PEDOT
Principles and Applications of an Intrinsically Conductive Polymer
6911X.indb 1
10/1/10 8:58:45 PM
6911X.ind...
355 downloads
1165 Views
5MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
PEDOT
Principles and Applications of an Intrinsically Conductive Polymer
6911X.indb 1
10/1/10 8:58:45 PM
6911X.indb 2
10/1/10 8:58:45 PM
PEDOT
Principles and Applications of an Intrinsically Conductive Polymer
Andreas Elschner Stephan Kirchmeyer Wilfried Lövenich Udo Merker Knud Reuter
Boca Raton London New York
CRC Press is an imprint of the Taylor & Francis Group, an informa business
6911X.indb 3
10/1/10 8:58:45 PM
CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2011 by Taylor and Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-13: 978-1-4200-6912-9 (Ebook-PDF) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright. com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com
Contents Foreword..................................................................................................................xi Preface.................................................................................................................... xiii Acknowledgments.............................................................................................. xvii Authors.................................................................................................................. xix Abbreviations....................................................................................................... xxi 1. The Discovery and Development of Conducting Polymers...................1 1.1 The Scope of This Historical Overview..............................................1 1.2 Introduction............................................................................................2 1.3 An Early Example: Polyaniline............................................................4 1.4 The First Electrically Conductive Poly(Heterocycle): Polypyrrole..............................................................................................9 1.5 The Fundamental Breakthrough: Doped Polyacetylene................ 10 References........................................................................................................ 15 2. Conductive Polymers versus Metals and Insulators.............................. 21 2.1 Metals, Semiconductors, and Insulators........................................... 21 2.2 Conjugated Polymers...........................................................................22 2.3 Temperature-Dependent Conductivity............................................. 24 2.4 Order and Disorder............................................................................. 26 References........................................................................................................ 29 3. Polythiophenes: A Chance for Maximum Conductivity?..................... 31 3.1 Introduction.......................................................................................... 31 3.2 Oxygen-Substituted Polythiophenes................................................. 33 References........................................................................................................ 38 4. A Short History of the PEDOT Invention................................................ 41 References........................................................................................................ 46 5. The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties.............................................................................. 47 5.1 Monomer Synthesis............................................................................. 47 5.2 Physical Properties............................................................................... 50 5.3 Chemical Properties............................................................................ 53 References........................................................................................................63
v
6911X.indb 5
10/1/10 8:58:48 PM
vi
Contents
6. From EDOT to PEDOT: Oxidative Polymerization and Other Routes........................................................................................... 67 6.1 Oxidative Polymerization and Doping............................................. 67 6.2 “Self-Oxidation” of EDOT Halogen Derivatives............................. 72 6.3 The Organometallic Route to PEDOT............................................... 74 6.4 Neutral, Undoped PEDOT by Oxidative Polymerization.............. 76 References........................................................................................................ 79 7. Counterions for PEDOT...............................................................................83 7.1 Counterions in Electrochemically Polymerized PEDOT...............83 7.2 Counterions in Chemically Polymerized PEDOT........................... 86 References........................................................................................................ 87 8. The In Situ Polymerization of EDOT to PEDOT.................................... 91 8.1 Synthesis of In Situ PEDOT................................................................ 91 8.2 Properties of In Situ PEDOT............................................................... 97 8.3 In Situ Polymerization of EDOT Derivatives and Relatives......... 102 References...................................................................................................... 109 9. PEDOT:PSS................................................................................................... 113 9.1 PEDOT:PSS Dispersions.................................................................... 113 9.1.1 Introduction........................................................................... 113 9.1.2 Polyelectrolyte Complexes................................................... 113 9.1.3 Synthesis of a PEDOT:PSS Complex.................................. 117 9.1.4 Commercial PEDOT:PSS Types and Their Properties..... 122 9.2 Properties of PEDOT:PSS.................................................................. 123 9.2.1 Deposition of PEDOT:PSS.................................................... 123 9.2.2 Thin-Film Properties............................................................ 123 9.2.2.1 Thermal and Lifetime Stability........................... 123 9.2.2.2 UV Stability............................................................ 126 9.2.2.3 Water Uptake......................................................... 128 9.2.2.4 Mechanical Properties.......................................... 130 9.2.2.5 Morphology: Surface and Bulk........................... 131 9.2.3 Electronic States.................................................................... 136 9.2.3.1 UV-Vis (Ultraviolet-Visible) Spectra................... 136 9.2.3.2 Energy Levels in PEDOT...................................... 137 9.2.3.3 Optical Constants.................................................. 142 9.2.3.4 Vibrational Spectra................................................ 143 9.2.4 Electrical Properties............................................................. 144 9.2.4.1 Conductivity.......................................................... 144 9.2.4.2 Microscopic Model for Conductivity in PEDOT:PSS........................................................ 144 9.2.4.3 Free Charge Carrier Mobility.............................. 147 9.2.4.4 Threshold Currents............................................... 148
6911X.indb 6
10/1/10 8:58:48 PM
Contents
vii
9.3
Secondary Doping............................................................................. 149 9.3.1 Introduction........................................................................... 149 9.3.2 The Chemical Nature of Secondary Dopants in PEDOT:PSS........................................................................ 150 9.3.3 Properties of PEDOT:PSS Films Including Secondary Dopants............................................................... 152 9.3.3.1 Conductivity as a Function of Temperature...... 152 9.3.3.2 X-Ray Diffraction................................................... 152 9.3.3.3 Optical Characterization of PEDOT:PSS Films........................................................................ 153 9.3.3.4 Surface Analysis of PEDOT:PSS Films............... 153 9.3.3.5 Atomic Force Microscopy and Scanning Tunnel Microscopy................................................ 154 9.3.3.6 Work Function and Electron Paramagnetic Resonance............................................................... 155 9.3.4 Discussion.............................................................................. 156 References...................................................................................................... 158 10. Applications.................................................................................................. 167 10.1 Solid Electrolyte Capacitors.............................................................. 167 10.1.1 Introduction........................................................................... 167 10.1.2 Capacitor Basics..................................................................... 169 10.1.3 Design of Solid Electrolyte Capacitors............................... 170 10.1.4 Deposition Methods for PEDOT Cathode......................... 172 10.1.4.1 Electrochemical Oxidative Polymerization....... 174 10.1.4.2 Chemical Oxidative Polymerization.................. 175 10.1.4.3 Conducting Polymer Dispersions....................... 177 10.1.5 Reformation and High Voltage Application..................... 179 10.1.6 Self-Healing and Thermal Runaway................................. 182 10.1.7 Conclusions............................................................................ 183 10.2 Through Hole Plating for Printed Wiring Boards........................ 184 10.3 ITO Substitution................................................................................. 188 10.4 Antistatic Coatings............................................................................ 194 10.4.1 Solvents................................................................................... 198 10.4.2 Surfactants............................................................................. 199 10.4.3 Binders.................................................................................... 199 10.4.4 Hardness and Abrasion....................................................... 200 10.4.5 Conductivity-Enhancing Additives................................... 201 10.4.6 Use of PEDOT in Antistatic Coatings................................ 202 10.5 Electroluminescent Lamps............................................................... 204 10.6 Organic Light Emitting Diodes (OLEDs)........................................ 205 10.6.1 Introduction........................................................................... 205 10.6.2 PEDOT:PSS as a Hole-Injection Layer................................ 207 10.6.3 The PEDOT:PSS–Semiconductor Interface........................ 209
6911X.indb 7
10/1/10 8:58:49 PM
viii
Contents
10.6.4 Lifetime Restraints by PEDOT:PSS Hole Injection Layers..................................................................... 213 10.6.5 Modified PEDOT-Based Materials for HILs..................... 214 10.7 PEDOT:PSS in Organic Solar Cells.................................................. 216 10.7.1 Introduction........................................................................... 216 10.7.2 PEDOT:PSS as a Transparent Anode in OSCs.................. 217 10.7.3 PEDOT:PSS as a Buffer Layer in OSCs.............................. 218 10.7.4 PEDOT:PSS in Dye-Sensitized Solar Cells........................ 221 10.8 Electrochromic Behavior...................................................................222 10.8.1 Introduction...........................................................................222 10.8.2 Control of Optical Properties.............................................. 226 10.8.3 EDOT Derivatives................................................................. 229 10.8.4 Copolymers............................................................................ 231 10.8.5 Organic–Inorganic Hybrid Polymers................................. 233 10.8.6 PEDOT with Pendant Electrochromic Dyes..................... 233 10.8.7 Blends and Layer-by-Layer Deposition.............................234 10.8.8 Electrolytes............................................................................. 235 10.8.9 Ion Storage Materials............................................................ 236 10.8.10 Dual Polymer Cells............................................................... 237 10.8.11 Substrates and Patterning.................................................... 237 10.9 Organic Field-Effect Transistors...................................................... 238 10.9.1 Introduction........................................................................... 238 10.9.2 PEDOT:PSS as Electrodes.................................................... 239 10.9.3 PEDOT:PSS as an Interlayer................................................ 241 10.9.4 PEDOT:PSS as an Active Layer........................................... 242 References...................................................................................................... 244 11 Technical Use and Commercial Aspects................................................. 265 References...................................................................................................... 269 12 EDOT and PEDOT Derivatives with Covalently Attached Side Groups............................................................................................................ 271 12.1 EDOT-CH2OH and Its Derivatives.................................................. 271 12.2 EDOT-CH2Cl and Its Follow-Up Products..................................... 280 12.3 Alkyl EDOTs....................................................................................... 282 12.4 Water Soluble, “Self-Doping” EDOT Derivatives.......................... 286 References...................................................................................................... 289 13 XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs: Ring Size Variations, Heteroanalogs, and Derivatives of EDOT with Substituents at the Thiophene Ring.............................................. 293 13.1 3,4-Methylenedioxythiophene (MDOT)......................................... 293 13.2 ProDOT (Propylenedioxythiophene) Derivatives......................... 295 13.3 Vinylenedioxythiophene (VDOT) and Benzo-EDOT................... 299
6911X.indb 8
10/1/10 8:58:49 PM
Contents
ix
13.4 3,4-Ethyleneoxythiathiophene (EOTT)........................................... 301 13.5 3,4-Ethylenedithiathiophene (EDTT)..............................................304 13.6 3,4-Ethylenedioxypyrrole (EDOP) and Its Derivatives.................306 13.7 3,4-Ethylenedioxyselenophene (EDOS)...........................................309 13.8 2,5-Disubstituted EDOT Derivatives [2(,5)-X(2)-EDOTs]................ 311 References...................................................................................................... 322 14 The Electrochemical Behavior of EDOT and PEDOT......................... 329 References......................................................................................................340 Index......................................................................................................................345
6911X.indb 9
10/1/10 8:58:49 PM
6911X.indb 10
10/1/10 8:58:49 PM
Foreword After the discovery of electrically conducting polymers in 1977 by Professors Heeger, MacDiarmid, and Shirakawa (Nobel Prize in Chemistry 2000), and more than 30 years of worldwide intense research and huge efforts, PEDOT, or poly(3,4-ethylenedioxythiophene), sets various standards for the entire field. PEDOT, which was invented in 1988 by Bayer AG, Leverkusen, is probably the best conducting polymer available in terms of conductivity, processability, and stability. Furthermore, PEDOT is the only conducting polymer that is commercially produced on a large-scale (nowadays mainly by H.C. Starck Clevios GmbH, Leverkusen) and sold for many applications. H.C. Starck Clevios GmbH and its predecessors (Bayer AG and H.C. Starck GmbH) in a consequent manner advanced PEDOT to the highly developed commercial product that is presently available in various formulations and conductivities adapted to the needs and specific industrial applications of the customer. The scientists at H.C. Starck Clevios succeeded in making an originally inherently insoluble polymer processable, mostly as dispersion, by optimizing monomers, polymerization route, composition of the dispersion, counteranions, and by secondary doping. The text covers all relevant aspects of PEDOT beginning with a historical view on conducting polymers and polythiophenes, in particular. The story continues by describing the invention of PEDOT based on the development of the suitable monomer EDOT and subsequent important polymerization routes to the conducting polymer. The properties of PEDOT depend on counterions, which led to the development of PEDOT:PSS, or poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), dispersions, which is the basic form of the commercial product. In the second part of the book, important applications in electronics and organic electronics concomitant with technical and commercial aspects are extensively described. This comprehensive book about PEDOT, written by experts from H.C. Starck Clevios GmbH, Leverkusen, will represent an indisputable and valuable source for researchers, developers, and users of PEDOT. I wish the book great success. Peter Bäuerle
xi
6911X.indb 11
10/1/10 8:58:49 PM
6911X.indb 12
10/1/10 8:58:49 PM
Preface In 2000, the Nobel Prize for Chemistry was awarded to Alan J. Heeger, Alan G. MacDiarmid, and Hideki Shirakawa “for their discovery and development of conducting polymers” (as written on their Nobel Prize diploma). This prize not only appreciated the scientific work of the Nobel Laureates and the universal importance of their research; a new class of chemical compounds, not broadly known by the public before, came into the limelight. Since then, the conductive polymers have won growing attention in the scientific world, and the public now has more benefits from innovations due to the increasing technical usage of conductive polymers. Poly(3,4-ethylenedioxythiophene), abbreviated PEDOT or PEDT, belongs to the nevertheless moderate amount of conductive polymers, which have not only attracted remarkable scientific interest but also serve as technically used materials in different products of modern life. The story of this book started several years ago with a lecture during the Fπ5 Conference in Ulm/Neu-Ulm (Germany) at the 5th International Symposium on Functional π-Electron Systems. S.K. was invited to hold a plenary lecture about PEDOT from the industrial point of view or, more precisely, from the standpoint of a PEDOT manufacturer. In this situation, the lecturer had to bring together the presentation of fundamental scientific investigations and the description of a technical product with advantageous properties covering not only the facts, but also a little “technical forecast.” This seemed to have been fulfilled rather successfully, as the demand for more information about PEDOT steadily grew within the scientific community during the years that followed. The authors were repeatedly asked for further lectures, reviews, and book chapters about PEDOT. The growing commercial product paralleled a growing need for easily available and concise, yet comprehensive publications. When CRC Press invited us to write an entire book about PEDOT, we were initially not too excited by the suggestion. Refusing this kind invitation was a seriously discussed topic. PEDOT was the focus for all of us: Producing commercial quantities of PEDOT; serving customers; investigating PEDOT chemistry and physics, especially in PEDOT-based devices; writing patent applications regarding new PEDOT-relating inventions; publishing full or preliminary papers, giving conference lectures about PEDOT; and so on. PEDOT was a full-time job for all of us, without the added task of writing a book. On the other hand, PEDOT is, roughly speaking, only one chemical and known for only about two decades. How could one fill more than 250 book pages with information a reader would be interested in? Could there be enough material worth a deeper discussion in a book? xiii
6911X.indb 13
10/1/10 8:58:49 PM
xiv
Preface
Most arguments against this project were extinguished as we looked at the number of PEDOT patents and scientific papers published every year. Between 1989 and 2005, these numbers nearly followed a steadily increasing exponential function, every year exceeding the numbers of the last one. Since then, a slight slowing occurred. Yet at its high level of more than 1000 documents, every year outnumbered the previous year by 10% or more until 2008. And for 2008–2009, the last years with complete figures, an approximately constant value of about 1500 documents per year was reached. It was obvious that a remarkable interest in PEDOT within the scientific community exists, and since about 40% of these figures represent patent applications, an additional intense industrial interest in PEDOT was also demonstrated. Meanwhile, a three-digit number of companies had generated inventions utilizing PEDOT, a persuasive fact symbolizing the industrial success of PEDOT. When the first conductive polymers were created, technical applications were soon discussed, but a big commercial success was not foreseeable. When PEDOT was invented in 1988, not a single technical application for conductive polymers existed or, to be more precise, no realized application existed although a huge number of potential applications had been announced. Since then this has changed dramatically, to a large extent due to various utilizations of PEDOT. Several other highly conductive polymers, known even earlier than PEDOT, were also introduced into the market and have found some technical usage, such as polyaniline and polypyrrole. But PEDOT still remains the preeminent example because of its very pale color and high transparency in combination with its high conductivity and stability. Another important point is that PEDOT has stimulated basic scientific research in many fields. Improvements and progress in the fields of, for instance, light-emissive display or semiconductor research, were substantially facilitated by the incorporation of PEDOT layers into the devices. So, after some pros and cons, we dared to write this book. The concept of the book is to meet the requirements of readers from different directions. Great emphasis has been presented on the technical usage of PEDOT. We try to demonstrate the enormous and steadily growing applicatory relevance of PEDOT. Much space is dedicated to the applications of PEDOT including the chemical and physical background for technical utilization. If we can inform the reader about the numerous and distinct technical products containing PEDOT, an important goal of this book will have been met. In other words, explaining the chemistry behind, as far it concerns the conductive polymer PEDOT, cellular phones, LED lamps, and other chemical products is one objective of this book. Another intention of the book is to provide broad information about the chemistry and physics of PEDOT. There has been a vast amount of interesting new chemistry with EDOT, PEDOT, and its derivatives that has been published in the last 20 years, only to some extent covered by comprehensive reviews until now. Perhaps specialists in the field of conductive polymers will not need the repetition of extremely detailed information, but it
6911X.indb 14
10/1/10 8:58:49 PM
Preface
xv
is hoped that they too will appreciate the compilation now available in one book. But with the goal to meet the expectations of specialists, we added several unpublished results from our work and also included a lot of information from patents, which are sometimes not intensely taken into consideration in scientific papers. Therefore, we often tried not to go into too much detail and provided enough references for further study for chemists and physicists who have only marginally engaged in conductive polymers. We apologize to all scientists active in (P)EDOT chemistry who may miss major contributions: Several thousand (P)EDOT publications were checked by the authors, and assessment or even locating the most important ones may have been inaccurate due to the huge number of scientific papers and patents, and the human factor. Last but not least, the historical development of conductive polymers from laboratory curiosities of unknown structures to small-scale chemicals of paramount scientific importance to multiton commercial technical products will be presented. Only a few publications until now have dealt with the science history of conductive polymers, and often only parts of the story have been told. The viewpoint of an industrial research group with the background of more than 25 years of continuous conductive polymer research will be incorporated into the historical description, it is hoped, without expanding this part of the book and ending at the time when PEDOT was commercialized. The reader will find that this book is not as complete as a handbook, not always as detailed as a specialized review, not always as up-to-date as a rapid communication, not always as scientifically deep as a peer-reviewed full paper, and not always as readable as a popular science (for instance, history of science) article. But the book tries to combine all these aspects for PEDOT, with additional input of formerly unpublished results and personal opinions from the authors. We hope the readers will enjoy this book as much as we enjoy PEDOT chemistry and physics.
6911X.indb 15
10/1/10 8:58:49 PM
6911X.indb 16
10/1/10 8:58:49 PM
Acknowledgments This book is based on extensive work performed both within and outside the authors’ company, H.C. Starck. Of course, without the fundamental discovery of highly conductive doped polyacetylene nothing would have happened. But it was the merit of Jürgen Hocker to start a Conductive Polymer Project within Bayer AG, at that time a rather ambitious project for an industrial company. It is not typical for a profit-oriented global player, busy in nearly every field of industrial chemistry, to fund such a basic research project in the long-term. Here, the vision of Rolf Dhein to believe in the future of conductive polymers on an industrial scale was decisive for overcoming a lot of years without any commercial success. Friedrich Jonas and Gerhard Heywang invented PEDOT after seven years of conductive polymer research within Bayer AG, and this innovative breakthrough cannot be appreciated enough. The complete work in the industrial and the scientific world with PEDOT is based on this innovation and unthinkable without it. Jürgen Heinze was the first researcher outside Bayer AG to recognize the enormous scientific potential of the PEDOT invention, and his contributions were a strong support for the further development in the initial years. This book would not have been possible without the encouragement of Jill Jurgensen and Allison Shatkin from Taylor & Francis/CRC Press, and their patience with the authors. The management of the authors’ company, particularly Gerhard Gille (H.C. Starck GmbH) and Aloys Eiling (H.C. Starck Clevios GmbH), accepted and supported the work of their employees, which had to be done in addition to the day-to-day business—not self-evident in this era of economic challenges. Thanks also to Sonja Raida and Aynur Cansay of our company, who transformed the figures into an acceptable shape. Last but not least, the scientific and professional contributions of Friedrich Jonas, Matthias Intelmann (H.C. Starck Clevios GmbH), and Peter Bäuerle (University of Ulm) and their continuous discussions are highly appreciated.
xvii
6911X.indb 17
10/1/10 8:58:49 PM
6911X.indb 18
10/1/10 8:58:49 PM
Authors Andreas Elschner, Ph.D., was educated as a solid-state physicist at the University of Marburg (Germany) where he received his Ph.D. in 1988. Following a postdoctoral year at Stanford University (California) he joined Bayer AG in 1990, and has been with H.C. Starck since 2002. Dr. Elschner’s research focus is on organic electronics and he is responsible for testing and characterizing organic devices and conducting polymers. Stephan Kirchmeyer, Ph.D., studied chemistry from 1978 to 1984 at the University of Hamburg (Germany) and at the University of Southern California in Los Angeles. Until 2001, Dr. Kirchmeyer worked as a researcher for IBM and Bayer AG. In 2002, he joined H.C. Starck GmbH and since then has held several responsible positions for H.C. Starck’s business with conductive polymers and electronic materials. Wilfried Lövenich, Ph.D., received his diploma in chemistry from the Technical University of Aachen (Germany). He then went to the University of Durham, Great Britain, to obtain his Ph.D. In 2002, Dr. Lövenich joined H.C. Starck, working as an R&D chemist on the development and pilot plant production of the conductive polymer PEDOT. Since 2009, Dr. Lövenich has been the head of the R&D group of H.C. Starck Clevios GmbH. Udo Merker, Ph.D., studied physics at the University of Bonn (Germany) from 1989 to 1994. He received his Ph.D. in 1998 for studies in molecular spectroscopy at the University of Bonn and Princeton University (New Jersey). From 1998 to 1999, Dr. Merker was a postdoctorate at the Chemistry Department of Princeton University. In 1999, he joined the corporate research division of Bayer AG to work on the development of electronic materials. From 2002 until 2008, Dr. Merker was responsible for the development of new materials and processes for electrolytic capacitors in the central R&D division of H.C. Starck GmbH. Since 2009, he has been the head of the application technology group of H.C. Starck Clevios GmbH. Knud Reuter, Ph.D., studied chemistry from 1969 to 1974 at the University of Dortmund (Germany) where he received his doctoral degree with a thesis in organometallic chemistry in 1977. In the same year, Dr. Reuter started his professional work as a member of a polymer research group at Bayer AG. Since 2000, he has worked on PEDOT chemistry, joining H.C. Starck GmbH in 2002.
xix
6911X.indb 19
10/1/10 8:58:50 PM
6911X.indb 20
10/1/10 8:58:50 PM
Abbreviations AFM DMF DMSO EDOT EPR ESR GPE HPE MDOT μC/g NMP Ohm/sq Ω/sq PEC PEDOT ProDOT PSS SEM STM S/cm TCNQ THF UPS VRH XRD
Atomic force microscopy N,N-Dimethylformamide Dimethyl sulfoxide 3,4-Ethylenedioxythiophene Electron paramagnetic resonance Equivalent series resistance Guest polyelectrolyte Host polyelectrolyte 3,4-Methylenedioxythiophene Microcoulomb/gram N-Methyl-2-pyrrolidone Ohm/square Ohm/square Polyelectrolyte complex Poly(3,4-ethylenedioxythiophene) 3,4-Propylenedioxythiophene Polystyrenesulfonic acid Scanning electron microscopy Scanning tunnel microscopy Siemens/centimeter Tetracyanoquinodimethane Tetrahydrofurane Ultraviolet photoelectron spectroscopy Variable range hopping X-ray diffraction
xxi
6911X.indb 21
10/1/10 8:58:50 PM
6911X.indb 22
10/1/10 8:58:50 PM
1 The Discovery and Development of Conducting Polymers
1.1 The Scope of This Historical Overview A lot of fragments of the conducting polymers scientific history can be found in the chemical literature or in Internet articles. There are several specialized chapters in monographs, and also remarks in publications, dealing with early investigations in the special field of the respective paper, including a more or less adequate number of citations. Also anecdotes, dealing with the influence of the human factor in the history of conductive polymers can be found, particularly on the Internet. The following chapter tries to give an overview from the perception of the authors. A sufficient compromise between details, completeness, understandability for nonspecialists, and novelty for specialists is not easy to find. Furthermore, historical aspects can be a matter of subjective assessments. As a result, a combination of hard facts and information on one side mixed with personal opinion, and older information, replenished by more recent developments, will be presented. To be readable for nonspecialists of conductive polymer chemistry and to give references to more detailed information for interested readers are further goals. As this book is a PEDOT, or poly(3,4-ethylenedioxythiophene), monograph and PEDOT is one of the most highly conductive polymers, the historical overview focuses on highly conductive polymers directly competing with PEDOT. Hence, this chapter is not extended to all π-conjugated polymers, as it is often found in reviews in this field. A short overview of polythiophenes other than PEDOT—mainly to be classified as semiconductors—will be given in Chapter 3, where the development of PEDOT is described in the context of thiophene chemistry. Last but not least the chapter follows the wonderful advice cited by W. James Feast in his contribution “Synthesis of Conducting Polymers” in the second edition of the Handbook of Conducting Polymers, although not used in the very same sense1:
1
6911X.indb 1
10/1/10 8:58:50 PM
2
PEDOT
”Where shall I begin, your Majesty?” he asked. “Begin at the beginning,” the King said, gravely, “and go on till you come to the end: then stop.”
—Lewis Carroll, Alice’s Adventures in Wonderland So the historical overview will begin—after the inevitable introduction—at the true beginning.
1.2 Introduction What is a polymer? This was a controversial question just before synthetic polymers were prepared for the first time, not to speak of “conducting polymers.” The character of macromolecules was the topic of fundamental discussions in the first half of the 20th century—one of the most fascinating scientific debates in the history of chemistry. After Hermann Staudinger’s concept of covalent bonds between the building blocks of macromolecules was accepted by the scientific community, the tremendous scientific and industrial development of synthetic polymers got a new and even more expansive dimension. The Internet allows a quick and easy answer to the question, What is a polymer? A naturally occurring or synthetic compound consisting of large molecules made up of a linked series of repeated simple monomers. (The Free Dictionary, Farlex, July 2010) A polymer is a substance composed of molecules with a large molecular mass composed of repeating structural units, or monomers, connected by covalent chemical bonds. (Wikipedia, The Free Encyclopedia, September 30, 2007) To give a clear and precise definition of the term polymer is the intention of IUPAC: “A molecule of high relative molecular mass, the structure of which essentially comprises the multiple repetition of units derived, actually or conceptually, from molecules of low relative molecular mass.” On a second glimpse, this definition lacks a little bit from clarity, because high, low, and multiple are not very well defined. We will see in the course of this book that several conductive polymers are not really “poly”mers. Nevertheless, they have been attributed as “the 4th generation of polymeric materials,” clearly demonstrating the enormous importance of this class of chemical compounds.2–4 Besides the problems of clearly defining a polymer, the complete term conducting polymer also can be misunderstood, because it is used with two
6911X.indb 2
10/1/10 8:58:50 PM
The Discovery and Development of Conducting Polymers
3
different meanings in the scientific literature. The blends of electrically conductive additives, like metallic fibers or carbon in its graphite modification, with duromers or thermoplastic polymers sometimes are denoted conductive polymers.1,5 Often they are called extrinsically conductive polymers. Because the polymer itself behaves as an insulator, this can be misleading. The term conductive polymers used in this book deals only with intrinsically conductive polymers (ICPs). Conductivity borderlines between electrically isolating, semiconducting, and conductive materials are fluent and not precisely defined. An overview with typical, widely accepted ranges of conductivity for these three, not very sharply separated material classes is given in Figure 1.1. Conjugated polymers and their conductivity obtainable today are inserted into Figure 1.1. The conductivity range of these polymers has been extended widely in the last few decades. It is obvious that adjusting the desired (particularly this means medium to high) conductivity for a polymeric material is a very difficult challenge; some sort of “molecular engineering” is required. This chapter describes the intriguing story of how this challenge has been met in the last 150 years, starting with the first tentative experiments without
Conductivity (S/cm) 106 Metallic conductors
104 102 100
Semiconductors
10–2 10–4 10–6
Materials
Copper Iron Graphite Bismuth
Indium/Antimony Gallium/Arsenic Germanium
Conjugated polymers
Silicon
10–8 10–10
Glass
10–12 Isolators
10–14 10–16 10–18 10–20
Diamond Sulfur Polyethylene Polystyrene Teflon Quartz
Figure 1.1 Electric conductivity of isolators, semiconductors, and conductive materials.
6911X.indb 3
10/1/10 8:58:52 PM
4
PEDOT
a defined goal in material properties to more systematic chemical syntheses and at the end to real technical breakthroughs in the last quarter of the 20th century. It is not possible to present an exhaustive survey over the vast amount of original literature dealing with all types of conductive polymers. But a first access for readers particularly interested in the historical development is given, and more details of the actual developments should be traceable by the references and the following chapters. The topic of this book, poly(3,4-ethylenedioxythiophene), abbreviated PEDT or, more common, PEDOT, is regarded as one of the highlights, placed at the end of this short history. The huge number of scientific PEDOT publications and patents (more than 1000 per annum), the large quantities of PEDOT-derived products commercially sold every year and the remarkable impact of PEDOT on daily-life goods show the importance of PEDOT as an ICP. Of course, there have been many other ICPs in the past, culminating in the Nobel Prize dedicated to conductive polymer chemistry in 2000. PEDOT:PSS, or poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), was already mentioned as a prominent example in the Advanced Information of the Nobel Committee for Chemistry and the Royal Swedish Academy of Sciences regarding the year 2000 prize.
1.3 An Early Example: Polyaniline One of the most important moments in the discovery and investigation of conducting polymers was the publication of the doped polyacetylene in 1977.6,7 The fundamental discovery of the simplest organic conjugated hydrocarbon (CH)x, combined with its enormous conductivity obtained by “doping” with, for instance, halogens, was honored by the Nobel Prize in 2000.3,4,8–11 The Nobel Laureates—Alan J. Heeger, Hideki Shirakawa, and Alan G. MacDiarmid—initiated a tremendous development in 1977, leading to a huge and steadily growing number of scientific publications, promising technical results, and even various commercial industrial uses. But the conductive polymer story in its widest sense apparently started as early as 1862, when H. Letheby, a chemistry professor at the College of the London Hospital, tried to check the behavior and selected chemical reactions of aniline. He was motivated by two cases of fatal poisoning by nitrobenzene, where aniline had been found as a metabolite in the stomach of the victims.12 Letheby electropolymerized aniline sulfate to a bluish-black solid layer on a platinum electrode and published his results in the Journal of the Chemical Society.12 The chemical nature of the colored, aniline-derived layer essentially remained more or less unknown at that time. The same is true for several
6911X.indb 4
10/1/10 8:58:52 PM
The Discovery and Development of Conducting Polymers
5
colored products, which were also described in Letheby’s paper, obtained by the reaction of aniline with different chemical oxidants. A brilliant bluishgreen oxidation product of aniline (later called emeraldine) had been first described as early as 1834 by F. F. Runge.13 It is not easy to track the complete literature from those early days, but the chemical oxidation of aniline to intensely colored, nearly black pigments was checked also by J. Lightfoot around 1860. He utilized aniline black for dying textiles or printing on fabrics in a U.S. Patent in 186314 (perhaps based on experiments dating from 1859).15 After some technical adjustments, aniline black was then used in the 19th century in a large-scale for textile printing and dyeing.15 An intriguing marginal note shall not be omitted:15 When the young William Henry Perkin in 1856 oxidized toluidine-contaminated aniline with potassium dichromate, he found mauvein, the famous first synthetic dyestuff. The byproduct besides the purple mauvein, is separated as the undesirable residue after alcohol extraction and isolation of the mauvein and discarded, obviously was a (toluidine modified) aniline black, considered useless! Letheby’s publication—the actual birthday of polyaniline and an early landmark in electropolymerization, although its potential remained unidentified at that time—was frequently ignored by more recent papers. More than 100 years later another, possibly fundamental investigation that did not receive much attention was published during the dawn of conducting polymers out of the medical scene. A paper by McGinness, Corry, and Proctor of the University of Texas Cancer Center in Science dealt with the biological pigment melanin, isolated from human tumor material, and its tunable electric conductivity.16,17 A closer look at this macromolecule (Figure 1.2) shows that it is indeed combining structural moieties of (oxidized) aniline and polyacetylene in its expanded conjugated π-system (biologically formed from an indole precursor).
H
O
N
O O
H
O N
N O
H
O n Figure 1.2 Simplified structure of melanin.
6911X.indb 5
10/1/10 8:58:52 PM
6
PEDOT
The response from the scientific community seemed to be very unsatisfactory for the authors, as traceable by the frustrated debate after the Nobel Prize decision in the year 2000.18 In a few words: Both publications with medical background remained “curiosities,” only moderately recognized. Right? To complete the exciting polyaniline story, we have to go back to the 19th century and to the early experiments. New results in polyaniline research after Letheby’s first investigations only sporadically turned up in the scientific literature. In 1891 Goppelsroeder published polyaniline (“aniline black”) again in an illustrated special issue of the German Elektrotechnische Rundschau (Electrotechnical Review).19 Goppelsroeder used the electrolytic oxidation of “aniline chloro hydrate” to aniline black for writing, as demonstrated by a picture from 1891.19 In his own words, Goppelsroeder performed “electrochemical writing or painting.” Several other publications appeared around the turn of the century, sometimes in less common journals (see Liechti and Suida,20 Dobroserdoff,21 Grandmougin,22 BÖttinger and Petzold,23 and Nover24), and with some erroneous chemical interpretations. They do not need to be discussed here in detail. The mistakes are excused by the complicated chemistry and the state of analytical methods at that time. A deeper knowledge of polyaniline evolved between 1907 and 1911, when future Nobel Laureate Richard Willstätter in his typical, strictly methodic way of research, characterized the oligomeric oxidation products of aniline.25–28 Starting only a short time later Arthur G. Green et al. at the Department of Tinctorial Chemistry of the University of Leeds also studied polyaniline and completed, corrected, and reinterpreted Willstätter’s results.29,30 Willstätter replied controversially,31 and Green answered once more32—a rather typical scientific dispute in those days. The names for completely oxidized aniline black (pernigraniline), for the pale reduced form of polyaniline (leucoemeraldine), and for the green, half-oxidized intermediate (emeraldine), and their assignment to chemical structures were finally confirmed. At the end of these discussions some well-founded knowledge about polyaniline had been achieved, and the formulae presented by Green were accepted and established in the scientific literature. It should be noted that in this time the macromolecular character of polyaniline or aniline black was not recognized. For all these products, defined monodisperse oligomers with 8 aniline moieties (n = 2 in Figure 1.3) were formulated. Leucoemeraldine, emeraldine, and pernigraniline as the three most important oxidation states of polyaniline were drawn in popular textbooks (for example, see Beyer33 and Fieser and Fieser34) in the 1960s as depicted in Figure 1.3 (n = 2). A more detailed overview with all five oxidation states to be formulated for oligomeric (n = 2) polyanilines is given by Groenendaal et al.35 The concept of defined octamers was used until the early 1970s, when polyaniline for the first time came into the focus of the chemical industry due to its (semi)conducting properties, for example, in Eastman Kodak patents.36,37 But this is in anticipation of the subsequent seminal and dramatic changes. Industrially, aniline
6911X.indb 6
10/1/10 8:58:52 PM
7
The Discovery and Development of Conducting Polymers
H
H
N
N
N
N
H
H
n
*
n
*
Leucoemeraldine H N
N N
N H
Emeraldine
N
N
n
*
N
N Pernigraniline Figure 1.3 Polyaniline oxidation states (free bases only).
black remained to a large extent an affordable and light-fast pigment (for cotton, silk, and synthetic fibers like polyesters) and is used to date for textile printing. Other applications are dyeing of lacquers, plastics, or paper.38,39 A closer look from the viewpoint of a dyestuff chemist must include subsequent reactions of emeraldine—condensations and oxidations—to phenazine dyes of technical relevance as the structurally modified top grades of aniline black.38 Since the Lightfoot patent of 1863, the industrial use and scientific interest regarding polyaniline were mainly concentrated on dyestuff chemistry and applications. From a scientific point of view this changed dramatically in the late 1960s when polyaniline was recognized as an electrically conducting organic material. All compounds depicted in Figure 1.3 with their more or less defined structures are electric insulators. A real progress on the way to highly conductive organic compounds was achieved, when polyaniline and in particular its salts were studied with more emphasis on its unusual electric properties in France. It was mainly the work of Marcel Jozefowicz and his group in Paris that clearly demonstrated the electronic conductivity of polyaniline salts more than 40 years ago. In several papers polyaniline was presented as a conductive polymer, with the conductivity dependent from the protic doping status. From the paper “Conductivité Electronique et Propriétés Chimiques
6911X.indb 7
10/1/10 8:58:53 PM
8
PEDOT
de Polyanilines Oligomères” the following passage in the abstract may be quoted as particularly illustrative: Conductivity results obtained on an emeraldine complex class (sulfates) indicate a reproducible electronic conductivity. The conductivity varies with the hydration and acid-base parameters of different derivatives; its value is very high for an organic material and ranges from 10 to 10 –4 ohm–1 cm–1 depending on the values of the parameters indicated.40
Although the moisture dependence is significant, a remarkable electronic conductivity was found to remain in completely dried samples of emeraldine sulfate. A lot of papers were published by the Jozefowicz group on conductive polyaniline,40–48 including an early review49 and first suggestions for technical applications.50 But the experimental findings presented to the public “did not give rise to great excitement at that time,” as György Inzelt comments, a little bit amused, in 2008 in his book Conducting Polymers: A New Era in Electrochemistry.51 When R. Buvet, who coauthored several Jozefowicz papers, gave a lecture at the 18th meeting of CITCE at Elmau in 1967, he had to reply to only two questions in the discussion, as can be traced by the published version in Electrochimica Acta.45,51 Although relevant and of technical interest, the questions did not disclose that the audience realized the true scientific breakthrough behind the new results with polyaniline, and the same seems to be true for the scientific community of those days. In the same era of the 1960s, work from Czechoslovak researchers established the concept of iodine doping for polyaniline. A conductivity of up to 1 S/cm for polyaniline–iodine complexes was obtained.52 In 1974, a few years before the spectacular progress for polyacetylene was published, the electronic conductivity of polyaniline was confirmed again in an interesting paper by investigations parallel with another conductive polymer, which had become known in the meanwhile as polypyrrole (see Section 1.4). A remarkable specific conductivity in the range of 5 to 30 S/cm was achieved.53 Surprisingly, the aforementioned basic work was disregarded later with the phrase “a few scattered papers by other groups.”54 MacDiarmid, Epstein, and their colleagues reinvestigated polyaniline in the 1980s, and broadly extended the knowledge about it and remarkably increased the achievable conductivity of PAni (published in many fundamental papers55; easily traceable in several comprehensive articles and reviews).54,56,57 Consequently, the industrial use of conductive polyaniline marked the successful end of a long development, actual suppliers being Ormecon GmbH58,59 in Germany and Panipol Oy60 in Finland. However, the intense color of polyaniline can be a serious drawback in technical applications, when transparency in the visible range of the spectrum and a pale color is required.61,62 So, preferred applications are conductive blends (extrinsically conductive polymers) with thermoplastic resins60,63–65 and active additive in corrosion primers.66
6911X.indb 8
10/1/10 8:58:53 PM
9
The Discovery and Development of Conducting Polymers
1.4 The First Electrically Conductive Poly(Heterocycle): Polypyrrole Not as early as polyaniline itself, described first in the 19th century, but shortly before the discovery of polyaniline conductivity, another interesting group of electrically conducting polymers emerged. The polypyrroles appeared in this field as the first example for conductive poly(heterocycles). From the early 1960s, marked by the work of D. E. Weiss, B. A. Bolto, and coauthors in Australia, until today, the polypyrroles remained one of the most interesting groups of conducting polymers, also with respect to having potential applications. In a series of publications in 1963, Weiss et al. described the thermolysis of 2,3,4,5-tetraiodopyrrole to macromolecular networks.67–69 These polypyrroles exhibit an electric conductivity depending on the degree of doping by iodine, as symbolized by the nonstoichiometric equation in Figure 1.4. The term doping was not used by the authors for the charge transfer complexes between the polypyrrole network and iodine eliminated from the heterocycle. As the result of the polyfunctional starting material, the polypyrrole of the Australian researchers could not exhibit the essentially linear structure of polypyrroles developed later on by other workgroups. A specific conductivity of about 1 S/cm was achieved. In 1968/1969 Dall’Olio et al. in Parma, Italy, revitalized polypyrrole chemistry by oxidizing pyrrole itself electrolytically to pyrrole black (“noir d’oxypyrrol”).70 In contrast to the investigations of Weiss et al., Dall’Olio and his coauthors focused on the electrochemical behavior of pyrrole and its electropolymerization. The paramagnetic behavior of the polymer was studied (the g-factor of the free radical was measured to 2.0026 by electron spin resonance spectroscopy), and a remarkable electric conductivity of 7.54 S/cm at ambient temperature was found. About 10 years later, industrial research activities at the IBM Research Laboratories (San Jose, California) again demonstrated the high conductivity of electrolytically deposited polypyrrole films.71,72 A period of broad investigations followed, traceable, for instance, in the first edition of the Handbook of Conducting Polymers in two polypyrrole chapters.73,74
I n I
....
I
N H
I
(–I2)
....
+
.... ....
I3–
N H
n
Figure 1.4 Conductive, doped polypyrrole network from tetraiodopyrrole.
6911X.indb 9
10/1/10 8:58:53 PM
10
PEDOT
The technical development of the polypyrrole led to several applications. Actually, the electrochemical polymerization is used in tantalum capacitors by Sanyo Electric Co.75 and in aluminum capacitors by Panasonic.76 The Dutch company DSM has commercialized polypyrrole as a prefabricated polymer in the form of a core-shell system with a polypyrrole outer layer and a core of mica or polyurethane particles (ConQuest®). These products are, as to the best knowledge of the authors, no longer available by DSM.
1.5 The Fundamental Breakthrough: Doped Polyacetylene The decisive breakthrough of 1977 by Alan J. Heeger, Alan G. MacDiarmid, and Hideki Shirakawa has its own history. The experimental work— and the intriguing scientific discussions—culminated in the two seminal publications: Electrical Conductivity in Doped Polyacetylene, C. K. Chiang, C. R. Fincher, Jr., Y. W. Park, A. J. Heeger, H. Shirakawa, E. J. Louis, S. C. Gau, and A. G. MacDiarmid, Physical Review Letters, 39(17), 1098–1101 (1977).6 Synthesis of Electrically Conducting Organic Polymers: Halogen Derivatives of Polyacetylene, (CH)x, H. Shirakawa, E. J. Louis, A. G. MacDiarmid, C. K. Chiang, and A. J. Heeger, Journal of the Chemical Society, Chemical Communications, 578–580 (1977).7 The authors classified their work in the tradition of solid state physics and chemistry: The formerly existing nonmetallic conductors compared to doped polyacetylene by the authors in their texts are, as the example for a highly conductive organic material, single crystals of the charge–transfer complex tetrathiafulvalene-tetracyanoquinodimethane (TTF-TCNQ), first described in 1972/197377–79 (see Figure 1.5) and, as the example for a highly conductive polymeric material, the inorganic compound (SN)x.80,81 A closer look at the scientific way Heeger, MacDiarmid, and Shirakawa went, resulting in the discovery of doped polyacetylene, demonstrates the preeminent role of (SN)x in the course of this development. A second, very important aspect is the interdisciplinarity and internationality of the work. Third, the role of serendipity in chemical research should not be underestimated; and doped, highly conductive polyacetylene is another example. In so far, polyacetylene (PAc) is in a line with penicillin, x-rays, Teflon, and high density polyethylene (HDPE)—all of them being invented with the aid of accidents; this line can be prolonged without major difficulties.
