inert atmosphere to avoid oxidation of aluminium metal particles. An image of the 3D surface profile and printed aluminium cathode are shown in Figure 3.4(a) and (b). Further optimisation of the screen printing process will significantly smooth the aluminium layer [7].

Figure 3.4 (a) WLI 3D surface profile from the edge of screen printed aluminium cathode, and (b) image of aluminium cathode (after Size)

26


Characteristics of All-Printed OLEDS

The all-printed OLED light sources described previously were characterised in an inert atmosphere. The devices had a turn-on voltage of approximately 5 V and an average brightness of 79 cd/m2 was achieved at the voltage of 16 V, as shown in Figure 3.5(a).

Figure 3.5 Brightness (a) and current density (b) versus voltage characteristics of flexible all-printed OLED with the structure of ITO/PEDOT: PSS/PFBT/aluminium

27


Measuring current density versus voltage characteristics in Figure 3.5(b) illustrates that there was no leakage of current at low operating voltages, indicating good compatibility of printed organic and metal structures. However, OLED performance can be significantly enhanced by adding small amounts of low work function metal into the aluminium ink.

Roll-to-Roll Printed OLED Demonstrators

The objective was to roll-to-roll manufacture an OLED tamper indicator as a demonstrator application. The demonstrator consisted of a label with two OLED red and green light-emitters and a radio-frequency identification (RFID) antenna for the power supply. The roll-to-roll manufacturing process comprised of printing several OLED layers on the ITO coated barrier foil, including: PEDOT:PSS, red and green light-emitters, an insulator, silver electric contacts, an aluminium ink cathode, and an adhesive glue for laminating the back side barrier foil. The device structure is illustrated in Figure 3.6(a). Figure 3.6(b) shows the all-printed OLED roll produced using the roll-to-roll manufacturing process.

The [box] (shown in Figure 3.7) contains an electrical seal in the form of a fuse. If the seal is broken a red OLED indicator will be light up when the circuit is powered by an external radio-frequency (RF) field, as illustrated in Figure 3.7(a). Otherwise, if the seal is intact, a green OLED indicator is light during RF activation, as shown in Figure 3.7(b).

Figure 3.6 (a) The device structure used in the OLED demonstrators and (b) the photograph of roll-to-roll manufactured OLED foil

28


Summary

The study demonstrated the successful roll-to-roll manufacturing of all-printed flexible OLED, in which the entire OLED stack, including: PEDOT:PSS, green and red light-emitting polymers, an aluminium ink cathode, insulator and silver electric contacts were gravure and screen printed onto plastic foil. This shows that roll-to-roll printing techniques can be used for processing smooth and pinhole-free thin film polymer layers. Furthermore, the study screen printed homogenous insulator, metal cathode and electric contact layers for the OLED demonstrators.

Acknowledgement

The European Commission ROLLED Project (FP6-2003-IST-2-004315) is highly acknowledged for funding the work presented in this paper. We thank ROLLED partners for their guidance in the demonstrator issues.

References

1. R. Das and P. Harrop in Organic & Printed Electronics Forecasts, Players & Opportunities, 2007-2027, IDTechEx Ltd., Cambridge, UK, 2007.

Figure 3.7 Photographs of the operating (a) red and (b) green OLED indicators

29


2. T. Kololuoma, M. Tuomikoski, T. Mäkelä, J. Heilmann, T. Haring, J. Kallioinen, J. Hagberg, I. Kettunen and H. Kopola in Emerging Optoelectronic Applications, Eds., G.E. Jabbour and J. T. Rantala, SPIE, Bellingham, WA, USA, 2004, SPIE Conference Proceedings No.5363, p.77.

3. ROLLED Project Website, http://www.vtt.fi/proj/rolled

4. H. Kipphan, Handbook of Print Media, Springer Verlag, Heidelberg, Germany, 2001.

5. S.B. Hoff, Screen Printing: A Contemporary Approach, Delmar Publishers, Albany, NY, USA, 1997.

6. A. Blayo and B. Pineaux in Proceedings of the 2005 Joint Conference on Smart Objects and Ambient Intelligence: Innovative Context-aware Services - Usages and Technologies, Grenoble, France, 2005, p.27.

7. T. Maaninen, M. Tuomikoski and A. Maaninen in Proceedings of the 7th International Meeting on Information Display, Daegu, South Korea, 2007, p.721

30


31

4 Roadmap for Organic and Printed Electronics

Peter C. Wilson, Dimitris Schoinas, Constantina Lekakou and John F. Watts

Introduction

The search for cheaper, lighter, and more flexible electronic devices has caused a surge of interest in the field of organic polymer electronics. Conducting -conjugated polymer films, which possess flexibility, semiconducting electronic properties and transparency, are opening the door to novel electronic equipment including roll up screens, photovoltaic coatings and smart windows. Central to this new push is the ability to use low-cost, precise deposition techniques like inkjet printing and electrospinning. Further developments, to ensure the electronic stability of the organic materials over their life cycle, are bringing closer the possibility for mass-produced, low cost electronics. For example, photovoltaic devices are showing significant potential to rival current energy sources. In any semiconducting material application, a transport layer is required to serve as the charge conductor. In many applications it is important that this layer possesses a high level of transparency. Of all the options, the use of poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) is demonstrating the most favourable results, not least due to its high conductivity (10 S/cm) [1], high work function (~ 5.2 eV) and high mechanical and thermal resistance.

The primary focus for current investigation is the optimisation of inkjet printed PEDOT:PSS layers. The two common ink jet printing techniques are thermal ink jet printing [2], and piezoelectric ink jet printing [3, 4, 5]. Undoped PEDOT can achieve conductivities of over 200 S/cm [6] although, like most conducting polymers, pure PEDOT is insoluble in water and hence must be doped with PSS. Analysis of inkjet printed, spin cast and drop spread deposited PEDOT:PSS layers has shown a phase separation between the conducting PEDOT and insulating PSS layers. To increase conductivity, the addition of a polar solvent becomes necessary [7]. The increased disorder in the morphology leads to an increase in conductivity by a factor of over 10, whilst also increasing the surface tension by 10-20%. Since inkjet printers

32


require fluids with surface tension within the range of 0-70 mN/m, with viscosities below 20 mPas, it is necessary to employ a surfactant to lower the surface tension to a more reasonable level. In this paper, it is shown how the optimum dimethyl sulfoxide (DMSO) concentration results in a surface tension above the upper boundary of working inkjet requirements. Investigations [8] into further organic solvents have shown that the addition of ethylene glycol, 2-nitroethanol, methyl sulfoxide or 1-methyl-2-pyrrolidinone have lead to similar conductivity enhancement. It is proposed that the driving force behind this is the interaction between the dipole of the organic compound, and the charges on the PEDOT chain.

Ink jet printing is a drop deposition technique, which results in precise patterns. Electrospinning, on the other hand, is a fibre deposition technique where a continuous fibre mat, or fibre assembly, may be deposited to cover a specified area. Electrospun films are highly porous and it is possible to fabricate them with a preferred fibre orientation and, hence, anisotropic properties.

4.1 Process Modelling and Simulation of Ink Jet Printing

Table 4.1 presents the dimensionless groups that are important in jet flow and jet break up times, where Rn is the nozzle radius and Uo is the velocity at the nozzle, , μ and are the density, viscosity and surface tension of the ink.

Table 4.1 Dimensionless groups important in jet flow and jet break-up

Reynolds Number (Re) = nRU 20 = cesViscousForcesInertiaFor

Capillary Number (Ca) = oU = sionSurfaceTencesViscousFor

Weber No (We) = sionsurfaceTencesInertiaForRU no 2

2

Also Rayleigh time scale for capillary break up (tr) = )2( nR

33


It is considered that the drop formation in the ink jet printing process is e governed by the timing of the electric pulse. This piezoelectrically controls the ink flow, inertia forces, viscous forces, surface tension and gravity. The final process model consists of two, one-dimensional equations originating from the kinematic boundary condition, the normal and tangential components of the traction boundary condition and the z-component of the momentum equation, thus:

gzuR

zRzH

zuu

tu 2

23)2(

(4.1)

zuR

zRu

tR

21

(4.2)

where R is the radius of the jet (or drop) at an axial position z, and time t, u is the axial velocity, g is the gravity and 2H is given by:

2/32

22

2/12 ])/(1[/

])/(1[12

zRzR

zRRH

(4.3)

Figure 4.1 Predictions and experiment results of, flow of carbon nanotube – PEDOT: PSS solution at a constant flow rate of 60 ml/h

34


- 400

0- 80 0 80

Figure 4.2 Constant flowrate of 43 ml/h: Predictions of drop formation at 80 μs (dimensions in μm)

Equations 4.1-4.3 describe an ideal constant flow rate problem. The explicit finite difference technique was applied for the numerical solution of Equations 4.1-4.3. An experimental study was carried out of the flow of carbon nanotube-PEDOT:PSS solution (μ 8.8 mPa-s and = 68 mN/m) through a needle of 60 μm internal diameter at a constant flow rate to validate the computer simulations. Flow rates of 60 ml/h resulted in a continuous jet stream, and this was also predicted by the computer simulation, see Figure 4.1. On the other hand, as the flow rate was lowered below Wecritical = 15, drop formation occurred. This was also predicted by the computer simulation as is illustrated in Figure 4.2.

An experimental case-study [9] of ink jet printing was considered to validate the computer simulations, where =1000 kg/m3, μ=1.3 mPa-s, = 60 mN/m, Rn = 20 μm, flowrate = 38.4 ml/h, tpulse = 100 μs. Drop formation and break up were predicted at about 50 μm distance from the nozzle. The falling droplet grew and evolved to a cap. The experimental case study showed many similarities, including the droplet break up time of about 123 μs, which agrees very well with the predictions. However, the predicted travelled distance is smaller than in the experiment. This is attributed to the fact that the experiment is not under constant flow rate during the pulse. As a result, the computer model needs modifications to account for a pressure waveform at inlet.

35


-400

-350

-300

-250

-200

-150

-100

-50

0-80 0 80

z

120 μs

-400

-350

-300

-250

-200

-150

-100

-50

0-80 0 80

z

130 μs

Figure 4.3 Predictions at a constant flowrate of 38.4 ml/h and a pulse time of 100 μs,

and comparison with experiment

36


4.2 Characterisation of PEDOT-PSS Inks and Products

Materials used for the case study included PEDOT:PSS poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) 1.3%wt in water (Sigma Aldrich), Dimethyl Sulfoxide (Fischer Scientific) and Surfynol (AirProducts). For drop cast conductivity and surface tension analysis, PEDOT:PSS/DMSO(0-5 wt%) and PEDOT:PSS/DMSO(0-5 wt%)/Surfynol mixtures were prepared at ambient conditions and allowed to mix in a sonication bath for 24 hours Samples were drop spread onto cleaned glass slides between the ends of conducting aluminium tapes. All samples were allowed to anneal in ambient conditions for 30 minutes at 130 oC.

Surface tension was determined via the pendant drop method using a Kruss EasyDrop. The liquid was suspended from an 1832 μm needle with equatorial and neck diameters measured for the largest stable drop. The surface tension was calculated using the relationship:

gaB2

(4.4)

where is the effective radius of the drop base, g is gravity, is the density and is the surface tension. Similarly:

e

s

DDS

(4.5)

where De is the equatorial diameter of the drop, and Ds is the diameter of the drop at a distance De from the base.

(4.6)

The results in Figure 4.4 demonstrate that surface tension increases when DMSO is added, which is undesirable in ink jet printing. When 1 wt% Surfynol is added, it lowers the surface tension below 30 mN/m for 0 – 5 wt% DMSO. Figure 4.5 shows that for 0 wt% DMSO the addition of 1 wt% Surfynol decreases resistively from 66 k / to 52 k /. It has been shown [6] that the use of a surfactant has a similar effect on conductivity as the solvent, suggesting that the long chain molecule of the Surfynol causes a similar disruption to the phase separation, and hence leads to a greater PEDOT to PSS surface ratio [10]. However, higher concentrations of DMSO bring down electrical resistivity at a lower rate in the presence of Surfynol. For example, for 5 wt% DMSO, a = 314 / with 1% Surfynol and a = 81 / without any Surfynol.

37


0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 1 2 3 4 5 6DMSO (% wt)

Surf

ace

Tens

ion

(N/m

)0% Surfynol1% Surfynol

Figure 4.4 Surface tension of PEDOT:PSS/DMAO solutions, with and without Surfynol, as a function of DMSO concentration.

1

10

100

1000

10000

100000

1000000

0 1 2 3 4 5 6DMSO (% wt)

Surf

ace

resi

stiv

ity (

/squ

are)

PEDOT / DMSO

PEDOT / DMSO / Surf

Figure 4.5 Surface resistivity of PEDOT:PSS/DMSO solutions, without and

with 1 wt% Surfynol, as a function of DMSO concentration.

4.3 Characterisation of Carbon Nanotube - PEDOT:PSS Nanocomposite Products

An amount of 0.023 g Elicarb multiwall carbon nanotubes (MWNT), supplied by Thomas Swan Ltd., was added to 25.47 g of 1.3 wt% PEDOT:PSS solution in water. The mixture was sonicated for 1 hour and it was then subjected to high shear mixing at 18,000 rpm for 1 hour. The resulting ink had a viscosity of 50 mPa-s. Before further processing, it was further diluted with de-ionised water by 5:1, yielding a reduced viscosity of 8.8 mPa-s. The diluted PEDOT:PSS ink had a surface tension of 68.3 mN/m either with or without the prescribed concentration of MWNT.

38


The fabrication method involved drop spreading on a Melinex film substrate. The drops were produced using a syringe pump at a constant flow rate of 86.5 ml/h, and a pulse time of 0.6 s. The syringe had a 36 gauge steel needle with a 60 μm inner diameter, and 180 μm outer diameter. The experimental procedure was repeated for paper and cardboard substrates. In all cases, printing was carried out incorporating: 1 print pass (1 printed drop), 2 print passes (2 printed drops overlaid), 3 print passes (3 printed drops overlaid) and 4 print passes (4 printed drops overlaid).

1.E-01

1.E+01

1.E+03

1.E+05

1.E+07

1 2 3 4Number of drops

Surf

ace

resi

stiv

ity

(Moh

m/s

quar

e) MelinexPaperCardboard

Figure 4.6 Surface resistivity as a function of the number of printed drops for different substrates

0

20

40

60

80

100

1 2 3 4

Number of drops

Tran

spar

ency

(%)

Figure 4.7 Transparency as a function of the number of printed drops on Melinex substrate

39


Figure 4.6 presents measurements of the surface resistivity, a, of the products. Starting from the Melinex substrate, it can be seen that a is high. This is due to (a) the dilution of the MWNT-PEDOT:PSS ink, and, (b) the fact that no DMSO was used in this experimental section. It can also be seen that the products on the Melinex substrate exhibited the lowest electrical resistance compared to the paper and cardboard substrates. This was especially found to be the case for 1 print pass (1 printed drop only). The reason for this is that paper and cardboard substrates are highly porous. Therefore, the ink permeates and spreads, where as it stays on the surface of the Melinex substrate. However, a is reduced as the number of printed drops is increased, which also reduces differences between products with different substrates. Figure 4.7 presents the transparency of drop prints on the Melinex substrate for different numbers of print passes. For example, the transparency of a print with 2 printed overlaid drops is 71%, while the value of a is fairly close to that of prints with a higher number of printed overlaid drops (Figure 4.6). Transparency reduces almost linearly, by approximately 7-9%, for each additional print pass.

(a) (b)

Figure 4.8 Optical micrograph of the border of printed drop on paper substrate: (a)1 printed drop; (b) 3 printed drops

Figures 4.8 and 4.9 illustrate the microstructure of printed paper and cardboard substrates. The random fibre structure, with micropores, is evident in both substrates. The cardboard has larger micropores, facilitating ink impregnation. For both porous substrates, there is a darker ink ring at the perimeter of the drop.

40


(a) (b)

Figure 4.9 Optical micrograph of the border of printed drop on paper cardboard: (a) 1 printed drop; (b) 3 printed drops

4.4 Electrospinning of Carbon Nanotube Inks

An amount of 0.032 g of Elicarb MWNT, supplied by Thomas Swan Ltd., was added to 10 ml of distilled water that contained 0.1 g of dodecylbenzenesulfonate as a surfactant. The mixture was sonicated for 2 hours. After this, 40 ml of 12% polyvinyl alcohol (PVA) in water was then added. The mixture was then subjected to shear mixing at 15,000 rpm for 1 hour. The suspension had a viscosity of 48 mPa-s.

The resultant ink was electrospun from an 18 gauge steel needle at a distance of 150 mm from the ground collector, which was a multiple wire drum rotating at 2 rpm. The electrospinning took place under a potential of 20 kV, while the MWNT – PVA ink was fed by a syringe pump at a constant flow rate of 4 ml/h. After 20 min of electrospinning, a fibrous mat of 300 x 810 mm was fabricated and left to dry for 24 hours.

Figure 4.10a presents an optical micrograph of the fibrous nanocomposite product, which shows the formation of continuous oriented nanofibres of about 200 nm diameter (Figure 4.10b). This yielded anisotropic electrical properties for the fibrous nanocomposite product, with an electrical conductivity of about 400 times higher in the direction of fibre orientation compared to the in-plane direction perpendicular to the fibres.

41


(a) (b)

Figure 4.10 Carbon nanotube – PVA nanocomposite fibre assembly deposited on Melinex substrate by electrospinning: (a) optical micrograph (b) scanning electron

microscope (SEM) micrograph

4.5 Conclusions

The study presents the first results of an investigation on the ink jet printing and electrospinning of PEDOT:PSS, and related carbon nanotube nanocomposites. A flow model has been assembled for continuous, and pulse flow, under constant flow rate. This was successfully validated under constant flow rate conditions. A comparison of the predictions with the experimental data of a study of piezoelectric ink jet printing [9] led to the conclusion that the pressure drop term needs to be included in the model to cater for varying flow rate under a pressure waveform inlet condition. Furthermore, electric field terms need be included for the modelling of electrospinning. However, the example of the computer simulation of ink jet printing indicated that a surface tension below 60 mN/m might be required for the formation of small droplets, since the predicted drop width of 150 μm for a surface tension of 60 mN/m was rather large. The surface tension of PEDOT:PSS/DMSO solutions was successfully reduced from 60-70 mN/m to below 30 mN/m for 1-5 wt% DMSO through the use of 1 wt% Surfynol. The addition of DMSO was essential to lower the surface resistivity of PEDOT:PSS in the presence of 1 wt% Surfynol, from 52 k / (0% DMSO) to 314

/ (5 wt% DMSO). The addition of MWNT to PEDOT:PSS resulted in increased surface resistivity.

The carbon nanotube – PEDOT:PSS ink was deposited as printed drops on three different substrates: Melinex, paper and cardboard. The printed drops on paper and

42


cardboard had higher resistivity than the drop printed on Melinex. This was due to the high permeability and porosity of paper and cardboard. However, the resistivity of two overlaid printed drops on paper and cardboard fell substantially to levels similar to that of the resistivity of two printed drops on Melinex. Therefore, it might be concluded that two print passes on paper and cardboard are needed to achieve a reduced resistivity. On the other hand, the transparency of two overlaid drops on Melinex is 71%, against a 79% transparency of one drop. An electrospinning run of MWNT – PVA ink resulted in an oriented, continuous, fibre mat deposited on Melinex. This had an average 200 nm fibre diameter and an electrical conductivity 400 times higher in the direction of the preferred fibre orientation, compared to the in-plane directi on perpendicular to that preferred fibre orientation.

