Handbook of Polymers in Electronics

Handbook of Polymers in Electronics

Bansi D. Malhotra

Handbook of Polymers in Electronics

Bansi D. Malhotra

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net

First Published in 2002 by

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

©2002, Rapra Technology Limited

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

ISBN: 1-85957-286-3

Typeset by Rapra Technology Limited Printed and bound by Polestar Scientifica, Exeter, UK

Contents

1

Charge Transport in Conjugated Polymers ...................................................... 3 1.1

Introduction ........................................................................................... 3

1.2

The Electronic Ground State .................................................................. 4

1.3

Charge Transport Carriers ..................................................................... 6 1.3.1

Soliton ....................................................................................... 7

1.3.2

Polaron and Bipolaron ............................................................... 9

1.4

Transport Properties of Polymers ......................................................... 12

1.5

Factors Influencing the Transport Properties of Polymers .................... 14 1.5.1

Disorder ................................................................................... 14

1.5.2

Doping ..................................................................................... 15

1.5.3

Interchain Coupling ................................................................. 17

1.6

Models of Charge Transport in Conducting Polymers ......................... 18

1.7

Conclusions ......................................................................................... 27

Acknowledgements ....................................................................................... 28 References ..................................................................................................... 28 2

Electrical Properties of Doped Conjugated Polymers .................................... 37 2.1

Introduction ......................................................................................... 37

2.2

Metallic State ....................................................................................... 39

2.3

2.2.1

Conductivity ............................................................................ 40

2.2.2

Magnetoconductance ............................................................... 50

2.2.3

Thermoelectric Power .............................................................. 55

2.2.4

Magnetic Susceptibility and Specific Heat................................ 56

Critical and Insulating States ............................................................... 58 i

Polymers in Electronics

2.4

Summary .............................................................................................. 63

References ..................................................................................................... 65 3

Non Linear Optical Properties of Polymers for Electronics ........................... 69 3.1

Introduction ......................................................................................... 69

3.2

NLO Polymer Issues for Device Applications ...................................... 70

3.3

Properties of Third-Order NLO Polymers ........................................... 71 3.3.1

Background of Third-Order NLO Polymer Research .............. 71

3.3.2

Poly(arylenevinylene), PAV ...................................................... 72

3.3.3

n-BCMU-PDA ......................................................................... 75

3.3.4

PT ............................................................................................ 76

3.3.5

Processible π-Conjugated Polymers .......................................... 76

3.3.6

Third-Order NLO Polymer Waveguides .................................. 82

3.4. Properties of Second-Order NLO Polymers ......................................... 84 3.4.1

Azo-Dye-Functionalised, Poled Polymers for Second-Order Non Linear Optics ................................................................... 84

3.4.2

EO Polymers ............................................................................ 87

3.4.3

Serially-Grafted Polymer Waveguides ...................................... 88

3.4.4

Refractive Index Grating Fabrication into Azo-Dye- ................... Functionalised Polymer Waveguides ........................................ 90

3.5

Future Targets of NLO Polymers for Optical Device Applications ...... 93

3.6

Conclusions ......................................................................................... 94

Acknowledgements ....................................................................................... 94 References ..................................................................................................... 94 4

ii

Luminescence Studies of Polymers ................................................................ 99 4.1

Introduction ......................................................................................... 99

4.2

Basic Photophysical Deactivation Processes ....................................... 100 4.2.1

Luminescence ......................................................................... 101

4.2.2

Bimolecular Photophysical Processes ..................................... 103

Contents

4.2.3 4.3

4.4

4.5

Quenching Processes .............................................................. 104

Methods for Fluorescence Studies ...................................................... 105 4.3.1

Time-Correlated Single-Photon Counting Studies .................. 105

4.3.2

Quantum Yields ..................................................................... 105

Fluorescence of Polymers, Excimer Fluorescence ............................... 106 4.4.1

Fluorescence of Polymers in Solution ..................................... 110

4.4.2

Fluorescence of Polymers in Gel State .................................... 122

Conclusions ....................................................................................... 136

Acknowledgements ..................................................................................... 136 References ................................................................................................... 136 5

Polymers for Light Emitting Diodes ............................................................ 141 5.1

Introduction ....................................................................................... 141

5.2

The Physics of Electroluminescent Devices ........................................ 142

5.3

5.4

5.5

5.2.1

The Physics of Conjugated Polymers ..................................... 142

5.2.2

The Physics of the Device....................................................... 144

5.2.3

LED Characterisation ............................................................ 147

Polymeric Structures for LED ............................................................ 148 5.3.1

Polyphenylenes ...................................................................... 148

5.3.2

Polythiophenes ....................................................................... 164

Recent Developments ......................................................................... 168 5.4.1

Polarised Electroluminescence ............................................... 168

5.4.2

Lifetime and Degradation in LEDs ........................................ 170

5.4.3

Microcavities ......................................................................... 170

Concluding Remarks ......................................................................... 171

References ................................................................................................... 172 6

Photopolymers and Photoresists for Electronics .......................................... 185 6.1

Introduction ....................................................................................... 185

iii

Polymers in Electronics

6.2

6.3

6.4

6.5

Microlithography Process .................................................................. 187 6.2.1

Resist Coating ........................................................................ 188

6.2.2

Exposure ................................................................................ 188

6.2.3

Development .......................................................................... 189

6.2.4

Post Baking ............................................................................ 190

6.2.5

Etching .................................................................................. 190

6.2.6

Resist Removal (Stripping) .................................................... 190

6.2.7

Doping ................................................................................... 190

Resist Requirements ........................................................................... 190 6.3.1

Solubility ............................................................................... 191

6.3.2

Adhesion ................................................................................ 191

6.3.3

Etching Resistance ................................................................. 191

6.3.4

Sensitivity and Contrast ......................................................... 191

Resist Materials ................................................................................. 193 6.4.1

Conventional Photoresists ..................................................... 193

6.4.2

Deep-UV Photoresists ............................................................ 197

6.4.3

Electron-Beam Resists ............................................................ 202

6.4.4

X-Ray Resists ........................................................................ 206

6.4.5

Special Resists ........................................................................ 208

Conclusions ....................................................................................... 212

References ................................................................................................... 212 7

Polymer Batteries for Electronics ................................................................. 217 7.1

Introduction ....................................................................................... 217

7.2

Ionically Conducting Polymers .......................................................... 218

7.3

7.2.1

Lithium Polymer Electrolytes and Lithium Batteries .............. 218

7.2.2

Proton Polymer Electrolytes ................................................... 239

Electronically Conducting Polymers .................................................. 242 7.3.1

iv

Lithium-Doped Conducting Polymer and LithiumPolymer Batteries ................................................................... 243

Contents

Acknowledgements ..................................................................................... 245 References ................................................................................................... 245 8

Polymer Microactuators .............................................................................. 255 8.1

Introduction ....................................................................................... 255

8.2

Sample Preparation and Measurements of Electrolytic Deformation . 257

8.3

Electrochemistry and Expansion Behaviour in Polyaniline Film ......... 259

8.4

Dependencies of the Expansion Ratio on the Degree of Oxidation and Dopant Ions ................................................................................ 260

8.5

pH Dependence of Electrolytic Expansion ......................................... 262

8.6

Time Response of the Electrolytic Expansion..................................... 265

8.7

Anisotropy of Electrolytic Expansion in Polyaniline Films ................. 266

8.8

Contraction Under Strain in Stretched Polyaniline Films ................... 267

8.9

Electrolytic Expansion in Other Conducting Polymers ...................... 267

8.10 Applications of Electrolytic Expansion .............................................. 268 8.11 Conclusions ....................................................................................... 269 References ................................................................................................... 269 9

Membranes for Electronics ......................................................................... 271 9.1

Introduction ....................................................................................... 271

9.2

Plasma Polymerisation ....................................................................... 276

9.3

9.4

9.2.1

History .................................................................................. 277

9.2.2

General Characteristics .......................................................... 277

9.2.3

Synthesis of Plasma Polymers ................................................ 278

Characterisation of Plasma Polymers ................................................. 282 9.3.1

IR Spectroscopy ..................................................................... 283

9.3.2

XPS ........................................................................................ 283

Applications of Plasma Polymers ....................................................... 283

v

Polymers in Electronics

9.4.1

Packaging .............................................................................. 284

9.4.2

Insulator ................................................................................ 284

9.4.3

Semiconductive Films ............................................................. 285

9.4.4

Conductive Films ................................................................... 286

9.4.5

Resist Films ............................................................................ 286

9.4.6

Ultrathin Polymer Films ......................................................... 286

9.4.7

Chemical Sensors ................................................................... 287

9.4.8

Biosensors .............................................................................. 287

Acknowledgements ..................................................................................... 290 References ................................................................................................... 291 10 Conducting Polymer-Based Biosensors ........................................................ 297 10.1 Introduction ....................................................................................... 297 10.1.1 Biosensors .............................................................................. 298 10.1.2 Construction of Biosensors .................................................... 299 10.1.3 Transducers ........................................................................... 301 10.1.4 Biological Component ........................................................... 301 10.1.5 Importance of Conducting Polymers to Biosensors ................ 302 10.2 Preparation of Electrodes ................................................................... 304 10.2.1 Synthesis of Conducting Polymers ......................................... 304 10.2.2 Conduction Mechanism in Conducting Polymers .................. 305 10.3 Immobilisation of Biomolecules/Enzymes .......................................... 305 10.3.1 Methods of Immobilisation.................................................... 305 10.3.2 Advantages of Immobilisation ............................................... 309 10.4 Characterisation of Enzyme Electrodes .............................................. 309 10.4.1 Determination of Enzyme Activity ......................................... 309 10.4.2 Effect of pH ........................................................................... 310 10.4.3 Effect of Temperature ............................................................ 311 10.4.4 Effect of Storage Time ........................................................... 312 10.4.5 Response Measurements ........................................................ 313

vi

Contents

10.5 Types of Biosensors ............................................................................ 313 10.5.1 Optical Biosensors ................................................................. 314 10.5.2 Electrochemical Biosensors .................................................... 315 10.6 Biosensors for Healthcare .................................................................. 318 10.6.1 Glucose Biosensor .................................................................. 318 10.6.2 Urea Biosensor ....................................................................... 320 10.6.3 Lactate Biosensor ................................................................... 321 10.6.4 Cholesterol Biosensor ............................................................ 323 10.6.5 DNA Biosensor ...................................................................... 324 10.7 Immunosensor ................................................................................... 326 10.8 Biosensors for Environmental Monitoring ......................................... 326 10.9 Conclusions ....................................................................................... 326 Acknowledgements ..................................................................................... 327 References ................................................................................................... 327 11 Nanoparticle-Dispersed Semiconducting Polymers for Electronics .............. 341 11.1 Introduction ....................................................................................... 341 11.2 Material Preparation Methods ........................................................... 344 11.3 Photophysics of Charge Separation Nanoparticle-Polymer Systems ... 346 11.3.1 TiO2-Conjugated Polymer Composites .................................. 348 11.3.2 Nanoparticle Semiconductors-Polymer Systems ..................... 353 11.3.3 Gold-Polythiophene Blends .................................................... 357 11.4 Summary ............................................................................................ 360 Acknowledgements ..................................................................................... 361 References ................................................................................................... 361 12 Polymers for Electronics .............................................................................. 367 12.1 Introduction ....................................................................................... 367 12.2 Polymer Electroluminescence ............................................................. 368 12.3 Conduction in Polymers ..................................................................... 375

vii

Polymers in Electronics

12.4 Molecular Electronics ........................................................................ 379 12.5 Polymer Deposition Technologies ...................................................... 379 12.6 Summary ............................................................................................ 388 Acknowledgements ..................................................................................... 388 References ................................................................................................... 388 13 Conducting Polymers in Molecular Electronics ........................................... 393 13.1 Introduction ....................................................................................... 393 13.2 Synthesis of Conducting Polymers ..................................................... 397 13.3 Preparation of Ultrathin Conducting Polymer Films .......................... 399 13.3.1 Langmuir-Blodgett Films ........................................................ 399 13.3.2 Self-Assembly Monolayers ..................................................... 404 13.4 Characterisation of Conducting Polymers .......................................... 404 13.5 Molecular Devices Based on Conducting Polymers ............................ 406 13.5.1 13.5.2 13.5.3 13.5.4 13.5.5 13.5.6 13.5.7 13.5.8

Diodes ................................................................................... 406 Field-Effect Transistor ............................................................ 409 Biosensors .............................................................................. 411 Electronic Tongue .................................................................. 414 Electronic Nose ...................................................................... 415 Nanowires ............................................................................. 418 Electroluminescent Displays .................................................. 419 Microactuators ...................................................................... 423

13.6 Conclusions ....................................................................................... 424 Acknowledgements ..................................................................................... 425 References ................................................................................................... 425 Abbreviations and Acronyms............................................................................. 441 Contributors ...................................................................................................... 449 Index ................................................................................................................. 453

viii

Preface

There is a global effort towards the applications of polymers in electronics. The demand for new polymeric materials that can replace the widely used semi-conductor silicon in microelectronics has recently intensified. This has essentially been due to the continuing drive towards higher circuit density of the micro-electronic components and the muchneeded very high speed processing of the data being continuously generated in various research, manufacturing and commercial establishments located worldwide. It is anticipated that polymers may perhaps offer viable solutions to the problems presently being confronted by the modern electronics industry. Among the various polymeric materials, conjugated polymers have been projected to have innumerable applications in electronics and are thus presently at the centre-stage of research and development. Conducting polymers have been found to have applications in a wide range of emerging areas such as light-emitting diodes, photonics, micro-actuators, light-weight batteries, biosensors and molecular electronics. However, it may be noted that development of polymers for electronics is still an open field wherein polymers are used not only as insulators but can also be tailored for the desired electronic properties for specific applications. It was thus thought that a Handbook dedicated entirely to the preparation, characterisation and potential applications of polymers coupled with the fundamentals of the electrical, optical and photo-physical properties will go a long way in bridging a long-felt industrial need and motivate the dedicated and younger researchers to venture into new experiments. ‘Handbook of Polymers in Electronics’ has been designed to discuss novel ways polymers can be used in the rapidly growing electronics industry. Recent developments in microelectronics have prompted enhanced interest towards the search for new molecular materials that can be utilised for increased density of packaging. Vibha Saxena and cowriters (Chapter 1) discuss the phenomenon of charge transport in electrically conducting polymers, considered to be a direct consequence of conjugation, i.e., chemical un-saturation of the carbon atoms in the polymer chain. It is indicated that an improved understanding of the mechanism of charge transport in these materials is likely to unravel new hidden phenomena having implications in polymer electronics. Reghu Menon (Chapter 2) discusses the role of easily polarisable delocalised p-electrons in determining the electrical properties of conducting polymers. Toshikuni Kaino (Chapter 3) focuses on the transmission and processing of digital information using conjugated non-linear optical devices based on polymers. Barbara Wandelt (Chapter 4) in her extensive coverage of the luminescence properties of polymers reveals how fluorescence probes can provide an insight into the nature of intermolecular interactions in these systems. Alberto Bolognesi and cowriters (Chapter 5) reveal that polymers offer a 1

Handbook of Polymers in Electronics unique possibility of working with cheaper technology giving flexible films that can be used to emit light. Jean-Claude Dubois (Chapter 6) discusses the technological developments of photopolymers and photo-resists presently being used in microelectronics industry. Bruno Scrosati (Chapter 7) has provided an interesting insight into the potential of polymeric electrodes for lightweight batteries for applications in electronics. Keiichi Kaneto and cowriters, in their outstanding appraisal (Chapter 8), have shown that the changes in molecular conformations arising due to the localisation of p-electrons and electronic repulsion between the polycations influence the operation of a conducting polymer micro-actuator. Isao Karube and cowriters (Chapter 9) have given an excellent review of the preparation, characterisation of the plasma-polymerised membranes for application in electronics. Conducting polymers have been predicted to play decisive role towards the fabrication of third generation biosensors. Keeping this in view, Asha Chaubey and cowriters (Chapter 10) have shown that redox polymers can be advantageously used to combine both the role of protein immobilisation matrices and the physical transducer resulting in improved response characteristics and miniaturisation. K.S.Narayan (Chapter 11) reveals a synergistic approach towards the nanoparticle dispersed particles semi-conducting polymers for application in miniaturised electronic devices. Tim Richardson (Chapter 12) deals with the various options available to a device engineer associated with the technological development of polymer based electronic devices. Chapter 13 contains a comprehensive review on molecular electronic applications of conducting polymers. ‘Polymers in Electronics’ is the result of the invaluable contributions of many celebrated researchers who have been active in their respective fields for many decades. I am grateful to all of them for their active participation in this important project. Special thanks are due to Ms Frances Powers of Rapra for her timely suggestion that this project should be undertaken. Ms Claire Griffiths, Dr Arshad Makhdum and Dr Sarah Ward of the editorial staff at Rapra have worked extremely hard to check that everything in the Handbook is correct and that the project is completed in time. Mr Steve Barnfield is thanked for the typesetting and the excellent cover design of the Handbook. I would also like to extend my thanks to Geoffrey Jones of Information Index who so skilfully produced the index. I am thankful to all the members of my research group (Biomolecular Electronics & Conducting Polymers) of the National Physical Laboratory (NPL), New Delhi for the many discussions and suggestions during the operation of the project. The Handbook would not have been possible without the invaluable advice received on many occasions from a number of eminent scientists including Professor S.K. Joshi, Dr R.A. Mashelkar, FRS, Dr W. Hayes, Professor A.P.F. Turner, Professor E.S.R. Gopal, Professor S. Slomkowskii, Dr A.K. Raychaudhri, Dr Krishan Lal and Dr Howard H. Weetall. I am thankful to all my colleagues especially Dr K.K. Saini, Dr. S.S. Bawa and Dr Subhas Chandra of NPL, New Delhi for many discussions held during the implementation of the project. Finally, it would have been difficult to complete the project without the emotional support I received from Shashi (wife), Aditi (daughter) and Rajat (son). Bansi D.Malhotra

2

1

Charge Transport in Conjugated Polymers V. Saxena and B.D. Malhotra

1.1 Introduction Since the early 1950s, polymers have been used extensively as passive components in electronic devices because of their light weight, flexibility, corrosion resistance, high chemical inertness, electrical insulation and ease of processing. In 1975, an inorganic conjugated polymer, polythiazyl, (SN)x, was discovered, which possesses metallic conductivity and becomes a superconductor at 0.29 K [1]. However, the idea of using polymers for their electrical conducting properties actually emerged in 1977 with the findings of Shirakawa and co-workers [2], that the iodine-doped transpolyacetylene, (CH)x, exhibits conductivity of 103 S cm-1. Since then, an active interest in synthesising other organic polymers possessing this property has been initiated. As a result, other polymers having a π-electron conjugated structure, such as polyaniline (PANI), polypyrrole (PPy), polythiophene (PT), polyfuran (PFu), poly(pphenylene) (PPP) and polycarbazole (PCz) [3-6] have been synthesised and studied. Some important conducting polymers and their energy gaps are shown in Table 1.1. Since the beginning of the last decade, these polymers (hereafter called conducting or conjugated polymers) have been extensively investigated for an understanding of their physical and chemical properties.

Table 1.1 Important conducting polymers Polymer

Optical absorption edge (eV)

Trans-polyacetylene

1.4

Cis-polyacetylene

2.0

Polypyrrole

2.5

Polythiophene

2.0

Poly(p-phenylene)

3.0

Poly(p-phenylenevinylene)

2.4

Polyaniline

1.6

3

Handbook of Polymers in Electronics The charge transfer process is one of the most intriguing properties of conducting polymers because the electrical conductivities of this class of polymers vary over many orders of magnitude due to chemical or electrochemical doping. It is understood that a wide variety of phenomena are involved in charge transport in this group of materials. A major source of this phenomenon originates from the quasi-one-dimensional (q-1D) nature of the materials. A polyconjugated chain can be considered as a q-1D metal, having one charge carrier per carbon atom. It is a well-established fact that such a half-filled system gives rise to Peierls instability by opening up an insulating gap at the Fermi level. This leads to a band structure responsible for the important electronic properties in polymers and thereby results in the existence of a non linear excitation called a soliton. This excitation and other excitations, such as polarons and bipolarons found in non degenerate groundstate systems, are produced due to the chain relaxation or deformation that results from adding/removing an electron from the polymeric chain. Under the influence of an applied electric field these non linear defects become mobile, resulting in an increased electrical conductivity. Each of these particles possesses its own characteristic transport properties. A clear understanding of the intrinsic excitations in doped and undoped conjugated polymers is still lacking [7-8]. A considerable amount of work has been carried out by several researchers focusing on this fundamental problem. The charge transfer properties as a function of temperature, pressure, magnetic fields, etc., for various polymeric samples have also been reported in the literature. The collective contributions from various parameters, such as electron-phonon interaction, electron-electron interaction, quantum lattice fluctuations, interchain interactions, etc., make it difficult to estimate the contribution from individual parameters quantitatively. Moreover, the contributions from disorder and doping, etc., make it rather difficult to envisage a microscopic mechanism for charge transport in doped conducting polymers. Therefore, the theoretical modelling of transport properties in conducting polymers is still a challenging problem due to the extreme complexity of the system. However, recent developments in reducing the extent of disorder have explained many phenomena regarding charge transport in doped conducting polymers. In this chapter, an overview of the past few years, a study of charge transport in conducting polymers is presented.

1.2 The Electronic Ground State It is well known that the accessible energy levels of an electron in a crystal are grouped into bands, which may be visualised as originating from the electronic levels of the atom. The bands form by the splitting of the atomic levels when the atoms approach one another and obtain their equilibrium positions in the crystal. The bands are separated by forbidden energy ranges called the energy gap. In a semiconductor, this gap separates the band which is completely filled (valence band) from the lowest energy band which is completely empty at absolute zero (conduction band) and accounts for the conduction processes in

4

Charge Transport in Conjugated Polymers this class of materials. In metals, the conduction band is partially filled, implying that a finite density of states exists at the Fermi level. Conjugated polymers differ from crystalline semiconductors and metals in several aspects and are often treated theoretically as a one-dimensional system. The formation of the band gap is explained taking into account either electron-phonon interactions or electronelectron interactions among π-electrons. If electron-phonon interaction dominates in real π-conjugated polymers, these systems could be treated using Peierls theory. In contrast, when electron-electron interactions dominate, the Hubbard model could be used to explain the physical properties of polymers. The Peierls model explains why a chain of unsaturated carbon atoms with one conduction electron per atom does not exhibit metallic properties. If all the atoms are spaced at equal distance, a, the basic cell in reciprocal space is the Brillouin zone in the interval –π/a
Figure 1.1 One-dimensional electronic system with a half-filled band; band structure (a) before and (b) after Peierls distortion, where E is energy and EF is the Fermi energy

5

Handbook of Polymers in Electronics occupied with electrons, a reduction of the energy will occur and the distorted state will be more favourable, implying that the semiconducting state is more stable than the metallic state. The size of the single-particle gap (δ) is proportional to the amplitude of the lattice distortion (u). However, in a three-dimensional array of one-dimensional chains, the quantum lattice fluctuations and the interchain coupling tend to reduce the Peierls gap. The Peierls model clearly indicates that a one-dimensional chain of unsaturated carbon atoms leads invariably to a semiconducting state [9]. However, Frohlich showed that the Peierls state is semiconducting only if the periodic lattice distortion is commensurate with the lattice [10]. If it is incommensurate, the phase of the periodic lattice distortion can move through the lattice carrying a charge density wave. In such a case, the Peierls state is conducting. The increased Frohlich conductivity in the Peierls state is rather sensitive to various extrinsic features such as disorder, chain interaction, pinning-depinning processes, etc. A considerable amount of work has been reported in this area [11-13]. The Peierls model completely neglects the coulomb repulsion for an electron that is transferred to a state already occupied. In the simple Hubbard model, electron correlation is taken into account, but electron-phonon interaction is assumed to be negligible [14]. This model yields a gap in the absence of a Peierls distortion. For an exactly half-filled band this model gives insulating behaviour. However, for the case deviating from the half-filled band condition, the conductivity behaviour is observed. This model provides an excellent intuitive framework and analytic benchmark for understanding the role of electron-electron interactions in a conducting polymer. However, to obtain a realistic description of these systems one needs to consider a model incorporating both electron-phonon and electron-electron interactions. The Peierls-Hubbard model incorporates both the coulomb interaction among π-electrons and their coupling to lattice degrees of freedom [15]. These models have been applied to simple polymers and found to agree with the experimental findings [16-19]. However, none of these models was found to satisfactorily explain all the physical properties of conducting polymers. Some other important models are the Su-Schrieffer-Heeger model [20, 21], the Pariser-Parr-Pople model [22], and the Mott-Hubbard model, which have been used extensively to explain the features of these materials.

1.3 Charge Transport Carriers The nature of the charge carriers in conducting polymers is not very well understood. It is believed that intrinsic excitations, such as solitons, polarons and bipolarons, do play the role of charge carriers [23, 24]. Considerable theoretical and experimental work has been carried out in this area [25-27]. The properties of these excitations are discussed in detail in the following sections.

6

Charge Transport in Conjugated Polymers

1.3.1 Soliton The probability of finding this excitation for conduction holds true for conjugated polymers with degenerate ground state structures such as trans-polyacetylene (PAc) (Figure 1.2a). In a degenerate ground state, the system has isoenergetic regions. For a neutral chain, the soliton can be thought of as a 2pz unhybridised orbital of a sp2 hybridised carbon atom, which is a non bonding orbital occupied by a single electron. Neutral, negative and positive solitons in a trans-PAc chain are shown in Figure 1.3. This excitation, called a soliton, has a 1/2 spin with zero charge, which can move along the chain without a distortion [25]. The non bonding orbital corresponds to a soliton level at the middle of the forbidden gap. It may contain zero, one or two electrons. The addition or removal of one electron to the neutral state corresponds to a negative or positive soliton with zero spin (Figure 1.4). Further oxidation of the polymer creates a dication. However, because of the two-fold degeneracy of PAc, these cations are not bound to each other by any lattice distortion and can freely separate along the chain. In the case of PAc, it is reported that solitons are delocalised over about 12 (CH) units [26]. Therefore, solitons are isolated, non interacting charged defects that form domain walls separating two phases of opposite orientation with identical energies.

Figure 1.2 Ground state energy (E) as a function of the configuration co-ordinate (Δ) for a system with a (a) degenerate and (b) non degenerate state

7

Handbook of Polymers in Electronics

(a)

(b)

(c) Figure 1.3 Formation of solitons in trans-polyacetylene: (a) neutral, (b) positive and (c) negative soliton

So

S-

S+

(a)

(b)

(c)

Figure 1.4 Energy levels and localised level occupations of the (a) neutral, (b) negative and (c) positive solitons

Several experiments have been carried out to confirm the physical properties of solitons in trans-polyacetylene [27]. Lately, this excitation has also been studied in another degenerate ground state conjugate polymer, poly(1,6-heptadiene) [28]. The onedimensional spin diffusion and associated spin dynamics are verified from electron magnetic resonance spectroscopy, nuclear magnetic resonance (NMR) spectroscopy and electron nuclear double resonance (ENDOR) measurements [13]. The density of neutral solitons has been estimated by Motsovoy and co-workers [29]. For more details on the physical properties of solitons, the reader is referred to a review article by Heeger and co-workers [13]. However, more theoretical and experimental work is

8

Charge Transport in Conjugated Polymers required in order to understand the influence of chain interaction, coulomb interaction, disorder, etc., on the physical properties of this excitation.

1.3.2 Polaron and Bipolaron In the case of a non degenerate ground state, the energies of the two chains on either side of the defect are different (Figure 1.2b). Therefore, a single bond alternation defect in such a chain cannot behave as a charged soliton. A charge injected on a chain is accompanied by a distortion in the chain and forms a polaron [30]. The removal of an electron from a polymeric system creates a free radical and a positive charge. The radical cation is then coupled by a local bond rearrangement and a quinoid-like bond sequence is formed. However, because of the higher lattice energy of quinoid compared to benzenoid, these distortions are limited. In the case of PPy, the lattice distortion is believed to extend over about four pyrrole rings. This combination of charged site coupled with a free radical via a local lattice distortion is called a polaron (Figure 1.5). A polaron may be a radical cation (oxidation) or a radical anion (reduction). Like a free carrier, a polaron has a spin of 1/2 and a charge of ±e. Polaron formation creates new localised electronic states in the band gap (Figure 1.6). Theoretical studies indicate that the polaron states of PPy are symmetrically located about 0.5 eV from the band edges [31]. The lower energy states are occupied by a single unpaired electron.

Figure 1.5 Formation of a polaron and a bipolaron in polypyrrole

9

Handbook of Polymers in Electronics

Q=+e

Q=-e

S = 1/2

S = 1/2

(a)

(b)

Figure 1.6 Energy levels and localised states for the (a) positive and (b) negative polarons

In many cases, it has been found that the conductivity of the system is spinless, which suggests that charge carriers other than polarons would be appropriate in these cases [32, 33]. Therefore, it was proposed that polaron interaction would produce a new charge carrier with no spin and 2e charge corresponding to a positive bipolaron (Figure 1.5). When a polymeric chain having a polaron is subjected to further oxidation, an electron is removed from either the polaron or the rest of the chain. In the former case, a polaron radical is removed and two new positive charges result, which are coupled through lattice distortion. In the latter case, two polarons are formed. However, the formation of a bipolaron causes a further decrease in ionisation compared to two polarons, indicating that bipolaron formation is thermodynamically more favourable. It has been reported through quantum chemical calculations that bipolaron energies are lower than those of polarons by 0.4 eV [13]. Bipolaron states are located symmetrically within the band gap, about 0.75 eV away from the band edges in case of PPy (Figure 1.7). Continuous doping of the polymer creates additional localised bipolaron states, which overlap to form continuous bipolaron bands. During doping, the polymer band gap also increases, but bipolaron bands tend to merge with the conduction band (CB) and valence band (VB), resulting in metal-like conductivity. The experimental evidence of formation of polarons and bipolarons has been given by Bredas and co-workers by examination of the optical spectra of many doped conjugated polymers [32]. The spectral and spin signatures of polaron pairs in πconjugated polymers have been reported by Lane and co-workers using photoinduced absorption and optically detected magnetic resonance (ODMR) [34]. Later, explicit

10

Charge Transport in Conjugated Polymers

Q=2e

Q=-2e

S=0

S=0

(a)

(b)

Figure 1.7 Energy levels and occupied localised states for the (a) positive and (b) negative bipolaron

evidence for bipolaron formation in conducting polymers has been provided by Ramsey and co-workers using ultraviolet photoelectron spectroscopy and electron-energy-loss spectroscopy [35]. A number of experimental and theoretical studies have dealt with the physical properties of polarons and bipolarons [36-40]. The effect of an electric field on polarons and bipolarons in a single polymer chain has been investigated by Magela [41]. It is found that dynamic effects do not reverse the stability relation between polarons and bipolarons. Conversely, as the electric field intensity gets stronger, the moving polaron structure is destroyed faster than the structure of moving bipolarons. The stability of a bipolaron as a function of the strength of the long-range coulomb interaction with and without impurities has been studied in detail [42]. It is found that in a free state a bipolaron is stable only when the coulomb interaction is small (weak coupling limit) and it becomes unstable for strong coulomb interactions. Being bound to a dopant, the bipolaron becomes stable in a wide range of coulomb interactions. Mizes and co-workers reported a study on the stability of polaron in trans-polyacetylene and poly(phenylene vinylene) [43]. They pointed out that many defects and the short conjugation length in these materials would tend to stabilise the polaron [43]. Three-dimensional band structure calculations of these polymers indicate the destabilisation of polaron. Nevertheless, the existence of chain endings and other conjugation breaks are found to stabilise the polaron. However, later, Vogl and Campbell concluded (from local density functional calculations for polyacetylene) that the effect of interchain interaction is sufficient to destabilise a polaron, thus making the electron into a conduction band electron of the kind usually found in conventional semiconductors [44]. In contrast, Emin and Nagai suggested that defect and disorder would tend to localise the polaron. The effect of conjugation is apparently

11

Handbook of Polymers in Electronics sufficient to prevent such localisation [45]. However, the roles of disorder, interchain interaction, dopants, etc., in pinning and stabilising these excitations are not precisely known. A detailed study of polarons and bipolarons is given by Conwell and Mizes [46]. The reports of these studies have implications in the emerging field of molecular electronics (Chapter 13).

1.4 Transport Properties of Polymers The strong variation of the electrical conductivity of conjugated polymers upon doping, as observed in the metal-insulator (M-I) transition, was first observed in the case of polyacetylene [47]. Since then, a great amount of work has been devoted to elucidating the conduction mechanism involved. The charge transport in these materials has been explained by dividing these materials into three regimes according to their reduced activation energy, defined as: W(T) = -T[dlnρ(T)/dT] = d(lnσ)/d(lnT)

(1.1)

where σ is the conductivity, T is the temperature and ρ is the resistivity. (i) when W(T) is greater than 0 at low temperature, the system is near the metallic side of the M-I transition. (ii) when W(T) is independent of temperature for a wide range of temperatures, the system is on the critical regime of the M-I transition (i.e., where the M-I transition is feasible). (iii)when W(T) is greater than 0 at low temperatures, the system is near the insulating side of the M-I transition. Another important parameter characterising the M-I transition is ρr, the resistivity ratio, ρr~ρ/(1.4 K)/ρ/(300 K). Being amorphous materials, the M-I transition in conducting polymers is determined by the extent of disorder. In early publications, the data analysis was not focused on the precise identification of the metallic, critical and insulating regimes. This hindered the clear understanding of the charge transport phenomenon in conducting polymers. Recent reports on the improved quality of the polymeric samples have substantially reduced the dominant role of the disorder. Therefore, the current emphasis is on the metallic properties of these systems. The transport properties of doped conducting polymers are given in Chapter 2 of this book. We herein give a brief account of the metallic conductivity observed in these polymers.

12

Charge Transport in Conjugated Polymers Metallic polymers are defined as polymers having a finite conductivity at temperatures approaching absolute zero and a room temperature conductivity in the range of that of conventional metals, such as copper [48]. Apart from this, temperature independent Pauli susceptibility down to 10 K, linear temperature dependence of the thermoelectric power down to 10 K, linear term in the specific heat at low temperatures, and large metallic reflectance in the infrared are other features observed in metallic polymers [49]. These features show the presence of a continuous density of states with a well-defined Fermi energy. Although a positive temperature coefficient of resistance (TCR) is desirable for a good metal, it is not an essential criterion for metallic behaviour. The extensive studies conducted on disordered metals in the past few years have shown that the absence of a significant positive TCR does not necessarily imply the system being non metallic. Another alternative method to determine the metallic state of polymers is the reduced activation energy as defined above. The conductivity in the disordered metals at low temperature is given by [50, 51]: σ(T) = σ(0) + mT1/2 + BT ρ/2

(1.2)

where B is a constant depending on the localisation effects, and m is a constant. σ(0) is the conductivity at absolute temperature. Here the second term in the equation arises from the electron-electron interaction and third term is the correction to σ(0) due to localisation effects. The value of ρ is determined by the temperature dependence of the scattering rate of the dominant dephasing mechanism. In the clean limit:

ρ = 3 for electron-phonon scattering ρ = 2 for inelastic electron-electron scattering

In the dirty limit:

ρ = 3/2

Among the various metallic conducting polymers studied so far, oriented transpolyacetylene has been studied the most [48, 52-56]. The maximum room temperature conductivity parallel to the chain axis obtained by Naarman and subsequently by others is approximately 105 S cm-1, being roughly one-fifth of that of copper [57-60]. The maximum value of intrinsic conductivity and stretchability of iodine-doped transpolyacetylene is very much dependent on the film thickness; the thinner the film, the higher the conductivity in both stretched and unstretched films [61]. The main difference between polyacetylene synthesised by Shirakawa and Naarman is the higher density and degree of chain orientation in the latter case, the spin dynamics being the same for both kind of polymers.

13

Handbook of Polymers in Electronics The transport properties of heavily doped Naarman polyacetylene have been reviewed by Paasch [62]. The sign of temperature dependence indicates that the conductivity is limited by material imperfections such as tunneling/hopping regions and disordered metal regions. Heavily doped and highly disordered polyacetylene is almost metallic (the Peierls distortion is suppressed) resulting in a mean free path of about 100 Å, consistent with the observed coherence length. Recently, Park and co-workers have found a metallic positive TCR at low temperature in ClO4--doped polyacetylene [63]. The temperature dependence of highly oriented iodine-doped polyacetylene has been compared with that of ClO4-- and FeCl3-doped polyacetylene and it has been found that a positive TCR in conducting polymers is highly influenced by the extent of disorder present in the sample [64]. Thummes and coworkers have emphasised that the conductivity of doped polyacetylene at low temperatures follows a T1/2 law indicating that electron-electron interactions play a dominant role at very low temperatures [65 ,66]. Similar T1/2 dependence was found in the case of Naarman polyacetylene [67]. For an intermediate temperature range (4-40 K), inelastic electronphonon scattering is found to be the dominant scattering mechanism. These facts suggest that both interaction and localisation play dominant roles in determining conductivity at low temperature in metallic polyacetylene.

1.5 Factors Influencing the Transport Properties of Polymers It may be remarked that the actual charge transport process in a conducting polymer is dependent on several parameters such as disorder (e.g., presence of vacancies, clusters, inhomogeneities), interchain coupling, the degree of doping, and the distribution and nature of dopant ions, etc.

1.5.1 Disorder It is known that disorder is an inherent feature of polymeric systems. The different sources of disorder in a polymer include inhomogeneous doping that arises due to the nature of the catalyst used and also the processing routes. Besides this, the process of preparation modifies the morphology and may lead to partial crystallinity, sp3 defects, chain termination, crosslinks, cis-segments within trans-chains and impurities, etc. Chemical defects such as non conjugated carbon atoms inserted in the chain or impurities may result in localised and strong potential, i.e., strong disorder. On the other hand, static fluctuations in the conformation of chain along its length are spread over some distance, leading to weak disorder. Both types of disorder may limit the conjugation length and hence the transport properties of polymers. Disorder effects are known to produce tails in the bands of regular systems and this usually reduces the energy gap between valence and conduction bands [68].

14

Charge Transport in Conjugated Polymers Several studies on the effect of the disorder have been based on either the Huckel or SuSchrieffer-Heeger model with perturbation added in the form of either a screened coulomb potential representing a charge impurity or a completely localised potential acting on a single site [69]. It has been argued in literature that disorder can either act as a barrier to the quasi-particle movement, or trap it. The rotation of rings or bonds owing to geometrical fluctuation is another disorder and has been shown to reduce the conjugation length [70]. The effect of disorder in interchain hopping has been investigated by Wolf and Fesser [71]. They showed that dimerisation decreases and the density of states in the gap increases as a function of increasing disorder in interchain coupling. The perfectly dimerised Peierls ground state breaks down towards a ‘metal-like’ state at a critical random distribution of interchain coupling. The random interchain coupling changes the properties of the band gap to a pseudo gap, with a small but non zero density of states. Harigaya and co-workers have reviewed the doping induced disorder in conducting polymers. Both theoretical and experimental studies suggest that it is difficult to reduce the Peierls gap to zero without taking account of the effects of disorder [72]. Lately, the effect of spatial disorder and anisotropy on the mobility of charge carriers has been reported using a dynamical Monte Carlo simulation. It is found that small disorder decreases the mobility of low external fields whereas a considerable increase in mobility is found when spatial disorder reaches value about 5%. Such studies are essential to optimise the performance of devices using these materials. [73]. A model was proposed by Levy and co-workers for understanding the experimental features observed due to disorder near the M-I transition in a q-1D conducting polymer. The polymer is modelled as a composite medium consisting of spherical regions of ordered polymers, randomly distributed in a much more disordered polymer host. Within each spherical region, the polymer chains are highly oriented, but the axis of orientation varies randomly from sphere to sphere [74]. It can be concluded that, in general, disorder is detrimental to transport phenomena in solids. The higher the disorder, the lower will be the dimensionality of the system. Although a considerable amount of work has been devoted to investigating the role of disorder in transport properties, systematic investigation of the physical properties by controlling and varying different types of disorder has not been accomplished yet. Therefore, more extensive studies are required in order to understand the effect of different types of disorder on structural, electronic and transport properties of conducting polymers. The main problem in understanding the conduction mechanism pertains to the correlation of disorder with that of the charge carrier transport.

1.5.2 Doping During the doping process, the charge is injected or removed from the polymer chain and dopant ions sit in the polymer matrix in order to maintain charge neutrality. In general, the doping process is inhomogeneous and the distribution of dopant ions in the

15

Handbook of Polymers in Electronics polymer matrix is not uniform. This is mainly due to the complex morphology of the polymer matrix, which consists of both crystalline and amorphous regions [75, 76]. The dopants easily diffuse into the vacancies in the amorphous region till saturation level is reached, after which they slowly migrate into the crystalline regions. According to the Kivelson and Heeger model [77], after adding or removing electrons to a chain, the dopant ions have a negligible effect on the electronic properties of the system. This has been attributed to high conductivity within the metallic regions and their weak dependence on the dopant species. Taking this fact into account, doped polymers can be described within an idealised π-electron model with variable electron density. However, neglecting the effect of impurity potentials arising out of dopant ions is a drastic assumption because random weak potential can sometimes produce bound states in one dimension. A theoretical study has revealed that disorder effectively quenches the Peierls distortion [78]. The influence of dopant ions on interchain coupling may either weaken it by placing chains farther away or vice versa. Cohen and Glick found that in the neighbourhood of a dopant ion the intrachain hopping is enhanced by 10% and the interchain hopping is enhanced by 100% [79]. Thus, the dopant may act as a bridge for the interchain transport. The interchain hopping strength at sites in the presence and absence of dopant ions are estimated to be nearly 0.17 eV and 0.10 eV, respectively. The electron spin resonance (ESR) measurements in conducting polymers studied by Bernier and co-workers have demonstrated that the dopants play a major role in the charge transport process [80]. Salkola and Kivelson suggest that the counterions affect the energy gap [81]. It is suggested that a minor change in the arrangement of counterions relative to the polymer chain is capable of influencing the energy gap of the polymer lattice. If the position of the soliton is taken as centre, then it is observed that if counterions are located near to odd and even sites, then the gap is enhanced. Conversely, if the counterions are located near to even or odd sites then the gap is reduced. Yamashiro and co-workers have estimated the dopant-chain interaction and its role in interchain transfer [82]. They showed that the dopants mediate the largest interchain transfer of about 0.3-0.1 eV with five to seven carbon atoms in another chain that is in contact with a common dopant column. The interchain transfer via dopants has little effect on interchain states but yields a modification of the orbital energy spectrum. It has also been reported that the electronic gap between a soliton and the conduction band is decreased due to coulomb potential of the dopant ions and vanishes at a sufficiently high dopant concentration. A random distribution of dopants can have a strong effect on disordering. The aggregation of dopants into an ordered structure suggests that finite metallic segments exist which increase in length upon doping. As a result, the density of states increases smoothly as a function of dopant concentration. Increase in dopant concentration beyond a critical level leads to the formation of a superlattice in many doped polymers [83].

16

Charge Transport in Conjugated Polymers

1.5.3 Interchain Coupling In contrast to the extensive efforts devoted to the studies of a one-dimensional model of π-conjugated polymers, no attention has been paid to account for the effect of the three dimensions of the real materials, being isotropic in nature. The interchain couplings have considerable influence on major sources affecting the solid state of polymers and thereby lead to substantial differences between the exact form of conducting polymers and their oligomers. For example, in an infinite idealised one-dimensional polymer material, bond alteration can exist in the ground state, but it does not persist at finite temperatures. Moreover, the domain walls (kink/solitons) connecting the two degenerate bond alteration regions will always be generated by thermal fluctuations at finite temperatures and even a small density of these kinks will destroy the long-range order in one dimension. Nevertheless, in real materials, even a weak interchain coupling is sufficient to sustain the long ranges order up to a certain temperature. Three-dimensional coupling is characterised by the transfer integral, t, which is the ratio of electron-phonon coupling to the electronic intersite coupling. Generally, only nearest neighbour interactions are taken into account, so only one or two transfer integrals are relevant. Another parameter, tanisotropy, can be expressed as the ratio of t⊥ to t//, t⊥ and t// being the interchain and on-chain transfer integral, respectively. This helps in determining the magnitude of the band structure anisotropy. It has been found that the threedimensional effect dominates if t⊥/t// > 10-2 [84]. Three-dimensional band calculations performed on trans-polyacetylene and polyphenylvinylene yield similar results if t⊥/t// is at least 3 x 10-2 [85]. In an anisotropic Landau-Ginzberg model, it has been argued that interchain coupling can be detrimental to polarons and bipolarons [86]. The result is confirmed by theoretical calculations with the Su-Schrieffer-Heeger (SSH) model. It is found that a polaron is destablised and charge is spread over the whole crystal if the energy gained by the lattice relaxation of a single chain is of the order of 4t. Since the polaron binding energy is small, a small t is sufficient to destabilise the polaron. Bipolarons are stable with respect to transverse delocalisation as long as coulombic effects are negligible. Ab initio calculations in crystalline polyacetylene also confirm the possibility of instability of polarons and bipolarons [87]. The role of interchain interaction in the case of oriented trans-polyacetylene has been investigated extensively by Leising and co-workers [88, 89]. Their study has shown that for a highly conducting polymer there is no gap around the Fermi energy consistent with the metallic properties of polymers. The effect of interchain coupling on excitation has been studied by Shi-Jie using a tight-binding model [90]. A new spinless charged polaronic excitation (q=±e, s=0) was obtained due to the transverse interchain coupling, which is different from the general magnetic polaron (q=±e, s=1/2) obtained in the perfect 1-D

17

Handbook of Polymers in Electronics model. A polaron is more easily stimulated energetically in a longer chain than in a shorter one. Additionally, interchain coupling decreases the energy of creation of a polaron. The effect of interchain coupling on the electronic structure of both doped and undoped trans-polyacetylene has been studied by Conwell and co-workers [91]. They suggested that the interchain coupling is energy dependent, decreasing at a constant rate from a maximum value at the bottom of the valence band. They suggested that the interchain interaction by itself is not enough for transport into the metallic state in doped conducting polymer samples. Although, in principle, the interchain coupling is sufficient to shrink the energy gap between the soliton band and the conduction band to zero, in reality the coulomb interaction with dopant ions plays a significant role in giving rise to the metallic density of states. The detailed theoretical modelling is carried out only for transpolyacetylene and it should be extended to other polymeric systems in order to delineate a comprehensive picture about the effect of interchain interaction in both doped and undoped conjugated polymers [92, 93]. Recently, Prigodin and Efetov have developed a theoretical model for interchain interaction at the M-I transition in a random network of coupled metallic chains [94]. In this model, the interchain disorder due to intrinsic defects and the randomness in the distribution of interchain contacts induce localisation. The M-I transition in such a system is determined by the critical concentration of interchain crosslinks, which in turn depends on the localisation lengths and interchain coupling. Moreover, a metallic state can exist in such a random network of coupled metallic chains only if the concentration of interfibril contacts is large enough to overcome the percolation threshold. As mentioned above, the interchain coupling plays a significant role in charge delocalisation, screening and coulomb interaction and thereby determines the nature and stability of intrinsic excitations. The transport properties of conducting polymers are highly dependent on the way the chains are organised and arranged with respect to each other. In spite of innumerable theoretical and experimental work, a quantitative estimation of the effect of interchain coupling on transport properties of various polymeric systems has not yet emerged.

1.6 Models of Charge Transport in Conducting Polymers As discussed in previous sections, the mechanisms for the formation of metallic state and charge conduction in conducting polymers have been the subject of intensive study since the occurrence of an insulator to metal transition was reported upon doping. It is proposed that non linear defects such as polarons, solitons and bipolarons have a major role in these systems [36, 95-97]. In earlier synthesised conducting polymers, inhomogeneities often dominated the transport properties, and metallic island models were proposed to

18

Charge Transport in Conjugated Polymers explain such features. Most of the transport measurements in such conducting polymers were in the insulating regime. However, recent advances in synthesising ordered and highly conducting polymers have signalled the onset of a new generation of polymers [57]. Improved homogeneity and a reduced degree of disorder in new kinds of polymers have provided a new opportunity for investigating metallic features through transport and optical measurements [49]. More recently, experimental results obtained in polyaniline and polyaniline derivatives were interpreted by invoking a q-1D variable range hopping model of Nakhmedov, Prigodin and Samukhin [98, 99]. In spite of the rapid progress in observing high quality samples of conducting polymers, the subtle details of the mechanism of charge transport are not yet understood. Firstly, the role of solitons, polarons and bipolarons in the charge transport is not very clear. The Kivelson model pertains to a phonon assisted hopping between soliton states in the case of lightly doped polyacetylene [100, 101]. However, the interpretation of experimental data in the framework of alternative models cannot be ruled out [102, 103]. Secondly, the difference between microscopic and macroscopic transport properties has to be understood clearly. Thirdly, the connection between the measured conductivity and structural parameters of the sample has to be sorted out. The present section deals with the various models reported so far in addressing these problems. In Kivelson’s model [100, 101], it is suggested that the low-energy charge excitations introduced into polyacetylene by light doping are charged solitons because of the large binding energy. At reasonable temperatures the population of solitons is extremely small and consequently thermally activated conduction due to free, charged soliton movement is difficult. At low temperatures, electron hopping between solitons is a less strongly activated conduction mechanism and it is a dominant process at low temperatures (Figure 1.8a). For example, the activation energy for phonon-assisted hopping or transition of an electron from the charged soliton to the neutral soliton is small if the neutral soliton happens to be near another impurity. The hopping conductivity is determined by the rate at which an electron hops between a pair of solitons. Because of the disorder (namely the random distribution of impurities), the conduction pathways are essentially three-dimensional, with interchain hops. Kivelson’s model predicts a steep power law dependence for the conductivity, σ=T9, which is close to that observed experimentally. The conductivity versus temperature data can be explained reasonably well in terms of the variable-range hopping (VRH) model, especially in the case of iodine-doped polyacetylene [104]. The temperature and frequency dependence of conductivity is found to be in agreement with VRH among soliton-like states in the lightly doped regime [105, 106]. Similar to Kivelson’s model, Chance and co-workers proposed interchain hopping for spinless conductivity in doped polyacetylene, doped poly(p-phenylene) and other doped polymers [102]. The mechanism accounts for the observed dopant concentration dependence of the conductivity in trans-polyacetylene and the observation of anomalously

19

Handbook of Polymers in Electronics

Figure 1.8 Interchain transport of solitons and bipolarons in (a) polyacetylene and (b) poly(p-phenylene)

low magnetic susceptibilities in the highly conducting regimes of several doped polymers (Figure 1.8b). However, this model does not explain the transition from spinless carriers to carriers with spins upon doping. Pietronero suggested that the main contribution to resistivity in conjugated polymers is due to scattering between the conduction electron and the phonon of the conjugated polymer chains [107]. The observed high conductivity of heavily doped polyacetylene has been explained using the one-dimensional model. The only possible scattering for polymers is from kF (the fermi vector) to –kF or vice versa involving large momentum (2k F) phonons of high energy (h ν~0.2 eV for

20

Charge Transport in Conjugated Polymers polyacetylene-type systems). With first-order scattering one obtains a strong conductivity increase even at room temperature, because of phonon freezing effects. High-order scattering processes with low energy phonons may then become less important. In the case of elastic scattering (kBT >> hν), where kB is the Boltzmann constant, the electrical conductivity is estimated as 1.6 x 105 S cm-1 which is nearly an order of magnitude less than the value estimated for graphite intercalation compounds. However, this estimation is not realistic since, in polymers, only 2kF phonons scatter, i.e., hν0 >> kBT. At room temperature this analysis gives high intrinsic conductivity, σ = 1.4 x 107 S cm-1. Kivelson and Heeger carried out a detailed study on the intrinsic conductivity of conducting polymers [77]. They described the expression for the conductivity as: 2πnν0e 2a 2t0

2

2 2

α h

⎛ hν ⎞ exp⎜ ⎟ ⎝ kBT ⎠

(1.3)

where α (~4.1 eV/Å) is the electron-phonon coupling constant, ν0 is the phonon frequency, t0 is the electron hopping matrix element [107], h is Planck’s constant, n is the conduction electron density and a is the C-C distance in the chain direction. A large t0, and a small number of phonons gives the conductivity at room temperature as 107 S cm-1. This value increases exponentially at low temperatures. Since the charged ions are spatially removed from the q-1D conduction path, the usual scattering of phonons is reduced. This is because phonons of wave vector 2kF are required to backscatter electrons. Since these phonons are thermally excited only at higher temperatures, the resistivity is very small at low temperatures, with a rapid rise when the thermal energy kBT approaches the energy of the 2kF phonons. At 0.12 eV, the rapid increase in resistivity can account for the change to metallic temperature dependence, which is a prominent feature of the conductivity in conducting polymers. As discussed earlier, the interchain couplings are necessary to prevent 1D localisation. Even a small interchain coupling, t⊥ ~0.1 eV, may give rise to the suppression of the Peierls distortion. The condition necessary for the system to remain three-dimensional is L/a >> 2t0/t⊥

(1.4)

where L characterises the distance between the chain interruptions or sp3 defects or crosslinks. When the concentration of chain interruptions is sufficiently high such that the left hand side of equation 1.3 is small, then the wave function will be localised. The possible limits for the conductivity arise from the chain interruptions and/or phonon scattering. All the above factors suggest that in high-quality conducting polymers the electronic mean free path could be much larger than the structural coherence length and real metallic features could be observed.

21

Handbook of Polymers in Electronics Similar charge transport models have been proposed by Prigodin and Firsov [108, 109]. According to their model, an abrupt transmission from extended to localised states is expected at the critical interchain exchange integral t⊥c~(3h/2πτ), τ being the scattering time. Therefore, the delocalisation in a q-1D metallic chain appears only if t⊥ is larger than the threshold value, t⊥~0.3/τ. Later, Nakhmdeov and co-workers carried out a study of hopping transport in q-1D systems near the M-I transition with weak disorder [98]. They set the cut-off temperature regimes in which band transport crosses over to hopping transport. In a high temperature regime, the band transport is governed by phonon scattering and disorder while at low temperatures, the hopping has been ascribed to Mott’s VRH. In the intermediate temperature region, the temperature dependence of conductivity shows a power law behaviour. However, these temperature regions are different in a q-1D system when there is a weak intrachain coupling. In these cases, the temperature dependence of conductivity gradually varies from 1D behaviour at high temperatures to that of an isotropic 3D behaviour at low temperatures. In the intermediate range, the conductivity follows exp(-T0/2T)exp[-(T0/2T)]1/2 dependence. Joo and coworkers extended this model, suggesting that the effect of finite temperature emerges through phonon scattering, which is expected to be highly anisotropic [110-112]. The role of phonon forward scattering is to break the phase coherence of the impurity scattering and thereby destroy the weak localisation. Moreover, the conductivity versus temperature curve exhibits a characteristic maximum at temperatures where the phonon backward scattering time becomes comparable with the impurity scattering time. Epstein and co-workers have widely used q-1D models to interpret transport properties in both metallic and insulating polymer systems [113, 115]. Conducting polymers were considered as an inhomogeneous system, which consists of partially crystalline and amorphous regions. The overlap of π-orbitals gives rise to crystalline regions whereas amorphous regions emerge from weak chain interactions. When the size and volume fraction of the crystalline region increases with respect to the amorphous region, the system is expected to undergo a transition from insulator to metal. The charge carriers are subjected to 1D localisation while passing through the amorphous region in between the crystalline region and thereby the movement of charge carriers occurs through the 1D localised regions and often dominates throughout the 3D extended states in crystalline regions. When the volume fraction of 3D extended states increases, the probability of movement of charge carriers is through the path of least resistance and therefore, a percolative metallic transport is expected. Microscopic properties in crystalline and amorphous regions are different and hence the usual Anderson localisation in the homogeneous disorder limit is not appropriate for conducting polymers. More recently, Samukhin and co-workers proposed [99] a q-1D fractals model in order to explain the experimental data obtained for poorly conducting polymers. It is found that at low temperatures, the VRH conductivity obeys a q-1D Mott’s law, σdc ∝ exp-(T1/T)1/2, but

22

Charge Transport in Conjugated Polymers the characteristic temperature T1 is greater than T0 for a 1D chain by a factor 1/D-1, D being the dimensionality of the system. Similar temperature dependence was obtained for dc conductivity. Low frequency conductivity is entirely controlled by the weak charge transfer between clusters, each cluster being very dense and well isolated. In contrast to q-1D models, Qiming and co-workers proposed a granular-rod model for the metallic state of the conducting polymer [116]. In this model, the metallic islands correspond to single strands of polymer. The macroscopic conductivity results from anisotropic threedimensional VRH in the network of metallic rods. This model explains very well the temperature dependence of the conductivity, σ = σ0exp(-T0/T1/2), the doping dependence of T0, the anomalous 1/T dependence of the thermoelectric power as well as the linear increase of Pauli susceptibility with dopant concentration. A temperature range, where the variable range hopping is valid, is decreased below the experimentally observed temperature range (over which the above equation holds) if the metallic islands correspond to 3D bundles of the polymer strands. Sheng’s fluctuation induced tunnelling (FIT) model was used extensively to interpret transport properties in metallic polymers. This model was originally developed for granular metal [117, 118] and polymers filled with carbon black or alumina flakes. The polymer system is described in terms of highly conducting regions separated by much less conducting or insulating areas. The electrical conduction is dominated by electron transfer between large conducting segments. Since the electrons tend to tunnel between conducting regions at points of their closest approach, the relevant tunnel junctions are usually small in size and are therefore subject to large, thermally activated voltage fluctuations across the junction. By modulating the potential barrier, the voltage fluctuations directly influence the tunneling probability and introduce a characteristic temperature variation to the normally temperature independent tunnelling conductivity. A non metallic feature of doped polyacetylene was explained in the framework of this model [119]. A similar model was used to interpret the transport properties of emeraldine polymer as a function of the protonation level, x [120]. At no composition level does the conductivity appear truly metallic. For all compositions, the conductivity behaviour is similar to that of a granular metal and this data fits transport via charging energy limited tunnelling between conductivity islands. The data were found to be consistent with percolation among these islands for x ≥ 0.3 with the presence of an insulating layer surrounding each island above the percolation threshold. The size of these islands is estimated to be 200-300 Å. Lux and co-workers reported scanning electron microscopy (SEM) and transmission electron microscopy (TEM) pictures of highly conducting polyaniline in support of the conducting island concepts [121]. Experimental measurements of TEM, X-ray diffraction (XRD), temperature dependence of dc conductivity and magnetic susceptibility indicate the applicability of this model. However, electron paramagnetic resonance (EPR) spectroscopy and magnetic susceptibility studies also suggest that pristine and doped polyaniline contain at least two types of spin carriers. It was suggested by Conwell and Mizes that the

23

Handbook of Polymers in Electronics conduction mechanism is not due to FIT dependence in conducting polymers since the metallic regions in the FIT model have negligible temperature dependence of conductivity. Also, the phonon scattering, scattering due to imperfections, and defects in the metallic regions are not considered in the FIT model. They determined the band motion versus diffusive hopping transport in oriented metallic conducting polymers to understand the effective dimensionality of the system. They showed that for band motion t⊥τ// is much greater than h/2π, while for diffusive motion t⊥τ// is much less than h/2π, τ// being the average time for the electron scattering along the chains (τ//~10-14 s from σ// = ne2τ//m*, if σ~105 S cm-1, n = 1022 cm-3, m* = free electron mass = 9.1 x 10-28 g, e = 1.6 x 10-19 C). If the system consists of high conducting regions mixed with low conducting regions, then σ⊥ is the mixture of band motion and hopping. However, the measurements σ// /σ⊥ as a function of temperature in oriented metallic trans-polyacetylene shows that σ// /σ⊥ is nearly temperature independent and hence it is impossible that σ⊥ is due to diffusive hopping in both cases [122, 123]. In contrast, experimental results demonstrated by Park and coworkers [124] indicate that σ// /σ⊥ increases with σ// and therefore, the issue of transport in highly oriented metallic conducting polymers is still debatable. Voit and Buttner examined the FIT model critically and it was concluded that physical parameters obtained from this model do not allow consistent description of highly doped polyacetylene [125]. Kaiser and Graham extended the FIT model for heterogeneous systems by introducing geometric factors to the insulating barriers. Such barriers could be due to material imperfections that dominate the total resistance of the sample [48]. If the intrinsic conductivity is very large and barriers form only short segments in the conduction path, the temperature dependence of the measured conductivity reflects that of the barriers, but the magnitude is very much larger than the barrier conductivity due to geometrical factors. Furthermore, if the barriers are reasonably good heat conductors, the temperature differences across them (and therefore, their contribution to thermoelectric power) will be small [126]. In contrast to conductivity, the thermoelectric power could then follow the intrinsic metallic behaviour. In addition to the tunnelling transport across the insulating barrier, a parallel phonon assisted hopping transport was included in the Kaiser and Graham model. The bulk conductivity is the combination of Kivelson and Heeger q-1D transport, hopping/ tunnelling transport and 3D disordered metallic transport (Figure 1.9). This heterogeneous model was reviewed by Kaiser for understanding the experimental data obtained on Naarman-polyacetylene. Localisation effects can appear at low temperatures despite the high conductivity. These could be due to charging energy effects in interchain transfer at low temperatures or to quantum corrections. The linear thermoelectric power behaviour observed indicates a smaller interaction between electrons and phonons than in normal metals, which is consistent with the remarkably high conductivities observed [127]. Paasch was of the opinion that this model contains seven independent data and so may describe

24

Charge Transport in Conjugated Polymers

Figure 1.9 Schematic representation of the conduction processes in (a) microscopic or (b) macroscopic pictures that are in agreement with transport data in Naarman-polyacetylene

the smooth temperature dependence. He argued that multiparameter fits of simple dependencies sometimes make the picture seem unambiguous whether the relative influences of the different contributions are reliable or not [128]. He modified the FIT model, arguing that one of the main barriers for the tunneling process is the chain segments with residual dimerisation. These segments exhibit a dimerisation gap which acts as a tunnelling barrier for the charge carriers. In general, however, the multiple parameter fitting procedure in FIT model has not been found satisfactory to explain the physical properties of conducting polymers. The exponential dependence of conductivity in the insulating regime of conducting polymers has been usually attributed to VRH. However, a wide range of exponent (d = 0.25-1) has been observed [49, 129, 130]. Moreover, T–1/2 behaviour of conductivity is often observed in granular metallic systems. Schreiber and Grussbach [131] suggested that the fluctuations in mesoscopic systems could give a wide range of values of the exponent due to the fractal nature of wavefunctions near the mobility edge. The transition of polymer system from q-1D to 3D was studied as a function of doping level [132].

25

Handbook of Polymers in Electronics Stafstrom [133, 134] reported that the enhancement in interchain interaction with increased dopant concentration can induce 3D localisation of the electronic states. He used the many-channel Buttiker-Landaur conductance formula [135] to study the conductance as a function of length of the system. Each channel contains several polymer segments represented by chain interruptions. The chain interruptions along the channels can be caused by sp3 defects, crosslinking between chains, etc. The hopping across such a chain interruption should, therefore, be reduced considerably compared to the interchain hopping. He showed that the most relevant parameter for causing localisation in the number of chain interruptions is the number of chain interruptions and not the channel length. The conductance of the system is unaffected by the presence of chain interruptions up to a critical value and, therefore, the critical chain length for which the conductance begins to drop should be used to characterise the transition between diffusive and non diffusive conductance. Moreover, if the interchain hopping term is equal to the interchain hopping strength (3D case) or if the number of channels is very small (1D case), the usual type of disorder-induced exponentially localised wave functions appear. Therefore, the q-1D nature of conjugated polymeric systems provides an example of a class of materials that differ from previously studied materials in the way the conductance responds to disorder in the form of chain interruptions. Recently, Schon and co-workers suggested the possibility of band-like transport in oligothiophene as a result of strong observed dispersion of the valence band based on band structure calculations [136]. Phillips and co-workers proposed a random dimmer model (RDM) [137-139] which has a set of delocalised conducting states, even in 1D, that initially allow a localised particle to move through the lattice almost ballistically. They showed that any disordered bipolaron lattice can be mapped onto a RDM. The model is found to be applicable to the M-I transition in a wide class of conducting polymers, such as polyaniline and heavily doped polyacetylene. Calculations performed on polyaniline demonstrate explicitly that the conducting state of the RDM is coincident with a recent calculation of the location of the Fermi level in the metallic region [140, 141]. A RDM analysis on poly(p-phenylene) also indicates the presence of a set of conducting states in the vicinity of the band edge. In highly disordered conducting polymers, the usual exponential dependence of temperature is explained by phonon assisted hopping or tunnelling or both. Zuppiroli and co-workers have proposed a model which describes adiabatically, the dimensionality, homogeneity, coulomb interactions and multi-phonon character in the framework of hopping conduction [142, 143]. It was shown that the electron transport in these materials is due to correlated hopping between polaronic clusters. They showed, both theoretically and experimentally, that the charging energy is the principal barrier to hopping. Polaronic clusters which originate from fluctuations in the dopant concentration function as metallic grains in the granular metal hopping model. The final temperature dependence is the same as in the case of models of Sheng,

26

Charge Transport in Conjugated Polymers Abeles, Arie or Efros and Shklovskii [144, 145]. Nagashima and co-workers recently described ac transport studies in polymers by using a statistical model of resistor network [146]. The model takes into account the polydispersiveness of the material as well as intrachain and interchain charge transport processes. The real and imaginary part of the resistivity was determined using a transfer-matrix technique. At low frequencies, interchain processes are more important and determine the transport mechanism. On the other hand, at high frequencies charge transport should be restricted along the polymer chains, as interchain processes should be dominant. Both regimes described by the model reproduce the experimental results in a remarkable way. Among various models proposed by several researchers for conducting polymers in the insulating side, the theoretical work by Ovchinnikov and Pronin [147] and Lewis [148] are slightly different from other models. In the former model, a q-1D percolation model was proposed for explaining the conductivity. According to this model, an impurity captures an electron from one of the adjacent chains and forms a charged impurity centre. Such a carrier can detrap by an activated process and diffuse along the chain. This polaron can recombine with another impurity centre near the chain and then escape to an arbitrary chain adjacent to the second impurity centre. Thus, conduction by percolation is possible in such a system if an infinite cluster of chains can be connected by impurity centres. Lewis and co-workers suggested that the charge transport occurs by tunnel transitions between localised states and lattice fluctuations, and electronlattice coupling tends to broaden and reorganise the energy levels at each site. They estimated the electric field, frequency and temperature dependences of charge transport in both conducting and non conducting polymers. Moreover, in this model the localised states correspond to the Urbach states in amorphous systems since the optical absorption from a distribution of states extend out from the fundamental absorption band edge.

1.7 Conclusions It can be concluded that interchain interaction and disorder play a leading role in governing the transport properties of polymers. Nevertheless, the theoretical understanding of charge transport phenomenon is not yet completely developed in these polymers. None of the models for charge transport in these polymers is able to explain all the features of the conducting polymers. The major hurdle is in quantifying the intrinsic and extrinsic parameters, for example, interchain and intrachain interaction, coulomb interaction, electron-phonon interaction, charge delocalisation, extent of conjugation length and disorder, etc., whose contributions to charge transport are greatly intermixed. Therefore, better understanding of the charge transport in these materials is essential to determine the various physical phenomena operating in the polymerbased electronic devices.

27

Handbook of Polymers in Electronics

Acknowledgements The authors are grateful to Dr. K. Lal, Director, NPL, New Delhi, India, for his interest in this work. Viba Saxena thanks CSIR, India, for the award of a Research Associateship.

References 1.

R.L. Greene, G.B. Street and L.J. Suter, Physical Review Letters, 1975, 34, 577.

2.

H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang and A.J. Heeger, Journal of Chemical Society Chemical Communications, 1977, 578.

3.

R.B. Seymour, Conducting Polymers, Plenum Press, New York, USA, 1981.

4.

T.A. Skotheim, Handbook of Conducting Polymers, Volumes 1 and 2, Marcel Dekker, New York, USA, 1986.

5.

P.T. Landsberg, Handbook on Semiconductors, Volume 1, North Holland, Amsterdam, The Netherlands, 1992.

6.

K.S.V. Srinivasan, Macromolecules: New Frontiers, Volumes 1 and 2, Allied Publishers Group, New Delhi, India, 1998.

7.

W.P. Su in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, 1986, 757.

8.

Y. Lu, Soliton and Polarons in Conducting Polymers, World Scientific, Singapore, 1988.

9.

R.E. Peierls, Quantum Theory of Solids, Oxford University Press, London, UK, 1955.

10. H. Frohlich, Proceedings of the Royal Society A, 1954, 223, 296. 11. G. Gruner, Reviews of Modern Physics, 1988, 60, 1129. 12. G. Gruner, Reviews of Modern Physics, 1994, 66, 1. 13. A.J. Heeger, S. Kivelson, J.R. Schrieffer and W.P. Su, Reviews of Modern Physics, 1988, 60, 781. 14. J. Hubbard, Proceedings of the Royal Society A, 1963, 276, 238.

28

Charge Transport in Conjugated Polymers 15. E. Jeckelmann and D. Baeriswyl, Synthetic Metals, 1994, 65, 211. 16. T.C. Clarke, R.D. Kendrick and C.S. Yannoni, Journal de Physique Colloques, 1983, 51, 2136. 17. S. Kivelson in Solitons, Eds., S. Trullinger, V. Zakharov and V.L. Pokroxsky, North-Holland, Amsterdam, 1986, 17, 301. 18. D.K. Campbell, A.R. Bishop and M.J. Rice in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, 1986, 937. 19. R.R. Chance, D.S. Boudreaux, J.L. Bredas and R. Silbey in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, 1986, 825. 20. A.J. Heeger in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, 1986, 729. 21. D. Baeriswyl, D.K. Campbell and S. Mazumadar in Conjugated Conducting Polymers, Ed., H.G. Kiess, Springer-Verlag, Berlin, 1992. 22. J.E. Frommer and R.R. Chance in Electrically Conducting Polymers, Encyclopedia Reprint Series, John Wiley & Sons, New York, 1991. 23. K. Pakbaz, C.H. Lee, A.J. Heeger, T.W. Hagler and D. McBranch, Synthetic Metals, 1994, 64, 295. 24. V.I. Arkhipov, H. Bässler, M. Deussen, E.O. Gobel, R. Kersting, H. Kurz, U. Lemmer and R.F. Mahrt, Physical Review B, 1995, 52, 4932. 25. J.A. Pople and S.H. Walmsley, Molecular Physics, 1962, 5, 15. 26. W.P. Su, J.R. Schrieffer and A.J. Heeger, Physical Review B, 1980, 22, 2099. 27. Y. Cao and A.J. Heeger, Synthetic Metals, 1990, 39, 205. 28. K. Pakbaz, R. Wu, F. Wudl and A.J. Heeger, Journal of Chemical Physics, 1993, 99, 590. 29. M.V. Mostovoy, M.T. Figge and J. Knoester, Physical Review B, 1998, 57, 2861. 30. S. Roth and H. Bleier, Advanced Physics, 1987, 36, 385. 31. D. Emin and T. Holstein, Physical Review Letters, 1976, 36, 323.

29

Handbook of Polymers in Electronics 32. J. Bredas and G.B. Street, Account of Chemical Research, 1985, 18, 309. 33

J. Chen, A.J. Heeger and F. Wudl, Solid State Communications, 1986, 58, 251.

34. P.A. Lane, X. Wei and Z.V. Vardeny, Physical Review B, 1997, 56, 4626. 35. M.G. Ramsey, D. Steinmuller and F.P. Netzer, Physical Review B, 1990, 42, 5902. 36. S.A. Brazovskii and N.N. Kirova, Pisma Zhurnal Eksperimental noi I Teoreticheskoi Fiziki, 1981, 33, 6. 37. K. Fesser, A.R. Bishop and D.K. Campbell, Physical Review B, 1983, 27, 4804. 38. M.N. Bussac and L. Zuppiroli, Physical Review B, 1993, 47, 5493. 39. F.L. Pratt, S.J. Blundell and B.W. Lovett, Synthetic Metals, 1999, 101, 323. 40. Y. Arikabe, M. Kuwabara and Y. Ono, Journal of the Physical Society of Japan, 1996, 65, 1317. 41. G. Magela e Silva, Physical Review B, 2000, 61, 10777. 42. S. Brazovskii, N. Kirova, Z.G. Yu, A.R. Bishop and A. Saxena, Optical Materials, 1998, 9, 502. 43. H.A. Mizes and E.M. Conwell, Physical Review Letters, 1993, 70, 1505. 44. P. Vogl and D.K. Campbell, Physical Review B, 1990, 41, 12797. 45. D. Emin and K.L. Nagai, Journal de Physique Colloque, C3, 1983, 471. 46. E.M. Conwell and H.A. Mizes in Handbook on Semiconductors, Ed., P.T. Landsberg, North Holland, Amsterdam, 1992, 583. 47. B. Derrida, J.G. Zabolitzky, J. Vannimenus and D. Stauffer, Journal of Statistical Physics, 1984, 36, 34. 48. A.B. Kaiser and S.C. Graham, Synthetic Metals, 1990, 36, 367. 49. R. Menon, C.O. Yoon, D. Moses and A.J. Heeger in Handbook of Conducting Polymers, Eds., T.A. Skotheim, R.L. Elsenbaumer and J.R. Reynolds, Marcel Dekker, New York, 1998, 27. 50. P. Dai, Y. Zhang and M.P. Sarachik, Physical Review B, 1992, 45, 3984,

30

Charge Transport in Conjugated Polymers 51. P. Dai, Y. Zhang and M.P. Sarachik, Physical Review B, 1992, 46, 6724. 52. Th. Schimmel, G. Denninger, W. Riess, J. Voit, M. Schwoerer, W. Schoepe and H. Naarmann, Synthetic Metals, 1989, 28D, 11. 53. R.H. Baughman and L.W. Shacklette, Physical Review B, 1989, 39, 5872. 54. G. Paasch, G. Lehmann and L. Wuckel, Synthetic Metals, 1990, 37, 23. 55. C.O. Yoon, M. Reghu, A.J. Heeger, E.B. Park, Y.W. Park, K. Akagi and H. Shirakawa, Synthetic Metals, 1995, 69, 79. 56. J. Tsukamoto, Advanced Physics, 1992, 41, 509 and references therein. 57. H. Naarman, Synthetic Metals, 1987, 17, 223. 58. J. Tsukamoto, A. Takahashi and K. Kawasaki, Japanese Journal of Applied Physics, 1989, 29, 125. 59. Y. Nogami, H. Kaneko, T. Ishiguro, A. Takahashi, J. Tsukamoto and N. Hosoito, Solid State Communications, 1990, 76, 583. 60. W. Pukacki, R. Zuzok and S. Roth in Electronic Properties of Polymers, Eds., H. Kuzmany, M. Mehring and S. Roth, Springer-Verlag, Berlin, 1992, 12. 61. H. Shirakawa, Y.X. Zhang, T. Okuda, K. Sakamaki and K. Akagi, Synthetic Metals, 1999, 65, 93. 62. G. Paasch, Synthetic Metals, 1992, 51, 7. 63. Y.W. Park, E.S. Choi and D.S. Suh, Synthetic Metals, 1998, 96, 81. 64. M. Ahlskog, R. Menon and A.J. Heeger, Journal of Physics: Condensed Matter, 1997, 9, 4145. 65. G. Thummes, U. Zimmer, F. Korner and J. Kotzler, Japanese Journal of Applied Physics Supplement, 1987, 26, 3713. 66. G. Thummes, F. Korner and J. Kotzler, Solid State Communications, 1988, 67, 215. 67. Y. Nogami, H. Kaneko, H. Ito, T. Ishiguro, T. Sasaki, N. Toyota, A. Takahashi and J. Tsukamoto, Physical Review B, 1991, 43, 11829. 68. S. Stafstrom, Physical Review B, 1993, 47, 12437.

31

Handbook of Polymers in Electronics 69. G.W. Bryant and A.J. Glick, Physical Review B, 1982, 26, 5855. 70. J.L. Bredas, G.B. Street, B. Themans and J.M. Andre, Journal of Chemical Physics, 1985, 83, 1323. 71. M. Wolf and K. Fesser, Journal of Physics: Condensed Matter, 1991, 3, 29, 5489. 72. H. Harigaya, Y. Eada and K. Fesser, Proceedings of Yamada Conference XXIV, Japan, 1989, 255. 73. J. Stephan, A. Liemant, F. Albrecht and L. Brehmer, Synthetic Metals, 2000, 109, 327. 74. O. Levy and D. Stroud, Journal of Physics: Condensed Matter, 1997, 9, L599. 75. J.P. Pouget, Z. Oblakowski, Y. Nogami, P. A. Albany, M. Laridjani, E.J. Oh, Y. Min, A.G. MacDiarmid, T. Tsukamato, T. Ishguro and A.J. Epstein, Synthetic Metals, 1994, 65, 131. 76. Y. Nagomi, J.P. Pouget and T. Ishguro, Synthetic Metals, 1994, 62, 257. 77. S. Kivelson, and A.J. Heeger, Synthetic Metals, 1989, 22, 371. 78. E.J. Mele and M.J. Rice, Physical Review B, 1989, 40, 1630. 79. R.J. Cohen and A.J. Glick, Physical Review B, 1990, 42, 7659. 80. P. Bernier, A. El-khodary, F. Rachdi and C. Fite, Synthetic Metals, 1987, 17, 413. 81. M.I. Salkola and S.A. Kivelson, Physical Review B, 1994, 50, 13962. 82. A. Yamashiro, A. Ikawa and H. Fukutome, Synthetic Metals, 1994, 65, 233. 83. J.P. Pouget in Electronic Processes of Conjugated Polymers and Related Compounds, Eds., H. Kuzmany, M. Mehring and S. Roth, Springer, Heidelberg, 1985, 26. 84. Yu. N. Gartstein and A.A. Zakhidov, Solid State Communications, 1985, 60, 105. 85. P. Gomes da Costa, R.G. Dandrea and E.M. Conwell, Physical Review B, 1993, 47, 1800. 86. D. Emin, Physical Review B, 1986, 33, 3973. 87. P. Vogl and D.K. Campbell, Physical Review Letters, 1989, 62, 2012.

32

Charge Transport in Conjugated Polymers 88. G. Leising, Physical Review B, 1988, 38, 10313. 89. G. Leising, Synthetic Metals, 1989, 28D, 215. 90. S.-J. Xie, Journal of Physics: Condensed Matter, 1996, 8, 2185. 91. E.M. Conwell and H.A. Mizes, Synthetic Metals, 1993, 55-57, 4284. 92. E.M. Conwell, H.Y. Choi and S. Jeyadev, Synthetic Metals, 1992, 49-50, 359. 93. H.A. Mizes and E.M. Conwell, Physical Review B, 1991, 43, 9053. 94. V.N. Prigodin and K.B. Efetov, Synthetic Metals, 1994, 65, 195. 95. M.J. Rice, Physics Letters, 1979, 71A, 152. 96. D.K. Campbell and A.R. Bishop, Physical Review B, 1981, 24, 4859. 97. Y. Onodera, Physical Review B, 1984, 30, 775. 98. E.P. Nakhmedov, V.N. Prigodin and A.N. Samukhin, Soviet Physics Solid State, 1989, 31, 368. 99. A.N. Samukhin, V.N. Prigodin, L. Jastrabik and A.J. Epstein, 1998, Physical Review B, 58, 11354. 100. S. Kivelson, Physical Review Letters, 1981, 46, 1344, 101. S. Kivelson, Physical Review B, 1982, 25, 3798. 102. R.R. Chance, J.L. Bredas and R. Silbey, Physical Review B, 1984, 29, 4491. 103. P. Kuivalainen, H. Stubb, H. Isotalo, P. Yli-Lahti and C. Holmstrom, Physical Review B, 1985, 31, 7901. 104. A.J. Epstein, H. Rommelmann, R. Bigelow, H.W. Gibson, D.M. Hoffmann and D.B. Tanner, Physical Review Letters, 1983, 50, 1866. 105. A.J. Epstein in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, 1986, 1041. 106. A.J. Epstein, R.W. Bigelow, A. Feldblum, H.W. Gibson, D.M. Hoffmann and D.B. Tanner, Synthetic Metals, 1984, 9, 155. 107. L. Pietronero, Synthetic Metals, 1983, 8, 225.

33

Handbook of Polymers in Electronics 108. V.N. Prigodin and Y.A. Firsov, Pisma Zhurnal Eksperimental noi i Teoreticheskoi Fiziki, 1983, 38, 241. 109. Y.A. Firsov in Localisation and Metal-insulator Transition, Eds., H. Fritzsche and D. Adler, Plenum, New York, 1985, 477. 110. J. Joo, V.N. Prigodin, Y.G. Min, A.G. MacDiarmid and A.J. Epstein, Physical Review B, 1994, 50, 12226. 111. J. Joo, Z. Oblakowski, G. Du, J.P. Pouget, E.J. Oh, J.M. Wiesinger, Y. Min, A.G. MacDiarmid and A.J. Epstein, Physical Review B, 1994, 49, 2977. 112. J. Joo and A.J. Epstein, Review of Scientific Instruments, 1994, 65, 2653. 113. A.J. Epstein, J. Joo, R.S. Kohlman, G. Du, A.G. MacDiarmid, E.J. Oh, Y. Min, J. Tsukamoto, H. Kaneko and J.P. Pouget, Synthetic Metals, 1994, 65, 149. 114. Z. Wang, A. Ray, A.G. MacDiarmid and A.J. Epstein, Physical Review B, 1991, 43, 4373. 115. A.J. Epstein, J.M. Ginder, F. Zuo, H.S. Woo, D.R. Atnner, A.F. Richter, M. Angeloupolos, W.S. Huang and A.G. MacDiarmid, Synthetic Metals, 1987, 21, 63. 116. Q. Li, L. Cruz and P. Phillips, Physical Review B, 1993, 47, 1840. 117. P. Sheng, Physical Review, B, 1980, 21, 2180. 118. P. Sheng and J. Klafter, Physical Review, B, 1983, 27, 2583. 119. Th. Schimmel, G. Denninger, W. Riess, J. Voit, M. Schwderer, Synthetic Metals, 1989, 28, D11. 120. F. Zuo, M. Angelopoulos, A.G. MacDiarmid and A.J. Epstein, Physical Review B, 1987, 36, 3475. 121. F. Lux, G. Hinnichsen, V.I. Krinichnyi, I.B. Nazarova, S.D. Cheremisov, M.M. Pohl, Synthetic Metals, 1993, 55-57, 347. 122. E.M. Conwell and H.A. Mizes, Synthetic Metals, 1990, 38, 319. 123. H.A. Mizes and E.M. Conwell, Physical Review B, 1991, 44, 3963. 124. Y.W. Park, C.O. Yoon, C.H. Lee, H. Shirakawa, Y. Suezaki and K. Akagi, Synthetic Metals, 1989, 28, D27.

34

Charge Transport in Conjugated Polymers 125. J. Voit and H. Buttner, Solid State Communications, 1988, 67, 1233. 126. A.B. Kaiser, Physical Review B, 1989, 40, 2806. 127. A.B. Kaiser, Synthetic Metals, 1991, 45, 183. 128. G. Paasch, Synthetic Metals, 1992, 51, 7. 129. S. Roth and H. Bleier, Advanced Physics, 1985, 36, 385. 130. S. Roth in Hopping Transport in Solids, Eds., M. Pollak, B. Shklovskii, NorthHolland, Amsterdam, 1995, 377 and references therein. 131. M. Schreiber and H. Grussbach, Philosophical Magazine, 1992, B65, 707. 132. J.A. Reedijk, H.C.F. Martens, S.M.C. van Bohemen, O. Hilt, H.B. Brom and M.A.J. Michels, Synthetic Metals, 1999, 475, 101. 133. S. Stafstrom, Synthetic Metals, 1994, 65, 185. 134. S. Stafstrom, Physical Review B, 1995, 51, 4137. 135. M. Buttiker, Y. Imry, R. Landauer and S. Pinhas, Physical Review B, 1985, 31, 6201. 136. J.H. Schon, Ch. Kloc, B. Batolgg, Synthetic Metals, 2000, 115, 75. 137. P. Phillips and H.L. Wu, Science, 1991, 252, 1805. 138. P. Phillips, Annual Review Physical Chemistry, 1993, 44, 115. 139. D.H. Dunlap, H-L. Wu and P.W. Phillips, Physical Review Letters, 1990, 65, 88. 140. D.S. Galvao, D.A. dos Santos, B. Laks, C.P. de Melo and M.J. Caldas, Physical Review Letters, 1989, 63, 786. 141. D.S. Galvao, D.A. dos Santos, B. Laks, C.P. de Melo and M.J. Caldas, Physical Review Letters, 1990, 65, 527. 142. L. Zuppiroli, M.N. Bussac, S. Paschen, O. Chauvet and L. Forro, Physical Review B, 1994, 50, 5196. 143. O. Chauvet, S. Paschen, L. Forro, L. Zuppiroli, P. Bujard, K. Kai and W. Wernet, Synthetic Metals, 1994, 63, 115.

35

Handbook of Polymers in Electronics 144. P. Sheng, B. Abeles and Y. Arie, Physical Review Letters, 1973, 31, 44. 145. A.L. Efros and B.I. Sklovskii, Journal of Physics C, 1975, 8, L49. 146. H.N. Nagashima, R.N. Onody and R.M. Faria, Physical Review B, 1999, 59, 905. 147. A.A. Ovchinnikov and K.A. Pronin, Synthetic Metals, 1991, 41-43, 3373. 148. T.J. Lewis, Faraday Discussions Chemical Society, 1989, 88, 189.

36

2

Electrical Properties of Doped Conjugated Polymers R. Menon

2.1 Introduction In the past three decades, several types of π-electron systems have shown very interesting features in electrical transport properties [1-4]. Charge-transfer complexes, intercalated graphite, conjugated polymers, carbon-60, carbon nanotubes, etc., are some of the wellknown π-electron systems. Polymeric materials were considered as insulators before the discovery of metallic poly(sulfur nitride), [SN]x, and the enhancement of conductivity in doped polyacetylene, (CH)x, by several orders of magnitude [4, 5]. The polyconjugated chains -(C=C-C=C-C=C)n- consist of alternating single (σ-bonds) and double bonds (π-bonds). The π-electrons are highly delocalised and easily polarisable, and these features play important roles in the electrical and optical properties of polyconjugated systems. It also makes the latter rather different from conventional electronic systems [6-8]. Moreover, the intrinsic q-1D nature and the extent of both intra- and interchain delocalisation of π-electrons play significant roles in the structural, electrical and optical properties of polyconjugated systems. Nevertheless, the complex morphology of polymeric systems, which are partially crystalline and partially amorphous in nature, plays a crucial role in the physical properties. In general, the conjugation length, the strength of the interchain interaction and the extent of disorder are some of the significant parameters that govern the physical properties of polyconjugated systems. The electrical and optical properties of (CH)x, PANI, PPy, PT, poly(p-phenylenevinylene) (PPV), PPP and polythienylene vinylene (PTV) are some of the extensively studied conjugated polymers [7, 9]. In first-generation conducting polymers (1976-1986), the maximum possible values of electrical conductivity were limited due to the presence of structural and morphological disorder, disorder-induced localisation, etc. The metallic features were rather weak. This was mainly due to the presence of strong structural and morphological disorder, as a result the π-electrons were not very well delocalised to facilitate intra- and interchain charge transport [10]. In the past decade, significant improvement in reducing the structural and morphological disorder has helped to create the new generation of conducting polymers in which the metallic features are predominantly observed in transport measurements. For example, in iodine-doped Tsukomoto (CH)x, the conductivity was around 105 S cm-1 [11].

37

Handbook of Polymers in Electronics By the early 1990s, several groups had started making high quality materials of PPV, PPy, PANI and polyalkylthiophene (PAT) [9]. In doped, oriented PPV samples the conductivity is of the order of 104 S cm-1 [12]. In high quality, PF6-doped PPy and PT samples (prepared by low temperature electrochemical polymerisation), the conductivity is nearly 500 S cm-1 [13]. With the development of counterion-induced processibility of PANI by dodecylbenzoyl sulfonic acid (DBSA) and camphor sulfonic acid (CSA) dopants, the conductivity was enhanced to nearly 500 S cm-1, and its temperature dependence showed a significant metallic positive TCR in the range 150-350 K [14-16]. The conductivity was enhanced to 103 S cm-1 in the case of regioregular PAT [9, 17]. The conductivity of undoped polyconjugated systems is 10-6-10-10 S cm-1, hence it can be considered at the semiconductor-insulator boundary [18]. The band gaps of known polyconjugated systems vary from 0.8 to 4 eV [9]. The charge carrier density in conducting polymers can be varied by several orders of magnitude (nearly 8 orders) by doping. In fully doped systems, the carrier density could be as high as 1022/cm3. The carrier mobility in doped conducting polymers is much lower with respect to that in inorganic semiconductors, and this is largely due to the presence of strong disorder in polymeric systems. Transient charge carriers can be generated by photoexcitation in conjugated polymers [7, 8, 19]. The maximum level of doping in conjugated polymers could be as high as 50%, and that corresponds to one dopant per two monomers. In conducting polymers the doping process can generate various types of charge carriers like polarons, bipolarons, solitons, free carriers, etc., and this to a large extent depends on the doping level, the structure of the polyconjugated chain, interchain interactions, disorder, etc., [7, 8, 19]. In degenerate systems like (CH)x, solitons are formed, especially at doping levels below 6%. However, in non degenerate systems like PPy, PT, etc., both polarons and bipolarons are formed depending upon the energetics. However, as the interchain interactions and the carrier density increases and the extent of disorder decreases, these excitations could behave more like free carriers. The M-I transition in doped conducting polymers is mainly governed by the extent of disorder, interchain interaction and doping level [18, 20]. The main source of disorder in conducting polymers are the sp3 defects in the chain, chain ends, chain entanglements, voids, morphological and doping induced defects, etc., [9]. In fibrillar morphology, the chains are extended, hence it is possible to have delocalised states along the chain length direction. In globular morphology, the chains are coiled up, and this tends to localise the electronic states. In unoriented conducting polymer systems, the chains are randomly dispersed and the physical properties are isotropic. However, by orienting the polymer chains by mechanical stretching it is possible to enhance the conductivity along the orienting axis, and an anisotropy of conductivity of the order of 100 can be

38

Electrical Properties of Doped Conjugated Polymers easily achieved. Since conducting polymers are partially crystalline and partially amorphous, the volume fraction of crystalline regions and the size of the crystalline coherence length play dominant roles in the charge transport. In general, the disorderinduced localisation plays a dominant role in the M-I transition and in the transport properties of conducting polymers.

2.2 Metallic State The metallic state in doped conducting polymers is inferred from the following: a large finite dc conductivity as the temperature (T) goes to 0 K, temperature independent Pauli spin susceptibility down to 10 K, linear temperature dependence of the thermoelectric power down to 10 K, linear term in the specific heat at low temperatures, free carrier absorption and large metallic reflectance in the infrared, etc., [20]. This evidence indicates the presence of a continuous density of states with a well-defined Fermi energy. In some conducting polymers, the typical metallic positive TCR was observed from 300 to 150 K, and in some others it was only below 20 K. However, recently Park and co-workers reported a metallic positive TCR in doped (CH)x, from 300 to 1.5 K, which is quite exceptional [21]. Although the typical negative TCR in high quality conducting polymers is indicative of non metallic behaviour, its temperature dependence was rather weak so that the logarithmic derivative of conductivity, σ, (W = dlnσ/dlnT) has a positive temperature coefficient. This implies a finite value of conductivity and a finite density of states at the Fermi level at very low temperatures, as expected in the case of disordered metallic systems. This evidence indicates that in spite of the disordered q-1D nature of polymer chains, it is possible to have a metallic state in these systems [22]. In the metallic state, the average size of the delocalised states is considerably larger than that of the structural coherence length, hence the carrier transport is slightly hindered by the presence of disorder potentials in the amorphous region. Although conducting polymers are intrinsically q-1D electronic systems, the interchain coupling can be sufficiently large to enable the formation of three-dimensional metals. The critical parameter in the M-I transition is the statistical average of a wide range of values of the correlation/localisation length, Lc. If Lc is greater than the average structural coherence length (which characterises the size of the crystalline regions), then the disorder can be considered as within the weak limit, which means the system sees only an average of the random fluctuations of the disorder potentials. Conversely, if Lc is less than the average structural coherence length, then the extent of disorder is considerably higher. The values of Lc and structural coherence length can be determined by transport property measurements and X-ray diffraction, respectively [9, 20].

39

Handbook of Polymers in Electronics

2.2.1 Conductivity The electrical conductivity is mainly determined by the carrier density, n, relaxation time, τ, and effective mass, m, of the carrier (electrical conductivity, σ = ne2τ/m, where e is the electron charge). According to the Ioffe-Regel criterion, the interatomic distance is considered as the lower limit for the mean free path, λ, in a metallic system. Hence, for a metallic system kFλ is greater than 1, where kFλ = [h(3π2)2/3] / (e2ρn1/3), kF is the Fermi wavevector and ρ is the electrical resistivity [23-25]. In highly doped conducting polymers, n ≈1021 per cm3, λ is around 10 Å, and kFλ ≈1-10, at room temperature. The details about metallic conducting polymers are shown in Table 2.1 [26].

Table 2.1 The details of various doped conducting polymers in the metallic state Abbreviation

Repeat Unit

Orientability

Crystallinitya

Conductivityb

(CH)x

C2H2

High

80%

104-105

PPV

(C6H4)C2H2

High

80%

104

Polyaniline

PANI

(C6H4)NH

Low

50%

400

Polypyrrole

PPy

C5H2N

Low

50%

400

PEDOT

C7H4O2S

Low

40%

300

PMeT

(C5H2S)CH3

Low

40%

400

Polymer Polyacetylene Poly(p-phenylene vinylene)

Poly(3,4-ethylene dioxythiophene Poly(3-methyl thiophene) a

Approximate values for high-quality samples.

b

In S cm-1. The values give approximately the highest observed values. In the cases of (CH)x and PPV, the conductivity in the highly oriented state along the direction of the chain alignment is given. Reproduced by permission from M. Ahlskog, R. Menon, Journal of Physics: Condensed Matter, IOP Publishing, 1998, 10, 32, 31-33, 7171.

Pietronero suggested that in a one-dimensional chain the only possible source of scattering for charge carriers is from +kF to –kF, involving the high-energy 2kF phonons [27]. Moreover, due to phonon freezing effects, possible even at room temperature, the first-order scattering should induce a strong enhancement in conductivity in one-dimensional chains. The conductivity in the chain direction, σ⎜⎜, in a one-dimensional chain is given by:

σ⎜⎜ = (ne2a/πh)vFτ = (ne2a2/πh) / (λ/a)

40

(2.1)

Electrical Properties of Doped Conjugated Polymers where a is the carbon-carbon distance, vF (=2t0a/h) is the Fermi velocity, t0 (= 2-3 eV) is the π-electron hopping matrix element and τ is the backscattering lifetime. In the limit of elastic scattering for a half-filled band system, σ⎜⎜(300 K)≈105 S cm-1. However, since the main scattering involved in a conducting polymer chain is only due to the 2kF phonons, and by including the inelastic scattering process below the characteristic temperature (kBT∼hω0/4 and T∼600 K), another two orders of magnitude of enhancement in conductivity should be possible, i.e., ≈ 107 S cm-1. Similar estimates of conductivity have also been obtained by the Kivelson and Heeger model [28]. In later models, the conductivity is expected to increase exponentially at low temperatures. Hence, the theoretical studies suggest that the intrinsic conductivity in one-dimensional models of conjugated polymers is expected to be even larger than that of conventional metals. The low temperature conductivity measurement is a simple and sensitive method to get a qualitative level of understanding about the extent of disorder present in the system. Since conductivity is directly related to the mean free path (which is rather sensitive to the presence of any disorder) the variation in low temperature conductivity is quite dramatic as disorder varies. The characteristic behaviour of the temperature dependence of conductivity can be understood in detail by defining the reduced activation energy, W, as the logarithmic derivative of the temperature dependence of conductivity, i.e., W = d(lnσ)/d(lnT) [29, 20]. If the system has a finite value of conductivity with a negative TCR, then W shows a positive temperature coefficient at low temperatures. Moreover, this ensures that there is a finite conductivity as T→0. In general, as the resistivity ratio, ρr [(= ρ(1.4 K)/ρ(300 K)], increases the temperature dependence of W gradually moves from a positive (metallic) to a negative (insulating) temperature coefficient at low temperatures. The approximate values of ρr for various conducting polymers in the metallic (M), critical (C) and insulating (I) regimes are shown in Table 2.2 [30]. The conductivity in the disordered metallic regime is expressed by [25, 31]

σ(T) = σ (0) + m´T1/2 + BTp/2

(2.2)

where m´ = α[4/3 – γ Fσ/2], α is a parameter depending on the diffusion coefficient, γFσ is the interaction parameter, p is determined by the scattering rate (for electron-phonon scattering, p = 3; for inelastic electron-electron (e-e) scattering, p = 2 in the clean (weakly disordered) limit or 3/2 in the dirty (strongly disordered) limit). The second term in Equation 2.2 results from the e-e interactions and the third term is the correction to σ(0) due to the localisation effects. In disordered metals, e-e interactions play an important role in the low temperature transport. Usually, the sign of m is negative when γFσ > 8/9, and this results in the change of sign (from negative to positive) in the TCR at low temperatures.

41

Handbook of Polymers in Electronics

Table 2.2 The σ(300 K) and ρr ≈ [ρ ρ(1.3 K)/ρ(300 K)] of various conducting polymers in the metallic (M), critical (C), and insulating regime (I) M

(CH)x-I2

C σ(300 K)

ρr

σ(300 K)

ρr

σ(300 K)

<10

>5,000

10-20

3-5 x 104

>20

<3,000

4

9.8-165

2-5 x 10

>400

<2 x 104

2.6-11.4

1-2 x 104

>27

<104

9.7-34

100-300

>50

<100

(CH)x-I2

<5

>5 x 10

(CH)x-FeCl3

<2

>2 x 104

PPV-AsF5b

<2

300-2,400 3

PPV-H2SO4

<2

4 x 10 -10

PPy

<2

PANI

<2

a

I

ρr

a

4

4

3

4.7-27

1-4 x 10

>60

<1,000

300-400

2-10

200-300

>10

<200

250-350

2-5

200-250

>10

<200

-1

Conductivity data is given in S cm .

b

The data comes from samples with different degrees of orientation and therefore does not give an entirely consistent picture of the M-I transition in this system. Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, Journal of Physics: Condensed Matter, IOP Publishing, 1997, 9, 20-22, 4145.

Iodine-doped polyacetylene I-(CH)x is one of the most extensively studied systems among doped conducting polymers [11, 32]. The maximum room temperature conductivity parallel to chain axis in highly oriented I-(CH)x is nearly 105 S cm-1, and the anisotropy of conductivity is 100-200. The stretchability and the maximum obtainable conductivity in I-(CH)x is very much dependent on the film thickness. The conductivity is higher in thinner films (below 10 μm), in both stretched and unstretched films. The structural and physical properties of I-(CH)x have been reviewed by Tsukamoto [11]. The average crystalline coherence lengths, parallel and perpendicular to the chain axis, are 150 and 50 Å, respectively. In highly conducting samples, the carrier density is nearly 1022 cm-3 and the mean free path is around 500 Å. The density of states at the Fermi level is approximately 0.3 states (eV C)-1, (1.8 x 1018 states (J A S)-1). Recently Park and co-workers [21] have observed, for the first time in conducting polymers, a metallic positive TCR, from 300 to 1.5 K, in ClO4-doped (CH)x; although in earlier work the positive TCR was observed in FeCl3-(CH)x and PANI-CSA down to 180 K. The conductivity for a ClO4-(CH)x sample at 300 K is nearly 40,000 S cm-1. Its conductivity increases by a factor of two at 1.5 K, as shown in Figure 2.1. For samples with conductivity in the range 2,000-20,000 S cm-1, the positive TCR was observed at temperatures above 150 K. The room temperature conductivity of highly oriented I-(CH)x is nearly a factor of two

42

Electrical Properties of Doped Conjugated Polymers

Figure 2.1 Normalised resistivity (ρ) versus temperature for ClO4-doped polyacetylene samples doped by: (1) ClO4 (Fe), ρ(300 K) ~ 2.41 x 10-5 (Ω cm); (2) ClO4 (Fe), ρ(300 K) ~ 5.01 x 10-5 (Ω cm); (3) ClO4 (Fe), ρ(300 K) ~ 6.58 x 10-5 (Ω cm); (4) ClO4 (Fe), ρ(300 K) ~ 8.84 x 10-5 (Ω cm); (5) ClO4 (Cu), ρ(300 K) ~ 5.88 x 10-5 (Ω cm); (6) ClO4 (Cu), ρ(300 K) ~ 1.67 x 10-4 (Ω cm). (Reproduced with permission from Y.W. Park, E.S. Choi, and D.S. Suh, Synthetic Metals, Elsevier Science SA, 1998, 96, 1-3, 81)

larger than that of ClO4-(CH)x and FeCl3-(CH)x; yet, it does not show any metallic positive TCR [30]. The temperature dependence of resistivity of I-(CH)x samples is shown in Figure 2.2a [30, 33]. The metallic samples have a rather weak negative TCR, and it shows a large finite conductivity at temperatures below 1 K. The absence of a positive TCR in the case of I-(CH)x suggests that the doping induced disorder is higher with respect to that in ClO4(CH)x and FeCl3-(CH)x systems. The sample B2 is systematically aged up to B6, so that it continuously moves from the metallic to insulating regime. The W versus T plot of the ρ(T) is shown in Figure 2.2b. The metallic B2 shows the expected positive temperature coefficient of W(T), and upon ageing it tends towards a negative temperature coefficient, as expected for insulating systems. This indicates that the positive TCR in conducting polymers is quite sensitive to subtle variations in the extent of disorder present in the system.

43

Handbook of Polymers in Electronics

Figure 2.2 (a) Resistivity versus temperature for an iodine-doped polyacetylene sample aged from metallic B2 to insulating B6; (b) W versus T for the same data. The dotted lines indicate the power law regime. (Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, Journal of Physics: Condensed Matter, IOP Publishing, 1997, 9, 20-22, 4145)

The σ(T) of I-(CH)x samples at various stretching ratios (l/l0) is shown in Figure 2.3 [18]. The unstretched sample has a σ(300 K) ∼ 800 S cm-1, and it shows a strong negative TCR. The anisotropy of conductivity increases upon increasing the stretching ratio. Even in highly stretched samples the number of misaligned and criss-crossed chains is rather high so that the σ(T) is nearly identical for both parallel and perpendicular directions to the chain axis, and the interchain transport plays an important role in both cases. Hence, further enhancement in chain orientation (σ⎜⎜/σ⊥ >103) is of considerable importance in order to observe the intrinsic anisotropic features in the charge transport properties in conducting polymers. As described in previous works [18, 21], the σ(T) below 4.2 K follows a T1/2 dependence (see Equation 2.2), both parallel and perpendicular to the chain axis in oriented I-(CH)x samples. The T1/2 dependence indicates that the contribution from e-e interactions plays 44

Electrical Properties of Doped Conjugated Polymers

Figure 2.3 Conductivity (both parallel and perpendicular) versus temperature for iodine-doped polyacetylene samples at various stretching ratios (l/l0) (Reproduced by permission from C.O. Yoon, R. Menon, A.J. Heeger, E.B. Park, Y.W. Park, K. Akagi, and H. Shirakawa, Synthetic Metals, Elsevier Science SA, 1995, 69, 1-3, 79)

a dominant role at very low temperatures. This is also consistent with the enhanced negative contribution to magnetoconductance (MC), as explained in detail in Section 2.2. For an intermediate temperature range (4-40 K), where σ ∝ T3/4, the inelastic electronphonon scattering (p = 3/2) is the dominant scattering mechanism for both parallel and perpendicular directions to chain axis [20, 32]. This is also consistent with the enhanced positive contribution to MC at temperatures above 4 K. This suggests that both interaction and localisation play dominant roles in σ(T) at low temperatures in metallic (CH)x samples. The temperature dependence of resistivity of a sulfuric acid doped PPV (PPV- H2SO4) sample (A) as it gradually aged to sample number (L) is shown in Figure 2.4a [34, 35]. Free-standing films of PPV (thickness 4-8 μm) were stretch-aligned to 10:1 ratio [12]. The optical anisotropy is nearly 50, as obtained from the dichroic ratio measurement at 1520 cm-1. This indicates that the PPV chains are quite well oriented after tensile drawing. The W versus T plot of the same data is shown in Figure 2.4b. The metallic samples follow a positive temperature coefficient of W at low temperatures; as ρr increases, W gradually moves towards the critical and insulating regimes, similar to that observed in the case of I-(CH)x. 45

Handbook of Polymers in Electronics

(a)

(b)

Figure 2.4 (a) Resistivity versus temperature for a PPV-H2SO4 sample from metallic (highly doped) C to insulating (less doped) L and (b) W versus T for the same data. (A-E on the metallic side, G-L on the insulating side) (Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi, and T. Ohnishi, Physical Review B, American Physical Society, 1997, 55, 11, 6777)

In PPV-H2SO4 samples, the low temperature conductivity of metallic samples in both parallel and perpendicular directions to the chain axis can also be fitted to Equation 2.2 [34, 35]. However, at temperatures below 4 K, the T1/2 fit is rather good even in the presence of a magnetic field, as shown in earlier works [34]. Hence, in metallic PPV samples, the localisation-interaction model is valid at low temperatures, as observed in the case of metallic (CH)x samples. Moreover, the MC data in both systems are consistent with the localisation-interaction model, as explained below. Although the anisotropy of conductivity in metallic, oriented (CH)x and PPV samples is nearly 100, the σ(T) is rather similar in both parallel and perpendicular directions to the chain axis, indicating that an anisotropic three-dimensional model is appropriate in these systems. The temperature dependence of conductivity of PPy doped with PF6 (PPy-PF6) is shown in Figure 2.5a [36]. The W versus T plot of the same data is shown in Figure 2.5b. The

46

Electrical Properties of Doped Conjugated Polymers

(a)

(b)

Figure 2.5 (a) Normalised resistivity versus temperature for various PPy-PF6 samples (M→ for metallic, I→ for insulating, and ‘c’ for critical) and (b) W versus T for the same data (Reproduced by permission from C.O. Yoon, R. Menon, D. Moses, and A.J. Heeger, Physical Review B, American Physical Society, 1994, 49, 16, 10851)

metallic sample (M1) shows a positive TCR below 12 K, and m′ in Equation 2.2 is negative. The behaviour of W(T) is consistent with that observed in metallic (CH)x and PPV samples. The temperature and field dependence of the resistivity of metallic PPy-PF6 samples at temperatures below 1 K shows the presence of a large finite conductivity (∼150 S cm-1) at 75 mK and 8 T field [37]. This suggests that the three-dimensional transport is quite robust and that the intrinsic one-dimensional nature of disordered polymer chains is not causing any severe localisation, although the conductivity at mK temperatures in these systems is around the minimum value of conductivity of a metallic substance according to Mott’s theory [24]. Hence, the interchain interactions are sufficiently large enough to prevent any severe localisation in polymer chains. Furthermore, PPy-PF6 samples can be stretched by a factor of two, and their conductivity increases up to 3000 S cm-1 [13]. This is one of the best examples of conductivity in a doped conjugated polymer that has a long-term stability at ambient conditions.