6911X.indb 10
10/1/10 8:58:54 PM
The Discovery and Development of Conducting Polymers
11
– + S
S
S
S
NC
CN
NC
CN
Figure 1.5 TTF-TCNQ complex.
Polyacetylene as a chemical individuum was observed as early as 1874,82,83 when the tendency of acetylene to decompose to untractable solids was first described. Later, photo- or radiochemical methods to form acetylene polymers (or oligomers) were described.84 A more practical way to PAc was found and intensely investigated in the 1920 and 1930s. The first examples were the work of Kaufmann and colleagues.85,86 As obtained by the catalytic action of cuprous or cupric oxide (Cu2O or CuO) or by copper in the presence of oxygen, the name cuprene for the solid PAc was created. The exact nature of the cuprene materials remained relatively ambiguous because of their unsolubility and hence very difficult analyses. A synthetic breakthrough in the synthesis of PAc, especially of PAc as a linear high polymer, was achieved by the future Nobel Laureate Giulio Natta in 1958 when he applied Ziegler–Natta catalysts in the polymerization of acetylene for the first time.87 As a potentially conductive polymer, polyacetylene was first recognized in 1961 by researchers of the Tokyo Institute of Technology.88 Like Giulio Natta, Hatano, et al. polymerized acetylene by Ziegler–Natta catalysts; mixtures of triethylaluminium and titanium tetrachloride or titanium tetra-n-butoxide were used. Investigation of the structure and measurements of the electric conductivity were performed. The PAc was suggested to be essentially transconjugated; the electric conductivity of the greenish-black powder did not exceed the semiconducting range and was in the order of 10 –5 S/cm in the best cases—the more crystalline, the higher. The temperature dependence of the resistance ρ followed the usual equation of an intrinsic semiconductor:
ρ = ρ0 exp (ΔE/2kT)
This was the state of the art when the group of Hideki Shirakawa in Tokyo started their work with polyacetylene. The now following first step in the technical revolution regarding the electric conductivity of plastics was decisive, but not very spectacular. The full consequences were not recognized
6911X.indb 11
10/1/10 8:58:54 PM
12
PEDOT
immediately, and it was done serendipitously. What had happened in the laboratory of Shirakawa? The following description mainly follows the personal reports of the Nobel Prize winners.89–91 Acetylene had been polymerized in the Shirakawa group with Ziegler– Natta catalyst [Ti(OnC4H9)4–Al(C2H5)3] as usual, and of course intractable black-brown powdery solids had been precipitated. Shirakawa was interested in the influence of different catalyst concentrations on the properties of polyacetylene. When a new, visiting foreign student attended the laboratory, Shirakawa advised him to check millimolar concentrations of the catalyst. It is not totally clear afterward what kind of mistake then took place: a language misunderstanding, or a missing letter m in the written instructions of Shirakawa, or a misreading of these instructions. Anyway, as the result of the misunderstanding, the student applied molar instead of millimolar concentrations of the two components of the Ziegler–Natta catalyst.89-92 The appearance of the polyacetylene product dramatically changed. Instead of being unsightly dark brown, the PAc obtained consisted of silvery lumps. In spite of this metallic look, the conductivity of the new modification of PAc was not significantly enhanced, but the material was far better processable into films, which were accessible for spectroscopic investigations and so forth. Several interesting papers published by the Shirakawa group described the synthesis and properties of this special new polyacetylene with metallic luster that could elucidate a lot of structural features.93–96 But until the mid-1970s, nothing else happened. Despite the progress Shirakawa had achieved, polyacetylene continued as a laboratory curiosity. The electric conductivity was not thrilling, especially compared to the conductive organic polymers already known, polyaniline and polypyrrole, which showed up to about five orders of magnitude higher conductivities. Nevertheless, and perhaps just therefore, this “1000-fold” experiment was one of the magic moments in chemical history, comparable to several other remarkable breakthroughs in science. But before getting aware of this fact, another scientific group at the opposite end of the world had to play a decisive role, and a second magic moment had to come about. At the University of Pennsylvania (Penn) in Philadelphia, the physicist professor Alan J. Heeger had a discussion with Alan G. MacDiarmid, a chemistry professor, about the inorganic conductor poly(sulfurnitride) (SN)x, a polymer with a typical golden-metallic luster. Heeger was intrigued by several quasimetallic properties of this novel compound. A fruitful collaboration about (SN)x was established in the mid-1970s, crossing the chemistry–physics border after some initial “language” problems (confusing Sn, metallic, element tin; and SN, sulfur nitride).89–92,97 When MacDiarmid later—at that time in 1975 he was a visiting professor at Kyoto University in Japan—visited the Tokyo Institute of Technology, he reported about the Heeger and MacDiarmid work on (SN)x. Shirakawa and MacDiarmid met “over a cup of green tea” and showed each other their lustrous samples: golden (SN)x and the silvery polyacetylene in the form of
6911X.indb 12
10/1/10 8:58:54 PM
The Discovery and Development of Conducting Polymers
13
shiny films. Both researchers were fascinated by the unusual appearance of (CH)x, since a formally simple hydrocarbon with a metallic appearance had never been seen before. So they decided to get more knowledge about this compound, which was supposed to be perhaps even more remarkable than already known, and to further extend the studies of Shirakawa. Shirakawa was invited to Penn as a visiting scientist. When the (CH)x synthesis was repeated in Philadelphia, a rather low conductivity was observed again, in spite of the metallic appearance. The product was not very pure, and so the researchers decided to strictly remove all impurities from the polyacetylene, with the opposite effect than expected: the conductivity decreased. Obviously, small halogen impurities measured by elemental analysis had a positive influence. So, an experiment analog to (SN)x chemistry, where the addition of bromine is known to increase the conductivity by one order of magnitude, was made. The effect achieved by addition of bromine was not as moderate as with sulfur nitride but really dramatic. An extremely high conductivity was immediately observed, and the electrometer was destroyed. On this November 23, 1976, a 10-million-times-enhanced conductivity for the socalled doped polyacetylene (Figure 1.6) was obtained—the birthday of the first highly conducting hydrocarbon! The enormous significance of this experimental breakthrough was quickly realized, as traceable by a letter to Kenneth Wynne, the program officer in the U.S. Office of Naval Research responsible for funding the visit of Shirakawa at Penn. With this letter, written by MacDiarmid, a copy of the seminal paper submitted to Chemical Communications was forwarded to Wynne.92 A short passage from this letter, demonstrating the enormous relevance attributed to this discovery by their originators, is cited by Hall: As you will doubtless observe, we believe this is an extremely important and exciting new area and, although we would not say it in public, we have other information which leads us to believe that some of the species are indeed metallic!92
Twenty-three years later, the Nobel Prize in Chemistry was awarded to Heeger, MacDiarmid, and Shirakawa.98 Their fundamental research had kicked off a tremendous development, from a scientific point of view as well as in the chemical industry. But all expectations regarding practical,
H
H
H
H
H
....
+
....
+ H
H
H
H
H
Figure 1.6 Polyacetylene in its highly conductive dicationic form (one mesomeric structure).
6911X.indb 13
10/1/10 8:58:55 PM
14
PEDOT
industrial applications for doped polyacetylene were in vain. The difficult processability and relative instability—measured at long-term standards in technical use—played the pivotal role in this failure. At the end of the 1980s, vast scientific progress in the field of conducting polymers had been achieved, several applications were suggested, and the world record in electric conductivity for organic materials was set in 1987 with stretched, iodine-doped polyacetylene by Naarmann and Theophilou to about 100,000 S/cm.99 Nevertheless, from a technical point of view, the statement of Billingham and Calvert written in 1988 in their article “Electrically Conducting Polymers: A Polymer Science Viewpoint,”100 was still valid: “If one asks what are the applications of conducting polymers, the short answer is ‘none.’” In the same review,100 the (apparently) experimentally well supported opinion of Münstedt is cited: That the whole range of common conducting polymers is unstable in accelerated tests of conductivity decay. He suggests that the carbenium ion structures, which are required to permit conduction in conjugated polymers, are incompatible with the presence of oxygen and water and that the only practical route towards conducting polymers which have environmental stability comparable to graphite, will be to seek structures whose band gap is intrinsically small enough to allow thermal excitation without the need for doping.101
At the end of the 1980s, the technical situation began to change. The aforementioned investigations101 clearly demonstrated insufficient conductivity half-life values of polyacetylene, polypyrrole, polythiophene, polyaniline, and TTF-TCNQ under accelerated testing in humid air or at inert conditions. But the ink used for writing the previous statements had not dried yet, when the invention of PEDOT (the polymer based on 3,4-ethylendioxythiophene [EDOT], see Figure 1.7) just demonstrated the opposite: doped PEDOT emerged as a highly conducting polymer stable in air up to very high temperatures, to humidity including moist air, also at elevated temperatures, and, after only a few years of further development, even processable in water. Several fundamental patent applications regarding the utilization of special conducting polymers were filed. In the following 20 years, polyaniline,
O
O
S Figure 1.7 3,4-Ethylenedioxythiophene (EDOT).
6911X.indb 14
10/1/10 8:58:56 PM
The Discovery and Development of Conducting Polymers
15
polypyrrole, and special polythiophenes found their way into the market. Some details regarding polythiophenes will be presented later. The most successful and widely used polythiophene, PEDOT, produced and technically applied in steadily growing multiton amounts per year as the only commercialized example for a highly conductive polythiophene, is the focus of this book.
References
6911X.indb 15
1. W. J. Feast. 1986. Synthesis of conducting polymers. In: Handbook of Conducting Polymers, ed. T. A. Skotheim, 1–43. New York: Marcel Dekker. 2. B. Rånby. 1993. Conjugated Polymers and Related Materials: The Interconnection of Chemical and Electronic Structures, ed. W. R. Salaneck, I. Lundström, and B. Rånby, Chapter 3. Oxford: Oxford University Press. 3. A. J. Heeger. 2001. Semiconducting and metallic polymers: The fourth generation of polymeric materials (Nobel Lecture). Angew Chem 113(14):2660–2682; Angew Chem Int Ed 40(14):2591–2611. 4. A. J. Heeger. 2001 (Volume date: 2002). Semiconducting and metallic polymers: The fourth generation of polymeric materials. Synth Met 125(1):23–42. 5. R. Hanselmann. 2008. Elektrisch leitfähige Polymere. In Thieme ROEMPP Online, www.roempp.com. 6. C. K. Chiang, C. R. Fincher, Jr., Y. W. Park, A. J. Heeger, H. Shirakawa, E. J. Louis, S. C. Gau, and A. G. MacDiarmid. 1977. Electrical conductivity in doped polyacetylene. Phys Rev Lett 39(17):1098–1101. 7. H. Shirakawa, E. J. Louis, A. G. MacDiarmid, C. K. Chiang, and A. J. Heeger. 1977. Synthesis of electrically conducting organic polymers: Halogen derivatives of polyacetylene, (CH)x. J Chem Soc, Chem Commun 1977:578–580. 8. H. Shirakawa. 2001. The discovery of polyacetylene film: The dawning of an era of conducting polymers (Nobel Lecture). Angew Chem 113(14):2642–2648; Angew Chem Int Ed 40(14):2575–2580. 9. H. Shirakawa. 2001 (Volume date: 2002). The discovery of polyacetylene film: The dawning of an era of conducting polymers. Synth Met 125(1):3–10. 10. A. G. MacDiarmid. 2001. Synthetic Metals: A novel role for organic polymers (Nobel Lecture). Angew Chem 113(14):2649–2659; Angew Chem Int Ed 40(14):2581–2590. 11. A. G. MacDiarmid. 2001 (Volume date: 2002). Synthetic Metals: A novel role for organic polymers. Synth Met 125(1):11–22. 12. H. Letheby. 1862. On the production of a blue substance by the electrolysis of sulphate of aniline. J Chem Soc XV:161–163. 13. F. F. Runge. 1834. Poggendorfs Ann. Phys. u. Chemie 31:513–524. 14. J. Lightfoot. U.S. Patent 38,589, Prior: February 21, 1863. 15. P. J. T. Morris and A. S. Travis. 1992. A History of the International Dyestuff Industry. http://colorantshistory.org/HistoryInternationalDyeIndustryRev1/ HistoryInternationalDyestuffIndustryFirefox/dyestuffs.html. 16. J. McGinness, P. Corry, and P. Proctor. 1974. Amorphous semiconductor switching in melanins. Science 183:853–855.
10/1/10 8:58:56 PM
16
PEDOT
17. News and Views: Semiconductors in the human body? 1974. Nature 248:475. 18. P. H. Proctor. 2000. http://www.organicsemiconductors.com/whine.htm. 19. F. Goppelsroeder. 1891. Studien über die Anwendung der Elektrolyse zur Darstellung, zur Veränderung und zur Zerstörung der Farbstoffe ohne oder in Gegenwart von vegetabilischen oder animalischen Fasern. Elektrotechnische Rundschau 19:1047–1051. 20. L. Liechti and W. Suida. (n.d.). On aniline black. Dinglers Polytechnisches J. 254:265, From: J Am Chem Soc 7(2):63–64 (1885). 21. D. K. Dobroserdoff. 1904. Conditions of the interaction between aniline vapour and aluminium chlorate solution. Zh Russ Fiziko-Khim Obshch 36:483–485, From: J Chem Soc, Abstr 86, I, 661 (1904). 22. E. Grandmougin. 1906. Over aniline blacks. Ztschr f Farbenindustrie 5:286–287. 23. E. Böttiger and G. Petzold. 1907. Contribution to the knowledge of technical oxidation blacks. Färber-Zeitung 18:8–10. 24. W. Nover. 1907. Emeraldine. Ber Dtsch Chem Ges 40:288–297. 25. R. Willstätter and C. W. Moore. 1907. Über Anilinschwarz I. (Aniline black I.). Ber Dtsch Chem Ges 40:2665–2689. 26. R. Willstätter and S. Dorogi. 1909. Über Anilinschwarz II. (Aniline black II.). Ber Dtsch Chem Ges 42:2147–2168. 27. R. Willstätter and S. Dorogi. 1910. Über Anilinschwarz III. (Aniline black III.). Ber Dtsch Chem Ges 42(3):4118–4135. 28. R. Willstätter and C. Cramer. 1911. Über Anilinschwarz IV. (Aniline black IV.). Ber Dtsch Chem Ges 43(3):2976–2988. 29. A. G. Green and A. E. Woodhead. 1910. Aniline black and allied compounds I. J Chem Soc, Trans 97:2388–2403. 30. A. G. Green and S. Wolff. 1911. Aniline black and its intermediate compounds. Ber Dtsch Chem Ges 44(3):2570–2579. 31. R. Willstätter and C. Cramer. 1911. Über Anilinschwarz V. (Aniline black V.). Ber Dtsch Chem Ges 44:2162–2171. 32. A. G. Green and A. E. Woodhead. 1912. Aniline black and allied compounds II. J Chem Soc, Trans 101:1117–1123. 33. H. Beyer. 1966. Lehrbuch der Organischen Chemie, 11th/12th ed. Leipzig: S. Hirzel Verlag 34. L. Fieser and M. Fieser. 1968. Organische Chemie, 2nd ed. Weinheim: Verlag Chemie. Translated from: 1961, Advanced Organic Chemistry. New York: Reinhold Publishing Corp.; and 1963, Topics in Organic Chemistry. New York: Reinhold Publishing Corp. 35. L. Groenendaal, E. W. Meijer, and J. A. J. M. Vekemans. 1998. Nitrogen-containing oligomers. In: Electronic Materials: The Oligomer Approach, ed. K. Müllen and G. Wegner, 235–272. Weinheim: Wiley-VCH. 36. D. J. Trevoy. DE 2 262 743 (Eastman Kodak Co.), Prior: December 27, 1971. 37. D. J. Trevoy. U.S. Patent 3,963,498 (Eastman Kodak Co.), Prior: December 27, 1971; March 28, 1974; June 25, 1974. 38. H. Berneth. 2008. Azine dyes. In: Ullmann’s Encyclopedia of Industrial Chemistry. Weinheim: Wiley-VCH Verlag GmbH & Co. KgaA. DOI: 10.1002/14356007. a03_213.pub3. http://mrw.interscience.wiley.com/emrw/9783527306732/ueic/ article/a03_213/current/pdf. 39. N. Welsch. 2006. Anilinschwarz. In Thieme ROEMPP Online, www.roempp.com.
6911X.indb 16
10/1/10 8:58:56 PM
The Discovery and Development of Conducting Polymers
17
40. M. Jozefowicz, L.-T. Yu, G. Belorgey, and R. Buvet. 1967. Conductivité electronique et propriétés chimiques de polyanilines oligomères (Conductivity and chemical properties of oligomeric polyanilines). J Polym Sci C (Polymer Symposia) 16 (Pt. 5):2943–2954. 41. L.-T. Yu, J. Petit, M. Jozefowicz, G. Belorgey, and R. Buvet. 1965. Conductivité en courant continu des sulfates acides d’éméraldine (Conductivity of emeraldine acid sulfate for continuous current). Compt Rend 260(19) (Groupe 7):5026–5029. 42. M. F. Combarel, G. Belorgey, M. Jozefowicz, L.-T. Yu, and R. Buvet. 1966. Conductivité en courant continu des polyanilines oligomères: Influence de l’état acide-base sur la conductivité électronique (Conductivity in a continuous stream of oligomeric polyanilines: Influence of the acid-base state on the electronic conductivity). Compt Rend C (Sciences Chimiques) 262(6):459–462. 43. L.-T. Yu and M. Jozefowicz. 1966. Conductivity and chemical composition of macromolecular semiconductors. Revue Generale de l’Electricite 75(9):1014–1018. 44. L.-T. Yu, M. S. Borredon, M. Jozefowicz, G. Belorgey, and R. Buvet. 1967. Experimental study of the direct current conductivity of macromolecular compounds. J Polym Sci C (Polymer Symposia) 16(5):2931–2942. 45. R. De Surville, M. Jozefowicz, L.-T. Yu, J. Perichon, and R. Buvet. 1968. Electro chemical cells using protolytic organic semiconductors. Electrochim Acta 13(6): 1451–1458. 46. F. Cristofini, R. De Surville, M. Jozefowicz, L.-T. Yu, and R. Buvet. 1969. Electrochemical properties of Poly(aniline sulfates). Compt Rend C (Sciences Chimiques) 268(15):1346–1349. 47. D. Labarre and M. Jozefowicz. 1969. Polymères conducteurs organiques filmogènes à base de polyanilines (Polyaniline-based filmogenic organic-conductor polymers). Compt Rend C (Sciences Chimiques) 269(17):946–966. 48. M. Doriomedoff, F. Hautiere-Cristofini, R. De Surville, M. Jozefowicz, L.-T. Yu, and R. Buvet. 1971. Direct current conductivity of polyaniline sulfates. J Chim Phys Physico-Chimie Biol 68(7–8):1055–1069. 49. M. Jozefowicz, L.-T. Yu, J. Perichon, and R. Buvet. 1967. Recently discovered properties of semiconducting properties. J Polym Sci C (Polymer Symposia) 22 (Pt. 2):1187–1195. 50. M. Jozefowicz, L.-T. Yu, J. Perichon, and R. Buvet. FR 1.519.729 (CNRS), Prior: February 20, 1967. 51. G. Inzelt. 2008. Historical background (Or: There is nothing new under the sun). In: Conducting Polymers: A New Era in Electrochemistry, Chapter 8. Berlin: Springer Verlag. 52. J. Honzl and M. Tlustáková. 1968. Polyaniline compounds. II. The linear oligoaniline derivatives tri-, tetra-, and hexaanilinobenzene and their conductive complexes. J Polymer Sci C 22(1):451–462. 53. I. Mamadou, L.-T. Yu, and R. Buvet. 1974. Conductivity of polyaniline and polypyrrole composites under point-plane geometry electronic injection. Compt Rend C (Sciences Chimiques) 279(23):931–934. 54. A. G. MacDiarmid. 1997. Polyaniline and polypyrrole: Where are we headed? Synth Met 84(1–3):27–34. 55. A. G. MacDiarmid, J. C. Chiang, A. F. Richter, N. L. D. Somasiri, and A. J. Epstein. 1987. Polyaniline: Synthesis and characterisation of the emeraldine oxidation state by elemental analysis. In: Conducting Polymers, Proc. Workshop, ed. L. Alcácer, 105–120. Dordrecht: D. Reidel.
6911X.indb 17
10/1/10 8:58:56 PM
18
PEDOT
56. D. C. Trivedi. 1997. Polyanilines. In: Handbook of Organic Conductive Molecules and Polymers, ed. H. S. Nalwa, Vol. 2, 505–572. Chichester: John Wiley & Sons. 57. E. T. Kang, K. G. Neoh, and K. L. Tan. 1998. Polyaniline: A polymer with many interesting intrinsic redox states. Prog Polym Sci 23(2):277–324. 58. Conductive Coatings—ORMECON® Lacquers and Coatings. www.enthone. com/ormecon/ormecon-conductive.aspx (accessed August 2010). 59. B. Wessling. 2007. Conducting polymers as organic nanometals. In: Handbook of Conducting Polymers, 3rd ed., ed. T. A. Skotheim and J. R. Reynolds, Vol. 2, 1-3–1-75. Boca Raton, FL: CRC Press. 60. Panipol—Inherently Conductive Polymers. www.panipol.com (accessed August 2010). 61. Y. Cao, G. M. Treacy, P. Smith, and A. J. Heeger. 1993. Optical-quality transparent conductive polyaniline films. Synth Met 57(1):3526–3531. 62. Y. Chang, K. Lee, R. Kiebooms, A. Aleshin, and A. J. Heeger. 1999. Reflectance of conducting poly(3,4-ethylenedioxythiophene). Synth Met 105(3):203–206. 63. O. Ikkala, J. Laakso, K. Väkiparta, E. Virtanen, H. Ruohonen, H. Järvinen, T. Taka, P. Passiniemi, J.-E. Österhelm, Y. Cao, A. Andreatta, P. Smith, and A. J. Heeger. 1995. Counter-ion induced processibility of polyaniline: Conducting melt processible polymer blends. Synth Met 69(1–3):97–100. 64. E. Virtanen, J. Laakso, H. Ruohonen, K. Väkiparta, H. Järvinen, M. Jussila, P. Passiniemi, and J.-E. Österholm. 1997. Electrically conductive compositions based on processible polyanilines - PANIPOLTM. Synth Met 84(1–3):113–114. 65. M. Saroop, A. K. Ghosh, and G. N. Mathur. 2003. Polyaniline based conductive polymers: An overview. Int J Plastics Techn 7:41–61. 66. T. Schauer, A. Joos, L. Dulog, and C. D. Eisenbach. 1999. Wirkungsprinzip von Polyanilin. Drei Phasen bestimmen den Schutzmechanismus. Farbe Lack 105(6):52–63. 67. R. McNeill, R. Siudak, J. H. Wardlaw, and D. E. Weiss. 1963. Electronic conduction in polymers I. The chemical structure of polypyrrole. Aust J Chem 16(6):1056–1075. 68. B. A. Bolto and D. E. Weiss. 1963. Electronic conduction in polymers II. The electrochemical reduction of polypyrrole at controlled potential. Aust J Chem 16(6):1076–1089. 69. B. A. Bolto, R. McNeill, and D. E. Weiss. 1963. Electronic conduction in polymers III. The electronic properties of polypyrrole. Aust J Chem 16(6):1090–1103. 70. A. Dall’Olio, G. Dascola, V. Varacca, and V. Bocchi. 1968. Résonance paramagnétique électronique et conductivité d’un noir d’oxypyrrol électrolytique. Compt Rend C (Sciences Chimiques) 267:433–435. 71. A. F. Diaz, K. K. Kanazawa, and G. P. Gardini. 1979. J Chem Soc, Chem Commun 1979:635–636. 72. K. K. Kanazawa, A. F. Diaz, R. H. Geiss, W. D. Gill, J. F. Kwak, J. A. Logan, J. F. Rabolt, and G. B. Street. 1979. “Organic metals”: Polypyrrole, a stable synthetic “metallic” polymer. J Chem Soc, Chem Commun 1979:854–855. 73. P. Burgmayer and R. W. Murray. 1986. Ionic conductivity of polypyrrole. In: Handbook of Conducting Polymers, ed. T. A. Skotheim, 507–523. New York: Marcel Dekker. 74. P. Pfluger, G. Weiser, J. C. Scott, and G. B. Street. 1986. Electronic structure and transport in the organic “amorphous semiconductor” polypyrrole. In: Handbook of Conducting Polymers, ed. T. A. Skotheim, 1369–1381. New York: Marcel Dekker.
6911X.indb 18
10/1/10 8:58:56 PM
The Discovery and Development of Conducting Polymers
19
75. POSCAP—A Capacitor Lecture (Basic). www.edc.sanyo.com/english/pdf/ poscap/E_poscap_basic.pdf. (accessed August 2010). 76. SP-Cap Aluminum Electrolytic Capacitor/Panasonic. www.ttiinc.com/object/ fp_panasonic_spcap.html (accessed August 2010). 77. J. H. Perlstein, J. P. Ferraris, V. V. Walatka, D. O. Cowan and G. A. Candela. 1973. Electron transport and magnetic properties of new highly conducting TCNQ complexes. AIP Conference Proceedings (1973), Volume Date 1972, No. 10 (Pt. 2):1494–1498. 78. P. W. Anderson, P. A. Lee, and M. Saitoh. 1973. Giant conductivity in TTFTCNQ [tetrathiofulvalinium tetracyanoquinodimethanide]. Solid State Commun 13(5):595–598. 79. A. A. Bright, A. F. Garito, and A. J. Heeger. 1973. Optical properties of TTFTCNQ [tetrathiofulvalenium 7,7,8,8-tetracyanoquinodimethanide] in the visible and infrared. Solid State Commun 13(7):943–948. 80. M. Goehring, D. Voigt. 1953. Sulfur nitrides (SN)2 and (SN)x. Naturwissenschaften 40:482. 81. M. M. Labes. 1966. Conductivity in polymeric solids. Pure Appl Chem 12 (1–4):275–285. 82. M. P. von Wilde. 1874. Über die Darstellung des Acetylens. Ber Dtsch Chem Ges 7:352–357. 83. P. Thenard and A. Thenard. 1874. Acetylene liquifie et solidife sous l’influence de l’effluve electrique. Compt Rend 78:219. 84. D. Berthelot and H. Gaudechon. 1912. Effective radiations in the photochemical synthesis of quaternary compounds, in the polymerization of various gases, and in the photolysis of acetone. Compt Rend 155:207–210. 85. H. P. Kaufmann and M. Schneider. 1922. Acetylene condensations. I. Attempts to determine the constitution of cuprene. Ber Dtsch Chem Ges 55B:267–282. 86. H. P. Kaufmann and W. Mohnhaupt. 1923. Acetylene condensations. II. Theory of the formation of cuprene. Ber Dtsch Chem Ges 56B:2533–2536. 87. G. Natta, G. Mazzanti, and B. Corradini. 1958. Atti Acad Nazl Lincei Rend Classe Sci Fis Mat Nat 25(8):3. 88. M. Hatano, S. Kambara, and S. Okamoto. 1961. Paramagnetic and electric properties of polyacetylene. J Polym Sci 51(156):S26–S29. 89. Nobelprize.org—The official Web site of the Nobel Prize. http://nobelprize. org/nobel_prizes/chemistry/laureates/2000/macdiarmid-autobio.html (accessed August 2010). 90. Nobelprize.org—The official Web site of the Nobel Prize. http://nobelprize. org/nobel_prizes/chemistry/laureates/2000/shirakawa-autobio.html (accessed August 2010). 91. Nobelprize.org—The official Web site of the Nobel Prize. http://nobelprize.org/ nobel_prizes/chemistry/laureates/2000/heeger-autobio.html (accessed August 2010). 92. N. Hall. 2003. Focus article: Twenty-five years of conducting polymers. Chem Commun 2003:1–4. 93. H. Shirakawa and S. Ikeda. 1971. Infrared spectra of poly(acetylene). Polym J 2(2):231–244. 94. H. Shirakawa, T. Ito, and S. Ikeda. 1973. Raman scattering and electronic spectra of poly(acetylene). Polym J 4(4):460–462.
6911X.indb 19
10/1/10 8:58:56 PM
20
PEDOT
95. T. Ito, H. Shirakawa, and S. Ikeda. 1974. Simultaneous polymerization and formation of polyacetylene film on the surface of concentrated soluble ziegler-type catalyst solution. J Polym Sci (P Chem Ed) 12(1):11–20. 96. T. Ito, H. Shirakawa, and S. Ikeda. 1975. Thermal cis-trans isomerization and decomposition of polyacetylene. J Polym Sci (P Chem Ed) 13(8):1943–1950. 97. C. K. Chiang, M. J. Cohen, A. F. Garito, A. J. Heeger, C. M. Mikulski, and A. G. MacDiarmid. 1976. Electrical conductivity of (SN)x. Solid State Commun 18(11–12):1451–1455. 98. Nobelprize.org—The official Web site of the Nobel Prize. http://nobelprize.org/ nobel_prizes/chemistry/laureates/2000/public.html (accessed August 2010). 99. H. Naarmann and N. Theophilou. 1987. New process for the production of metal-like, stable polyacetylene. Synth Met 22(1):1–8. 100. N. C. Billingham and P. D. Calvert. 1989. Electrically conducting polymers: A polymer science viewpoint. Adv Polym Sci 90:1–104. 101. H. Münstedt. 1988. Aging of electrically conducting organic materials. Polymer 29(2):296–302.
6911X.indb 20
10/1/10 8:58:56 PM
2 Conductive Polymers versus Metals and Insulators This chapter is an introduction into the origin of the conductivity in intrinsically conducting polymers. The properties of metals, semiconductors, and insulators are first described in general. The description of undoped and doped conjugated polymers then leads to a description of the charge transport in those materials. However, there is no model fully describing the conductive properties of all intrinsically conductive polymers. Nonetheless, significant progress in understanding conductive polymers has been made over the last 20 years and many features can now be described. For more detailed descriptions the reader is referred to the literature.1,2
2.1 Metals, Semiconductors, and Insulators The properties of metals, semiconductors, and insulators can be most easily described using simple, nonpolymeric solids as an example. The changes in orbitals and energy levels starting from individual atoms to small molecules and then to three-dimensional solids are well understood.3 In small molecules with a well-defined number of atoms the atomic orbitals merge to form molecular orbitals with discrete and well-defined energy levels. The low lying orbitals are filled with electrons and typically have a bonding character, whereas high lying orbitals are often unfilled and have an antibonding character. If a very large number of atoms is arranged in a three-dimensional lattice, the energy levels of neighboring states rearrange into bonding and antibonding states, and a continuous band is formed. In certain solids such as metals, the orbitals in the bands are continuous and electrons close to the top of the filled orbitals can be excited into unoccupied levels requiring hardly any energy. To describe the energy levels in such metals, it is useful to consider the distribution of electrons first at temperature T = 0 without thermal energy. The highest occupied energy level in these metals at T = 0 is called the Fermi level. The extension of electronic states is defined by the boundaries of the crystalline domains and the charge carrier transport is only limited by scattering processes. At temperatures above T = 0 there is no sharp distinction between occupied and unoccupied levels, since the electrons are excited by the thermal energy.3 The higher the temperature the more electrons are excited into 21
6911X.indb 21
10/1/10 8:58:56 PM
22
PEDOT
higher unoccupied energy levels. The Fermi–Dirac distribution describes this population of orbitals at different temperatures. It is a version of the Boltzmann distribution relating the energy of individual electrons to the thermal energy kT. The mean value in the Fermi–Dirac distribution is the Fermi energy Ef, which is assigned to the energy level that is half occupied. The Fermi energy itself is therefore temperature dependent. The electrical conductivity of metallic solids decreases with increasing temperature although more electrons are excited. This is because of the thermal motion of the atoms that lead to collisions between electrons and atoms. Because of these collisions the electrons become less efficient in transporting charge. In semiconductors and insulators the relevant bands for charge transport are separated by an energy gap.3 At T = 0 the electronic states of the energetically lower lying band, called the valence band, are completely filled, whereas the states forming the conduction band are empty. As the temperature is increased, certain electrons gain enough energy that they can occupy empty orbitals in the conduction band. These electrons are now mobile and the solid becomes electrically conductive. It can now be described as a semiconductor. If the energy gap is very large, only a few electrons will reach the conductive band and the overall conductivity is close to zero, and the material is described as an insulator. The distinction between insulator and semiconductor is therefore a gradual distinction, which depends on the band gap. The distinction between a semiconductor and an insulator is, therefore, less well defined as that between a metal (incomplete bands at T = 0) and a semiconductor (complete bands at T = 0). In semiconductors the conductivity increases with increasing temperature since more electrons are able to reach the valence band. The numbers of charge carriers in semiconducting solids, such as silicon, can also be increased by the addition of small amounts of foreign atoms into the otherwise pure material. If these so-called dopants carry fewer electrons than the host, an additional narrow band is formed in the energy gap, which accepts electrons. The electrons from the valence band can move easily into this acceptor band and the semiconductor is called p-doped or a p-type semiconductor. The introduction of such a dopant does not inject a charge into the system. It does, however, lead to the formation of additional states between the valence band and the conduction band, which were not present before. Alternatively, a dopant with more electrons than the host leads to a donor band in the energy gap and an n-type semiconductor.
2.2 Conjugated Polymers In saturated polymers, such as polyethylene, all valence electrons are used in covalent σ-bonds. Hence, the gap between the valence band and the conduction band is very large and the material shows typical insulating properties.
6911X.indb 22
10/1/10 8:58:57 PM
Conductive Polymers versus Metals and Insulators
23
In conjugated polymers a π system is formed along the polymer backbone.4 The carbon atoms typically involved in the formation of the polymer backbone form three σ-bonds with neighboring atoms and the remaining p orbitals—typically described as pz orbitals—engage in the π system. In some conjugated polymers, such as polyaniline, nitrogen atoms are also involved in the conjugation path. Polyacetylene is the simplest conjugated polymer. Each carbon is σ-bonded to two neighboring carbon atoms and one hydrogen atom. One π electron per carbon remains. If the carbon–carbon bonds were equally long, the remaining π electrons would be found in one half filled continuous band.4 Based on the considerations of the previous section, such a material would show metallic behavior such as conductivity at T = 0 and decreasing conductivity with increasing temperature. However, Peierls’ theorem states that a onedimensional, equally spaced chain with one electron per ion is unstable.2,4 The Peierls’ distortion in chain compounds is the analog to the Jahn–Teller effect in molecules. In both cases the symmetry of the system is reduced and the orbital levels are rearranged in a way that filled orbitals are lowered. In the case of polyacetylene, a lattice distortion leads to a repeat unit with two carbon atoms closer together and the next two carbon atoms further apart. Hence, the repeat unit can be described as −CH=CH− instead of −CH−. The electronic result is an energy gap between a completely filled π band and an empty π* band. The energy savings due to the new band gap outweighs the energy cost of rearranging the carbon atoms. This bond-alternating structure is typical for all conjugated polymers. Since there are no partially filled bands, pristine conjugated polymers are typically semiconductors.4 Similar to inorganic semiconductors, such as silicon, conjugated polymers show very low conductivities in their pristine state. However, since they are organic molecules, there is a variety of ways to introduce charges into this material and to radically change their electronic properties. The first way of charge introduction is so-called chemical doping. If, for example, polyacetylene is treated with iodine, the latter oxidizes the polymer chain.5 The polymer becomes positively charged and iodide is formed as counterion. The term doping is in some way misleading since it was originally used in solid-state physics for the introduction of a foreign neutral atom into a host lattice, changing the electronic structure in that lattice. In the context of conductive polymers, doping refers to a chemical reaction—oxidation or reduction. However, in both cases new electronic states are created and a previously semiconducting material becomes conductive. Both electron donating (n-type) and electron accepting (p-type) dopants, that is, reducing agents and oxidants, have been used to introduce charges into conjugated polymers and render them conductive. The charges can also be introduced electrochemically. The necessary counterion is drawn from the surrounding solution. In the case of photo-doping the conjugated polymer is locally oxidized and reduced by forming an electron–hole pair due to light absorption followed by charge diffusion.
6911X.indb 23
10/1/10 8:58:57 PM
24
PEDOT
Polyaniline is a specific case in which acid–base chemistry introduces a structural change. The resulting spin-unpairing without a change in the number of electrons involved then leads to new mobile charged states.6 Charges that are introduced in polymers and oligomers are stored in novel states, called polarons, bipolarons, or solitons.1 These states include a charge and a lattice distortion. A polaron is formed when an electron is added to or removed from the conjugated chain. This change results in a chain deformation, which is in the case of polyacetylene approximately 20 sites long, and a change in the energy level structure.1 One electronic level is moved from the valence band into the gap with its two electrons and an addition level is moved from the conducting band into the gap. In the case of an electron polaron, the added electron is stored in the newly created level drawn from the conduction band. In the case of a hole polaron, an electron is removed from the newly created level moved up from the valence band. In both cases a half-filled level is created with spin ½. The energy difference between the band edge and the newly created states depends on the band gap and chain length. Bipolarons are formed by the combination of two polarons with the same charge. The bipolaron also has two levels in the energy gap. In the case of a negative bipolaron both levels are fully occupied and for a positive bipolaron both levels are empty. In either case the spin is zero. Because of their charge, bipolarons are assumed to be in close proximity to their counterions. Solitons are a third type of excited species, which only occur in degenerate polymers.7 Degenerate polymers are those where an interchange of single and double bonds along the polymer chain results in the same structure. Solitons are states in the center of the band gap associated with an interchange of single and double bonds. The most prominent example for a degenerate conjugated polymer is trans-polyacetylene. Solitons can be filled with one electron (neutral soliton, spin ½), with two electrons (negatively charged soliton, spin zero) or empty (positively charged soliton, spin zero). PEDOT is not a degenerate conjugated polymer.
2.3 Temperature-Dependent Conductivity To find models and physical descriptions of the charge transport in conjugated polymers, it is useful to consider the temperature dependence of parameters, such as conductivity and thermopower, of those polymers. Some typical features of metals and semiconductors have already been pointed out. In a metal the conductivity remains finite as the temperature approaches 0. This is due to the fact that there are delocalized electronic states at the Fermi
6911X.indb 24
10/1/10 8:58:57 PM
Conductive Polymers versus Metals and Insulators
25
level, which allow conductivity without thermal activation.8 Furthermore, in metals the conductivity decreases with increasing temperature. In semiconductors the opposite is true. An exponential increase in conductivity is a typical behavior for a semiconductor where the hopping of charges occurs between localized states. Polyacetylene is a good example to demonstrate the changes in conductivity and thermopower as a function of temperature and doping. At low doping level the conductivity approaches 0 for T = 0 and increases exponentially with increasing temperature.9 Hence, a typical semiconducting behavior is observed. The newly created charge carriers that are due to the small amounts of dopants added are not mobile, but they are strongly localized to defect sites within the electronic gap.10,11 With increasing doping levels a so-called low temperature semiconductor–metal transition occurs resulting in nonzero conductivities at T = 0. In these cases there is still a continuous increase in conductivity with increasing temperature. However, the fact that the conductivity remains above 0 for T = 0 shows that the sample contains delocalized states at the Fermi level. In highly doped polyacetylene samples, for instance with FeCl3, the term dσ/dT changes its sign at higher temperatures.8 Above this temperature the conductivity decreases with increasing temperature. This so-called crossover temperature is a characteristic feature of many highly doped conducting polymers. There are only very few cases in which a decrease in conductivity has been observed over the whole temperature range. Park et al. reported in 1998 a perchlorate doped polyacetylene, which showed a decrease of conductivity with increasing temperature over the whole temperature range for the first time.12 Hence this sample shows two features of metallic behavior: a nonzero conductivity at T = 0 and a decrease of conductivity with increasing temperature. A second important parameter for the understanding of conductivity in conjugated polymers is the thermoelectric power or thermopower. When a temperature difference is applied to solid, mobile charge carriers—electrons or holes—migrate from the hot side to the cold side similar to a gas that expands during heating.13 If only one type of carrier is able to migrate or migrates stronger than the other one, an electric field is created within the material. The thermopower describes the strength of that field per temperature difference. In the case of iodine doped polyacetylene, the thermopower has a positive sign, indicating that holes are responsible for the charge transport. The thermopower of intrinsically conductive polymers decreases with increasing doping due to the increased charge density. The magnitude of thermopower for semiconductors is typically higher than that of conventional metals. For fully doped conjugated polymers the thermopower shows a linear increase with increasing temperature from 0 to 300 K.14 This is a typical behavior for metallic diffusion.
6911X.indb 25
10/1/10 8:58:57 PM
26
PEDOT
2.4 Order and Disorder In contrast to a semiconductor solid, such as silicon, the structure of a conjugated polymer is by far less regular. Polymers contain individual molecules with different chain lengths, varying amounts of defects, and chain ends; furthermore, they can be amorphous or partially crystalline. Additional differences occur due to aging. Finally, the polymer chains can have orientation in x, y, and z direction resulting in different electronic properties. As a consequence, the disorder in conjugated polymers has a strong effect on the electronic properties. Generally speaking, disorder leads to the localization of charges.15 However, order itself is not a sufficient condition for charge transport, since even in a highly ordered system, macroscopic charge transport is not possible unless the charges can hop or diffuse from one chain to another. High percentages of crystallinity up to 80% to 90% could be obtained for certain polyacetylene samples.2 For polypyrrole, crystallinity of up to 50% has been shown. For polyaniline it was shown that the crystallinity can be increased by stretching the sample after doping with HCl.16 As a consequence of this increased crystallinity the conductivity of the sample is also greatly increased, confirming the general trend that disorder leads to localization of charge. For PEDOT the crystalline order is limited. In the case of PEDOT:Tosylate films, evidence is found for a paracrystalline structure, where the order decays at long distances.17 In the case of PEDOT:PSS films no crystalline structure was observed by x-ray analysis, but the existence of a lower degree of order has been shown that is not detected by x-ray analysis but does have a strong effect on the charge transport (see Chapter 9). Hence, models for disordered conjugated polymers are of particular interest to describe the conductivity in PEDOT. New charge carriers can be introduced into conjugated polymers by the different types of doping described earlier. To participate in charge transport these charge carriers need to be mobile. Already in 1958 Anderson proposed a model for charge transport in systems that are randomly disordered.15 Such systems with a disorder length scale that is equal to or smaller than the electronic correlation length are also described as homogeneously disordered.18 Anderson suggested that at a low concentration of impurities charge transport does not occur by motion of free charge carriers, scattered in the medium, but via quantum mechanical jumps between individual local sites. He then defined conditions under which the wave functions of charge carriers become localized and no charge transport can occur. A system with fully localized states at the Fermi level is also called a Fermi glass.4 Mott proposed that in a homogeneously disordered system a metal–insulator transition must occur when the disorder is sufficiently large. He named this transition Anderson transition due to Anderson’s earlier work.4 Systems in which the disorder length scale is large compared to the electronic correlation length are described as heterogeneously disordered.18
6911X.indb 26
10/1/10 8:58:57 PM
27
Conductive Polymers versus Metals and Insulators
Heterogeneous disorder has been proposed for various types of polyacetylene, and also for PEDOT:PSS a heterogeneous disorder model is appropriate.19 It has been pointed out that the concepts of homogeneous or heterogeneous disorder on their own are insufficient to describe the special carrier behavior in conjugated polymers.18 Nonetheless, these models can give the reader a first understanding for the charge transport in intrinsically conductive polymers, and the calculated data based on these models gives good fits to the experimental results.8 The localization effects in the inhomogeneously disordered conducting polymers are proposed to originate from rod-like quasi one-dimensional chains.2,18,20 Prigodin and Efetov proposed a model for the charge transport for such disordered systems that takes into account ordered regions in which conduction electrons are delocalized and disordered regions in which the electrons must diffuse along isolated chains and become readily localized.22 This model is similar to a percolation model since the conductivity depends on the coupling between individual ordered regions. The difference is, however, that the ordered regions in conjugated polymers do not have sharp boundaries like metal particles in an insulating matrix. Also, a conjugated polymer chain may well be part of two ordered regions. A schematic picture is shown in Figure 2.1.21 The resistance in such a heterogeneous system can be written as the sum of the more ordered parts and the barrier parts, and the overall mathematical description correlates well to the experimental data.8 The crossover temperature observed in the intrinsically conductive polymers can be well described based on the contributions from highly conductive, ordered regions as well as poorly conductive disordered regions. This model also explains the linear increase of the thermoelectric power with temperature since the heat is carried by lattice vibration and charge carriers, but the carrier diffusion itself is dominated by the ordered regions.8
dot 2 dot 1
Figure 2.1 Schematic illustration of spatial relationship of two metallic “dots” made of conducting polymers separated by insulating (disordered) regions. (Reprinted from Physica B: Condensed Matter, 338(114): V. N. Prigodin and A. J. Epstein, Quantum hopping in metallic polymers. Copyright 2003, with permission from Elsevier.)