References

1. BL.Groenendaal, F.Jonas, D.Freitag and H. Pielartzik, Advanced Materials, 2000, 12, 7, 481.

2. B. Ballarin and A. Fraleoni-Morgera, Synthetic Metals, 2004, 146, 201.

3. M.A. Lopez, J.C. Sanchez and M. Estrada in Proceedings of the 7th International Caribbean Conference on Devices, Circuits and Systems, Cancun, Mexico, 2008, p.165.

4. J. Bharathan and Y. Yang, Applied Physics Letters, 1998, 72, 21, 2660.

5. Z. Liu, Y. Su and K. Varahramyan, Thin Solid Films, 2005, 478, 275.

6. S. Sapp, S. Luebben, Y.B. Losowi, P. Jeppson, D.L. Schulz and A.N. Caruzo, Applied Physics Letters, 2006, 88, 15.

7. E.Garnett and D.Ginley, Journal of Undergraduate Research, 2005, 5, 24.

8. J. Ouyanga, Q. Xua, C.-W Chua, Y. Yanga, G. Lib and J. Shinarb, Polymer, 2004, 45, 8443.

9. A-S Yang, C-H Cheng and C-T Lin in Proceedings of the Institute of Mechanical Engineering Part C: Journal of Mechanical Engineering Science, 2006, 220, 4, 435.

10. S.K.M. Jonsson, J. Brigerson and X. Chrispin, Synthetic Metals, 2003, 139, 1.

43

5 Highly Conductive Plastics – Custom-formulated Functional Materials for Injection Mouldable Electronic Applications

Walter Michaeli, Tobias Pfefferkorn and Jan Fragner

5.1 Introduction

Electric and electronic products have gained increasing importance in a number of industries. Becoming ever more complex, they are being used in highly integrated assembly groups. In addition, there is a growing trend for miniaturisation, particularly in the fields of electronics, electrical engineering, communications engineering and automotive engineering [1].

Conductive components in electric and electronic products have to fulfil different requirements depending on their field of application. For example, a very high electrical conductivity is needed for plug-in connections, elements in transmission and engine control systems. Defined ranges of conductivity are required for control units, sensors and enclosures. For the latter application, electromagnetic shielding is also of importance. Lower electrical conductivity values are demanded for applications that have to prevent electrostatic charge.

Polymers are typical insulation materials due to their extremely low electrical conductivity. Interest in using polymers for other electrical applications has increased due to advantageous properties such as weight, processibility and resistance to chemicals. Over the years, thermally and electrically conducting polymers have been developed by the addition of common fillers such as carbon black, graphite, metallic fibres, flakes or carbon fibres and, increasingly, nano-fillers such as carbon nano-tubes. These compounds have already been deployed successfully in a range of antistatic and electromagnetic shielding applications. In order to ensure a high degree of electrical conductivity, a high content of conductive fillers is required, which forms a close percolation network [2, 3].

For each material composition, a compromise has to be found regarding the amount of filler. Higher filler contents commonly have a negative influence on the mechanical properties and processability due to the considerable increase in melt viscosity.

44


Concurrently, the wear on machinery and mould is higher than for unfilled polymers. The material compositions used so far are unable to fulfil the demands of upcoming component functionality, particularly for thin-walled and miniaturised elements. Hence, functional components that require high electrical conductivity, are still typically produced in cost-intensive processing and assembly steps (e.g., insert moulding, hot stamping or metallising) [2, 4-7].

The filler content, and thus the electrical conductivity, can be increased significantly without decreasing the processability by using metal alloys that exhibit a low melting point [8-10]. These metal alloys are liquid in the processing phase and will not solidify before the cooling phase. This allows the production of complex moulded parts with definite electrical and thermal properties. As a result, material related disadvantages are reduced in comparison with highly filled moulding compounds.

The reference material under investigation consists of 15 wt% (56 vol%) polyamide 6 (PA6) (type 6NV12 by A. Schulman Inc, Akron, USA), 33 wt% of a low-melting tin/zinc alloy with a melting point of 199 °C and 52 wt% of fine copper fibres (average length: 0.65 mm, diameter: 35 μm), developed in cooperation with Siemens AG, Germany. This highly filled polymer/metal hybrid material allows passage conductivities of 105-7 106 S/m and conductivities over the wall thickness of 103-105

Figure 5.1 Electrical conductivity in comparison to other materials and potential applications

45

Highly Conductive Plastics – Custom-formulated Functional Materials for Injection Mouldable Electronic Applications

S/m [10]. These values are comparable with conductivities of pure metals (Figure 5.1).

Thus, it becomes possible to produce conducting structures and junctions for plug-in connections and/or cables in a single processing step by means of an injection moulding process. Time-consuming joining and soldering processes can be avoided. Figure 5.1 shows some possible applications. Contact pins can be directly integrated into electrically conducting elements, and electromagnetic shielding, achieving modules that are more efficient at cooling devices and motors.

5.2 Characteristics of the Hybrid Compound Within the Injection Moulding Process

5.2.1 The Cavity Filling

The thermoplastic/metal compounds can be processed in conventional single or multi-component injection moulding processes. Due to the high metal content, the filling and freezing behaviour of the materials differs from that of unfilled thermoplastics. That is caused by both the solid and liquid filler. The tests were carried out using an injection moulding machine of the type Allrounder 320 S 500 - 150 (Arburg GmbH + Co KG, Lossburg, Germany), which has a screw diameter of 30 mm (L/D = 20). First, the effects on the flow properties of the new material are discussed, followed by the characterisation of local filler distribution. Unfilled and low-filled thermoplastics show a typical frontal flow during filling of the mould. This is typical for viscoelastic fluids, whereby a parabolic velocity profile is formed over the parts thickness [11, 12]. With increasing filler content, the velocity profile generally become more block-shaped [13-15]. These typical flow patterns can be determined by mould filling studies of unfilled PA6 and material exclusively filled with copper fibres (Figure 5.2). The evenness of the filling is slightly reduced by copper fibres due to the decreased melt elasticity. The highly filled thermoplastic/metal compound shows a significantly altered filling behaviour due to the high filler content of 85 wt%. Beside the significantly reduced melt elasticity of the compound; the rapid heat flow, increased thermal conductivity and the specific phase transition of the metal alloy are the main reasons for this behaviour.

46


5.2.2 The Morphological Structure

In addition to the flow behaviour, the morphological structure of the highly filled compound directly influences the property of the part being produced. The combination of copper fibres and metal alloys with low melting points forms a dense, three-dimensional metal network, as can be seen in Figure 5.3. The positive effect of the metal alloy can be demonstrated in a close up view of a cross section. Due to the fine dispersion of the alloy, and good affinity of copper fibres and alloy, the distance between metal particles can be significantly reduced. This explains the high passage conductivity observed in the range of 106 S/m (Figure 5.3). By contrast, polyamide exclusively filled with copper fibres achieved conductivity values that were approximately three powers of ten lower, even when the filling degree was the same.

Figure 5.3 also shows that there is no linear relationship between the quantity of filler content and conductivity. At a low filler content, the conductivity of the matrix polymer is dominant. With increasing fibre content, the contacts between the fibres increase and the specific conductivity escalates in a small window. The critical concentrations of volume required for this escalation – called the percolation threshold – range between 20-25 vol% for both the metal/thermoplastic hybrid material, and the polymer exclusively filled with copper. Above this concentration, the metal network becomes more dense to an extent where conductivity can be increased only marginally [3, 16-18].

Figure 5.2 Comparison of mould filling studies of different materials

47


5.2.3 Dependence of the Conductivity on Geometry, Material and

Process Condition

The measured conductivity is not constant and changes considerably if the resistance

is measured through the interior or over its thickness, thus including the surface.

Figure 5.3 Formation of a 3-dimensional network and the improved passage conductivity (CT = computed tomography)

Figure 5.4 Influence of the geometry (wall thickness) on the conductivity (h = wall thickness of the plate geometry shown in the figure on the right)

48


This is explained by the surface layer effect. Due to the flow conditions of solid filled polymers, the surfaces of the injection moulded parts contain hardly any filler material [17, 19]. This effect can be reduced significantly with a low-melting metal alloy but cannot be avoided completely. Further on, the conductivity depends on the flow path due to the flow conditions in the gating system and in the moulded part. Figure 5.4 shows that the electrical conductivity over the wall thickness rises almost linearly with increasing flow length, which correlates well with the local density of the part. Besides the increasing filler content, the outer layer effect is reduced towards the end of the flow path. The rate of increase varies depending on the geometry, the composition of the material and the process conditions. With an increasing plate thickness, the conductivity over the wall thickness shows less dependence on the part’s position.

Furthermore, the electrical properties can be adjusted by varying the matrix polymer and the filler distribution. Since the material concept allows for the use of almost any polymer matrix, the required material properties, for instance a high temperature resistance, can be provided by the use of polyphenylene sulfide (PPS) as a matrix polymer. The morphological structure, and thus the measured conductivity over the thickness, depend on the ratio of the melt temperatures and viscosities of both the polymer and metal alloy, whereas the passage conductivity is at a similarly high level of > 4 105 S/m.

In the case of the polyethylene (PE) matrix, the metal alloy solidifies first leading to a very even and fine distribution of the metal on the surface as well as in the interior (Figure 5.5). Therefore, the conductivities measured over the thickness are

Figure 5.5 Influence of various matrix materials on morphological structure and conductivity (LV = low viscosity and HV = high viscosity)

49


Figure 5.6 Influence of the injection velocity on the electrical conductivity over the wall thickness

almost constant along the flow path. By contrast the polymer freezes first in the PPS compound. Thus, the pressure is transmitted by the metal alloy resulting in sizeable metal alloy agglomerates in the part’s centre and a high amount of the metallic phase that can be detected near the surface. Due to the big difference between polymer melting and mould temperature, the free flow cross-section decreases rapidly. This leads to a higher orientation of the filler material at the surface. The influence of the shear viscosity is analysed by comparing a low and highly viscous polyamide.

The latter shows very distinctive shear sections in the outer layers leading to conductivities over the thickness that are strongly dependent on the flow path. However, the viscosity itself has no significant influence on the distribution of the metal alloy. Freezing of the polymer at the same time or later than the metal alloy and a low viscosity of the material can considerably improve the level and homogeneity of conductivity.

The material properties are not only influenced by the composition of the material but also by the injection moulding process. Therefore, both the dosing and the injection parameters were systematically varied in the design of the experiments. The electrical passage conductivity is nearly constant in a broad process window due to the dense metallic network, whereas the conductivity measured over the thickness changes significantly by varying both the dosing and the injection parameters. For example, Figure 5.6 shows the influence of a low and high injection velocity on conductivity.

50


A lower injection velocity allows a longer filling phase, which reduces the formation of shear zones and hence improves the conductivity over the wall thickness.

5.2.4 Temperature Dependency on the Conductivity

Up to this point, the material properties have only been discussed at room temperature. However, in many fields of applications conductive polymers have to fulfil their functionalities (e.g., electrical conductivity) at elevated temperatures. In certain situations heating is produced from the outside, such as occur in ‘under the hood’ applications. Material warming can also occur under current load where the material is used as a conductor path in order to conduct electricity at a reasonably high current. In this process, the material is heated due to the electrical resistance of the material. It is important to consider if the conductive network would be damaged locally under thermal loading. Therefore, the time-dependent passage conductivity was determined by means of test specimens. The highest values measured at room temperature are slightly reduced until the glass transition temperature of polyamide (50-60 °C) is reached (Figure 5.7). Afterwards, the values remain constant at a high level and start decreasing at approximately 170 °C. The change in conductivity under heating is reversible, since the initial values are reached again after cooling.

The charging of samples with electric current leads to increasing material temperatures when choosing either smaller cross sections or higher currents (Figure 5.8). However, this stabilises after an initial steep increase in temperature. The final temperatures

Figure 5.7 Electrical conductivity in dependence on the part temperature

51


range between room temperature and approximately 50 °C. Higher temperatures can only be reached where a high current is combined with cross-sections under 5 mm2. Thereby, the test specimens could be made to fail when heated strongly. In order to estimate both the geometry and current load together, the current density is built showing a linear relationship with the material warming. In addition, sheathed conductors reached higher end temperatures (approximately 15%) throughout the tests due to the reduced heat dissipation of plastic sheathing.

5.2.5 Electromagnetic Shielding Effectiveness

In many applications, electrically conductive plastic compounds are also used for shielding of electromagnetic waves. Electromagnetic compatibility is becoming increasingly important as electrical systems extend further into daily life. Depending on the field of application the product must fulfil different kinds of marks, such as the CE mark [20]. These aims can be facilitated by electromagnetic shielded enclosures.

Plastics cannot alleviate the electromagnetic fields and waves due to their non-magnetic, insulating physical properties. A good shielding can only be achieved by conductive polymers, such as those filled with long steel fibres of 20 wt% [20, 21]. The combining of copper fibres with metal alloy is not only advantageous for electrical conductivity but also for electromagnetic shielding (Figure 5.9). To show this effect, the shielding of plates with a wall thickness of 2 mm was measured using transmission-line measuring equipment. Due to the pronounced metallic network of the hybrid material, a very

Figure 5.8 Material warming under current load

52


high shielding effectiveness can be reach compared to that obtained with other filled polymers. The measured values are in the range of the resolution of the equipment and on the same level as a metallic plate. In addition, the values are practically independent of the frequency.

As long as the filler content enables a dense, conductive network, the shielding effectiveness remains at a high level. Despite the lack of homogeneity in the metallic network of a test place with a wall thickness of 1 mm (Figure 5.4), a shielding effectiveness of approximately 58 dB can be achieved. Thus, the hybrid material can be considered suitable for shielding applications and can enable the substitution of metallic plates or coatings.

5.2.6 Injection Moulding of Conductor Paths

High thermal conductivity changes not only the flow behaviour but can also lead to the premature freezing of the material. As a result, the maximum filling length is reduced compared to the unfilled matrix polymer. It is often not necessary to take this behaviour into consideration for the production of shielding housings, and two-dimensional components with short and medium flow lengths. However, it plays an important role in the realisation of high flow path/wall thickness ratios. In these situations it may be necessary to directly manufacture multifunctional electronic components like electrically conducting structures on electrically insulating polymer carriers.

Figure 5.9 Electromagnetic shielding of conductive polymers

53


Altering the mould temperature can significantly influence the flow and freezing behaviour of polymer melts in the mould cavity [22]. In order to analyse the influence of the conductor cross-section on the flow length, conductor parts with different cross sections were injection moulded on unfilled polymer plates. In addition, the mould temperature was varied in a wide process window (Figure 5.10). It can be seen that a flow length of considerably more than 100 mm can be realised for a cross-section greater than 1 mm² with an oil-heated mould. Conductor cross-section sizes under 1 mm² only allow a flow length of less than 50 mm. Thus, the limit of the lowest conductor cross-section size is approximately 1 mm² due to the filler size used (length of copper fibres L = 0.3-0.8 mm). Furthermore, conductor paths with a greater height than width offer increased contact area with the mould, and therefore offer the preferred option. By means of an increased mould temperature of 180 °C the flow length can also be disproportionately improved.

Especially for the application of the material in conducting and/or shielding structures, the conductive elements and non-conductive carriers must be integrated in the same electronic system. In first preliminary tests the bond strength is tested using two-component tensile bars made of conducting compound and unfilled matrix polymer PA6. When applying the filled component to the unfilled polymer the stability only decreases marginally. The two components show good bond strength compared to conventional highly filled melts. Therefore, the compound is suitable for the employment in multi-component moulded assemblies.

Figure 5.10 Flow length in dependence on the conductor cross-section and the mould temperature

54


5.3 Conclusions and Outlook

The investigations presented show the high potential of the novel compounds for the use in moulded parts with high electrical conductivity. The material range stretches from carbon black compounds to metals and forms the transition between semiconductors and conductors.

The described thermoplastic/metal hybrid materials have a high potential for being employed in the production of complex electronic components with very high demands on the electrical conductivity.

Due to the injection moulding process moulded parts of highly filled material show a dependence of the distribution and direction of the filler material on the local position along the flow path and over the part’s thickness. Therefore, in the design of the moulded part it is important to consider not only global characteristic values but also the influence of the material composition, the process parameters as well as the geometries of gating system and cavity. This also makes it possible to systematically influence morphology and part properties during the compounding and injection moulding of the polymer/metal hybrid materials.

It can be briefly shown that the matrix polymer can be varied widely in order to adjust the compound properties for specific applications’ requirements. Thereby, a freezing of the polymer at the same time or later than the metal alloy and a low viscosity of the material can considerably improve the level and homogeneity of conductivity.

Moreover, the conductivities of the hybrid material are only slightly influenced by elevated temperatures. The conductivity is not considerably reduced until the softening temperature of approximately 200 °C is reached. As long as the cross-section is greater than 5 mm² at a comparatively low current load of approximately 10 A, the hybrid material does not heat up critically and can be used for conductor paths. Besides an outstanding electrical conductivity, an excellent shielding effectiveness is provided by the hybrid material due to the pronounced metallic network. This allows the material for being employed in the field of electromagnetic shielding applications.

High filling degrees and high thermal conductivity change the filling and freezing behavior of the material. This is particularly important for the production of conducting structures with high flow path/wall thickness ratios and may require an adjustment of the mould temperature control. First tests already point out that the minimum cross-section size is approximately 1 mm². Current research projects at the IKV are aimed at developing mould and process solutions for an optimised material processing of complex multi-component parts. Thereby, the use of inductive heating should be investigated at the IKV in order to further increase the flow distances.

55


Acknowledgements

The investigations set out in this report received financial support from the Federal Ministry of Economics and Technology (BMWi) by the Arbeitsgemeinschaft Industrieller Forschungsvereinigungen eV (AiF No. 15258 N), to whom we extend our thanks.

References

1. D. Drummer and R.Dörfler in Spritzgießen 2007, VDI-Verlag, Düsseldorf, Germany, 2007, p.141-157.

2. H.J. Mair and S. Roth, Elektrisch Leitende Kunststoffe, Carl Hanser Publishing, Munich, Germany, 1989.

3. A.K. Bledzki, L. Subocz and J. Subocz, Kunststoffe, 1992, 82, 3, 243.

4. B. Pfeiffer in the Proceedings of the OTTI Conference: Elektrisch Leitfaehige Kunststoffe, Regensburg, Germany, 2006.

5. J. Versieck in the Proceedings of the SPE Annual Technical Conference - ANTEC, Orlando, FL, USA, 2000, Paper No.290.

6. D.J. Jendritza, Auto & Elektronik, 2002, 3, 12.

7. K. Feldmann, 3D-MID Technologie, Carl Hanser Publishing, Munich, Germany, 2004.

8. S. Prollius and C. Hopmann, inventors; Foerderung Institut Kunststoff, assignee; DE 19,962,408, 2001.

9. E. Haberstroh, M. Hoelzel, M. Koch and E. Krampe, Kunststoffe, 2004, 94, 3, 106.

10. W. Michaeli and T. Pfefferkorn, Plastics, Rubber and Composites: Macromolecular Engineering, 2006, 35, 9, 380.

11. P. Thienel, Der Formfüllvorgang beim Spritzgießen von Thermoplasten, Rheinisch-Westfälische RWTH Aachen University, 1996. [PhD thesis]

12. G. Menges, W. Michaeli and P. Mohren, Anleitung zum Bau von Spritzgieß-Werkzeugen, Carl Hanser Verlag, Munich, Germany, 1999.