47

Handbook of Polymers in Electronics The σ(T) of 2-acrylamido-2-methyl-1-propane-sulfonic acid (AMPSA) samples are shown in Figure 2.6 [38, 39]. These samples show a positive TCR at temperatures above 80 K. Although the σ(300 K) of PANI-AMPSA samples is slightly lower with respect to PANICSA, its positive TCR is observed down to much lower temperatures. The wet spun PANI-AMPSA fibres have shown an impressive room temperature conductivity of nearly 2000 S cm-1 [40]. The difference in σ(300 K) and σ(T) between CSA- and AMPSA-doped PANI samples is possibly due to the variations in microstructure and its contribution to disorder-induced localisation. This shows that the charge transport in doped conducting polymers is rather sensitive to slight variations in the morphological features.

Figure 2.6 Conductivity versus temperature for PANI-AMPSA samples at various doping levels mentioned in subscripts (Reproduced by permission from P.N. Adams, P. Devasagayam, S.J. Pomfret, L. Abell, and A.P. Monkman, Journal of Physics: Condensed Matter, IOP Publishing, 1998, 10, 37-38, 8293) 48

Electrical Properties of Doped Conjugated Polymers The σ(T) of doped poly(3,4-ethylenedioxy-thiophene), PEDOT, samples are shown in Figure 2.7 [41, 42]. In some doped PEDOT samples, a positive TCR have been observed at temperatures below 10 K. Similar positive TCR and weak σ(T) values have been observed in poly(3-methyl)thiophene (PMeT)-PF6 samples having σ(300 K) ≈ 200 S cm-1. In these systems, Masubuchi and co-workers have observed a pressure tuning of the M-I transition [43, 44]. In doped PEDOT samples, the σ(T) can be exceptionally weak, as shown in Figure 2.7. For example, even for samples with σ(300 K) ≈10 S cm-1 (i.e., much below the Mott minimum value), the conductivity at 1 K can be around 4 S cm-1. One possible explanation is that the structural disorder in PEDOT samples is considerably less with respect to other systems; since the β-positions in the thiopene rings are blocked by an ethylenedioxy group, the chain extension is possible only through the α-positions. A large finite conductivity [~150 S cm-1] has been observed in metallic PEDOT and PMeT systems too. In summary, the conductivity and its temperature dependence in doped (CH)x, PPV, PPy, PANI and PEDOT samples suggests that by reducing the role of disorder-induced localisation in charge transport, it is possible to observe the intrinsic metallic positive TCR in doped conducting polymers.

Figure 2.7 Normalised resistivity versus temperature for various doped PEDOT samples (from metallic PF1 to insulating PF6, from metallic BF1 to insulating BF5) (Reproduced by permission from A. Aleshin, R. Kiebooms, R. Menon, and A.J. Heeger, Synthetic Metals, Elsevier Science SA , 1997, 90, 1-3, 61)

49

Handbook of Polymers in Electronics

2.2.2 Magnetoconductance It is well known that MC is a sensitive local probe for investigating the microscopic transport parameters (e.g., scattering process, relaxation mechanism, etc.) in metallic and semiconducting systems [18]. The quantitative level of understanding of MC for disordered systems obtained using the localisation-interaction model is rather useful to check the appropriateness of this model for metallic conducting polymers. Moreover, the consistency of using this model can be verified by comparing the results from the temperature dependence of conductivity and MC measurements. It is well known that in an ideal one-dimensional conductor the transverse orbital motion is restricted, thus the carriers cannot make circular motion in the presence of a magnetic field. Hence, one could hardly expect any MC in an ideal one-dimensional conductor. However, in the presence of any finite interchain transfer integral, as in several q-1D conductors, the MC can be used as a powerful tool to investigate the intrachain versus interchain transport. Nevertheless, the fine features in anisotropic MC can be made inconspicuous in the presence of disorder. In disordered metallic systems, it is well known that the quantum corrections due to weak localisation (WL) and e-e interactions contribute to the MC at low temperatures [18, 32]. Usually, the WL contribution (positive MC) dominates at temperatures greater than 4 K and low fields (below 3 T) and the contribution from e-e interaction (negative MC) dominates at temperature below 4 K and higher fields (above 3 T). However, as the extent of disorder increases, both WL and e-e interaction contributions decrease, and the hopping contribution to MC (large negative MC) increases. Although the theoretical estimate for the upper limit to the quantum corrections to MC in conventional disordered systems is below 3% (i.e., Δσ/σ < ± 3%), the observed MC in oriented metallic conducting polymers is slightly higher, which is probably associated with the anisotropic diffusion coefficient, anisotropic effective mass, etc. There are several detailed studies of MC in conducting polymers [18, 32, 45]. The behaviour of MC in oriented and unoriented conducting polymers shows a clear difference. Highly conducting, oriented samples (e.g., (CH)x and PPV) exhibit a positive MC due to the WL contribution (especially, when the field is perpendicular to the chain axis) at temperatures greater than 2 K; whereas, in unoriented samples (e.g., PPy, PANI, PEDOT) the MC is observed to be negative at all temperatures and fields [31]. In oriented samples, the MC is an interplay between WL (positive MC at low fields and temperatures greater than 2 K) and e-e interaction (negative MC at high fields and temperature less than 2 K) contributions; the sign of the MC depends on the angle between the magnetic field and chain axis. This anisotropic MC in oriented samples is due to the anisotropy in the WL contribution, since the positive MC due to the WL contribution is maximised when the field is perpendicular to the chain axis, and it is minimised when the field is parallel to the chain axis [32, 34, 35].

50

Electrical Properties of Doped Conjugated Polymers

(a)

(b)

Figure 2.8 (a) Magnetoconductance versus field for a metallic I-(CH)x sample at 4.2 K (dot), 2 K (square) and 1.2 K (triangle), upper is transverse and lower is longitudinal (Reproduced by permission from R. Menon, K. Vakiparta, Y. Cao, and D. Moses, Physical Review B, American Physical Society, 1994, 49, 23, 16162) (b) Magnetoconductance versus field for a metallic PPV-H2SO4 sample, upper is transverse and lower is longitudinal (Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi, and T. Ohnishi, Physical Review B, American Physical Society, 1996, 53, 23, 15529)

The MC for oriented metallic I-(CH)x and PPV-H2SO4 samples are shown in Figure 2.8a [32] and (b) [34, 35], respectively. In both systems, the anisotropy of conductivity is nearly 100. As shown in these figures, when the field is perpendicular to the chain axis, the sign of the MC is positive. However, when the field is parallel to the chain axis, the sign of the MC is negative. The experimental results in both systems show that the WL contribution is negligible when the field is parallel to the chain axis. Although the aged sample remains metallic down to 1.4 K, the positive MC due to the WL contribution vanishes owing to the increasing extent of disorder caused by the aging. This indicates that even a marginal increase in the extent of disorder in metallic conducting polymers 51

Handbook of Polymers in Electronics can suppress the quantum transport involved in the WL contribution to a positive MC. Hence, the coherent interchain transport in intrinsically q-1D polymer chains can be easily affected by minute variations in the interchain alignments, disorder, etc. The MC due to the e-e interaction can be distinguished from the WL by scaling the total value [34, 46]. The scaling plot for a metallic oriented PPV-H2SO4 sample is shown in Figure 2.9. The expected universal scaling behaviour in the longitudinal MC, due to the dominant contribution from e-e interaction, is clearly shown in Figure 2.9b. However, in the case of transverse MC, the scaling behaviour deviates, as shown in Figure 2.9a. This clearly shows the importance of the WL contribution to the transverse MC. This anisotropic MC can be used to probe the microscopic anisotropy at the molecular length

(a)

(b)

Figure 2.9 Universal scaling plot of magnetoconductance for a metallic PPV-H2SO4 sample: (a) transverse (b) longitudinal (Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi and T. Ohnishi al., Physical Review B, American Physical Society, 1996, 53, 23, 15529)

52

Electrical Properties of Doped Conjugated Polymers

Figure 2.10 Universal scaling plot of magnetoconductance for a metallic PEDOT- PF6 sample at various temperatures (dots 1.48 K; diamonds 2.02 K; open triangles 3.03 K; closed triangles 4.24 K) (Reproduced by permission from A. Aleshin, R. Kiebooms, R. Menon, F. Wudl, and A.J. Heeger, Physical Review B, American Physical Society, 1997, 56, 7, 3659)

scale which is usually masked by the morphological features in the bulk conductivity measurements. Hence, modelling the anisotropic MC can provide a quantitative estimate of the number of misaligned chains in oriented samples. In slightly less conducting (~ 100 S cm-1) metallic samples of unoriented PEDOT a universal scaling of the MC has been observed, as shown in Figure 2.10 [41, 42]. In this case, since the chains are not oriented, the scaling behaviour is nearly identical in both transverse and longitudinal MC, indicating that the e-e interaction is the dominant contribution to the MC. Although the anisotropy of conductivity in metallic oriented-(CH)x or PPV-H2SO4 samples is nearly 100, the behaviour of the MC is identical whether the current is parallel or perpendicular to the chain axis. This suggests that high quality oriented conducting polymers behave as anisotropic three-dimensional systems in which the charge transport mechanism is nearly identical in both parallel and perpendicular directions to the chain axis. The MC for a metallic PANI-CSA sample, at low and high fields, is shown in Figure 2.11 [15]. The MC shows H2 and H1/2 dependence at low and high fields, respectively. Similar results have been observed for metallic PPy [36] and PEDOT [41] samples. This field dependence is consistent with the localisation-interaction model. These systems are just

53

Handbook of Polymers in Electronics

(a)

(b)

Figure 2.11 Magnetoconductance versus field for a metallic PANI-CSA sample: (a) low field fit and (b) high field fit (Reproduced by permission from R. Menon, C.O. Yoon, D. Moses, A.J. Heeger, and Y. Cao, Physical Review B, American Physical Society, 1993, 48, 24, 17685)

on the metallic side of the M-I boundary, and their values of conductivity are not high enough to observe any positive MC due to the WL contribution. Hence, in PANI, PPy and PEDOT systems, the MC remains negative at all temperatures and fields. When the extent of disorder increases and the system moves to the critical and insulating regimes the negative MC increases dramatically. In summary, the MC in metallic conducting polymers is a sensitive probe for investigating the microscopic charge transport mechanism. In oriented metallic conducting polymers, the anisotropic MC depends on the angle between the chain axis and the applied field, and this can give qualitative information regarding the molecular scale anisotropy in these systems. In less conducting and unoriented systems, the magnitude of the negative MC increases as ρr increases, and this can be used to discover the extent of disorder present in the system.

54

Electrical Properties of Doped Conjugated Polymers

2.2.3 Thermoelectric Power In doped conducting polymers, the thermoelectric power is not as sensitive to disorder as electrical conductivity, since the mean free path involved in the electrical transport is very much affected by the extent of disorder present in the system [18]. Although the metallic positive TCR is rather unusual in highly conducting polymers for a wide range of temperatures, the metallic linear temperature dependence of thermoelectric power, S(T), is quite usual in all conducting polymers for a wide range of temperatures (10-300 K). The quasi-linear temperature dependence of thermoelectric power is observed to persist well into the insulating regime. The remarkable linearity of S(T) and the negligible non linear contribution to thermoelectric power in high quality metallic (CH)x indicate that the lattice contribution to thermoelectric power due to phonons is less significant. The S(T) is quite linear even in the case of samples on the insulating side of the M-I transition [18]. Kaiser [47] has proposed a heterogeneous model to explain the apparent difference in the behaviour of S(T) and σ(T). In such a model, the less conducting regions that limit the motion of charge carriers, determine the bulk transport properties. If the thermal current carried by phonons is impeded less by thin insulating barriers than the electrical current carried by electrons or holes, then the system indeed shows a metallic thermoelectric power. In other words, the majority of the temperature gradient occurs across the highly conducting regions and the majority of the electrical potential drop occurs across the thin insulating barriers. A large enhancement of thermoelectric power is expected if the barriers play any significant role to the total value of the thermoelectric power. However, in conducting polymers, on the metallic and critical regimes of the MI transition, no such enhancement has been observed; this indicates that the barriers contribute little to the thermoelectric power. The S(T) of various PPy-PF6 samples on both sides of the M-I transition is shown in Figure 2.12 [36]. Similar results have been observed in PANI-CSA samples too [48]. In these systems the quasi-linear thermoelectric power is relatively insensitive to the variations in ρr near the disorder-induced M-I transition. The density of states estimated from S(T) is around one state per eV per two rings for PANI-CSA, and nearly one state per eV per four rings for PPy-PF6, for samples near the M-I transition. The conventional notion suggests that the sign of the thermoelectric power depends on the sign of the charge carrier. However, Park and co-workers [49] have observed a positive thermoelectric power for both p- and n-type doped (CH)x. Since the sign of the thermoelectric power depends on the band structure, etc., it may not be that straightforward in complex systems such as doped conducting polymers. More work is required to fully understand this anomaly regarding the sign of thermoelectric power in conducting polymers.

55

Handbook of Polymers in Electronics

Figure 2.12 Thermoelectric power versus temperature for various PPy-PF6 samples from metallic side (dot) to insulating side (square) (Reproduced by permission from C.O. Yoon, R. Menon, D. Moses, and A.J. Heeger, Physical Review B, American Physical Society, 1994, 49, 16, 10851)

2.2.4 Magnetic Susceptibility and Specific Heat In metallic systems, the temperature-independent Pauli susceptibility (χP) is a characteristic feature for delocalised carriers [18]. The Pauli susceptibility is directly proportional to the density of states at the Fermi level, i.e. χP = μ2BN(EF). where μB is the Bohr magneton and N(EF) is the density of states at the Fermi level. Hence, it is possible to determine the N(EF) from the temperature-independent χP for metallic systems. Usually, in disordered systems, the measured χP is the sum of both Curie and Pauli terms; the Curie term gives an estimate of the localised spins present in the system, and this in turn is a measure of the extent of disorder. A small Curie term has been observed in all metallic conducting polymers at very low temperatures (T < 20 K) [18]. This indicates the presence of localised spins due to impurities, defects, etc. The χ(T) of PANI-CSA samples near the M-I transition show this behaviour [50]. The density of states at the Fermi level for metallic PANI-CSA and PPyPF6 samples are one states per eV per two rings and three states per eV per four rings, respectively [51]. These values are rather similar to that obtained from the thermoelectric power measurements. The Curie term at low temperatures is lower for metallic samples than for insulating samples. The magnetic properties and spin dynamics in doped conducting polymers are described in recent review articles [51].

56

Electrical Properties of Doped Conjugated Polymers

Figure 2.13 Heat capacity (C/T) versus temperature for doped PPy samples from metallic PPy-PF6 (circles) to insulating PPy(TSO) [triangles] (Reproduced by permission from T.H. Gilani and T. Ishiguro, Journal of the Physical Society of Japan, Physical Society of Japan, 1997, 66, 3, 727)

There are few reports of thermal property measurements (e.g., thermal conductivity, specific heat, etc.) [52, 53]. The linear term in specific heat at low temperatures is evidence of the continuous density of states with a well-defined Fermi energy for any metallic system. The low temperature specific heat, C, for a metallic PPy-PF6 sample and for an insulating PPy-p-toluenesulfonate (TSO) sample is shown in Figure 2.13 [54]. The data for both samples fit to the equation C/T = γ + βT2, where γ and β are the electronic and lattice contributions, respectively. From the values of β and γ, the calculated density of states for metallic and insulating samples are 3.6±0.12 and 1.2±0.04 states per eV per unit, and the corresponding Debye temperatures are 210±7 and 191±6.3 K, respectively. These values are comparable to those obtained from the spin susceptibility data. Although the resistivity ratio of the insulating PPy-TSO sample is an order of magnitude larger with respect to the metallic PPy-PF6 sample, both systems show a linear term in specific heat, and the density of states of the insulating system is only a factor of three lower with respect to that in the metallic sample [51]. Moreover, from the specific heat data [54] it seems that both systems have a finite density of states at the Fermi level, hence the insulating system can be considered as slightly less metallic (less density of states due to localisation) with respect to the ‘real’ metallic system (i.e., finite conductivity as the temperature tends to 0 K). The comparison of specific heat and conductivity data

57

Handbook of Polymers in Electronics indicates that the latter is more sensitive for determining the extent of disorder, and for discerning the metallic and insulating systems near the M-I transition. Furthermore, both systems do not show any anomalies due to glassy behaviour that can occur due to the localisation of charge carriers in the amorphous regions, one-dimensional localisation, etc. Surprisingly, the linear term in the specific heat is strongly present in both metallic and insulating systems down to 2 K, although a Curie term in the spin susceptibility was observed for all conducting polymer samples at low temperatures. This indicates that for conducting polymer samples near the M-I transition, the extent of disorder is not severe enough to induce any drastic localisation of charge carriers, contrary to the conventional view that all states are localised in a disordered one-dimensional conductor.

2.3 Critical and Insulating States The intrinsic metallic state in high quality doped conducting polymers is suppressed by disorder-induced localisation [18, 20]. As the extent of disorder increases, the effective conjugation length and the interchain transport decreases; the system gradually moves from the metallic to the insulating state via the M-I transition. From the previous section it is known that even slight variations in the extent of disorder have dramatic effects in the temperature dependence of conductivity near the M-I transition. However, in other transport property measurements like thermoelectric power, specific heat, etc., the effect of disorder is not that conspicuous near the M-I transition, especially in conducting polymer systems. The power law behaviour of conductivity is universal for systems at the critical regime, and the temperature dependence of conductivity is given by σ(T) ∝ Tβ, where β is between 0.33 and 1 [20]. In the power law regime (temperature-independent W(T) regime) W(T) = β. The value of β and the temperature range of the power law regime can be determined from the log-log plots of W versus T. As a typical example, the W versus T plot of various conducting polymer systems near the critical regime is shown in Figure 2.14 [55]. Usually, the value of β is lower for systems with a smaller value of ρr, and as ρr increases the system moves towards the insulating regime in which σ(T) becomes exponential, i.e., σ(T) ∝ exp(-T 0/T) γ, where γ = 1/(d +1) and d is the dimensionality of the system. The W versus T plots for various conducting polymers are shown in Figures 2.2b, 2.4b and 2.5b. In all these figures, the temperature coefficient of W(T) varies distinctly for metallic (positive), critical (temperature-independent) and insulating (negative) regimes. These figures clearly show that the critical regime is rather robust in conducting polymers. Moreover, the critical regime can be easily tuned to the metallic and insulating regimes by pressure and magnetic fields, respectively, as shown in case of a PPy-PF6 sample in

58

Electrical Properties of Doped Conjugated Polymers

Figure 2.14 W versus temperature plot for various conducting polymers in the critical regime of M-I transition: unoriented I-(CH) x (open circle), oriented I-(CH) x (open square), oriented K-(CH) x (dark square), PPy-PF6 (dark triangle) and PANI-CSA (dark circle) (Reproduced by permission from R. Menon, C.O. Yoon, D. Moses, Y. Cao and A.J. Heeger, Synthetic Metals, Elsevier Science SA, 1995, 69, 1-3, 329)

Figure 2.15 [55]. The enhanced interchain transport under pressure is driving the system towards the metallic state. Alternatively, the field-induced transition to the critical and insulating regimes occurs when the localisation length is comparable to the magnetic length, and the field shrinks the overlap of the wavefunctions of the delocalised states. The field-induced transition, from metallic to critical regime, for a doped PPV sample is shown in Figure 2.16 [35]. Although this sample has a large finite conductivity in zero field, it approaches zero at 8 T. Another typical example of a field-induced transition, from power law to exponential law behaviour of conductivity at low temperatures, is shown for a PANI-CSA sample in Figure 2.17 [15, 16]. These field-induced transitions in conducting polymer systems near the M-I transition show that the mobility edge and the Fermi level are situated rather close to each other, and slight variations in the overlap of the wavefunctions of the delocalised states by magnetic field, disorder, etc., could alter the electronic properties of the system. The negative magnetoresistance (MR) in metallic (CH)x and PPV samples is quite sensitive to the extent of disorder [18, 20]. Even in the case of oriented metallic samples of (CH)x

59

Handbook of Polymers in Electronics

Figure 2.15 W versus temperature plot for a PPy-PF6 sample in the critical regime: pressureinduced transition to the metallic side and the field-induced transition to the insulating side (Reproduced by permission from R. Menon, C.O. Yoon, D. Moses, Y. Cao and A.J. Heeger, Synthetic Metals, Elsevier Science SA, 1995, 69, 1-3, 329)

Figure 2.16 Field-induced transition from metallic to critical regime: (a) W versus temperature for a metallic PPV-H2SO4 sample (E from Figure 2.4) at 0, 5 and 8 T fields and (b) conductivity versus T0.1 for same data (Reproduced by permission from M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi, and T. Ohnishi, Physical Review B, American Physical Society, 1997, 55, 11, 6777)

60

Electrical Properties of Doped Conjugated Polymers

Figure 2.17 Field-induced transition from critical to insulating regime for PANI-CSA samples: resistivity versus T–1/4 plot. H = 0 T (dark dot) and H = 10 T (dark diamond) for a sample with ρ(T) ∝ T–0.26 and H = 0 T (dark square) and H = 8 T (dark triangle) for a sample with ρ(T) ∝ T–0.36 (Reproduced by permission from R. Menon, C.O. Yoon, D. Moses, A.J. Heeger, and Y. Cao, Physical Review B, American Physical Society, 1993, 48, 24, 17685)

and PPV, a marginal increase in disorder can easily suppress the quantum interference process involved in the WL contribution to negative MR [35, 45]. As disorder increases and the system moves towards the critical and insulating regimes, the positive MR contribution increases due to the shrinkage of the overlap of the wave function of the localised states. In conducting polymers, as the chain orientation decreases, the contribution to positive MR increases. The MR is positive at all temperatures and fields in PANI, PPy, PMET, PEDOT, etc., since the chains are less oriented in these systems with respect to (CH)x and PPV [20]. In these systems, as ρr increases the positive MR increases, as shown in the case of PPy-PF6 samples in Figure 2.18 [36]. Hence, the magnitude of the positive MR can be used to obtain a rough estimate of the extent of disorder present in the system. Usually the magnitude of the positive MR at very low temperatures (T < 4 K) is as sensitive as the temperature dependence of conductivity for probing the extent of disorder present in the sample. The typical values of the magnitude of positive MR at 1.4 K and 8 T field, in the metallic, critical and insulating regimes are less than 5%, between 5% and 20%, and larger than 20%, respectively [18]. The large positive MR in the VRH regime is due to the shrinkage of the overlap of the tails of the wavefunctions of the localised states, and the hopping transport becomes more difficult at higher fields and

61

Handbook of Polymers in Electronics

Figure 2.18 Magnetoresistance versus resistivity ratio (ρr) for various PPy-PF6 samples from metallic to insulating side (Reproduced by permission from C.O. Yoon, R. Menon, D. Moses, and A.J. Heeger, Physical Review B, American Physical Society, 1994, 49, 16, 10851)

lower temperatures. In the VRH regime, the low field MR follows a H2 dependence, and this can be used to determine the localisation length in the insulating regime. If the localisation length is of the order of few hundreds (few tens) of Angstrom then the extend of disorder is relatively low (high) [56]. Hence, the MR data is complementary to the low temperature conductivity data in probing the extent of disorder present in critical and insulating systems. Although transport property measurements, like thermoelectric power, specific heat, etc., are important in the critical and insulating regimes, they are not as sensitive as the low temperature conductivity and MR. The linear temperature dependence of thermoelectric power and a linear term in the specific heat (which are typical for metallic systems) have been observed in the insulating regime too. As the system moves into the insulating side the hopping contribution to thermoelectric power (Shop ∝ T1/2) dominates over the metallic diffusion thermoelectric power (Sdif ∝ T) [56]. Usually the negative hopping contribution can be estimated by subtracting the diffusion contribution, and this gives an estimate of the extent of disorder present in the macroscopic level. The gradual variation of S(T) from the positive linear temperature dependence to the negative temperature dependence can be correlated with the microstructure, as the system moves to the highly disordered granular, metallic islands. A typical example of this feature is shown in Figure 2.19 [20, 56]. These features in S(T) can be used to obtain a qualitative picture about the macroscopic scale disorder present in conducting polymers.

62

Electrical Properties of Doped Conjugated Polymers

Figure 2.19 Hopping contribution to thermoelectric power for various doped PANI samples (Reproduced by permission from C.O. Yoon, R. Menon, D. Moses, A.J. Heeger, Y. Cao, T.A. Chen, X. Wu and R. D. Reike, Synthetic Metals, Elsevier Science SA, 1995, 75, 1-3, 229)

The Curie term in χ(T) is rather dominant at low temperatures for systems in the critical and insulating regimes, and the temperature independent Pauli term is usually observed at temperatures greater than 100 K. However, χ(T) is not as sensitive as σ(T) and MR for identifying the metallic, critical and insulating regimes near the M-I transition [17].

2.4 Summary This brief overview of the electrical transport properties in doped conducting polymers highlights the following points: (1) Conducting polymers are rather complex systems, in which the structural and morphological features influence the electrical and optical properties significantly. The charge transport properties, the M-I transition, etc., are strongly governed by both microscopic and macroscopic level structural disorder, doping-induced disorder, etc.

63

Handbook of Polymers in Electronics (2) In general, the effective conjugation length, interchain interactions and morphology are the important parameters that influence the physical properties, disorder-induced localisation, charge transport mechanism, etc. (3) In doped conducting polymers, both localised and delocalised states coexist as an interpenetrating network. In oriented materials, the electronic states are delocalised along the chain direction, as a result the conductivity, the carrier mobility, etc., are higher with respect to unoriented systems. In unoriented globular or granular conducting polymer systems, the localised states determine the charge transport properties. (4) The molecular structure, the doping level, interchain interaction, the extent of disorder, etc., determine the stability of solitons, polarons, bipolarons, free carriers, etc., in doped conducting polymers. (5) A metallic state has been observed in high quality samples of doped (CH)x, PPV, PANI, PPy, PMeT and PEDOT. The experimental evidence indicates the following: finite conductivity at mK temperatures, linear temperature dependence of thermoelectric power, linear term in specific heat, temperature independent Pauli susceptibility, quantum corrections (weak localisation and e-e interaction) to MC, metallic reflectance and free carrier absorption in the infrared. (6) The metallic positive TCR has been observed from 300 to 1.5 K in ClO4-doped (CH)x. In PANI-AMPSA (CSA) the positive TCR is observed at temperatures greater than 70 (150) K. In several metallic conducting polymer samples a positive TCR has been observed at temperature below 20 K. These features show the intrinsic metallic nature of doped conducting polymers. (7) The √T dependence of conductivity, at low temperatures, in metallic conducting polymers indicates that the e-e interaction contribution is significant. The universal scaling behaviour of the MC confirms the dominant role of the e-e interaction contribution. (8) In oriented metallic systems, the anisotropic MC due to the interplay of WL and e-e interaction contributions (i.e., the sign of MC is positive (negative) when the field is perpendicular (parallel) to the chain axis) can be used to probe the misaligned chains. (9) The behaviour of σ(T) and MC is nearly identical in both parallel and perpendicular directions to the chain axis in highly oriented metallic (CH)x and PPV. This suggests that the charge transport mechanism is nearly identical in both parallel and perpendicular directions to chain axis, and the system behaves more like an anisotropic 3D system.

64

Electrical Properties of Doped Conjugated Polymers (10) The metallic, critical and insulating regimes can be identified from the W versus T plots. The positive, temperature-independent and negative temperature coefficients of W(T) corresponds to metallic, critical and insulating regimes, respectively. (11) In the critical and insulating regimes, the resistivity ratio and positive MR increases as the extent of disorder increases. The field-induced transitions, from metallic to critical and from critical to insulating regimes, show that the mobility edge and Fermi level are situated rather close together. Hence, due to interchain transport and disorder, conducting polymers are at the M-I boundary. (12) From the hopping contribution to the S(T) and the Curie term in χ(T), a semiquantitative level of information about the extent of disorder can be obtained.

References 1.

Handbook of Organic Conductive Molecules and Polymers, Volumes 1-4, Ed., H.S. Nalwa, Wiley, New York, NY, USA, 1997.

2.

A. Graja, Low-Dimensional Organic Conductors, World Scientific, Singapore, 1992.

3.

Solid State Properties of Fullerene, Eds., H. Ehrenriech and F. Spaepan, Academic, Boston, MA, USA, 1994.

4.

K. Kaneto, K. Yoshino and Y. Inushi in Electronic Properties of Inorganic QuasiOne-Dimensional Materials, Volume II, Ed., P. Monceau, Dordrecht, Netherlands, 1985, 69.

5.

H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang and A.J. Heeger, Journal of the Chemical Society, Chemical Communications, 578 (1977).

6.

Handbook of Conducting Polymers, Volumes 1 and 2, Ed., T.A. Skotheim, Marcel Dekker, New York, NY, USA, 1986.

7.

For a review see A.J. Heeger, S. Kivelson, J.R. Schrieffer and W.P. Su, Reviews in Modern Physics, 1988, 60, 3-4, 781.

8.

Conjugated Conducting Polymers, Springer Series in Solid State Sciences, Volume 102, Ed., H.G. Kiess, Springer, Berlin, Germany, 1992.

9.

Handbook of Conducting Polymers, 2nd Edition, Eds., T.A. Skotheim, R.L. Elsenbaumer and J.R. Reynolds, Marcel Dekker, New York, NY, USA, 1998. 65

Handbook of Polymers in Electronics 10. S. Roth and H. Bleier, Advanced Physics, 1987, 36, 5-8, 385. 11. J. Tsukamoto, Advanced Physics, 1992, 41, 4-6, 509. 12. T. Ohnishi, T. Noguchi, T. Nakano, M. Hirooka and I. Murase, Synthetic Metals, 1991, 41-43, 1-2, 309. 13. T. Hagiwara, M. Hirasaka, K. Sato and M. Yamaura, Synthetic Metals, 1990, 36, 1-3, 241. 14. Y. Cao, P. Smith and A.J. Heeger, Synthetic Metals, 1992, 48, 1-3, 91. 15. R. Menon, Y. Cao, D. Moses and A.J. Heeger, Physical Review B, 1993, 47, 4, 1758. 16. R. Menon, C.O. Yoon, D. Moses, A.J. Heeger and Y. Cao, Physical Review B, 1993, 48, 24, 17685. 17. R.D. McCullough and R.D. Lowe, Journal of Organic Chemistry, 1993, 70, 3-4, 904. 18. R. Menon in Handbook of Organic Conductive Molecules and Polymers, Volume 4, Ed., H.S. Nalwa, Wiley, New York, NY, USA, 1997, 47. 19. Primary Photoexcitations in Conjugated Polymers, Ed., N.S. Sariciftci, World Scientific, Singapore, 1997. 20. R. Menon, C.O. Yoon, D. Moses and A.J. Heeger in Handbook of Conducting Polymers, 2nd Edition, Eds., T.A. Skotheim, R.L. Elsenbaumer and J.R. Reynolds, Marcel Dekker, New York, NY, USA, 1998, 27. 21. Y.W. Park, E.S. Choi and D.S. Suh, Synthetic Metals, 1998, 96, 1-3, 81. 22. S. Stafstrom, Physical Review B, 1995, 51, 7, 4137. 23. P.W. Anderson, Physical Review, 1958, 109, 5, 1492. 24. N.F. Mott, Metal-Insulator Transition, 2nd Edition, Taylor & Francis, London, UK, 1990. 25. P.A. Lee and T.V. Ramakrishnan, Reviews of Modern Physics, 1985, 57, 1-2, 287. 26. M. Ahlskog and R. Menon, Journal of Physics: Condensed Matter, 1998, 10, 32, 31-33, 7171.

66

Electrical Properties of Doped Conjugated Polymers 27. L. Pietronero, Synthetic Metals, 1983, 8, 1-4, 225. 28. S. Kievelson and A.J. Heeger, Synthetic Metals, 1989, 22, 1, 371. 29. A.G. Zabrodskii and K.N. Zeninova, Zhurnal Technicheskoi Fiziki, 1984, 86, 727; [Sov. Phys. JETP, 1984, 59, 425.] 30. M. Ahlskog, R. Menon and A.J. Heeger, Journal of Physics: Condensed Matter, 1997, 9, 20-22, 4145. 31. P. Dai, Y. Zhang and M.P. Sarachik, Physical Review B, 1992, 46, 11, 6724. 32. R. Menon, K. Vakiparta, Y. Cao and D. Moses, Physical Review B, 1994, 49, 23, 16162. 33. H. Kaneko and T. Ishiguro, Synthetic Metals, 1994, 65, 1-3, 141. 34. M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi and T. Ohnishi, Physical Review B, 1996, 53, 23, 15529. 35. M. Ahlskog, R. Menon, A.J. Heeger, T. Noguchi and T. Ohnishi, Physical Review B, 1997, 55, 11, 6777. 36. C.O. Yoon, R. Menon, D. Moses and A.J. Heeger, Physical Review B, 1994, 49, 16, 10851. 37. J.C. Clark, G.G. Ihas, A.J. Rafanello, M.W. Meisel, R. Menon, C.O. Yoon, Y. Cao and A.J. Heeger, Synthetic Metals, 1995, 69, 1-3, 215. 38. P.N. Adams, P. Devasagayam, S.J. Pomfret, L. Abell and A.P. Monkman, Journal of Physics: Condensed Matter, 1998, 10, 37-38, 8293. 39. E.R. Holland, S.J. Pomfret, P.N. Adams and A.P. Monkman, Journal of Physics: Condensed Matter, 1996, 8, 15-17, 2991. 40. S.J. Pomfret, P.N. Adams, N.P. Comfort and A.P. Monkman, Advanced Materials, 1998, 10, 16, 1351. 41. A. Aleshin, R. Kiebooms, R. Menon, F. Wudl and A.J. Heeger, Physical Review B, 1997, 56, 7, 3659. 42. A. Aleshin, R. Kiebooms, R. Menon and A.J. Heeger, Synthetic Metals, 1997, 90, 1-3, 61.

67

Handbook of Polymers in Electronics 43. S. Masubuchi, Fukuhara and S. Kazama, Synthetic Metals, 1997, 84, 1-3, 601. 44. T. Fukuhara, S. Masubuchi and S. Kazama, Synthetic Metals, 1998, 92, 1-3, 229. 45. Y. Nogami, H. Kaneko, H. Ito, T. Ishiguro, T. Sasaki, N. Toyota, A. Takahashi, and J. Tsukamoto, Physical Review B, 1991, 43, 14, 11829. 46. S. Bogdanovich, P. Dai, M.P. Sarachik and V. Dobrosavljevic, Physical Review Letters, 1995, 74, 13-14, 2543. 47. A.B. Kaiser, Physical Review B, 1989, 40, 5, 2806. 48. C.O. Yoon, R. Menon, D. Moses, A.J. Heeger and Y. Cao, Physical Review B, 1993, 48, 19, 14080. 49. E.B. Park, Y.S. Yoo, J.Y. Park, Y.W. Park, K. Akagi and H. Shirakawa, Synthetic Metals, 1995, 69, 1-3, 61. 50. N.S. Sariciftci, A.J. Heeger and Y. Cao, Physical Review B, 1994, 49, 9, 5988. 51. K. Mizoguchi and S. Kuroda in Handbook of Organic Conductive Molecules and Polymers, Volume 3, Ed., H.S. Nalwa, Wiley, New York, USA, 1997. 52. D. Moses, A. Denenstein, A. Pron, A.J. Heeger and A.G. MacDiarmid, Solid State Commun., 1980, 36, 219. 53. L. Piraux, E. Ducarme, J.P. Issi, D. Begin and D. Billaud, Synthetic Metals, 1991, 41-43, 1-2, 129. 54. T.H. Gilani and T. Ishiguro, Journal of the Physics Society of Japan, 1997, 66, 3, 727. 55. R. Menon, C.O. Yoon, D. Moses, Y. Cao and A.J. Heeger, Synthetic Metals, 1995, 69, 1-3, 329. 56. C.O. Yoon, R. Menon, D. Moses, A.J. Heeger, Y. Cao, T.A. Chen, X. Wu and R.D. Reike, Synthetic Metals, 1995, 75, 1-3, 229.

68

3

Non Linear Optical Properties of Polymers for Electronics T. Kaino

3.1 Introduction The upcoming information technology (IT) era of the 21st century requires processing of large amounts of information accurately at fast speed. Signal processing by optical technology will be a key for that requirement because high-quality and high-speed digital signals can be processed effectively using optical systems. Optical technologies are not only for long haul signal transmission and processing applications, but also for short distance mutual communication applications, like local area networks. Within a couple of years, individual subscribers will receive information transmitted via optical fibre cable, moveable terminal unit, satellite, and cable network. To process such information, high-speed optical switching devices and parallel signal processing devices are required. So, there is a growing interest in the transmission and processing of digital information using non linear optical (NLO) devices. In these optical systems, optical interconnection and optical device installation in an integrated substrate is important to overcome bottlenecks. For practical optical signal processing applications, optical devices should be gathered together in an integrated substrate with optical waveguide structure to control optical signals with low input power. Among the many types of NLO materials, NLO polymers are the most attractive because they may be used to fabricate optical waveguides by standard photoprocesses and it is possible to enhance optical non linearities by synthesising a variety of NLO chromophores (dyes) and attaching them to a polymer backbone. NLO polymer devices can be fabricated on planar passive waveguides, such as glass or polymer waveguides, since these can be used to process optical signals with a high data transmission rate. NLO polymer waveguides offer the potential to create highly complex integrated optical devices and optical interconnections on a planar polymer substrate because their optical properties can be tailored by using different types of passive materials and NLO active materials [1]. The other advantage of using polymers is that they are low in cost and it is possible to fabricate monolithically integrated optical circuits where all the optical devices are combined in one step. The potential of the combination of a signal processing function (active function) using NLO polymers with a signal transmission function (passive function) of transparent

69

Handbook of Polymers in Electronics polymers is the most appealing prospect for waveguides made from polymers. For example, a silicon wafer area network using polymer-integrated optics has already been described [2]. This device is based on a combination of an active reconfiguration function with a passive transmission function. In this chapter, the development of NLO polymers with thin film or channel waveguide structures directed toward practical optical devices will be discussed. All optical signal processing polymers and electro-optical polymers are presented. Hybrid polymer optical devices for future applications will also be presented.

3.2 NLO Polymer Issues for Device Applications The chemical and physical properties of polymers, the type of optically active materials (i.e., chromophores or dyes) and the polymer device fabrication processes used decide the applicability, advantages and limitations for creating practical optical devices using NLO polymers. If NLO polymers are used as waveguide device materials, there exists a trade-off between their optical non linearity, the optical transmission length and their processibility [3]. Since a high concentration of NLO dyes in a polymer backbone are needed to achieve a large NLO function, optical loss will increase due to π-π* transitional absorption of the dyes. Hence, the reduction of transmission loss of the dye-functionalised polymer must be balanced against the enhancement of optical non linearity. Processibility of the NLO polymer will also be influenced by the concentration of the dye, and a highly NLO dye-functionalised polymer usually becomes difficult to process. Spin coating or casting techniques on substrates such as silicon, glass and transparent polymer are typically used to create polymer thin films for waveguides. The inherent chemical and physical properties of the matrix polymer decide the fabrication process. To fabricate polymer channel waveguides with optical non linearity, the photolithography technique is a basic method to define device patterns in the polymer. After patterning, reactive ion etching (RIE) is an excellent example of waveguide fabrication technology that can be compared to the solvent etching process. Waveguides with buried structure created via these techniques comprise of two types of compositions: (1) an etched groove backfilled with a NLO polymer, and (2) an NLO ridge waveguide surrounded by a lower refractive index cladding polymer. The problem of trying to reduce the waveguide fabrication cost is a highly complicated process, because many steps are required, including a mask fabrication process. To solve the problem, direct patterning of the waveguides by using electron beam or laser beam has already been investigated [4]. The embossing process is also used to fabricate polymer waveguides through an inexpensive moulding technique [5].

70

Non Linear Optical Properties of Polymers for Electronics High transparency with excellent processibility is needed for NLO dye-functionalised polymers for waveguide applications, though they suffer from high optical loss in the near-IR region. This is because NLO chromophores usually have π-π* transition absorption at visible wavelength which influences the near-IR absorption of the waveguides. When discussing the transparency of a polymer, not only the absorption of dyes should be considered, but also the vibrational higher harmonics of the polymer IR absorption [6]. Almost all the passive polymer optical waveguides developed so far have an optical loss of higher than 0.5 dB/cm in this region (except for deuterated and/or fluorinated polymer waveguides) [7]. The deuteration and fluorination of the hydrogen in the polymer is very effective for reducing the vibrational higher harmonic absorption loss of the polymer in the visible to near-IR region [8]. Transparency in the near-IR region rather than in the visible region is sometimes important because wavelengths of 1.3 and 1.55 μm are used in optical telecommunication systems. Thus the reduction of the loss in the near-IR is a critical issue for the telecommunication applications of the NLO polymer waveguide devices. The stability of the NLO polymer waveguides in high-temperature conditions is also a main issue for practical applications. As will be discussed in Section 3.4, the alignment of the dye in the polymer matrix which induces the second-order optical non linearity will be lost in high-temperature conditions. Reliability, reproducibility, acceptable cost performance, and compatibility with other optical systems are also important factors for the development of a NLO polymer waveguide device.

3.3 Properties of Third-Order NLO Polymers 3.3.1 Background of Third-Order NLO Polymer Research Liquid crystal devices and thermo-optical polymer devices are already being used as switching components with a millisecond response time. Practical optical switching devices with nanosecond to picosecond response times have not been developed so far. For pico- to subpicosecond all-optical signal processing systems, the development of highly efficient thirdorder NLO materials with processibility is required. Delocalised π-electron conjugated polymers are of great interest because of their potentially fast response time: faster than tens of femtoseconds. In view of practical use, polymers are expected to overcome the disadvantages of molecular crystals in mechanical properties and processibility. Various kinds of π-electron conjugated polymers have been energetically studied to achieve higher third-order NLO efficiency because many π-conjugated structures are possible and hence absorption wavelengths can be controlled to achieve high efficiency via resonant effect. Highly efficient third-order NLO polymers can be applied to many optical devices such as thin film and fibre waveguides. In particular, thin films with large optical non linearities

71

Handbook of Polymers in Electronics have many useful applications in integrated optics. These include optical switching and optical signal processing that show applied optical field intensity dependent refractive indices. Among the most studied films are polyacetylene and polydiacetylene (PDA), which have been reported to show third-order NLO susceptibilities, χ(3), of the order of 10-10 to 10-9 esu [9, 10]. The non linear electronic polarisation of π-electron conjugated polymers originates from mesomeric effects that depend on the size of the π-conjugated systems. The length dependence of χ(3) for polyenes and polyynes has already been discussed and it has been revealed that the hyperpolarisabilities increase rapidly with chain length up to 10 repeating units and approach an asymptotic value after about 15 repeating units [9]. Although large χ(3) values have been observed in the crystalline states or uniaxially oriented films of these polymers, the high crystallinity or uniaxial domain results in large light scattering losses. These polymers are usually difficult to process because they are rarely fusible and are insoluble in almost all solvents. For example, one of the representative πelectron conjugated polymers, poly [bis(p-toluenesulfonate) of 2,4-hexadiyne-1,6-diol] (PTSPDA), shows a large χ(3) of around 10-9 esu [10]. However, it has problems in processibility and stability. High crystallinity and low processibility prevent its use in optical waveguide devices. Therefore, other types of amorphous π-conjugated polymers with large χ(3) and good processibility are needed. Jenekhe and co-workers have reported several novel main chain χ(3) polymers with aromatic Schiffs base structure [11]. Spangler and co-workers have also reported several main chain χ(3) polymers with vinyl structures containing πelectron conjugations in their backbones [12]. By considering this previous research, several amorphous or low crystallinity conjugated polymeric NLO materials are presented in the following sections.

3.3.2 Poly(arylenevinylene), PAV PAV electrically conductive, tough, good optical quality, thin films with good transparency are easily obtained from the precursor polymers. PAVs exist with variations in chemical structure: PPV, PTV and poly (2,5-dimethoxy-p-phenylene vinylene) (MOPPV) [13, 14]. PAVs are of special interest as electroluminescent polymers because, by changing their chemical structure, the luminescent wavelength can be dramatically controlled from green to red [15]. NLO properties are also an interesting feature of the PAVs because their precursor polymers are soluble in organic solvents or water. Hence, processibility for fabricating thin films by spin coating can be achieved without any modifications of the conjugated length. Good optical quality thin films of PAV have been prepared through a new precursor polymer that is soluble in organic solvents [16]. For example, PTV thin film can be fabricated as follows. Polymerisation of a sulfonium monomer, 2,5-thienylene bis(methylene-dimethyl-sulfonium chloride), is carried out in a methanol-water mixture

72

Non Linear Optical Properties of Polymers for Electronics at –2 °C by adding a methanol solution of tetramethylammonium hydroxide. The reaction is quenched by the addition of hydrochloric acid. A yellow precipitate (precursor polymer) appears as the solution is warmed to room temperature. A precursor polymer thin film is obtained by spin coating of the dichloromethane solution of the precursor polymer onto a substrate under inert atmosphere to prevent oxidation with air. The film is heated at 200-250 °C in a vacuum of 10-2 Torr for 5 hours, to give a tough, flexible PTV film. The resulting PTV thin film is chemically stable in air below 10 °C. The thermal conversion process of PAV precursor polymers to fully π-conjugated final structure is shown in Figure 3.1, where A is an arylene unit and X is a halogen. The conversion proceeds from a non π-conjugated precursor polymer to a fully converted polymer via partially converted copolymers. At the initial conversion level, short linear sequences are formed. These copolymers have both π-conjugated and non π-conjugated sequences for various conversion levels. At the higher partial conversion level, the short sequences have developed into long rigid sequences. At the final conversion level, the development of the long rigid π-conjugated sequences is completed, and even the highly bent non π-conjugated parts between the rigid π-conjugated sequences are converted. Controlling temperature and heating time can vary the π-conjugation lengths in PAV. The conversion dependence of the MOPPV χ(3) spectra superimposed onto that of the absorption spectra is shown in Figure 3.2 [17]. As the conversion proceeds from A to F, an increase in absorption intensities and red shift of the absorption maxima is observed. Conversely, the χ(3) peaks are fixed over all conversion levels. The wavelengths of the χ(3) peaks are 30 nm longer than the absorption peak wavelength of the fully converted film. The steep enhancement of χ(3) from D to F associated with the development of π-conjugated sequences is notable. The magnitude of the χ(3) values of the fully converted MOPPV film (F) reaches 1.6 x 10-10 esu. The χ(3) increases due to the three-photon resonant effect at the final conversion level where the long rigid π-conjugated sequences are formed. The position of the absorption peak is decided by the maximum sequence present, but the χ(3) peak always appears at the same wavelength. The χ(3) peak and the absorption peak match each other only when the fully converted state is obtained.

Figure 3.1 Thermal conversion process of PAV precursor polymers to fully π-conjugated final structure

73

Handbook of Polymers in Electronics

Figure 3.2 Conversion dependence of the MOPPV χ (3) spectra superimposed on their absorption spectra (Reproduced with permission from [17], copyright 1991, Elsevier Science Publishers)

The optical absorption maxima are revealed to be at 520, 500 and 420 nm for PTV, MOPPV and PPV films, respectively. X-ray diffraction patterns indicate that the PAV films have low crystallinity. The χ(3) values of these PAV films evaluated by thirdharmonic generation (THG) measurement at 1.85 μm are found to be 5.85 x 10-11 esu, 3.2 x 10 -11 esu, and 7.8 x 10-12 esu for PTV, MOPPV and PPV films , respectively. These values are non resonant χ(3) values and are almost the same as that reported for a processible polydiacetylene (n-butoxycarbonylmethyl urethane) (n-BCMU-PDA, n = 4,3) thin film [18]. The incident light wavelength dependence of χ(3) revealed that as the wavelength decreases, χ(3) increases. This was caused by three-photon resonant enhancement effects due to the overlap of the third harmonic wavelength with a polymer absorption band. The maximum χ(3) of PPV film, 1.4 x 10-10 esu at 1.475 μm, is about one order higher than that of non resonant values. The χ(3) values are almost the same for PPV, MOPPV and PTV in the absorption wavelength region. Thus, dispersion measurements are required in the absorption energy range to clarify the nature of the resonance and to determine the maximum resonantly enhanced χ(3). The evaluated three-photon resonant values of χ(3) of the PAV films are almost of the same order of magnitude as that of crystalline PDA films. This result indicates that PAV thin films will be effective for NLO device application, due to their ease in processing films as well as possible good optical quality.

74

Non Linear Optical Properties of Polymers for Electronics

3.3.3 n-BCMU-PDA PDAs have been widely investigated as potential materials for all-optical switching devices. For these purposes a large non linear refractive index (n2) and processibility are two of the most important features. Soluble PDA, such as 4BCMU-PDA and 3BCMU-PDA are of special interest for the application of polymeric waveguides with high optical non linearity. The χ(3) values of amorphous and crystalline films of 4BCMU-PDA have been compared using THG measurements. The orientation effect due to the crystallisation of the polymer is revealed to be approximately a factor of five, i.e., the χ(3) values of the spin coated film is about 10-11 esu at 1.06 μm and that of the crystalline film is 5 x 10-11 esu [19]. Using a degenerated four wave mixing (DFWM) technique with backward interaction geometry, χ(3) values of the red form and the yellow form of 4 μm thick 4BCMU-PDA film are revealed to be 4 x 10-10 esu and 2.5 x 10-11 esu, respectively. The difference of these values is explained to be due to a conformational transition from an extended rigid coil in the red form to a random coil in the yellow form [20]. Using an inverted rib design, single mode channel waveguides of 4BCMU-PDA and 3BCMUPDA spun films have been fabricated. Guided light is predominantly in the PDA layer with the glass channels, which provides lateral confinement [21]. Single-mode planar waveguides made from these films have propagation losses as low as 1 dB/cm. Using this technique, directional couplers and grating couplers are fabricated, which are potentially attractive for applications in all-optical signal processing. At a wavelength near the two-photon absorbance, the large imaginary component of n2 dominates the response of the directional coupler [22]. Assanto and co-workers fabricated an efficient grating guided-wave coupler using 4BCMUPDA slab waveguides, and a coupling efficiency of 45% at 1.064 μm was obtained [23]. Using this grating coupler, waveguide losses measured by the total waveguide throughput over given distances (longer than 1 cm) were revealed to be larger than 5 dB/cm for 0.65 μm thick film at 1.064 μm. This large loss may be due to non uniformities in film thickness and refractive index. Minimisation of the loss is needed for actual device applications. Strong self-lensing effects have been observed through thin film slab waveguides formed with 4BCMU-PDA using a 1.0 μm mode-locked Nd:YAG laser which provides single 35 ps pulses. The estimated value for n2 is approximately 10-13 cm2/W [24]. Using a pump and probe technique for measuring n2 of thin films of spun 4BCMU-PDA at 630 nm, the real part of n2 is reported to be negative with a magnitude of around 6 x 10-12 esu. If the alignment of the polymer chain is attained, the value is expected to be almost the same as PTS-PDA. The mechanism for the non linear index is thought to be phase space filling by excitons [25]. Channel waveguides of PTS-PDA single crystal on a Si/SiO2 substrate have been fabricated by a photolithography technique [26]. The value of n2 in the PTS-PDA chain axis direction is found to be 3 x 10-11 cm2/W at 1.06 μm. There will be a considerable contribution from two-photon resonance in this system.

75

Handbook of Polymers in Electronics

3.3.4 PT Among the class of π-conjugated polymers, PT is of considerable interest because a sulfurcontaining heterocyclic backbone increases the rigidity of the polymer and hence the πelectron localisation. Soluble PT is synthesised by adding alkyloxy or alkyl groups into a thiophene ring, such as 3-alkyloxymethoxy thiophene or 3-methyl-4´octyl-2,2´-bithiophene5,5´-diyl (3MOT). THG measured as a function of wavelength for poly(3MOT) is revealed to be 4.4 x 10-11 esu at 2.4 μm, i.e., in the preresonant region [27]. The NLO properties of poly(3-butylthiophene), which is soluble in common organic solvents, were investigated using DFWM at 1.06 μm. The response time was less than 70 ps, limited by the laser pulse width. In the case of the solution of the polymer in chloroform, the position of a threephoton resonance was revealed to depend on the concentration of the polymer [28]. Prasad and co-workers studied poly(3-dodecylthiophene) using a 60 fs pulse DFWM and obtained a resonant χ(3) of 5.5 x 10-11 esu at 660 nm, which is about six times smaller than that obtained with 400 fs pulses [29]. The effective χ(3) is less when the excitation pulse duration is shorter than the relaxation time of the excitation which gives rise to the non linear response. They concluded that the short non linear response obtained with pulses less than 200 fs is derived from initial photogenerated excitons. Jenekhe and co-workers reported an organic polymer superlattice using a soluble conjugated polythiophene derivative [30]. From DFWM measurements in a phase conjugate geometry for the sulfuric acid solution of the polymer, a χ(3) of 2.7 x 10-7 esu at 532 nm was estimated. For the experiment 250 mJ pulses with 25 ps full width at half-maxima were used. The response time of the polymer is said to be faster than the normal instrument response of 25 ps.

3.3.5 Processible π-Conjugated Polymers Apart from typical π-electron conjugated polymers such as PAV, n-BCMU-PDA and PT, processible main-chain polymers with different types of π-conjugation structure have been investigated. Three types of main-chain polymers will be discussed: symmetrically substituted benzylidene aniline with chloride (SBAC) polymer, azo-dye polyester, and heteroaromatic polymers. The main chains of these polymers possess donor substituents at both ends of their π-electron conjugated unit [31], a donor-acceptor structure in their π-electron conjugated unit [32], and a π-deficient aromatic ring that is alternately connected with a π-rich aromatic ring [33], respectively. They are easy-to-fabricate amorphous solids made by conventional spin coating or casting methods from solution. By using THG measurements, the wavelength dependence of χ(3) for these main chain polymers was investigated and the resonant enhancement of χ(3) revealed. Waveguiding properties of the polymers were also investigated.

76

Non Linear Optical Properties of Polymers for Electronics SBAC polymer is a novel symmetrical molecule with a large χ(3). SBAC is a benzylidene aniline derivative in which donor groups with nitrogen atoms symmetrically substitute both ends of the π-conjugated unit. This novel molecule was designed to control the transition dipole moment, P, between the excited states and enhance the molecular hyperpolarisability, γ, as shown in Figure 3.3. The transition dipole moment between the 1Bu and 2Ag excited states is enhanced by symmetrical donor substitution of the relatively short π-conjugation of benzylidene aniline [34]. This idea is different from donor-acceptor substitution based on a two-level model, such as the azo-dye polyesters. In this donoracceptor system, the χ(3) of short π-conjugation molecules depends on dipole moment change between the ground state (G) and the 1Bu excited state due to the intramolecular charge-transfer effect from donor to acceptor.

Figure 3.3 Transition dipole moment between the 1Bu and 2Ag excited states of SBAC

A strong χ(3) enhancement of the vacuum-deposited SBAC polymer thin films was observed when the incident wavelength is three times the wavelength of the lowest energy peak in the absorption spectra. This three-photon resonant χ(3) reached approximately 2 x 10-10 esu. Even in the off-resonant regions, the χ(3) was 4 x 10-11 esu. These values are comparable to those of non crystalline or non oriented π-conjugated polymers. The control of transition dipole moments by donor/acceptor substitution or heteroatom introduction into π-conjugated systems is an effective way to enhance the χ(3) of the materials. However, it is difficult to obtain thin films of the vacuum-deposited SBAC polymer with good optical quality. Therefore, SBAC polymers were synthesised where SBAC molecules are used as a part of the polymer main chain [35]. The synthetic scheme for the SBAC polymers is shown in Figure 3.4. SBAC polymers are easily obtained by polycondensation reaction from terephthalic aldehyde and N, N-dibutylamino derivative of SBAC. As a result, the SBAC molecule was combined with an alkyl chain or with

77

Handbook of Polymers in Electronics

Figure 3.4 SBAC polymer synthetic scheme

an ester chain. Since the SBAC polymers are soluble in organic solvents, good optical quality thin films were obtained. Since SBAC polymers have an absorption in the third-harmonic wavelength region (l = 0.50 to 0.60 μm) of the laser beam used here, the χ(3) values in the resonant region were evaluated. All polymers exhibit χ(3) values of 10-10-10-11 esu in this resonant region. The wavelength dependence of χ(3) of polyester SBAC(II-c) thin film is shown in Figure 3.5. The maximum χ(3) value obtained is about 7 x 10-11 esu at 1.57 μm.

Figure 3.5 χ(3) wavelength dependence of SBAC polymer (II-c) 78

Non Linear Optical Properties of Polymers for Electronics In an azo-dye-introduced main-chain polymer, the dye possesses an electron donor and an acceptor along the backbone (electric dipoles are arranged in a head-to-tail configuration) [36]. A mono-azo dye may easily be incorporated into the main chain of a polymer system. The synthetic scheme of such a polymer is shown in Figure 3.6. The polymer is polyester with N, N-diethylaminonitroazobenzene introduced into its backbone. The polymer is soluble in common organic solvents and fabricated into a thin film using a spinning technique. The χ (3) spectrum of the polymer is shown in Figure 3.7 along with its absorption spectrum. The largest χ (3) is in the resonant region with a value of 2.0 x 10 -11 esu. For the polyimide (PI) system with π-electron conjugation units in the backbone without a donor/acceptor, χ (3) was of the order of 10-11 esu even for the four benzene ring π-electron conjugation system [32]. The value thus obtained for the azo-dye-introduced main-chain polymer is very high even though the dye used is a short π-electron conjugation system (i.e., a two benzene ring mono-azo dye). This is because the main-chain polymer has a very high dye density, as recognised from its chemical structure. An intramolecular charge transfer derived from a substituted donor and acceptor also contributes to the increase of χ (3). A value of χ (3) larger than 10 -10 esu was achieved for an intramolecular charge transfer dye-attached three benzene ring polymer system [37]. Thus, it should be emphasised that a large χ (3) is possible not only in π-electron conjugated polymers, but also in short π-electron conjugated systems like an azodye-introduced main-chain polymer by modification of molecular structure. These main-chain polymers show promise for the development of processible materials with a large χ (3).

Figure 3.6 Azo-dye-functionalised main-chain polymer synthetic scheme

79

Handbook of Polymers in Electronics

Figure 3.7 χ(3) wavelength dependence of azo-dye-functionalised main-chain polymer

A different type of processible π-conjugated polymer is a heteroaromatic polymer with π-rich (electron donating) heteroaromatic rings and π-deficient (electron accepting) heteroaromatic rings which are alternately combined to form the polymer [33]. These heteroaromatic polymers are periodically sequenced copolymers with well-defined polar monomer units, and include polythiophenediyl-pyridinediyl (PTPY), polythiophenediylbipyridinediyl (PTBPY) and polyphenylenediyl-pyridinediyl (PPPY). Their chemical structures are shown in Figure 3.8. The most important feature of these polymers is the intramolecular charge transfer from the π-rich rings to the π-deficient rings [38]. The charge transfer between heteroaromatic rings in the π-conjugated sequences is thought to influence certain excited states that contribute to the χ(3) enhancement. PTPY, in

Figure 3.8 Molecular structure of heteroaromatic polymers

80

Non Linear Optical Properties of Polymers for Electronics particular, shows a red shift of the absorption maximum compared with poly(thiophene2,5-diyl) and poly(pyridine-2,5-diyl), which suggests the occurrence of an intramolecular charge transfer from thiophene rings to pyridine rings. The orbital distortion through heteroatoms in the polymer breaks electron-hole symmetry to contribute to the χ(3) enhancement. The existence of different heteroaromatic rings that induce the charge transfer interaction on the π-conjugated sequences was thought to influence the properties of certain excited states, such as 1Bu and 2Ag, which contribute to the χ(3) enhancement. These heteroaromatic polymers are synthesised by nickel(0) complex polymerisation as reported elsewhere [39]. The synthetic scheme of PTPY is shown in Figure 3.9. Dissolving the polymers in formic acids and spin coating on a glass substrate allowed amorphous polymer films to be prepared. Typical film thickness was 0.01-0.3 μm. THG measurements of the polymer films were performed between 1.25 and 2.1 μm. The χ(3) value was four times larger than the resonance value of amorphous 3BCMU-PDA and was comparable to those for PPV and the promising π-conjugated polymers for optical switching application. The absorption and χ(3) spectra obtained for PTPY are shown in Figure 3.10. A steep enhancement of χ(3) was observed at the fundamental wavelengths close to three times the wavelength of the absorption maximum. In the off-resonance region, 3hv/Eg, where hv is the photon energy of the fundamental wavelength and Eg is the optical band gap energy, the χ(3) values for PTPY and PTBPY are larger than those of PPPY, which has no π-rich thiophene ring. The maximum χ(3) value of PTPY is 2.8 x 10-10 esu at 1.525 μm fundamental wavelength (hn= 0.813 eV). Poly(1,4-di(2-thienyl)benzene), PDTB, another type of heteroaromatic polymer with no π-deficient pyridine ring, was reported to have a χ(3) value of 1.1 x 10-11 esu in the three-photon resonant region and a value of 0.2 x 10-11 esu at the non resonant wavelength [39]. This result, where χ(3) values of PDTB are one order smaller than those of PTPY, suggests that the orbital distortion due to the intramolecular charge transfer between the π-rich thiophene ring and the π-deficient pyridine ring enhances the non resonant χ(3). The resonance χ(3) value of PTPY is larger than that of PTBPY. Usually, the χ(3) value of π-electron conjugated polymers depends on the π-conjugation length, and elongation of the π-conjugation

Figure 3.9 Synthetic scheme for PTPY

81

Handbook of Polymers in Electronics

Figure 3.10 χ(3) wavelength dependence of heteroaromatic polymers

sequence reduces the optical band gap energy. The χ(3) value has been reported to be proportional to the minus 6th power of the optical band gap energy. Taking into account that Eg = 2.7 eV and 3.0 eV for PTPY and PTBPY, respectively, the χ(3) value for the (PTPY)/(PTBPY) system has been calculated to be 1.9. This value is consistent with the experimental value of 1.8 at the near resonance wavelength corresponding to 3hv/Eg = 0.825. Bipyridine units on the PTBPY polymer chain, which are twisted more strongly than the periodic sequence of the five-membered thiophene ring and the six-membered pyridine ring in PTPY, are considered to shorten the conjugation length and result in a decreased χ(3) value.

3.3.6 Third-Order NLO Polymer Waveguides Among the many types of processible NLO polymers, polyurethane with symmetrical substituted tris-azo dye with fluorinated alkyl units (PSTF for short) was selected as a waveguide material. PSTF is a novel third-order NLO main-chain polyurethane with tris-azo dye incorporated into the main chain with a fluorinated alkyl backbone [40]. This polymer is similar to the SBAC polymer family whereby a symmetrically substituted π-conjugated molecule was incorporated in a main-chain polymer. Azo dye is an effective NLO dye investigated by many researchers and the tris-azo dye is very special for NLO applications. Fluorinated alkyl units are expected to be effective for reducing the polymer

82

Non Linear Optical Properties of Polymers for Electronics

Figure 3.11 Molecular structure of PSTF polymer (Reproduced with permission from [17], copyright 1992, American Institute of Physics)

waveguide loss. The chemical structure of PSTF is shown in Figure 3.11, where R represents fluorinated alkyl units. The refractive index of the NLO polymers should be well controlled when single-mode waveguides are to be fabricated using the NLO polymers. Although third-order NLO polymers generally have higher refractive indices than those of transparent polymers for the cladding layer, the PSTF has a moderate refractive index because of its fluorinated alkyl chain. As a cladding material, UV-cured fluorinated epoxy resin was selected. It has an appropriate refractive index for fabricating PSTF single-mode waveguides [41]. The channel waveguide of the PSTF was fabricated applying standard photolithography and RIE techniques. By using a silicone negativetype photoresist and a mask aligner, the mask pattern was transferred to the photoresist and a ridge waveguide was fabricated through oxygen plasma etching. By applying the UV-cured epoxy-cladding polymer on top of the ridge, a buried PSTF waveguide was fabricated. The far-field pattern of the guided mode at 1.3 μm reveals that it is a quasisinglemode waveguide. The losses of the waveguide at 1.3 and 1.55 μm were slightly large, 3.5 and 4.5 dB/cm, respectively, even though a fluorinated alkyl chain was used as part of the matrix polymer. This was due to the high tri-azo dye content of the PSTF, 44 to 66 mol percent, to the backbone polyurethane units. The non resonant χ(3) of the PSTF polymer film measured by THG was approximately 2 x 10–11 esu at 1.55 μm. The non linear optical effect of the waveguide was also confirmed by detecting self phase modulation (SPM) of the waveguide at this wavelength. From the result, the non linear refractive index, n2, of the polymer was calculated to be 2.8 x 10-14 cm2/W [42]. This value was approximately an order of magnitude smaller than the value obtained by THG measurement. This result was due to the strong two-photon absorption of the polymer at 1.55 μm. With respect to optical devices using NLO materials, an optical Kerr switch that uses the third-order optical non linearity of the material is a simple method to attain ultrafast response optical switching. In this device configuration, switching gate power, Pπ, for the π-phase shift of the signal beam at the signal wavelength λ, in the NLO waveguide with the core area A, and the medium length L, is expressed as: Pπ = 3 λ A/4Ln2

(3.1)

83

Handbook of Polymers in Electronics It is important to switch optical signals with small gate power. To reduce the gate power, the core area should be decreased, or the medium length and n2 of the medium should be increased. The characteristics of all-optical switching materials and the required optical switching power for the materials are shown in Table 3.1. Optical switching using the PSTF waveguide has not been attained so far owing to its strong linear and non linear absorption. This absorption limits the wavelength at which optical switching can be operated. It is thus important to think about the trade-off between linear and non linear absorption and the χ(3) value of polymer waveguides when selecting the wavelength for optical switching.

Table 3.1 Characteristics of all-optical switching materials Sample

n2 (cm2/W)

Loss* (dB/m)

Switching power (W)

Length

SiO2 fibre

3x10-16

2x10-4

1

>km

3

0.47

4 cm

-

1 cm (target)

Chalcogenide fibre PSTF

1x10

-14

3x10

-13

4x10

2

*at 1.55 mm

In the case of chalcogenide glass fibre with an n2 value of 2 x 10-14 cm2/W, a peak switching power of less than 1 W was obtained using a 3.2 metre fibre [43]. To realise the alloptical switching using polymer waveguides with less than 1 W switching power, the polymer waveguide should have an n2 value two orders of magnitude higher than that of the PSTF waveguide or it must have one-order lower absorption loss as well as oneorder higher n2. Much effort should be concentrated in developing NLO polymers with higher optical non linearity with low loss for all-optical switching with allied voltage similar to that of glass optical fibre switches.

3.4. Properties of Second-Order NLO Polymers 3.4.1 Azo-Dye-Functionalised, Poled Polymers for Second-Order Non Linear Optics Azobenzene-dye-, stilbene-dye-, or polyene-dye-functionalised polymers have been investigated as second-order NLO materials (χ(2) materials), showing efficient NLO characteristics with low absorption loss [44]. In this case, the dye in the polymer should be aligned in a specific direction to possess second-order optical non linearity. For that purpose, a high electric field

84

Non Linear Optical Properties of Polymers for Electronics will be applied to the dye-functionalised polymers, i.e., electric field poling is necessary to obtain χ(2) polymers. The non linear electronic polarisation of the dye-functionalised polymers originates from mesomeric effects that usually depend on the size of the π-conjugated systems of the dye. The π-conjugation length dependence of the second-order NLO hyperpolarisability, β, has been discussed by many researchers [45]. It has been revealed that β of the poled polymer increases rapidly with chain length if the molecular planarity of the dye is maintained and there will be no aggregation of dye in the polymer network [46]. These dye-functionalised polymers are easy to process as high-quality thin films for fabricating optical waveguides and a large χ(2) value (~10-7 esu) is easily obtainable. From these points, dye-functionalised polymers are promising χ(2) materials. The polymer investigated here is a polymethylmethacrylate (PMMA) copolymerised with methacrylate esters of a dicyanovinyl-terminated bisazo dye derivative. A nitro-terminated version of the bisazo dye derivative and a typical monoazo dye, Disperse Red 1 (DR1), derivative is also discussed in [47]. These azo dyes are hereafter referred as 3RDCVXY, 3RNO2, and 2RNO2, respectively. The molecular structure of 3RDCVXY is shown in Figure 3.12a.

Figure 3.12 (a) Chemical structure of 3RDCVXY bis-azo dye (b) Chemical structure of deuterated 3RDCVXY polymer

85

Handbook of Polymers in Electronics Each dye molecule contains a dicyanovinyl group or a nitro group as an electron acceptor, and an ethylethoxyamino group as an electron donor. This donor-acceptor charge transfer will greatly contribute to β. The 3RDCVXY and 3RNO2 dye contain three benzene rings connected with azo groups, so their conjugated structure is longer than that of 2RNO2. The glass transition temperatures (Tg) were 135, 100 and 80 °C for 3RDCVXY, 3RNO2 and 2RNO2, respectively. The maximum absorption wavelength of 515 nm (2.41 eV) for 3RDCVXY was longer than those of 3RNO2 and 2RNO2, which were 500 nm (2.48 eV) and 470 nm (2.64 eV), respectively. The methyl substitution of the 3RDCVXY was very effective for increasing the dye content in the copolymer and as a result the dye content was nearly 2 times higher than those of the 3RNO2 and 2RNO2. These polymer films on glass substrates were poled using a parallel electrode poling technique. The poling direction was vertical to the film surface. The χ(2) was determined according to the standard procedure and its wavelength dependence is shown in Figure 3.13. The χ(2) increases as the fundamental wavelength decreases, corresponding to the absorption spectrum. This is caused by the χ(2) enhancement effect of two-photon resonance near the absorption band. It was revealed that the maximum χ(2) of the 3RDCVXY reaches 1.0 x 10-6 esu. This is 3 and 7 times larger than those of the 3RNO2 and a typical inorganic NLO crystal, LiNbO3, respectively.

Figure 3.13 χ(3) wavelength dependence of 2R, 3R, and 3RDCVXY polymers

A thermal ageing test shows that the χ(2) of 3RDCVXY is stable even when kept at 80 °C for more than 6 months. The thermal and temporal stability make 3RDCVXY a viable substitute for the current inorganic electro-optical (EO) materials. The stability of these

86

Non Linear Optical Properties of Polymers for Electronics EO polymers at high temperatures is an important area of NLO polymer research. One way to increase the thermal stability of NLO polymers is a crosslinked polymer system where NLO dye is covalently attached to the polymer network at more than one site. A bifunctional molecule, such as a molecule with an amino group and a (N-ethyl, Nhydroxyethyl)amino group, is one example, which reacts with a trifunctional isocyanurate comonomer. This polymer is reported to be stable at 75 °C for more than three months. Deuteration or fluorination will be needed even for NLO waveguides. It is necessary to fabricate channel waveguides with low loss to make NLO polymer devices such as phase modulators that can be driven with low input power. To this end, a deuterated 3RDCVXY polymer was developed for device application; its chemical structure is shown in Figure 3.12(b) [48]. This polymer is composed of 3RDCVXY-attached PMMA and a transparent PMMA where all the hydrogen atoms are deuterated.

3.4.2 EO Polymers One of the important features of second-order NLO materials for obtaining optically active devices is an EO characteristic. EO polymer waveguide devices such as modulators and couplers are of special interest because response time faster than a nanosecond will be obtained using the device, more than 5 orders higher than conventional liquid crystal optical devices or thermo-optical waveguide switching devices. For such applications, azo-dye-functionalised poled polymers are promising materials and have been investigated [49]. Using an electric-field poling technique, efficient EO property with moderate absorption loss is attained. By using the deuterated 3RDCVXY polymer, a channel waveguide was fabricated. The waveguide fabrication process is shown in Figure 3.14 where standard photolithography and RIE is used. The selection of cladding layer is important to obtain flatness of the waveguide surface where the metal electrode for supplying the driving voltage will be deposited. Through this process, a Mach-Zehnder (MZ) interferometer was fabricated. The EO coefficient (r-coefficient) of the waveguide was 26 pm/V at about 70 MV/m poling voltage, and a half-wave driving voltage of 12 V of the interferometer was obtained [50]. The value is low compared to the LiNbO3 interferometer, whose r-coefficient was approximately 32 pm/V, though it is possible to enhance the value by applying a higher poling electric field to the waveguide or by using a longer waveguide. From a practical viewpoint, the value should be increased to more than 40 pm/V to reduce the driving voltage with transistor-transistor logic (TTL) level. By using the deuterated 3RDCVXY polymer, a vertical coupling MZ interferometer was also fabricated [51]. These azo-dyefunctionalised poled polymers have the potential to be applied to a variety of optical devices due to their processibility and low potential.

87

Handbook of Polymers in Electronics

Figure 3.14 Fabrication method for a core channel waveguide based on 3RDCVXY

3.4.3 Serially-Grafted Polymer Waveguides To realise frequency conversion devices by using a second-order NLO process, the quasiphase-matching (QPM) technique is an attractive method as it is easy to achieve frequency conversion, such as second-harmonic generation (SHG), with high efficiency and to obtain a collimating light beam from the device [52]. Almost all the QPM devices reported to date use inorganic materials and there have been very limited reports on QPM frequency conversion devices using second-order NLO polymers [53]. This is because the latter are difficult to fabricate into a specific structure with precise periodic poling pitches, which is an important factor in realising high efficient frequency conversion [54]. To obtain high efficiency from the QPM devices, a channel waveguide structure is desirable. A serially grafted waveguide fabrication technique, where two types of polymer core are aligned by using photolithography and RIE techniques, is an effective method for fabricating a QPM waveguide. The fabrication process and fabricated polymer QPM waveguide structure are shown in Figure 3.15. Electric poling is necessary to make the polymer film NLO active. For fabrication of the QPM waveguide, deuterated 3RDCVXY (as NLO and active part) and a UV-cured epoxy resin (as a transparent part) are used as a waveguide core. The deuterated 3RDCVXY polymer exhibited a d33 (= 0.5 χ(2)) value of 80 pm/V at 1.3 μm [49]. Changing the fluorination content of the resin can easily control the refractive index of the UV-cured epoxy resin. This controllability made it easy to form a periodically aligned structure composed of a poled 3RDCVXY polymer

88

Non Linear Optical Properties of Polymers for Electronics

Figure 3.15 Schematic of polymeric QPM waveguide fabrication method

and a UV-cured epoxy resin with almost the same refractive index. This refractive index matching between two polymers is important to reduce the Fresnel reflection at the interfaces between two polymer cores. The process shown in Figure 3.15 was used to fabricate the serially grafted polymer waveguide. On top of a silicon substrate, a low refractive index UV-cured epoxy resin was first spin coated as an undercladding layer of the waveguide. Then, after fabricating a thin film layer of the NLO polymer and patterning it by photolithography, a core ridge of NLO polymer was fabricated by an RIE process. After spin coating of the UV-cured epoxy resin on top of the NLO polymer core, the epoxy resin core ridge was fabricated using photolithography and a RIE process. The refractive index difference between the UV-cured epoxy resin and the NLO polymer was controlled to within 0.001. Then, a buried channel waveguide with periodic structure was fabricated by covering these core ridges with a low refractive index UV-cured resin. Although the oxygen plasma etching rates of these two polymers are different, this overcladding layer covered the groove under the NLO polymer core ridge that was formed during the fabrication of the serially grafted waveguide structure. Finally, the NLO polymer in the waveguide was poled by supplying a high electric field by using parallel electrodes. This poling process provided large second-order optical non linearity only to the periodical parts of NLO polymer in the core layer. The periodic length (Λ) in the waveguide was controlled easily by changing the mask pattern for the photoprocess. Using this technique, a QPM frequency conversion polymer waveguide was successfully fabricated. The loss of the NLO polymer and UV-cured resin of this QPM waveguide were 0.67 dB/ cm and 0.53 dB/cm, respectively. The Fresnel reflection loss of the grafted interface was

89

Handbook of Polymers in Electronics

Figure 3.16 Relationship between laser power and SHG efficiency of the QPM waveguide (Reproduced with permission from [57], copyright 1996, American Institute of Physics)

0.005 dB/point [55]. By changing the periodic length, in the 10 to 90 μm range at intervals of 1 μm, the phase-matched wavelength for SHG was controlled. SHG experiments were performed with 5 mm long QPM waveguides using a colour centre laser with 1.48-1.65 μm as fundamental wavelengths. When the fundamental wavelength was 1.586 μm, the SHG intensity was strongest in the waveguide with a periodic length of 32 μm. The relationship between the applied voltage and the SHG intensity at the wavelength is shown in Figure 3.16. The QPM waveguide had an efficiency of 4x10-1%/W/cm2, the largest ever reported for a QPM polymer waveguide [56]. This hybrid waveguide fabrication technique can be applied for an optical sampling device and a compact all-optical switching device. The application of hybrid waveguides will be key in optical signal processing technologies.

3.4.4 Refractive Index Grating Fabrication into Azo-Dye-Functionalised Polymer Waveguides Waveguide grating (an important device in optical communication and signal processing systems) has functions such as input-output coupler, wavelength filter and wavelength division multiplexer. A variety of polymer waveguide gratings have already been proposed [57, 58]. An azo-dye-functionalised polymer is one of the candidates for fabricating

90

Non Linear Optical Properties of Polymers for Electronics gratings using the photochemical reaction of the azo dye in the interface region of two laser beams. This reaction can be explained by following two mechanisms. The first mechanism is the reversible trans-cis-trans isomerisation of the dye where a strong interference beam was irradiated [59, 60]. A surface relief grating can be fabricated through the movement of the azo-dye-functionalised matrix polymer through dye conformational change due to isomerisation. These surface relief gratings based on the isomerisation mechanism were optically and thermally erasable. The second mechanism is irreversible photobleaching of the dye that occurs under higher energy irradiation. In this case, a π-conjugated system in the azo dye is broken and a permanent refractive index change of the polymer occurs [61, 62]. By the two laser beam interference process using 532 nm light from SHG of a continuous wave (CW) Nd:YVO4 laser as a light source, a high output efficiency, 35% or more, was attained using 3RDCVXY polymer thin film, as shown in Figure 3.17 [64]. The efficiency was monitored using a 0.5 mW 633 nm He-Ne laser by measuring the first-order diffracted light power. Using atomic force microscopy, the thickness of the relief structure of the film was evaluated to be about 30 nm. This grating has a very thin relief structure, onetenth or less than those of typical surface relief gratings with the same diffraction efficiency [59, 60]. So, the fabricated structure was not a relief grating but a refractive index grating. From the IR absorption measurement of the 3RDCVXY polymer film before and after laser irradiation, it was revealed that 532 nm laser light could effectively induce the photobleaching of 3RDCVXY. The refractive index changes of the film by photobleaching were measured at 633 nm, 830 nm, 1300 nm and 1550 nm wavelengths.

Figure 3.17 Diffraction efficiency of the grating with 1.6 μm spacing

91

Handbook of Polymers in Electronics The refractive index decreased with increase of irradiation time at each wavelength. The refractive index difference (Dn) of the film before and after 20 minutes irradiation was 0.022, 0.009, 0.005, and 0.005 at 633 nm, 830 nm, 1300 nm, and 1550 nm, respectively. To fabricate a waveguide grating, two simple methods were applied [63]. One was waveguide fabrication by using an etched groove of an epoxy resin substrate shaped by using a dicing saw. An azo-dye-functionalised PMMA solution was spin coated into this groove, then a UV-cured resin was spin coated on top of the waveguide, and a buried waveguide of the azo-dye-functionalised polymer was fabricated. The other method was grating formation on the buried waveguide through photobleaching by using two beam interference light from a 532 nm SHG laser. A periodical refractive index grating structure was fabricated onto this buried channel waveguide. In designing and fabricating the grating, the combination of these simple techniques is an easier method than that of typical fabrication processes such as photolithography and RIE method. PMMA was used as a host polymer of the refractive index buried waveguide core and DR1 was added to functionalise this polymer. PMMA and DR1 were dissolved in chlorobenzene and the solution was filtrated through a 0.2 μm filter. A channel waveguide was fabricated using the above-mentioned method. The thickness of the substrate was approximately 90 μm and the groove size was 41 μm in width and 44 μm in height. PMMA containing 10 wt% DR1 (DR1(10%)/PMMA) solution was spin coated on the groove-shaped substrate. A UV-curable epoxy resin of equal composition to the substrate was spin coated on this DR1(10%)/PMMA buried waveguide layer, about 80 μm in depth. The refractive indices of the waveguide materials measured using a Metricon Model 2010 Prism Coupler at 1.3 μm wavelength were 1.500 and 1.466 for DR1(10%)/ PMMA and the UV resin, respectively. To evaluate the optical propagation loss of the fabricated buried waveguide a cutback method was applied. A schematic of the waveguide endface cut by a dicing saw is shown in Figure 3.18. Light from a 1.3 μm laser diode was coupled to optical fibre, the core/ cladding of which was 50/125 μm in diameter, and the fibre was butt-jointed to the waveguide end face. A Ge photodiode was used as a detector. By varying the waveguide length from 0.7 to 1.6 cm, the loss was evaluated to be 1.3 dB/cm. This loss was practically available for a few centimetres application of the waveguide, with a loss budget of ~3 dB/cm. A refractive index grating was formed in a DR1 (10%)/PMMA thin film on a glass substrate. The light source and its intensity were the same as those of the 3RDCVXY photobleaching reaction. The diffraction efficiency of the grating with 1.6 μm period and saturated after irradiation for 20 minutes reached 6.2%. The refractive index decreased with increase of irradiation time at each wavelength measured. The change in refractive index of the film before and after 20 minutes irradiation was 0.014, 0.006, 0.005, and 0.004 at 633 nm, 830 nm, 1300 nm, and 1550 nm, respectively.

92

Non Linear Optical Properties of Polymers for Electronics

Figure 3.18 A schematic of the end-face of a buried waveguide

By using the above procedure, a refractive index grating was inscribed into the buried waveguide. Irradiation intensity, time and grating period were the same as those for the thin film grating. To detect output-coupled light via the grating, 1.3 μm laser diode light was coupled into the waveguide through optical fibre. By rotating a photodiode from the position perpendicular to the grating (0 degrees) to the end face where light comes out (90 degrees), output-coupled light angle and its intensity were measured. The outputcoupled light angle was detected at approximately 70 degrees and the output-coupling efficiency was 10.2%, which was defined as the intensity ratio of light output from the grating to light from the end face. The input-coupling characteristic was also measured and the coupled light was detected at approximately 70 degrees. The coupling efficiency was less than 1% because the waveguide surface flatness was not good. The polymer waveguide grating was fabricated by the combination of using a grooveshaped substrate for buried waveguide fabrication and a photobleaching technique for the grating fabrication. These simple fabrication methods are widely applicable to a variety of polymer systems in which refractive index change can be induced by photobleaching.

3.5 Future Targets of NLO Polymers for Optical Device Applications Currently, several technical problems exist in second- and third-order NLO polymers. As mentioned above, the trade-off between χ(2) and χ(3) values of polymer waveguides, linear and non linear absorption, processibility, and reliability should be carefully considered. To address these issues, novel hybrid waveguides, i.e., a waveguide with the combination of a signal processing function of NLO polymer and a signal transmission function of a transparent polymer, should also be considered. Hybrid waveguides can be fabricated by constructing the waveguide such that many optical functions are serially

93

Handbook of Polymers in Electronics grafted or vertically integrated. The vertically integrated polymer device fabricated to date was a vertically stacked second-order NLO polymer directional coupler [54]. The non linear optical material used was a 3RDCVXY polymer. Standard photolithography and RIE techniques were applied to fabricate the vertically integrated waveguides. Using the vertically stacked directional coupler, optical coupling characteristics similar to that of an in-plane directional coupler were obtained. These hybrid techniques may be extended to many types of functional device fabrication. The hybrid waveguide design can be applied not only for second-order NLO waveguides but also for all optical switching waveguides by considering the role of the integrated polymers.

3.6 Conclusions NLO waveguides for optical switches and optical modulators have been discussed, mainly focusing on the azo-dye-functionalised NLO polymers. To make the most of polymer processibility, a hybrid waveguide structure is proposed. It is anticipated that this new approach of the application of hybrid waveguides will be a key for future advances in optical signal processing technologies. These hybrid waveguide fabrication techniques can be applied for EO waveguides and for χ(3) waveguides. The successful development of polymeric optical waveguiding devices will require much effort not only by NLO material and device researchers, but also by optical system researchers.

Acknowledgements The author would like to thank S. Tomaru, T. Kurihara, M. Amano, T. Watanabe, M. Hikita, Y. Shuto, M. Asobe, T. Hattori and T. Shibata for their contribution to the research work.

References 1.

Nonlinear Optical Properties of Organic Molecules and Crystals, Ed., D. S. Chemla and J. Zyss, Academic Press, Orlando, FL, USA, 1987.

2.

Photonic Switching II, Springer Series in Electronics and Photonics, Volume 29, Ed., K. Toda and H.S. Hinton, Springer-Verlag, Berlin, Germany, 1990.

3.

T. Kaino, Journal of Optics A: Pure and Applied Optics, 2000, 2, 4, R1.

4.

Y. Maruo, S. Sasaki and T. Tamamura, Journal of Lightwave Technology, 1995, 13, 8, 1718.

94

Non Linear Optical Properties of Polymers for Electronics 5.

A. Neyer, T. Knoche and L.Muller, Electronics Letters, 1993, 29, 399.

6.

T. Kaino in Polymers for Lightwave and Integrated Optics: Technology and Applications, Ed., L. A. Hornak, Marcel Dekker, New York, NY, USA, 1992, Chapter 1, 1.

7.

S. Imamura, R. Yoshimura and T. Isawa, Electronics Letters, 1991, 27, 1342.

8.

T. Matsuura, S. Ando, S. Matsui, S. Sasaki and F. Yamamoto, Electronics Letters, 1993, 29, 2107.

9.

A.J. Heeger, D. Moses and M. Sinclair, Synthetic Metals, 1986, 15, 95.

10. G.M. Carter, M.K. Thakur, Y.J. Chen and J.V. Hryniewicz, Applied Physics Letters, 1985 47, 5, 457. 11. C-J. Yang, S.A. Jenekhe, H. Vanherzeele and J.S. Meth in Electrical, Optical, and Magnetic Properties of Organic Solid State Materials, Materials Research Society Symposia Proceedings, Volume 247, Eds., L.Y. Chiang, A.F. Garito and D.J. Sandman, Materials Research Society, Warrendale, PA, USA, 1992, 247. 12. C.W. Spangler, T.J. Hall, K.O. Havelka, D.W. Polis, L.S. Sapochak and L.R. Dalton in Nonlinear Optical Properties of Organic Materials III, Proceedings of the SPIE, No. 1337, Ed. G. Khanarian, 1990, 125. 13. T. Kaino, K. Kubodera, H. Kobayashi, T. Kurihara, S. Saito, T. Tsutsui, S. Tokito and S. Murata, Applied Physics Letters, 1988, 53, 21, 2002. 14. T. Kaino, T. Kurihara, K. Kubodera and H. Kanbara in Materials for Nonlinear Optics, ACS Symposium Series, No. 455, Ed., S.R. Marder, G.D. Stucky and J.E. Sohn, American Chemical Society, Washington, DC, 1991, 704. 15. J.H. Burroughes, Nature, 1990, 347, 539. 16. I. Murase, T. Ohnishi, T. Noguchi and M. Hirooka, Polymer Communications, 1985, 26, 12, 362. 17. T. Kurihara, Y. Mori, and T. Kaino, Chemical Physics Letters, 1991, 183, 6, 534. 18. G.L. Baker, S. Etemad and F. Kajzer in Advances in Nonlinear Polymers and Inorganic Crystals, Liquid Crystals and Laser Media, Ed., S. Musikant, Proceedings of the SPIE, No. 824, 1987, 102. 19. G.E. Wnek, J.C.W. Chien, F.E. Karasz and C.P. Lillta, Polymer, 1979, 20, 12, 1441. 95

Handbook of Polymers in Electronics 20. D.N. Rao, P. Chopra, S.K. Ghoshal, J. Switakiewicz and P.N. Prasad, Journal of Chemical Physics, 1986, 84, 12, 7049. 21. N.E. Schlotter, J.L. Jackel, P.D. Townsend and G.L. Baker, Applied Physics Letters, 1990, 56, 1, 13. 22. P.D. Townsend, G.L. Baker, J.L. Jackel, J.A. Schelburne III and S. Etemad in Nonlinear Optical Properties of Organic Materials II, Proceedings of the SPIE, No. 1147, 1989, 256. 23. G. Assanto, Q. Gong, R. Zanoni, G.I. Stegeman, R. Burzynski and P.N. Prasad in Proceedings of the SPIE, No. 1216, 1990, 222. 24. J. Valera, A. Darzi, A.C. Walker, W. Krug, E. Miao, M. Derstine and J.N. Polky, Electronics Letters, 1990, 26, 222. 25. M. Thakur and S. Meyler, Macromolecules, 1985, 18, 11, 2341. 26. D.M. Krol and M. Thakur, Applied Physics Letters, 1990, 56, 15, 1406. 27. V.H. Houlding, A. Nahata, T. Yadley and R.L. Elsenbaumer, Chemical Matter, 1990, 169. 28. H.J. Byrne, W. Blau and K.-Y. Jen, Synthetic Metals, 1989, 32, 229. 29. Y. Pan and P.N. Prasad, Journal of Chemical Physics, 1990, 93, 4, 2201. 30. S.A. Jenekhe, W-C. Chen, S. Lo and S.R. Flom, Applied Physics Letters, 1990, 57, 2, 126. 31. T. Kurihara, N. Oba, Y. Mori, S. Tomaru and T. Kaino, Journal of Applied Physics, 1991, 70, 1, 17. 32. S. Matsumoto, T. Kurihara, K. Kubodera and T. Kaino, Molecular Crystals and Liquid Crystals, 1990, 182A, 115. 33. T. Kurihara, T. Kaino, Z.-H. Zhou, T. Kanbara and T. Yamamoto, Electronic Letters, 1992, 28, 681. 34. Y. Mori, T. Kurihara, T. Kaino and S. Tomaru, Japanese Journal of Applied Physics, 1992, 31, 896. 35. T. Kaino, T. Kurihara and S. Tomaru, Nonlinear Optics, 1993, 4, 319.

96

Non Linear Optical Properties of Polymers for Electronics 36. T. Kaino, S. Tomaru, T. Kurihara and M. Amano, Materials Research Society Symposium Proceedings, Eds., L.Y. Chiang, A.F. Garito and D.J. Sandman, MRS, Pittsburg, 1992, Volume 247, 179. 37. M. Amano and T. Kaino, Chemical Physics Letters, 1990, 170, 515. 38. Z.-H. Zhou, T. Maruyama, T. Kanbara, T. Ikeda, K. Ichimura, T. Yamamoto and K. Tokuda, Journal of the Chemical Society, Chemical Communications, 1991, 1210. 39. F. Kajzar, G. Ruani, C. Ttaliani and R. Zambori, Synthetic Metals, 1990, 37, 223. 40. T. Kurihara, S. Tomaru, Y. Mori, M. Hikita and T. Kaino, Applied Physics Letters, 1992, 61, 1901. 41. N. Murata, K. Nakamura, Journal of Adhesion, 1991, 35, 251. 42. M. Asobe, I. Yokohama, T. Kaino, S. Tomaru and T. Kurihara, Applied Physics Letters, 1995, 67, 891. 43. M. Asobe, T. Ohara, I. Yokohama and T. Kaino, Electronic Letters, 1996, 32, 1396. 44. K.D. Singer, M.G. Kuzyk, W.R. Holland, J.E. Sohn, S.J. Lalama, R.B. Comizzoli, H.E. Katz and M.L. Schilling, Applied Physics Letters, 1988, 53, 19, 800. 45. L.R. Dalton, L.P.U. Yu, M.Chen, L.S. Sapochak and C. Xu, Synthetic Metals, 1993, 54, 156. 46. I. Liakatas, C. Cai, M. Bosch, M. Jager, C. Bosshard, P. Gunter, C. Zhang and L.R. Dalton, Applied Physics Letters, 2000, 76, 11, 1368. 47. Y. Shuto, M. Amano and T. Kaino in Nonlinear Optical Properties of Organic Materials IV, Ed., K.D. Singer, Proceedings of the SPIE, No.1560, 1991, 184. 48. M. Amano, M. Hikita, Y. Shuto, T. Watanabe, S. Tomaru, H. Yaita and T. Nagatsuma, in Organic, Metallo-Organic, and Polymeric Materials for Nonlinear Optical Applications, Eds., S.R. Marder and J.W. Perry, Proceedings of the SPIE, No.2143, 1994, 68. 49. Y. Shuto, M. Amano and T. Kaino, IEEE Photonic Technology Letters, 1991, 3, 1003.

97

Handbook of Polymers in Electronics 50. Y. Shuto, S. Tomaru, M. Amano and M. Hikita, IEEE Journal of Quantum Electronics, 1995, 31, 1451. 51. M. Hikita, Y. Shuto, M. Amano, R. Yoshimura, S. Tomaru and H. Kozawaguchi, Applied Physics Letters, 1993, 63, 9, 1161. 52. E. Lim, M.M. Fejer and R.L. Byer, Electronics Letters, 1989, 25, 174. 53. V. Taggi, F. Michelotti, M. Bertolotti, G. Petrocco, V. Foglietti, A. Donval, E. Toussaere and J. Zyss, Applied Physics Letters, 1998, 72, 22, 2794. 54. G. Khanarian, R.A. Norwood, D. Haas, B. Feuer and D. Karim, Applied Physics Letters, 1990, 57, 977. 55. T. Watanabe, M. Amano, M. Hikita, Y. Shuto and S. Tomaru, Applied Physics Letters, 1994, 65, 10, 1205. 56. S. Tomaru, T. Watanabe, M. Hikita, M. Amano, Y. Shuto, I. Yokohama, T. Kaino and M. Asobe, Applied Physics Letters, 1996, 68, 13, 1760. 57. M.-C. Oh, M.-H. Lee. J.H. Ahn, H.J. Lee and S.G. Han, Applied Physics Letters, 1998, 72, 13, 1559. 58. W.W. Ng, C. Hong and A. Yariv, IEEE Transactions on Electronic Devices, 1978, ED-25, 1193. 59. A. Natansohn, P. Rochon, M-S. Ho and C. Barrett, Macromolecules, 1995, 28, 12, 4179. 60. D.Y. Kim, L. Li, X.L. Jiang, V. Shivshankar, J. Kumar and S.K. Tripathy, Macromolecules, 1995, 28, 26, 8835. 61. T. Hattori, T. Shibata, S. Onodera and T. Kaino, Journal of Applied Physics, 2000, 87, 7, 3240. 62. T. Shibata, T. Hattori, S. Onodera and T. Kaino, Journal of the Chemical Society of Japan, 1998, 831. 63. T. Hattori, T. Shibata, S. Onodera and T. Kaino in Organic Nonlinear Optical Materials, Eds., M. Eich and M.G. Kusyk, Proceedings of the SPIE, No. 3796, 2000, 320.

98

4

Luminescence Studies of Polymers B. Wandelt

4.1 Introduction When radiation is incident on a material some of the energy absorbed by a chromophore is re-emitted as light of a longer wavelength, which is in agreement with Stokes law [13]. There are many pathways for the deactivation of the absorbed energy. Two important domains characterise the radiation: intensity and frequency. The intensity depends on spectroscopic selection rules: the efficiency of competing non radiative, quenching and energy transfer processes. The frequency depends on the nature of the emitting states, i.e., whether it is a singlet or a triplet state, and whether this is a complex formatted by an interaction with some other chromophore. Most of the significant features in luminescence studies have their origin in the fact that the electronically excited states are affected by molecular interaction and molecular motion. The photochemists [3, 4] suggested classifying polymers to distinguish between those in which the repeat unit contains a chromophore and those in which isolated chromophores are attached to a polymer chain as an end group or minor component of a copolymer. That is important, since the main deactivation processes are bimolecular complexes through dimer formation, which is dependent on segmental motion of the polymer chain. In the solution state, the translational and segmental motions are very extensive but are much reduced in concentrated or bulk polymer systems. The macromolecules that become expanded in a good solvent tend to contract in a poor solvent below the value of its unperturbed dimensions, defined by Flory [5]. However, the translation and segmental motion of a polymer chain are temperature dependent, so that the thermally activated dynamics of the polymer chain can affect the mobility of the probe molecule. It is one of the purposes of this chapter to demonstrate, by reviewing and summarising the data already available in the literature, that, due to photophysical features of polymers, fluorescence investigations are the methodologies suited to the study of environmental effects in a polymeric network. Since the fluorescence properties of a probe molecule, such as decay kinetics, are strongly affected by microenvironments, the fluorescence probe method can be used to probe the local mobility of polymer chains both in solution and in the solid state. Fluorescence probe methodology can be applied for studying microstructures and the morphology of a polymeric medium [6-7]. It can also be applied

99

Handbook of Polymers in Electronics in microheterogeneous systems, imaging processes, microlithography [8], latex beads [9], silica [10] and imprinted polymers [11]. Various fluorescent species (such as monomers, ground-state and excited complexes) have been observed owing to the differences in the formation of molecular interaction between moieties of polymers, and only some of these will be discussed here. Obviously the most important molecular probe in polymer fluorescence is an excimer: an excited complex of two identical species, one of which is in the excited state prior to complexation. The present review begins with a brief description of the basic principles of luminescence [1-3], after which it deals with the fluorescence studies of polymers in solution and energy transfer in polymeric systems. Spectral and time dependent studies of polymers in the gel state are discussed later, with an emphasis on phase transition and phase separation studies.

4.2 Basic Photophysical Deactivation Processes The excess of energy taken up by light absorption is known to dissipate through photochemical and photophysical processes. Some of these are unimolecular whereas others are bimolecular. Some of the unimolecular processes of energy dissipation following the absorption of light are depicted in a Jablonski diagram (Figure 4.1).

Figure 4.1 Jablonski diagram of the important unimolecular photophysical processes of excitation energy dissipation: absorption (A), fluorescence (F), phosphorescence (P), internal conversion (IC), intersystem crossing (ISC), vibrational relaxation (VR). S0 is the singlet ground state, S1 and S2 are excited singlet states and T1 is the excited triplet state.

100

Luminescence Studies of Polymers The initial absorption raises the molecule to an excited singlet state (S2). However, the internal conversion (IC) between excited singlet states is usually very rapid, so that the molecule relaxes to the ground vibrational level of the first excited singlet, S1, in times of the order of a picosecond. The energy may then be emitted as fluorescence, or intersystem crossing (ISC) may take place to the triplet state, T1. The triplet state may then lose its energy and return to the ground state either by emitting phosphorescence or by radiationless transition to the ground state or by ISC from T1 back to S1 when the energy difference between them is small enough to allow the thermal activation. These processes, involving changes in multiplicity, are generally forbidden and require times from millisecond to seconds. The relative content of all the processes is dependent on the molecular and supermolecular structure of the material as well as on the molecular motions, which in turn are affected by temperature. Obviously, there will be higher energy singlet and triplet states, but in condensed phases these are more involved in photochemical reactions.

4.2.1 Luminescence Fluorescence is defined as the emission from a transition between states of the same multiplicity, usually from the lowest vibrational level of the first excited singlet state to the ground state. A mirror-image relationship usually exists between absorption and emission spectra; the longest absorbed and shortest emitted wavelength usually corresponds to the 0-0 transition. If the spontaneous emission of radiation of the appropriate energy is the only pathway for return to the ground state, the average statistical time that the molecule spends in the excited state is called the natural radiative lifetime, τ0, which relates to the rate constant, k0, of the fluorescence in the following way: τ0 =

1 k0

(4.1)

Each process competing with the spontaneous emission reduces the observed lifetime, τ, relative to the natural lifetime: τ=

1 ( kF +

∑k )

(4.2)

i

i

where kF is the rate constant of fluorescence and for all processes competing with emission.

∑k

i

is the sum of the rate constants

i

101

Handbook of Polymers in Electronics Equation 4.1 can be approximated to a useful expression relating τ for a molecule to the maximum extinction coefficient εmax (λ), determined from the absorption spectrum as a function of wavelength (λ): τ F = 10

−4

ε max ( λ )

(4.3)

This equation predicts an approximate value of 10-9 s at εmax ≈ 105 l mol-1 cm-2. The measured fluorescence lifetime is usually less than the predicted value. This is attributed to the presence of quenchers like oxygen and many other impurities occurring at a concentration of about 10-3 M in the commonly used spectroscopic solvents. Fluorescence is a comparatively fast process and the radiative fluorescence lifetime is usually in the range of 10-6 to 10-12 s. Delayed fluorescence differs from ‘normal’ fluorescence in that the measured rate of decay of emission is less than that expected from the transition giving rise to the emission. Spectral distribution of the delayed fluorescence is similar to that of ‘normal’ fluorescence. However, its lifetime corresponds to the excited triplet state. Phosphorescence emission is the result of a transition between states of different multiplicity (typically T1 to S0) that has a much smaller rate constant than that for fluorescence. Consequently, the natural lifetime, τP0 , of the triplet state is long, varying between 10-6 s and seconds. The natural lifetime can be formulated according to Equation 4.4. τP0 =

1 kP

(4.4)

Vibrational relaxation (VR) from a vibrational level of a higher electronic state such as S2 to the vibrational ground state is very rapid. Before the molecule can react, the vibrational energy is quickly distributed among the various vibrations and can be partially or completely dissipated by collisions into heat. IC takes place as a transition between two isoenergetic vibrational levels of different electronic states of the same multiplicity, which may have quite different energies at their equilibrium geometries. The IC is used in a wider sense, encompassing vibratrional relaxation. It denotes radiationless transitions from the first excited singlet state into the vibrationally equilibrated ground states. The radiationless processes from the higher to lower excited states are usually very fast. The lifetimes of the higher excited states are very short and quantum yields of emission from higher excited states are negligible.

102

Luminescence Studies of Polymers ISC differs from IC in that the transition takes place between states of different multiplicity as can be seen in Figure 4.1. IC and ISC processes are similar in that they involve a conversion of electronic energy into vibrational energy, which is followed by rapid relaxation to the lowest vibrational level of the lowest excited state. These all are known as radiationless processes.

4.2.2 Bimolecular Photophysical Processes Considering the excitation originating from the singlet ground state of a molecule, which is able to interact with a similar or a different molecule, there is a further possibility of energy dissipation involving the other molecule. This may result in the energy transfer between the molecules. Additionally, it may cause interaction of the excited state of a molecule with the ground state of the other, resulting in the formation of an excited complex. The competition between the processes of energy dissipation can be controlled by either thermodynamic or kinetic parameters, but the latter are of prime importance in a consideration of molecular motion and luminescence. As significant competition to radiation can come only from processes capable of occurring in the appropriate timescales, changes in fluorescence are likely to be caused primarily by the fastest molecular motions, whereas phosphorescence can be significantly affected by a very wide variety of relatively slower processes. Excimers are complexes/dimers of electronically excited molecules with molecules of the same type in their ground state. They only exist in the excited state and they dissociate into monomers upon radiative or non radiative deactivation in agreement with scheme shown in Figure 4.2. This phenomenon of association of chromophores is called concentration quenching. Since the discovery of the pyrene excimer by Förster and Kasper in 1954 [12],

Figure 4.2 Birks’ scheme [1] of excitation and deactivation by bimolecular processes: kEM and kME are the rate constants of excimer association and dissociation, respectively, kFM and kFE are the rate constants of fluorescence emission of excited monomer chromophore and excimer, respectively, and kIM and kIE are the rate constants for non radiative energy decay of the monomer end excimer, respectively.

103

Handbook of Polymers in Electronics these complexes were observed frequently with aromatic hydrocarbons. Excimers of aromatic molecules adopt a sandwich structure with a separation distance of 0.30-0.35 nm between them. Their fluorescence emission spectrum from the broad and structure-less band has been found to be shifted to the lower energy (relative to the molecular emission) by 50-60 nm. Excimer fluorescence can be observed in solutions and in solids, if the crystal and/or material structure allow a close overlap of the molecular planes. Exciplexes are complexes of two different molecules usually of 1:1 stoichiometry. Their fluorescence phenomena are similar to those described for excimers, but their formation is not restricted to aromatic systems. If the sum of effective rate constants of the non radiative processes is so high such that the lifetime of emission is undetectable, these molecules do not necessarily luminesce. In contrast to the excimers, which are non polar, the exciplexes are polar entities. It was shown by Beens and co-workers [13] that exciplexes from aromatic hydrocarbons and aromatic tertiary amines demonstrate the charge-transfer character of the complexes as reported by Knibbe and co-workers [14], and their dipole moments were greater than 3.3 x 10-29 C m (10 D). Birks [1] and Klessinger and Michl [2] have given a more detailed discussion of the photophysics of organic molecules.

4.2.3 Quenching Processes There are many quenching processes. Quenching by photochemical reaction refers to organic photochemistry and will not be discussed here. Photophysical quenching, which does not lead to a new chemical compound, includes self-quenching through excimers and exciplex formation. The last photophysical process includes exciplex formation, but there are other possibilities like quenching by electron transfer or by energy transfer due to presence of heavy atom. In complicated systems like polymers, the competition between the possible processes is controlled by either thermodynamic or kinetic parameters, and naturally both are important and interactive. Rate constants for energy transfer can be measured by exciting a chromophore with a fast light pulse followed by monitoring the dynamics of the excited species. It can be measured from the competition between emission and the deactivation by an added quencher or acceptor. The Stern-Volmer equation (Equation 4.5) relates the experimental values of the quantum yields of the process, Φ0 and Φ, in the absence and presence of acceptor respectively, to the molar concentration of the quencher Q as: Φ0 = 1 + kq τ Q Φ

[ ]

(4.5)

where τ is the lifetime of the donor in the absence of additives, and kq is the quenching rate constant.

104

Luminescence Studies of Polymers

4.3 Methods for Fluorescence Studies The phenomenon of fluorescence spectroscopy contains two domains: spectral studies and time-dependent studies. Although fluorescence spectroscopy cannot be classified as a ‘finger print’ type, the spectral studies are widely used as analytical applications of fluorescence. Although the time-dependent studies of fluorescence pertaining to the fluorescence lifetime measurements are not sufficiently specific, they are used for many purposes, such as determination of reaction kinetics, rates of competitive processes, and to probe local environments. There are essentially two types of methods for measuring fluorescence lifetimes: pulse fluorometry (relating to measurements performed in the time domain), and phase and modulation fluorometry (relating to the frequency domain) [15]. Among the several publications devoted to the methods of studies of time dependent fluorescence, the reader is referred to the book edited by Lakowicz [15].

4.3.1 Time-Correlated Single-Photon Counting Studies A single photon fluorometer consists of the following subsystems: an optical spectrometer (including a pulsed light source), a data acquisition system (consisting of timing electronics) and a data analyser (composed of a computer with re-convolution software). Birch and Imhof [16] have widely published on instrumentation design and analytical methods. The data analysis software is used to extract the kinetic information in a given system from the experimental intensity decay curve. The recorded decay curve of emission intensity, I(t), is generally synthesised by means of an exponential re-convolution function of the form: I( t ) =

∑B

n

exp( −t / τ n )

n

(4.6)

with n = 1,2,3,..., where τn and Bn are the lifetime parameter and pre-exponential function of the nth component in the decay, respectively. An important feature of the equation is that the existence of n decay components suggests the existence of n excited states. Moreover, it is usually recommended that the number of decay components chosen for least-squares analysis should always be the minimum necessary to give a satisfactory fit. The last-squares analysis uses a quantity χ2 as a measure of discrepancies between data and fitted function and, for a good fit, χ 2 less than 1.2 is recommended [16].