6911X.indb 27
10/1/10 8:58:58 PM
28
PEDOT
For a random network of coupled one-dimensional metallic wires, Prigodin and Efetov predicted four different conductivity modes dependent on the temperature.22 Initially the material shows a conductivity of variable range hopping. The next transport mode is the hopping over nearest states followed by regions of localization correlations. Finally band transport is reached. For a one-dimensional system the variable range hopping is described by Mott’s temperature law:
1 e2 T0 2 σ DG (T ) = σ 0 exp − with T0 = , T εL
where e is the electron charge, ε is the dielectric constant, L is the localization length, and σ is the conductivity. In highly doped conjugated polymers, localized and extended states exist in parallel. A critical energy Ec exists that separates localized states from extended states. The resulting electronic behavior depends on the position of the Fermi level EF relative to the mobility edge Ec. If the Fermi level lies within the extended states, then the conductivity remains positive even as T approaches 0, since no thermal activation is required for the charge transport. The number of mobile charge carriers may be very small but they still dominate the electronic behavior of the material.23 Based on this assumption Mott calculated a minimum metallic conductivity of 100 S/cm.2 A further feature to determine whether a conjugated polymer shows semiconducting or metallic behavior is the slope of the reduced activation energy, W(T), over the temperature:
W (T ) =
∆E(T ) d ln σ = . kT d ln T
If the Fermi level lies in the region of extended states, the plot of W against T has a positive slope, σdc is finite as T approaches 0, and the material is in the metallic regime.24 If the disorder is so strong that the Fermi level lies in the region of localized states, the carriers show hopping behavior, σ approaches 0 for low temperatures, and the material is in the insulating or semiconducting state. In this case the plot of W against T has a negative slope. PEDOT:PSS can be described as a heterogeneously disordered conjugated polymer.19 The large disorder in PEDOT:PSS is a major difference to other conductive polymers such as polyacetylene or polypyrrole. The temperature dependence of conductivity in PEDOT:PSS films is discussed in Chapter 9 with a particular focus on the results obtained using the variable range hopping model. There are different conclusions about the dimensionality of the hopping mode.19,25 The electronic states in PEDOT are also discussed in Chapter 9
6911X.indb 28
10/1/10 8:58:58 PM
Conductive Polymers versus Metals and Insulators
29
based on its optical and electronic properties. Kim and Bredas have calculated the band structure of PEDOT using density functional theory methods26 based on a PEDOT:Tosylate crystal structure.17 Their calculation showed that in such a crystal there are states available at the Fermi level, opening the possibility for metallic behavior in PEDOT. However, in Chapter 9 the reduced activation energy for PEDOT films of various conductivities is discussed, and as yet no metallic behavior has been observed for PEDOT films.
References
6911X.indb 29
1. E. M. Conwell. 1997. Transport in conducting polymers. In: Handbook of Organic Conductive Molecules and Polymers, ed. H. S. Nalwa, Vol. 4, 1–45. Chichester: John Wiley & Sons. 2. A. J. Epstein. 2007. Insulator-metal transition and transition and metallic state in conducting polymers. In: Handbook of Conducting Polymers, 3rd ed., ed. T. A. Skotheim and J. R. Reynolds, 15-1–15-75. Boca Raton, FL: CRC Press. 3. P. W. Atkins. 1994. Physical Chemistry, 5th ed., 501–505. Oxford University Press. 4. A. J. Heeger. 2001. Semiconducting and metallic polymers: The fourth generation of polymeric materials. J Phys Chem B 105(36):8475–8491. 5. S. Curran, A. Starkhauser, and S. Roth. 1997. Polyacetylene. In: Handbook of Organic Conductive Molecules and Polymers, ed. H. S. Nalwa, Volume 2, 1–60. Chichester: John Wiley & Sons. 6. D. C. Trivedi. 1997. Polyanilines. In: Handbook of Organic Conductive Molecules and Polymers, ed. H. S. Nalwa, Volume 2, 505–572. Chichester: John Wiley & Sons. 7. W. P. Su, J. R. Schrieffer, and A. J. Heeger. 1979. Solitons in polyacetylene. Phys Rev Lett 42(25):1698–1701. 8. A. B. Kaiser. 2001. Systematic conductivity behaviour in conjugated polymers: Effects of heterogeneous disorder. Adv Mater 13(12–13):927–941. 9. K. Ehinger and S. Roth. Non-solitonic conductivity in polyacetylene. Philosophical Magazine B 53(4):301–320. 10. A. J. Heeger, S. Kivelson, J. R. Schrieffer, and W.-P. Su. 1988. Solitons in conducting polymers Rev Mod Phys 60(3):781–850. 11. V. N. Prigodin, F. C. Hsu, J. H. Park, O. Waldmann, and A. J. Epstein. 2008. Electronion interaction in doped conducting polymers. Phy Rev B 78(3):035203-1–035203-9. 12. Y. W. Park, E. S. Choi, and D. S. Suh. 1998 Metallic temperature dependence of resistivity in perchlorate doped polyacetylene. Synth Met 96(1):81–86. 13. “Thermopower.” www.wikipedia.org (accessed August 2010). 14. R. Zuzok, A. B. Kaiser, W. Pukacki, and S. Roth. 1991. Thermoelectric power and conductivity of iodine-doped “new” polyacetylene. J Chem Phys 95(2): 1270–1275. 15. P. W. Anderson. 1958. Absence of diffusion in certain random lattices. Phys Rev 108(5):1492–1505. 16. A. G. MacDiarmid and A. J. Epstein. 1994. The concept of secondary doping as applied to polyaniline. Synth Met 65(2–3):103–116.
10/1/10 8:58:58 PM
30
PEDOT
17. K. E. Aasmundtveit, E. J. Samuelsen, L. A. A. Petterson, O. Inganäs, T. Johansson, and R. Feidenhans’l. 1999. Structure of thin films of poly(3,4-ethylenedioxythiophene). Synth Met 101(1–3):561–564. 18. H. C. F. Martens, J. A. Reedijk, H. B. Brom, D. M. de Leeuw, and R. Menon. 2001. Metallic state in disordered quasi-one-dimensional conducturs. Phys Rev B 63(7):073203-1–073203-4. 19. A. N. Aleshin, S. R. Williams, and A. J. Heeger. 1998. Transport properties of poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate). Synth Met 94(2):173–177. 20. J. Joo, Z. Oblakowski, G. Du, J. P. Pouget, E. J. Oh, J. M. Wiesinger, Y. Min, A. G. MacDiarmid, and A. J. Epstein. 1994. Microwave dielectric response of mesoscopic metallic regions and the intrinsic metallic state of polyaniline. Phys Rev B 49(4):2977–2980. 21. V. N. Prigodin and A. J. Epstein. 2003. Quantum hopping in metallic polymers. Physica B: Condensed Matter 338(1–4):310–317. 22. V. N. Prigodin and K. B. Efetov. 1994. Metal-insulator transition in an irregular structure of metallic chains. Synth Met 65(2–3):195–201. 23. V. N. Prigodin, F. C. Hsu, Y. M. Kim, J. H. Park, O. Waldmann, and A. J. Epstein. 2005. Electric field control of charge transport in doped polymers. Synth Met 153(1–3):157–160. 24. J. Y. Kim, J. H. Jung, D. E. Lee, and J. Joo. 2002. Enhancement of electrical conductivity of poly(3,4-ethylenedioxythiophene)/poly(4-styrenesulfonate) by a change of solvents. Synth Met 126(2–3):311–316. 25. A. M. Nardes, R. A. J. Jansen, and M. Kemerink. 2008. A morphological model for the solvent-enhanced conductivity of PEDOT:PSS thin films. Adv Funct Mater 18(6):865–871. 26. E.-G. Kim and J.-L. Bredas. 2008. Electronic evolution of poly(3,4-ethylenedioxythiophene) (PEDOT): From the isolated chain to the pristine and heavily doped crystals. J Am Chem Soc 130(50):16880–16889.
6911X.indb 30
10/1/10 8:58:58 PM
3 Polythiophenes: A Chance for Maximum Conductivity?
3.1 Introduction Starting with doped polyacetylene as the “prototype” and the most highly conductive polymer known, the main challenges in the development of a technically useful industrial product are given by the intrinsic drawbacks of polyacetylene: its air sensitivity and the missing processability of the doped, π-conjugated polymer. Practically from the beginning the most promising approach to overcome these drawbacks was to stabilize the sensitive π-electron system by heteroatoms in electron-donating substituents or as polymer chain atoms. The latter has been realized only in polyaniline, whereas the first approach is offering more different options—the most interesting substituents being N and S, incorporated in a heterocyclic conjugated structure (Figure 3.1). As pointed out in Chapter 1, polypyrrole has been broadly investigated and found certain industrial usage as conductive polymer anode in capacitors and corrosion inhibitor in base coatings. Both polymers suffer from their intense color and therefore insufficient transparency in thin films, which prohibits several potential applications. By integrating the polyacetylene moiety into a polyheterocyclic system, the electron donating function of N or S ring atoms stabilizes the conjugated system, especially in the doped, positively charged highly conductive state (bipolaron state). Although oligothiophenes were synthesized by Ullmann’s aryl coupling as early as the 1930s and 1940s,1–3 for a long time there was no knowledge about the potential of oligo- and polythiophenes in conducting electric current. A new era was started when Gronowitz and Karlsson opened a novel organometallic pathway for linking thiophene rings in 1960.4 The oxidative coupling reaction was performed by lithiation of the parent thiophene, followed by oxidation with cupric chloride to the corresponding bithiophene. This new reaction initiated a lot of synthetic work, mainly focused on bi- and terthiophenes. A comprehensive review was given by P. Bäuerle.5
31
6911X.indb 31
10/1/10 8:58:59 PM
32
PEDOT
R +O
R O + S
R
R O
O
S
+
+S Figure 3.1 Mesomeric charge delocalization in O-substituted, doped polythiophene moiety. S
The first mention of polythiophenes as potential conducting polymers can be found in the literature in 1967. A. G. Davies and coauthors investigated the acid catalyzed polymerization of furan, pyrrole, and thiophene heterocycles.6 Although this polymerization reaction was known before, several new and interesting facts could be elucidated. Besides the confirmation of the cyclic structure of the polymer units (instead of ring opened), the electric conductivity was investigated. In contrast to the already known types of electronically conducting polymers (polyaniline and polypyrrole, see Chapter 1), the conductivity of all polyheterocycles in Armour’s paper, isolated as trichloro- or trifluoroacetates (triflate), is ionic. In the case of the polythiophene, ion pairs formed by the reaction of polythiophene with trifluoro acetic acid dissociated in methylene chloride solution to the protonated polymer and triflate anions. The polythiophene triflate decomposed at relatively low temperatures (60°C–80°C).6 Tourillon and Garnier first observed true electronic conductivity in polythiophenes in 1982.7 Thiophene was electropolymerized on platinum electrodes in acetonitrile with perchlorate or tetrafluoroborate counterions. A remarkable conductivity of 10–100 S/cm was obtained. By this experiment, a new era of polythiophene chemistry had started, and within a few years polythiophene(s) belonged to the most prominent classes of polymers with electric conductance. Polythiophenes obviously had an enormous potential for technical applications. The development culminated in regioregular poly(3-hexylthiophene) as a semiconductor. Regioregular poly(3-alkylthiophenes) could be synthesized via Kumada coupling8 or a Negishi organozinc protocol,9 starting with 2,5-dibromo-3-hexylthiophene. Several similar cross-coupling methods like Stille10 and Suzuki11 coupling methods were also applied. A further breakthrough was achieved by McCullough and his group by introducing Grignard metathesis (GRIM).12,13 A very detailed review can be read in the Handbook of Thiophene-Based Materials.14 Although highly conductive polythiophenes were achievable from the beginning with Garnier and Tourillon’s fundamental work,7 long-term environmental stability against air and humidity of the doped, bipolaron state— one of the prerequisites for technical use as a truly conductive polymer (not as a semiconductor)—was not fulfilled. Technical processability of doped poly(alkylthiophene)s was also an issue not solved satisfyingly. For example, highly conductive, electrochemically prepared doped poly(alkylthiophene)s
6911X.indb 32
10/1/10 8:58:59 PM
Polythiophenes: A Chance for Maximum Conductivity?
33
can hardly be dissolved in common solvents, like ketones or halogenated hydrocarbons, to a sufficient extent.15 Doped poly(alkylthiophene)s with enhanced alkyl chain length equal to or greater than n-butyl can be obtained via solution processing when utilizing the undoped polymer state as the soluble intermediate, subsequently doping electrochemically (or chemically).16–20 The conductivity of solution-processed films of these polythiophenes is remarkably lower than that of electrochemically prepared ones.15,17,19,20 So soluble or vaporizable oligothiophenes and soluble poly-3-alkylthi ophenes (among other molecules containing thiophene rings, like thiophene-anellated aromatic polycyclics) mostly attracted great attention as semiconductors, and only to a lesser extent as organic “metallic” conductors with high conductivity. This topic will not be discussed here in detail. A lot of comprehensive reviews and also monographs are available. The references should be considered as suggestions.20–24 Thinking about eliminating undesirable α,β′-coupling reactions completely and creating perfectly stereoregular polythiophenes also led to the development of β,β′-disubstituted monomers. Although structural defects from grafting at and cross-linking between polythiophene chains are suppressed, the polymerization of 3,4-dialkylthiophenes suffers from several limitations. The steric hindrance by disubstitution results in distortion of the conjugated polymer, so the effective conjugation length is remarkably reduced. Conductivity, therefore, is lower, and band gap and oxidation potential are higher than for the monosubstituted analogs.20,25 Cyclization of the 3,4-alkyl groups decreases the adverse steric effects, but, for example, 5,6-dihydro-4H-cyclopenta[c]thiophene was found to electropolymerize less smoothly than 3-alkylthiophenes, and the corresponding polymer exhibited inferior electrical and electrochemical properties.20,26 Enhancing the size to 6, 7, or 12 C-atoms in the anellated cycloalkane ring further increases the steric hindrance and, consequently, the band gap. As a result of the even higher positive oxidation potentials of the polymers, the doped films are rather unstable under ambient conditions. Additionally, a sharp decrease in the conductivity of the doped state is observed, compared to poly(cyclopenta[c]thiophene).27 We will see in the following section that some of these findings can be transferred, with several constraints, to oxygen-bearing substituted thiophenes. But besides various analogies there is a remarkable exception.
3.2 Oxygen-Substituted Polythiophenes Oxygen substituents at the 3- and 4-position in the thiophene moiety further stabilize the doped, bipolaronic state in polythiophenes by their electron donating properties (Figure 3.1).
6911X.indb 33
10/1/10 8:58:59 PM
34
PEDOT
Consequently, with the first patents and papers describing oxygen-bearing polythiophenes, a new era on the way to stable, technically useful, highly conductive polythiophenes started. At the end of this development, the EDOT, known for a long time, emerged as the most attractive monomer in 1987–1988. In this section, this development will be outlined until 1988 and the years shortly thereafter, when the first PEDOT publications were published. Several more recent publications from about 1990, dealing with oxygen-substituted polythiophenes, which presented more information about the mechanisms resulting in the excellent conductance and stability of these polymers, are also included. Alkoxy-substituted thiophenes first succeeded as monomers for highly conductive polymers by the work of chemists from Hoechst AG. Several patent applications were filed: the earliest and most important ones from 1986 to 1988.28–34 Mainly, the Hoechst chemists investigated the electrochemical polymerization of 3-alkoxythiophenes.28–32 The alkoxythiophenes were electrolytically polymerized with tetrafluoroborate counterions (from tetraalkyl tetrafluoroborate used as conducting salt). At the beginning of the experiments the best conductivity obtained with pressed powder pellets was 10–2 S/cm (typical values 10 –4 to 10 –3 S/cm).28,35 Only the monosubstituted compounds 3-methoxy-, 3-ethoxy-, and 3-n-butoxythiophene were investigated first,28 and so the conductance of the typically deep blue, insoluble powders was limited. By mass spectroscopic investigations of undoped samples, the oligomeric character of the polythiophenes resulting in mainly molecule ions of penta- and hexamers was observed.28–30 Remarkable progress was achieved, when the Hoechst group modified 3-alkoxythiophenes by additional substitution in the 4-position, especially by methyl groups (Figure 3.2).32,35 Table 3.1 summarizes the conductivity achievable with short-chain 3-alkoxy- and a few 3-alkoxy-4-methylthiophenes.35 Several years later the Mario Leclerc group in Canada also investigated alkoxythiophenes and could elaborate some more structure–property relationships.36 The relatively low conductivity of poly-3-alkoxythiophenes was confirmed. Chemically, with iron-III chloride as the oxidant, polymerized 3-n-butoxythiophene exhibited 8 × 10 –4 S/cm. Although the level of all Leclerc data was lower than that of comparable compounds in Table 3.1, the relatively
R1
R2 R1 = H, CH3 R2 = Alkoxy S
Figure 3.2 Alkoxythiophenes investigated by Hoechst AG.
6911X.indb 34
10/1/10 8:59:00 PM
35
Polythiophenes: A Chance for Maximum Conductivity?
Table 3.1 Electropolymerized Polyalkoxythiophene Conductivity R1 H H H CH3 CH3 CH3
R2
Counterion − 4 − 4 − 4 − 4 − 4 − 4
CH3O C2H5O C3H7O CH3O CH3OCH2CH2O C12H25O
BF BF BF BF BF BF
(S/cm) 0.01 0.01 0.006 220 30 5
Source: Adapted from M. Feldhues, G. Kämpf, H. Litterer, T. Mecklenburg, and P. Wegener, 1989, Synth Met 28: C487–C493.
high conductivity achievable with 3-alkoxy-4-methylthiophenes was also confirmed by the chemically polymerized materials. Additionally, 3,4-di-nbutoxythiophene was studied as a representative of 3,4-dialkoxythiophenes. The structure–property relationship established by Daoust and Leclerc36 can be easily extracted from Table 3.2 (R1 and R2 have the same meaning as in Figure 3.2 or Table 3.1, respectively). Other 3,4-dialkoxythiophenes, as published earlier by Japanese researchers,37,38 had shown better conductivity results by electrochemical polymerization, see Table 3.3 (R1 and R2 corresponding to the formula in Figure 3.2). To summarize, the data for 3,4-dimethoxythiophene and 3-alkoxy-4-methylthiophenes were especially very promising with respect to technical usage in high conductivity applications, and the structural prerequisites necessary for the development of such materials seemed to be understood. But this knowledge, which describes the research state of around 1990, did not result in a single technical product on the basis of any 3-alkoxythiophene derivative, whether substituted by a further alkoxy residue or methyl group or not. In particular, intrinsically conductive polymers (ICP) based on the Hoechst AG patents could not be commercialized at all, and Hoechst finished this development. Table 3.2 Chemically Polymerized Polyalkoxythiophenes Conductivity R1 H CH3 CH3 n-BuO
R2 n-BuO n-BuO n-OctO n-BuO
Counterion − 4 − 4 − 4 − 4
FeCl FeCl FeCl FeCl
(S/cm) 8 × 10−4 2 1 10−5
Source: Adapted from G. Daoust and M. Leclerc, 1991, Macromolecules 24(2):455–459.
6911X.indb 35
10/1/10 8:59:00 PM
36
PEDOT
Table 3.3 Electropolymerized Polymethoxythiophene Conductivity R1 H CH3O CH3O CH3O
R2 CH3O CH3O CH3O CH3O
Counterion − 6 − 4 − 6 − 4
PF ClO PF BF
Reference
(S/cm)
38 37 38 38
0.2 64 1.5 30
Sources: Adapted from T. Hagiwara, M. Yamaura, K. Sato, M. Hirasaka, and K. Iwata, 1989, Synth Met 32(3):367–379; T. Yamamoto, A. Kashiwazaki, and K. Kato, 1989, Makromol Chem 190(7): 1649–1654.
Although highly conductive, the best ICPs derived from monomers des cribed earlier suffer from several intrinsic shortcomings. The most important requirements, besides the high conductivity, for an industrial ICP are long-term stability, including against humidity and air oxygen, and processability. The compound type described in Figure 3.2 and Table 3.1 through Table 3.3 cannot fulfill both requirements. Films of and doped polyalkoxy- and polydialkoxythiophenes chemically prepared from these monomers do not withstand ambient conditions, that is, humid air. After only a few days at room temperature significant dedoping takes place, easily traceable by a typical change in color from blue to magenta or red, accompanied by a sharp drop in conductivity by some orders of magnitude. There seems to be no exceptions with R1 and R2 being discrete substituents. Several examples of the aforementioned polyalkoxythiophenes and their doping/dedoping behavior have been described in more detail in the paper by Daoust and Leclerc.36 The color change in humid air parallels partial or complete dedoping of the tetrachloroferrates, observed with methanol. Whereas poly-3,4-dibutoxythiophene is completely dedoped to the brown-red neutral state by contact with methanol, poly-3-butoxythiophene, poly-3-butoxythiophene, poly-3-butoxy-4-methylthiophene, and poly-3-octyloxy-4-methylthiophene remain partially doped under similar conditions with methanol. Poly-3-alkoxy-4-methylthiophenes exhibit significant thermo- and solvatochromism. Films of these polymers in their neutral, undoped form, are red-violet at room temperature. When heated above approximately 100°C, they become yellow in a reversible transition. When dissolved in a good solvent like chloroform, they give clear yellow solutions. Upon addition of methanol as an example for a poor solvent, the color of these solutions changes to red.36 The difference between the light absorption of violet-red solid poly-3-octyloxy-4-methylthiophene and its yellow chloroform solution is exemplified by the following data of poly[(3-octyloxy)-4-methylthiophene] (POMT): The absorption maximum of POMT in the solid state is found at 545 nm as the result of a highly conjugated backbone.36 In chloroform solution
6911X.indb 36
10/1/10 8:59:00 PM
Polythiophenes: A Chance for Maximum Conductivity?
37
the maximum at 440 nm indicates a remarkably smaller effective conjugation length in a coil conformation.36 Besides the large number of interesting and scientifically important publications regarding poly-3-alkoxythiophenes and poly-3,4-dialkoxythiophenes, quoted only to a minimal extent here, the mandatory conclusion from their behavior was that there is no way to stable, technically useful, highly conductive polymers on this structural basis. The stability of the doped state of all these polymers is not suitable for mainly one reason: The average, effective conjugation length necessary for sufficient mesomeric stabilization of the highly conductive bipolaron state is not adequate. A closer look can distinguish this statement for the polyalkoxythiophene types described earlier. Poly-3-butoxythiophene has a low conductivity because of its low molecular weight and irregular polymer structure (which both implicate a short effective conjugation length).36 This is the result of the reactive C−H in 4-position of the thiophene ring. The same reasons can be claimed for poly-3-methoxythiophene, and last for all poly-3-alkoxythiophenes, unsubstituted at ring atom 4. The low molecular weight of several electrochemically synthesized poly-3-alkoxythiophenes, consisting mainly of oligomers with about six repeating units, has also been described.35 Poly-3,4-dibutoxythiophene is assumed to adopt a nonplanar conformation because of steric interactions. Additionally, due to the poor interchain contacts enforced by the bulky butoxy substituents, a strongly reduced electric conductivity results.36 Here also, similar effects with short-chain 3,4dialkoxythiophenes have to be discussed. The drawbacks of mono- and dialkylthiophene and monoalkoxy thiophene monomers as well are reduced in 3-alkoxy-4-methylthiophenes. Blocking the reactive 4-position by methyl completely inhibits possible adverse structural effects by branching, and, furthermore, a regioregular polymer structure is induced.36 This effect could be correlated with the calculated spin densities in the intermediate radical cations.39 The small methyl group does not adversely influence the interchain contact. The conformational situation in poly-3-alkoxy-4-methylthiophenes was checked by Leclerc via cyclovoltammetric measurements and by UV-Vis spectra in the solid state. The oxidation potentials of poly-3-butoxy-4-methylthiophene (0.60 V) and poly-3-octyloxy-4-methylthiophene (0.64 V) in acetonitril/Bu4NPF6 are significantly lower than that of poly-3,4-dibutoxythiophene (0.70 V).36 A nearly coplanar structure of these polymers in the solid state, in contrast to poly-3, 4-dibutoxythiophene, is concluded, which is responsible for the relatively high conductivity. But obviously the single oxygen atom in 3-alkoxy-4-methylthiophenes is not sufficient for an adequate mesomeric stabilization of the bipolaronic state in the doped polymer. For example, the poly-3-alkoxy-4-methylthiophenes described by Leclerc and Daoust (alkoxy = methoxy, n-butoxy) tend to be partially dedoped even by treatment with methanol, whereas poly-3,4-di-n-butoxythiophene
6911X.indb 37
10/1/10 8:59:00 PM
38
PEDOT
is completely dedoped under these conditions.36 So high conductivity is achieved, accompanied by only moderate stability. The crucial modification step for alkoxythiophenes, supposedly improving the polymers to technically useful products, had to be found. Surprisingly, the “simple” ring closure of two alkoxy substituents completely changes the situation: High conductivity, stability, and processability were combined in the unique polymer molecule poly-3,4-ethylenedioxythiophene (PEDOT).
References
6911X.indb 38
1. W. Steinkopf and W. Köhler. 1936. Thiophene series. XXX. Derivatives of 2,2′bithienyl and α,α,α-tetrathienyl. Liebigs Ann Chem 522:17–27. 2. W. Steinkopf, H.-J. von Petersdorf, and R. Gording. 1937. Thiophene series. XXXV. α,α,α-Quaterthienyls. Liebigs Ann Chem 527:272–278. 3. W. Steinkopf, R. Leitsmann, and K.-H. Hofmann. 1941. Thiophene series. LVII. α-Polythienyls. Liebigs Ann Chem 546:180–199. 4. S. Gronowitz and H.-O. Karlsson. 1960. New syntheses of 2,2′- and 3,3′-bithienyl. Arkiv Kemi 17:89–92. 5. P. Bäuerle. 1998. Oligothiophenes. In: Electronic Materials: The Oligomer Approach, ed. K. Müllen and G. Wegner. 105-197. Weinheim: Wiley-VCH; P. Bäuerle. 1999. The synthesis of oligothiophenes. In: Handbook of Oligo- and Polythiophenes, ed. D. Fichou, 89–181. Weinheim: Wiley-VCH. 6. M. Armour, A. G. Davies, J. Upadhyay, and A. Wassermann. 1967. Colored electrically conducting polymers from furan, pyrrole, and thiophene. J Polym Sci A-1 5:1527–1538. 7. G. Tourillon and F. Garnier. 1982. New electrochemically generated organic conducting polymers. J Electroanal Chem 135:173–178. 8. R. D. McCullough and R. D. Lowe. 1992. Enhanced electrical conductivity in regioselectively synthesized poly(3-alkylthiophenes). Chem Commun 1992:70–72. 9. T. A. Chen and R. D. Rieke. 1992. The first regioregular head-to-tail poly (3-hexylthiophene-2,5-diyl) and a regiorandom isopolymer: Nickel versus palladium catalysis of 2(5)-bromo-5(2)-(bromozincio)-3-hexylthiophene polymerization. J Am Chem Soc 114:10087–10088. 10. A. Iraqi and G. W. Barker. 1998. Synthesis and characterization of telechelic regioregular poly(3-alkylthiophenes). J Mater Chem 8(1):25–29. 11. S. Guillerez and G. Bidan. 1998. New convenient synthesis of highly regioregular poly(3-octylthiophene) based on Suzuki coupling reaction. Synth Met 93(2):123–126. 12. R. S. Loewe, S. M. Khersonysky, and R. D. McCullough. 1999. A simple method to prepare head-to-tail coupled, regioregular poly(3-alkylthiophenes) using Grignard metathesis. Adv Mater 11(3):250–253. 13. R. S. Loewe, P. C. Ewbank, J. Liu, L. Zhai, and R. D. McCullough. 2001. Regioregular, head-to-tail coupled poly(3-alkylthiophenes) made easy by the GRIM method: Investigation of the reaction and the origin of regioselectivity. Macromolecules 34(13):4324–4333.
10/1/10 8:59:00 PM
Polythiophenes: A Chance for Maximum Conductivity?
39
14. P. C. Ewbank, M. C. Stefan, G. Sauvé, and R. D. McCullough. 2009. Synthesis, characterization and properties of regioregular polythiophene-based materials. In: Handbook of Thiophene-Based Materials: Applications in Organic Electronics and Photonics, ed. I. F. Perepichka and D. F. Perepichka, 157–217. Chichester: John Wiley & Sons. 15. J. Roncali, R. Garreau, A. Yassar, P. Marque, F. Garnier, and M. Lemaire. 1987. Effects of steric factors on the electrosynthesis and properties of conducting poly(3-alkylthiophenes). J Phys Chem 91(27):6706–6714. 16. S. Hotta, S. D. D. V. Rughooputh, A. J. Heeger, and F. Wudl. 1987. Spectroscopic studies of soluble poly(3-alkylthienylenes). Macromolecules 20(1):212–215. 17. S. Hotta, M. Soga, and N. Sonoda. 1988. Novel organosynthetic routes to polythiophene and its derivatives. Synth Met 26(3):267–279. 18. R. L. Elsenbaumer, K. Y. Yen, and R. Oboodi. 1986. Processible and environmentally stable conducting polymers. Synth Met 15(2–3):169–174. 19. M. Sato, S. Tanaka, and K. Kaeriyama. 1986. Soluble conducting polythiophenes. J Chem Soc Chem Commun 1986:873. 20. J. Roncali. 1992. Conjugated poly(thiophenes): Synthesis, functionalization, and applications. Chem Rev 92(4):711–738. 21. K. Müllen and G. Wegner, ed. 1998. Electronic Materials: The Oligomer Approach. Weinheim: Wiley-VCH. 22. D. Fichou, ed. 1999. Handbook of Oligo- and Polythiophenes. Weinheim: Wiley-VCH. 23. I. F. Perepichka and D. F. Perepichka, ed. 2009. Handbook of Thiophene-Based Materials: Applications in Organic Electronics and Photonics. Chichester: John Wiley & Sons. 24. M. Jeffries-El and R. D. McCullough. 2007. Regioregular polythiophenes. In: Handbook of Conducting Polymers, 3rd ed., ed. T. A. Skotheim and J. R. Reynolds. Boca Raton, FL: CRC Press. 25. G. Tourillon and F. Garnier. 1984. Structural effect on the electrochemical properties of polythiophene and derivatives. J Electroanal Chem 161:51–58. 26. J. Roncali, F. Garnier, R. Garreau, and M. Lemaire. 1987. Reduction of the steric hindrance to conjugation in 3,4-disubstituted poly(thiophenes); cyclopenta[c]thiophene and thieno[c]thiophene as precursors of electrogenerated conducting polymers. J Chem Soc Chem Commun 1987:1500–1502. 27. J. Rühe, A. Berlin, and G. Wegner. 1995. Poly(cycloalkyl[c]thiophene)s syntheses, electrical properties and charge transport mechanism. Macromol Chem Phys 196:225–242. 28. M. Feldhues, T. Mecklenburg, P. Wegener, and G. Kämpf. EP 257 573 (Hoechst AG), Prior: August 26, 1986/May 26, 1987. 29. G. Kämpf and M. Feldhues. EP 292 905 (Hoechst AG), Prior: May 2, 1987. 30. M. Feldhues, G. Kämpf and T. Mecklenburg. EP 313 998 (Hoechst AG), Prior: October 26, 1987. 31. G. Kämpf and M. Feldhues. EP 328 981 (Hoechst AG), Prior: February 13, 1988. 32. M. Feldhues and G. Kämpf. EP 328 982 (Hoechst AG), Prior: February 13, 1988. 33. M. Feldhues, G. Kämpf, and T. Mecklenburg. EP 328 983 (Hoechst AG), Prior: February 13, 1988. 34. P. Wegener, M. Feldhues, and H. Litterer. EP 328 984 (Hoechst AG), Prior: February 13, 1988. 35. M. Feldhues, G. Kämpf, H. Litterer, T. Mecklenburg, and P. Wegener. 1989. Polyalkoxythiophenes. Soluble electrically conducting polymers. Synth Met 28:C487–C493.
6911X.indb 39
10/1/10 8:59:00 PM
40
PEDOT
36. G. Daoust and M. Leclerc. 1991. Structure-property relationships in alkoxy-substituted polythiophenes. Macromolecules 24(2):455–459. 37. T. Hagiwara, M. Yamaura, K. Sato, M. Hirasaka, and K. Iwata. 1989. Synthesis and properties of poly(3,4-dimethoxythiophene). Synth Met 32(3):367–379. 38. T. Yamamoto, A. Kashiwazaki, and K. Kato. 1989. Polymers and oligomers with substituted 2,5-thienylene units. Preparation and electrical conductivity properties. Makromol Chem 190(7):1649–1654. 39. M. Fréchette, M. Belletête, J.-Y. Bergeron, G. Durocher, and M. Leclerc. 1997. Monomer reactivity vs. regioregularity in polythiophene derivatives. Macromol Chem Phys 198:1709–1722.
6911X.indb 40
10/1/10 8:59:00 PM
4 A Short History of the PEDOT Invention Soon after the discovery of conducting polyacetylene attempts to utilize these exciting new compounds in technical applications started.1–3 Bayer’s Central Research Department also initially focused on polyacetylenes beginning as early as 1980.4,5 Several promising attempts to stabilize polyacetylene in its highly doped form and to achieve processability ultimately failed.6 As a result, Bayer rather promptly decided to abandon this research direction. Attempts to utilize polyacetylene in polarizers several years later failed commercially. A bit of work on conductive tetracyano-quinodimethan complexes followed, but a major technical or even commercial breakthrough was not achieved. A few other charge–transfer complexes were investigated to a limited extent, but they also were not commercialized successfully by Bayer AG.7,8 Therefore the polycondensates workgroup within the Bayer Central Rese arch Department started several projects regarding polyheterocycles in the second half of the 1980s. First, work was focused on polypyrrole, particularly for antistatic layers on thermoplastics like polycarbonate (molded parts or sheets).9–12 After less than two years, the Bayer workgroup was forced to terminate these activities by the intrinsic drawbacks of polypyrrole, like toxicity and high vapor pressure of the monomer, and the intense color and poor transparency of polypyrrole layers itself. A lot of poly(3-alkyl-thiophenes) were known at that time, and some of these special polythiophenes exhibited a remarkable conductivity in the doped state. A comprehensive overview about polythiophenes and their electronic properties available at that time can be found in the Handbook of Oligo- and Polythiophenes.13 The fundamental drawback of all these polythiophenes was the instability of the highly conductive form. In particular, humid air was destructive. Some of the most promising alternative candidates were oxygen substituted, for example, for the polymers from monoalkoxy- and 3,4-dialkoxysubstituted thiophenes (see Chapter 3). The oxygen-bearing substituent is able to stabilize free radical and positive charge carrying forms by further delocalization, as depicted in Figures 4.1 and 4.2. Although a solution seemed to be close by with these thiophenes, a technical breakthrough was achieved not until the Bayer researchers Friedrich Jonas and Gerhard Heywang decided to extend the thiophene structures to bicyclic ring systems. In other words, ring closure of two alkoxy substituents had to be accomplished to form dioxolane-, dioxane-, or dioxepane 3,4anellated thiophenes (see Figure 4.3). 41
6911X.indb 41
10/1/10 8:59:00 PM
42
PEDOT
R
O+
.
+
.
.
.
R
.
+
O +
.S
S
S
S
S
O
O +
+
+
R
R
. O
+
S
S
S
O
O
O
.
R
R
R
R
Figure 4.1 Radical cation (polaron) mesomeric stabilization with oxygen contribution.
The goal was to combine extended mesomeric stabilization by two oxygen atoms with less steric strain, compared to molecules like 3,4-dibutoxythio phene (see Chapter 3). Obviously, the first and most interesting candidate was the dioxolane derivative 3,4-methylenedioxythiophene (I, MDOT). Surprisingly, Jonas and Heywang were not able to isolate appreciable amounts of 3,4-methylenedioxythiophene, although following a logical and straightforward synthetic pathway. They reproduced the synthesis of 3,4methylenedioxy-thiophene-2,5-dicarboxylic acid, published by Dallacker and Mues (see Figure 4.4).14,15 All experiments to completely decarboxylate this dicarboxylic acid seemed to have failed then, and only traces of the target compound could be isolated, not manageable preparatively. Dallacker and Mues were also not able to decarboxylate 3,4-methylenedioxy-thiophene-2,5-dicarboxylic acid (only 3,4methylenedioxy-thiophene-2-carboxylic acid had been prepared via stepwise esterhydrolysis/decarboxylation).15 The reason of these failures is not completely clear, but it was good luck for the Bayer researchers. When Ahonen
R O
R
+
+ S
S
R
R O
O
O + S
+S
Figure 4.2 Cation (bipolaron) mesomeric stabilization with oxygen contribution. (Only one half of the bipolaronic structure shown.)
6911X.indb 42
10/1/10 8:59:02 PM
43
A Short History of the PEDOT Invention
O
O
O
O
O
O
S
S
S
I
II
III
Figure 4.3 Bicyclic dialkoxythiophenes: candidates for stable conducting polythiophenes.
et al. electrochemically polymerized 3,4-methylenedioxythiophene (PMDOT) later, a lot of drawbacks of the polymer were disclosed, especially low conductivity and poor stability of the doped state, all properties correlated with a low conjugation length of the polymer.16 Ahonen et al. did not report about the synthetic procedure to 3,4-methylenedioxythiophene. A lot of years later Lomas and coauthors could first reproducibly synthesize the unsubstituted 3,4-methylenedioxythiophene in quite moderate yields.17 They also note from their results with sterically strained 3,4-alkylenedioxy-2-thienyl-methanols and their intramolecular hydrogen bonding behavior that “from a chemical standpoint … EDOT is situated on a continuum between 3,4-methylenedioxythiophene and other 3,4-alkylenedioxy-, 3-alkoxy- and 3,4-dialkoxythiophenes”; in other words, MDOT would be a very promising candidate for highly conducting polymers.17 But all hopes were dashed when MDOT and its EtO
OEt
O
O +
EtOOC
COOEt
S
NaOEt
EtOH
HO
OH
O
O + CH2BrCl
EtOOC
S
O
1. NaOH 2. HCl
EtOOC
COOEt
HOOC
O
S
–2 CO2
COOEt
S
O
O
COOH
S
Figure 4.4 Attempted (and later realized) synthesis of 3,4-methylenedioxythiophene.