56


13. R. Schulze-Kadelbach, Fließverhalten Gefüllter Polymerschmelzen, Rheinisch-Westfälische Technische Hochschule Aachen, 1978. [PhD thesis]

14. P. Barth, Pulvermetallspritzgießen – Ein Beitrag zur Verarbeitung Ultrahoch-Fefüllter Kunststoffe, Rheinisch-Westfälische Technische Hochschule Aachen, 1988. [PhD thesis]

15. J.P. Greene and J.O. Wilkes, Polymer Engineering and Science, 1995, 35, 21, 1670.

16. G. Ruschau and R. Newnham, Journal of Composite Materials, 1992, 26, 18, 2727.

17. J.M. Knothe, Elektrische Eigenschaften von Spritzgegossenen Kunststofformteilen aus Leitfaehigen Compounds, RWTH Aachen University, Dissertation, 1996. [PhD thesis]

18. W-Z. Cai, S-T. Tu and J-M. Gong, Journal of Composite Materials, 2006, 40, 23, 2131.

19. D.M. Bigg, Advances in Polymer Technology, 1984, 4, 255.

20. U. Leute in the Proceedings of the OTTI Conference: Elektrisch Leitfaehige Kunststoffe, Regensburg, Germany, 2007.

21. S. Roth, Spritzgegossene Abschirmgehäuse aus Stahlfasergefüllten Thermoplasten – Materialeigenschaften, Verarbeitung und Gestaltung, Technical University Chemnitz, 2006. [PhD Thesis]

22. G. Wübken, Einfluss der Verarbeitungsbedingungen auf die innere Struktur thermoplastischer Spritzgussteile unter besonderer Berücksichtigung der Abkühlverhältnisse, RWTH Aachen University, 1978. [PhD thesis]

57

6 Additives in Polymer Electronics

Daisuke Fujiki

6.1 Introduction

Ciba is global specialty chemicals company. The roots of Ciba date abck to 1758 when JR Geigy Ltd., began trading in chemicals and dyes in Basel, Switzerland. In 1970, Geigy merged with Ciba - a Basel based chemical company founded in 1884 - to form Ciba Geigy Ltd.,. Ciba Geigy and the pharmaceutical compant Sandoz merged in 1996 and formed one of the world’s largest life science groups - Novartis. As a result of this merger, the specialty chemical division were spun off as Ciba Specialty Chemicals, Inc., in 1997. The company adopted the name Ciba, Inc in 2007. In April 2009, Ciba was aquired by BASF SE.

Ciba has three segments: Plastic Additives, Coating Effects and a Water and Paper. The Plastic Additives segment deals with colour and additives for plastics, lubricants and home & personal care. The Coating Effect segment is dedicated primarily towards the coating and imaging industries, while the Water and Paper segment provides solutions for paper and water treatment businesses.

6.2 Degradation and Stabilisation of Polymer

Many kinds of polymers are used in the electronics industries, from polyethylene to so-called super engineering plastics, such as polyethersulfone or polyimide. Almost all of them require additives. The reason for this is either to retain intrinsic characteristics or to extend those characteristics. In order to retain properties, polymers need process and heat stability, thermal stability or light stability. For the acquisition of new function, many kinds of functional additives can be added. Metal deactivators, antistatic agents and flame retardants are just some examples.

Each polymer has different degradation pattern and here shows representative degradation pattern of polyolefin, the so-called autoxidation cycle (Figure 6.1).

58


Polymer is degraded by heat, energy, UV or residues of catalyst and generates alkyl radicals. This alkyl radical reacts with oxygen and form peroxy radicals. These peroxy radicals abstract hydrogen from other polymer and forms alkyl radicals and hydroperoxide. The decomposition of hydroperoxide to alkoxy and hydroxyl radicals induces additional decomposition of the polymer chain. In order to stop the radical chain reaction of degradation, stabilisers such as phenolic antioxidant, phosphites, thioether and hindered amine light stabilisers (HALS) are added.

Figure 6.2 shows some representative processes for degradation and stabilisation of polymers. Phenolic antioxidants react with peroxy radical and alkoxy radical by donation of hydrogen. Phosphites and thioether act as reducing agent for hydroperoxide, which is converted to alcohol. Hydroxylamine, which is a relatively new stabiliser, acts as hydrogen donor and hydroperoxide decomposer.

As illustrated in Figure 6.3, there are two kind of light stabilisers, HALS and UV absorbers. HALS react with alkyl radicals and peroxy radicals after activation by oxygen and light. The radical stabilisation reaction should be cycled as shown.

UV absorbers absorb harmful light and the absorbed energy is converted to harmless heat. The absorbance of UV absorbers depends on the sample thickness and concentration. This law is called Beer-Lambert law.

Figure 6.1 Degradation and stabilisation of polymer

59

Additives in Polymer Electronics

Figure 6.2 Process and heat stabilisers

Figure 6.3 Light and heat stabilisers

60


Table 6.1 shows the wavelength that specific bond absorbs, which leads to the decompostion of the bond. For example, the ether bond absorbs at 320 nm wavelength and amine absorbs at 360 nm wavelength. This means the UV absorber should absorb those corresponding wavelengths of light in order to protect the polymer against UV radiation.

6.3 Stabilisers for Polymer Electronics

Normally UV comes from the sun. The polymers for outdoor applications; here is an example of photovoltaic module, need protection against UV radiation (Figure 6.4). In electrical and electronics applications, not only the sun but also the light can be the source of the UV radiation. For example, the modules in liquid crystal display (LCD) also need to be protected against UV radiation emitted by the back light. The fluorescent light possess specific sharp emission of UV radiation and the materials need protection against it.

For UV management, proper understanding for UV absorbers is necessary. Most UV absorbers, independent on the type of chromophor, absorb the damaging light and emit non destructive heat. This is the examples of benzotriazole type of UV absorbers (Figure 6.5). In the market, different kinds of substitution of benzotriazole UV absorbers are

Table 6.1 Harmful wavelength for polymers

Wavelength and hemolytic bond dissociation energy E (AB some organic molecules

, nm Bond Bond type Bond energy, kJ/mol

UV-B 230 –C–C– aromatic 520

290-310 R–O–H alcohol 420-385

290 R–CR2–H prim / sec / tert H 410 / 395 / 385

320 –C–O– ether 365-390

UV-A 240 R–CH2–CH2 aliphatic 335-370

350 CR2–Cl aliphatic chloride 330-350

360 CH2–NR2 amine 330

400 –O–O– peroxide 270

prim: primary, sec: secondary, tert: tertiary

61


Figure 6.4 UV management in polymer electronics

Figure 6.5 Side chain effect on benzotriazole UV absorbers

Figure 6.6 TGA of benzotriazole UV absorbers

available. The reason is different side chains provide different physical influence, such as volatility, compatibility to the polymer, melting point and others. One example is thermal gravity analysis (TGA) of UV absorbers (Figure 6.6). All of the UV absorber have benzotriazole chromophor but different substitutions. This gives molecules of differing molecular weight. Consequently, they also then have different volatilities. If the processing temperature is quite high, low volatility UV absorbers should be used.

62


Figure 6.7 Spectra of different UV absorbers

Figure 6.8 Thickness and concentration dependency

63


The chromophor is also an important factor to understand the UV absorber because different chromophors have different absorption patterns (Figure 6.7). Benzotriazole type has two peaks around 310 nm and 350 nm. Hydroxyphenyl-triazine has a strong peak around 270 nm and relatively weak peak around 340 nm. Benzoxazinone has specific peaks and cyanoacrylate have single peak around 310 nm. Those absorptions should correspond to the sensitive regions of polymers for better protection.

The performance of UV absorbers follows Beer-Lambert law. It means the absorbance is proportionate to the extinction coefficient of the UV absorber, concentration and the sample thickness. Figure 6.8 shows the dependency of thickness and concentration on transmission. This means that the appropriate concentration should be chosen for the thickness of the article if whole UV should be cut by the article.

As showed so far, several factors should be considered in order to choose the right UV absorber. The first factor is the thickness of the article. This determines the concentration of UV absorber. If the article is quite thin, a UV absorber with high coefficient is needed. Otherwise too much UV absorber must be added, which leads to a migration problem if the solubility in the polymer is poor. The second factor is processing temperature. Each polymer and process has specific processing temperature and the UV absorber must be stable at the processing temperature. If the UV absorber is not enough heat stable, volatiles causes several problems. The durability of UV absorber is also important factor. If the lifetime of the application is one year, most UV absorbers can do the job. But if the application has over 10 years service time, a stable UV absorber must be chosen. The last factor is the wavelength that UV absorber absorbs. Specific wavelength has two meanings. One is to absorb the wavelength of light that degrades the polymer. For example, polyesters or polycarbonates absorb around 300 nm and UV absorbers for those polymers should absorb 300 nm in order to protect polymers. The other purpose is to cut certain wavelengths. If the application needs to cut 380 nm, the UV absorber should have strong absorbance at 380 nm.

6.4 Functional Additives for Polymer Electronics

So far the retention or stabilisation of polymer characteristics has been discussed. The last part is functional additives in polymer electronics. In electronics applications, polymer tends to have contact with metals because metals are used as conductors, solder, springs, screws and other small parts. If the polymer is in contact with metals, especially transition metals, the degradation of polymer is accelerated. In that case, metal deactivation provides the solution.

Standard application of metal deactivator is jacketing of the cable because it contacts to copper wire. Copper ions can accelerate the degradation by changing valance.

64


Metal deactivators act as electron donors to form a stable complex with copper ions and the degradation reaction is retarded. Here is an example of adding 0.2% copper ion and with/without 0.2% IRGANOX® MD 1024 (Figure 6.9). IRGANOX® MD 1024 helps to retain the mechanical properties.

Figure 6.9 Function of metal deactivator

Electricity induces static charge. This causese dust pick up, which interferes with the calrity, handling problems and electrical discharges which may induce electrical shocks for fire or dust explosion. Antistatic agents can solkve such an issue.

Classical antistatic agents such as glycerine mono-stearate or amines migrate to the surface of the polymer and anchor in the polymer while the polar groups project out of polymer. The polar groups absorb the humidity in the air and form a conductive path at the surface. The drawbacks are performance over time, thermal stability, humidity dependency and migration itself, which causes the stickiness of the film. On the other hand, the discharging mechanism of IRGASTAT® is to build a conductive network in the polymer matrix and release the charge (Figure 6.10). The performance is not affected by humidity and the performance lasts a long time until the network breaks. The benefits are to minimise the risk of fire or explosion and damage to the product by electrostatic discharge during handling or transportation, on top of dust pick-up.

65


The last topic is halogen free flame retardants. Needless to say, the benefit of flame retardancy is to minimise the risk of fire accidents. Melamine derivatives are a niche in the flame retardant market, but have significant advantages in specific polymers.

Figure 6.9 Performance of IRGASTAT®

Figure 6.10 Advantages of MELAPUR® 200

66


With MELAPUR® technology, UL-94 V-0 with glass fibre reinforced polyamide is achievable (Figure 6.10). Melamine derivatives influence on heat, oxygen and fuel. Endothermic decomposition of melamine reduces the heat and nitrogen dilutes the oxygen and char formation insulates the polymer. Compared to brominated flame retardants, MELAPUR® 200 shows much less risk for corrosion. Less smoke density and less toxicity are also benefits against conventional flame retardants.

As a summary, for the best performance of polymer in electronic applications, the right combination of tailored additives must be optimised for the best protection. This allows you to differentiate your products.

Further Reading

1. H. Zweilfel, Plastics Additives Handbook, 5th Edition, Hanser Gardner Publishers, Cincinnati, OH, USA, 2001.

67

7 A Facile Route to Organic Nanocomposite Dispersions of Polyaniline – single Wall Carbon Nanotubes

Sainath G. Vaidya and Sanjay Rastogi

7.1 Introduction

Among available conducting polymers, polyaniline (PANI) has been a subject of interest due to its high conductivity, ease of synthesis, environmental stability, interesting doping and de-doping characteristics. In general the conducting form of polyaniline is regarded as intractable in nature, which cannot be processed via conventional processing routes. This is because emeraldine salt of PANI degrades before melting and the surface tension is too high to dissolve in available solvents [1]. It is known that use of functional protonic acids e.g., dodecyl benzenesulfonic acid (DBSA) facilitates polyaniline processing in a range of organic solvents [2]. Similarly, Han and co-workers [3] used DBSA micellar solution to synthesise polyaniline with enhanced processibility. Kohut-Svelko and co-workers prepared polyaniline in presence of non-ionic surfactants in water [4]. The authors reported that compared to DBSA doped PANI, polymer particle formation is much slower owing to the lengthening of the induction period. Using micellar solutions of sodium dodecyl sulfate (SDS) Kim and co-workers obtained polyaniline dispersions with the particle size of 10-20 nm [5]. It is claimed that electrical properties are affected by the presence of the SDS micelles and does not depend either upon molecular weight or morphology. The effect of surfactant anions is also investigated by Cai and co-workers [6]. The authors reported the preparation of polyaniline film in presence of sodium dodecylbenzene sulfonate and SDS with higher conductivity than that of parent polyaniline.

On the other hand carbon nanotubes (CNT) have been the subject of extensive investigation due to their remarkable properties such as very high aspect ratio, excellent mechanical, electrical, optical and magnetic properties. To exploit the properties of carbon nanotubes, efficient exfoliation of the CNT bundles in the polymer matrix is a prerequisite. There are various methods aimed at efficient exfoliation of CNT either in polymer or solvent [7-10]. In addition, nanocomposites

68


of polyaniline-single wall carbon nanotubes (SWNT) are pursued with an interest to use in variety of electronic devices [11-16].

Despite of the advances in preparations, the processing of such nanocomposites still remains a challenge due to intractability of both PANI as well as SWNT. Up until now there is no simple route to process such nanocomposite that will facilitate the use in electronic devices. Here, we used anionic surfactant as a common stabiliser for polyaniline as well as for SWNT. Based on this work, we report a facile route to obtain organic nanocomposite dispersions of polyaniline with variable concentrations of (up to 10 wt%) SWNT. From the processing point of view, this route eliminates the need for polyaniline to be de-doped followed by dissolution in suitable solvent and re-doping to regain the conductivity.

7.2 Materials

Highly purified HiPCO single wall carbon nanotubes were purchased from CNI Inc., USA. Xylene (in ortho and para forms) was purchased from VWR, UK and used as obtained. Aniline was purchased from Riedel de Haen, Germany. All other chemicals were purchased from Sigma Aldrich.

7.3 Experimental Section

7.3.1 Scheme - 1 Polyaniline Synthesis [17]

At room temperature, polyaniline nanofibres are synthesised using interfacial polymerisation between 1 M HCl in water (25 ml) containing ammonium persulfate and xylene (25 ml) containing aniline. The molar ratio of aniline to ammonium persulfate is kept at 4:1. Upon polymerization, xylene containing unreacted aniline is removed from the top and replaced with an equal volume of xylene.

7.3.2 Scheme - 2 Organic Dispersions of PANI

A portion (1 wt%) of either sodium dodecylbenzenesulfonate (NaDBS) or SDS are added to the previously mentioned system and rigorously stirred for a minimum of three hours. The system is then left idle, during which phase inversion occurs (Figure 7.1). After phase inversion, organic phase containing surfactant modified polyaniline is taken from the top and washed with excess amount of xylene. Throughout this chapter the surfactant percentage mentioned is relative to the aqueous phase.

69

A Facile Route to Organic Nanocomposite Dispersions of Polyaniline–single Wall Carbon Nanotubes

Figure 7.1 Phase inversion phenomenon; from left PANI, as prepared intermediate state and organic dispersion

7.3.3 Scheme - 3 Exfoliation of Single Wall Carbon Nanotubes

For exfoliation of SWNT, the reported method from the literature is followed [9]. In the first step 0.2 wt% SWNT is dispersed in deionised water containing 1% anionic surfactant. With the help of ultrasonication and UV-Vis spectroscopy, the optimum exfoliation of the SWNT is obtained. The final wt% of SWNT in organic nanocomposite is calculated based on the yield of PANI. The weight % of SWNT mentioned in this chapter is relative to the yield of PANI.

7.3.4 Scheme - 4 Organic ‘Nanocomposite’ Dispersions of PANI-SWNT

Organic nanocomposite dispersions of PANI-SWNT are prepared by in situ synthesis of polyaniline as shown in Scheme – 3. Ammonium persulfate is added to aqueous phase and aniline in xylene is added from the top. After rigorous stirring, the system is then left idle, during which phase inversion occurs. No traces of SWNT were observed in the aqueous phase. The nanocomposite dispersion is then taken from the top and washed with excess amount of xylene. The filtered material is then dried under dynamic vacuum at 60 C for more than 24 hours.

70


7.4 Characterisation

Fourier transform infrared (FT-IR) spectra are recorded using a potassium bromide pellet on Infrared Excalibur instrument. UV-Vis spectroscopy is done on a Hewlett-Packard 8453 spectrometer. Co-linear four probe conductivity measurements were performed on the compressed pellets (20 MPa, 10 min) using nanovoltmeter (Keithley 2182A) and programmable current source (Keithley 6220) at room temperature.

7.5 Results and Discussions

7.5.1 Phase Inversion Phenomenon

Figure 7.1 demonstrates the steps in obtaining organic dispersions of PANI.

Interestingly, an addition of surfactant to as synthesised PANI without presence of organic phase gives viscous gel-like system. Similarly, if an organic solvent is added to aqueous suspension of surfactant stabilised SWNT, no phase inversion is observed. It is observed that upon phase inversion (Figure 7.1), the left over aqueous phase was pink in colour throughout. The pink color arises due to the pernigraniline form of polyaniline, which is confirmed by UV-Vis spectroscopy and is in accordance with the observations of other authors [18-20]. Hence it can be said that this facile route to obtain organic polyaniline dispersion possess self filtering capability.

7.5.2 FT-IR Spectroscopy

Transmission mode FT-IR spectra of PANI, surfactant modified polyaniline and PANI-surfactant-SWNT is shown in Figure 7.2.

The spectra are normalised with respect to intensity at 1300 cm-1 in neat PANI. The major peaks in PANI agreed well with the literature [21]. Some similarities are observed in samples (b) to (e). The peak at 1374 cm-1, which is assigned to quinoid=N-benzene stretching mode [22] is absent. In addition, quinoid vibration peak at ~1600 cm-1

became broad, which could be resulting from the interaction of the bulky surfactant headgroups at the quinoid part of polyaniline. It is known that sulfonic acid absorb strongly in the 1050 and 1200 cm-1 region. Although this region is masked by an ‘electron like band’ of polyaniline, a weak shoulder of symmetric SO2 stretch at 1128 cm-1 is observed in samples (b) to (d) [21]. In sample (e) the electronic like band of PANI at 1140 cm-1 became broad and shifted to 1116 cm-1. On the other hand no such shift is observed in nanocomposites prepared using NaDBS [sample (d)] except that it became broader as compared to sample (b), which contained NaDBS only.

71


Figure 7.2(a) FT-IR spectra of PANI, PANI-Surfactant, PANI-Surfactant-SWNT in region of 1800-600 cm-1

Figure 7.2(b) FT-IR spectra of PANI, PANI-Surfactant, PANI-Surfactant-SWNT in region 4000-2800 cm-1

72


Therefore, it can be convincingly deduced that combination of SWNT-SDS perpetuates more restrictions on electronic like band of PANI as compared to SWNT-NaDBS.