4.3.2 Quantum Yields A useful quantity in the description of photophysical and photochemical processes is the quantum yield. The quantum yield, Φi, of a process ‘i’ is defined as the number, nA, of molecules, 105

Handbook of Polymers in Electronics A, undergoing that process divided by the number, Q, of light quanta absorbed [2]: nA Q

Φi =

(4.7)

For practical use, we measure the quantum yield relative to the quantum yield of a known standard compound. The following materials can be used as standards: •

Rhodamine B in ethanol at 22 °C; ΦF = 0.69 with λexc. = 366 nm [17];

•

Quinine sulfate in 1N H2SO4, 25 °C; ΦF = 0.546 with λexc. = 328 nm [18];

•

2-Aminopyridine in 1N H2SO4, 25 °C; ΦF = 0.60 with λexc. = 290 nm [19].

It will be useful for many purposes to distinguish between quantum yield related to the absorbed radiation and efficiency, ηi, related to the number of molecules in a given state [2]: ηi =

ni nA

(4.8)

which is the ratio of the number, ni, of molecules undergoing a specific reaction to the number, nA, in the excited state. If only processes that obey a purely exponential rate law such as fluorescence are involved in deactivating the singlet state, S1, the quantum yield of fluorescence may be written as the ratio of the observed to the natural lifetime: ΦF =

τ τ0

(4.9)

where the natural lifetime, τ0, and the observed lifetime, τ, of the singlet state are given by equations (4.1) and (4.2).

4.4 Fluorescence of Polymers, Excimer Fluorescence The typical synthetic vinyl polymers prepared by the free radical polymerisation of vinyl monomers are shown in Figure 4.3. They have a linear chain structure in which the substituents R and R´ are separated by three carbons. If the substituents R are aromatic chromophores, the vinyl polymers will be rich in excimer structures and strong bimolecular quenching of the fluorescence will occur. This is in agreement with the Hirayama rule [20], which says that if the chromophores are separated by 3 carbon atoms, the probability of excimer formation is the highest, assuming that the chain is flexible enough to rotate

106

Luminescence Studies of Polymers R'

R

R'

R

R'

R

R'

R

R

R' =

H

R=

H

polyethylene (PE)

R' =

H

R=

CH3

polypropylene (PP)

R' =

H

R=

polystyrene (PS)

O R' =

CH3

R=

C

OCH3

polymethyl methacrylate (PMMA)

O R' =

CH3

R=

C

O

polymethyl methacrylate (PMMA)

Figure 4.3 Schematic representation of vinyl polymers with substituents R and R´

around the carbon-carbon bond. The majority of literature in polymer studies in the last two decades has originated due to excimer emission and formation. Excimer emission has been detected from a number of aromatic vinyl polymers [21-24]. The aromatic chromophores have included phenyl, naphthalene, pyrene and carbazole. The vinyl polymers and copolymers that have been widely studied are: polystyrene (PS) [22], polyvinylnaphthalene (PVN) [23], polyvinylcarbazole (PVCZ) [24]. Many photophysical processes in polymers depend on whether the process occurs in the solution or solid state. This is because the rotational diffusion in the polymer chain can control the kinetics of the processes. Therefore, the conformation of the chain backbone required to bring neighbouring aromatic chromophores into sandwich geometry is one of unfavourably high energy. This is almost entirely absent from these systems, where the low-temperature distribution of chain conformations becomes trapped in a rigid medium, but it can have a transient existence when segmental motion and rapidly rotating chain backbones undergo a variety of conformation states. According to this model, the polymer chains dissolved in fluid solution should have a Boltzmann distribution of suitable excimerforming sites, the concentration of which will be the factor determining excimer fluorescence intensity. Harrah [25] has illustrated these dependencies for poly-2vinylnaphthalene. The emission characteristics of a solid-state polymer will then depend on the conformational equilibrium in solution existing at the temperature solidification. A demonstration of this phenomenon in solid solutions of poly-2-vinylnaphthalene and poly-4-vinylbiphenyl, as a relation between the ratio of excimer to monomer emissions

107

Handbook of Polymers in Electronics and temperature of solidification, has been shown by Frank and Harrah [26] in agreement with the following equation: ⎛ − ΔE E ⎞ IE ∝ exp⎜ ⎟ IM ⎝ RTs ⎠

(4.11)

where ΔEE represents the energy barrier between the normal and excimer creating conformations at the temperature at which the film was cast, and Ts is the temperature of solidification. The analysis of excimer and monomer emission in fluid systems is complex. A general kinetic scheme for excimer formation, including the inter- and intramolecular complex formation in terms of a singlet ground and excited states, and the various photophysical processes occurring in polymers, is given in Table 4.1. The three routes of energy

Table 4.1 Photophysical processes occurring in polymers Photophysical processes

Initial species 1

Energy absorption

M + hν

Fluorescence

1

IC

1

ISC

1

Ia

M + hνFM

kFM

1

M

kIM

M*

kTI

M + Q*

kQM

M + hνPM

kPM

3

M*

1

1

M* + Q 3

ISC

3

1

M*

1

M*

Singlet migration

1

Excimer formation

1

1

1

M* + M

Excimer fluorescence

1

(MM)*

ISC

1

(MM)*

Non radiative process

1

kMM

(MM)*

kEM

M + 1(MM)*

kEM

1

1

(MM)* 2

M + 1M*

kME

(1M) + hνFE

kFE

3

(MM)* 1

(MM)*

(MM)* + Q

kIT

1

M* + M

1

1

M + M*

1

M* + 1(MM)

1

M

1

Excimer dissociation

108

1

Rate constant

M*

M*

Phosphorescence

Excimer quenching

1

M*

Monomer quenching

Excimer formation via preformed trap

Products

kIE

2( M)

kIE

2( M) + Q*

kQE

1

Luminescence Studies of Polymers deactivation illustrated in Figure 4.1 are fluorescence, phosphorescence and non radiative transition, and represent the competitive relaxation processes if the system is unimolecular. The situation becomes rather more complex when bimolecular interactions are included. There are further possibilities involving both the molecules. As an example, some competing radiative, migration, and energy transfer processes, which can occur after singlet excitation in a polymer chain, are shown in Figure 4.4.

Figure 4.4 Schematic illustration for competing radiative, transfer and migration processes which can occur after singlet excitation. Energy quantum absorbed by the chromophore can be dissipated by single chromophore, monomer emission, hνM, or by dimer chromophore, excimer emission, hνE, or it can migrate along the polymer chain, and/or the excitation energy can be transferred to an acceptor molecule.

When two identical chromophores are brought into proximity in a suitable relative orientation, there is a possibility that excitation energy may move as an exciton from one to the other by a non radiative process. In a polymer chain containing a large number of chromophores, the exciton can migrate down the chain until a suitable trap appears or it can be transferred to a chemically different molecule (quencher). Much of the evidence for energy migration comes from observations of excimer emission. In many polymers, the intensity of excimer emission is greater than can be explained by absorption at the pre-formed sites, as would be necessary in a solid, and neither by conversion of the absorption site to excimer geometry by conformation change of a flexible chain. The explanation is that absorption occurs at monomer sites, and then the energy migrates until the appropriate exciton trap appears. Among the many papers devoted to energy

109

Handbook of Polymers in Electronics migration studies, a wide description of particular mechanisms of energy migration and energy transfer can be found in [3]. The competition between the photochemical processes in polymers, as shown in Figure 4.4, can be controlled by either thermodynamic or kinetic parameters. The latter are of prime importance if we consider the molecular motion in the polymer. The various photochemical processes occurring in polymers and the related rate constants are given in Table 4.1. Since IM = kFMM* and IE = kFEE*, inserting a stationary-state of excited monomer and excimer allowed the following expressions for monomer and excimer emission intensities (IM and IE, respectively) to be obtained by David and co-workers [23, 24]: I M = kFM

I E = kFE

I a ( kE + kQEQ + kME ) ( kM + kEM M + kQMQ )( kE + kQEQ + kME ) − kEM MkME

I a kEM M ( kM + kEM M + kQMQ )( kE + kQEQ + kME ) − kEM MkME

(4.12)

(4.13)

From the practical point of view, the ratio of monomer to excimer intensity has been used very frequently: I M kFM ⎛ kE + kME + kQEQ ⎞ = ⎟ IE kFE ⎜⎝ kEM M ⎠

(4.14)

Application of these equations and the scheme of photophysical processes to fluorescence observed under variety of conditions have been used in a number of papers on processes in polymer solutions and in solids [3, 23, 24, 33].

4.4.1 Fluorescence of Polymers in Solution 4.4.1.1 Effect of Conformation of the Polymer Chain In solution, the rate of the many bimolecular photoprocesses due to a polymer chain may be limited by the rate of mutual diffusion of the interacting species, but the most important factor in the case of a polymeric system is the chain conformation and flexibility. Guillet and coworkers [27, 28] studied photophysical properties for naphthyl-substituted polymethacrylate (PNMA) solutions very extensively. The spectral properties of absorption and emission of PNMA studied using different solvents under different conditions are shown in Figure 4.5. Somersall and Guillet [27] observed delayed fluorescence, involving

110

Luminescence Studies of Polymers

Figure 4.5 Ultraviolet absorption and emission spectra of polynaphthyl methacrylate in chloroform at 298 K: (1) absorption, (2) fluorescence, (3) and (4) delayed emission in tetrahydrofuran-ether at 77 K and phosphorescence (Reprinted with permission from Macromolecules, 1973, 2, 219, copyright 1973, American Chemical Society)

the triplet state, with a lifetime of 0.1 s, when PNMA in tetrahydrofuran-ether glass at 77 K was excited at 313 nm. Similarly, Cozzens and Fox observed delayed fluorescence in glassy PVN [29]. These studies provide evidence for complicated chromophore interactions involving the excited states. Somersall and co-workers [27] studied the fluorescence behaviour of polynaphthyl methacrylate in different solvents and they have shown that the ratio of excimer to monomer emission varied with the solvent quality, being high in poor solvents and low in good solvents, as illustrated in Figure 4.6. Fluorescence spectra of PNMA at 298 K contain two main bands, a mirror image of the absorption band due to monomer emission and a broad band shifted about 60 nm to longer wavelengths which is attributed to excimer fluorescence. When chloroform is used as a solvent for PNMA, the monomer emission is predominant, being visible with benzene as a solvent. However, when the solvent is ethyl acetate, monomer emission almost disappears and the excimer emission rises. Aspler and co-workers [28] demonstrated that the fluorescence properties correlate with a change in the effective volume of the random coil in solution. An increase in excimer emission was observed if a non solvent was added to the solution to compress the polymer random coils. These observations did not depend on whether they used a polar non solvent (such as alcohol), or a non polar solvent (such as cyclohexane). In Figure 4.6, one can observe an increase in the ratio of emission intensity from excimer to monomer bands when PNMA was in chloroform and then cyclohexane was added to a 1:1 mixture of the solvents; the ratio, IE/IM, changed from 1.44 to 2.81 when the viscosity decreases from 0.184 x 10-3 to 0.125 x 10-3 N s m-2 [28]. The relationship between the fluorescence and viscosity was explained according to the classical Flory theory [5] of intrinsic viscosity:

111

Handbook of Polymers in Electronics

Figure 4.6 Fluorescence of PNMA in different solvents at 298 K: (1) in benzene, (2) in ethyl acetate, (3) in chloroform, (4) in chloroform-cyclohexane (1:1). The intensities of the spectra are not comparative (Reprinted with permission from Macromolecules, 1973, 2, 221, copyright 1973, American Chemical Society)

[ η] =

63 / 2 φ < S2 > 3 / 2 φ < h2 > 3 / 2 or [ η] = M M

(4.15)

where ø is constant, 2.1 x 1021, and < S2 > is the mean square radius of gyration, < h2 >1/2 is the root mean square end-to-end distance and M is the molecular weight. The average density of polymer segments, ρ, is proportional to M / < S2 >3/2 and the relationship is: ρ∝

1 [ η]

(4.16)

The rigidity of the chain has been characterised by the value of the steric factor, σ, which is defined as: σ2 =

< h2 > < h2 > f

(4.17)

where < h2 >f is the theoretical distance between chain ends with completely free rotation around all carbon-carbon bonds. Chain length is important in providing a sufficient

112

Luminescence Studies of Polymers number of chromophore interactions in intramolecular excimer formation behaviour. Equation 4.16 can be used to explain molecular weight effects on excimer formation as a chain densification effect. Aspler and co-workers [28] observed the influence of molecular weight of PNMA in the range from 40,000 to 360,000 on increase of excimer emission. Similarly, Nishijima and co-workers [30] found a strong effect of molecular weight on increase of excimer formation when they studied low molecular weight polyvinyl naphthalenes series, range 1,400-6,000. The conformation of a polymer chain can be affected by the temperature of the polymer solution. The conformation change of poly(phenyl methacrylate) (PPMA) in tetrahydrofuran (THF) solution was observed in dilute solutions using fluorescence emission, viscosity, density and dipole moment measurements by Wandelt and Szumilewicz [31]. The fluorescence spectra of PPMA in THF solution when the temperature rises from 266 K to 318 K is shown in Figure 4.7. We can observe the shift of the maximum

Figure 4.7 Fluorescence spectra of poly(phenyl methacrylate) in THF, 4.8 x 10-3 M, at different temperatures: (1) 266 K, (2) 273 K, (3) 279 K, (4) 289 K, (5)295 K, (6) 308 K, (7) 318 K: Excitation at 265 nm.

113

Handbook of Polymers in Electronics of fluorescence from λmax ≅ 335 nm to emission centred at λmax ≅ 358 nm. The emission spectrum of PPMA becomes very broad at 295 K where two bands are visible and comparative. Similar behaviour of fluorescence spectra was observed by Somersall and co-workers [27] for other methacrylate polymers PNMA, for which the monomer emission is relatively strong in comparison with vinyl polymers under the same conditions. The fluorescence emission spectrum of PPMA at 266 K fits 80% of the fluorescence spectrum of phenyl methacrylate, as is shown in Figure 4.8. Abuin and co-workers [32] reported similar fluorescence spectra in phenyl-containing methacrylate polymers solutions in ethyl acetate and in dichloromethane. As a first approximation, the phenyl chromophore in PPMA can be considered as isolated, because these polymers do not conform to the Hirayama 3 carbons rule. But we can observe considerable excimer emission from other polymers where the chromophores are separated by more than three atoms, for example in the spectra of naphthyl-containing methacrylates by Somersall and co-workers [27]. The spectrum of PPMA obtained at 318 K fits the Gaussian distribution function, as is shown in Figure 4.9, which would suggest that the majority of emission is excimer emission. Although the monomer and excimer bands from PPMA are strongly overlapping, the bands are distinguishable. The steric factor values obtained from viscosity studies [31] correspond to the values obtained for polyvinylcarbazole which was established as a stiff polymer [23, 24]. These studies emphasise the role of chain conformation and excimer binding energies in determining the extent of excimer formation. PS and PVN are examples of polymers where excimers can exist at nearly every point of the chain and the concentration of excimer sites could be quite high, especially when the mobility of

Figure 4.8 Fluorescence spectra of poly(phenyl methacrylate) in THF, 4.8 x 10-3 M, at 266 K (1) and of phenyl methacrylate in THF (2)

114

Luminescence Studies of Polymers

Figure 4.9 Fluorescence spectrum of poly(phenyl methacrylate) in dilute solution in THF, 4.8 x 10-3 M, at 318 K (1) and the fitting Gaussian curve (2)

the chain allows the necessary rotation to form an excimeric structure. Naphthyl esters are polymers where the probability of adjacent excimer formation is low and the high intensity of excimer emission would be due to the high efficiency of energy migration. The effect of temperature on excimer formation in a polymer chain is mostly due to thermally-activated conformational changes in the polymer chain. It was established that in dilute polymer solutions the singlet excimer formation is an intramolecular process since the ratio of excimer to monomer emissions intensity is independent of polymer concentration [23, 24]. However, it was shown by David and co-workers that in vinyl polymers, excimers usually occur between neighbouring chromophores and their formation is activated by temperature. Starting from 77 K, the intensity of excimer fluorescence from vinyl polymers is initially constant when it is in the glass state, and then increases, whereas the monomer intensity decreases. This is due to the fact that the rate constant for excimer dissociation, kME, is much lower than the rate constant for excimer formation, kEM, in the low temperature range. At a temperature higher than 202 K, excimer dissociation is very efficient for the vinyl polymers, and the intensity of excimer fluorescence decreases. This observation corresponds to a minimum which was observed by David and co-workers [23, 24] at 202 K in a graph of ln(IM/IE) versus 1/T. Values of energy of activation of excimers formation in poly-1-vinylnaphthalene and polyacenaphthylene in methyl tetrahydrofuran (MTHF) solution, obtained by David and co-workers [23, 24], were respectively 11.3 and 3.4 kJ mol-1. But the energy value of 14.7 kJ mol-1 was reported for activation of excimer formation in poly-2-vinylnaphthalene in solution by Al-Wattar and Lumb [33] and 15.1 kJ mol-1 by Harrah [25]. Chandross and Dempster [34] reported values of energy activation 13.8, 16.8 and 12.6 kJ mol-1 for

115

Handbook of Polymers in Electronics excimers association in, respectively, 1,3-bis-1-naphthylpropane, 1,3-bis-2naphthylpropane and 1-methylnaphthalene. The fluorescence behaviour of PVCZ solutions is quite different from that of the other vinyl polymer solutions. The emission spectrum of PVCZ results from two distinct excimers: a low-energy sandwich type structure with emission centred at 420 nm and a high-energy excimer centred at 375 nm. Itaya and co-workers [35] have proposed that the high-energy excimer has a structure in which only one pair of phenyl rings from the two-carbazole chromophores overlap. In the temperature-dependent curve for ln(IM/IE) as a function of 1/T, two minima corresponding to the formation and dissociation of two excimer structures were observed for PVCZ by David and co-workers [23, 24]. These two different excimer emissions have been assigned to different conformations of the chain, and a change in the conformation occurs between 143 and 295 K. Further information on these structures was obtained with use of time-resolved fluorescence spectroscopy. Time-resolved fluorescence spectra for PVCZ in benzene obtained by Hoyle and co-workers [21] are presented in Figure 4.10. This experiment showed that the high energy excimer, in a partially eclipsed form, appeared at the earliest times of observation,

Figure 4.10 Time-resolved fluorescence spectra of polyvinylcarbazole, 5 x10-4 M, in degassed benzene at 296 K, excitation wavelength 313 nm. Spectra obtained with time intervals from the lamp maximum: (1) 0-0.23 ns, (2) 8-9.4 ns, (3) 19-35 ns, (4) 182323 ns. Spectra are adjusted to the same intensity scale. (Reprinted with permission from Macromolecules, 1979, 5, 957, copyright 1979, American Chemical Society)

116

Luminescence Studies of Polymers (i.e., at a 0.2 ns interval from a photon absorption) and the emission from the lowenergy sandwich excimer was absent at that time. The low-energy excimer due to the normally eclipsed form appeared after 8 ns to be present after 182 ns as well. These studies suggest that the chain of PVCZ requires significant rotational time to form the sandwich excimer. The PVCZ chain was reported [23, 24] as a stiff polymer for which the steric factor from Equation 4.17 was 2.8, whereas it was 2.2 for PS [23, 24], and has a dielectric segmental rotational relaxation time of about 180 ns for a chain of Mw greater than 4 x 104 [36]. An interesting experiment was done by Cuniberti and Perico [37]. They synthesised a number of monodisperse polyethylene oxide samples terminated with pyrene groups. This experiment, which was related to the cyclisation of the polymer chain as it is schematically presented in Figure 4.11, showed that intramolecular excimer fluorescence increases inversely proportionally to the molecular weight of the polyethylene oxide chain separating the two chromophores. Similar experiments were continued by Winnik and co-workers [38] and they obtained the particular rate constants for cyclisation for PS and polydimethyl siloxane chains terminated by pyrenes. The cyclization activation energy, 14 kJ mol-1, was three times lower than that for excimer dissociation, 44 kJ mol-1, in pyrene-labelled PS. The difference corresponds to the binding energy of the excimers for solution of pyrene in cyclohexane, but approximate calculations have shown that the probability of cyclisation is strongly sensitive on the length of PS chain separating the two chromophores forming the excimer [38, 39]. It was shown that the addition of non labelled PS to pyrene-labelled polydimethyl siloxane reduces the rate of cyclisation. These effects were interpreted in terms of the decrease of the conformational mobility of pyrene-labelled polydimethyl siloxane chain through interaction with PS molecules.

Figure 4.11 Intramolecular excimer formation through end-to-end cyclisation of polymer chain

117

Handbook of Polymers in Electronics It has been demonstrated that the excimer emission intensity from chromophores incorporated into the vinyl polymer chain have been correlated with a change in the effective volume of random polymer coil in solution, and the volume of random polymer coil was correlated with viscosity, which is dependent on temperature. But a temperature change of a polymer solution does not always lead to expansion or contraction of the chain coil; it can cause conformation change to a more or less packed polymer chain and to a more or less excimer forming conformation. Both of these structural changes in a polymer chain can affect the fluorescence, but each polymer needs to be considered individually.

4.4.1.2 Energy Transfer and Migration The energy transfer and migration processes play an extremely important role in the photochemistry and photophysics of polymers. But, there is no direct measurement that allows conclusions to be made about these processes. Three types of energy transfer can be distinguished. Firstly, there is a transfer from the chromophore absorbing the incident radiation to the photochemically active site. A second type is collisional energy transfer, where the electronic excitation energy localised on the absorbing chromophore is transferred to another group on the same or different molecule as a result of the overlap of electronic charge clouds which occur during a collision. As an indicator of energy transfer efficiency we can use the Smoluchowski equation [3]: kq =

4πrAB DAB 10 3

(4.18)

where rAB is the energy capture radius and DAB = DA + DB is the diffusion coefficient for relative movement of energy donor and acceptor. The third type of energy transfer takes place by resonance transfer over an extensive range of space, as a result of dipole-dipole interaction, as was proposed by Förster [40]. Migration of energy along polymer chains can occur by two mechanisms. When the electronic interactions between adjacent chromophores are weak, and the orientation correlations are low, the energy undergoes a statistical sequence of jumps from one chromophore to another as schematically shown in Figure 4.4. Each of the jumps has the characteristics of non radiative resonance transfer as proposed by Förster [40] and Dexter [41], and characterised by an effective energy diffusion distance. When there are strong interactions between neighbouring chromophores, and orientation correlations are appropriate, the energy may be delocalised as a wave over a number of chromophore units. The most convincing quantitative evaluation of down-chain energy migration comes from efficiency of polymer luminescence quenching by acceptor species present in concentrations which would be most inefficient if the absorbed energy were localised.

118

Luminescence Studies of Polymers When a system contains two fluorescing chromophores such that the emission spectrum of the donor overlaps the absorption spectrum of the acceptor, excitation energy from the donor can be transferred to the acceptor over a considerable separation distance, R. The efficiency, E, of this energy transfer would be governed by the equation proposed by Förster [40]: E=

R06 R06

+ R6

(4.19)

where R0 is a characteristic distance for a rotationally averaged pair for which half of the excitation energy is transferred. This characteristic distance depends on the overlap integral between the emission spectrum of the donor and the absorption spectrum of the acceptor [40]. There is always uncertainty if we adopt this equation for polymeric systems; the most important consideration seems to be the inhomogeneity of the polymer solution when the polymer segments are constrained into local high and low concentrations by the distribution of macromolecules. An incident photon absorbed by a chromophore is then transferred to the site of chemical activity. This is frequently followed by chain scission reactions, which in photooxidation depends upon the solubility and diffusivity of oxygen and the onset of localised molecular motion in the solid state [42-44]. To protect the polymer from the photooxidation and photodegradation, some additives are frequently used. Kryszewski and co-workers [45, 46] presented fluorescence intensity and lifetime studies as indicators of the efficiency of energy transfer between two polymers: PS and poly-2,6-dimethylphenylene oxide (PPO) and correlation between energy transfer and photochemical stabilisation. Thus transfer of energy from PS to PPO can be seen in Figure 4.12 in the diminution of PS monomer emission at about 285 nm, and the replacement of PS excimer emission with a maximum at 330 nm by PPO emission with a maximum at 315 nm. Overlap of the PPO emission with PS emission makes estimates of energy transfer from intensity measurements unreliable; time-resolved studies of fluorescence decay and resulting lifetime data have been used for this purpose. The decay curves have been fitted by two exponential function in agreement with Equation 4.6. The fluorescence lifetime data of the mixture of PS with PPO in solution are presented in Table 4.2. We can observe emission from both PS excimer and PPO. The τ1 = 12 ns due to pure PS in solution decreases with PPO increase and the relative intensity of the short-lived PPO component increases in the presence of PS. The decrease in the lifetime of PS subjected to Stern-Volmer analysis gave the slope 5.2 x 103 l mol-1. Semi-quantitative analysis, assuming the mutual diffusion coefficient of 2 x 10-11 m2 s-1, gave the distance of interaction in the order of magnitude 103 nm. Quenching efficiencies of this magnitude would be possible only

119

Handbook of Polymers in Electronics

Figure 4.12 Fluorescence of polystyrene quenched by poly-(2,6-dimethyl-pphenylene oxide) in tetrahydrofuran at 298 K, PS concentration 2.7 x 10-3 mol l-1 (styrene units): (1) no PPO, (2) 0.43 x 10-4, (3) 0.83 x 10-4, (4) 4.3 x 10-4, (5) 1.74 x 10-4, (6) 2.61 x 10-4 mol l-1 (Reprinted from Polymer, Volume 23, M. Kryszewski, B. Wandelt, D.J.S. Birch, R.E. Imhof, A.M. North and R.A. Pethrick, Photo-energy transfer in polystyrene-polyphenylene oxide blends, 926, copyright 1982, with permission from Elsevier Science)

Table 4.2 Lifetime parameters of PS in tetrahydrofuran at 298 K, concentration 2.9 x 10-3 mol l-1 of styrene units, quenched by PPO; λexc. = 265 nm, λem. = 330 nm PPO concentration (10,000 mol l-1 of monomer units)

τ1 (ns)

τ2 (ns)

χ2

0.00

12.0

-

0.9

0.24

11.3

1.1

1.3

1.30

8.4

1.2

1.5

2.00

6.4

1.1

1.6

3.80

4.9

1.1

1.2

-

0.9

0.9

12 (no PS)

120

Luminescence Studies of Polymers if the donor energy is effectively available over the whole of the PS chain coil and all the monomer segments of the PPO chain are able to quench the excited PS. However, studies of fluorescence in the solid state of the system [45] have shown that energy transfer from PS to PPO is efficient in the solid blend, and at 15 wt.% of PPO, almost all the emission is from PPO. However, the magnitude of PS stabilisation in the presence of PPO, calculated by analysis of the polymer chain scission processes of the polymer mixture in solution, is smaller than it would be from the energy transfer characteristics by fluorescence lifetime measurements. This suggests that chain scission arises from energy more localised than the mobile excitons in the quenching of fluorescence. Jensen and co-workers [47] reported that the decrease in PS excimer fluorescence is greater than can be explained by absorption of PPO at the excitation wavelength. Kryszewski and co-workers [46] showed the molecular compatibility of PSPPO blends is important. After studies of the fluorescence intensity and fluorescence lifetime measurements during storage and annealing of the polymers and blends, they suggested that annealing and storage permit chain packing rearrangements that favour non radiative energy conversion and transfer processes, and in the case of blends lead to the PS to PPO transfer which quenches the PS emission and photodegradation. Amrani and co-workers [48] have used fluorescence measurements of energy transfer from donor to acceptor for studies of polymer compatibility. They labelled methyl methacrylate-ethyl methacrylate copolymer and/or methyl methacrylate-butyl methacrylate copolymer with donor-naphthalene and polymethyl methacrylate with acceptor-anthracene. The variation in the ratio of donor to acceptor fluorescence was plotted as a function of butyl methacrylate and ethyl methacrylate in the copolymer and gradual increase of the ratio corresponded to gradual transition from two-phase to a one-phase system. The fluorescence technique was found to be more sensitive to small changes of compatibility of the polymers. Winnik [49] used fluorescence measurements of transfer of the electronic excitation between donor-naphthalene and acceptor-pyrene chromophores attached to the same polymer chain for studies of thermoreversible phase separation of aqueous solutions of poly(N-isopropylacrylamide) (PNIPAM). Dilute solutions of the doubly labelled polymer PNIPAM were heated from 277 K to 313 K, and the fluorescence emission intensity of pyrene (integrated spectrum) was measured when the system was excited with 290 nm, donor excitation, and when excited with 328 nm, acceptor excitation. Non radiative energy transfer between excited naphthalene and pyrene occurred in aqueous solution of the polymer. The increase in intensity of pyrene fluorescence when the solution was excited at 290 nm, shown in Figure 4.13, is due to a phase separation process at lower critical solution temperature (LCST). When the LCST was reached, the phase separation into polymer-rich and polymer-lean phases occurred. It was concluded that the collapse of the polymer chain leading to densification of polymer phase is followed by domination of intramolecular contributions to the energy transfer process.

121

Handbook of Polymers in Electronics

Figure 4.13 Plot of pyrene emission intensity as a function of temperature for aqueous solution of poly-(N-isopropylacrylamide) (PNIPAM) labelled with naphthalene(N)-donor and pyrene(Py)-acceptor; PNIPAM-Py/366-N/50, 44 ppm in water. Wavelength excitation at 290 nm due to N excitation, at 328 nm due to Py excitation (Reprinted from Polymer, Volume 31, F.M. Winnik, Phase transition of aqueous poly-(N-isopropylacrylamide) solutions: a study by non radiative energy transfer, 2132, copyright 1990, with permission from Elsevier Science)

4.4.2 Fluorescence of Polymers in Gel State 4.4.2.1 Effect of Chain Conformation on Excimer Emission Excimer formation can be explained in terms of a combination of local rotational isomeric and longer range diffusional motions of the chain. In polymer solution we observed and discuss the role of both; in solid film any type of movement is generally restricted, and the mobility of a polymer in a gel state is rather between the solution and solid, a more precise description of which was published by de Gennes [50]. The freely-rotating chain of a polymer in dilute solution is represented by a statistical distribution of conformational structures of the chain. The resultant excimer fluorescence emission is a broad Gaussian band. The excimer fluorescence spectrum from a solid polymer results in a very broad spectrum which is characteristic of the distribution of conformations adopted by the polymer in the preparation process. This distribution depends upon the thermal history of the sample and the conditions used for the casting; these dependencies were presented by Frank and co-workers [26]. The complexity of photophysical processes accompanying the excimer formation and the inherent complexity

122

Luminescence Studies of Polymers of polymer systems (e.g., tacticity) together with the conformational sensitivity of the structures make for analytical problems. For some purposes, we can stimulate some of the local movement by photo and/or thermal cure [6, 50]. Isotactic polystyrene (iPS) in the gel state was used by Wandelt [7] to obtain a selected, homogeneous helical conformation of the vinyl polymer chain. A solution of iPS in benzyl alcohol (BA) quenched from 443 K to 273 K, with a rate of 0.42 K s-1, leads to the formation of a gel [51]. BA appeared to be a good solvent for the gel formation and transparent in the infrared regions where the characteristic absorption of the helical conformations occur; additionally it is not photoactive in the region of PS excimer emission [51]. The iPS chain of the fresh gel exists in the extended conformation defined by Atkins and co-workers [52] and Sundararajan and co-workers [53, 54], which can easily be converted to the three-fold helical conformation by annealing at a temperature higher than the Tg [52, 54], i.e., between 383 to 393 K. The sample of crystalline iPS was obtained from iPS/BA gel by heating at 293 K in the presence of nitrogen. The sample showed 68% crystallinity obtained from differential scanning calorimetry (DSC) thermograms. Characteristic three-fold helical conformation of the chain and spherulitic morphological forms were observed by infrared spectroscopy and microscopy [7]. The fluorescence emission spectra from the 68% crystalline iPS, and from atactic PS (aPS) film cast from chromophore, both excited at 257 nm are shown in Figure 4.14. One can see that the emission band from crystalline iPS is narrower then that of atactic PS, and the crystalline iPS band is shifted a few nm to the red. Examples of time-resolved fluorescence decays from the crystalline iPS are shown in Figure 4.15. We can observe similarities of the decays obtained for the 330 and 350 nm wavelengths of emission they are both due to excimer emission. Some differences in the decay can be observed for emission wavelength 310 nm, where possibly some monomer emission at the shorter time of the decay was collected. The recorded decay curves of emission intensity from the crystalline and atactic PS were analysed using a non linear least-squares analytical method [16], and the resulting photophysical parameters of the decays are shown in Table 4.3. A two-exponential fitting function was used for fluorescence decay from the crystalline iPS sample. Some amount of short-lived component in the parameters fitting the fluorescence decays from crystalline iPS correspond to some presence of monomer emission, which strongly overlaps the excimer band. The long-lived component of 21 ns is in the majority, more than 80% of the emission intensity is due to the component. However, this lifetime parameter does not change with the wavelength. There is evidence for homogeneity of the excimer structures in the crystalline polymer, especially if we compare the data with the same from aPS film, gathered in Table 4.3. A three-exponential function, in agreement to Equation 4.6, was found to be necessary to fit the fluorescence decay from aPS film. Considerable changes in the excimer lifetime parameters and their

123

Handbook of Polymers in Electronics

Figure 4.14 Steady-state fluorescence emission spectra of crystalline iPS obtained from gel and atactic PS film cast from chloroform at room temperature, excitation wavelength 257 nm (Reprinted from Polymer, Volume 32, B. Wandelt, Correlation of photophysical parameters with conformational structure of crystalline iPS and comparison with data of atactic PS, 2708, copyright 1991, with permission from Elsevier Science)

Figure 4.15 Fluorescence emission decays of crystalline iPS with excitation at 257 nm and different emission wavelengths: (1) 310 nm, (2) 330 nm, (3) 350 nm, (4) lamp pulse (Reprinted from Polymer, Volume 32, B. Wandelt, Correlation of photophysical parameters with conformational structure of crystalline iPS and comparison with data of atactic PS, 2709, copyright 1991, with permission from Elsevier Science) 124

Luminescence Studies of Polymers

Table 4.3 Fluorescence decay data for 68% crystalline iPS and for aPS λem (nm)

τ1 (ns)

τ2 (ns)

τ3 (ns)

B1 (%)

B2 (%)

B3 (%)

χ2

310

3.7±0.4

19.6±0.2

-

27.2

72.8

-

1.27

320

4.3±0.3

20.5±0.3

-

16.7

83.3

-

1.16

330

5.2±0.3

21.0±0.3

-

11.9

88.1

-

0.99

340

5.1±0.9

21.3±0.4

-

9.9

90.0

-

1.11

350

4.7±0.6

21.6±0.2

-

6.9

93.1

-

1.09

310

1.6±0.4

7.4±0.6

19.4±0.4

15.4

41.5

43.1

1.18

320

1.8±0.6

8.5±0.9

21.3±0.4

12.2

39.1

48.7

1.22

330

2.6±0.3

13.4±0.9

29.2±0.9

14.2

54.5

31.3

1.27

340

2.1±0.3

13.8±0.9

30.0±0.9

12.7

51.8

35.5

1.19

350

2.1±0.3

15.0±1.2

32.8±1.2

16.2

53.0

30.8

1.23

iPS

aPS

contributions with emission wavelength were observed in the case of aPS. This suggests that the excimeric band is very complex. The aPS film was cast from chloroform solution at room temperature, and it included microstructural domains with frozen polymer chains of different conformations. Some examples of fluorescence lifetime measurements for PS from different laboratories are gathered in Table 4.4. The lifetime data of PS differ from laboratory to laboratory, because the fluorescence decay curve reflects not only the polymer type and solvent, but the processes of preparation, i.e., temperature, quality of the solvent, etc. Basically, in the extended conformation as well as in the three-fold helical conformation of the iPS chain, the phenyls do not form excimer states, because the distance between parallel phenyls in the three-fold helical conformation is 0.665 nm, as reported by Natta [62] and Sundararajan and co-workers [54]. This distance is too large for excimer formation of the sandwich type (which have a distance of 0.3-0.35 nm). In this situation, the excimer can be formed only outside the crystalline region, e.g., in the region of the lamellar borders, because in these areas some deformation of the helical conformation of the PS chain makes the excimer structure formation more probable. The excitation energy can effectively migrate along the helical structure [60, 63] to the lamellar border, where

125

Handbook of Polymers in Electronics

Table 4.4 PS lifetime data from different laboratories Physical state of PS

Solvent

λexc, λem (nm)

Lifetimes (ns)

solution

CH2Cl2

266, 285

1.3

55

solution

CH2Cl2

266, 365

12.7

55

film

CH2Cl2

266, 365

22.0

55

solution

toluene

257, 340

15.3

4

56

solution

toluene

257, 290

0.9 2.0 14.6

<1.3

56

solution

toluene

257, 290

1.3 4.2 15.2

<1.3

56

solution

CH2Cl2

257, 270

0.72 13.5

<1.3

57

solution

CH2Cl2

257, 385

1.02 18.9

58

solution

C2H4Cl2

255, 330

15.5

59

solution Mw = 3,800

C2H4Cl2

255, 360

17

59

cyclohexane

250, 360

21

60

solution

C2H4Cl2

253, 333

1.85 17.5

61

solution

CH2Cl2

253, 333

<1 15.5

61

solution oligomers

χ2

Reference

the pre-excimer sites are formed. Although the energy migration coefficient in PS is lower than, for example, in PVN [64, 65], only high effectiveness of energy migration would explain the strong increase of the excimer fluorescence emission from the crystalline iPS in comparison with the amorphous aPS observed by David and co-workers [23, 24]. However, only higher effectiveness of the energy migration in the three-fold helical homogeneous conformation than in the extended conformation, can explain the 10 times higher excimer emission from the crystalline state than from the fresh gel of iPS (extended conformation of the chain) [51].

4.4.2.2 Effect of Phase Separation and Collapse Transition The phase separation process was observed in iPS gel [66, 67]. The gel was formed by cooling a solution of iPS in BA (0%-20% w/w) from a temperature of 443 K, in which clear solution is obtained, to 273 K, at a maximum cooling rate of 0.42 K s-1, which lead

126

Luminescence Studies of Polymers to a meta-stable state in which the chains adopt an extended conformation [52-54]. Generally, a polymer gel is a network of flexible chains, for which a two-dimensional view of the structure is similar to a fish net, but it is filled with solvent. We consider a gel to be a series of chains which are associated by a physical process and where the crosslinks are not strong, thus under any weak and finite stress the crosslinks will eventually split, and the long-term behaviour of the material will always be liquid-like. Thus, in this case, gelation is conceptually similar to a glass transition. It is not an equilibrium process, but it corresponds to the freezing of a certain number of degrees of freedoms [50]. There are no strict physical parameters in such a system, with respect to temperature or time of the process. In the case of the iPS/BA system, the BA is a poor solvent for iPS, and hence, phenyl-solvent interactions are unfavourable in comparison with intramolecular phenylphenyl interactions. Under these interactions, some segregation into small regions high in chain concentration and others which are solvent rich (i.e., polymer-rich and lean phases) may occur [50]. Then the polymer chain structure in the polymer-rich regions can be converted to the more stable form. The rate at which this process occurs depends upon the temperature and time used. Moreover, it has been found [51-54] that the solvent plays an important role in determining the conformational structure of the gel. Annealing allows the iPS chains to undergo phase separation into PS-rich regions, which can be transformed into a crystalline phase with a characteristic, three-fold helix conformation [7, 52, 54]. The phase separation step is determined at a fixed concentration by two factors: temperature and time. The annealing measurements were carried out at 318 K. This temperature was chosen because it lies sufficiently close to the Tg, reported to be 323 K [52], and allows effective production of the heterogeneous two-phase system. After annealing for 30 minutes at 318 K, characteristic morphological structures were observed using polarised optical microscopy and a crystallinity value of 23% was obtained from DSC thermograms [66]. The change of the steady-state emission spectrum of iPS gel during annealing at 318 K over a period of 30 minutes is shown in Figure 4.16. The changes observed indicate an increase in the intensity of a red-shifted excimer emission with time of heating. The fluorescence emission decays of a sample of iPS annealed for half an hour required two exponential functions to obtain a good fit of the experimental curve. The photophysical parameters obtained at a series of wavelengths are presented in Table 4.5 together with the same for freshly prepared gel. Changes in the lifetime parameters with emission wavelength suggest that the excimer emission band is rather complex and more than one type of excimer structure may be contributing to the emission spectrum. The lifetime parameter of about 26 ns in amount over 43%, at the emission wavelength 340 nm, shows that the long-living components of the decay correspond to the less energetic, red-shifted side of the fluorescence band. For comparison the lifetime parameters of the freshly prepared gel suggest clearly that one type of excimer structure is present: similar contribution at the emission wavelengths 330 and 350 nm, and the same lifetime of 19 ns (if we take into account the error for

127

Handbook of Polymers in Electronics

Figure 4.16 Changes of the fluorescence emission spectrum of iPS gel (10% w/w) with time of annealing at 318 K: (1) 0 min, (2) 4 min, (3) 9 min, (4) 14 min, (5) 20 min, (6) 30 min (Reprinted with permission from Macromolecules, 1991, 24, 5144, copyright 1991, American Chemical Society)

Table 4.5 Fluorescence decay parameters for iPS gel annealed for 0.5 hour at 318 K λem (nm)

τ1 (ns)

τ2 (ns)

B1 (%)

B2 (%)

χ2

iPS gel after 30 min of annealing at 318 K 310

8.6±0.5

18.5±0.6

59.6

40.4

1.35

330

11.3±0.3

24.5±0.2

60.1

39.9

0.98

340

12.7±0.4

25.9±0.6

56.8

43.2

1.26

310

5.4±0.3

18.3±0.2

31.6

68.4

1.17

330

5.9±0.5

18.5±0.2

19.8

80.2

1.15

350

11.6±1.3

19.6±0.5

18.4

81.6

1.19

Fresh iPS gel

both emission wavelengths). It was suggested by Atkins and co-workers [52] and Sundararajan and co-workers [54] that the extended conformation of iPS chains exists in the fresh gel and that it is stabilised by the solvent. If the quenching process of the pre-gelled solution is comparatively fast (of the order of minutes) the elimination of

128

Luminescence Studies of Polymers solvent cannot occur and the polymer is frozen in a themodynamically non equilibrium state. The rate at which the conformational transition occurs will depend on the conditions of the gel. When the polymer chain undergoes the volume phase transition, the excess of solvent is expelled from the polymer chains. Density measurements of the iPS gel during the volume phase transition were performed and correlation with the fluorescence lifetimes for the long-term storage process of the iPS gel at 283 K was presented by Wandelt and co-workers [67]. Changes in the lifetimes for excimer emission at the red side of the spectrum (at 330 and 350 nm) are shown in Figure 4.17. After 400 hours of storage, the gel exhibits an emission spectrum that contains a red-shifted excimer component, curve 2, with a fluorescence lifetime of 27 ns. After 528 hours of storage, a fluorescence lifetime of 29 ns is achieved. The long-lived fluorescence excimeric structure vanishes after longer periods of storage and a majority of component of 20 ns can be observed after 864 hours. This suggests that the long-living excimer component is due to an unstable transition state.

Figure 4.17 Dependence of the excimeric fluorescence lifetimes of iPS gel on time of storage at 283 K: (1) from emission decays at 330 nm and (2) at 350 nm: Excitation at 257 nm.

The equation for observed lifetime (Equation 4.2) can be rewritten: τ = ( kf + kn )−1

(4.20)

129

Handbook of Polymers in Electronics where kf is the fluorescence rate constant (which is independent of temperature), and kn is the rate constant of the radiationless processes. The latter appears to be responsible for the lifetime changes. The lifetime increase accompanies the decrease of the non radiative deactivation rate constant at around 400 hours of storage, when the new redshifted excimer emission appears. From measurements of time-resolved emission spectra for iPS/BA gel stored for 240 hours, shown in Figure 4.18, it may be concluded that two phases exist in the system. A high-energy excimer band with a maximum at 325 nm obtained with no delay after excitation and with a time interval of 9 ns can be identified with one phase; the same spectrum was obtained with time interval from 6 ns to 11 ns. A close correspondence of the band with the steady-state fluorescence spectrum from the fresh gel, the dotted curve, suggests correlation of the band with a short-lived component in the excimer spectrum. The red-shifted emission band with maximum at about 345 nm in Figure 4.18 was obtained using a time interval of 79 to

Figure 4.18 Time-resolved fluorescence emission spectra of iPS/BA gel stored for 240 hours at 283 K; λexc.=257 nm; slit width = 20 nm; obtained with time intervals: (1) 0-9 ns, (2) 6-11 ns, (3) 79-152 ns; dotted curve is steady-state fluorescence spectrum for iPS/BA gel after preparation. (Reprinted from Polymer, Volume 33, B. Wandelt, D.J.S. Birch, R.E. Imhof, R.A. Pethrick, Time-resolved excimer fluorescence studies as a probe of the coil collapse transition and phase separation in isotactic PS/BA gel, 3561, copyright 1992, with permission from Elsevier Science)

130

Luminescence Studies of Polymers 152 ns. This red-shifted band appeared after phase separation into polymer-lean and polymer-rich phases, and it can be ascribed to the polymer-rich phase. After longer times of storage, the red-shifted excimer structure of lifetime 29 ns was transformed into an excimer structure of lifetime 20 ns, as seen in Figure 4.17. To observe the three time-resolved bands, due to different excimeric structures in an iPS gel, a model iPS gel was prepared [51] by placing a crystalline, previously nucleated material into a hot solution of iPS in BA, which was immediately quenched to a solid gel. Time-resolved emission spectra shown in Figure 4.19 indicate the existence of three excimer states. The high-energy band 1 obtained with no delay from excitation pulse, with a maximum at about 325 nm, corresponds to spectra 1 and 2 in Figure 4.18. This band, obtained with time interval 0-5 ns, corresponds to a short fluorescence lifetime, 17 ns, which is characteristic for polystyrene in solution. The band 2 with a maximum at 330 nm, obtained at a time interval of 7-16 ns, is due to an excimer structure clearly identified as characteristic for crystalline iPS [7] of 21 ns lifetime. This was completely invisible after 240 hours of storage but it was observed [67] after 744 hours of storage when the

Figure 4.19 Time-resolved fluorescence emission spectra for modelled fresh iPS gel with suspended crystalline material. Spectra obtained with time intervals: (1) 0-5 ns, (2) 7-16 ns, (3) 71-149 ns (Reprinted with permission from Macromolecules, 1991, 24, 5144. Copyright 1991, American Chemical Society)

131

Handbook of Polymers in Electronics polymer chain was transformed to the three-fold helical conformation. Figure 4.19 pertains to the crystalline material suspended in the last step of gel preparation having a memory of chain conformation. The band (3) in Figure 4.19 was obtained with a time interval of 71-149 ns, with a maximum at about 345 nm, and is due to the redshifted excimer fluorescence of 29 ns lifetime. The red-shifted fluorescence component was identified as an unstable excimer structure which finally undergoes transformation into a stable excimeric structure with emission at 330 nm, represented by band 2 in Figure 4.19. The time-resolved spectrum of the component 2 was shown [66] to be the same as the steady-state fluorescence spectrum of the 68% crystalline iPS gel. A linear combination of components 2 and 3 using the pre-exponential parameters obtained from the analysis of the decay at 330 nm emission wavelength in Table 4.5 has been used to predict the theoretical spectrum for the two-phase excimer emission [66]. The total intensity in Figure 4.20, curve 1 was calculated in agreement with Equation 4.21.

Figure 4.20 The spectrum estimated using the equation I = 0.60(comp.2 + 0.40(comp.3) is shown as curve 1; the steady-state fluorescence emission spectrum of iPS gel after annealing for 30 min at 318 K as curve 2; comp.2 and comp.3 correspond to spectra 2 and 3 in Figure 4.19. (Reprinted from Polymer, Volume 33, B. Wandelt, D.J.S. Birch, R.E. Imhof and R.A. Pethrick, Correlation of steady state and time-dependent studies of a two-phase iPS gel, 3556, copyright 1992, with permission from Elsevier Science)

132

Luminescence Studies of Polymers IE = 0.6(comp.2) + 0.4(comp.3)

(4.21)

where comp.2 and comp.3 corresponds to spectra 2 and 3 in Figure 4.19. This was compared with the experimental spectrum of the annealed iPS gel. Comparison of the experimental and theoretical curves in Figure 4.20 indicates that differences are observed at short wavelength due to the monomer and to some degree the extended conformation contributing to the spectrum, contributions which are not included in the theoretical curves. The generally good agreement between the theoretical and experimental curve supports the assumption that the excimer emission arises from a two-phase system. Description of the excimer photophysics for a two-phase system presented by Wandelt and co-workers [66] are based on both the two-phase model assumptions and the experimental results. The two-phase model describes the results of the experimental studies of photoenergy migration in heterogeneous solid-state polymer blends by Frank and collaborators [68, 69]. Tao and Frank [69] used three-dimensional electronic excitation transport to interpret the ratio of excimer to monomer fluorescence for poly-2vinylnaphthalene with polycyclohexyl methacrylate. The assumptions of the two-phase model are: •

There is no energy transfer between the different domains, and

•

The exciton can migrate inside a particular phase but becomes trapped at interfaces without escaping.

The two-phase photophysical process described by Wandelt and co-workers [66] was similar to that previously used for a single phase [3]. It is worth considering that the emission arises from not only different states but also different phases, and the processes are simultaneous and independent. Thus the two-phase system excimeric fluorescence can be calculated: IE (two-phase) = IE (1) + IE (2)

(4.22)

where 1 and 2 refer to particular phases. The generally good agreement obtained in Figure 4.20 between the experimental spectrum and that arising from the two-phase model calculated by Equation 4.22 supports the model assumptions. Investigation of the fluorescence emission from the crystalline iPS shows that it is very effective, but the three-fold helical conformation of the chains ensures that the parallel phenyl groups are separated by a distance of 0.665 nm [54] and packed into a lamellar structure about 0.57 nm long. The distance between phenyl groups is large for excimer formation, but the very regular and periodic chain microstructure incorporated into the lamellar macrostructure gives the chance for efficient energy migration. The structure of the crystalline phase favours exciton energy migration along the helix to the lamellae boundary

133

Handbook of Polymers in Electronics where it is finally trapped in the interface region. It has been established [7] that all of the chromophores associated with the crystalline (and only 20% of those in the amorphous) state participate in the excimer emission. Fluorescence studies of volume phase transition of polyacrylamide (PAAM) in mixed acetone/ water solvent, with incorporated dansyl group and pyrenyl probe were reported by Hu and co-workers [70, 71]. They observed an increase in the fluorescence lifetime of the probe with increase of acetone content in the solvent. They reported that gradual increase of fluorescence lifetime accompanies the volume phase transition of the PAAM gels from the swollen to collapsed state with increasing hydrophobicity of the microenvironment. Picosecond fluorescence studies were applied by Winnik and co-workers [72] for studies of temperature-induced phase transition of pyrene-labelled hydroxypropylcellulose (HPCPy) in water. Temperature dependence of the fluorescence emission ratio of excimer to monomer emission (IE/IM) showed a significant increase of excimer emission in a temperature range 283-313 K, then a decrease to a constant value at 319 K. Two excimer bands were observed when time-resolved spectroscopy was used: i) a broad, structureless band with a maximum at 420 nm and a corresponding lifetime of 250 ps and ii) the wellknown band of pyrene excimer, with a maximum at 470 nm and a lifetime of 68 ns. In the initial time region, 0-150 ps, monomer emission was observed, with a simulation by a superposition of three components (377, 398 and 421 nm). They observed only one excimer emission above the LCST and that was with a maximum at 470 nm. They concluded that the LCST implies a complete disruption of the ordered microstructures, which were created in cold water. An interesting application of fluorescence studies was reported by Huang and co-workers [73]. They studied intermolecular complex formation between mesogenic terphenyldiimide moieties of a thermotropic liquid-crystalline (LC) polyimide (P-11TPE). The temperature dependence of the complex fluorescence peak wavelength is shown in Figure 4.21. It shifts slightly to a shorter wavelength below the Tg and then shifts to a longer wavelength around crystallisation, then again shifts to shorter wavelength during a complicated crystal-crystal transition around 200 °C, and around the phase transition from crystalline to smectic phase at 240 °C. An Arrhenius-type plot for changes in fluorescence lifetime of the complex during heating is shown in Figure 4.22. The lifetime decreases with temperature until the Tg after which it slightly increases to again decrease gradually with temperature until the material reaches the smectic phase. The decrease of lifetime with temperature seen in Figure 4.22 indicate that the radiationless deactivation process which accompanies the observed fluorescence is in agreement with Equation 4.20 and is involved in the complicated phase transitions of LC polyimides. It was also concluded that these apparent changes in wavelength of fluorescence and lifetime behaviour are dependent on temperature and indicate the complexity of the nature of intermolecular complexes and radiationless deactivation processes in various phases of the thermotropic LC polyimides.

134

Luminescence Studies of Polymers

Figure 4.21 Temperature dependence of the fluorescence peak wavelength during heating of thermotropic liquid-crystalline polyimide. Excitation at 320 nm. (Reprinted from Polymer, Volume 40, H.W. Huang, T.I. Kaneko, K. Horie and J. Watanabe, Fluorescence study on intermolecular complex formation between mesogenic terphenyldiimide moieties of a thermotropic liquid-crystalline polyimide, 3826, copyright 1999, with permission from Elsevier Science)

Figure 4.22 Arrhenius-type plot for the change in lifetime of fluorescence of liquidcrystalline polyimide during heating (Reprinted from Polymer, Volume 40, H.W. Huang, T.I. Kaneko, K. Horie and J. Watanabe, Fluorescence study on intermolecular complex formation between mesogenic terphenyldiimide moieties of a thermo-tropic liquid-crystalline polyimide, 3826, copyright 1999, with permission from Elsevier Science)

135

Handbook of Polymers in Electronics

4.5 Conclusions Many of the significant features observed in the photophysics and photochemistry of polymer systems have been shown to have their origin in the way the behaviour of electronically excited states are affected by molecular motion. Most of the phenomena involve a competition between physical processes leading to non radiative deactivation and emission of the excitation energy, and thus are most easily observed in the characteristics of fluorescence. The competition between processes can be controlled by either thermodynamic or kinetic parameters. Which of the parameters is of prime importance depends largely on: (i) the environment of the luminescent species, (ii) if the luminescent species is attached to the polymer or detached from the polymer, or (iii) if the luminescent species is randomly distributed along the polymer chain or not. For example, measurement of the relative intensities yields information concerning the ordering of the polymer chain and similarly studies of the kinetics of fluorescence decay yield information on energy migration and segmental motion. The changes in fluorescence intensity and fluorescence lifetime around phase transition temperatures can yield information on packing pattern and variation with thermal treatment of the polymer during the phase transition. The excited state properties, principally fluorescence, of probe molecules incorporated in the polymer chain can be used to a large extent to study molecular motion, order, and energy migration in polymeric systems. The present review has revealed how fluorescence methodologies can provide an insight into the nature of the intramolecular and intermolecular interactions, which are responsible for thermally initiated variations as well as the formation of microstructures and the morphology of a polymeric medium.

Acknowledgements This work was supported by KBN grant No. 3 T09B 058 14 (Poland).

References 1.

J.B. Birks, Organic Molecular Photophysics, John Wiley & Sons, London, 1973.

2.

M. Klessinger and J. Michl, Excited States and Photochemistry of Organic Molecules, VCH Publishers, Inc., New York, NY, USA, 1995.

3.

J.E. Guillet, Polymer Photophysics and Photochemistry, Cambridge University Press, Cambridge, UK, 1985.

136

Luminescence Studies of Polymers 4.

D. Phillips in Photochemistry and Polymeric Systems, Eds., J.M. Kelly, C.B. McArdle and M.J.de F. Maunder, The Royal Society of Chemistry, Cambridge, UK, 1993, 87.

5.

P.J. Flory, Principles of Polymer Chemistry, Cornel University Press, Ithaca, NY, USA, 1953.

6.

R.A. Pethrick, B. Wandelt, D.J.S. Birch, R.E. Imhof and S. Radhakrishnan in Photochemistry and Polymeric Systems, Eds., J.M. Kelly, C.B. McArdle and M.J. de F. Maunder, The Royal Society of Chemistry, Bath, UK, 1993, 68.

7.

B. Wandelt, Polymer, 1991, 32, 2707.

8.

A. Ledwith in Photochemistry and Polymeric Systems, Eds., J.M. Kelly, C.B. McArdle and M.J.de F. Maunder, The Royal Society of Chemistry, Cambridge, UK, 1993, 1.

9.

K. Nakashima, J. Duhamel and M.A. Winnik, Journal of Physical Chemistry, 1993, 97, 10702.

10. C. McDonagh, B.D. MacCraith and A.K. McEvoy, Analytical Chemistry, 1998, 70, 45. 11. P. Turkewitch, B. Wandelt, G.D. Darling and W.S. Powell, Analytical Chemistry, 1998, 70, 2025. 12. Th. Förster and K. Kasper, Zeitschrift für Physicalische Chemie, Neue Folge, 1954, 1, 275. 13. H. Beens, H. Knibbe and A. Weller, Journal of Chemical Physics, 1967, 47, 1183. 14. H. Knibbe, D. Rehm and A. Weller, Zeitschrift für Physicalische Chemie, Neue Folge, 1967, 56, 95. 15. J.R. Lakowicz and J. Gryczynski in Topics in Fluorescence Spectroscopy Techniques, Ed., J.R. Lakowicz, Plenum Press, New York, NY, 1991, Volume 1, 293. 16. D.J.S. Birch and R.E. Imhof in Topics in Fluorescence Spectroscopy Techniques, Ed., J.R. Lakowicz, Plenum Press, New York, 1991, Volume 1, 1. 17. C.A. Parker and W.T. Rees, Analyst, 1960, 85, 587. 18. W.H. Melhuish, Journal of Physical Chemistry, 1961, 65, 229. 137

Handbook of Polymers in Electronics 19. R. Rusakowicz and A.C. Testa, Journal of Physical Chemistry, 1968, 72, 2680. 20. F. Hirayama, Journal of Chemical Physics, 1965, 42, 3163. 21. C.E. Hoyle and J.E. Guillet, Macromolecules, 1979, 12, 956. 22. D. Phillips, A.J. Roberts and I. Soutar, Journal of Polymer Science, Polymer Physics, 1980, 18, 2401. 23. C. David, M. Piens and G. Geuskens, European Polymer Journal, 1972, 8, 1019. 24. C. David, M. Piens and G. Geuskens, European Polymer Journal, 1972, 8, 1291. 25. L.A. Harrah, Journal of Chemical Physics, 1972, 56, 385. 26. C.W. Frank and L.A. Harrah, Journal of Chemical Physics, 1974, 61, 1526. 27. A.C. Somersall and J.E. Guillet, Macromolecules, 1973, 6, 218. 28. J. Aspler and J.E. Guillet, Macromolecules, 1979, 12, 1082. 29. R.F. Cozzens and R.B. Fox, Journal of Chemical Physics, 1969, 50, 1532. 30. Y. Nishijima, K. Mitani, S. Katayama and M. Yamamoto, Report on Progress in Polymer Physics, Japan, 1970, 13, 421. 31. B. Wandelt and J. Szumilewicz, Polymer, 1987, 28, 1791. 32. E.A. Abuin, E.A. Lissi, L. Gargallo and D. Radic, European Polymer Journal, 1980, 16, 793. 33. A.J.H. Al-Watter and M.D. Lumb, Chemical Physics Letters, 1971, 2, 374. 34. E.A. Chandross and C.J. Dempster, Journal of the American Chemical Society, 1970, 92, 3586. 35. A. Itaya, K. Okamoto and S. Kubayashi, Bulletins of the Chemical Society of Japan, 1976, 49, 2083. 36. A.M. North and J. Soutar, Journal of the Chemical Society Faraday Transactions I, 1972, 68, 1101. 37. C. Cuniberti and A. Perico, European Polymer Journal, 1977, 13, 369. 38. M.A. Winnik, A.E.C. Redpath, P. Svirskaya and A. Mar, Polymer, 1983, 24, 473.

138

Luminescence Studies of Polymers 39. A.E.C. Redpath and M.A. Winnik, Journal of the American Chemical Society, 1982, 104, 5604. 40. Th. Förster, Discussions of the Faraday Society, 1959, 27, 9. 41. D.L. Dexter, Journal of Chemical Physics, 1953, 21, 836. 42. B. Wandelt, Polymer Bulletin, 1981, 4, 199. 43. B. Wandelt, European Polymer Journal, 1986, 22, 755. 44. B. Wandelt, J. Jachowicz and M. Kryszewski, Acta Polymerica, 1981, 32, 637. 45. M. Kryszewski, B. Wandelt, D.J.S. Birch, R.E. Imhof, A.M. North and R.A. Pethrick, Polymer, 1982, 23, 924. 46. M. Kryszewski, B. Wandelt, D.J.S. Birch, R.E. Imhof, A.M. North and R.A. Pethrick, Polymer Communications, 1983, 24, 73. 47. T. Jensen and J. Kops, Journal of Polymer Science, Polymer Chemistry, 1981, 19, 2765. 48. F. Amrani, J.M. Hung and H. Morawetz, Macromolecules, 1980, 13, 649. 49. F.M. Winnik, Polymer, 1990, 31, 2125. 50. P.G. de Gennes, Scaling Concepts in Polymer Physics, Cornell University Press, London, UK, 1979. 51. B. Wandelt, D.J.S. Birch, R.E. Imhof, A.S. Holmes and R.A. Pethrick, Macromolecules, 1991, 24, 5141. 52. E.D. Atkins, M.J. Hill, D.A. Jarvis, A. Keller, E. Sarhene and J.S. Shapiro, Colloid Polymer Science, 1984, 262, 22. 53. P.R. Sundararajan, Macromolecules, 1979, 12, 575. 54. N.J. Tyrer, L.T. Bluhm and P.R. Sundararajan, Macromolecules, 1984, 17, 2296. 55. M.O. Gupta, A. Gupta, J. Horwitz and D. Kliger, Macromolecules, 1982, 15, 1372. 56. J. Soutar, D. Phillips, A.J. Roberts and G. Rumbles, Journal of Polymer Science, Polymer Physics, 1982, 20, 1759.

139

Handbook of Polymers in Electronics 57. D. Phillips, A.J. Roberts, G. Rumbles and J. Soutar, Macromolecules, 1983, 16, 1597. 58. K.P. Ghiggino, R.D. Wright and D. Phillips, Journal of Polymer Science, Polymer Physics, 1978, 16, 1499. 59. T. Ishii, T. Handa and S. Matsunaga, Macromolecules, 1978, 11, 40. 60. H. Itagaki, K. Horie, I. Mita, M. Washio, S. Tagawa and Y. Tabata, Journal of Chemical Physics, 1983, 79, 3996. 61. F. Heisel and G. Laustriat, Journal of Chimie Physique, 1969, 66, 1881. 62. G. Natta, Macromolecular Chemistry, 1960, 35, 94. 63. J. R. MacCallum, European Polymer Journal, 1981, 17, 209. 64. A.M. North and D.A. Ross, Journal of Polymer Science C Symposia, 1976, 55, 259. 65. S.B. Dew, R.Y. Lockhead and A.M. North, Discussions of the Faraday Society, 1970, 49, 244. 66. B. Wandelt, D.J.S. Birch, R.E. Imhof and R.A. Pethrick, Polymer, 1992, 33, 3552. 67. B. Wandelt, D.J.S. Birch, R.E. Imhof and R.A. Pethrick, Polymer, 1992, 33, 3558. 68. M.A. Gashgari and C.W. Frank, Macromolecules, 1988, 21, 2782. 69. W.C. Tao and C.W. Frank, Macromolecules, 1990, 23, 2782. 70. Y. Hu, K. Horie, H. Ushiki, T. Yamashita and F. Tsunomori, Macromolecules, 1993, 26, 1761. 71. Y. Hu, K. Horie, H. Ushiki, F. Tsunomori and T. Yamashita, Macromolecules, 1992, 25, 7324. 72. F.M. Winnik, N. Tamai, J. Yonezawa, Y. Nishimura and I. Yamazaki, Journal of Physical Chemistry, 1992, 96, 1967. 73. H.W. Huang, T.I. Kaneko, K. Horie and J. Watanabe, Polymer, 1999, 40, 3821.

140

5

Polymers for Light Emitting Diodes A. Bolognesi and C. Botta

5.1 Introduction Polymer science, which is a young branch of chemistry, has been the subject of great development both as a basic and applied science in the last 30 years. In the first instance, the great interest in polymers was due to the good mechanical properties associated with materials having low density and low processability costs. Later the possibility of combining new chemical functions in a backbone opened new fields of applications for macromolecules. So structures with specific uses were developed creating new interfaces with technological fields that, at the beginning, were unusual for polymers: pharmacology, biology, medicine, optics, electronics. In particular, the development of polymers for electronics is still an open field where polymers are used not only as insulators, but where the electronic properties of conjugated macromolecules can be tailored for specific applications. The problem which confined the use of polymers only in the area of insulators was overcome in 1977 with the discovery by McDiarmid and Shirakawa [1] that iodinedoped polyacetylene possesses metallic conductivity. After this discovery, many polymeric structures were synthesised with the aim of improving both the electrical conductivity and the stability of the materials. At the end of the 1980s, work by Tang and Van Slyke [2] on the electroluminescence (EL) of aluminium 8hydroxyquinoline (Alq3) stimulated a great amount of work worldwide on the generation of light by electrical excitation in organic materials. It was in 1990 that Friend and coworkers [3] presented the first electroluminescent polymeric device able to emit at 2.2 eV with an external quantum efficiency of 0.05%. The light emitting diode (LED) was formed by a single layer of PPV sandwiched between an indium tin oxide (ITO)-coated glass and an aluminium cathode which was vacuum evaporated on the top of the polymer film. Injection of electrons takes place at the cathode, while holes are injected at the anode. In the following years, the number of scientific contributions on polymeric LEDs increased consistently. Different strategies were developed, all addressed to understanding the basic mechanism of the injection of opposite charges in the polymers, their travelling in the thickness of the layer, and their radiative recombination, all these factors being fundamental for the increase of efficiency and lifetime.

141

Handbook of Polymers in Electronics The great effort that is still pursued in this field is responsible for both the optimisation of new polymeric structures and for the preparation of new and more reliable devices. The fast development of this branch of polymer science has stimulated the interest of the industrial world and nowadays there are several small, medium and big enterprises, both in the chemistry area and in the microelectronics sector, developing LED prototypes. It has been recently reported that Philips will be able to produce backlighting and segmented displays very soon [4], Seiko-Epson and Cambridge Displays Technology are developing a technology using ink-jet printing of electroluminescent material for colour patterning. The development in this area is extremely fast; in reference [5] some of the web sites continuously updating in this field are reported. It is suggested that the reader visit these sites for the very latest news.

5.2 The Physics of Electroluminescent Devices 5.2.1 The Physics of Conjugated Polymers Several reviews have been reported on the electronic properties of conjugated polymers [69] as well as on the physical mechanisms occurring in LED devices [10-12]. Here, a few aspects of this very wide field are summarised, focusing on the properties of the electronic excitations which are more relevant in the physics of an electroluminescent device. Conjugated polymers possess strongly anisotropic electrical and optical properties since the π-electron delocalisation occurs along the chain direction. An ideal conjugated polymer chain possesses a one-dimensional structure which, due to the Peierls instability, leads to the electronic structure of a monodimensional semiconductor with an energy gap typically in the 1.5-3.5 eV range. The Su, Schrieffer and Heeger theory, often referred to as SSH, [6, 13], where electron-electron (e-e) interactions are neglected, describes the properties of the polymeric semiconductors by introducing a strong electron-phonon interaction. According to this theory, for non degenerate ground state polymers, injection of charges (by chemical doping, through an electrode or by photoinduced interchain electron-hole (e-h) separation) leads to the formation of singly charged polarons or doubly charged bipolarons. Neutral species (exciton-polarons) can be generated by photoexcitation via e-h separation or by the fusion of two polarons with opposite charges, obtained via charge injection from the electrodes, and are responsible for photoluminescence (PL) and EL, respectively. Both the charged and the neutral excitations are stabilised through a local modification of the bond alternation. This produces two localised electronic states within the forbidden energy gap. The energy position of these levels, which are symmetrically displaced from the midgap, depends on the spatial extent of the defect (polaron or bipolaron) [14]. The more extended the defect, the closer the levels to the midgap. The optical transitions occurring between these levels and the conduction and

142

Polymers for Light Emitting Diodes

Figure 5.1 Scheme of the generation and radiative recombination of the singlet exciton S by photoexcitation (a), and by charge injection (b) mechanisms

valence bands (see Figure 5.1) account for the sub-gap optical absorption bands observed by chemical doping, by photoinduced absorption, and by charge injection. According to this model, the PL is due to the spin-allowed transition from the upper to the lower of the polaron-exciton levels. However, the PL emission is generally observed at energies higher than the spacing of the two polaronic levels, measured with photoinduced absorption techniques. Indeed, there are several areas of the physics of conjugated polymers which cannot be explained without taking into account the e-e interaction effects. In order to describe the phenomena related to the neutral excited states of conjugated polymers (which are primarly involved in the physics of LEDs), models generally applied to the physics of molecular organic crystals [15] are more appropriate. In conjugated polymeric systems, the coulombically bound eh couple can be described as a Frenkel exciton (localised within a molecule) or a Wannier exciton (extended over many molecular units), according to its binding energy or degree of localisation [16-18]. When interchain interactions are strong, charge transfer excitons are formed, as excimers and exciplexes [15, 19, 20]. Only singlet excitons, S, are created by photon absorption (direct triplet generation is forbidden from the spin selection) and recombine radiatively with a certain probability, expressed by the PL quantum efficiency, ηPL (the ratio between the number of photons emitted and the number of photons absorbed). When the exciton is created by the coalescence of two polarons, the probability of forming a singlet exciton is 1/4 while

143

Handbook of Polymers in Electronics for a triplet exciton (which does not contribute to the EL emission) it is 3/4 [21, 22]. After an exciton has been created, it moves in the material by hopping or by resonant transfer processes [15, 23-26]. If more than one material is used (blends or heterostructures), by a proper selection of the materials, the exciton migration processes may red shift the emission and also increase the emission efficiency [25-27]. If only one polymer is used, exciton migration effects decrease the ηPL of the material since the probability of exciton quenching by impurities is increased. The degree of order and crystallinity of the material plays an important role in these processes as well as in determining the charge mobilities [28, 29].

5.2.2 The Physics of the Device In this section, some basic information on operative guidelines is provided for readers who are new to this field. LEDs are devices that transform electrical signals into optical signals. The typical structure of a single-layer polymeric LED is shown in Figure 5.2. A glass substrate coated by ITO is used as a transparent (positive) electrode. A polymer layer is deposited onto the substrate by a spin coating technique. On top of the polymer is deposited, by vacuum evaporation, a metallic layer which forms the negative electrode. By applying a bias voltage at the electrodes, light emission is obtained through the transparent electrode.

Figure 5.2 Scheme of a single layer LED

The working mechanism of the device is shown schematically in Figure 5.3, where the levels of the electrodes and the HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) of the polymers are reported, together with the intragap levels of the polarons and the exciton-polaron. The position of the HOMO and

144

Polymers for Light Emitting Diodes

Figure 5.3 Working mechanism of a LED: P+ and P- are positive and negative polarons; φ is the metal work function; EA is the electronic affinity; IP is the ionisation potential; Δe and Δh are potential barriers for negative and positive charges; HOMO is the highest occupied molecular orbital; LUMO is the lowest unoccupied molecular orbital.

LUMO of the polymer, relative to the Fermi levels of the electrode, are determined by the ionisation potential (IP), the electronic affinity (EA) of the polymer and the workfunction (φ) of the metals, as shown in Figure 5.3. The working mechanism of a LED consists of four main steps: (1) By applying a bias voltage, charges of opposite signs are injected in the active material and form positive and negative polarons. (2) The charges P+ (P-) move in the material towards the negative (positive) electrode, driven by the applied electric field. (3) Polarons of opposite sign couple to generate the excitons, S. (4) S recombine radiatively by photon emission. The efficiency of an EL device (ηEL) is the ratio between the number of the emitted photons and the number of injected charges. It can be expressed as

ηEL = γ ηEX ηPL

(5.1)

where γ is a factor dependent on the double charge injection (γ = 1 for perfect charge balance), ηEX is the efficiency of singlet exciton generation (maximum ηEX is 1/4) and ηPL is the measurable PL quantum efficiency [30].

145

Handbook of Polymers in Electronics Some considerations can be made on each of the four steps in order to increase the device performances: (i) Charge injection. The number of injected positive and negative charges must be similar. This is obtained if the electrode metals have φ such that the potential barriers for positive (Δh) and negative (Δe) charge injection are similar (Δe≈Δh). Positive electrodes generally used are ITO, PANI [31, 32] and silver, with φ in the 4.5-4.8 eV range [33-36]. Recently, the introduction of a conducting polymer interfacial layer between the ITO and the active polymer has been reported to provide φ determined by the intermediate polymer [37]. For the negative electrode, calcium (2.9 eV), indium, aluminium (4.2-4.3 eV), silver, copper and several alloys are used [38]. Calcium usually gives the best results since the device efficiency can be increased by reducing Δe, which controls the injection of the minority carriers [33, 34, 39]. (ii) Charge transport. A good balance of positive and negative charge transport is necessary. Unfortunately, conjugated polymers are better hole than electron transporters, since P- are easily trapped. The high trapping of negative charges is responsible for bias-dependent efficiencies with low efficiency near the onset voltage. Improvement of charge transport can be obtained by inserting an electron transporter layer (ETL) and a hole transporter layer (HTL) between the active material and the electrodes. The introduction of an ETL also has the advantage of reducing the hole currents, as it acts as a hole blocking layer, thus increasing the charge balance [40, 41]. iii) Exciton generation. The efficiency of exciton generation can be optimised by increasing the probability of polaron coalescence, through a good balance of positive/ negative charges in a restricted active volume (heterostructures) [40]. (iv) Radiative recombination. The efficiency of radiative recombination of the S exciton is directly related to ηPL, and can be maximised by reducing the quenching centres and by localising the exciton (polymer blends, conjugated copolymers, conjugatednon conjugated copolymers [42, 43]). The active area of the polymer must be far from the electrodes, where a high concentration of quenching defects are present [44-46]. This is generally obtained by confining the active layer at the interface between two materials (heterostructures). Another way to increase the radiative recombination probability is to transfer the excitation from one material to another (more efficient) material (lower gap polymer or dye) by resonant energy processes such as Foerster transfer. These processes also red shift the emission. Factors which limit the radiative recombination are an excess of density of charges (exciton-polaron quenching) and quenching by the electric field [47-49]. These effects can be reduced by improving the charge balance and by reducing the working bias voltage (V).

146

Polymers for Light Emitting Diodes

5.2.3 LED Characterisation In order to characterise a device, the I-V (current intensity versus bias voltage) characteristic together with the EL-V (EL intensity versus bias voltage) curves are measured, and, through their ratio, the external efficiency of the device is deduced taking into account the geometrical factors [50]. In Figure 5.4 a typical behaviour of a LED is reported. When a good charge balance is obtained in the device, the two curves are similar and possess the same onset voltage, Von. If the current onset is at lower voltages than the EL (as in the case of Figure 5.4), an excess of positive charges is usually responsible for biasdependent efficiencies. The spectral shape of the EL is then compared with the PL of the active material. The same spectral shape is expected when the active region is in the bulk of the material, while emission influenced by chemical degradation of the polymer [44] or by interference effects [45] is often observed when the emitting material lies at the polymer-electrode interface. Many LEDs have been studied in order to gain information on the detailed physical mechanisms involved in their operation. The studies on fluorescence quenching induced by electric fields, as well as internal electric field distribution analysis, have provided deep insight into these phenomena [16, 21, 47, 51-59].

Figure 5.4 Typical I-V and EL-V curves for a single layer LED based on poly(3-alkylthiophene). The onset voltage (Von) is shown.

147

Handbook of Polymers in Electronics

5.3 Polymeric Structures for LED The tailoring of polymeric structures plays a significant role in the development of this new technology. Polymer chemists have synthesised from the beginning new structures in order to obtain the desired properties of conjugated macromolecules. The role of many factors are almost completely understood, but there is still a lot of work to do because improvements both in lifetime and in electroemission efficiencies are possible. The structural control can be extended to several physicochemical parameters: •

The band gap of a conjugated polymer is responsible for the PL and EL peak position,

•

The EA and the IP of the polymer strongly affect the charge injections from the electrodes into the polymers, and this is related to the PL and EL efficiency [33, 34],

•

The solid state packing of the macromolecules influences the stability and the emission efficiencies [42],

•

The surface polarity of the polymer is responsible for the adhesion between the active polymer and the electrodes, which is an important factor in charge injection, and

•

Resistance to oxidation and to temperature of the active polymeric layer are fundamental to the lifetime of the device.

Optimisation of all the above parameters is extremely difficult on a chemical synthesis basis, but the great efforts provided by many groups in this area have lead to exciting results for both improvements in macromolecular chemistry and in technological application.

5.3.1 Polyphenylenes 5.3.1.1 PPV PPV (Figure 5.5) was one of the first polymers studied, for its good PL and EL efficiencies. Several synthetic routes have been reported for the preparation of this polymer. Its insolubility is a great problem in the preparation of high molecular weight materials. For example, step growth polymerisation, such as Wittig condensation between terephthaldicarboxaldehydes and arylene-bisphosphylidenes gives very low molecular weight polymers, because when the growing chain is formed by 6-10 repeating units, it becomes insoluble and chain growth stops. The low molecular weight polymers synthesised in this way cannot be used for thin film preparation because they are insoluble. To overcome these difficulties, a soluble form of a precursor was synthesised. A thin film of the precursor was formed by spin coating techniques; a subsequent thermal treatment of the film lead to the insoluble conjugated thin film of PPV.

148

Polymers for Light Emitting Diodes

n

Figure 5.5 Structure of PPV

The Wessling method, consisting of the preparation of a sulfonium precursor polymer, was one of the first reported [60]. In Figure 5.6 the procedure followed to prepare PPV according to [61] is reported. Several modifications of this general procedure have been introduced, consisting of changing the conditions of the transformation process [62-64] and/or the nature of the chemical species to be eliminated [65-67]. Another procedure, introduced by Gilch [68] in 1966, consists of the polymerisation of dichloro-p-xylene with potassium tert-butoxide in organic solvents.

XH2C

X=

CH2X

S Cl- +

or

+ ClX = S(CH3)2

(0.4 M / MeOH) 1. -OH (0.4 M / H2O), 1h, 0°C, N2 2. H+ (0.4 M / H2O), to pH = 5-7 3. Dialysis, 3 days (dist. H2O), r.t., N2

H 2C

X (I) X =

300 °C, 1 h

CH2

S Cl- +

n

vacuum

+ Cl(II) X = S(CH3)2 n

Solution in MeOH H 2C

PPV 200 °C, 2 h

CH (III)

OMe

n

HCl (g)

Solution in CHCl3

Figure 5.6 Synthetic procedure for PPV (Reprinted from Synthetic Materials, 41-43, 261-264, P.L. Burn, D.D.C. Bradley, A.R. Brown, R.H. Friend, A.B. Holmes; Studies of the Efficient Synthesis of PPV and Dimethoxy-PPV, 1991, with permission from Elsevier Science)

149

Handbook of Polymers in Electronics

Figure 5.7 Electroluminescence spectrum of PPV

The EL spectrum of a LED prepared with PPV is reported in Figure 5.7, exhibiting an emission centred at 525 nm that corresponds to yellow-green light. The external efficiency of the simple LED formed by ITO, PPV, and aluminium (usually reported as ITO/PPV/ Al), was 0.05% [3]. This value increases by one order of magnitude if calcium is used instead of aluminium [69]. The use of a calcium electrode increases the efficiencies, but this metal is very reactive to humidity and its use may be allowed only in an inert atmosphere or when the electrode is well covered with a protecting layer. The first EL from PPV stimulated the work of chemists to develop new alternative routes to improve the preparation of this polymer. The idea of having a polymer soluble in organic solvent already in its conjugated form led to the introduction of solubilising groups in the PPV backbone. Some of the most studied structures that were developed are reported in Table 5.1. A detailed description of the synthesis for their preparation is not given; references are reported for each structure [70-85]. Some of the general criteria followed to prepare these materials are discussed. Highly ordered structures (i.e., highly packed in three dimensions) have the tendency to decrease the PL and consequently the EL efficiency. In fact, strong interchain interactions, which are much more relevant in the solid state than in solution, increase both exciton migrations to quenching sites [42, 43] and the formation of low energy excited states, mainly decaying non radiatively [86]. Consequently, it must be noted that polymer PL efficiencies can be very high in solution, but generally decrease by one order of magnitude in the solid state. The introduction of branched side chains, rather than linear residues, was found to be effective in lowering the degree of order of the material, preventing the tight packing of the macromolecules. This is the case of structures 1a, 1b, 2a, 2b, 4a, 4b in Table 5.1. For example,

150

Polymers for Light Emitting Diodes Son and co-workers [87] found that the introduction of cis double bonds in a PPV interrupts the conjugation and disturbs the polymer chain packing, resulting in the formation of an almost amorphous PPV with an EL efficiency of 0.22% for a simple ITO/PPV/Al LED. According to these findings, a simple ITO/MEH-PPV (1a)/Ca LED was reported to give an external quantum efficiency of 1% at 620 nm [70]. Changing substituents also has the effect of modulating the gap, resulting in a change in colour emission, as shown in Table 5.1. It is worthwhile mentioning that in [70] a simple flexible device of PANI-polyethylene terephthalate (PET) is described. PANI is a soluble conducting polymer which acts as a hole injecting electrode substituting ITO; on the top of the PANI layer a film of MEH-PPV (poly[2-(2´-ethylhexyloxy)-5-methoxy-1,4-phenylenevinylene]) was deposited. The electron injecting electrode was formed by calcium. The advantage of this LED with respect to other ones (where glass is covered by ITO) is that PET and PANI are flexible, hence the device can be completely bent without failure. This example shows the potential of polymeric materials with respect to inorganic compounds which cannot give large flexible electroluminescent areas. The preparation of a suitable film by spin coating or by casting from solution is the key step for development of new interesting devices. An example of a needle-like LED is described in [88], where the MEH-PPV active layer is directly formed onto a thin aluminium wire. The cathode is then formed by deposition onto the active layer of a thin transparent film of PANI thus obtaining an electroluminescent wire. Soluble electroluminescent polymers can be dissolved with non electroluminescent polymers; by spin coating the mixed solution, a blend film is obtained. This is a reliable method for improving the efficiencies both in PL and EL. In fact, blending allows a decrease of the interchain interactions. The drawbacks of blending mainly consist of the higher working voltage and a decreased lifetime of the devices [89]. Moreover, the mixing of two different polymeric materials often leads to phase segregation which has to be avoided if a decrease in the interchain interactions is required. The synthesis of conjugated copolymers is another approach followed to prepare new structures useful for device preparation. Moreover, the copolymeric approach is a suitable method for further tailoring the electrooptical properties of the active layer. In fact, the combination of different monomeric units, in different ratios, in the same structure, allows a fine modulation of the properties of the materials [90]. In Table 5.2 some of the many soluble copolymers based on the PPV structure are reported as examples. The structures so far prepared are mainly amorphous; the glass transitions have to be taken into account for the preparation of devices where the dimensional stability is very important. Some of the materials are block copolymeric structures (9, 12 and 16), while in other structures, although a repeating unit can be identified, the term copolymer is preferred because more than one chemical function appears in the repeating unit. As in the previous table, for each structure references are reported [69, 91-101].

151

Handbook of Polymers in Electronics

Table 5.1 PPV derivatives used as active materials in LED. For some structures the external and/or internal efficiencies are reported together with the kind of cathode used and wavelength of the EL emission

OR (a) MEH-PPV

R=

n

Elmax = 620 nm; ηel = 1% (Ca cathode) [70, 71, 72, 73]

OCH3

(b) OC1C10

R=

[74, 75, 76]

(c)

R=

[77]

1

OR (a)

R=

n

OR

Elmax = 560 [78] (b)

Elmax = 560

2

152

[79]

R=

Polymers for Light Emitting Diodes

Table 5.1 Continued

OR CH3 (a)

R=

Si

n

C6H12

CH3 OR

R' =

[80]

CH3 (b)

R=

Si

CH3

R' = H

CH3 Elmax = 540 nm; ηel = 0.0003% (Al cathode) [81] CH3 (c)

R=

Si

C8H18

R' = H

CH3 Elmax = 564 nm; ηel = 0.05% (Al cathode);

3

0.08% (Ca cathode) [82]

R

R (a)

R=

[83]

n

R'

R' = –(CH2)nH (b)

n = 6, 8, 10

R=

Elmax = 510 nm; ηel = 0.1% (Mg cathode); [84]

4

0.002% (Al cathode)

153

Handbook of Polymers in Electronics

Table 5.1 Continued

5

Z

(a)

Z=H R=

(b)

n

Z = SO2CH3

Elmax = 540-560 nm; ηel = 0.01% (Al cathode) [85]

OR

Table 5.2 Examples of complex polymeric structures used as active materials in LED

C12H25

6

OC6H13

CN

S

n

CN OC6H13

Elmax = 750 nm; ηel = 0.2% (internal, Ca cathode); 0.07% (internal, Al cathode) [91]

OC6H13

7

OC6H13 CN C6H13O

NC

n

C6H13O

Elmax = 710 nm; ηel = 0.2% (internal, both Ca and Al cathode) [92, 93]

154

Polymers for Light Emitting Diodes

Table 5.2 Continued

C10H21

8

C10H21 N N

N

Elmax = 540 nm; ηel = 0.015% (Al cathode) [94]

OCH3

9

m

n

OCH3

ηel = 0.02-0.3% (Al cathode) [69]

OCH3

10

OC7H15

OCH3

n

OC7H15

Elmax = 590 nm [95]

R

11

OCH3

OCH3 O(CH2)6

O

R

n

OCH3

OCH3

Elmax = 455 nm [96]

155

Handbook of Polymers in Electronics

Table 5.2 Continued 12 C8H17 Si

O

* n

OCH3

m

Elmax = 576-618 nm; ηel = 0.1-0.02% (internal, Al cathode) [97]

13

n

[98]

14

R Si R n

[99]

15 n

CN N

[100]

156

Table 5.2 Continued OC10H21

16

OC10H21 OC10H21

*

*

* n

m

p

OC10H21

Elmax = 515-567 nm [101]

Structure modulation, through a proper choice of the monomeric units forming the copolymer, has a great influence on its electronic properties; altering the copolymer composition allows tuning of the band gap, i.e., the control of the emission colour extending from the red, near IR region (700-1100 nm, peaked at 850 nm) of copolymer 6 in Table 5.2 [91], to the orange/red of the cyano-PPV [92] (structure 7) or of its derivatives [93] and to the blue of poly(1,20-(10,13)-didecyl)distyrylbenzene-co-1,2-(4-(p-ethylphenyl))triazole (TRIDSB) [94] (structure 8). Another example of modulation of the gap is shown in copolymers 11 and 14. The insertion of a non conjugated segment or of a silicon atom, respectively, interrupts the conjugation path giving emission peaked at around 450 nm. One of the factors that induces a decrease in the EL efficiency is the poor suitability of polymers to transport one of the injected charges, resulting in an imbalance in the recombination. A suitable chemical approach to increase the injection of electrons consists of increasing the electron affinity of the polymer for the minority charge carriers. As most of the semiconducting polymers are easily p-doped and not n-doped, the increase of the electron affinity through the introduction of suitable electron withdrawing substituents has been found to be very promising. This is the case of structure 7 (Table 5.2) showing emission at 710 nm; internal quantum efficiencies up to 0.2% were reached both with a calcium and aluminium cathode in a single layer device [92]. It is important to remark that, due to the presence of cyano groups, the lowest unoccupied orbitals, forming the conduction band, are lowered, so that good injection of electrons both from calcium and aluminium occurs and equivalent efficiencies are obtained even though the two metals have different work functions. However, for PPV single-layer devices, it was reported [69] that the quantum efficiency with the calcium electrode is one order of magnitude higher than the corresponding device where aluminium is used.

157

Handbook of Polymers in Electronics An even better balance in the injection of the opposite charges was found when another layer more suitable for hole transport, such as unsubstituted PPV, is used [70]. A doublelayer LED, formed by ITO/PPV/cyano-PPV/Al has an internal quantum efficiency as high as 4% with emission at 610 nm [70]. The cyano-PPV approach was extremely useful and other withdrawing substituents were introduced in the PPV backbone with the aim of increasing the electron affinity; other electron-deficient nitrogen containing groups such as oxadiazole [102, 103], triazole [94], pyridine [104, 105] and quinoxaline [106] were also introduced in the polymer structure. Copolymer 10 was found to give bright EL either with a dc forward bias or with a reverse bias voltage, and also with alternate field. The EL spectra are almost equivalent for a simple ITO/10/Al configuration. As previously reported, the balance between the two opposite injected charges is one of the main goals of this area. Apart from the increase in the electron affinity of the material, the adding, in the proper position (i.e., near the cathode or near the anode) of an additional layer formed by polymeric material and/or small molecules as electron transporters or hole transporters was found to be very useful. An example has already been reported above: PPV has been used as a hole transporter in a double-layer configuration LED [92]. The number of low molecular weight molecules synthesised as hole or electron transporters is extremely high. Two of the simplest molecules are shown in Figure 5.8. Good electron transporting molecules include 1,3,4-oxadiazole derivatives, while triphenylamine and related structures (TPD, Figure 5.8) are good hole transporters. A simple method to obtain a better electron injection consists of the deposition of a layer of 1,3,4-oxadiazole on the top of the polymeric emitting layer.

N

N

TPD CH3

N

N

H3C CH3

O PBD

Figure 5.8 Hole transporter (tetraphenyl diaminobiphenyl (TPD)) and electron transporter (2-(4-biphenyl)-5-(tert-butylphenyl-1,3,4-oxadiazole (PBD))

158

Polymers for Light Emitting Diodes It was found, however, that devices having 1,3,4,-oxadiazole derivatives, or other low molecular weight molecules, as electron transporters in the form of a vacuum evaporated thin film, have a short lifetime, a feature probably associated with recrystallisation or aggregation phenomena [107]. Incorporating hole or electron transporting molecules in the polymer backbone is a general procedure that has also been applied to other emitting polymers [108-110]; polymeric hole or electron transporters have also been prepared [111, 112]. There are examples in the literature where the balance between holes and electrons is reached by inserting a hole blocking material between the ITO and the active polymers. For example, a simple polymethylmethacrylate (PMMA) monolayer prepared by means of the Langmuir Blodgett technique was reported to give a four-fold efficiency increase if introduced between the ITO and the active layer [113]. Greenham, Nüesch and Yang [114-116] followed similar approaches to control the performance of a MEH-PPV based device and a polyparaphenylene-based device respectively. PANI was reported to improve the performance of a PPV LED [116]. The high interface area between the PANI film and the active luminescent layer enhances the local electric field increasing charge carrier injection [117]. Other materials were found to be very efficient if introduced between the ITO and the active layer. Many authors [118, 119] reported that poly(3,4-ethylenedioxythiophene) (PEDOT), mixed with polystyrene sulfonated acid, PSSA, (Figure 5.9) has a meaningful effect both on the lifetime and in the EL efficiency if spin coated between ITO and the active polymeric layer. Very recently this subject has been studied by means of the ultraviolet photoelectron spectroscopy (UPS) technique [37].

n

O

O

S PEDOT

n

SO3H PSSA

Figure 5.9 Poly(3,4-ethylenedioxythiophene) (PEDOT) and polystyrene sulfonated acid (PSSA)

159

Handbook of Polymers in Electronics The problem of the interfaces between the conjugated systems and the electrodes has been deeply studied, indicating that chemical reactions limiting the charge injections occur at the interfaces [120]. Degradation of the polymer near to the surface occurs and this is one of the limiting factors reducing the lifetime of the devices.

5.3.1.2 Other Polyphenylene Derivatives The deep and detailed work performed by different groups on the many PPV derivatives is only a part of the huge amount of other structures which were investigated with the aim of reaching valuable performances of polymeric LEDs. As the double bonds in PPV are reactive chemical functions which can be oxidised, the polyparaphenylene (PPP) structures were intensively investigated with the aim of obtaining more robust materials. PPP (Figure 5.10a) is insoluble and only the introduction of alkyl side chains (Figure 5.10b), linear or branched, has enabled the polymers to be solubilised, obtaining thin homogeneous films suitable for LED applications, even though one of the first LEDs emitting in the blue region was prepared from conversion of unsaturated PPP [121].

n

(a)

n

(b)

Figure 5.10 Polyparaphenylene (PPP) (a) and substituted polyparaphenylene (b)

PPP, substituted or unsubstituted, is not planar; the band gap is relatively high and emission is shifted towards a higher energy with respect to PPV. So, while the emission of PPV is mostly centered in the yellow-green region, PPP emissions can reach the blue region of the spectrum.

160

Polymers for Light Emitting Diodes Soluble PPP obtained through the Suzuki coupling, as reported by Schlüter and Wegner [122], gives high molecular weight compared to previously reported soluble PPP synthesis [123] and an LED prepared with an alkoxy PPP with a two-layer configuration, ITO/ PVCZ/alkoxy PPP/Ca reaches a quite respectable EL efficiency of up to 3% [124]. With the introduction of alkoxy-branched side chains, it was possible to prepare the active thin layer by means of the Langmuir Blodgett technique [125]. During the transfer process, orientation of the macromolecule axis in the dipping direction occurs. LEDs having these oriented layers as the active film give polarised EL with a polarisation ratio between the two perpendicular directions of about 5 and an external emission efficiency value of 0.05%. A block of substituted phenyl sequences connected with double bonds has allowed for the preparation of copolymer structures that can be used as a thin active layer in an LED, with emission in the 460 nm region and internal quantum efficiencies from 2% to 4% [126]. The Mullen and Scherf approach [127, 128] to ladder PPP (L-PPP) provided a good structural improvement for the preparation of stable and tuneable light emitting polymers. An example of an L-PPP is reported in Figure 5.11. As the introduction of the bridges between adjacent monomers induces planarisation of the backbone, a remarkable shift towards higher wavelengths was observed for both PL and EL that reached a value of 600-620 nm at the maximum of the peak. This shift was also attributed to the formation of excimers due to interchain interactions in a highly planarised, conjugated segment [128]. The introduction of suitable substituents has led to an L-PPP [129] having bluegreen emission with external EL efficiencies of 4% in a single-layer configuration with aluminium as the cathode. It was also reported that a certain modulation in the emitted light through the applied field variation was possible in the blue-green range [129].

R

H

C6H13

n

C6H13

C6H13

R

H

Figure 5.11 An example of an L-PPP

An interesting result in preparing devices with red-green-blue (RGB) emission was shown by Leising and co-workers [130] using parahexaphenyl (PHP) as an active emitting layer. The blue emission of PHP was converted by means of a suitable dye layer absorbing in the blue region into green light which excited another dye layer absorbing green and emitting red. The light was also spectrally purified with a suitable dielectric mirror/filter. A representation of the RGB LED architecture is reported in Figure 5.12, as from [130].

161

Handbook of Polymers in Electronics

Figure 5.12 RGB LED architecture [130] (Reproduced with permission from Advanced Materials, 1997, 9, 1, 35, published by Wiley-VCH, 1997)

5.3.1.3 Poly(phenylene ethynylenes) The desire of polymer chemists to test new structures in the family of phenylenes has led to the preparation of polymers having triple bonds in between two phenyl rings; poly(phenylene ethynylenes), alkoxy or alkyl substituted, have been prepared and studied for their electrooptical properties. PL emissions in the 550-600 nm region were found [131] in polymeric structures having an average degree of polymerisation between 23 and 28. Alternate copolymers containing thiophene units [132] having PL efficiencies up to 38% in CHCl3 solution at 460 nm were also described. These polymers were recently studied for their liquid crystalline behaviour [133] and for the possibility to be oriented and give polarised EL [134, 135].

5.3.1.4 Polyfluorenes Another class of semiconducting polymers containing the phenylene ring has recently attracted the attention of many researchers for its promising properties. In Figure 5.13 the structure of poly(9,9-dihexylfluorene) is shown as an example of this class of polymers [136, 137]. The possibility of introducing, as in the case of substituted PPV, alkyl side chains of different chemical nature and branching level without seriously affecting the electronic properties of the backbone is one of the advantages of this polymeric family [138].

162

Polymers for Light Emitting Diodes H13C6

C6H13

n

Figure 5.13 Poly(9,9-dihexylfluorene)

Polymerisation of 2,7-dibromo-9,9-bis(3,6-dioxaheptyl)fluorene with Ni(0) catalyst is reported to be efficient, giving polymers with molecular weight of 215,000 [139]. Other synthetic approaches through the Suzuki coupling gave lower molecular weight [140]. Pei and Yang introduced 2-methoxyethoxyethane in the 9 position of a fluorene monomer [139]. The corresponding polymer, obtained by polymerising the 2,7 dibromoderivatives of the monomer by Ni(0) catalyst, poly[9,9-bis(3,6-dioxaheptyl)-fluorene-2,7-diyl] (BDOH-PF) had a PL efficiency in solution of 77%, which remains almost constant (73%) in the solid state, with two sharp peaks in the blue region and a broad intense peak in the green region attributed to interchain excimer emission. A simple single-layer LED, ITO/BDOH-PF/Ca has an external quantum efficiency of 0.3%. This low EL efficiency, almost unexpected on the basis of the PL quantum yield, was attributed to the poor electron injection. The introduction in the active polymeric layer of lithium triflate [139], in a given ratio with respect to the polymer, increased the EL up to an external quantum efficiency of 4% in the blue-green region. Addition of inorganic salts, as described by Pei and co-workers [141, 142] induces, upon applying the field, a migration of the cations and anions in opposite directions causing ndoping at the cathode and p-doping at the anode. This doping is responsible for a more efficient injecting contact at the interfaces between cathode/anode and the active layer. The polar nature of the side chains solvates the ions of lithium triflate that were introduced in the polymer and promote their transport. By adding polyethylene oxide in the matrix, the device gave a white emission [142]. The copolymerisation of fluorene units with other phenylene monomers was studied by Miller and co-workers [143]. Copolymers having molecular weight ranging from 24,000 up to 204,000 were obtained with a Ni(0) polymerisation. High PL efficiencies in solution and the solid state were found for these copolymers, while external EL efficiencies were around 0.15% at 525 nm when tetraethyl ammonium tetrafluorate was added (2%) in the active layer. Introduction of different amounts of anthracene units in poly(di-nhexylfluorene) did not have a meaningful improvement in the device performance [144]. Soluble polyfluorenes were found to give a liquid crystalline behaviour. For example, poly(9,9-di(ethylhexyl)fluorene (PDF2/6) (Figure 5.14) was found to have a transition into a birefringent fluid phase [145] which can be obtained in a highly oriented arrangement by

163

Handbook of Polymers in Electronics

n

Figure 5.14 Structure of poly(9,9-di(ethylhexyl)fluorene) – PDF2/6

deposition onto a highly preoriented polyimide substrate [145]. The degree of orientation which can be reached [145] seems to be dependent on the kind of side chain, (linear or branched) or on the different packing reached by changing the side chains.

5.3.2 Polythiophenes The wide field of research in the area of electroluminescent polymers is not limited to phenyl-like structures. Polythiophene derivatives have also been the subject of many studies for their interesting electrooptical properties. Polythiophene was one of the first conducting polymers to be studied, but only when its soluble form was obtained by introduction of alkyl side chains. The chance to prepare thin film by solution casting or by spin coating drastically increased the interest in these organic semiconductors. From the very beginning it was clear that the modulation of the band gap was easy through a proper choice of the substituents. Poly(3-alkylthiophenes) (PATs) in a head-to-tail configuration are almost planar; the so called trans conformation (Figure 5.15) minimises the steric hindrance between adjacent monomer units. However, the introduction of head-to-head connections leads to a distortion from the planarity of the conjugated system; bulky substituents [146] in the 3 position of the thiophene ring or 3,4-dialkyl substituted thiophene [147] inserted in a regular head-to-tail connection have the same effect. With this simple method it was easy, for example, by copolymerising 3,4-disubstituted thiophene and 3-alkyithiophene in different ratios [147], to obtain copolymers with different absorptions from 320 nm to 450 nm. Berggren [148] used a series of polymeric structures to prepare LEDs covering all the visible range (Figure 5.16). By blending the different copolymers in a matrix such as PMMA, (which acts as an energy transfer blocking material), the colour emission could be varied by a proper applied voltage [148] from orange to blue (Figure 5.17). Tuning of the EL was also obtained by preparing copolymers of 3-alkylthiophene and unsubstituted thiophenes in different molar ratios [149].

164

Polymers for Light Emitting Diodes R

R S

S

S R

R

Figure 5.15 Structure of a trans conformation of a regioregular poly(3-alkylthiophene)

S S S IV. PCHMT

n

S III. PCHT

n

n

II. PTOPT

S

n

I. POPT

Figure 5.16 Polymeric and copolymeric structures emitting from 450 nm to 800 nm (Reproduced with permission from O. Inganäs, Nature, 1994, 372, 444, published by MacMillan)

165

Handbook of Polymers in Electronics

Figure 5.17 Electroluminescence at two different applied field of a blend of different PAT structures (Reproduced with permission from O. Inganäs, Nature, 1994, 372, 446, published by MacMillan)

Polarised luminescence was obtained with a simple stretched film of poly(3-(4octylphenyl)thiophene) as an active film. The anisotropy (the ratio of the intensity of the emitted light parallel versus the intensity of the emitted light perpendicular to the drawing direction) was up to 3.1 [150]. A suitably functionalised PAT was found to give polarised electroemission in the yellow/orange region when the active layer of the LED was prepared by means of the Langmuir Blodgett technique [151]. In many works, the tailoring of the electronic properties of PAT was achieved by controlling the regioregularity of the backbone. Barta and co-workers [152] prepared a poly(4,4´-dialkyl-2,2´-bithiophenes) (PDABT) having both head-to-tail connections and tail-to-tail connections, where different lengths of the side chains do not drastically affect the optical properties. EL was observed at 541 nm with an efficiency of 0.5%, which is two orders of magnitude higher with respect to previous results on regioregular PAT [153, 154]. The higher EL in PDABT was attributed to the head-to-head connection between two 3-alkylthiophones forming exciton traps thereby hindering their migration to quenching sites [155]. It was suggested [155] that in PDABT, the head-to-head connection, which is responsible for a reduced planarity in the backbone, influences the relative position of singlet and triplet states. The non planar situation favours the creation of singlet exciton with respect to triplet states leading to an improvement in EL. The photophysics of many substituted polythiophenes has been recently summarised by Theander and co-workers [156] who showed that PL efficiency, in the solid state, as high as 24% can only be obtained with the polymer shown in Figure 5.18 at 599 nm.

166

Polymers for Light Emitting Diodes

C8H17

C8H17

S

n

Figure 5.18 PAT structure exhibiting PL at 599 nm with ηpl = 24%

Heeger and co-workers [157] found that high EL efficiencies can be reached by dispersing surfactant-like additives in the active layer. They reported that a device prepared with surfactant-like compounds in poly(3-octylthiophene) gives, with an aluminium electrode, 0.03% external efficiency, much higher than the corresponding device without additives but with calcium as the cathode. The procedure reported in [157] has been successfully applied to other electroluminescent polymers by the same authors. PATs are available by three synthetic procedures: (i) Electrochemical synthesis allows the preparation of thin doped films on an electrode. The subsequent reduction of the polymer gives the neutral state [158]. (ii) FeCl3 polymerisation [156] is a simple method to obtain the polymer without many problems. Careful cleaning of the polymer by many precipitations in methanol has to be done in order to avoid the presence of the iron impurity (acting as dopant). The starting point is the monomer, 3-alkylthiophene, which is easily obtained. The regioregular control with this kind of polymerisation is not extremely high. However, when the substituent in the 3 position has a high steric hindrance, a polymer having head-to-tail regularity higher than 90% may be obtained [156]. (iii) With Ni catalyst polymerisation a higher control of the regioregularity can be obtained [154]. A very simple method for reaching high values of regioregularity has recently been reported by McCullough [159] and Bolognesi [160]. Regioregularity as high as 98%-100% has been reached, giving emission in the 700-720 nm range. Other synthetic procedures for highly regioregular polymers based on the Suzuki coupling [161], Stille reaction [162] and using Rieke zinc [163] have been reported. As in the case of polymers of the phenyl family, a great amount of work has been done concerning the introduction of electron/hole transporter molecules in the main conjugated backbone and by modifying the architecture of the devices [164-167]. Introduction of

167

Handbook of Polymers in Electronics Alq3 between the PAT thin film and the cathode increases the EL efficiency up to 0.05%, using aluminium as the cathode [167]. It was also observed that a certain tuning of the EL by applying different electric fields may be reached [168].

5.4 Recent Developments 5.4.1 Polarised Electroluminescence The opportunities offered by polymeric structures for LED fabrication are manifold. Besides the tuning of the properties through proper tailoring of the structures and the cheap technique used to prepare thin and large area films, polymeric materials have the advantage, with respect to other low molecular weight material, that they can be stretched to give highly oriented structures. Since conjugated polymers are monodimensional semiconductors, their electrical and optical properties are strongly anisotropic and the chain orientation leads to extremely high electrical and optical anisotropy. Stretched (CH)X was reported [169] to have highly anisotropic electrical behaviour with a ratio between the conductivity parallel and perpendicular to the drawing direction in excess of 100. Besides the electrical anisotropy, optical anisotropy can be observed in oriented absorbing polymers. Dyereklev and co-workers [150] and Ohmari [164] showed that an LED with an oriented electroluminescent polymer as an active layer emits polarised light. The development of polarised electroemission is extremely important for optical applications; in backlighting liquid crystal displays (LCDs) the light passes through a set of polarisers in the front and in the back of the active LCD layer [170]. These polarisers absorb 50% of the light so that higher light intensities are reached by increasing power consumption. For this reason, polarised EL has recently been the subject of attention for many research groups. Different methods can be applied to reach a high orientation in the polymers to be used in an LED. It has already been mentioned that the Langmuir Blodgett technique allows orientation of rod-like systems during the transfer process of the monolayer onto the substrates [125, 151]. So a multilayered structure directly deposited onto the ITO electrode can be used as the active layer of a LED. The highest dichroic ratios in EL were around 5 with a PPP derivative [125]. Though the Langmuir Blodgett technique can be used to prepare very well-defined architecture suitable for fundamental studies, [171], this technique has the limit of requiring long preparation procedures before reaching an appropriate thickness. The mechanical stretching of a thin film which has to be assembled between the cathode and anode is the simplest method of orienting polymeric materials. The work by Dyereklev [150] is an example of this approach. External efficiencies between 0.1% and 0.01%

168

Polymers for Light Emitting Diodes were obtained with an EL dichroic ratio of 2. Hamaguchi was able to orient alkoxy substituted PPV through the rubbing procedure [172], reaching a polarisation ratio up to 4. More recently, Jandkhe and co-workers [173] were able to gain, through a particular procedure applied to a partially converted PPV precursor, a dichroic ratio in EL of about 12 at 511 nm with a brightness of 200 cd/m2. This result is very promising and near to the value required for technological applications. However, the authors mentioned that lifetimes of the devices are strongly decreased. In a series of very recent works, Grell [145] and Grell and Bradley [174] reported very promising results on the polarised EL obtained with dialkylsubstituted polyfluorene derivatives. High orientation was reached [175] by heating the polyfluorene films, deposited onto a preoriented polyimide (PI) film, at the temperature where it exhibits a birefringent fluid phase. PI film orientation is achieved by rubbing its surface with a cloth as described in literature following a well-established procedure [176]. PI is a good insulator, so the oriented PI film, deposited on ITO, contained a good hole transporter, 4,4´,4´´-tris(1naphthyl)-(N-phenyl-amino)triphenylamine to promote hole injection from the ITO to the polyfluorene. It has been reported that the procedure followed to reach the high orientation of the macromolecules [174] produces a highly oriented monodomain structure [177] whose morphology has been described as a glassy liquid crystalline monodomain. In Figure 5.19, the polarised EL of a device is shown. The EL anisotropy for the peak at 477 nm is 15.