6911X.indb 43
10/1/10 8:59:03 PM
44
PEDOT
ability to polymerize via chemical oxidation was completely checked unsuccessfully in the laboratory of one of the authors, see Chapter 13.18 When the Bayer researchers after the initial failure synthesizing MDOT extended the ring size of the anellated dioxolane to the six-membered dioxane ring in 3,4-ethylenedioxythiophene (see Figure 4.3, molecule II = EDOT), an immediate success was achieved. In sharp contrast to the behavior of cycloalkane-anellated thiophenes, the six-membered ring in EDOT yielded a technical breakthrough. The polymerization to PEDOT by the action of iron-III chloride manifested the particular properties of the resulting polythiophene practically instantaneously, especially the high conductivity and stability of the doped state (only doped PEDOT was available at that time). The patent application DE 38 13 589 A1 of the inventors Jonas, Heywang, and Werner Schmidtberg was filed in Germany on April 22, 1988.19 An extraordinary story of technical success had been started, leading into extensive and readily spreading scientific investigations internally and also, after becoming known to the scientific community, in a lot of university workgroups. Only one week later, on April 30, the same inventors filed a German application (DE 38 14 730 A1) disclosing an important technical application: the use of PEDOT in capacitors.20 Although not the first commercially realized technical application of PEDOT, PEDOT containing capacitors later became one of the largest applications ever. Details are given in Chapter 10. Shortly after the first synthesis of PEDOT by oxidative polymerization, the electrochemical process for the synthesis of PEDOT was found and filed as patent application DE 38 43 412 A1 at the end of 1988 as an extension of the basic patent.21 The group of J. Heinze at the University of Freiburg joined in this invention by important electrochemical, especially cyclovoltammetric, investigations. Although as early as in this patent a first technical application— rechargeable batteries—was claimed, this was not realized up to now (to the best knowledge of the authors). The so-called in situ polymerization and the electrochemical polymerization, both invented in 1988, are special types of processing, utilizing monomeric EDOT to circumvent the completely insoluble doped PEDOT during processing. Regarding the processing of PEDOT, the following question became more important in the following years: Would it be possible to create a processable doped PEDOT? No example for a truly organically soluble doped polythiophene had been reported then, and water was known to degrade the doped state of conducting polythiophenes. But this was not the case for PEDOT, which exhibited remarkable stability against air humidity. So a good chance to find a technical solution for this problem was obvious. Indeed, a scientifically very interesting and—here this wording is even more than true—unexpected technical solution for “persons skilled in the art,” was realized. Agfa-Gevaert AG, a Bayer subsidiary in those days, asked for new antistatics for photographic films. Known antistatics suffered from several drawbacks, for example vanadium pentoxide, utilized by Kodak in
6911X.indb 44
10/1/10 8:59:03 PM
A Short History of the PEDOT Invention
45
Kodachrome films, then for toxicological and environmental reasons, and the sodium salt of poly(styrenesulfonic acid) (PSS-Na), used by Agfa, from the distinct humidity dependence of its (ionic) conductivity. Of course, a permanent, humidity-independent antistatic was required for advanced films. Jonas suggested using the new conductive PEDOT as an antistatic material, and a collaboration of Jonas in the Central Research Department of Bayer AG and his colleague from Agfa-Gevaert (Leverkusen), Werner Krafft, an expert in photographic films, started in 1989. The collaboration rapidly resulted in an intelligent and innovative combination of the Bayer PEDOT invention with the known Agfa film antistatic agent. Surprisingly, waterborne poly(styrenesulfonic acid) (PSS) and the corresponding polyanion, respectively, is able to sufficiently function as the counterion for positively charged doped PEDOT, forming a new PEDOT:PSS complex. During in situ polymerization, the oxidant does not only operate oxidatively for polymerization and doping, but also as the source for the counterion, for example, tetrachloroferrate-III (FeCl4–) from FeCl3 or tosylate from iron-III tosylate. As PSS does not have an oxidative effect, it cannot be the dopant (as sometimes it is called erroneously in the literature), and counterion is its only function. So polymerization and doping had to be performed by an oxidant distinct from PSS. Iron-III compounds are not suitable due to the salt-forming potential with PSS moieties (which results in undesirable precipitation at the end of the polymerization process). Additionally they are objectionable due to their known photographic activity with respect to the scheduled application in photographic films. Jonas and Krafft found that peroxodisulfates (persulfates) like potassium peroxodisulfate (K2S2O8) are very efficient oxidants for polymerizing EDOT and doping PEDOT in the presence of PSS anions as the counterion in water. Furthermore—and this was recognized as one aspect for a real breakthrough—the PEDOT:PSS complex was found to be processable from the resulting very stable microdispersion in water. A lot of details will be presented in following chapters. Early in the year 1990, the PEDOT:PSS, its manufacturing process, the use as antistatic agent and photographic materials therefrom were filed by Bayer in a patent application in Germany (DE 40 03 720, later EP 440 957).22 Parallel to the introduction of PEDOT:PSS as an industrial product for antistatic layers in photographic films on the basis of EP 440 957 (published 1991) and after Bayer had released its first publications on PEDOT early in the 1990s,23,24 widespread investigations of PEDOT and PEDOT:PSS, respectively, at several universities began. The big scientific interest on this highly conductive and stable polymer is documented by the continually growing number of up to several hundred publications per year today, mentioning PEDOT (following SciFinder statistics). Interestingly, PEDOT:PSS is also mentioned in the Advanced Information by the Nobel Committee regarding the 2000 Chemistry Nobel Prize to Heeger, MacDiarmid, and Shirakawa.25
6911X.indb 45
10/1/10 8:59:03 PM
46
PEDOT
References
6911X.indb 46
1. A. G. MacDiarmid, A. J. Heeger, and P. J. Nigrey. EP 36118 (University Patents, Inc.), Prior: March 11, 1980. 2. V. Münch, H. Naarmann, and K. Penzien. EP 44411 (BASF AG), Prior: July 19, 1980. 3. V. Münch, H. Naarmann, and K. Penzien. EP 44935 (BASF AG), Prior: July 11, 1980. 4. J. Hocker, W. Wieder, R. Merten, and J. Witte. EP 45905 (Bayer AG.), Prior: August 9, 1980. 5. J. Hocker, W. Wieder, and R. Dhein. EP 45908 (Bayer AG), Prior: August 9, 1980. 6. H. K. Müller, J. Hocker, K. Menke, K. Ehinger, and S. Roth. 1985. Long-term conductivity decrease in polyacetylene samples. Synth Met 10(4):273–280. 7. G. Heywang and F. Jonas. EP 339419 (Bayer AG), Prior: April 29, 1988. 8. G. Heywang and T. Hassel. DE 4016535 (Bayer AG), Prior: May 23, 1990. 9. K. Sirinyan and F. Jonas. DE 3625272 (Bayer AG), Prior: July 25, 1986. 10. F. Jonas and W. Waldenrath. DE 3725575 (Bayer AG), Prior: August 1, 1987. 11. F. Jonas and W. Waldenrath. DE 3729875 (Bayer AG), Prior: September 5, 1987. 12. F. Jonas. DE 3802472 (Bayer AG), Prior: January 28, 1988. 13. D. Fichou, ed. 1999. Handbook of Oligo- and Polythiophenes. Weinheim: Wiley-VCH. 14. F. Dallacker and V. Mues.1975. Zur Darstellung von 3,4-Methylendioxythiophen-, -furan- und -pyrrol-Abkömmlingen. Chem Ber 108(2):569–575. 15. F. Dallacker and V. Mues. 1975. Reaktionen des 3,4-Methylendioxy-2,5-thioph endicarbonsäurediethylesters. Chem Ber 108(2):576–581. 16. H. J. Ahonen, J. Kankare, J. Lukkari, and P. Pasanen. 1997. Electrochemical synthesis and spectroscopic study of poly(3,4-methylenedioxythiophene). Synth Met 84(3):215–216. 17. J. S. Lomas, A. Adenier, K. Gao, F. Maurel and J. Vaisserman. 2002. Hydrogen bonding and steric effects on rotamerization in 3,4-alkylenedioxy, 3-alkoxy- and 3,4-dialkoxy-2-thienyldi(tert-butyl)-methanols: An NMR, IR and X-ray crystallographic study. J Chem Soc, Perkin Trans 2 2002(2):216–224. 18. K. Reuter. 2002. Unpublished results. 19. F. Jonas, G. Heywang and W. Schmidtberg. DE 38 13 589 A1 (Bayer AG), Prior: April 22, 1988. 20. F. Jonas, G. Heywang and W. Schmidtberg. DE 38 14 730 A1 (Bayer AG), Prior: April 30, 1988. 21. G. Heywang, F. Jonas, J. Heinze and M. Dietrich. DE 38 43 412 A1 (Bayer AG), Prior: December 23, 1988. 22. F. Jonas and W. Krafft. EP 440 957 (Bayer AG), Prior: February 8, 1990 (DE 40 03 720). 23. F. Jonas and L. Schrader. 1991. Conductive modifications of polymers with polypyrroles and polythiophenes. Synth Met 41–43:831–836 (presented on ICSM’90 [International conference on science and technology of synthetic metals], Tübingen, Germany, September 2, 1990). 24. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 25. B. Nordén and E. Krutmeijer. 2000. The Nobel Prize in Chemistry, 2000: Conductive Polymers (Advanced Information). http://nobelprize.org/nobel_prizes/chemistry/ laureates/2000/chemadv.pdf.
10/1/10 8:59:03 PM
5 The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties
5.1 Monomer Synthesis EDOT (3,4-ethylenedioxythiophene) chemistry started as early as the 1930s, when the corresponding 2,5-dicarboxylic acid esters were synthesized.1,2 This was the first description of the special biheterocyclic EDOT system consisting of one 1,4-dioxane ring and one thiophene moiety, annelated over the carbon single ([c]-) bond of the thiophene. One of these early investigations originally was a spinoff from studies in the cantharidin chemistry.2 In a more elaborate study the synthesis of the basic 3,4-dioxy substituted thiophene ring was improved, yielding closely related compounds, for example, 3,4-dimethoxythiophene and several derivatives.3 A detailed synthesis description for 3,4-ethylenedioxythiophene-2,5-dicarboxylic acid (EDOT-2,5dicarboxylic acid) was published by Gogte et al. in 1967.4 The Gogte synthesis started with a Hunsdiecker condensation reaction of oxalic acid diester with thiodiacetic diester to 3,4-dihydroxythiophene-2,5-dicarboxylic acid diester as the first step. 3,4-Dihydroxythiophene-2,5-dicarboxylic acid diester was alkylated with 1,2-dichloro- or 1,2-dibromoethane and then saponified. Decarboxylation of EDOT-2,5-dicarboxylic acid led to EDOT.5 Since its introduction6 into the chemistry of intrinsically conductive polymers (ICPs), the industrial manufacture is based on the Gogte pathway with minor changes, utilizing copper-catalyzed decarboxylation in the last step (Figure 5.1).5,7,8 Sufficient copper catalysts are basic copper carbonate (CuCO3Cu(OH)2) or copper quinoline complexes. Several alternative routes have been suggested, which in some cases are especially useful to prepare alkyl derivatives of EDOT with substitution at the dioxane ring. The most important of these alternative pathways appears to be the acid catalyzed transetherification of 3,4-dimethoxythiophen (or other lower alkoxythiophenes) with vicinal diols.9,10 The Williamson ether synthesis can lead to low yields particularly in the case of long chain 1,2dibromoalkanes due to the competing elimination reactions instead of nucleophilic substitution, resulting in α-olefins or α-acetylenes. Although 47
6911X.indb 47
10/1/10 8:59:03 PM
48
PEDOT
S
HOOC
HO
EtOH
COOH
OH
O COOR
S
O
O
O
ROOC
K2CO3
1) NaOH 2) HCl
COOR
S
O
O
Cu-Cat. HOOC
R = Me, Et
COOH
S
COOEt
NaOMe
Cl
Cl ROOC
EtOOC
COOEt
S
EtOOC
S
Figure 5.1 Synthesis of EDOT from oxalic acid ester and thio-diacetic acid ester.
the Williamson synthesis for sterically hindered EDOT derivatives could be improved by using aliphatic amines or mixtures of DMF and aliphatic amines as a solvent–base combination,11 in such cases transetherification often is the synthetic strategy of choice. For several EDOT derivatives—the benzo-EDOT,12,13 for instance—the transetherification reaction is the very best synthetic access. Transetherification proved to be particularly useful in synthesizing enantiomerically pure chiral disubstituted EDOTs (Figure 5.2), 9,14 where the configurations of the diol carbon atoms are fully retained. An interesting alternative, which uses diols but starts from the 3,4dihydroxythiophene-2,5-dicarboxylic acid diethyl ester (the same intermediate as it is used in the Gogte pathway), was developed independently by Reynolds and colleagues15 and Bäuerle and Caras-Quintero,16 who utilized the Mitsunobu reaction with azodicarboxylic acid ester/phosphane as the etherification agent. The Mitsunobu synthesis also presents good access to chiral EDOT derivatives,16 as demonstrated by the example in Figure 5.3. Since the Mitsunobu reaction is a pure SN2 reaction, the configuration at the chiral C-atom of the secondary alcohol (marked by an asterisk in Figure 5.3) is inverted, leading to an enantiomeric excess of typically more than 97%. The following steps to the substituted EDOT derivative are those known
OR
RO
S
R´ +
HO
R´ OH
(H+) –2 ROH
R´
R´
O
O
S
Figure 5.2 Transetherification as a synthetic route to EDOT derivatives.
6911X.indb 48
10/1/10 8:59:04 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 49
CH3 *
HO HO C2H5OOC
CH3
OH
DIAD, Bu3P
OH
THF (40°C) H5C2OOC
COOC2H5
S
*
O
COOC2H5
S
O
O N
DIAD =
O
O
N O
(diisopropylazodicarboxylate)
Figure 5.3 Mitsunobu reaction for the synthesis of EDOT derivatives. (Adapted from D. Caras-Quintero and P. Bäuerle, 2002, Chem Commun 22:2690–2691.)
from the general procedure shown in Figure 5.1. These conventional reactions (ester hydrolysis and decarboxylation) do not change the configuration at the chiral carbon atom. A very inventive synthesis concept was realized by Roncali and colleagues.17 for the closely related 3,4-vinylenedioxythiophene (VDOT) (Figure 5.4). This elegant pathway was used because no direct nucleophilic substitution route is accessible. Starting by a transetherification reaction with 3,4- dimethoxythiophene, an iodine substituted ether, capable for elimination of HI, was synthesized. After double elimination to 3,4-divinyloxythiophene olefin metathesis with ethylene and a second generation Grubbs ruthenium catalyst was utilized, leading to an overall yield of 55% VDOT. A completely distinct synthesis concept for the thiophene heterocycle has been published, starting with 2,3-dimethoxybutadien and then carrying out the thiophene ring closure reaction with SCl2 (see Figure 5.5).18 In MeO
OMe 2 HOCH2CH2I
ICH2CH2O
OCH2CH2I
–2 HI
(H+)
S
S
O
O
Metathesis with ethylene
O
O
(Ru-catalyst) S
S
Figure 5.4 Synthesis of VDOT.
6911X.indb 49
10/1/10 8:59:06 PM
50
PEDOT
H3CO
OCH3
OCH3
+ SCl2 (NaOAc) OCH3
S
Figure 5.5 3,4-Dimethoxythiophene synthesis from dimethoxybutadiene/SCl2.
the last step of this EDOT synthesis the transetherification reaction of 3,4dimethoxythiophene to EDOT is necessary, using ethylene glycol. The exchange of the furan-oxygen in 2,5-dimethoxy-3,4-ethylendioxytetrahydrofuran by sulfur using NaSH and the elimination of two moles of methanol demonstrate another alternative route to EDOT, as described in a patent application.19 All characteristic last steps in the EDOT synthesis, starting with the ring closure to the dioxane structure, are also sufficient for the synthesis of the analogous seven-membered rings (1,3-dioxepanes), the 3,4-propylenedioxythiophenes (ProDOTs). Williamson ether synthesis,5 transetherification,20 and Mitsunobu reaction have been utilized for dioxepane ring formation.15 The analogous parent five-membered ring compound 3,4-methylenedioxythiophene (MDOT, a 1,3-dioxolane derivative) is also accessible by the Williamson ether synthesis, using bromochloromethane/3,4-dihydroxythiophene-2,5-dicarboxylic acid diester and the subsequent ester hydrolysis and decarboxylation.21
5.2 Physical Properties Pure, freshly distilled EDOT is a nearly colorless liquid with an unpleasant odor. A small change in color to pale yellow is possible after extended storage, especially in the daylight, without significantly affecting the analytical purity. The inhibition or decrease of these effects by base treatment has been described in a patent application.22 Some physical data, including flash point and ignition temperature, are given in Table 5.1.23 The purification procedure of choice is vacuum distillation. Additionally, the melting point of 10.5°C allows a very efficient low temperature recrystallization from solvents like methanol, ethanol or mixtures thereof.24 Several spectroscopic properties are depicted in Figure 5.6 through Figure 5.9. The 1H-NMR spectrum of EDOT (solvent: CDCl3; 400 MHz) consists of two singlets at 4.17 ppm (4H, −OCH2CH2O−) and 6.30 ppm (2H, CH-2,5) (Figure 5.6). The 1H-decoupled 13C−NMR spectrum in CDCl3 (Figure 5.7) exhibits three signals at 141.8 ppm (C-3,4), 99.5 ppm (C-2,5), and
6911X.indb 50
10/1/10 8:59:07 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 51
Table 5.1 Selected Physical Properties of EDOT Viscosity (20°C) Density (20°C) Melting point Boiling point (1013 mbar) Vapor pressure (20°C) Vapor pressure (90°C) Solubility in water (20°C) Flash point Ignition temperature
11 mPa·s 1.34 g/cm³ 10.5°C 225°C 0.05 mbar 10 mbar 2.1 g/l 104°C 360°C
64.6 ppm (−OCH2CH2O−). The EI-mass spectrum (Figure 5.8) is dominated by the M+ peak (m/z = 142) with 100% relative abundance. Using infrared (IR) spectroscopy (Figure 5.9), the typical strong absorption of the C=C-stretching vibration in the thiophene heteroaromatic ring and the ether C−O-stretching band are observed besides the common aromatic and aliphatic C−H vibrations close to 3000 cm–1. The UV-Vis (ultraviolet-visible) spectrum will be discussed in the next section (also see Chapter 9).
120000 110000 100000 90000 6.30 ppm
4.17 ppm
80000 70000 60000 50000 40000 30000 20000 10000 0 –10000
7.4 7.2 7 6.8 6.6 6.4 6.2 6 5.8 5.6 5.4 5.2 5 4.8 4.6 4.4 4.2 4 3.8 3.6 3.4 f2 (ppm) Figure 5.6 1H-NMR spectrum of EDOT (400 MHz; solvent: CDCl ; δ/TMS). 3
6911X.indb 51
10/1/10 8:59:09 PM
52
PEDOT
64.6 ppm
90000
99.5 ppm
80000 70000 60000 50000 40000 30000
141.8 ppm
20000 10000 0
155 150 145 140 135 130 125 120 115 110 105 100 95 90 85 80 75 70 65 60 55 50 45 f1 (ppm)
Figure 5.7 13C-NMR spectrum of EDOT (CDCl ; 400 MHz; 1H-decoupled). 3
3.5×106
142
3.0×106
Abundance
2.5×106 45 2.0×106 1.5×106
68 58
1.0×106
0.0
95
27
5.0×105
0
20
116
81 40
60
80
100
120
140
160
Mass/Charge
Figure 5.8 EI-Mass spectrum of EDOT (minor peaks deleted).
6911X.indb 52
10/1/10 8:59:14 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 53
1487
1.2
555
677
764
892 833
1136
1247
1584
1949
1446
3112 2982 2925 2873
0.2
1422
0.4
1022 935
0.6
1057
1186
0.8 1367
Absorbance
1.0
0.0 3500 3000 2500
2000
1600
1200
800
400
Wave Number [cm–1]
Figure 5.9 IR-spectrum of EDOT (neat film between KBr windows).
5.3 Chemical Properties The most remarkable EDOT reactions are its oxidation reactions, typically resulting in conductive oligomeric to polymeric materials in the presence of charge balancing, so-called doping counterions (anions). These reactions and syntheses will be discussed in detail later (Chapters 7 through 9). There are several other reaction pathways not leading to conductive polythiophenes, which will be in the focus of this chapter. Nevertheless, a lot of them are closely related to the essential EDOT chemistry, that is, the tendency to form electrically active oligomers and polymers. A simple, but mechanistically important feature is the ability of EDOT and a limited number of derivatives to be protonated in α-position of the thiophene ring by strong acids. The protonation—for example, performed by sulfuric acid or organic sulfonic acids, and more efficiently by trifluoro acetic acid—results in the formation of an active, electrophilic [EDOT-H]+ intermediate. Hydrochloric acid leads to additional side reactions; trichloro acetic acid is far less active than the fluoro analog. The [EDOT-H]+ is able to reversibly add to the basic C-2 of another EDOT molecule. The now formed intermediate may deprotonate to a dimeric structure, a 1,4-dihydro-thiophene derivative (see Figure 5.10).25 Because the C-5 in the dimeric product in Figure 5.10 is also easily proto nated, the reaction does not stop at the stage of this dimer but adds one further EDOT molecule to the trimeric structure. Due to the reversibility of all
6911X.indb 53
10/1/10 8:59:14 PM
54
PEDOT
O
O
+ H+ H
δ–
S
O +
H
O H H
S
+
S
O
O
O
O
–
H
S
–H+
S S
O
O
O
O
H O
H
H
H
H
S
H O
O
O
S
S
H H
H O
O
Figure 5.10 Acid catalyzed dimerization and trimerization of EDOT.
steps, a quantitative yield of EDOT dimer or trimer is not achievable, and an equilibrium state with the trimer as the minor component is established. Once isolated, separated, and purified by, for example, column chromatography, the solid dimeric and trimeric compounds are stable but tend to change color in air. Standard cond.: EDT; 5 wt-% cat; RT; CH2Cl2; 8 h Dimer
Trimer
70
Maximum dimer yield: 25–50% Trimer up to 20%
60 wt-%
50 40 30 20
**
H2SO4
p-Ts
CH3SO3H
CF3COOH
SbCl5
ZnCl2
AlCl3
TiCl4
SnCl4
0
BF3xOEt2
10
Equilibrium reaction!
*
*: Conc. 1/10; **: 10 d in n-Butanol Figure 5.11 Catalysts for EDOT dimerization. (p-Ts = Toluenesulfonic acid.)
6911X.indb 54
10/1/10 8:59:15 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 55
Other efficient catalysts for this EDOT di- and trimerization are nonoxidative Lewis acids like BF3, AlCl3, TiCl4, CuCl2, ZnCl2, SnCl4, SbCl5, or TaCl5. AlBr3 or PBr3 are less effective. Several representative examples are shown in Figure 5.11. The reasons for the formation of different dimer:trimer ratios by different catalyst have not been examined. For the sake of simplification, only one stereoisomer of each dimeric and trimeric structure is depicted in Figure 5.10. All possible isomers are formed and can be detected by 1H-NMR-spectroscopy (see Figure 5.12). The characteristic 1H-NMR-spectrum of EDOT-dimer is depicted in Figure 5.13. (The spectrum shows the 1:1-mixture of R- and S-isomer; only the formula of the S-isomer is drawn for the sake of clarity.)
O
O
O H
S
O H
H H
S O
H
S-EDOT-Dimer
O
O S
S
H
S
H O
H
O
O
H
H
H
H
H O
O
S
H
RS-EDOT-Trimer
H
H O
O
O
O
O
H
SS-EDOT-Trimer
S
H
S
O
O
S
H
S H
O
O
O
S
RR-EDOT-Trimer
O
O
O
R-EDOT-Dimer
H
H
S
O
O
S
O
H
H
H
H
S
S H O
S
O
H H
H O
O
SR-EDOT-Trimer
Figure 5.12 All isomeric products of acid-catalyzed EDOT-di- and trimerization.
6911X.indb 55
10/1/10 8:59:15 PM
56
PEDOT
C O A
O
C D
B H
S
H F H
S O
E
E
O
F
D B CH2Cl2 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 3.8 3.6 3.4 (ppm) Figure 5.13 1H-NMR spectrum of EDOT-dimer (400 MHz; solvent: CDCl ; δ/TMS). 3
The importance of dimer (and trimer) formation for basic mechanistic considerations lies in the delocalization of the positive charge. A more detailed look at the mesomeric structures of the protonated EDOT shows the massive influence of the oxygen atom(s) (Figure 5.14). The extension of the mesomeric system on to the oxygen atoms stabilizes the protonated EDOT and thereby facilitates dimerization and subsequent trimerization. The same stabilization is efficient in the polaronic or bipolaronic state of PEDOT (Figure 5.15). Hence, the electronic reasons for the facilitation of the acid catalyzed dimerization (and trimerization) reactions of EDOT are very closely related to the prerequisites for being oxidizable to a stable conducting state, with special emphasis to the word stable in the meaning of electronically and environmentally stable. The latter means stable to air, including oxygen and humidity, and also to water and solvents like aliphatic alcohols. Therefore one can expect that only thiophene monomers prone to dimerization form completely stable, highly conducting, bipolaronic states of the
+O
O
H H
S
O
O H
H H
O
+ S
H
H H
O
O
S
+
H
O
H H
+
S
H
Figure 5.14 Mesomeric charge delocalization in monomeric EDOT.
6911X.indb 56
10/1/10 8:59:16 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 57
+O
O
O
O +
S
S
S
O
O
O
O
+
+S
Figure 5.15 Mesomeric charge delocalization in PEDOT subunit.
polymer. Until now no contradicting examples are known. EDOT and simple, substituted derivatives like EDOT-alkyl (EDOT-CH3 and EDOT-C14H29) or EDOT-CH2OH, ProDOT (to a small extent), and also related compounds like ethylene-3,4-oxythiathiophene (EOTT) undergo these dimerization reactions. A more recent example undergoing both dimerization and formation of a stable conductive state as the polymer is thieno-[3,4-b]-thiophene.26–29 The situation is summarized in Figure 5.16, where monomers representing both features are drawn in comparison to monomers neither dimerizing nor forming highly conductive ICPs. The oxidation (dehydrogenation) of the EDOT-dimer by quinones like chloroanil or 2,3-dichloro-4,5-dicyano-benzoquinone (see Figure 5.17) presents a rapid and easy access to the bis-EDOT (BEDOT) 2,2′-di(3,4-ethylenedioxythiophene),30 R O
O
O
S
O
O
S
S
S
S
S
R = H, Alk, CH2OH Thiophenes undergoing dimerization and polymerization to an air- and water-stable conductive state of the polymer
R
S
R´
R
R
R
O
O
O
S R´ = H, CH3
S
O
O
S
Thiophenes neither undergoing dimerization nor exhibiting a permanently air- and water-stable conductive state of the polymer Figure 5.16 Thiophenes with and without dimerization reactivity.
6911X.indb 57
10/1/10 8:59:17 PM
58
PEDOT
Cl
O
O
Cl
O
H
S
Cl
H H
S O
–
S
Cl Cl
O
O
O
O
S
Cl
O
OH
HO Cl
O
Cl
Figure 5.17 BEDOT synthesis by dehydrogenation of EDOT-dimer.
which can be isolated in a very high purity grade more easily than by conventional organometallic routes via lithiated EDOT. The organometallic routes to BEDOT are outlined in Figure 5.18. They all utilize monolithiation of EDOT in 2-position, followed by oxidative coupling. Viable oxidants are copper-II chlorid31–33 or iron-III acetylacetonate.34 The reactivity of BEDOT with oxidants is significantly enhanced, compared to EDOT (see Chapter 8). However, no preparative dimerization is achieved with strong Brønsted or Lewis acids.35 Nonetheless, primary protonation is detectable. The investigations of Bäuerle and Reinold36 by UV-Vis spectroscopy clearly demonstrate the complete transformation of BEDOT and its next homologues up to the quaterthiophene-derivative (ter-EDOT, quater-EDOT; see Figure 5.19; n = 0 – 2) into α-protonated species by trifluoro acetic acid or trifluoromethane sulfonic acid in dichloromethane. The protonated species exhibit a remarkable bathochromic shift, steadily growing with increasing number of EDOT units. The aforementioned lithiation of EDOT by, for example, butyllithium, is one of the key steps to 2-substituted EDOT derivatives. Some typical subsequent reactions of the 2-Li-EDOT (except the oxidative coupling to BEDOT described earlier) are summarized in Figure 5.20. The transformations of 2-Li-EDOT as a key intermediate to various EDOT derivatives comprise (a) formation of EDOT-Grignard compounds37; (b)
O
O O
O 2
+ 2 BuLi
O
O
CuCl2 30-32
2
S
S
Li
Fe(acac)3 33
S S O
O
Figure 5.18 BEDOT by organometallic synthesis.
6911X.indb 58
10/1/10 8:59:18 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 59
O
O O
O O
O +
H
n
S
S
H H
S
Figure 5.19 Protonated EDOT oligomers.
alkylation37; (c) formation of thienylstannanes,34,38 useful for Stille coupling reactions; (d) addition of ketones to form 2-hydroxymethyl derivatives of EDOT39; (e) transmetalation to zinc compounds,37 which are useful educts for Negishi coupling37; (f) carbonylation with dimethyl formamide (DMF)33; and (g) carboxylation with CO2.37 Dilithiation of EDOT to 2,5-Li2-EDOT is also possible, and the corresponding 3,4-ethylenedioxythiophene-2,5-dicarboxaldehyde is readily available by the
O
O
MgBr
S
O
a
O
O COOH
S
Alkyl
S
b
O
g
O
O
O
c
O
Li
S
S d
f e O
S
O
O
O
Alkyl CHO
O
O
S
S
OH Alkyl
SnR3
a: + Mg/C2H4Br2 b: + Alkyl-Hal c: + R3SnCl d: + Alkyl-CO-Alkyl e: + ZnCl2 f: + DMF g: + CO2
ZnCl
Figure 5.20 Lithiated EDOT as a key intermediate.
6911X.indb 59
10/1/10 8:59:19 PM
60
PEDOT
O
O
O O
Hg
Hg O S
O
Figure 5.21 Dimercury EDOT derivative.
same carbonylation procedure with DMF as mentioned earlier. The transmetalation of 2,5-Li2-EDOT to 2,5-distannyl-EDOT has also been reported.33 Another organometallic compound achievable from 2,5-Li2-EDOT, useful for further syntheses (see below in this section), is the 2,5-dimercury derivative (Figure 5.21).40 Monofunctional organometallic EDOT derivatives have widely been used for the synthesis of tricyclic π-conjugated systems. These functional heteroaromatics with terminal EDOT moieties are of major interest for electro optical applications and as precursors for functional polymers. This topic is too specialized to be discussed here. A comprehensive review covering these compounds and their properties, among others, has been published by Roncali et al.41 Figure 5.22 gives an impression of the large variety of tricyclic conjugated molecules that have been synthesized, covered by the general formula: EDOT-(hetero)arylene group-EDOT. Without going into too many synthetic details, several selected tricyclic compounds from Figure 5.22 shall be exemplarily assigned to the different 2-Li-EDOT reactions, depicted in Figure 5.20. Pyrazine derivatives (1) with R = H or hexyl have been prepared via Pd-0 catalyzed Stille coupling (c in Figure 5.20) with the corresponding dibromo pyrazinothiophenes.42 Stille coupling was also utilized for the synthesis of (3) from 4,7-dibromo-2,1,3 benzothiadiazole.43 The substituted thiophenes (8) were synthesized by nickel(II) chloride catalyzed Grignard cross-coupling according to (a) with 2,5-dibromo thiophenes and 3,4-ethylenedioxy-2-thienyl magnesium bromide (EDOT-2-MgBr).37 The same procedure was utilized for different arylene compounds (9).37 A special Grignard reaction with EDOT2-MgBr was published for the synthesis of (2): 1,2-bis[S-(2-pyridyl)]-benzene dithionate (from 2-mercaptopyridine and phthaloyl chloride) was reacted to the diketone 2-EDOT-CO-(o-phenylene)-CO-2-EDOT. Ring closure with the Lawesson reagent yielded (2).43 Negishi cross-coupling via zinc derived EDOT compound according to (e) with 2,5-dibromothiazole yielded compound (10).37 The carboxylation of 2-lithiated EDOT with carbon dioxide (g) was used as a starting reaction for the preparation of 1,3,4-oxadiazole derivative (11): EDOT-2-carboxylic acid was transferred to its hydrazide via EDOT-2carboxylic acid chloride and the corresponding methyl ester with hydrazine hydrate. Condensation of the hydrazide with EDOT-2-carboxylic acid chloride yielded 1,2-bis[(3,4-ethylenedioxy)thiophene-2-carbonyl]-hydrazine. This precursor was cyclized to (11) with POCl3; achieving an overall yield of 24%.37
6911X.indb 60
10/1/10 8:59:19 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 61
O
O
O
O
Ar
S
(1) Ar =
R
R
N
N
S R = H, n-Hex, pyrrol-2-yl
S
N (2) Ar =
S
N
(3) Ar = S S
N
(4) Ar =
N
Ph
(5) Ar =
Si
S
R
R N
O
Ph
O
R
R
(7) Ar =
(6) Ar = S
S
R = = O, = C(CN)2 S R = 2-Ethylhexyl R
(8) Ar =
(9) Ar =
R
S R = Bu, Oct
R = NO2, F, CO2Me, OBu
(11) Ar =
(10) Ar =
N
N S
N O
Figure 5.22 Tricyclic, EDOT-terminated compounds.
6911X.indb 61
10/1/10 8:59:20 PM
62
PEDOT
O
O
S
O
O
2 n-BuLi Li
S
O
(EtO)3SiCl Li
(EtO)3Si
O
S
Si(OEt)3
Figure 5.23 Synthesis of BTES-EDOT via dilithiated EDOT.
Several basic organic reactions are also feasible with EDOT,33 but, due to the smooth di- and trimerization in the presence of Brønsted and Lewis acids, some problems with acid catalyzed reactions occur, and special conditions are sometimes required. For example, the diacetylation with acetic anhydride to 2,5-diacetyl-3,4-ethylenedioxythiophene was only achieved in good yield by the addition of SnCl4/acetonitril to the CH2Cl2 reaction medium.33 A small amount of the mono-acetylation product was also isolated. Obviously the monoketone can be readily acetylated further, because the electron donating capacity of the ethylenedioxy function predominates over the deactivating nature of the first keto group.33 Also 2,5 -dilithiated EDOT is accessible and a valuable intermediate. So, just recently, EDOT has been dilithiated with n-butyllithium and then reacted with triethoxychlorosilane to 2,5-bis(triethoxysilyl)-3,4-ethylenedioxythiophene (BTES-EDOT) (Figure 5.23).44 BTES-EDOT was copolymerized with EDOT by the action of iron(III)-tosylate in the presence of prepolymerized PEDOT to a hybrid precursor, which then could be cured to conducting films on poly(ethyleneterephthalate) sheets. After rinsing residual iron salts with water, anorganic-organic hybrid conducting films with acceptable conductivity and transparency, but improved hardness and solvent resistance, could be obtained.44 EDOT halogenation easily takes place at positions 2 and 5 of the thiophene ring. N-halogen succinimides are sufficient halogenation agents for the synthesis of the chlorine and bromine derivatives. The reaction can be stopped at the monohalogenation step, only with difficulties. The di-halogenation is the predominant reaction, as demonstrated with the bromo and iodo compounds. So, for example, the bromination of EDOT even with substoichiometric amounts of N-bromo succinimide yielded in a 1:1 mixture of 2-bromo- and 2,5-dibromo EDOT plus starting material.45 The 2-iodo-EDOT has been synthesized from EDOT and N-iodo-succinimide as a mixture with EDOT and 2,5-diiodo EDOT.46 Efficient syntheses for 2,5-diiodo EDOT by iodination of the dilithio-EDOT or of the mercury compound mentioned earlier (Figure 5.21) with elemental iodine have also been published.40 All halogenation products slowly decompose even at ambient temperature and are highly reactive. An interesting new pathway to PEDOT has been opened by these compounds,40,47–49 which will be discussed in detail later (see Chapter 6). As mentioned earlier, EDOT oxidation reactions typically lead to the conducting polymer PEDOT in its different doping states. These technically very
6911X.indb 62
10/1/10 8:59:20 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 63
O
O
m-CPBA
O
O
O
O +
S
S O
S
O
O
Figure 5.24 Oxidation of EDOT with peroxy carboxylic acids.
important, characteristic reactions will be presented broadly and discussed in depth elsewhere (see Chapters 6 through 10). Besides the industrially useful oxidative polymerization, there are several oxidation reactions of EDOT not yielding polymers. This behavior is dominated by the ability of the thiophene sulfur atom to be oxidized. However, special reaction conditions can also result in species oxidized at ring-carbon atom 2. Oxidation of EDOT by m-chloroperbenzoic acid (m-CPBA) in halogenated hydrocarbons results in mixtures of the EDOT-derived sulfone (via sulfoxide intermediate) and the thiolactone heterocyclus 3,4-ethylenedioxy-2(5H)thiophenone (Figure 5.24). By using inorganic persulfates (peroxodisulfates) like Na2S2O8 as oxidants in aqueous medium, the thiophenone derivative, depicted in Figure 5.24, can be isolated.50 By adding a sulfonic acid to the aqueous persulfate system, the EDOT reaction pathway changes: the sulfonic acid functions as a dopant, that is, it provides the counterion for positively charged oligo- and poly-EDOT molecules, and the conductive polymer PEDOT in the form of its sulfonic acid complex is isolated. This type of chemistry will be the topic of the following chapters.
References
6911X.indb 63
1. H. Kondo, S. Ono, and S. Irie. 1937. Pyrrole derivatives. IV. Condensation of β-bromolevulinic ester with acetoacetic ester and ammonia. Yakugaku Zasshi 57:404–406. 2. P. C. Guha and B. H. Iyer. 1938. Attempts towards the synthesis of cantharidin, part II. J Indian Inst Sci 21A:115–118. 3. E. W. Fager. 1945. Some derivatives of 3,4-dioxythiophene. J Am Chem Soc 67:2217–2218. 4. V. N. Gogte, L. G. Shah, B. D. Tilak, K. N. Gadekar, and M. B. Sahasrabudhe. 1967. Synthesis of potential anticancer agents-I, synthesis of substituted thiophenes. Tetrahedron 23(5):2437–2441. 5. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118.
10/1/10 8:59:21 PM
64
6911X.indb 64
PEDOT
6. F. Jonas, G. Heywang, W. Schmidtberg, J. Heinze, and M. Dietrich. EP 339 340 (Bayer AG), Prior: April 22, 1988. 7. M. Coffey, B. R. McKellar, B. A. Reinhardt, T. Nijakowski, and W. A. Feld. 1996. A facile synthesis of 3,4-dialkoxythiophenes, Synth Commun 26(11):2205–2212. 8. G. Rauchschwalbe and F. Jonas. EP 1 142 888 (Bayer AG), Prior: April 4, 2000. 9. L. Groenendaal, F. Louwet, and G. Zotti. WO 02/079295 (Agfa-Gevaert), Prior: March 29, 2001. 10. K. Reuter. DE 10 162 746 (Bayer AG), Prior: January 20, 2001. 11. B. A. Frontana-Uribe and J. Heinze. 2006. Efficient route for the synthesis of 3,4cycloalkoxy-2,5-diethoxycarbonyl-thiophenes obtained with bulky alkyl dibromides using trialkylamines as base–solvent. Tetrahedron Lett 47(27):4635–4640. 12. G. Rauchschwalbe, A. Klausener, S. Kirchmeyer, and K. Reuter. EP 1 275 649 (Bayer AG), Prior: July 12, 2001. 13. S. Roquet, P. Leriche, I. Perepichka, B. Jousselme, E. Levillain, P. Frère, and J. Roncali. 2004. 3,4-Phenylenedioxythiophene (PheDOT): A novel platform for the synthesis of planar substituted π-donor conjugated systems. J Mater Chem 14:1396–1400. 14. D. Caras-Quintero and P. Bäuerle. 2004. Synthesis of the first enantiomerically pure and chiral, disubstituted 3,4-ethylenedioxythiophenes (EDOTs) and corresponding stereo- and regioregular PEDOTs. Chem Commun 8:926–927. 15. K. Zong, L. Madrigal, L. Groenendaal, and J. R. Reynolds. 2002. 3,4-Alkylenedioxy ring formation via double Mitsunobu reactions: An efficient route for the synthesis of 3,4-ethylenedioxythiophene (EDOT) and 3,4-propylenedioxythiophene (ProDOT) derivatives as monomers for electron-rich conducting polymers. Chem Commun 21:2498–2499. 16. D. Caras-Quintero and P. Bäuerle. 2002. Efficient synthesis of 3,4-ethylenedioxythiophenes (EDOT) by Mitsunobu reaction. Chem Commun 22:2690–2691. 17. P. Leriche, P. Blanchard, P. Frère, E. Levillain, G. Mabon, and J. Roncali. 2006. 3,4Vinylenedioxythiophene (VDOT): A new building block for thiophene-based π-conjugated systems. Chem Commun 2006:275–277. 18. F. von Kieseritzky, F. Allared, E. Dahlstedt, and J. Hellberg. 2004. Simple onestep synthesis of 3,4-dimethoxythiophene and its conversion into 3,4-ethylenedioxythiophene (EDOT). Tetrahedron Lett 45:6049–6050. 19. E. J. Bergner and K. Ebel. DE 10 2004 029774 (BASF AG), Prior: June 21, 2004. 20. D. M. Welsh, A. Kumar, E. W. Meijer, and J. R. Reynolds. 1999. Enhanced contrast ratios and rapid switching in electrochromics based on poly(3,4-propylenedioxy-thiophene) derivatives. Adv Mater 11(16):1379–1382. 21. J. S. Lomas, A. Adenier, K. Gao, F. Maurel, and J. Vaissermann. 2002. Hydrogen bonding and steric effects on rotamerization in 3,4-alkylenedioxy-, 3-alkoxyand 3,4-dialkoxy-2-thienyldi(tert-butyl)-methanols: An NMR, IR and X-ray crystallographic study. J Chem Soc, Perkin Trans 2 2:216–224. 22. F. Jonas, K. Wussow, and K. Reuter. DE 10 2006 020744 (H.C. Starck GmbH), Prior: May 4, 2006. 23. CLEVIOSTM—The Ultimate Conductive Polymer. http://www.clevios.com (accessed August 2010). 24. L. Brassat and S. Kirchmeyer. US 2005 065352 (H.C. Starck GmbH), Prior: September 23, 2003. 25. K. Reuter, V. A. Nikanorov, and V. M. Bazhenov. EP 1 375 560 (H.C. Starck GmbH), Prior: June 28, 2002.
10/1/10 8:59:21 PM
The Synthesis of EDOT Monomer, and Its Physical and Chemical Properties 65
26. K. Lee and G. A. Sotzing. 2001. Poly(thieno[3,4-b]thiophene: A new stable low band gap conducting polymer. Macromolecules 34(17):5746–5747. 27. G. A. Sotzing and K. Lee. 2001. Poly(thieno[3,4-b]thiophene as a low band gap conducting polymer and electrochromic material. Poly Mater Sci Eng 85:604–605. 28. G. Sotzing and K. Lee. 2002. A p- and n-dopable polythiophene exhibiting high optical transparency in the semiconducting state. Macromolecules 35(19): 7281–7286. 29. B. Lee, V. Seshadri, and G. A. Sotzing. 2005. Water dispersible low band gap conductive polymer based on thieno[3,4-b]thiophene. Synth Met 152(1-3):177–180. 30. K. Reuter. EP 1 428 827 (H.C. Starck GmbH), Prior: December 10, 2002. 31. G. A. Sotzing, J. R. Reynolds, and P. J. Steel. 1997. Poly(3,4-ethylenedioxythiophene) (PEDOT) prepared via electrochemical polymerization of EDOT, 2,2′-bis(3,4ethylenedioxythiophene) (BiEDOT), and their TMS derivatives. Adv Mater 9(10):795–798. 32. A. Donat-Bouillud, I. Lévesque, Y. Tao, M. D‘Iorio, S. Beaupré, P. Blondin, M. Ranger, J. Bouchard, and M. Leclerc. 2000. Light-emitting diodes from fluorenebased π-conjugated polymers. Chem Mater 12:1931–1936. 33. A. K. Mohanakrishnan, A. Hucke, M. A. Lyon, M. V. Lakshmikantham, and M. P. Cava. 1999. Functionalization of 3,4-ethylenedioxythiophene. Tetrahedron 55:11745–11754. 34. S. S. Zhu and T. M. Swager. 1997. Conducting polymetallorotaxanes: Metal ion mediated enhancements in conductivity and charge localization. J Am Chem Soc 119:12568–12577. 35. K. Reuter. 2002. Unpublished results. 36. E. Reinold and P. Bäuerle. 2005. Personal communication. 37. M. F. Pepitone, S. S. Hardaker, and R. V. Gregory. 2003. Synthesis and characterization of photoluminescent 3,4-ethylenedioxythiophene derivatives. Chem Mater 2003(15):557–563. 38. F. Larmat, J. R. Reynolds, B. Reinhardt, and L. L. Brott. 1996. Electrochemical and electronic properties of poly[bis(2-thienyl)-9,9’-didecylfluorene] and poly[bis(2(3,4-ethylenedioxy)thienyl)-9,9′-didecylfluorene]. Polymer Preprints 37(1):799–800. 39. J. S. Lomas. 2001. Competing inter- and intramolecular hydrogen bonding: solvent-driven rotamerization in 3,4-(ethylenedioxy)-2-thienyldi(tert-alkyl) methanols. J Chem Soc, Perkin Trans 2 5:754–757. 40. H. Meng, D. F. Perepichka, M. Bendikov, F. Wudl, G. Z. Pan, W. Yu, W. Dong, and S. Brown. 2003. Solid-state synthesis of a conducting polythiophene via an unprecedented heterocyclic coupling reaction. J Am Chem Soc 125(49):15151–15162. 41. J. Roncali, P. Blanchard, and P. Frère. 2005. 3,4-Ethylenedioxythiophene (EDOT) as a versatile building block for advanced functional π-conjugated systems. J Mater Chem 15:1589–1610. 42. J. Casado, R. Ponce Ortiz, M. C. Ruiz Delgado, V. Hernández, J. T. López Navarrete, J.-M. Raimundo, P. Blanchard, M. Allain, and J. Roncali. 2005. Alternated quinoid/aromatic units in terthiophenes building blocks for electroactive narrow band gap polymers. Extended spectroscopic, solid state electrochemical, and theoretical study. J Phys Chem B 109(35):16616–16627.