7.5.3 UV-Vis Spectroscopy

UV-Vis spectroscopy can be used to probe the degree of doping in polyaniline [3]. As can be inferred from Figure 7.3, PANI showed only weak polaron (quasiparticle composed of an electron and its accompanying polarisation field) absorption at ~430 nm while - * transition of benzoid segment at 360 nm was absent. It is known that these two peaks often combine in single distorted peak with local maxima between 360 and 420 nm [3, 5]. Upon addition of either NaDBS or SDS, the polaron band shifted significantly to 400 nm and became sharp. This observation can be explained in terms of transformation from compact coil to expanded coil conformation, which leads to the reduction of -conjugation defects [3, 23]. With further addition of 2 wt% SWNT the polaron absorption band shifted from 400 nm to 385 nm and new weak polaron band at ~775 nm is observed. This band is related to the localisation of the cations on PANI backbone. This band is absent in pristine polyaniline as well as in surfactant modified polyaniline. Therefore the previous observations show that complex interactions exist between PANI and SWNT which are further influenced by the structure of anionic surfactant.

Figure 7.3 Absorbance spectra of PANI, PANI-Surfactant and PANI-SWNT-surfactant

73


7.5.4 Four-probe Conductivity

Four probe conductivity measurements on polyaniline-SWNT pellets are summarised in Figure 7.4. As observed from UV-Vis spectroscopy, reduction in -conjugation defects suggests higher conductivity in surfactant modified polyaniline. Indeed, the conductivity of PANI-1% NaDBS and PANI-1%SDS is 3.35 S/cm and 5.53 S/cm, respectively.

Figure 7.4 Conductivity of organic nanocomposite dispersions

This result confirms the role of surfactant as an additional dopant to PANI. In presence of 2% SWNT, the conductivity of the nanocomposite dispersions containing NaDBS is much higher (11.08 S/cm) compared to its SDS counterpart 8.01 S/cm. It is known from the literature that both NaDBS and SDS show strong interactions with SWNT and self organisation on SWNT surface [10, 24]. However, from FT-IR and UV-spectroscopy, it can be suggested that combination of SWNT-SDS stimulate more restrictions on chain conformations of PANI as compared to NaDBS counterpart. As a result, higher conductivity is observed in nanocomposites prepared using NaDBS.

7.6 Conclusions

In conclusion, we have demonstrated a facile route to obtain processable organic dispersions of polyaniline with and without SWNT. The dispersions are studied with

74


FT-IR and UV-Vis spectroscopy and four-probe conductivity measurements. The role of surfactants as a secondary dopant as well as effective stabiliser for polyaniline is recognised. This work is then extended to obtain organic ‘nanocomposite’ dispersions where it is shown that the anionic surfactants can be successfully used as an ‘interfacial link’ between polyaniline as well as SWNT. It is observed that SWNT in polyaniline matrix enhances the conductivity of nanocomposite through complex interactions in presence of the surfactant. However, the type of anionic surfactant plays a crucial role to influence the chain conformations of PANI in the presence of SWNT. The investigation of complex interactions between PANI-Surfactant-SWNT forms the part of our future work. The nanocomposite dispersion thus obtained is easy to disperse in a range of polymer matrices through either solution processing or coating route. This can also be used in coating of the inorganic particles thus facilitating development of electronic devices.

7.7 Acknowledgements

We gratefully thank Dutch Polymer Institute (DPI) for the financial support for this work. We are also thankful to Dr. Hari Upadhyaya of CREST (Centre for Renewable Energy and Systems Technology, Loughborough) and Dr. John.A.E.H.van Haare of DPI for helpful discussions and comments.

References

1. B. Wessling, Synthetic Metals, 2005, 152, 5.

2. Y. Cao, P. Smith and A.J. Heeger, Synthetic Metals, 1993, 48, 91.

3. M-G. Han, S.K. Cho and S-G. Oh, Synthetic Metals, 2002, 126, 53.

4. N. Kohut-Svelko, S. Reynaud and J. Francois, Synthetic Metals, 2005, 150, 107.

5. B-J. Kim, S-G. Oh and M-G. Han, Synthetic Metals, 2001, 122, 297.

6. L-T. Cai, S-B. Yao and S-M. Zhou, Synthetic Metals, 1997, 88, 209.

7. M.S.P. Shaffer, X.F. Fan and A.H. Windle, Carbon, 1998, 36, 1603.

8. A.R. Bhattacharya, P. Potschke, M. Adbel-Goad and D. Fischer, Chemical Physics Letters, 2004, 392, 28.

75


9. O. Regev, P.N.B. Elkati, J. Loos and C.E. Koning, Advanced Materials, 2004, 16, 248.

10. O. Mattaredona, H. Rhoads, Z. Li, J.H. Harwell, L. Balzano and D.E. Resasco, Journal of Physics Chemisty B., 2003, 107, 13, 357.

11. G.B. Blanchet, C.R. Fincher and F. Gao, Applied Physics Letters, 2003, 82, 1290.

12. P.C. Ramamurthy, A.M. Malshe, W.R. Harrell, R.V. Gregory, K. McGuire and A.M. Rao, Solid- State Electronics, 2004, 48, 2019.

13. S. Lefrant, M. Baibarc, I. Baltog, C. Godon, J.Y. Mevellec, J. Wery, E. Falques, L. Milhut, H. Aarab and O. Chauvet, Synthetic Metals, 2005, 155, 666.

14. Y. Long, Z. Chen, X. Zhang, J. Zhang and Z. Liu, Applied Physics Letters, 2004, 85, 1796.

15. M.R. Karim, C.J. Lee, Y-T. Park and M.S. Lee, Synthetic Metals, 2005, 151, 131.

16. H. Zengin, W. Zhou, J. Jin, R. Czerw, D.W. Smith, Jr., L. Echegoyen, D.L. Carroll, S.H. Foulger and J. Ballato, Advanced Materials, 2002, 14, 1480.

17. J. Huang and R.B. Kaner, Journal of the Amerian Chemical Society, 2004, 126, 851.

18. A.G. MacDiarmid, S.K. Manohar, J.G. Masters, Y. Sun and H. Weiss, Synthetic Metals, 1991, 41-43, 621.

19. K.G. Neoh and E.T. Kang, Synthetic Metals, 1993, 60, 13.

20. H.S. Kolla, S.P. Surwade, Z. Zhang, A.G. MacDiarmid and S.K. Manohar, Journal of the American Chemical Society, 2005, 126, 16770.

21. Z. Lu, H.Y. Ng, J. Xu and C. He, Synthetic Metals, 2002, 128, 167.

22. M. Trchova, J. Stejskal and J. Prokes, Synthetic Metals, 1999, 101, 840.

23. N. Kuramoto and A. Tomita, Synthetic Metals, 1997, 88, 147.

24. M.F. Islam, E. Rojas, D.M. Bergery, A.T. Johnson and A.G. Yodh, Nanoletters, 2003, 3, 269.

76


77

8 Preparation and Characterisation of Novel Electrical Conductive Rubber Blends

Kok Chong Yong

8.1 Introduction

This chapter describes an efficient method (i.e., solution mixing) for producing blends with good compatibilities and also high electrical conductivities by combining poly(butadiene-co-acrylonitrile) NBR [grade with 48 wt% acrylonitrile (ACN) content, as the rubbery host] with different proportions of polyaniline dodecylbenzenesulfonate (PAni.DBSA) (as the only electrical conductive filler). In this method, both NBR and PAni.DBSA are required to be soluble in a common shared solvent, e.g., chloroform.

8.2 Experimental

8.2.1 Chemicals and Raw Materials

For the PAni.DBSA synthesis, aniline monomer, ammonium persulfate [as the antioxidant (APS)], HCl solution (as the first doping agent), anhydrous ferric chloride [(FeCl3) as the catalyst], ammonia solution (as the dedoping agent), DBSA solution in 2-propanol (as the final doping agent), were used. The NBR with 48 wt% ACN content was washed with methanol for 24 hours using Soxhlet extraction in order to remove chemical contaminants [1].

8.2.2 Synthesis of PANI.DBSA and Blends Preparation

The PAni.DBSA with a 42% protonation level on the basis of the S:N atomic ratio was prepared by the following procedures as described in literature [1]. Masterbatch solutions of the pure NBR (50 mg NBR/ml solvent) and PAni.DBSA (16.7 mg PAni.DBSA/ml solvent) in chloroform were prepared and filtered. The pure NBR solution

78


was mixed with the PANI.DBSA solution to obtain the following compositions (wt% NBR:wt% PANI.DBSA): 50:50, 60:40, 70:30, 80:20, 90:10, 95:5, 97.5:2.5 and 99:1. Each of the blend solutions was magnetically stirred for 24 hours at room temperature prior to casting in order to enhance its homogeneity [1].

8.3 Characterisation of NBR-PAni.DBSA Blends

8.3.1 Methods of Characterisation

Characterisations of the resulting NBR-PAni.DBSA blends were performed by using:

I. Optical microscope (thickness of each sample, ~3.0 μm and 200 magnification) and transmission electron microscope (TEM) (microtome sample, 40,000 × magnification) at accelerating voltage, 80 kV [1].

II. FT-IR spectrometer, by casting a thin film (~3.0 μm) on a KBr window [1].

III. Differential scanning calorimeter for 30 to 400 °C analyses (with heating rate 20 °C/min) and -60 to 0 °C analyses (with heating rate 10 °C/min); each sample sealed with aluminium pan and analysed under a N2 atmosphere [1].

IV. Keithley electrometer to measure the electrical resistances of pure polymers and their blends cast films (thickness of each sample, ~6.0 μm) on microscope slides (6.25 cm2 0.1 cm), based on the two-probe method [1] and four-probe Van der Pauw technique [1-2].

8.3.2 Calculation of Solubility Parameter Values

Solubility parameters for pure NBR and PANI.DBSA were estimated by using the molar attraction constants values, Fi as calculated by Hoy [3]. Chloroform has a solubility parameter of 19.0 (MJ.m-3)0.5 [1]. The calculated values of solubility parameter for both NBR 48 wt% ACN and PAni.DBSA were 20.7 and 20.8 (MJ.m-3)0.5, respectively [1]. The pure NBR (48 wt% ACN) and PAni.DBSA were highly soluble in it.

8.3.3 Morphological Studies (Optical Microscopy and TEM)

Optical micrographs of blends containing 5 and 40 wt% of PANI.DBSA are shown in Figure 8.1. Two coloured regions were observed, i.e., the bright ones (bright-green in actual colour) and the dark ones (dark green in actual colour). The bright regions

79

Preparation and Characterisation of Novel Electrical Conductive Rubber Blends

(richer in NBR) are regions with well-blended NBR and PAni.DBSA [1]. The dark regions (rich in PAni.DBSA) mainly consist of large PAni.DBSA particles and their agglomerates, resulting from some degree of phase separation [1].

Figure 8.1 Optical Micrograph (200 × magnification) for blends of NBR (48 ACN wt%) - PANI.DBSA containing (a) 5 wt% and (b) 40 wt% of

PANI.DBSA. (Reproduced with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer Journal, 2006, 42, 8, 1716,

Figure 2a and 2d. ©2006, Elsevier)

Figure 8.2 TEM (40,000 × magnification) for blends of NBR 41 ACN wt%- PANI.DBSA containing (a) 5 wt% and (b) 10 wt% of PANI.DBSA. (Reproduced

with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer Journal, 2006, 42, 8, 1716, Figure 3. ©2006, Elsevier)

80


Transmission electron micrographs of blends containing 5 wt% (below percolation threshold) and 10 wt% (above percolation threshold) of PANI.DBSA are shown in Figure 8.2. Larger, discrete polyaniline particles are observed in Figure 8.2(a), which is consistent with the fairly low electrical conductivity of the corresponding blend [1]. In Figure 8.2(b), the PAni.DBSA particles are starting to form conductive pathways, which is consistent with the much higher conductivity for the related blend [1].

8.3.4 FT-IR Spectroscopy

Examples of FT-IR spectra for pure NBR (48 wt% ACN), PAni.DBSA and their blends (i.e., with 10 wt% and 40 wt% of PAni.DBSA) are shown in Figure 8.3. All the spectra of pure materials obtained corresponded well with those found in the literature [1, 4, 5].

Figure 8.3 FT-IR spectra for the pure NBR (48 ACN wt%), PAni.DBSA and their blends containing 10 wt% and 40 wt% of PAni.DBSA. (Reproduced with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker,

European Polymer Journal, 2006, 42, 8, 1716, Figure 4. ©2006, Elsevier)

81


In general, the FT-IR spectra of the blend films are merely superpositions of the spectra for both polymers with predominance to the richer phase absorption bands [1]. However, some significant shifts of certain key bands were also successfully observed, i.e., a decrease from 3447 to 3411 cm-1 for the N-H stretch and an increase from 1030 to 1080 cm-1 for the S=O stretch (which are solely due to the PAni.DBSA), attributable to changes in the intermolecular interactions with the NBR [1].

8.3.5 DSC Thermal Analysis

The sub-ambient temperature DSC thermograms (see example, Figure 8.4 and Table 8.1) of the blends had positive glass transition (Tg) shifts that increased with the proportion of PANI.DBSA up to 30 wt%, and then decreased when 40 wt% or more is added (due to more complete phase separation between the two components at the higher levels of PAni.DBSA) [1].

The above-ambient temperature DSC thermograms for the blends (see Figure 8.5 and Table 8.2) show thermal processes that are combinations of events recorded for pure NBR and PAni.DBSA, but the events show some temperature shifts relative to the corresponding processes in the pure polymers [1]. Some indication of the high-temperature miscibility of PAni.DBSA with NBR in their blends may also be gained from the onset temperatures of the major exotherm [1]. For high proportions of PAni.DBSA (40 to 50 wt%), only small shifts in the onset temperature were observed [1]. The blends containing low to moderate amounts of PANI.DBSA (10 to 30 wt%) showed large shifts in the onset temperature of the major exotherm, echoing the evidence for better mixing in these blends at room temperature and below [1].

8.3.6 Electrical Conductivity Determination

Electrical conductivities were calculated from the mean resistance values using the Van der Pauw Equation (8.1) for the samples measured by the four-probe technique [1].

2ln2R R df1 2

(8.1)

Where: σ is the electrical conductivity (S/cm), R1 and R2 are the mean values of measured resistance for a cast blend in its two perpendicular contact configurations (Ω) d is thickness of the sample (cm), and f is a geometric factor (approximately 1.00 for square-shaped plaques).

82


Table 8.1 The mean glass transition temperature (Tg) values for NBR-PAni.DBSA blends [1]

Composition (wt% NBR:wt% PAni.DBSA)

Tg (°C)

Pure NBR -11

90:10 -8

80:20 -8

70:30 -7

60:40 -8

50:50 -8

Figure 8.4 Sub-ambient temperature DSC thermograms for NBR 48 ACN wt%-PANI.DBSA blends of various compositions (wt% NBR:wt% PANI.DBSA).

(Reproduced with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer Journal, 2006, 42, 8, 1716, Figure 6. ©2006, Elsevier)

83


Table 8.2 The mean onset temperature values of major exotherms for the above-ambient DSC thermograms of PAni.DBSA, pure NBR and their blends [1]

Composition (wt% NBR:wt% PAni.DBSA)

Onset Temperature of Major Exotherm (°C)

Pure NBR 331

90:10 263

80:20 240

70:30 222

60:40 246

50:50 229

PAni.DBSA 233

Figure 8.5 Above-ambient temperature DSC thermograms for NBR 48 ACN wt%-PANI.DBSA blends of various compositions (wt% NBR:wt% PANI.DBSA)

(Reproduced with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer Journal, 2006, 42, 8, 1716, Figure 8. ©2006, Elsevier)

84


Electrical conductivities (< 10-7 S/cm) were also calculated from the mean resistance values obtained by the two-probe method using Equation 8.2 [1]:

1R

LA

(8.2)

where: is the electrical conductivity (S/cm), R is the mean value of measured resistance (voltage/current) for the cast blend (Ω), L is the measured electrode spacing distance [cm] and A is the cross-sectional area (cm2) of cast film between the electrodes.

Pure NBR has conductivity in the region of 10-14 S/cm. The PAni.DBSA had an electrical conductivity of 1.2 ± 0.5 S/cm. The electrical conductivity of all the blends increased with the proportion of PAni.DBSA. The conductivity percolation threshold for the blends was estimated by fitting the data from the curve of log blend electrical conductivity versus PAni.DBSA content (see Figure 8.6) to a simple percolation model as defined by Equation 8.3 [1, 6-7]:

f c f fp

t (8.3)

where: c is a constant, t is the critical exponent, f is the volume fraction of the conductive medium and fp is the volume fraction at the percolation threshold. All the weight fractions referred to previously were converted into volume fractions for this analysis. The values of t and correlation coefficient (R) for each case were estimated by fitting the data to a plot of log electrical conductivity ( ) versus log (f-fp).

Figure 8.6 Electrical conductivities of NBR 48 ACN wt%-PANI.DBSA blends as a function of PANI.DBSA content (wt%) [1] (Reproduced with permission from K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer

Journal, 2006, 42, 8, 1716. ©2006, Elsevier)

85


Reasonably low electrical conductivity percolation threshold, i.e., 6.0 wt%/5.4 volume % of PANI.DBSA content can evidently be achieved here (with value of t = 3.6 and R = 0.99 for blends based on NBR 48 ACN wt%).

8.4 Conclusion

NBR-PAni.DBSA blends (48 wt% ACN) with reasonably good compatibility and useful levels of electrical conductivity (up to 10-2 S/cm) could be produced by solution mixing [1]. It is also proposed that for an unstrained NBR-PAni.DBSA blend sample (as discussed in this chapter) the PAni.DBSA fillers in both of the well-blended and phase-separated-regions are all contributing to the electrical conductivity of the blend, especially the one that was below the main percolation threshold.

Acknowledgements

The author would like to acknowledge the Malaysian Rubber Board for the financial support of this project. The author also thanks P. J. S. Foot, H. Morgan, S. Cook, A. J. Tinker, C. Hull, R. Davies and K. Lawrence for their supervision, support and assistance in this part of work.

References

1. K.C. Yong, P.J.S. Foot, H. Morgan, S. Cook and A.J. Tinker, European Polymer Journal, 2006, 42, 8, 1716.

2. L.J. Van der Pauw, Philips Research Reports, 1958, 13, 1.

3. K.L. Hoy, Journal of Paint Technology, 1970, 42, 76.

4. E.L. Tassi and M-A. De Paoli, Polymer, 1992, 33, 11, 2427.

5. H. Morgan, P.J.S. Foot and N.W. Brooks, Journal of Materials Science, 2001, 36, 5369.

6. M.E. Leyva, G.M.O. Barra and B.G. Soares, Synthetic Metals, 2001, 123, 443.

7. D. Stauffer, Introduction to Percolation Theory, Taylor and Francis, London, UK, 1985.

86


87

9 Solar Textiles

Robert Mather and John Wilson

9.1 Introduction

This Chapter brings together two very important technologies: one old, the other much newer. The old technology concerns textiles: materials that have been in use for several thousand years. There are, however, many exciting developments still occurring in textile processing and application, and indeed this Chapter discusses such a development. The newer technology concerns sources of energy, alternative to those predominantly in use today. This is currently a topic of immense importance and urgency.

It is now almost universally accepted that the dependence on oil worldwide has to cease within the next few decades. Coal resources are still vast, but there are concerns about the potentially damaging environmental problems that the extraction and use of coal entail. Many people are concerned too about nuclear fission as a source of energy. Controlled nuclear fusion has not yet been achieved. On the other hand, there are numerous sources of alternative energy: geothermal, wind power, tidal power and renewable natural products. These sources, taken together, are more than adequate for the world’s current needs. Their problems are chiefly their uneven distribution around the world, and also their seasonal or diurnal variations. Many are best utilised as localised, rather than centralised sources of energy. One particularly important energy source is sunlight, in that this source is ‘endless’, and through the application of photovoltaic technology, it can be converted directly to electricity.