Figure 5.19 Polarized EL from a LED having as active layer a highly oriented polyfluorene [145]. Perpendicular and parallel refer to the chain orientation (Reproduced with permission from Advanced Materials, 1999, 11, 8, 671, published by Wiley-VCH, 1999)

169

Handbook of Polymers in Electronics The research in this field is very active and even higher polarisation ratios could be obtained in the near future.

5.4.2 Lifetime and Degradation in LEDs All the ‘components’ of a device are prepared with materials that can be chemically reactive. Metals constituting the cathode electrodes may react with the active layer causing a change in the structure of the conjugated system. Moreover, charge injection in polymers introduces reactive species in the backbone and creation of crosslinks between adjacent chains may not be excluded. Lifetime is strictly related to this degradation phenomena. In addition, the presence of impurities or traces of oxygen has a strong effect in decreasing the overall EL efficiency as well as the lifetime [178]. In fact, oxidation of conjugated polymers in the presence of singlet oxygen is reported by many research groups [179, 180]. Parker and co-workers recently studied the mechanism of degradation in a PPV LED [181], in a single-layer device packaged with a glass cover in a nitrogen atmosphere. They were able to measure lifetimes of around 20,000 hours at luminance greater than 100 cd/m2 when operating in constant current mode. However, during continuous operation the luminance of the devices changed in a non monotonic way and the operating voltage increased in a linear fashion. The obtained data were explained attributing the main degradation process of the polymer to the passage of electrons, while hole currents seem not to lead to degradation.

5.4.3 Microcavities Apart from the nature of the polymeric active material, the structure and size of the LED device play an important role since they determine the coupling of the polymer excitations to the electromagnetic resonant modes. In fact, a device of the type reported in Figure 5.2 acts as a microcavity, and simply by adding a mirror (metallic or dielectric, distributed Bragg reflector) between the glass and the ITO layer, a good resonant microcavity is obtained for typical polymer thickness of the order of 100 nm [182-185]. The second mirror of the cavity is already formed by the top metal contact (Al or Ag) [186]. The effect of the microcavity is both to modulate the intensity of the PL or EL of the polymer, according to its spectral position, and to modify the spatial distribution of the emitted light. In fact the polymer emission is enhanced at the wavelengths corresponding to the resonant modes of the cavity, while it is quenched at the wavelengths which are not compatible with the cavity modes. The result is a sharp narrowing of the emission spectrum (see Figure 5.20). For a planar structure the spatial distribution of the emission considerably narrows, increasing the intensity of the light emitted in the forward direction [185]. The dependence of the emission wavelength on the viewing angle [185, 187] can

170

Polymers for Light Emitting Diodes

Figure 5.20 Narrowing of the emission spectrum in a microcavity. (Reproduced by permission from Journal of Applied Physics, 1996, 80, 1, 209, published by the American Physical Society, 1996)

be suppressed by introducing proper cavity optical length dispersion [188]. The spectral position of the enhanced emission can be varied (within the spectral range of the free space emission) by varying the optical parameters of the cavity (thickness, refractive index, and quality factor), thus obtaining different colours with the same material [185]. Thus, microcavity structures have been fabricated in order to control the emission energy, linewidth, intensity and directionality of many polymeric LEDs. Microcavities have been very intensively studied because they are good structures for optically pumped lasers [189-196]. In fact, many polymers (see Table 1 in reference [195]) showed gain narrowing (spectral narrowing of the PL above an energy threshold of the optical pump excitation) in solution, in blends, and also in undiluted sub-micron films [197] for microcavity and planar waveguide structures [198]. Further efforts are needed in order to obtain electronically pumped lasers due to the high current densities and problems related to excited state absorption from injected charges [199].

5.5 Concluding Remarks The rapid growth of this new area of polymer science has stimulated the work of chemists, physicists, and engineers. The exciting results that have been obtained are due to a strong cooperation among different group of scientists used to working in their own field. Now there is a common area and this is the reason why this field is growing very rapidly.

171

Handbook of Polymers in Electronics The research presented in this chapter is only a part of the field of EL in organic materials. A lot of work has been done with low molecular weight molecules that have been used successfully to prepare good LEDs with very respectable performances. High vacuum evaporation techniques are generally used for the preparation of the active layer. Polymers offer the possibility of working with a cheap technology giving flexible films that can be bent without breaking and that can be oriented to emit polarised light. These are the real advantages of working with polymers and in the near future even better results will be reached.

References 1.

C.K. Chiang, C.R. Fincher, Y.W. Park, A.J. Heeger, H. Shirakawa, E.J. Luis, S.C. Gau and A.G. MacDiarmid, Physical Review Letters, 1977, 39, 1098.

2.

C.W. Tang and S.A. Van Slyke, Applied Physics Letters, 1987, 51, 913.

3.

J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. MacKay, R.H. Friend, P.L. Burn and A.B. Holmes, Nature, 1990, 347, 539.

4.

R. H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughes, R.N. Marks, C. Taliani, D.D.C. Bradley, D.A. Dos Santos, J.L. Brèdas, M. Logdlünd and W.R. Salaneck, Nature, 1999, 397, 121.

5.

www.kodak.com, www.covion.com/index.htm, www.cdtltd.co.uk, www.uniax.com, www.philips.com

6.

A.J. Heeger, S. Kivelson, J.R. Schrieffer and W.-P. Su, Reviews in Modern Physics, 1988, 60, 3, 781.

7.

R.H. Friend, D.D.C. Bradley and P.D. Towsend, Journal of Physics D: Applied Physics, 1987, 20, 1367.

8.

Z.G. Soos, D.S. Galvão and S. Etemad, Advanced Materials, 1994, 6, 4, 280.

9.

R.H. Friend, G.J. Denton, J.J.M. Halls, N.T. Harrison, A.B. Holmes, A. Köhler, A. Lux, S.C. Moratti, K. Pichler, N. Tessler, K. Towns and H.F. Wittmann, Solid State Communications, 1997, 102, 249.

10. R.H. Friend in Handbook of Conducting Polymers, Eds., T.A. Skotheim, R.L. Elesnbaumer and J.R. Reynolds, Marcel Dekker, Inc., New York, 1998, 823. 11. G. Leising, S. Tasch and W. Graupner in Handbook of Conducting Polymers, Eds., T.A. Skotheim, R.L. Elesnbaumer and J.R. Reynolds, Marcel Dekker, Inc., New York, 1998, 847.

172

Polymers for Light Emitting Diodes 12. R.H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughes, R.N. Marks, C. Taliani, D.D.C. Bradley, D.A. Dos Santos, J.L. Brédas, M. Lögdlund and W.R. Salaneck, Nature, 1999, 397, 121. 13. W.-P. Su, J.R. Schrieffer and A.J. Heeger, Physical Review B, 1983, 28, 1138R. 14. Z. Shuai, J.L. Brédas and W.P. Su, Chem. Physical Letters, 1994, 228, 301. 15. M. Pope and C.E. Swenberg, Electronic Processes in Organic Crystals, Oxford University Press, New York, NY, USA, 1982. 16. H. Bässler, Macromolecular Symposia, 1996, 104, 269 17. J-L. Brédas, J. Cornil and A.J. Heeger, Advanced Materials, 1996, 8, 5, 447. 18. M. Yan, L. Rothberg, F. Papadimitrakopoulos, M.E. Galvin and T.M. Miller, Physical Review Letters, 1994, 72, 7, 1104. 19. B.J. Sun, Y.-J. Miao, G.C. Bazan and E.M. Conwell, Chemical Physics Letters, 1996, 260, 186. 20. R. Jakubiak, Ch.J. Collison, W.C. Wan, L.J. Rothberg and B.R. Hsieh, Journal of Physical Chemistry A, 1999, 103, 2394. 21. N.C. Greenham, J. Shinar, J. Partee, P.A. Lane, O. Amir, F. Lu and R.H. Friend, Physical Review B, 1996, 53, 20, 13528. 22. A.R. Brown, K. Pichler, N.C. Greenham, D.D.C. Bradley, R.H. Friend and A.B. Holmes, Chemical Physics Letters, 1993, 210, 61. 23. U. Rauscher, H. Bässler, D.D.C. Bradley and M. Hennecke, Physical Review B, 1990, 42, 9830. 24. S. Speiser, Chemical Reviews, 1996, 96, 1953. 25. U. Lemmer, A. Ochse, M. Deussen, R.F. Mahrt, E.O. Göbel, H. Bässler, P. Haring Bolivar, G. Wegmann and H. Kurtz, Synthetic Metals, 1996, 78, 289. 26. A. Dogariu, R. Gupta, A.J. Heeger and H. Wang, Synthetic Metals, 1999, 100, 1, 95. 27. J.J.M. Halls, J. Cornil, D.A. dos Santos, R. Silbey, D.-H. Hwang, A.B. Holmes, J.L. Brédas and R.H. Friend, Physical Review B, 1999, 60, 8, 5721. 28. M.D. Joswick, I.H. Campbell, N.N. Barashkov and J.P. Ferraris, Journal of Applied Physics, 1996, 80, 5, 2883. 173

Handbook of Polymers in Electronics 29. D.J. Pinner, R.H. Friend and N. Tessler, Journal of Applied Physics, 1999, 86, 9, 5116. 30. J.C. de Mello, H.F. Wittmann and R.H. Friend, Advanced Materials, 1997, 9, 3, 230. 31. Y. Yang and A.J. Heeger, Applied Physics Letters, 1994, 64, 10, 1245. 32. S. Karg, J.C. Scott, J.R. Salem and M. Angelopoulus, Synthetic Metals, 1996, 80, 2, 111. 33. A.J. Heeger, I.D. Parker and Y. Yang, Synthetic Metals, 1994, 67, 23 34. I.D. Parker, Journal of Applied Physics, 1994, 75, 3, 1656. 35. V.R. Nikitenko and H. Bässler, Journal of Applied Physics, 1999, 85, 9, 6515. 36. C. Giebler, H. Antoniadis, D.D.C. Bradley and Y. Shirota, Journal of Applied Physics, 1999, 85, 1, 608. 37. Th. Kugler, W.R. Salaneck, H. Rost and A.B. Holmes, Chemical Physics Letters, 1999, 310, 391. 38. Ch. Jonda, A.B.R. Mayer and W. Grothe, Journal of Applied Physics, 1999, 85, 9, 6884. 39. S. Karg, M. Meier and W. Riess, Journal of Applied Physics, 1997, 82, 4, 1951. 40. D.R. Baigent, N.C. Greenham, J. Grüner, R.N. Marks, R.H. Friend, S.C. Moratti and A.B. Holmes, Synthetic Metals, 1994, 67, 1/3, 3. 41. F. Cacialli, R.H. Friend, N. Haylett, R. Daik, W.J. Feast, D.A. dos Santos and J.L. Brédas, Synthetic Metals, 1997, 84, 643. 42. W. Tachelet, S. Jacobis, H. Ndayikengurukiye, H.J. Geise and J. Grüner, Applied Physics Letters, 1994, 64, 2, 2364. 43. F. Garten, A. Hilberer, F. Cacialli, E. Esselink, Y. van Dam, B. Schlatmann, R.H. Friend, T.M. Klapwijk and G. Hadziioannou, Advanced Materials, 1997, 9, 2, 127. 44. P. Dannetum, M. Boman, S. Stafström, W. Salaneck, R. Lazzaroni, C. Fredriksson, J.L. Brédas, R. Zamboni and C. Taliani, Journal of Chemical Physics, 1993, 99, 1, 664. 45. H. Becker, S.E. Burns and R.H. Friend, Physical Reviews B, 1997, 56, 4, 1893.

174

Polymers for Light Emitting Diodes 46. J. Birgerson, M. Fahlman, P. Bröms and W.R. Salaneck, Synthetic Metals, 1996, 80, 2, 125 and references therein. 47. M. Deussen, M. Scheidler and H. Bässler, Synthetic Metals, 1995, 73, 2, 123. 48. S. Tash, G. Kranzelbinder, G. Leising and U. Scherf, Physical Reviews B, 1997, 55, 8, 5079. 49. H. Bässler, Macromolecular Symposia, 1996, 104, 269. 50. N.C. Greenham, R.H. Friend and D.D.C. Bradley, Advanced Materials, 1994, 6, 6, 491. 51. P. Destruel, P. Jolinat, R. Clergereaux and J Farenec, Journal of Applied Physics, 1999, 85, 397. 52. M.G. Harrison, J. Grüner and G.C.W. Spencer, Synthetic Metals, 1996, 76, 1/3, 71. 53. D.G. Lidzey, S.F. Alvarado, P.F. Seidler, A. Bleyer and D.D.C. Bradley, Applied Physics Letters, 1997, 71, 2008. 54. G.E. O’Keefe, J.J.M. Halls, C.A. Walsh, G.J. Denton, R.H. Friend and H.L. Anderson, Chemical Physics Letters, 1997, 275, 78. 55. R. Kersting, U. Lemmer, M. Deussen, H.J. Bakker, R.F. Mahrt, H. Kurz, V.I. Arkhipov, H. Bässler and E.O. Göbel, Physical Reviews Letters, 1994, 73, 1440. 56. N. Ohta, S. Umeuchi, T. Kanada, Y. Nishimura and I. Yamazaki, Chemical Physics Letters, 1997, 279, 215. 57. S. Großmann, T. Weyraauch, S. Saal and W. Haase, Optical Materials, 1998, 9, 236. 58. M.I. Khan, M.L. Renak, G.C. Bazan and Z. Popovic, Journal of the American Chemical Society, 1997, 119, 5344. 59. N. Pfeffer, D. Neher, M. Remmers, C. Poga, M. Hopmeier and R. Mahrt, Chemical Physics, 1998, 227, 167. 60. R.A. Wessling, Journal of Polymer Science, Polymer Symposia, 1985, 72, 55. 61. P.L. Burn, D.D.C. Bradley, A.R. Brown, R.H. Friend and A.B. Holmes, Synthetic Metals, 1991, 41-43, 261. 62. P.L. Burn, D.D.C. Bradley, R. H. Friend, D.A. Halliday, A.B. Holmes, R.W. Jackson and A. Kraft, Journal of the Chemical Society Perkin Transactions, I, 1992, 3225.

175

Handbook of Polymers in Electronics 63. R.O. Garay, U. Baier, C. Bubeck and K. Mullen, Advanced Materials, 1993, 5, 7/ 8, 561. 64. J. Gmeiner, S. Karg, M. Meier, W. Rieß, P. Strohriegl and M. Schwoer, Acta Polymerica, 1993, 44, 201. 65. A. Beerden, D. Vandenzande and J. Gelan, Synthetic Metals, 1992, 52, 387. 66. M. Herold, J. Gmeiner, W. Rieb and M. Schwoerer, Synthetic Metals, 1996, 76, 1/ 3, 109. 67. F. Louwet, D. Vardenzande, J. Gelan and J. Mullens Macromolecules, 1995, 28, 4, 1330. 68. H.G. Gilch and W.L. Wheelwright, Journal of Polymer Science A1, 1966, 4, 1337. 69. P.L. Burn, A.B. Holmes, A. Kraft, A.R. Brown, D.D.C. Bradley and R.H. Friend, Material Research Society Symposia Proceedings, 1992, 247, 647. 70. G. Gustafsson, Y. Cao, G.M. Treacy, F. Klavetter, N. Colaneri and A.Y. Heeger, Nature, 1992, 357, 477. 71. A.J. Heeger and D. Braun, Chemical Abstracts, 1993, 118, 157401j. 72. F. Wudl and G. Srdanov, Chemical Abstracts, 1993, 118, 255575p. 73. F. Wudl, P.M. Allemand, G. Srdanov, Z. Ni and D. McBranch in Materials for Non Linear Optics, Chemical Perspectives, Ed., S.R. Marder, J.E. Sohn and G.D. Stucky, A.C.S., Washington 1991. Symposium Series 455, 683. 74. D. Braun, E.G.J. Staring, R.C.J.E. Demandt, G.L.J. Rikken, Y.A.R. R. Kessener and A.H.J. Venhuizen, Synthetic Metals, 1994, 66, 1, 75. 75. L. Lutsen, P. Adriaensen, H. Becker, A.J. Van Breemen, D. Vardenzande and J. Gelan, Macromolecules, 1999, 32, 6517. 76. J. Salbeck, Berichte Bunsengesellschaft für Physikalische Chemie, 1996, 100, 1666. 77. G.J. Sarnecki, P.L. Burn, A. Kraft, R.H. Friend and A.B. Holmes, Synthetic Metals, 1993, 55, 2/3, 914. 78. F. Wudl, S. Höger, C. Zhang, K. Pakbaz and A.J. Heeger, Polymer Preprints, 1993, 34, 197. 79. S. Klingelhöfer, C. Schellenberg, J. Pommerehne, H. Bässler, A. Greiner and W. Heitz, Makromolecular Chemistry and Physics, 1997, 198, 5, 1511.

176

Polymers for Light Emitting Diodes 80. S. Höger, J.J. McNamara, S. Schricker and F. Wudl, Chemistry of Materials, 1994, 6, 171. 81. D.H. Hwang, I.N. Kang, M.S. Jang, H.K. Shim and T. Zyung, Polymer Bulletin, 1996, 36, 3, 383. 82. S.T. Kim, D.-H. Hwang, X.C. Li, J. Grüner, R.H. Friend, A.B. Holmes and H.K. Shim, Advanced Materials, 1996, 8, 12, 979. 83. B.R. Hsieh, Y. Yu, E.W. Forsythe, G.M. Schaaf and W.A. Feld, Journal of the American Chemical Society, 1998, 120, 231. 84. B.R. Hsieh, H. Antoniadis, D.C. Bland and W.A. Feld, Advanced Materials, 1995, 7, 1, 36. 85. F.H. Boardman, A.W. Grice, M.G. Rüther, T.J. Sheldon, D.D.C. Bradley and P.L. Burn, Macromolecules, 1999, 32, 1, 111. 86. J. Cornil, D.A. dos Santos, X. Crispin, R. Silbey and J.L. Brédas, Journal of the American Chemical Society, 1998, 120, 1289. 87. S. Son, A. Dodabalapur, A.J. Lovinger and M.E. Galvin, Science, 1995, 269, 376. 88. E. Westerweele, P. Smith and A.J. Heeger, Advanced Materials, 1995, 7, 9, 788. 89. H. Vestweber, J. Oberski, A. Greiner, W. Heitz, R.F. Mahrt and H. Bässler, Advanced Materials for Optics and Electronics, 1993, 2, 197. 90. X.L. Chen and S.A. Jeneke, Applied Physics Letters, 1997, 70, 4, 487. 91. D.R. Baigent, P.J. Hamer, R.H. Friends, S.C. Moratti and A.B. Holmes, Synthetic Metals, 1995, 71, 1/3, 2175. 92. N.C. Greenham, S.C. Moratti, D.D.C. Bradley, R.H. Friend and A.B. Holmes, Nature, 1993, 365, 628. 93. S.C. Moratti, R. Cervini, A.B. Holmes, D.R. Baigent, R.H. Friend, N.C. Greenham, J. Grüner and P.J. Hamer, Synthetic Metals, 1995, 71, 1/3, 2117. 94. A.W. Grice, A. Tajbakhsh, P.L. Burn and D.D.C. Bradley, Advanced Materials, 1997, 9, 15, 1174. 95. Z. Yang, B. Hu and F.E. Karasz, Macromolecules, 1995, 28, 6151. 96. S.T. Pasco, P.M. Lahti and F.E. Karasz, Macromolecules, 1999, 32, 6993.

177

Handbook of Polymers in Electronics 97. D.H. Hwang, S.T. Kim, H.K. Shim, A.B. Holmes, S.C. Moratti and R.H. Friend, Synthetic Metals, 1997, 84, 615. 98. Z. Peng, J. Zhang and B. Xu; Macromolecules, 1999, 32, 5162. 99. H. Kim, M. Ryu and S.M. Lee, Macromolecules, 1997, 30, 1236. 100. D.J. Kim, S.H. Kim, T. Zyung, J.J. Kim, I. Cho and S.K. Choi, Macromolecules, 1996, 29, 3657. 101. H. Spreitzer, H. Becker, E. Kluge, W. Krueder, H. Shenk, R. Demandt and H. Schoo, Advanced Materials, 1998, 10, 16, 1340. 102. Z. Peng, Z. Bao and M. E. Galvin, Advanced Materials, 1998, 10, 9, 681. 103. Z.K. Chen, H. Meng, Y.H. Lai and W. Huang, Macromolecules, 1999, 32, 13, 4351. 104. M. Halim, I.D.W. Samuel, E. Rebourt and A.P. Monkman, Synthetic Metals, 1997, 84, 951. 105. X.C. Li, F. Cacialli, R. Cervini, A.H. Holmes, S.C. Moratti, A.C. Grimsdale and R.H. Friend, Synthetic Metals, 1997, 84, 159. 106. D.O. Brien, M.S. Weaver, D.G. Lidzey and D.D.C. Bradley, Applied Physics Letters, 1996, 69, 881. 107. F. Cacialli, X.C. Li, R.H. Friend, S.C. Moratti and A.B. Holmes, Synthetic Metals, 1995, 75, 161. 108. W. Huang, H. Meng, W.L. Yu, J. Cao and A.J. Heeger, Advanced Materials, 1998, 10, 8, 593. 109. V. Boucard, D. Adès, A. Siove, D. Romero, M. Schaer and L. Zuppiroli, Macromolecules, 1999, 32, 4729. 110. M. Redecker, D.D.C. Bradley, M. Inbasekaran, W.W. Wu and E.P. Woo, Advanced Materials, 1999, 11, 3, 241. 111. A. Bacher, C.H. Erdelen, W. Paulus, H. Ringsdorf, H.W. Schmidt and P. Schuhmacher, Macromolecules, 1999, 32, 4551. 112. W.J. Feast, R.J. Peace, I.C. Sage and E.L. Wood, Polymer Bulletin, 1999, 42, 167. 113. Y.E. Kim, H. Park and J.J. Kim, Applied Physics Letters, 1996, 69, 599.

178

Polymers for Light Emitting Diodes 114. P.K.H. Ho, M. Ganström, R.H. Friend and N.C. Greenham, Advanced Materials, 1998, 10, 10, 769. 115. F. Nüesch, L. Si-Ahmed, B. Francois and L. Zuppiroli, Advanced Materials, 1997, 9, 3, 222. 116. Y. Yang and A.J. Heeger, Applied Physics Letters, 1994, 64, 6245. 117. Y. Yang, E. Westerweele, C. Zhang, P. Smith and A.J. Heeger, Journal of Applied Physics, 1995, 77, 694. 118. G.A. Sotzing, J.R. Reynolds and P.J. Steel, Advanced Materials, 1997, 9, 10, 795. 119. W. Bantikassegn and O. Inganäs, Thin Solid Films, 1997, 293, 138. 120. W.R. Salaneck, S. Strafstrom and J.L. Brédas, Conjugated Polymer Surfaces and Interfaces: Electronic and Chemical Structure of Interfaces for Light Emitting Devices, Cambridge University Press, Cambridge, 1996. 121. G. Grem, G. Leditsky, B. Ulrich and G. Leising, Advanced Materials, 1992, 4, 36. 122. M. Rehahn, A.D. Schlüter, G. Wegner and J. Feast, Polymer, 1989, 30, 1060. 123. M. Rehahn, A.D. Schlüter, G. Wegner and J. Feast, Polymer, 1989, 30, 1054. 124. Y. Yang, Q. Pei and A.J. Heeger, Journal of Applied Physics, 1995, 34, 1587. 125. V. Cimrova, M. Remmers, D. Neher and G. Wegner, Advanced Materials, 1996, 8, 2, 146. 126. M. Remmers, D. Neher, J. Grüner, R.H. Friend, G.H. Gelinck, J.M. Warman, C. Quattrocchi, D.A. dos Santos and J.L. Brédas, Macromolecules 1996, 29, 7432. 127. U. Scherf and K. Mullen, Macromolecules 1992, 25, 3546. 128. G. Grem, C. Paar, J. Stampfl, and G. Leising, J. Huber and U. Scherf, Chemistry of Materials, 1995, 7, 1, 2. 129. A. Tasch, A. Niko, G. Leising and U. Scherf, Materials Research Society Symposium Proceedings, 1996, 413, 71. 130. S. Tasch, C. Brandstätter, F. Meghdadi, G. Leising, G. Froyer and L. Athouel, Advanced Materials, 1997, 9, 1, 33. 131. C. Weder and M.S. Wrighton, Macromolecules, 1996, 29, 5157.

179

Handbook of Polymers in Electronics 132. T. Yamamoto, K. Honda, N. Ooba and S. Tomaru, Macromolecules, 1998, 31, 7. 133. L. Kloppenburg, D. Jones, J.B. Claridge, H.C. Loyle and U.H. Bunz, Macromolecules, 1999, 32, 13, 4460. 134. C. Weder, C. Sarwa, A. Montali and P. Smith, Advanced Materials, 1997, 9, 1035. 135. B-C. Weder, C. Sarwa, A. Montali, C. Bastainnsnsenn and P. Smith, Science, 1998, 279, 835. 136. E.P. Woo, M. Inbasekaran, W.R. Shiang and G.R. Roof, Chemical Abstracts, 1997, 126, 225700y. 137. E.P. Woo, M. Inbasekaran, W.R. Shiang and G.R. Roof, inventors, Dow Chemical Company, WO 97 05,184, 1997. 138. M. Fukuda, K. Sawada and K. Yoshino, Journal of Polymer Science A, Polymer Chemistry, 1993, 31, 2465. 139. Q. Pei and Y. Yang, Journal of the American Chemical Society, 1996, 118, 7416. 140. M. Ranger, D. Rondeau and M. Leclerc, Macromolecules, 1997, 30, 7686. 141. Q. Pei, G. Yu, C. Zhang, Y. Yang and A.J. Heeger, Science, 1995, 269, 1086. 142. Y. Yang and Q. Pei, Polymer Prepprints, 1997, 38, 335. 143. M. Kreyenschmidt, G. Klaerner, T. Fuhrer, J. Ashnerhurst, S. Karg, W.D. Chen, V.Y. Lee, J.C. Scott and R.D. Miller, Macromolecules, 1998, 31, 4, 1099. 144. G. Klärner, M.H. Davey, W.D. Chen, J.C. Scott and R.D. Miller, Advanced Materials, 1998, 10, 993. 145. M. Grell, W. Knoll, D.M. Lupo, A. Meisel, T. Miteva, D. Neher, H.G. Nothofer, U. Scherf and A. Yasuda, Advanced Materials, 1999, 11, 8, 671. 146. M. Granström, Polymer for Advanced Technology, 1997, 8, 7, 424. 147. M. Catellani, S. Luzzati, R. Mendichi and A.G. Schieroni, Polymer, 1996, 37, 3, 105. 148. M. Berggren, O. Inganäs, G. Gustafsson, J. Rasmusson, M.R. Andersson, J. Rasmusson, M.R. Andersson, T. Hjertber and O. Wennerström, Nature, 1994, 372, 444. 149. R.E. Gill, G.G. Malliaras, J. W.Wilderman and G. Hadziioannou, Advanced Materials, 1994, 6, 2, 132.

180

Polymers for Light Emitting Diodes 150. P. Dyereklev, D. Berggren, O. Inganäs, M.R. Andersson, T. Hjertberg and O. Wennerström, Advanced Materials, 1994, 7, 1, 43. 151. A. Bolognesi, G. Bajo, J. Paloheimo, T. Ostergard and H. Stubb, Advanced Materials, 1997, 9, 2, 121. 152. P. Barta, J. Sanetra and M. Zagòrska, Synthetic Metals, 1998, 94, 119. 153. N. Nishino, G. Yu, A.J. Heeger, T.A. Chen and R.D. Rieke, Synthetic Metals, 1995, 68, 3, 243. 154. F. Chen, P.G. Mehta, L. Takiff and R.D. McCullogh, Journal of Materials Chemistry, 1996, 6, 11, 1763. 155. P. Barta, W.R. Salaneck, M. Zagòrska, A. Pròn and S. Niziol, Advanced Materials for Optics and Electronics, 1996, 6, 406. 156. M. Theander, O. Inganäs, W. Mammo, T. Olinga, M. Svensson and M.R. Andersson, Journal of Physical Chemistry B, 1999, 103, 7771. 157. Y. Cao, G. Yu and A.J. Heeger, Advanced Materials, 1998, 10, 12, 917. 158. R.M. Souto Maior, K. Hinkelmann, H. Eckert and F. Wudl, Macromolecules, 1990, 23, 1268. 159. R.S. Loewe, S.M. Khersonsky and R.D. McCullough, Advanced Materials, 1999, 11, 3, 250. 160. A. Bolognesi, W. Porzio, G. Bajo, G. Zannoni and L. Fannig, Acta Polymerica, 1999, 50, 4, 151. 161. S. Guillerez and G. Bidan, Synthetic Metals, 1998, 93, 123. 162. A. Iraqi and G.W. Barker, Journal of Materials Chemistry, 1998, 8, 1, 25. 163. X. Wu, T.A. Chen and R.D. Rieke, Macromolecules, 1996, 29, 24, 7671. 164. Y. Ohmori, Y. Hironaka, M. Yoshida, A. Fuji, N. Tada and K. Yoshino, Polymers for Advanced Technologies, 1997, 8, 7, 403. 165. K.S. Narayan and G.L. Murthy, Chemical Physics Letters, 1997, 276, 441. 166. S.D. Jung, T. Zyung, W.H. Kim, C.J.L. Lee and S.K. Tripathy, Synthetic Metals, 1999, 100, 223.

181

Handbook of Polymers in Electronics 167. D.B. Romero, M. Schaer, M. Leclerc, D. Adès, A. Sdiove, L. Zuppiroli, Synthetic Metals, 1996, 80, 271. 168. A. Bolognesi, C. Botta, L. Cecchinato, V. Fattori and M. Cocchi, Synthetic Metals, 1999, 106, 183. 169. C.O. Yoon, M. Reghu, A.J. Heeger, E.B. Park, Y.W. Park, K. Akagi, and H. Shirakawa, Synthetic Metals, 1995, 69, 79. 170. A. Moutali, C. Bastiaansen, P. Smith and C. Weder, Nature, 1998, 392, 261. 171. J. Grüner, M. Remmers, D. Neher, Advanced Materials, 1997, 9, 964. 172. M. Hamaguchi and K. Yoshino, Applied Physics Letters, 1995, 67, 3381. 173. M. Jandke, P. Strohriegl, J. Gmeiner, W. Brütting and M. Schoerer, Synthetic Metals, 2000, 111, 177. 174. M. Grell and D.D.C. Bradley, Advanced Materials, 1999, 11, 895. 175. M. Grell, D.D.C. Bradley, M. Inbasekaran and E.P. Woo, Advanced Materials, 1997, 9, 10, 798. 176. N.A.J.M. van Aerle, M. Barmentlo and R.W.J. Hollering, Journal of Applied Physics, 1993, 74, 3111. 177. B. Schartel, V. Wachtendorf, M. Grell, D.D.C. Bradley and M. Hennecke, Physical Reviews B, 1999, 60, 277. 178. Y. Kaminorz, E. Smela, O. Inganäs and L. Brehmer, Advanced Materials, 1998, 10, 10, 765. 179. B.H. Cumpston, I.D. Parker and F. Jensen, Journal of Applied Physics, 1997, 81, 3716. 180. M.S.A. Abdou, F.P. Orfino, Z.W. Xie, M.J. Deen and S. Holdcroft, Advanced Materials, 1994, 6, 830. 181. I.D. Parker, Y.Cao and C.Y. Yang, Journal of Applied Physics, 1999, 85, 2441. 182. D.G. Lidzey, D.D.C. Bradley, M.S. Skolnick, T. Virgili, S. Walker and D.M. Whittaker, Nature, 1998, 395, 53. 183. G.R. Hayes, F. Cacialli and R.T. Phillips, Physical Reviews B, 1997, 56, 8, 4798. 184. A. Dodabalapur, L.J. Rothberg, R.H. Jordan, T.M. Miller, R.E. Slusher and J.M. Phillips, Applied Physics Letters, 1996, 80, 12, 6954.

182

Polymers for Light Emitting Diodes 185. J. Grüner, F. Cacialli and R.H. Friend, Journal of Applied Physics, 1996, 80, 207. 186. F. Cacialli, G.R. Hayes, J. Grüner, R.T. Phillips and R.H. Friend, Synthetic Metals, 1997, 84, 533. 187. A. Dodabalapur, L.J. Rothberg and T.M. Miller, Applied Physics Letters, 1994, 65, 2308. 188. N. Tessler, S. Burns, H. Becker and R.H. Friend, Applied Physics Letters, 1997, 70, 5, 556. 189. R. Sastre and A. Costela, Advanced Materials, 1995, 7, 2, 198. 190. N. Tessler, G.J. Denton and R.H. Friend, Nature 1996, 382, 695. 191. W. Holzer, A. Penzkofer, S-H. Gong, A. Bleyer and D.D.C. Bradley, Advanced Materials, 1996, 8, 12, 974. 192. G.J. Denton, N. Tessler, M.A. Stevens and R.H. Friend, Advanced Materials, 1997, 9, 7, 547. 193. S.V. Frolov, M. Shkunov, Z.V. Vardeny and K. Yoshino, Physical Reviews B, 1997, 56, 8, R4363. 194. V.G. Kozlov, V. Bulovic, P.E. Burrows and S.R. Forrest, Nature, 1997, 389 195. F. Hide, M.A. Díaz-García, B.J. Schwartz and A.J. Heeger, Accounts of Chemical Research, 1997, 30, 10, 430. 196. S. Stagira, M. Zavelani-Rossi, M. Nisoli, S. De Silvestri, G. Lanzani, C. Zenz, P. Mataloni and G. Leising, Applied Physics Letters, 1998, 73, 20, 2860. 197. V.G. Kozlov, V. Bulovic, P.E. Burrows, M. Baldo, V.B. Khalfin, G. Paethasarathy, S.R. Forrest, Y. You and M.E. Thompson, Journal of Applied Physics, 1998, 84, 8, 4096. 198. Gupta, M. Stevenson, A. Dogariu, M.D. McGehee, J.Y. Park, V. Srdanov and H. Wang, Applied Physics Letters, 1998, 73, 24, 3492. 199. D.G. Lidzey, D.D.C. Bradley, S.F. Alvarado and P.F. Seidler, Nature, 1997, 386, 135.

183

Handbook of Polymers in Electronics

184

6

Photopolymers and Photoresists for Electronics J.-C. Dubois

6.1 Introduction There has been tremendous progress in integrated circuit technology since its beginning about 25 years ago. This has given rise to a continuous increase in chip size and in the density of components combined with a reduction in their cost. In addition, other characteristic parameters, such as power consumption, switching speed and reliability, have been improved. Integrated circuits consist of patterned, thin films of semiconductors, metals and dielectrics on a semiconducting substrate such as silicon or gallium arsenide. Today the so-called very large scale integration (VLSI) technology is used, permitting the manufacture of integrated circuits with several millions of discrete transistors working at frequencies reaching GHz. This trend to miniaturisation is continuing, leading to a demand for sub-micron device geometry. This evolution of the minimum size of the VLSI is known as More’s law and can be represented by Figure 6.1.

Figure 6.1 More’s law showing the evolution of the minimum size of VLSI versus year

185

Handbook of Polymers in Electronics Such resolution cannot be reached by using conventional (or near-UV) photolithography because of diffraction phenomena. Here other technologies with potentially higher resolution such as deep-UV photolithography, electron-beam lithography, X-ray lithography and ion beam lithography are under development. Electron-beam lithography is already used commercially for fabricating master masks for photolithography. The formation of the desired patterns on the substrates requires the use of a resist coating in which the pattern is first generated. Depending on the wavelength of the irradiation, different minimum linewidth are obtained. The different techniques of lithography are summarised in Table 6.1. One can expect the development of deep-UV lithography techniques at 193 nm giving line dimensions as small as 0.12 μm, necessary for the latest VLSI. The resist layer, usually a polymeric film (0.5-1.0 μm thick) is spin-coated onto the substrates and then exposed to radiation. One of the key points for obtaining high resolution is to obtain polymers with sensitivities well adapted to the chosen type of radiation and with response speeds which allow high production rates.

Table 6.1 Microlithography techniques λ (nm)

E(eV)

Pros

Cons

Resolution (μm)

320-480

3.9-2.6

Global exposure

Sensitivity to visible

>0.7

g-line

436

2.8

Steppers

Limited resolution

>0.35

i-line

365

4

Steppers

Deep-UV

248

(6.2-4.8)

No vacuum necessary

Limited resolution

>0.24

Deep-UV

193

No vacuum necessary

Resist to be optimised

>0.12

Technique Contact (vis.UV)

X-ray E-beam

Ion-beam

186

>0.30

0.5-5

2500-250

Global exposure

Resist sensitivity

>0.1

0.01-0.02 (corresponding to an associated wavelength λ)

0.5-2 x 103

Direct writing

Vacuum necessary

<0.1

High resolution

Vacuum necessary

<0.1

Photopolymers and Photoresists for Electronics This chapter describes some aspects of polymeric resist materials for microlithography allowing the necessary dimensions of VLSI to be reached.

6.2 Microlithography Process [1, 2] In order to have a better understanding of a resist’s requirements, a description of the different processing steps used in microlithography is necessary. In the example shown in Figure 6.2, the resist is applied as a thin film to the substrates, consisting of SiO2 on Si [3, 4].

Figure 6.2 Microlithography principle

187

Handbook of Polymers in Electronics

6.2.1 Resist Coating Polymeric resists are usually deposited on the substrates by spin coating, followed by baking in order to eliminate the residual solvent and to suppress mechanical stress. Some attempts have been made to deposit polymeric films by the plasma deposition technique [5].

6.2.2 Exposure In this step, the resist layer is exposed, generally through a mask, to one of the following types of radiation: UV light (including recently deep-UV), ions, electrons or X-rays. The lithographic method could be contact, proximity or projection mode [6], Figure 6.3. In the contact mode the mask is in contact with the substrate to be treated. In the proximity mode there is a gap between the mask and the substrate, while in the projection mode the image of the mask is projected on the substrate. The projection mode is the best method for VLSI production since it allows more precision and involves less deterioration of the mask. The source of radiation is generally a mercury lamp or a mercury-rare gas (xenon) discharge lamp. This gives a maximum of radiation in the 350-450 nm range. The steppers use classically a high-pressure mercury lamp with two radiations: i-line (436 nm) and gline (365 nm). The recent development of excimer lasers emitting in the deep-UV region has introduced the use of deep-UV radiation (150-300 nm), and especially the 248 and 193 nm wavelengths [7, 8, 9].

Figure 6.3 The different lithographic methods: contact, proximity or projection mode

188

Photopolymers and Photoresists for Electronics The resist contains radiation-sensitive groups which undergo chemical reactions when exposed, leading to the formation of a latent image which closely matches the pattern of the mask. Organic-based resists have been classified according to whether they consist of one or two components. In one-component systems, the structure of the polymer contains the radiation-sensitive groups either in the main chain or in the side chain. In two-component systems, the resist consists of an inert matrix resin and a radiation-sensitive molecule called a sensitiser, which is usually, a low molecular weight compound.

6.2.3 Development Development is an important step of the process, since it is one of the keys for obtaining a well-defined pattern on the substrates. Development transforms the latent image into an image serving as a mask for etching of the substrates. Two technologies are now available: wet development, which is widely used in circuit manufacture and dry development, which is still under study. Traditionally, resists have been divided into two classes depending on their behaviour upon irradiation: positive and negative resists (Figure 6.2). Positive resists become more soluble in the irradiated area relative to the unexposed area whereas negative resists become less soluble in the irradiated area relative to the unexposed area.

6.2.3.1 Wet Development Wet development can be based on three different types of radiation-induced changes: (i) variation in molecular weights of the polymers (by crosslinking or by chain scission), (ii) reactivity change, and (iii) polarity change of the polymer. However, the use of a solvent may cause swelling and may lead to a lack of adhesion of the resist to the substrate. These problems may be solved by using dry development techniques which have recently been introduced.

6.2.3.2 Dry Development Dry development may be achieved using either a vapour phase process or a plasma (usually oxygen plasma).

189

Handbook of Polymers in Electronics

6.2.4 Post Baking Post baking is used as a post-development step to eliminate the residual development solvent, to improve adhesion between the resist and the substrate and chiefly to increase the resistance of the resist to the subsequent etching process.

6.2.5 Etching [1, 3] The most common method of pattern transfer to the substrate is by wet chemical etching. However, all semiconductor wet etching processes exhibit the same basic limitation. This limitation is due to the isotropy of the process, which makes linewidth control difficult for features less than 2 μm when thick substrate layers are used. The need to transfer fine features in thick substrates has led to the development of anisotropic etching techniques such as plasma etching, reactive ion etching (RIE) and sputter etching. Dry etching is more economical, gives rise to less pollution and leads to integrated circuits with higher densities than does wet etching. As a rule, aromatic polymers resist plasma etching better than aliphatic polymers.

6.2.6 Resist Removal (Stripping) The main goal of this step is the complete removal of the resist without affecting the wafer surface. Two methods are used: wet stripping (the use of either solvents, which dissolve the resists, or oxidisers that transform the resist into carbon dioxide and waste) and plasma resist stripping.

6.2.7 Doping During this step very small amounts of ‘impurities’ (e.g., boron, arsenic) diffuse into the stripped regions of the semiconductor substrates. The cycle described in this section has to be repeated several times to obtain the integrated circuit component.

6.3 Resist Requirements The main requirements of a resist are solubility, adhesion, etch resistance, sensitivity to radiation and contrast.

190

Photopolymers and Photoresists for Electronics

6.3.1 Solubility Resists are generally deposited onto substrates by spin coating; therefore, solubility in organic solvents is necessary.

6.3.2 Adhesion The resist must possess good adhesion properties to various substrates, such as metal, silicon dioxide, silicon nitride or semiconductors, throughout the various steps of integrated circuit manufacture. Poor adhesion leads to loss of resolution.

6.3.3 Etching Resistance Wet chemical etching requires good adhesion and chemical stability of the resist towards acidic or basic etching solutions. Most dry etching techniques involve a high radiation flux and temperatures often higher than 80 °C. Polymers exhibiting high Tg values and containing radiation stable groups (e.g., aromatic structures) have higher dry etch resistance.

6.3.4 Sensitivity and Contrast The sensitivity, σ, is related to the ability of a polymer to undergo structural modification on irradiation. Sensitivity is said to increase as the dose required to produce the lithographic image decreases. The sensitivity of a positive resist, σ0, is the dose required to achieve complete solubility of the exposed region under conditions where the unexposed region remains completely insoluble. The sensitivity of a negative resist is conventionally defined as the dose at which 70% of the original film thickness has been retained after development, σ0,7. The required sensitivity varies with the type of irradiation and is expressed as energy/surface (e.g., J/m2). The contrast, γ, is related to the ability of a polymer to give vertical sidewalls. Resolution (defined as the smallest linewidth which can be achieved) depends on the contrast. The parameters are determined from the sensitivity curve of the resist, which expresses the normalised film thickness, er/e0 (er is the thickness after development and irradiation, e0 is the initial thickness of the resist), as a function of log10 (Dose) (Figures 6.4(a) and 6.4(b)). For a negative resist: γ = (log10 σ1/σ0)-1

(6.1)

191

Handbook of Polymers in Electronics

(a)

(b)

Figure 6.4 Typical sensitivity curves for negative and positive resists

and for a positive resist: γ = ⎜(log10 σ1/σ0)-1⎜

(6.2)

where σ1 is the dose required to reach the required thickness after development. For narrow-UV radiation, values of the doses for the i- and g-line vary from 10-50 mJ/cm2 and for deep-UV, from 10-20 mJ/cm2. For electron beam lithography, this dose is expressed as a current density around 5-10 mC/cm2. Negative resists are in general more sensitive than positive resists but they exhibit a lower contrast (γ<1) (Table 6.2). A contrast higher than 3 is generally required for the high circuit density of new generation technologies such as the 256 Mb DRAM.

Table 6.2 Parameters affecting the characteristics of photoresists Characteristics

Positive

Negative

Radiation sensitivity

low

high

Contrast

high

low

Resolution capability

high

low

Resistance to etching

low

high

Stripping capability

high

low

192

Photopolymers and Photoresists for Electronics

6.4 Resist Materials 6.4.1 Conventional Photoresists Photolithography using light in the wavelength range 350-450 nm is the technology currently used in the manufacture of integrated circuit devices.

6.4.1.1 Positive Photoresists All near-UV positive photoresists are two-component systems; the polymeric material is a low molecular weight novolac polymer (Figure 6.5) and the sensitiser is a derivative of a 1,2-diazonaphthoquinone (DNQ) (20%-50% by weight). DNQ forms a complex with the phenol groups of the novolac resin and prevents the dissolution of the latter in an aqueous base.

Figure 6.5 Novolac polymer

Exposure of the resist to UV light results in photodecomposition of the sensitiser to an unstable ketocarbene. This reacts with water to produce the base-soluble indene carboxylic acid, which no longer inhibits dissolution of the novolac polymer in aqueous base [4] (Figure 6.6). There are many commercially available positive photoresists, which differ slightly from one another. The most significant advantages of this family of photoresists are the lack of swelling during the development step, a good resolution in thick coating and wellknown film forming properties. Since swelling does not take place during the development of positive photoresists, several process variations have been reported aimed at reversing the tone of the image so that the resist would act as a negative resist [11, 12]. The most interesting process is probably that described by Moritz and Paal [13] and known as the Monazoline process. 193

Handbook of Polymers in Electronics

Figure 6.6 Photodecomposition of 1,2-diazonaphthoquinone

The process is outlined in Figure 6.7; it is based on the thermal decomposition of indene carboxylic acids in the presence of amines such as monazolines, a range of commercially available detergents (1-hydroxyethyl-2-alkyl-imidazolines).

6.4.1.2 Negative Photoresists 6.4.1.2.1 One-Component Systems In this type of photosensitive system, UV sensitive groups, such as chalcone, cinnamate, styrylacrylate, diphenyl-cyclopropene carboxylate, cinnamilidene malonate, p-carboxycinnamate, p-phenylene-bis acrylate and styrylpyridinium, etc., are included in the main chain or in the side chain of the polymer. UV irradiation gives rise to crosslinking. The crosslinking of poly(vinylcinnamate) [14], which is the key process occurring in the KPR resist (Kodak), is shown in Figure 6.8.

194

Photopolymers and Photoresists for Electronics

Figure 6.7 Monazoline process: the thermal decomposition of the carboxylic acid in presence of the monazoline base transforms the acid into an insoluble indene derivative leading to a negative resist (reverse tone process)

Figure 6.8 The KPR resist, a negative resist which crosslinks through the double bonds under irradiation

195

Handbook of Polymers in Electronics 6.4.1.2.2 Two-Component Systems Negative near-UV resists are derived mainly from photoinduced crosslinking of partially cyclized cis-1,4-polyisoprene with bisazido sensitisers [15]. Cyclised cis 1,4-polyisoprene is obtained by heating 1,4-polyisoprene in the presence of a Lewis acid (ZnO2, AICI3, SnCl4) (Figure 6.9).

Figure 6.9 Formation of cyclised cis-1,4-polyisoprene

The photochemical transformations of a bisazide sensitiser leading to the crosslinking of a cyclised cis-1,4-polyisoprene are shown in Figure 6.10. An insoluble three-dimensional crosslinked network is obtained.

Figure 6.10 Photochemical transformations of a bisazide sensitiser in a carbene leading to the cross-linking of cyclized cis-1,4-polyisoprene. There are several different possible reaction mechanisms.

196

Photopolymers and Photoresists for Electronics These resists are commercially available and are able to meet the requirements of most integrated circuit designs. Their sensitivity is high, allowing good masking speed. For example, KTFR (made by Kodak) is a negative resist of this type. The main limitation of negative resists of this type is their dependence on organic solvent developers, which cause resist image distortion as a result of swelling. However, for high resolution circuits, positive resists are usually preferable.

6.4.2 Deep-UV Photoresists Interest in deep-UV photolithography (150-280 nm range) lies in the possibility of obtaining higher resolutions than cannot be achieved with conventional optical technology, since the diffraction-limited resolution of optical projection printing tools is proportional to the exposure wavelength. Resolution limits in the 0.5-0.25 mm range may be expected. One of the problems of deep-UV photolithography is the low brightness of mercury arc lamps, which are the most common exposure sources. That is to say that new exposure sources and/or very sensitive resist materials must be developed. That is why there is now a great interest in excimer lasers [1, 16, 17] as intense deep-UV sources. The excimer lasers are mainly lasers based on KrCI, KrF and ArF, which have output wavelengths of 222, 249 and 193 nm, respectively. KrF exposure systems (step and repeat) already exist commercially and are used for 0.25 μm linewidth generation. But most emphasis has been recently put on the 195 nm exposure systems for 0.18 μm linewidth generation. The next generation of components will require linewidths below 0.12 μm.

6.4.2.1 Positive Deep-UV Resists 6.4.2.1.1 One-Component Systems One-component positive resists are essentially copolymers or terpolymers derived from PMMA. Copolymer structures are chosen in order to increase the low value of the absorption coefficient of PMMA at 215 nm. Comonomers are selected either to absorb in the 230-280 nm range or because they contain a photosensitive chromophoric group. Poly(methyl methacrylate-co-3-oximino-2-butanone methacrylate-co-methacrylonitrile), p(PMMA-OM-MAN) (Figure 6.11), sensitised with t-butyl benzoic acid requires an exposure dose of less than 30 mJ/cm2 [18]. The sensitivity of this terpolymer is 170 times that of PMMA and is capable of sub-micron resolution. Poly(aromatic sulfones) may also be used as positive resists [19]. Their quantum efficiency is low, however. Aromatic moieties ensure a useful plasma etch resistance.

197

Handbook of Polymers in Electronics

Figure 6.11 Terpolymer of PMMA sensitive to 240 nm radiation

6.4.2.1.2 Two-Component Systems Several attempts have been made to design two-component deep-UV positive systems with increased sensitivity. An example is a system based on a novolac polymer and 5diazo Meldrum’s acid as a dissolution inhibitor [20]. The Meldrum’s acid derivative provides a bleachable chromophore at 250 nm, which is converted to volatile compounds on irradiation according to the scheme shown in Figure 6.12.

Figure 6.12 Meldrum’s acid is converted to a volatile compound under irradiation at 250 nm

The resist exhibits reasonable sensitivity. However, the novolac resin itself has a strong absorption around 310 nm, which is bad for the sensitivity. Wilkins and others [21] have described another system. In this case, the resin is a copolymer of methyl methacrylate and methacrylic acid (which is transparent above 260 nm and which is soluble in aqueous alkali) and the dissolution inhibitor is an o-nitrobenzyl carboxylate. This system exhibits a good photosensitivity (100 mJ/cm2 in the 230-300 nm range) with a very high contrast (g>5).

198

Photopolymers and Photoresists for Electronics 6.4.2.1.2.1 t-Butyl Carbonate (T-BOC) Photoresists The progress in two-component systems is due to the combination of two things: •

an acid photogeneration and

•

a polymer with labile groups sensitive to the acid (e.g., T-BOC) leading to positive resists [22]. The i-line and g-line resists, such as novolac resists, absorb too much and give bad contrast when used as deep-UV resists in the 248-193 nm domains.

Very good results are obtained with the use of a different approach to photoactivation using the t-butyl carbonate (T-BOC) group. This group is fixed on a polymer and used with an acid photogenerator. The acid generated decomposes the T-BOC group, leading to a soluble polymer. 6.4.2.1.2.2 Acid Photogenerators A typical acid photogenerator is triphenyl sulfonium triflate (Figure 6.13), which generates trifluoromethane sulfonic acid. Onium salt photoinitiators of this type can induce cationic photopolymerisation under photoirradiation. The reaction is complex and several products of this type are used with Lewis acids, such as AsPF6–, AsF6– and BF4–. See, for example, [44].

Figure 6.13 Example of an acid photogenerator

Other acid photogenerators are used, such as diphenyliodonium hexafluoropropane sulfonate, or bis-(diphenylsulfonyl)diazomethane. The latter (shown in Figure 6.14) generates benzene sulfonic acid.

199

Handbook of Polymers in Electronics

Figure 6.14 Bis(diphenylsulfonyl)diazomethane

6.4.2.1.2.3 T-BOC Polymers Different types of T-BOC polymers have been synthesised and are used in association with the photoacid generators. A resist made with poly(p-tert-butoxycarbonyloxystyrene) (T-BOC-PHS) and an onium salt was first described by IBM Laboratories [49]. The TBOC-PHS is decomposed by the acid generated by the photoacid generator, CO2 and isobutylene is evolved (Figure 6.15).

Figure 6.15 Decomposition of T-BOC-PHS in an acid medium

The resolution of the system is moderate and the stability after irradiation is poor. However, it is used for the 243 nm domain due to its sensitivity. Derivatives of this resist system have also been proposed, e.g., a copolymer of 4-hydroxystyrene with a t-butyl acrylate and a phenyloxysuccinimid as the acid photogenerator. This resist is more stable during processing (Figure 6.16). An evolution toward the lower wavelength is in progress with the use of the 193 nm ArF laser source. Optical absorption requirements demand new resist polymers with no

200

Photopolymers and Photoresists for Electronics

Figure 6.16 Copolymer of t-butyl acrylate and 4-hydroxystyrene used as a deep-UV resist

aromatic structures. Polyacrylates seem adapted to this wavelength, but the lack of aromatic structure gives poor resistance to plasma etching. For this reason alicyclic radicals, such as adamantine, are used on the polymer [49-56]. AZ Electronic Materials, a subsidiary of Clariant Corporation, has proposed a resist with alicyclic radicals for use in 193 nm microlithography (Figures 6.17a and 6.17b).

Figure 6.17a Mechanism of degradation of the photoresist AX 1000

6.4.2.2 Negative Resists Microresists for shorter wavelengths (MRS) are developed in an aqueous base, which avoids the swelling phenomena. Resolution of 0.16 μm lines and spaces has been demonstrated. The MRS is composed of a phenolic resin and a bisazide [24].

201

Handbook of Polymers in Electronics

Figure 6.17b Lithography results with AX 1000 at 193 nm. The resolution is better than 0.15 μm. (Picture courtesy of AZ Electronic Materials, a business unit of Clariant Corporation)

Many other deep-UV negative resists have been reported. Examples are summarised in Table 6.3.

6.4.3 Electron-Beam Resists The major advantages of electron-beam lithography over conventional photolithography are a higher potential resolution and the possibility of direct beam writing on the resist surface. In addition to the usual requirements discussed earlier, positive or negative electron-beam resists have to possess the following properties: •

Be able to exhibit sub-micron resolution (resolution being affected more by backscattering of electrons than by diffraction effects since the de Broglie wavelength of electron beams is of the order of a few tenths of an Angstrom),

202

Photopolymers and Photoresists for Electronics

Table 6.3 Some examples of deep-UV negative resists Resist

Cyclised polyisoprene

Structure of repeat unit

Sensitiser

See Figure 6.9

Bis-azide

Sensitivity relative to PMMA

Tradename

Ref.

60-450

UR 110ODWR

23

Polyvinylphenol (MRS)

3,3′ Diazidodiphenyl sulfone

50

RD2000N

24

Poly(methylisopropenylketone) (PMIPK)

2,6-Di(4azido benzylidone)4-methylcyclohexanone

100

ONR20

25

None

40

Chloromethylated polystyrene

•

Have a sensitivity of the order of 10-6 C/cm2 (at 10-30 kV), and

•

Be sufficiently stable to withstand dry etching as well as wet etching.

26

6.4.3.1 Positive Electron-Beam Resists PMMA was one of the first positive resists to be investigated. When subjected to electronbeam irradiation, PMMA suffers extensive main chain scission, owing to the presence of quaternary carbon atoms. PMMA is an excellent positive resist from the viewpoint of adhesion and resolution, but it exhibits rather low sensitivity (50 μC/cm2 at 15 kV) and poor resistance to dry etching techniques.

203

Handbook of Polymers in Electronics

Table 6.4 Some electron beam positive resists Resist

Structure of repeat unit

Molecular weight

Tg Sensitivity (°C) 20 kV (μC m-2)

Ref.

Poly(hexafluorobutylmethacrylate) (FBPM-110, Daikin Kogyo)

5x105

50

0.4

27

Poly(dimethyltetrafluoropropyl methacrylate) (FPM, Daikin Kogyo)

106

93

3-12

29

Poly(trichloroethylmethacrylate (EBR-1, Toray Industries)

5.5x105

138

1.25

30

1.5

31

Poly(ethylcyanoacrylate)

2x105

One way to increase the sensitivity of PMMA in electron-beam lithography is to weaken the main chain stability of the polymer. Substitution either on the quaternary carbon (using polar electronegative substituents) or in the side chain may do this. Some examples are given in Table 6.4. In order to improve the thermal stability of PMMA, Roberts [32] has described a twocomponent electron-beam resist. By copolymerising methacryloyl chloride and methacrylic acid with methyl methacrylate, it is possible to form intermolecular crosslinks (acid anhydride bonds) in situ on the appropriate substrate by heating the resist after spin coating (Figure 6.18).

204

Photopolymers and Photoresists for Electronics

Figure 6.18 Crosslinking through an anhydride [32]

Electron-beam irradiation breaks the acid anhydride linkage, thus restoring solubility to the irradiated areas. Several systems derived from that of Roberts have been studied [33, 34]. Resistance to plasma and to developer solvent is improved by the anhydride crosslinking. Poly(olefin sulfones), alternating copolymers of sulfur dioxide and an olefin, are another important class of positive resists, which exhibit a very high sensitivity [35] owing to the weak carbon-sulfur bond (see Figure 6.19). Poly(butene-1-sulfone) shows the best properties (σ = 1.6 μC/cm2 at 20 kV) and is commercially available.

Figure 6.19 Mechanism of the degradation of polysulfone by electron-beam irradiation

Unfortunately, poly(olefin sulfones) are not sufficiently stable towards dry etching. In order to improve the etch resistance, a two-component system consisting of a novolac resin (known for its excellent dry etching resistance) and poly(2-methyl-1-pentene) (acting as a dissolution inhibitor) has been designed by Bowden and others [35, 36] at Lucent Technologies, USA. Poly(2-methyl-1-pentene) undergoes spontaneous depolymerisation during irradiation leading to gaseous compounds, and the exposed regions become soluble. Resists having the structure [37] shown in Figure 6.20 have good oxygen plasma resistance, presumably due to the conversion of Si to SiO2.

205

Handbook of Polymers in Electronics

Figure 6.20 Poly(olefin sulfone) resistant to plasma etching

6.4.3.2 Negative Electron-Beam Resists The chemistry of negative electron-beam resists derives from the ability of epoxy, episulfide, vinyl(allyl) and halogen (particularly chlorine) groups to promote crosslinking. However, the inherent sensitivity of these reactive moieties incorporated into the polymer is well within the required exposure range of most electron beam exposure machines. Incorporation of aromatic groups into the polymer improves the dry etch resistance and the resolution of negative electron-beam resists. Characteristics of some examples of negative electron-beam resists, including poly(acrylate) and polystyrene derivatives, are given in Table 6.5. The different structures of these polymers are given in Table 6.6.

Table 6.5 Some negative electron-beam resists Resist

Structure of repeat unit

Molecular weight

Sensitivity 20 kV (μC m-2)

Ref.

COP

Acrylate

1.8x105

0.6

38

GMC

Epoxy acrylate

2x105

4

39

Thiirane acrylate

5.5x105

0.6

40

PCMS

Polystyrene

4.5x105

0.3

41

PVMS

Polysiloxane

2.9x105

1.5

3, 42

P(ETMA-co-MMA)

COP = copolymer of ethylacrylate and methylmethacrylate GMC = poly(glycidylmethacrylate-co-m-chlorostyrene) P(ETMA-co-MMA) = poly(ethylthiomethacrylate-co-methylmethacrylate) PCMS = poly(m-chloromethylstyrene) PVMS = poly(methylvinyl siloxane)

6.4.4 X-Ray Resists X-ray lithography is an extension of near-contact lithography for replication of 1 μm and smaller geometry. X-ray masks are difficult to fabricate with sufficient yields. Resolution in X-ray lithography will never do better than the resolution of the electronbeam exposures used to make the masks, since X-rays cannot be focused or deflected.

206

Photopolymers and Photoresists for Electronics

Table 6.6 Formulae of some negative electron-beam resists

COP

GMC PVMS

P(ETMA-co-MMA)

PCS

For technological reasons, the energy flux deposited at the wafer is low, so highly sensitive resists (σ<10 mJ/cm 2) are required. A typical sensitivity curve of fluorinated polymethacrylate (FPB) in comparison to the curve of PMMA is shown in Figure 6.21.

Figure 6.21 Comparison of the sensitivity curve of a FPB resist (σ = 70 mJ/cm2) with that of PMMA (σ = 360 mJ/cm2) at X-ray irradiation of 13.34 Å [48]

207

Handbook of Polymers in Electronics It can be said that positive or negative resists developed for electron-beam lithography can be used for X-ray lithography. However, the sensitivity of conventional electronbeam resists is not sufficient from an economical point of view. The highest X-ray resist sensitivity has been obtained by using an elegant resist technique referred to as ‘photo-locking’ [47]. The resist consists of a plasma-degradable acrylic polymer (poly(2,3-dichloropropyl acrylate)) and a volatile silicon-containing acrylate (bis-acryloxybutyl tetramethyl disiloxane, Figure 6.22) which can be readily polymerised by the absorbed radiation.

Figure 6.22 Bis-acryloxybutyl tetramethyl disiloxane (BABTDS) used for photolocking

The sensitivity of this system is 1.5 mJ/cm2. Highly sensitive positive X-ray resists are not really well adapted at this time since they do not possess all the required sensitivity and masking properties. However, they offer interesting advantages for including inorganic atoms which could be transformed into oxide (e.g., silica). The X-ray masking technique itself suffers from problems such as the difficulty of mask manufacture.

6.4.5 Special Resists 6.4.5.1 Multilayer Resists The multilayer resist (MLR) strategy was introduced by Lin and co-workers [16, 17] and called ‘portable conformable masking’ (PCM). The aim of the multilayer resist technology is the simultaneous achievement of good linewidth control, high resolution and good step coverage. Generally, these requirements cannot be met simultaneously since good step coverage and leveling of topography variations on the wafer surface require a thick resist while high resolution is obtained with a thin resist. The main feature of multilayer systems is the separation of imaging and step-coverage functions in different layers. Highresolution patterns are generated in the thin top layer followed by pattern transfer into the thick bottom layer (bilayer resist systems). Sometimes, a third layer, called the isolation layer, is introduced between the two, preventing mixing of these products. This process is referred to as a tri-layer resist system (Figure 6.23).

208

Photopolymers and Photoresists for Electronics

Figure 6.23 Multilayer resist system

Some resists that have been previously described may be used in MLR technology, e.g., a novolac-type resist for the imaging layer and PMMA for the planarising layer. However, most polymers do not possess sufficient etching resistance to withstand oxygen plasma etching of the thick hard-baked resist layer during the process of a multilevel system. This has led to an extensive use of silicon-containing polymers since these materials are known to convert into SiO2 in oxygen plasma and thereby to exhibit high tolerance to oxygenreactive plasma etching. Examples of some silicone resists are shown in Table 6.7.

6.4.5.2 Polyimide-Based Photoresists [57, 58] Polyimides are interesting thermally-stable polymers which could resist at high temperatures (more than 300 °C). These are used as insulation layers in microelectronics due to their good dielectric properties. Photoresists based on polyimides are very attractive because they allow direct patterning of the polymer.

6.4.5.2.1 Negative Polyimides Most of the negative polyimides are based on the principles of the pioneering work of Rubner [58]. A soluble polyamic acid precursor carries a photo-reactive group sensitive in the 250-450 nm range able to crosslink under irradiation. After development, the crosslinked patterns are converted into polyimides by heating, as shown in Figure 6.24. In the system of Rubner, the sensitive group R allowing photocrosslinking is an acrylate ester. This product is now developed by CIBA (Probimide, Selectilux HTR) or DuPont. The acrylate is eliminated during subsequent heating.

209

Handbook of Polymers in Electronics

Table 6.7 Silicon-containing polymers used in multilayer resists Structure

Radiation

Company

Reference

Electron-beam

IBM

43

Deep UV

IBM

44

Electron-beam X-ray Deep UV

NTT

45

Near UV

NTT

46

Other types of photosensitive groups have been used. Yoda and Hiramoto [60] reported a negative photopolyimide having photosensitive groups introduced to polyimide precursors through acid-amine ion linkage. Toray developed this product.

6.4.5.2.2 Positive Polyimide Resists Positive polyimide resists have also been developed. One example starts from a soluble polymer [60] with a formulation analogous to a novolac positive resist; o-naphthoquinone diazide was used as a photoreactant linked to the polymer (Figure 6.25). The different characteristics of polyimide-based photoresists are represented in Table 6.8 [57].

210

Photopolymers and Photoresists for Electronics

Figure 6.24 Formation of polyimide, the polyamic acid is crosslinked by radiation through the acrylate and cyclised into polyimide by heating at 200 °C after development

Figure 6.25 Example positive photopolyimide resist material

211

Handbook of Polymers in Electronics

Table 6.8 Examples of polyimide-based photoresists [57] Manufacturer

Trademark

Type(1)

Dose (mJ cm-2)

Shrinkage (%)(2)

Thickness (μm)

Asahi (Pimel)

G -7610A TL500A IX - 3

E E E

400 400 300

48 50 40

10 5 5

Du Pont Electronics (Pyralin)

PI 2722 PI 2732 PI 2741

E E E

300 <500 80-200

50 50 45

5 5 10

Ciba-Geigy (Probimid)

Probimid 343 Probimid 400

E PI

200 700

40 <9

10 7.5

Amoco (Ultradel)

Ultradel 7501

PI

300

8

7.5

UR 3800 UR 5100

S S

150-250

30-40 50-62

10 10

Toray (Photoneece)

(1) Type E = ether, PI = precyclised polyimide, S = ester (2) Decrease of the film thickness during baking

6.5 Conclusions It is clear that the progress in VLSI technology is closely linked to lithographic techniques, as the size of the circuit decreases, the speed and the power consumption decreases. These techniques are dependent on the photochemistry and the resists. The minimum linewidth permitted is already around 0.1 μm or less with sources like electron-beam, Xrays or the 193 nm excimer laser. No doubt these limits will progress and one already speaks of 157 nm projection lithography. The requirements placed on photoresists will probably increase in the near future.

References 1.

S. Nonogaki, T. Ueno and T. Ito, Microlithography Fundamentals in Semiconductor Devices and Fabrication Technology, Marcel Dekker, New York, NY, USA, 1998.

2.

A. Eranian and J.C. Dubois, Revue Technique Thomson-CSF, 1987, 19, 1, 95.

3.

J.C. Dubois and M. Gazard, Revue Technique Thomson-CSF, 1974, 6, 4, 1169.

212

Photopolymers and Photoresists for Electronics 4.

J.C. Dubois, L’Actualité Chimique, 1999, 11, 86.

5.

S. Morita, J. Tamano, S. Hatori and M. Ieda, Journal of Applied Physics, 1980, 51, 3938.

6.

M. King in VSLI Electronics Microstructure Science, Volume 1, Ed., N.G. Einspruch, Academic Press, New York, NY, USA, 1981, Chapter 2.

7.

W.M. Moreau, Semiconductor Lithography: Principle, Practices and Materials, Plenum Press, New York, NY, USA, 1988.

8.

Introduction to Microlithography, 2nd Edition, Eds., L.F. Thompson, C.G. Willson and M.J. Bowden, American Chemical Society, Washington, DC, USA, 1994.

9.

Handbook of Microlithography, Micromachining and Microfabrication, Ed., P. Rai-Choudhury, SPIE Optical Engineering Press, Bellingham, WA, USA, 1997, 7.

10. J. Pacansky, Polymer Engineering and Science, 1980, 20, 1049. 11. W. Neugebauer, BR Patent 844,039, 1960. 12. W.G. Oldham and E. Heike, IEEE Electron Device Letters, 1980, 1, 217. 13. H. Moritz and G. Paal, inventors; International Business Machines Corporation, assignee; US Patent 4,104,070, 1978. 14. K.G. Clark, Microelectronics, 1971, 3, 11, 23. 15. J. Sagura and J.A. van Allan, inventors; US Patent 2,940,853, 1960. 16. K. Jain, C.G. Willson and B.J. Lin, IBM Journal of Research and Development, 1982, 26, 2, 151. 17. K. Jain, C.G. Willson and B.J. Lin, IEEE Electron Device Letters, 1982, 3, 3, 53. 18. E. Reichmanis and C.W. Wilkins, Journal of the American Chemical Society, Division of Organic Coatings and Plastics Chemistry Preprints, 1980, 43, 243. 19. M.J. Bowden and E.A. Chandross, Journal of the Electrochemical Society, 1975, 122, 10, 1371. 20. B.D. Grant, N.H. Clecak, R.J. Twieg and C.G. Willson, IEEE Transactions on Electron Devices, 1981, 28, 1300. 21. C.W. Wilkins, Jr., E. Reichmams and E.A. Chandross, Journal of the Electrochemical Society, 1982, 129, 11, 2552.

213

Handbook of Polymers in Electronics 22. N. Hacker, J. Fouassier and J. Rebek in Radiation Curing in Polymer Science and Technology, Volume 2, Elsevier, New York, NY, USA, 1993, Chapter 9. 23. T. Iwayanagi, T. Kohashi and S. Nonogaki, Journal of the Electrochemical Society, 1980, 127, 2759. 24. T. Iwayanagi, T. Kohashi, S. Nonogaki, T. Matsuzawa, K. Douta and H. Yanazawa, IEEE Transactions on Electron Devices, 1981, 28, 1306. 25. H. Nakane, A. Yokota, S. Yamamoto and W. Kanai, Proceedings of the Regional Technical Conference on Photopolymers: Principles, Processes and Materials, Mid-Hudson Section, SPE, Ellenville, NY, USA, 1982, 43. 26. S. Imamura and S. Sugarawa, Japanese Journal of Applied Physics, 1982, 21, 5, 776. 27. M. Kakuchi, S. Sugawara, K. Murase and K. Matsuyama, Journal of the Electrochemical Society, 1977, 124, 1648. 28. S. Sugawara, O. Kogure, K. Harada, M. Kakuchi, K. Sugegawa. S. Imamura and K. Miyoshi, presented at the 9th International Conference on Electron and Ion Beam Science and Technology, Extended Abstract, 680. (Part of the Electrochemical Society Spring Meeting, St. Louis, MO, USA, 1980) 30. T. Tada, Journal of the Electrochemical Society, 1979, 126, 9, 1635. 31. S. Matsuda, S. Tsuchiya, M. Honma and G. Nagamatsu, inventors; Matsushita Electrical Co., Ltd, assignee; US Patent 4, 279, 984, 1981. 32. E.D. Roberts, Applied Polymer Symposia, 1974, 23, 87; Journal of the American Chemical Society, Division of Organic Coatings and Plastics Chemistry Preprints, 1973, 33, 1, 359; 1975, 35, 2, 281; 1977, 37, 2, 36. 33. H. Saeki and M. Kohda, Proceedings of the 17th Symposium on Semiconductor and Integrated Circuit Technology, Tokyo, Japan, 1979, 48. 34. J.H. Lai and L.T. Shepherd, Proceedings of the 10th International Symposium on Electron and Ion Beam Science and Technology, 1983, 82-83, 185. 35. M.J. Bowden and L.F. Thompson, Solid State Technology, 1979, 22, 72. 36. M.J. Bowden, L.F. Thompson, S.R. Fahrenholtz and E.M. Doerris, Journal of the Electrochemical Society, 1981, 128, 1304. 37. K.B. Kilichowski and T.R. Pampalone, inventors; RCA Corporation, assignee; US Patent 4,357,369, 1982.

214

Photopolymers and Photoresists for Electronics 38. L.F. Thompson, J.P. Ballantyne and E.D. Feit, Journal of Vacuum Science and Technology, 1975, 12, 6, 1280. 39. L.F. Thompson, L.D. Yau and E.M. Doerries, Journal of the Electrochemical Society, 1979, 126, 10, 1703. 40. M. Gazard, A. Eranian, F. Barre and C. Duchesne, inventors; Thomson CSF, assignee; FR Patent 7,615,520, 1976. 41. H.S. Choong and F.J. Kahn, Journal of Vacuum Science and Technology, 1981, 19, 4, 1121. 42. J.M. Shaw. M. Hatzakis, J. Paraszczak. J. Liutkus and E. Babich. Proceedings of the Regional Technical Conference on Photopolymers: Principles, Processes and Materials, Mid-Hudson Section, SPE, Ellenville, NY, USA, 1982, 285. 43. M. Hatzakis, J. Paraszczak and J. Shaw, Preprints from Microcircuit Engineering, Lausanne, Switzerland, 1998, 1, 386. 44. Photoinitiation, Photopolymerisation and Photocuring, Fundamentals and Applications, Ed., J.P. Fouassier, Hanser, Munich, Germany, 1995. 45. M. Monta, A. Tanaka, S. Imamura, T. Tamamura and O. Kogure, Japanese Journal of Applied Physics, 1983, 22, 10, L659. 46. A. Tawaka, M. Monta and O. Kogure, Proceedings of the 1st International Polymer Symposium of the Polymer Society of Japan, 1984, 63. 47. G.N. Taylor, M.Y. Heilman, M.D. Feather and W.E. Willenbrock, Proceedings of the Regional Technical Conference on Photopolymers: Principles, Processes and Materials, Mid-Hudson Section, SPE, Ellenville, NY, USA, 1982, 355. 48. A. Eranian, F. Bernard and J.C. Dubois, Die Makromolekulare ChemieMacromolecular Symposia, 1989, 24, 41. 49. R.D. Allen, G.M. Wallraff, W.D. Hinsberg and L.L. Simpson, Journal of Vacuum Science and Technology, 1991, B9, 3357. 50. Y. Kaimoto, K. Nozaki, S. Takechi, and N. Abe, Proceedings of the SPIE, 1992, 66, 1672. 51. K. Nozaki, Y. Kaimoto, M. Takahashi, S. Takechi and N. Abe, Chemistry of Materials, 1994, 6, 1492.

215

Handbook of Polymers in Electronics 52. S. Takechi, M. Takahashi, A. Kotachi, K. Nozaki, E. Yano and I. Hanyu, Journal of Photopolymer Science and Technology, 1996, 9, 475. 53. K. Nozaki, K. Watanabe,E.Yano, A. Kotachi, S. Takechi and I. Hanyu, Journal of Photopolymer Science and Technology, 1996, 9, 1996, 509. 54. K. Yamashita, M. Endo, M. Sasago, N. Nomura, H. Nagano, S. Mizuguchi, T. Ono and T. Sato, Journal of Vacuum Science and Technology B, 1993, 11, 2692. 55. K. Patterson, U. Okoroanyanwu, T. Shimokawa, S. Cho, J. Byers and C.G. Wilson, Proceedings of the SPIE, 1998, 3333, 425. 56. W. Hinsberg, Materials Research Society Fall Meeting Tutorial Symposium, 28 November 1999. 57. J.C. Dubois and J.M. Bureau, Proceedings of the 2nd European Technical Symposium on Polyimides (STEPI), Eds., J.M. Abadie and B. Sillion, Elsevier, 1991, 464. 58. J.C. Dubois and G. Rabilloud, Techniques de l’Ingénieur, 1995, E1855. 59. H. Ahne, R. Leuschner and R. Rubner, Polymers for Advanced Technologies, 1993, 4, 217. 60. N. Yoda and H.J. Hiramoto, Journal of Macromolecular Science A, 1984, 21, 1641. 61. H. Mochizuki, T. Omote, K. Koseki and T. Yamacka, Polymer Preprints, Japan, 1989, 38, 3, 788.

216

7

Polymer Batteries for Electronics B. Scrosati

7.1 Introduction Conductive polymers are a class of materials that play a key role in the modern technology for energy storage and conversion. This prominent position results from the latest developments in the field that have produced a series of new polymers with exceptional electric properties. Depending on the nature of the electric carriers, these polymers can be broadly divided into two classes, namely the electronically and the ionically conducting polymers. Both classes of materials are of interest in the energy area. The electronic polymers may find use as electrodes in advanced-design devices, such as high-energy batteries and super capacitors. The ionic polymers are widely studied as electrolytes for the development of high performance fuel cells and of high energy density batteries, with particular focus on lithium batteries. Indeed, novel-design, high energy density batteries are in increasing demand. The growing market for portable electronic products, and the stringent environmental necessity for zero emission vehicles, e.g., electric vehicles (EV), has motivated research on the development of electrochemical power sources characterised by high energy density, long cyclability, reliability and safety. A recent breakthrough has been the commercialisation of rechargeable lithium batteries, the so-called lithium-ion batteries [1-6] that are now produced at a rate of a million units per month for the consumer electronics market. Another important innovation in the field is the development of redox supercapacitors [7, 8]. These high power devices are of increasing importance in the EV technology designed as a support to the energy power sources, such as the batteries or the fuel cells, to assure well-balanced vehicle operation. In this combined action, the supercapacitor provides the peak needs for the vehicle acceleration while the battery assures its longrange running. Most of the lithium-ion battery and supercapacitor research and development projects are focused on the fabrication of prototypes using liquid electrolytes. An important step forward in this technology is the replacement of the liquid electrolyte with an ionic membrane and, eventually, of the common inorganic-type electrode materials with advanced electronically conducting polymers, in order to produce novel devices having a full polymeric configuration. This is an interesting concept since it provides the prospect of a favourable combination of

217

Handbook of Polymers in Electronics the high energy and long life typical of lithium or lithium-ion cells with the reliability and easy manufacturing typical of all-polymer structures. The practical exploitation of this concept requires the availability of polymer membranes having ionic conductivity approaching those of the conventional liquid solutions and of polymer films having charge storage capabilities approaching those of the conventional electrode materials. This chapter attempts to describe the various classes of polymers presently under investigation to meet these requirements. The ionically conducting polymer membranes, which are designed as electrolyte separators, are first discussed. After this, the properties and the application potential of the electronically conducting membranes, which are designed as innovative electrodes, are reviewed.

7.2 Ionically Conducting Polymers These polymers have been studied and developed in view of their application as electrolyte separators in energy conversion and storage devices, such as batteries and fuel cells. Accordingly, these materials have been generally termed ‘polymer electrolytes’. With respect to batteries, the development of polymer electrolytes has been mainly confined to lithium ion conducting membranes, i.e., membranes compatible with the electrochemical reactions driving high-energy lithium batteries. Indeed, these are the batteries in continuous commercial evolution as ideal power sources for the consumer electronics market. Modern fuel cell technology has relied on polymer electrolytes with proton transport, i.e., on membranes capable of assuring simple yet efficient structures to power sources mainly directed to EV applications. Some recent advances in the R&D of these two classes of polymer electrolytes will be discussed in the following sections.

7.2.1 Lithium Polymer Electrolytes and Lithium Batteries Historically, the first types of lithium polymer electrolytes to be considered for battery applications were those formed by blending high molecular weight poly(ethylene oxide) (PEO), with a lithium salt LiX, where X is preferably a large anion [9-12]. Various books and extensive review articles have been published on these polymer electrolytes [5, 1316] and thus only their essential properties will be briefly recalled. The PEO-LiX membranes are typically prepared by casting an acetonitrile solution of the two components or by directly hot-pressing their intimate mixture. The polar oxygen atoms in the sequential oxyethylene groups of the PEO chains coordinate the Li+ cations, separating them from their X- counteranions. Accordingly, the structure of the PEO-LiX complexes may be broadly discussed as a sequence of polymer chains coiled around the lithium ions while the anions are more loosely coordinated [14]. A pictorial model of this structural sequence is shown in Figure 7.1.

218

Polymer Batteries for Electronics

Figure 7.1 Schematic structural arrangements of the PEO chains coiling around Li+ cations

This oversimplified picture progressively diverges from reality depending upon the relative concentration of the two components. As the concentration of the LiX, expressed as the ratio between the oxygen atoms in PEO and of the lithium ions in LiX, increases, the overall structure becomes more and more complicated due to ion-ion and association phenomena [17]. As for all conductors, the conductivity of the PEO-LiX polymer electrolytes depends on the number of the ionic carriers and on their mobility. The number of the Li+ carriers increases as the LiX concentration increases, but their mobility is greatly depressed by the progressive occurrence of ion-ion interaction which may even lead to large ion cluster formation [14]. Due to their particular structural position (Figure 7.1), the Li+ ions can be released to transport the current only upon unfolding of the coordinating PEO chains. Thus, high conductivity and fast Li+ ion transport are restricted to the amorphous state of the PEO component, which on average occurs at temperatures above 70 °C. This is shown by Figure 7.2, which illustrates the phase diagram and the ionic conductivity Arrhenius plot of a typical electrolyte in the PEO-LiClO4 system. It may be clearly noticed that at ambient temperature, when the system is in its crystalline state, the conductivity is very low, i.e., in the 10-8-10-7 S cm-1 range. However, at around 70 °C, i.e., at the crystalline to amorphous PEO transition, the conductivity increases by

219

Handbook of Polymers in Electronics

Figure 7.2 Phase diagram and conductivity Arrhenius plot of the PEO-LiClO4 polymer electrolyte. x is percent fraction of LiClO4.

several orders of magnitude to reach values of practical interest (i.e., around 10-3 S cm-1) at 100 °C. This implies that the use of the PEO-LiX electrolytes is restricted to batteries for which a relatively high temperature of operation does not represent a major problem, e.g., batteries designed for EV traction. Indeed, various R&D projects aimed at the production of PEO-based, EV lithium batteries are in progress worldwide [18-20]. These batteries typically use a lithium metal anode and a Li-intercalation cathode. The latter may be basically described as a compound with an ‘open structure’, i.e., a layered (e.g., TiS2, V2O5, LiCoO2,) or a tunnel (e.g., V6O13, LiMn2O4) structure, which provides channels for the reversible insertion-deinsertion of lithium ions [21-23]. It is possible to name this intercalation cathode by the general AyBz notation. A schematic diagram of an electrochemical process of the battery reported in the literature [2, 20] is given in Figure 7.3. Upon discharge, the Li+ ions, produced at the Li metal anode, travel across the electrolyte to reach the AyBz cathode and are inserted into its structure, while the electrons travel through the external circuit to reach the cathode and modify its electronic density of states [23]. The charging process is the exact opposite and thus the overall process may be written as shown in Equation 7.1. xLi + AyBz ⇔ LixAyBz

220

(7.1)

Polymer Batteries for Electronics

Figure 7.3 Schematic diagram of the Li-AyBz lithium battery (Reproduced with permission from B. Scrosati, La Chimica e l’Industria, 1997, 5, 464, published by Societa Chimica Italiana)

The open circuit voltage of this battery is associated with the difference in the Fermi levels of the two electrodes. Accordingly, if the overall process does not induce phase changes in the host AyBz cathode, the discharge voltage decreases upon increasing lithium intercalation level (x in Equation 7.1). A typical example, where AyBz is TiS2, is given in Figure 7.4. Some selected examples of R&D projects presently in progress for EV lithium polymer batteries [24-26] are listed in Table 7.1. Although these projects are very relevant, the high temperature operation of the batteries obviously limits their practical output. Accordingly, many studies have been carried out with the goal of improving the conductivity of the PEO-based polymer electrolytes at ambient temperature. Various approaches have been considered, e.g., the use of modified PEO polymer architectures to achieve low crystallinity at room temperature. These include block copolymers, crosslinked polymer networks and comb-shaped polymers having short oligooxyethylene chains attached to the polymer

221

Handbook of Polymers in Electronics

Figure 7.4 The Li/TiS2 cell. Open circuit voltage (a) and voltage discharge profile (b) upon development of the xLi + TiS2 ⇒ LixTiS2 discharge process. Derived from [23]. DOS = density of states, Eg = band gap, EF = Fermi level.

backbone [14, 27-29]. Other approaches have considered the addition of plasticisers, such as organic liquids, e.g., propylene carbonate (PC) or ethylene carbonate (EC) [30, 31] or low molecular weight ethylene glycols [32]. However, these modified PEO electrolytes (in particular the ones with added plasticiser), although reaching high levels of conductivity,

222

Polymer Batteries for Electronics

Table 7.1 Some examples of R&D projects for the lithium polymer batteries for EV application Project

System

Status

Support

3M-Hydroquebec-Argonne (North America)

Li/PEO-LiX/VOx

EV module prototypes

USABC

Bolloré-EDF-CEA/CEREM (France)

Li/PEO-LiX/VOx

Prototypes

Li/PEO-LiXcomposite/LiMn2O4

Demonstration prototypes

ENEA-ARCOTRONICS Universities (Italy)

Italian Government

USABC: United States Battery Consortium

suffer from serious drawbacks such as the loss of mechanical stability and, particularly, reactivity toward the lithium metal electrode [20, 33, 34] that may rule them out from practical applications. Indeed, to our knowledge, no practical lithium battery using a plasticiser-added polymer electrolyte has so far reached large-scale production. Therefore, the ideal solution in this field would be the use of ‘solid plasticisers’, namely of solid additives which would promote amorphicity at ambient temperature without affecting the mechanical and the interfacial properties of the electrolyte. A result that approaches this ideal condition has been obtained by dispersing selected ceramic powders, such as TiO2, Al2O3 and SiO2, at the nanoscale particle size, in the PEO-LiX matrix [3541]. The conductivity behaviour of a selected example of these ‘nanocomposite’ polymer electrolytes is shown in Figure 7.5. It may be seen that the related Arrhenius plot does not break around 70 °C, as typically expected for PEO-based electrolyes. It clearly suggests that the added ceramics prevent PEO crystallisation, with a resulting enhancement in conductivity which at room temperature may reach values of the order of 10-5 S cm-1, compared to the 10-7 S cm1 of the standard, ceramic-free electrolytes. This outstanding effect is explained by assuming that, once the composite electrolyte is annealed at temperatures higher than the PEO crystalline to amorphous transition (i.e., above 70 °C), the ceramic additive, due to its large surface area, prevents local PEO chain reorganisation with the result of freezing a high degree of disorder which favours the Li+ ion transport [38, 42, 43]. It has been proposed [44] that the Lewis acid character of the added ceramic may compete with that of the lithium cations for the formation of complexes with the PEO chains. Thus, the ceramics may act as crosslinking centres for the PEO segments, lowering the polymer chain reorganisation tendency and promoting

223

Handbook of Polymers in Electronics

Figure 7.5 Conductivity Arrhenius plots of ceramic-free and composite PEO-based, polymer electrolytes. w/o = weight percentage.

Li+ conducting pathways at the ceramic surface [44-46]. Therefore, according to this model, the structural modifications at microscopic levels promote consistent enhancement in the transport properties of the electrolyte. In addition, the all-solid configuration (no addition of liquids) gives to these nanocomposite electrolytes a high compatibility with the lithium metal electrode [47-50], all these properties making them suitable for use as safe and efficient separators in rechargeable lithium batteries [51]. The fabrication and test of prototypes based on the preferred composite electrolyte (a PEO-LiCF3SO3 system with dispersed γ-LiAlO2 ceramic powders) has been demonstrated (Table 7.1). The cathode was LiMn2O4, i.e., the lithium manganese spinel operating in its medium (3V vs. Li) voltage range [52, 53]. The voltage profile of this battery is shown in Figure 7.6. Long cycle life and an energy density of the order of 110 Wh kg-1 are expected for this battery [26]. Peled and co-workers [54-56] have reported another interesting battery application of the PEO-based composite electrolytes in a cell of the following structure:

Li/(PEO)20LiI-EC-Al2O3/FeS2

224

Polymer Batteries for Electronics

Figure 7.6 Voltage profile of a typical charge-discharge cycle of a Li /PEO-LiCF3SO3 + γLiAlO2 / LiMn2O4 polymer battery at 94 °C and at C/10 rate. Derived from [26]. ΔV is the charge (upper curve) and discharge (lower curve) voltage.

The cell operates at 135 °C on the basis of the overall process shown in Equation 7.2. 4 Li + FeS2 ⇔ Fe + 2Li2S

(7.2)

The overall process is associated with a theoretical energy density of 800 Wh kg-1 and a voltage slope between 2.5 V and 1.8 V. The need to extend the use of polymer electrolytes to devices capable of operating at ambient and sub-ambient temperatures has motivated the search for materials capable of offering values of conductivity even higher than those of the PEO-LiX composites. The most successful result has been obtained with the development of electrolyte membranes formed by trapping liquid solutions (e.g., solutions of a lithium salt in organic solvent mixtures) in a polymer, e.g., polyacrylonitrile (PAN), PMMA or polyvinylidene fluoride (PVdF) matrices [20, 5759]. The immobilisation procedure varies from case to case and includes UV crosslinking, casting and gelification, the latter being the most commonly adopted. The gelification procedure may involve a sequence of steps including: (i) the dissolution of the lithium salt in the given organic solvent mixture, (ii) the addition of the polymer component and its dispersion in the solution, (iii) the short-time heating of the slurry at around 90-100 °C for promoting complete homogenisation, and (iv) the cooling of the resulting solution to room temperature to promote gelification [60]. These gel-type membranes will be hereafter simply indicated by listing their components in sequence. For instance, the gel electrolyte formed by immobilising an ethylene carbonate-dimethyl carbonate lithium hexafluorophosfate solution in PAN, will be here indicated as LiPF6-EC-DMC-PAN. Strictly speaking, the gel membranes cannot be classified as ‘true’ polymer electrolytes, but rather as hybrid systems where a liquid phase is contained within a polymer matrix. A schematic view of this structure is represented in Figure 7.7.

225

Handbook of Polymers in Electronics

Figure 7.7 Schematic representation of gel-type polymer electrolyte configuration

226

Polymer Batteries for Electronics However, there has been some discussion as to whether these gel electrolytes are indeed simple two-phase materials where the polymer is a passive component acting as a rigid framework for regions of liquid solutions or whether they are integrated systems where the polymer provides the stability of the gel network down to the immediate vicinity of the Li+ ions [61]. Recent papers reporting NMR [62, 63] and Raman [64-66] spectroscopy studies show that some interactions do occur between the polymer backbone and the electrolyte solution, to an extent that depends upon the nature of the two components. For instance, in the case of electrolytes using a PAN matrix, clear evidence of coordination between the polymer backbone and both the solvated ions and the solvent molecules, has been shown, whereas in those using the PMMA matrix the interaction is weak [65]. Thus, while the latter can be regarded as hybrid systems where the liquid solutions are imbedded in a ‘passive’ host polymer, the former are systems characterised by an ‘active’ polymer experiencing strong interactions with the solutions. These structural differences may be of importance when a selection has to be made in view of battery application. In general, the PAN-based membranes, having characteristics which approach those of a ‘true’ polymer electrolyte entity, are expected to be more stable than the PMMA-based counterparts and thus to offer a higher reliability as practical battery separators. Feullade and Perche [67] originally introduced gel-type electrolytes. However, interest in these materials has increased recently with the development and the characterisation of many new types of membranes having different properties [59, 68-74]. Some examples of these membranes with their composition and their electrochemical properties are listed in Table 7.2 [72]. Clearly, these gel-type electrolytes have quite promising properties in terms of conductivity, approaching that of liquid solutions. This can be seen in Figure 7.8, which shows the Arrhenius plots of some selected examples, and Figure 7.5 which compares the conductivity of gels with that of PEO-based membranes. However, a high ionic conductivity, while obviously an important feature, is not sufficient to make a given electrolyte suitable for battery applications. Additionally, the chemical and electrochemical stability are key parameters. The time evolution of the conductivity of a LiPF6-EC-PC-PAN electrolyte obtained at 25 °C, the storage temperature, is shown in Figure 7.9 [74]. The trend shows that the conductivity is quite high, approaching 10-2 S cm-1 and remains stable for many days, indicating that the membrane does not degrade by crystallisation or phase separation phenomena either upon thermal or time excursions. In general, the electrochemical stability of an electrolyte is typically established by determining its breakdown voltage. This test can be carried out by running sweep voltametry on cells using the given membrane as electrolyte, a ‘blocking’ (i.e., not reversible to the membrane’s mobile ion) working electrode and a third reference electrode. Under

227

Handbook of Polymers in Electronics

Table 7.2 Composition and electrochemical properties at 25 °C of some selected examples of gel-type polymer electrolytes (average thickness = 100 mm) Electrolyte membrane

Molar composition

LiClO4-EC-PC-PAN

8.0-38.0-33.0-21.0*

1.1 x 10-3

5.0

4.5-56.5-23.0-16.0*

-3

4.9

-3

LiClO4-EC-PC-PAN

Conductivity Anodic (S cm-1) breakdown voltage vs. Li+/Lio (V) 1.1 x 10

LiClO4-EC-DMC -PAN

4.5-56.5-23.0-16.0*

3.9 x 10

5.1

LiClO4-EC-DEC-PAN

4.5-53.5-19.0-23.0*

4.0 x 10-3

4.8

4.5-79.5-16.0*

2.8 x 10-3

5.0

-3

4.3

-3

LiClO4-γBL-PAN LiAsF6-EC-PC-PAN

4.5-56.5-23.0-16.0*

0.9 x 10

LiAsF6-γBL-PAN

4.5-79.5-16.0*

4.1 x 10

4.6

LiPF6-γBL-PAN

4.5-79.5-16.0*

4.4 x 10-3

5.1

4.5-56.5-23.0-16.0*

5.5 x 10-3

4.6

4.0-20.0-62.0-14.0*

-3

4.4

-3

LiPF6-EC-γBL-PAN LiPF6-EC-DMC-PAN

4.2 x 10

LiN(SO2CF3)2-EC-PC-PAN

4.5-56.5-23.0-16.0*

1.0 x 10

4.6

LiN(SO2CF3)2-EC-γBL-PAN

4.5-56.5-23.0-16.0*

2.6 x 10-3

4.7

LiClO4-EC-PC-PMMA

4.5-46.5-19.0-30.0*

0.7 x 10-3

4.6

4.5-46.5-19.0-30.0*

-3

4.8

-3

4.9

LiAsF6-EC-PC-PMMA

0.8 x 10

LiN(SO2CF3)2-EC-PC-PMMA

4.5-46.5-19.0-30.0*

0.7 x 10

LiN(SO2CF3)2-EC-DMC-PMMA

5.0-50.0-20.0-25.0*

1.1 x 10-3

LiC(CF3SO2)3-EC-DBP-PVdF LiC(CF3SO2)3-EC-DBP-PVdF(C3F6) LiC(CF3SO2)3-EC-PC-PVdF(CTFE)

3.5-36.5-30.0-30.0*

-

-3

0.017 x 10

3.5-36.5-30.0-30.0*

0.035 x 10

4.8

1.2-42.0-16.8-40.0*

-3

4.6

(* = this value is referred to monomer) DEC = diethylene carbonate γ BL = γ -butyrolactone CTFE = chlorotrifluoroethylene DBP = dibutylphthalate

228

4.8 -3

0.1 x 10

Polymer Batteries for Electronics

Figure 7.8 Conductivity Arrhenius plots of various gel-type electrolytes. The plot of a typical liquid solution is also reported for comparison purposes.

Figure 7.9 Time evolution of the conductivity of the LiPF6-EC-PC-PAN electrolyte at 25 °C

229

Handbook of Polymers in Electronics

Figure 7.10 Sweep voltammetry of a Ni electrode in a LiPF6-EC-PC-PAN electrolyte cell at 25 °C. Li counter electrode. Scan rate: 0.5 V s-1.

these conditions, the voltage at which the current starts to flow through the cell may be assumed as the decomposition of the electrolyte. In the case of the gel-type membranes under discussion here, the test has been run by using a Ni blocking electrode and a Li reference electrode. Results obtained for the LiPF6-EC-PC-PAN electrolyte are shown in Figure 7.10. The current onset is detected around 4.3 V vs. Li, this being high enough to allow the safe use of the electrolyte membrane with a large selection of electrode couples. One may then conclude that, the gel-type electrolytes, and the PAN-based ones in particular, have electrochemical properties that in principle make them suitable for application in versatile, high-energy lithium batteries. In practice, their use may be limited by the reactivity towards the lithium electrodes induced by the high content of the liquid component. Indeed, severe passivation phenomenon occurs when the lithium metal electrode is kept in contact with the gel electrolytes [60, 69]. This confirms the general rule that if from one side the wet-like configuration is essential to confer high conductivity to a given polymer electrolyte, from the other it unavoidably affects its interfacial stability with the lithium metal electrode. On the other hand, the high conductivity of the gel electrolytes may be exploited in an effective way by directing them to the development of new-design, plastic-like batteries where the lithium metal anode is replaced by a lithium-accepting compound, such as a carbon or graphite [75]. These are the so-called ‘rocking chair’ or, more commonly ‘lithium-ion’ batteries [76]. Basically, these batteries operate on the cyclic transport of lithium ions from one lithium-

230

Polymer Batteries for Electronics rich cathode (e.g., LiCoO2 or, alternatively, LiNiyCo1-yO2 or LiMn2O4) to a lithium-poor anode [1, 6, 20] (e.g., C), according to the general overall process shown in Equation 7.3 6C + LiCoO2 ⇔ LixC6 + Li1-xCoO2

(7.3)

These lithium-ion batteries are quite effective and are presently produced at a rate of several millions of units per months, prevalently by Japanese manufacturing companies [20] and directed to the consumer electronics market where they have now assumed a prominent position [2]. These commercial batteries are commonly prepared in the so-called ‘bobbin-type’ configuration, by layering in sequence a thin layer of the carbon anode (backed on a metal, usually copper) foil collector, a microporous separator felt and the composite (blend of active material, conductive carbon and binder) cathode (backed on a metal, usually aluminum) foil. In a typical cylindrical design (Figure 7.11), the three contacted layers are coiled one on top of the other to obtain a spirally wound geometry having a high surface area. This assembly is then housed in a suitable container which, after tapping with the liquid electrolyte, is hermetically sealed to prevent either leakage or

Figure 7.11 Schematic representation of a cylindrical lithium-ion battery (Reproduced with permission from B. Scrosati, La Chimica e l’Industria, 1997, 5, 465, published by Societa Chimica Italiana)

231

Handbook of Polymers in Electronics contact with the external environment. This procedure is commonly used to produce a series of different geometries that includes prismatic as well as cylindrical cases. The most common, commercially available lithium-ion batteries use a liquid electrolyte, generally formed by a solution of a lithium salt (e.g., LiPF6) in an organic solvent mixture (e.g., EC-DMC). The next important step in this technology is the replacement of the fibre separator embedded with the liquid electrolyte with a polymer membrane that can act both as the separator and the electrolyte. This is expected to be an important technological progress because it provides the prospect of a favourable combination of the high energy and long life typical of the lithium-ion concept, with the reliability and diversified design that the plastic configuration may offer. Accordingly, many attempts to reach this goal are presently underway. One of the requirements for a successful result is the availability of polymer electrolyte membranes having electrical and chemical properties comparable with those of the common liquid electrolytes. An approach in this direction is based on the use of an elasticised electrolyte membrane separator formed by a copolymer of vinylidene fluoride and hexafluoropropylene [77, 78]. This membrane is capable of absorbing large quantities of liquid electrolytes and this feature has been exploited for a battery fabrication process which initially involves the lamination of three battery components, i.e., the anode film, the membrane soaked with a low vapour pressure plasticiser and the cathode film. A second process in which the plasticiser is eliminated to precondition the separator into a highly porous, liquidphilic membrane follows this first step. Finally, activation is accomplished by spreading a suitable lithiumion liquid electrolyte solution (e.g., the LiPF6-EC-DMC solution) throughout the separator membrane and the electrode films [79]. Alternative routes to obtain lithium-ion plastic batteries have considered the use of PANbased gel-type polymer electrolytes as separators. These electrolyte membranes, although macroscopically solid, contain in their structure the active liquid electrolyte (Figure 7.7). Therefore, they have a configuration which in principle allows a single lamination process for the fabrication of the lithium-ion battery, i.e., a process that avoids intermediate liquid extraction-soaking activation steps. The feasibility of the gel electrolytes for lithium-ion batteries development has been tested by first examining their compatibility with appropriate electrode materials, i.e., the carbonaceous anode and the lithium metal oxide cathode. This has been carried out by examining the characteristics of the lithium intercalation-deintercalation processes in the electrode materials using cells based on the given polymer as the electrolyte and lithium metal as the counter electrode. As an example, the voltage response of a graphite electrode cycled in a LiClO4-ECDMC-PAN electrolyte cell is shown in Figure 7.12. The response follows Equation 7.4. 232

Polymer Batteries for Electronics

Figure 7.12 Typical voltage response of the Li intercalation-deintercalation process of a graphite electrode in a LiClO4-EC-DMC-PAN electrolyte cell. Temperature: 25 °C. Lithium counter electrode. Cycling rate: C/4.

xLi+ + 6C ⇔ LixC6

(7.4)

It has been seen that the response approaches that observed in liquid electrolyte cells, namely a voltage profile which, during the intercalation process, decreases along a series of distinguishable plateaus corresponding to the progressively occupied staging graphite phases [80, 81]. A similar trend is reproduced upon the reverse, lithium deintercalation process, although with an apparent loss in capacity. This is also expected on the basis of the results obtained in liquid electrolytes, which have demonstrated that the formal excess capacity during the initial cycles is due to the side reactions involving the decomposition of the electrolyte with the formation of a passivation layer on the graphite electrode surface [82, 83]. It may be noted that, in contrast with the case of the lithium metal electrode, the occurrence of this passivation layer, which is electronically insulating but lithium ion conducting, is essential for assuring the proper cyclability to the graphite electrode [1]. The layer, often called the solid electrolyte interface (SEI) [84] prevents the decomposition of the electrolyte, thus allowing the continuation of the electrochemical intercalation process down to very low voltage levels, i.e., around 50 mV vs. Li (Figure 7.12) and with a reversible capacity which cycles around 300 mAh g-1. These two features are important to maintain high battery voltage and a long cyclability, respectively, when the graphite anode is coupled with a lithium metal oxide cathode. As in the case of the graphite anode, the electrochemical response of the cathodes can also be evaluated by following the lithium intercalation-deintercalation processes, again

233

Handbook of Polymers in Electronics

Figure 7.13 Typical voltage response of the Li intercalation-deintercalation process of a LiCryMn(2-y)O4 electrode in a LiClO4-EC-DMC-PAN electrolyte cell. Temperature: 25 °C. Lithium counter electrode. Cycling rate: C/4.

using cells with a gel electrolyte and a lithium metal counter electrode. The results obtained in the case of a lithium manganese spinel cathode cycled in a LiClO4-EC-DMC-PAN electrolyte cell are reported in Figure 7.13, the data collected to promote and evaluate the process as shown in Equation 7.5. LiCryMn2-yO4 ⇔ Li1-xCryMn2-yO4 + xLi+ + xe

(7.5)

Instead of the stoichiometric LiMn2O4, a Cr-added spinel has been used in this test. The use of a metal-doped manganese spinel is common in the lithium-ion technology since it has been demonstrated that a partial substitution of Mn (III) with M (III) metals, such as Cr, may consistently stabilise the spinel structure, conferring to the cathode a good capacity retention upon cycling [53, 85-87]. The trend seen in Figure 7.13 demonstrates that the voltage profile and the cycling capacity (about 130 mAh g-1) match those expected for these cathode materials in liquid electrolyte cells. Once the compatibility of the gel-type electrolyte with both anode and cathode materials is ascertained, one can proceed with the combination of the two for the fabrication of polymer-based lithium-ion battery prototypes. A few examples of these prototypes have been reported at the laboratory level scale. One is provided by a battery of the type C/ LiClO4-EC-PC-PAN/LiCryMn2-yO4. This battery was fabricated and tested under a coin-type cell configuration [80, 81]. The overall process of this battery is shown in Equation 7.6.

234

Polymer Batteries for Electronics

Figure 7.14 Typical voltage profile of the charge-discharge cycles of the Li/LiClO4-EC-DMC-PAN/LiCryMn(2-y)O4 polymer battery at 25 °C. Cycling rate: C/10.

6C + LiCryMn2-yO4 ⇔ LixC6+ Li1-xCryMn2-yO4

(7.6)

A typical charge-discharge cycle is shown in Figure 7.14, and confirms the feasibility of the PAN-based gel electrolytes as separators in lithium-ion batteries by showing that the cell can indeed be cycled with a good capacity delivery. Another lithium-ion polymer cell recently tested as a laboratory-scale prototype [88] has the configuration KC8/LiClO4-EC-PC-PAN/LiMn2O4. In this case a potassium-graphite (KC8) electrode has been used as the carbonaceous anode material. Upon anodic polarisation this electrode irreversibly deintercalates potassium resulting in a graphite-like compound, which on subsequent cycles performs with fast kinetics of its lithium intercalation-deintercalation process [89, 90]. Accordingly, the first charging process of the battery may be written as shown in Equation 7.7 at the anode. KC8 ⇒ K+ + e- + 8C

(7.7)

The reaction at the cathode is shown in Equation 7.8. LiMn2O4 + e- + Li+ ⇒ Li2Mn2O4

(7.8)

The charge balance in the electrolyte is assured since the amount of Li+ intercalated in Li2MnO4 is compensated by the amount of K+ deintercalated from KC8. The discharge 235

Handbook of Polymers in Electronics process and all the other cycles involve the cycling of the Li+ ions between the two electrodes, according to the typical lithium-ion process shown in Equation 7.9. 8C + Li2Mn2O4 ⇔ 4/3LiC6 + Li2/3Mn2O4

(7.9)

In this system, the maximum achievable specific capacity is 372 mA g-1 referred to the anode. These processes have been confirmed by the experimental response of the cell [8] which showed a voltage profile developing around the 3V characteristic of the lithiumrich manganese spinel phase [91] and delivered about 80% of the theoretical capacity at 0.1 mA cm-2 cycling rate [88]. Another interesting application of the lithium-ion battery concept has been applied to the SnO2/LiNi0.8Co0.2O2 electrodic couple [92]. Convertible oxides, and tin oxide in particular, first proposed as alternative anode materials by the Japanese Fuji Photo Film Company [93, 94], are presently the object of considerable attention in the lithium-ion battery community [95-98]. When negatively polarised in a lithium cell, tin oxide first undergoes an irreversible reaction shown in Equation 7.10. SnO2 + 4Li ⇒ Sn + 2Li2O

(7.10)

This results in metallic Sn particles that remain finely dispersed in the Li2O matrix. This produces a favourable geometry and facilitates a subsequent reversible reaction as indicated by the lithium alloying-dealloying process shown in Equation 7.11. Sn + 4.4Li ⇔ Li4.4Sn

(7.11)

The lithium oxide, surrounded by the tin particles, creates a sufficient amount of free volume to accommodate the mechanical stresses experienced by the metal particles during the course of the process. This greatly improves the cyclability of the electrode [94]. The interest in this electrode lies in its high specific capacity, which reaches values approaching 700 mAh g-1, i.e., almost double that offered by the more conventional graphite electrode. This feature is somewhat contrasted by a large, initial irreversible capacity associated with reaction 7.10 and by a certain tendency to lose capacity upon cycling, although the latter issue can be largely controlled. A Ni-Co mixed compound can be used as the cathode. This material is presently considered as a valid alternative to the more common LiCoO2 cathode, both in terms of cost and environmental compatibility [103, 104]. The electrochemical process is similar in both cases, i.e., the reversible release and uptake of lithium shown in Equation 7.12. LiNi0.8Co0.2O2 ⇔ xLi+ + Li1-xNi0.8Co0.2O2

236

(7.12)

Polymer Batteries for Electronics Considering the reactions of the two electrodes, the overall process of the SnO2/ LiNi0.8Co0.2O2 cell may, under steady conditions, be written as shown in Equation 7.13. LiNi0.8Co0.2O2 + Sn ⇔ LixSn + Li1-xNi0.8Co0.2O2

(7.13)

This process has been exploited for running a laboratory prototype polymer lithium-ion cell based on the LiClO4-EC-DMC-PAN electrolyte. The voltage profile of a typical discharge-charge cycle of this cell is shown in Figure 7.15, along with anode and cathode voltage variations. It may be seen that the SnO2 anode cycles with a trend comparable to that usually obtained in more conventional liquid electrolyte cells [105], delivering a reversible capacity approaching 300 mAh g-1. This capacity level is maintained upon further cycling, although with a certain progressive fade, this again being experienced in liquid-electrolyte cells as well.