6911X.indb 65
10/1/10 8:59:21 PM
66
PEDOT
43. J.-M. Raimundo, P. Blanchard, H. Brisset, S. Akoudad, and J. Roncali. 2000. Proquinoid acceptors as building blocks for the design of efficient π-conjugated fluorophores with high electron affinity. Chem Commun 11:939–940. 44. J. S. Choi, J.-K. Jeon, Y. S. Ko, Y.-K. Park, S.-G. Kim, and J.-H. Yim. 2009. Electrical and physicochemical properties of poly(3,4-ethylenedioxythiophene)-based organic-inorganic hybrid conductive thin films. Thin Solid Films 517:3645–3649. 45. M. Belletête, S. Beaupré, J. Bouchard, P. Blondin, M. Leclerc, and G. Durocher. 2000. Theoretical and experimental investigations of the spectroscopic and photophysical properties of fluorene-phenylene and fluorene-thiophene derivatives: Precursors of light-emitting polymers. J Phys Chem B 104(39):9118–9125. 46. G. Zotti, G. Schiavon, S. Zecchin, and A. Berlin. 1998. Conducting polymers from anodic coupling of dithienylacetylenes: Electrochemistry and potentialdriven conductive and magnetic properties. Synth Met 97:245–254. 47. H. Meng, D. F. Perepichka, and F. Wudl. 2003. Facile solid-state synthesis of highly conducting poly(ethylenedioxythiophene). Angew Chem 115(6):682–685; Angew Chem Int Ed 42(6):658–661. 48. W.-P. Baik, Y.-S. Kim, J.-H. Park, and S.-G. Jung. US 7,034,104 (Myongji University; Woon-Phil Baik), Prior: December 6, 2002. 49. K. Reuter, S. Kirchmeyer, and F. Jonas. EP 1 741 737 (H.C. Starck GmbH), Prior: July 5, 2005. 50. K. Reuter. 2002. Unpublished results.
6911X.indb 66
10/1/10 8:59:21 PM
6 From EDOT to PEDOT: Oxidative Polymerization and Other Routes This chapter intends to give an overview about all known chemical methods yielding the PEDOT, or poly(3,4-ethylenedioxythiophene), structure. Details for relevant methods for in situ polymerization and the synthesis of PEDOT:PSS, or poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), are presented in Chapters 8 and 9, respectively. Electrochemical synthesis is described in Chapter 14.
6.1 Oxidative Polymerization and Doping Although known for decades, the particular reactivity of 3,4-ethylenedioxythiophene (EDOT) with chemical oxidants was found not earlier than 1988.1 First investigations about the electrochemical oxidation followed soon within the same year.2 The story of the PEDOT invention is disclosed in more detail in Chapter 4. The chemical and technical facts with emphasis on chemical methods will be summarized here. Generally speaking, EDOT is not too stable against oxidation, and it may be completely decomposed by very strong oxidants. So, for example, mixtures of EDOT and nitration agents like concentrated nitric acid quickly exhibit a dark blue color, which after a short time disappears. No defined nitration products can be isolated from these reaction mixtures.3 Logically, first systematic experiments for oxidizing EDOT to PEDOT followed similar routes known for pyrrole. Use of iron(III)-chloride as the oxidant of choice (Figure 6.1) results in insoluble powders with a tremendously high conductivity, compared to all other polyheterocycles known.4 To get a clear stoichiometric description, the equation in Figure 6.1 is depicted for the EDOT hexamer. Synthesized in boiling acetonitrile (bp 82°C), the PEDOT tetrachloroferrate exhibits a conductivity 3000-fold to polypyrrole × FeCl4–, prepared under the same conditions: 15 S/cm versus 5 × 10 –3 S/cm, measured with pressed pellets. Surprisingly, the conductivity of PEDOT-tetrachloroferrate can be further enhanced by an extended reaction of EDOT and FeCl3 in boiling benzonitrile
67
6911X.indb 67
10/1/10 8:59:21 PM
68
PEDOT
O
O 6
– 10 HCl
14 FeCl3
+
– 12 FeCl2
S [FeCl4 ]
_
O
+ S
H
O
S S
O
O
O
S S
O
O
O
O
O
+
[FeCl4 ]
_
S O
H
O
Figure 6.1 PEDOT-tetrachloroferrate synthesis.
(6 h at bp 188°C). This interesting effect clearly demonstrated—as one of the first hints in this direction—the enormous temperature stability of PEDOT. Several other metal ions have been used with different success to oxidize EDOT. Manganese in higher oxidation states, especially manganese dioxide, is commercially used in a special application. PEDOT can be deposited on MnO2 layers to be further used as conductive base for galvanostatic copper plating.5–7 This will be discussed in detail in Chapters 8 and 10. Cerium(IV) was checked in the form of its sulfate Ce(SO4)2 and, because of its better solubility, also with ammonium cerium nitrate (NH4)2Ce(NO3)6 in aqueous solution.8 Despite the high reactivity of cerium(IV) resulting in a very rapid reaction, the PEDOT from oxidation by Ce(IV) had a disappointing low conductivity of not more than 0.075 S/cm.8 CuCl2 was used successfully as a new oxidant recently, yielding PEDOT:Cl in a special oxidative chemical vapor deposition (oCVD) process discussed later.9 Technically useful, highly conductive PEDOT layers in lieu of more or less intractable powders are accessible by using iron(III)-sulfonates as oxidants.1,10 Iron(III)-sulfonates, for example, the salts of toluene sulfonic acid, are soluble in several organic solvents. Aliphatic alcohols like ethanol or n-butanol proved to be particularly useful, dissolving large amounts of iron(III)toluenesulfonate without being oxidized by Fe(III). Mixing EDOT and Fe(III)-tosylate in alcoholic solution is possible without immediate reaction or forming undesired precipitates. Nevertheless, the pot life of these mixtures is limited. More practice-oriented details are discussed in Chapter 8. Pot-life enhancement by addition of organic bases, accompanied by an increased conductivity, was found with imidazole11,12 and pyridine.13 One of the backgrounds for these observations is the formation of toluenesulfonic acid during the course of the EDOT oxidation. Toluenesulfonic acid and its anion, respectively, not only functions as the counterion reservoir for the
6911X.indb 68
10/1/10 8:59:22 PM
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
O
O
6
+
69
– 10 TosH
12 FeTos3
– 12 FeTos2
S Tos– + H
O
O
O
S
S
S
S O
O
S O
O
O
O
O
+
Tos–
S O
H
O
Figure 6.2 PEDOT-tosylate synthesis.
positively charged (doped) EDOT moieties in the polymer, but also is the byproduct of the oxidative polymerization (see the stoichiometric equation in Figure 6.2, depicted for the EDOT hexamer). Although the acid-catalyzed side-reaction of EDOT to EDOT-dimers and -trimers,14 induced by toluenesulfonic acid, is no dead end in the reaction route, because these intermediates are also leading to doped PEDOT at the end of the reaction (see Chapter 8), detrimental effects cannot be excluded and may be one of the reasons for improvements by the addition of bases. Additionally, catalytic effects on EDOT polymerization by protic acids— reducing pot life—are suppressed, and premature precipitation of doped PEDOT is diminished. Similar positive effects could also be achieved by reducing the acid content of iron(III)-toluenesulfonate by ion exchanging.15,16 Layers of PEDOT toluenesulfonate are formed in the course of alcohol evaporation even at room temperature. These layers are practically insoluble and can be rinsed with water to remove iron(II)-toluenesulfonate and excess iron(III) salt. After rinsing and drying, they exhibit an electrical conductivity of up to 1000 S/cm. Besides iron-III salts, special peroxides are the most important group of oxidants, which are able to polymerize EDOT and subsequently dope PEDOT to yield highly conductive PEDOT cations (bipolarons). There are a lot of peroxidic compounds—decomposed thermally or by the catalytic action of metal cations—which, by the reaction of intermediate free oxyradicals, produce typically blue dispersions of PEDOT in water. Most of them are not sufficient to achieve optimal high conductivity. Hydrogen peroxide, alkyl hydroperoxides like tBu-OOH, and diacyl peroxides like dibenzoyl peroxide have all been described as EDOT oxidants, without becoming technically important.17,18 EDOT is oxidized by m-chloroperbenzoic acid to a mixture of the corresponding sulfone and 3,4-ethylenedioxy-2(5H)-thiophenone in
6911X.indb 69
10/1/10 8:59:22 PM
70
PEDOT
SO3H O
O
+
6
6 Na2S2O8
+
–6 Na2SO4 *
*
x
6 H2SO4
S
O + H
S S
O
O
– SO3
O
O
S
S S
O
O
O
O
O
SO3H
+
SO3
S
H
O
O
–
x
SO3H
SO3H
SO3H
Figure 6.3 EDOT oxidation to PEDOT:PSS.
organic solvents (see Chapter 5 and Figure 5.24). The same thiophenone is found in considerable amounts when EDOT is oxidized by potassium peroxodisulfate in water in the absence of additional counter ions. Peroxodisulfates (persulfates) are the best peroxidic oxidants for EDOT to form conductive PEDOT complexes. The overall chemical equation can be written as follows with the example of polystyrenesulfonic acid as the counterion (Figure 6.3, depicted for the hexamer for simplification). Other peroxodisulfates than the sodium salt are also sufficient, like ammonium or potassium peroxodisulfate. Small amounts of metal salts of Fe(II) or Fe(III), for example Fe2(SO4)3, must be added to the reaction mixture. They provide the catalytic decomposition of peroxodisulfate at defined reaction rates and so are pivotal for a high and reproducible conductivity of the PEDOT:PSS complex formed. A detailed description of the PEDOT:PSS complex and its synthetic parameters will follow in Chapter 9. An interesting group of oxidants for EDOT, but nevertheless without practical relevance, are so-called hypervalent iodine compounds. These iodine
6911X.indb 70
10/1/10 8:59:23 PM
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
I
R R´
R = OH, R´ = Ph-SO2-O
Koser's Reagent
R = R´ = CH3-COO
Di(acetoxy)iodobenzene
R = R´ = CF3-COO
Bis(trifluoroacetoxy)iodobenzene
71
Figure 6.4 Hypervalent iodine compounds for EDOT oxidation.
compounds in an oxidation state of +3 or more can be separated into the two groups of inorganic and organic hypervalent iodine19 compounds. Both have been claimed in a patent to yield doped PEDOT when reacted with the monomer.20 Sodium iodate, iodic acid (iodine-V), and sodium periodate (iodineVII) could be used to oxidize EDOT in aqueous medium to the conductive PEDOT:PSS complex. In situ polymerization of EDOT with Koser’s reagent (hydroxy-tosyloxy-iodobenzene), diacetoxy-iodobenzene, or bis(trifluorace tocy)iodobenzene (Figure 6.4) yielded conductive PEDOT layers.20 Beginning with first attempts by J. Kim et al.,21 a new route to highly conductive PEDOT layers by vapor phase polymerization (VPP) was been established in 2003 to 2005.13,22 Substrates coated with iron(III)-tosylate as the oxidant were deposited in a polymerization chamber, where the evaporation of EDOT and its polymerization on the substrate was performed by bubbling different gases (air, nitrogen, argon) through the EDOT reservoir. An acid catalyzed side reaction forming nonconductive material was observed. Obviously, EDOT-dimers and trimers were formed initially, following the typical equilibrium reaction of EDOT with strong acids,14 and then transformed to hardly soluble, nonconductive oxidation products. Therefore the authors did not identify the intermediate EDOT dimers and trimers, but postulated similar oligo- to polymeric structures with dihydroEDOT moieties. This reaction products could be postoxidized by iron(III)tosylate solutions to conductive films. Suppressing the acid catalysis by the action of pyridine, the authors could inhibit the side reaction and obtain wellconductive PEDOT:Tos materials. High conductivities up to about 1000 S/cm were observed after rinsing with ethanol to remove the byproduct iron(II)tosylate from the deposited films. Water vapor instead of pyridine was reported later to play the same role of proton scavenging when introduced into the reaction chamber for the vapor phase polymerization of EDOT.23 The detrimental effect of water on the film structure, resulting in hole formation, could be overcome by the addition of amphiphilic copolymers like poly(ethylene glycol)-ran-poly(propylene glycol).24 An advanced process, completely eliminating all solution processing steps, was proposed by Lock et al.25 and Gleason and Lock.26 A true oxidative chemical vapor deposition process (oCVD) was established by depositing an iron(III)-chloride layer on a substrate by sublimating the FeCl3 in a special reaction chamber.
6911X.indb 71
10/1/10 8:59:25 PM
72
PEDOT
Conductivities as high as 105 S/cm were achieved. Detailed investigations by x-ray photoelectron spectroscopy (XPS), scanning transmission electron microscopy (STEM) with energy dispersive x-ray analysis (EDX), and photoelectron spectroscopy revealed some interesting information about the conductive PEDOT layers.27 Fe could not be detected by XPS; Cl– counterions instead of FeCl4– were incorporated into the PEDOT+-film matrix. Every sixth thiophene moiety of the PEDOT:Cl was found to be positively charged (doping level ca 17%) when deposited at 15°C. By increasing the substrate temperature the doping level could be enhanced to 33%, corresponding to a good conductivity of about 350 S/cm.27 In parallel, the work function—an important parameter for charge injection—is improved from 5.1 eV to 5.4 eV. Bradley’s group of at Imperial College London investigated the vapor phase polymerization of EDOT on to iron(III)-tosylate coated glass or plastic substrates.28 A very high conductivity of about 1200 S/cm was reported, one of the highest values ever found for PEDOT. By introducing a new oxidant for EDOT–CuCl2–PEDOT:Cl films with tunable nanoporosity could be prepared by oCVD, also on difficult or fragile surfaces.9
6.2 “Self-Oxidation” of EDOT Halogen Derivatives A surprising pathway to PEDOT was opened by chance. The low stability of 2,5-dibromo-EDOT (DBEDOT, Figure 6.5) was already known, but nobody was interested in the degradation route and decomposition products. This changed when the Wudl group at the University of California had a closer look at the decomposed material they found after storing 2,5- dibromo-3,4ethylenedioxythiophene unobserved for two years. After this time the originally nearly colorless crystals had been transformed into a dark solid. The dark color indeed was a very intense dark blue, and associating the typical deep blue of doped PEDOT powders was correct. The insoluble and intractable powder had a surprisingly high conductivity: 80 S/cm–1.29 In systematic experiments, time and temperature proved to be the decisive factors for the O Cl
O
S DCEDOT
O
O Cl
Br
S DBEDOT
O
O Br
I
S
I
DIEDOT
Figure 6.5 2,5-Dihalogeno-EDOTs.
6911X.indb 72
10/1/10 8:59:25 PM
73
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
Table 6.1 PEDOT from 2,5-Dibromo-EDOT: Conductivity (S/cm) Temperature (°C) Reaction Time
ca 20
60
80
120
2y
24 h
4h
24 h
Conductivity (S/cm) Crystals/fibers Pellets as synthesized Pellets after I2-doping Thin films Films after I2-doping
80 30 53
Conductivity Conductivity (S/cm) (S/cm) 33 18 30 23 48
20 16 27
Conductivity (S/cm) 0.1 5.8
Sources: Data extracted from H. Meng, D. F. Perepichka, and F. Wudl. 2003. Angew Chem Int Ed 42(6):658–661; H. Meng, D. F. Perepichka, and F. Wudl. 2003. Angew Chem 42(6):682–685.
accessible conductivity. A prolonged reaction time at lower temperatures was found to increase the conductivity. Table 6.1 illustrates time and temperature dependence of the solid-state conductivity of PEDOT samples, prepared from 2,5-dibromo-3,4-ethylenedioxy-thiophene.29 Temperatures above the melting point of the dibromo-EDOT (96°C–97°C) clearly inhibit the formation of the conductive polymer, as observed by the measured conductivity. In the original article the conductivity of PEDOT from dibromo-EDOT is compared to FeCl3-doped PEDOT. The low value for the PEDOT tetrachloroferrate indicated is a little bit misleading, because a conductivity improvement by the “dibromo-EDOT way” is therefore suggested for PEDOT. This is not the case, as can be traced by the various higher values for different PEDOT grades cited within this book up to 1000 S/cm for in situ PEDOT and for PEDOT:PSS, respectively. The polymerization and doping mechanisms are not fully understood, but some deeper knowledge has been obtained by x-ray analysis of the crystalline monomer and EPR spectroscopy investigations during the reaction.30 In situ monitoring of the polymerization showed the development of an EPR signal (g = 2.004).30 This signal reached an intensity maximum, corresponding to polaron = radical cation formation, and then changed its shape, which can be associated with bipolaron formation, as discussed by the authors. A closer look at the kinetic curves between 60°C and 80°C allowed the estimation of the activation energy of this solid-state polymerization to ≈26 kcal/mol. DSC and gravimetric experiments could confirm this value with only minor deviations in the Arrhenius plots. Br3– counterions are postulated by Wudl et al., doping on an average every 2.5th thiophene moiety. Surprisingly, the analog 2,5-dichloro and 2,5-diiodo compounds (DCEDOT and DIEDOT, see Figure 6.5) are both significantly
6911X.indb 73
10/1/10 8:59:26 PM
74
PEDOT
less reactive. The authors were not able to polymerize DCEDOT at temperatures up to the melting point of DCEDOT (ca. 60°C). Following the observations of Wudl’s groups, the self-oxidation of 2,5-diha logeno-EDOTs was assumed solely as a solid-state reaction, which does not occur in solution.29,30 Even molten 2,5-dibromo-EDOT is only extremely slowly reacting to PEDOT, if at all.29,30 Nevertheless, the solution reaction of 2,5-dihalogenoEDOTs to doped PEDOT could be achieved by acid catalysis, as described in a patent application.31 Lewis acids, like BF3 and ZnCl2, are used as well as protic acids like phosphoric acid with or without poly(styrenesulfonic acid). A further type of the Hal2-EDOT polymerization is disclosed in another patent application.32 Here the copolymerization of 2,5-dibromo-EDOT with EDOT in organic solution is described, but the conductivity of the obtained PEDOT powder is low.
6.3 The Organometallic Route to PEDOT Several 2,5-dihalogeno-EDOT compounds as described earlier do not only tend to undergo self-reaction, but also are versatile educts for organometallic reactions due to their high reactivity. The synthesis of neutral, undoped PEDOT was in the focus of most papers uncovering these coupling reactions. The first example was reported by Yamamoto in 1999.33 2,5-Dichloro-EDOT can be dehalogenated by a particular nickel(0) complex to an essentially neutral, apparently undoped PEDOT (Figure 6.6). The careful analysis performed by Yamamoto and Abla revealed a few drawbacks of this method. Although IR spectroscopy demonstrated an essentially undoped PEDOT, and residual Cl could not be found by elemental analysis, the material was insoluble in all organic media, and a molecular weight determination therefore was not possible. Unfortunately, the color of the product has not been disclosed in the Yamamoto communication. So an exact assignment to a definite polymer structure remains difficult, especially in the light of more recent results that confirmed the good solubility of neutral, undoped PEDOT. A very high molecular weight alone should not be the reason for complete insolubility, because medium length polymers of
O
O
n Cl
S
Cl
O
O
[Ni(COD)2/bipy/COD] H
S
n
H
(COD = 1,5-cyclooctadiene; bipy = 2,2‘-bipyridine) Figure 6.6 First organometallic PEDOT-synthesis from 2,5-dichloro-EDOT.
6911X.indb 74
10/1/10 8:59:26 PM
75
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
O
O
+
n Br
S
Br
n Zn
– n ZnBr2
O
O
(NiBr2.bipy/Ph3P) X
S
n
X
X = H, Br Figure 6.7 Organometallic PEDOT-synthesis from 2,5-dibromo-EDOT.
the same basic structure and identical repeating units are well soluble (see Section 6.4). In contrast to the analytical findings above, a small residual doping or some crosslinking has to be discussed for this material. In a subsequent publication, more details about the PEDOT, prepared by Ni(COD)-catalyzed reductive polycondensation of DCEDOT were reported.34 Although an infrared spectrum similar to the spectra of other more or less undoped PEDOT materials was obtained, the nature of the isolated material remains unclear. The black color and its insolubility do not correspond to a typical, completely undoped PEDOT. Garreau, Chevrot, and their colleagues in 2001 described a modified way to a neutral PEDOT not completely halogen free but soluble (Figure 6.7).35,36 The product described in detail by Tran-Van et al.36 contained 1.84% residual bromine in the polymer. So an indetermined part of the PEDOT molecules is constituted by PEDOT oligomers with bromine end groups. Three distinct oligomers with different molecular weights are described by these authors. The reasons for the formation of three—in the course of the reaction obviously preferred—PEDOT oligomer types remain unclear. All were soluble, for example, in N,N-dimethyl-acetamide (DMA). They therefore allowed more detailed analyses than the insoluble product from Yamamoto and Abla.33 Three different signals could be identified by high performance liquid chromatography (HPLC).36 In the absorption spectrum of the oligomer mixture in DMA, three main peaks at 423, 451 (maximum), and 482 nm were observed and assigned to the three oligomers detected by HPLC. No molecular weights were reported. Interestingly, the ultraviolet-visible (UV-Vis) spectrum of the oxidatively prepared undoped PEDOT37 (see Section 6.4) exhibits a λmax value of 451 nm,38 exactly compatible to the main peak.36 Raman and infrared spectra have been recorded35,36 and compared to calculated frequencies from a vibrational analysis in good accordance.36,39 For more details regarding the IR data, see later. The Raman spectra, particularly the main band due to Cα=Cβ-stretching in the thiophene ring, confirmed the low homogeneity of the conjugation length, due to the presence of different oligomers. The organometallic route—in this case dechlorination polycondensation of the 2,5-dichlorothiophene, catalyzed by a nickel(0)-COD-2,2′-bipyridyl complex—was also applied in the synthesis of n-hexyl-PEDOT.34,40 A practically fully undoped material was obtained. So 1H-nuclear magnetic resonance (NMR)
6911X.indb 75
10/1/10 8:59:27 PM
76
PEDOT
analysis and GPC could be performed. By gel permeation chromatography (GPC) against polystyrene standard, a number average molecular weight (Mn) of 5400 and a weight average molecular weight (Mw) of 8500 were found.
6.4 Neutral, Undoped PEDOT by Oxidative Polymerization All former attempts described earlier to chemically synthesize a truly undoped PEDOT were performed by reductive or organometallic means. Another “reductive pathway”—the electrochemical reduction—is not suitable to yield completely undoped, neutral PEDOT. The electropolymerization of EDOT with subsequent electrochemical reduction did not result in the neutral polymer.39 The attenuated total reflection (ATR)-infrared spectrum of such a material clearly exhibited residual-doped segments.39 Electrochemical dedoping of in situ polymerized EDOT (by means of iron(III)-p-toluenesulfonate) at –2 V yielded PEDOT with a residual conductivity of 5 × 10–4 S/cm, at –3 V a material with 6 × 10–5 S/cm was achieved.41 Although quite low, a truly undoped PEDOT should have a conductivity orders of magnitude lower since other semiconducting polythiophenes have a conductivity in the range of 10–6 to 10–10 S/cm in their neutral form, depending on the purity (the lower, the purer). The reduction of in situ polymerized material by the same authors with hydrazine only resulted in a dedoped PEDOT with a conductivity of 1 mS/cm—too high for complete dedoping.41 This is in good accordance with the reduction of PEDOT:FeCl4, investigated by Yamamoto, which also did not yield truly undoped PEDOT.33 Surprisingly, oxidative polymerization with iron(III) chloride is a sufficient method for preparing neutral, fully undoped PEDOT molecules, although only moderate yields are achievable.37 Neutral PEDOT (PEDOT0) is clearly the intermediate in the course to doped, highly conductive PEDOT and can be isolated by adjusting special reaction conditions (Figure 6.8). To suppress premature doping, the presence of a stoichiometric excess of iron(III) chloride over EDOT during the reaction must be avoided. O
O
(FeCl3)
O
O
(FeCl3) S
(–FeCl2, –HCl) S
S
S O
O
(FeCl3)
S
(–FeCl2, –HCl)
S
O
O
O
O
O
O
(–FeCl2, –HCl)
PEDOT
(FeCl3)
PEDOT + FeCl4–
Figure 6.8 Synthesis of neutral PEDOT (PEDOT0).
6911X.indb 76
10/1/10 8:59:27 PM
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
77
Furthermore, the use of an appropriate solvent—for example, a halogenated hydrocarbon—and the presence of an acid acceptor like calcium carbonate are helpful to isolate neutral PEDOT in acceptable yields in the lower twodigit percent range. Because even under optimal conditions doping cannot be excluded completely, the yield of undoped PEDOT is limited. As iron(III) chloride additionally functions as Lewis acid, trace amounts of the EDOT dimer and trimer (see Chapter 5) could be detected in the product 1H-NMR-spectroscopically by the clearly separated peak at 5.42 ppm (σ).38 Logically, the neutral PEDOT synthesized oxidatively by the method of EP 1,327,64537 is not a high polymer. Due to its good solubility, GPCs are easily obtained and demonstrate typical low molecular weights of about 1500 (weight average). This value corresponds well to that obtained by 1H-NMR spectroscopic end group analysis. Comparing other spectroscopic data with those in the Chevrot and Garreou articles,35,36 the conclusion may be considered that similar molecular weights have been achieved by both methods. Infrared and absorption spectra are very similar, the crystalline products and solutions in organic solvents (halogenated or aromatic hydrocarbons) exhibit a characteristic deep purple color. Higher molecular weight fractions are less soluble in aromatic solvents like toluene, but remain chloroform soluble. The characteristic infrared spectrum of neutral PEDOT, synthesized by chemical oxidation of EDOT (Figure 6.9), can be compared to data in the literature and shows the chemical similarity (Table 6.2). The peaks between 1350 and 1490 cm–1 are assigned to the ring vibration modes of the thiophene ring (C=C 1.4 1.2
Absorbance
1.0 0.8 0.6 0.4 0.2 0.0 4000 3500 3000 2500
1800
1500
Wave Number [cm–1]
1200
900
600
Figure 6.9 Infrared spectrum of oxidatively prepared undoped, neutral PEDOT. (Adapted from K. Reuter, 2001. Unpublished results.)
6911X.indb 77
10/1/10 8:59:28 PM
78
PEDOT
Table 6.2 Comparison of IR Data of Different Undoped PEDOT Grades Reference 36, 37 35 33a 38b a b
IR Peaks 1470 1465 1470 1488
1436 1433 1430 1427
1363 1359 1360 1366
Synthesis 1060 1069 1070 1051
899
Oxidatively prepared Ni(PPh3)2/bipy/DBEDOT Ni(COD)2/bipy/DCEDOT Electrochemically, then reduced
925 892
Estimated from Figure 1 in T. Yamamoto and M. Abla, 1999, Synth Met 100(2):237–239. Additional peaks from residual doped material.
antisymmetric stretching, C=C symmetric stretching and C−C stretching36), whereas at about 1070 cm–1 the C−O stretching and at about 900 an O−C−C deformation (tentative assignment) of the dioxane ring are observed.36 Under ambient conditions, the neutral, undoped PEDOT is a stable powder, and the material does not tend to discolor in air. This is, although surprising on the first glimpse, not inconsistent with the very facile oxidability of ter-EDOT described by Reynolds and his colleagues.42 The parallels may be found in the sensitivity of all conjugated EDOT oligomers—not to be confused with the true dimers and trimers with dihydrothiophene structures described earlier—to oxidants, which is remarkable enhanced by acids. Solutions of neutral PEDOT in chloroform therefore slowly generate a blue precipitate, especially when catalytically active traces of acids are present. The oxidation sensitivity in solution is high: typical EDOT oxidants like iron(III) tosylate immediately precipitate doped material from organic solutions. So utilization similar to EDOT/Fe(III)tosylate via in situ doping is practically impossible due to the very short pot life and rapid reaction of all neutral PEDOT/oxidant mixtures. Applying PEDOT0 solutions on an active ground of iron(III)-tosylate layers results in the formation of blue “films” of doped PEDOT:Tos, but the conductivity is very low and the films are inhomogeneous, substantially consisting of small particles. Another route to blue, more or less doped PEDOT:PSS films is doctor blading PEDOT0 solutions on to a poly(styrenesulfonic acid) base ground and oxidizing by air during the drying process. Here the obtained films are more homogeneous, but doping is not sufficient to achieve good conductivities. Similar to the effect demonstrated by the Bäuerle group for low EDOT “oligomers” (2, 3, or 4 EDOT units; n = 0, 1, 2 in Figure 6.10; see also Chapter 5),43
H
O O
O O
O
O
+ H S
S
n
S
H
Figure 6.10 Protonated EDOT oligomers and polymers.
6911X.indb 78
10/1/10 8:59:28 PM
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
79
a bathochromic shift can be observed for the neutral PEDOT when acidified in organic solution by, for instance, trifluoro acetic acid. This effect results from the protonation of EDOT end groups. Neutral PEDOT thereby changes its color from purple to a greenish blue. In contrast to the monomeric EDOT,14 protonation of all EDOT homologs like bis-EDOT (BEDOT), ter-EDOT, and so on, does not yield dimerization products in preparative experiments.
References
6911X.indb 79
1. F. Jonas, G. Heywang, and W. Schmidtberg. DE 38 13 589 (Bayer AG), Prior: April 22, 1988. 2. G. Heywang, F. Jonas, J. Heinze, and M. Dietrich. DE 38 43 412 (Bayer AG), Prior: December 23, 1988. 3. K. Reuter. 2003. Unpublished results. 4. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 5. J. Hupe, G. D. Wolf, and F. Jonas. 1995. DMS-E: Bekanntes Prinzip mit neuer Basis. Galvanotechnik 86(10):3404–3411. 6. G. D. Wolf, F. Jonas, and R. Schomaecker. EP 707 440 (Bayer AG), Prior: October 12, 1994. 7. S. Kirchmeyer and F. Jonas. WO 2000 045625 (Bayer AG), Prior: January 27, 1999. 8. R. Corradi and S. P. Armes. 1997. Chemical synthesis of poly(3,4-ethylenedioxythiophene). Synth Met 84(1-3):453–454. 9. S. G. Im, D. Kusters, W. Choi, S. H. Baxamusa, M. C. M. van de Sanden, and K. K. Gleason. 2008. Conformal coverage of poly(3,4-ethylenedioxythiophene) films with tunable nanoporosity via oxidative chemical vapor deposition. ACS Nano 2(9):1959–1967. 10. F. Jonas, G. Heywang, W. Schmidtberg, J. Heinze, and M. Dietrich. EP 339 340 (Bayer AG), Prior: April 22, 1988. 11. D. M. de Leeuw, P. A. Krakman, P. F. G. Bongaerts, C. M. J. Mutsaers and D. B. M. Klaassen. 1994. Electroplating of conductive polymers for the metallization of insulators. Synth Met 66(1–3):263–273. 12. C. M. J. Mutsaers, D. M. de Leeuw, and M. M. J. Simenon. EP 615 256 (Koninklijke Philips Electronics N. V.). Prior: March 9, 1993. 13. B. Winther-Jensen, D. W. Breiby and K. West. 2005. Base inhibited oxidative polymerization of 3,4-ethylenedioxythiophene with iron(III)tosylate. Synth Met 152(1-3):1–4. 14. K. Reuter, V. A. Nikanorov, and V. M. Bazhenov. EP 1 375 560 (H.C. Starck GmbH). Prior: June 28, 2002. 15. U. Merker, S. Kirchmeyer, and K. Wussow. WO 2004/088672 (H.C. Starck GmbH). Prior: April 2, 2003; May 28, 2003. 16. U. Merker, K. Wussow, S. Kirchmeyer, C. Schnitter, and K. Lerch. 2003. Manufacturing process for low ESR polymer electrolyte capacitors. Proceedings of the 17th Passive Components Conference CARTS Europe, Stuttgart, p. 79.
10/1/10 8:59:28 PM
80
PEDOT
17. K. Reuter, S. Kirchmeyer, U. Merker, P. W. Lövenich, and T. Meyer-Friedrichsen. WO 2008/034848 (H.C. Starck GmbH), Prior: September 20, 2006 18. P.-M. Allemand. US 2006/0065889, Prior: September 30, 2004. 19. A. Varvoglis. 1997. Hypervalent Iodine in Organic Synthesis. London: Academic Press. 20. K. Reuter and S. Kirchmeyer. EP 1 979 393 (H.C. Starck GmbH), Prior: January 20, 2006. 21. J. Kim, E. Kim, Y. Won, H. Lee, and K. Suh. 2003. The preparation and characteristics of conductive poly(3,4-ethylenedioxythiophene) thin film by vapor-phase polymerization. Synth Met 139(2):485–489. 22. B. Winther-Jensen and K. West. 2004. Vapor-phase polymerization of 3,4ethylenedioxythiophene: A route to highly conducting polymer surface layers. Macromolecules 37:4538–4543. 23. M. Fabretto, K. Zuber, C. Hall, and P. Murphy. 2008. High conductivity PEDOT using humidity facilitated vacuum vapour phase polymerisation. Macromol Rapid Commun 29(16):1403–1409. 24. K. Zuber, M. Fabretto, C. Hall, and P. Murphy. 2008. Improved PEDOT conductivity via suppression of crystallite formation in Fe(III) tosylate during vapor phase polymerization. Macromol Rapid Commun 29(18):1503–1508. 25. J. P. Lock, S. G. Im, and K. K. Gleason. 2006. Oxidative chemical vapor deposition of electrically conducting poly(3,4-ethylenedioxythiophene) films. Macro molecules 39:5326–5329. 26. K. K. Gleason and J. Lock. US 2006/0269664 (Massachusetts Institute of Technology), Prior: 31 May 2005 27. S. G. Im, K. K. Gleason, and E. A. Olivetti. 2007. Doping level and work function control in oxidative chemical vapor deposited poly(3,4-ethylenedioxythiophene). Appl Phys Lett 90:152112-1–152112-3. 28. P. A. Levermore, L. Chen, X. Wang, R. Das, and D. D. C. Bradley. 2007. Highly conductive poly(3,4-ethylenedioxythiophene) films by vapor phase poly merization for application in efficient organic light-emitting diodes. Adv Mater 19(17):2379–2385. 29. H. Meng, D. F. Perepichka, and F. Wudl. 2003. Facile solid-state synthesis of highly conducting poly(ethylenedioxythiophene). Angew Chem Int Ed 42(6): 658–661; H. Meng, D. F. Perepichka, and F. Wudl. 2003. Facile solid-state synthesis of highly conducting poly(ethylenedioxythiophene). Angew Chem 42(6):682–685. 30. H. Meng, D. F. Perepichka, M. Bendikov, F. Wudl, G. Z. Pan, W. Yu, W. Dong, and S. Brown. 2003. Solid-state synthesis of a conducting polythiophene via an unprecedented heterocyclic coupling reaction. J Am Chem Soc 125(49):15151–15162. 31. W.-P. Baik, Y.-S. Kim, J.-H. Park, and S.-G. Jung. US 7,034,104 (Myongji Univ. Seoul/Woon-Phil Baik), Prior: December 6, 2002. 32. K. Reuter, S. Kirchmeyer, and F. Jonas. EP 1 741 737 (H.C. Starck GmbH & Co. KG), Prior: May 5, 2005. 33. T. Yamamoto and M. Abla. 1999. Synthesis of non-doped poly(3,4-ethylenedioxythiophene) and its spectroscopic data. Synth Met 100(2):237–239. 34. T. Yamamoto, K. Shiraishi, M. Abla, I. Yamaguchi, and L. Groenendaal. 2002. Neutral poly(3,4-ethylenedioxythiophene-2,5-diyl)s: Preparation by organometallic polycondensation and their unique p-doping behavior. Polymer 43(3):711–719.
6911X.indb 80
10/1/10 8:59:28 PM
From EDOT to PEDOT: Oxidative Polymerization and Other Routes
81
35. F. Tran-Van, S. Garreau, G. Louarn, G. Froyer, and C. Chevrot. 2001. A fully undoped oligo(3,4-ethylenedioxythiophene): Spectroscopic properties. Synth Met 119(1–3):381–382. 36. F. Tran-Van, S. Garreau, G. Louarn, G. Froyer, and C. Chevrot. 2001. Fully undoped and soluble oligo(3,4-ethylenedioxythiophene)s: Spectroscopic study and electrochemical characterization. J Mater Chem 11:1378–1382. 37. K. Reuter and S. Kirchmeyer. EP 1 327 645 (Bayer AG), Prior: December 27, 2001. 38. K. Reuter. 2001. Unpublished results. 39. S. Garreau, G. Louarn, J. P. Buisson, G. Froyer, and S. Lefrant. 1999. In situ spectroelectrochemical Raman studies of poly (3,4-ethylenedioxythiophene) (PEDT). Macromolecules 32(20):6807–6812. 40. K. Shiraishi, T. Kanbara, T. Yamamoto, and L. Groenendaal. 2001. Preparation of a soluble and neutral alkyl derivative of poly(3,4-ethylenedioxythiophene) and its optical properties. Polymer 42(16):7229–7232. 41. T. Johansson, L. A. A. Pettersson, and O. Inganäs. 2002. Conductivity of dedoped poly(3,4-ethylenedioxythiophene). Synth Met 129(3):269–274. 42. G. A. Sotzing, J. R. Reynolds, and P. J. Steel. 1996. Electrochromic Conducting polymers via electrochemical polymerization of bis(2-(3,4-ethylenedioxy)thienyl) monomers. Chem Mater 8(4):882–889. 43. U. Reinold and P. Bäuerle. 2005. Unpublished results.
6911X.indb 81
10/1/10 8:59:28 PM
6911X.indb 82
10/1/10 8:59:28 PM
7 Counterions for PEDOT This chapter describes the effect of counterions on poly(3,4-ethylenedioxythio phene) (PEDOT). Counterions are present in all oxidized forms of PEDOT for charge balancing. Hence, they are also mentioned in the previous and following chapters, dealing with the charged form of PEDOT. This chapter aims to give an overview and to compare the effects of different counterions. Since chemistry and physics of electrochemical and chemical polymerization are different, the effect of counterions on these two techniques is discussed in two sections.
7.1 Counterions in Electrochemically Polymerized PEDOT The electrochemical polymerization of PEDOT allows the introduction of a wide range of counterions since the latter can be added in the form of salts to the reaction mixture. The choice of counterions is limited only by their solubility and stability under the reaction conditions. A list of selected counterions used in the electrochemical polymerization is shown in Table 7.1 together with the methods used and the conductivities obtained. Different polymerization methods such as potentiostatic, galvanostatic, and repetitive multisweep result in different PEDOT films with differing properties.1 The in situ photoelectropolymerization uses light as well as a constant potential.2,3 Brief descriptions of polymerization methods can be found in the description of Table 7.1. The highest conductivities are obtained in the presence of perchlorate, using acetonitrile as solvent (method C and F with 650 and 780 S/cm, respectively).6,8 The results obtained by El Moustafid et al. show that water as a solvent leads to lower conductivities than organic solvents such as propylene carbonate or acetonitrile for the same counterions.7 Using water as solvent, the highest conductivity was found in the presence of nitrate. The series of measurements using bis(perfluorosulfonyl)imides or perfluorosulfonates with different chain lengths performed by Xia et al. using in situ photoelectropolymerization indicates that long perfluoroalkyl chains decrease the conductivity of the resulting films.2 Several groups performed the electrochemical polymerization of EDOT in the presence of polymeric counterions. Yamato et al. were the first to report 83
6911X.indb 83
10/1/10 8:59:29 PM
84
PEDOT
Table 7.1 Conductivities Obtained in Electrochemical Polymerization of EDOT Method
A
B
Counterion
D
E
F
G
780
650
Conductivity at RT (S/cm)
– 4
ClO BF4– PF6– NO3– SO4– Tos– N(SO2CF3)2− N(SO2C2F5)2− N(SO2C3F7)2− N(SO2C4F9)2− CF3SO3− C4F9SO3− C8F17SO3− PSS−
C
105
30 200
400–650 280 120–150
8.5
95
50
200
50–80
55 24 49 79 41 36
30 121
125–450 130 108 106 25 86 91 8 50–80
Sources: (A) I. Giurgiu et al., 2001, Synth Met 119(1–3):405–406.4 (B) A. Aleshin et al., 1997, Synth Met 90(1):61–68.5 (C) L. Groenendaal et al., 2003, Adv Mater 15 (11):855–8791; L. Groenendaal, G. Zotti, and F. Jonas, 2001, Synth Met 118(1–3):105–109.6 (D) T. El Moustafid et al., 2002, Polymer Preprints 43(2):1320.7 (E) J. Xia et al., 2008, J Phys Chem C 112:11569–115742; J. Xia et al., 2008, J Am Chem Soc 130(4):1258–1263.3 (F) M. Granström and O. Inganäs, 1995, Polymer 36(15):2867–2872.8 (G) G. Zotti et al., 2003, Macromolecules 36(9):3337–3344.9 Notes: Method A—polymerization in propylene carbonate (PC) at –30°C with I = 0.04 mA/cm2; method B—polymerization in PC at –30°C with I = 0.01 – 0.06mA/cm2; method C—polymerization in acetonitrile at RT with E = 0.85 – 1.1 eV; method D— polymerization in water with tenside at RT with sweeps between –1 and +1 V; method E—in situ photoelectropolymerization in acetonitrile with a series of potentials above the half-wave oxidation potential; method F—polymerization in microfiltration membranes in acetonitrile at 1.3 V; method G—polymerization at 0.8 V in acetonitrile (in the case of PSS in acetonitrile and water).
the electrochemical polymerization of EDOT in aqueous polystyrenesulfonic acid (PSS) solution and showed that the oxidation occurs at a similar potential as those in organic solvents.10 The use of polyelectrolyte dopants improves the mechanical stability of the film.11 The conductivity of films using PSS as the counterion in electropolymerization is lower compared to those using low molecular weight counterions (50–80 S/cm). Yamato et al. furthermore used sulfonated poly(β-hydroxyethers) as counterions and obtained conductivities of 150 to 180 S/cm.11 These films could be used as immunosensors for antigens and antibodies.12 Sonmez et al. used poly(2-acrylamido-2methyl-1propane sulfonate) as the counterion and obtained similar conductivities to those obtained with PSS (80 S/cm).13 The authors were able to show that such films have electrochromic and cation exchange properties.