This Chapter sets out to make the case for the use of textiles, and the polymeric fibres that comprise them, as substrates for photovoltaic devices – or solar cells. Apart from the use of an endless energy source, there are other advantages too: a clean, silent technology, very small maintenance costs, and an attractive technology for remote areas where there is no electrical supply from a grid. It could also be a key technology for quick electrical supply to regions struck by sudden natural disasters, such as earthquakes, hurricanes and floods.

88


9.2 Why Textiles?

Solar cells are typically sandwiched between glass plates, but glass plates are heavy, fragile and inflexible. Thus, care has to be taken with their storage, transport and assembly. For these reasons, there has been increasing attention commercially to the construction of lighter, flexible cells, which can still maintain the durability required of them, even in demanding environments. There are now a number of examples of solar cells embedded onto plastic films. Amongst these examples is the ‘PowerPlastic ’, developed by Konarka Technologies Inc., where organic solar cells are printed or coated onto a flexible plastic substrate using a roll-to-roll manufacturing process. Another example is the development by PowerFilm Inc., of a roll-to-roll process for depositing layers of amorphous silicon onto a polyimide film, backed by a metal layer. Flexcell Ltd., have developed a process in which a very high frequency plasma deposition technique is used for the deposition of silicon onto a plastic roll, which has first been coated with a thin metal layer. In a process developed by Nanosolar Inc., copper indium gallium selenide ink is printed onto a metal foil.

The successful integration of solar cells on plastics is, without doubt, an important expansion of photovoltaic technology. Nevertheless, another expansion would be achieved by the integration of solar cells on textile fabrics, provided the physical and mechanical properties of the fabrics were not impaired. Textiles have a vast range of applications and markets, they can be produced by a wide variety of fabrication processes (weaving, knitting, braiding, felting), and these fabrication processes offer enormous versatility for tailoring fabric shape and properties. Textile fabrics can, therefore, be readily installed into structures with complex geometries, a feature that invites a number of potential applications.

There are already a number of examples of the use of solar cells with textiles. Both Konarka Technologies Inc., and PowerFilm Inc., have applied their solar plastic films to textiles. Zabetakis and co-workers have described the design of an awning, which provides both protection from the sun and converts the incident sunlight into electricity [1]. A sail has been designed in which flexible solar cells are attached to the sailcloth [2]. The design allows for the expansion of the sailcloth as wind blows on it. There is considerable interest too in attaching solar films to apparel.

9.3 Solar Cells

These are essentially large area semiconductor diodes that absorb sunlight to generate free electrical charges. The photovoltaic effect relies on the in-built electric field that is produced at the junction between two dissimilar semiconducting layers: the size of this field determines the voltage that may be produced by the solar cell and it is

89

Solar Textiles

responsible for separating the pairs of positive and negative photogenerated charges. These are passed to the external electrical load by electrical contacts on the top and bottom of the cell: one of these contacts is transparent to allow the sunlight to reach the semiconductor.

The choice of semiconductors is limited by the requirement to match the optical absorption to the solar spectrum: the semiconductor characteristic that defines this feature is the bandgap energy, the lowest energy that may be absorbed. If this is too high (wavelengths towards the ultraviolet (UV)) then most sunlight passes through the cell without any effect and generates a low current; if it is too low (wavelengths towards the infrared) then the cell has a weak in-built field and so generates a low voltage. Fortunately one of the commonest and most developed semiconductors, silicon, is a reasonable match for sunlight and may be produced in a thin-film form that is an even better optical absorber than the single-crystal silicon used in microelectronics. It may also be synthesised in p-type and n-type forms to make up a typical diode.

When these fundamental limits are taken into consideration, the maximum solar power conversion efficiency for a simple cell with one in-built junction is ~25%. Triple junction cells have a limiting efficiency of 32% but are more difficult to manufacture. To attain high conversion efficiency a cell must be designed with trade-offs between the various conflicting requirements, giving the variety of configurations and additional layers that are used by different manufacturers. One of the obvious features is the need to reduce cost without compromising performance: thin-film cells lie at the low cost end of the range, but have lower efficiencies. Another consideration is the energy embedded in each cell, because not only is this reflected in the manufacturing cost but also it determines the energy pay-back period for a real renewable energy source. Again, most thin-film cells have lower pay-back periods than do ‘bulk’ crystalline cells.

9.4 Technological Specifications

There are a number of key technological specifications that have to be met, for a solar textile to prove viable. The performance of the cells must not be compromised by the textile fabric: in particular, there must be efficient and durable conducting paths within the fabric to deliver the current from the solar cells. Moreover, the textile is very likely to contain additives which assist the processing of the fabric, and indeed of the fibres themselves, and additionally enhance the fabric’s performance as a product. Not only should the additives not compromise the performance of the solar cells, but also conversely, the solar cells should not impair the performance of the fabric or its associated additives. Where solar cells are bonded to a fabric surface, the surface properties will necessarily be changed. These changes will have implications for the

90


resistance of the fabric surface to abrasion and even to the adverse environments which the fabric may encounter in use. It is, therefore, essential that a protective coating is deposited over the solar textile, such as are already used for many textile fabric applications. Finally, the flexibility of a fabric and its conformability to a desired shape must not be severely compromised.

One of the most important technological considerations is matching the load to be powered by the solar textile to the current-voltage characteristics of the solar cells it contains. The current that can be drawn from the cells is proportional to the illuminated area. It also depends on the intensity of illumination and the resistance of the external load [3]. In contrast, the voltage of the cell depends on its construction. On an open circuit, a cell may produce approximately 0.5 V, but by stacking up to three cells on top of one another, the voltage is appreciably increased. Nevertheless, when the cell is connected to the external load, its voltage may be considerably reduced. Overall, however, the matching of the current-voltage characteristics to the external load is far less demanding if the area occupied by the cells is sufficiently large, and hence accessible to sufficient sunlight.

9.5 Suitable Textile Constructions

9.5.1 Fibres

Successful construction of textile fabrics as substrates for the integration of solar cells has to take into account the type and grade of polymer used, the type of fibre extruded from it, and subsequent fabric construction. Fibre selection is very much determined by its ability to withstand successfully the elevated temperatures required for the deposition of the thin layers comprising a solar cell. Where these temperatures are as high as 400 oC or more, as in the deposition of many types of solar cell, selection is restricted to high-performance fibres, such as glass fibres, polyimides, polybenzimidazoles and polybenzoxazoles, but these types of fibre are expensive. However, it has been demonstrated that nanocrystalline silicon films can be successfully laid down at temperatures as low as 200 oC [4, 5], a factor that then broadens fibre choice to include a range of commodity fibres.

The fibre must also be able to withstand prolonged exposure to sunlight, specifically irradiation by UV light. The twin considerations of thermomechanical stability up to 200 °C and resistance to UV radiation do rule out several types of commodity fibre. These fibres include most natural fibres, as well as commercial polyolefin and acrylic fibres, all of which melt or begin to decompose below 200 °C. Nevertheless, polyethylene terephthalate (PET) fibres are potentially suitable substrates: they melt at 260-270 °C and exhibit good stability to UV radiation [6]. They are also commercially

91

Solar Textiles

attractive on account of their widespread use. A huge range of different PET fabrics is available. They possess good mechanical properties and, with the exception of alkalis, are resistant to chemical attack. Fabrics made from Nylon 6:6 fibres are also widely available, although they are prone to degradation in strongly acidic environments. However, to be sufficiently UV-resistant, they need to contain suitable light-stabilisers, but it should equally be noted that commercial Nylon 6:6 fabrics are extensively used in the production of tents, hot-air balloons and parachutes!

Fabrics made of E-glass fibres could also be viable, in that the price of E-glass is similar to or only slightly higher than that of PET [7]. One advantage could be their transparency, as with plate glass in conventional solar cells. E-glass fibres are, however, prone to flexural rupture, and possess poor resistance to environments of extreme pH. Some other glasses, such as R- and S-glass, which are more stable to these extreme pHs, are also approximately 6-8 times more expensive [8].

9.5.2 Fabrics

The type of fabric construction is also important, in that it affects the fabric’s physical and mechanical properties, and its effectiveness as an electrical conductor. Whatever type of construction is used, it is important that the yarns constituting the fabric are sufficiently close together. If the fabric is too porous, the thin layers comprising the solar cells cannot be successfully deposited onto it.

For good electrical conduction, woven fabrics are generally favoured [9, 10], because they possess the best dimensional stability and can be constructed to render desired flexibilities and conformations. Furthermore, the yarn paths in woven structures are well ordered, which permits the design of complex fabric-based electrical circuits [9]. In addition to woven fabrics, embroidered fabrics may also in the future be attractive options for circuit design, since they too possess ordered yarn paths, but with much more complex patterns than in woven fabrics. Knitted fabrics, by contrast, do not retain their shapes so well, and the rupture of a yarn is liable to induce laddering. These problems are exacerbated if the shape of the fabric is continually changing, as in apparel usage. Nonwoven fabrics do not, so far, possess the strength and dimensional stability of woven fabrics; and the scope for construction of electrical circuits in them is limited, as their yarn paths are highly unoriented.

9.6 Electrical Conductivity

It is evident that, as with any type of solar cell, the top and bottom layers must be electrically conductive. The transparent top layer will consist of a transparent oxide,

92


such as indium tin oxide or zinc oxide. The bottom layer, formed by the textile fabric, must also be made conductive, and several strategies are available for achieving this goal. One approach is to process fibres from electrically conducting polymers. There has been particular emphasis on the fibre processing of poly-heterocyclic compounds such as polyaniline, polypyrrole and polythiophene. Conductivities of up to the order of 100 S/cm have been reported for these polymers [11-13]. For comparison, the conductivities of copper and graphite are approximately 6 105 S/cm and 700 S/cm, respectively. However, the mechanical properties of these conductive polymers are often inadequate [14], and they possess limited flexibility [15]. To obviate these difficulties, some fibre blends with conventional polymers have been produced, which aim to exploit the desirable properties of both polymeric components. This approach has enjoyed a measure of success [15], although the conductive polymer has to withstand the conditions, e.g., elevated temperatures, required for processing the conventional polymer fibre. However, one report has been published of conducting fibres arising from the addition of solutions of conducting polymer to strands of freshly melt extruded synthetic fibre, before the fibre has solidified [13].

Another approach is to incorporate metal filaments during fibre processing, prior to fabric construction. Whilst conducting fabrics can certainly be achieved by this approach, it necessitates the production of specially constructed fabrics for supporting the solar cells rather than the use of fabrics already available on the market. In addition, these fabrics are generally much stiffer. Nevertheless, a number of electrically conductive fabrics have been produced, particularly for military use and where safety and security are important. Deposition of conductive polymers offers another route. The conductive polymer may, for example, be deposited from a suspension. Alternatively, the conductive polymer may be formed by bulk polymerisation in the presence of the textile [14]. As polymerisation proceeds, the resulting polymer is precipitated as an insoluble solid. The polymer deposited onto the textile surface must adhere to it sufficiently strongly during the stages of integration of the solar cell and subsequent use [16]. Better control is achieved if polymerisation occurs on the textile fibre surfaces themselves. A layer of monomer is first adsorbed onto the fibre surfaces. On exposure to a suitable oxidising agent, adsorbed monomer is converted to the corresponding polymer [14]. For this approach to be successful, there must be sufficient adsorption of monomer to the textile substrate, and the substrate must be resistant to oxidising agent.

The pigmentation of synthetic fibres with carbon black has been practised for a good number of years. Latterly, its potential for promoting electrical conductivity in fibres has been explored. Nowadays, there is rapidly growing interest too within the textiles community in the incorporation of carbon nanotubes into fibres, particularly as a means of reinforcing them. However, their incorporation at a sufficient level would also render the fibres electrically conducting, and no doubt this property will be fully explored over the coming years.

93

Solar Textiles

9.7 The Future

The successful integration of solar cells on textiles opens up a whole range of possible applications. Many of these applications would exploit the light weight, flexibility and conformability of textile fabrics. For example, they can be placed over curved surfaces in buildings, and be installed in spaces that would be otherwise inaccessible, such as in automotive, marine and aerospace equipment. Moreover, textile fabrics can be rolled up, transported to a desired location and then unrolled at that location. The technology would, therefore, be valuable to those living in remote areas, and could also be used to get power quickly to disaster areas, hit by earthquakes, hurricanes, floods or fire.

In the meantime, organic semiconductors are becoming better developed, and these semiconductors are likely to become increasingly attractive alternatives to inorganic semiconductors. Many organic semiconductors are polymeric. Their use would reduce the cost of solar cell deposition. Dye-sensitisation for matching with the solar spectrum may too one day become a reality, though this day is likely to be far into the future. Both these developments would assist the future technological and commercial progress of solar textiles.

References

1. A. Zabetakis, A. Stamelaki and T. Teloniati in Proceedings of a Conference on Fibrous Assemblies at the Design and Engineering Interface – INTERDEC, 2003, Edinburgh, UK.

2. H-F. Muller, inventor; no assignee; US 6237521, 2001.

3. R.R. Mather and J.I.B. Wilson in Intelligent Textiles and Clothing, Ed., H.R. Mattila, Woodhead Publishing Limited, Cambridge, UK, 2006, p.206.

4. H. Shirai, Y. Sakuma, Y. Moriya, C. Fukai and H. Ueyama, Japanese Journal of Applied Physics, 1999, 38, 6629.

5. J. Löffler, C. Devilee, M. Geusebroek, W.J. Soppe and H-J. Muffler in Proceedings of the 21st European Photovoltaic Solar Energy Conference, Dresden, Germany, 2006, p.1596.

6. R.W. Moncrieff, Man-made Fibres, 6th Edition, Newnes-Butterworths, London, UK, 1975.

7. J.W.S. Hearle in High-performance Fibres, Ed., J.W.S. Hearle, Woodhead Publishing Limited, Cambridge, UK, 2001, p.1.

94


8. F.R. Jones in High-performance Fibres, Ed., J.W.S. Hearle, Woodhead Publishing Limited, Cambridge, UK, 2001, p.191.

9. M.S. Abdelfattah in Electronics on Unconventional Substrates – Electrotextiles and Giant-Area Flexible Circuits, Eds., M.S. Shur, P.M. Wilson and D. Urban, MRS Symposium Proceedings Volume 736, Materials Research Society, Warrendale, PA, USA, 2003, p.25.

10. E. Bonderover, S.Wagner and Z. Suo in Electronics on Unconventional Substrates – Electrotextiles and Giant-Area Flexible Circuits, Eds., M.S. Shur, P.M. Wilson and D. Urban, MRS Symposium Proceedings Volume 736, Materials Research Society, Warrendale, PA, USA, 2003, p.109.

11. A.G. MacDiarmid and A.J. Epstein, Journal of the Chemical Society, Faraday Discussions, 1989, 88, 317.

12. J-A. Pomposo, E. Ochoteco, C. Pozo, P-M. Carrasco, H-J. Grande and F-J. Rodriguez, Polymers for Advanced Technologies, 2006, 17, 26.

13. Chemical Fibres International, 2001, 51, 361.

14. A. Malinauskas, Polymer, 2001, 42, 3957.

15. D. Akbarov, B. Baymuratov, R. Akbarov, P. Westbroek, K. de Clerck and P. Kiekens, Textile Research Journal, 2005, 75, 197.

16. J.P. Boutrois, R. Jolly and C. Petrescu, Synthetic Metals, 1997, 85, 1405.

95

10 Flexible Sensor Array for a Robotic Fingertip Using Organic Thin Film Transistors (OTFT) with Minimum Interconnects and Improved Noise Tolerance

Ahmed A. Maatouk and Darek Korzec

10.1 Introduction

Organic technology has quickly developed in the past few years and has been employed in the manufacturing of several applications such as radio frequency identification (RFID) tags and thin film transistor (TFT) displays. Mainly because of its simpler fabrication process and low costs. Recently scientists have developed what is called artificial skin which mainly relies on organic polymer materials and organic devices which is OTFT. Due to its high flexibility engineers have been able to develop large area sensors array in the form of what might be a robot’s skin. This skin would be able to detect tactile forces and temperature as well as giving the robots the ability to manipulate objects and physically interact with the surrounding environment. This could also be promising in the field of medicine to people who have lost limbs as this new technology could provide some prosthetic limbs. It also could help in the diagnoses of breast cancer and remotely monitoring patients’ vital signs.

Touch sensing is the ability to detect the force of contact between an object and the sensing area at a defined point on this object and on the sensor. While tactile sensing is the ability to detect the mechanical forces directly perpendicular to the surface of the sensing area. Applications for sensor arrays are not present only in the robotics field but also in medical fields such as breast manipulation for breast cancer diagnoses and for monitoring patients’ status. Also in the automotive industry it’s used in defining tire patterns to enhance tire manufacturing for racing cars and in car seats to monitor the driver’s comfort and status in order to decrease the probability of car crashes.

There are different technologies available for the development of tactile sensors such as optical [1], capacitive [2], resistive [3] and micro electro mechanical systems (MEMS) [4] technologies. Each suffers from a limitation, for example capacitive sensors are

96


very sensitive to any external noise source, optical sensors are highly sensitive to the deformation of the source as it would cause a change in the light orientation receive by the optical receiver and last but not least in resistive technologies which mainly relies on rubber, the rubbery materials get deformed over time with constant usage.

10.2 General Description of the System

These piezoresistive materials are polymeric materials, which are mechanically flexible and could get tailored to any surface, such a tactile skin would be very thin because it’s fabricated on a plastic substrate by deposition of thin films. This artificial smart skin would incorporate its addressing electronics which is designed using flexible OTFT in this way the number of the interfacing terminals would be decreased to two terminals, one for the supply voltage while the other is for the output data and the data would be serially sent to the read out interface. the system under investigation would send the output signal in the form of an oscillating signal, the frequency of oscillation of this signal varies with the variation of the applied pressure on the addressed sensor cell and this would greatly decrease the effect of any external noise source more than if the output was sent in the form of just a voltage level. In order to proceed with specifications of the finger-tip sensors array we should first gather the specifications of a human’s tactile sensors. These data are gathered from [1-9] and are presented in Table 10.1.

Table 10.1 Specification of a human’s finger-tip tactile sensors

Frequency Response 0-400 Hz

Response Range 0-100 g/mm2

Sensitivity 0.2 g/mm2

Spatial Resolution 1.8 mm

10.3 Operation Mechanism

The sensor arrays considered in this paper are fabricated using a piezoresistive material. Piezoresistive materials are materials which exhibit a change in its resistivity upon an application of a mechanical stress on its surface. This phenomenon is called piezoresistivity. It has been found that the contact resistance could be described by the following equation [10]:

97

Flexible Sensor Array for a Robotic Fingertip Using Organic Thin Film Transistors (OTFT) with Minimum Interconnects and Improved Noise Tolerance

RcF

K (10.1)

Where Rc is the resistivity of the material and F is the stress force applied while K is function of the roughness and elasticity of the material [10]. A gauge factor to characterise the sensitivity of the piezoresistive material to the applied stress force has been reported [10] which could be described using the following equation:

GFdS

dF Kcc 1

(10.2)

Where Sc is the electrical conductance of the contact and F is the applied force. It is shown by the above equation that the material is sensitive to the variation of the applied stress forces and the lower the material resistivity (rho) and the roughness and elastic properties K the higher the sensitivity to such forces.