Figure 7.15 Typical voltage profile of a charge-discharge cycle of the SnO2 / LiClO4EC-DMC-PAN / LiNi0.8Co0.2O2 polymer battery at 25 °C. Cycling rate: 0.25 mA cm-2.

237

Handbook of Polymers in Electronics Peramunage and Abraham have recently reported an advanced lithium-ion polymer cell [106, 107]. In this case, a material of the Li [Li1/3Ti5/3]O4 family [108, 109], e.g., the Li4Ti5O12, intercalation compound, has been used as an anode. The lithium intercalationdeintercalation process in this compound is shown in Equation 7.14. Li4Ti5O12 + 3Li ⇔ Li7Ti5O12

(7.14)

This system evolves around 1.5 V versus Li and is accompanied by very little change in lattice dimension. Thus, as opposed to the lithium-metal alloy cases discussed above, the absence of structural deformation makes Li4Ti5O12 an almost ‘zero strain’ electrode material characterised by a very good cyclability and by very little capacity fade upon cycling. This important feature has been experimentally confirmed [107] by determining the response of a lithium-ion polymer prototype cell such as Li4Ti5O12/LiPF6 -EC-PC-PAN/LiMn2O4. Some charge-discharge cycles obtained for the cell are shown in Figure 7.16. An apparent drawback of this cell is its relatively low overall voltage, which stabilises around 2.5 V, which is about 1.5 V less than that of more conventional lithium-ion systems based on graphite anodes. The difference is in the voltage levels of the two anode materials, i.e., 1.5 V versus Li for Li4Ti5O12 versus the 0.050 V vs. Li for graphite.

Figure 7.16 Voltage profile of a charge-discharge cycles of the Li4Ti5O12/LiPF6-EC- PC- PAN/ LiMn2O4 polymer battery at 25 °C. Cycling current densities are shown in the figure. (From reference 107, reproduced by permission of the Electrochemical Society, Inc.)

238

Polymer Batteries for Electronics However, this may not be an issue in the near future considering that the trend in electronics circuitry, wherein the use of the batteries is directed, is that the voltage-powering request will be progressively reduced from the initial 4 V to 3 V and, in the future, to an even lower range. This may somewhat eliminate the need for high-voltage power sources and thus, the constraint of choosing low-voltage anodes. This in turn opens new avenues for electrode materials operating within the stability window of the electrolyte, such as the Li4Ti5O12 discussed above, with important advantages in terms of the cyclability and safety of the battery. The results so far described, although demonstrating the feasibility of the lithium-ion concept in polymer electrolyte batteries, are somewhat limited to cases involving laboratory cells. More recently, various battery manufacturer companies worldwide have announced their involvement in the large-scale, commercial production of lithium-ion polymer batteries [110]. However, the level of released information is very scarce and, it is not possible at this stage to evaluate the effective status of the development of these advanced types of batteries. One can only cite that the Japanese Matsushita Battery Company has announced the completion of the facility for the large-scale production of thin-film lithium-ion polymer batteries [111]. However, no information is available on the nature of the polymer electrolyte or on the type of electrodes selected for this development. Other Japanese companies, including Sony Energy Technology and Toshiba, have announced similar activity, with even less details of the type of electrode and electrolyte materials used for their products [110]. Shipments of lithium-ion polymer electrolyte batteries have also been reported by Ultralife Batteries in the USA [112]. According to the released information, these batteries use a graphite anode, a Li1+xMn2-xO4 cathode and a polymer electrolyte which has not yet been disclosed. Apparently, these commercial lithium-ion polymer batteries have characteristics, such as reduction in thickness and improvements in safety, which make them very appealing for the modern consumer electronics markets, particularly the new generation of cellular phones. This may lead to the conclusion that the evolution of the lithium-ion battery technology will be focused on polymer configurations with an output that is expected to soon experience a substantial share in the electronics market.

7.2.2 Proton Polymer Electrolytes The research into proton conducting polymer electrolytes has consistently increased in recent years due to the transport characteristics which make them promising for various electrochemical applications of interest for the electronics market, including sensors and, particularly, fuel cells [113]. Nevertheless, the proton conductivity of the known polymer systems still remains below the upper limit of proton conductivity in liquids. The major problems arise from the numerous additional requirements, other than proton conductivity, which must be met for any specific application.

239

Handbook of Polymers in Electronics Nafion is presently the material of choice, because it acts as a good, chemically inert and stable polymer electrolyte [114]. As is well known, the hydrophobic part of this polymer provides a relatively good mechanical stability even in the presence of water, while the hydrated hydrophilic domains provide high proton conductivity (i.e., higher than 10-3 S cm-1 at ambient temperature) at water contents exceeding 30% wt. In fact, due to the hydrophilic nature of the sulfonic groups attached to the polymer backbone, the conductivity strongly depends upon relative humidity [115]. Although widely used, Nafion is somewhat affected by some operational problems and thus, new, optimised proton conducting membranes would be highly welcome. Indeed, intense research efforts are presently directed to reach this goal. Accordingly, many studies have been performed on proton conducting electrolytes based on polar polymers having basic sites. These sites form compounds with strong acids, such as H2SO4 or H3PO4. Particularly, polybenzimidazole (PBI), polyvinylpyrrolidone (PVP) and polyacrylamide (PAAM) form complexes with inorganic acids [116, 117]. As is well known, the amide group has basicity comparable to that of water and thus it can be protonated at either the oxygen or the nitrogen atoms. Indeed, the high conductivity (around 10-2 S cm-1) observed at room temperature for the PAAM-H2SO4 electrolyte has been attributed to the protonation capability of the amide groups [118]. It may be noted that it is not only important to adjust the polymeric host to the requirements of a particular application as described above, but also to optimise the environment of the proton. In fact, adding protonated and unprotonated solvent species can increase proton transport. As suggested by Kreuer [119], these species result in the generation of protonic defects in the non polar gel environment. In the same way, the inclusion of basic nitrogen groups in branched polyethyleneimine-H2SO4 or imidazole sulfonated polyaromatic membranes is expected to produce an increase in conductivity. This effect could be attributed to the proton exchange between two amine groups. In general, the ionic transport in linear or crosslinked swollen polymers containing a low molecular weight polar or ion-chelating additive mainly occurs in the solvent phase [118, 120]. This concept has been applied to develop proton conducting polymeric gel or hydrogel membranes [121-123] which reach conductivity values around 10-3 S cm-1 at room temperature and are not destroyed or dissolved even at high humidity levels. The question is whether the procedure successfully used for the fabrication of gel-type, lithium conducting polymer electrolytes discussed in the previous paragraph may be successfully extended to other ion conducting systems. In this respect, one may consider that ionic transport in crosslinked swollen polymers containing a low molecular weight polar or ion-chelating additive may indeed occur in the solvent phase and thus, that the gel concept can be extended to the proton conducting systems. Effectively, proton conducting polymeric gel or hydrogel membranes with conductivity values around 10-3 S cm-1 have been developed in the past [122]. Recently, a new approach aimed at

240

Polymer Batteries for Electronics developing proton conducting gel membranes having transport properties not dramatically dependent on the humidity level has been reported [124]. These novel types of proton gel electrolyte membranes have been obtained by incorporating salicylic acid (SA) or benzoic acid in a highly plasticised PMMA matrix. The synthesis involved the mixing of solutions of organic acids in protophobic solvents, such as PC and EC, with low contents of protophilic solvents (such as dimethylformamide (DMF), methylformamide or formamide) into a highly plasticised PMMA matrix [124, 125]. Here the carboxylic groups are expected to be able to act as proton donors and to show low levels of hydration, while the ring itself is rather non polar. In addition, the melting point of these acid molecules is generally higher than the boiling point of water, which makes them interesting candidates for supporting high proton conductivity at room temperature. These membranes feature as robust electrolyte systems with a thermal stability that extends up to 70-90 °C [124]. It is also important to point out that the conductivity values of the membranes were found to be higher than those of the plain liquid solutions. This effect was in part attributed to a specific interaction between the polymer and the non polar fragment of the SA molecule, thus enhancing ionic transport. In addition, it was assumed that the presence of DMF induces a more polar environment inside the gel matrix due to their strongly polar amide groups [124].

Figure 7.17 Conductivity of proton membranes as function of the humidity content. (From reference 124, reproduced by permission of the Electrochemical Society, Inc.)

241

Handbook of Polymers in Electronics The effect of the environmental humidity on the conductivity of SA-based PMMA membranes compared with that of a typical Nafion membrane, the presently most used proton-conducting membrane, is shown in Figure 7.17. The high conductivity of the membranes confirms that their transport mechanism is much less influenced by the humidity level then that of Nafion. As for most of the other advanced systems, these new protonic gel-type membranes appear to be still under evaluation. Presently, Nafion and related membranes still dominate the fuel cell market. The limited length of this chapter and its focus on batteries does not allow inclusion of a description of the present status of the polymer electrolyte fuel cells. The interested reader is referred to the many books and review articles available on the subject [e.g., 113, 126, 127].

7.3 Electronically Conducting Polymers The discovery in the late 1970s that certain types of polymers, though intrinsically poor conductors, can acquire electronic conductivity approaching that of metals following chemical or electrochemical treatment has triggered intensive research interest in the field. The essential structural characteristic needed by polymers to attain this significant conductivity change is a conjugated π-system extending over a large number of monomeric units, a trait common to polyheterocycles, such as polypyrroles and polythiophenes, and to polyanilines. The processes that switch conjugated polymers from the insulating to the conducting state are redox reactions, whether chemically or electrochemically driven. They are called doping processes, p-doping or n-doping, in relation to the positive or the negative sign of the injected charge. Many extensive review articles have been published on the synthesis procedure of these conducting polymers, as well as on their electronic structure and its evolution upon the doping processes [129-131]. The reader is referred to these references for detailed information. The most important feature of conducting polymers is in the reversibility of the doping process, which involves the formation of charged complexes which include polycations or polyanions, and to maintain the electrical balance, the corresponding structural insertion of adversely charged ions from the redox medium. Thus, conducting polymers can be switched repeatedly between their doped and undoped states by electrochemical oxidation and reduction processes that may involve a relatively large amount of electronic and ionic charge. This makes conjugated polymers an interesting class of high capacity electrode materials where the charge injected or released is accompanied by a corresponding ion motion within their structure. Thus, conducting polymers may be regarded as ion-insertion electrodes that act during the charge-discharge (i.e., dopingundoping) process as mixed electronic-ionic conductors, somewhat similar to the insertion inorganic compounds discussed in Section 7.2.1. Consequently, much effort has been devoted to their use as cathodes in batteries, with particular attention to lithium batteries.

242

Polymer Batteries for Electronics

7.3.1 Lithium-Doped Conducting Polymer and Lithium-Polymer Batteries Although, in principle, conducting polymers can be used either as anodes (by exploiting their reduction or n-doping process) or as cathodes (by exploiting their oxidation or pdoping process), most battery applications have been confined to the latter. Considerable interest was devoted in the early 1980s to the use of polyheterocycles and polyanilines as cathodes in lithium batteries. The initial efforts were mostly concentrated in cells using a liquid electrolyte, i.e., a solution of a lithium salt in an organic solvent. More recently, attention has been focused on polyheterocycles. In the typical case of a cell based on a PPy cathode, a LiClO4-PC electrolyte and a lithium metal anode, the electrochemical process can be written as: (PPy)n + nyLiClO4 ⇔ [(PPy)y+ (ClO4-)y ]n + nyLi

(7.15)

The charge process, schematically drawn in Figure 7.18, involves the p-doping (oxidation) of the polymer with the formation of a polycation whose positive charge is counterbalanced by the ClO4- electrolyte dopant anion (depicted as A– in Figure 7.18) which diffuses into the polymer matrix. The PPy oxidation at the cathode is accompanied at the anodic electrode by the reduction of lithium ions, which deposit as lithium metal on the given substrate. In the discharge process the electroactive polymer cathode releases the ClO4- anions and the Li+ cations are stripped from the metal anode to restore the initial electrolyte concentration. The extent of the process is defined by the term y, generally called ‘doping level’ which, representing the percentage of moles of the dopant anion over the moles of pyrrole monomer units, is proportional to the charge involved and thus to the battery capacity. Considerable effort was paid in the early 1990s to the marketing of lithium-polymer batteries, especially of prototypes using PPy or polyaniline as the cathode [132-135]. However, although benefiting from some interesting features, such as a reasonably high voltage and a low cost, these batteries suffered from some major drawbacks, including a relatively low energy density, limited power density and, particularly, severe self-discharge [135-137]. The self-discharge issue has been addressed by considering fully solid-state configurations where the liquid electrolyte was replaced by a polymer electrolyte, such as the previously discussed PEO-LiX blends or the PAN- or PMMA-based gels. This important concept was first exploited using the PEO-based polymer electrolytes to fabricate thin-film Li/ PPy or Li/PT batteries [138, 139]. Significant results have also been obtained using the highly conducting, gel-type electrolyte membranes described in Section 7.2.1. Cells formed by laminating a lithium metal anode, a PMMA-based electrolyte membrane and a PPy cathode have been successfully assembled and tested [140, 141]. The most relevant features of these cells were a very high charge-discharge coulombic efficiency and a long cyclability.

243

Handbook of Polymers in Electronics

Figure 7.18 Schematic representation of the doping process of heterocyclic conducting polymers, e.g., polypyrrole

It is interesting to use the polymer cathodes in lithium-ion cell types. This concept has been exploited by assembling cells having the structure C/LiClO4-EC-PC-PMMA/PPy. Laboratory prototypes have been fabricated by preparing the electrodes in the form of thin films backed on metallic substrates and separating them by the polymer electrolyte membrane [142, 143]. By charging the cell, Li+ cations enter the graphite structure and ClO4- anions simultaneously inject into the PPy structure: n6C + (PPy)n + nyLiClO4 ⇔ [(PPy)y+ (ClO4-)y ]n + nLiyC6

(7.16)

where n is the number of pyrrole units corresponding to one positive doping charge and y is the lithium intercalation level in graphite. Since both electrodes experience intercalation from different electrolyte species, this particular cell has been named a ‘dual lithium ion’ battery [142]. The cell is characterised by good cycling performance and by a total energy density of 300 Wh kg-1. Other advantages of the dual battery are the low cost, the compatibility with the environment and high power capabilities. Drawbacks in respect to the standard C/LiMO2 polymer lithium-ion batteries are in the lower capacity and lower operational voltage [144].

244

Polymer Batteries for Electronics The results reported above indicate that a proper choice of the battery components enables the intrinsic potentialities of the polymer electrode and electrolyte materials to be exploited for the development of revolutionary electrochemical devices. Accordingly, a number of laboratories are currently seeking to enhance the electrochemical properties of conducting polymers by designing suitable materials, the final goal being to optimise their response in advanced, plastic-like batteries. Undoubtedly, this will be the type of batteries that will dominate the electronic market in the new millennium.

Acknowledgements I would like to thank my students and collaborators, Giovanni Battista Appetecchi, Franco Bonino, Fausto Croce, Simona D’Andrea, Alessandra D’Epifanio, Stefania Panero, Luigi Persi, Priscilla Reale, Paola Romagnoli, Fabio Ronci and Giuseppe Savo, for their important and dedicated research work, some results of which are reported in this chapter.

References 1.

S. Megahed and B. Scrosati, Journal of Power Sources, 1994, 51,79.

2.

S. Megahed and B. Scrosati, Interface, 1995, 4, 4 34.

3.

Lithium Batteries, Ed., G. Pistoia, Elsevier Science, London, UK, 1997.

4.

D. Linden, Handbook of Batteries, 2nd Edition, McGraw Hill, Inc, New York, USA, 1995.

5.

C.A. Vincent and B. Scrosati, Modern Batteries: An Introduction to Electrochemical Power Sources, 2nd Edition, Butterworth-Heinemann, London, UK, 1998.

6.

Lithium Ion Batteries, Eds., M. Wakihara and O. Yamamoto, Wiley-VCH, Wenheim, Germany, 1998.

7.

B.E. Conway, Journal of the Electrochemical Society, 1991, 138, 6, 1539.

8.

B.E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Technological Applications, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1999.

9.

D.E. Fenton, J.M. Parker and P.V. Wright, Polymer, 1973, 14, 11, 589.

245

Handbook of Polymers in Electronics 10. M.B. Armand, S.M. Chabagno and M. Duclot, Second International Meeting on Solid Electrolytes, St. Andrews, Extended Abstracts, 1978, Paper 6.5.1. 11. B. Scrosati, Materials Science and Engineering B, Solid State Materials for Advanced Technology, 1993, 12, 369. 12. M. Armand, Advanced Materials 1990, 2, 6/7, 278. 13. Polymer Electrolyte Reviews I & II, Eds., J.R. MacCallum and C.A. Vincent, Elsevier Science, London, UK, 1987 & 1989. 14. F.M. Gray, Polymer Electrolytes, Royal Society of Chemistry, Cambridge, UK, 1997. 15. R. Neat and B. Scrosati in Applications of Electroactive Polymers, Ed., B. Scrosati, Chapman & Hall, London, UK, 1993, 182. 16. B. Scrosati, Chimica e l’Industria, 1995, 77, 5, 285. 17. C.A Vincent, IEE Review, 1999, March, 65. 18. M. Salomon and B. Scrosati, Gazzetta Chimica Italiana, 1996, 126, 415. 19. Lithium Polymer Batteries, Eds., J. Broadhead and B. Scrosati, Proceedings of the ECS, Volume 96-17, The Electrochemical Society, Pennington, NJ, USA, 1997. 20. B. Scrosati, Gazzetta Chimica Italiana, 1997, 79, 463. 21. B. Scrosati in the Electrochemistry of Novel Materials-Frontiers of Electrochemistry, Eds., J. Lipkowski and P. N. Ross, VCH, Wenheim, Germany, 1994, 111. 22. W.R. McKinnon in Solid State Electrochemistry, Ed., P.G. Bruce, Cambridge University Press, Cambridge, UK, 1995, 163. 23. J.B. Goodenough in Progress in Solid State Chemistry, Volume 5, Ed., H. Resiss, Pergamon Press, Oxford, UK, 1971, 279. 24. C. Donnelly, L. Christensen and D. Kuller, Electric & Hybrid Vehicle Technology, SAE, Warrendale, PA, USA, 1996. 25. P. Baudry, S. Lascaud, H. Majastre and D. Bloch, Journal of Power Sources, 1997, 68, 2, 432.

246

Polymer Batteries for Electronics 26. M.C. Borghini, M. Mastragostino, A. Zanelli, G.B. Appetecchi, F. Croce, B. Scrosati, M. Conte, S. Passerini, P.P. Prosini, P.G. Bernini and S. Saguatti, Associatione Elettrotecnica Italiana, 1998, 85, 222. 27. A. Bouridah, F. Dalard, D. Deroo, H. Cheradame and J.F. LeNest, Solid State Ionics, 1985, 15, 6854. 28. K.M. Abraham and M. Alamgir, Chemistry of Materials, 1991, 3, 2, 339. 29. L. Marchese, M. Andrei, A. Roggero, S. Passerini, P. Prosperi and B. Scrosati, Electrochimica Acta, 1992, 37, 1559. 30. M.Z.A. Munshi and B.B. Owens, Solid State Ionics, 1998, 26, 41. 31. S. Chintapalli and R. Frech, Solid State Ionics, 1996, 86/88, 341. 32. G. Dautzenberg, F. Croce, S. Passerini and B. Scrosati, Chemistry of Materials, 1994, 6, 4, 538. 33. F. Croce and B. Scrosati, Journal of Power Sources, 1993, 43/45, 9. 34. D. Fauteaux, Solid State Ionics, 1985, 17, 133. 35. B. Scrosati and F. Croce, Polymers for Advanced Technology, 1993, 4, 2/3, 198. 36. F. Capuano, F. Croce and B. Scrosati, inventors; L’Energie e L’Ambiente (ENEA), assignee; US Patent 5,576,115, 1996. 37. B. Kumar and L.G. Scanlon, Journal of Power Sources, 1994, 52, 2, 261. 38. F. Croce, G.B. Appetecchi, L. Persi and B. Scrosati, Nature, 1998, 394, 6692, 456. 39. B. Kumar and L.G. Scanlon, Solid State Ionics, 1999, 124, 3-4, 239. 40. C. Capiglia, P. Mustarelli, E. Quartarone, C. Tomasi and A. Magistris, Solid State Ionics, 1999, 118, 1-2, 73. 41. A.S. Best, A. Ferry, D.R. MacFarlane and M. Forsyth, Solid State Ionics, 1999, 126, 3-4, 269. 42. M. Siekierski, J. Przyluski and W. Wieczorek, Electrochimica Acta, 1995 40, 1314, 2101. 43. W. Wieczorek, Z. Florjanczyk and J.R. Stevens, Electrochimica Acta, 1995, 40, 1314, 2251.

247

Handbook of Polymers in Electronics 44. F. Croce, R. Curvini, A. Martinelli, L. Persi, F. Ronci and B. Scrosati, The Journal of Physical Chemistry B, 1999, 103, 48, 10632. 45. F. Croce, L. Persi, F. Ronci and B. Scrosati, Solid State Ionics, 2000, 135, 1-4, 47. 46. G.B. Appetecchi, F. Croce, L. Persi, F. Ronci and B. Scrosati, Electrochimica Acta, 2000, 45, 8-9, 1481. 47. B. Scrosati, Journal of the Electrochemical Society, 1990, 136, 2774. 48. F. Capuano, F. Croce and B. Scrosati, Journal of the Electrochemical Society, 1991, 138, 7, 1918. 49. G.B. Appetecchi, F. Croce, M. Mastragostino, B. Scrosati, F. Soavi and F. Zanelli, Journal of the Electrochemical Society, 1998, 145, 4133. 50. M.C. Borghini, M. Mastragostino, S. Passerini and B. Scrosati, Journal of the Electrochemical Society, 1995, 142, 7, 2118. 51. G.B. Appetecchi, F. Croce, G. Dautzenberg, M. Mastragostino, F. Ronci, B. Scrosati, F. Soavi, F. Zanelli. F. Alessandrini and P.P. Prosini, Journal of the Electrochemical Society, 1998, 145, 4126. 52. M.M. Thackeray, P.J. Johnson, L.A. de Picciotto, P.G. Bruce and J.B. Goodenough, Materials Research Bulletin, 1984, 19, 179. 53. M.M Thackeray, Progress in Solid State Chemistry, 1997, 25, 1, 1. 54. E. Strauss, D. Golodnitsky, Y. Lavi, E. Peld, L. Burstein and Y. Lareah, 192nd Meeting of the Electrochemical Society, Paris, France, 1997, Abstract No. 137 55. S. Kostov, M. denBoer, E. Strauss, D. Golodnitsky, S.G. Greenbaum and E. Peled, Journal of Power Sources, 1999, 81/82, 1-2, 709. 56. G.B. Appetecchi, P. Romagnoli, B. Scrosati, G. Ardel, D. Golodnitsky and E. Peled, 196th Meeting of the Electrochemical Society, Honolulu, USA, 1999, Abstract No. 365 57. B. Scrosati, Polymer International, 1998, 47, 1, 50 58. K.M. Abraham in Applications of Electroactive Polymers, Ed., B. Scrosati, Chapman & Hall, London, 1993. 59. S. Slane and M. Salomon, Journal of Power Sources, 1995, 55, 1, 7.

248

Polymer Batteries for Electronics 60. G.B. Appetecchi, F. Croce, G. Dautzenberg, F. Gerace, S. Panero, F. Ronci, E. Spila and B. Scrosati, Gazzetta Chimica Italiana, 1996, 126, 405. 61. F. Croce, S. Panero, S. Passerini and B. Scrosati, Electrochimica Acta, 1994, 39, 225, 1. 62. P.E. Stallworth, S.G. Li, S.G. Greenbaum, F. Croce, S.Slane and M.Salomon, Solid State Ionics, 1994, 73, 119. 63. C.A. Edmondson, M.G. Wintersgill, J.L. Fontanella, F. Gerace, B. Scrosati and S.G. Greenbaum, Solid State Ionics, 1996, 85, 1-4, 173. 64. D. Ostrovskii, L.M. Torrell, G.B. Appetecchi and B. Scrosati, Solid State Ionics, 1998, 106, 1-2, 19. 65. D. Ostrovskii, A. Brodin, L.M. Torell, G.B. Appetecchi and B. Scrosati, Journal of Physical Chemistry, 1998, 109, 7618. 66. C. Svanberg, J. Adebahr, H. Ericson, L. Borjesson, L.M. Torell and B. Scrosati, Journal of Chemical Physics, in press. G. Feuillade and Ph. Perche, Journal of Applied Electrochemistry, 1975, 5, 63. 68. K.M. Abraham, H.S. Choe and D. Pasquariello, Electrochimica Acta, 1998, 43, 16/ 17, 2399. 69. H.S. Choe, B.G. Carroll, D.M. Pasquariello and K.M. Abraham, Chemistry of Materials, 1997, 9, 1, 369. 70. G.B. Appetecchi, F. Croce, G. Dautzenberg, F. Gerace, S. Panero, F. Ronci, E. Spila and B. Scrosati, Gazzetta Chimica Italiana, 1996, 126, 405. 71. G.B. Appetecchi, F. Croce and B. Scrosati, Journal of Power Sources, 1997, 66, 1-2, 77. 72. G.B. Appetecchi, F. Croce, F. Gerace, S. Panero, E. Spila and B. Scrosati, Gazzetta Chimica Italiana, 1997, 127, 325. 73. G.B. Appetecchi, F. Croce, A. de Paolis and B. Scrosati, Journal of Electroanalytical Chemistry, 1999, 463, 248. 74. G.B. Appetecchi, F. Croce, P. Romagnoli, B. Scrosati, U. Heider and R. Oesten, Electrochemistry Communications, 1999, 1, 2, 83.

249

Handbook of Polymers in Electronics 75. B. Scrosati in Lithium Ion Batteries, Eds., M. Wakihara and O. Yamamoto, Wiley-VCH, Wenheim, Germany, 1998, 218. 76. B. Scrosati, Journal of the Electrochemical Society, 1992, 139, 2776. 77. G.C. Amatucci, C.N. Schmutz, A. Blyr, C. Sigala, A.S. Gozdz, D. Larcher and J.M. Tarascon, Journal of Power Sources, 1997, 69, 1-2, 11. 78. A.S. Gozdz, C.N. Schmutz and J.M. Tarascon, inventors; Bell Communications Research, assignee; US Patent 5,296,318, 1994. 79. F. Shokoohi, P.C. Warren, S.J. Greaney, J.M. Tarascon, A.S. Gozdz, G.C. Amatucci, Proceedings of the 37th International Power Sources Symposium, Cherry Hill, New Jersey, USA, 1996, 243. 80. G.B. Appetecchi and B. Scrosati, Electrochimica Acta, 1998, 43, 10/11, 1105. 81. G.B. Appetecchi and B. Scrosati, Denki Kagacu, 1998, 66, 1299. 82. F.A.C. Chu, J.Y. Josefowicz and G.C. Farrington, Journal of the Electrochemical Society, 1997, 144, 12, 4161. 83. Z. Ogumi and M.Inaba, Bulletin of the Chemical Society of Japan, 1998,71, 3, 521. 84. E. Peled in Lithium Batteries, Ed., J.P. Gabano, Academic Press, New York, USA, 1983. 85. L. Guoha, H. Ikuta, T. Uchida and M. Wakihara, Journal of the Electrochemical Society, 1996, 143, 1, 178. 86. G. Pistoia, A. Antonini, R. Rosato, C. Bellitto and G.M. Ingo, Chemistry of Materials, 1997, 9, 6, 1443. J.M. Tarascon, E. Wang, F.K. Shokooki, W.R. Mc Kinnon, S. Colson, Journal of the Electrochemical Society, 1991, 138, 10, 2859. 88. S. Sconocchia, R. Tossici, R. Marassi, F. Croce and B. Scrosati, Electrochemical and Solid State Letters, 1998, 1, 4,159. 89. R. Tossici, M. Barrettoni, V. Nalimova, R. Marassi and B. Scrosati, Journal of the Electrochemical Society, 1996, 143, 3, L64. 90. R. Tossici, M. Barrettoni, M. Rosolen, R. Marassi and B. Scrosati, Journal of the Electrochemical Society, 1997, 144, 1, 186.

250

Polymer Batteries for Electronics 91. M.M Thackeray, W.I.F. David. P.G. Bruce and J.B. Goodenough, Materials Research Bulletin, 1982, 17, 785. 92. S. Panero, G. Savo and B. Scrosati, Electrochemical and Solid State Letters, 1999, 2, 8, 365. 93. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 1997, 276, 5317, 1395. 94. Y. Idota, M. Mishima, Y. Miyaki, T. Kubota and T. Miyasaka, inventors; Fuiji Photo Film Co., Ltd., assignee; US Patent 5,618,640, 1997. 95. R.A. Huggins, Solid State Ionics, 1998, 113/115, 57. 96. I.A. Courtney and J.R. Dahn, Journal of the Electrochemical Society, 1997, 144, 6, 2045. 97. T. Brousse, S.M. Lee, L. Pasquerau, D. Defives and D.M. Schleich, Solid State Ionics, 1998, 113/115, 51. 98. Handbook of Battery Materials, Ed., J.O. Besenhard, Wiley-VCH, Weinheim, Germany, 1999. 99. J.O. Besenhard, J. Yang and M. Winter, Journal of Power Sources, 1997, 68, 1, 87. 100. J. Yang, M. Winter and J.O. Besenhard, Solid State Ionics, 1996, 90, 1-4, 281. 101. Brousse, R. Retoux, U. Herterich and D.M. Schleich, Journal of the Electrochemical Society, 1998, 145, 1, 1. 102. N. Li, C.R. Martin and B. Scrosati, Electrochemical and Solid State Letters, in press. 103. C. Delmas and I. Saadounne, Solid State Ionics, 1992, 53/56, 370. 104. C. Delmas, I. Saadounne and A. Rougier, Journal of Power Sources, 1993, 44/ 45, 595. 105. W.R. MacKinnon and J.R. Dahn, Journal of the American Chemical Society, 1999, 59. 106. D. Peramunage and K.M. Abraham, Journal of the Electrochemical Society, 1998, 145, 8, 2609. 107. D. Peramunage and K.M. Abraham, Journal of the Electrochemical Society, 1998, 145, 8, 2615.

251

Handbook of Polymers in Electronics 108. T. Ohzuku, A. Ueda and N. Yamamoto, Journal of the Electrochemical Society, 1995, 142, 5, 1431. 109. T. Ohzuku, A. Ueda and M. Kouguchi, Journal of the Electrochemical Society, 1995, 142, 12, 4033. 110. T. Osaka, Interface, 1999, 8, 3, 9. 111. www.mbi.panasonic.co.jp 112. J. Simon Xue, R.D. Wise, X. Zhang, M.E. Manna, Y. Lu, G. Ducharme and E.A. Cuellar, Journal of Power Sources, 1999, 80, 1-2, 119. 113. K. Kordesch and G. Simader, Fuel Cells and Their Applications, VCH, Weinheim, Germany, 1996. 114. F.M. Gray, Solid Polymer Electrolytes: Fundamentals and Technological Applications, Wiley-VCH, Weinheim, Germany, 1991, Chapter 7, p.125. 115. S. Srinivasan, D.J. Manko, H. Koch, M.A. Enayetullah and J.A. Appleby, Journal of Power Sources, 1990, 29, 367. 116. J.S. Wainright, J.T. Wang, D. Weng, R.F. Salvinell and M. Litt, Journal of the Electrochemical Society, 1995, 142, 7, L-121. 117. M.F. Daniel, B. Desbat and J.C. Lassegues, Solid State Ionics, 1988, 28-30, 632. 118. E. Zygadlo-Monikowska, Z. Florjanczyk and W. Wieczorek, Journal of Macromolecular Science A, 1994, 31, 9, 1121. 119. K.D. Kreuer, Chemical Materials, 1996, 8, 610. 120. D.G.H. Ballard, P. Cheshire, T.S. Mann and J.E. Przeworski, Macromolecules, 1990, 23, 5, 1256. 121. J. Przyluski and W. Wieczorek, Synthetic Metals, 1991, 45, 323. 122. J.R. Stevens, W. Wieczorek, D. Raducha and K.R. Jeffrey, Solid State Ionics, 1997, 97, 1-4, 347. 123. G. Vaivars, J. Kleperis, A. Azens, C.G. Granqvist and A. Lusis, Solid State Ionics, 1997, 97, 1-4, 365. 124. A.M. Grillone, S. Panero, B.A. Retamal and B. Scrosati, Journal of the Electrochemical Society, 1999, 146,27.

252

Polymer Batteries for Electronics 125. S. Panero and B. Scrosati, Journal of Power Sources, 2000, 90, 13. 126. J. Przyluski, W. Wieczorek, Z. Poltarzewski, P. Staiti and N. Giordano, in Recent Advances in Fast Ion Conducting Materials and Devices, Eds., B.V.R. Chowdari, Q.G. Liu and L.Q. Chen, 1990, World Scientific Publishing, River Edge, NJ, USA, 1990, 307. 127. A.J. Appelby and F.R. Foulkes, Fuel Cell Handbook, Van Nostrand Reinhold, New York, NY, USA, 1989. 128. T.A. Skotheim, Handbook of Conducting Polymers, Volumes 1 and 2, Marcel Dekker, New York, NY, USA, 1986. 129. B. Scrosati in Solid State Electrochemistry, Ed., P.G. Bruce, Cambridge University Press, Cambridge, UK, 1996, 229. 130. M.G. Kanatzidis, Chemistry and Engineering News, 1990, 68, 49, 36. 131. C. Arbizzani, M. Mastragostino and B. Scrosati in Handbook of Organic Conductive Molecules and Polymers, Volume 4, Ed., H. S. Nalwa, John Wiley & Sons Ltd., New York, NY, USA, 1997, 595. 132. N. Furukawa and K. Nishio in Applications of Electroactive Polymers, Ed., B. Scrosati, Chapman & Hall, London, UK, 1993, 150. 133. K. Nishio, M. Fujimoto, N. Yoshinaga, O. Ando, H. Ono and T. Suzuchi, Journal of Power Sources, 1991, 34, 153. 134. H. Mustedt, G. Kohler, H. Mohwald, D. Naegele, R. Bittin, G. Ely and H. Meissner, Synthetic Metals, 1987, 18, 259. 135. M. Mastragostino, A.M. Marinangeli, A. Corradini and C. Arbizzani, Electrochimica Acta, 1987, 32, 1589. 136. S. Panero, P. Prosperi and B. Scrosati, Electrochimica Acta, 1987, 32, 1461. 137. P. Novak, O. Inganas and R. Bjorklund, Journal of Power Sources, 1987, 21, 17. 138. C. Arbizzani, M. Mastragostino, S. Panero, P. Prosperi and B. Scrosati, Synthetic Metals, 1988, 28, C663. 139. C. Arbizzani and M. Mastragostino, Electrochimica Acta, 1990, 35, 251. 140. S. Kakuda, T. Momma, T. Osaka, G.B. Appetecchi and B. Scrosati, Journal of the Electrochemical Society, 1995, 142, 1, L1.

253

Handbook of Polymers in Electronics 141. T. Osaka, T. Momma, H. Ito and B. Scrosati, Journal of Power Sources, 1997, 68, 2, 392. 142. S. Panero, E. Spila and B. Scrosati, Journal of the Electrochemical Society, 1996, 143, L29. 143. A. Clemente, S. Panero, E. Spila and B. Scrosati, Solid State Ionics, 1996, 85, 273. 144. G.B. Appetecchi, S. Panero, E. Spila and B. Scrosati, Journal of Applied Electrochemistry, 1998, 28, 12, 1299.

254

8

Polymer Microactuators K. Kaneto and M. Kaneko

8.1 Introduction Actuators that generate movements and forces, such as bending, expansion and contraction driven by stimulation of electrical, chemical, thermal and optical energies, are different from rotating machines such as electric motors and internal combustion engines. There are many sorts of soft actuators made of polymers [1-3], gels [4] and nanotubes [5]. Particularly, biomimetic actuators are interesting because of the application to artificial muscles that will be demanded for medical equipment, robotics and replacement of human muscle in the future. In natural muscles, nowadays, the mechanism of actuation has being revealed to some extent; however, the details are not known. In fact, the muscles are constructed in an ordered structure from molecular level to macroscopic level [6]. The driving force originates from the conformational change of peptide molecules by the chemical energy cycles. Similarly, various sorts of stimulating energies can change the conformation of molecules. For example, the cis-trans photoisomerisation in azobenzene, as shown in Figure 8.1a, is a well-known phenomenon [7]. The cis form converts to the longer trans form upon illumination by UV light, and the trans form reverts to the cis form upon illumination by visible light. The electrical conductivity in polyacetylene (Figure 8.1b), the representative of conducting polymers, dramatically increases from insulator to conductor upon chemical or electrochemical oxidation [8]. This results from the delocalisation of π-electrons and in the change of polymer conformation. At the pristine stage (or in the reduced state), the polyacetylene is flexible because of the single bond in the bond alternation. The single bond has more freedom in the rotational and bending modes than that of the double bond. In the oxidised state, the bond alternation is reduced and the molecular structure becomes more planar than that of the reduced state. Similarly, polyaniline, shown in Figure 8.1c and discussed mainly in this chapter, changes its bond alternation from the benzenoid form to quinoid form upon oxidation [9, 10]. This results in the deformation of polymer conformation. In fact, films, fibres or blocks of conducting polymer expand and contract upon electrochemical oxidation and reduction [1, 2, 11-16], respectively. This process is

255

Handbook of Polymers in Electronics

Figure 8.1 Conformational changes of molecular structure, (a) photoisomerisation of azobenzene, (b) and (c) extension and contraction of polyacetylene and polyaniline, respectively, upon oxidation and reduction

tentatively named ‘electrolytic deformation’ or ‘electrolytic expansion’ in this chapter. The mechanisms of electrolytic deformation have been classified by three principal mechanisms [17, 18]: (1) the insertion and removal of bulky ions, (2) conformational change of polymer structure due to the delocalisation of π-electrons, and (3) electrostatic repulsion between likely charged polycations (polarons and/or bipolarons). In mechanism (1), the insertion is induced by the neutralisation of polymers for oxidation and reduction. The magnitude of expansion or contraction depends on the volume of ions and cannot be larger than the total volume of the inserted ions. For mechanism (2), the expansion of the polymer is associated with an expanding spring and depends on the morphology of polymer structure. The electroexpansion of hydrogels [4] is explained by mechanism (3). To accomplish the larger expansion ratio in the electrolytic deformation of conducting polymers, it is desirable to utilise mechanisms (2) and (3). In this section, the behaviour of the electrolytic expansion in conducting polymers, especially polyaniline and poly(o-methoxyaniline) (PMAN) are described, with discussion of the basic redox reaction of polyaniline, the dependence of the expansion ratios on oxidation levels, the kind of anions, strain, the pH of the electrolyte and anisotropy.

256

Polymer Microactuators

8.2 Sample Preparation and Measurements of Electrolytic Deformation Conducting polymers are prepared by either chemical or electrochemical oxidation of monomers, like pyrrole, thiophene and aniline, following the methods described in the literature [9, 19]. For the measurement of electrolytic deformation, it may be preferable to use soluble polymers, such as polyaniline (soluble in N-methyl-2-pyrrolidinone (NMP)). Others are hardly soluble in usual organic solvents. However, polymers with substituted long alkyl chains have been found to be soluble in organic solvents. Usually, the conducting polymers prepared by the electrochemical methods are obtained as thin films. Such conducting polymers are, however, difficult to process. The polyaniline prepared by the chemical oxidation is a powder and the base form is named emeraldine base and is soluble in NMP. The emeraldine film can be prepared by casting the NMP solution containing its concentration of 2-10 wt.% on a glass plate. The cast film obtained by this method can be stretched mechanically to more than 3 times the original length. The films have been examined along the stretch direction, the perpendicular direction [14] and also the thickness direction [20] to investigate the anisotropic behaviour. For a measurement of electrolytic deformation, a bimorph actuator [11] has been fabricated, and the ratio of expansion is estimated from the bent curvature of the actuators. This method is effective for qualitative measurement and demonstration, since the expansion is tremendously magnified. Even with an expansion rate at the level of 1%, the bending is clearly observed. For the direct measurement of the displacement, a balance [12], an Instron testing machine [17, 21] and a special cell with a pinhole at the bottom [13, 14] were employed. In this last method, the change of the film length is picked up by a thread through the pinhole as shown in Figure 8.2. The results obtained using the cell with the pinhole will be discussed. From the simultaneous measurements of the redox current (cyclic voltammogram, CV) and the change of film length (Δl) by the application of linear voltage sweep cycles, the relationship between the degree of oxidation and the rate of expansion (Δl/l0, l0 is the original length) is obtained. By the application of a stepwise voltage, the diffusion constant of ions in the film is estimated from the time response of the electrolytic deformation. The diffusion constant is estimated [12] from the initial time dependence of injected charges after the voltage application: f = 4D 1/2 t1/2 π-1/2 d-1

(8.1)

where f is the injected charge normalised to the saturated value, d is the thickness of the film, D is the difussion coefficient and t is the time. The diffusion constant obtained from Equation 8.1 is based on the model that ionic species diffuse from the surface of the film. Equation 8.1 may also be applicable to the time response of the deformation.

257

Handbook of Polymers in Electronics

(a)

(b) Figure 8.2 Schematic diagrams for the measurement of electrolytic expansion along (a) the film length and (b) the thickness direction. WE, RE and CE are the working electrode, reference electrode and counter electrode, respectively.

258

Polymer Microactuators

8.3 Electrochemistry and Expansion Behaviour in Polyaniline Film Polyaniline in an aqueous acid solution takes three typical redox stages [9] depending on the degree of oxidation, as shown by the cyclic voltammogram curve in Figure 8.3. The half redox potential, E1/2 is defined as E1/2 = (Ea + Ec)/2, where Ea and Ec are anodic and cathodic potentials at the peak currents for oxidation and reduction, respectively.

Figure 8.3 Typical cyclic voltammogram (upper), redox behavior in chemical structures (middle) and expansion and contraction of polyaniline film along the stretched direction (lower). Ea and Ec are the anodic peak and cathode peak, respectively, E1/2, LS-ES = 1/2 (Ea + Ec).

259

Handbook of Polymers in Electronics The most electrically conductive state is the emeraldine salt (ES) that is between pernigraniline salt (Pas) at the high potential side and the leuco-emeraldine salt (LS) at the low potential side. The LS and the Pas are the most reduced and oxidised states, respectively, and have been found to be insulating. In an oxidation process from the LS to the ES, two electrons are withdrawn and two chloride ions are doped for every four benzene units. For the oxidation from the ES to the Pas, two electrons are withdrawn and two protons are released. The LS/ ES reaction is reversible, however, the hydrolysis occurs at the higher oxidised Pas states. The bottom of Figure 8.3 shows a typical expansion behaviour in a polyaniline film along the stretched direction. By the potential sweep from the LS to higher potentials, the film starts to expand and shows the maximum expansion at the ES. Then the film contracts slightly. When the potential was returned from the Pas state, the film expands slightly, and then contracts below the potential of E1/2, LS↔ES and returns to the original length. This extension and contraction behaviour is very similar to the weight change measured by a quartz crystal microbalance [22]. The result indicates that the electrolytic expansion closely relates to the insertion and exclusion of dopant ions in the film. It may be noted that this does not violate the mechanisms (2) and (3) described in Section 8.1. The unstretched film and the stretched film perpendicular to the stretched direction show monotonous expansion during the oxidation LS→ES→Pas and vice versa [14]. The result possibly indicates that the dopant ions settle between the polymer chains. In the expansion of the film perpendicular to the stretched direction, irreversible expansion, namely, a creeping effect, was observed even under light load during the redox cycles [14].

8.4 Dependencies of the Expansion Ratio on the Degree of Oxidation and Dopant Ions The dependence of the expansion ratio [13] on the level of reduction, y, for various kinds of electrolytes in polyaniline films at pH = 0 are shown in Figure 8.4. The level of reduction was determined from the amount of electrically injected electrons in the cyclic voltammogram process of test materials, and the definition is shown as the inset of Figure 8.4. Here, y = 0 is the basis of the ES state, and y = –0.5 and 0.5 are the LS and Pas states, respectively. The polyaniline film contracts 2%-3% by the reduction from the ES state to LS state at around y = –0.2. The expansion ratio of polyaniline films at pH = 0 strongly depends on the kind of negative ions as seen in Figure 8.4. The hysteresis of the curve originates from the non equilibrium condition of the system due to the slow rate of diffusion and also due to thermodynamics. Though the radius of benzenesulfonic acid (BSA) is extremely large there is no large contraction since the molecule is too large to penetrate into the film. The anion radius dependence of the contraction ratio and the diffusion coefficients in polyaniline film at pH = 0 are shown in Figure 8.5, showing that the larger the ion

260

Polymer Microactuators

Figure 8.4 Dependency of the expansion ratio on the level of reduction, y, for various kinds of electrolytes in polyaniline films at pH = 0. The definition of y is shown by the inset, and y = 0 is taken as the ES state. l0 is the length of film at y = 0. BSA is benzene sulfonic acid.

Figure 8.5 Anion radius dependence of the contraction ratio in polyaniline film and poly(o-methoxyaniline) at pH = 0.

261

Handbook of Polymers in Electronics the more it expands and the smaller the diffusion coefficient. Sulfuric acid is considered to be a divalent negative ion. The results proved that the electrolytic expansion is surely due to the doping and dedoping of bulky ions. However, in the PMAN, the expansion ratio does not show any dependence on the kind of anions at pH = 0, which will be discussed later. It is important to note that the expansion ratio at the extrapolation to the zero ion radius is a finite value of about 1% and not zero for both films. The finite value results from the conformational change of polymer structure and the electrostatic repulsion.

8.5 pH Dependence of Electrolytic Expansion The observed shift of the redox peaks with change in pH of the solution in an aqueous electrolyte solution indicates that protons are sometimes involved in electrochemical reaction. The pH dependence of the E 1/2 and electrolytic expansion in polyaniline and poly(o-methoxyaniline) [20] are shown in Figures 8.6a and 8.6b, respectively. As is evident from these results, E1/2 is pH dependent at pH below 0 for polyaniline and below 1.5 for poly(o-methoxyaniline). The gradient of the pH dependence is approximately 60 mV/pH for both curves, indicating that the oxidation takes place by the ejection of a proton per one electron [9]. On the other hand, at pH greater than 0 for polyaniline and greater than 1.5 for poly(o-methoxyaniline), the E1/2 is independent of the pH, suggesting that anions are injected by the oxidation. Taking this into account, the electrolytic expansion at pH < 0 (polyaniline) or < 1.5 (poly(omethoxyaniline), where protons are ejected by the oxidation, results from the conformational change and electrostatic repulsion. The difference of the magnitude in electrolytic expansion rates at lower and higher pH regions may be attributed to the inserted bulky anions plus protons. The dependence of the electrolytic expansion rates in poly(o-methoxyaniline) film on the type of anions is shown in Figures 8.7a and 8.7b for pH = 0 and pH = 2, respectively. It should be noted that at pH = 0, the expansion rate scarcely depends on the kind of anion, whereas at pH = 2 the remarkable dependence is observed. The result indicates that at pH = 0 or pH < 1.5 in poly(o-methoxyaniline) the electrolytic expansion and contraction are certainly driven by the change of polymer conformation and/or the electrostatic repulsion.

262

Polymer Microactuators

(a)

(b) Figure 8.6 The pH dependency of the E1/2 and electrolytic expansion in (a) polyaniline, and (b) poly(o-methoxyaniline)

263

Handbook of Polymers in Electronics

(a)

(b) Figure 8.7 The dependency of electrolytic expansions in poly(o-methoxyaniline) films on the kind of anions at (a) pH 0, and (b) pH2. TSA is toluene sulfonic acid.

264

Polymer Microactuators

8.6 Time Response of the Electrolytic Expansion Usually the diffusion constant in a solution is much larger than in a solid, as the rate of oxidation and reduction are determined by the diffusion of ions in the film. The typical time responses [13, 23] of an applied step potential, current and the expansion and contraction in polyaniline film at pH = 0 are shown in Figure 8.8. The larger the applied potential, the faster is the response of expansion. However, the response of the expansion is slower than the current response, since the film expansion takes place after the oxidation of film. From the response of the injecting charge normalised to the saturated value based on Equation 8.1, the diffusion coefficient of dopant ions can be estimated. The diffusion constants in polyaniline [13] and poly(o-methoxyaniline) film [23] in various electrolytes at pH = 0 have been summarised. The diffusion constants of poly (o-methoxyaniline) are always larger than that of polyaniline, as the oxidation occurs by ejection of protons in poly(o-methoxyaniline), while in the case of polyaniline the oxidation occurs by injection of bulky anions as discussed previously.

Figure 8.8 Typical time responses of an applied step potential (upper), current (middle) and the expansion and contraction (bottom) in polyaniline film at pH 0

265

Handbook of Polymers in Electronics

8.7 Anisotropy of Electrolytic Expansion in Polyaniline Films Polymers are quasi-one-dimensional; therefore, anisotropic behaviour of the electrolytic expansion is expected [14, 24]. Uniaxial pulling can stretch the polyaniline films by casting from the NMP (used as plasticiser) solution. In the stretched polyaniline film, chains align in the stretched direction. The anisotropic expansion of uniaxially stretched film parallel and perpendicular to the stretched direction and of unstretched film is shown in Figure 8.9 [14]. The electrolytic expansion perpendicular to the stretched direction is a rough estimation, because of the creeping effect caused by the continuous elongation arising as a consequence of repeated electrochemical oxidation and reduction. The electrolytic expansion for the perpendicular direction is larger than that of the unstretched film, resulting from the fact that the dopant ions are intercalated between the main chains.

Figure 8.9 The anisotropic expansion of uniaxially stretched film

The electrolytic expansion for the thickness direction in polyaniline cast film [20] shows an extremely large expansion ratio of more than 25% as shown in Figure 8.10, and is comparable to that of natural muscles [6]. A similar result was also obtained in the cast film of poly(o-methoxyaniline) for the thickness direction. The large expansion ratio for the thickness direction is conjectured to relate to the condensation process of the cast film. It may be remarked that the evaporation of NMP solution results in shrinkage only in the thickness direction, but not in the area. Therefore, the cast film has more freedom to expand in the thickness direction than that parallel to the film surface.

266

Polymer Microactuators

Figure 8.10 CV curve (upper) and electrolytic expansion (lower) ratio for the thickness direction in a polyaniline cast film

8.8 Contraction Under Strain in Stretched Polyaniline Films The contraction ratio of polyaniline film as a function of strain obtained in the stretched polyaniline film is shown in Figure 8.11 [25]. The result indicates that the contraction force is of the order of 1-2 MPa, which is about 10 times larger than that of natural muscles. Under smaller strains, the contraction ratio for the perpendicular direction is slightly larger than that observed in the stretched direction. At larger strains, however, the contraction ratio decreases more rapidly in the stretched direction than in the perpendicular direction, indicating that the mechanical strength is weaker in the direction perpendicular to polymer chains.

8.9 Electrolytic Expansion in Other Conducting Polymers Apart from polyaniline, other conducting polymers that are being studied for electrolytic expansion include polypyrrole [11, 15-17], poly(alkylthiophene) [26] and carbon nanotubes [5]. For example, electrochemically prepared polypyrrole films were used to study the qualitative movement of electrolytic expansion by fabricating a bimorph actuator. The movement of bending and stretching of the actuator was demonstrated in electrolyte solution [15]. Actuators fabricated by electrodeposition on gold-coated polyethylene films were studied [11] for the evaluation of expansion ratio and response time. Also, a microactuator of several tens of microns made from two layers of gold and

267

Handbook of Polymers in Electronics

Figure 8.11 The typical contraction ratio of polyaniline film against the strain for the stretched direction (//) and the perpendicular direction (⊥)

polypyrrole [16] has been demonstrated. Using polypyrrole films [17], the expansion ratio and the force were evaluated by an Instron pulling test machine. The expansion behaviours of bimorph actuators using poly(alkylthiophene) solids and gels [26] have also been studied. In carbon nanotube actuators [5], backbone types [24] were fabricated for demonstration of the bending motion. In these actuators even electrolyte solutions were used; the origin of actuation was explained by the mechanism of non faradic charging and discharging on the enormous surface area of the carbon nanotubes.

8.10 Applications of Electrolytic Expansion Actuators fabricated by conducting polymers are soft, flexible, and lightweight, with low voltage drive and strong contraction force. Tweezers, microvalves, and directors of optical fibre [27] are some of the technological applications. Two types of bimorph actuators have been proposed [24]. One is the backbone type, which consists of two conducting polymer films stuck together on a double-sided adhesive tape, which can also be replaced by a solid polymer electrolyte. The other is the shell type, in which conducting polymer films stuck on adhesive tapes are sandwiched onto a sheet of electrolyte media. The shell-type actuator is self-standing and works in air. There are some advantages in the use of these bimorph structures. One of these relates to the oxidation and the reduction of the polymer film process resulting in the bending force doubling. Since dopants transfer from one film to the

268

Polymer Microactuators other film through the electrolyte media, the electrolyte can be minimised. The electrolytic actuator can also be used both for positioning and rechargeable battery.

8.11 Conclusions The fundamentals of electrolytic expansion in polyaniline films have been discussed. Ion insertion and exclusion by electrolytic oxidation and reduction are the primary mechanisms. However, it is also evident that the changes in molecular conformations, arising due to the delocalisation of π-electrons and the electrostatic repulsion between the polycations, are other mechanisms operating in a conducting polymer microactuator. By investigating the molecular structure and the higher order structure to optimise the electrolytic expansion, it should be possible to improve the expansion ratio and the force for practical usage.

References 1.

R.H. Baughman, Makromolekulare Chemie, Macromolecular Symposium, 1991, 4, 277.

2.

K. Oguro, Y. Kawami and H. Takenaka, Journal of Micromachine Society, 1992, 5, 1, 27.

3.

K. Kaneto, M. Kaneko, Y. Min and A.G. MacDiarmid, Synthetic Metals, 1995, 71, 1-3, 2211.

4.

Y. Osada, H. Okuzaki and H. Hori, Nature, 1992, 6357, 242.

5.

R.H. Baughman, C. Cui, A.A. Zakhidov, Z. Iqubal, J.N. Barisci, G.M. Spinks, G.G. Wallace, A. Mazzoldi, D.D. Rossi, A.G. Rinzler, O. Jaschinski, S. Roth and M. Kertesz, Science, 1999, 284, 1340.

6.

R.M. Alexander, Nikkei Science, 1992, 13 (in Japanese).

7.

Y. Hirshberg, Journal of the American Chemical Society, 1956, 78, 2304.

8.

C.K. Chiang, M.A. Drug, S.C. Gau, A.J. Heeger, E.J. Louis, A.G. MacDiarmid, Y.W. Park and H. Shirakawa, Journal of the American Chemical Society, 1978, 100, 1013.

9.

W.-S. Huang, B.D. Humphrey and A.G. MacDiarmid, Journal of the Chemical Society, Faraday Transactions, 1986, 82, 2385.

10. L.W. Shacklette, J.F. Wolf, S. Gould and R.H. Baughman, Journal of Chemical Physics, 1988, 88, 6, 3955.

269

Handbook of Polymers in Electronics 11. Q. Pei and O. Inganäs, Synthetic Metals, 1993, 55-57, 1, 3718. 12. T.H. Herod and J.B. Schlenoff, Chemisty of Materials, 1993, 5, 7, 951. 13. K. Kaneto, M. Kaneko and W. Takashima, Japanese Journal of Applied Physics, 1995, 34, 7A, L837. 14. W. Takashima, M. Fukui, M. Kaneko and K. Kaneto, Japanese Journal of Applied Physics, 1995, 34, 7B, 3786. 15. T.F. Otero, J. Rodriguez, E. Angulo and C. Santamaria, Synthetic Metals, 1993, 55-57, 1, 3713. 16. E. Smela, O. Inganäs and I. Lundström, Science, 1995, 268, 1735. 17. M.R. Gandhi, P. Murray, G.M. Sprinks and G.G. Wallace, Synthetic Metals, 1995, 73, 247. 18. K. Kaneto, K. Kudo, Y. Ohmori, M. Onoda and M. Iwamoto, IEICE Transactions, 1998, E81-C, 7, 1009. 19. K. Kaneto, S. Hayashi, S. Ura and K. Yoshino, Journal of the Physical Society of Japan, 1985, 54, 3, 1146. 20. M. Kaneko and K. Kaneto, Synthetic Metals, 1999, 102, 1350. 21. P. Murray, G.M. Spinks, G.G. Wallace and R.P. Burford, Synthetic Metals, 1997, 84, 1-3, 847. 22. H. Daifuku, T. Kawagoe, N. Yamamoto, T. Ohsaka and N. Oyama, Journal of Electroanalytical Chemisty, 1989, 274, 313. 23. M. Kaneko and K. Kaneto, IEICE Transactions, 1998, E81-C, 7, 1064. 24. K. Kaneto, Y. Min and A.G. MacDiarmid, inventors; University of Pennsylvania, assignee; US5,556,700, 1996. 25. K. Kaneto in Gel Handbook, Eds., Y. Osada and K. Kajihara, N.T.S., Japan, 1997, 354. 26. X. Chen and O. Inganäs, Synthetic Metals, 1995, 74, 159. 27. R.H. Baughman, L.W. Shacklette, R.L. Elsenbaumer, E.J. Plichta and C. Bechtin in Molecular Electronics, Ed., P.I. Lazarev, Kluwer Academic Publishers, The Netherlands, 1991, 267.

270

9

Membranes for Electronics I. Karube and A. Hiratsuka

9.1 Introduction In materials science, thin films are designed to have a desired molecular order. The films should also have several different material properties in a restricted geometry. Organised films are now being designed to perform new and special functions. In the past, organic films were considered to be too fragile with insufficient purity to give reliable and consistent properties to make thin films for practical use. Extensive scientific studies have generated a wide variety of useful knowledge for the development of new thin film materials that exhibit specifically desired behaviour. A number of books, reviews, and general articles have been published about organic thin films, such as Langmuir-Blodgett (LB) films, self-assembled films [1, 2], conducting films [3, 4] and imprinted polymers. Other related topics include polymer surfaces and interfaces [5]. However, this trend is changing with the discovery of new materials. Recently, many new compounds and polymers have been synthesised and made into thin films by a variety of techniques. These organic films are carefully constructed to avoid the common problems. Although the preparation and study of LB films are still very active areas of research, polymeric films fabricated by either oxidative deposition or polymerised in situ are rapidly becoming more popular. Many intensive scientific investigations are focused on the preparation and characterisation of new polymeric organic films. Many newer surface science techniques, designed and developed for semiconductors and dielectric materials, address specific details about the structure and morphology of these organic films. Fundamental studies are also in progress for the characterisation of the optical, spectroscopic, and electrical properties, including energy transfer between molecules in the film and between layers. Clearly, knowledge of the morphology of the substrate is needed. Significantly, cooperative effects can change the behaviour of any film. The extent of interaction, both laterally and vertically between the layers and substrates, could influence behaviour. With this base of knowledge, it is hoped that this leads to our understanding of intermolecular interactions, energy transfer, and dynamic behaviour for best optimising a film for a specific research study or specific application.

271

Handbook of Polymers in Electronics Another field of scientific research is ‘interface materials’. These materials interface between two layers of different materials. This is often studied between two different polymeric layers. In this regard, surface pre- and post-treatment creating specific interactions are active areas of research. This leads to wetting phenomena, adsorption, adhesion, and lubrication. The nature of the interfacial organic thin films contributes to the orientation of the molecules and their packing. It also affects molecular motion, including lateral diffusion, gas or ionic diffusion through the film, phase transitions, melting and other processes. Polymeric materials for the purpose of biomedical applications are presently of growing concern. One can find applications in such diverse fields as tissue engineering, implants, therapeutic devices and diagnostic assays. Polymers can be prepared with several compositions and modifications while their physical properties including morphology can be regulated by variation in their composition and modifications. Polymer surface-based chemistries for biosensors are gaining importance and will continue to see growth in demand in the near future. Polymers are used to enhance the speed, sensitivity and versatility of biosensors and in medical diagnostics to measure vital analytes. Today’s diagnostic medicine is placing demands on technology for new materials and specific applications. Polymers are thus finding ever-increasing use as diagnostic medical reagents [6]. One example of the application of polymers in coatings is use as a binder, which is necessary to integrate the system chemistry. The reagent matrix must be carefully selected to mitigate or eliminate non uniformity in a reagent’s concentration due to improper mixing, setting or non uniform coating thickness. Therefore, aqueous-based emulsion polymers and watersoluble polymers are being extensively used. Polymers must be carefully screened and selected to avoid interference with the chemistry. The properties of polymers, e.g., solubility, viscosity, solid content, surfactants, residual initiators, film forming temperature and particle size should be carefully considered. In general, the polymer should have good adhesion to the support substrate and should show little or no change during handling or manufacture of films. The coated matrix must have the desired pore size and pore distribution to allow penetration of the analyte being measured as well as having the desired gloss, swelling characteristics and surface energetics. Depending on the system, swelling of the polymer binder due to the absorption of the liquid sample may or may not be advantageous. Emulsion polymers have a distinct advantage over soluble polymers due to their high molecular weight, superior mechanical properties and potential for adsorbing enzymes properties in a restricted geometry. Polymeric binders used in multilayered coatings include various emulsion polymers, gelatine, polyacrylamide, agarose, polyvinylpyrrolidone, polyvinyl alcohol, copolymers of vinylpyrrolidone and acrylamine, and hydrophilic cellulose derivatives such as hydroxyethylcellulose and methyl cellulose. Biosensors have been widely researched and developed as a tool for medical and environmental monitoring. They are designed to produce a digital electronic signal that is proportional to the concentration of a specific chemical or a set of chemicals.

272

Membranes for Electronics The biological recognition element is generally chosen from the enzymes, antibodies, receptors, tissues and microorganisms due to their excellent selectivity for target substances. The transducers could be electrodes, photon counters, thermisters, quartz crystal microbalances or semiconducting devices. Ion-selective, field-effect transistors (ISFET) and surface plasmon resonance (SPR) techniques have also been utilised. An essential technique in developing biosensors includes immobilisation of the biological components to the surface of the transducers. The performance of the biosensors is governed by the techniques used in combining these two components. The recent trend in biosensors is towards miniaturisation with semiconductor microfabrication or micromachining techniques [7-9]. The immobilisation matrix (interfacial design) should feature a thin film in order to maintain the desired sensor characteristics such as response time, sensitivity, reproducibility and reusability. There has been an increasing utilisation of organic thin films in many new electronic, optical and mechanical devices. Organic photoconductors are used in copiers and printers. The first organic photoconductors were charge transfer polymers that can generate and transport charges. In later versions, these functions were separated into different polymer layers, each of which could be handled easily. Polymers as liquid crystal displays are now ubiquitous in watches and flat panel displays. Not only have the liquid crystals been improved, but also the surface treatment and manufacture processes have been significantly advanced. This has led to the surface modification of organic materials for specific applications. Photoresists and electron-beam resists are the key to the success of VLSI electronic circuits. Without these resists, most electronic equipment would not exist. These polymers are spun onto the semiconductor and exposed to the circuit pattern leading to main chain scission or crosslinking. Subsequently, unpolymerised sections are removed. This process is employed either in wet or in dry conditions. This is known as the photolithographic process, which is part of the semiconductor fabrication technology. Further treatment includes diffusion of various semiconductor elements and metallisation for conduction lines. Layer by layer, the total package is developed. Current research is now directed toward finer features in the patterns and changes in the surface characteristics for subsequent layers. High temperature polymers, such as polyimide and other related polymers, are also used as insulating layers or for packaging. Further, sputtered or thermally oxidised SiO2 layers were used. This process required a large sputtering chamber and the films obtained were not uniform resulting in the development of cracks. However, polyimides did not cause problems. The adhesion of metal layers to polyimide was at times not so good. Both of these problems now seem to be under control. In the case of magnetic disks, storage density is demanded. Efforts to reduce the bit size by manoeuvring the read-write head closer to the disk surface tends to generate new problems such as stick or friction. To overcome these problems, lubricants were added

273

Handbook of Polymers in Electronics to the disk surfaces. The most popular have been fluorinated polymeric ethers, usually on top of an amorphous carbon coating, allowing the read-write head to be much closer to the magnetic storage disk surface. A number of groups are studying friction, lubrication, and adhesion from a fundamental point of view [10-12]. Sanyo introduced electrolytic capacitors in 1983 based on tetracyanoquinodimethane (TCNQ). Since that times several other capacitors have been introduced into the marketplace. Polypyrrole, polyaniline and polythiophene have all been used. These capacitors range in values from 0.1 μF to 200 μF and have low equivalent series resistance and high frequency impedance with good reliability and lifetimes. In 1967, Updike and co-workers investigated a promising approach to glucose monitoring in the form of an enzyme electrode [13]. Glucose was oxidised to glucono-lactone by the enzyme of glucose oxidase (GOD); the enzyme was incorporated and immobilised in the polymer gel matrices. This application of polymeric gel materials is growing very fast. Polymers have found applications in such diverse biomedical fields as tissue engineering, implantation of medical devices and artificial organs, prostheses and many other medical fields. Various approaches are used for diagnostic applications of these biomaterials. In the first approach, enzymes are used. As for GOD, oxygen consumption is measured. An enzyme-immobilised electrode is not essential in this system but an oxygen electrode is required for measuring changes in the dissolved oxygen concentration. The second approach uses an enzyme-immobilised electrode and the commonly used immobilising matrices are polymers. The enzyme is immobilised close to the electrode, and so only a small amount of reaction product is needed. Thus response time and sensitivity are improved. This system is polarised at a suitable redox potential. In the third approach, the polymers are functionalised not only as immobilised matrices but also as electron mediators. In this approach, the nature of the enzyme reaction is exploited. For example, in the case of GOD, the enzyme is reduced (or oxidised) by Dglucose (substrate) and then this reduced (oxidised) enzyme is oxidised (reduced) by an acceptor. The mediator could also be a redox polymer. Direct electron transfer between GOD and the electrode can also occur. In the past decade, much work has been done on the exploration and development of redox polymers that can rapidly and efficiently shuttle electrons. Several research groups have wired the enzyme to the electrode with a long chain polymer having a dense array of electron relays. The polymer penetrates and binds the enzyme to the electrode. Gregg and Heller have done extensive work on osmium-containing polymers. They have made a large number of such polymers and evaluated their electrochemical characteristics [14]. Their most stable reproducible redox polymer was poly(4-vinyl pyridine) to which Os(bpy)3Cl2

274

Membranes for Electronics (tris-(2,2´-bipyridyl) osmium chloride) had been attached to the pendant pyridine group. The resultant redox polymer was water insoluble. To make it water soluble and biologically compatible, Heller and co-workers partially quaternised the remaining pyridine pendants with 2-bromoethylamine. This polymer is water soluble and the newly introduced amine groups can react with a water-soluble epoxy, e.g., polyethylene glycol diglycidyl ether, and GOD to produce a crosslinked biosensor coating film. Such coating films produce high current densities and a linear response to glucose up to 600 mg/dL. In contrast, Boguslavsky and co-workers used a flexible polymer chain to put electron relays [15]. Their polymers provide communication between redox centres in glucose oxidase (GOD) and the electrode. No mediation occurred when ferrocene was attached to a non silicone backbone. Their ferrocene-modified siloxane polymers are stable and non diffusing. Therefore, biosensors based on these redox polymers give good response and stability. In order to transfer electrons directly between the electrodes and enzymes, an electron relay that transfers redox equivalents (electrons) from the active site of GOD’s cofactor to the electrode surface is needed. The choice for an artificial electron relay depends on a molecule’s ability to reach the reduced flavin adenine dinucleotide, FADH2 (in close proximity to the GOD active site), undergo fast electron transfer, and then transport electrons to the electrodes as rapidly as possible. Surridge and co-workers have carried out electron-transport rate studies on an enzyme electrode for glucose using interdigited array electrodes [16]. For integrating biosensors with electronics, the enzyme electrode should be fabricated in a mass fabrication oriented technique and should be capable of being miniaturised. Integration of enzymes and mediators simultaneously should improve the electron transfer pathway from the active site of the enzyme to the electrode. Conductive polymers such as polypyrrole, polyaniline and polythiophene are formed at the anode by electrochemical polymerisation [17]. For integration of bioselective compounds and/or redox polymers into conductive polymers, functionalisation of conductive polymer films is essential. There is a pressing need for an implantable glucose sensor for optimal control of blood glucose concentration in diabetics. A biosensor providing continuous readings of blood glucose would be most useful at the onset of hyper- or hypoglycaemia, enabling a patient to take corrective measures. Furthermore, incorporating such a biosensor into a closed-loop system with a microprocessor and an insulin infusion pump could provide automatic regulation of the patient’s blood glucose. Johnson and co-workers used two novel technologies in the fabrication of a miniature sensor for implantation in the subcutaneous tissues of humans with diabetes [18]. They developed an electrodeposition technique to electrically attract GOD and albumin onto the surface of the working electrode. The resultant enzyme/albumin layer was crosslinked by butraldehyde. They also developed a biocompatible polyethylene glycol/polyurethane copolymer to serve as the outer membrane of the sensor to provide differential permeability of oxygen relative to glucose to avoid the oxygen deficit encountered in physiological tissues.

275

Handbook of Polymers in Electronics Recently, commercial electrochemical microbiosensors, e.g., Exactach (Medisense), and silicon-base arrayed systems (I-Stat) have appeared in the market. These new technologies will certainly have an impact on rapid chemical analysis by the turn of this century. A typically important example is that of blood glucose determination on very small blood volumes (<5 μL) obtained by a finger pricker. This is possible as detection instruments can be designed compactly. Certainly, there is a market for small, disposable electrochemical tests in the emergency room, surgical and critical care units as well as in homes. Fabrication of polymers for these small and integrated sensors should be by the new processing technologies, which can produce accurate, mass reproducible and thin polymers. The polymers fabricated by conventional methods may have potential problems such as the difficulty of preparing thin (<1 μm) and homogeneous films. A plasma-polymerised film offers a new alternative [19]. The plasma-polymerised film is achieved in a glow discharge or plasma in the vapour phase. Such films are thin (< 1 μm), pinhole-free, flat-surface structures and are chemically and mechanically stable. In the following sections, the characterisation techniques, properties and applications for both electronical and biological aspects of plasma-polymerised films are described.

9.2 Plasma Polymerisation The polymers employed for integration between biomaterials and electronics should be mass-producible, very thin, and stable. In general, organic thin films have received a great deal of interest due to their extensive applications in the fields of mechanics, electronics and optics [20, 21]. Applications also include chemical, physical and biological sensors, microelectronic devices, non linear optical (NLO) and molecular devices [22]. In spite of a large number of studies in this area, only a few of these materials have been successfully used, even for electronic and optic-based applications. This is mainly because organic films often show poor thermal and chemical stability and poor mechanical toughness [23]. Therefore, it is of interest to develop polymer thin films of high quality for a variety of industrial applications. Ultrathin polymer films can be prepared using two kinds of technology. The first includes wet processes like LB, spreading, dipping or solvent casting methods. The other is dry processing, such as physical vapour deposition (PVD) and chemical vapour deposition (CVD). Of these methods, the CVD methods, such as plasma polymerisation, are frequently used to make polymer thin films [24-26]. Comparing these two technologies, dry processing is more advantageous, mainly because its technology originated from semiconductor or VLSI processes. VLSI and related

276

Membranes for Electronics technologies are very compatible with mass-production miniaturisation and integration processes. Among all the VLSI technologies, plasma polymerisation is a process of preparing thin polymer films. For the time being, there have only been a few examples of plasma polymerisation compared with other ordinary polymer processing techniques such as chemical or electrochemical synthesis. This is partly because of the difficulty of controlling the polymeric reactions. In spite of the difficulty, extensive work has been done on characteristic studies and improvement of instrumental and processing technologies. Hence, applications to electronic fields have been carried out and several practical examples for surface coatings and corrosion protections can be found. Plasmapolymerised films for chemical and biochemical applications are still rare, but the number of reports related to this field is currently on the increase.

9.2.1 History Organic chemical reactions induced by discharge have been used and studied since the 1800s [27, 28]. In 1874, Thenard [29] and Wilde [30] reported that solid materials were produced on the reaction wall after discharging in hydrocarbon vapours. In 1931, Brewer and coworkers [31] reported that insoluble yellow coloured materials were produced when applying the glow discharge to methane at room temperature. Stewart [32] produced insulating film from hydrocarbon gas under a high-pressure condition of 10-5 Torr. However, most of these materials were not intended to control the character or the structures of the reaction products. The initial study on the nature of well-defined adhesion and uniformity structures for metal coating films was reported by Goodman [33]. Following this work, plasma polymerisation has been utilised for the processing of pinhole-free and sub-micron films [34].

9.2.2 General Characteristics Plasma polymers are deposited as a thin film and/or as a powder on surfaces contacting a glow discharge of organic or organometallic feed gases. Plasma polymerisation is a specific type of plasma chemistry, which involves reactions between plasma species, between plasma on the surface [35]. Among the several possible mechanisms which have been expressed, a free-radical reaction could be a dominant process for plasma depositions [36, 37]. Hence, two types of reaction, i.e., plasma-induced polymerisation and plasma-state polymerisation are presumed. The plasma-induced polymerisation is the conventional free-radical induced polymerisation of molecules containing unsaturated carbon-carbon bonds. The plasma-state polymerisation depends on the presence of ions, electrons and other species which are energetic enough to break any bond. The resulting decomposition products of the plasma recombine by a free-

277

Handbook of Polymers in Electronics radical termination reaction. This allows the polymerisation of unconventional starting materials, such as saturated alkanes or benzene [38]. The most notable aspect of plasma polymerisation is that the thin polymer film can be prepared from almost any kind of organic vapour. Unlike conventional organic polymers, plasma polymers do not consist of chains with a regular repeat unit, but tend to form an irregular three-dimensional crosslinked network. The chemical structure and physical properties may be quite different from the conventional polymer derived from the same starting materials. Plasma-polymerised films are in general chemically inert, insoluble, mechanically tough, and thermally stable. So these films have been used in a variety of applications such as permeability selective membranes, protective coatings, and electrical, optical and biomedical films. There are several advantages of plasma-polymerised films over conventional polymers: •

The starting feed gases used do not have to contain the type of functional groups normally associated with conventional polymerisation.

•

Films are often highly coherent and adherent to a variety of substrates, including conventional polymers, glasses, and metals.

•

Polymerisation may be employed without the use of solvents.

•

Plasma-polymerised films can be easily produced with thickness of 10 nm to 1 μm.

•

Ultrathin, pinhole-free films may be prepared.

•

With careful control of the polymerisation parameters, it is possible to fabricate a chemically functionalised surface with a desired thickness.

9.2.3 Synthesis of Plasma Polymers The physical and chemical properties and the deposition rate for plasma-polymerised films depend on the following factors: •

Reactor type,

•

Feed gas composition,

•

Frequency and power of the excitation signal,

•

Flow rate of feed gases,

•

Plasma temperature, and

•

Substrate position.

Several review articles have already discussed the effect of such properties. The plasma polymerisation reactions are especially influenced by the monomer gases and the power supplied.

278

Membranes for Electronics

9.2.3.1 Reactors Many authors [39-42] have provided detailed studies of experimental configurations. The most widely used reactor configurations for plasma polymerisation can be divided into the following types (see Figure 9.1):

Internal electrode reactor with dc power supply

Internal electrode reactor with RF power supply

External electrode reactor with RF power supply

Electrodeless microwave (2.45 GHz) reactor

Figure 9.1 Reactors for plasma polymerisation

279

Handbook of Polymers in Electronics •

Internal electrode reactors with direct current (dc) power supply Internal electrode reactors have several names, e.g., flat bed, parallel plate, planar, diode, etc. However, this system is rarely used to develop the plasma polymerisation because it breaks the insulation between the electrodes due to polymerisation on the cathode by the dc glow discharge.

•

Internal electrode reactors with radio frequency (RF) power supply The RF power supply is coupled to the system by means of a blocking capacitor (capacitive coupling). An applied electrode potential oscillates around a cathode selfbias potential, which is prone to being negative. The working conditions and apparatus geometry can significantly influence the extent of ion bombardment on the substrate, the electron energy distribution and the production of active species [43]. Metal plate electrodes are aligned parallel with the reactor. Either an alternating current (ac) (1-50 kHz) or RF field is used. The vacuum chambers can either be made of glass or conductive materials, such as metals. In the case of bell jar reactors, no particular care is essential about the grounded electrode, but the design and arrangement of the cathode requires special attention. On the other hand, the metallic shield reactor surrounding the electrode improves the glow confinement inside the inter-electrode space. However, metallic reactor material can be sputtered, contaminating the target.

•

External electrode reactors with RF power supply External electrode reactors can be used either capacitively or inductively coupled. Many experimental arrangements have been reported. Each differs in power supply, reactor geometry, and sample position. The working frequency of the most commonly used power supplies is 13.56 MHz. Power is transmitted from the power supply to the monomer material by a capacitor and a coil. The tube geometry is variable [4446] and the position of the sample stage may vary from upstream to downstream of the monomer gas flow in order to obtain different polymer composition and properties. Insulating tubular reactors may be made of glass, quartz, ceramics or alumina as reactor materials. Usually inductively coupled tubular reactors cannot be uniformly coupled to the power supply while operating at low pressure (<< 1 Torr). However, coupling uniformity increases with increasing working pressure [47].

•

Electrode-less microwave or high frequency reactors The name of these reactors implies that no impurities can be sputtered off and incorporated into the growing films. A microwave (MW)-powered system is characterised by tubular quartz or Pyrex reactors and by a resonant cavity coupled with a power supply typically in the resonant cavity. The polymer is generally collected outside the glow region [48].

280

Membranes for Electronics The differences between the MW discharge and the RF discharge have been discussed [49]. The paper said that the large differences between the two systems could be the production of different active species and the different roles of ions and electrons.

9.2.3.2 Gas Parameters The main parameters of the gas are flow rate and pressure. The flow rate is controlled by either a mass flow or a needle valve, which connects between a gas reservoir and the reaction chamber. The deposition rate is limited by the supply of the feed gas. At a high flow rate, the deposition rate decreases and the activated species may be prevented from reaching the substrate. D’Agostino and co-workers showed that unsaturation of the feed gases always causes an increase of the maximum deposition rate by one order of magnitude and a shift to higher flow rate [50, 51]. The effect of the direction of flow on deposition rate distribution has received some attention. The effect of the position of the monomer inlet in a capacitively coupled belljar reactor has been studied [41, 52]. In the latter work, the position of the inlet was found to affect the efficacy of the resultant plasma polymer as a reverse osmosis membrane. The effects of the pressure on the plasma polymerisation process include: •

The effect on residence time: the residence time is directly proportional to the pressure.

•

The effect on average electron energy: for RF plasma the average electron energy is proportional to E/pg, where E is the electric field activating the plasma and pg is the pressure in the plasma. The chemical affects associated with low power plasma are observed at high pressures.

•

The effect on mean free path: the mean free path of a molecule, λ, in a gas is expressed as:

λ = (π r2 N) / 4

(9.1)

where r is the radius of the molecule and N is the gas density [53]. The pressure can affect the mean pathway by affecting the gas density. The polymer powders are formed between 200 mTorr to 2 Torr under relatively low flow rate conditions. The exclusion of powder formation will occur at above 2 Torr in which low mean free path, long residence time and relatively high electron energy are required.

281

Handbook of Polymers in Electronics However, inhomogenity in deposition rate is caused by an increase in pressure. Thus most plasma polymerisation is carried out at pressures below 1 Torr, in order to obtain the increased interaction of the plasma with a polymer surface to obtain homogeneous films [54, 55].

9.2.3.3 Power Parameter Increase of power will result in an increased density of energetic electrons and more frequent ion bombardment on the electrode. At constant pressure and flow rate, the deposition rate of the film increases with power at first and then becomes independent at higher values of power [43]. For a dc or ac glow discharge, an increase of power is obtained by both an increase in potential drop between electrodes and an increase in current density on the electrodes. Both effects result in an increased density of energetic electrons and increase bombardment of the electrode by energetic ions. For an RF discharge, the power has been reported to increase with an increase in current, which results in an increase in the energetic electrons.

9.3 Characterisation of Plasma Polymers Plasma polymer films are generally considered to be amorphous. The chemical composition and the nature of the internal bonding of plasma polymers are usually different from conventional polymers synthesised from the same monomer units. Plasma polymers may have a high degree of crosslinking and may contain unsaturated bonds. The relationship between the deposition condition and the resulting structure of the compositions has been investigated and reviewed [56, 57]. In most plasma polymerisation experiments, the quantities of solid plasma films are very small, typically milligrams, and the polymers are generally insoluble in organic solvents due to their high degree of crosslinking. These factors are essential for the characterisation of plasma polymers. Therefore, sophisticated tools should be used instead of the analytical methods generally used for conventional polymers. The structures of plasma polymer films have been studied by numerous modern analytical techniques. These include IR and Fourier Transform IR (FTIR) spectroscopy, UV absorption spectroscopy, electroluminescent spectroscopy, Auger electron spectroscopy (AES), X-ray photoelectron spectroscopy (XPS) or electron spectroscopy for chemical analysis (ESCA), secondary ion mass spectroscopy, ion scattering spectroscopy, solid state NMR spectroscopy, chromatography, combined pyrolysis/gas chromatography/mass spectroscopy (P/GC/MS), atomic absorption

282

Membranes for Electronics spectroscopy, neutron activation analysis, nuclear elastic recoil detection, elemental analysis, electron spin resonance (ESR) spectroscopy, X-ray diffraction, reflection high energy electron diffraction, differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), transmission electron microscopy (TEM), scanning electron microscopy (SEM) and atomic force microscopy (AFM). Among these analytical tools, IR and FTIR spectroscopy and XPS are frequently used to investigate the chemical and physical structures of films [54].

9.3.1 IR Spectroscopy IR spectroscopy can be used for the identification of functional groups in plasma polymer films. Although primarily a qualitative analytical tool, it has been used to quantitatively measure the concentrations of functional groups and the crosslinking density of plasma polymer films [42-44]. IR spectra might be obtained by depositing the plasma polymer on IR transparent substances or on other surfaces by using attenuated total reflectance (ATR).

9.3.2 XPS XPS (or ESCA) is a powerful tool for analysing the surface of plasma polymer films. Since plasma polymers are deposited as extremely thin films, XPS is ideally suited for the determination of the chemical properties of these films. XPS is a technique in which the number and energy of core level electrons expelled from atoms are analysed on their absorption of X-rays. The penetration depth of the X-ray is about 5 nm and this yields the relative population of elements at the surface. The electron energy observed for a given atom will also be influenced by the electron withdrawing power of the nearest neighbour atom. Thus, XPS is also suitable for functional group identification. This method is most effective when the nearest neighbour atom is highly electron withdrawing. Consequently, fluorinated plasma polymers are most widely studied by XPS [38]. XPS can also be used to obtain information regarding variations in depth by means of angle-resolved spectra and ion milling.

9.4 Applications of Plasma Polymers Both electronic and medical applications of plasma polymers have been reported [5461]. Most of these investigations are on the interface between polymers and inorganic materials, for instance, metal/polymer interfaces in structural adhesive joints, and cation diffusion along polymer/metal interfaces under an applied electric potential. In another reference, more specific aspects for electrical and electronic applications [59] were treated, wherein protective films for microcircuitry, and for wettability were explained. The use of such film for surface treatment has also been examined.

283

Handbook of Polymers in Electronics

9.4.1 Packaging Due to their excellent electrical and thermal properties, polysiloxanes are some of the most prevailing materials used in electronic encapsulation applications. Elastic thin films have been developed using some proprietary silicon-containing compounds [60]. Most of the conventional polysiloxanes are either gel-like or rubbery materials, and have limited applications in areas demanding high mechanical strength of the coating material in addition to passivation properties. However, plasma-polymerised films have good mechanical strength and resist cracking up to 25 microns thickness. These thin films are useful for conformal coatings in hermetic packages and as dielectrics for integrated circuits or multichip modules. Another packaging material is plasma-polymerised hexamethyldisilazane, which has a character of moisture impermeability and which could be used for protection of thin film nichrome resistors [61]. Plasma polymerisation/vapour deposition (PP/VD) was applied to provide functional hermetic encapsulation of integrated circuits [62]. The plasma-polymerised films combine the features of surface cleaning, surface treatment, and hermetic encapsulation into a series of in-situ vacuum processes, functional hermetic encapsulating polymer films with a total thickness of 5 μm have been produced, independent of the composition, dimensions or geometry of the substrates. Complementary metal oxide semiconductor (CMOS) circuits fabricated by a plasma process immersed in Ringer’s solution were tested for a year. The result showed unchanging device performance parameters characteristic of hermetically encapsulated integrated circuits.

9.4.2 Insulator Thin films of high thermal resistivity and electrical insulation were prepared by plasma polymerisation of silazane and subsequent pyrolysis in air. The film had strong adhesion, high thermal resistivity, and very good insulating properties in a broad range of temperatures. The author described how the chemical constitution of the films transformed into a silicon oxide type and a polycrystalline structure after the pyrolysis process [63]. The hybrid films for the pinhole-free electrical insulators were prepared using hexamethyldisiloxane (HMDSO) and silicon monoxide (SiO) [64]. The HMDSO hybrid films were prepared on the substrates by evaporating SiO during HMDSO plasma polymerisation in RF discharge. SiO was evaporated by heating in RF plasma consisting of HMDSO and oxygen at a pressure of 10-4 Torr. The films for insulators can be fabricated not only by inorganic compounds as mentioned above, but also by organic compounds. Plasma-polymerised films with uniform thin

284

Membranes for Electronics organic layers [65] can be prepared from almost any organic compounds. It has been reported that a plasma polymer copolymerised from allylamine and ethylene could be used as an orientation layer for a liquid crystal [69]. The synthesised films had good resistance to heat and chemicals. These plasma polymers have been considered as passivation or insulation layers of ICs and other electronic devices. Parylene C (poly(chloro-p-xylene) was used as an electrical insulator for the polymerisation [66]. A novel reactor was designed to deposit thin adhering and insulating films to common semiconductor materials. The reactor incorporates an electrode-less or inductively coupled power source and a thermally activated sublimation and pyrolysis chamber for vapour deposition of Parylene C. This design allows for the in-situ application of a thin adherent coating of plasma-polymerised methane followed by a thicker coating of Parylene C as well as serving as a ‘bio’ interface. Adhesion between the plasma polymer of methane and Parylene C could be improved by incorporating an intermediate layer of plasma polymerised para-xylyene film.

9.4.3 Semiconductive Films A doped material was prepared by plasma polymerisation using 3,4,9,10perylenetetracarboxylic dianhydride (PTCDA) [67]. Electrical conductivity of the iodinedoped material decreased by five orders of magnitude and the iodine doping caused a serious structural change. A structural and electrical survey for conductive phenomena was performed using acetonitrile plasma polymerised film [68]. This polymer was discovered to have two states of electrical conduction. Electrical conduction in the high conduction state was sensitive to the nature of electrode material. Electrical conduction in the low conduction state was unaffected by the electrode material. Organometallic compounds were used to fabricate semiconductive thin films on different substrates by glow discharge polymerisation [69]. Tetramethyltin (TMT) and diethylzinc (DEZ) were deposited on several substrates such as polypropylene, SnO2, quartz and glass. The physicochemical properties of the deposited films were characterised by FTIR, XPS, SEM and X-ray diffraction. Plasma-polymerised thiophene for passivating the surface defects on GaAs has been employed [70]. The paper showed the passivation of GaAs surface was made possible by sulfur present in an overlayer, provided by the thin film of plasma-polymerised thiophene. The deposition of polythiophene lowered the barrier height, reduced the surface recombination velocity and increased diffusion length.

285

Handbook of Polymers in Electronics

9.4.4 Conductive Films A conducting plasma polymer was processed using TCNQ with quinoline [71]. The results showed that the plasma-polymerised quinoline and TCNQ had conductivities of 10-12 Ω-1 cm-1 and 10-9 Ω-1 cm-1, respectively. However, a high conductivity of 10-5 Ω-1 cm-1 was measured for the plasma polymers mixed with TCNQ and quinoline. p-Xylene polymer was fabricated by plasma polymerisation and investigated for highfield conduction and photoconduction [72]. The electronic properties of the plasma polymer were similar to those of semicrystalline poly-p-xylene, influenced by the shortrange molecular order, while the charge transport influenced by the long-range order did not occur in these systems. An organic conductive plasma polymer was produced from fumaronitrile and paminobenzonitrile in a heated system [73]. The films were very smooth and pinhole-free, with electrical conductivity ranging from 10-6 to 10-7 Ω-1 cm-1.

9.4.5 Resist Films Lithography is indispensable for the fabrication of integrated circuits. Dry resist coating is a key technology to replace a conventional wet lithography. There have been several papers and reports about plasma-polymerised resists [76-78]. PMMA has been used as a plasma-polymerised resist [74, 75]. Copolymerised resists for high sensitivity have been produced [76-78], using metal-containing monomers blended in the resist polymer as resist sensitiser. Multilayered resist film was produced by forming alternating multiple layers. The resist properties were improved by including the plasma etching resistive layer in the multilayerd resist system [79].

9.4.6 Ultrathin Polymer Films It is well known that thin polymer films with thickness about 100–1000 nm can be prepared by plasma polymerisation or electron beam and UV irradiation of organic vapour. However, modern electronic applications require films of 30 nm or even thinner. In 1992, a new system was developed [80] which could plasma polymerise a pinhole-free ultrathin polymer film (2-10 nm) in a RF glow discharge. Ultrathin hydrocarbon films (C2H2 as feed gas) were deposited on electronic grade n-type silicon wafers. A capacitively coupled RF discharge was initiated in a bos-type reactor in a flow of pure hydrocarbon vapour, or a hydrocarbon/argon mixture. The deposition conditions included a discharge frequency of 13.56 MHz and a pressure range of 0.5-0.8 Torr.

286

Membranes for Electronics

9.4.7 Chemical Sensors Integration of sensors on silicon and/or glass is an emerging technology for fabrication of an intelligent sensing system. Low temperature plasma polymerisation processes are useful for the integration and show various unique characteristics of the sensors. ISFETs are used for pH measurement with a reference field-effect transistor. Sensing elements of the ISFETs were covered by the plasma polymerised styrene and were characterised [81]. An NO2 gas sensor was fabricated using plasma polymerisation of copper phthalocyanines. The materials processed were copper phthalocyanines without (CuPc) and with a chlorine substituent (CUPc-Cl), and with hydroxymethyl (CuPcCH2OH) and phthalimidomethyl substituents [82]. It has also been reported that the tetramethyl plasma-polymerised films can be used for sensing propane gas [83]. Plasma-polymerised humidity sensors were proposed using styrene (PPS) [84]. The sensor was constructed with a 12 nm Au - 164 nm PPS - 12 nm Au sandwich structure. The response time of the capacitance was less than one minute.

9.4.8 Biosensors The promising areas for applying plasma deposition technology to biomedical device fabrication are sensor electrodes and the packaging of implantable integrated sensors. Insulation and packaging materials on implantable electronic devices must be biocompatible. They must have good mechanical durability and good cohesive and adhesive properties, to prevent leakage from the implant surroundings.

9.4.8.1 Enzyme Support Plasma-polymerised HMDSO film was used to produce a biocompatible surface and an enzyme support system [85]. The adsorption of urease onto a well-defined solid support, petroleum-based activated charcoal, has been achieved to provide the enzymatic hydrolysis of urea. The adsorption of urease, and the activity and stability of the enzyme on the support were studied and optimised, improving its availability for clinical applications.

9.4.8.2 Catalytic Biosensors The most popular catalytic biosensor in clinical analysis is a glucose sensor, used in monitoring the blood sugar of diabetes patients. The biological component often used GOD.

287

Handbook of Polymers in Electronics GOD specifically catalyses the oxidation of glucose: D-glucose + O2  D-glucono-1,5-lacton + H2O2

(9.2)

An amperometric oxygen electrode or a noble metal electrode (such as platinum, silver and gold) detects the biochemical reaction event. The electrode current increases with an increase in the concentration of glucose. This event is primarily an electron transfer process mediated by a specific enzymatic reaction. In order to achieve the best sensor performance, the distance between the electrode and GOD should be as short as possible, with rejection of interfering compounds. The plasma-polymerised film is an alternative material for the immobilisation matrix that allows for such operation. There have been several typical examples shown in papers. A plasma-polymerised glucose microsensor using acrylic acid, methacrylic acid and 2-amino-benzotrifluoride was prepared [86]. The plasma-polymerised films were deposited on the electrode, which was fabricated on a silicon substrate. Subsequently, oxygen and ammonia plasma treatment was carried out to introduce carboxyl and amino groups, respectively, which were coupled to GOD. A large amount of GOD was immobilised onto the surface of the film. Due to the thickness of the film, the diffusional resistance for the analyte and reaction product is low and a response time of about 4 seconds was obtained. The calibration curves of the sensor were linear from 0.05 mmol/l glucose to 2 mmol/l glucose and the variance for a series of 10 injections of 2 mmol/l glucose was found to be 2%. Another sensor using plasma-polymerised films has been reported [87]. This overcomes the poor adhesion on slide glass or silicon substrates without a chromium layer (often used between a glass slide and platinum to promote adhesion but causes undesirable electrochemical side reactions with alloy formation). The calibration curves of the sensor were linear from 0.5 mmol/l glucose to 100 mmol/l glucose and a response time of about 2 seconds was obtained. Perfluoroallylphosphonic acid plasma-polymerised film was prepared as a charge rejection membrane [88]. The film deposited on a working electrode showed the ability to reject a negative organic interfering material such as ascorbate (vitamin C). This overcomes the problem with wire electrodes, which were found to be difficult to homogeneously coat with Nafion films. Encapsulation of GOD by plasma-polymerised film has also been proposed [89, 90]. GOD is encapsulated between a Millpore membrane filter and a propargyl alcohol plasmapolymerised layer. The pores of the plasma-polymerised matrix are small enough to prevent loss of GOD but large enough to allow the diffusion of glucose and electron. The polymerisation condition needed relatively low power discharge, and hence, the denaturisation of GOD frequently caused by the high energy of plasma species was minimised. The film also played a role as a barrier for interference from Cu2+. Nylon membranes have been widely used in such a sensor assembly for conventional applications. These sensors, however, have slow response times because the thickness of the nylon membranes was more than 50 μm whereas those of the layers of plasma-polymerised films were less than 100 nm.

288

Membranes for Electronics

9.4.8.3 Affinity Biosensors Affinity biosensors take advantage of biomolecular interactions such as antigen-antibody, receptor-ligand and nucleic acid base pair association, collectively referred as ‘biospecific interaction’ in the biochemical and medical field. Transducers detecting the binding events are mostly quartz crystal microbalances (QCM) and surface plasmon resonance (SPR) devices. Affinity biosensors employing QCM are principally based on the measurement of the change in resonant frequency of a QCM according to mass changes on its surface. The QCM enables one to detect the mass change with a biospecific-binding event. Ethylenediamine plasma-polymerised films formed on gold electrodes covering the surface of quartz crystals are extremely thin and homogeneous. A QCM coated with such a film showed stable oscillation in phosphate buffer saline (PBS) compared with those coated with polyethylenimine (PEI) and (3-aminopropyl) trimethoxysilane (APTES). The noise levels of the plasma-polymerised film, PEI, and APTES were 12, 54 and 50 Hz, respectively [91]. This indicated that the plasma-polymerised film on QCM was more homogeneous than APTES and PEI. Moreover, there are a large number of amino groups, which enabled antibodies to be immobilised on the surface of the film. Therefore, sensors produced using this method were more reproducible from sample to sample and exhibited lower noise and higher sensitivity than sensors made using conventional APTES and PEI immobilisation methods. Standard deviations (sample amount = 10) in the frequency due to antigen-antibody bindings were investigated for three kinds of coating methods. The values found for coatings made from plasma-polymerised film, APTES, and PEI were 5%, 47%, and 38%, respectively. The time-course profile of the sensor response demonstrates a good reusability. The calibration plot of frequency change against concentration of the antigen, human serum albumin (HSA), was linear for concentrations ranging from 10-2 to 10 mg/l with 15 minutes of incubation. They also reported on the ‘red tide’ sensor [92]. Antibodies to the red tide-causing plankton, Alexandrium affine, were immobilised onto a piezoelectric device coated with the plasma-polymerised film. Since the sensor is often used in seawater, the plasma-polymerised film matrix played a role as a coating against the seawater. Affinity biosensors using SPR detection have been widely reported [93, 94]. BIAcore (Uppsala, Sweden) has developed BIAcore 2000 Biosensor AB into a widely used commercial instrument [95]. A surface plasmon wave is a non propagating evanescent wave formed at a metal-coated (mainly gold-coated) surface when light is directed toward the interface at a very specific reflection angle. It extends from the metal surface into the sample solution, decaying exponentially as a function of distance. Refractive index changes localised near the glass/gold interface resulting from the interaction between biomolecules perturb the evanescent wave and alter the propagation characteristics of the surface plasmon. Therefore, the resonance angle changes. Since the evanescent wave decays exponentially as a function of distance, the interaction of biomolecule interactions should be carried out close to the gold surface to obtain high sensitivity. A typical penetration depth of the evanescent wave from the gold surface is

289

Handbook of Polymers in Electronics several hundred nanometres. When antibodies are directly immobilised on the gold surface, it is often difficult to attach a large amount of antibodies with good adhesion. Therefore, antibodies are bound to be denatured and the sensor surface suffers from low surface coverage as well as non specific adsorption. Plasma-polymerised films might be suitable for the coating matrix between the antibodies and the gold. The sensor chip made with the film showed a better sensor response (higher sensitivity) than a conventional design, e.g., carboxylated dextran designed by BIAcore AB. The reason is that the antibodies immobilised onto the plasma-polymerised films are two-dimensional and are present at a high density whereas antibodies immobilised onto carboxylated dextran are three-dimensional and are packed at a lower density. The calibration plot of change of resonance angle against concentration of HSA was linear for concentrations ranging from 10-2 to 50 mg/l with 7 minutes of incubation [96]. Similarly, another type of sensor chip was also reported [97] using a plasma-polymerised film and it was shown that it was effective for use in detection of the pesticide, etfenprox, which is a low molecular weight molecule. SPR detection for low weight molecules (molecular weight < 3000) is difficult with conventional sensor chips.

9.4.8.4 DNA Sensor In spite of the usefulness of plasma-polymerised film in biosensors, only a few cases have been reported on how such materials are used. Recently, a gene chip has been reported [98]. This chip is a high density, oligonucleotide probe array for easy detection or identification of genetic information from a virus. Due to the demand for immobilising thousands of different kinds of oligonucleotides in mm2 areas, the processes for fabricating these chips have employed semiconductor process technology. Interface materials between oligonucleotides and substrates need to be well characterised in terms of structural properties such as thickness, pinhole density, strong adhesion to a variety of materials, chemical resistivity and hydrophobicity. Miyachi and co-workers deposited a plasmapolymerised film of HMDSO as a support layer for deoxyribonucleic acid (DNA) oligonucleotide arrays [99]. Non specific DNA was avoided due to the thin, hydrophobic properties of plasma-polymerised film. Hence, the background signal of the DNA array was lower than that of a poly-L-lysine-coated [100] using a conventional interface layer. DNA array detection systems are widely used. Plasma-polymerised films and/or the plasma process as an interfacial design between substrates and DNA may thus help towards the increased the use of both biosensors and DNA arrays.

Acknowledgements The authors are grateful to Dr. Hitoshi Muguruma for many helpful discussions. We also acknowledge Dr. Scott J. McNiven and Dr. Kyong-Hoon Lee for their helpful suggestions.

290

Membranes for Electronics

References 1.

Langmuir-Blodgett Films, Ed., G. Roberts, Plenum, New York, NY, USA, 1990.

2.

M. Maskus, J. Tirado, J. Hudson, R. Bretz, H.D. Abruna, NATO Advanced Science Institutes Series, Series C: Mathematical and Physical Series, 1996, 485, 337.

3.

J.W. Schultze, T. Morgenstern, D. Schattka and S. Winkels, Electrochimica Acta, 1999, 44, 1847.

4.

N. Arsalani and K.E. Geckeker, Journal fur Praktische Chemie/ChemikerZeitung, 1995, 337, 1.

5.

Physics of Polymer Surfaces and Interfaces, Ed., I.C. Sanchez, ButterworthHeinemann, Boston, MA, USA, 1992.

6.

J. Jagur-Grodzinski, Reactive and Functional Polymers, 1999, 39, 99.

7.

M. Alvarez-Icaza and U. Bilitewski, Analytical Chemistry, 1993, 65, 11, 525A.

8.

G.T.A. Kovacs, K. Petersen and M. Albin, Analytical Chemistry, 1996, July 1, 9, 407A.

9.

A. Hiratsuka, S. Saski and I. Karube, Electroanalysis, 1998, 10, 49, 231.

10. K. Kendall, Science, 1994, 263, 1720. 11. H. Yoshizawa and J. Israelachvili, Thin Solid Films, 1994, 246. 12. J. Peanasky, L. Cai, S. Granick, C.R. Kessel, Langmuir, 1994, 10, 3874. 13. S.J. Updike and G.P. Hicks, Nature, 1967, 214, June 3, 986. 14. B.A. Gregg and A.J. Heller, Journal of Physical Chemistry, 1991, 95, 5970. 15. L. Boguslavsly, P. Hale, T. Skotheim, H. Karan, H. Lee, Y. Okamoto, Polymer Materials Science and Engineering, 1991, 64, 322. 16. N.A. Surridge, E.R. Diebold, J. Chang and G.W. Neudeck, ACS Symposium Series 556, ACS, Washington D.C., USA, 1994, 47. 17. S. Cosnier, Biosensors and Bioelectronics, 1999, 14, 443. 18. K.W. Johnson, D.J. Allen, J.J. Mastrototaro, R.J. Morff and R.S. Nevin, ACS Symposium Series 556, ACS, Washington, D.C., USA, 1994, 84.

291

Handbook of Polymers in Electronics 19. H. Yasuda, Plasma Polymerization, Academic Press, New York, NY, USA, 1985. 20. G.G. Roberts, Advances in Physics, 1985, 34, 475. 21. H. Biederman, Vacuum, 1987, 37, 367. 22. A. Kubono and N. Okui, Progress in Polymer Science, 1994, 19, 389. 23. S.T. Kowel, R. Selfridge, C. Eldering, N. Matloff, P. Stroeve, B.G. Higgins, M.P. Strinivasan and L. B. Coleman, Thin Solid Films, 1987, 152, 377. 24. Y. Takahashi, M. Iijima, K. Inagawa and A. Ito, Journal of Vacuum Science and Technology A, 1987, 5, 2253. 25. Techniques and Applications of Plasma Chemistry, Eds., R. Hollahan, A.T. Bell, Wiley, New York, 1974. 26. W. Gombotz and A. Hoffman, Journal of Applied Polymer Science, 1989, 37, 91. 27. O. Henry, Annales de Chimie Science, 1798, 25, 175. 28. M. Berthelot, Annales de Chimie Science des Physique, 1863, 67, 70. 29. A. Thenard, Comptes Rendus de l’Academie des Science, 1874, 78, 219. 30. P. De Wilde, Berichte Deutsche Chemische Gesellschaft, 1908, 41, 2683. 31. A. Brewer and R. Kveck, Journal of Physical Chemistry, 1931, 35, 1293. 32. R.L. Stewart, Physical Review, 1934, 45, 488. 33. J. Goodman, Journal of Polymer Science, 1960, 44, 551. 34. H. Pignia, Physica Status Solidi, 1961, 1, 499. 35. N. Morosoff, An Introduction to Plasma Polymerization, Plasma Deposition, Treatment, and Etching of Polymers, Academic Press, New York, USA, 1990. 36. A.K. Sharma, A. Hahn, F. Millich and E.W. Hellmuth, Plasma Deposition, Treatment and Etching of Polymers, Academic Press, New York, USA, 1980. 37. C.J.M. Striling, Radicals in Organic Chemistry, Oldbourn Press, London, UK, 1965. 38. H. Yasuda, Glow Discharge Polymerization, Thin Film Processes, Academic Press, New York, USA, 1978.

292

Membranes for Electronics 39. H. Konig and G. Helwig, Zeitschrift fur Physik, 1951, 129, 491. 40. T. Hirai and O. Nakada, Japanese Journal of Applied Physics, 1968, 7, 112. 41. H. Kobayashi, A.T. Bell and M. Shen, Journal of Macromolecular Science– Chemistry, 1976, A10, 3, 491. 42. R.A. Conell and L.V. Grggor, Journal of the Electrochemical Society, 1965, 9, 1198. 43. H. Biederman, Vacuum, 1987, 37, 367. 44. H. Yasuda and T. Hsu, Surface Science, 1978, 76, 232. 45. D.W. Rice and D.F. O’Kane, Journal of Macromolecular Science–Chemistry, 1976, A10, 3, 567. 46. H. Yasuda, Contemporary Topics in Polymer Science, 1979, 3, 103. 47. J. Vossen, Journal of the Electrochemical Society, 1979, 126, 319. 48. M. Duval and A. Theoret, Journal of the Electrochemical Society, 1975, 122, 581. 49. R. Claude, M. Moisan, M.R. Wertherimer and Z. Zakrzewski, Plasma Chemistry and Plasma Processing, 1987, 7, 451. 50. R. d’Agostino, F. Cramarossa and F. Illuzzi, Journal of Applied Physics, 1987, 61, 2754. 51. H. Yasuda, Journal of Polymer Science Macromolecular Review, 1981, 16, 199 52. D. Peric, A.T. Bell and M. Shen, Journal of Applied Polymer Science, 1977, 21, 2661. 53. E. Nasser, Fundamentals of Gaseous Ionization and Plasma Electronics, Wiley, New York, NY, USA, 1971. 54. R. d’Agostino, Plasma Deposition, Treatment and Etching of Polymers, Academic Press, New York, NY, USA, 1987. 55. R.V. Hinman, A.T. Bell and M. Shen, Journal of Applied Polymer Science, 1979, 23, 3651. 56. Techniques and Applications of Plasma Chemistry, Eds., R. Hollahan and A.T. Bell, Wiley, New York, NY, USA, 1974.