6911X.indb 84
10/1/10 8:59:29 PM
Counterions for PEDOT
85
The choice of the counterion may also have an effect on the oxidation level (doping level). In 1995 Granström and Inganäs modeled the x-ray diffraction pattern for PEDOT with perchlorate as the counterion as obtained electrochemically in well-defined pores.8 The best fit was obtained for a crystal with one perchlorate counterion for every four thiophene repeat units in an ortho rhombic cell. Niu et al. polymerized 3,4-ethylenedioxythiophene (EDOT) in the presence of tetra-n-butyl ammonium hexafluorophosphate on platinum foils and were able to obtain the resulting polymer as a powder.14 Similar to the results of Granström et al., the x-ray diffraction pattern revealed an orthorhombic unit cell with four thiophene units and one hexafluorophosphate ion, in which stacks of PEDOT are separated by layers of counterions. A different doping level using perchlorate was obtained by Zotti et al., who used electrochemical quartz crystal microbalance analysis to relate the charge stored in the polymer to the dried mass.9 Using tosylate and perchlorate they obtained a ratio of one counterion for every three thiophene units. Using the same technique, Zotti examined the electropolymerization in the presence of PSS. In this case the resulting film contained more sulfonate moieties than required for the charge balance simply due to the polymeric nature of the counterion. Interestingly, it was found that the polymeric composition in the film was independent from the PSS concentration in the solution. Furthermore, Zotti found that at low concentrations of EDOT the presence of tosylate hinders the polymerization.9 No polymerization is observed at EDOT concentrations below 0.1 M in the presence of tosylate, whereas the polymerization of EDOT in the presence of perchlorate proceeds readily. This can be explained by the chemical nature of the counterion. The tosylate ion is a base strong enough to deprotonate the EDOT radical cation, and the neutral radical then undergoes further reactions to nonconductive products. This pathway plays a negligible role at higher concentrations but limits the electrochemical polymerization at low concentrations.9 On the other hand, the tosylate ion is better at forming ion pairs and shielding the radical cation and thereby speeds up the polymerization processes. Perchlorate on the other hand is very poor in forming ion pairs.15 This was shown in the polymerization of 2,2′-bis-EDOT in the presence of tetra-n-butylammonium perchlorate or tosylate. The number of exchanged electrons as extracted from the peak current in the case of perchlorate as counterion is only 1.6 indicating a very low degree of polymerization. In the presence of tosylate, 2.5 exchanged electrons per 2,2′-bis-EDOT are observed, indicating a higher degree of polymerization.9 Danielson et al. were able to electrochemically polymerize EDOT in a ionic liquid consisting of octyl sulfate as the anion and 1-butyl-3-methylimidazolium as the cation.16 When diethyleneglycol monomethylether sulfate was used as anion, the polymerization only proceeded in the presence of additional water. Sakmeche et al. used combinations of anions such as dodecyl sulfate and perchlorate.17 In this case the deposition occurred for an aqueous micellar medium due to the presence of the dodecylsulfate ion. As a consequence,
6911X.indb 85
10/1/10 8:59:29 PM
86
PEDOT
the solubility of EDOT is increased and the oxidation potential is reduced, compared to solutions in acetonitrile. Adamczyk and Kulesza polymerized EDOT in the presence of phosphodecamolybdate and silicododekamolybdate as bulky counterions as potential anti-corrosion layers.18 David et al. used cobaltabisdicarbollide as a highly voluminous and redox-active counterion in electrochemical polymerization.19
7.2 Counterions in Chemically Polymerized PEDOT The number of PEDOT:counterion complexes generated by chemical poly merization is smaller than that for electrochemical polymerization. This is due to the fact that the oxidation agent itself is typically a salt and its anion becomes the counterion for the PEDOT complex. If additional ions are added to the reaction mixture, mixed complexes are generated. The original work of Jonas and Heywang points to some of the most important oxidation agents.20 The authors used iron(III) chloride, iron(III) tosylate, or ammoniumperoxodisulfate resulting in chloride, tetrachloroferrate, tosylate, or sulfate as counterions. Iron(III) salts have proven to be particularly useful oxidation agents for PEDOT.21,22 Iron(III) tosylate is the most prominent case, since this salt is well soluble in organic solvents such as alcohols and readily available. Mixtures of iron(III) tosylate solutions with EDOT are easily prepared and result in a well controlled deposition of conductive films (see Chapter 8). Furthermore, the tosylate ion has proven to promote the chemical reaction of EDOT as shown in the previous section. PEDOT:tosylate films prepared by chemical polymerization show conductivities as high as 1000 S/cm.23 Aasmundtveit et al. reported on the x-ray diffraction of PEDOT:tosylate films, which showed paracrystalline regions in the range of 50 Å.24 The crystalline order increased with increasing temperature at which the film was dried. The authors proposed a model structure with an orthorhombic unit cell containing four thiophene units and one tosylate ion similar to those obtained with perchlorate or PF6 – in the electrochemically polymerized PEDOT. The similarity between the x-ray diffraction results for the different counterions in different polymerization techniques indicates that the counterion has only a limited influence on the structure, as long as it is small enough to fit between separate PEDOT stacks. Kirchmeyer and Jonas have reported the use of iron(III) camphor sulfonate, iron(III) phenol sulfonate, and iron(III) methanesulfonate as oxidation agents.25 The use of camphor sulfonate resulted in highly conductive films with a specific conductivity of more than 1000 S/cm. Kudoh et al. have polymerized EDOT in the presence of sodium alkyl naphthalenesulfonate with different alkyl chain lengths and an average
6911X.indb 86
10/1/10 8:59:29 PM
Counterions for PEDOT
87
molecular weight of 338 g/mol using iron(III) sulfate as the oxidation agent.26 The resulting emulsion polymerization yielded conductive films with alkyl naphthalenesulfonate as counterion and conductivities of 10 to 100 S/cm. Furthermore, the authors used iron(III) alkylnaphthalenesulfonate as the oxidation agent. The resulting films showed lower conductivities than those obtained with iron(III) sulfate and sodium alkylnaphthalenesulfonate as additives. The use of peroxodisulfate as oxidation agent results in films with sulfate as counterion as reported in the original work by Jonas and Heywang.20 The use of sulfate as counterion is limited as Kirchmeyer and Reuter later showed;27 at lower concentrations of EDOT sulfate is insufficient as the counterion for the polymerization to proceed and additional counterions such as PSS are required. Polystyrenesulfonic acid (PSS) is the most prominent polymeric counterion for PEDOT, and the complex of PEDOT:PSS is described in detail in Chapter 9. As shown in the early work of Jonas and Krafft on PEDOT dispersions, PEDOT:PSS is, strictly speaking, a complex with two types of counterions.28 The complex contains sulfate ions as obtained from the oxidation agents iron(III) sulfate and potassium peroxodisulfate as well as polystyrenesulfonic acid. However, the sulfate ion can easily be removed using, for example, ion exchange resins and a true PEDOT:PSS complex is prepared.29 Dai et al. have described the preparation of PEDOT complexes with poly[2-(3-thienyl)-ethoxy-4-butylsulfonate].30 They obtained the complex in the form of spherical particles with an average size in the range of 100 nm. They were also able to prepare films with conductivities in the range of 0.2 to 4 S/cm. Pickup has shown that PEDOT can be polymerized in the presence of sulfonated tetrafluoroethylene based polymers such as Nafion®, resulting in stable dispersion with a conductivity of 0.01 S/cm.31 Furthermore, PEDOT has reportedly been polymerized on top of surfaces—including nanoparticles— typically with a sulfonic acid functionality.31,32, 33
References
6911X.indb 87
1. L. Groenendaal, G. Zotti, P.-H. Aubert, S. M. Waybbright, and J. R. Reynolds. 2003. Electrochemistry of Poly(3,4-alkylenedioxythiophene) Derivatives. Adv Mater 15 (11):855–879. 2. J. Xia, N. Masaki, M. Lira-Cantu, Y. Kim, K. Jiang, and S. Yanagida. 2008. Effect of doping anions’ structures on poly(3,4-ethylenedioxythiophene) as hole conductors in solid-state dye-sensitized solar cells. J Phys Chem C 112:11569–11574. 3. J. Xia, N. Masaki, M. Lira-Cantu, Y. Kim, K. Jiang, and S. Yanagida. 2008. Poly(3,4-ethylenedioxythiophene) as hole conductors for iodine-free solid state dye-sensitized solar cells. J Am Chem Soc 130(4):1258–1263.
10/1/10 8:59:29 PM
88
6911X.indb 88
PEDOT
4. I. Giurgiu, K. Zong, J. R. Reynolds, W.-P. Lee, K. R. Brenneman, A. V. Saprigin, A. J. Epstein, J. Hwang, and D. B. Tanner. 2001. Dioxypyrrole and dioxythiophene based conducting polymers: Properties and applications. Synth Met 119(1–3):405–406. 5. A. Aleshin, R. Kiebooms, R. Menon, and A. J. Heeger. 1997. Electronic transport in doped poly(3,4-ethylenedioxythiophene) near the metal-insulator transition. Synth Met 90(1):61–68. 6. L. Groenendaal, G. Zotti, and F. Jonas. 2001. Optical, conductive and magnetic properties of electrochemically prepared alkylated poly(3,4-ethylenedioxythio phene)s. Synth Met 118(1–3):105–109. 7. T. El Moustafid, R. V. Gregory, K. R. Brenneman, and P. M. Lessner. 2002. Influence of the anion of the supporting electrolyte on the formation and the electrochemical properties of poly(3,4-ethylenedioxythiophene) films. Polymer Preprints 43(2):1320. 8. M. Granström and O. Inganäs. 1995. Electrically conductive polymer fibres with mesoscopic diameters: 1. Studies of structure and electrical properties. Polymer 36(15):2867–2872. 9. G. Zotti, S. Zecchin, G. Schiavon, F. Louwet, L. Groenendaal, X. Crispin, W. Osikowicz, W. Salaneck, and M. Fahlman. 2003. Electrochemical and XPS studies toward the role of monomeric and polymeric sulfonate counterions in the synthesis, composition, and properties of poly(3,4-ethylenedioxythiophene). Macromolecules 36(9):3337–3344. 10. H. Yamato, M. Ohwa, and W. Wernet. 1995. Stability of polypyrrole and poly(3,4-ethylenedioxythiophene) for biosensor application. J Electroanal Chem 397(1–2):163–170. 11. H. Yamato, K. Kai, M. Ohwa, W. Wernet, and M. Matsumura. 1997. Mechanical, electrochemical and optical properties of poly(3,4-ethylenedioxythiophene)/ sulfonated poly(β-hydroxyethers) composite films. Electrochim Acta 42(16): 2517–2523. 12. W. Wernet and G. Khan. EP 0658906 (Ciba-Geigy AG), Prior: December 18, 1993. 13. G. Sonmez, P. Schottland and J. R. Reynolds. 2005. PEDOT/PAMPS: An electrically conductive polymer composite with electrochromic and cation exchange properties. Synth Met 155(1):130–137 14. L. Niu, C. Kvarnström, K. Fröberg, and A. Ivaska. 2001. Electrochemically controlled surface morphology and crystallinity in poly(3,4-ethylenedioxythio phene) films. Synth Met (2)122:425–429. 15. R. H. Erlich and A. I. Popov. 1971. Spectroscopic studies of ionic solvation. X. Study of the solvation of sodium ions in nonaqueous solvents by sodium-23 nuclear magnetic resonance. J Am Chem Soc 93(22):5620–5623. 16. P. Danielsson, J. Bobacka, and A. Ivaska. 2004. Electrochemical synthesis and characterization of poly(3,4-ethylenedioxythiophene) in ionic liquids with bulky organic anions. J Solid State Electrochem 8(10):809–817. 17. N. Sakmeche, J. J. Aaron, M. Fall, S. Aeiyach, M. Jouini, J. C. Lacroix, and P. C. Lacaze. 1996. Anionic micelles; a new aqueous medium for electropolymerization of poly(3,4-ethylenedioxythiophene) films on Pt electrodes. Chem Comm 24:2723–2724. 18. L. Adamczyk and P. J. Kulesza. 2008. Preparation and protective properties of composite films of poly(3,4-ethylenedioxythiophene) with heteropolyanions on stainless steel. ECS Transactions 13(27):85–93.
10/1/10 8:59:29 PM
Counterions for PEDOT
89
19. V. David, C. Vinas, and F. Teixidor. 2006. Poly(3,4-ethylenedioxythiophene) doped with a non-extrudable metallacarborane anion electroactive during synthesis. Polymer 47(13):4694–4702. 20. F. Jonas and G. Heywang. DE 3813589 (Bayer AG), Prior: April 22, 1988. 21. F. Jonas and L. Schrader. 1991. Conductive modifications of polymers with polypyrroles and polythiophenes. Synth Met 41–43(3):831–836. 22. L. Groenendaal, F. Jonas, D. Freitag, H. Pielartzik, and J. R. Reynolds. 2000. Poly(3,4-ethylenedioxythiophene) and its derivatives: Past, present, and future. Adv Mater 12(7):481–494. 23. B. Winther-Jensen, D. W. Breibym, and K. West. 2005. Base inhibited oxidative polymerization of 3,4-ethylenedioxythiophene with iron(III) tosylate. Synth Met 152(1–3):1–4. 24. K. E. Aasmundtveit, E. J. Samuelsen, L. A. A. Petterson, O. Inganäs, T. Johansson, and R. Feidenhans’l. 1999. Structure of thin films of poly(3,4-ethylenedioxythio phene). Synth Met 101(1–3):561–564. 25. S. Kirchmeyer and F. Jonas. EP 1338617 (Bayer AG), Prior: February 15, 2002. 26. Y. Kudoh, K. Akami, and Y. Matsuya. 1998. Chemical polymerization of 3,4ethylenedioxythiophene using an aqueous medium containing an anionic surfactant. Synth Met 98(1):65–70. 27. S. Kirchmeyer and K. Reuter. 2005. Scientific importance, properties and growing applications of poly(3,4-ethylenedioxythiophene). J Mater Chem 15(21):2077–2088. 28. F. Jonas and W. Krafft. EP 440957 (Bayer AG), Prior: February 8, 1990. 29. F. Jonas, K. Lerch, and W. Fischer. 1996. EP 752454 (Bayer AG), Prior: July 3, 1995. 30. C.-A. Dai, C. J. Chang, H.-Y. Chi, H.-T. Chien, W.-F. Su, and W.-Y. Chiu. 2008. Emulsion synthesis of nanoparticles containing PEDOT using conducting polymeric surfactant: Synergy for colloid stability and intercalation doping. J Polym Sci A: Polym Chem 46(7):2536–2548. 31. P. G. Pickup, C. L. Kean, M. C. Lefebvre, G. Li, Z. Qi, and J. Shan. 2000. Electronically conducting cation-exchange polymer powders: Synthesis, characterization and applications in PEM fuel cells and supercapacitors. J New Mat Electrochem Systems 3(1):21–26. 32. L. Puppe, W. Loevenich, A. Elschner, and S. Kirchmeyer. WO 2009/043648 (H.C. Starck GmbH), Prior: September 28, 2007. 33. M. G. Han and S. H. Fougler. 2004. Preparation of poly(3,4-ethylenedioxythio phene) (PEDOT) coated silica core-shell particles and PEDOT hollow particles. Chem Comm 24:2514–2155.
6911X.indb 89
10/1/10 8:59:29 PM
6911X.indb 90
10/1/10 8:59:29 PM
8 The In Situ Polymerization of EDOT to PEDOT
8.1 Synthesis of In Situ PEDOT In situ poly(3,4-ethylenedioxythiophene) (PEDOT) was the first example for PEDOT as synthesized under film-forming conditions.1 Up to now in situ PEDOT is the polythiophene with the highest achievable electric conductivity, although PEDOT:PSS (poly(styrenesulfonate)) is catching up. That is why in situ PEDOT is of great practical and commercial value. These industrial aspects are described in Chapters 10 and 11. This chapter will focus on chemical issues, including some practical advice for performing oxidative in situ polymerization. There are several investigations of PEDOT covered by the term in situ in the literature, which deal with electrochemically synthesized PEDOT layers; these studies are discussed in Chapter 14. Here the focus is on the preparation of doped PEDOT films from EDOT solutions with a chemical oxidant. The oxidative in situ polymerization is best carried out with ionic oxidants like iron-III manganese-IV, or similar metal ions in a suitable higher oxidation state. There are various other metal salt oxidants and also completely distinct compounds, for example, peroxides, suggested for oxidative poly merization in the literature or claimed in patents. These alternatives have been discussed in more detail in Chapter 6. Iron-III is the oxidant preferred in most experiments in literature and also by far predominant in industrial use. The solubility requirements for iron-III oxidants are determined by the limited solubility of EDOT in water in contrast to its miscibility with alcohols like ethanol or n-butanol in any ratio. So the well alcohol-soluble iron salts of sulfonic acids are preferred oxidants. Especially p-toluenesulfonate has been established as a very suitable anion with respect to solubility and reactivity of the corresponding iron-III salt. Iron-III toluenesulfonate has become the most widely used oxidant for EDOT in the preparation of in situ PEDOT layers, scientifically and commercially. The overall reaction is described in Figure 8.1. To get a clearer picture, the equation is drawn for the EDOT hexamer, typically representing every 91
6911X.indb 91
10/1/10 8:59:30 PM
92
PEDOT
O
O
6
+
– 10 TosH
12 FeTos3
– 12 FeTos2
S Tos + H
– O
O
S
O
S S
O
O
O
S S
O
O
O
O
+
Tos –
S O
H
O
Figure 8.1 In situ PEDOT synthesis.
third thiophene ring as a cationic moiety (of course, this is only one mesomeric structure). The overall polymerization reaction can be separated into two steps. First, the oxidative polymerization of EDOT to the neutral, undoped polythiophene occurs. p-Toluenesulfonic acid and iron-II p-toluenesulfonate are formed as byproducts stoichiometrically. Therefore two mole equivalents of Fe-III tosylate are necessary to perform this reaction. In the second step, the neutral PEDOT is doped by the action of “excess” Fe-III tosylate. As every third or fourth thiophene moiety loses one electron to form the cationic structure, about 0.25 to 0.33 additional equivalents of iron(III) toluenesulfonate are needed for efficient doping. Here also the corresponding stoichiometric amount of Fe-II tosylate is formed. The third tosylate anion is incorporated into the polymer structure as the counterion. As the result, the in situ PEDOT (more exactly, PEDOT:pTs) layers formed by this two-step reaction are initially containing about 2.3 moles of iron-II tosylate and 2.0 moles of free toluenesulfonic acid. Both can be removed by rinsing with water, for example. More detailed investigations reveal not only the two-step character of this reaction, but also an even more complex reaction mechanism. The neutral PEDOT can be isolated in traces to moderate amounts. For example, using iron-III chloride instead of the Fe tosylate, which is not suitable for technical applications due to the formation of corrosive HCl and of deeply colored tetrachloroferrate counterions, neutral PEDOT can be synthesized in moderate yields.2 The prerequisites necessary for cancelling the reaction before it comes to complete doping are discussed in more detail in Chapter 6. Kinetic parameters obtained in a study carried out at Bayer AG explain the polymerization rate and stoichiometric effects when oxidizing EDOT (see Figure 8.2).3 The first step in this reaction is the EDOT oxidation to the corresponding radical cation. This is the slowest rate-determining step with a rate constant k1 = 0.16 L3mol–3h–1 (all kinetic parameters given in the following
6911X.indb 92
10/1/10 8:59:30 PM
The In Situ Polymerization of EDOT to PEDOT
O
O 2 Fe3+ O
O HC
CH
S
2
k1
2
+
HC
k2
CH
S
O
O O
O
2 H+
+
k3
O
O
H+
HC
HC
k4
S
2Fe3+
C
S
O
O
S
CH
CH2 C H S
O
O
S +
CH 2 Fe3+
HC
O
O
k5
O
O
C
+
CH
S
HC
+ S
C
k2
HC
O
O
S
S
O
O
S
n
CH
+ 2H+
O
O
S
O
O
S S O
O
+ Fe(III)(p-Tos)3
O
S O
O
O S
+
n
O
O S
S
S
S O
+
O
O
O
O
S
O
O
93
O
S O
O– O S O
O
n
O
O
+ Fe(II)(p-Tos)2
Figure 8.2 Proposed reaction mechanism for the oxidation of EDOT to PEDOT:pTS.
description have been determined at 30°C). The free radical cation then very rapidly dimerizes with k2 = 10 8–10 9 Lmol–1h–1 (this parameter is found in literature4 in accordance to the value for a terthiophene given there and, due to its magnitude, does not significantly influence the overall reaction rate). End group oxidation of EDOT oligomers follows, starting with dimer oxidation, which is remarkably faster than monomer oxidation (k5 = 3 × 103 Lmol–1h–1; the same value as proposed for all chain lengths). Recombination of two radical cationic end groups results in oligomer formation with the same rate constant for recombination as for the monomeric cations (k2 = 109 Lmol–1h–1). Finally, oligomers or polymers are doped by further oxidation. Figure 8.2 shows the paramagnetic polaron state as the first step of doping. This paramagnetic intermediate step was proved to be plausible by electron spin resonance studies in electrochemical investigations (more details are outlined in Chapter 14).5–7 The polaron state is then further oxidized to the extremely stable, highly conductive and diamagnetic bipolaron state of PEDOT.
6911X.indb 93
10/1/10 8:59:31 PM
94
PEDOT
O
O
O S
O
S
S
S
S O
EDOT-Dimer
O
O
O
O
O
EDOT-Trimer
Figure 8.3 EDOT dimer and trimer.
The intermediate formation of dimeric and trimeric EDOT (see Figure 8.3, formulae without stereochemical aspects) is a slow equilibrium reaction,8 but does not influence the overall reaction rate. For the sake of clarity, in Figure 8.2 the trimer formation has been omitted. This reaction is a side reaction, so at the end of the overall reaction, the main yield of PEDOT molecules has been formed via EDOT radical cation dimerization and a small part via dimer formation. The parameters were measured independently by experiments with toluenesulfonic acid and EDOT and the EDOT dimer, respectively. The rate constant k3 for the dimer formation was found to be 1.5 × 10 –3 L2mol–2h–1; the value for the back reaction (k-3) is 1.2 × 10 –2 Lmol–1h–1. The oxidation rate for the dimer to BEDOT was determined to 20 Lmol–1h–1. Here also, the trimer formation has not been taken into account; the overall reaction is not influenced much by the additional trimer formation. Obviously, all reaction steps in this route to PEDOT are significantly slower than via EDOT radical cation. For a more detailed discussion of EDOT dimer and trimer formation, see Chapter 5. The EDOT dimerization was discussed as the reason for a remarkable catalytic effect of protons on the EDOT oxidation to PEDOT. Addition of p-toluenesulfonic acid clearly accelerates the overall reaction, but, due to the aforementioned facts, this cannot be a result of the increased dimer (and trimer) formation effected thereby. The best fit of the experimental results with a rate equation including the accelerating influence of protons could be achieved with the following equation for the rate-determining first step, the oxidation of EDOT to the EDOT+. radical cation:
r1 = k1 c2EDOT c2FeIII + k11 cH + cEDOT cFeIII,
where r is the reaction rate, k is the rate constant, and c is the concentration. Although this formal kinetic approach does not exactly correspond to the reaction scheme of Figure 8.2, it is in complete accordance to all experimental data. k11 has been estimated to 0.026 L2mol–1h–1.
6911X.indb 94
10/1/10 8:59:32 PM
The In Situ Polymerization of EDOT to PEDOT
95
A rather analog adjustment to represent the accelerating influence of protons can be made for the oxidation of oligomer end groups:
r5,i = k5 cpolymer(i) cFeIII + k51 cH+ cpolymer(i) cFeIII
,
where i is the degree of polymerization. k51 could be determined to 105 L2mol–2h–1. The activation energy of the rate determining step was estimated to 67 ± 5 kJ/mol derived from an Arrhenius plot of the rate constants k1 at five different temperatures between 10°C and 50°C. Just when the in situ polymerization of EDOT to PEDOT was established, most of the pivotal requirements for the preparation of highly conductive layers already were known to the PEDOT inventors and recorded in several patent examples.1 Iron(III)-toluenesulfonate as an optimal oxidant and lower alcohols (up to butanol) as suitable solvents remained typical components in practically used recipes and formulations. Also early investigations on in situ polymerized PEDOT in the scientific literature9–11 did not broadly deviate from the principles disclosed by Jonas, Heywang, et al.1 This did not change over the years, but several modifications and improvements have been made, for example, the addition of imidazole to the EDOT/iron(III)tosylate formulation.9,12 A typical recipe for the in situ polymerization is given here in detail.13
6911X.indb 95
1. Chemicals—For the in situ PEDOT layer, the following chemicals are needed: EDOT; iron(III)-tosylate in n-butanol, typically as a 40% by weight solution; and imidazole. For the deposition of an adhesion layer on the substrate under the in situ PEDOT layer is needed: PEDOT:PSS, typically as a 1.3% dispersion in water; or γ-glycidoxy propyltrimethoxysilane. 2. Deposition of an adhesion layer on glass substrates—The adhesion layer is recommended, especially for the deposition on glass. Without adhesion layer the in situ PEDOT layers may peel off most probably during the washing step. The adhesion layer should have a thickness of 10 to 20 nm. a. PEDOT:PSS as the adhesion layer—PEDOT:PSS is spin-coated as a 10 to 20 nm thick film onto a glass substrate and subsequently baked at 200°C for about 10 min on a hot plate. b. γ-Glycidoxypropyltrimethoxysilane as the adhesion layer—γ-Glyci doxypropyltrimethoxysilane is spin-coated onto a glass substrate (1000 rpm, t = 120 sec, Acc. 2, lid closed), or the substrate simply dipped in diluted epoxysilane solution (0.1 % in isopropanol) and subsequently dried at 50°C for 5 min in air. Humidity has an impact on the cross-linking of the epoxysilane. Medium humidity of 30% to 70% is preferred.
10/1/10 8:59:32 PM
96
PEDOT
3. Mixing the solution—EDOT:iron(III)-tosylate solution (40%):imidazole are mixed in a weight-ratio of 1:25:0.5. In a first step imidazole is dissolved in the iron(III)-tosylate solution. In a second step EDOT is added. The mixture can be filtered by using a syringe filter (0.45 µm PVDF). 4. Coating the solution—The solution has to be processed immediately after the addition of EDOT. After 5 to 10 min the solution starts to gel due to polymerization, and the films will become inhomogeneous. The solution can be spin coated at spin speeds of 200 to 3000 rpm and thin homogeneous films are formed. On a 50 mm × 50 mm large substrate the film thickness will be 170 nm at 1000 rpm, t = 30 sec, Acc. 2, lid closed. 5. Drying the film—The first step is 5 min at 50°C on a hot plate in air. The second step is 60 min at room temperature in air. The air humidity level is 30% to 70%. The film changes color as EDT polymerizes while drying. 6. Washing—The films are washed by shaking the substrate in a beaker filled with distilled H2O thoroughly to remove Fe salts. Adhering water is blown off in a N2 flow. 7. Conductivity—The conductivity can be easily checked by printing two parallel conducting bars (silver paste) on top of the layer and by using an Ohm meter to determine the resistivity. The conductivity will typically be in the range of 400 to 600 S/cm.
Another recipe from in the literature shall be given here in a shortened form14 to give an impression of the substantial similarity of practically all sufficient recipes with small differences to obtain particular effects: A solution of 6.5 ml iron(III)-tosylate (40% in n-butanol), 2.0 ml water, 0.22 ml EDOT, and 0.15 ml pyridine was applied by spin coating (250 to 4000 rpm for 30 s) on to a poly(methyl methacrylate), or PMMA, sheet. Then the PMMA sheet was baked for 10 min at 65°C. Afterward, the PEDOT layers were washed by applying 3 ml of a 1:1 mixture of n-butanol and anisole while spinning at 1000 rpm for 20 s, and the washed substrates were further dried for 5 min at 65°C. In this recipe, water was added to prolong the pot life of the solution and to decrease the viscosity. The butanol/anisole mixture as a less convenient washing fluid was utilized to integrate parts of the PEDOT layer into the plastic substrate (after this procedure the top part of the network consisted of 30 to 40 vol.% PEDOT and 60 to 70 vol.% PMMA). As a result, very good mechanical properties and a particular, surprisingly high conductivity of about 1000 S/cm were found.14 Alternative cerium salt oxidants with limited suitability have been discussed.15 Whereas cerium-IV sulfate Ce(SO4)2 resulted in reasonable conductivities, but Ce-contaminated PEDOT, (NH4)2Ce(NO3)6 yielded PEDOT with only low conductivity. So cerium-based oxidants do not represent advantages compared to iron-III.
6911X.indb 96
10/1/10 8:59:32 PM
The In Situ Polymerization of EDOT to PEDOT
I
OH O
SO2
Hydroxy-tosyloxy iodobenzene "Koser's Reagent"
97
I
O CO –CF3 O
CO–CF3
Bis(trifluoroacetoxy)iodobenzene
Figure 8.4 Organic hypervalent iodine compounds for EDOT oxidation. (Data from K. Reuter and S. Kirchmeyer, WO 2007 085371, Prior: January 20, 2006.8)
Instead of metal salts, peroxides and other more sophisticated oxidants were suggested in several publications and patent applications. Rather closely related to the preferred method for the manufacture of PEDOT:PSS— the oxidative polymerization with peroxodisulfates—in situ polymerization by peroxidic compounds has been claimed by several patent applications.16,17 Another patent application utilizes so-called hypervalent iodine compounds as oxidants.18 Examples for inorganic hypervalent iodine compounds are iodic acid, sodium iodate (iodine-V), and sodium periodate (iodine-VII). Typical organic hypervalent iodine compounds19 were of the iodine-III type, for instance, Koser’s reagent = hydroxy-tosyloxy iodobenzene or bis(trifluoroacetoxy) iodobenzene (see Figure 8.4). Several interesting properties may be a motivation for the use of hypervalent iodine compounds. For example, EDOT derivatives with thioether function like 3,4-ethyleneoxythiathiophene cannot be oxidatively polymerized with peroxides due to the oxidation of the thioether function to a sulfone group. Use of periodate excludes this reaction and facilitates oxidative polymerization. Nevertheless, achievable conductivities are only very moderate, and so the practical value is limited.
8.2 Properties of In Situ PEDOT Especially because of its high conductivity combined with its very good overall performance, large industrial use of in situ PEDOT in its widest sense has been established. This is discussed in later chapters. As pointed out in the previous section, with standard recipes13 conductive layers with around 500 S/cm are easily achievable. There are several publications and patents demonstrating the great potential of in situ PEDOT to achieve four-digit values under special conditions. For example, 1000 to 1200 S/cm have been disclosed in a patent application, where iron(III) camphorsulfonate was used as the oxidant instead of the Fe(III) tosylate.20 Nearly 1200 S/cm has also been published by Levermore et al. as an upper value.21
6911X.indb 97
10/1/10 8:59:32 PM
98
PEDOT
Here this extreme conductivity has been achieved by depositing the PEDOT in a vapor phase process (VPP PEDOT) with EDOT vapor on to a spin-coated active layer of iron(III) tosylate in a vacuum chamber at about 10 mbar. In another publication a four-digit number (1025 S/cm) for the conductivity of a vapor phase deposited in situ PEDOT was reported by Winther-Jensen and West.22 The transmission spectrum of thin in situ PEDOT films is displayed in Figure 8.5a. The films were prepared on glass substrates according to the preparation guide outlined in Section 8.1. In one of the last processing steps the films are rinsed with H2O. The absorption of the film will increase immediately leading to a significant color change. The rinsing step is necessary to remove residues of the oxidant, mainly iron ions, and will cause simultaneous the shrinkage of the film. Most probably this morphological reorganization triggers final polymerization reactions of so far sterically hindered components. In Figure 8.5b the normalized transmission of in situ PEDOT films is shown as a function of layer thickness. Owing to the higher absorption in the red part of the spectrum, the films exhibit a blue appearance. With increasing layer thickness the absorption of the films increases too while simultaneously the sheet resistance decreases. The relation between layer thickness, transmission, and sheet resistance is discussed in detail in Chapter 10. An important parameter that determines the conductivity of in situ PEDOT films is the humidity during processing and drying (or, optionally, baking). In the very early days of PEDOT in the 1980s, when PEDOT was established on a laboratory scale as a humidity-independent conductive material, this was a little bit surprising. Of course, PEDOT was far more conductive than typical humidity-dependent antistatics like poly(styrenesulfonic acid), and, after preparation, the conductivity was stable against varying ambient conditions during the measurement. But in situ layers of PEDOT prepared in winter had another surface resistance than their summer analogs; a factor of two or more between both seasonal types was observed. So, standard conditions are necessary to get comparable results. A typical picture (see Figure 8.6), although obtained in vapor phase experiments, by Zuber et al., clearly demonstrates the humidity dependence of in situ PEDOT (to emphasize again: humidity not during measurement, but during preparation of the film).23 This phenomenon seems to correspond to the fact that water in the monomer/oxidant solutions prolongs the pot life of these mixtures by reducing their reactivity. Regarding their visual appearance, in situ PEDOT:Tos layers do not differ very much from PEDOT:PSS films or electrochemically prepared PEDOT films with other counterions. Thin films are light blue, transparent and visually homogeneous. The transparency clearly follows Beer’s law.24 A very detailed investigation regarding the influence of a vast amount of parameters onto the properties of in situ PEDOT has been published 2004.24
6911X.indb 98
10/1/10 8:59:32 PM
The In Situ Polymerization of EDOT to PEDOT
99
80 As prepared After H2O-purge
70
Transmission [%]
60 50 40 30 20 10 0
500
1000
1500
2000
2500
Wavelength [nm] (a) Energy (eV)
Internal Transmission T/T0 [%]
4 3.6
3.2
2.8
2
2.4
1.6
80 60 40 20 0 300
50 nm 90 nm 165 nm 400
500
600
700
800
Wavelength [nm] (b) Figure 8.5 (a) Transmission spectra of in situ PEDOT films on glass substrates before and after being purged with H2O. (b) Film transmission of in situ PEDOT normalized to the substrate for different layer thicknesses.
6911X.indb 99
10/1/10 8:59:33 PM
100
PEDOT
Conductivity (S cm–1)
900 800 700 600 500 400 300
0
20
40
60
80
%RH Figure 8.6 Humidity dependence of in situ PEDOT conductivity. (From K. Zuber, M. Fabretto, C. Hall, and P. Murphy, 2008, Improved PEDOT Conductivity via Suppression of Crystalline Formation in Fe(III) Tosylate during Vapor Phase Polymerization, Macromol Rapid Commun 29(18):1503–1508. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
The reaction mechanism depicted Ha et al. is in accordance to our scheme in Figure 8.2,24 without mentioning the proton-catalyzed side reaction to the EDOT dimer (and trimer). The authors explain the beneficial action of imidazole by attenuation of the Fe(Tos)3 reactivity, and coordination of imidazole to the Fe is supposed as one of the reasons for this. This may be a matter of discussion: other good complexing agents like acetylacetone or acetoacetates, which do not have the typical nitrogen basicity of imidazole (or pyridin), do not work similarly.25 The perception of de Leeuw et al., regarding the pH change as the main function of imidazole,9 is supported by the kinetic investigations described in Section 8.1, especially by the observation that the reaction rate is significantly influenced by the proton concentration, expressed by the equation r1 = k1 c2EDOT c2FeIII + k11 cH+ cEDOT cFeIII. Additionally, the disadvantageous formation of EDOT dimer and trimer is suppressed, and the base-assisted deprotonation of the intermediate radical dication is supported. Independent from this discussion, the positive action of imidazole onto the conductivity of in situ PEDOT (PEDOT:Tos) is well documented, and Ha et al.,24 achieve a specific conductivity of 750 S/cm by optimizing the imidazole–Fe–EDOT stoichiometry (2:1.75:1) and changing the solvent to n-pentanol. The observed values of 300 Ω/square for the surface resistance combined with a transparency of 84% are in good accordance to the data presented in Chapter 10. In situ PEDOT:Tos layers tend to be mechanically brittle. Further improvement of film formation and lowering of brittleness are achievable by the addition of sufficient alcohol soluble polymers, like poly(vinylacetate),1 which function as a plasticizer. A lot of other polymers have been suggested for this purpose.1 Polymer components in PEDOT films are not inevitably detrimental for their conductivity; see also, for example, the findings for the mechanically improved PEDOT/PMMA by Hansen et al.14
6911X.indb 100
10/1/10 8:59:34 PM
The In Situ Polymerization of EDOT to PEDOT
101
Intensity (arb. unit)
UPS VEH
30
25 20 15 10 Binding Energy (eV)
5
Figure 8.7 UPS spectrum of PEDOT:Tos (top) and corresponding VEH DOVS curve. (Reprinted from K. Z. Xing, M. Fahlmen, X. W. Chen, O. Inganäs, and W. R. Salaneck, Synth Met 89(3):161–165, The Electronic Structure of Poly(3,4-ethylene-dioxythiophene): Studied by XPS and UPS. Copyright 1997, with permission from Elsevier.)
The electronic structure of in situ PEDOT:Tos has been investigated by Inganäs, Salaneck, and their group.11 They used x-ray photoelectron spectroscopy (XPS) and UV photoelectron spectroscopy (UPS) and compared the data to quantum chemical calculations (AM1, MNDO, and VEH methods). In Figure 8.7, the UPS spectrum of in situ PEDOT:Tos compared to the VEH (valence effective Hamiltonian) DOVS (density of valence electronic states) curve is shown.11 No temperature-dependent effects were found in the range from liquid nitrogen to 250°C in the UPS spectrum, indicating the enormous stability of the polymer at elevated temperatures. The work function of PEDOT:Tos (on the basis of 20% doping, which corresponds to one tosylate counterion per five thiophene rings) was determined to 4.4 ± 0.2 eV.11 In situ PEDOT has been suggested as glucose sensor material. EDOT has been polymerized inside the pores of a track-etch membrane by the action of iron(III) chloride in the presence of the positively charged poly(N-methyl-4vinylpyridin).26,27 The in situ PEDOT with the resulting, very special properties regarding an efficient electron transfer ability and optimal morphology could be utilized as an amperometric glucose sensor.26 Figure 8.8 demonstrates the proposed working mechanism of the PEDOT-based glucose sensor, working with the aid of the enzyme glucose oxidase (GOx).26,27 GOx is immobilized inside the pores of the track-etch membrane in close proximity to the conducting polymer.
6911X.indb 101
10/1/10 8:59:34 PM
102
PEDOT
e–
+ PEDOTox –PEDOTred Glucose
+ GOxox
–GOxred
Gluconolactone
Figure 8.8 Proposed working mechanism of the PEDOT-based glucose sensor. (Adapted from A. Kros et al., 2001, Adv Mater 13(20):1555–1557; A. Kros, R. J. M. Nolte, and N. A. J. M. Sommerdijk, 2002, Adv Mater 14(23):1779–1782.)
8.3 In Situ Polymerization of EDOT Derivatives and Relatives The oxidative polymerization with the aid of persulfate in water is limited to EDOT and a few derivatives, which are water-soluble to a certain extent (for example, EDOT-CH3, EDOT-CH2OH, sulfonate-modified EDOT or oligoethylenglycol-substituted EDOT). In contrast, the in situ polymerization is of far broader range with respect to the EDOT derivatives applicable. The minimum prerequisite is a moderate to good solubility in alcohols like ethanol or n-butanol at slightly elevated or room temperature. Another important aspect is the reactivity of the monomer, which must not be too high with iron-III toluenesulfonates to avoid instantaneous polymerization. A lot of substituted derivatives behave very similar to EDOT during in situ polymerization (see Figure 8.9). Alkyl-EDOTs (R = alkyl, R′ = H), EDOTCH2OH (R = CH2OH), and the corresponding alkylethers (R = CH2O-alkyl) or alkylurethanes (R = CH2O−OC−NH−alkyl), and also several dialkyl-EDOTs (R = R′ = alkyl) can easily be in situ polymerized with Fe-III-tosylate. Lower conductivities than for the unsubstituted in situ PEDOT are resulting in most cases. There are virtually no comprehensive investigations in the literature to learn more about conductivities of the polymers from these R´
R
O
O
S Figure 8.9 EDOT derivatives for in situ polymerization (for a detailed explanation of R and R′, see Table 8.1 and Figure 8.11).
6911X.indb 102
10/1/10 8:59:35 PM
The In Situ Polymerization of EDOT to PEDOT
103
Table 8.1 In Situ Polymerized EDOT Alkyl Derivatives (Corresponds with Figure 8.10) Entry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
R H CH3 CH3 CH3 n–C3H7 n–C6H13 n–C8H17 n–C8H17 n–C10H21 n–C14H29 CH3 n–C6H13 n–C6H13
R′
Surface Resistance (Ohm/sq)
H H H H H H H H H H CH3 n–C6H13 n–C6H13
ca. 100 90 100 100 150 160 100 290 310 450 990 10000 32000 160 160
R + R′ = –(CH2)4– R + R′ = –(CH2)4–
Remarks
Reference
Comparison R,S-mixture S-isomer R-isomer
32 32 32
R,S-mixture S-isomer
32 32
R,S/S,R-mixture S,S-isomer
32 32
S,S-isomer R,R-isomer
32 32
molecules, with one considerable exception to be discussed later: EDOTCH2OH.24 Therefore, scientifically unobjectionable, truly comparable values are essentially not available. In the following, some exploratory experiments will be presented, which give basic information regarding trends in the conductivity of in situ PEDOT derivatives. The significance of these results is certainly acceptable; nevertheless all values should not be regarded as being set in stone. The surface resistance for in situ PEDOT, obtained under the same conditions and as comparative test, is incorporated. Since the in situ PEDOT conductivity can depend of several external factors, including humidity during film preparation, it is difficult to compare results obtained in differently conducted experiments. As a consequence, results should only be compared within one series as done in Table 8.1 but not among different series. All films were made with the following simple recipe: 14 mmol EDOT derivative, 50 g iron(III)-tosylate as a 41% (wt/wt) solution in n-butanol and 212 g n-butanol were mixed. The solution was coated on glass by a doctor blade with wet film thickness 60 µm at room temperature and then dried at room temperature or slightly enhanced temperature (for example, at 23°C for 20 min or at 40°C for 10 min). After drying, the films were rinsed with deionized water to remove iron salts and toluenesulfonic acid. Surface resistances were measured by a two-point method. Several alkyl-EDOT derivatives (synthesis via Williamson ether synthesis, transetherification,28 or Mitsunobu reaction,29,30 see Chapter 5) have been
6911X.indb 103
10/1/10 8:59:36 PM
104
PEDOT
investigated. Table 8.1 presents the surface resistances from in situ polymerized layers. The trends observable by these experiments may be summarized as follows: The more bulky the substituent(s) are, the lower the conductivity will be. Therefore, from methyl-EDOT to n-tetradecyl-EDOT the surface resistance continuously increases (entry 1 to 10). This is not consistent to observations in electrochemical experiments, where the polymer from tetradecyl-EDOT has been found to be more conductive than PEDOT or the medium chain length analogs.31 The conductivity is very rapidly decreased by disubstitution at the dioxane ring with a by far more drastic steric effect than observed for monosubstitution (entry 11 to 13). Obviously, the coplanar orientation of the thiophene rings is massively hindered by the bulky groups, and so the effective conjugation length is reduced. The red-violet color of the in situ polymerized and doped layers of both di-hexyl-derivatives is in good accordance to this phenomenon. Ring closure between two alkyl substituents decreases the steric effects of disubstitution remarkably, and so—not unexpectedly—both cyclohexanoderivatives (entry 14, 15)32 have conductivities in the range of medium alkyl substituted PEDOTs (entry 6 to 8).32 Surprising data are obtained from the chiral EDOT derivatives (entry 2 to 4, 7 to 8, and 12 to 13).32 In all cases the enantiomerically pure compounds result in higher surface resistances than mixtures of the enantiomers, contrary to expectations. Also ethers of EDOT-CH2OH have been checked with very similar results; a trend to lower conductivity with growing alkyl chainlength can be observed (see Figure 8.10). All points are corresponding to the n-alkylether with a number of C-atoms as marked in the legend of the x-axis, with two exceptions: the lower point for C8 represents the 2-ethylhexyl ether and the lower point for C5 the 3-pentyl ether. In both cases, the linear ethers exhibit a slightly higher surface resistance, but the trend in these rather close-by results is not very significant. 4 Arbitrary Units
3.5 3 2.5 2 1.5 1 0.5 0
0
2
4
6
8
10
12
14
16
C Number in R Figure 8.10 Trends in surface resistance: alkylethers of EDOT-CH2OH (R = 0: EDOT-CH2OH; surface resistance of EDOT-CH2OH = 1).
6911X.indb 104
10/1/10 8:59:36 PM
The In Situ Polymerization of EDOT to PEDOT
105
O
O O
N H
O
O
O
R
O
S
N H
R
O
S A
R = n-Bu R = tert-Bu R = n-Hex R = cyclo-Hex R = n-Oct R = n-Dodec R = n-Octadec R = Ph
B A/B = 80/20 A/B = 93/7 A/B = 80/20 A/B = 80/20 A/B = 80/20 A/B = 78/22 A/B = 80/20 A/B = 80/20
R = 25 Ohm/sq* R = 60 Ohm/sq* R = 30 Ohm/sq* R = 70 Ohm/sq* R = 25 Ohm/sq** R = 65 Ohm/sq* R = 60 Ohm/sq* R = 410 Ohm/sq*
*: 10 min/40°C */**: 60 µm wet film on glass **: 20 min/23°C R = surface resistance Figure 8.11 EDOT-CH2OH/ProDOT-OH urethanes. (Adapted from K. Reuter, EP 1 352 918, Prior: April 10, 2002; K. Reuter. 2000. Unpublished results.)