10.4 Robotic Finger Tip Specifications

In order to start working on the design of the electronic blocks of the system, the specifications needed for this system to function efficiently for its intended application need to be set. Based on the criteria found in [6], the specification of the robotic finger tip’s artificial skin was set, and is found in Table 10.2.

Table 10.2 Robotic fingertip artificial skin specifications

Spatial Resolution 1–2 mm2

Sensitivity 0.01-10 N

Frequency 100 Hz - 10 kHz

Area 20 20 mm2

10.5 System Design

10.5.1 OTFT Devices

All the circuits were designed using only a positive channel metal oxide semiconductor (PMOS) devices as in contrast to their silicon candidates OTFT exhibit high mobility

98


in P-type than in N-type [11]. Also, design techniques for organic circuits that were presented in [12] were taken into consideration when implementing the circuits. OTFT with holes’ mobility values up to 15 cm2/V.s were reported [13], therefore, devices used in this paper are of holes mobility 2 cm2/V.s with on/off ratio 107, Vt = -2.5 V and maximum drift velocity of 5 104 m/s.

10.5.2 Sensor Array

The sensor array under investigation which is presented in [10] consists of two polyester sheets adhered together with an adhesive at the non sensing areas each sheet has a deposited 25 m Ag-filled conducting polymer lines, upon adhering the two sheets together they form a grid of lines crossing each other forming the sensor cells at the point of contact. Such structures of these sensor arrays could be viewed in Figure 10.1(a) while Figure 10.1(b) shows the circuit model of the piezoresistive sensor [10]. The resistance of one sencel (sensor cell) is modelled with a variable resistance that varies with application of any stress force on the sencel, while the capacitance is due to the fact that there are two conductors overlapping over each other.

Figure 10.1 Electric model of a 2 2 matrix of sencels

(a) (b)

10.5.3 Active Matrix Addressing Electronics

The sensor cells are addressed using the active matrix addressing technique, using two decoders one that acts as a column driver and the other that acts as a row selector. These two decoders are synchronised using two binary counters one that iterates

99


the select inputs of the column driver decoder having its last bit connected to the clock input of the second counter which iterates the select inputs of the row selector decoder. Figure 10.2(a) and (b) shows the output simulation of the primarily designed decoder that exhibits glitches and the secondary improved one using OTFT devices both operating at a 20 kHz. Figure 10.3 shows the output simulation of a 3-bit binary counter using OTFT devices also operating at 20 kHz. It should be noted here that unlike silicon devices, different paths delays in the 3-bit counter were very significant even at a frequency of 20 kHz as it was in the order of 2 s.

10.5.4 Astable Multivibrator

A multivibrator is a circuit that oscillates between two states, if it’s an astable multivibrator then it does not settle at any of the two states and keeps on oscillating. An example of an astable multivibrator circuit is the ring oscillator circuit. The astable multivibrator is the block responsible for converting our output signal from a voltage level to a frequency modulated signal. It consists of two inverters connected with a resistance and capacitance (RC) network. Figure 10.4 shows the configuration of the multivibrator and its connection with the sensors array, the addressed cell in the sensors array is the element responsible for modulating the frequency of the output signal depending on the resistance of the addressed cell which is in turn modulated by the applied mechanical force. Figure 10.5 shows the output simulation of the output of the multivibrator running at 100 kHz designed the address of a sensor cell, using OTFT. Figure 10.6 is the all the blocks integrated together to form the whole system.

Figure 10.4 The multivibrator consists of a RC network and 2 inverters, while the frequency modulation of the signal is achieved by the variations in the load of the

first inverter which is the sensor array

100


Figure 10.2 (a) Non-improved output (b) improved output

Figure 10.3 Simulation output of a 3-bit binary counter operating at 20 kHz using OTFT

101


10.6 Important Design Features

The multivibrator should be sensitive enough to the change in the resistance of the addressed sensor cell to be able to modulate the frequency of the output signal significantly with every incremental small change in the applied force. The last sensor cell to be scanned should have a unique resistivity value so as the read out interface can mark the end of the scanning frame and the beginning of a new one. Also the read out interface should be able to mark the start and end of the scanning frame of each element as it’s only reading the data out and not addressing the sensor cells, so a time based algorithm to identify each cell should be employed because as mentioned before the addressing electronics are on the same substrate as the sensor array, and are not controlled by the read out interface as is usually done.

Figure 10.6 System level design of the finger-tip artificial skin

Figure 10.5 The oscillating output voltage of the astable multivibrator

102


Conclusion

We investigated the design of addressing electronics for an array of sensors by implementing the circuits for and simulating it on ORCAD Pspice®. Circuits were implemented using an OTFT model, and the circuits showed good performance when run at a frequency 20 kHz. OTFT are still in need of major development and research in order to increase the charges mobility and thus increasing the operation frequency.

References

1. J.L. Banks, Design and Control of an Anthropomorphic Robotic Finger with Multi-point Tactile Sensation, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Boston, MA, USA, 2001. [MSc thesis]

2. L.K. Baxter, Capacitive Sensors: Design and Applications, IEEE Press, Piscataway NJ, USA, 1997.

3. T. Someya, T. Sakurai and T. Saketani in the Proceedings of the IEEE Electron Devices Meeting, 2005, Washington, DC, USA, p.445.

4. F. Jiang, Y. Tai, K. Walsh, T. Tsao, G. Lee and C. Ho in the Proceedings of the IEEE Tenth Annual International Workshop on Micro Electro Mechanical Systems - MEMS ‘97, Nagoya, Japan, 1997, p.465.

5. P. Coiffet, Robot Technology – Interaction with the Environment, Volume 2, Prentice-Hall, Englewood Cliffs, NJ, USA, 1981.

6. Active Touch: the Mechanism of Recognition of Objects by Manipulation, Ed., G. Gordon, Pergamon, Oxford, UK, 1987.

7. L. Heimer, The Human Brain and Spinal Cord, Springer-Verlag, New York, NY, USA, 1983.

8. L.L. Langley, I.R. Telford and J.B. Christensen, Dynamic Anatomy and Physiology: Functional Neuroanatomy and Dissection Guide, McGraw-Hill, New York, NY, USA, 1974.

9. E.R. Kandel, J.H. Schwartz and T.M. Jessell, Principles of Neural Science, 4th Edition, The McGraw-Hill Companies, Inc., New York, NY, USA, 2000.

103


10. T.V. Papakostas, J. Lima and M. Lowe in Proceedings of the IEEE Sensors Meeting, 2002, Volume 2, p.1620.

11. Q. Wu, Zhang and Q. Qiu in Proceedings of the IEEE International Symposium - Circuits and Systems (ISCAS 2006), 2006, p.1270.

12. C.D. Dimitrakopoulos and D.J. Mascaro, IBM Journal of Resaerch and Development, 2001, 45,

13. J.N Tegin and J. Wikander, Industrial Robot, 2005, 32, 1, 64.

Further Reading

a) T. Someya, H. Kawaguchi and T. Sakurai in Proceedings of the IEEE International Solid-State Circuits Conference (ISSCC 2004), San Francisco, CA, USA, p.288.

b) R.S. Dahiya, and M. Valle, and G. Metta and L. Lorenzelli in Proceedings of the MRS Fall Meeting, Boston, MA, USA, 2007, Volume 3, p.1.

c) C.M.A Ashruf, Sensors Review, 2002, 22, 4, 322.

d) R.S. Dahiya, M. Valle and G. Metta in Proceedings of PRIME 2007 - 4th Conference on PhD Research on Microelectronics and Electronics, Istanbul, Turkey, 2007, p.201.

e) T. Sekitani, Y. Noguchi, U. Zschieschang, H. Klauk and T. Someya, Proceedings of the National Academy of Sciences, 2008, 105, 13,

f) R.D. Howe, Journal of Advanced Robotics, 2004, 8, 3, 245.

g) V.J. Lumelsky, M.S. Shur and S. Wagner, IEEE Sensors Journal, 2001, 1, 1.

h) R. Brederlow, S. Briole, H. Klauk, M. Halik, U. Zschieschang, G. Schmid, J-M. Gorriz-Saez, C. Pacha, R. Thewes and W. Weber in Proceedings of the IEEE Symposium on Computers and Communications, Kemer – Antalya, Turkey, Session 21/TD: Organic and Nanoscale Technologies, Paper No.21.6.

i) O. Kerpa, K. Weiss and H. Wörn in Proceedings of the 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems, Las Vegas, NV, USA, 2003, Volume 1, p.1.

104


j) N. Wettels, D. Popovic and G.E. Loeb in the Proceedings of the 2nd BioMed - Frontiers in Biomedical Devices Conference, Irvine, CA, USA, 2007, p.829.

k) P. Dario, R. Lcazzarini and R. Magni in Proceedings of the IEEE 7th International Symposium on Micro Machine and Human Science, Nagouya, Japan, 1996, p.91.

105

11 An Organic Thin Film Transistor Pixel Circuit for Active-Matrix Organic Light Emitting Diode Flat Panel Display

Mohamed M. Montasser and Darek Korzec

11.1 Introduction

Organic light emitting diode (OLED) displays have promising characteristics that will make them the dominant technology in the display market in the future. These characteristics are high efficiency, high contrast ratio, wide viewing angle and low fabrication cost [1]. However, the challenges in the development of these displays are the instability of the organic materials, as voltage fluctuations and performance degradation may occur during the operation of the organic device.

The organic thin film transistor (OTFT) is a thin film transistor which uses an organic material for its channel instead of a-Si:H or poly-Si. It is also known as an organic field effect transistor (OFET). The main purpose of fabricating the OTFT was to realise low cost, large area electronic products such as, high quality, large, flat panel displays. The OTFT is a low cost fabrication, as the deposition temperature of the organic materials is very low which allows use of inexpensive substrate materials such as glass, plastic and foils. There is great interest in OTFT nowadays from researchers, since the performance of the pentacene TFT has been improved so that it has become close to the performance of the hydrogenated amorphous silicon, a-Si:H or even better [2]. Although it became very attractive, it still has problems of instability such as the threshold voltage shift, and mobility degradation. On the other hand, OLED also has some limitations such as the formation of dark spots and the OLED luminance degradation.

In this Chapter, an electronic solution for the instability conditions of the organic devices is provided. The circuit is an example of the voltage programming technique for the compensation of the threshold voltage variation and mobility degradation of the OTFT and the OLED luminance degradation. The circuit is also simulated using PSPICE (circuit simulation tool) to verify the results.

106


11.2 Pixel Circuits Designs

The pixel circuit is a main part in the OLED flat panel display; it is the circuit that provides the OLED with sufficient current according to the data current or voltage. The challenge in the pixel circuit design is to provide high quality picture and to overcome the problems of the organic devices. Figure 11.1 shows the conventional active matrix (AM-OLED) pixel circuit; M1 is a switch transistor, M2 is the driver transistor and Cs is the storage capacitor. Although this pixel circuit has a built in memory, it suffers from the unstable OTFT conditions such as the threshold voltage variation and mobility degradation. Therefore, compensation methods need to be added to the circuit in order to overcome the problems of the organic transistors. In previous studies [3-5], an optical feedback technique was used in order to compensate for the decrease of current caused by the instabilities of the organic devices, however, this technique is inefficient as it uses an organic optical sensor which adds another source of instability to the whole system. There are several pixel driving methods that were investigated by previous studies [1, 2, 6]; they were divided between the current and the voltage driving methods. The current programming scheme can overcome the problems of threshold variation and mobility degradation of the OTFT however, the current programming method needs very high addressing speed, also the constant current source is difficult to be designed for controlling the small current levels. On the other hand, the voltage driving scheme is very efficient in compensating for the threshold voltage variation and the mobility degradation of the OTFT, but the voltage driving method become slow when a large number of gray levels are demanded.

Figure 11.1 The conventional pixel circuit [6]. Reproduced with permission from W.F. Aerts, S. Verlaak and P. Heremans, IEEE Transactions on Electron Devices,

2002, 49, 2124. ©2002, IEEE

107

An Organic Thin Film Transistor Pixel Circuit for Active-Matrix Organic LightEmitting Diode Flat Panel Display

11.3 The Voltage Programming Pixel Circuit

The voltage programming pixel circuit is the circuit that uses the output voltage (programming voltage) of the column driver to set the gate voltage of the driver transistor. The pixel circuit shown in Figure 11.2 is an example of a voltage driven pixel. This design was presented by Ashtiani and co-workers [1] and was based on a-Si:H TFT. In this Chapter the same design is developed by using the OTFT for driving this circuit and the system is converted into a pull-up network as the transistors used in the design are p-type transistors. The design consists of five p-type OTFT, three select lines, data line, power supply and an OLED. M1, M3, M4, M5 are switching transistors, M2 is the driver transistor and C1 is the storage capacitor.

This design is compensating for the threshold voltage variation and the mobility degradation of the OTFT, and also compensates for OLED luminance degradation. It operates in three different states; pre-discharging state, compensation state and driving state. During the pre-discharging period, SEL1 and SEL2 are low therefore M1, M3 and M5 are ON. Therefore the gate of M2 is discharged to a voltage near to the power supply VDD (negative supply) and stored in the storage capacitor C1. The data voltage is chosen to be –VP + VOLEDI, where VP is the programming voltage, and VOLEDI is the initial ON voltage of the OLED.

During the compensation period, SEL1 goes high and M1 is turned off, therefore C1 charges through M2 until the gate voltage of M2 reaches VOLED+VT2, where VT2 is the threshold voltage of the driving transistor, and VOLED is the shifted ON voltage of the OLED. Finally in the driving period, when SEL1 and SEL3 go low and SEL2 goes high, thus M1 and M4 are on. Then the capacitor voltage VC1 is applied to the gate source of M2, and the current of the driver transistor is supplied to the OLED:

2222 )(

2 TPOLEDIOLEDT VVVVVKI (11.1)

22 )(

2 POLED VVKI (11.2)

where K is the trans-conductance and VOLED is the shift in the OLED voltage.

From (11.2) we can find that the OLED current in independent on the threshold voltage VT2, and at the same time it is increasing as the VOLED increases over time. The OLED current is also dependent on the programming voltage, so the current range of the OLED is determined by taking a proper value for VP, thus this circuit is a voltage programming circuit. On the other hand the sensitivity of the IOLED to the

VOLED is defined as S, where:

108


(11.3)

(11.4)

From Equations (11.2) and (11.4)

(11.5)

From Equation (11.5) the value of the trans-conductance (K) could be calculated and then the aspect-ratio (W/L) of the driver transistor M2 could be obtained.

OLED

SEL3

SEL3

VDATA

VDD

SEL3

M5

M3

M2

M4

M1

CI

Figure 11.2 The voltage programming circuit

Table 11.1 shows the design parameters of the design. Figure 11.3 shows the timing diagram of the select signals and the programming voltage.

11.4 Simulation Results

The circuit was simulated using PSPICE to verify the operation of the pixel circuits. The simulation was based on the following models for the OTFT and OLED.

109


Table 11.1 The design parameters of the voltage programming circuit

Parameter Value

(W/L)1 (μm) 400/5

(W/L)2 (μm) 400/5

(W/L)3 (μm) 100/5

(W/L)4 (μm) 50/5

(W/L)5 (μm) 100/5

C1 (pF) 5

Vsel1 (V) -15 – 15

Vsel2 (V) -15 – 15

Vsel3 (V) -15 – 15

VOLEDI (V) 4

Figure 11.3 The timing diagram of the select signals and the input voltage

110


11.4.1 OTFT model

Since the OTFT is a special type of field effect transistor, then it can be modelled using a metal oxide semiconductor field effect transistor (MOSFET) model but by changing the values of some parameters in order to achieve the DC response and to have the IV curves of the OTFT. In this Chapter the model is taken from [2], this model is for p-type pentacene transistors. In previous literature, it was found that the field effect mobility of holes in pentacene is exceeding 1 cm2/V.s, but this model is based on hole mobility of 0.5 cm2/V.s for practical results [2]. The threshold voltage is –2.5 V, the maximum drift velocity is 5 x 104 m/s with an on/off ratio of 107. The dielectric thickness is 0.4 μm, the gate-source and gate-drain overlap capacitance are 124 pF and the mobility degradation factor is 0.07 V-1. Figure 11.5 shows the IV characteristics of the OTFT.

11.4.2 OLED model

Figure 11.4 shows the OLED model used in this Chapter. The model consists of a series resistance and a diode parallel with a capacitor. The capacitor models the total capacitance of the layers, the series resistance models the total resistance of the device and the diode models the rectifying nature of the OLED, the model is based on the model used in [6].

11.4.3 The results of the voltage programming circuit

A set of conditions were used for testing the voltage programming circuit. Figure 11.6 shows the time response of the OLED current with OTFT threshold voltages of –2.5, –1.5, and –3.5. As shown in the figures, there are slight changes in the OLED current which will not affect the brightness of the image. The voltage programming circuit also compensates for the mobility degradation as well as the current programming circuit - Figure 11.7 shows the OLED current with respect to the mobility degradation. Furthermore, due to the dependence of the OLED current on the change in the OLED on voltage, the circuit compensates for the OLED degradation. Figure 11.8 shows the OLED current with respect to the change in the OLED voltage. As shown in the figures the OLED current increases by increasing the shift in the OLED voltage which enhances the lifetime of the display.

111


R1

950

C15p

D1n=45.8

Figure 11.4 The OLED model

Figure 11.5 The IV curves of the OTFT

112


V T= -1.5 V V T= -2.5 V

V T= -3.5 V

Figure 11.6 The transient response of the OLED current with respect to the threshold voltage variation for voltage programming circuit at VT = -2.5V (middle

curve), VT = -1.5V (top curve), VT = -3.5V (bottom curve)

Figure 11.7 The transient response of the OLED current with respect to the mobility degradation for voltage programming circuit

113


I OLED after the shift in the V OLED

I OLED after a higher shift in the V OLED

Figure 11.8 The transient response of the OLED current after the shift in the OLED voltage. As shown in the figure, the bottom curve is after the shift in the

OLED voltage and the top curve is after a higher shift in the OLED voltage

11.5 Conclusion

In conclusion, this Chapter is about designing a fully organic pixel circuit for AM-OLED flat panel display. A fully organic means that the circuit consists of organic transistors (OTFT) and an OLED. The advantage of having a fully organic pixel circuit is the low cost of fabrication, easy to be fabricated on flexible substrates due to low temperature fabrication, thus flexible displays are adoptable and this is beside the high brightness, the great range of colors, the high contrast ratio and the wide viewing angle of the OLED displays.

However, the organic devices are still unstable and their performance degrades gradually with time. Therefore, a voltage programming pixel circuit was developed using OTFT in order to overcome the problems of organic devices and have a fully organic design. The design was studied and simulated using PSPICE. The design showed an impressive response with respect to the degradation of the organic devices, as it provides an electronic solution for these problems. The simulation proofs the capability of the design in compensating for the threshold voltage variation and mobility degradation of the OTFT. On the other hand, it compensates for the OLED luminance degradation and increase the life time of the display.

114


References

1. S.J. Ashtiani, G.C. Reza and N. Arokia, Journal of Display Technology, 2007, 3, 36.

2. M.R. Powell, Integrated Feedback Circuit for Organic LED Display Driver, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Boston, MA, USA, 2004. [MSc Thesis]

3. Y. He, H. Reiji and K. Jerzy, IEEE Transactions on Electron Devices, 2001, 48, 1322.

4. K.S. Lamba, An Integrated Circuit for Feedback Control and Compensation of an Organic LED Display, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Boston, MA, USA, 2006. [MSc Thesis]

5. C-L. Lin and K-C. Liao in Proceedings of the IEEE Conference on Electron Devices and Solid-State Circuits, IEEE, Taiwan, 2007, p.385.