293

Handbook of Polymers in Electronics 57. Plasma Polymerisation, American Chemical Society, Eds., T. Bell and M. Shen, Washington, DC, 1979. 58. D.T. Clark and H.R. Thomas, Journal of Polymer Science, Polymer Chemistry, 1978, 16, 791. 59. R.K. Sadhir and H.E. Saunders, Plasma Processes for Electrical and Electronics Applications, IEEE, Electrical Insulation Magazine, 1986, 2, 6, 8. 60. A. Krishnan, J.C. Lee and N. Kumar, International SAMPE Electronics Conference, 1992, 6, 528. 61. B.L. Rathbun and P.W. Schuessler, Polymer Science and Technology, Plenum Press, New York, NY, USA, 785. 62. F. Nichols and A.W. Hahn, Biomedical Sciences Instrumentation, 27. ISA Services Inc., Research Triangle Park, NC, USA, 331. 63. J. Tyczkowski, P. Kazimierski and H. Szymanowski, Proceedings of the 3rd IEEE International Conference on Properties and Applications of Dielectric Materials, Tokyo, Japan, July 1991. 64. K. Kashiwagi, Y. Yoshida and Y. Murayama, Japanese Journal of Applied Physics, Part 1, 1991, 30, 8, 1803. 65. T. Nakano, O. Nakamura, N. Shibayama and Y. Ohki, Electrical Engineering in Japan, 1988, 108, 4, 8. 66. M.F. Nichols and A.W. Hahn, Proceedings of the ACS Division of Polymeric Materials Science and Engineering, 56, ACS, Washington, DC, USA, 608. 67. K. Tanaka, Y. Matsuura, S. Nishio and T. Yamabe, Synthetic Metals, 1994, 65, 1, 81. 68. B.A. Suleimanov, M.M. Akhmedov, Y.I. Suleimanova and M.K. Kerimov, Thin Solid Films, 1991, 197, 1, 319. 69. F. Tchoubineh, S. Mondin, F. Arefi and J. Amouroux, Proceedings of the ACS Division of Polymeric Materials: Science and Engineering, 1990, ACS, Washington, DC, USA, 62, 548. 70. M. Vardhireddy, R.S. Bhide, S.V. Bhoraskar and V.J. Rao, Proceedings of the ACS Division of Polymeric Materials: Science and Engineering, 1990, ACS, Washington, DC, USA, 62, 373.

294

Membranes for Electronics 71. X. Xie, J.U. Thiele, R. Steiner and P. Oelhafen, Synthetic Metals, 1994, 63, 3, 221. 72. Y. Takai, T. Mizutani and M. Ieda, Japanese Journal of Applied Physics, Part 1, 1987, 26, 6, 812. 73. Y.H. Park, H. Tsutsumi, S. Tasaka and S. Miyata, Polymer Journal, 1986, 18, 10, 713. 74. S. Morita, J. Tamano, S. Hattori and M. Ikeda, Journal of Applied Physics, 1980, 51, 7, 3938. 75. L. Martinu, H. Biederman, I. Haller, R. Feder, M. Hatzakis and E. Spiller, Journal of the Electrochemical Society, 1979, 126, 1, 154. 76. I. Heller, R. Feder, M. Hatzakis and E. Spiller, Journal of the Electrochemical Society, 1979, 126, 1, 154. 77. D.J. Webb and M. Htzakis, Journal of Vacuum Science and Technology, 1979, 16, 6, 2008. 78. M. Yamada, J. Tamano, K. Yoneda, S. Morita and S. Hattori, Japanese Journal of Applied Physics, 1982, 21, 5, 768. 79. J. Tamano, T. Kato, S. Morita and S. Hattori, Dry Process Symposium 3 1, Tokyo, October 11-12, 1984, 55. 80. K. Senda, G.K. Vinogradov, S. Gorwadkar and S. Morita, Journal of Applied Physics, 1993, 74, 6425. 81. S. Tahara, M. Yoshii and S. Oka, Chemical Letters, 1982, 307, 310. 82. N. Inagaki, S. Tasaka and Y. Ikeda, Journal of Applied Polymer Science, 1995, 55, 10, 1451. 83. N. Inagaki, Y. Hashimoto, Kobunshi Ronbunshu, 1986, 43, 10, 711. 84. S. Yakeda, Japanese Journal of Applied Physics, 1981, 20, 7, 1219. 85. K. Gunay and A. Guneri, Biomaterials, 1996, 17, 15, 1473. 86. G. Kampfrath and R. Hintsche, Analytical Letters, 1989, 22, 2423. 87. A. Hiratsuka, H. Muguruma, S. Sasaki, K. Ikebukuro and I. Karube, Electroanalysis, 1999, 11, 15, 1098.

295

Handbook of Polymers in Electronics 88. R.C. Tucker, I. Song, J.H. Payer and R.E. Marchant, Journal of Applied Electrochemistry, 1997, 27, 1079. 89. K. Yoshimura, T. Kitade, K. Kitamura and K. Hozumi, Microchemical Journal, 1991, 43, 133. 90. K. Yoshimura and K. Hozumi, Microchemical Journal, 1996, 53, 404. 91. K. Nakanishi, H. Muguruma and I. Karube, Analytical Chemistry, 1996, 68, 1695. 92. K. Nakanishi, M. Adachi, Y. Sako, Y. Ishida, H. Muguruma and I. Karube, Analytical Letters, 1996, 29, 1247. 93. B. Liedberg, C. Nylander and I. Lungstrom, Sensors and Actuators, 1983, 4, 299. 94. M.T. Flanagan and R.H. Pantell, Electronic Letters, 1984, 20, 968. 95. See: www.biacore.com/products/products.shtml 96. R. Nakamura, H. Muguruma, K. Ikebukuro, S. Sasaki, R. Nagata, I. Karube and H. Pedersen, Analytical Chemistry, 1997, 69, 4649. 97. S. Sasaki, E. Kai, H. Miyachi, H. Muguruma, K. Ikebukuro, H. Ohkawa and I. Karube, Analytica Chimica Acta, in press. 98. See: www.affymetrix.com/products/index.html 99. H. Miyachi, A. Hiratsuka, K. Ikebukuro, K. Yano, H. Muguruma and I. Karube, to be published. 100. M. Schena, D. Shalon, R. Heller, A. Chai, P.O. Brown and R.W. Davis, Proceedings of the National Academy of Sciences in the United States of America, 93, 1996, 10614.

296

10

Conducting Polymer-Based Biosensors A. Chaubey, M. Gerard and B.D. Malhotra

10.1 Introduction Current research and development in bioanalytical problems aims at improved stability, selectivity and sensitivity of sensing devices [1-6]. With the advent of computerisation and increasing sophistication in many technological areas, dramatic changes are being seen in the way scientific experiments are being controlled. Industrial applications of biochemical and morphological processes, in fields such as production of pharmaceuticals, food manufacturing, wastewater treatment and energy production, are on the increase. This has led to the development of biosensors. It is now widely accepted that sensing devices such as biosensors are likely to revolutionise several areas of analytical biotechnology, such as healthcare, veterinary medicine, agricultural, petrochemical and pollution monitoring, by providing valuable real-time information [7-12]. It has been observed that about 80% of the non medical biosensor application market (Figure 10.1a) is covered by food industries. The rest of the market (20%) involves environmental, agriculture, defence, veterinary medicine and general industrial processing, etc. The medical diagnostic biosensor application areas (Figure 10.1b) involve centralised testing (75%), decentralised testing (20%) and consumer testing (5%). Clark and Lyons were the first to demonstrate that an enzyme could be integrated into an electrode to construct a biosensor [13]. Updike and Hicks described the first functional enzyme electrode based on glucose oxidase deposited onto an oxygen electrode [14]. Technical developments of biosensors for medical care have demanded the greatest attention. Research on biosensors has been motivated by a strong practical instinct with clear applications in sight. The key attraction of biosensors for the applied sciences is that they offer the prospect of simplified, virtually non destructive analysis of turbid biological fluids. There is an increasing demand of biosensors for the determination of substances in biological fluids such as blood, urine, serum, etc. [15, 16]. A common requirement of all these is the need for on-site chemical information on dynamic or rapidly evolving processes, preferably on a real-time basis. There is a distinct trend in clinical analysis from a centralised laboratory to a doctor’s clinic and to a patient’s self testing at home. A medical practitioner requires correct information about various biological parameters, such as glucose, urea, haemoglobin, albumin, creatinine, amylase, lactate, etc. Many companies, including MediSense (USA), Boehringer (Germany) and Yellow Springs

297

Handbook of Polymers in Electronics

Figure 10.1 The biosensor market: (a) non medical, (b) medical

Instrumentation Company (USA), have marketed a hand-operated blood glucose biosensor. The National Physical Laboratory (India) has also patented a device that can be used for estimating glucose in blood.

10.1.1 Biosensors A biosensor is a synergic combination of analytical biochemistry and microelectronics. Biosensors have recently been considered as a highly potential field of scientific research. This is because development of biosensors is necessary for the realisation of implantable, integrated and intelligent devices for biochemical information [17]. Biosensors are analytical devices that respond selectively to analytes in a given sample and convert their

298

Conducting Polymer-Based Biosensors

Figure 10.2 Schematic of a typical biosensing device comprising of immobilised biomolecule and a transducer

respective concentration into an electrical signal via a combination of a biological recognition system and a physicochemical transducer. The schematic of a biosensor is shown in Figure 10.2.

10.1.2 Construction of Biosensors In analytical devices, the biologically active molecules, such as enzymes, cells and antibodies, are used repeatedly. These are, therefore, fixed to the carrier materials. There are several advantages for immobilising the enzymes for application in analytical chemistry. These include stabilisation of enzymes and retention of enzyme activity for long periods of time. Biosensors provide a powerful and inexpensive alternative to conventional analytical strategies for assaying chemical species in complex matrices. A biosensing device incorporates a biological molecular recognition component connected to a transducer. The main aim of a transducer is to produce a continuous or discrete electronic signal, which is directly proportional to the concentration of an analyte. In biosensors, the following sequence of events take place: •

Specific recognition of the analyte,

•

Transduction of the physicochemical effect caused by the interaction with the receptor into an electrical signal, and

•

Signal processing and amplification.

299

Handbook of Polymers in Electronics Depending on the level of integration, biosensors can be subdivided into four generations (Figure 10.3). In first-generation biosensors (Figure 10.3a), the biocatalyst is either bound to or entrapped between the membranes, which in turn are fixed on the surface of the transducer [13,14]. The second-generation biosensors (Figure 10.3b) involve the adsorption or covalent fixation of the biologically active component to the transducer surface and permits the elimination of the semipermeable membrane [18, 19]. These were based either on the potentiometric or amperometric methods of detection. The commercially available kits for estimation of blood glucose are based on amperometry. In the third-generation biosensors (Figure 10.3c), conducting polymers, such as polyaniline, polypyrrole, etc., have been used as the immobilisation matrix [20, 21]. These biosensors are found to be cost effective, easily available and more stable. The fourth-generation biosensors (Figure 10.3d) are based on direct incorporation of the biological elements into the matrices [2223]. The most commonly used matrices are the conducting polymers and silicon. These biosensors are expected to involve the application of the micro-electrode technology for in-vitro as well as in-vivo applications.

Figure 10.3 Different generations of biosensors (a) first-generation (b) secondgeneration (c) third-generation (d) fourth-generation

300

Conducting Polymer-Based Biosensors

10.1.3 Transducers A transducer converts the biochemical signal to an electronic signal. The biochemical transducer or biocomponent gives the biosensor selectivity or specificity. The transducer of an electrical device responds in such a way that a signal can be electronically amplified and displayed. The physical transducers vary from electrochemical, spectroscopic, thermal, piezoelectric and surface acoustic wave technology [24, 25]. The most common electrochemical transducers being utilised are based on amperometric and potentiometric techniques [26-28]. Amperometric biosensors measure the current produced during the oxidation or reduction of a product or reactant usually at a constant applied potential. Such sensors have fast response times and good sensitivity. However, the excellent specificity of the biological component can be compromised by the partial selectivity of the electrode. This lack of specificity requires sample preparation and separation, or some compensation for interfacing signals. Potentiometric biosensors relate electrical potentials to the concentration of analyte by using an ion-selective electrode or a gas-sensing electrode as the physical transducer [29-33]. These are selective, have large dynamic ranges and are non destructive.

10.1.4 Biological Component Biocomponents are typically enzymes, tissues, bacteria, yeast, antibodies/antigens, liposomes, organelles, cell membrane components, etc. [23, 33-35]. Although the biomolecule incorporated within a biosensor possesses an exquisite level of selectivity, it remains a structurally weak component of the system and is vulnerable to extreme conditions such as pH, temperature and ionic strength [36]. Most of the biological molecules have a very short lifetime in solution phase and thus have to be fixed in a suitable matrix. The immobilisation of the biological component decreases its activity, but imparts stability to the biological component against the environmental conditions [37, 38]. For the economic utilisation of the biocatalysts, their immobilisation in a suitable matrix is an important practice in biomedical, industrial and basic enzymology for repetitive and continuous processes. The reaction conditions and the methodology chosen for immobilisation are important in determining the activity of biomolecules. The activity also depends on the surface area, porosity and hydrophilic character of the immobilising matrix. Immobilisation of the biological component can be done in a number of ways, depending on the type of the component. Techniques such as physical adsorption, crosslinking, gel entrapment, covalent coupling and electrochemical entrapment, etc., have been used for this purpose [39-42]. A number of matrices have been used for the immobilisation of enzymes on various matrices, such as polymeric films, membranes, gels, carbon and silica, etc. [43-46]. To overcome the problems of slow electron transfer reactions of biological molecules at ordinary electrodes,

301

Handbook of Polymers in Electronics various modified electrodes have been designed to facilitate direct coupling of biological redox reaction to the electrodes [40, 47, 48]. These include electrodes modified by deposition of polymer species and electrodes based on conducting salts or conducting polymers [49-51].

10.1.5 Importance of Conducting Polymers to Biosensors Conducting polymers have recently evolved as an important research area with diverse scientific problems of fundamental investigations and a potential for commercial exploitation [52-55]. This important development has implications in interdisciplinary research. Conducting polymers contain π-electrons in the backbone and these are responsible for their unusual electronic properties such as electrical conductivity, low energy optical transition, low ionisation potential and high electron affinity. The electrically conducting polymers contain extended π-conjugated systems with single and double bonds alternating along the polymer chains. The higher values of the electrical conductivity obtained in such organic polymers have led to their being called ‘synthetic metals’. The application of conducting polymers in the field of analytical chemistry and biosensing devices has been reviewed by various researchers [43, 48, 56-58]. Conducting polymers have attracted much interest as a suitable matrix for the entrapment of enzymes [59, 60]. The techniques for incorporating enzymes into electrodepositable conducting polymeric films permit the localisation of biologically active molecules on electrodes of any size or geometry and are particularly appropriate for the fabrication of multianalyte amperometric microbiosensors [61]. Electrically conducting polymers have considerable flexibility in the available chemical structure, which can be modified as required. By chemical modeling and synthesis, it is possible to modulate the required electronic and mechanical properties. Moreover, the polymer itself can be modified to bind protein molecules [62-64]. Another advantage offered by conducting polymers is that the electrochemical synthesis allows the direct deposition of the polymer on the electrode surface, while simultaneously trapping the protein molecules [65, 66]. It is thus possible to control the spatial distribution of the immobilised enzymes and the film thickness, and to modulate the enzyme activity by changing the state of the polymer. The development of any kind of technology in this field heavily depends on the understanding of the interaction at the molecular level with the biologically active protein, either as a simple composite or through chemical grafting. For efficient relay of electrons from the surface of an electrode to the enzyme active site, the concept of ‘electrical wiring’ has been reported [67, 68]. Conducting polymers are likely to provide a 3-D electrically conducting structure for this purpose. Conducting polymers are also known to be compatible with biological molecules in neutral aqueous solutions. They can be reversibly doped and undoped electrochemically, processes which are accompanied by significant changes in conductivity and spectroscopic properties of the films that can be used as a signal for the biochemical reaction [69, 70]. The electronic conductivity of conducting polymers changes over several orders of magnitude in response to changes in pH and redox potential of their environment [71].

302

Conducting Polymer-Based Biosensors Conducting polymers have the ability to efficiently transfer the electric charge produced by the biochemical reaction to the electronic circuit [72, 73]. Moreover, conducting polymers can be deposited over defined areas of the electrodes. This unique property of conducting polymers along with the possibility of entraping enzymes during electrochemical polymerisation has been exploited for the fabrication of amperometric biosensors [74-79]. Besides this, conducting polymers exhibit ion-exchange and size-exclusion properties due to which they are highly sensitive and specific towards the substrate [43, 80-82]. Numerous published papers conclude that conducting polymers can be used as a medium for the

Figure 10.4 Modes of electron transfer in a conducting polymer-based (CP) amperometric biosensor: (a) enzyme and mediator immobilised on conducting polymer, (b) enzyme linked with conducting polymer through mediator, (c) enzyme directly linked to conducting polymers without any mediator

303

Handbook of Polymers in Electronics immobilisation of enzymes [11, 43, 83, 84]. They provide good detectability and fast response as the redox reaction of the substrate, catalysed by an appropriate enzyme, takes place in the bulk of the polymer layer. A schematic of the different pathways suggested for electron transfer in the conducting polymer-based amperometric biosensors is given in Figure 10.4. This chapter provides an overview of the importance of conducting polymers and their application to biosensors. In this context many new conducting polymers have been synthesised and studied. These include PPV, PT, PPy, PFu, PANI and PVCZ.

10.2 Preparation of Electrodes 10.2.1 Synthesis of Conducting Polymers Various methods have been reported in the literature for the synthesis of conducting polymers [85-87]. However, the most widely used method is oxidative coupling, which can be either chemical or electrochemical. The electrochemical technique has recently gained popularity due to the ease with which it is possible to produce homogenous and coherent films, to achieve uniform doping and to control the thickness of the film [88]. The methods of electrochemical polymerisation of conducting polymers generally employed are (a) constant current, or galvanostatic, (b) constant potential, or potentiostatic, or (c) potential scanning/ cycling or sweeping [89]. The charge consumption accompanying the rate of polymer formation is linearly dependent on the time and is independent of concentration of monomer. Generally, potentiostatic conditions are recommended to obtain these films; galvanostatic conditions are employed to obtain thick films. Electrodes should be chosen carefully so that they are not oxidised during the electrochemical oxidation process. Metals used successfully as anodes are chromium, gold, nickel, palladium, platinum and titanium. Platinum plate and glass coated with a conductive ITO layer are the two most popular electrodes. Metals such as aluminium, indium, iron and silver are unsuitable for polymerisation of PPy and PT as they will get oxidised before the polymerisation occurs. Polymer films have also been grown on a number of semiconducting materials such as ndoped silicon [90], gallium arsenide and cadmium sulphide [91] and the semimetal, graphite [92]. Since Diaz and co-workers reported the electrochemical synthesis of PPy [93], many conducting polymers, such as PANI, polyindole, PT, polyazulene, PVCZ, polypyrene, PFu, etc., have been reported in the literature [54, 74, 94]. Polyaniline can be electrochemically grown in the form of a blue-green powdery pigment on a platinum anode during electrolysis of a solution of aniline in sulphuric acid [95-97]. This powdery product was known as ‘aniline black’. Genies and co-workers have presented a historical survey of PANI [94]. Many researchers have presented detailed reports on the electropolymerisation mechanism, properties and applications of PANI. Details relating to the various methods for the synthesis of conducting polymers have been given in Chapter 13. 304

Conducting Polymer-Based Biosensors

10.2.2 Conduction Mechanism in Conducting Polymers On doping, conducting polymers show enhanced electrical conductivity by several orders of magnitude [98]. To explain the electronic phenomenon in these polymeric systems, a new concept involving solitons, polarons and bipolarons has been used [99]. These are structural deformations produced as a consequence of doping. These defects are delocalised over 4-5 monomeric units and are mobile, resulting in the enhanced conductivity. The conduction in PA having two-fold degeneracy has been understood in terms of the motion of charged solitons [100]. However, for highly conducting PPy and PT, which are energetically non degenerate, the conduction occurs via spinless positive charge carriers termed as bipolarons [101]. At lower concentrations, a polaron is envisaged which has a positive charge at one end and a free electron at the other end. The coalescence of polarons may lead to the formation of bipolarons with increasing charge density. Conductivity in conducting polymers is influenced by a variety of factors including polaron length, the conjugation length and the overall chain length and by the charge transfer to adjacent molecules [102].

10.3 Immobilisation of Biomolecules/Enzymes Stable immobilisation of macromolecular biomolecules on conducting microsurfaces with complete retention of their biological recognition properties is a crucial problem for the commercial development of miniaturised biosensors. Various conducting polymers have been utilised for immobilisation of enzymes at an electrode surface including PPy [7476, 103-106], polyindole [79], PANI [77, 107, 108], poly (N-methylpyrrole) [65] and copolymers of N-substituted pyrroles [34].

10.3.1 Methods of Immobilisation Most of the conventional procedures for biomolecule immobilisation (such as crosslinking, covalent binding and entrapment in gels or a membrane) suffer from low reproducibility and a poor spatially controlled deposition. A few biosensors based on insulating electropolymerised films of polyphenol, poly(o-phenylenediamine) and overoxidised polypyrrole have been reported. Malitesta and co-workers [109], Bartlett and Caruana [110], Groom and Luong [111] and Ramanathan and co-workers [112] have studied the changes in the dielectric constant in polypyrrole immobilised glucose oxidase films as a function of glucose concentration. Among the conducting polymers, PPy and its derivatives play a leading role due to their versatile applicability and the wide variety of molecular (redox) species which may be covalently linked to a pyrrole group. Many recent reports have been published on the immobilisation of biological reagents [3, 113-115]. The different methods which can be utilised for immobilisation of the enzyme/biomolecule on conducting polymers are shown in Figure 10.5.

305

Handbook of Polymers in Electronics

Figure 10.5 Different methods of enzyme immobilisation on conducting polymers (a) physical adsorption (b) crosslinking by bifunctional reagents (c) covalent bonding to the matrix (d) entrapment within the polymer matrix (e) immobilisation in a conducting polymer polyvinyl carbazole (PVCZ) / stearic acid (StA) monolayer using Langmuir-Blodgett film technique

10.3.1.1 Physical Adsorption A number of biomolecules have been physically immobilised on conducting polymers [66, 112, 116-119]. This is the simplest method of enzyme immobilisation. Since the binding forces involved are hydrogen bonds, van der Waals forces, etc., porous conducting polymer surfaces are most commonly used. The pre-adsorption of an enzyme monolayer prior to the electrodeposition of the polymer, [120] and two-step enzyme adsorption on the bare electrode surface and then on PPy film [121] have also been investigated.

306

Conducting Polymer-Based Biosensors

10.3.1.2 Crosslinking The crosslinking of the biomolecules via bifunctional reagents, e.g., glutaraldehyde and bovine serum albumin, has been utilised to stabilise the biomolecules. In this case, a crosslinked network of enzyme is formed resulting in the formation of a bigger molecule on the polymer surface [122, 123]. Enzymes have also been immobilised on the surface of electrodeposited polymer by applying the bovine serum albumin and glutaraldehyde procedure followed by the electrodeposition of the polymers [1, 124, 125]. This method has certain limitations since it may cause a drastic loss in the activity of the biomolecules.

10.3.1.3 Covalent Bonding The biomolecules are attached directly to the matrix by chemical/covalent linkage, which is not reversed by pH or changes in ionic strength. Carbodiimide coupling to form peptide bonds has been extensively used for the covalent coupling of the enzyme with polymers [126-128]. Since chemical modification is involved, this method results in the drastic loss of activity of the enzymes/biomolecules. Covalently attached redox biomolecules on polymers have been utilised as highly electron transfer mediators in flavin adenine dinucleotide (FAD) centres of oxidases [73].

10.3.1.4 Electrochemical Entrapment A number of enzymes may be incorporated into conducting polymer films during electrochemical deposition on appropriate electrodes. This is a mild procedure which does not affect the activity of the biomolecules. However, it creates a diffusion barrier to the transport of substrate/products in and out of the matrix. Conducting polymers provide a facile means for ensuring proximity between the active site of an enzyme and the conducting surface of the electrode with considerable potential for biosensor construction. This method provides a facile and controllable method for the deposition of biologically active molecules to defined areas on the electrodes. For electrochemical entrapment, a solution of the electropolymerisable monomer and the aqueous buffer containing the enzyme is used for the deposition process. The polymer film can be deposited electrochemically using a potentiostatic or galvanostatic method. It has been established that glucose oxidase can be successfully entrapped in PPy films [74, 75] and the morphology of the film alters to a more fibrillar form. Polymerisation based on the electrochemical oxidation of a given monomer from a solution containing the enzyme is the simplest method of enzyme immobilisation in a polymer at the working electrode surface and results in the formation of conducting or non conducting

307

Handbook of Polymers in Electronics polymer layers containing entrapped enzyme molecules. The electropolymerisation is often done in aqueous solution at a pH close to neutral in order to immobilise an enzyme without loss of its activity [1, 66, 74, 75, 107]. During electropolymerisation, mediators can be immobilised along with an enzyme [107, 129, 130] or can be conjugated with the monomer before polymerisation [131]. This well-controlled procedure of enzyme immobilisation is of great significance in the fabrication of microsensors [107, 124, 125, 130, 132] used in the preparation of multilayer devices [130, 133, 134] and multienzyme biosensors [125, 135-138]. Many theoretical models have been proposed with the electrochemical entrapment of enzyme for the evaluation of the role of the thickness of polymeric layers, the enzyme location, the enzyme loading, etc., on the functioning of a biosensor [65, 139, 140, 141]. Foulds and Lowe [75], Umana and Waller [74] and Bartlett and Whitaker [77] have described the entrapment of glucose oxidase within the growing network of conducting PPy. This method is inconsistent with the application of an appropriate potential to the working electrode soaked in aqueous solution containing the enzyme and monomer molecules.

10.3.1.5 Immobilisation by the Langmuir-Blodgett Technique It has been observed that the surface of the conducting polymer plays an important role in the effective immobilisation of the desired enzyme. The Langmuir-Blodgett (LB) technique can be successfully applied to deposit a desired monolayer with the desired orientation of the biomolecules/enzymes [142-145]. Ramanathan and co-workers [146] have utilised the polyemeraldine base LB films for the immobilisation of GOD. These films have been shown to function as amperometric glucose biosensors and have a linear range from 5 to 50 mM. LB films of PT immobilised with GOD and urease have also been prepared for application to respective biosensors [147, 148]. The electrochemical entrapment stabilises the enzyme in a polymer layer [149] by protecting against environmental humidity and biological contaminants. However, covalent binding of an enzyme to a suitably modified polymer with the formation of peptide bonds after carbodiimide activation has widely been utilised [126-128, 150]. It has also been observed that the original biocatalyst sensitivity remained unchanged when the electrode surface was renewed after five months of controlled storage [151]. Artificial mediators such as ferrocene carboxylic acid or quinones have been utilised to reoxidise the enzymes in conducting polymers [76, 103-104, 152]. The most popular ways to immobilise enzymes on conducting polymers are either to entrap the enzyme within the growing polymeric films [56, 59, 153, 154] or to use a two-step procedure based on the formation of a functionalised conducting polymer film followed by the covalent binding of the enzyme at the functional groups at the polymer surface [125, 155, 156].

308

Conducting Polymer-Based Biosensors

10.3.2 Advantages of Immobilisation Biosensors require highly active enzymes/biomolecules; therefore, the immobilisation methods must be chosen in such a way that they can achieve a high sensitivity and functional stability. This is important for economic reasons also. The measurable activity gives an idea about the biocatalytic efficiency of an immobilised enzyme. The rate of substrate conversion should rise linearly with enzyme concentration. The measured reaction rates depend not only on the substrate concentration and the kinetic constants Km (Michaelis Menten constant) and Vmax (maximum velocity of the reaction) but also on the immobilisation effects. The following effects have been observed [157] due to the immobilisation process: •

A decrease in the affinity of the enzyme to the substrate (increase in Km) is observed due to conformational changes of the enzyme on immobilisation.

•

A partial/complete inactivation of a part of the enzyme molecule may occur (decrease in Vmax).

•

Ionic, hydrophobic or other interactions between the enzyme and the matrix (microenvironmental effects) may result in changed Km and Vmax values. These effects are reversible and are caused by variations in the dissociation equilibria of charged groups of the active centre.

•

A non uniform distribution of substrate/product between the enzyme matrix and the surrounding solution affects the measured (approximate) kinetic constants.

•

In biosensors the biocatalyst and the signal transducer are spatially combined, i.e., the enzyme reaction proceeds in a layer separated from the measuring solution. In such a system, the substrate and products are transformed by diffusion. Thus diffusion and the enzyme reaction cannot proceed independently of one another; they are coupled in a complex manner [158].

It has been predicted [159] that, in the future, the immobilisation techniques compatible with microelectronic mass production processes will be utilised for the fabrication of amperometric enzyme electrodes.

10.4 Characterisation of Enzyme Electrodes 10.4.1 Determination of Enzyme Activity During the course of immobilisation, some portion of an enzyme gets inactivated. Therefore, it becomes essential to determine the portion of enzyme remaining active

309

Handbook of Polymers in Electronics after immobilisation. The determination of the initial rate of product formation or substrate consumption in the measuring cell using the sensor gives the measure of the enzyme activity taking part in the measuring process. The following formula is used to calculate the apparent enzyme activity (U, units per cm2) of the enzyme electrode: U = AV / ∈st

(10.1)

where A is the difference in the absorbance at a particular wavelength before and after incubation, V is the total volume of assay, ∈ is a millimolar extinction coefficient, s is the surface area of the film and t is the incubation time. One unit can be defined as the amount of enzyme, which converts 1 μM of reactant into product per minute.

10.4.2 Effect of pH It has been observed that the enzymes are active over a limited range of pH and most of the enzymes have a definite optimum pH. The optimum pH is obtained due to the effect on the affinity of the enzyme with the substrate; stability may be irreversibly destroyed on either side of the optimum. The pH profile in the linear measuring range of a biosensor is a diffusion controlled phenomenon and, therefore, is less sharp than those of the respective enzymes in solution [159] and it shifts towards higher pH in solution due to the decrease in local pH [160]. Usually, the enzymes in dilute solution exhibit a bell-shaped pH activity curve. If the immobilised enzyme is studied in essentially buffer-free conditions, considerable change in pH activity profile may be observed. These changes will depend partially on the intrinsic catalytic activity of the enzyme and on the substrate concentration, the reaction rate and the rate of H+ liberation for acid releasing reactions. Reactions catalysed by GOD will be low and the internal pH of the matrix will be a bit different from that of the bulk solution. With relatively high substrate concentrations, the reaction rate will be high enough to cause a marked difference between the pH of the internal matrix and the bulk phase. For such a case, the change in pH of a microenvironment of the enzyme will also affect the reaction rate and the increase in substrate concentration. Thus if the bulk phase pH is higher than the optimum pH of the enzyme, these factors will act synergistically (positive cooperation) or if the bulk phase pH is less than the pH optimum, they will act antagonistically (negative cooperation). A schematic representation of the pH/activity profile of an enzyme (urease) in dilute solution and in the immobilised state in PVCZ films is shown in Figure 10.6.

310

Conducting Polymer-Based Biosensors

Figure 10.6 Activity profile of an enzyme (urease) in (a) native state (■) and (b) after immobilisation in the polymer matrix (PVCZ) at different pH in phosphate buffer (●)

10.4.3 Effect of Temperature The overall enzyme-catalysed reaction succeeds in three steps: formation of the enzymesubstrate complex, conversion of the complex to an enzyme-product complex, and dissociation of the enzyme-product complex to products and free enzyme. There are at least 18 thermodynamic parameters for the forward reaction, i.e., the heat, free energy and entropy of activation, and the heat, free energy and entropy of the process for each of these stages. The increase in the enzyme activity with temperature follows the Arrhenius relation: log k = log A - Ea / 2.303 RT

(10.2)

where k is the reaction rate constant, A is a numerical constant, Ea is the activation energy, R is the thermodynamic gas constant and T is the temperature (in K). The slope of the Arrhenius plot (log k against 1/T) is given by Slope = Ea / 2.303 R

(10.3)

311

Handbook of Polymers in Electronics It has been observed that the rate of enzyme reaction rises with temperature up to a certain maximum above which, thermal inactivation of the enzyme takes place. The inactivation of enzymes by heat is due to the denaturation of the enzyme protein. The effect of the instability of the enzyme, free and in the immobilised state, can be studied by exposing the enzyme to thermal treatment for a defined period prior to measuring its activity at a temperature at which it is stable. Chaubey and co-workers obtained 40 °C as the critical temperature of PPy-polyvinyl sulfonate films immobilised with crosslinked lactate dehydrogenase [123]. The activation energies below and above the critical temperature were found to be 93.3 and 22.4 kJ/mole, respectively.

10.4.4 Effect of Storage Time A biosensor should have excellent stability, reproducibility and a long-term storage time. Biosensors with an electron transfer mechanism based on shuttling of the free diffusing artificial mediators between the enzyme and electrode surface have poor long-term stability due to continuous loss of mediators [163]. In order to increase the long-term stability, it is essential to utilise a new electron transfer pathway between the enzyme and the electrode surface [164]. In this context, conducting polymer based biosensors have been reported to undergo direct electron transfer between the enzyme and the electrode surface [165, 166].

Figure 10.7 Effect of storage time on the response of a conducting polyaniline based lactate biosensor: ■ LOD in solution phase and ■ LOD in immobilised state 312

Conducting Polymer-Based Biosensors It has been observed that enzymes are highly unstable in the solution phase and lose their activity rapidly with time. In the immobilised state, the enzyme is observed to lose its activity very slowly. The immobilisation of enzymes on conducting polymers provides increased stability but decreased electron transfer in the medium resulting in the reduced response of the biosensor. The response of the enzyme lactate oxidase (LOD) to the analyte in solution and in the immobilised state as an enzyme electrode after storing at 4-10 °C is shown in Figure 10.7.

10.4.5 Response Measurements Use of electropolymerised conducting films in the development of small sensing devices is found to be important because electropolymerisation allows control over the thickness and spatial location of electrode modification. These interfacing electroactive materials have inherent oxidation-reduction abilities. These materials are important for direct and rapid electron transfer at the electrode surface [73, 119, 161-163]. Physical and chemical properties can be modified by appropriate polarisation and doping with counter ions. When a biologically active material interacts with an analyte, a physicochemical change takes place that is converted into an electrical output signal using a suitable transducer. Based on various types of transducers, biosensors may be classified into optical, calorimetric, piezoelectric and electrochemical biosensors. Cooper and Hall have reported the electrochemical response of a GOD loaded PANI film [84]. Mu and co-workers have studied the bioelectrochemical response of the PANI-uricase electrode [162]. The factors affecting the overall response of amperometric biosensors involve current limiting processes. There is a diffusion of substrate from the bulk solution to the outer surface of the membrane, enzyme diffusion of the product through the enzyme layer, and quiescence of the solution (if the enzyme layer is thin). For thick enzyme layers and slower diffusion of substrate through the enzyme layer compared to external mass transfer, simultaneous limitation of higher enzymatic reaction and diffusion must be considered. The electrochemical synthesis of conducting polymers allows the direct deposition of the polymer on the electrode surface, while simultaneously trapping the problem molecule. It is thus possible to control the spatial distribution of the immobilised enzyme and the film thickness, and to modulate the enzyme activity by changing the state of the polymer.

10.5 Types of Biosensors Based on the type of transducer used, biosensors can be classified as optical, electrochemical (amperometric or potentiometric), piezoelectric or thermal. Among all these the electrochemical biosensors are most widely used. 313

Handbook of Polymers in Electronics

10.5.1 Optical Biosensors Optical biosensors are based on the measurement of light absorbed or emitted as a consequence of a biochemical reaction. In such biosensors, light waves are guided by means of optical fibres to suitable detectors. Such biosensors have been used for the detection of pH, O2 and CO2. Gerard and co-workers have optically measured the pyruvate concentration (Figure 10.8) by immobilising lactate dehydrogenase (LDH) onto a PANI electrode [163]. These electrodes are stable for about 15 days and have a response time of about 90 seconds. Chaubey and coworkers have immobilised LDH on polypyrrole-polyvinyl sulfonate (PPy-PVS) electrodes on ITO plates and have estimated the lactate concentration by optical measurements [123]. Singhal and co-workers immobilised glucose oxidase on polyhexyl thiophene by the LB technique and fabricated an optical glucose sensor using UV-visible spectroscopy [147]. Urease and glutamate dehydrogenase (GLDH) have been co-immobilised in electrochemically prepared PPy-PVS films for application in a urea biosensor [66]. Quantification of urea by kinetic methods was carried out by measuring the oxidation of NADH in the presence of glutamate dehydrogenase and urease at 350 nm: Urea + H2O → NH4+ + 2HCO3–

(10.4)

α-ketoglutarate + NH4+ + NADH → L-glutamate + NAD+ + H2O

(10.5)

Figure 10.8 Calibration graph of an optical pyruvate biosensor based on polyaniline

314

Conducting Polymer-Based Biosensors

10.5.2 Electrochemical Biosensors The interaction between the enzyme and substrate produces an electrochemical signal that can be detected by an electrochemical detector. These are based on mediated or unmediated electrochemistry.

10.5.2.1 Conductometric Biosensors The conductometric biosensors measure the changes in the conductance of the biological component arising between a pair of metal electrodes. Contractor and co-workers constructed biosensors for glucose/GOD, urea/urease, neutral lipid/lipase and haemoglobin/pepsin by monitoring the change in the electronic conductivity, which is a result of a change in redox potential and/or the pH of the microenvironment in the polymer matrix [167]. The response of a urea conductometric biosensor is shown in Figure 10.9. Ramanathan and co-workers studied the application of polyaniline LB films as glucose biosensors [146]. They have also investigated the dielectric spectroscopic measurements of a PPy/glucose biosensor. Conductivity biosensors based on conducting polymers have been developed for penicillin [168].

Figure 10.9 Response curve of a conductometric urea biosensor

10.5.2.2 Potentiometric Sensors Potentiometry is a rarely used detection method employed in biosensors, with enzymes immobilised in an electrodeposited polymer layer, although certain advantages over 315

Handbook of Polymers in Electronics amperometric detection for a PPy-based electrode with immobilised GOD have been demonstrated [84]. Potentiometric biosensors with conducting polymers can be produced, using the pH sensitivity of polymers [169]. The sensitivity of PPy to NH3 was used to produce such biosensors [170, 171]. LB films of poly(N-vinyl carbazole) have been utilised for the fabrication of a potentiometric urea biosensor (Figure 10.10). Conducting polypyrrole molecular interfaces have also been used to modulate the biological function of enzymes and living cells at the electrode surface by the adjustment of electrode potentials. Often, additionally obtained discrimination of electrochemical interfaces, by using conducting polymers as matrices for enzymes, allows the use of such biosensors for analysis of natural samples, e.g., flow injection determination of lactate in whole blood. Apart from the advantage of a well-defined formation of the polymer layer, this methodology has some difficulties associated with a significant chemical and electrochemical activity of the polymer matrix due to the similarity of these materials towards ion-exchange processes and redox equilibrium. Karyakin and co-workers have described a potentiometric pH response of electrodes modified with PANI for application as a glucose biosensor [172]. These electrodes exhibit a fully reversible potentiometric response of about 90 mV/pH in the range of pH 3-9. This biosensor has a response of about 5-6 minutes and has been shown to be useful

Figure 10.10 Potentiometric response of PVCZ/urease electrodes with different urea concentration

316

Conducting Polymer-Based Biosensors for estimation of glucose in the range of 0.1-30 mM. A creatinine electrode was fabricated by co-immobilisation of creatininase, creatinase and sarcosine oxidase in a PPy matrix [173]. Difficulties concerning the immobilisation of enzymes in the electrodeposited polymer layer or the mechanism of the entrapment and dynamics effects on a number of biosensors have been widely discussed. However, no conclusive reports on the comparison of different polymer matrices for immobilisation of the same enzyme are currently available.

10.5.2.3 Amperometric Biosensors Amperometric biosensors measure the current produced during the oxidation or reduction of a product or reactant usually at a constant applied potential. The most important factor affecting the functioning of amperometric biosensors pertains to the electron transfer between catalytic molecules, usually oxidase or dehydrogenase, and the electrode surface, this transfer most often involving mediation or a conducting polymer. Although the role of the electrodeposited conducting polymer films is not fully understood and explained, various polymers used for the enzyme immobilisation may significantly affect the response of biosensors sensitive to given species. Much work has been done in recent years on the application of electrochemically grown conducting polymer layers in amperometric biosensors. A theoretical model describing the operation of a conducting polymer based biosensor has been reported. In such a model, two different mechanisms have been suggested. An enzymatic reaction may take place at the polymer/solution interface without any diffusion of the substrate molecules into the polymer, or diffusion of substrate into the bulk of the polymer layer due to its porosity may occur [174]. It is shown that the quinone-hydroquinone reaction at a PPy modified electrode is accompanied by the diffusion of the substrate into a porous polymer layer. A general opinion is that no electron transfer occurs between the working electrode surface and the enzyme molecules entrapped in the polymer layer. However, de Taxis du Poet and co-workers have suggested a direct electron transfer between the redox centre of GOD immobilised in poly(N-methylpyrrole) and a gold electrode surface [73]. The immobilisation of GOD via manipulation of pore size in PPy to enhance enzyme loading, has also been reported [175]. Verghese and co-workers have conducted similar experiments in PANI/ GOD films [176]. Biosensors based on conducting polymers have found potential applications in healthcare, veterinary medicine, environmental monitoring, immunosensing, etc.

317

Handbook of Polymers in Electronics

10.6 Biosensors for Healthcare 10.6.1 Glucose Biosensor Selective determination of blood glucose is of utmost importance for the screening and treatment of diabetes. About 4% of the world population is presently suffering from diabetes. Normal blood glucose concentration varies from 4.2-5.2 mM. A number of glucose biosensors are available in the market for blood glucose estimation, but none of these is based on conducting polymers. Numerous reports have been published for the immobilisation of GOD in conducting polymers for glucose estimation [75, 110, 175]. GOD catalyses the oxidation of glucose in the following sequence of reactions: D-Glucose + GOD (FAD) → Gluconolactone + GOD (FADH2)

(10.6)

GOD (FADH2) + O2 → GOD (FAD) + H2O2

(10.7)

Conducting PPy is one of the first conducting polymers to be used for the fabrication of a glucose biosensor [75, 84]. These studies have indicated that up to 125 mU of GOD activity can be incorporated. Umana and co-workers have fabricated a PPy-based glucose biosensor for detection of hydrogen peroxide [74]. The response current of these electrodes was found to be diminished after about one week indicating gradual leaching of the GOD from the PPy matrix. This has been attributed to the electrostatic repulsion between the cationically charged enzyme and the polycationic PPy as well as the porosity of the PPy. It has been investigated if the polymers containing para- and ortho-quinone groups as electron-transfer relay systems for oxido-reductases can effectively catalyse the electrooxidation of glucose [177, 178]. Ramanathan and co-workers have immobilised GOD after manipulation of the pore size in PPy to improve the loading of the enzyme [175]. The incorporation of large size dopant anions such as paratoluene sulfonate (PTS) and ferricyanide into PPy films during electropolymerisation and subsequent exchange of these ions with a smaller ion like chloride in solution, ion self-exchange or electrochemical switching renders the PPy more porous and the immobilisation of GOD more facile. The response of such polypyrrole-GOD electrodes is shown in Figure 10.11. L-B films of polyemeraldine base have been deposited on ITO glass substrates by injecting a solution of 60% CHCl3 in N-methyl phenazine containing 100 μl of GOD. The activity of GOD immobilised in these polyemeraldine base films determined by the o-dianisidine procedure has been found to be 5 IU cm-2 [146]. A few reports are available wherein conducting polymers have been incorporated in materials such as graphite [179] and carbon paste [180]. However, these have not yet been commercialised. A glucose biosensor utilising covalently coupled GOD to poly(o-amino benzoic acid) (PAB, a carboxy group functionalised polyaniline) has been described [42].

318

Conducting Polymer-Based Biosensors

Figure 10.11 Amperometric response of GOD immobilised PPy based glucose biosensors for PPy-ferricyanide unexchanged (▲) and PPy-ferricyanide exchanged with Cl– (■)

Amperometric response measurements conducted via unmediated and mediated (with ferrocene carboxylic acid and tetrathiafulvaline) reoxidation of GOD have shown that glucose can be detected over a wide range of concentrations. An enzyme-conducting polymermediator model provides for better charge transport in a biosensor. The screen-printed electrodes consisted of two silver tracks with an active working area of 4 mm x 2 mm (Figure 10.12), printed on a polyvinyl chloride (PVC) substrate. On one of the tracks, the Ag/AgCl reference electrode was screen-printed and on the other electrode, the PAB/GOD

Figure 10.12 A two electrode set-up used for an amperometric glucose biosensor based on poly(o-amino benzoic acid). Each electrode consists of two screen-printed tracks of Ag paste. On one of the tracks, Ag/AgCl (reference electrode) was printed and on the other, a PAB/GOD complex was adsorbed (working electrode). The non working area was covered by the PMMA mask.

319

Handbook of Polymers in Electronics complex was adsorbed. The area of the electrode exposed to the solution was always kept at 8 mm2 by masking it with a PMMA layer. A nylon membrane shielded the working area of the screen-printed PAB/GOD electrode. The optimal response obtained at pH 5.5 at 300 K lies in the range from 1 mM to 40 mM of glucose. The operational stability of the PABbased glucose biosensor has been experimentally determined to be about six days.

10.6.2 Urea Biosensor Urea is the most important end product of protein degradation in the body. Its concentration in blood depends on the protein catabolism and nutritive protein intake and is regulated by renal excretion. Thus the estimation of blood urea nitrogen is important in the assessment of kidney failure. The normal level of urea ranges from 3.6 mM to 8.9 mM. All enzymatic methods for urea determination are based on the principle of urea hydrolysis by urease: (NH2)2CO + 3H2O → 2 NH4+ + OH- + HCO3–

(10.8)

Most biosensors described in the literature for the determination of urea are potentiometric based on NH4+ or HCO3- sensitive electrodes [181, 182]. Osaka and co-workers constructed a highly sensitive and rapid flow injection system for urea analysis with a composite film of electropolymerised inactive PPy and a polyion complex [183]. Pandey and co-workers fabricated a urea biosensor based on immobilised urease on the tip of an ammonia gas electrode (diameter: 10 μm) made from a PPy film coated onto a platinum wire [170]. The enzymatic response was achieved in the wide range of 0.001-0.05 M with a stability of more than 32 days. Cho and co-workers [184] developed a procedure for urea determination by crosslinking urease onto PANI-Nafion composite electrodes, which could sense the ammonium ions efficiently. Such a urea biosensor has a detection limit of about 0.5 μM and a response time of 40 seconds. Immobilisation of urease and glutamate dehydrogenase enzymes in electrochemically prepared PPy/PVS film has been carried out using physical adsorption and electrochemical entrapment techniques [66]. Hydrolysis of urea is catalysed by the enzyme as in Reaction 10.4. The ammonium ion released in the first reaction is coupled with α-ketoglutarate in the presence of GLDH; NADH acts as a cofactor (Reaction 10.5). A kinetic method was utilised for the quantification of urea (Figure 10.13) by measuring the oxidation of NADH in the presence of glutamate dehydrogenase and urease at 350 nm. The leaching was found to be 5%-10% in the case of entrapped enzymes over a period of one hour in solution. However, in the case of adsorbed enzymes, this was found to be 15%-20%. The half-life of this electrode was found to be about 30 days for adsorbed enzymes and 35 days for the entrapped enzymes.

320

Conducting Polymer-Based Biosensors

Figure 10.13 Lineweaver-Burke plot for the urea biosensor based on PPy/PVS films adsorbed with urease

10.6.3 Lactate Biosensor L-lactate is the intermediate product of carbohydrate metabolism. The rapid, accurate and selective assay of L-lactate and pyruvate is necessary in clinical biology where it is important in the growth of certain cells and in fermenters. L-lactate also plays an important role in food industries engaged in fermentation of wine and dairy products. The lactate concentration level in blood indicates various pathological states, including shock, respiratory insufficiencies and heart diseases, and finds immense importance in neonatology and sports medicine. Different lactate oxidising enzymes use different co-substrates and, therefore, a variety of electrochemical indicator reactions in biosensors can be utilised. Most of the lactate biosensors are based on enzymes like lactate oxidase (LOD) and lactate dehydrogenase (LDH). A needle-type lactate biosensor has been recently developed by Yang and coworkers who fabricated poly (1,3-phenylenediamine) electrodes immobilised with LOD for continuous intravascular lactate monitoring [185]. In the enzyme electrodes based on LDH, the biochemical reaction has been coupled to the electrode via NADH oxidation, either directly [119, 123, 163], or by using mediators [186] or additional enzymes [119]. This may lead to a shift of the unfavourable reaction equilibrium by partial trapping of the reduced cofactor. Direct oxidation of NADH requires potentials of more than 0.4 V;

321

Handbook of Polymers in Electronics other interfering substances get oxidised at this potential. In order to avoid this, mediators or pretreatment of the electrode have been utilised. Potassium ferricyanide has been extensively used for spontaneous oxidation of NADH in potentiometric LDH electrodes [186]. Yao and co-workers observed that the co-immobilisation of LOD and LDH in poly(1,2-diaminobenzene) films show a highly sensitive detection of L-lactate due to amplification of the signal by substrate recycling [187]. LOD and LDH have been co-immobilised on electrochemically prepared PANI films by physical adsorption [119]. Regeneration of L-lactate by substrate recycling (Figure 10.14) is shown to provide an amplification of the sensor response. The linearity for the LOD/LDH immobilised electrodes has been found to be from 0.1 mM to 1 mM of lactate with a detection limit of 5x10-5 M. These enzyme electrodes are stable for about 3 weeks at 4-10 °C implying that these can be used for estimation of L-lactate in cells and fermentation. A comparison of the results obtained with the PANI electrodes immobilised with either LOD or LDH, indicates that the co-immobilisation provides the possibility to detect L-lactate at lower concentrations. Chaubey and co-workers have reported the results relating to the electrochemical preparation and characterisation of the PPy-PVS/LDH electrodes. NAD+ is reduced to NADH as a result of its reaction with L-lactate in the presence of LDH [123]. NADH in turn is oxidised at the working electrode releasing two electrons at 0.2 V (bias voltage). The PPy-PVS film acts as an electron acceptor and hence gets reduced. An attempt has also been made to experimentally investigate the effect of film thickness, pH, temperature and enzyme concentration on the activity of the PPy-PVS/LDH electrodes.

Figure 10.14 Substrate recycling of lactate/pyruvate in PANI/LOD-LDH electrodes

322

Conducting Polymer-Based Biosensors

10.6.4 Cholesterol Biosensor Cholesterol and its fatty acid esters are important components of nerve and brain cells and are precursors of the biological materials such as bile acids and steroid hormones. Accumulation of cholesterol in blood leads to fatal diseases such as arteriosclerosis, cerebral thrombosis and coronary diseases. Kajiya and co-workers immobilised cholesterol oxidase (ChOx) and ferrocene carboxylate in PPy electrochemically to describe the sensitivity of the resulting films [188]. The response was proportional to the cholesterol concentration up to 0.05 mM. It has been demonstrated that ferrocene attached to polymer chains can mediate electron transfer from horseradish peroxidase (HRP) to a conventional electrode surface [189]. In the case of immobilised HRP and ChOx the sensor yields 0.35 μA to 10 mM of cholesterol whereas 3 μA was obtained in the case of free ChOx. It was therefore suggested that the sensor response is limited by the interfacial transport or reaction rate of H2O2. The sensor response was also found to be independent of the applied potential between –100 and 100 mV. Trettnak and co-workers prepared a cholesterol biosensor by electropolymerising pyrrole in solutions of 0.1 M phosphate buffer containing NaClO4 (10 mM) and 15 to 55 IU/ ml of ChOx at 0.8 V vs SCE (standard calomel electrode) [190]. The sensor shows high reproducibility in the range 0 to 250 μM with the detection limit of about 5 μM of cholesterol. The amperometric cholesterol biosensor based on PPy/ferrocene carboxylic acid electrodes is shown in Figure 10.15. Kumar and co-workers have used

Figure 10.15 Response curve for the polypyrrole-ferrocene carboxylic acid based cholesterol biosensor

323

Handbook of Polymers in Electronics dodecylbenzene sulfonate (DBS) doped PPy films for immobilisation of ChOx by physical adsorption [191]. These ChOx/DBS-PPy/ITO electrodes exhibit a response time of about 60 seconds, linearity from 2 to 8 mM of cholesterol and stability of about three months at 4 °C. The cyclic voltammetry studies were performed in 0.1 M phosphate buffer (pH 7.0) using enzyme immobilised DBS-PPy/ITO films with and without ferricyanide ion mediator as a working electrode, Ag/AgCl as reference electrode and Pt as the counter electrode. Cholesterol + ChOx(ox) → Cholestenone + ChOx(red) ChOx (red) + Fe3+ → ChOx + Fe2+

(10.9) (10.10)

0.4V

Fe2+ → Fe3+ + e-

(10.11)

The oxidation peak at 0.75 V (without ferricyanide) was observed to shift cathodically by about 350 mV (observed at 0.4 V in the presence of ferricyanide). In such a case the oxidation of other elements at 0.75 V, which could result in an increase in the resulting current, can be avoided.

10.6.5 DNA Biosensor DNA biosensors are considered to be important for the clinical diagnosis of inherited diseases, rapid detection of pathogenic infections and screening of complementary DNA colonies required in the field of molecular biology research. Present methods of genetic analysis are dependent upon the ability to detect specific DNA sequences in a heterogeneous mixture. More recently, DNA integrated electroactive polymers (thin films or two-dimensional L-B monolayers) have created a novel class of intelligent materials, which possess superior intelligent material properties of self assembly, self multiplication, self repair, self degradation and self diagnosis, etc. [192, 193]. Very few reports related to the interaction of DNA with conducting polymers are available [194, 195]. Livache and co-workers have described one-step electrodeposition of a PPy film functionalised by a covalently linked oligonucleotide [196]. Earlier they had described electrochemical copolymerisation of pyrrole and oligonucleotides having a pyrrole moiety. Introduction of a pyrrole moiety to the 5´ end of the oligonucleotide was explained by phosphoramidite chemistry [197]. Most of these rely on measuring changes in the peak current of a redox active marker that preferentially binds to the duplex formed in the hybridisation [198]. Marazza and co-workers have described detection by differential pulse voltammetry and

324

Conducting Polymer-Based Biosensors chronopotentiometric stripping analysis by the use of an electroactive indicator, daunomycin hydrochloride, which interacts with the double-stranded DNA [199]. Doping of DNA probes within electropolymerised PPy films and monitoring the current changes provoked by the hybridisation event is another possibility, as described by Wang and co-workers [200]. Immobilisation of DNA on a conducting polymer matrix facilitates detection of a signal (amperometric/potentiometric) generated as a result of interaction of proteins or drugs with DNA. Possibilities of detecting the signal generated as a result of hybridisation of DNA strands can also be explored. Gambhir and co-workers have recently described adsorption characteristics of DNA on electrochemically generated PPy-PVS films as a function of pH. Adsorption on PPy doped with an anion proceeds by anion exchange by the adsorbed molecule. DNA possesses a fixed negative charge due to the presence of PO4-. Therefore, PVS displacement favours strong binding due to energetic interactions with PPy [201]. Characterisation of adsorbed DNA on the PPy-PVS films has been carried out by UV-visible spectroscopy, FTIR spectroscopy and cyclic voltammetric studies (Figure 10.16).

Figure 10.16 Adsorption of DNA onto conducting PPy-PVS films as a function of pH.6.0 (◆), 6.5 (■), 7.0 (▲), 8.0 (● ) on ITO glass plates

325

Handbook of Polymers in Electronics

10.7 Immunosensor Immunosensors are small, portable instruments for analysis of complex fluids and are designed for ease of use by untrained personnel, for rapid assay with sensitivity comparable to that of enzyme linked immunosorbant assay (ELISA). During the past decade, a number of methods for analysing microorganisms, viruses, pesticides and industrial pollutants have been developed. Immunosensors are based on immunochemical principles that can automatically carry out estimation of the desired parameter. Barnett and co-workers have detected thaumatin using antibody containing PPy electrodes [202]. In the recent past, immunoassays have relied on complex indirect enzyme methods in which the resultant product of the enzyme immuno reaction can be measured [203]. Recently, antibodies have been raised against the conducting polymer, carbazole as a hapten, which may react to modulate the polymer electrochemistry. It has been observed by cyclic voltammetry that the reaction of the antiserum influences the polymer matrix electrochemistry by an amperometric response. This system has been shown to form the basis of a direct sensor for immunoassay [204].

10.8 Biosensors for Environmental Monitoring Environmental monitoring of hazardous pollutants has become important to regulatory agencies and the general public. A number of biosensors for environmental pollutants such as sulfite [205], nitrite [206], phenolic compounds [207], etc., have been fabricated. Cao and co-workers have reported amperometric gas sensors that can be used for the estimation of gases such as CO, NO2, and O2 [208]. Responsive chemoresistors for gas analysis have been reported where the organic polymer modified electrodes were processed by pattern recognition [209, 210]. Yadong and co-workers have performed studies on the ammonia gas sensitive properties and sensing mechanism of PPy [211]. The redox enzyme polyphenol oxidase has been utilised to oxidise phenolic compounds in a two-step reaction to the electrochemically reducible quinone [212]. Pesticide contamination in developing and developed countries is rapidly growing due to the increasing and indiscriminate use of these organic chemicals in agricultural fields. Since pesticides remaining in the food crops are transferred through the food chain system, the residues have been cumulatively building up in the human body to alarming proportions necessitating drastic control measures. Accurate and efficient monitoring of these residues is therefore of vital concern to the survival of our society.

10.9 Conclusions Marked progress has been made in the last decade towards the application of conducting polymers to biosensors. In this context, the researchers working in this field of research 326

Conducting Polymer-Based Biosensors have extensively investigated the immobilisation of enzymes in conducting polymer films. The self-assembly of redox enzymes on the surface of an electrode followed by the electrochemical deposition of conducting polymers can be used as a molecular interface to enable the redox enzyme to communicate electronically with the electrodes. Thus conducting polymers are advantageous in the fact that they combine the role of the enzyme entrapment matrix and chemicophysical transducer, resulting in substantial miniaturisation and reduced response time. The structure of the polymer affects the sensitivity and the detection limit of the biosensor obtained, as well as the ion-exchange and size-exclusion properties leading to improved detection. Conducting polymers are highly effective in minimising the effect of interferents when used for entrapment of redox enzymes and as protection against electrode fouling by non specific adsorption of high molecular compounds from natural samples. It has also been revealed that the conducting polymer matrix serves as an electron shuttling medium between the enzyme and the electrode, and is thus capable of functioning as an electron transfer mediator. Simultaneous integration of enzymes and mediators may improve the electron transfer pathway from the active site of enzyme to the electrode. Functionalised conducting polymers will play an important role in the integration of biological molecules for development of improved biosensing devices.

Acknowledgements The authors are thankful to the Director, NPL, for his interest and constant encouragement. A. Chaubey gratefully acknowledges CSIR, India, for the award of a Senior Research Fellowship.

References 1.

S.V. Sasso, R.J. Pierce and R. Walla and A.M. Yacynych, Analytical Chemistry, 1990, 62, 1111.

2.

D. Diamond, Electroanalysis, 1993, 5, 795.

3.

G. Bidan, Sensors and Actuators B, 1992, 6, 45.

4.

D.J. Clarke, M.R. Calder, R.J.G. Carr, B.C. Blake-Coleman and S.C. Moody, Public Health Library Service Microbiology Digest, 1984, 1, 3, 25.

5.

A.M. Usmani, Polymer News, 1995, 20, 170.

6.

H.H. Weetall, Biosensors and Bioelectronics, 1999, 14, 237. 327

Handbook of Polymers in Electronics 7.

A. Boyle, E.M. Genies and M. Lapkowski, Synthetic Metals, 1989, 28, 769.

8.

Biosensors: Fundamentals and Applications, Eds., A.P.F. Turner, I. Karube and G.S. Wilson, Oxford University Press, Oxford, UK, 1987.

9.

L.J. Blum and P.R. Coulet in Biosensors: Principles and Applications, Marcel Dekker, New York, NY, USA, 1987.

10. S.F. White and A.P.F. Turner in Encyclopedia of Bioprocess Technology: Fermentation, Biocatalysis and Bioseparations, Eds., M.C. Flickinger and S.W. Drew, Wiley, New York, NY, USA, 1997. 11. W. Schuhmann in Immobilised Biomolecules in Analysis: A Practical Approach, Eds., T. Cass and F. Ligler, Oxford University Press, Oxford, 1998, 187. 12. E. Kress-Rogers, Handbook of Biosensors and Electronic Noses: Medicine, Food and Environment, CRC Press, Boca Raton, FL, USA, 1996. 13. L.C. Clark, Jr., and C. Lyons, Annals of the New York Academy of Sciences, 1962, 102, 29. 14. S.J. Updike and G.P. Hicks, Nature, 1967, 214, 986. 15. A.P.F. Turner, Annali Di Chimica, 1997, 87, 255. 16. H.H. Weetall, Biosensors and Bioelectronics, Guest Editorial, 1996, 11, i. 17. D. Griffiths and G. Hall, Trends in Biotechnology, 1993, 11, 122. 18. R.K. Kobos, J.W. Eveleigh and R. Arentzen, Trends in Biotechnology, 1989, 7, 101. 19. A. Miyabayashi and D.J. O’Shannessy, Biotechnology and Applied Biochemistry, 1989, 11, 101. 20. C.G.J. Koopal, A.A.C.M. Bos and R.J.M. Nolte, Sensors and Actuators B, 1994, 18-19, 166. 21. T.W. Lewis, G.G. Wallace and M.R. Smyth, Analyst, 1999, 124, 213. 22. M. Situmorang, D.B. Hibbert, J.J. Gooding and D. Barnett, Analyst, 1999, 124, 1775. 23. O. Sadik and G.G. Wallace, Analytica Chimica Acta, 1993, 279, 209.

328

Conducting Polymer-Based Biosensors 24. J. Svorc, S. Miertu, J. Katrlik and M. Stred’ansk, Analytical Chemistry, 1997, 69, 2086. 25. G. Urban, A. Jachimowicz, F. Kohl, H. Kuttner, F. Olcaytug, H. Kamper, F. Pittner, E. Mann-Buxbaum, T. Schalkhammer, O. Prohaska and M. Schonauer, Sensors and Actuators A, 1990, 21-23, 650. 26. H. Habermuller, M. Mosbach, W. Schuhmann and J. Fresenius, Analytical Chemistry, 2000, 366, 560. 27. S.B. Adeloju, S.J. Shaw and G.G. Wallace, Analytica Chimica Acta, 1993, 281, 611. 28. A. Mulchandani, P. Mulchandani, S. Chauhan, I. Kaneva and W. Chen, Electroanalysis, 1998, 10, 733. 29. J. Bobacka, Analytical Chemistry, 1999, 71, 4932. 30. A. Lewenstam, J. Bobacka and A. Ivaska, Journal of Electroanalytical Chemistry, 1994, 368, 23. 31. K. Domansky, D.L. Baldwin, J.W. Girate, T.B. Hall, J. Li, M. Josowicz and J. Janata, Analytical Chemistry, 1988, 70, 473. 32. J. Anzai, J. Hashimoto, T. Osa and T. Matsuo, Analytical Sciences, 1988, 4, 247. 33. I. Karube, K. Hiramoto, M. Kawarai and K. Sode, Membrane, 1989, 14, 311. 34

N.C. Foulds and C.R. Lowe, Analytical Chemistry, 1988, 60, 2473.

35. R.D. Schmidt, Biosensors International Workshop, Ed., R.D. Schmidt, 10, VCH, Weinheim, Federal Republic of Germany, 1987, Preface. 36. A.A. Karyakin, M. Vuki, L.V. Lukachova, E.E. Karyakina, A.V. Orlov, G.P. Karpachova and J. Wang, Analytical Chemistry, 1999, 71, 2534. 37. W. Schuhmann, C. Lehn, H.L. Schmidt and B. Grundig, Sensors and Actuators B, 1992, 7, 393. 38. G.A. Evtugyn, H.C. Budnikov and E.B. Nikolskaya, Talanta, 1998, 46, 465. 39. G.G. Guilbault, Analytical Uses of Immobilized Enzymes, Marcel Dekker Inc., New York, USA, 1984, 78.

329

Handbook of Polymers in Electronics 40. P.W. Stoecker and A.M. Yacynych, Selective Electrode Review, 1990, 12, 137. 41. A. Kumar, Rajesh, S.K. Grover and B.D. Malhotra, Analytica Chimica Acta, 2000, 414, 43. 42. K. Ramanathan, S.S. Pandey, R. Kumar, A. Gulati, A.S.N. Murthy and B.D. Malhotra, Journal of Applied Polymer Science, 2000, 78, 662. 43. M. Trojanowicz and T.K. vel Krawczyk, Mikrochimica Acta, 1995, 121, 167. 44. P.C Pandey, Bulletin of Electrochemistry, 1992, 8, 5, 212. 45. T. Ikeda, H. Hamada, K. Miki and M. Senda, Agricultural and Biological Chemistry, 1985, 49, 541. 46. J. Gun and O. Lev, Analytica Chimica Acta, 1995, 336, 95. 47. T. Lotzbeyer, W. Schuhmann and H.L. Schmidt, Sensors and Actuators B, 1996, 33, 50. 48. M. Situmorang, J.J. Gooding, D.B. Hibbert and D. Barnett, Biosensors and Bioelectronics, 1998, 13, 953. 49. S.A.M. Marzouk, V.V. Cosofret, R.P. Buck, H. Yang and S.S.M. Hassan, Analytical Chemistry, 1997, 69, 2646. 50. W.J. Albery, P.N. Bartlett and D.H. Craston, Journal of Electroanalytical Chemistry, 1985, 194, 223. 51. P.N. Bartlett and P.R. Birkin, Synthetic Metals, 1993, 61, 15. 52. W. Schuhmann in Methods in Biotechnology, Volume 6, Eds., A. Mulchandani and K. Rogers, Humana Press, Totowa, NJ, USA, 1998, 143. 53. V. Saxena and R. Prakash, Polymer Bulletin, 2000, 45, 267. 54. B.D. Malhotra, N. Kumar and S. Chandra, Progress in Polymer Science, 1986, 12, 179. 55. T.A. Skotheim, Electroresponsive Molecular and Polymeric Systems, Volume 2, Marcel Dekker, New York, NY, USA, 1992. 56. W. Schuhmann, Mikrochimica Acta, 1995, 121, 1. 57. S.A. Wring and J.P. Hart, Analyst, 1992, 117, 1215.

330

Conducting Polymer-Based Biosensors 58. A. Guiseppi-Elie, G.G. Wallace and T. Matsue in Handbook of Conducting Polymers, Second Edition, Eds., T. Skotheim, R. Elsenbaumer and J.R. Reynolds, Marcel Dekker, New York, NY, USA, 1997, 963. 59. P.N. Bartlett and J.M. Cooper, Journal of Electroanalytical Chemistry, 1993, 362, 1. 60. W.J. Sung and Y.H. Bae, Analytical Chemistry, 2000, 72, 2177. 61. P.R. Unwin and A.J. Bard, Analytical Chemistry, 1994, 64, 113. 62. W. Lu, H. Zhao and G.G. Wallace, Analytica Chimica Acta, 1995, 315, 27. 63. A. Mulchandani and C.L. Wang, Electroanalysis, 1996, 8, 414. 64. M. Situmorang, D.B. Hibbert and J.J. Gooding, Electroanalysis, 2000, 12, 111. 65. P.N. Bartlett and R.G. Whitaker, Journal of Electroanalytical Chemistry, 1987, 224, 37. 66. A. Gambhir, M. Gerard, A. Mulchandani and B.D. Malhotra, Applied Biochemistry and Biotechnology, 2001, 96, 249. 67. A. Heller, Accounts of Chemical Research, 1990, 23, 128. 68. B.A. Gregg and A. Heller, Analytical Chemistry, 1990, 62, 258. 69. M.S. Wrighton, Science, 1986, 231, 32. 70. G.P. Kittlesen, H.S. White and M.S. Wrighton, Journal of the American Chemical Society, 1984, 106, 7389. 71. E.W. Paul, A.J. Ricco and M.S. Wrighton, Journal of Physical Chemistry, 1985, 89, 1441. 72. M.E.G. Lyons, W. Breen and J. Cassidy, Journal of the Chemical Society, Faraday Transactions I, 1991, 87, 115. 73. P. de Taxis du Poet, S. Miyamoto, T. Murakami, J. Kimura and I. Karube, Analytica Chimica Acta, 1990, 235, 255. 74. M. Umana and J. Waller, Analytical Chemistry, 1986, 58, 2979. 75. N.C. Foulds and C.R. Lowe, Journal of the Chemical Society, Faraday Transactions I, 1986, 82, 1259.

331

Handbook of Polymers in Electronics 76. C. Iwakura, Y. Kajiya and H. Yoneyama, Journal of the Chemical Society, Chemical Communications, 1988, 1019. 77. P.N. Bartlett and R.G. Whitaker, Biosensors, 1987/88, 3, 359. 78. M. Shaolin, X. Huaiguo and Q. Bidong, Journal of Electroanalytical Chemistry, 1991, 304, 7. 79. P.N. Barlett in Biosensors: A Practical Approach, Ed., A.E.G. Cass, IRL press, Oxford, UK, 1990, 47. 80. A. Mirmohseni, W.E. Price and G.G. Wallace, Polymer Gels & Networks, 1993, 1, 61. 81. P.R. Teasdale and G.G. Wallace, Analyst, 1993, 118, 329. 82. W.E. Price, G.G. Wallace, H. Zhao and J. Membrane, Science, 1994, 87, 47. 83. C. Kranz, H. Wohlschlager, H.L. Schmidt and W. Schuhmann, Electroanalysis, 1988, 10, 546. 84. J.C. Cooper and E.A.H. Hall, Biosensors and Bioelectronics, 1992, 7, 473. 85. S.S. Pandey, S. Annapoorni and B.D. Malhotra, Macromolecules, 1993, 26, 3190. 86. P. Kovacic and A. Kyriakis, Journal of the American Chemical Society, 1963, 85, 454. 87. J.H. Edwards and W.J. Feast, Polymer Communications, 1980, 21, 595. 88. M.V. Deshpande and D.P. Amalnerkar, Progress in Polymer Science, 1993, 18, 623. 89. E.M. Genies and C. Tsintavis, Journal of Electroanalytical Chemistry, 1985, 195, 109. 90. R. Noufi, A.J. Frank and A.J. Nozik, Journal of the American Chemical Society, 1981, 103, 1849. 91. R. Noufi, D. Tenchana and L.F. Warren, Journal of Electroanalytical Chemistry, 1980, 127, 2310. 92. R.A. Bull, F.R. Fan and A.J. Bard, Journal of Electroanalytical Chemistry, 1983, 130, 1636.

332

Conducting Polymer-Based Biosensors 93. A.F. Diaz, K.K. Kanazawa and G.P. Gardini, Journal of the Chemical Society, Chemical Communications, 1979, 635. 94. E.M. Genies, A. Boyle, M. Lapkowski and C. Tsintavis, Synthetic Metals, 1990, 36, 139. 95. D.M. Mohilner, R.N. Adams and W.J. Argersinger, Journal of the American Chemical Society, 1962, 84, 3618. 96. J. Bacon and R.N. Adams, Journal of the American Chemical Society, 1968, 90, 6596. 97. A.G. MacDiarmid, J.C. Chiang, A.F. Richter and A.J Epstein, Synthetic Metals, 1987, 18, 285. 98. W.S. Huang, A.G. MacDiarmid and A.J. Epstein, Journal of the Chemical Society, Chemical Communications, 1987, 1784. 99. A.J. Heeger in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, NY, USA, 1986, 729. 100. J. Simon and J.J. Andre, Molecular Semiconductors, Springer Verlag, Berlin, Germany, 1985, 166. 101. W.P. Su, in Handbook of Conducting Polymers, Volume 2, Ed., T.A. Skotheim, Marcel Dekker, New York, NY, USA, 1986, 757. 102. Ed., J.I. Kroschwitz, Electrical & Electronic Properties of Polymers, John Wiley & Sons, Somerset, NJ, USA, 1988. 103. P. Janda and J. Weber, Journal of Electroanalytical Chemistry, 1991, 300, 119. 104. T. Matsue, N. Kasai, M. Narumi, M. Nishizawa, H. Yamada and I. Uchida, Journal of Electroanalytical Chemistry, 1991, 300, 111. 105. D. Belanger, E. Brassard and G. Fortier, Analytica Chimica Acta, 1990, 228, 311. 106. M. Hiller, C. Kranz, J. Huber, P. Bauerle and W. Schuhmann, Advanced Materials, 1996, 8, 219. 107. H. Shinohara, T. Chiba and M. Aizawa, Sensors and Actuators B, 1988, 13, 79. 108. M. Shaolin, X. Hiaiguo and Q. Bidong, Journal of Electroanalytical Chemistry, 1991, 304, 7.

333

Handbook of Polymers in Electronics 109. C. Malitesta, F. Palmisano, L. Torsi and P.G. Zambonin, Analytical Chemistry, 1990, 62, 2735. 110. P.N. Bartlett and D.J. Caruana, Analyst, 1992, 117, 1287. 111. C.A. Groom and J.H.T. Luong, Analytical Letters, 1993, 26, 1383. 112. K. Ramanathan, M.K. Ram, M.M. Verghese and B.D. Malhotra, Journal of Applied Polymer Science, 1996, 60, 2309. 113. R. John, M. Spencer, G.G. Wallace and M.R. Smyth, Analytica Chimica Acta, 1991, 249, 381. 114. M. Shaolin, Journal of Electroanalytical Chemistry, 1994, 370, 135. 115. I. Moser, T. Schalkhammer, F. Pittner and G. Urban, Biosensors and Bioelectronics, 1997, 12, 729. 116. K. Ramanathan, R. Mehrotra, B. Jayaram, A.S.N. Murthy and B.D. Malhotra, Anal. Lett., 1996, 29, 1477. 117. K. Ramanathan, N.S. Sundaresan and B.D. Malhotra, Electroanalysis, 1995, 7, 579. 118. M.M. Verghese, K.Ramanathan, S.M. Ashraf and B.D. Malhotra, Journal of Applied Polymer Science, 1988, 70, 1447. 119. A. Chaubey, K.K. Pande, V.S. Singh and B.D. Malhotra, Analytica Chimica Acta, 2000, 407, 97. 120. G.F. Khan, E. Kobatake, H. Shinohara, Y. Ikariyama and M. Aizawa, Analytical Chemistry, 1992, 64, 1254. 121. D.R. Yaniv, L. McCormick, J. Wang and N. Naser, Journal of Electroanalytical Chemistry, 1991, 314, 353. 122. Y. Xu, G.G. Guilbault and S.S. Kuan, Enzyme and Microbial Technology, 1990 12, 104. 123. A. Chaubey, M. Gerard, R. Singhal, V.S. Singh and B.D. Malhotra, Electrochimica Acta, 2000, 46, 723. 124. D.J. Strike, N.F. de Rooij and M. Koudelka-Hep, Sensors and Actuators B, 1993, 13, 61.

334

Conducting Polymer-Based Biosensors 125. B.F.Y. Yon-Hin, R.S. Sethi and C.R. Lowe, Sensors and Actuators B, 1990, 1, 550. 126. B.F.Y. Yon-Hin, M. Smolander, T. Crompton and C.R. Lowe, Analytical Chemistry, 1993, 65, 2067. 127. W. Schuhmann, R. Lammert, B. Uhe and H.L. Schmidt, Sensors and Actuators B, 1990, 1, 537. 128. T. Schalkhammer, E. Mann. Buxbaum, F. Pittner and G. Urban, Sensors and Actuators B, 1991, 4, 273. 129. Y. Kajiya, H. Sugai, C. Iwakura and H. Yoneyama, Denki Kagaku, 1988, 56, 1110. 130. Z. Sun and H. Tachikawa, Analytical Chemistry, 1992, 64, 1112. 131. I. Taniguchi, K. Matsushita, M. Okamoto, J.P. Collin and J.P. Sauvage, Journal of Electroanalytical Chemistry, 1990, 280, 221. 132. Y. Kajiya, H. Sugai, C. Iwakura and H. Yoneyama, Analytical Chemistry, 1991, 63, 49. 133. D.T. Hoa, T.N. Suresh Kumar, N.S. Punekar, R.S. Srinivasa, R. Lal and A.Q. Contractor, Analytical Chemistry, 1992, 64, 2645. 134. L. Coche-Guerente, S. Cosnier, C. Innocent, P. Mailley, J.C. Moutet, R.M. Morelis, B. Leca and P.R. Coulet, Electroanalysis, 1993, 5, 647. 135. T. Tatsuma, T. Watanabe and T. Watanabe, Sensors and Actuators B, 1993, 14, 752. 136. T. Tatsuma, T. Watanabe and T. Watanabe, Journal of Electroanalytical Chemistry, 1993, 356, 245. 137. B.F.Y. Yon Hin and C.R. Lowe, Sensors and Actuators B, 1992, 7, 339. 138. M.G. Garguilo, N. Huynh, A. Proctor, A.C. Michael Chael, Analytical Chemistry, 1993, 65, 523. 139. M. Maschesielf and E. Genies, Journal of Electroanalytical Chemistry, 1993, 358, 35. 140. P. Gross and A. Bergel, Journal of Electroanalytical Chemistry, 1995, 386, 65.

335

Handbook of Polymers in Electronics 141. J.J. Gooding, E.A.H. Hall, D.B. Hibbert, Electroanalysis, 1998, 10, 1130. 142. D.G. Zhu, M.C. Petty, H. Ancelin and J. Yarwood, Thin Solid Films, 1989, 176, 151. 143. T. Dubrovsky, S. Vakula and C. Nicolini, Sensors and Actuators B, 1994, 22, 69. 144. A.P. Girard-Egrot, R.M. Morelis and P.R. Coulet, Thin Solid Films, 1997, 292, 282. 145. M. Yasuzawa, M. Hashimoto, S. Fujii, A. Kunuji and T. Nakaya, Sensors and Actuators B, 2000, 65, 241. 146 K. Ramanathan, M.K. Ram, B.D. Malhotra and A.S.N. Murthy, Materials Science and Engineering C, 1995, 3, 159. 147. R. Singhal, A. Chaubey, M.K. Pandey and B.D. Malhotra, Presented at the 4th National Conference on Solid State Ionics, Mumbai, India, 2000. 148. R. Singhal, A. Gambhir, M.K. Pandey, S. Annapoorni and B.D. Malhotra, Biosensors and Bioelectronics, (submitted) 149. G. Fortier and D. Belanger, Biotechnology and Bioengineering, 1991, 37, 854. 150. S.E. Wolowacz, B.F.Y. Yon Hin and C.R. Lowe, Analytical Chemistry, 1992, 64, 1541. 151. J. Katrlik, A. Pizzariello, V. Mastitube, J. Svore, M. Stredansky and S. Miertus, Analytica Chimica Acta, 1999, 379, 193. 152. J.M. Dicks, S.Hattori, I. Karube A.P.F. Turner and T. Yokozawa, Annales de Biologie Clinique, 1989, 47, 607. 153. C. Iwakura, Y. Kajiya and H. Yoneyama, Journal of the Chemical Society, Chemical Communications, 1988, 1010. 154. B.F.Y. Yon Hin, M. Smolander, T. Crompton and C.R. Lowe, Analytical Chemistry, 1993, 65, 2067. 155. S. Koide and K. Yokoyama, Journal of Electroanalytical Chemistry, 1999, 468, 193. 156. W. Schuhmann, Synthetic Metals, 1991, 41, 429. 157. T. Kobayashi and K. Laidler, Biotechnology and Bioengineering, 1974, 16, 77. 158. P.W. Carr and L.D. Bowers in Immobilised Enzymes in Analytical and Clinical Chemistry, John Wiley, New York, NY, USA, 1980.

336

Conducting Polymer-Based Biosensors 159. W. Schumann in Diagnostic Biosensor Polymers, Eds., A.M. Usmani and N. Akmal, ACS Symposium Series 556, Washington, DC, USA, 1994, 110. 160. F. Scheller, R. Renneberg and F. Schubert in Methods Emzymology, Ed., K. Mosbach, Academic Press, New York, NY, USA, 1988, 29. 161. M.J. Green and H.A.O. Hill, Journal of the Chemical Society, Faraday Transactions I, 1986, 82, 1237. 162. S. Mu, J. Kan and J. Zhou, Journal of Electroanalytical Chemistry, 1992, 334, 121. 163. M. Gerard, K. Ramanathan, A. Chaubey and B.D. Malhotra, Electroanalysis, 1999, 11, 450. 164. H.L. Schmidt, Sensors and Actuators B, 1993, 13-14, 366. 165. C.G.J. Koopal, B. De Ruiter and R.J.M. Nolte, Journal of the Chemical Society, Chemical Communications, 1991, 1691. 166. C.G.J. Koopal, M.C. Feiters and R.J.M. Nolte, Synthetic Metals, 1992, 51, 397. 167. A.Q. Contractor, T.N. Sureshkumar, R. Narayanan, S. Sukeerthi, R. Lal and R.S. Srinivasa, Electrochimica Acta, 1994, 39, 1321. 168. M. Nishizawa, T. Matsue and I. Uchida, Analytical Chemistry, 1992, 64, 2642. 169. N.R. Ratcliffe, Analytica Chimica Acta, 1990, 239, 257. 170. P.C. Pandey and A.P. Mishra, Analyst, 1988, 113, 329. 171. M. Trojanowicz, W. Matuszewski, B. Szczepanczyk and A. Lewenstam in Uses of Immobilised Biological Compounds, Eds., G.G. Guilbault and M. Mascini, Kluwer, Dordrecht, Germany, 1993, 141. 172. A.A. Karyakin, L.V. Lukachova, E.E. Karyakina, A.V. Orlov and G. P. Karpachova, Analytical Communications, 1999, 36, 153. 173. H. Yamato, M. Ohwa and W. Wernet, Analytical Chemistry, 1995, 67, 2776. 174. M.E.G. Lyons, C.H. Lyons, C. Fitzgerald and T. Bannon, Analyst, 1993, 118, 361. 175. K. Ramanathan, N.S. Sundaresan and B.D. Malhotra, Electroanalysis, 1995, 7, 579. 176. M.M. Verghese, K. Ramanathan, S.M. Ashraf and B.D. Malhotra, Journal of Applied Polymer Science, 1998, 170, 1447.

337

Handbook of Polymers in Electronics 177. N.K. Cenas, A.K. Pocius, and J.J. Kulys, Biotechnology and Bioengineering, 1984, 12, 583. 178. J.J. Kulys, Biosensors, 1986, 2, 3. 179. A.A. Karyakin, A.K. Strakhova, E.E. Karyakina, S.D. Varfolomeyev and A.K. Yatsimirsky, Biotechnology and Bioengineering, 1993, 32, 35. 180. S. Alegret, F. Cespedes, E.M. Fabregas, D. Martorell and A. Morales, Biosensors and Bioelectronics, 1996, 11, 35. 181. R.Koncki, A. Radomska and S. Glab, Talanta, 2000, 52, 13. 182. M. Jurkiewicz, M. del Valle, S. Alegret and E. Martincz-Fabregas, Analytica Chimica Acta, 1996, 327, 243. 183. T. Osaka, S. Komaba, Y. Fujino, T. Matsuda and I. Satoh, Journal of Electroanalytical Chemistry, 1999, 146, 615. 184. W.J. Cho and H.J. Huang, Analytical Chemistry, 1998, 70, 3946. 185. Q. Yang, P. Atanasov and E. Wilkins, Biosensors and Bioelectronics, 1999, 14, 203. 186. H. Durliat, C. Causser and M. Comtat, Analytica Chimica Acta, 1990, 231, 309. 187. T. Yao, M. Satomura and T. Nakahara, Electroanalysis, 1994, 7, 395. 188. Y.Kajiya, R. Tsuda and H. Yoneyama, Journal of Electroanalytical Chemistry, 1991, 301, 155. 189. L. Boguslavsky, P.D. Hale, L. Geng, T.A. Skotheim and H.S. Lee, Solid State Ionics, 1993, 60, 189. 190. W. Trettnak, I. Lionti and M. Mascini, Electroanalysis, 1993, 5, 753. 191. A. Kumar, Rajesh, A. Chaubey, S.K. Grover and B.D. Malhotra, Journal of Applied Polymer Science, 2001, 82, 3486. 192. S. Minehan, K.A. Marx and S.K. Tripathy, Macromolecules, 1994, 27, 777. 193. F. Garnier, H. KorriYoussoufi, P. Srivastava, B. Mandrand and T. Delair, Synthetic Metals, 1999, 100, 89. 194. N. Saoudi, N. Jammul, M.L. Abel, M.M. Chehimi and G. Dodin, Synthetic Metals, 1997, 87, 97.

338

Conducting Polymer-Based Biosensors 195. K.A. Marx, J.O. Lim, D.S. Minehan, R. Pande, M.N. Kamath, S.K. Tripathi and D.L. Kaplan, Journal of Intelligent Material Systems and Structures, 1994, 5, 447. 196. T. Livache, A. Roget, E. Dejean, C. Barthet, G. Bidan and R. Teoule, Synthetic Metals, 1995, 71, 2143. 197. T. Livache, A. Roget, E. Dejean, C. Bathet, G. Bidan and R. Teoule, Nucleic Acid Research, 1992, 22, 2915. 198. K. Hoshimoto, K. Ito and Y. Ishimori, Analytica Chimica Acta, 1994, 286, 219. 199. G. Marrazza, I. Chianella and M. Mascini, Biosensors and Bioelectronics, 1999, 14, 43. 200. J. Wang, M. Jiang, F. Aantonio and B. Mukherjee, Analytica Chimica Acta, 1999, 402, 7. 201. A. Gambhir, M. Gerard, S.K. Jain and B.D. Malhotra, Applied Biochemistry and Biotechnology, 2001, 96, 303. 202. O.A. Sadik, M.J. John, G.G. Wallace and D. Barnett, Analyst, 1994, 119, 1997. 203. Electrochemical Sensors in Immunological Analysis, Volume I, Ed., T.T. Ngo, Plenum Publishing Corporation, USA, 1987, 103. 204. R.A. Porter, Journal of Immunoassay, 2000, 21, 51. 205. S.B. Adeloju, S.J. Shaw and G.G. Wallace, Electroanalysis, 1994, 6, 865. 206. Q.Wu, G.D. Storrier, F. Pariente, Y. Wang, J.P. Shapleigh and H.D. Abruna, Analytical Chemistry, 1997, 69, 4856. 207. D. Barnett, D.G. Laing, S. Skopec and G.G. Wallace, Analytical Letters, 1994, 27, 2417. 208. Z. Cao, W.J. Buttner and J.R. Stetter, Electoanalysis, 1992, 4, 253. 209. T.C. Pearce, J.W. Gardner, S. Friel, P.N. Bartlett and N. Blair, Analyst, 1993, 118, 371. 210. J.M. Slater, J. Paynter and E.J. Watt, Analyst, 1993, 118, 379. 211. J. Yadong, W. Tao, W. Zhiming, L. Dan, C. Xiangdong and X. Dan, Sensors and Actuators B, 2000, 66, 280. 212. R.Z. Kazandjian and A.M. Klibanov, Journal of the American Chemical Society, 1985, 107, 5448. 339

Handbook of Polymers in Electronics

340

11

Nanoparticle-Dispersed Semiconducting Polymers for Electronics K.S. Narayan

11.1 Introduction A composite, a blend or a bilayer consisting of a conjugated semiconducting polymer and a nanoparticle component combine to form an interesting system due to a variety of features. In these binary systems, profound changes are observed in the structural, optical and electronic properties with the degree of interaction between the two components varying from a weak electrostatic interaction to a strong chemical link. Studies of such hybrid systems add further insight into the individual characteristics of the different components. The distinctiveness of the nanoparticle properties arises from size-dependent fundamental properties such as ionisation potential, melting point, optical fluoroscence and absorbance [1-4]. The typical range of size for observing variation in these properties depends upon the solid being periodic over a nanometre lengthscale. In the case of nanoparticles of semiconductors such as cadmium sulphide, CdS, and cadmium selenide, CdSe, when the particle size is smaller than that of the exciton in the bulk semiconductor the lowest energy optical transition significantly increases due to quantum confinement [5]. By precisely controlling the size and the surface of a nanocrystal, its properties can be tuned. However, due to the large surface to volume ratio, the radiative luminescence from these nanoparticles is significantly reduced through non radiative processes mediated by the surface states [6]. These non radiative losses may be minimised by an improved method of core-shell synthesis [7]. By enclosing a core nanocrystal of one material with a shell of another having a larger bandgap, one can efficiently confine the excitation to the core, eliminating non radiative relaxation pathways and preventing photochemical degradation [7, 8]. In fact, nanocrystals prepared using these routes were demonstrated to be useful as fluorescent probes in biological staining and diagnostics [9]. Compared with conventional fluorophores, the nanocrystals have a narrow, tunable, symmetric emission spectrum and are photochemically stable [9]. These nanocrystal probes are thus complementary to and in some cases better than existing fluorophores. Such nanoparticles when dispersed in a non conductive polymer matrix show improved long-term stability. This is due to the fact that the polymer matrix to some extent prevents interdiffusion of inorganic substituents, which is observed to be a drawback in heterojunction-based inorganic semiconductors.

341

Handbook of Polymers in Electronics Conjugated polymers on the other hand derive their semiconducting properties by having delocalised π-electron bonding along the polymer chain. The π (bonding) and π* (antibonding) orbitals form delocalised valence and conduction wavefunctions, which support mobile charge carriers. The classic model polymers which have been studied extensively from this perspective of electrical transport are polyacetylene, PANI and PPy. Conjugated polymers such as PPV, PT and their derivatives have been studied for their light-emitting and photodetection properties, and can be classified as second-generation polymers. They have recently been shown to be a viable option for large area display devices and image sensors [10-13]. The tunability of the properties can be achieved by attaching pendant groups to the main chain or by control of conjugation length. There has been considerable effort in estimating the different lengthscales and the timescales of the different photo-induced excitations prior to culmination as free carriers in these polymers. This knowledge is especially required to fully exploit the potential and obtain highly efficient photoactive materials [14, 15]. The concept of size-dependent effects has also been investigated in conjugated copolymer heterostructures, exclusively based on polymers without any nanoparticle component [16]. Theoretical studies of polydiacetylene-polyacetylene-polydiacetylene triblock polymer chains led to prediction of results similar to those of multiple quantum wells in inorganic low-dimensional semiconductors [17]. Discrete split-off exciton states as well as localisation of electronic states were predicted. Theoretical studies of various aspects of the electronic and excitonic properties of quasi-one-dimensional semiconducting polymer super lattices, which are periodic copolymers, have been reported [18]. The electronic structure was systematically studied as a function of copolymer composition and block sizes, AmBn. However, the experimental efforts to test these concepts have not shown evidence of such quantum effects. Recent results have been reported on blends of ABA triblock conjugated copolymer, poly(2,5-benzoxazole)-block-poly(benzobisthiazole2-hydroxy-1,4-phenylene)-block-poly(2,5-benzoxazole). The higher energy gap poly(2,5benzoxazole) shows quantum confinement effects at room temperature based on PL emission and excitation, electric field-modulated PL and electric field and picosecond absorption spectroscopy results [16]. This chapter focuses on the blends and multilayers of a variety of nanoparticles and conjugated polymers. However, it must be mentioned that there has been a large amount of research in the last two decades on nanocomposites of conventional polymers [19]. Polymer nanocomposites in this context are generally defined as the combination of a polymer matrix resin and inorganic particles that have at least one dimension, i.e., length, width, or thickness, in the nanometre size range. Typical of this class of materials is the nanocomposite which researchers at Toyota Co., discovered in the 1980s: polyamide 6

342

Nanoparticle-Dispersed Semiconducting Polymers for Electronics (from caprolactam), with dispersed ion-exchanged montmorillonite as the reinforcement. Several benefits of such a nanocomposite that have been identified include: •

Efficient reinforcement with minimal loss of ductility and impact strength,

•

Heat stability and flame resistance,

•

Improved gas barrier properties,

•

Improved abrasion resistance,

•

Reduced shrinkage, and

•

Residual stress.

The shapes of the particles used in nanocomposites can be roughly spherical, fibrillar or platelets, and each shape will result in different properties. For maximum reinforcement, platelets or fibrillar particles would be used, since reinforcement efficiency is related to the aspect ratio (length/diameter, L/d). The most extensive research has been performed with layered silicates, which provide a platelet reinforcement [19]. The blends with conducting polymers go beyond the physical aspects and tap in to the quantum electronic processes. Dispersion of electron acceptor nanoparticles such as TiO2 in these second-generation conducting polymers has proven to be an effective methodology for enhancing photoinduced charge separation in a layered device [20]. Pristine conjugated polymers with absorption in the visible range do not have good photoconducting properties, primarily arising from the constraints in both the yield of free carriers due to the neutral excitation as a dominant species and the low carrier mobility prevailing in these disordered systems. Since this neutral excitation can be dissociated at an interface between the polymer and an electron accepting species, charge separation is often facilitated via inclusion of a high electron affinity substance such as C60 [21, 22]. Device fabrication with composites of conjugated polymers and C60 as the active layer, with efficient photoinduced charge transfer preventing the initial e-h recombination, was a significant advancement in this field [21, 22]. A prerequisite for such an enhancement are materials with high electron affinity with a distribution in the polymer matrix such that the interparticle distance is of the order of the exciton diffusion length. In addition, the charge separation process must be fast enough to compete with the radiative and non radiative decay pathways of the singlet exciton which is in the range of 100 ps - 1 ns [23]. The large surface area to volume ratio of the nanoparticle enhances the probability for these charge separation processes. The first-generation conducting polymers have been used as the supporting matrix in different composites for intercalation of catalytically important nanoparticles so that the catalytic activity can be retained in the composite. A few such hybrid materials have been synthesized by Qi and Pickup, who have incorporated Pt and PtO in a matrix of

343

Handbook of Polymers in Electronics PEDOT as the conducting polymer component. This has been shown to work well as an electrode of a fuel cell and its performance is comparable to the commercial carbon supported catalyst at a higher Pt loading [24-26]. PPy and PANI have the ability to reduce and precipitate some metals from their respective acid solutions. PPy-SiO2 and PANI-SiO2 hybrid nanocomposites offer the advantage of a material with large surface area. These two important characteristics have enabled the above mentioned nanocomposites to be an attractive host for entrapping a substantial amount of metal particles with important catalytic activity and these metal-rich composites can be used for subsequent catalytic application, as postulated recently [27-29]. The nanocomposites of metal nanoparticles in a matrix of a passive non conducting polymer matrix have a percolation threshold for transport when the nanoparticle concentration reaches values high enough to provide conduction along the chain of connected nanoparticles [30]. At the threshold values of the filler concentration, the conductivity of the composite changes by several orders of magnitude analogous to the metal-insulator transition. However, it has been shown that there are subtle size-dependent percolation thresholds, with the nanoparticle systems of sizes 5 nm and 12 nm having much lower threshold values in terms of relative concentration compared to micronsized nanoparticles [31]. Metallic levels of conductivity have been achieved in the past by filling polymers with conductive particles (5-10 μm). However, the loadings required for percolation are large enough to seriously compromise the weight and flexibility advantages of polymers, while intrinsically conductive polymers are not able to reach the necessary levels of conductivity. The decrease in the loading density without sacrificing the conductance behaviour of the composite can be achieved if the surface area of the unit weight of the filler is increased. On the other hand, insertion of insulating nanoparticles (oxides) in an active conducting polymer matrix also leads to various interesting properties. Introduction of these insulating particles has shown to increase conductivity in certain cases, such as PPy-ZrO2, PPy-Fe2O3, and PANI-TiO2. Models based on macroscopic structural rearrangements and packing density have been used to explain these properties. PPy and PANI nanocomposites using SiO2, latex bead and polyethyelene oxide have been fabricated. Intensive efforts are also being made to develop hybrid materials with appreciable magnetic and electrical properties, such as Fe2O3 containing doped PANI and PPy-silica coated magnetic Fe2O3 particles (5-30 nm) [34]. However, at this stage, very few examples are available regarding the mechanism of electrical transport in the nanocomposites.

11.2 Material Preparation Methods A brief summary of the synthesis adopted for nanoparticles, conjugated polymers and composites is discussed in this section. The last decade has seen an explosion of activity in

344

Nanoparticle-Dispersed Semiconducting Polymers for Electronics the preparation of various kinds of nanoparticles. No attempt is made to exhaustively cover the various strategies used for fabricating these structures but a few representative examples are provided. Common techniques for preparing semiconductor nanoparticles include arrested precipitation of colloidal particles from homogenous solution by controlled release of ions or forced hydrolysis in the presence of surfactants [2-4]. For example, controlled mixing of Cd2+ with sulfide ions yields colloidal particles of CdS. Surfactants are often used to stabilise the particles. For metal nanoparticles, chemical reduction of metal ions is a common approach. To narrow the particle size distribution resulting from the homogenous nucleation step, the colloidal suspension is typically fractionated using size-selective precipitation. A homogeneous monodispersion of semiconducting nanocrystallites has been obtained after pyrolysis of organometallic reagents by injection into hot coordinating solvents [35]. This method provides discrete nucleation and permits controlled growth of macroscopic quantities of the nanoparticles [35]. Physical approaches are also used for synthesis of metal oxide nanoparticles. In physical vapour synthesis, a plasma is used to heat a precursor metal. The metal atoms boil off, creating a vapour. A gas is introduced to cool the vapour, which condenses into liquid molecular clusters. As the cooling process continues, the molecular clusters are frozen into solid nanoparticles. The metal atoms in the molecular clusters mix with oxygen atoms, forming metal oxides, such as aluminum oxide, smaller than 100 nm. Another commonly used application-specific method is discrete particle encapsulation (DPE). In this method, selected chemicals are used to form a thin polymeric shell around each nanoparticle providing the characteristics a user needs. Then a second thin-shell coating is added, so the nanoparticle will disperse in the best needed format. This shell contains spacer molecules that prevent the nanoparticles from coming into contact with each other. The result is steric stabilisation for nanoparticles in non liquid solvents and polymers, and electrosteric stabilisation for those needing to disperse in a fluid. Methods for preparing soluble PPV and PT are standardised to a great extent and are commercially available. The issue of structural homogeneity and purity of conjugated polymers becomes particularly important when the target structure is obtained as a difficult to characterise solid. An example which demonstrates these difficulties is the WesslingZimmermann route to synthesise PPV [36, 37], where the elimination of small molecules from a so-called precursor polymer leads to a molecular structure with extended conjugation. Failure to perform this polymer-analogous reaction quantitatively will leave sp3 carbon centres within the chain, which would thus interrupt the π-system. Selectivity is a general concern within chemical synthesis that will, of course, raise questions at different levels of sophistication. Oxidative coupling of 3-alkyl-substituted thiophenes leaves one with an ambiguity since the coupling can occur in different fashions producing sequences such as 2,5′ (head to tail), 2,2′ (head to head) and 5,5′ (tail to tail) [38]. More subtle procedures

345

Handbook of Polymers in Electronics have been devised to regioselectively synthesise head-to-tail polyalkylthiophenes Many such routes for obtaining high purity polymers with regioregularity are now well established. Inorganic nanoparticles can be introduced into the matrix of a host conducting polymer either by some suitable chemical route or by an electrochemical incorporation technique. An example of this route for hybrid nanocomposite materials is to polymerise aniline or pyrrole in the presence of some preformed inorganic particles, using an appropriate soluble oxidant. PPy (colloidal) gold nanocomposites have recently been introduced by Marinakos and others [39]. Although the work starts with template-guided polymerisation, it ultimately provides template-free nanocomposite particles, tubes or wires. Gold nanoparticles were arranged within the pores of an Al2O3 membrane using a vacuum filtration technique. Fe(ClO4)3 is passed through the membrane and encounters rising pyrrole vapour within the membrane. PPy is grown within the pores of the respective film supporting the gold particles and the Au-PPy composite grows inside. The template membrane can then be got rid of by dissolving in a solvent [39]. PPy-coated gold nanoparticles were also synthesised within the microdomain of a diblock copolymer, providing an excellent means of formation of such dispersions [40]. Diblock copolymers, owing to their ability to form microdomains and to associate in solution in the form of micelles, can provide small compartments inside which particles of a finite size can be generated and stabilised [40]. Since this review essentially focuses on optical and electronic properties and not the structural or chemical aspects of the hybrid nanoparticle-polymer composites, the reader is referred to a recent review which exhaustively covers recent efforts from this perspective [41]. A crucial factor in processing films and the hybrid composites is obtaining homogenous solutions before utilising the various options for deposition of these solutions [42]. For example, spin coating of a nanoparticle with poor dispersion properties in a polymer solution can produce a thin film of small, randomly ordered nanoparticle crystallites interspersed with amorphous areas. Introduction of stabilisers has been known to improve the homogeneity of these systems. For example PNVC has been combined with SiO2 to form a nanocomposite in the presence of a polymeric stabiliser, PVP [43]. Key parameters for the coating are concentration, solvent evaporation rates, and the spinning rates. Phase separation always exists for multicomponents but if the lengthscale over which this phase separation occurs is more of the order of the relevant lengthscales of the charge carriers/ excitons diffusion processes, then the hybrid system can be considered to be fairly uniform from the perspective of charge separation strategies.

11.3 Photophysics of Charge Separation Nanoparticle-Polymer Systems The mechanisms for charge generation and separation upon optical excitation and light emission are different for the spherical nanoparticles and essentially linear chain polymers.

346

Nanoparticle-Dispersed Semiconducting Polymers for Electronics Hybrid systems based on these different class of materials reveal interesting electronic and optical properties and add further insight into the individual characteristics of the different components. The photovoltaic properties rely on efficient charge separation upon photoexcitation. Pristine polymers have a problem with large amounts of disorder and low mobilities, in spite of large absorption. Many of the properties of polymer systems will be examined with three different varieties of nanoparticles: •

High dielectric, insulating TiO2,

•

Semiconducting CdS, and

•

Metallic Au.

The various properties of these systems are explored and provide a glimpse of the richness in terms of photophysical and electronic processes. Charge generation and charge transport processes have been extensively studied in pristine PPV and its derivatives [44]. Based on the coincidence of the onsets of the photoconductivity and absorption in MEH-PPV, it was concluded that photoexcitation of PPV leads to a direct generation of mobile charges through an interband π-π* transition [45, 46]. On the other hand, based on a variety of complementary experiments, it is argued that the photocurrent (Iph) originates from secondary processes, where the initial intrachain excitons dissociate to free charges [44]. These two points of view differ in two respects. Whereas the band model [45] assumes a small binding energy for photoexcited electron-hole pairs and immediate charge separation, the exciton model requires larger exciton binding energy and extrinsic charge separation mechanisms [47]. Exciton dissociation could arise from exciton dissociation at defect sites [48], field and thermal ionisation [49], exciton interaction with trapped carriers, or exciton-exciton interaction. The evolution of Iph upon photoexcitation and the subsequent decay is, therefore, a valuable tool for exploring various processes and developing a deeper understanding of phenomena such as electroluminescence in these systems. Extrinsic mechanisms are most likely to affect the relatively long (>nanosecond) transient photoconductivity measurements [50]. This relatively slow decay process is interpreted as being extrinsic due to dissociation of polaron pairs (or interchain excitons) through interaction with oxygenated defects to create positive polarons. The slow component has been modelled in terms of a recombination limited dispersive decay. A typical configuration for the studies of these properties involves an active layer (polymer, polymer-nanoparticle blend) in a sandwich configuration, as shown in Figure 11.1, between a transparent electrode and a metal electrode of suitable work function. EL requires the injection of electrons from one electrode and holes from the other, the capture of oppositely charged carriers (so-called recombination), and the radiative decay of the

347

Handbook of Polymers in Electronics (a)

(b)

Figure 11.1 Typical device measurement for (a) electroluminescence and (b) photocurrent

excited electron–hole state (exciton) produced by this recombination process. The complementary process of charge separation upon photoexcitation is evaluated on the basis of the charge transport and transfer to the electrodes to constitute a photocurrent, which depends on the incident photon rate, the charge generation efficiency, the recombination rate and parameters which control the trap-limited transport in these disordered polymer systems. Routes for exciton and charge transfer processes in these mixed systems are shown in Figure 11.2. In most cases, the polymer acts as the holetransporting medium, with the nanoparticles as a sink for electrons. In addition it is normally required to synchronise the charge generation rate, the transfer rate, the transit time to the electrodes (which is governed by the transport parameters) and the duration for the decay process for the electron acceptors to come down to an uncharged state.