Similar circumstances are obvious are for R = Alkyl-NH-CO in the urethanes derived from EDOT-CH2OH (see formula A in Figure 8.11).33,34 A trend to lower conductivities with longer alkyl chains is observed. In contrast to the two examples for EDOT-CH2OH-ethers in Figure 8.9, the conductivity of highly branched or cyclic urethane polymers is lower than that for openchain derivatives with the same C number. In Figure 8.11 the isomeric urethane derived from ProDOT-OH (hydroxy3,4-propylenedioxy-thiophene, 3,4-dihydro-2H-thieno[3,4-b][1,4]dioxepin-3-ol) instead of EDOT-CH2OH is also depicted (formula B). ProDOT-OH is a typical byproduct of EDOT-CH2OH when synthesized from epibromohydrin and 3,4dihydroxythiophene-2,5-dicarboxylic acid esters, and it is difficult to separate. More synthetic details are given in Chapter 12. Due to the mixture of two components, the results must be considered as only tentative. However, very detailed and comprehensive investigations for the underivatized EDOT-CH2OH have been made, where the small part of ProDOT-OH, which is present as standard impurity from the preparative procedure, is also neglected with good arguments.24 In Ha et al.,24 the optimization of the in situ polymerization of EDOT is also applied for EDOT-CH2OH. These optimizations (described in Section 8.1 for EDOT), regarding stoichiometry, concentration, imidazole content, solvent, and so on, resulted in PEDOT-CH2OH layers
6911X.indb 105
10/1/10 8:59:37 PM
106
PEDOT
with the very high conductivity of 900 S/cm, combined with good transparency (82% at a surface resistance of 140 Ω/square). These results, gradually better than for EDOT, are fully consistent with the experiences made at H.C. Starck GmbH. That is why a patent application for the use of EDOT-CH2OH in capacitors had been filed, where the trend to higher conductivity also had been disclosed.35 The reason for the good performance of in situ layers of PEDOTCH2OH is considered in the tendency of EDOT-CH2OH to inhibit crystallite formation of iron salts and to facilitate the formation of very homogeneous films. Interestingly, a similar effect has led to improved PEDOT conductivity in vapor phase deposited films: The addition of an ethylenoxid-propylenoxid copolymer (PEG-ran-PPG) to the active layer of ferric p-toluenesulfonate suppressed the formation of crystallites and so allowed better film formation of the VPP PEDOT with therefore enhanced conductivity.23 Comparing the different groups of substituted EDOTs or even very similar compounds of the same group, another effect has to be kept in mind. For every compound there is an optimum temperature for in situ polymerization. This means that the best conductivity is achieved for EDOT and simple derivatives at room temperature or slightly elevated temperature, for example, at 40°C. EDOTs with longer alkyl chains tend to be less reactive and so need higher temperatures up to 80°C to achieve lowest surface resistances. A very surprising effect was observed with a special group of EDOT derivatives with bulky substituents. A lot of EDOT-CH2OH based ethers with mesogenic side chains were synthesized and in situ polymerized to study potential liquid crystalline (LC) behavior.36,37 Figure 8.12 shows the general formula for some typical monomers. In most cases the products investigated in this study contained around 5% of the analogous isomeric ProDOT-OH ether (compare Figure 8.11). The results were not affected by this impurity. For the ease of readability, these structures are omitted in the following figures. In Figure 8.13 the mesogen and compound types are specified. The results of surface resistance measurements with in situ polymerized films from these compounds are tabulated in Table 8.2. In full accordance with several other EDOTs with bulky substituents, moderate conductivities of all films after curing between 80°C and 120°C are
O
O
Mesogen
n
O
O n = 4; 5; 6; 8 S
Figure 8.12 EDOT derivatives with mesogenic side chains/general formula.
6911X.indb 106
10/1/10 8:59:37 PM
The In Situ Polymerization of EDOT to PEDOT
107
(I) X = CN (II) X = Br (III) X = COOCH3
X
(IV) X = COOC2H5 (V) X = COO-nC4H9 (VI) X = O-nC5H11
Y
(VII) Y = nC3H7
O
COOCH3
(VIII); not mesogenic
Figure 8.13 EDOT derivatives with mesogenic side chains/compound types.
observed. Dependent on the spacer (n between 4 and 8, compound/mesogen type II), the conductivity decreases with growing spacer length. This behavior is very similar to all EDOT derivatives bearing medium- and long-chain alkyl substituents (see earlier in this section). By annealing the films at 150°C, the conductivity is improved significantly in all cases. Because iron(III)-tosylate is practically completely reacted and residual traces are removed by rinsing the films with water, a chemical postreaction after
Table 8.2 Surface Resistance Data from Mesogen Substituted EDOTs Mesogen Typea
Nb
(I) (II) (II) (II) (II) (III) (IV) (V) (VI) (VII) (VIII)
5 4 5 6 8 5 5 5 5 5 5
a b
6911X.indb 107
Surface Resistance before Annealing (Ω/sq) 1200 390 600 710 2600 3700 170 190 330 180 380
Surface Resistance after Annealing (Ω/sq) 130 100 190 320 790 760 60 70 80 90 90
See Figure 8.13. See Figure 8.12.
10/1/10 8:59:38 PM
108
PEDOT
O O
O *
O O
n
S
*
PEDOT-CH2O-Bz
*
O
S
n
*
PEDOT-CH2O-CH2-cycHex
Figure 8.14 Aromatic and analog nonaromatic PEDOT derivatives with and without annealing effect on the conductivity.
curing can be excluded. This annealing effect cannot be observed with EDOT or simple EDOT derivatives like EDOT-C14 or EDOT-CH2OH. A typical LC behavior of all EDOTs with mesogenic side chains could be presumed, resulting in structural rearrangements to a greater order and therefore better charge transport. Surprisingly, the EDOT bearing the nonmesogenic phenyl-4-carboxy methyl group (VIII) also shows a distinct effect. A closer look to this phenomenon revealed that the effect indeed is not limited to liquid crystalline EDOT monomers. Not surprisingly, the cyclohexylmethyl-ether of PEDOT (Figure 8.14) as a nonmesogenically substituted EDOT does not exhibit a significant annealing effect; the surface resistance of doped films was measured to about 1000 Ω/square before and after annealing. In contrast, the benzyl ether as the aromatic, but also not mesogenic, analog (Figure 8.14) does show a clearly improved conductivity after annealing. From about 450 Ω/square after depositing, baking, and rinsing the film, the surface resistance could be decreased to 180 Ω/sq. Obviously, the LC character of the EDOT monomer is not a prerequisite for the occurrence of an annealing effect. More probably the annealing effect should be considered as a rearrangement induced by π-stacking resulting from the aromatic side groups. O O
Br
O O
O
S Figure 8.15 EDOT monomer for x-ray study after in situ polymerization. (Adapted from N. Wrubbel, 2004, Ph.D. Thesis, Heinrick-Heine-Universität Düsseldorf36; N. Wrubbel, H. Ritter, K. Reuter, A. Karbach, and D. Dreschsler, 2006, e-Polymers37.)
6911X.indb 108
10/1/10 8:59:40 PM
The In Situ Polymerization of EDOT to PEDOT
109
A pronounced rearrangement for an in situ polymerized and then annealed film of the monomer depicted in Figure 8.15, compared to the film before the annealing step, was verified by x-ray diffractometry measurements.36,37 Four reflections could be attributed to the ordered state of the polymer film. The intensities of two of these reflections were clearly enhanced upon annealing. Having a look on EDOT polymerization kinetics with Fe(III) tosylate, it is easily understood that an enormous acceleration for the in situ polymerization is observed by advancing from monomeric EDOT to bis-EDOT (BEDOT) or to longer conjugated structures. The reaction rate is enhanced by a factor of about 20,000 for the first rate determining step, that is, the one-electron transfer from the EDOT molecule to the oxidant, as the consequence of changing from the thiophene to the bithiophenic structure. So it is practically impossible to get a stable, processable solution of BEDOT or ter-EDOT containing an oxidant. For example, an alcoholic solution of BEDOT instantaneously precipitates blue, insoluble flocculation products when mixed with an iron(III) tosylate solution. Moreover, typical solvents for in situ polymerization, such as lower alcohols, are poor solvents for BEDOT and other lower oligomers. So the practical application of in situ polymerization is limited to derivatives with only one isolated EDOT moiety without further conjugation to EDOT residues.
References
6911X.indb 109
1. F. Jonas, G. Heywang, W. Schmidtberg, J. Heinze, and M. Dietrich. EP 339 340 (Bayer AG), Prior: April 22, 1988. 2. K. Reuter and S. Kirchmeyer. EP 1 327 645 (Bayer AG), Prior: December 27, 2001. 3. U. Tracht. 2001. Bayer AG, Leverkusen. Personal communication; S. Kirchmeyer, K. Reuter, and J. Simpson. 2007. Poly(3,4-Ethylenedioxythiophene): Scientific importance, remarkable properties, and applications. In: Handbook of Conducting Polymers, 3rd ed., ed. T. A. Skotheim and J. A. Reynolds, 10-1–10-22. Boca Raton, FL: CRC Press. 4. P. Audebert, J.-M. Catel, V. Duchenet, L. Guyard, P. Hapiot, and G. Le Coustumer. 1999. Redox chemistry of thiophene, pyrrole and thiophene-pyrrole-thiophene oligomers. Synth Met 101(1–3):642–645. 5. G. Zotti, S. Zecchin, G. Schiavon, and L. Groenendaal. 2000. Conductive and magnetic properties of 3,4-dimethoxy- and 3,4-ethylenedioxy-capped polypyrrole and polythiophene. Chem Mater 12(10):2996–3005. 6. A. Zykwinska, W. Domagala, A. Czardybon, B. Pilawa, and M. Lapkowski. 2003. In situ EPR spectrochemical studies of paramagnetic centres in poly(3,4ethylenedioxythiophene) (PEDOT) and poly(3,4-butylenedioxythiophene) (PBuDOT) films. Chem Phys 292(1):31–45. 7. A. Zykwinska, W. Domagala, and M. Lapkowski. 2003. ESR spectroelectrochemistry of poly(3,4-ethylenedioxythiophene) (PEDOT). Electrochem Commun 5(7):603–608.
10/1/10 8:59:41 PM
110
6911X.indb 110
PEDOT
8. K. Reuter, V. A. Nikanorov, and V. M. Bazhenov. EP 1 375 560 (H.C. Starck GmbH.), Prior: June 28, 2002. 9. D. M. de Leeuw, P. A. Kraakman, P. F. G. Bongaerts, C. M. J. Mutsaers, and D. B. M. Klaassen. 1994. Electroplating of conductive polymers for the metallization of insulators. Synth Met 66(1–3):263–273. 10. J. C. Carlberg and O. Inganäs. 1997. Poly(3,4-ethylenedioxythiophene) as electrode material in electrochemical capacitors. J Electrochem Soc 144(4):L61–L64. 11. K. Z. Xing, M. Fahlman, X. W. Chen, O. Inganäs, and W. R. Salaneck. 1997. The electronic structure of poly(3,4-ethylene-dioxythiophene): Studied by XPS and UPS. Synth Met 89(3):161–165. 12. D. M. de Leeuw, C. M. J. Mutsaers, and M. M. J. Simenon. EP 615 257 (Koninklijke Philips Electronics N. V.), Prior: March 9, 1993. 13. CLEVIOSTM—The Ultimate Conductive Polymer. http://www.clevios.com (accessed August 2010). 14. T. S. Hansen, K. West, O. Hassager, and N. B. Larsen. 2006. Integration of conducting polymer network in non-conductive polymer substrates. Synth Met 156(18–20):1203–1207. 15. R. Corradi and S. P. Armes. 1997. Chemical synthesis of poly(3,4-ethylenedioxythiophene). Synth Met 84(1–3):453–454. 16. K. Reuter, S. Kirchmeyer, U. Merker, P. W. Lövenich, and T. Meyer-Friedrichsen. WO 2008/034848 (H.C. Starck GmbH), Prior: September 20, 2006. 17. P.-M. Allemand. US 2006 0065889, Prior: September 30, 2004. 18. K. Reuter and S. Kirchmeyer. WO 2007 085371 (H.C. Starck GmbH Co. KG), Prior: January 20, 2006. 19. A. Varvoglis. 1997. Hypervalent iodine in organic synthesis. London: Academic Press. 20. S. Kirchmeyer and F. Jonas. EP 1 338 617 (Bayer AG), Prior: February 15, 2002. 21. P. A. Levermore, L. Chen, X. Wang, R. Das, and D. D. C. Bradley. 2007. Highly conductive poly(3,4-ethylenedioxythiophene) films by vapor phase polymerization for application in efficient organic light-emitting diodes. Adv Mater 19(17):2379–2385. 22. B. Winther-Jensen and K. West. 2004. Vapor-phase polymerization of 3,4ethylenedioxythiophene: a route to highly conducting polymer surface layers. Macromolecules 37:4538–4543. 23. K. Zuber, M. Fabretto, C. Hall, and P. Murphy. 2008. Improved PEDOT conductivity via suppression of crystallite formation in Fe(III) tosylate during vapor phase polymerization. Macromol Rapid Commun 29(18):1503–1508. 24. Y.-H. Ha, N. Nikolov, S. K. Pollack, J. Mastrangelo, B. D. Martin, and R. Shashidar. 2004. Towards a transparent, highly conductive poly(3,4-ethylenedioxythiophene). Adv Funct Mater 14(6):615–622. 25. U. Merker and K. Reuter. 2003. Unpublished results. 26. A. Kros, S. W. F. M. van Hövell, N. A. J. M. Sommerdijk, and R. J. M. Nolte. 2001. Poly(3,4-ethylenedioxythiophene)-based glucose biosensors. Adv Mater 13(20):1555–1557. 27. A. Kros, R. J. M. Nolte, and N. A. J. M. Sommerdijk. 2002. Conducting polymers with confined dimensions: Track-etch membranes for amperometric biosensor applications. Adv Mater 14(23):1779–1782. 28. D. Caras-Quintero and P. Bäuerle. 2004. Synthesis of the first enantiomerically pure and chiral, disubstituted 3,4-ethylenedioxythiophenes (EDOTs) and corresponding stereo- and regioregular PEDOTs. Chem Commun 8:926–927.
10/1/10 8:59:41 PM
The In Situ Polymerization of EDOT to PEDOT
111
29. K. Zong, L. Madrigal, L. Groenendaal, and J. R. Reynolds. 2002. 3,4-Alkylenedioxy ring formation via double Mitsunobu reactions: an efficient route for the synthesis of 3,4-ethylenedioxythiophene (EDOT) and 3,4-propylenedioxythiophene (ProDOT) derivatives as monomers for electron-rich conducting polymers. Chem Commun 21:2498–2499. 30. D. Caras-Quintero and P. Bäuerle. 2002. Efficient synthesis of 3,4-ethylenedioxythiophenes (EDOT) by Mitsunobu reaction. Chem Commun 22:2690–2691. 31. L. Groenendaal, G. Zotti, and F. Jonas. 2001. Optical, conductive and magnetic properties of electrochemically prepared alkylated poly(3,4-ethylenedioxythiophene)s. Synth Met 118(1–3):105–109. 32. P. Bäuerle, D. Caras-Quintero, and K. Reuter. 2004. Unpublished results. 33. K. Reuter. EP 1 352 918 (H.C. Starck GmbH), Prior: April 10, 2002. 34. K. Reuter. 2000. Unpublished results. 35. U. Merker, K. Reuter, and K. Lerch. EP 1 391 474 (H.C. Starck GmbH), Prior: August 16, 2002. 36. N. Wrubbel. 2004. Elektrisch leitfähige Polymere auf Basis von flüssigkristallinen Thiophenderivaten. Ph.D. thesis, Heinrich-Heine-Universität Düsseldorf. 37. N. Wrubbel, H. Ritter, K. Reuter, A. Karbach, and D. Drechsler. 2006. Annealing effects on the electrical conductivity of polymerized liquid crystalline 3,4ethylenedioxythiophene derivatives. e-Polymers 2; http://www.e-polymers. org/papers/ritter_030206.pdf.
6911X.indb 111
10/1/10 8:59:41 PM
6911X.indb 112
10/1/10 8:59:41 PM
9 PEDOT:PSS
9.1 PEDOT:PSS Dispersions 9.1.1 Introduction One of the reasons why poly(3,4-ethylenedioxythiophene) (PEDOT) has become a successful conductive polymer is the availability as a polymer dispersion. In combination with poly(styrenesulfonic acid) (PSS) as a counter ion, a polyelectrolyte complex (PEC) can be prepared that forms a stable dispersion, which is producible on an industrial scale and can be used in many deposition techniques. To understand the function of PSS and the requirements for the formation of a stable PEC, this chapter will start with a general view on polyelectrolyte complexes. This is followed by a section on the synthesis and properties of PEDOT:PSS dispersions, the properties of PEDOT:PSS films, and the function of conductivity enhancement agents. 9.1.2 Polyelectrolyte Complexes Polyelectrolyte complexes are typically formed by mixing aqueous solutions of polyanions and polycations. Depending on parameters, which are described in the following section, this mixing can result in a water-soluble complex or an insoluble precipitate. The first experiments on polyelectrolytes were performed in the 1930s by mixing oppositely charged natural polyelectrolytes such as gelatin or gum arabic.1 The kinetics, thermodynamics and the mechanisms that led to the formation of stable PEC complexes were investigated in the 1970s by Kabanov,2–3 Bakeev et al.,4 Karibyants et al.,5 Dautzenberg,6 Zitchenko et al.,7 Thünemann et al.,8 and others.9–10 The difference between charged complexes of low molecular weight compounds and those of macromolecules lies in the cooperative inter- and intramolecular interactions between the polymer chains. A comparison of the forces that occur around one, two, or three charges gives a first impression of that effect.12 For a single ion, such as a cation or an anion, there is no dependence of its potential energy on its position. For an ion pair cation–anion the gain in potential energy is –e2/r, where e is the electric charge and r is the distance 113
6911X.indb 113
10/1/10 8:59:41 PM
114
PEDOT
Figure 9.1 PEC arrangements: (left) ladder-type and (right) scrambled egg type. (Adapted from B. Philipp, W. Dawydoff, and K.-J. Linow, 1982, Z Chem 22(1):1–13; H.-G. Elias, 2001, Makromoleküle Physikalische Strukturen und Eigenschaften, Vol. 2, 6th ed. Weinheim: Wiley-VCH.)
between the two ions. For a triple ion with an arrangement such as cation– anion–cation, the potential gain is –2e2/r for the two cation–anion interactions and the potential loss is +e2/2r for the interaction between the two cations, resulting in a gain of –1.5 e2/r. Hence, a triple ion is more stable than an ion pair or the single ions.12 The same argument holds for even bigger aggregates, and therefore the arrangement of PECs is dominated by the Coulomb forces. On a molecular level, two arrangements of PECs have been discussed in the literature.10 The ladder type (Figure 9.1) shows a pairing of most of the polar groups from one macromolecule with those of the opposite charge of another macromolecule. The ladder type is found particularly in dilute solutions and in cases where the spacing of the charged groups along the chain is similar for both polyelectrolytes. Furthermore, the ladder type is found when polyelectrolytes with very different molecular weights are used.8 The so-called scrambled egg type is based on random interactions between polar groups of one macromolecule and various other polar groups of many other polymer chains with no order on the molecular or supermolecular level. Of course, many gradual variations from the ordered ladder type toward the disordered scrambled egg type are also possible. It is worthwhile noting that the degree of order for both cases is far lower than that in natural polyelectrolytes, such as the double helix in DNA with its precise spacing between the nucleic acids. The solvent used for most PECs described in literature is water, since it dissolves polyelectrolytes well due to its high dielectric constant; furthermore, it is easily accessible, nontoxic, and therefore the “natural” solvent for polyelectrolytes.13 To assess the stability of a PEC, it is necessary to consider
6911X.indb 114
10/1/10 8:59:41 PM
PEDOT:PSS
115
the number of effective polar groups of each polyelectrolyte. In an aqueous system that number is governed by the pH value since the groups change from their dissociate state to the nondissociate state. Therefore, to determine the concentration of effective polar groups [ceff] one needs to consider the concentration of the relevant repeating unit [ci] as well as their dissociation constants αi.12 In some polyelectrolytes with more than one type of functional group, such as gelatin, the overall charge of the polymer can change from a positive charge, to the neutral case, and then to an overall negative charge, simply by changing the pH.12 The mixing of solutions containing stoichiometric amounts of polyanions and polycations, which can be described in the equation [cpos] αpos = [cneg] αneg, generally leads to the precipitation of both species.3 This is due to the fact that the polar groups screen each other and the overall solubility is lost. Only at very low concentrations a few cases of soluble complexes with stoichiometric amounts of polyanions and polycations have been demonstrated, for example, in the case of poly(styrenesulfonate) and Polybrene® (1,5-dimethyl1,5-diazaundecamethylene polymethobromide; hexadimethrine bromide).14 However, when polyanions and polycations are mixed in nonstoichiometric ratios, soluble complexes can be formed. These polyelectrolyte complexes are found in the form of discrete soluble gel particles in the aqueous medium.15 For nonstoichiometric polyelectrolyte complexes the major component can also be described as host polyelectrolyte (HPE), whereas the oppositely charged minor component can be described as guest polyelectrolyte (GPE).16 The latter joins the repeat units of the HPE via electrostatic interactions, so that a network is formed. For the formation of soluble particles, it is beneficial if the HPE consists of high molecular weight material, whereas the GPE has a low molecular weight.6 Furthermore, it is beneficial for a stable PEC if at least one of the polyelectrolytes has weak ionic groups. PECs are typically found in the form of gel particles also described as microgel particles.6 The term gel is generally used to describe a chemically or physically cross-linked polymer network that is strongly swollen by a solvent.12 Gels show little deformation due to hydrostatic pressure but strong deformation by shear energy. The gel particles in the PEC dispersion are discrete particles and their size can be determined by static light scattering, dynamic light scattering, potentiometry or viscometry, or ultracentrifugation. Dautzenberg6 and Pogodina and Tsvetkov11 have investigated PEC by static and dynamic light scattering and found that PEC particles are highly polydisperse systems of nearly spherical structures. The sections where the charges are neutralized due to the HPE–GPE complex formation are less hydrophilic and are found in the center of the PEC.16 On the other hand, the HPE is found overproportionately on the outside, in the form of tails and loops, forming a shell around the gel particle. This excess of the major component results in an electrostatic repulsion between particles, which stabilizes the particles against further coagulation.5
6911X.indb 115
10/1/10 8:59:41 PM
116
PEDOT
The amount of excess HPE required for a stable dispersion is not universal but depends on the individual polyelectrolytes. Typically, some of the HPE is also found as free polymer in the solution, while the GPE is completely included in the particles. Another important parameter for the stability of PECs is the concentration of salt in the solvent. Dautzenberg investigated the response to different salt concentrations for the system PSS and poly(diallyldimethylammonium chloride) (PDADMAC) via static light scattering. PDADMAC was added to PSS in a molar ratio of 1:2 and the response to different salt concentration and hence the ionic(I) strength of the solution on the size of the particle (radius of gyration RG) was determined (Figure 9.2). In deionized water, large particles were observed. Such PDADMAC/PSS complexes contained more than 1000 chains.6 With salt concentration the size of the particles first decreases and then increases again. This is due to two effects. The first effect deals with the conformation of polyelectrolyte chains depending on the ion concentration. In the absence of mobile ions they are stiff due to the Coulomb repulsion between the charged functional groups.17 The charge compensation between stiff chains is strongly hindered and large particles are formed.9 In the presence of salt this repulsion is screened and the polyelectrolytes can take a more random coil structure. When the chains get flexible in the presence of salt then the charge compensation between two oppositely charged chains becomes easier resulting in a lower degree of aggregation.6
RG3 [nm]3
107
106
105
10–5
10–4
10–3
10–2
10–1
100
I [mol/L] Figure 9.2 Particle size (RG) of the complex NaPSS/PDADMAC at a molar ratio of 2:1 as a function of the salt concentration (ionic strength I). (Reprinted with permission from H. Dautzenberg, Polyelectrolyte Complex Formation in Highly Aggregating Systems. 1. Effect of Salt: Polyelectrolyte Complex Formation in the Presence of NaCl, Macromolecules 30(25):7810–7815. Copyright 1997 American Chemical Society.)
6911X.indb 116
10/1/10 8:59:42 PM
PEDOT:PSS
117
The second effect influencing the particle size is the repulsion between two individual microgel particles. The excess of the HPE and the resulting charged shells lead to the repulsion of individual particles. As the salt concentration is increased, this repulsion between different shells is reduced and aggregation of particles occurs. As a consequence, the size of the particles starts to increase again. When a critical salt concentration is exceeded, the dispersion collapses. This collapse can also be described as disproportionation reaction, where some fraction of the HPE is precipitated with the GPE, whereas another fraction of HPE is released into solution as free polyelectrolyte.8 The salt concentration also shows an effect on the exchange reaction between polyelectrolyte chains. At low ionic strengths PECs show few exchange reactions. At higher ion strengths polyion exchange and substitution reactions occur. These exchange reactions can be particularly observed in the first stages of PEC formation. Bakeev et al. have used pyrene labeled sodium poly(methacrylate) and poly(N-ethyl-N-vinylpyridinium bromide) to investigate the kinetics of the HPE–GPE complex formation.4 They found that in a first stage a polyelectrolyte complex is formed completely at random. In a second, slower stage the chains rearrange in the complex toward a thermodynamic equilibrium. Exchange reactions can lead to the formation of a soluble complex. In this case a polyelectrolyte is added in excess to a previously insoluble stoichiometric complex resulting in soluble PEC particles. Exchange reactions can particularly be observed when a compound with higher molecular weight is added. The exchange reaction is then driven by entropy.8 Saltybaeva at al. have found for a combination of polymers that a minimum molecular weight is required before a polyelectrolyte complex is formed.18 This minimum energy is not universal but depends on the PEC combination. 9.1.3 Synthesis of a PEDOT:PSS Complex Polystyrenesulfonic acid (PSS) was the first polyelectrolyte used for a PEC with PEDOT in 1990 and has remained the industrial standard ever since.19,20 PSS is commercially available in a large range of molecular weights with different polydispersities. Further to its commercial availability and its solubility in water, PSS forms durable films and shows no absorption in the visible range of light, resulting in transparent films.21 The sulfonic acid group is strongly acidic and hence highly polar. Sulfonic acids typically have a low pKa value, for example, for benzenesulfonic acid the pKa is 0.70.22 PSS as a counterion for PEDOT is always used in excess, that is, as host polyelectrolyte (HPE). The molar ratio of thiophene groups to sulfonic acid groups in standard PEDOT:PSS dispersions is in the range of 1:1.9 to 1:15.2, which corresponds to a weight ratio range of 1:2.5 up to 1:20. Since only one
6911X.indb 117
10/1/10 8:59:42 PM
118
PEDOT
charge is found for every three to four thiophene rings,23–25 the charge excess of PSS is between 6-fold and 46-fold. More details on standard commercial PEDOT:PSS dispersions are given at the end of Section 9.1. Due to the delocalization of positive charges in PEDOT, the resulting weak polar groups and the different spacing of charges in PEDOT compared to PSS, it is reasonable to a assume that the structure of PEDOT:PSS shows the form of a scrambled egg type. A pairing of charges as required in the ladder type is not possible. Although most of the research on PECs has been done by mixing the two polyelectrolyte types, this approach is not possible for the PEDOT:PSS complex. Similar to most charged conjugated polymers, the PEDOT polycation is not soluble in any solvent. Therefore the synthesis of this polycation needs to be performed in the presence of PSS so that the PEC complex is formed in situ as the polycation chain grows. This requirement forms a strong limitation for the analysis of PEDOT since a cationic PEDOT chain can only be obtained in combination with a suitable polyanion as PEC. The solvent of choice for the synthesis of PEDOT:PSS complex is water.19 Water is inert with respect to most oxidation or reducing agents. It is highly polar and a good solvent for PSS. However, water is a poor solvent for the monomer EDOT. At 20°C 0.21g of EDOT can be dissolved in 100 mL water. The solubility increases in the presence of PSS to 0.30% as shown in Figure 9.3.26 Furthermore, the figure shows that solubility of EDOT in the presence of PSS increases with temperature. In order to overcome the limited solubility, Lefebvre et al. have used mixtures of water and acetonitrile to increase the solubility of EDOT.27 An important parameter for the EDOT polymerization is the pH value. The presence of PSS leads to a low pH of less than 3. This is beneficial since acid acts as a catalyst for the reaction. In the absence of acid, the oxidation of EDOT can result in keto-functionalized side products.28 As described later, the reaction turns more acidic as it progresses since further protons
Temperature [°C]
40 30 20 10 0
0.2
0.25
0.3
0.35
0.4
Solubility of EDOT [%] Figure 9.3 Solubility of EDOT in a 1% solution of PSS.
6911X.indb 118
10/1/10 8:59:42 PM
119
PEDOT:PSS
are released. A range of oxidizing agents is available for the polymerization of EDOT in water. Iron(III) salts such as iron(III) nitrate and iron(III) chloride can be used as described in Chapter 7. However, the presence of stoichiometric amounts of iron(III) or iron(II) ions leads to the precipitation of the PEDOT:PSS complex in the latest stages of the reaction.27 This is because multivalent cations such as Fe(III) form a complex with PSS and reduce the stability of the PEDOT:PSS complex. The most versatile oxidizing agents are peroxodisulfates, in particular those with monovalent cations such as sodium, potassium, or ammonium.29–31 These peroxodisulfates are water-soluble and the oxidation potential of 2.12 V in acidic medium is sufficient for the oxidative polymerization of EDOT to the positively charged PEDOT polyelectrolyte.22 Peroxodisulfate has also been used in the form of the free acid, which is, however, less stable than the salts.32–34 In most cases peroxodisulfate is used in combination with an Fe(III) salt as a catalyst. From a stoichiometric point of view, one equivalent of peroxodisulfate is needed for each thiophene unit for the oxidative polymerization, taking two electrons from each thiophene. Additional peroxodisulfate is consumed in the oxidation of the polymer chain resulting in a polycation. Approximately one positive charge is found for every three to four thiophene rings.23–25 This ratio is also found experimentally, when the consumption of EDOT and peroxodisulfate is monitored during the polymerization.26 For simplicity, the reaction depicted in Figure 9.4 shows a PEDOT repeat unit with six thiophene rings and two positive charges. The exact chain length (n can be multiples of one-sixth) and the exact charge of the PEDOT chains have not been determined since these chains cannot be obtained as a separated material without counterions. Figure 9.4 shows that the reaction mixture turns more acidic as the reaction progresses since each mol of EDOT releases two moles of protons.
O +
6n O
Fe2(SO4)3
7 n Na2S2O8
PSS
O
O
O
O
O
O S
2+
S
S
S
S O
O
O
+ 14 nSO42– + 14 n Na+ + 12 nH+
S O
O
n
O
O
Figure 9.4 Reaction scheme for the PEDOT synthesis using sodium peroxodisulfate as an oxidant.
6911X.indb 119
10/1/10 8:59:43 PM
120
PEDOT
Oxygen has also been used as oxidizing agent for PEDOT together with small amounts of iron(III) sulfate as catalyst.19,29 The conductivity of films prepared from such PEDOT:PPS complexes was rather poor compared to those prepared using peroxodisulfate. Hydrogen peroxide has also been used as an oxidation agent in combination with enzymes that catalyze the oxidation.35 In the absence of such enzymes, hydrogen peroxide leads to the destruction of the conductive properties of PEDOT probably due to overoxidation.36 To remove byproducts from the reaction, the PEC complex can be precipitated and washed.27 However, the redispersion of the precipitate is difficult due to the agglomeration of the gel particles. Instead, inorganic ions such as sodium and sulfate can be removed using ion exchange resins.29 Furthermore, dialysis can be used to remove salts as well as organic low-molecular weight impurities. Similar to the case of PDADMAC/PSS the salt concentration in a PEDOT:PSS dispersion has a strong effect on the gel particles. In the case of PEDOT:PSS this can easily be detected by measuring the viscosity of the dispersion. Figure 9.5 shows the viscosity of PEDOT:PSS dispersions as a function of the salt concentration. The effects of salt on the viscosity in PEDOT:PSS is similar to the effect of salt on the radius of gyration in PDADMAC/PSS shown in Figure 9.2. It is reasonable to assume that the changes in viscosity for PEDOT:PSS dispersions are at least partial also due to changes in the particle size. In this particular PEDOT:PSS complex the weight ratio is 1:2.5 and concentration in water is 2.0%.26 At low salt concentrations viscosities 350
0 days
Viscosity [mPas]
300
4 days 11 days
250
18 days
200 150 100 50 0
0.0
2.2
4.3
6.5
8.7
10.9
13.0
15.2
17.4
19.6
Sodiumsulfate Concentration [mmol/L] Figure 9.5 Viscosity dependency of a 2.0% PEDOT:PSS dispersion in water on the salt concentration of sodium sulfate and storage time at 20°C. (Data from W. Lövenich, H.C. Starck Clevios GmbH, 2008. Unpublished results.)
6911X.indb 120
10/1/10 8:59:43 PM
121
PEDOT:PSS
140 Viscosity [mPas]
120 100 80 60 40 20 0
1
1.5
2
2.5
Solids Content [%] Figure 9.6 Solids content and viscosity relationship for a PEDOT:PSS complex at 20°C. (Data from W. Lövenich, H.C. Starck Clevios GmbH, 2008. Unpublished results.)
are approximately 100 mPas. With increasing salt concentration the viscosity is initially reduced; for example, when 6.9 mmol/L sodium sulfate are added the viscosity is reduced to 59 mPas. When 19.6 mmol/L sodium sulfate are added, the viscosity is increased to 76 mPas, again due to beginning agglomeration. The latter is a slow process in the case of PEDOT:PSS, as can be seen on the viscosity change over time. The higher the salt concentration the faster the agglomeration occurs. In the case of a 19.6 mmol/L concentration of sodium sulfate the viscosity quadruples within 18 days from 76 mPas to 320 mPas (Figure 9.5). Furthermore, the viscosity of a PEDOT:PSS dispersion strongly depends on its concentration. Figure 9.6 shows the increase of the PEDOT:PSS dispersion viscosity with dispersion contents between 1% and 2.5%. Due to this strong increase of viscosity PEDOT:PSS dispersions are typically prepared in concentrations below 5%. The size of the gel particles in the PEDOT:PSS dispersion can be detected using ultracentrifugation.37 In this analytical technique the PEC particles are precipitated using very high centrifugation speeds. Larger particles precipitate faster than smaller ones. The intensity of a laser beam passing through the dispersion detects the decrease in concentration and hence the particle size distribution of the dispersion can be determined. The average particle size of PEDOT:PSS dispersions is in the range of 10 nm to 1 µm. The particle size distribution is further characterized by values that indicate which percentage of particles is below a certain size. For example a d50 value of 100 nm indicates that 50% of all particles are smaller than 100 nm. The particle size distribution for PEDOT:PSS dispersion is typically non-Gaussian. The gel particles of PEC dispersions can be altered after the synthesis using shear energy. In this process the particles are deformed, so that smaller particles are generated. Figure 9.7 shows the particle size distribution of a
6911X.indb 121
10/1/10 8:59:44 PM
122
PEDOT
Fraction [rel. units]
1 Before shearing
0.8
After shearing
0.6 0.4 0.2 0
0
0.1
0.2
0.3
0.4
0.5
0.6
Particle Size [µm] Figure 9.7 Particle size distribution of a PEDOT:PSS PEC before and after a shear treatment. (Data from W. Lövenich 2007. Unpublished results.)
PEDOT:PSS dispersion before and after a shearing process. In this case, the average particle size is reduced from 380 nm to 23 nm. 9.1.4 Commercial PEDOT:PSS Types and Their Properties Aqueous PEDOT:PSS dispersions are commercially available from H.C. Starck Clevios GmbH under the trade name CleviosTM. As reference Table 9.1 summarizes important properties of established Clevios dispersions. Table 9.1 Commercial PEDOT:PSS Dispersions in Water and Their Properties
Trade Name Clevios P Clevios PH Clevios P VP AI 4083 Clevios P VP CH 8000 Clevios PH 500 Clevios PH 750 Clevios PH 1000 a
b
6911X.indb 122
Solids Content in Water (w/w) (%)a
PEDOT:PSS Ratio (w/w)
Viscosity at 20°C (mPas)a
Particle Size d50 (nm)a
Conductivity (S/cm)a
1.3 1.3 1.5
1:2.5 1:2.5 1:6
80 20 10
80 30 40
<10 <10 10–3
2.8
1:20
15
25
10–5
1.1 1.1 1.1
1:2.5 1:2.5 1:2.5
25 25 30
30 30 30
500b 750b 1000b
Typical values for solids content, viscosity, particle size, and conductivity are given; no specification. Conductivities for Clevios PH 500, PH 750, and PH 1000 are measured for dispersions containing 5% dimethyl sulfoxide.
10/1/10 8:59:44 PM
PEDOT:PSS
123
9.2 Properties of PEDOT:PSS 9.2.1 Deposition of PEDOT:PSS PEDOT:PSS dispersed in water can be deposited in principle by all common techniques employed for the deposition of waterborne coatings. Common deposition techniques to obtain uniform coatings are slit coating, drop casting, bar coating, spin coating, electrospinning, and spraying. In case a structured deposition is required, other techniques are commonly employed such as screen printing, inkjet, nozzle printing, and various forms of contact printing (that is, relief, flexo, gravure, and offset printing).38 Other ways for structured deposition of PEDOT:PSS have been realized by modifying the wetting properties of the surface, that is by depositing water repellants39 or by introducing photo-lithographical techniques.40 The pristine PEDOT:PSS dispersion in water has to be adjusted to meet the requirements of the specific deposition technique and to obtain uniform films. Important properties that determine the film quality are the viscosity, the surface tension, and the adhesion to the substrate. This can be done by using different grades of PEDOT:PSS, which differ in solids content, the ratio of PEDOT to PSS, the gel particle distribution, or by the addition of water soluble or dispersible additives (i.e. surfactants, stabilizers, and cross-linking agents or inert polymers as binders). Several ready-to-use formulations optimized for specific applications and deposition techniques are commercially available.41 For a discussion of common components in formulations and their function refer to Chapter 10, Section 10.4. The deposition of PEDOT:PSS dispersions by spin coating has proven to be an easily accessible technique to obtain uniform films in a thickness range of 0 to 300 nm. Typical spin curves for two commercial grades of PEDOT:PSS (CLEVIOS P AI 4083 designed for hole injection in OLEDs and CLEVIOS PH 500 designed for transparent electrodes41) are illustrated in Figure 9.8a,b.42,43 The root-mean-square roughness Ra of these films are in the order of Ra ≈ 1 nm44 making these films attractive as cladding layer for thin film device applications. 9.2.2 Thin-Film Properties 9.2.2.1 Thermal and Lifetime Stability PEDOT:PSS films have to be dried prior to further processing. The water is removed by baking the layers at elevated temperatures, under infrared (IR) radiation, or by applying a vacuum. These different drying processes are often also combined, for example, in a vacuum oven. Thin films are typically dry after leaving the coated substrates on a hot plate set to temperatures above 100°C for some seconds.
6911X.indb 123
10/1/10 8:59:44 PM
124
PEDOT
150
Layer Thickness [nm]
125 100
Lid closed Lid opened
75 50
Layer Thickness [nm]
300 CLEVIOS P Al 4083 Viscosity (@ 700 s–1): 7.0 mPas Solid content: 1.34% Substrate size: 2" square Spin time: 30 sec Acceleration: 200 rpm/s
CLEVIOS PH 500 Viscosity (@ 100 s–1): 38 mPas Solid content: 1.56% Substrate size: 2" square Spin time: 30 sec Acceleration: 200 rpm/s
200
Lid closed Lid opened
100
25 0
0
1000
2000
0
0
500
1000 1500 2000 2500 3000
Spin Speed [rpm]
Spin Speed [rpm]
(a)
(b)
Figure 9.8 Spin curves of CLEVIOS P AI 4083 and CLEVIOS PH 500 taken on a Carl Süss RC8 spincoater, equipped with a rotating lid. Spin curves were obtained with the lid in opened and closed mode. The insets disclose parameter settings. (From A. Elschner and S. Kirchmeyer, Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies, ed. C. Brabec, V. Dyakonov, and U. Scherf, 2008. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
One major advantage of PEDOT:PSS films is their thermal stability. The thermal stability has been studied by thermogravimetrical analysis (TGA) of thick layers.45 Figure 9.9 depicts the weight loss and the ion currents of the relevant masses over time while the sample is heated at a constant rate in helium. Between 100°C and 200°C the weight loss is solely determined by evaporation of remaining water. At about 250°C the sample weight decreases significantly. Simultaneously the mass m/e = 64 increases, attributed to SO2 ions and indicating the fragmentation of the PSS sulfonate group. At higher temperatures of T > 350°C other fragments due to carbon oxidation are detected. Following the analytical data obtained by TGA the material is considered to be thermally stable up to temperatures of T = 200°C. In contrast to many other conjugated highly conductive polymers, PEDOT exhibits a very stable conductivity.46,47 However, like all carbon compounds, PEDOT-type polymers will be subject to degradation, especially when exposed to harsher conditions. The light stability of PEDOT and its derivatives have been studied in detail.48 The overall decay mechanism seems to be oxidation by oxygen49–52 enhanced by light. Attack to the sulfur atom of the thiophene ring will yield nonconducting sulfoxide and sulfone structures, whereas the attack on the α-carbon atom next to the thiophene sulfur will yield a hydroxyl-group which rearranges subsequently (Figure 9.10).53
6911X.indb 124
10/1/10 8:59:45 PM
125
PEDOT:PSS
100
Weight of sample: 11.752 mg, 80 ml min–1 He, 5 K min–1
Weight [%]
90 80 70 60 50 40
0
100
200
300 400 Temperature [°C] (a)
500
600
Ion Current [–]
m/z = 18(H2O) m/z = 28(CO,N2) m/z = 44(CO2) m/z = 76 (same as m/e = 60) m/z = 64(SO2) (same as m/e = 48)
0
100
200
300 400 Temperature [°C] (b)
500
600
Figure 9.9 Thermogravimetric analysis (TGA) of PEDOT:PSS with a ratio of PEDOT to PSS as 1:20 by weight. (a) The weight loss is monitored while heating the sample at a constant rate of 5 K/min. (b) The ion currents of several relevant masses are recorded simultaneously. (Data from Bayer Technology Services, Leverkusen, Germany.)
To maintain the conductivity of PEDOT:PSS films over time, an ultraviolet (UV)-light exposure as well as elevated temperatures above 70°C in combination with the exposure to oxygen need to be avoided. One way to protect layers in devices is the hermetic encapsulation of the device. Commercial formulations were found to prolong lifetime stability as depicted in Figure 9.11 by blending PEDOT:PSS with additives as stabilizers.41 In many practical
6911X.indb 125
10/1/10 8:59:45 PM
126
PEDOT
O
O
O
O
O
S
S
O
O O
O
S
S
S
OH
O O
O
O
O
O
S
O
Figure 9.10 Possible degradation reactions of PEDOT.
applications such as in capacitors or as antistatic layers, PEDOT has proven to fulfill its function over the years. 9.2.2.2 UV Stability Like most polymers pure PEDOT:PSS films will degrade over time when exposed to UV light. This fact might limit the use of the conducting polymers in outdoor applications if proper protection means are not taken. Figure 9.12 displays the resistivity of thin films of pristine PEDOT:PSS with a ratio of PEDOT to PSS as 1:2.5 including 5 wt% DMSO under continuous 500
CLEVIOS PH1000 CLEVIOS PH1000 stabilized
Sheet Resistance [Ωsq]
400 300 200 100 0
0
200
400
600 Time [h]
800
1000
Figure 9.11 Sheet resistance of films of PEDOT:PSS including 5% ethylene-glycole stored at 85°C and 85% relative humidity as a function of time. (Data from H.C. Starck Clevios GmbH, Leverkusen, Germany.)