6. W.F. Aerts, S. Verlaak and P. Heremans, IEEE Transactions on Electron Devices, 2002, 49, 2124.

7. H. Aziz, Z.D. Popovic, N-X. Hu, A-M. Hor and G. Xu, Science, 1999, 283, 1900.

8. G.R. Chaji, C. Ng, A. Nathan, A. Werner, J, Birnstock, O. Schneider and J. Blochwitz-Nimroth, IEEE Electron Device Letters, 2007, 28, 1108.

9. A. Lin, A Silicon Current Sensing Amplifier and Organic Imager for an Optical Feedback OLED Display, Massachusetts Institute of Technology, Boston, MA, USA, 2006. [MSc Thesis]

10. P. Melpignano, A. Baron-Toaldo, V. Biondo, S.Priante, R. Zamboni, M. Murgia, S. Caria, L. Gregoratti, A. Barinov and M. Kiskinova, Research Highlights, 95-98.

11. M. Mizukami, N. Hiorhata, T.Iseki, K. Ohtawara, T. Tada, S. Yagyu, T. Abe, T. Suzuki, Y. Fujisaki, Y, Inoue, S. Tokito and T. Kurita, IEEE Electron Device Letters, 2006, 27, 249.

12. Ch. Pannemann, T. Diekmann and U. Hilleringmann in Proceedings of an IEEE Conference on Microelectronics – ICM, Tunisia, 2004, p.76.

115


13. A. Shin, S.J. Hwang, S.W. Yu and M.Y. Sung, Journal of Computers, 2008, 3, 3, 1.

14. S.M. Sze, Semiconductor Devices: Physics and Technology, 2nd Edition, Wiley, New York, NY, USA, 2001.

116


117

12 Intelligent Packaging for the Food Industry

Salvatore Parisi

12.1 Introduction

So-called ‘intelligent packaging’, often confused with the better ‘active packaging’, is singularly differentiated from the other fields of the manufacturing industry. The main reason for this is the extraordinary growth of the related technology in recent times. Consequently, there is the concrete possibility of high profits in the future depending on the predictable request for ‘intelligent’ devices. First of all, some distinction between the active and intelligent devices has to be made.

Both instruments, generically called ‘smart packaging’, are different from conventional packages [1]. In detail, the requested function is the capacity to interact with the product if the device is associated with the packaging. The particular type of interaction - active mode or passive detection - is the real difference between the two solutions. This interaction is dependent on the particular exigency of the asking subject with concern to pharmaceutical companies and food and beverage industries [1]. The aim of this Chapter is to show the explicit and unknown necessities of the food and beverage industry.

Basically, the active packaging is able to interact with the inner atmosphere of the package. The declared aim is the modification of packaged products and related headspaces by means of antimicrobial and/or antioxidant substances. This alteration tends to improve the shelf life and other important features (colour, smell, texture).

On the other hand, ‘intelligent packaging’ is designed to collect, store and communicate particular information about the story of packaged products [1]. In other words, some kind of external or internal instrument is connected to the package and becomes able to:

1) Detect storage conditions (temperature, relative humidity) in function of the time progression, or,

118


2) Make evident the quality decay of the packaged food, or,

3) Detect storage conditions and quality decrement at the same time, and

4) Store all the relevant information and/or modifications.

As a consequence, the previous situation may be expressed as the interaction between the ‘Food and Package’ system [2] and the aware Consumer [3]. In effect, every food company should be able to collect, satisfy and anticipate the exigencies of Consumers, as requested by British Retail Consortium (BRC) and International Food Standard (IFS) protocols. These requests are often in accord with existing Regulations with concern to food safety. However, the necessity to foresee the needs of normal people and sensible consumers is highly recommended because of the risk of expensive and intricate class actions.

From a general viewpoint, the current ‘intelligent’ solutions can be subdivided in two categories with concern to food and beverage applications.

The first group is related to all the devices – radio frequency identification (RFID) tags above all - that are able to record and store important variations of quality and functional features. In other words, the monitoring activity is linked to the durability of food commodities and the good storage during all the shelf life. In general terms, the storage temperature is one of the important parameters that should be monitored. Other variables are the partial pressure of inner oxygen, the relative humidity, and so on.

The second category of ‘intelligent’ tools are those used to highlight imperceptible anomalies concerning the storage. In effect, 90% or more of food complaints are linked to the incorrect storage. So, the monitoring activity should be dedicated to the measure of storage conditions and Remaining Shelf Life (RSL). This problem is surely evident when Japanese or South-Korean Retailers import notable quantities of perishable foods under strict storage conditions (temperature: < –20 °C) because of the known degradability [4]. It should to be noted that these indicators are easily understandable by Retailers and normal Consumers without specific training.

The second category of ‘intelligent’ devices can be easily divided in to four subclasses with concern to the particular use [5]:

a) Temperature indicators

b) Time-temperature indicators (TTI)

c) Leakage indicators

d) Freshness indicators

119

Intelligent Packaging for the Food Industry

These instruments should be:

1) Easily activated

2) Able to exhibit measurable and reproducible variations that are linked to the remaining quality of foods

3) Absolutely irreversible

In recent times, the world of smart packaging seems to have changed, in comparison with traditional packaging, because of the enormous contributions of RFID and other PC-based technologies [1]. Actually, this impetuous evolution is mainly related to the sub-field of intelligent applications. In effect, new types of microchips have opened up new possibilities for commercial electronics and wireless technologies. New approaches have also been used to upgrade or explore other applications for non-food applications. At the same time, a quantity of possible connections between IT and other aspects of the social life has been tempted with different results. One of these attempts has been the development of complicated ‘multi-agent’ networks based on a number of programs with very low processing capacity. However, the most promising opportunities are in the food and beverage field, according to Food Retailers [1]. The declared aim is the reduction of ‘wastage’ by means of the improvement of supply chains. With regard to logistics, all foodstuffs may be distributed even more quickly and cost effectively [1]. As a consequence, intelligent packaging sales are predicted to grow four-fold to reach more than $80 million by 2010 [1] because of the growing power of retail chains in emerging markets (Asia, Latin America). However, the incumbent economic crisis and the consequent doubts about the fast expansion of supermarket chains in emerging countries like China do not allow confirmation of these predictions, particularly with concern to RFID tags.

For these reasons, a sort of premise about the present situation of Intelligent Packaging should be necessary.

12.2 RFID and Packaging Connection - The Age of Connectivity

Intelligent monitoring is able to give some information about the presumptive conditions of packaged foods until the expiration date. Clearly, the most critical control point (or CCP, in Hazard Analysis and Critical Control Points language (HACCP)) is represented by refrigerated transports. Thus, the so-called TTI have been created and continually updated. However, several observers have highlighted that improvements in supply chain management and innovations in food processing may reduce or eliminate the necessity of smart packages [1]. This risk has been

120


denied after new developments in RFID technologies; moreover, food retailers tend to encourage suppliers to boost the efficiency of their stores [1]. In effect, the supply chain should be seen as a series of intermediate steps from manufacturing to final distribution and catering services. These steps are generally continuous and carried out by different operators, but the constant factor is the identity of the product (lot, batch, expiry date, sell-by-date). Probably, this level of identification is difficult to enforce on a massive scale.

Anyway, the key factor for the implementation of intelligent packages and RFID tags is still cost. This assertion was the best argument for use of RFID detractors until recently. Nowadays, the global lack of liquid assets imposes shared strategies for the retailers and producers. Otherwise, RFID tags may have serious difficulties in the food market because of their relatively high cost. Probably, new ‘multi-factor tags’ (a combination of chemicals, inks and coatings with RFID) may help to obtain the desired advantages at low cost [1]. With regards to existing devices, the implementation of new IT systems should correspond to 20-30% of the total cost [1], while normal tags do not exceed 40 Eurocents (minimum: 5 Eurocents). On the other hand, the impetuous growth of RFID tags is not comparable with other Smart applications.

From a technical viewpoint, the RFID is based on three components compressed in an 1 cm2 adhesive tag:

1) A transponder;

2) An ID code which contains thousands of bits of potential information;

3) A mini-antenna.

In detail, the transponder is a sort of radio transmitter-receiver activated for transmission by reception of a predetermined signal.

With regard to shapes and dimensions, RFID tags may be produced as adhesive labels, bottoms or micro-ampules. The final target is the food or pharmaceutical product, but some in vivo applications have been realised. Anyway, the difference between different RFID applications is related to the ‘active’ or ‘passive’ role of tags with regard to the hardware and software tracing system. In effect, ‘active’ tags are able to detect and record temperature, relative humidity and abnormal shocks [6]. Successively, these data are transmitted automatically because of the presence of low batteries or solar micro-cells [7]. In addition, the reader/receiver distance can be extended to 30 metres (Figure 12.1). These features explain the high costs ($100 per tag).

121


On the other hand, ‘passive’ - or battery assisted passive (BAP) - RFID tags are provided with very low cost batteries (cost: $5) or completely inactive (up to 40 Eurocents). The difference is the impossibility of activating the reading/receiving system (distance: up to 10 cm; frequency: 1 Hz, resolution: 100 μm). This apparent limitation is the key of the success of RFID passive tags (Figure 12.2). Other advantages are:

1) Virtual absence of problems about reading angles.

2) Easy reading during movement.

3) Unlimited durability, except for massive crashes.

4) Possibility of transmission on 865 and 868 MHz (EU, US, etc.).

5) Practical codification, reading and writing standards (Electronic Product Code (EPC), EU).

6) Possibility of tracing the position and the whole history of food pallets (for ‘n’ units with 96 bit-RFID tags, the whole pallet represents ‘n 96’ bits of information).

Other important applications are related to the problem of damage during transport. It is known that deliveries are critical not only for the storage temperatures but also

Figure 12.1 Active RFID tags contain small batteries (or solar microcells). As a consequence, they are able to ‘alert’ receiving and tracing systems

(distance up to 30 m)

122


for possible mechanical shocks [6]. The total cost of claims may be up to 10% of delivered commodities. So, some operators would like to employ RFID tags with integrated accelerometers to monitor the shipment step.

With concern to the management of warehouses the so-called ‘First In, First Out’ strategy (FIFO) may be replaced with more practical procedures that are based on RSL [8]. One of these methods has been already proposed some year ago [7] with the Products Released on Properties Obviously Selected for Additional Life (PROPOSAL).

The previously mentioned advantages are not ‘the end of the story’ about RFID tags. In effect, there are two points that should be considered as inconveniences. Firstly, food companies are reluctant to employ this technology because of the possible substitution of RFID software with newer products. So, RFID technology should be able to manage the information by means of a single writing/receiving protocol that may be recognised by existing and newer software agents. In addition, this exigency is strictly linked to the possibility of reading errors and/or data cancellations.

The second problem is related to possible attacks to RFID tags by computer hackers. So, it is necessary that the electronic devices are provided with protection. This

Figure 12.2 Passive RFID tags are not able to activate the reading/receiving system (distance: up to 10 cm; frequency: 1 Hz, Resolution; 100 μm). On the other hand,

cost are very low

123


argument is really important with concern to traceability requirements and BRC/IFS standards about the preservation of food identity (GMO-free or melanin-free products, etc.). It should be noted that every computer hacker might attack RFID tags in wireless mode by means of normal mobile networks. This risk can be very high in case of regional or local RFID networks, managed by private companies. The age of total connectivity allows this type of actions, thus the problem has to be resolved from the IT point of view.

This section should not be concluded without some word about new intelligent tracking solutions, also called nanobarcodes and deoxyribonucleic acid (DNA) codes. The first technology [9] is related to the possibility of producing particular barcodes on a nano-scale (100 to 1,000 nm). These codes can be easily read by means of spectrophotometers and normal scanners. DNA codes, on the other hand, are simply constituted of deoxyribonucleic acid fragments that may be inserted into labels, traditional packages and glass materials. This system allows storage of unlimited quantities of information because of the presence of a key code inserted into the DNA sequence. Clearly, the alteration of similar codes is not workable with simple tools. So, the only possibility of sabotage is the physical destruction - by abrasion - of the molecular sequence.

12.3 TTI and Other Solutions - The Viewpoint of Food Industries

The second category of intelligent devices is approximately subdivided in four sub-classes with concern to the particular information. In detail, the group is often confused with the so-called time-temperature indicators or TTI. In recent times, researchers have studied possible applications and real benefits of TTI in terms of food safety and hygiene [10]. At the same time, the detractors of intelligent devices have always preferred to attack the whole category targeting TTI.

First of all, the TTI acronym has not to be attributed to simple temperature indicators. These instruments are exclusively able to record excessive values of storage temperatures in an irreversible way [3]. The TTI evolution is weightier because of the possibility of correlating thermal values at the period n which the variations have been recorded in. The visual result can be recognised as the average sum of all thermal variations during the whole shelf life, or RSL. In other words, TTI record storage temperatures and are able to add cumulatively their fluctuations by integration. It has to be noted that thermal conditions are related to the Food and Packaging integrated system. As a consequence, all conclusions about RSL are reliable assuming that food degradation is comparable to TTI reactions [11].

In general, RSL is displayed in a colorimetric way. The chromatic alteration is always associated to the evolution of the general quality of products, assuming that this is a

124


synergetic sum of n factors [12]. In effect, the main argument against the use of TTI is the reliable correlation between three factors:

1) The chromatic decay of TTI devices.

2) The most degrading reaction - or chain of reactions - in foods.

3) The storage temperature at the heart of products.

From a technical viewpoint, there are three main types of TTI:

1) ‘Bull’s Eye’ solutions (Figure 12.3). The label shows a circular zone filled with diacetylenic monomers that are ready to initiate long polymerisation chains. The process, which is temperature-dependent, allows the chromatic variation of coloured circles to darker tints [10]. This variation is appreciable with or without optical scanners and may be correlated to different sensitivities. In addition, the activation has to be started exactly when foods are packaged. So, Bull’s Eye indicators are simply frozen until the final use. Normally, the main targets are fresh and frozen products.

2) Chromatographic solutions (Figure 12.4). These instruments tend to reproduce qualitative columns of chemical analysis. The aim is to induce the progression of coloured solutions - fat substances - throughout transparent tubes or columns. The progression can be intended as time and temperature dependent. However, food operators have some doubt about the correlation between the viscosity of fluid substances and the behaviour of food fats.

3) Enzymic indicators (Figure 12.5). The synthetic mixture is composed of a lipidic substrate, a pH indicator and a specific lipase [5] that catalyses the hydrolysis of fats into glycerol and fatty acids. The colour variation is extremely simple, reliable and immediately interpretable because of different coloured windows.

Nowadays, the main advantages of TTI are related to:

1) The easy use and reading by normal consumers.

2) The possibility of just-in-time (JIT) and shortest remaining shelf life (SRSL) strategies instead of FIFO.

3) The objective compliance with Codex Alimentarius, Step 9, Principle 4 (every CCP has to be continually inspected) and related BRC and IFS Protocols.

4) The observed trend in favour of minimally processed and assembled foods.

125


Figure 3 Bull’s Eye indicators show the presumptive degradation in a colorimetric way. The darkening of the circular zone is appreciable with and

without optical scanners

Figure 12.4 ‘Chromatographic’ indicators show the progression of coloured substances throughout transparent tubes. This shift can be correlated with the

decrease of remaining shelf life

126


The last point is very evident in the EU where active packages, in direct contact with food atmospheres, are considered extraneous or incompatible with fresh foods [1]. In contrast, Japanese consumers tend to prefer active packaging while RFID tags sales are not comparable.

On the other hand, there are three objections against the use of TTI. First of all, every monitoring instrument should be adapted to inform the consumer about the intrinsic quality of packaged foods. From the hygienic viewpoint, this quality is related to the reduction or the complete elimination of microbiological, chemical and physical risks, according to Codex Alimentarius principles. However, it is known that the so-called risk-analysis may give different conclusions, depending on the particular product and the process [11]. Consequently, TTI should be realised on the basis of much information. The associated hazard is the design of standardised indicators with very different results, product by product. This situation is clearly an important inconvenience because different risks have different weights.

The second argument is emerging by the behaviour of some food companies. Actually, there is the jeopardised tendency to reduce shelf life values to the shortest term in relation to perishable foods and dairy products. Manufacturers prefer this strategy because of the consumeristic approach. Consequently, TTI sales could diminish in the future [1]. However, the role of these instruments from the hygienic point of view should be considered. In detail, the safety of foods is strictly dependent on the correct storage [13]. In addition, all edible products are subjected to some type of alteration according to the 1st Law of Food Degradation [11]. So, every official verdict or

Figure 12.5 Enzymic solutions contain a synthetic mixture. Lipids are demolished by a specific lipase. The acidification is displayed by normal pH indicators

127


analysis about unacceptable products (clear degradation, microbial spreading, etc.) should be considered on the basis of the related thermal history.

Other intelligent indicators are designed to perceive the presence of undesired oxygen and/or carbon dioxide into modified atmosphere packaging (MAP) packages [14]. These devices are named leakage indicators [3]. The basic idea is the insertion of Redox indicators (methylene blue, and so on) or acid/ base systems and the consequent revelation of active components. It has to be noted that the most important failure of these devices is strictly related to the microbial spoilage and consequent variations in the oxygen/carbon dioxide ratio. This interference has to be expected before considering eventual failures in the MAP process. In addition, the correct storage of these indicators is generally fixed at 5-8 °C in order to prevent anticipated reactions [5].

Finally, a particular mention should be done about ‘freshness’ indicators. Actually, these devices are able to detect the production and consequent accumulation of gaseous substances by microbial spoilage. These detectable targets are hydrogen sulfide (by reaction on myoglobin saturated substrates), ammonia, tri-methylamine N-oxide [15], and acetic acid [3]. In relation to consumeristic approaches, these systems are considered extraneous or incompatible with fresh foods, like active devices. Consequently, related sales are not good in the EU in respect of other solutions [1, 3].

12.4 New Proposals - The Consumer’s Viewpoint

The previous paragraphs have displayed the current offerings for intelligent packaging. Actually, these offers are well known, however food companies would like to make use of slightly different devices. Basically, the final consumer and the manufacturer cannot be well qualified about smart or intelligent packaging. On the other hand, the opinion of consumers (and food companies) is strictly necessary in order to design and implement new solutions. The explicit confirmation comes from BRC Global Standard Food, Version 5, point 3.4, about the necessity to determine requests and expectations of customers.

In a general way, recent market research tends to highlight the opinion that large groups of normal people (age: 30-65) are not well disposed towards ‘assembled’ foods. The European situation has been discussed with concern over active packages. A particular case study has been verified in Italy some year ago. After the discovery of a food scandal, the main networks have proposed radio and TV commercials related to different brands or food types. These messages were to assure the consumer about the substantial difference between farm-made and ‘assembled’ foods.

128


On this basis, food technologists can surely help with the development of new intelligent devices that may be largely appreciated. These replica indicators (Figure 12.6) may be realised with different substances that should mime the composition of packaged foods (fermentable sugars, and so on). In detail, a hypothetic multifunctional indicator may contain an edible substrate composed of:

a) Food-grade rennet casein and agar agar (as blocking agent).

b) Simple sugars, lactose or biodegradable polysaccharides (starch and methyl-cellulose), able to turn brown because of simple enzymic reactions [16-19].

c) Unsaturated triglycerides or fat acids (oleic and ricinoleic acids); food-grade oils (soya, sunflower).

d) Carotenoids and/or riboflavine, very susceptible to sunlight [3, 20, 21].

e) Gelatine (or similar colloids) as a co-indicator of degradation because of its hydrophilicity [22].