11.3.1 TiO2-Conjugated Polymer Composites Studies of charge separation at the interface between organic molecules and nanocrystals, particularly systems of organic dyes adsorbed on TiO2 nanocrystalline films, as a basis for efficient photovoltaic devices have generated considerable interest [51]. It has been shown that in a nanocrystalline TiO2/PPV composite, excitons photogenerated in the polymer could be dissociated at the interface between the components with the electrons transferred to the nanocrystals [52] (Scheme b in Figure 11.2). Considerable attention has been directed towards dye-sensitised nanocrystalline semiconductor films since O’Regan and Grätzel [51] reported a high-efficiency dye-sensitised solar cell. Nanocrystalline semiconductor films are highly porous, thus having a large internal surface area. Only the first monolayer of adsorbed dye results in efficient electron injection into

348

Nanoparticle-Dispersed Semiconducting Polymers for Electronics

Figure 11.2 Scheme for efficient charge separation in a nanoparticle-polymer composite (adapted from [13]): (a) absorption in the polymer is followed by an electron transfer process to the nanoparticle (electron acceptor), (b) absorption in the polymer is followed by an exciton transfer process to the nanoparticle, followed by the hole transfer to the polymer, (c) absorption in the nanoparticle is followed by hole transfer to the polymer [23]

the semiconductor, but the light-harvesting efficiency of a single dye monolayer is very small. In a mesoporous film consisting of nanometre-sized TiO2 particles, the effective surface area can be enhanced about a thousand-fold theoretically, thus making light absorption efficient even with only a dye monolayer on each particle. Recent work on dye-sensitised solar cells is centred on ruthenium-bipyridine complexes sensitising nanocrystalline TiO2 films, which has proven to be highly efficient in photon-to-electron conversion. However, in order to study the photoelectric conversion mechanism and develop a more efficient dye-sensitised solar cell, it is necessary to probe the sensitisation of other kinds of dyes, such as organic molecules, which are easily modified [53-55]. TiO2 particles have been traditionally interesting to study due to a wide variety of applications. TiO2 occurs as the mineral rutile, anatase, octahedrite, ilmenite, and brookite, and exhibits distinct size effects. Anatase TiO2 is used widely for welding-rod coatings, acid-resistant vitreous enamel, specific paints, etc. Below a critical size, TiO2 clusters can 349

Handbook of Polymers in Electronics absorb the energy of UV light to release electrons and radicals by oxidation. So it is also used to protect against external irradiation and sunlight. The absorbed organic compounds on TiO2 clusters can be decomposed by oxidation due to the presence of a radical released by irradiation. Therefore, such nanoclusters of TiO2 are known as photocatalysts. Because of the unique properties, anatase TiO2 nanoclusters have high potential for applications in diverse areas of environmental purification, such as purification of water and air. TiO2 has also been studied for preparation of composites with conducting polymers, such as poly(3-methylthiophene) supported on TiO2, for solid-state photoelectrochemical devices [56]. Because of the combination of electrical conductivity of PANI and the UV sensitivity of anatase TiO2, such nanomaterials are expected to find applications in electrochromic devices, NLO systems, and photoelectrochemical devices. Another class of application of TiO2 dispersed polymers that has been explored is solidstate polymer laser diodes and photonic crystals. The high refractive index TiO2 particles can scatter the emitted photons in the active polymer medium such that the gain exceeds loss above a critical excitation threshold. Scattering from the randomly distributed high refractive index nanoparticles greatly increases the pathlength traversed by the emitted light. Solid-state lasing has been observed from freestanding films of MEH-PPV and TiO2, with greatly reduced threshold pump powers [57]. The tunable photonic crystals, TPCs, consist of periodic particle arrays in which either the particles or a matrix component is either optically or electrically tunable, thereby enabling tunability for the properties of photonic crystals, particularly the width and position of the photonic band gap. The basic PC will be a three-dimensionally periodic array of spherical nanoparticles, which are structurally similar to those of naturally occurring opals. Conducting organic polymers have been heavily used as the tunable component. TPCs are expected to combine the advantageous properties of conventional PCs with the tunability of such materials as conducting polymers, photoresponsive materials, and ferroelectric materials. A significant property of nanocrystalline TiO2/PPV composites is that the excitons photogenerated in the polymer could be dissociated at the interface between the components with the electrons transferred to the nanocrystals as shown in Figure 11.2. This feature has been exploited to form efficient photodiodes. TiO2, even in dilute quantities, acts as a charge separator, with the primary photogeneration and carrier transport essentially in the polymer backbone. TiO2 at such low levels of concentration can be introduced with a fair degree of uniformity in an MEH-PPV matrix as observed in the SEM images shown in Figure 11.3. Another important aspect which has not been addressed in these devices is that photodiodes are typically used in the reverse bias in the high rectification ratio range for maximum sensitivity, and under these bias conditions the spectral range of interest should not be a limitation [20, 58]. In pristine MEH-PPV based devices, the photocurrrent spectral response Iph(I) in the reverse bias peaks at the absorption edge, (indicative of an antibatic response) where Iph(I) is

350

Nanoparticle-Dispersed Semiconducting Polymers for Electronics

Figure 11.3 100 nm TiO2 particle uniformly dispersed in matrix of MEH-PPV

asymmetric with respect to the absorption response, with practically no photocurrent in the absorbing region, and this was explained qualitatively on the basis of exciton quenching at the Al interface coupled with the low electron mobility of the polymer. Parameters such as exciton/free carrier diffusion lengths, barrier depths, and film thickness come into play to model the spectra [58]. With the presence of electron acceptor moieties such as TiO2, the spectral range can be controlled by the magnitude of the bias voltage [20]. The results also highlight important differences in the switching response of the devices in the forward and reverse bias and its implication in photodetector devices [20]. An Iph/Idark (photocurrent/dark current) ratio as high as 104 at a reverse bias of –4.5 V and a responsivity of ~50 mA/W at –9 V with a dark current of 10 nA has been observed in these blended materials [23]. The open circuit voltage (as shown in the bottom part of Figure 11.4) was in the range of ~0.8 V with a short circuit current density of ~5 nA/cm2 for a photon density of 10 μW/cm2. The intensity dependence of Iph for input power 0.1 μW/ cm2 to 1 mW/cm2 is linear in the entire voltage range. In the case of TiO2 dispersed samples, the spectral features are sensitive to the magnitude of the reverse bias as shown in Figure 11.4. TiO2 nanoparticles play the role of electron acceptors since the interparticle distances, with large surface area, are in the same order of magnitude as the singlet exciton diffusion length (50 Å-150 Å) observed in MEH-PPV [58]. At higher absorption, λ < 3400 Å, where e-h pairs are generated in TiO2 nanocrystallites, there may be a hole transfer process onto the polymer from TiO2.

351

Handbook of Polymers in Electronics (a)

(b)

Figure 11.4 (a) Spectral response of TiO2 dispersed in MEH-PPV in forward and reverse bias. Note that the response at –3.3V is scaled down by a factor of 10. (b) Typical photodiode characteristic Iph – V, with Iph measured using the lock-in technique.

The lengthscales of the device, which may decide the spectral width of the different regions, are the thickness of the polymer layer, l, 1/α(λ) where α(λ) is the absorption coefficient, the exciton/free carrier diffusion length, 1/β, and the barrier width, lb. The bias dependendency of Iph(λ) can be qualitatively understood in terms of the model by Ghosh and others [59]: I ph ∝ ∫0l − l b ae ( − ax )e

352

− β( l − l b − x )

dx + ∫ll− l ae ( − ax )dx . b

Nanoparticle-Dispersed Semiconducting Polymers for Electronics Iph in this model depends on the ability of the minority carriers (electrons generated in the bulk) to reach (diffuse) the interface and is characterised essentially in terms of 1/β and lb. This argument is also applicable for excitons which diffuse and undergo dissociation. The barrier normally in these sandwich devices is primarily at the metalpolymer interface and is estimated to be ~200 Å [58]. For illumination from the side opposite the barrier, i.e., the ITO side, Iph results in an antibatic response; the numerical estimates of Iph(λ) for different diffusion length yield profiles which are remarkably similar to the experimental results [20]. It is seen that the experimental Iph(λ) at different voltages in the reverse bias can be explained by the diffusion length having a stronger dependency on the voltage than the barrier length within the framework of this model. An improvement in processing methods whereby a higher concentration of nanoparticles can be dispersed in the polymer matrix without a phase separation could lead to the realisation of the full potential of these systems for attractive technological applications. Dye-sensitised TiO2 particles blended with different hole-transporting polymer matrices have also been investigated [55]. The dye-sensitised layer acts effectively as an intermediary for the charge transfer processes. Host polymers with different HOMO levels have been used to prove that the Iph enhancement requires hole transfer from the dye to the polymer [55]. TiO2 can also be introduced as a porous nanocrystalline layer into which the polymer penetrates. A monolayer of TiO2 can be self-assembled onto the ITO surface using activation of the ITO surface with 3-aminopropyltriethoxysilane [53]. The forward-bias currents and open-circuit voltages are determined by the conduction band energy of TiO2 [53]. Apart from a single-layer film containing dispersed TiO 2 particles, a heterojunction of dye-sensitised TiO2 with an amorphous organic hole transport material, 2,2´,7,7´-tetrakis(N,N-di-p-methoxyphenyl-amine)9,9´-spirobifluorine, has been demonstrated to be an efficient solar cell with a high yield of 33% [60].

11.3.2 Nanoparticle Semiconductors-Polymer Systems Semiconductor nanoclusters, such as CdS, CdSe and ZnSe, have also been demonstrated as active layers for emission and detection devices with the optical range of interest tuned by size selection of the nanoparticles. The nanoparticle has shown to reveal efficient photocurrent [61] and EL properties with the spectral response governed by the particle size. A clear systematic blue shift in the photocurrent onset with the decrease in the particle size has been observed, see Figure 11.5a. Reports on a heterostructure device with an inorganic layer of CdS and a polymer light emitter show the advantage of the high charge transporting properties of an inorganic semiconductor and the enhanced emission efficiency of the polymer emitter [62]. It was found that the work function of the nanoparticles could be matched with that of the metal contacts improving the electron injection efficiency and it was also observed that the emission colour governed by the

353

Handbook of Polymers in Electronics (a)

(b)

Figure 11.5 (a) Nanoparticle photocurrent spectral response for different particle sizes, and (b) Typical electroluminescence response of 6FPBO and MEH-PPV

recombination zone could be controlled by the magnitude of external voltage. At low voltages, emission from the nanoparticle layer occurs, and at higher voltages the emission from the polymer was reported to predominate [62]. The possibility of synergistically combining the efficient photoconducting, PC properties of the inorganic component and the EL properties of the polymer in a single device was explored recently [63]. Results on such hybrid devices, where photoconductivity characteristics are dominated by the quantum size of the inorganic semiconductor and the enhanced EL spectral feature is attributable predominantly to the polymer component, were reported [63]. The potential of such devices, where windows for the detection and emission can be selected by nanoparticle-size engineering and/or modifying the chemical structure of the polymer, was demonstrated [63]. The EL response of such hybrid devices can essentially highlight the emissive polymer component, for example, a maximum at 6200 Å, 5400 Å, and 5250 Å for CdS-44/P3HT (polyhexylthiopene), CdS-44/MEH-PPV and CdS-44/6FPBO (6-fluoro poly(benzoxazolebioxydecyl)) devices, respectively. (CdS-44 = 44Å CdS nanoparticle.) The EL spectra of

354

Nanoparticle-Dispersed Semiconducting Polymers for Electronics the pure polymer devices are shown in Figure 11.5b. The EL spectra in the multilayer hybrid devices are closer to the pure polymer responses as shown in Figure 11.5 and Figure 11.6 with efficiency nearly an order of magnitude higher than CdS-only devices and polymer-only devices. 6FPBO is an electron transporting emitter [64] while MEHPPV and P3HT are predominantly hole transporting. The recombination in the 6FPBO device is closer to the 6FPBO-hole injecting layer interface with a minimum trap emission contribution from the nanocluster layer, resulting in a sharper emission. It has been observed in LED devices based on nanocrystals, which have high band-edge PL yield, that the insertion of a hole transporting polymer layer enhances the nanocrystal band-edge emission characteristics [62]. In the present case the results can be interpreted in terms of the CdS-44 layer enhancing the polymer emission characteristics. It is expected that the absorbance and the emission from the polymer layer will modulate the emission from the CdS-44 layer. The issues of PL quenching, charge separation and transport in nanocrystal CdS-44/MEH-PPV composite blends have been studied in detail [23, 65]. The results correlate quenching with improved quantum yield for charge separation. In the present case of bilayers with completely segregated phases, the possibility of Förster tranfer of the exciton to the nanoparticles will be effective only at the interface leading to a possible emission quenching. Initial results on PL with excitation energies corresponding to that of the polymer in CdS-44/P3HT do not show substantial quenching effects. Quenching effects can be minimised by reducing Förster transfer rates, by choosing polymers of appropriate electron affinity values and by using suitable capping materials for the nanocrystals. The EL efficiency can also be enhanced by optimising device parameters, such as thickness, with prior information on values of mobility and electron-hole pair diffusion length, to have radiative recombination away from the interface [63, 65]. In stark contrast, the short circuit photocurrent spectral response in all the three devices is identical to that of a CdS-44 layer independent of the type of the polymer. The Iph(λ) and EL response of the CdS-44/P3HT device is shown in Figure 11.6. The Iph(λ) response shown is representative of any CdS-44-based hybrid devices while the EL(λ) response is representative of any P3HT based hybrid device. The polymer signature is observed in the photocurrent, but the Iph(λ) in the non absorbing region of CdS-44 and absorbing region of the polymer is less than 3-4 orders in magnitude than the Iph(λ) in the CdS-44 absorption region. A significant separation width of nearly 2000 Å (2.0 eV) in the visible range between the emission maxima and the photocurrent region is observed for the CdS-44/P3HT device as shown in Figure 11.6. The external zero bias photocurrent efficiency in all the hybrid devices is in a similar range as that of pure nanoparticle devices with a responsivity of ~100 mA/W. It is expected that if the polymer thickness is increased to be comparable to the absorption depth, the spectral response would be modified. The large photocurrent primarily reflects the high efficiencies commonly observed in CdS-based detectors due to high absorption and efficient charge carrier

355

Handbook of Polymers in Electronics

Figure 11.6 Depiction of a bifunctional device with a spectral window for detection and emission controlled by independent components with photocurrent and EL of a multilayer device ITO/PVCZ/P3HT/CdS/A1: The nanoparticle size is 4.4 nm, which corresponds to a band edge of ~450 nm.

356

Nanoparticle-Dispersed Semiconducting Polymers for Electronics separation. The novelty of such a dual functional bilayer device with independent control of spectral windows for emission and detection response is depicted in Figure 11.6. The tunability of a Shottky diode based on a semiconductor/conjugated polymer (doped) interface has been explored electrochemically manipulating the work function of the conjugated polymer [66]. In the above case, the work function along with the charge carrier concentration and the nature of interface decide the LED characteristics. Another concept that has been explored is that of ‘diffused’ and ‘fractal’ p-n junctions in nanoparticle-polymer composites with both p- and n-type nanoparticles. Diffused junctions are obtained upon converting p-type particles into n-type. One can obtain an area consisting of pure p-type and pure n-type particles close to the electrodes, with an intermixed region inbetween the electrodes. Peculiar effects are observed in the intermixed layers since the nanoparticles sizes are similar or smaller than the Debye screening length [30, 31]. Results similar to that of the TiO2 dispersed MEH-PPV have also been observed for CdSe dispersed in MEH-PPV [23, 64]. In this case, a clear case of PL quenching is also observed as an evidence of charge transfer. Results have shown that the electron affinity of even the smallest CdSe nanocrystals is sufficient to allow electron transfer from MEHPPV. A general rule of thumb, which has been followed to enhance transfer process, is Eananocrystal – Eapolymer > Upolymer – Vcharge transfer, where Upolymer is the binding energy of the singlet exciton on the polymer and Vcharge transfer is the coulombic energy associated with attraction between the electron and hole in the final, charge separated state. At high concentration of the nanocrystals, where both the nanocrystals and polymer components provide continuous pathways to the electrodes, quantum efficiency up to 12% has been reported [64].

11.3.3 Gold-Polythiophene Blends Another class of materials, alkanethiol-stabilised metal nanoparticles, display electronic, optical and structural features that are tunable via particle size [67]. The theme of this section is to demonstrate the effects of interfacial chemistry and material heterogeneity on electronic and optical properties of luminescent conjugated polymers at metal interfaces. Hybrid systems comprising of these materials reveal several interesting features. Reports on blends of gold nanoparticles and conducting oligomers have demonstrated concepts such as the construction of mesoscopic oligomer bridges between metal nanoparticles [68]. Gold nanoparticles have surface reactivity amenable to immobilisation at chemically functionalised surfaces and can bind to positively charged polymers via electrostatic interaction [69] or covalently to amine, thiol and phosphine functional groups [70]. The

357

Handbook of Polymers in Electronics electronic structure of the polymer chains is also expected to be strongly perturbed in an environment of these metal nanoparticles. The striking features in this case are the complex structures that spontaneously appear upon film formation of gold nanoparticles and thiophene-based polymers [71, 72]. Fluorescence centres of ~1 μm diameter, containing the pure polymer, are dispersed throughout the film, see Figure 11.7. At the periphery of the fluorescent centres, a blue shift of ~120 nm in the fluorescence is observed, tunable via the gold nanoparticle concentration. The different component-specific features present in the film can be distinguished by the contrasting PL properties along with the direct TEM observation of the nanoparticles. The source of the PL in the polymer region is known to arise from the radiative decay of the intrachain singlet exciton level. Fluorescence microscopy and near-field scanning optical microscopy (NSOM) have been used to study the PL variation, with TEM for morphological studies of the phase separated system. NSOM is used to acquire spectra from regions of size less than that of the wavelength of the light source employed and the results demonstrate the capability of this technique to differentiate the components in the sub-micron lengthscale of these multiphase systems. In fact, Au-P3OT (polyoctylthiophene) systems can be viewed as model systems for probing using NSOM due to the spatial gradient of the spectral shifts occurring at nanometre lengthscales. The shear-force feedback technique in NSOM provides a topographic image in parallel with the optical image. A typical confocal fluorescence image of the Au-P3OT blended film is shown in Figure 11.7. Randomly distributed circular features with sharp intensity contrast are present throughout the film, indicating distinct phases. The circular features are essentially a signature of the

Figure 11.7 Confocal fluorescence microscope image of gold nanoparticles dispersed in P3OT

358

Nanoparticle-Dispersed Semiconducting Polymers for Electronics differences in the evaporation/dewetting rates of the two systems during the film formation, resulting in the creation of local domains. The emission of these circular regions resembled the emission from a pristine polymer film which has a PL maximum centred at ~700 nm. NSOM fluorescence images are consistent with the simultaneously obtained topographic image for these PT/gold nanoparticle films. As the proportion of polymer is increased, the size of the circular regions increases and the contrast between the bright and dark regions decreases. It is also to be noted that these polymer/Au nanoparticle film patterns, as shown in Figure 11.7 and Figure 11.8 are obtained only with polythiophene derivatives, such as

(a)

(b)

Figure 11.8 NSOM fluorescence contrast image (1 μm x 1 μm) of a bright spot/sphere on the Au blended P3OT film (a) along with the emission spectra at different distances away from the centre of the spot (b)

359

Handbook of Polymers in Electronics P3OT and P3HT. The PPV-based polymers such as MEH-PPV do not show such distinct phase formation. The high density of the nanoparticles in the periphery can be explained by the non uniform evaporation rate and a finite mobility of the nanoparticles, which get pinned at the rim. Another factor which comes into play is the aggregation of polymer chains, which can act as an effective drag on the nanoparticle motion during the drying process, giving rise to the inhomogenous radial distribution of the nanoparticles. The ringlike structures have been observed in alkanethiol-coated gold nanoparticles [74]. The nanoparticle mobility in that case was high enough to allow most particles to accumulate in the ring during the receding process of the contact line created by the interface between the solvent and the substrate [74]. Spatially-resolved PL spectra are obtained with the fibre tip held at several positions in the vicinity of a typical fluorescent centre of the blended system [72]. The emission spectra from different regions are shown in Figure 11.8. Spectra are blue-shifted as the tip moved from the centre towards the periphery. The PL from these fluorescent centres can be interpreted in terms of a crystalline (maximum ~700 nm) and an amorphous phase (maximum ~565 nm) of the polymer. The macrophase separation of these nanoparticles and the polymer in the films arises due to the non formation of a homogenous solution. An important issue which needs to be resolved is the determination of whether the electronic structural changes of the polymer in the blended system are secondary effects caused by a nanoparticle-induced physical process such as rearrangement of the polymer chains or by a nanoparticle-specific, weak metal-polymer interaction as the driving factor for the appearance of different phases [72]. It is noted that a typical nanoparticle stochastically dispersed in a polymer forms a system which is disordered. These disordered systems are generally parameterised by the presence of a percolation threshold, a change in structure and the dependence of topology on interconnected chains. In terms of optoelectronic properties, the entire advantages of amorphous silicon technology are applicable here with a much greater flexibility, a far wider range of control and better functionality, such as spectral distribution, switching speed and substrate choices.

11.4 Summary In summary, nanoparticles and polymers can form two distinct components in a composite with a diverse set of properties. The origin of these properties is distinctly different in the two components. One can tailor properties in composites by exploiting the various attributes. Nanoparticle-polymer composites demonstrate a synergistic approach to enhance efficiency in a variety of phenomena and in certain instances to achieve a complete set of novel properties. Considerable efforts to sort out issues such as processibility and stability should lead to the use of such combinations of materials as potential routes to many nanotechnology-based device applications.

360

Nanoparticle-Dispersed Semiconducting Polymers for Electronics

Acknowledgements K.S. Narayan thanks A.G. Manoj, Th.B. Singh, A. Alagiriswamy, G.L. Murthy, N. Kumar, K. Vijaya Sarathy and collaborators - Professor D.D. Sarma, Dr. J.W. White, Dr. R.J. Spry, Professor S. Ramakrishnan and Professor C.N.R. Rao. He also acknowledges the Department of Science and Technology and Council of Scientific and Industrial Research, Government of India, for partly funding the project.

References 1.

H. Weller, Angewandte Chemie International Edition, 1993, 32, 41, 53.

2.

A. Hengelin, Chemical Reviews, 1989, 89, 1861.

3.

L.E. Brus, Applied Physics A, 1991, 53, 465.

4.

J.Z. Zhang, Accounts of Chemical Research, 1997, 30, 10, 423.

5.

V.L. Colvin, A.N. Goldstein and A.P. Alivisatos, Journal of the American Chemical Society, 1992, 259, 1426.

6.

S. Okamoto, Y. Kanemitsu, H. Hosokawa, K. Murakoshi and S. Yanagida, Solid State Communications, 1998, 105, 1, 7.

7.

X. Peng, M.C. Schlamp, A.V. Kadavanich and A.P. Alivisatos, Journal of the American Chemical Society, 1997, 119, 30, 7019.

8.

M.C. Schlamp, X. Peng and P. Alivisatos, Journal of Applied Physics, 1997, 82, 4, 5837.

9.

M. Bruchez, Jr., M. Moronne, P. Gin, S. Weiss and A.P. Alivisatos, Science, 1998, 281, 5385, 2013.

10. J.H. Burroughs, D.C.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burn and A.B. Holmes, Nature, 1990, 347, 539. 11. D. Braun and A.J. Heeger, Applied Physics Letters, 1991, 58, 1982. 12. R.H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughes, R.N. Marks, C. Taliani, D.D.C. Bradley, D.A. Dos Santos, J.L. Bredas, M. Lögdlund and W.R. Salaneck, Nature, 1999, 397, 6715, 121. 13. G. Yu, J. Wang, J. McElvain and A.J. Heeger, Advanced Materials 1998, 10, 17, 1431.

361

Handbook of Polymers in Electronics 14. Q. Pei, G. Yu, C. Zhang, Y. Yang and A.J. Heeger, Science, 1995, 269, 5227, 1086. 15. G. Yu, J. Gao, J.C. Hummelen, F. Wudl and A.J. Heeger, Science, 1995, 270, 5243, 1789. 16. X.L. Chen and S. Jenekhe, Organic Thin Films, ACS Symposium Series 695, Ed., C.W. Frank, ACS, Washington, DC, USA, 1997, 161. 17. M. Seel, C.M. Liegner, W. Forner and J. Ladik, Physical Review B, 1988, 37, 2, 956. 18. A.K. Bakshi, Journal of Chemical Physics, 1992, 96, 2239. 19. R.A. Via and E.P. Giannelis, MRS Bulletin, 2001, 26, 394. 20. K.S. Narayan and Th. B. Singh, Applied Physics Letters, 1999, 74, 3, 345. 21. N.S. Sariciftci, L. Smilowitz, A.J. Heeger and F. Wudl, Science, 1992, 258, 1474. 22. C.H. Lee, G. Yu, N.S. Sariciftci, A.J. Heeger and C. Zhang, Synthetic Metals, 1995, 75, 2, 127. 23. N.C. Greenham, X. Peng and A.P. Alivisatos, Physical Review B, 1996, 54, 24, 17628. 24. Z. Qi and P.G. Pickup, Chemical Communications, 1998, 1, 15. 25. Z. Qi and P.G. Pickup, Chemical Communications, 1998, 21, 2299. 26. Z. Qi, M.C. Lefebvre and P.G. Pickup, Electroanalytical Chemistry, 1998, 459, 1, 9. 27. S.W. Huang, K.G. Neoh, E.T. Kang, H.S. Han and K.L. Tan, Journal of Materials Chemistry, 1998, 8, 8, 1743. 28. S.W. Huang, K.G. Neoh, C.W. Shih, D.S. Lim, E.T. Kang, H.S. Han and K.L. Tan, Synthetic Metals, 1998, 96, 2, 117. 29. K.G. Neoh, K.K. Tan, P.L. Goh, S.W. Huang, E.T. Kang and K.L. Tan, Polymer, 1999, 40, 4, 887. 30. D.Y. Godovsky, Advances in Polymer Science, Volume 153, Springer, Berlin, 2000, 163. 31. D. Godovsky, V. Sukhorev, A. Volkov and M. Moskvina, Journal of Physics and Chemistry of Solids, 1993, 54, 1613.

362

Nanoparticle-Dispersed Semiconducting Polymers for Electronics 32. R. Gangopadhyay and A. De, European Polymer Journal, 1999, 35, 1985. 33. R. Gangopadhyay, A. De and S.N. Das, Journal of Applied Physics, 2000, 87, 2363. 34. B.Z. Tang, Y. Geng, J.W.Y. Lam, B. Li, X. Jing, X. Wang, F. Wang, A.B. Pakhomov and X.X. Zhang, Chemistry of Materials, 1999, 11, 6, 1581. 35. C.B. Murray, D.J. Norris and M.G. Bawendi, Journal of the American Chemical Society, 1993, 115, 19, 8707. 36. R.A. Wessling and R.G. Zimmermann, inventors; Dow Chemical Company, assignee; US Patent B 3401152, 1968. 37. R.A. Wessling, Journal of Polymer Science - Polymer Symposia, 1985, 72, 55. 38. R.D. McCullough, Advanced Materials, 1998, 10, 2, 93. 39. S.M. Marinakos, L.C. Brousseau, A. Jones and D.L. Feldheim, Chemistry of Materials, 1998, 10, 5, 1214. 40. S.T. Selvan, J.P. Spatz, H.A. Klok and M. Moller, Advanced Materials, 1998, 10, 2, 132. 41. R. Gangopadhyay and A. De, Chemistry of Materials, 2000, 12, 3, 608. 42. A. Chevreau, B. Philips, B.G. Higgins and S. Risbud, Journal of Materials Chemistry, 1996, 6, 10, 1643. 43. S.S. Ray and M. Biswas, MRS, 1998, 33, 4, 533. 44. Primary Photoexcitations in Conjugated Polymers: Molecular Excitons versus Semiconductor Band Model, Ed., N.S. Sariciftci, World Scientific Publishers, Singapore, 1997. 45. A.J. Heeger in Primary Photoexcitations in Conjugated Polymers: Molecular Excitons versus Semiconductor Band Model, Ed., N.S. Sariciftci, World Scientific Publishers, Singapore, 1997, 20. 46. D. Moses in Primary Photoexcitations in Conjugated Polymers: Molecular Excitons versus Semiconductor Band Model, Ed., N.S. Sariciftci, World Scientific Publishers, Singapore, 1997, 174. 47. H. Bassler in Primary Photoexcitations in Conjugated Polymers: Molecular Excitons versus Semiconductor Band Model, Ed., N.S. Sariciftci, World Scientific Publishers, Singapore, 1997, 51.

363

Handbook of Polymers in Electronics 48. I.D.W. Samuel, G. Rumbles and R.H. Friend in Primary Photoexcitations in Conjugated Polymers: Molecular Excitons versus Semiconductor Band Model, Ed., N.S. Sariciftci, World Scientific Publishers, Singapore, 1997, 140. 49. M. Gailberger and H. Bassler, Physical Review B, 1991, 44, 16, 8643. 50. B. Dulieu, J. Wery, S. Lefrant and J. Bullot, Physical Review B, 1998, 57, 15, 9118. 51. B. O’Regan and M. Gratzel, Nature, 1991, 353, 737. 52. J.S. Salafsky, W.H. Lubberhuizen and R.E.I. Schropp, Chemical Physics Letters, 1998, 290, 4, 297. 53. A.C. Arango, S.A. Carter and P.J. Brock, Applied Physics Letters, 1999, 74, 12, 1698. 54. S.A. Carter, J.C. Scott and P.J. Brock, Applied Physics Letters, 1997, 71, 9, 1145. 55. T.K. Daubler, I. Glowacki, U. Scherf, J. Ulanski, H.-H. Horhold and D. Neher, Journal of Applied Physics, 1999, 86, 12, 6915. 56. A.F. Nogueira, L. Micaroni, W.A. Gazotti and M-A. De Paoli, Electrochemistry Communications, 1999, 1, 7, 262. 57. F. Hide, J.C. Schwartz, B.J. Diaz-Garcia and A.J. Heeger, Chemical Physics Letters, 1996, 256, 4, 424. 58. M.G. Harrison, J. Gruner and G.C.W. Spencer, Physical Review B, 1997, 55, 12, 7831. 59. A.K. Ghosh, D.L. Morel, T. Feng, R.F. Shaw and C.A. Rowe, Journal of Applied Physics, 1974, 45, 230. 60. U. Bach, D. Lupo, P. Comte, J.E. Moser, F. Weissortel, J. Salbeck, H. Spreitzer and M. Gratzel, Nature, 1998, 395, 583. 61. J. Nanda, K.S. Narayan, B.A. Kuruvilla, G.L. Murthy and D.D. Sarma, Applied Physics Letters, 1998, 72, 11, 1335. 62. V.L. Colvin, M.C. Schlamp and A.P. Alivisatos, Nature, 1994, 370, 354. 63. K.S. Narayan, A.G. Manoj, J. Nanda and D.D. Sarma, Applied Physics Letters, 1999, 74, 6, 871.

364

Nanoparticle-Dispersed Semiconducting Polymers for Electronics 64. G. Du, B. Taylor, R.J. Spry, M. Alexander, C. Grayson, J. Ferguson, B. Rienhardt and J. Burkett, Synthetic Metals, 1998, 97, 2, 135. 65. D.S. Ginger and N.C. Greenham, Physical Review B, 1999, 59, 16, 10622. 66. M.C.A. Lonergan, Science, 1997, 278, 5346, 2103. 67. U. Kreibig and M. Vollmer, Optical Properties of Metal Clusters, Springer Series in Materials Science-2502, 1995. 68. L.C. Brousseau, III, J.P. Novak, S.M. Marinakos and D.L. Feldheim, Advanced Materials, 1999, 11, 6, 447. 69. R.G. Freeman, K.C. Grabar, K.J. Allison, R.M. Bright, J.A. Davis, M.A. Jackson, P.C. Smith, D.G. Walter and M.J. Natan, Science, 1995, 267, 1629. 70. G. Schmid, Chemical Reviews, 1992, 2, 1709. 71. K. Vijaya Sarathy and K.S. Narayan, Current Science India, 1999, 77, 678. 72. V. Sarathy, K.S. Narayan, J. Kim and J.O. White, Chemical Physics Letters, 2000, 318, 6, 543. 73. R. Brunner, A. Bietsch, O. Hollricher and O. Marti, Review of Scientific Instruments, 1997, 68, 4, 1769. 74. P.C. Ohara, J.R. Heath and W.M. Gelbart, Angewandte Chemie International Edition, 1997, 36, 10, 1078.

365

Handbook of Polymers in Electronics

366

12

Polymers for Electronics T.H. Richardson

12.1 Introduction Electronics at the beginning of the third millennium is still massively dominated by traditional inorganic semiconductors, metals and ceramics. However, the last 20 years, and in particular the last decade, have seen enormous developments in the use of organic materials that can process electric charge or photons [1-4]. In one area at least, that of electroluminescent light-emitting diodes, organic materials are poised to make a huge impact on consumer electronics. This has arisen as a result of three main factors: •

The global research effort towards the realisation of cheap, lightweight, flat-panel, flexible, low power displays,

•

The insatiable consumer demand for ever-increasing visual display quality, and

•

The information technology revolution that has seen the enormous increase in the numbers of computer monitors, mobile phones and paging devices in the last 5 years alone.

It is noteworthy here, at the beginning of this chapter, to mention the awful truth about light-emitting polymers – their discovery was entirely serendipitous! Whilst measuring the breakdown voltage of a conjugated polymer [5], researchers noticed the emission of faint green light from their device. So was born the modern era of organic electroluminescence. More surprising still, perhaps, is the fact that organic electroluminescence was actually first observed as early as 1963 in single crystals of anthracene [6]. Whilst certainly taking the limelight, luminescent organic compounds are by no means the only group of materials that are continuing to attract research interest in electronics and photonics. The most used organic component in the electronics industry has been around for several decades already: the humble polymer photoresist [7]. This is the shy, retiring partner that is essential for the success of every integrated device in every circuit board. As device feature size has gradually decreased since the invention of the first transistor, refinements in photoresist technology have had to keep pace [8-9]. However, the real excitement will begin in the next decade or two when polymer transistors become commonplace, and when these transistors eventually comprise only a few tens of molecules! In a world in

367

Handbook of Polymers in Electronics which a surgeon can transplant hearts, lungs, livers and kidneys, not to mention the recent arm transplants, the idea of molecular-scale transistors and other devices is really not far-fetched. This chapter will include an introduction to some of the topics mentioned above with the intention of giving the reader some insight into the world of molecular electronics. It is less meaningful to discuss complex theories and device designs without devoting some thought to the deposition techniques available for the fabrication of polymer membranes for electronics devices. Ordered polymer layers of well-defined thickness and orientation do not appear by magic and so this chapter is closed with a survey of the fabrication technologies at the disposal of the organic device engineer.

12.2 Polymer Electroluminescence Liquid crystal materials [10] currently dominate the electronic display market since they are cheap and easy to produce and are extremely reliable passive devices. They produce no light of their own but utilise ambient light that is reflected from the displays in which they are housed. Whilst such displays are extremely effective in a wide range of singleuser applications (e.g., wristwatches, lap-top computers, miniature TVs, video camera displays), there are two major drawbacks for multi-user applications (e.g., full-size TVs, bulletin boards), because they process incident light from the surroundings, their brightness effectively is dependent on ambient lighting conditions (except for slight enhancements made possible with backlighting), and the display image quality is highly dependent on the viewing angle of the observer. Polymer electroluminescent devices [11] offer the possibility of efficient full-colour, lowvoltage displays with an improved brightness and viewing angle dependence over liquid crystal displays. Such devices contain several layers of organic materials interspersed between a metal cathode and a transparent (most often ITO) electrode to allow the luminescence to exit the device. Arguably the most important breakthrough in organic LEDs came in 1987 when Tang and co-workers [12] used monopolar charge transporting layers alone or either side of an emitter layer, as depicted in Figure 12.1. Although much of the early organic LED work focused on low molar mass organic molecules [13-16], researchers at Cambridge made the serendipitous discovery that the conjugated polymer PPV was electroluminescent. Whilst studying the insulating properties of the polymer in an Al-polymer-Al device configuration, they noticed green light emerging from the device. They subsequently realised that one of their electrodes had partially oxidised and was acting as a good hole-injecting contact. Thus holes were being favourably injected at one side of the polymer and electrons from the metal contact. Radiative recombination was occurring within the PPV, and the research field of polymer electroluminescence was born [17]!

368

Polymers for Electronics

Figure 12.1 Typical organic electroluminescence device configurations showing the arrangement of electrodes, carrier transport and emitting layers. (ETL and HTL refer to electron and hole transporting layers, respectively.)

The advances since this first discovery are summarised in Figure 12.2 which indicates how the drive voltage has been gradually reduced from around 100 V to a few, whilst the luminous efficiency has increased to around 100 lm/W. Many advanced display devices are expected to be marketed during the first decade of the new millennium, but the Pioneer Corporation was the first to produce a working 64 x 256 pixel device commercially, in 1996 [18].

369

Handbook of Polymers in Electronics

Figure 12.2 (a) Increase in luminous efficiency and (b) reduction in drive voltage since the beginning of research into organic electroluminescence

The simple two-layer device shown in Figure 12.1a contains two organic layers between a metal cathode and an optically transparent anode. The cathode is normally a low work function metal such as calcium, magnesium or aluminium. Calcium and magnesium, and to a lesser extent aluminium, are relatively unstable in air and oxidise quite rapidly leading to short device lifetimes. This problem is overcome in the research environment by carrying out device testing in an evacuated glove box with very low

370

Polymers for Electronics levels of water vapour and oxygen (typically sub-ppm). Clearly, efficient encapsulation of devices will be extremely important for obtaining long-life, stable operation in commercial devices. The almost universal anode material is ITO owing to its optical transparency and high conductivity. Furthermore it can be easily patterned using standard photolithography and a hydrochloric acid etchant. The application of a voltage across the device results in the injection of electrons and holes (at the electrodes) which then move towards the centre of the device. Excited state organic molecules are formed as a result of the recombination of electron-hole pairs; visible light is emitted as these excited species decay to their ground state. The wavelength of the emitted light depends on the region in which recombination takes place, which in turn is governed by the respective electron and hole mobilities of the electron-transport and hole-transport layers. A typical set of device characteristics for a device containing an emitting polymer material is shown in Figure 12.3. The current-voltage characteristic is similar to that of a traditional diode except that the turn-on voltage is much higher. The luminance output of the device is seen to be proportional to the current (density) through the device and the electroluminescence spectrum indicates that the emitted radiation is green in colour and is relatively broadband. A selected range of emitter materials as a function of the peak emission wavelength is shown in Figure 12.4 and demonstrates that polymers are available for red, green and blue light outputs. Also shown are some common low molar mass materials that have been developed in parallel with their polymeric counterparts. It is tempting to believe that organic molecules exhibiting high photoluminescent efficiencies in solution (the often preferred screening medium) will therefore be useful candidates for electroluminescent devices. Although this is sometimes true, there are many examples of little or no electroluminescence being produced. This is due to concentration quenching in the solid state through the formation of exciplexes and/or quenching due to interactions with oxygen or the electrode materials [19]. One disappointing feature of most electroluminescent polymers is that the emission spectrum tends to be relatively broad (full wave half maximum (FWHM) ~100-150 nm). Ideally for display applications, high colour purity is desirable. Several approaches to sharpening the emission spectrum of electroluminescent light output have been made. These include the use of rare earth containing materials in order to utilise the atomic levels in the metal ions themselves as the electronic transition sites [20-21]. More success, however, has been achieved using microcavities in which both spectral sharpening and luminance enhancement can be achieved simultaneously [22]. A typical microcavity structure comprising a dielectric mirror at one end of the usual device configuration is depicted in Figure 12.5. It must be noted here that the thickness of the transport and emitting layers are crucial in tuning the peak emission wavelength

371

Handbook of Polymers in Electronics

Figure 12.3 Device characteristics for a polymer light-emitting device showing (a) the current-voltage characteristic (open circles) and the luminance-voltage characteristic (filled squares) and (b) the absorbance (Abs) and photoluminescence (PL) spectra. The electroluminescence is very similar to the PL.

372

Polymers for Electronics

Figure 12.4 A selected range of organic emitter materials as a function of the peak emission wavelength

and intensity. Spectral bandwidths have been reduced typically to 25 nm using this approach. One present difficulty is the angular dependence of the emission wavelength; the apparent colour of the emitted light changes slightly depending on the viewing angle as a result of the modified pathlength of the cavity. One possible way of overcoming this problem is to move from a planar to a radial geometry in which each layer of the device is deposited onto a substrate containing an array of hemispherical lens-type features. The future for electroluminescent devices is certainly bright.

373

Handbook of Polymers in Electronics (a)

(b)

Figure 12.5 (a) Typical microcavity-based device structure for colour purity enhancement and (b) the resulting narrow spectral output

374

Polymers for Electronics

12.3 Conduction in Polymers Conjugated polymers have attracted great interest for a number of devices including light-emitting diodes, photovoltaic cells, field-effect transistors and photocopiers [23]. The common property linking all these applications is that the polymer must be able to support the transport of electrical charge and accept or relinquish this charge at its interface with other media. The density of this current, J, is given by: J = q {n(x)μn(E) E(x) + Dn δn(x)/δx}

(12.1)

where q is the electronic unit charge of the carriers involved in conduction, n(x) is the density of the carrier, μn(E) is the carrier mobility, E(x) is the electric field and Dn is the carrier diffusion coefficient. This equation is shown for a single carrier but in the general case where electrons and holes are involved in the conduction process, terms can be added such that both carriers are represented. Equation 12.1 shows that the current density is spatially dependent but the steady-state bulk conductivity, σ, is independent of position: σ = J/E = q n μn

(12.2)

The density of the charge carriers and their mobilities therefore determine the conductivity of a material. The majority of conjugated molecular materials have energy gaps in the range 1.5-3.0 eV and are thus viewed as wide band gap semiconductors. They are very different to traditional inorganic semiconductors, the main difference being their much lower carrier mobility (typically 10-5-10-6 cm2 V-1 s-1 compared to 103-104 cm2 V-1 s-1 for GaAs). In silicon, for example, the atoms are held together by covalent bonds formed by the overlapping electron sp3-hybridised orbitals. These electrons are delocalised resulting in giant extended ‘orbitals’ referred to as ‘energy bands’ that are responsible for giving charge carriers their freedom to move easily through the solid under the influence of an electric field. For a small organic molecule, the electrons do not have it so easy. Carboncarbon single bonds (σ bonds) form as a result of the overlap of two sp3 orbitals, one from each carbon atom. The electrons in these bonds are strongly localised between the two atom centres and cannot effectively participate in any conduction process. Carboncarbon double bonds, on the other hand, involve a 2pz orbital from each carbon atom as well as two overlapping sp2 orbitals. The overlapping 2pz orbitals are referred to as a π bond; the σ bond and the π bond together are termed the ‘double bond’. Electrons in π bonds are much more delocalised and can roam over an extended region far beyond the vicinity of the σ bond. In a conjugated molecule, an alternating sequence of carboncarbon single and double bonds exists; this leads effectively to another giant (although

375

Handbook of Polymers in Electronics not as giant as in the case of silicon) orbital stretching along the length of the conjugated region of the molecule. This region is effectively a ‘cavity’ along which π electrons can move relatively easily and contribute to conduction. As will be mentioned later, however, the conductivity of a conjugated material is not merely governed by the length of the molecule; indeed, the main bottleneck to the process arises from the large energy barrier for intermolecular charge transport. The energy bandgap reduces as the length of conjugation increases as seen in Figure 12.6. This explains why small organic molecules are usually colourless (appearing as white crystals due to scattering) whereas longer conjugated molecules are coloured; hence carrots are orange and grass is green due to their principal component molecules being β -carotene and chlorophyll, both molecules having regions of extended conjugation [24]. As the conjugation length increases, the energy difference between the HOMO and the LUMO decreases according to: Egopt (n) ~ Egopt (infinity) - K/n

(12.3)

where Egopt (n) is the band gap (optical exciton or carrier energy gap) for a conjugated chain of n repeat units, Egopt (infinity) is the energy gap for an infinitely long conjugated chain and K is a constant. It should be noted, however, that very long conjugated chains twist, bend and distort quite easily; this disrupts the π-electron system so the molecule is said to possess an effective conjugation length that is dependent on the particular polymer chain environment.

Figure 12.6 The effect of conjugation length in polymers on the optical exciton gap

376

Polymers for Electronics Electrons and holes on conjugated polymer chains transfer from chain to chain (molecule to molecule) in response to an applied field at a rate that determines the carrier mobility and hence the conductivity of the material. Many models have been developed to explain the detailed conductivity mechanisms for molecular materials but their description is beyond the scope of this chapter. The interested reader should refer to the excellent review by Campbell [25]. High conductivities are usually achieved by doping the polymer in order to add carriers that will subsequently be able to move through the polymer material under the influence of an electric field. If an electron accepting dopant is used, then the p-type doping results in oxidation of the polymer and the formation of a positive polaron, also known as a radical cation. Further doping results in a positive bipolar or radical dication. This process is shown in Figure 12.7 for PPP. These polarons or bipolarons can move along the polymer chain relatively easily and can hop from one polymer chain to another as a result of redox (polaron-transfer) reactions between neighbouring chains. The bulk electrical conductivity is primarily dependent on the product of the number and mobility of the charge carriers. Doping increases the carrier density, long (but not too long) conjugation regions enhance the intrachain transport, and alignment of the polymer chains (by stretching or poling for example) augments the interchain transport which is usually the limiting step of the conduction process.

Figure 12.7 Polaron and bipolaron formation as a result of doping

377

Handbook of Polymers in Electronics

Figure 12.8 Well-known conducting polymers

Even when doped, most ‘conducting’ polymers do not show metallic-level conductivity. An exception is polyacetylene whose conductivity has been measured to be as high as ~105 S cm-1 [26]. PPV is commonly labelled as a conducting polymer but when undoped typically has a conductivity of around 10-14 S cm-1 and is insulating. Most conducting polymers have conductivities in the range 0.1-1 S cm-1 when fully doped [27]. A selected range of well-known conducting polymers is shown in Figure 12.8. The research field is advancing rapidly and this selection represents only the best-known families of polymers rather than the most conducting individual examples.

378

Polymers for Electronics

12.4 Molecular Electronics It is important to understand the driving forces behind the elucidation and development of the optical and electrical properties of polymeric materials. These are varied and fall into two main categories: low-technology and futuristic high-technology opportunities. Imagine being able to produce large, thin sheets of a highly conducting polymer exhibiting conductivity comparable to that of copper. There are many applications for such a material, the most obvious ones being coatings for electromagnetic shielding (instead of using expensive, heavy metal boxes) and electronic circuit interconnecting tracking (to replace the miles of copper or gold currently used). Speculate further about circuits in which each silicon transistor is replaced by a polymer version; again this is not farfetched. Organic field electron transistors (FETs) have been developed in research groups already with mobilities around 0.1-0.5 cm2 V-1 s-1 [28-30]. Their performance is poor compared to silicon transistors but it must be noted that it took just 40 years to progress from the discrete silicon transistor to a global microelectronics revolution in which the transistor gate geometry is currently 0.18 μm and photolithography occurs using deep ultraviolet radiation of around 190 nm. The fascination with organic electronic devices stems from the lure of miniaturisation. Consideration of the gradual reduction in transistor dimensions since the 1960s suggests that molecular scale transistors may be achievable shortly. Researchers have minimised the z-dimension (thickness) for devices such as unimolecular rectifiers [31]. Metzger’s and Ashwell’s research in this area has shown that charge-transfer TTF:TCNQ-type (tetrathiafulvalene:tetracyanoquinodimethane) molecules indeed show directional conduction [32-33]. Little progress has been made in shrinking the x- and y-dimensions to similar values to the z-dimension (~2-10 nm) in order to achieve a truly molecular scale device. Furthermore, the question of making contacts at this scale has received little attention. In the humble opinion of this author, however, the switch to organic-based electronic diodes and transistors is unlikely; the more probable event will be the invention of completely novel electronic devices that utilise complex biological molecules as their active ingredients. Computing based on DNA [34] is a growing discipline far removed from present-day electronic circuit approaches.

12.5 Polymer Deposition Technologies As mentioned earlier, it is misleading to describe advances in the properties of polymers without also describing how such polymers are processed into thin films or crystals. Moreover, many of the physical properties of polymers are inextricably linked to their structural and orientational order. An excellent example of this is the piezoelectric and pyroelectric coefficients of the well-known polymer, polyvinylidene difluoride (PVdF)

379

Handbook of Polymers in Electronics [35]. When this polymer is processed into thin sheets, these coefficients are close to zero, yet after stretch aligning (which causes the ordered alignment of the long polymer chains) the coefficients rise to the highest known values for organic materials, rendering PVdF extremely useful for pressure and heat sensing applications. Whilst single crystals of polymers can sometimes be grown, the most common geometrical configuration of a polymer in an electronic device is a thin planar film. This results from the advanced planar microelectronics and integrated optics technologies that have grown up over the last 30 years. A number of techniques exist for the deposition of polymers in thin film form, the most well-known and well-used method being polymer spin coating. Spin coating is used most often for depositing layers of photoresist as part of the patterning process in integrated circuit fabrication. This simple, yet effective, technique is illustrated in Figure 12.9. A relatively viscous polymer solution is placed onto the substrate to be coated which is then rotated at a fixed angular speed in the range 500-4000 rpm.

Figure 12.9 The polymer spin-coating process

The polymer solution flows radially outwards to form a thin solution layer that subsequently ‘sets’ as the solvent evaporates. The uniformity of the layer depends on a number of factors including the initial acceleration of the substrate and the rate of solvent evaporation, both of which can be easily controlled. The film thickness, d, depends on the solution viscosity, η, rotation speed ω, solution density ρ and spinning time t and is given by: d = {η /4π ρ ω2 t}1/2

(12.4)

More involved models to relate the thickness to the spinning conditions have been developed [36]. Generally, the polymer molecules within spin-coated films are relatively disordered and order has to be induced after deposition via stretch aligning or more commonly via electrical poling at elevated temperature [37].

380

Polymers for Electronics A favoured technique by those who require highly ordered or partially ordered polymer films is LB deposition [38]. An LB film is an ultrathin organic assembly formed by the sequential transfer of floating Langmuir layers from a water surface onto a solid substrate. In its conventional form, it comprises two-dimensional solid sheets of well-packed organic molecules deposited in a predetermined, controllable sequence. The first materials investigated by Pockels, Langmuir and Blodgett were generally rod-shaped amphiphilic molecules such as octadecanoic acid [39]. There is today a huge variety in the shapes and types of molecules that can be deposited as LB films, many of them being polymers [40]. Before the LB film can be fabricated, a floating Langmuir film must first be created at an air-water interface. Octadecanoic acid (Figure 12.10a) is an amphiphilic molecule, which illustrates the LB process nicely. The polar carboxylic group is attracted to other polar substances such as water, whereas the long alkyl chain is hydrophobic and is repelled by water. This amphiphilicity is responsible for giving octadecanoic acid its unique orientational properties; the polar headgroup is effectively dissolved in water, yet the alkyl chain protrudes from the water surface in a near normal direction. The carboxyl functionality is so strong that many alkanoic acids will spontaneously spread from a bulk crystallite placed in contact with a water surface. Usually, however, the material is dissolved in a solvent such as chloroform, minute droplets (each containing ~2μl) of which are then dropped carefully on the water surface. The solution spreads over the available water surface rapidly, the solvent evaporating fully over a period of a few minutes, leaving behind the randomly distributed octadecanoic acid solute molecules. Indeed, even at this point in the procedure, a monolayer film exists, albeit an inhomogeneous one. Uniformity in the surface density and thickness of the monolayer is improved by now reducing the water surface area available to the floating solute molecules, thus compressing the initially expanded monolayer to form eventually a closed-packed, two-dimensional sheet of octadecanoic acid. Typically its molecular surface density in the compressed state is ~5 x 1014 cm-2 and its thickness is ~2.5 nm. In practice, an apparatus known as a Langmuir trough is used to perform the compression. There are many different commercial designs of Langmuir troughs available, the most well-known being the NIMA [41] version. Each provides a means of confining the floating monolayer within an accurately controlled surface area that can be varied from a maximum value (at which the spreading process is usually performed) to a minimum value corresponding to maximum compression. After spreading and evaporation of the solvent, the area available to the monolayer can be continually reduced in order to gradually compress the molecules. The presence of a full or partial layer of molecules at a pure water surface dramatically modifies the surface tension. A change in the surface tension, known as surface pressure, Π, can be measured easily using a sensor that probes the water surface. A highly wettable plate (known as a Wilhelmy plate) is suspended from the microbalance using a fine, light and inextensible thread. A surface pressure-

381

Handbook of Polymers in Electronics (a)

(b)

(c)

Figure 12.10 (a) Octadecanoic acid, a typical amphiphilic molecule, (b) a schematic surface pressure-area isotherm for octadecanoic acid and (c) the orientation of octadecanoic acid molecules in a monolayer

382

Polymers for Electronics area isotherm, which describes the change in surface tension of the water surface as the floating monolayer is compressed, is shown for an alkanoic acid in Figure 12.10b. Inspection of this isotherm is useful to obtain an insight into the orientation of the molecules within the monolayer. At large area in its expanded state, the surface pressure is close to zero indicating that only weak interactions are occurring between molecules. As the confinement area of the monolayer is decreased further, the surface pressure begins to rise – this region is referred to as the liquid-condensed region (L2) and corresponds to phases in which the molecules are tilted relative to the plane of the water surface. Further compression results in the formation of the superliquid (LS) phase in which the alkyl chains protrude almost orthogonally from the plane of the water surface. The LS region represents a state of high incompressibility such as found in a traditional three-dimensional solid. Indeed the LS region used to be referred to as the ‘solid’ region of the isotherm [42]. The area occupied on the water surface by a single alkanoic acid molecule within the monolayer is about 0.2 nm2. This value corresponds closely to the cross-sectional area of the carboxyl end of the molecule as shown in Figure 12.10c. This implies that the floating film is indeed monomolecular in dimension. Such measurements have been made for a huge number of different materials, providing insight into the relationships between the detailed molecular structure and the molecular orientation at the air-water interface. The transfer of floating Langmuir films onto solid supports is in principle a straightforward process involving the successive insertion and withdrawal of a substrate through the monolayer film at the air-water interface. Its success depends on the optimisation of a number of controlling variables such as the deposition surface pressure, the deposition rate (rate of withdrawal and/or rate of insertion), substrate cleanliness and surface treatment, monolayer fluidity, pH and ion content of the sub-phase and temperature. The most fundamental requirement is that of a highly stable feedback mechanism between the surface pressure monitors and the motors, which drive the confinement belts responsible for confining the monolayer. This ensures that the small decrease in the surface pressure that results when part of the monolayer is transferred onto a solid substrate stimulates further compression of the monolayer in order to restore the original preset deposition surface pressure. This ensures that the molecular surface density within the transferred monolayer remains constant. A schematic diagram of a typical Langmuir trough is shown in Figure 12.11. The substrate insertion and withdrawal is normally achieved using a motor-driven micrometer screw to which is attached the clamp for the substrate. The sequential transfer process for a hydrophilic substrate withdrawn and inserted through an alkanoic acid monolayer is shown in Figure 12.12. The transfer of the first monolayer occurs due to the strong interaction between the carboxylic acid groups of the alkanoic acid molecules and polar sites (usually taking the form of free hydroxyl groups) on the hydrophilic substrate surface.

383

Handbook of Polymers in Electronics

Figure 12.11 A typical Langmuir trough for the fabrication of LB films showing the surface pressure monitor (P), the wettable plate (W), the substrate (S) attached to the deposition mechanism (D) via the clamp (C), the moveable constant perimeter barriers (MB), the barrier drive motors (BD) and the water bath (B).

Figure 12.12 The sequential monolayer transfer process for the formation of LB films showing the formation of Y-type assemblies. Inset: Y, X and Z-type architectures.

384

Polymers for Electronics Reinsertion of the monolayer-coated substrate results in the transfer of the second monolayer. There is a strong hydrophobic interaction between the alkyl chains within the first monolayer (the new substrate surface for the second layer) and those protruding from the water surface in the floating monolayer. A second layer is thus adsorbed and the surface of the coated substrate becomes hydrophilic again and is thus ready to accept the third monolayer upon its next withdrawal through the Langmuir film. This process can be repeated over many cycles and is referred to as Y-type deposition. Most rodshaped amphiphilic molecules follow this mode of deposition as do most pendant-chain polymers, like polysiloxanes [43-44]. This deposition mode results essentially (except for the first monolayer) in a non centrosymmetric structure. Such assemblies are generally very stable over time owing to the strong stabilising interactions between hydrophobic chains from adjacent monolayers, and between hydrophilic carboxyl groups that can form hydrogen bonds across the interface between layers and in-plane sideways dimers that serve to strengthen the layers laterally. Certain materials, such as many phthalocyanines [45], can only be deposited during each upstroke (Z-type deposition) and others only during each insertion (X-type). The type of deposition followed by a particular material affects the properties of the resulting LB films due to symmetry restrictions for second-order physical effects such as piezoelectricity or optical second-harmonic generation. The bilayer unit (equivalent to the unit cell in a crystal) is symmetric for Y-type deposition but non centric for Z- and Xtype modes, as indicated in Figure 12.12. Unfortunately, several researchers have found that Z- and X-mode multilayers are often temporally unstable, although the zwitterionic dyes of Ashwell prove to be one of the exceptions [46]. One solution to achieving non centrosymmetry within stable multilayer films is to use the alternate layer LB deposition technique. This employs a special kind of Langmuir trough known as an alternate layer trough that is effectively two single-compartment troughs hybridised into one unit. Such an apparatus is depicted in Figure 12.13 and this shows how two independent water surfaces are used to carry two different monolayers, A and B. The deposition sequence is ABABABA…. so that the resulting architecture is pseudo Y-type. Thus, alkyl chains from adjacent monolayers interact hydrophobically, as do the hydrophilic, polar groups; the difference here is that the molecules within layers A and B are inherently different. Therefore, dipoles μ1 and μ2 would not cancel completely and indeed, if the molecules have been designed such that μ1 and μ2 act in opposite directions with respect to their alkyl chains, then an overall electric polarisation results which is an additive function of μ1 + μ2. The resultant stability of this system is clearly a trade-off between the structural stability arising from the pseudo Y-type architecture and the energetic stability that is expected to gradually reduce as the number of transferred layers increases owing to the large build-up in electric polarisation [47]. However, the alkyl chain regions act as screening layers meaning that multilayer films

385

Handbook of Polymers in Electronics

Figure 12.13 Alternate layer Langmuir trough showing the substrate attached to the rotating drum (R). All other symbols have the same meaning as in Figure 12.11.

containing up to ~200 layers can often be deposited without structural degradation. Alternate layer LB films have been adopted most commonly for studies concerning non linear physical phenomena such as piezoelectricity, pyroelectricity and optical secondharmonic generation.

386

Polymers for Electronics Although the LB approach is applicable to an ever-increasing range of materials, water-soluble compounds are excluded. In a much lower technology, yet elegant, deposition process, water-soluble polymers can be deposited layer by layer to form similar molecular assemblies to LB films [48]. This method, generally known as the polyelectrolyte layer-by-layer self-assembly process, makes use of the ionic attraction between opposite charges on anionic and cationic electrolytes. A solid substrate is positively charged by chemical cleaning methods and is placed in a solution of an anionic polyelectrolyte, such as poly (sodium 4-styrenesulfonate) [49]. The deposition process is depicted in Figure 12.14. During a period of around 10 minutes, a monolayer of the polymer is adsorbed onto the substrate surface via the electrostatic attraction. The process is self-regulating since a second monolayer of the anion cannot be adsorbed due to electrostatic repulsion since there are many unpaired anionic charges on the polymer which remain electrostatically unsatisfied. Therefore, after monolayer adsorption, the coated substrate is washed thoroughly in water and then placed in a solution of a cationic polyelectrolyte, such as poly(alylamine hydrochloride [50]. Again ionic attractions result in the adsorption of a cationic monolayer. This procedure can be repeated many times to build up alternate layer polymer films. The advantages of this method over LB deposition are that it is very easy to perform and applicable to water-soluble polymers; the drawback, however, is that the rate of deposition is limited to around 5 monolayers per hour compared to conventional LB deposition rates of 20 monolayers per hour. Recent developments in LB technology have led to much higher rates approaching 30 monolayers per minute [51].

Figure 12.14 The layer-by-layer polyelectrolyte self-assembly process for charged polymers

387

Handbook of Polymers in Electronics

12.6 Summary Much progress has been made in the development of polymers for electronics applications over the last 20 years. The next few years will see the impact of flat-screen displays based on organic materials and the continued development of biological electronics for biosensing and biocomputing applications. The amazing developments in the microelectronics sector over the last few decades has been driven by the insatiable demand for technology to improve mankind’s quality of life and to accelerate further progress. A similar development in the fabrication of organic circuits may or may not occur depending on the perceived advantages of moving away from silicon. Certainly if it were to happen, it would take much less than 40 years. Undoubtedly, this research area is set to attract increasing attention and to produce devices that are as of yet totally inconceivable.

Acknowledgements The author would like to thank Al Campbell, Dave Lidzey, Rob Fletcher, Colin Dooling, Pierre Oliviere and O. Worsfold for helpful discussions during the preparation of this chapter.

References 1.

G.G. Roberts, Langmuir-Blodgett Films, Plenum Press, New York, NY, USA, 1990.

2.

M.C. Petty, M.R. Bryce and D. Bloor, Introduction to Molecular Electronics, Edward Arnold, London, UK, 1995.

3.

William Jones, Organic Molecular Solids: Properties and Applications, CRC Press, Boca Raton, FL, USA, 1997.

4.

T.H. Richardson, Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, John Wiley & Sons, Chichester, UK, 2000.

5.

J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burns and A.B. Holmes, Nature, 1990, 347, 6293, 539.

6.

M. Pope, H.P. Kallmann and P. Magnante, Journal of Chemical Physics, 1963, 38, 2042.

7.

T. Iwayanagi, T. Ueno, S. Nonogaki, H. Ito and C.G. Wilson in Electronic and Photonic Applications of Polymers, Eds., M.J. Bowden and S.R. Turner, ACS, Washington, DC, USA, 1988, 109.

388

Polymers for Electronics 8.

M. McCallum, K.R. Dean and J.D.Byers, Microelectronic Engineering, 1999, 46, 1-4, 335.

9.

M.W. Poulter, National Semiconductor Inc., Santa Clara, California, USA, 2000 (Private Communication).

10. D.A. Dunmur and L.D. Farrand in Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, Ed., T.H. Richardson, John Wiley & Sons, Chichester, UK, 2000, 31. 11. D.D.C. Bradley, Current Opinion in Solid State Materials Science, 1996, 1, 789. 12. C.W. Tang and S.A. VanSlyke, Applied Physics Letters, 1987, 51, 913. 13. P.S. Vincett, W.A. Barlow, R.H. Hann and G.G. Roberts, Thin Solid Films, 1982, 94, 171. 14. C. Adachi, S. Tokito, T. Tsutsui and S. Saito, Japanese Journal of Applied Physics, 1988, 27, L269. 15. R.H. Partridge, Polymer, 1983, 24, 733. 16. C.W. Tang, S.A. VanSlyke and C.H. Chen, Journal of Applied Physics, 1989, 65, 3610. 17. A.J. Hudson and M.S. Weaver in Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, Ed., T.H. Richardson, John Wiley & Sons, Chichester, UK, 2000, 365. 18. H. Inada, Y. Yonemoto, T. Wakimoto, K. Imai and Y. Shirota, Molecular Crystals and Liquid Crystals Science and Technology Section A, 1996, 280, 331-336. 19. C. Adachi, T. Tsutsui and S. Saito, Applied Physics Letters, 1989, 56, 799. 20. J. Kido, K. Nagai, Y. Okamoto and T. Skotheim, Chemical Letters, 1991, 1267. 21. M. Weaver, S. Martin, D.D.C. Bradley, M. Pavier, T. Searle and T.H. Richardson, Synthetic Metals, 1995, 76, 1-3, 91. 22. D.G. Lidzey, M.S. Weaver, T.A. Fisher, M.A. Pate, D.M. Whittaker, M.S. Skolnick and D.D.C. Bradley, Synthetic Metals, 1996, 76, 1-3, 129. 23. M. Stolka in Special Polymers for Electronics and Optoelectronics, Eds., J.A. Chilton and M.T. Goosey, Chapman and Hall, London, UK, 1995, 284.

389

Handbook of Polymers in Electronics 24. L.R. Milgrom, The Colours of Life: An Introduction to the Chemistry of Porphyrins and Related Compounds, Oxford University Press, Oxford, UK, 1997. 25. A.J. Campbell in Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, Ed., T.H. Richardson, John Wiley & Sons, Chichester, UK, 2000, 201. 26. H. Naarmann and N. Theophilou, Synthetic Metals, 1987, 22, 1. 27. S. Roth, One-Dimensional Metals: Physics and Materials Science, VCH, Weinheim, Germany, 1995. 28. W.P. Hu, Y.Q. Liu, Y. Xu, S.G. Liu, S.Q. Zhou and D.B. Zhu, Molecular Crystals and Liquid Crystals Science and Technology Section A: Liquid Crystals and Molecular Crystals, 1999, 337, 511. 29. Y.Y. Lin, D.J. Gundlach, S.F. Nelson and T.N. Jackson, IEEE Electronic Device Letters, 1997, 18, 12, 606. 30. J.G. Laquindanum, H.E. Katz, A. Dodabalapur and A.J. Lovinger, Journal of the American Chemical Society, 1996, 118, 45, 11331. 31. A. Aviram and M.A. Ratner, Chemical Physics Letters, 1974, 29, 277. 32. R.M. Metzger, B. Chen, U. Hopfer, M.V. Lakshmikantham, D. Vuillaume, T. Kawai, X. Wu, H. Tachibana, T.V. Hughes, H. Sakurai, J.W. Baldwin, C. Hosch, M.P. Cava, L. Brehmer and G.J. Ashwell, Journal of the American Chemical Society, 1997, 119, 10455. 33. R.M. Metzger, Advanced Materials for Optics and Electronics, 1998, 8, 229. 34. M. Ogihara and A. Ray, Nature, 2000, 403, 143. 35. T. Furukawa, M. Date and E. Fukada, Journal of Applied Physics, 1980, 51, 1135. 36. W.W. Flack, D.S. Soong, A.T. Bell and D.W. Hess, Journal of Applied Physics, 1984, 56, 1199. 37. M. Petty, J. Tsibouklis, M.C. Petty and W.J. Feast, Ferroelectrics, 1993, 150, 267. 38. T.H. Richardson in Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, Ed., T.H. Richardson, John Wiley & Sons, Chichester, UK, 2000, 181.

390

Polymers for Electronics 39. I.R. Peterson, Thin Solid Films, 1984, 116, 357. 40. R.A. Hann in Langmuir-Blodgett Films, Ed., G.G. Roberts, Plenum Press, New York, NY, USA, 1990, 68. 41. Nima Technology Limited, http://www.nima.co.uk 42. I.R. Peterson in Functional Organic and Polymeric Materials: Molecular Functionality – Macroscopic Reality, Ed., T.H. Richardson, John Wiley & Sons, Chichester, UK, 2000, 109. 43. T.H. Richardson, W.H.A. Majid, R.Capan, D. Lacey and S. Holder, Supramolecular Science, 1994, 1, 1, 1. 44. D. Lacey, S. Holder, W.H.A. Majid, R. Capan and T.H. Richardson, Materials Science & Engineering, 1995, C99. 45. C.C. Leznoff and A.B.P. Lever, Phthalocyanines: Volume 1 – Properties and Applications, VCH, New York, NY, USA, 1989. 46. G.J. Ashwell, P.D. Jackson and W.A. Crossland, Nature, 1994, 368, 438. 47. P. Christie, G.G. Roberts and M.C. Petty, Applied Physics Letters, 1986, 48, 1101. 48. G. Decher, J.D. Hong and J. Schmitt, Thin Solid Films, 1992, 210/211, 831. 49. G. Decher, Y. Lvov and J. Schmitt, Thin Solid Films, 1994, 244, 772. 50. W.B. Stockton and M.F. Rubner, Macromolecules, 1997, 30, 2717. 51. T.H. Richardson, C.M. Dooling, O. Worsfold, L.T. Jones, K. Kato, K. Shinbo, F. Kaneko, R. Tregonning, M.O. Vysotsky and C.A. Hunter, Colloid and Surfaces A, 2002, 198-200, 843.

391

Handbook of Polymers in Electronics

392

13

Conducting Polymers in Molecular Electronics B. D. Malhotra, V. Saxena, A. Gambhir, R. Singhal, S. Annapoorni and A. Mukhopadhyay

13.1 Introduction Electronic technology has developed enormously over the past decades. The trend is to make better, faster and smaller electronic devices for use in our daily living. Most of the electronic devices are fabricated directly on semiconductor silicon. Today’s advanced silicon chip can store sixteen million bits of information within an area less than 1 cm2. However, there is a practical limit to the capacity of the storage of information within the chip. It is being conjectured that as chip density increases, crosstalk between them tends to degrade their performance. If the components are pushed further, they may short-circuit. This inherent technical difficulty has led to the evolution of the field of molecular electronics. Molecular electronics (ME) is so named because it uses molecules to function as ‘switches‘ and ‘wires’. ME is a term that refers both to the use of molecular materials in electronics and to electronics at molecular level. It is as yet not very clear how molecular electronic devices will operate, but it is conjectured that active molecules are needed, either in isolation or becoming active by association with other molecules. It is thought that electronics is likely to imitate some of the basic functions of macroscopic devices such as memories, sensors and logic circuits. Organic molecules such as conducting polymers, proteins and pigments are being considered as alternatives for carrying out the same functions that are presently performed by semiconductors (e.g., silicon) and metals. Among them, conducting polymers (or conjugated polymers) have been considered as highly promising for molecular electronics [1-5]. These conducting polymers offer a unique combination of properties that make them attractive materials for use in molecular devices (Figure 13.1). The conductivity of these polymers can be tuned by chemical manipulation of the polymer backbone, by the nature of the dopant, by the degree of doping and by blending with other polymers. In addition they offer lightweight, processibility and flexibility. Because of these advantages, the use of conducting polymers in molecular electronics is rapidly evolving from physics, chemistry, biology, electronics and information technology. These molecular electronic materials differ from conventional polymers by having a delocalised electronic structure that can accommodate charge carriers such as electrons and holes. Besides this, these organic materials

393

Handbook of Polymers in Electronics

Figure 13.1 Applications of conducting polymers in molecular electronics

exhibit Peierl’s instabilities due to the built-in high anisotropic interactions and undergo substantial geometric modifications due to electronic excitations. This results in various charge transfer processes and a substantial degree of disorder leading to various localised states in the forbidden gap due to localisation. Conducting polymers exhibit the behaviour of both metal and semiconductor. Polyacetylene, the first and simplest electrically conducting polymer, doped with iodine (I2) has been shown to have an electrical conductivity of 1000 S cm-1 [2]. Some of the conducting polymers that have recently generated much interest are shown in Figure 13.2. It can be seen that conducting polymers behave as anisotropic semiconductors with band gaps in the range 1.4-3.2 eV. It has been suggested that electrical conduction in the conjugated polymers occurs via non linear (or topological) defects (solitons or polarons/biploarons) generated either during polymerisation or as a consequence of doping [6, 7]. Solitons can be produced as a result of an interruption in the structure of a degenerate conducting polymer such as trans-polyacetylene (Figure 13.3a). In other conducting polymers such as polypyrroles, polyanilines and polythiophenes, etc., the charge

394

Conducting Polymers in Molecular Electronics

Figure 13.2 Structures and band gaps of some important conducting polymers

395

Handbook of Polymers in Electronics

Figure 13.3 Energy band diagram of (a) soliton (b) polaron and (c) bipolaron

carrier is a polaron [8], a non linear defect that occurs as a result of localisation of charge in a polarised lattice of a conjugated polymer (Figure 13.3b). Due to increased doping, the concentration of non linear defects (solitons/polarons) increases resulting in an enhanced electrical conductivity. It has been shown that increased electrical conductivity in highly conducting polymers is spinless which results due to the formation of bipolarons (a bipolaron is a bound state of two polarons) [9, 10]. The concentration of bipolarons can be increased as a consequence of external stimuli (doping/heat/irradiation) on a conducting polymer. Solitons and polarons have recently been shown to have implications in the technical development of molecular electronic devices [11]. The development of molecular electronics is dependent upon the synthesis and tailoring of ‘active molecules’ and is a great challenge to researchers. Conducting polymers can be prepared both by chemical and electrochemical techniques. The nature of a monomer

396

Conducting Polymers in Molecular Electronics governs the selection of a method. For example, polyacetylene is prepared by polymerisation of acetylene using a Ziegler-Natta catalyst at low temperatures [12]. Heterocyclic polymers, such as polyaniline, polypyrrole and poly(p-phenylene), have been synthesised using the electrochemical technique [13-15]. Due to ease of handling and control of thickness, the electrochemical technique has been extensively used for technical development of molecular electronic devices. By varying the nature of the groups, specific interactions with external physical and chemical phenomena can be developed in these materials, leading to molecular devices such as transducers, memories and logic operators. Characterisation of conducting polymers is very important for investigating the electronic processes occurring in molecular electronic materials. A variety of techniques (electrochemical, optical, ESR, SEM, AFM, gel permeation chromatography (GPC)) have been widely used to delineate the physical properties of the conjugated polymers [16-20]. For example, the changes in the optical spectra accompanied with doping have been considered very significant in elucidating the mechanism of structural changes in the polymer chains. Information on morphological changes has been found very helpful towards the fabrication of lightweight batteries [21, 22]. Electrochemical characterisation provides information regarding redox behaviour, the number of electrons in the redox reaction and diffusion coefficient estimation of conducting polymers [23-25]. Thermal techniques such as DSC and TGA reveal valuable information on the thermal stability and degradation of these organic molecular electronic materials [26-28]. Time-of-flight (TOF) measurements have recently been used to estimate the magnitude of charge carrier mobility in conducting polymer systems [29]. It is emphasised that the experimental data accumulated as a result of characterisation plays a significant role in the application of a desired conducting polymer to molecular electronic applications. Conducting polymers have been shown to have a number of potential technological and commercial applications in optical, drug delivery, memory and biosensing devices [3034]. Among these, application of conducting polymers to molecular electronics has attracted the maximum attention. The major challenge confronting the materials scientists, including biochemists and physicists, is how the properties of these electronic materials differ from those of conventional semiconductors. Our group has been actively engaged in the research and development of conducting polymer-based molecular electronic devices for the past fifteen years [35-39]. It was, therefore, thought that a review based on the recent research findings will prove helpful to the researchers venturing to enter this highly fascinating field of molecular electronics.

13.2 Synthesis of Conducting Polymers Conducting polymers have been prepared by both chemical and electrochemical techniques. Conducting polyheptadiene has been obtained by cycling 1,4-heptadiene in the gas phase

397

Handbook of Polymers in Electronics by selective deposition on polyethylene and Teflon. Polyacetylene has been prepared by allowing acetylene to polymerise on the wall of a reaction flask on which the catalyst solution containing Ti-[O-n-C2H9]4 and [C2H5]3Al is coated [40]. Polymerisation on the wall of a reaction flask depends on the catalyst activity and preparative conditions such as the Al/Ti ratio, the polymerisation temperature, the acetylene pressure and the ageing condition. Polyacetylene can also be polymerised by the Durham route [41, 42], in which 7,8-bis(trifluoromethyl)tricyclodeca-3,7,9-triene is used as a monomer and CuCl6:(C6H5)4Sn (1:2) and TiCl4:(C2H5)3Al (1:2) are used as catalyst. Synthesis of poly(arylenevinylene) involves preparation of a water-soluble dimethyl sulfonium salt as a precursor monomer, its base-catalysed polymerisation and subsequent thermal elimination to yield the conducting poly(arylenevinylene). A solid-state polymerisation method makes use of monomer crystals using heat or light. Polydiacetylene and polysulfur nitride with perfect stereoregularity have been prepared using this technique. Besides this, polyaniline can be prepared using aqueous and non aqueous routes using chemical technique [43]. Photopolymerisation utilises photons to initiate a desired polymerisation reaction in the presence of photosynthesisers. Polypyrrole has been reportedly synthesised from pyrrole using a tris (2,2´-bipyridyl ruthenium) complex as a photosynthesiser. The advantage of this technique lies in the easier control of the polymerisation reaction to a desired surface. The oxidative coupling method is one of the best methods for obtaining high molecular weight poly(paraphenylene) [44]. Plasma polymerisation makes use of molecules occurring in various plasma environments. This technique results in very thin but uniform polymeric layers that strongly adhere to a desired substrate. It has been found that the plasmapolymerised films are usually highly crosslinked and are resistant to higher temperatures and chemicals. The method makes use of monomers to form cations followed by their coupling to form dications, and repetition of this process produces a conducting polymer. Polythiophene has been synthesised by both electrochemical [45] and chemical techniques [46]. Pandey and co-workers [47] have reported the chemical synthesis of poly(anilineco-orthoanisidine) that has been found to be soluble in common organic solvents such as acetone, chloroform and n-methylpyrrolidone. Wang and co-workers [48] later systematically investigated the effect of temperature on the synthesis and properties of poly(aniline-co-orthoanisidine). It has been revealed that this processable conducting polymer has application in molecular electronic devices. Electrochemical oxidative polymerisation is known to be an effective method for obtaining conducting polymers [49]. There are three methods: galvanostatic, potentiostatic and potentiodynamic (or cyclic sweep). In galvanostatic mode, a conducting polymer is generated by supplying a constant current between a working electrode and a counter electrode. Films produced by this method are smooth and adherent to the anodic surfaces. In the potentiostatic mode, the conducting polymer is grown at a predefined constant potential maintained between the counter and working electrodes. The selection of the

398

Conducting Polymers in Molecular Electronics constant potential for the synthesis of a conducting polymer (e.g., polyaniline) by this method is obtained from polarisation studies. It can be remarked that this method produces powdery deposits that are non adherent to the electrode surface. In the case of the potential cycling (i.e., cyclic voltammetry) method, the potential is cycled between two predefined potentials with a definite sweep rate. It has been found that this method gives homogeneous conducting polymer films that strongly adhere to the electrode surface. It can be seen that the electrochemical method is an advancement over chemical techniques because the resulting product does not need to be extracted. The flexible and smooth films can be prepared by the judicious selection of the conducting salt acting as an electrolyte. It has been ascertained that the charge and the geometry of the anions greatly affect the properties of a given conducting polymer. Apart from these, some general considerations such as the choice of the solvent, counterions [organic/inorganic] and also the substrate play a key role in the mechanical and electrical properties of conducting polymers prepared using electrochemical techniques.

13.3 Preparation of Ultrathin Conducting Polymer Films Ultrathin films of conducting polymers have been projected to have applications in molecular electronics [50, 51]. The approach lies in fabricating devices starting from atoms and molecules. This has been considered to be an attractive alternative if ordered structures are required. It is thought that an understanding of the physical properties of conducting polymer monolayers will prove very valuable in the evolution of the area of molecular electronics. It has been shown that it is possible to obtain by these techniques a system of ‘wires’ and ‘switches‘ comprising of conducting polymer monolayers [52, 53].

13.3.1 Langmuir-Blodgett Films The preparation of polymeric Langmuir-Blodgett films with different characteristics is of great scientific and technological interest. Materials that form monolayers on the surface of water comprise of molecules that contains both hydrophilic (water-attracting) and hydrophobic (water-repelling) chemical groups. These materials are known as amphiphiles. Langmuir-Blodgett films are formed by first depositing a small quantity of an amphiphilic material (stearic acid) dissolved in a volatile organic solvent onto the surface of purified water (sub-phase). On evaporation of the solvent, the pressure-area isotherm (Figure 13.4) is recorded by compressing the monolayer mechanically [54, 55]. This results in the formation of the floating organic material into a ‘two-dimensional solid’. And since the amphiphile has hydrophobic and hydrophilic ends, the compression causes the molecules to be aligned in the same way on the surface of water (Figure 13.5).

399

Handbook of Polymers in Electronics

Figure 13.4 Pressure-area isotherm of an Langmuir-Blodgett film of poly(3-hexylthiophene)

Figure 13.5 Possible orientations of a molecule deposited by Langmuir-Blodgett technique

The molecules in their closest packed arrangement (solid phase) are removed from the surface of water by suitably dipping and raising a suitable plate (substrate) through the air/water interface in three ways (Figure 13.6). If the substrate is hydrophilic the first monolayer is transferred as the substrate is raised through the sub-phase, these stack in a head-to-head and tail-to-tail configuration. This deposition mode is referred to as Ytype deposition. This results in an odd number of monolayers being transferred onto the solid substrate. However, if the solid substrate is hydrophobic a monolayer will be deposited as it is first lowered into the sub-phase, thus a Y-type film containing an even number of monolayers can be fabricated (Figure 13.7). If a layer is deposited on the substrate only when the solid substrate enters the sub-phase, this is called X-type deposition. On the other hand if a layer is deposited on the substrate when withdrawn

400

Conducting Polymers in Molecular Electronics

X

Y

Z

Figure 13.6 X-, Y- and Z-type deposition in Langmuir-Blodgett film deposition

Figure 13.7 Schematic of Y-type deposition in a Langmuir trough. (a) Compressed monolayer of molecules on water surface, (b) Deposition of first monolayer on withdrawal of hydrophilic substrate, (c) Deposition of second monolayer on insertion of hydrophilic substrate, (d) Deposition of third monolayer in a head-to-head and tail-to-tail configuration

401

Handbook of Polymers in Electronics from the sub-phase, it is called Z-type deposition. Film deposition can be characterised by knowing the deposition ratio, τ, as given by Langmuir and co-workers [56]:

τ = Al / As

(13.1)

where Al is the decrease in the film area of the sub-phase and As is the contact area of the substrate. Various deposition modes can be obtained using a parameter φ given by Hoing and coworkers [57]:

φ = τu / τd

(13.2)

where τu and τd are the deposition ratio on the upward and downward passage, respectively. For the Y-type deposition, φ =1, for X-type, φ = 0, and for Z-type deposition, φ = ∞. Many factors such as temperature, humidity and pH (aqueous phase) can influence measurements on interfacial films. Langmuir-Blodgett films offer ways of studying molecular conformational changes and providing information on molecular packing, crosslinkage and denaturing. Langmuir-Blodgett films are ideal systems for spectroscopy of complex monolayers. Energy transfer between excited molecular states and model membranes has been used to mimic photosynthetic systems. It has recently been demonstrated that the Langmuir-Blodgett technique can be utilised to process a variety of conductive polymeric systems into multilayer films. The conjugated polymer backbone provides a route for the flow of electrons either along the molecule or between nearest neighbours, the latter being by way of charge-transfer interactions. Langmuir-Blodgett films of a number of conducting polymers such as polypyrroles, polyanilines and poly-(o-anisidines), etc., have recently been prepared. Hoing and co-workers [58] have shown that electrically conducting polypyrrole films can be formed at the airwater interface of a Langmuir-Blodgett trough using a solution containing a surface-active pyrrole monomer and a large excess of pyrrole on a sub-phase containing ferric chloride (FeCl3). They showed that the chemistry initiated at the air-water interface and the properties of the resultant polymer are strongly influenced by the type of the surface-active monomer used. Ultrathin films of 3-octadecylpyrrole (3ODP) and 3-octadecanoylpyrrole (3ODOP) were subsequently obtained using Langmuir-Blodgett technique. The pyrrole/3ODOP film was found to be highly aniosotropic with conductivity in the plane being 107 times greater than the conductivity across the film thickness. Bardosova and co-workers [59] have investigated the formation of polythiophene Langmuir-Blodgett films on silicon. The results of atomic force measurements conducted on a five-layer polythiophene film revealed that the rearrangement of molecules occurs resulting in a fibrous structure.

402

Conducting Polymers in Molecular Electronics Aghbor and co-workers have obtained films of deposition of preformed polyemeralidine base (PEB) by dissolving PEB in a N-methylpyrrolidone/CHCl3 mixture in an aqueous subphase containing acetic acid [60]. Ram and co-workers [61, 62] subsequently showed that it is possible to obtain quasi-ordered Langmuir-Blodgett films of emeraldine base without incorporating fatty acid tails in the molecule. However, the results of cyclic voltammetery and chronopotentiometry indicate that irregularities begin to form in the film and the ordered nature of the films is lost, resulting in its reduced electroactivity. Dabke and co-workers [63] studied the electrochemistry of polyaniline Langmuir-Blodgett films using cyclic voltammetry coupled with a quartz microbalance. It was found that the multilayer films exhibit poor electrochromic response. These results have implications in the fabrication of molecular devices. Recently, ultrathin films of poly(o-anisidine) and poly(ethoxyaniline) have been fabricated for application in nanotechnology [64-66]. Matsura and co-workers [67] have fabricated monolayers of β-carotene using a Langmuir-Blodgett film technique together with the flow-orientation method. They have utilised XRD, UV-visible and FTIR techniques to elucidate the film-structure of β-carotene indicating that β-carotene orients perpendicular to the air-water interface. It was found that the films are, however, well-ordered both in the stacking direction and the in-plane direction. The ability to tailor properties of conducting polymer films has been of advantage for several applications. The best-known example is of liquid crystalline polymers that have potential applications in electro-optic applications such as calculators, wristwatches, message boards, flat panel televisions, waveguide switches, real-time optical data processing systems, etc. The molecular order existing in Langmuir-Blodgett multilayers facilitates the formation of liquid crystalline polymer substances since some of the conditions are easily met. These conditions are: •

The volume contraction during the chain formation should not be in the direction of chain growth,

•

The overlapping volumes of monomer units in the polymer chain should not much differ from that of the monomer molecule, and

•

The chain period of the polymer must coincide with a transitional period of the monomer crystal lattice, which are required for a polymerisation.

Conducting polymer Langmuir-Blodgett films find applications in insulation, adhesion and encapsulation. The molecular electronic applications of conducting polymer Langmuir-Blodgett films include microactuators, high-density information storage, highdefinition television, Schottky devices, biosensors and chemical sensors. It may be emphasised that the industrial use of each of these applications is closely linked with manipulating the architecture of a conducting polymer via ‘molecular engineering’. It is

403

Handbook of Polymers in Electronics appropriate to mention here that the Langmuir-Blodgett technique provides an opportunity to control the thickness and supermolecular organisation of electroactive polymers at the molecular level for application to the potential field of molecular electronics.

13.3.2 Self-Assembly Monolayers One of the methods of formation of organised monolayer assemblies (OMAs) on the desired solid surface has been described in the preceding section. Another process is the method of self-assembly wherein the molecules are transferred to the surface of a solid from the liquid phase by a dipping process. These self-assembled monolayers have potential for technological and scientific applications ranging from microelectronics to biological sensors [68]. Monolayers of polythiophene and substituted polythiophene have been formed on gold substrates by a dipping method [69]. Multilayered thin films with precisely controlled thickness and layer sequences can also be obtained by this technique [70-76]. Another approach to fabricating self-assembled monolayers is by depositing alternating layers of p- and n-type polymers on a suitable substrate. Here the molecules are held by the electrostatic attraction of the alternatively charged polymers. Converting the polymer into a polycation or a polyanion by using the appropriate acidic solution produces the charges. For example, this technique has been utilised for preparing multilayers of polyaniline and polypyrrole samples with uniform ordering [68]. Different conjugated and non conjugated polymers in the form of bilayers have been used to make heterostructures.

13.4 Characterisation of Conducting Polymers Characterisation of a conducting polymer is important before it can be used for any technological application. Physical and chemical properties of conducting polymers have been investigated by a number of experimental techniques. The molecular weight of a conducting polymer has been estimated by GPC. Thermal techniques such as DSC/DTA have been utilised to check the stability of conducting polymers in air. ESR measurements have been found to reveal information on the phenomenon of doping in electrically conducting polymeric systems. It has been shown that the origin of magnetic properties in a conjugated polymer lies in its π-electron system, which has also been considered as the reason for the observed chemical properties [77]. Dielectric relaxation and low frequency conductivity measurements in the frequency range from 100 to 107 Hz have proven valuable in giving information on the mechanism of conductivity that cannot be obtained with dc conductivity data alone [78]. The morphology of conducting polymer films has been investigated using optical microscopy, SEM/TEM, scanning tunneling microscopy (STM) and AFM techniques [79-82]. Studies conducted on the morphology

404

Conducting Polymers in Molecular Electronics of conducting polypyrrole films containing different anions such as NO3–, F–, ClO4–, BF4– and CH3C6H4SO3–, respectively, have revealed that the topology of the growing conducting polypyrrole surface is influenced by the nature of the electrolyte [83-84]. For instance, it has been shown that the fibrillar structure of polyacetylene is often advantageous since it can store up to about 7% of electrical charge [85]. AFM and STM methods [86] have recently provided valuable information on the presence of microdomains in poly(3-hexylthiophene)/stearic acid films (Figures 13.8 and 13.9).

Figure 13.8 Atomic force micrograph of conducting poly(3-hexylthiophene)

Figure 13.9 Scanning electron micrograph of conducting poly(3-hexylthiophene)

405

Handbook of Polymers in Electronics Spectroscopic methods have been used to yield information on the charge transport behaviour in conjugated polymers. On interaction of a photon with a conducting polymer, electrons get excited to a higher potential with concomitant creation of electron-hole pairs. The application of spectroscopic techniques in the IR and UVvisible regions to conducting polymers has brought interesting information on the configurational changes arising as a consequence of doping. Photoelectron spectroscopy methods have been used as probes to investigate electronic and chemical structures of conducting polymers such as polypyrroles and polythiophenes. 13C NMR studies have been conducted on polyazulene, polybithiophene and polyfuran in their respective doped and undoped states [87]. The 13C NMR data obtained on the electrochemically prepared polypyrrole indicate the presence of α-α′ bonding in conducting polypyrrole films [88]. The ellipsometric technique has been used for the estimation of thickness, refractive index and optical dielectric constants of a conducting polymer film [89]. These conducting polymer films have been used for the fabrication of electrochromic and glucose biosensing devices [90, 91].

13.5 Molecular Devices Based on Conducting Polymers 13.5.1 Diodes There has been an increased interest towards the possible applications of conducting polymers as the active elements in electronics [92-95]. The characteristics of conducting polymer/inorganic semiconductor interfaces have been considered as very important since it has been indicated that restrictions on Schottky barrier devices can be overcome by using conducting polymer contact layers. Besides this, the ability to manipulate the interface characteristics by changing the polymer dopant allows switchable devices to be fabricated. Semiconducting polymers such as polyacetylene, polypyrrole and polyaniline have recently been used for the fabrication of Schottky barrier diodes, like metal-insulatorsemiconductor (MIS) diodes and p-n junction diodes [96-98]. A schematic diagram for a Schottky device is shown in Figure 13.10. Schottky diodes formed between metallic AsF5– doped (CH)X and n-type GaAs indicate high electronegativity [99]. The polyacetylene has been found to exhibit p-type behaviour when it is used for fabrication of Schottky diodes with low work function metals. Heterojunctions have recently been fabricated using electrochemically prepared polypyrrole and metal (indium, titanium, aluminium and tin). It has been revealed that carrier concentration, estimated as 1.5 x 1020 and 5 x 1017 in doped and undoped polypyrrole samples, respectively, plays an important role in controlling the junction

406

Conducting Polymers in Molecular Electronics

Figure 13.10 Schematic diagram of a conducting poly(3-alkylthiophene) based Schottky device

characteristics and hence the performance of these Schottky devices [100]. The electrical characteristics of the junctions have been found to be dependent on the work function of polypyrrole, estimated as 4.42-4.49 eV. It has been recently revealed that it is possible to fabricate all vacuum deposited metal (Pb, Al, In, Sn)/polyaniline/metal Schottky devices [101]. It has been shown that the barrier height and the ideality factors determined are dependent on the work function of the metal used in the fabrication of these devices. The improved ideality factor obtained as 1.2 for an Al/polyaniline/Ag device has been attributed to more intimate contact of the metal with the vacuum deposited polyaniline electrode. Poly(3-methylthiophene) was prepared using an electrochemical technique to fabricate silicon-based devices [102]. From the results obtained using current-voltage measurements, chronopotentiometery, SEM and FTIR techniques, it has been seen that the rectifying behaviour was induced by covalent bond formation between poly(3methylthiophene) and the silicon. Schottky devices have recently been fabricated by thermal evaporation of indium on polyaniline, poly(o-anisidine) and poly(aniline-co-orthoanisidine), respectively [103]. The values of the rectification ratio, the ideality factor and the barrier height of an indium/poly(o-anisidine) have been experimentally determined as 300, 4.41 and 0.4972, respectively. The observed deviation from the Schottky behaviour for these devices seen at higher voltages has been explained in terms of either the Poole-Frenkel effect or due to the presence of a large number of defects containing the trapped charges existing at the indium/poly (aniline-co-orthoanisidine) interface.

407

Handbook of Polymers in Electronics Recently, the junction properties of Schottky devices using films of chemically synthesised poly(3-cyclohexylthiophene) and poly(3-n-hexylthiophene) units and metals have also been studied [104]. Electrical properties of the poly(3-cyclohexylthiophene)/metal junctions were compared with those of the poly(3-n-hexylthiophene)/metal junctions (Figure 13.11). Better rectification properties of the poly(3-cyclohexylthiophene)/metal junctions were attributed to the decreased conductivity that perhaps results due to steric hindrance in the thiophene ring.

Figure 13.11 Current-voltage and capacitance-voltage characteristics of metal/ conducting polymer junctions

MIS structures were recently fabricated [105] by thermal deposition of metals (indium, aluminium and tin) on Langmuir-Blodgett films of cadmium stearate (CdSt2) obtained on polypyrrole films electrochemically deposited onto ITO glass. Junction parameters, like barrier height, rectification ratio and work function, of these devices were experimentally determined. The ideality factors of the CdSt2 layer/semiconducting polypyrrole structures have been estimated as 6.63, 6.57 and 6.54 for tin, aliminium and indium, respectively, in comparison to the values of 8.85, 8.82 and 8.20 obtained with a semiconducting polypyrrole interfaced with the same elements. It has been concluded that the passivation of the semiconducting polypyrrole results in the lower value of the ideality factor.

408

Conducting Polymers in Molecular Electronics

13.5.2 Field-Effect Transistor In conventional transistors, ‘field effect’ has been used to improve device characteristics [106]. This ‘field effect’ controls the current through a ‘gate’ electrode and thereby opens the possibility of transistor action without requiring the existence of p-n junctions. This phenomenon is useful not only for fabricating devices but also a very useful tool for studying semiconductor and surface states [107]. The application of field effect to fabricate conjugated polymer based device was first demonstrated by Koezuka and co-workers [108] using electropolymerised polythiophene. When charge is induced in the polymer layer by applying a voltage to the gate, the conductivity of polymer changes, which in turn controls the drain current flowing between ‘source’ and ‘drain’. After Koezuka, many researchers applied various conducting polymers to construct field-effect transistor (FET) devices [109-111]. Simultaneously, they exploited thin film fabrication methods [112]. Tsumura and co-workers have fabricated a molecular electronic device based on an electrochemically prepared semiconducting polythiophene thin film [113]. They have shown that this solid-state FET device is normally off (enhancement type) and the source (drain) current can be modulated by more than 102 by varying the gate voltage. The transconductance and the carrier mobility have been estimated as 3 nS and 10-5 cm2 V-1 S-1, respectively, using electrical measurement technique. MIS field-effect transistors (MISFETs) and enzyme field-effect transistors (ENFETs) based on conducting polymers have also been fabricated. A schematic diagram of such a device is shown in Figure 13.12. Janata and co-workers have investigated the electrical properties

Figure 13.12 Schematic diagram of an ion-selective field-effect transistor, VG = gate voltage, VD = drain voltage, ID = drain current

409

Handbook of Polymers in Electronics of insulated gate field-effect transistors (IGFETs) using a chemically prepared polyaniline as a gate electrode [114]. The response of the polyaniline insulated gate field-effect transistors (PANI-IGFETs) to step function potential has been experimentally determined. It has been found that the dc behaviour of these devices is the same as that of the corresponding MOSFETS (metal oxide semiconductor field effect transistors). Saxena and co-workers [115] reported an ion-selective microelectrochemical transistor (ISMET) specifically responsive to Cu(II) ions using electrochemically prepared polycarbazole films. The device turns on by adding 2.5 x 10-6 M Cu(II) ions and reaches a saturation region beyond 10-4 M Cu(II) ions (Figure 13.13). In the above concentration range, the device response is linear. The increase in drain current, I D, with increase in Cu(II) ion concentration was attributed to the conformational changes in the polymer matrix which arise due to the occupation of Cu(II) ions in the polymer matrix.

Figure 13.13 ID-[conc.] plot of an ion-selective microelectrochemical transistor

The field-effect mobilities in these FET devices are found to be around 10-5 cm2 V-1 s-1 depending on the applied voltages and the nature of the gate insulator. These values are significantly lower than that of inorganic semiconductor devices in which mobilities are found to be in the range of 0.1-1 cm2 V-1 s-1 [116]. Work on sexithienyl based FETs yielded more promising values, the field-effect mobility reaching 3 x 10-2 cm2 V-1 s-1

410

Conducting Polymers in Molecular Electronics [117], which is about one order of magnitude lower than that of an amorphous Si-based FET [118]. In the FET configurations, the interface of the gate insulator and organic semiconductor layer plays a crucial role in charge transport [110, 119]. It was also found that an oligothiophene (sexithiophene) showed a mobility of 0.46 cm2 V -1 s-1 in cyanoethylpullulan (CYEPL) which is used as a gate insulator, the mobility being almost a thousand times higher than that measured using SiO2 as the gate insulator [120]. Waragai and co-workers reported a similar trend; the highest mobility recorded was 3 cm2 V-1 s-1 for dimethylsexithiophene (CMSxT) on CYEPL, which was a hundred times more than that measured using SiO2 [121]. It is to be noted that these values of the mobility are comparable to or even surpass that of amorphous silicon [122]. Further, Garnier and coworkers have reported fully plastic FET by using a printing technique [123]. A recently proposed device comprising of ‘polymer grid triodes’ by Yang and Heeger is worth noting [124]. The performances of conjugated polymer based FETs are quite encouraging and could be used to control a pixel in a liquid crystal display. Lately, Friend and co-workers demonstrated a high mobility conjugated polymer FET driving a polymer LED [125]. It can be seen that conjugated polymers show a clear superiority over amorphous silicone if the cost is taken into account. However, lifetime and low response of the device restrict the competition of these materials with silicon-based devices. Nevertheless, the possibility of making a flexible panel with conjugated polymers opens a different field of large-area, low-cost plastic electronics.

13.5.3 Biosensors One of the most exciting areas of research in molecular electronics lies in the development of biosensing devices (usually called biosensors or receptrodes). The term has been defined as an analytical device incorporating a biological or biologically derived sensing element either integrated or intimately connected with a physicochemical transducer. The aim is to produce either discrete or continuous electronic signal(s) that is (are) specific to a single analyte (or a related group of analytes). A biosensor can be considered as a combination of an electrochemical or an electrical sensing device and a miniaturised reactor containing an immobilised biomolecule, and is used in most cases to measure the concentration of a substrate. A number of biomolecules such as enzymes, antibodies, organelles, cells and receptors have been used as sensing probes for the fabrication of biosensors. The transducer can either be an electrical, optical, thermal or piezoelectric device. The electrochemical method of detection has been at the centre of biosensor development. The potentiometric technique relates to the dependence of the potential on the analyte concentration, whereas in amperometric biosensors current is based on heterogeneous electron transfer reactions, i.e., the oxidation and reduction of electroactive substances. A schematic of a biosensor is shown in Figure 13.14.

411

Handbook of Polymers in Electronics

Figure 13.14 General principle of a biosensor

Much interest in biosensors has been attached to the potential application of this highly expanding field of research to the solution of a variety of problems that occur in clinical diagnostics, food industry, agriculture and environment industry. Some of the other reasons that have made this area very attractive include those (i) that provide quick, selective and reliable information on the measurement species, (ii) that yield measurable signal, (iii) that require minimum pretreatment of the samples, (iv) that are inexpensive and can be used repetitively. Various publications in recent years indicate organic conducting polymers as a convenient tool for the immobilisation of enzymes at the electrode surface and its interaction with metallic or carbon electrode surfaces. The application of conducting polymers in analytical chemistry has recently been reviewed [126-130]. Some other reviews have been devoted to their use in design of biosensors [131, 132]. Immobilisation of biomolecules on the surface of an effective matrix with maximum retention of their biological recognition properties is a crucial problem for the commercial development of a biosensor. Different methods of immobilisation have been used. One such method is electrochemical entrapment. Several conducting polymers can be deposited electrochemically and, in the process, a biological molecule can be entrapped. This process is also useful in the fabrication of microsensors in preparation of a multilayered structure with one or more enzymes/biomolecules layered within a multilayered copolymer for analysis of multiple analytes [133-135]. A number of reports have appeared on immobilisation of biomolecules using electrochemical entrapment [130, 131, 136-143]. Another procedure for the immobilisation of biomolecules is the covalent binding of enzyme to a conducting polymer film. This is essentially a two-step procedure based on the formation of a functionalised conducting polymer film followed by the covalent binding of the enzyme at the functional groups on the polymer surface. The major advantage associated with this procedure lies in the independent optimisation of conditions required for the synthesis of the polymer with respect to solvent, electrolyte salts, film thickness and side reactions. This is

412

Conducting Polymers in Molecular Electronics followed by immobilisation of the desired biomolecule (e.g., enzyme) using an appropriate procedure in order to preserve the active state of the enzyme. Covalent coupling of an enzyme with a polymer can be carried out using carbodiimide coupling with the formation of peptide bonds [126, 144-146]. Bovine serum albumin and glutaraldehyde coupling [147] and various bifunctional reagents have been used to form a crosslinked network of enzyme, with the crosslinking reagent on the surface of a conducting polymer [148-151]. Others report chemical grafting or affinity of the biomolecule at the functional group [152-155]. Conducting polymers such as polypyrrole [127] and its derivatives [156, 157], polyaniline [158-164], polyindole [137] and poly-o-aminobenzoic acid have recently been used for the fabrication of biosensors. A few biosensors based on insulating electropolymerised films like polyphenols, poly(o-phenylenediamine), poly(dichlorophenolindophenol) and overoxidised polypyrrole have also been elaborated [165-167] . The glucose biosensor remains the most extensively studied biosensor. Lowe and coworkers [156] entrapped glucose oxidase in a polypyrrole matrix electrochemically deposited on a printed platinum electrode. Cooper and co-workers have reported the electrochemical preparation of a glucose oxidase loaded polyaniline film [157]. It has been found that polyaniline films exhibit enhanced loading of glucose oxidase after a self-exchange and hence can be used for the fabrication of third-generation glucose biosensors [37]. Some recent reports on polyaniline-glucode oxidase interaction include the work of Ozden and co-workers [168] and Karyakin and co-workers [169]. Morphological studies have revealed that the surface of a conducting polymer is critically dependent on the method of preparation [170] and plays an important role in the effective immobilisation of desired enzyme [171, 172]. The Langmuir-Blodgett technique for monolayer deposition could be very successful in achieving the desired orientation of the molecule. Few reports have appeared on this subject [173]. Okhata and co-workers have used Langmuir-Blodgett films of GOD-stearic acid monolayers on a platinum electrode to fabricate an ultrathin glucose sensing membrane [174]. Besides this, direct preparation of Langmuir-Blodgett films of GOD without the need to add a lipid layer has also been demonstrated by crosslinking with glutaldehyde. Ramanathan and co-workers [171] have successfully used a physical adsorption technique to immobilise glucose oxidase in LangmuirBlodgett films of polyemeraldine base (PEB) deposited onto ITO glass. These GOD-PEB films have been seen to result in the linear increase of anodic current as a function of glucose concentration (5 to 50 mM). Other biomolecules, such as antibodies [175-177], DNA [178-180], etc., have attracted much interest in the development of biosensors useful for detection of viruses and genetically transmitted diseases. Few reports have appeared on immobilisation in conducting polymers. A full range of optical, electrochemical and piezoelectric transduction modes aimed at detecting the base pair hybridisation between the immobilised

413

Handbook of Polymers in Electronics cDNA probe and the target DNA have been developed. Mediated electron transfer reactions of DNA for detecting PCR (polymerase chain reaction) amplified genomic DNA and ferrocene mediated oligonucleotides for a sandwich-based electrochemical detection of DNA hybridisation have been described. Immobilisation of DNA onto glass, carbon and gold electrodes has also been reported [181, 182]. Chaubey and co-workers have given a detailed account of the application of conducting polymers to biosensors in Chapter 10.

13.5.4 Electronic Tongue Taste is produced when interactions between molecules and biological membranes are not specific to or characteristic of each molecule. The five kinds of basic taste qualities reflect the differences among these interactions. What is important in recognition of taste is not discrimination of minute differences in amounts of molecules but rather the transformation of molecular information contained in interactions with biological membranes into several kinds of groups, namely taste intensity and quality. Taste comprises of five basic qualities: •

Saltiness produced by NaCl,

•

Sourness produced by the hydrogen ions of HCl, acetic acid, citric acid, etc.,

•

Bitterness produced by quinine, caffeine, L-tryptophan and MgCl2,

•

Deliciousness produced by monosodium phosphate and disodium inosinate in meat and fish, and by disodium guanylate in mushrooms, and

•

Sweetness due to glucose, fructose and sucrose.

Substances producing taste are received by the biological membrane of gustatory cells in taste buds on the tongue. Taste is perceived when the information on the substances is transformed into electrical signals that are transmitted to the brain via nerve fibre. The taste sensor employs the concept of global selectivity, implying the ability to respond to many kinds of chemical substances at the same time. Wine has both taste and odour qualities due to the aromatic molecules in the liquid and vapour phases being different. Most wines contain about 8%-85% water and 500 other substances, some of which are very important to the flavour in spite of low concentrations. The overall perception of taste is known to be due to a combination of taste senses and smell, and also called flavour or trigeminal sensor. Sussi and co-workers have designed

414

Conducting Polymers in Molecular Electronics an odour-sensor array of different conducting polymers [183]. Polymerisation occurs when the monomer (25 mg) and the oxidising salt is sprayed onto four interdigited electrodes fabricated onto the alumina substrate. The electrical resistance measured between different inner electrodes has been found to vary from 1 to 100 kΩ when volatile molecules are adsorbed at the surface of the conducting polymer film. The average intensity, defined as the ratio of the resistance change to the base resistance value, was estimated to be less than 2% for elements related to wine sensing. Toko [184] has fabricated a multichannel taste sensor, with global selectivity comprising of several kinds of lipid/polymer membranes for transforming information about substances producing taste into electrical signals. The electronic tongues or taste sensors are based on the principle of potentiometry, i.e., the charging of the membrane is measured. A voltammetry technique has several advantages due to features such as very high sensitivity, simplicity, versatility and robustness. An electronic tongue, working on the principle of pulsed voltammetry based on an array of five working electrodes, has recently been designed to follow the deterioration in the quality of milk due to microbial growth when milk is stored at room temperature [185]. Depending on the purpose and on the object to be measured, it is possible to miniaturise the taste sensor. Accordingly, a taste-sensing fieldeffect transistor (TSFET) has recently been designed [186] for estimation of chemical substances in foodstuffs and organisms. A number of sensors for the measurement of odour and deliciousness based on conducting polymers are currently at various stages of technical development. It appears that we are gradually entering into a new age of ‘food culture’.

13.5.5 Electronic Nose The complicated olfactory system in humans and animals can detect and differentiate the presence of an odour even at trace levels [187]. Sensory evaluation is one of the important parameters for environmental monitoring, quality assessment for food, wine and beverages, and clinical diagnosis, as well as for the control of many cosmetics and fermentation processes [188-190]. Typically, sensory evaluation in odour as well as food/ wine testing is performed by a panel of well-trained professionals based upon their sense of smell, taste, experience and mood. However, the human olfactory system is very sensitive but not selective. The human nose neither tries to break the aroma into different constituents nor to quantify the constituents. Unfortunately, there is no effective instrumental analysis to replace the human sense. Analytical instruments such as gas chromatography (GC)/mass spectrometry (MS), high pressure liquid chromatography (HPLC) and NMR spectroscopy could be used to monitor the particular compounds present in a variety of samples. However, their use is

415

Handbook of Polymers in Electronics impractical for monitoring unknown odorous compounds. The processes suffer from tedious, time-consuming and expensive works, and more importantly, a lack of a direct correlation between the instrumental results and human perception. These instruments probably cannot detect threshold (10-9 ppb) levels of some odorous compounds, e.g., αterpinethiol in water [191]. Consequently, there is an enormous demand for an electronic instrument that can mimic the human sense of smell and provide low-cost, rapid, sensory information. The earliest work on the development of an instrument specifically to detect odour dates back to 1961, when Moncrieff reported a mechanical nose. The first report on an electronic nose was from Wilkens and Hatman in 1964 [192]. However, the term ‘electronic nose’ appeared around the late 1980s when one was demonstrated at a conference in 1987 [193]. The electronic nose can be used to monitor and measure odour anywhere, such as in environmental monitoring, mining, clinical diagnosis, food and drinks, household products, healthcare or pharmaceutical products, perfumery, tobacco and smoke. An electronic nose is often a bench-mounted instrument comprising of an array of electronic chemical sensors with partial specificity and an appropriate pattern-recognition system, capable of recognising simple or complex odours [194]. Several technologies such as tin oxide, quartz resonator and surface acoustic wave (SAW) devices have been proposed for the technical development of electronic nose for the analysis of vapours and different gases. These devices operate at elevated temperatures (e.g., 100-600 °C). They are quite sensitive to combustible materials such as alcohols but are generally poor at detecting sulfur- or nitrogen-based odours. Although there are some oxide materials that show a good specificity to certain odours, there are several potential advantages to employing organic materials in an electronic nose. Of these materials, the most interesting are conducting polymers, which are sensitive but not selective. Conducting polymers are highly suited as odour sensing devices: •

These sensors exhibit rapid adsorption and desorption kinetics at room temperature.

•

The polymer used can be made highly specific to chemical substances.

•

They are highly insensitive to poisoning from sulfur-containing compounds.

•

There is low power consumption, unlike tin-oxide sensors, and so no heater is required.

•

They are easier to process than oxides, and spin casting, electrochemical, screenprinting and Langmuir-Blodgett methods could be employed to fabricate sensors. Organic polymers like polypyrrole can be readily patterned using standard integrated circuit technology.

•

These sensors can be operated close to room temperature.

416

Conducting Polymers in Molecular Electronics Different conducting polymers produce different responses when exposed to complex vapours [195, 196]. The conducting polypyrrole is, however, the most commonly used, although some work on polyaniline has also been reported [197]. To date several electronic noses are available commercially or close to it. An example is the Odour Mapper (UMIST Ventures, UK), which consists of an array of twenty conducting polymer chemoresistors. Similarly, Neotronics, UK, has designed an electronic nose comprising of two conducting polypyrrole electrodes separated by 10 μm [198]. The anions diffuse between the chains due to the relatively weak interchain bonding of the polypyrrole whereas the cations form part of the chain. The size of the anion and the spacing between the polymer chains define how the solvent is dispersed around the chains. The resistance of the sensor (arising due to the motion of charge carriers along the chains) varies with the changes in the concentration of the exposed vapours. The variation of change in resistance recorded from 10 sensors when exposed to a 4% ethanol solution is shown in Figure 13.15. Attempts have also been made to use the polypyrrole nose to detect several other vapours such as isobutyl-3methoxypyrazine, trichloroanisole, vanillin, isoalleraldehyde and methanthiol [199]. The restriction to the electron flow when a trifluoromethyl molecule is attached directly to the polypyrrole chain is shown in Figure 13.16. It may thus be remarked that the polypyrrole sensors respond to vapours in a similar way, though not as effectively as the human nose. These electronic noses are for specific purposes and there is presently no universal nose that can solve all odour sensing problems. There is thus a need to develop specific electronic nose technology appropriate for the application. This means developing sensors, materials and appropriate pattern-recognition methods. There is thus a wide scope for the development of an artificial nose, based on conducting polymers, that can mimic the human nose.

Figure 13.15 Profile of the variation in resistance with 10 different sensors for 4% ethanol solution

417

Handbook of Polymers in Electronics

Figure 13.16 The restriction of electron flow in polypyrrole when trifluoromethyl molecule is attached directly to the polymer chain

13.5.6 Nanowires Nanomaterials have been found to have several technological and commercial applications including use in electronic, optical, drug-delivery and biosensing devices. The application of conducting polymers as nanomaterials for electronic communication between redox enzyme and electrode surfaces is one of the active areas in biomolecular electronics. Martin and co-workers have explored the electronic, optical and electrochemical properties of electrochemically synthesised polypyrrole within the pores of nanoporous polycarbonate filtration membrane walls [200]. Besides this, these researchers have accomplished the template synthesis of a number of polymers on a nanoscale [201, 202]. It has been suggested that organic microtubules can perhaps serve as a useful replica of various biological systems. In this context, Malbers and co-workers [203] have shown that heteroarene oligomers comprising of two pyridinium groups, linked by thiophene units of variable length, thienoviologens, are promising candidates for molecular wires. Glucose oxidase and choline oxidase exhibit strong adsorption to these conductive layers obtained on gold electrodes. Entrapment of conducting polymers such as polyacetylene, polypyrrole, polyaniline and polythiophene in sol-gel films has recently been proposed [204]. It has been shown that it is possible to entrap conducting polymers in porous structures such as alumina, sol-gel films and polycarbonate membranes. In this context, it has recently been demonstrated that polyaniline can be electrochemically entrapped into the tetraethylorthosilicate (TEOS) matrix obtained on an ITO glass [205]. Cyclic voltammetry, UV-visible and IR spectroscopy and SEM studies have been used to detect the presence of electroactive polyaniline in a TEOS-derived sol gel matrix. The results of these studies indicate that conducting polymers can be utilised as ‘molecular wires’.

418

Conducting Polymers in Molecular Electronics

13.5.7 Electroluminescent Displays The current interest in exploring the semiconducting properties of conducting polymers is for application to electroluminescence, i.e., using these materials as an emissive layer in LEDs [125, 206-211]. Electroluminescence is the emission of light by electrical excitation. Pope and co-workers [212] observed emission in single crystals of anthracene using silver paste electrodes at 400 V. Subsequently, it was established that the phenomenon of electroluminescence necessitates the injection of electrons from one electrode and holes from the other, the capture of one by the other (recombination) and the radioactive decay of the excited state (exciton) produced by the recombination process. Current flat-panel display technology primarily revolves around inorganic LEDs, backlit liquid crystal displays (LCDs) and vacuum fluorescent displays. Although these technologies are trim and efficient compared with cathode ray tubes, they can be bulkier and much more power consuming than required for many applications. In many batteryoperated devices, such as laptop computers, cellular telephones and other hand-held instruments, the illuminated display is the primary energy consumer. Furthermore, LED displays are expensive to fabricate because of the many individual diodes required to make up an alphabet, each with its own contacts and interconnections. These problems forced researchers to look for other materials. Conducting polymers have been considered to have outstanding potential for replacing inorganic light-emitting materials such as used in large area, lightweight, flexible displays. As compared to conventional fluorescent materials, conducting polymers offer the following advantages: •

The characteristics of conducting polymers can be altered either by modifying the electron structure or by chemically altering the polymer backbone.

•

They operate with low dc voltage and are less power consuming compared to conventional LEDs or LCDs.

•

They can be processed in the form of thin films and have potential for the production of flexible devices.

•

They can be uniformly illuminated over a large area.

•

Output colours can span the whole visible spectrum.

•

The polymeric materials are available at low cost.

419

Handbook of Polymers in Electronics Owing to the above advantages, a number of conducting polymers have been produced that emit light across the visible spectrum and which may be used to fabricate devices with greater quantum efficiencies [31, 213-215]. Poly (f-phenylene vinylene) (PPV) gives emission in the yellow-green region. Red emission can be observed by the substitution of electron-donating groups at the 2- and 5-positions on the phenyl ring [216]. Electroluminescence has been reported in different conjugated polymers such as poly (vinyl carbazole) (PVCZ), polyalkylfluorene [217, 218] and fluorinated polyquinoline [219]. Many efforts have been made to improve luminescence yields in conjugated polymers both through the use of copolymers formed with segments of the chain having different π-π* gaps and by the synthesis of higher purity polymers. Saxena and co-workers reported LEDs based on copolymers of substituted thiophenes [220]. The structures of the copolymer have been related to their electroluminescence property (Figure 13.17). The devices emit greenish-blue light in the wavelength region of 550-580 nm (Figures 13.18 and 13.19), which is easily visible in a poorly lighted room. The quantum efficiencies are in the range of 0.002% to 0.01% (photons per electron) at

Figure 13.17 Relation between the structure of copolymers of 3-cyclohexyl thiophene and 3-n-hexyl thiophene, and the quantum efficiency of light-emitting didoes based on these copolymers

420

Conducting Polymers in Molecular Electronics

Figure 13.18 Photoluminescence spectra of copolymers synthesised using ratios of 3-nhexyl thiophene: 3-cyclohexyl thiophene (I) 1:9, (II) 2:3, (III) 1:1, (IV) 9:1

Figure 13.19 Electroluminescence spectra of copolymers synthesised using ratios of 3n-hexyl thiophene: 3-cyclohexyl thiophene (I) 1:9, (II) 2:3, (III) 1:1, (IV) 9:1

room temperature, significantly higher than the corresponding values for poly(3cyclohexylthiophene)-based LEDs. Another approach to improve the performance is the use of additional semiconductor layers. These layers separate the emissive layer from the electrodes where undesirable non radiative recombination may occur, and they also transport both carrier types to the recombination zone more efficiently [221].

421

Handbook of Polymers in Electronics

Table 13.1 Six processable and fluorescent conjugated polymers covering the entire visible spectrum Polymer

Wavelength (nm) / Colour

Repeating Unit

Year

CH3O

Poly(9-9,dihexylfluorene) [236]

427 Violet

OCH3

CH

–O(CH2)3

CH

CH

1991

CH

CH3O

OCH3

Poly(p-phenylene) [237]

465 Blue

1992

Poly(p-phenylene vinylene) [238]

477 Green

1990

Poly[2-methoxy-5(2′-ethylhexyloxy)-pphenylene vinylene] [216]

O

590 Yellow

1991

OCH3

R

Poly(3-alkylthiophene) [217]

650 Orange

1991 S R - Alkyl Group

CH3O

Poly(3-alkoxy cyanoterephthalylidene) [239]

OCH3

710 Red

1993 CN

OCH3

422

CN

OCH3

Conducting Polymers in Molecular Electronics To date, these conjugated polymers have been reported to exhibit light emission in the entire visible spectrum [222], with quantum efficiency over 4% [223], response time of 1 μs [224], brightness of about 106 cd/m2 [225] and device lifetime of 1000 hours [226]. Recently, it has been shown that polymer displays can be fabricated using methods such as screen printing and ink-jet printing [227]. The six processable polymers covering the whole visible spectrum are summarised in Table 13.1. Considerable progress has been made towards the understanding of the electronic processes that control the properties of the conducting polymer based electroluminescent diodes [228-231]. One of the problems still confronting the technologists relates to the red-shifted blue emission in solid films of large-gap polymers including PPV-based ladder copolymers. It is expected that many other problems like durability under drive and under storage conditions, degradation at the polymer-metal interface and the formation of dark spot defects, etc., will soon get solved resulting in speedy commercialisation of the polymer electroluminescent displays. Owing to their advantages, the potential for electroluminescent devices is enormous in, for example, radio receiver, toys, small hand-held devices, large panel displays, stereo equipment and automobile dashboard, notebook computer screens, etc. Many major companies such as Eastman Kodak, Hewlett Packard and Phillips, along with several small entrepreneurial ventures, such as Uniax Corp. (USA) and Cambridge Display Technology (UK), are attempting to grasp the huge market for electroluminescent displays. Large-scale flexible displays, roll-up TV screens, full colour polymer displays and luminescent room lighting are still a long way off. One hurdle is that different polymers are needed to emit light in different colours, and they come with a range of properties. However, in the near future it appears that the two technologies will coexist until the market chooses the winner. Bolognesi and Botta have discussed in detail in Chapter 5 the application of polymers for light-emitting diodes.

13.5.8 Microactuators There is a considerable scope for the development of materials suitable for the fabrication of electromechanical actuators having very small dimensions. Microactuators are devices that convert electrical energy to mechanical energy. It has been suggested that conducting polymers offer attractive alternatives for actuators that function more analogously to natural muscle. Conducting polymer electromechanical actuators are based on the large dimensional changes that occur from the redox doping of any of the host conducting polymers such as polypyrrole, polyaniline and polyalkylthiophene [232]. These dimensional changes are largely due to the volume required to accommodate cations or anions, together with the co-intercalating solvating species. The electrochemical actuators

423

Handbook of Polymers in Electronics have direct similarities to both electrochromic displays and conducting polymer batteries, and the electromechanical cycle corresponds to both the chromatic switching cycle of an electrochromic display and the charge-discharge cycle of a battery. Reversible electrochemical actuators comprise of an anode, a cathode and a separating electrolyte. Either electrode or both can be conducting polymers. However, other redox materials such as graphite have also been used as electromechanical electrodes. The nature of redox processes for conducting polymer electrodes has been found to be rate dependent. For instance, an electrode that is anion doped can be reduced either during the addition of cations or the removal of anions. Conducting polymers have also been utilised for fabrication of hydrostatic and extensional actuators. The hydrostatic actuators provide mechanical work using net volume change during electrochemical redox processes. The net volume change is the overall volume change of the anode, the cathode and the electrolyte. The extensional actuators make use of either linear or biaxial changes in conducting polymers to obtain mechanical work. The dimensional changes can either be individual or relative changes in the dimensions of two or more elements. Compared to alternative actuator materials, the development of conducting polymer actuators is still in its infancy. The discovery that conducting polymer actuators can be operated at lower voltages indicates that these can be utilised for medical applications. Another advantage of conducting polymer lies in the fact that it is possible to use a conducting polymer actuator without an external source. However, subsequent cycles would require external recharging. The performance of conducting polymers is critically dependent on molecular diffusion, which restricts the response of these actuators. This is because fast response time can only be achieved using very thin electromechanical elements. Thus conducting polymers are of greatest interest for either large actuators or microactuators using parallel arrays of thin electromechanical elements. It may be emphasised that many of the proposed applications of microactuators, such as microrobotics for exploration and repair of the human body to microscopic machines for the manipulation and alteration of micron dimensional objects, are futuristic. Some of the shorter term goals, however, include microvalves, microtweezers and micropositioners. Kaneto and co-workers [233] have given (Chapter 8) an excellent review of conducting polymer based microactuators.

13.6 Conclusions It has been shown that electrically conducting polymers are versatile electronic materials that display a unique range of properties for application to molecular electronics. However, there are several problems associated with these materials, namely reproducibility, stability

424

Conducting Polymers in Molecular Electronics and processability. A better understanding of the relevant principles is gradually emerging to accelerate the drive towards the wide range of practical applications [234-239]. It may be remarked that the field is still expanding as new advances are transferred to commercial applications. Recent reports promise the use of regioregular conducting polymers and composites with nanoparticles in optoelectronic devices. Besides this, the use of conjugated polymers in integrated biosensors is another area which needs to be exploited.

Acknowledgements We are grateful to Dr. K. Lal, Director, NPL, India, for his interest in this work. Vibha Saxena is thankful to CSIR, India, for the award of Research Associateship.

References 1.

R.L. Greene, G.B. Street and L.J. Sutude, Physical Review Letters, 1975, 34, 577.

2.

H. Shirkawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang and A.J. Heeger, Journal of the Chemical Society, Chemical Communications, 1977, 578.

3.

T.A. Skotheim, Handbook of Conducting Polymers, Volumes 1 and 2, Marcel Dekker Inc., New York, NY, USA, 1986.

4.

K.S.V. Srinivasan, Macromolecules: New Frontiers, Volumes 1 and 2, Allied Publisher Ltd., India, 1998.

5.

M. Aldissi, Intrinsically Conducting Polymers: An Emerging Technology, Kluwer, Dordrecht, The Netherlands, 1993.

6.

J.A. Pople and S.H. Walmsley, Molecular Physics, 1962, 5, 15.

7.

W.P. Su, J.R. Schrieffer and A.J. Heeger, Physical Review B, 1980, 22, 2099.

8.

D. Emin and T. Holstein, Physical Review Letters, 1976, 36, 323.

9.

J. Chen and A.J. Heeger, Solid State Communications, 1986, 58, 251.

10. A.J. Heeger, S. Kivelson, J.R. Schrieffer and W.P. Su, Reviews of Modern Physics, 1988, 60, 781. 11. B.D. Malhotra, Bulletin of Materials Science, 1988, 10, 11, 85.

425

Handbook of Polymers in Electronics 12. M. Aldissi, F. Schue, L. Giral and M. Rolland, Polymer, 1982, 23, 276. 13. A.F. Diaz and J.A. Logan, Journal of Electroanalytical Chemistry, 1980, 111, 111. 14. A. Dall’olio, Y. Dascola, Y. Varacca and V. Bocchi, Comptes Rendus, 1968, C267, 433. 15. M. Satoh, M. Tabata, T. Takiguchi, K. Kaneto and K. Yoshino, Technological Report of Osaka University, 1986, 25, L73. 16. F. Genoud, M. Gugliemi, M. Nechtschtin, E. Genies and M. Salmon, Physical Review Letters, 1985, 55, 118. 17. M.G. Cross, U. Walton and D.J. Simmons, Journal of Electroanalytical Chemistry, 1985, 189, 389. 18. H.S. Nalwa, Journal of Materials Science, 1991, 26, 1683. 19. X. Zhang, E.T. Kang, K.G. Neoh, K.L. Tan, D.Y. Kim and C.Y. Kim, Journal of Applied Polymer Science, 1996, 60, 625. 20. S.Z. Lewin, Chemical Instrumentation, 1987, 43, A567. 21. P.C. Pandey and R. Prakash, Journal of the Electrochemical Society, 1998, 145, 999. 22. P. Novak, K. Muller, K.S.V. Santhanam and O. Hass, Chemical Review, 1997, 97, 207. 23. E.M. Genies, G. Bidari and A.F. Diaz, Journal of Electroanalytical Chemistry, 1983, 149, 101. 24. A.F. Diaz, A. Martinez, K.K. Kanazawa and M. Salmon, Journal of Electroanalytical Chemistry, 1980, 130, 181. 25. R.J. Waltmon, A.F. Diaz and J. Bargon, Journal of the Electrochemical Society, 1984, 131, 1402. 26. J. Chiu, Polymer Characterization of Thermal Methods of Analysis, Marcel Dekker, New York, NY, USA, 1974. 27. V.T. Traung, B.C. Ennis, T. Turner and C.M. Jender, Polymer International, 1992, 27, 187. 28. S.S. Pandey, M. Gerard, A.L. Sharma and B.D. Malhotra, Journal of Applied Polymer Science, 2000, 75,149.

426

Conducting Polymers in Molecular Electronics 29. E. Lebedev, Th. Dittrich, V. Petrova-Koch, S. Karg and W. Brutting, Applied Physical Letters, 1997, 71, 18, 2686. 30. D.G. Lidzey, D.D.C. Bradley, S.F. Alvarado and D.F. Sedler, Nature, 1997, 386, 135. 31. Z. Shen, P.E. Burrows, V. Bulovic, S.R. Forest and M.E. Thompson, Science, 1997, 276, 2009. 32. M. Granstrom, K. Petrisch, A.C. Areas, A. Lux, M.R. Anderson and R.H. Friend, Nature, 1996, 382, 695. 33. M.J. Marsella, R.J. Newland, P.J. Carroll and T.M. Swager, Journal of the American Chemical Society, 1995, 117, 9842. 34. B. Yon-Hin Bernadette, M. Smolander, T. Crompton and C.R. Lowe, Analytical Chemistry, 1993, 65, 2057. 35. A. Chaubey, K.K. Pandey, V.S. Singh and B.D. Malhotra, Analytica Chimica Acta, 2000, 407, 97. 36. M. Gerard, K. Ramanathan, A. Chaubey and B.D. Malhotra, Electroanalysis, 1999, 11, 6, 450. 37. M.M. Verghese, K. Ramanathan, S.M. Ashraf and B.D. Malhotra, Journal of Applied Polymer Science, 1998, 70, 1447. 38. A.L. Sharma, M. Gerard, R. Singhal, S. Annapoorni and B.D. Malhotra, Applied Biochemistry and Biotechnology, 2001, 96, 155. 39. A.L. Shrama, V. Saxena, S. Annapoorni and B.D. Malhotra, Journal of Applied Polymer Science, 2001, 81, 1460. 40. M. Aldissi, F. Schue, L. Giral and M. Rolland, Polymer, 1982, 23, 246. 41. J.H. Edwards and W.J. Feast, Polymer Communications, 1980, 21, 595. 42. M.M. Ahmed, A.B. Alimuniar, J.H. Edwards and W.J. Feast, Polymer Preprints, 1984, 25, 217. 43. A.G. MacDiarmid, J.C. Chiang, A.F. Richter and A.J. Epstein, Synthetic Metals, 1987, 18, 285. 44. P. Kovacic and A. Kyriakis, Journal of the American Chemical Society, 1963, 85, 454.

427

Handbook of Polymers in Electronics 45. G. Tourillon and F. Garnier, Journal of Electroanalytical Chemistry, 1982, 135, 173. 46. M. Kobayashi, J. Chen, T.C. Chung, F. Moracs, A.J. Heeger and F. Wudl, Synthetic Metals, 1984, 9, 77. 47. S.S. Pandey, S. Annapoorni and B.D. Malhotra, Macromolecules, 1993, 26, 3190. 48. G.Y. Wang and R.P. Quirk, Polymer Preprints, 1994, 35, 712. 49. P.G. Pickup and R.A. Osteryoung, Journal of the American Chemical Society, 1984, 106, 2294. 50. Z. Gao and K.S. Siow, Journal of Electroanalytical Chemistry, 1996, 412, 179. 51. G.M. Whitesides, J.P. Mathia and C.T. Seto, Science, 1991, 254, 1312. 52. M.K. Ram and B.D. Malhotra, Polymer, 1996, 37, 21, 4809. 53. R.C. Haddon and R.A. Lamola, Proceedings of the National Academy of Science, USA, 1985, 82, 1874. 54. K.B. Blodgett, Journal of the American Chemical Society, 1935, 57, 1007. 55. K.B. Blodgett, Review of Scientific Instruments, 1939, 12, 10. 56. I. Langmuir, V.K. Schaefer and H. Sobotka, Journal of the American Chemical Society, 1937, 59, 1751. 57. E.P. Hoing, J.H. Hengst and D. den Engelsen, Journal of Colloidal Interface Science, 1973, 45, 92. 58. K. Hong, R.B. Rosner and M.F. Rubner, Chemistry of Materials, 1990, 2, 82. 59. M. Bardosova, B. Stiller, R.H. Tredgold, M. Wooley, P. Hodge and L. Brehmer, Thin Solid Films, 1996, 284, 450. 60. N.E. Aghbor, M.C. Petty, A.P. Monkman and M. Harris, Synthetic Metals, 1993, 55, 3789. 61. M.K. Ram, N.S. Sudaresan and B.D. Malhotra, Journal of Physical Chemistry, 1993, 97, 11580. 62. M.K. Ram and B.D. Malhotra, Polymer, 1996, 37, 4809. 63. R.B. Dabke, A. Dhanabalan, S. Major, S.S. Talwar, R. Lal and A.Q. Contractor, Thin Solid Films, 1998, 335, 203.

428

Conducting Polymers in Molecular Electronics 64. J.F. Penneau, M. Lapkowski and E.M. Genies, New Journal of Chemistry, 1989, 13, 449. 65. M.K. Ram, E. Maccioni and C. Nicolini, Thin Solid Films, 1997, 303, 27. 66. J.H. Cheung, E. Punkka, M. Rikukawa, M.B. Rosner, A.J. Roraappa and M.F. Rubner, Thin Sold Films, 1992, 210, 246. 67. T. Matsura, A. Nishimura and Y. Shimoyama, Japanese Journal of Applied Physics, 2000, 39, 3357. 68. D.L. Allara, Biosensors and Bioelectronics, 1995, 10, 771. 69. L.F. Rozsnyai and M.S. Wrighton, Chemistry of Materials, 1996, 8, 2, 311. 70. G.M. Whitesides, Scientific American, September 1995, 114. 71. M.H. Dishner, J.C. Hermminger and F.J. Feher, Langmuir, 1996, 12, 26, 6176. 72. C.D. Baim and G.M. Whitesides, Angewandte Chemie International Edition in English, 1989, 28, 4, 506. 73. A. Ulman, An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self Assembly, Academic, San Diego, USA, 1991. 74. J.H. Cheng, A.F. Fou and M.F. Rubner, Thin Solid Films, 1994, 244, 985. 75. M. Ferreira, J.H. Cheung and M.F. Rubner, Thin Solid Films, 1994, 244, 806. 76. J. Paloheimo, K. Laakso, H. Isotalo and H. Stubb, Synthetic Metals, 1995, 68, 244. 77. S. Tanaka, K. Yoshino, A. Takata and T. Yamabe, Synthetic Metals, 1991, 41-43, 3297. 78. M.K. Ram, S. Annapoorni, S.S. Pandey and B.D. Malhotra, Polymer, 1998, 39, 3399. 79. G. Tourillon and F. Garnier, Journal of Applied Polymer Science, Polymer Physics Edition, 1984, 22, 33. 80. M. Onda, H. Nakayama, S. Morita and K. Yoshino, Synthetic Metals, 1993, 5557, 275. 81. S.J. Kamarava, M. Zagorska, B. Krische and S. Soderholm, Physica Scripta, 1991,

41, 112.

429

Handbook of Polymers in Electronics 82. K. Kaneto, K. Harada, W. Takashima, K. Endo and K. Rikukawa, Japanese Journal of Applied Physics, 1999, 38, L1062. 83. A.F. Diaz and B. Hall, IBM Journal of Research and Development, 1983, 27, 342. 84. K. Kaneto, K. Yoshino and Y. Inuishi, Japanese Journal of Applied Physics, 1984, 23, L189. 85. M. Rolland, M.J.M. Abadie and M. Cadene, Revue de Physique Appliquee, 1984, 19, 187. 86. B.D. Malhotra, W. Takashima, S.S. Pandey, R. Singhal, K. Endo, M. Rikukawa and K. Kaneto, Japanese Journal of Applied Physics, 1999, 38, 157. 87. M. Sato and H. Morri, Macromolecules, 1991, 24, 1196. 88. M. Sato and H. Morri, Polymer Communications, 1991, 32, 42. 89. F. Chao, M. Costa, E. Museux, E. Levart and L.M. Abrantes, Electrokhimya, 1993, 29, 57. 90. M.M. Verghese, M.K. Ram, H.Vardhan, B.D.Malhotra and S.M. Ashraf, Polymer, 1997, 38, 1625. 91. H. Sangodkar, R.S. Srinivasa, R. Lal and A.Q. Contractor, Analytical Chemistry, 1996, 68, 779. 92. R. Singh and A.K. Narula, Applied Physical Letters, 1997, 71, 2845. 93. M. Narsimhan, M. Hagler, V. Cammarata and M. Thakur, Applied Physical Letters, 1998, 72, 1063. 94. L. Torsi, A. Dadabalpur and H.E. Katz, Journal of Applied Physics, 1995, 78, 1088. 95. A. Dadabalpur, L. Torsi and H.E. Katz, Science, 1995, 268, 270. 96. F. Garnier, X.Z. Peng, G. Horowitz and D. Fichou, Advanced Materials, 1990, 2, 1592. 97. S.S. Pandey, S.C.K. Misra, B.D. Malhotra and S. Chandra, Journal of Applied Polymer Science, 1992, 44, 911. 98. H. Koezuka, H. Hyodo and A.G. MacDiarmid, Journal of Applied Physics, 1985, 58, 1279.

430

Conducting Polymers in Molecular Electronics 99.

M. Ozuki, D.L. Peebles, B.R. Wein berger, C.K. Chiang, S.C. Gau, A.J. Heeger and A.G. MacDiarmid, Applied Physical Letters, 1979, 35, 83.

100. R. Gupta, S.C.K. Misra, B.D. Malhotra and S. Chandra, Applied Physical Letters, 1991, 58, 51. 101. S.C.K. Misra, M.K. Ram, S.S. Pandey, B.D. Malhotra and S. Chandra, Applied Physical Letters, 1992, 61, 1220. 102. H. Kokado, F. Husokawa and K. Hoshino, Japanese Journal of Applied Physics, 1993, 32, 189. 103. S.S. Pandey, M.K. Ram, V.K. Srivastava and B.D. Malhotra, Journal of Applied Polymer Science, 1997, 65, 2745. 104. V. Saxena and R. Prakash, Polymer Bulletin, 2000, 45, 267. 105. M.K. Ram, S. Annapoorni and B.D. Malhotra, Journal of Applied Polymer Science, 1996, 60, 407. 106. S.M. Sze, Physics of Semiconductor Devices, Second Edition, John Wiley & Sons, Inc., New York, NY, USA, 1981. 107. A. Mary, Y. Goldstein and N.B. Grover, Semiconductor Surfaces, North-Holland, Amsterdam, The Netherlands, 1964. 108. A. Tsumura, M. Koezuka and T. Ando, Applied Physical Letters, 1986, 49, 1210. 109. J.H. Burroughes, C.A. Jones and R.H. Friend, Nature, 1988, 335, 137. 110. G. Guilaud, M. Al Sadoun, M. Maitrot, J. Simon and M. Bouvet, Chemical Physics Letters, 1990, 167, 503. 111. A. Tsumura, H. Fuchigami and H. Koezuka, Synthetic Metals, 1991, 41-43, 1181. 112. J. Paloheimo, P. Kuivalaiven, H. Stubb, E. Vuorimaa and P. Yli-Lahti, Applied Physical Letters, 1990, 56, 1157. 113. A. Tsumura, H. Koezuka and T. Ando, Applied Physical Letters, 1986, 49, 1210. 114. M. Liess, D. Chinn, D. Petelenz and J. Janta, Thin Solid Films, 1996, 286, 252. 115. V. Saxena, V. Shirodkar and R. Prakash, Applied Biochemistry and Biotechnology, 2001, 96, 63.

431

Handbook of Polymers in Electronics 116. J.C. Anderson, Thin Solid Films, 1976, 38, 151. 117. G. Horowitz, F. Deloffre, F. Garnier, R. Hajlaoui, M. Thruyene and A. Yassar, Synthetic Metals, 1993, 54, 435. 118. R.K. Willardson and A.C. Bear, Semiconductors and Semimetals, Volume 21D, Academic Press, Orlando, FL, USA, 1984. 119. J.H. Burroughes, C.A. Jones, R.A. Lawrence and R.H. Friend in Conjugated Polymeric Materials: Opportunities in Electronics, Opto-electronics and Molecular Electronics, Eds., J.L. Bredas and R.R. Chance, Kluwer Academic Publishers, Dordecht, Germany, 1990, 22. 120. X. Peng and D. Fichou, Advanced Materials, 1990, 2, 592. 121. K. Waragai, H. Akimichi, S. Hotta, H. Kano and H. Sakaki, Synthetic Metals, 1993, 57, 4053. 122. S.R. Elliott, Physics of Amorphous Materials, Second Edition, Longman, New York, NY, USA, Chapters 5 and 6, 1990. 123. F. Garneir, R. Hajlaoui, A. Yassar and P. Srivastava, Science, 1994, 265, 1684. 124. Y.Yang and A.J. Heeger, Nature, 1994, 372, 344. 125. H. Sirringhaus, N. Tessler and R.H. Friend, Science, 1998, 280, 1741. 126. M. Trojanowicz and K.T.K.V. Krawczy, Mikrochimica Acta, 1995, 121, 167. 127. M. Elmgren, S.E. Lundquist and M. Sharp, Journal of Electroanalytical Chemistry, 1993, 362, 227. 128. M. Trojanowicz and K.V. Krawczyk, Mikrochimica Acta, 1995, 121, 167. 129. S.A. Wring and J.P. Hart, Analyst, 1992, 117, 1215. 130. P.R. Teasdale and J.M. Wallace, Journal of Electroanalytical Chemistry, 1993, 362, 1. 131. S. Cosnier, Biosensors and Bioelectronics, 1999, 14, 443. 132. A.O. Patil, A.J. Heeger and F. Wudl, Chemical Review, 1988, 88, 183. 133. T. Tatsuma, T. Watanabe and T. Watanabe, Sensors and Actuators B, 1993, 14, 752.

432

Conducting Polymers in Molecular Electronics 134. C.A. Rowe-Taitt, J.P. Golden, M.J. Feldstein, J.J. Cras, K.E. Hoffman and F.S. Ligler, Biosensors and Bioelectronics, 2000, 14, 785. 135. C.A. Rowe, L.M. Tender, M.J. Feldstein, J.P. Golden, S.B. Scruggs, B.D. Maccraith, J.J. Cras and F.S. Ligler, Analytical Chemistry, 1999, 71, 3846. 136. G. Bidan, Sensors and Actuators B, 1992, 7, 383. 137. P.C. Pandey, Journal of the Chemical Society, Faraday Transactions 1, 1988, 84, 2259. 138. A. Deronzier and J.C.Moutet, Current Topics in Electrochemistry, 1994, 3, 159. 139. H. Shinohara, T. Chiba and M. Aizawa, Sensors and Actuators B, 1998, 13, 79. 140. P. Gros and A. Bergel, Journal of Electroanalytical Chemistry, 1995, 386, 65. 141. J.J. Gooding, E.A.H. Hall and D.B. Hibbert, Electroanalysis, 1998, 10, 1130. 142. M. Aizawa and S. Yabuki, Proceedings of the 51st Annual Meeting of the Japanese Chemical Society, 1985, 6. 143. N.C. Foulds and C.R. Lowe, Journal of the Chemical Society, Faraday Transactions 1, 1986, 82, 1259. 144. A. Senillou, N. Jaffrezic-Renault, C. Martelet and S. Cosnier, Talanta, 1999, 50, 219. 145. S. Milardovic, I. Kruhak and B.S. Grabaric, Laboratory Robotics and Automation, 1999, 11, 266. 146. J. Losada and M.P.G. Armada, Electroanalysis, 1997, 9, 1416. 147. E. Dinckaya, E. Akilmaz and S. Akgol, Indian Journal of Biochemistry and Biophysics, 2000, 37, 67. 148. S. Dong, Q. Deng and G. Cheng, Analytica Chimica Acta, 1993, 279, 235. 149. C.G.J. Koopal, M.C. Feiters, R.J.M. Notte, B. de Ruiter and R.B.M. Schasfoort, Biosensors and Bioelectronics, 1992, 7, 461. 150. C.G.J. Koopal and R.J.M. Notte, Bioelectrochemistry and Bioenergetics, 1994,

33, 45. 151. W.J. Cho and H.J. Huang, Analytical Chemistry, 1998, 70, 3946.

433

Handbook of Polymers in Electronics 152. C. Kranz, H. Wohlschlager, H.-L. Schmidt and W. Schuhmann, Electroanalysis, 1998, 110, 546. 153. W. Schuhmann, R. Lammert, B. Uhe and H.-L. Schmidt, Sensors and Actuators B, 1990, 1, 537. 154. S. Cosnier, B. Galland, C. Gondran and A. Le Pellec, Electroanalysis, 1998, 10, 808. 155. H. Korri-Youssoufi, F. Garnier, P. Srivasta, P. Godillot and A. Yassar, Journal of the American Chemical Society, 1997, 119, 7388. 156. N.C. Foulds and C.R. Lowe, Analytical Chemistry, 1988, 60, 2473. 157. A. Gambhir, M. Gerard, A.K. Mulchandani and B.D. Malhotra, Applied Biochemistry and Biotechnology, 2001, 96, 249. 158. J.C. Cooper and E.A.H. Hall, Biosensors, 1992, 7, 473. 159. E. Pringsheim, E. Terpetschnig and O.S. Wolfbeis, Analytica Chimica Acta, 1997, 357, 247. 160. L. Rover, G.B. Neto and L.T. Kubota, Quimica Nova, 1997, 20, 519. 161. E.E. Karyakina, L.V. Neftyakova and A.A. Karyakin, Analytical Letters, 1994, 27, 2871. 162. E.I. Woha, D.S. De Villaverde, N.P. Garcia, M.R. Smyth and J.M. Pingarron, Biosensors and Bioelectronics, 1997, 12, 749. 163. U.W. Grummt, A. Pron, M. Zagorska and S. Lefrant, Analytica Chimica Acta, 1997, 357, 253. 164. M.C. Shin and H.S. Kim, Biosensors and Bioelectronics, 1996, 11, 171. 165. P.N. Bartlett and D.J. Caruana, Analyst, 1992, 117, 1287. 166. C. Malitesta, F. Palmisano, L. Torsi and P. Zamborun, Analytical Chemistry, 1990, 62, 2735. 167. C.A. Groom and J.H.T. Luong, Analytical Letters, 1993, 26, 1383. 168. M. Ozden, E. Ekinci and A.E. Karagozier, Turkish Journal of Chemistry, 1999, 23, 89. 169. A.A. Karyakin, L.V. Lukachova, E.E. Karyakina, A.V. Orlov and G.P. Karpachova, Analytical Communications, 1999, 36, 153.

434

Conducting Polymers in Molecular Electronics 170. K. Ramanathan, M.K. Ram, B.D. Malhotra and A.S.N. Murthy, Materials Science and Engineering, 1995, C3, 159. 171. K. Ramanathan, S. Annapoorni and B.D. Malhotra, Sensors and Actuators B, 1994, 21, 165. 172. K. Ramanathan, M.N. Kamalasanan, B.D. Malhotra, D.R. Pradhan and S. Chandra, Journal of Sol-Gel Science and Technology, 1997, 10, 309. 173. M. Sriyudthsak, H. Vamagishi and T. Moriizumi, Thin Solid Films, 1988, 160, 463. 174. Y. Okhata, T. Surata, K. Jjiro and K. Ariya, Langmuir, 1988, 4, 1373. 175. T. Hianik, M. Snejdarkova, L. Sokilikova, E. Meszar, R. Krivanek, V. Tvarozek, I. Novotny and J. Wang, Sensors and Actuators B, 1999, 57, 201. 176. J. Rishpon and D. Ivnitski, Biosensors and Bioelectronics, 1997, 12, 195. 177. Y. Okahata, Y. Matsunobu, K. Jjiro, M. Mukai, A. Murakami and K. Makino, Journal of the American Chemical Society, 1992, 114, 8299. 178. N. Saoudi, N. Jammul, M.-L. Abel, M.M. Chehimi and G. Dodin, Synthetic Metals, 1997, 87, 97. 179. K.A. Marx, J.O. Lim, D.S. Minehan, R. Pande, M.N. Kamath, S.K. Tripathi and D.L. Kaplan, Journal of Intelligent Material Systems and Structures, 1994, 5, 447. 180. T. Livache, A. Roget, E. Dejean, C. Barthet, G. Bidan and R. Teoule, Synthetic Metals, 1995, 71, 2143. 181. T. Livache, A. Roget, E. Dejean, C. Bathet, G. Bidan and R. Teoule, Nucleic Acid Research, 1994, 22, 2915. 182. K. Hoshimoto, K. Ito and Y. Isimori, Analytica Chimica Acta, 1994, 286, 219. 183. E. Stussi, S. Cella, G. Serra and D. De Rossi, Materials Science and Engineering, 1996, C4, 27. 184. K. Toko, Measurement Science and Technology, 1998, 9, 1919. 185. F. Winquist, C. Krantz-Rulker, P. Wide and I. Lundstrom, Measurement Science and Technology, 1998, 9, 1937. 186. K. Toko and T. Fukusaka, Sensors and Materials, 1997, 9, 171.

435

Handbook of Polymers in Electronics 187. Z. Chen and D. Lancet, Proceedings of the National Academy of Science of the United States of America, 1984, 81, 1859. 188. Sensory Evaluation Techniques, Volumes 1 and 2, Eds., M. Meilgaard, G.V. Civille and B.T. Carr, CRS Press, Boca Raton, FL, USA, 1987. 189. M.T. Morales, M.V. Alonso, J.J. Rios and R. Aparicio, Journal of Food Science and Agriculture, 1995, 43, 2925. 190. J.W. Gardner and P.N. Barlett, Sensors & Actuators B, 1994, 221, 18119. 191. R.W. Moncrieff, Journal of Applied Physiology, 1961, 16, 742. 192. W.F. Wilkens and A.D. Hatman, Annals of the New York Academy of Science, 1964, 116, 608. 193. J.W. Gardner, Proceedings of the 8th International Congress of the European Chemoreception Research Organization, University of Warwick, UK, July 1987. 194. J.W. Gardner, E.L. Hines and H.C. Tang, Sensors & Actuators B, 1992, 9, 9. 195. P.N. Bartlett, J. Gradner and R.G. Whitaker, Sensors & Actuators B, 1992, A21A23, 911. 196. R. Gutierrez-Osuna, H. Troy Nagle and S.S. Schiffman, Sensors & Actuators B, 2000, B61, 170. 197. R. Stella, J.N. Barisci, G. Serra, G.G. Wallace and D. De Rossi, Sensors & Actuators B, 2000, 63, 1. 198. D. Hodgins, Sensors & Actuators B, 1995, 26-27, 255. 199. K.C. Persaud and P. Travers, Intelligent Instrumentation and Control, 1992, July/Aug, 147. 200. C.R. Martin, Advanced Materials, 1991, 3, 457. 201. Q. Zhou and T.M. Swager, Journal of the American Chemical Society, 1995, 117, 1200. 202. T.J. Gardner, C.D. Frisble and M.S. Wrighton, Journal of the American Chemical Society, 1995, 117, 6927. 203. W.M. Albers, J.O. Lekkala, L. Jeuken, G.W. Canters and A.P.F. Turner, Bioelectrochemistry and Bioenergetics, 1997, 42, 25.

436

Conducting Polymers in Molecular Electronics 204. M.M. Verghese, K. Ramanathan, S.M. Ashraf, M.N. Kamalasanan and B.D. Malhotra, Chemistry of Materials, 822, 1996. 205. K. Ramanathan, M.N. Kamalasanan, B.D. Malhotra, D.R. Pradhan and S. Chandra, Journal of Sol-Gel Science and Technology, 1996, 10, 822. 206. R.H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughes, R.N. Marks, C. Taliani, D.D.C. Bradley, D.A. Dos Santos, J.L. Bredas, M. Logd and and W.R. Salaneck, Nature, 1999, 397, 121. 207. J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackey, R.H. Friend, P.L. Burns and A.B. Holmes, Nature, 1990, 347, 539. 208. K. Tada, M. Onoda, A.A. Zakhidov and K. Yoshino, Japanese Journal of Applied Physics, 1997, 36, L306. 209. J. Bharathan and Y. Yang, Applied Physical Letters, 1998, 72, 2660. 210. J.A. Roges, Z. Bao and L. Dhar, Applied Physical Letters, 1998, 73, 294. 211. K. Tada, M. Onoda and H. Nakayama, Japanese Journal of Applied Physics, 1998, 37, L1181. 212. M. Pope, H. Kallmann and P. Magnante, Journal of Chemical Physics, 1965, 38, 2042. 213. J.J.M. Halls, C.A. Walsh, N.C. Greenham, E.A. Marceglia, R.H. Friend, S.C. Moratti and A.B. Holmes, Nature, 1995, 376, 498. 214. M. Granstrom, K. Petritsch, A.C. Areas, A. Lux, M.R. Anderson and R.H. Friend, Nature, 1998, 395, 257. 215. K. Yoshino, T. Kuwabara and T. Iwasa, Japanese Journal of Applied Physics, 1990, 29, L1514. 216. D. Braun and A.J. Heeger, Applied Physical Letters, 1991, 58, 1982. 217. Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Japanese Journal of Applied Physics, 1991, L1941, 218. J.F. Gruner, R.H. Friend, U. Scherf, J. Huber and A.B. Holmes, Advanced

Materials, 1994, 748. 219. I.D. Parker, Q. Pei and M. Marrocco, Applied Physical Letters, 1994, 65, 1272.

437

Handbook of Polymers in Electronics 220. V. Saxena and V. Shirodkar, Journal of Applied Polymer Science, 2000, 77, 1051. 221. D.R. Baigent, N.C. Greenham, J. Gruner, R.N. Marks, R.H. Friend, S.C. Morotii and A.B. Holmes, Synthetic Metals, 1994, 67, 3. 222. M. Hamaguchi, A. Fujii, Y. Ohmori and K. Yoshino, Synthetic Metals, 1997, 84, 557. 223. I.D.W. Samuel, B. Crystall, G. Rumbles, P.L. Burn, A.B. Holmes and R.H. Friend, Synthetic Metals, 1993, 54, 281. 224. R. Osterbacka, G. Juska, K. Arlauskas, A.J. Pal, K.-M. Kallman and H. Stubb, Journal of Applied Physics, 1998, 84, 6, 1. 225. N. Tessler, N.T. Harrison and R.H. Friend, Advanced Materials, 1998, 10, 64. 226. R.F. Service, Science, 1996, 273, 878. 227. K. Yoshimori, S. Naka, M. Shibata, H. Okada, H. Onnagawa, Proceedings of the 18th International Display Research Conference and Exhibition, Asia Display ’98, Seoul, Korea, 1998, 213. 228. M. Yan, L.J. Rothberg, F. Papadimitrakopoulos, M.E. Galvin and T.M. Miller, Physical Review Letters, 1994, 73, 744. 229. L.J. Rothberg, M. Yan, E. Kowck, T.M. Miller, M.E. Galvin, S. Son and F Papadimitrakopoulos, IEEE Transactions on Electron Devices, 1997, 44, 1262. 230. K. Tada and M. Onoda, Journal of Applied Physics, 1999, 86, 3139. 231. M. Redecker, D.D.C. Bradley, M. Inbasekaran and E.N. Woo, Applied Physical Letters, 1998, 73, 1565. 232. Molecular Electronics, Eds., R.H. Baughman, L.W. Shacklette, R.L. Elsenbaumer, E.J. Plichta, C. Becht and P.I. Lazarev, Kluwer Publishers, Dordrecht, Germany, 1991, 267. 233. M. Kaneko, M. Fukuei, W. Takashima and K. Kaneto, Synthetic Metals, 1997, 84, 795. 234. P.K.H. Ho, D.S. Thomas, R.H. Friend and N. Tessler, Science, 1999, 285, 233. 235. R. Osterbacka, C.P. An, X.M. Jiang and Z.V. Vardeny, Science, 2000, 287, 839.

438

Conducting Polymers in Molecular Electronics 236. Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Japanese Journal of Applied Physics, 1991, 30, L1971. 237. G. Girem, G. Leditzty, B. Ulrich and G. Leising, Synthetic Metals, 1992, 51, 383. 238. R.H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughs, R.N. Marks, T. Taliani, D.D.C. Bradley, D.A. Dos Santos, J.L. Bredas, M. Loglund and W.R. Salaneck, Nature, 1999, 397, 121. 239. N.C. Greenham, S.C. Moratti, D.D.C. Bradley, R.H. Friend and A.B. Holmes, Nature, 1993, 265, 628.

439

Handbook of Polymers in Electronics

440

Abbreviations and Acronyms

ηEL

Electroluminescence efficiency

ηPL

Photoluminescence efficiency

3MOT

3-methyl-4′-octyl-2,2-bithiophene-5,5′-diyl

3ODOP

3-Octadecanoylpyrrole

3ODP

3-Octadecylpyrrole

ac

Alternating current

AES

Auger electron spectroscopy

AFM

Atomic force microscopy

Alq3

Aluminium 8-hydroxyquinoline

AMPSA

2-Acryloamido-2-methyl-1-propane sulfonic acid

aPS

Atactic polystyrene

APTES

(3-aminopropyl)trimethoxysilane

ATR

Attenuated total reflectance

BA

Benzyl alcohol

BABTDS

Bisacryloxybutyl tetramethyl disiloxane

BDOH-PF

Poly(9,9-bis(3,8-dioxaheptyl-fluorine-2,7-diyl)

BSA

Benzenesulfonic acid

CB

Conduction band

ChOx

Cholesterol oxidase

CMOS

Complementary metal oxide semiconductor

CSA

Camphor sulfonic acid

CTFE

Chlorotrifluoroethylene

CV

Cyclic voltammagram

CVD

Chemical vapour deposition

CYEPL

Cyanoethylpullulan

441

Handbook of Polymers in Electronics DBP

Dibutylphthalate

DBS

Dodecylbenzene sulfonate

DBSA

Dodecylbenzoyl sulfonic acid

dc

Direct current

DEC

Diethylene carbonate

DEZ

Diethylzinc

DFWM

Degenerate four wave mixing

DMF

Dimethylformamide

DNA

Deoxyribonucleic acid

DNQ

1,2-Diazonaphthoquinone

DPE

Discrete particle encapsulation

DR1

Disperse Red 1

DSC

Differential scanning calorimetry

EA

Electron affinity

EC

Ethylene carbonate

e-e

Electron-electron

e-h

Electron-hole

EL

Electroluminescence

ELISA

Enzyme linked immunosorbant assay

ENDOR

Electron nuclear double resonance

ENFET

Enzyme field-effect transistor

EO

Electrooptical

EPR

Electron paramagnetic resonance

ES

Emeraldine salt

ESCA

Electron spectroscopy for chemical analysis

ESR

Electron spin resonance

ETL

Electron transporter layer

EV

Electric vehicle

FAD

Flavin adenine dinucleotide

FADH2

Flavin adenine dinucleotide (reduced form)

442

Abbreviations and Acronyms FBP

Fluorinated polymethacrylate

FET

Field-effect transistor

FIT

Fluctuation induced tunnelling

FTIR

Fourier transform infrared

FWHM

Full wave half maximum

GC

Gas chromatography

GLDH

Glutamate dehydrogenase

GOD

Glucose oxidase

GPC

Gel permeation chromatography

HMDSO

Hexamethyl disiloxane

HOMO

Highest occupied molecular orbital

HPC-Py

Pyrene-labelled hydroxypropylcellulose

HPLC

High pressure liquid chromatography

HRP

Horseradish peroxidase

HSA

Human serum albumin

HTL

Hole transporter layer

I(t)

Emission intensity

IC

Internal conversion

IEEE

Institute of Electrical and Electronics Engineers

IGFET

Insulated-gate field-effect transistor

IP

Ionisation potential

IPS

Isotactic polystyrene

IR

Infrared

ISC

Intersystem crossing

ISFET

Ion-selective field-effect transistor

ISMET

Ion-selective microelectrochemical transistor

IT

Information technology

ITO

Indium tin oxide

KC8

Potassium graphite

LB

Langmuir Blodgett

443

Handbook of Polymers in Electronics LC

Liquid crystalline

LCD

Liquid crystal display

LCST

Low critical solution temperature

LDH

Lactate dehydrogenase

LDM

Laser displacement meter

LED

Light-emitting diode

LOD

Lactate oxidase

L-PPP

Ladder PPP

LS

Leuco-emeraldine salt

LUMO

Lowest occupied molecular orbital

MAdMA

2-Methyl-2-adamananol methacrylate

MC

Magnetoconductance

ME

Molecular electronics

MEH-PPV

Poly[2-(2´-ethylhexyloxy)-5-methoxy-1,4-phenylenevinylene]

M-I

Metal-insulator

MIS

Metal-insulator semiconductor

MISFET

Metal-insulator semiconductor field-effect transistor

MLMA

Mevalonic lactone methacrylate

MLR

Multilayer resist

MOPPV

Poly(2,5-dimethoxy-p-phenylene vinylene)

MOSFET

Metal oxide semiconductor field-effect transistor

MR

Magnetoresistance

MRS

Microresists

MS

Mass spectrometry

MTHF

Methyltetrahydrofuran

MW

Microwave

MZ

Mach-Zehnder

NAD

Nicotinamide adenine dinucleotide

NADH

Nicotinamide adenine dinucleotide (reduced form)

n-BCMU-PDA n-Butoxycarbonylmethyl urethane

444

Abbreviations and Acronyms NLO

Non linear optical

NMP

N-methyl-2-pyrrolidinone

NMR

Nuclear magnetic resonance

NSOM

Near-field scanning optical microscopy

ODMR

Optically detected magnetic resonance

OMA

Organised monolayer assembly

P(PMMAOM-MAN)

Poly(methyl methacrylate-co-3-oximino-2-butanone methacrylate-comethacrylonitrile)

P/GC/MS

Pyrolysis/gas chromatography/mass spectroscopy

PAAM

Polyacrylamide

PAB

Poly(o-aminobenzoic acid)

PAN

Polyacrylonitrile

PANI

Polyaniline

PaS

Pernigraniline salt

PAT

Polyalkylthiophene

PAV

Poly(arylene vinylene)

PBI

Polybenzimidazole

PBS

Phosphate buffer saline

PC

Propylene carbonate

PCM

Portable conformable masking

PCz

Polycarbazole

PDA

Polydiacetylene

PDABT

Poly(4,4-dialkyl-2,2′-bithiophene)

PDF2/6

Poly(9,9-di(ethylhexyl)fluorine

PDTB

Poly(1,4-di(2-thienyl)benzene

PEB

Polyemeraldine base

PEDOT

Poly(3,4-ethylenedioxythiophene)

PEI

Polyethyleneimine

PEO

Polyethylene oxide

PET

Polyethylene terephthalate)

445

Handbook of Polymers in Electronics PFu

Polyfuran

PHP

Para-hexaphenyl

PI

Polyimide

PL

Photoluminescence

PMAN

Poly(o-methoxyaniline)

PMeT

Poly(3-methyl)thiophene

PMMA

Polymethyl methacrylate

PNIPAM

Poly(N-isopropylacrylamide)

PNMA

Naphthyl-substituted polymethacrylate

PP/VD

Plasma polymerisation/vapour deposition

PPMA

Poly(phenyl methacrylate)

PPO

Poly(2,6-dimethylphenylene oxide)

PPP

Poly(p-phenylene)

PPPY

Polyphenylenediyl-pyridinediyl

PPV

Poly(p-phenylene vinylene)

PPy

Polypyrrole

PS

Polystyrene

PSSA

Polystyrene sulfonated acid

PSTF

Polyurethane with symmetrical substituted tris-azo dye with fluorinated alkyl units

PT

Polythiophene

PTBPY

Polythiophenediyl-bipyridenediyl

PTCDA

3,4,9,10-Perylene tetracarboxylic dianhydride

PTPY

Polythiophenediyl-pyridinediyl

PTS

p-Toluene sulfonate

PTS-PDA

Poly[bis(p-toluenesulfonate) of 2,4-hexadiyne-1,6-diol]

PTV

Polythienylene vinylene

PVC

Polyvinyl chloride

PVCZ

Polyvinylcarbazole

PVD

Physical vapour deposition

446

Abbreviations and Acronyms PVdF

Polyvinylidene fluoride

PVN

Polyvinylnaphthalene

PVP

Polyvinylpyrrolidine

PVS

Polyvinylsulfonate

q-1D

Quasi one-dimensional

QCM

Quartz crystal microbalance

QPM

Quasi-phase-matching

RDM

Random dimmer model

RF

Radio frequency

RGB

Red green blue

RIE

Reactive ion etching

SA

Salicyclic acid

SAW

Surface acoustic wave

SBAC

Symmetrically substituted benzylidene aniline with chloride

SCE

Standard calomel electrode

SEI

Solid electrolyte interface

SEM

Scanning electron microscopy

SHG

Second-harmonic generation

SPIE

International Society for Optical Engineering

SPM

Self phase modulation

SPR

Surface plasmon resistance

SSH

Su Schrieffer and Heeger

StA

Stearic acid

STM

Scanning tunnelling microscopy

T-BOC

t-Butyl carbonate

T-BOC-PMS

Poly(p-t-butyloxycarbonyloxystyrene)

TCNQ

Tetracyanoquinodimethane

TCR

Temperature coefficient of resistance

TEM

Transmission electron microscopy

TEOS

Tetraethylorthosilicate

447

Handbook of Polymers in Electronics Tg

Glass transition temperature

TGA

Thermogravimetric analysis

THF

Tetrahydrofuran

THG

Third-harmonic generation

TMT

Tetramethyltin

TOF

Time of flight

TPD

Triphenylamine derivative

TRIDSB

Poly(1,20-(10,13)-didecyl)distyrylbenzene-co-1,2-(4-(pethylphenyl))triazole

TSFET

Taste-sensing field-effect transistor

TSO

p-Toluene sulfonate

TTF

Tetrathiafulvaline

TTL

Transistor-transistor logic

UPS

Ultraviolet photoelectron spectroscopy

UV

Ultraviolet

VB

Valence band

VLSI

Very large scale integration

Von

Onset voltage

VR

Vibrational relaxation

VRH

Variable-range hopping

WL

Weak localisation

XPS

X-ray photoelectron spectroscopy

XRD

X-ray diffraction

?BL

?-Butyrolactone

448

Contributors

S. Annapoorni Department of Physics and Astrophysics University of Delhi Delhi – 110007 India Alberto Bolognesi Istituto di Chimica delle Macromolecole Consiglio Nazionale delle Ricerche Via E. Bassini 15 20133 Milano Italy C. Botta Istituto di Chimica delle Macromolecole Consiglio Nazionale delle Ricerche Via E. Bassini 15 20133 Milano Italy Asha Chaubey Biomolecular Electronics & Conducting Polymer Research Group National Physical Laboratory Dr K.S. Krishnan Road New Delhi-110012 India Jean-Claude Dubois Université Pierre et Marie Curie Laboratoire de Physico-chimie des Polyméres Tour 44.1er et Case 185 4 Place Jussieu. 75252 Paris Cedex France

449

Handbook of Polymers in Electronics Anamika Gambhir Biomolecular Electronics and Conducting Polymer Research National Physical Laboratory Dr. K.S. Krishnan Road New Delhi – 110012 India Manju Gerard Department of Chemistry Allahabad Research Institute (Deemed University) Naini Allahabad-210007 India Atsunori Hiratsuka Health Care Group Living Environment Development Center Matsushita Electric Industrial Co., Ltd. 3-1-1 Yagumo-naka-machi Moriguchi City Osaka 570-8501 Japan Toshikuni Kaino Institute for Chemical Reaction Science Tohoku University 2-1-1 Katahira Aoba-ku Sendai-shi, Miyagi 980-77 Japan Masmitsu Kaneko Department of Computer Science and Electronics Kyushu Institute of Technology Iizuka, Fukuoka, 820-8502 Japan Keiichi Kaneto Department of Computer Science and Electronics Kyushu Institute of Technology Iizuka, Fukuoka, 820-8502 Japan

450

Contributors Isao Karube Karube Laboratory Biosensor Division Research Centre for Advanced Science and Technology University of Tokyo 4-6-1 Komaba Meguru-ku Tokyo 153-8904 Japan Bansi D. Malhotra Biomolecular Electronics & Conducting Polymer Research Group National Physical Laboratory Dr K.S. Krishnan Marg New Delhi-110012 India Reghu Menon Department of Physics Indian Institute of Science Bangalore India – 560012 Amalesh Mukhopadhyay Department of Science and Technology New Mehrauli Road New Delhi – 110016 India K.S. Narayan Chemistry and Physics of Materials Unit Jawaharlal Nehru for Advanced Scientific Research Jakkur PO Bangalore 560064 India Tim H. Richardson Applied Molecular Engineering Group Department of Physics & Astronomy University of Sheffield Hounsfield Road Sheffield S3 7RH UK

451

Handbook of Polymers in Electronics Vibha Saxena Biomolecular Electronics and Conducting Polymer Research National Physical Laboratory Dr. K.S. Krishnan Road New Delhi – 110012 India Bruno Scrosati Department of Chemistry University ‘La Sapienza’ 00185 Rome Italy Rahul Singhal Biomolecular Electronics and Conducting Polymer Research National Physical Laboratory Dr. K.S. Krishnan Road New Delhi – 110012 India Barbara Wandelt Technical University of Lodz Faculty of Chemistry Department of Molecular Physics Zeromskiego 116 90924 Lodz Poland

452

Main Index

Index

A absorbance 372 absorption 100, 111 acetone 398 acid-amine ion linkage 210 acid anhydride bonds 204-205 acid photogenerators 199 2-acrylamido-2-methyl-1-propanesulfonic acid (AMPSA) 48 acrylate ester 209 actuators 255, 267 affinity biosensors 289-290 Ag/AgCl electrode 319 Alexandrium affine 289 alkanethiol-stabilised metal nanoparticles 357 alkanoic acid 383 alkoxy substituted PPV 169 3-alkyloxymethoxy thiophene 76 3-alkyl-substituted thiophenes 345 3-alkylthiophene 164 Alq3 see aluminium 8-hydroxyquinolene (Alq3) alternate layer Langmuir trough 385, 386 alternate layer LB deposition technique 385 alternate layer LB films 386 aluminium 8-hydroxyquinolene (Alq3), electroluminescence 141 aluminium 370 3-aminopropyltriethoxysilane 352 3-aminopropyltrimethoxysilane (APTES) 289 2-aminopyridine 106

amperometric biosensors 301, 303, 313, 317 electron transfer in 303-304 response measurements 319 anisotropy conductivity 53 electroluminescence 169 electrolytic expansion in polyaniline films 266 etching techniques 190 anode 370, 371 anthracene 121, 367 antibodies 413 ArF 197, 200 aromatic hydrocarbons 104 Arrhenius relation 311 artificial muscles 255 arylene-bisphosphylidenes 148 atactic PS (aPS) 123-125 atomic force micrograph (AFM) 404, 405 attenuated total reflectance (ATR) 283 Au-P30T 358, 360 AX 1000 photoresist 201 AZ Electronic Materials 201 azo benzene 255, 256 azo-dye-functionalised main-chain polymer synthetic scheme 79 χ(3) wavelength dependence 80 azo-dye-functionalised poled polymers 84-87 azo-dye-functionalised polymer waveguides 90-93 azo-dye polyester 76

453

Handbook of Polymers in Electronics

B backscattering lifetime 41 band gap 5, 148, 394, 395 band structure 4, 5 batteries C/LiMO2 polymer lithium-ion 244 high energy density 217 Li-AyBz lithium 221 Li/LiClO4-EC-DMC-PAN/LiCryMn(2y) O4 235 Li/PEO-LiCF3SO3 + çLiAlO2/LiMn2O4 225 lithium 218-239, 243-245 lithium-ion 217, 230-231 rocking chair 230-231 SnO2/LiClO4-EC-DMC-PAN/ LiNi0.8Co0.2 O2 237 BDOH-PF 163 benzene 112, 116 benzenesulfonic acid (BSA) 260 benzyl alcohol (BA) 123 benzylidene aniline 77 β-carotene 376, 403 bias voltage 144, 146 bifunctional device 356 bifunctional reagents 307 bimolecular photophysical processes 103-104 bimolecular photoprocesses 110 bimorph actuator 257 binder 272 biocomponents 301-302 biological fluids, analysis 297 biological parameters 297 biomedical applications 272 biomimetic actuators 255 biomolecules, immobilisation 305-309, 412-413 biosensors 272-273, 275-276, 287, 309, 411-414 1st generation 300

454

2nd generation 300 3rd generation 300 4th generation 300 amperometric see amperometric biosensors blood glucose 298 cholesterol 323-324 classification 313 conducting polymer-based 297 conductometric 315 construction 299-300 design 412 development 297, 298 effects of storage time 312-313 electrochemical 313, 315 fabrication 413 general principle 412 generations 300 glucose 318, 413 healthcare 318 importance of conducting polymers 302-304 key attraction 297 market 298 medical diagnostic application 297 microbiosensors 276 non medical application market 297 optical 314 potential applications 412 potentiometric 301, 315-317 research and applications 297-300 response measurements 313 schematic 299, 411 selectivity 301 sequence of events 299-300 specificity 301 types 313-317 2-(4-biphenyl)-5-(tert-butylphenyl-1,3,4oxadiazole (PBD) 158 bipolaron 4, 6, 9-12, 142, 377, 396 energy levels and occupied localised states 11

Index formation 9, 11 instability 17 interchain transport 20 negative 11 1,3-bis-1-naphthylpropane 116 1,3-bis-2-naphthylpropane 116 bis-acryloxybutyl tetramethyl disiloxane (BABTDS) 208 bisazide 196, 201 blood glucose biosensor 298 bobbin-type configuration 231 Boltzmann constant 21 Boltzmann distribution 107 bovine serum albumin 413 branched side chains 150 breakdown voltage 227 Brillouin zone 5 2-bromoethylamine 275 buried waveguide, end-face 92-93 butraldehyde 275 Buttiker-Landaur conductance formula 26

C 13

C NMR studies 406 C60 343 cadmium selenide (CdSe) 341, 353, 357 cadmium sulphide (CdS) 341, 353-355 calcium 370 camphor sulfonic acid (CSA) 38 capacitance-voltage characteristics 408 carbazole 107 carbon-carbon bonds 112 carbon-carbon distance 41 carbon-carbon double bonds 375 carbon-carbon single bonds 375 carbon nanotubes 267 carboxylic acid 195 carrier density 377 carrier mobility 375 carrier transport 369

catalytic biosensors 287-288 cathode 370 centralised testing 297 chain conformation effect on excimer emission 122-126 chain densification effect 113 chain interaction 9 chain interruptions 21 chain length 112-113 chain orientation 13, 169 chain scission reactions 119 chalcogenide glass fibre 84 channel waveguides 75, 87 charge carriers 375 charge-discharge cycle 238 charge-discharge process 242 charge generation and separation 346347 charge injection 143, 146, 148 charge separation 349 charge separation nanoparticle-polymer systems, photophysics 346-360 charge transfer excitons 143 charge transfer process 4, 348, 394 charge transfer TTF:TCNQ-type molecules 379 charge transport 3-36, 146, 406 charge transport carriers 6-12 charge transport models 18-27 chemical doping 4, 143 chemical sensors 287 chemical vapour deposition (CVD) 276 chloroform 112, 124, 398 polynaphthyl methacrylate in 111 chloroform-cyclohexane 112 chloromethylated polystyrene 203 chlorophyll 376 cholesterol biosensors 323-324 chromophores 99, 107, 113, 115, 118, 119 association of 103 (CH)x-FeCl3 42

455

Handbook of Polymers in Electronics (CH)x-I2 42 cis-polyacetylene 3, 395 cis-1,4-polyisoprene 196 cis-trans photoisomerisation 255 Clariant Corporation 201 C/LiClO4-EC-PC-PAN/LiCryMn(2-y)O4 234 C/LiClO4-EC-PC-PMMA/PPy 244 C/LiMO2 polymer lithium-ion batteries 244 ClO4-(CH)x 43 ClO4-doped polyacetylene, normalised resistivity (ρ) versus temperature 43 ClO4--doped polyacetylene 14 coatings 272 coin-type cell configuration 234 collapse transition 126-135 collisional energy transfer 118 colour components 376 colour purity 371, 373 enhancement 374 complementary metal oxide semiconductor (CMOS) circuits 284 concentration quenching 103 conducting films 271, 286 applications 403 properties 403 conducting polymers 1, 6, 217, 255, 257, 343, 375-378 analytical chemistry 412 applications 397 as odour sensing devices 416-417 band gap 395 biosensors 297-239 characterisation 397, 404-406 characteristics 419 charge transport models 18-27 conduction mechanism 305 electrical transport properties, summary 63-65 electrolytic expansion 267-268

456

in molecular electronics 393-439 applications 394, 397 magnetic susceptibility 56-58 molecular devices based on 406-424 specific heat 56-58 structures 378, 395 synthesis 304, 397-399 see also specific types and applications conducting properties 3-4 conduction band 4, 5, 10, 16, 18, 142143 conductivity see electrical conductivity conductometric biosensors 315 configuration co-ordinate 7 conformation change of PPMA 113 conformation effect of polymer chain 110-118 conjugated polymers 1, 5, 342, 367, 375 anisotropic electrical and optical properties 142 band gap 148 charge transport 3-36 light emission in entire visible spectrum 422-423 physical properties 397 physics of 142-144 preparation methods 344-346 synthesis 151 conjugation length 376 consumer testing 297 contraction ratio 260, 261 contraction ratio of polyaniline films 267 COP 206, 207 core channel waveguide 88 core-shell synthesis 341 Coulomb interaction 9, 11 Coulomb potential 16 covalent binding 308, 412 covalent bonding 307 critical states 58-63 crosslinking 307 through anhydride 204-205

Index crosslinking centres 223 crystal-crysal transition 134 crystalline iPS 123-125, 131, 133 Cu(II) ions 410 current density 375 current-voltage characteristics 371, 372, 408 cyano-PPV 157 cyanothylpullulan (CYEPL) 411 cyclic voltammogram (CV) 257, 259, 260 cyclised polyisoprene 203 3-cyclohexyl thiophene 420, 421

D data acquisition system 105 data analyser 105 deactivation, Birks’ scheme 103 decentralised testing 297 deep-UV photoresists 197 degenerate state 7 delayed emission 111 delayed fluorescence 102, 110 density of states 16, 57 DFWM 76 1,2-diazonaphthoquinone (DNQ) 193 photodecomposition 194 diblock copolymers 346 2,7-dibromo-9,9-bis(3,6dioxaheptyl)fluorene 163 dichloro-p-xylene 149 dielectric materials 271 dielectric relaxation 404 N,N-diethylaminonitroazobenzene 79 diethylzinc (DEZ) 285 differential scanning calorimetry (DSC) 123, 397 diffraction efficiency of waveguide grating 91 diffusion coefficient 41, 257, 260, 262 diffusion constant 257

dimethylsexithiophene (CMSxT) 411 diodes 406-408 dipole-dipole interaction 118 discharge-charge cycle 237 discrete particle encapsulation (DPE) 345 disorder 9, 12, 41, 50 influence on transport properties 1415 disordered metallic regime 41 DNA 413-414 immobilisation 325 DNA biosensors 324-325 DNA sensor 290 dodecylbenzene sulfonate (DBS) 324 dodecylbenzoyl sulfonic acid (DBSA) 38 dopant-chain interaction 16 doping 4, 143, 190, 242, 377, 393, 396, 397, 404 influence on transport properties 1516 double-layer LED 158 down-chain energy migration 118 DR1 92 drive voltage 369, 370 dual lithium ion 244

E elastic scattering 41 electrical conductivity 21, 40-49, 141, 305, 394, 396 anisotropy 53 Arrhenius plots 224, 229 power law behaviour 58 temperature dependence 41, 46 electrical properties of doped conjugated polymers 37-68 electrical transport properties of conducting polymers, summary 63-65 electric field 377 intensity 11 poling technique 87

457

Handbook of Polymers in Electronics electric vehicles (EV) 217 electrochemical biosensors 313, 315 electrochemical doping 4 electrochemical entrapment 307-308, 412 electrochemical oxidative polymerisation 398 electrochemical stability 227 electrochemical synthesis 167 electrode-less microwave or high frequency reactors 280-281 electrodes 369 preparation 304-305 electroluminescence 166, 347 aluminium 8-hydroxyquinolene (Alq3) 141 anisotropy 169 efficiency 145, 150, 157, 159, 161, 166, 168, 170 little or no 371 measurement 348 organic 367, 369, 370 polarised 168-170 polymer 368-374 electroluminescence response 354 electroluminescence spectra 421 electroluminescent devices physics of 142-147 potential for 423 electroluminescent displays 419-423 electrolytic capacitors 274 electrolytic deformation 256 measurements of 257 electrolytic expansion 256 anisotropy of, in polyaniline films 266 applications 268-269 measurement of 258 pH dependence of 262 time response of 265 electrolytic expansion in conducting polymers 267-268

458

electromechanical actuators 423 electron-beam irradiation 205 electron-beam lithography 186 electron-beam resists 202-206, 273 electron correlation 6 electron-electron (e-e) interactions 4-6, 13, 52, 53, 142 electron-energy-loss spectroscopy 11 electron-hole (e-h) pairs, recombination of 371 electron-hole (e-h) separation 142 electron-hole (e-h) transporter 167 electronically conducting polymers 217, 242-245 electronic encapsulation 284 electronic excitation energy 118 electronic ground state 4-6 electronic nose 415-417 electronic tongue 414-415 electron magnetic resonance spectroscopy 8 electron nuclear double resonance (ENDOR) 8 electron paramagnetic resonance (EPR) spectroscopy 23 electron-phonon coupling constant 21 electron-phonon interactions 4, 5, 6, 142 electron-phonon scattering 13 electrons 24 electron spectroscopy for chemical analysis (ESCA) 282, 283 electron spin resonance (ESR) 16, 404 electron transfer in amperometric biosensor 303-304 electron transporter 158 electro-optical (EO) materials 86-87 electropolymerisation 308 electropolymerised conducting films 313 ellipsometric technique 406 EL-V (EL intensity versus bias voltage) 147 emeraldine salt (ES) 260

Index emission efficiency 148 emission spectrum 371 emission wavelength 373 emitter materials 371, 373 emitting layers 369 emulsion polymers 272 encapsulation 371 end-to-end cyclisation 117 energy band diagram 396 energy bandgap 376 energy gap 3, 4, 375, 376, 395 energy migration coefficient 126 energy migration processes 118-121 energy migration studies 109-110 energy storage and conversion 217 energy transfer 110, 118-121 efficiency 118 fluorescence measurements of 121 types 118 environmental monitoring 326 enzyme acitivity, effect of temperature 311-312 enzyme activitiy, determination of 309310 enzyme activity profile 311 enzyme electrodes 309-313 enzyme field-effect transistors (ENFETs) 409 enzyme immobilisation 304-309 enzyme-immobilised electrode 274 enzyme linked immunosorbant assay (ELISA) 326 enzymes 302, 303 optimum pH 310-311 enzyme support system 287 etching 190, 191 ethyl acetate 112 ethylene carbonate (EC) 222 excimer dissociation 115 excimer emission 113 chain conformation effect on 122-126 excimer fluorescence 106-135, 111

excimer fluorescence lifetimes, iPS gel 129 excimer formation 115 effect of temperature 115 molecular weight effects on 113 excimer photophysics for two-phase system 133 excimers 103, 104, 143 exciplexes 104, 143 excitation, Birks’ scheme 103 excitation energy 119 excited singlet state 100, 101 excited triplet state 100 exciton generation 146 excitons 347-348 exciton trap 109 expansion ratio, dependencies of 260262 extensional actuators 424 external electrode reactors with RF power supply 280

F FeCl3-(CH)x 43 FeCl3--doped polyacetylene 14 FeCl3 polymerisation 167 Fe2O3 344 Fermi energy 13 Fermi level 4, 5, 56, 57, 145 Fermi velocity 41 ferrocene carboxylic acid 319 ferrocene-modified siloxane polymers 275 field-effect mobilities 410 field-effect transistors (FETs) 379, 409411 field-induced transition 61 W versus temperature 60 films see conducting films; thin films; ultrathin films and specific types flavin adenine 307

459

Handbook of Polymers in Electronics flavin adenine dinucleotide (FADH2) 275 Flory theory 111 fluctuation induced tunnelling (FIT) model 23-25 fluorescence 100, 101, 109, 111 decay 125 definition 101 delayed 102, 110 investigations 99 lifetimes 105 ‘normal’ 102 of polymers 106-135 in gel state 122-135 in solution 110-121 study methods 105 fluorescence emission spectrum for iPS gel 132 fluorescence measurements of energy transfer 121 fluorescence probe method 99-100 fluorescence spectra 111 of phenyl methacrylate 114 of poly(phenyl methacrylate) 113-115 fluorescence spectroscopy 105 fluorinated polymeric ethers 274 fluorinated polymethacrylate (FPB) 207 sensitivity curve 207 fluorinated polyquinoline 420 fluorosphores 341 food industries 297 Frenkel exciton 143 Fresnel reflection loss 89-90 Frohlich conductivity 6

G GaAs 285 gas chromatography (GC)/mass spectrometry (MS) 415 gel electrolytes 230, 232 gel state, fluorescence of polymers in 122-135

460

gel-type electrolytes 227-229 configuration 226 glass transition temperatures 86 glow discharge 277 glucose biosensors 318, 413 glucose oxidase (GOD) 274, 275, 287288, 297, 317, 318 glutamate dehydrogenase (GLDH) 314 glutaraldehyde 413 GMC 206, 207 GOD-PEB films 413 GOD-stearic acid monolayers 413 gold nanoparticles 346, 357-360 gold-polythiophene blends 357-360 GPC 404 graphite anode 233 ground state energy 7

H 1,4-heptadiene 397 heteroaromatic polymers 76 molecular structure of 80 χ(3) wavelength dependence 82 heterojunctions 406 hexamethyldisilazane 284 hexamethyldisiloxane (HMDSO) 284, 287, 290 3-n-hexyl thiophene 420, 421 high energy density batteries 217 high pressure liquid chromatography (HPLC) 415 hole transporter 158, 169 HOMO (highest occupied molecular orbital) 144-145, 376 hopping/tunnelling transport 24 Hubbard model 5, 6 Huckel model 15 humidity content 241 humidity sensors 287 hybrid waveguides 93-94 hydrochloric acid etchant 371

Index hydrostatic actuators 424 8-hydroxyquinolene (Alq3) 168 electroluminescence of aluminium 141 4-hydroxystyrene 200, 201

I I-(CH)x 43, 44, 45 immobilisation biomolecules 305-309, 412-413 DNA 325 enzyme 304-309 methods 305-309 immunosensors 326 imprinted polymers 271 impurity potentials 16 indium tin oxide see ITO inelastic electron-electron scattering 13, 41 inelastic electron-phonon scattering 45 inelastic scattering 41 information technology 69, 367 inorganic nanoparticles 346 insulated gate field-effect transistors (IGFETs) 410 insulating states 58-63 insulators 284-285 interchain coupling, influence on transport properties 17-18 interchain hopping 19 interchain interactions 4, 143 interchain transfer 16 interface materials 272 internal conversion (IC) 100-102 internal electrode reactors with dc power supply 280 with RF power supply 280 intersystem crossing (ISC) 100, 101, 103 intramolecular excimer formation 117 iodine 394 iodine-doped polyacetylene 19, 42, 141 conductivity (both parallel and

perpendicular) versus temperature 45 resistivity versus temperature 44 iodine-doped trans-polyacetylene 3, 13 iodine-doped Tsukomoto (CH)x 37 ion-beam lithography 186 ionically conducting polymers 217-242 ionic attraction 387 ion-selective field-effect transistor (ISFET) 273, 287, 409 ion-selective microelectrochemical transistor (ISMET) 410 iPS/BA gel 126 time-resolved fluorescence emission spectra 130 iPS gel 123 crystalline 123-125, 131, 133 density measurements 129 excimeric fluorescence lifetimes 129 extended conformation 128 fluorescence decay parameters 128 fluorescence emission spectrum 128 steady-rate fluorescence emission spectrum 132 time-resolved fluorescence emission spectra 131 IR spectroscopy 283 ISC see intersystem crossing (ISC) isobutylene 200 isotactic polystyrene see iPS ITO 144, 151, 159, 170, 353, 371, 413, 418 ITO/BDOH-PF/Ca 163 ITO-coated glass 141 ITO/PPV/Al 150 ITO/PPV/cyano-PPV/Al 158 ITO/PVCZ/alkoxy PPP/Ca 161 I-V (current intensity versus bias voltage) 147

J Jablonski diagram 100

461

Handbook of Polymers in Electronics

K Kerr switch 83 Kivelson and Heeger model 16 KPR resist 194, 195 KrCl 197 KrF 197

L lactate biosensors 321-322 lactate dehydrogenase (LDH) 312, 314, 321-322 lactate oxidase (LOD) 312-313, 321-322 lactate/pyruvate substrate recycling 322 ladder PPP (L-PPP) 161 Landau-Ginzberg model 17 Langmuir-Blodgett deposition 381, 401 X-type 400-402 Y-type 400-402 Langmuir-Blodgett films 271, 318, 381, 384, 385, 387, 399-404, 408, 413 applications 403-404 pressure-area isotherm 400 Langmuir-Blodgett technique 159, 168, 308, 413 Langmuir films 381, 383 Langmuir trough 381, 383-385, 401 laser power and SHG efficiency 90 layer-by-layer polyelectrolyte selfassembly process 387 LCST see lower critical solution temperature (LCST) least-squares analysis 105 LEDs 141-183, 355, 357, 367, 411, 419421 characterisation 147-148 complex polymeric structures used as active materials 154-157 degradation 170 double layer 158 lifetime 170 mechanisms involved 147

462

microcavities 170-171 narrowing of emission spectrum in microcavity 171 needle-like 151 physical mechanisms occurring in 142 physics of 144-146 polymeric structures 148 PPV derivatives used as active materials in 152-154 prototypes 142 recent developments 168-171 single layer 144 structural control 148 working mechanism 145 leuco-emeraldine salt (LS) 260 Lewis acids 196, 199, 223 Li+ cations 219 Li+ ion 223 Li-AyBz lithium battery 221 LiClO4 220 LiClO4-EC-DMC-PAN 232-234, 237 LiClO4-PC 243 LiCoO2 231 LiCoO2 cathode 236 LiCryMn(2-y)O4 electrode 234 lifetime parameter 105, 127 light emitting diodes see LEDs light-emitting polymers 367 Li/LiClO4-EC-DMC-PAN/LiCryMn(2-y)O4 battery 235 Li2MnO4 235 LiMn2O4 231 lineweaver-Burke plot 321 LiNiy Co1-yO2 231 Li/PEO-LiCF3SO3 + γLiAlO2/LiMn2O4 battery 225 Li/(PEO)20LiI-EC-Al2O3/FeS2 cell 224-225 LiPF6-EC-DMC 232 LiPF6-EC-DMC-PAN 225 LiPF6-EC-PC-PAN 227, 229, 230 liquid crystal devices 71 liquid crystal displays (LCDs) 168, 419

Index liquid crystalline polyimide, change in lifetime during heating 135 liquid electrolytes 217, 232 lithium batteries 218-239 lithium battery 243-245 R&D projects 223 lithium-doped conducting polymer 243-245 lithium intercalation-deintercalation processes 232-234, 238 lithium-ion battery 217, 230-231 schematic representation 231 lithium metal anode 243 lithium metal oxide cathode 233 lithium polymer electrolytes 218-239 lithium triflate 163 lithographic methods 188 contact mode 188 projection mode 188 proximity mode 188 lithography 286 Li4Ti5O12/LiPF6-EC-PC-PAN/LiMn2O4 cell 238 Li/TiS2 cell 222 LiX 218-219 lower critical solution temperature (LCST) 121, 134 luminance enhancement 371 luminance output 371 luminance-voltage characteristic 372 luminescence 101-103 basic principles 100 luminescence quenching 118 luminescence studies 99-140 luminescent organic compounds 367 luminous efficiency 369, 370 LUMO (lowest occupied molecular orbital) 144-145, 376

M Mach-Zehnder (MZ) interferometer 87 magnesium 370

magnetic disks 273 magnetic susceptibility of conducting polymers 56-58 magnetic susceptibility studies 23 magnetoconductance (MC) 45 in conducting polymers 50-54 maximum extinction coefficient 102 medical diagnostic biosensor application 297 MEH-PPV 151, 159, 347, 350-352, 354, 360 Meldrum’s acid 198 membranes 271-296 mercury lamp 188 mercury-rare gas (xenon) discharge lamp 188 metal/conducting polymer junctions 408 metal-insulator (M-I) transition 12 metal-insulator-semiconductor (MIS) diodes 406 metal-insulator transition 15, 18, 38-39, 55, 58, 59 metallic I-(CH)x, magnetoconductance versus field 51 metallic oriented-(CH)x 53 metallic polymers 13 metallic state in doped conducting polymers 39-49 metal oxide nanoparticles 345 methacrylate esters, PMMA copolymerised with 85 methacrylic acid 204 methacryloyl chloride 204 3-methyl-4′octyl-2,2′-bithiophene-5,5′diyl (3MOT) 76 methyl methacrylate 204 methyl methacrylate-butyl methacrylate copolymer 121 methyl methacrylate-ethyl methacrylate copolymer 121 methyl tetrahydrofuran (MTHF) 115 microactivity-based device structure 374

463

Handbook of Polymers in Electronics microactivity structure 371 microactuators 255-270, 423-424 applications 424 microbiosensors 276 microlithography development 189 dry development 189 exposure 188-189 processing 187-190 techniques 186 wet development 189 microresists for shorter wavelengths (MRS) 201 microscopic charge transport mechanism 54 miniaturisation 379 MIS field-effect transistors (MISFETs) 409 molecular devices based on conducting polymers 406-424 molecular diffusion 424 molecular electronics (ME) 368, 379 conducting polymers in 393-439 applications 394, 397 use of term 393 molecular interaction 99 molecular motion 99 molecular scale anisotropy 54 molecular scale transistors 367-368, 379 molecular structure conformational changes 256 of heteroaromatic polymers 80 molecular weight effects on excimer formation 113 Monazoline process 193, 195 MOPPV see poly-(2,5-dimethoxy-pphenylene vinylene (MOPPV) More’s law 185 MOSFETS 410 Mott-Hubbard model 6 MRS see microresists for shorter wavelengths (MRS)

464

multilayer resists (MLR) 208-209, 209 silicon-containing polymers 210 technology 209 muscles 255

N Naarman polyacetylene 14, 24, 25 NADH 321-322 Nafion 240 nanocomposite polymer electrolytes 223 nanocomposites 342 benefits 343 metal nanoparticles 344 particle shape 343 preparation methods 344-346 nanocrystal probes 341 nanocrystals 341 nanoparticle components 341-365 nanoparticle photocurrent spectral response 354 nanoparticle properties 341 nanoparticles 342 preparation methods 344-346 size-dependence 344 nanoparticle semiconductor-polymer systems 353-357 nanotubes 255 nanowires 418 naphthalene 107, 121 naphthyl-substituted polymethacrylate (PNMA) 110 n-BCMU-PDA 75-76, 76 n-butoxycarbonylmethyl urethane) (nBCMU-PDA) 74 n-doping 242 near-field scanning optical microscopy (NSOM) 358-359 needle-like LED 151 negative bipolaron 11 negative deep-UV resists 201-202 examples 203

Index negative electron-beam resists characteristics 206 formulae 207 negative magnetoresistance 59 negative polaron 10 negative polyimides 209-210 negative soliton 8 Neotronics 417 neutral soliton 8 nickel catalyst polymerisation 167 NIMA 381 n-methylpyrrolidone 398 N-methyl-2-pyrrolidinone (NMP) 257 NMP solution 266 non degenerate ground state 7, 9 non labelled PS 117 non linear optical (NLO) chromophores 69, 71 non linear optical (NLO) devices 69 non linear optical (NLO) materials 69 non linear optical (NLO) polymers 69 future targets for optical device applications 93-94 third-order 71-72, 82-84 non linear optical (NLO) properties 6998 non radiative energy transfer 121 non radiative resonance transfer 118 non radiative transition 109 ‘normal’ fluorescence 102 novolac resin 193 nuclear magnetic resonance (NMR) spectroscopy 8, 227, 415

O octadecanoic acid 381, 382 3-octadecanoylpyrrole (3ODOP) 402 3-octadecylpyrrole (3ODP) 402 Odour Mapper 417 odour sensors 414-417 one-dimensional chain 40

one-dimensional electronic system 5 1-methylnaphthalene 116 optical absorption edge 3 optical biosensors 314 optically detected magnetic resonance (ODMR) 10 optical properties, non linear 69-98 optical second-harmonic generation 386 optical signal processing 72 optical spectrometer 105 optical switching devices 71, 72 materials characteristics 84 optical technology 69 organic electroluminescence 367, 369, 370 organic molecules 393 oriented I-(CH)x, W versus temperature 59 oriented K-(CH)x, W versus temperature 59 osmium-containing polymers 274 oxadiazole 158 1,3,4-oxadiazole 158, 159 oxidation 10 resistance 148 oxidative coupling method 398

P PAB/GOD 320 packaging materials 284 PAN 225, 227, 230, 232, 243 PANI 3, 37, 38, 40, 42, 54, 159, 243, 256, 257, 274, 314, 316, 342, 344, 394, 395, 397, 402, 413, 423 electrolytic expansion 263 thermoelectric power 63 PANI-AMPSA 48 conductivity versus temperature 48 PANI-CSA 55, 56, 61 magnetoconductance 53 magnetoconductance versus field 54

465

Handbook of Polymers in Electronics W versus temperature 59 PANI-Nafion composite electrodes 320 PANI-PET 151 PANI-SiO2 344 PANI-TiO2 344 PANI-uricase electrode 313 parahexaphenyl (PHP) 161 Pariser-Parr-Pople model 6 parylene C (poly(chloro-p-xylene) 285 Pauli susceptibility 13, 56 PAV see poly(arylenevinylene) (PAV) PC see propylene carbonate (PC) PCHMT 165 PCHT 165 PCMS 206, 207 PDABT 166 p-doping 242 PDTB 81 PEDOT 53, 54, 344 normalised resistivity versus temperature 49 PEDOT-PF6, magnetoconductance 53 Peierls distortion 6, 14, 16, 21 Peierls-Hubbard model 6 Peierls instability 4, 142, 394 Peierls model 5, 6 PEO see poly(ethylene oxide) (PEO) PEO-LiCF3SO3 224 PEO-LiClO4 219, 220 PEO-LiX 220, 223, 225, 243 membranes 218-219 pernigraniline salt (Pas) 260 3,4,9,10-perylenetetracarboxylic dianhydride (PTCDA) 285 P(ETMA-co-MMA) 206, 207 PF6-doped PPy 38 phase and modulation fluorometry 105 phase separation 126-135 pH dependence of electrolytic expansion 262 pH effects 302, 310-311 phenyl 107

466

phenyl methacrylate, fluorescence spectra of 114 phenyloxysuccinimid 200 phonon freezing 21, 40 phonons 20, 24 phonon scattering 21 phosphorescence 100, 102, 109, 111 photobleaching 91 photochemical processes 105 photoconductors 273 photocurrent/dark current ratio 351 photocurrent measurement 348 photoelectron spectroscopy 406 photoexcitation 142, 143 photoinduced absorption 10, 143 photoisomerisation 256 photolithography 75, 92, 186, 273, 371 photo-locking 208 photoluminescence (PL) 142-143 efficiency 166 quantum efficiency 143, 145 spectra 372, 421 photon absorption 143 photonic crystals 350 photonics 367 photons 119, 188, 367, 398, 406 photooxidation 119 photophysical deactivation processes 100-104 photophysical processes 105, 108 photophysical quenching 104 photopolyimide 210 photopolymerisation 398 photopolymers 185 photoresists 185, 273, 367 characteristics 192 conventional 193 negative 194-197 one-component systems 194-195 positive 193-194 t-butyl carbonate (T-BOC) 199 two-component systems 196-197

Index see also polimide-based photoresists; specific types photosynthesisers 398 photovoltaic properties 347 pH sensitivity 316 phthalocyanines 385 physical adsorption 306 physical vapour deposition (PVD) 276 physical vapour synthesis 345 π-bond 375 π-conjugated final structure 73 π-conjugated polymers 72 processible 76-82 π-electron bonding 342 π-electron conjugated polymers 71 π-electron conjugated structure 3 π-electron conjugation 79 π-electron delocalisation 142 π-electron hopping matrix element 41 π-electrons 5, 37, 255, 256, 376 piezoelectric coefficient 379-380 piezoelectricity 386 Pioneer Corporation 369 plasma deposition technique 188 plasma etching 190 plasma polymerisation 276-282, 398 advantages over conventional polymers 278 gas parameters 281-282 general characteristics 277-278 history 277 power parameter 282 reactors 279-281 synthesis 278-282 plasma polymerisation/vapour deposition (PP/VD) 284 plasma polymers applications 283-290 characterisation 282-283 films 282-283 platelet reinforcement 343 PMAN see poly(o-methoxyaniline)

(PMAN) PMMA 92, 164, 197, 203, 207, 209, 225, 227, 243, 286, 319, 320 sensitivity curve 207 PNIPAM 122 PNMA see polymethacrylate (PNMA) PNVC 346 polarisation ratio 170 polarised electroluminescence 168-170 polaron-exciton levels 143 polaronic clusters 26 polarons 4, 6, 9-12, 27, 377, 396 energy levels and localised states 10 formation 9 instability 17 negative 10 polyacenaphthalene 115 polyacetylene 12, 14, 19, 20, 37, 40, 72, 255, 256, 342, 394, 397 polyacrylamide (PAAM) 240, 241-242 volume phase transition 134 polyacrylonitrile see PAN poly(3-alkoxycyanoterephthalylidene) 422 polyalkylfluorene 420 polyalkylthiophene (PAT) 38, 147, 267, 346, 422, 423, 1164 structure 166, 167 synthetic procedures 167 trans conformation 165 polyamic acid 211 polyamide 6 342-343 poly-o-aminobenzoic acid 413 polyaniline (PANI) see PANI poly(aniline-co-ortho anisidine) 398 polyaniline film 265 anisotropy of electrolytic expansion in 266 contraction under strain 267 electrochemistry 259-260 expansion behaviour 259-260 expansion ratio 260-261

467

Handbook of Polymers in Electronics polyaniline insulated gate field-effect transistors (PANI-IGFETs) 410 poly(o-anisidine) 402, 403 poly(aromatic sulfones) 197 poly(arylenevinylene) (PAV) 72-74, 76, 398 poly(2,5-benzoxazole) 342 poly(2,5-benzoxazole)-blockpoly(benzobisthiazole-2-hydroxy-1,4phenylene)-block-poly(2,5benzoxazole) 342 polybenzimidazole (PBI) 240 poly[bis(p-toluenesulfonate) of 2,4hexadiyne-1,6-diol] (PTS-PDA) 72 polycarbazole (PCz) 3, 395 polyconjugated chains 4, 37 polycyclohexyl methacrylate 133 poly(3-cyclohexylthiophene) 408 polydiacetylene (PDA) 72, 398 polydiacetylene-polyacetylenepolydiacetylene triblock polymer chains 342 poly(9,9-di(ethylhexyl)fluorene (PDF2/6) 163, 164 poly(9-9,dihexylfluorene) 162, 422 poly-2,6-dimethylphenylene oxide (PPO) 119, 121 poly(dimethyltetrafluoropropylmethacrylate) 204 poly-(2,5-dimethoxy-p-phenylene vinylene (MOPPV) 72-74 poly-(2,6-dimethyl-p-phenylene oxide) 120 poly(3-dodecylthiophene) 76 polyemeraldine base (PEB) 318, 403, 413 poly(ethoxyaniline) 403 poly(3,4-ethyelene dioxythiphene) 40 poly(ethylcyanoacrylate) 204 polyethylene (PE) 107, 398 poly(3,4-ethylenedioxythiophene) (PEDOT) 49, 159 polyethylene glycol/polyurethane copolymer 275

468

polyethyleneimine-H2SO4 240 poly(ethylene oxide) (PEO) 218, 219, 221-224 see also PEO polyethylenimine (PEI) 289 polyfluorene 162-164, 169 polyfuran (PFu) 3 polyheptadiene 8, 397 polyheterocycles 243 poly(hexafluorobutyl-methacrylate) 204 poly(3-hexylthiophene) 400, 405 poly(3-n-hexylthiophene) 408 polyimide 209-210, 273 formation 211 polyimide-based photoresists 209-210 characteristics 210 examples 212 polyindole 395, 413 poly(N-isopropylacrylamide) (PNIPAM) 121 poly-L-lysine 290 polymer chain 10 conformation effect of 110-118 polymer compatibility 121 polymer deposition technologies 379-387 polymer electroluminescence 368-374 polymers, electronics applications 367-391 polymer science 141 polymer spin coating process 380 polymer transistors 367 polymethacrylate (PNMA) fluorescence in different solvents 112 naphthyl-substituted 110 poly(o-methoxyaniline) (PMAN) 256, 261, 262 electrolytic expansion 262-264 poly[2-methoxy-5-(2′-ethylhexyloxy)-pphenylene vinylene] 422 poly(2-methyl-1-pentene) 205 poly(methyl-isopropenyl-ketone) (PMIPK) 203 polymethyl methacrylate (PMMA) 107,

121, 159 copolymerised with methacrylate esters 85 poly(3-methylthiophene) 40, 407 polynaphthyl methacrylate in chloroform 111 poly(olefin sulfones) 205, 206 polyparaphenylene (PPP) 37, 159, 160, 377, 398 poly(o-phenylenediamine) 305 polyphenol 305 polyphenylene derivatives 160-161 polyphenylenediyl-pyridinediyl (PPPY) 80, 81 poly(phenylene ethynylenes) 162 polyphenylenes 148-164 poly(phenylene vinylene) 11 poly(phenyl methacrylate) (PPMA) conformation change of 113 fluorescence spectra of 113, 114, 115 poly(p-phenylene) (PPP) 3, 20, 395, 397, 422 poly(p-phenylenevinylene) see PPV polypropylene (PP) 107 poly(pyridine-2,5-diyl) 81 polypyrrole see PPy polysiloxanes 284 poly(sodium 4-styrenesulfonate) 387 polystyrene (PS) 107, 119 fluorescence 120 lifetime data 125, 126 lifetime parameters 120 polystyrene sulfonated acid (PSSA) 159 polysulfone, degradation 205 polysulfur nitride 398 polythiazyl (SN)x 3 polythienylene vinylene (PTV) 37, 72, 74 poly(thiophene-2,5-diyl) 81 polythiophenediyl-bipyridinediyl (PTBPY) 80-82 polythiophenediyl-pyridinediyl (PTPY) 80-82

polythiophene (PT) 3, 37, 38, 76, 164168, 274, 285, 342-345, 394, 395, 398, 402, 404, 406 poly-4-vinylbiphenyl 107 polyvinylcarbazole (PVCZ) 107, 116, 306, 420 time-resolved fluorescence spectra for 116 time-resolved fluorescence spectra of 116 polyvinylcarbazole (PVCZ)/urease electrodes 316 polyvinylchloride (PVC) 319 poly(vinylcinnamate) 194 polyvinylidene difluoride (PVdF) 379-380 polyvinylidene fluoride (PVdF) 225 polyvinylnaphthalene (PVN) 107, 115, 126, 133 polyvinylphenol (MRS) 203 poly(4-vinylpyridine) 274 polyvinylpyrrolidone (PVP) 240, 346 poly-p-xylene 286 Poole-Frenkel effect 407 POPT 165 portable conformable masking (PCM) 208 positive bipolaron 11 positive deep-UV photoresists one-component systems 197 two-component systems 198 positive electron-beam resists 203-205 positive MR 61 positive photopolyimide resists, example 211 positive polaron 10 positive polyimide resists 210 positive soliton 8 post baking 190 potassium tert-butoxide 149 potentiometric biosensors 301, 315-317 power law behaviour of conductivity 58 p(PMMA-OM-MAN) 197 PPO see poly-2,6-dimethylphenylene

469

Handbook of Polymers in Electronics oxide (PPO) PPP see polyparaphenylene (PPP) PPPY see polyphenylenediyl-pyridinediyl (PPPY) PPV 3, 17, 37, 38, 40, 72, 74, 81, 141, 148-160, 342, 345, 347, 420, 422 cis double bonds 151 derivatives used as active materials in LED 152-154 electroluminescence spectrum 150 soluble copolymers based on 151 structure 149 sulfuric acid doped 45 synthetic procedure 149 PPV-AsF5 42 PPV-based ladder copolymers 423 PPV-based polymers 360 PPV-H2SO4 42, 53, 60 magnetoconductance 52 magnetoconductance versus field 51 resistivity versus temperature 46 PPy 3, 9, 10, 37, 38, 40, 42, 53, 54, 244, 267, 268, 274, 305, 316, 317, 318, 320, 324, 326, 342, 344, 346, 394, 395, 397, 398, 402, 406, 408, 413, 417, 418, 423 cathode 243 doped with PF6 (PPy-PF6) 46 films 307 heat capacity versus temperature 57 PPy-Fe2O3 344 PPy-ferrocene carboxylic acid based cholesterol biosensor 323 PPy-PF6 56, 58 heat capacity versus temperature 57 magnetoresistance versus resistivity ratio 62 normalised resistivity versus temperature 47 S(T) 55 W versus temperature 59, 60 PPy-polyvinyl sulfonate films 312

470

PPy-PVS/LDH electrodes 322 PPy-SiO2 344 PPy-p-toluenesulfonate (TSO) 57 PPy-ZrO2 344 pre-exponential function 105 propylene carbonate (PC) 222 proton membranes, conductivity 241 proton polymer electrolytes 239-242 PSTF 82 molecular structure 83 waveguide 84 PT see polythiophene (PT) PTBPY see polythiophenediylbipyridinediyl (PTBPY) PtO 343-344 PTOPT 165 PTPY see polythiophenediyl-pyridinediyl (PTPY) PTV see polythienylene vinylene (PTV) pulse fluorometry 105 PVCZ see polyvinylcarbazole (PVCZ) PVMS 206, 207 PVN see polyvinylnaphthalene (PVN) PVP see polyvinylpyrrolidone (PVP) p-xylene polymer 286 pyrene 107, 121 emission intensity 122 pyrene-labelled PS 117 pyridine 158 pyridine rings 81 pyroelectric coefficient 379-380 pyroelectricity 386 pyrrole/3ODOP film 402 pyrrole 398 pyruvate biosensor, calibration graph 314

Q q-1D 4 conduction path 21 electronic systems 39 fractals model 22

metallic chain 22 models 22, 23 Mott’s law 22 percolation model 27 transport 24 variable range hopping model 19 quantum lattice fluctuations 4 quantum yield 105-106 definition 105-106 quartz crystal microbalances (QCM) 289 quasi-one-dimensional material see q-1D quasi-one-dimensional (q-1D) material 4 quasi-phase-matching (QPM) polymer waveguide 90 technique 88 waveguide fabrication method 89 quenching processes 103, 104, 118 quinine sulfate 106 quinone-hydroquinone reaction 317 quinoxaline 158

R radiation 99 frequency 99 intensity 99 types 188 radiationless processes 103 radiative recombination 146 Raman spectroscopy 227 random dimmer model (RDM) 26 rate constant 101, 103, 104, 108, 110, 115, 130 reactive ion etching (RIE) 89, 92, 94, 190 reactors for plasma polymerisation 279 recombination of electron-hole pairs 371 radiative 146 recombination process 347-348 red-green-blue (RGB) emission 161 redox behaviour 259

redox current 257 redox potential 302 redox reaction 304 refractive index grating fabrication 90 regioregularity 167 resistivity, temperature dependence 43, 45 resistivity ratio 12, 57 resists adhesion 191 coating process 188 contrast 191 etching resistance 191 films 286 materials 193-210 negative 189, 192 positive 189, 192 removal (stripping) 190 requirements 190-192 sensitivity 191 solubility 191 special 208-211 see also multilayer resists (MLR) reversible electrochemical actuators 424 rhodamine B 106 RIE see reactive ion etching (RIE) rocking chair battery 230-231

S salicylic acid (SA) 241 sample preparation 257 sandwich configuration 347 SBAC polymer 76-78 synthetic scheme 78 transition dipole moment between 1Bu and 2Ag excited states 77 wavelength dependence 78 scanning electron microscopy (SEM) 23, 405, 418 scanning transmission microscopy (STM) 404

471

Handbook of Polymers in Electronics scattering rate 41 Schottky devices 406-408 Schottky diode 357 Schrieffer-Heeger model 15 second-harmonic generation (SHG) 88, 91 efficiency and laser power 90 second-order NLO polymers 84-87 self-assembled films 271 self-assembly monolayers 404 self-phase modulation (SPM) 83 semiconducting polymers 341-365 semiconducting state 6 semiconductive films 285 semiconductor nanoparticles 345 semiconductors 4, 11, 271, 273, 276277, 393 sensing devices 297 sequential monolayer tranfer process 384 sequential transfer process 383 serially-grafted polymer waveguides 8890 sexithienyl based FETs 410 σ bond 375 signal processing by optical technology 69 signal processing function 69-70 signal transmission function 69-70 silicon 375 SiO2 on 187 silicon chip 393 silicon-containing polymers, multilayer resists 210 silicon monoxide 284 silicon transistor 379 single crystals 380 single layer LED 144 single photon fluorometer 105 singlet excitation, processes occurring after 109 singlet exciton 143 singlet ground state 100 SiO2, on Si 187

472

SiO2 fibre 84 6FPBO 354, 355 Smoluchowski equation 118 SnO2/LiClO4-EC-DMC-PAN/LiNi0.8 Co0.2 O2 battery 237 SnO2/LiNi0.8 Co0.2O2 cell 237 SnO2/LiNi0.8Co0.2O2 electrodic couple 236 solid electrolyte interface (SEI) 233 solid-state polymer laser diodes 350 soliton band 18 solitons 6-9, 16, 19, 38, 394, 396 energy levels and localised level occupations 8 formation 8 interchain transport 20 physical properties 8 special resists 208-211 specific heat of conducting polymers 5658 spectral bandwidths 373 spectral sharpening 371 spectral studies 105 spectroscopic methods 406 spin coating 151, 188, 346, 380 sputter etching 190 sputtering 273 stability 141, 148 steady-rate fluorescence emission spectrum for iPS gel 132 Stern-Volmer equation 104 Stokes law 99 structure modulation 157 sub-gap optical absorption bands 143 substituted benzylidene aniline with chloride polymer see SBAC substituted polyparaphenylene 160 sulfonium precursor polymer 149 sulfuric acid doped PPV 45 supercapacitor research and development 217 surface plasmon resonance (SPR) 273, 289

surface polarity 148 Su-Schrieffer-Heeger (SSH) model 6, 17, 142 sweep voltametry 227, 230

T T1/2 law 14, 44 taste-sensing field-effect transistor (TSFET) 415 taste sensors 414-415 T-BOC-PHS 200 decomposition in acid medium 200 T-BOC polymers 200 t-butyl acrylate 200, 201 t-butyl carbonate (T-BOC) photoresists 199 Teflon 398 temperature coefficient of resistance (TCR) 13, 14, 38, 39, 41-43, 47, 48, 49, 55 terephthaldicarboxaldehydes 148 terephthalic aldehyde 77 tetracyanoquinodimethane (TCNQ) 274, 286 tetraethylorthosilicate (TEOS) 418 tetrahydrofuran-ether 111 tetrahydrofuran (THF) 113, 120 2,2′,7,7′-tetrakis(N,N-di-pmethoxyphenyl-amine)9,9′spirobifluorine 353 tetramethylammonium hydroxide 73 tetramethyltin (TMT) 285 tetraphenyl diaminobiphenyl (TPD) 158 TGA 397 thermal ageing test 86 thermal property measurements 57 thermoelectric power, temperature dependence 55 thermoelectric power (S(T)) of conducting polymers 55 thermo-optical polymer devices 71

thermotropic liquid-crystallime polyimide 135 thermotropic liquid-crystalline (LC) polyimide (P-11TPE) 134 2,5-thienylene bis(methylene-dimethylsulfonium chloride) 72-73 thin films 71-74, 77, 91, 148, 271, 284 applications 273 new materials 271 thiophene, plasma-polymerised 285 thiophene rings 81 third-harmonic generation (THG) 74, 76, 83 third-order NLO polymers 71-84 research background 71-72 waveguides 82-84 three-dimensional band calculations 17 three-dimensional coupling 17 three-fold helical conformation 132 tight-binding model 17 time-correlated single-photon counting studies 105 time-dependent studies 105 time-of-flight (TOF) measurements 397 time-resolved fluorescence spectra iPS/BA gel 130 iPS gel 131 PVCZ 116 time-response of electrolytic expansion 265 TiO2 343 TiO2-conjugated polymer composites 348-353 TiS2 221 toluene sulfonic acid 264 trans-cis-trans isomerisation 91 transducers 299, 301, 313 transistors molecular-scale 367-368 polymer 367 transistor-transistor logic (TTL) 87 transmission electron microscopy (TEM) 23

473

Handbook of Polymers in Electronics trans-polyacetylene (PAc) 3, 7, 8, 11, 13, 17, 19, 394, 395 transport properties of polymers 12-14 factors influencing 14-18 transport property measurements 39 triazole 158 TRIDSB 157 trifluoromethane sulfonic acid 199 triphenyl sulfonium triflate 199 triplet exciton 144 triplet state 101, 111 tris (2,2′-bipyridyl ruthenium) complex 398 4,4′,4′′-tris(1-naphthyl)-(N-phenylamino)triphenylamine 169 tunable photonic crystals (TPCs) 350 two-phase system, excimer photophysics for 133 two-photon resonance 86

U ultrathin films 286, 399-404 ultraviolet photoelectron spectroscopy (UPS) 11, 159 unoriented I-(CH)x, W versus temperature 59 unsubstituted PPV 158 unsubstituted thiophenes 164 urea biosensor 320

very large scale integration see VLSI vibrational relaxation (VR) 100, 102 vinyl polymers with substituents R and R′ 106-107 VLSI, minimum size versus year 185 VLSI electronic circuits 273 VLSI technology 185, 212, 276-277 voltage profile 233 volume phase transition of PAAM 134

W Wannier exciton 143 water-soluble polymers 387 waveguide grating 92 diffraction efficiency of 91 waveguides 69, 71 third-order NLO polymer 82-84 weak localisation (WL) 50, 52, 54 Wessling method 149 Wilhelmy plate 381 Wittig condensation 148

X X-ray diffraction (XRD) 23, 39, 74 X-ray lithography 186, 206 X-ray photoelectron spectroscopy (XPS) 282, 283 X-ray resists 206-208

V

Z

vacuum-deposited SBAC polymer 77 valence band 4, 10, 143 variable-range hopping (VRH) model 19, 22, 23, 25, 61-62

Ziegler-Natta catalyst 397 ZnSe 353 Z-type deposition 401-402

474