Similar mixtures have already been devised for different purposes [23]. The following features have to be highlighted:

Figure 12.6 The visible degradation of ‘Replica’ tags ‘mimes’ exclusively the behaviour of packaged foods, according to the principle: ‘One food, one replica;

two foods, two replicas’

129


1) Visible degradation of a biodegradable matrix that is able to ‘mime’ exclusively the behaviour of packaged foods, according to the principle: ‘One food, one replica; two foods, two replicas’.

2) Predetermined composition of biodegradable mixtures.

3) Fixed composition of tag atmospheres.

4) Possibility of acid-base or Redox indicators and specific enzymes (lipase, amylase). In particular, Redox indicators can show immediately dangers in terms of favourable conditions for the spoilage of pathogen bacteria (Escherichia coli, Listeria monocytogenes), if they are present in the real food [14].

5) Reliability of results, on the basis of correlations between the decay of packaged foods and the visible degradation of like-food mixtures.

In relation to the necessity of different indicators for different products, it should be highlighted that every food has different pH, Redox potential (Eh) and water activity (Aw). These parameters depend on the different composition [11] and other factors (provenence, processing, presence of ‘wild’ cultures, etc.). Consequently, pathogenic contaminations are possible in particular conditions only [14, 24, 25]. So, it is strictly necessary to develop a good replica for every food type.

New approaches may be realised considering the biodegradable and ‘replicative’ nature of these indicators. An interesting development could be the insertion of inoffensive bacteria with the final goal to produce evident and reproducible reactions. These bacteria could be constituted of Lactobacillus casei, Lactobacillus bifidus or Lactobacillus bulgaricus cultures, because of the large applications in yoghurts. The previously-mentioned agents are capable of proteolytic and fermentative reactions producing acid substances (optimal temperature: 40 °C, with detectable activity at 10 °C). So, the related spoilage would be easily displayed by means of normal acid-base indicators and should demonstrate the real situation of a substrate that is similar to the packaged food.

For the same reason, the insertion of well known food yeasts and moulds is possible. Two examples are:

a) Penicillium roquefortii (inserted in blue and green cheeses). This mould is able to spread at 12-13 °C (optimal temperature: 25-30 °C) in the absence of light and produces grey mycelia [26].

b) Brevibacterium linens (present in ‘wet crust’ cheeses). This yeast is well known because of its ability to produce red pigments.

130


However, this solution could not be appreciated by normal consumers because of the hypothetical microbial dissemination – by breakage - in the environment. Naturally, the consumer should be informed about the absence of real risks associated to Lactobacillaceae or moulds, but the acceptance of a similar device is not sure. Existing regulations are not ready to accept the authorised release of ‘living’ indicators. In addition, there are two risks to be considered: (1) the eventuality of mutant bacteria and (2) the release of powerful bacteriophages. For these reasons, ‘living’ indicators may be proposed as monitoring instruments for manufacturers and official inspectors only.

12.5 Food Companies Under Attack - Sabotage and Class Actions

Another difficulty is represented by possible hostile actions against food companies and retailers. Basically, the related risk is double because all intelligent systems can be destroyed, ripped out, damaged and/or reprogrammed (RFID tags). As a consequence, every legal action against justified Class Actions (degradated foods and poor storage) may be very difficult.

RFID tags are vulnerable to PC attacks and mechanical damage. For these reasons, the matter has to be resolved on the IT ground. Actually, the thorny subject is the defence of food companies against the possible accusation of selling food and beverages without related control tags. Because of the difficulty of producing films (cans, bottles, and so on) with integrated intelligent devices, the best solution should be the creation of removable ‘multi-coupon’ tags. The basic idea is the possibility to trace out the thermal history of food commodities, step-by-step, allowing Warehouse managers to pull out (and refrigerate) one of the n tags of a packaged product or pallet. Thus, three objectives can be reached:

1) Evidence and record of the quality of packaged products (pallets), step-by-step.

2) Storage of traceability data, pallet by pallet, in case of RFID tags.

3) Evidence of the ‘tracking’ coupon, product by product, in case intelligent tags should be partially or totally destroyed.

In addition, it should be noted that the subdivision of intelligent tags in n independent coupons allows the option to store the totality of remaining devices near the last warehouse/market, lot by lot.

12.6 Conclusions

The aim of this Chapter has been to demonstrate that intelligent packaging can be very useful for all the components of food chains. Producers, intermediate distributors, retailers

131


have sufficient reasons to find a profitable or practical use for the previously mentioned systems. However, supply chains must work in strict association with the packaging industry to work out shared solutions. In effect, there are several ‘implicit’ proposals that must be discussed and satisfied. Some of these suggestions have been described here without too technical specifications. According to food producers, the realisation of these or similar ideas may help the growth of the whole manufacturing sector.

References

1. Flexible, 2005, 4, 4, 24.

2. S. Parisi, Alterazioni in Imballaggi Metallici Termicamente Processati, Gulotta Press, Palermo, Italy, 2004, p.9.

3. L. Piergiovanni, Dispensa di ‘Tecnologia del Condizionamento dei Prodotti Agro-Alimentari’, Edizione 2002, Lezioni di Food Packaging, Agosto, State University of Milan, Milan, Italy, 2002, p.1-5 and 174-179. http://www.distam.unimi.it/~packlab/download.htm

4. M. Shinkman and P. Lewis, Rich Pickings, Opportunities in South-east Asia’s Emerging Markets, Atradius Credit Insurance NV and The Economist Intelligence Unit, The Economist, NY, USA, 2008, p.15. http://global.atradius.com/images/stories/EmergMarkRep%20SEAsia_int.pdf

5. N. de Kruijf, M. van Beest, R. Rijk, T. Sipiläinen-Malm, P. Paseiro Losada and B. De Meuleunaer, Food Additives and Contaminants, 2002, 19, Supplement, 144.

6. B. Jarimopas, S.P. Singh and S. Wasan, Packaging Technology and Science, 2005, 18, 179.

7. K.S. Marsh, Italian Journal of Food Science, 2007, Special Issue, p.4.

8. I LearnShop, Missione Logistica, Food Packages - Quick Guide, Artek, Varese, Italy, 2006, 9, 3.

9. Flexible, 2005, 4, 2, 26.

10. M. Riva, L. Piergiovanni and A. Schiraldi, Packaging Technology and Science, 2001, 14, 1.

11. S. Parisi, Industrie Alimentari, 2002, 417, 905.

12. S. Parisi, Italian Journal of Food Science, 2004, Special issue, p.11.

132


13. S. Parisi, S. Delia and P. Laganà, Industrie Alimentari, 2004, 438, 735.

14. S. Delia, P. Laganà and S. Parisi in Proceedings of the XIV Conferenza Nazionale Microbiologia degli Alimenti Conservati in Stato di Refrigerazione, Bologna, Italy, 2005, p.93.

15. A.R. de Jong, H. Boumans, T. Slaghek, J. van Veen, R. Rijk and M. van Zandvoort, Food Additives and Contaminants, 2005, 22, 10, 975.

16. M.T. Bilancia, F. Caponio, E. Sikorska, A. Paqualone, T. Gomez and C. Summo, Italian Journal of Food Science, 2007, Special Issue, p.400.

17. W.D. van Dongen and J. Jetten, Italian Journal of Food Science, 2007, Special Issue, p.288.

18. G. Muratore, M.A. Del Nobile, L. Bongiovanni, G.G. Buonocore, C.M. Lanza and C. Nicolosi Asmundo, Italian Journal of Food Science, 2004, Special Issue, p.199.

19. P. Pittia, G. Sacchetti and D. Mastrocola, Italian Journal of Food Science, 2004, Special Issue, 315-326.

20. P. Baardseth, A. Veberg, O. Tomic, J.P. Wold, M. Augestadm and T. Eie, Italian Journal of Food Science, 2007, Special Issue, p.260.

21. L. Torri, L. Piergiovanni and S. Limbo, Italian Journal of Food Science, 2007, Special Issue, p.62.

22. H. Dommershuijzen, L. Hviid, H. den Hartog and J. Vereijken, Packaging Technology and Science, 2005, 18, 303.

23. P.V.A. Bergo, R.A. Carvalho, P.J.A. Sobral, R.M.C. Dos Santos, F.B.R. Da Silva, J.M. Prison, J. Solorza-Feria and A.M.Q.B. Habitante, Italian Journal of Food Science, 2007, Special Issue, p.167.

24. G. Andreis and F. Ottaviani, Manuale delle Sicurezza Microbiologica degli Alimenti e delle Acque, Oxoid SpA., Italy, 2002, p.67.

25. D.A.A. Mossel, J.E.L. Corry, C.B. Struijk and R.M. Baird, Essentials of the Microbiology of Foods: a Textbook for Advanced Studies, John Wiley and Sons, Chichester, UK 1995, p.185.

26. V. Bottazzi, Microbiologia Lattiero-casearia, Edagricole, Bologna, Italy, 1993, p.129.

133

A bbreviations for Electronics

3D Three dimensional

ACN Acrylonitrile

AFM Atomic force microscope

AM Active matrix

APS Ammonium persulfate

a-Si:H Hydrogenated amorphous silicon

B&W Black and white

BAP Battery assisted passive

BRC British Retail Consortium

CCP Critical control point

CMOS Complementary metal-oxide semi-conductor

CNT Carbon nanotube(s)

CO2 Carbon dioxide

CT Computed tomography

DBSA Dodecyl benzenesulfonic acid

DC Direct current

DIN Deutsches Institut für Normung eV - the German Institute for Standardisation

DMSO Dimethyl sulfoxide

DNA Deoxyribonucleic acid

DSC Differential scanning calorimetry

EPC Electronic Product Code

EU European Union

FIFO ‘First In, First Out’ strategy

FT-IR Fourier transform infrared spectra/spectroscopy

GMO Genetically-modified organism

GO Gate overlap

134


HACPP Hazard Analysis and Critical Control Points language

HALS Hindered amine light stabilisers

HV High viscosity

ID Identification

IFS International Food Standard

IKV German Plastics Processing Institute

IT Information technology

ITO Indium tin oxide

IV I stands for current, V stands for voltage

JIT Just-in-time

KBr Potassium bromide

L/D Length to diameter ratio

LCD Liquid crystal display(s)

LED Light emitting diode

LV Low viscosity

MAP Modified atmosphere packaging

MEMS Micro Electro Mechanical Systems

MOSFET Metal oxide semiconductor field effect transistor

MWNT Multiwall carbon nanotubes

N2 Nitrogen

NaDBS Sodium dodecylbenzenesulfonate

NBR Poly(butadiene-co-acrylonitrile)

NV-RAM Non volatile random access memory

O2 Oxygen

OE Organic electronics

OE-A Organic Electronics Association

OFET Organic field-effect transistor

OLED Organic light-emitting diode(s)

OPV Organic photovoltaic cells

OTFT Organic thin film transistor(s)

PA6 Polyamide 6

PANI Polyaniline

PAni.DBSA Polyaniline dodecylbenzenesulfonate

PC Personal computer

135

Abbreviations for Electronics

PE Polyethylene

PEDOT Poly(3,4-ethylenediocythiophene)

PET Polyethylene terephalate

PET Polyethylene terephthalate

PFBT Pentafluorobenzenethiol

PMOS Positive channel metal oxide semiconductor

PPS Polyphenylene sulfide

PROPOSAL Properties Obviously Selected for Additional Life

PSPICE Circuit simulation tool

PSS Poly(styrenesulfonate)

PVA Polyvinyl alcohol

PVP Poly(4-vinylphenol)

R&D Research and Development

RC Resistance and capacitance

RF Radio-frequency

RFID Radio frequency identification

RH Relative humidity

RMS Root-mean-square

ROLLED Roll-to-roll manufacturing technology

rpm Revolutions per minute

RSL Remaining shelf life

S/D Source/drain

SDS Sodium dodecyl sulfate

Sn/Pb Tin/lead

SnAgCu Tin silver copper

SRSL Shortest remaining shelf life

SWNT Single wall carbon nanotube(s)

TEM Transmission electron microscopy

TFT Thin film transistors

Tg Glass transition temperature

TGA Thermogravimetric analysis

TIPS 6,13-Bis (triisopropylsilylethynyl)

TTI Time-temperature indicator(s)

UV Ultraviolet

136


UV-Vis Ultraviolet-visible spectroscopy

VDMA German Engineering Foundation

WLI White light interferometer

WORM Write once, read many

137

Index

AActive matrix addressing technique 98Active matrix organic light emitting diode 106

Flat panel display 105Active packages 117, 126Assembled foods 127Astable multivibrator 99Atomic force microscope images 14Automotive engineering 43

BBacteria 129

Lactobacillus bifidus 129Lactobacillus bulgaricus 129Lactobacillus casei 129

Ball milling process 25Beer-Lambert law 58, 63Blends preparation 77BRC global standard food 127Bull’s eye indicators 125

CCable jacketing 63Carbon nanotube

Characterisation 37Nanotube inks 40, 67, 92

Characterisation of nitrile-butadiene rubber - polyaniline - dodecylbenzenesulfonate blends 78

Chromatographic indicators 125Codex alimentarius 124, 126Communications engineering 43Complementary metal-oxide-semiconductor

Circuits 4 Transistors 9

Conductivity, four-probe 73

138


Conductor paths, injection moulding of 52Copper indium gallium selenide ink 88Critical control point 119, 124

DDifferential scanning calorimeter 78

Thermal analysis 81Direct printing 13

Gravure printing 13, 22-24Imprinting 13Ink jet printing 13, 31-34Laser assisted patterning 13Microcontact printing 13 Screen printing 13, 23, 25

Dye sensitisation 93

EEdge waviness 15Electrical conductivity

Determination of 81Four-probe technique 81

Electrical engineering 43Electrodes, printed 1, 18, 31Electromagnetic shielding 45, 51-52, 54Electronic applications, injection mouldable 43Electrospinning 31-32, 40-41Enzymic indicators 124Exfoliation of single wall carbon nanotubes 69

FFabric construction 92Fibre processing 92Sensors, finger-tip 96First in, first out strategy 122First Law of Food Degradation 126Food and package system 118Food industry 117, 123

Packaging 117Food moulds 129

Penicillium roquefortii 129Food yeasts 129

Brevibacterium linens 129Fourier Transform-infrared

Spectra 70, 74Spectra of polyaniline 71

139

Index

Spectra of polyaniline - surfactant 71Spectra of polyaniline - surfactant-single walled nanotubes 71Spectrometer 78, 80

HHazard analysis and critical control points language 119Hindered amine light stabilisers 58-59

IInfrared Excalibur instrument 70Ink jet printing techniques 31

Piezoelectric ink jet printing 31, 41Thermal ink jet printing 31Injection moulding 45, 49, 54Intelligent devices 117-118

Freshness indicators 118, 127Leakage indicators 118Temperature indicators 118Time-temperature indicators 118, 123-124 Bull’s eye solutions 124 Chromatographic solutions 124 Enzymic solutions 126

Intelligent monitoring 119Intelligent packaging 117, 127, 130

KKnitted fabrics laddering of 91Kruss Easy Drop 36

LLactobacillaceae 130

MMaterial warming 50Melapur 200 65-66Metal oxide semiconductor field effect transistor 110Microelectromechanical systems 95Modified atmosphere packaging packages 127Multivibrator 101

OOptical micrograph 40-41, 79Optical microscope 78Organic electronics-A roadmap 3, 5, 10Organic electronics 1-4, 10, 21

140


Applications 2Materials 4Technology 4, 95

Organic field effect transistor 8, 105Organic light-emitting diodes 21, 22, 105, 106, 108, 110-113

Characteristics of all-printed 26Printing 22

Organic thin film transistor 2, 7, 17-18, 95-96, 98-99, 100, 102, 105-108, 110-111, 113

PPolyaniline 67-70, 72-74

Absorbance spectra of 72Organic dispersions of 68

Polyaniline - dodecylbenzenesulfonate, synthesis of 77Polyaniline - single walled nanotubes, organic nanocomposite dispersions of 69Polyaniline - single walled nanotubes - surfactant, absorbance spectra of 72Polyaniline - surfactant, absorbance spectra of 72Patterning techniques 6Poly(3,4-ethylenediocythiophene) - poly(styrenesulfonate) inks, characterisation of 36Pendant drop method 36Phase inversion phenomenon 70Photovoltaic technology 88Piezoresistive materials 96Piezoresistive sensor 98Piezoresistivity 96Pixel circuits designs 106Polyaniline synthesis 68Polyethylene matrix 48Polyethylene terephthalate 90

Fabrics 91Polymer degradation 57, 58Polymer electronics 57, 63Polymer matrix 64, 67Polymer stabilisation 57, 58Polymers, degradation and stabilisation of 58Polymerisation 68, 92Positive channel metal oxide semiconductor devices 97Printing techniques 6, 7, 21Products Released on Properties Obviously Selected For Additional Life (PROPOSAL)

122PSPICE 105, 113

RRadio frequency identification tags 13, 95, 118-122, 126, 130Redox indicators 127, 129

141

Index

Robotic finger tip specifications 97Roll-to-roll printed organic light-emitting diodes demonstrators 27Roll-to-roll printing 21-22Rotary screen printing 23Rubber blends 77

SScanning electron microscope micrograph 41Screen printing of aluminium cathode 25Sencel 98Sensor array 98Shelf life, remaining 118Shelf life, shortest remaining 124Single wall carbon nanotubes 67-68Single wall nanotubes 69-70, 72-74Smart packaging 117Solar cell 88-90, 92-93Solar textile 87, 90, 93Solution mixing 77Stabilisers for polymer electronics 60Surface roughness of printed electrodes 14

TTextile construction 90

Fibres 90Fabrics 91

Thermal gravity analysis 61Thin film transistor applications, all-printed 13Thin film transistor displays 95Touch sensing 95Transmission electron microscopy 78, 79

UUV absorbers 61-63UV-vis spectroscopy 70, 72-74

VVan der Pauw equation 81Van der Pauw technique, four-probe 78Voltage programming circuit 109-110, 112Voltage programming pixel circuit 107

142


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

The worldwide market for polymer electronic products has been estimated to be worth

up to £15 billion by 2015 and the opportunity for new markets could be as high as £125

billion by 2025. The rapid development of polymer electronics has revealed the possibility

for transforming the electronics market by offering lighter, flexible and more cost effective

alternatives to conventional materials and products. With applications ranging from printed,

flexible conductors and novel semiconductor components to intelligent labels, large area

displays and solar panels, products that were previously unimaginable are now beginning

to be commercialised.

Smithers Rapra is pleased to announce this new book on Polymer Electronics, designed

to inform researchers, material suppliers, component fabricators and electronics

manufacturers of the latest research and developments in this dynamic and rapidly

evolving field.

With a prestigious group of authors writing on a wide range of polymer electronics and

related technology topics, this book brings together the technologies and applications at

the heart of polymer electronics.

This book details current and future developments in the fast-moving polymer electronics

industry.

This book will be of interest to those involved in the following sectors:

photovoltaics and solar energy, displays, Intelligent packaging, smart labels, medical

electronics, electronic manufacturing, defence and aerospace, and sensors and transducers.

Published by Smithers Rapra, 2009ISBN: 978-1-84735-423-5

Documents

Polymer Electronics