6911X.indb 126
10/1/10 8:59:46 PM
127
PEDOT:PSS
2400
Irradiated, not encapsulated Irradiated, encapsulated Dark, not encapsulated
Sheet Resistance [Ωsq]
2000 1600 1200 800 400 0
0
5
10
15
20
25
Time of Exposure [days] Figure 9.12 Sheet resistance of pristine PEDOT PSS-films (Clevios PH 500 including 5 wt% DMSO, doctor bladed films, 18 µm wet) were monitored over time while the samples were exposed to the radiation of an Xe lamp (Atlas Suntester CPS+, 500 W/m², introduced glass-filter). When the films are encapsulated against ambient air by a glued-on glass cap the increase of resistivity is slowed down and resembles the change of a control sample being stored in the dark. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
illumination. The films were deposited on glass substrates and were exposed to the light of an Xe lamp while the resistivity was monitored.42 The films showed a steep increase of resistivity when left unprotected in air, whereas films being additionally encapsulated by a thin glass plate to avoid the contact with ambient air exhibited only a slow increase. The increase for the encapsulated layers was found to be similar to the films stored in air in the absence of light. These results demonstrate the necessity to protect thin PEDOT:PSS films from the impact of UV light when exposed to ambient air. The degradation is due to the oxidation of PEDOT, which is accelerated by the simultaneous absorption of UV light. It can be significantly slowed down when the films are properly encapsulated, that is, by covering PEDOT:PSS with glass plates or protective polymer coatings. Also the deposition of thick PEDOT:PSS layers or the addition of stabilizing agents will reduce the increase of resistivity over time (see Figure 9.11). The wavelength of radiation will affect the kinetic of degradation significantly. To determine the dependence of the resistivity increase on the photon energy, PEDOT:PSS films were exposed to monochromatic light of different wavelengths.42 This so-called activation spectrum is depicted in Figure 9.13. It relates to unprotected thin conductive films with a ratio of PEDOT to PSS
6911X.indb 127
10/1/10 8:59:47 PM
128
PEDOT
2000
∆S [Ω cm/Ws/cm2]
5
1500
4
× 10
1000
3 2
500
1 0
280
300
320
340
360
380
400
420
AM1.5 Irradiance [W/(m2µm]
6
0
λ (nm) Figure 9.13 Activation spectrum of PEDOT:PSS (Clevios P) depicts the change of resistivity on irradiation dose and wavelength. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.) For comparison reasons the spectrum of solar AM1.5 irradiance adapted from Bird/ Hulstom is inserted.
as 1:2.5. The curve indicates that the conductivity of PEDOT:PSS films will especially decrease for absorbed UV photons in the spectral range of λ < 320 nm. The absorption of photons stemming from this spectral range has to be especially inhibited, that is, by an overcoat containing efficient UV absorbers. 9.2.2.3 Water Uptake Like other polymers containing sulfonic acid groups, PEDOT:PSS is strongly hygroscopic and will take up moisture when handled under ambient conditions. The 15 wt% water loss of predried PEDOT:PSS after baking, as depicted in Figure 9.9a, can therefore be attributed to absorbed water. For the sake of better illustration the water absorption of a predried PEDOT:PSS sample of 1.5 g is monitored over time in Figure 9.14. The sample consisted of flakes peeled off from the underlying substrate with a thickness of several ten microns. Within the first 3 minutes the weight of the solid sample already increased by 10% when brought in contact with ambient air. The kinetics of water absorption depends, of course, on sample geometry and the level of humidity. Thin layers of PEDOT:PSS films up to a thickness of about 100 nm almost instantaneously absorb water from the environment. The picked up water will be incorporated into the films and hence the layer thickness will increase accordingly. The swelling of the films is shown in Figure 9.15 in dependence
6911X.indb 128
10/1/10 8:59:47 PM
129
PEDOT:PSS
Rel. Weight Increase
1.3 PEDOT: PSS (1:2.5) Weight of solid sample: 1.5 g Temp.: 24°C Humidity: 70%
1.2
1.1
1.0
0.9
0
200
400 600 Time [sec]
800
1000
Figure 9.14 PEDOT:PSS (weight ratio of PEDOT:PSS = 1:2.5) has been dried in a beaker in an oven to form a layer several micrometers thick and subsequently transferred into air with a relative humidity of 70%. The weight increase owing to the absorption of water is monitored as a function of time. (Data from Bayer Technology Services, Leverkusen, Germany.)
300 rH 0.6% rH 15.5% rH 41%
Layer Thickness [nm]
250
rH 62% rH 82%
200 150 100 50 0
1:2.5
1:6
1:2.5 + 5% DMSO
1:20
PEDOT:PSS Figure 9.15 Layer thickness increase of PEDOT:PSS as a function of relative humidity for different ratios of PEDOT to PSS. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
6911X.indb 129
10/1/10 8:59:49 PM
130
PEDOT
of the relative humidity level. An increase of layer thickness is especially pronounced for films with a high PSS content and can reach up to 30%. To avoid the water absorption, precautions have to be taken, for example, hermetic encapsulation of devices. 9.2.2.4 Mechanical Properties For transparent conductive coatings on polymer foils the flexibility of PEDOT:PSS films is considered to be a major advantage compared to inorganic materials like metal oxide films since PEDOT:PSS films do not crack upon bending.54 The mechanical properties of PEDOT:PSS have been investigated by tensile strength tests on free-standing PEDOT:PSS films.55 Young’s modulus and tensile strength were found to be strongly depending on the relative humidity (rH) level during the tests. The polymer exhibits brittle fracture behavior at rH = 23% which changes to plastic fracture behavior at an intermediate rH = 55% (Figure 9.16). The Young’s modulus of films increases from 0.9 GPa at 55% rH to 2.8 GPa at 23% rH. The ability to change film dimensions by altering the water content in the films has driven Okuzaki et al. to employ PEDOT:PSS as an electro-active polymer actuator.56 Film contractions of 2.4% to 4.5 % depending on relative humidity were realized by applying an electrical bias and removing water from the films due to Joule heating. 60 23% rH
Stress [MPa]
50 40 40% rH
30
55% rH
20 10 0 0.00
0.02
0.04
0.06
Strain [–] Figure 9.16 Stress–strain diagrams of PEDOT:PSS at different levels of relative humidity. (Reprinted from Synth Met 159(5–6): 473–479, U. Lang, N. Naujoks, and J. Dual, Mechanical Characterization of PEDOT:PSS Thin Films. Copyright 2009, with permission from Elsevier.)
6911X.indb 130
10/1/10 8:59:50 PM
PEDOT:PSS
131
9.2.2.5 Morphology: Surface and Bulk Owing to the film-forming properties of PSS, PEDOT:PSS films are amorphous.57 No evidence for crystalline regimes, except short-range structural order, was found by x-ray diffraction (XRD) even when high boiling solvents to change film morphology and enhance film conductivity were added.58 This is in contrast to chemically in situ polymerized PEDOT films, which do not contain PSS and exhibit crystalline ordering and fibril-like structures.59,60 No phase segregation occurs between PEDOT and PSS on a microscopic scale in thick dried films. The mixture is homogeneous even on the nanometer scale as determined by solid state 13C-CP/MAS NMR.61 Phase separation has been observed, however, on uniform thin films of PEDOT:PSS deposited by spin coating on glass substrates. PSS was found to segregate to the surface to form a PSS-enriched phase of 30 to 40 Å. The enrichment was first determined by Greczynski et al. investigating surfaces of PEDOT:PSS films with photoelectron spectroscopy.62 Photoelectrons from S(2p) and O(1s) core levels can be distinguished whether they stem from PEDOT or PSS due to differences in binding energy. The molar ratio of PSS to PEDOT was found to be 3.5 in contradiction to 1.2 anticipated from the overall composition of the investigated material, which suggests that PSS enriches at the film surface. By varying the photon energy it was possible to profile the PSS concentration in dependence of penetration depth and to confirm this suggestion.63 Because angle-dependent x-ray photoelectron spectroscopy (XPS) did result in data that point to variations in the PSSto-PEDOT ratio, it was concluded that PEDOT:PSS films exhibit a rather rough surface. A grainlike morphology of the films with a nonuniform distribution of PEDOT and PSS species within a single grain was proposed (Figure 9.17). A different experimental approach was pursued by Jukes et al. to determine the phase separation between PEDOT and PSS.64 PEDOT:PSS films with different componential ratios of PEDOT to PSS were investigated by small angle neutron spectroscopy. Deuterated PSS was used to achieve diffraction contrast between the two polymers. Again it was found that the volume fraction of PSS at the film surface was enriched relative to the bulk. With increasing PSS concentration in the film this phase separation becomes more and more pronounced, an effect that might be a consequence of an increasing amount of free PSS not incorporated into the PEDOT:PSS ionic complex. The length scale for the PSS-enriched region was found to be 5 to 15 nm and can be controlled by the baking conditions, especially by the annealing time. A variety of publications analyzed the surface roughness of PEDOT:PSS films with scanning atomic force microscopy. Especially for electro-optical devices comprising thin organic layers the smoothness is of upmost importance. The surface roughness depends on the weight ratio of PEDOT to PSS and on the specific distribution of gel particle sizes. PEDOT:PSS forms gel
6911X.indb 131
10/1/10 8:59:50 PM
132
PEDOT
hv (210 eV)
e– (S(2p))
hv (1000 eV)
e– (S(2p))
S(2p)
PSS
PEDOT
172
170 168 166 164 Binding Energy [eV]
PSS
Intensity [arb. units]
Intensity [arb. units]
S(2p)
PEDOT
172
170 168 166 164 Binding Energy [eV]
Figure 9.17 A schematic illustration on the studies of phase separation in PEDOT:PSS polymer blend by employing variable photon energy photoelectron spectroscopy. In the cross-section through a single PEDOT:PSS grain shown at the top of the figure the dark areas represent the regions with an excess of the PEDOT component, whereas light-gray regions correspond to the PSSrich part. The size and position of the arrows representing the photoelectrons ejected from the sample are correlated with the intensity of the signal from a particular blend component and the effective probing depth for a given photon energy. (Reprinted from J Electr Spectr Rel Phenom 121(1–3):1–17, G. Greczynski, T. Kugler, M. Keil et al., Photoelectron Spectroscopy of Thin Films of PEDOT:PSS Conjugated Polymer Blend: A Mini-Review and Some New Results. Copyright 2001, with permission from Elsevier.)
particles in aqueous dispersion as discussed in Section 9.1. As a general rule, the film’s surface smoothness will increase with a decreasing average gel particles size. PEDOT:PSS grades designed for hole injection, for example, Clevios P CH 8000,41 with a weight ratio of PEDOT to PSS of 1:20 by weight and an average particle size of 25 nm (d50) forms films with a root-mean-square surface roughness of Ra = 0.8 nm (Figure 9.18).65 For comparison Clevios PH 500, a dispersion for conductive coatings with PEDOT to PSS ratio of 1:2.5 and a similar particle size of 30 nm (d50) will lead to slightly rougher films of Ra = 1.4 nm. Different models exist to describe the morphology of PEDOT:PSS films. A lamellae model was proposed by Ionescu-Zanetti et al.66 With conductivity
6911X.indb 132
10/1/10 8:59:52 PM
133
PEDOT:PSS
PEDOT:PSS Surface Roughness 1:25 (Clevios PH 500)
0
0.25
0.50
0.75
Ra = 1.42 nm
1:6 (Clevios P Al 4083)
1.00
1:20 (Clevios P CH 8000)
1.00
1.00
0.75
0.75
0.75
0.50
0.50
0.50
0.25
0.25
0.25
0 1.00 µm
0
0.25
0.50
0.75
Ra = 0.91 nm
0 1.00 µm
0
0.25
0.50
0.75
Ra = 0.79 nm
0 1.00 µm
Figure 9.18 Surface roughness of various PEDOT:PSS-types differing in their ratio of PEDOT to PSS determined by atomic force microscopy. The layers were spin coated on ITO-coated glass (R a = 1.8 nm) at a spin speed of 33 rps and baked at T = 200°C for 5 min on a hot plate. The rootmean-square surface roughness R a was calculated for an area of 1 µm square.
and phase-shift atomic force microscopy (AFM) paracrystalline, terraced structures were detected, which were interpreted as molecular lamellae of PEDOT and PSS. The PEDOT lamellae were found to be conductive, whereas the PSS lamellae are electrically isolating. The interlamellae distance was estimated to be 3 nm. A different morphological model for the bulk properties of PEDOT:PSS films was proposed by Nardes et al.67 The authors suggest a film morphology resembling stacked pancakes (Figure 9.19a). This model is derived from the morphology of PEDOT:PSS in solution where gel particles with a PSSrich outer shell are formed.68,69 When these sphere-shaped gels dry to form a film, their original structure is maintained, but due to a shrinkage predominantly in the vertical direction, they form disc-like shapes, which exhibit PSS-rich regions in the contact areas (Figure 9.19b). This model will explain why the anisotropic conductivity of spin-cast Clevios P AI 4083 films, which was determined to be 1 × 10 –3 S/cm in the horizontal direction (parallel to the substrates surface at room temperature) and about 500 times lower in the vertical (perpendicular) direction.67,70 According to this model, free charge carriers passing a film composed of disc-like structures will travel easier along the long axis than along the short axis of the discs. The regions of PSS enrichment are believed to be poor electrical conductors due to a lack of PEDOT charge transport sites and hence form energetic charge barriers. Indeed, the density of PSS-enriched grain boundaries between the discs was found to be higher parallel to the substrate’s surface than in the vertical direction. Employing high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM), Lang et al. have evidenced the model of a
6911X.indb 133
10/1/10 8:59:53 PM
134
PEDOT
A
B
C
Figure 9.19 (A) 200 nm × 200 nm topographic scanning tunnel microscope (STM) image of PEDOT:PSS on ITO at 2.3 V, tunneling current 10 pA, and vertical scale 15 nm, the inset shows a line section. (B) 200 nm × 200 nm cross-sectional atomic force microscope phase image (X-AFM) of cleaved PEDOT:PSS on glass, vertical scale is 8°. The glass substrate is on the bottom side of the image, as shown by the inset of 530 nm × 580 nm and a vertical scale 70°. A pancake-like particle is highlighted by the ellipse. (C) Cross-sectional view of the schematic morphological model for PEDOT:PSS thin films derived from combined STM and X-AFM measurements. PSS-rich lamella is composed by several pancake-like particles as shown by the dotted lines. The typical diameter (d) of the particles is about 20 to 25 nm and a height (h) of 5 to 6 nm. (From A. M. Nardes, M. Kemerink, R. A. J. Janssen et al., Microscopic Understanding of the Anisotropic Conductivity of PEDOT:PSS Thin Films, Adv Mater, 2007, 19(9):1196–1200. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
granular structure of PEDOT:PSS gel particles maintained during film formation.71 The lateral size of the dried grains in the films was found to be in the order of 50 nm similar to the grain size of swollen particles in dispersion (Figure 9.20a). These grains forming larger aggregates and exhibiting brighter contrast were detected on the film’s surface. Energy dispersive x-ray (EDX) analysis of the films revealed that within the large grains the elemental composition differs between core and contact area with higher concentrations of sodium, potassium, and calcium atoms as well as a reduced concentration of sulfur in the contact area (Figure 9.20b). It is reasonable to assume that the
6911X.indb 134
10/1/10 8:59:56 PM
135
PEDOT:PSS
Agglomerate
Ring (a) Figure 9.20 (a) HAADF-STEM image of a 25 nm thick PEDOT:PSS film. The individual grains are clearly visible with an average diameter of about 50 nm. The grains are surrounded by a thin bright ring. In the lower right corner an agglomerate of larger grains with a diameter of about 200 nm is partly visible. The scalebar represents 100 nm. (b) HAADF-STEM image of a 25 nm thick solid PEDOT:PSS film with several agglomerates exemplarily showing different areas from which the elemental composition were acquired by EDX: Area 1 2 3
Contact area Core of grains Background
Na (%)
Si (%)
S (%)
K (%)
Ca (%)
15–20 7–13
0–4 3–8
~65 75–82
2–6 0–3
5–11 0–4
In principle the same as for cores of grains
Area 1 represents contact areas between individual grains of the agglomerate, area 2 stands for cores of grains of the agglomerate, and area 3 means that the mea surement was acquired from the background, that is, the 25 nm thick PEDOT:PSS film. (c) Schematic showing how aggregates lie on top of film. (From U. Lang, E. Müller, N. Naujoks, and J. Dual, Microscopical Investigations of PEDOT:PSS Thin Films, Adv Funct Mater, 2009, 19(8):1215–1220. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
6911X.indb 135
10/1/10 8:59:58 PM
136
PEDOT
3
1
2
100 nm (b) Film Agglomerate
(c) Figure 9.20 (Continued)
alkaline and alkaline earth ions accumulate here, in the PSS-enriched zone between the grains, to form ionic complexes. 9.2.3 Electronic States 9.2.3.1 UV-Vis (Ultraviolet-Visible) Spectra Ethylenedioxythiophene (EDOT), the monomer for PEDOT, is a transparent liquid. The absorption spectrum of EDOT peaks at 260 nm. The absorption maximum shifts considerably to lower photon energy when the π system of EDOT is extended by the formation of dimers and higher oligomers in which conjugation
6911X.indb 136
10/1/10 8:59:59 PM
PEDOT:PSS
137
is extended to 3 to 6 monomeric EDOT units. This effect has been extensively studied for EDOT oligomers up to the quater(3,4-ethylenedioxythiophene) in theory by quantum-chemical calculations72 and experimentally by cyclic volt ammetry.73 Besides doping charges, which might generate either radical ions or dications and yield a significant bathochromic shift of the absorption, another mechanism for a significant red shift has been identified: with protons the neutral oligomers form protonated oligo-EDOT (Figure 9.21).74 The shift anticipated due to the extension of the π system is in accordance with the model of an electron gas confined to a potential tray. A longer conjugation length of the molecule yields a narrower energy gap between the highest occupied and the lowest unoccupied molecular orbital (HOMO and LUMO, respectively), an effect which has been observed in many thiophene systems75–78 and understood based on theoretical molecular orbital calculations.79 Oligo-EDOTs represent a model system for undisturbed delocalized π systems due to their almost planar geometry. The rotation between the thiophene rings is hindered owing to the stabilizing interaction between oxygen and sulfur atoms of adjacent rings.80,81 Although the absorption spectrum of chemically polymerized PEDOT may result from various chemical species present, it is a common understanding that predominantly the main contribution stems from oxidatively charged oligomers consisting of 5 to 15 monomeric units. The absorption of PEDOT:PSS is almost identical to the absorption of in situ chemically polymerized PEDOT without PSS. Figure 9.22 depicts the relative transmission, absorption, and reflection spectra of CLEVIOS PH 1000 with 5% DMSO added. Strong absorbing features at 193 and 225 nm, not shown here, are present for pure PSS and PEDOT:PSS.82 These are attributed to π and π* transitions of the benzene rings in the PSS system. The broad absorption band in the visible and in the IR region can be interpreted as the contribution of free charge carriers to absorption or alternatively to excitations of midgap states (polarons or bipolaron states).82 The addition of additives like high boiling solvents will not effect the optical properties significantly,82 although the conductivity is enhanced by several orders of magnitude (see Section 9.3). Note that for high-conductive PEDOT:PSS films the IR reflection can reach 50%. 9.2.3.2 Energy Levels in PEDOT The absorption spectrum of PEDOT:PSS strongly depends on the PEDOT oxidation state, and this has motivated many groups to investigate this material in electrochromic devices (see Chapter 10). Gustafsson et al. have studied PEDOT:PSS films deposited on ITO as electrode in an electrochemical cell.83 Figure 9.23 depicts the optical absorption spectra as a function of photon energy at different bias. The peak at 2.2 eV (560 nm), the band gap of absorption, reduces as the applied bias is increased and two new peaks appear at lower energies. The isobestic point proves that
6911X.indb 137
10/1/10 8:59:59 PM
6911X.indb 138
Absorbance Coefficient [L mol–1 cm–1]
250
300
EDOT 2-EDOT 3-EDOT 4-EDOT
400
Wavelength [nm]
350
450
500
O
O
S
S
O
O
O
S
O
O
n
S
H+
+ O
O
H H
S
O
O
S
O
n
O
S
O
S
O
–m e–
O
O
S
O
n
O
S
O
m+
Figure 9.21 Absorption spectra of protonated oligo-ethylenedioxythiophenes dissolved in 10% AcOH/DCM and the corresponding chemical structures. (Diagram adapted from E. Reinoldi and P. Bäuerle, March 2005. Personal communication with permission of the authors.)
0
10000
20000
30000
40000
138 PEDOT
10/1/10 9:00:00 PM
139
PEDOT:PSS
90 Transmission Absorption Reflection
80 70
T, A, R [%]
60 50 40 30 20 10 0
500
1000
1500
2000
2500
Wavelength [nm] Figure 9.22 Relative transmission, absorption, and reflection spectra of a 190 nm thick film of CLEVIOS PH 1000 including 5% DMSO deposited on quartz substrate with a conductivity of approximate 1000 S/cm. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
2.0 a
1.5 Abs [A]
U a –1.5 V b –1.0 V c –0.5 V d 0V e 0.5 V
b
1.0
c
0.5
d e
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Energy [eV] Figure 9.23 Optical absorption spectra of a PEDOT:PSS electrode in an electro chemical cell for different applied voltages: (a) –1.5 V, (b) –1.0 V, (c) –0.5 V , (d) 0 V, and (e) + 0.5 V. (Reprinted from J. C. Gustafsson, B. Liedberg, and O. Inganäs, Solid State Ionics 69(2):145–152, In Situ Spectroscopic Investigations of Electrochromism and Ion Transport in a Poly(3,4-ethylenedioxythiophene) Electrode in a Solid State Electrochemical Cell. Copyright 1994, with permission from Elsevier.)
6911X.indb 139
10/1/10 9:00:00 PM
140
PEDOT
CB
VB Neutral state
q = +e , s = ½ polaron
q = +2e , s = 0 bipolaron
High doping level
Figure 9.24 Energy level diagram of conductive polymers. Dashed arrows indicate possible electronic transitions caused by light absorption. The polaron state is formed by a localized hole gaining energy due to the coulombic relaxation of its vicinity. In case of high current densities new levels are created as bipolarons or even new bands within the energetic gap opening multiple possibilities for subband absorption.
complete conversion takes place between the neutral and the oxidized states of the polymer without any side reactions as observed for in situ PEDOT.84 The partly neutralized polymer obtained at a bias of –1.5 V is strongly absorbing in the visible range, while PEDOT in its oxidized form at +0.5 V is almost transparent in this region. This behavior may be exploited in electrochromic applications, like displays and smart windows.85 The two peaks at 0.5 eV and 1.4 eV in Figure 9.23 indicate that the charge inserted electrochemically into the polymer electrode is stored as polarons or even bipolarons at high oxidation levels,77,86,87 creating new electronic states within the energy gap (see Figure 9.24). These new energy levels are broadened due to the amorphous nature of the film, the length distribution of PEDOT segments, and interaction of charge carriers at high doping levels forming bands without detailed structure. The high transparency of PEDOT:PSS films in the visible spectral range makes these films highly interesting as candidates for transparent polymeric electrodes (see Chapter 10). The transmission spectra of Clevios PH 500 and Clevios P AI 4083 deposited as 80 nm thick layers on glass substrates are shown in Figure 9.25a. The two materials vary in their composition ratio of PEDOT to PSS. AI 4083 with a ratio of 1:6 by weight is more transparent than PH 500 with a ratio of 1:2.5. The transmission curves are almost flat with a slight tilt to higher absorption in the red giving the films a light-blue appearance. The fine structure observable in the transmission spectra for both films is due to thin film interference as illustrated by the variation of layer thickness for a PEDOT to PSS ratio of 1:6 in Figure 9.25b.42
6911X.indb 140
10/1/10 9:00:01 PM
141
PEDOT:PSS
100
Transmission [%]
90 80 70
Reference + 80 nm PEDOT:PSS (1:6) + 80 nm PEDOT:PSS (1:2.5)
60 50
300
400
500
600
700
800
Wavelength [nm] (a) 100
Transmission [%]
90
80
70
60
50
300
400
500
Reference + 50 nm + 100 nm + 150 nm
Y(D65/10°) 88.7 88.9 83.7 81.6
600
700
800
Wavelength [nm] (b) Figure 9.25 Transmission spectra of PEDOT:PSS films on ITO-coated glass: (a) 80 nm thickness, different PEDOT-to-PSS ratios of 1:6 and 1:2.5; (b) PEDOT:PSS of ratio 1:6, layer thickness of 50, 100, and 150 nm, film’s luminous transmission Y determined for a D65-illuminant and a 10° observer. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
6911X.indb 141
10/1/10 9:00:02 PM
142
PEDOT
9.2.3.3 Optical Constants To calculate light out-coupling efficiency in optoelectronic devices comprising thin layers of PEDOT:PSS it will be necessary to determine the spectral dependence of the index of refraction n and the absorption constant k. This has been done in Figure 9.26a for the PEDOT grades. Figure 9.26b additionally
1.6 Refractive Index n
0.14
PEDOT:PSS (1:6) PEDOT:PSS (1:2.5)
0.12 0.10
1.5
0.08 0.06
1.4
0.04
1.3 1.2 300
Absorption Constant k
1.7
0.02 400
500
600
700
800
900
0.00
Wavelength [nm] (a) 0.0
0.2
0.6
In-Situ PEDOT CLEVIOS P (1:2.5) CLEVIOS PH V4 w/o DMSO (1:2.5) CLEVIOS PH V4 + DMSO (1:2.5) CLEVIOS P Al4083 (1:6) CLEVIOS P CH8000) (1:20)
1.6 n @ 550 nm
0.4
1.4
0.20
0.15
0.10
1.2
1.0
0.8
k @ 550 nm
1.8
0.05
0.0
0.2
0.4
0.6
0.00 0.8
PEDOT (wt%) (b) Figure 9.26 (a) Spectral dependence of refractive index (n) and absorption constant (k) for the PEDOT:PSS types 1:2.5 and 1:6; (b) dependence of n and k at λ = 550 nm on PEDOT concentration in the film by weight. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
6911X.indb 142
10/1/10 9:00:03 PM
143
PEDOT:PSS
shows the dependence of n and k on the PEDOT concentration in the films determined at a wavelength of 550 nm.42 The data have been obtained for incident light beam perpendicular to the polymer surface not taking into account optical anisotropy.82 The values for n and k scale linearly with the PEDOT content for PEDOT:PSS as anticipated. The optical constants of PH 500 films do not change significantly in the visible spectral range when high boiling solvents are added to increase the conductivity. The values of n and k for in situ PEDOT were determined for films of PEDOT polymerized from EDOT with Fe(III)-tosylate on quartz substrates following the route described by Gustafsson.83 The high value for k of in situ PEDOT is not in line with the absorption constants of the PEDOT:PSS layers deposited from solution. Most probable is that the oxidation level of in situ PEDOT films is lower than PEDOT that has been polymerized in water in the presence of PSS and this will lead to a higher absorption according to Figure 9.23. 9.2.3.4 Vibrational Spectra The vibrational spectra of in situ PEDOT and PEDOT:PSS have been investigated by IR and Raman spectroscopy.83 A number of well-defined bands assigned to PEDOT vibrations are found in the region of 500 to 2000 cm–1 as depicted in Figure 9.27a,b. The assignment of experimental and calculated Raman bands to intramolecular vibrations has been published by Garreau et al.89 As confirmed by others later in more detail,90–92 the form of spectra and the relative intensity of the peaks depend on the oxidation level of PEDOT. Especially the form of the Raman band associated with C = C symmetrical stretching at 1400 to 1500 cm–1 has been employed to distinguish PEDOT 100
100
80
Transmission [%]
Raman Signal [–]
90
70 60 50
90
80
40 30 3500 3000 2500 2000 1500 1000 500
0
70 2000
1500
1000
Wave Number [cm–1]
Wave Number [cm–1]
(a)
(b)
500
Figure 9.27 (a) Raman spectra for oxidized PEDOT with PSS– as a counterion. (b) IR spectra for PEDOT:PSS. (Reprinted with permission from S. Garreau, G. Louarn, J. P. Buisson et al. In Situ Spectroelectrochemical Raman Studies of Poly(3,4-ethylenedioxythiophene) (PEDOT), Mac romol 32(20):6807–6812. Copyright 1999 American Chemical Society.)
6911X.indb 143
10/1/10 9:00:03 PM
144
PEDOT
being in its benzoid and quinoid form93 and to draw conclusion on secondary doping as discussed in Section 9.3. 9.2.4 Electrical Properties As discussed in Chapter 2, PEDOT:PSS is an intrinsically conducting polymer with metal-like properties. In contrast to solid ion conductors the charge transported here stems from free charge carriers. The thiophene rings form a conjugated π system being heavily p-doped. Owing to the oxidative poly merization reaction triggered by radicals one net free positive charge per three to four thiophene rings is created (see Section 9.1). The poly(styrenesulfonic acid) does not contribute to charge transport directly, but acts as a template to keep PEDOT in the dispersed state and provides film-forming properties. The dissociated sulfonate groups balance the charges of the cationic PEDOT by forming a stable salt. As both the cation and the anion resemble polymeric ions, a complete dissociation does not occur, in accordance with results from the solid-state electrophoresis.94 9.2.4.1 Conductivity The conductivity of PEDOT:PSS layers is usually determined by depositing uniform thin films onto a nonconductive substrate. The sheet resistance, Rsq, is measured via four-point or two-point probes. The resistivity, ρ, or its inverse, the conductivity, σ, are calculated by multiplying Rsq times the layer thickness, d, according to ρ = σ –1 = Rsq· d. In case of a two-point probe special care has to be taken to distinguish between Rsq and the contact resistance, that is, by varying the electrode distance as described by Giraudet et al. for circular electrodes.95 The contact resistance will become of particular importance for high-conductive PEDOT:PSS films with σ > 100 S/cm. Vacuum deposited Al contacts on PEDOT:PSS should be avoided as the acidity of PSS will oxidize the metal to form a thin, poorly defined and insulating AlOx layer.96,97 No interface corrosion has been observed for other metal contacts on PEDOT:PSS like Au, Ag, Ni, and Ti.42 Typically the layer thickness, d, of thin PEDOT:PSS films is determined by scanning the stylus of a high-resolution mechanical profilometer across a scratch introduced to the film. Alternatively, without destroying the film, ellipsometry is often employed. This requires, however, a detailed knowledge on the optical constants, which make this method difficult to apply when the material’s composition is altered. 9.2.4.2 Microscopic Model for Conductivity in PEDOT:PSS The conductivity of PEDOT:PSS films can be altered by various means: The modification of the ratio of PEDOT to PSS will have direct impact on the
6911X.indb 144
10/1/10 9:00:04 PM
145
PEDOT:PSS
conductivity as it determines the density of charge transporting PEDOT sites. In commercial products the ratio of PEDOT to PSS varies usually between 1:2.5 to 1:20 by weight of compounds and the conductivity changes, as outlined in Table 9.1. The addition of other electrical inert components like polyvinylalcohol98 will reduce conductivity accordingly. The distribution of PEDOT:PSS gel particles in solution will determine the morphology of thin films and will consequently determine conductivity as discussed in Section 9.1. The addition of water miscible high boiling solvents like ethylene glycol or dimethyl sulfoxide to a PEDOT:PSS solution will significantly boost conductivity31 owing to morphological changes in the film as discussed in detail in Section 9.3. Changing the pH value of the solution will also have an impact on conductivity as reported by Aleshin et al.57 The highest conductivity was found at pH values between 0 and 3. The mechanism of charge transport in PEDOT:PSS is still poorly understood. A comprehensive theory describing the charge transport properties especially of strongly doped conjugated amorphous polymers is missing. Owing to its random ordering of charge transporting segments in these systems without having any translation symmetry, the quantum mechanical model of Bloch’s eigenstates commonly invoked to explain electrical properties of crystalline semiconductors and metals are not applicable. Following the concept of charge transport in amorphous inorganic semiconductors, the conduction mechanism of conjugated polymers is commonly discussed in terms of charge hopping between adjacent sites. Segments of conjugated polymers are preferentially forming electronic active sites due to their ability of being easily oxidized and reduced. The frequency of charge transport between adjacent sites depend on its relative energetic position, the distance, and the relative orientation, as outlined by Borsenberger and Weiss, to explain charge transport in molecularly doped polymers.99 First attempts to unravel the mechanisms of conductivity in PEDOT:PSS films have been made by Aleshin et al.57 The authors studied the conductivity and magnetoresistance of PEDOT:PSS as a function of temperature and found that both parameters increase with temperature. The temperature dependence of conductivity was discussed using the model of variable range hopping (VRH)100:
σ(T) = σ (0)·exp(–(T0/T)α),
where σ (0) is the infinite temperature and T0 = 16/(kbN(Ef) ξ³ for α = 1/4. Here ξ is the distance from a site for the electron wave function to decay to 1/e of its value, often referred to as localization length and N(Ef) is the density of states at the Fermi energy. In the meantime there have been several reports focusing on the temperature dependence of the conductivity of PEDOT:PSS.93,101–103 All results are discussed within the framework of the VRH model. The results differ with
6911X.indb 145
10/1/10 9:00:04 PM
146
PEDOT
Table 9.2 Temperature Dependence of PEDOT:PSS Films, Discussed in Terms of the Variable Range Hopping100 Model PEDOT: Layer PSS Ratioa Thickness 1:2.5 1:2.5 1:2.5 1:2.5 1:2.5 1:2.5 1:2.5 1:2.5 1:6 1:6 a b c d
25–40 µm 25–40 µm 10–30 µm 10–30 µm n.d. n.d 27 nm 34 nm 90–100 nm 90–100 nm
pH 1.23 5.2 ~2b ~2b ~2b ~2b ~2b ~2b ~1.8b ~1.8b
Conductivity Enhancing σ_(@RT)/ Agent (S/cm) None None None DMSOc (25%) None EGd (purge) None EGd (20%) None Sorbitol
20.6 0.077 0.8 80 0.4 200 0.015 3.0 0.0011 4.18
T0/K
α
Reference
610 3400 1700
0.52 0.43 ~0.5
57 57 101 101 93 93 102 102 103 103
σ(DC) ∝ T0.54 2927 ~0.5 903 ~0.5 4200 ~0.5 360 ~0.5 3200000 0.25 ± 0.1 2720 0.53 ± 0.02
Ratio of PEDOT to PSS by weight, following material producers specifications. pH-value following material producers specifications, not determined explicitly Dimethyl sulfoxide. Ethylene glycole
PEDOT:PSS grades used and experimental conditions employed as summarized in Table 9.2. It is obvious that all comparative data acquired indicate that the DC conductivity of pristine material is by two to three orders of magnitude lower compared to films treated with conductivity enhancing agents31 as discussed in detail in Section 9.3. Following Table 9.2, almost all data are best modeled with an exponent of α = 0.5. The exponent α refers to the dimensionality (D) of the system and is predicted to be α = 1/4 and α = 1/2 for 3-D systems and 1-D systems, respectively.104 As it is rather speculative to draw conclusions on the microscopic transport mechanism just by interpreting the slope of plot log(σ(T)) over log(T) in a limited temperature range, Nardes et al. have done a regression analysis of their data claiming the conductivity to be 1-D type for the low conductive and 3-D for highly conductive type. In the examples of Table 9.2 the parameter T0 decreases with increasing conductivity, independent of the absolute value. This is in accordance with the VRH model claiming that the localization length ξ will increase as conductivity decreases. All of the data summarized in Table 9.2 report an increase of conductivity with increasing temperature within the entire temperature regime. Although the conductivity of PEDOT:PSS reaches maximum levels of up to 1000 S/cm, its conductivity cannot be considered to be metallic.
6911X.indb 146
10/1/10 9:00:04 PM
147
PEDOT:PSS
9.2.4.3 Free Charge Carrier Mobility The conductivity σ is defined by the product of elemental charge (e), charge carrier mobility (µ), and density of charge carriers (n). In case of hole- and electronconducting materials, both charge carriers species contribute according to
σ = e · µp· np + e · µn · nn.
In case of PEDOT:PSS, only holes contribute to the charge transport. Injected free electrons will immediately recombine at oxidized PEDOT sites, hence the transport of electrons does therefore not contribute to the overall current. The density of holes in PEDOT:PSS can simply be calculated using a geometrical consideration. For highly conductive PEDOT:PSS, the ratio of PEDOT to PSS is 1:2.5 by weight. The density of solid films is approximately 1 g/cm³. Owing to the molecular weight of the monomeric units of PEDOT and poly(styrenesulfonic acid) with 140 and 182 g/mol respectively, the density of EDOT monomer can be estimated to be approximately 1·1021 cm–³. From electrochemical measurement, the level of oxidation per monomer unit is known to be approximately 1 charge per 3 EDOT units as outlined in Section 9.1. Consequently the density of holes in PEDOT:PSS films can be estimated to be np = 3·1020 cm–³. For highly conductive films a conductivity of 1000 S/cm has been obtained (see Table 9.1). The hole mobility in PEDOT:PSS can be calculated to be approximately µp = 20 cm²/Vs for the given conductivity and the estimated hole density. This value is well in accordance with the work of Winther-Jensen et al. who investigated vapor phase-polymerized EDOT.105 For these in situ deposited layers a mobility of µp = 11 cm²/Vs was extracted for PEDOT with ptoluenesulfonate as the counterion. This value was actually considered as a lower limit according to Prigodin and Epstein suggesting that only a part of the free charge carriers within a conductive polymer will contribute to charge transport.106 Significantly higher mobilities for PEDOT:PSS have been observed by Okuzaki and his groups.102,107 The authors exploited the material as channel polymer in field effect transistors (FETs) and extracted a mobility of 170 cm²/Vs. Hsu et al. considered data of PEDOT:PSS mobility extracted from FETs to be not of physical relevance108 as ion migration dictate the observed field effect (see Chapter 10). It should be pointed out that for organic semiconductors mobilities of up to 15 cm²/Vs at room temperature have been reported.109 However, these high values relate to single crystals with translation symmetry in contrast to PEDOT:PSS as an amorphous solid. Additionally, free-charge carrier interaction can be neglected in such crystalline systems, whereas in PEDOT:PSS such an interaction is anticipated due to the high density of oxidized states. This undetermined contribution makes it difficult to translate conductivity models applied for organic semiconductors to PEDOT:PSS.
6911X.indb 147
10/1/10 9:00:04 PM
148
PEDOT
9.2.4.4 Threshold Currents High-conductive PEDOT:PSS enables large current densities even in thin films. Stable current densities of approximately 10000 A/cm² can be obtained as long as the charge injecting contacts persist. In Figure 9.28 the current through a 140 nm thick film has been monitored while applying a constant bias of 8.0 V. When the PEDOT:PSS layer is maintained in inert atmosphere almost no change of resistivity occurs, whereas when left in ambient conditions a continuous decrease can be seen. The latter is related to oxidative degradation of PEDOT and not to current triggered processes (see Section 9.2.2.1). Simple device configurations have been proposed to employ the interfaces of PEDOT:PSS to metal as active components for write-once read-many-times electrically addressable memories (WORM). The interface acts as a microfuse changing from high to low conductivity when a voltage above threshold has been applied. Möller et al. discussed the process by dedoping of PEDOT at the interface.110 Later the results were reinterpreted as a migration of PSS chains to the anode creating a thin insulating layer.111 This result is not anticipated in account of early work by Inganäs and Ghosh claiming a tight attachment 120
Current Density [A/mm2]
100
In N2
80 60
In Air
40 Au
20 0
0
100
200
300
PEDOT:PSS Glass 400
Au
500
Time [h] Figure 9.28 A constant bias of U = 8 V is applied to a thin films of PEDOT:PSS including 5% DMSO and the current density is monitored over time. One sample has been kept in inert atmosphere, while the other was measured in plain air. The inset illustrates the experimental setup, the channel length, and width between evaporated gold contacts being L = 10 mm and W = 2 mm and polymer film thickness: d = 140 nm. (Data from A. Elschner, H.C. Starck Clevios GmbH. Unpublished results.)
6911X.indb 148
10/1/10 9:00:05 PM
PEDOT:PSS
149
of the PEDOT segment to the PSS counterion.94 Others found evidence of PEDOT dedoping at the cathode when PEDOT:PSS is sandwiched between two electrodes and high current densities are introduced.112,113 The change of interface conductivity of another WORM device comprising an approximately 250 nm thin layer of PEDOT:PSS sandwiched between Au electrodes was discussed by de Brito et al. in a different way. The authors observed delamination between Au and the polymer due to electrolysis of water and subsequent bubble formation.114
9.3 Secondary Doping 9.3.1 Introduction MacDiarmid and Epstein have used the term primary doping for the addition of small nonstoichiometric quantities of a material to conductive polymers that lead to a strong increase in conductivity of these polymers.115 This can be brought about by a redox agent that either oxidizes (p doping) or reduces (n doping) the polymer chain. Alternatively, the protonation of polyaniline is also referred to as doping. The term secondary doping as used by MacDiarmid and Epstein refers to an additive that further increases the conductivity of an already doped polymer by up to several orders of magnitude.115 The main difference between primary and secondary doping is that the first has a reversible effect, whereas the effect of the latter is permanent and remains even when the additive is removed. Occasionally, the term secondary dopant is used more loosely for a second redox treatment of a polymer chain that has already been subjected to a redox agent.116 Generally, the term is used, however, in the way as defined by MacDiarmid and Epstein and will therefore only be used in this more specific sense in this book. Also, the term conductivity enhancement agent is widely used as an alternative for the term secondary doping. In the case of polyaniline doped with d,l-camphorsulfonate the secondary dopant m-cresol leads to conductivities as high as 400 S/cm.115,117 This is due to a strong increase in crystallinity of the prepared films. Already in the dispersion an increase in viscosity was observed hinting at an increase in the hydrodynamic volume of the polymer chain. X-ray diffraction spectra of films clearly show the formation of crystals upon the addition of the secondary dopant. In the case of PEDOT:PSS, a range of different explanations can be found in the literature.93,118 The aim of the following section is first a summary of the observations found by various analytical techniques and followed by their discussion and interpretation resulting in a consistent model for the doping effect.
6911X.indb 149
10/1/10 9:00:06 PM
150
PEDOT
9.3.2 The Chemical Nature of Secondary Dopants in PEDOT:PSS Table 9.3 shows a range of chemicals and their effect on the conductivity of PEDOT:PSS films.31,58,101,119–128 It includes many organic solvents, sugars, polyols, ionic liquids, surfactants, and salts. Whereas some chemicals have no or only small effects, others increase the conductivity by several orders of magnitude. Table 9.3 Additives and Their Effect on the Conductivity of PEDOT:PSS Films
Substance
Example
ConductivityEnhancement Factor (Maximum of Different Values)
Reference
None
19
Monovalent alcohols
Methanol Ethanol Heptanol
Polyols
Glycerol Ethylene glycol Sorbitol meso-Erythritol
5 500 45 100
58, 120 93 31,121 119
Alcohols with a second polar group
2-Nitroethanol Methoxyphenol
100 50
93 93
Hydroxylated ethers
Diethylene glycol
5
58, 118
Hydroxylated thioethers
Thiodiethanol
5
58, 122
Ethers
Tetrahydrofurane
5
101
Amines
Pyridine
None
93
Amides
N,N-Dimethylformamide N,N-Dimethylacetamide N-Methylpyrrolidone
40 100 100
101 93 93, 123
Imide
Succinimide
330
124
Sulfoxides
Dimethyl sulfoxide
800
41, 93
Ketones
Cyclohexanone
None
93
Nitriles
Acetonitrile
None
119
Nitromethane
Nitromethane
None
93
Anionic surfactant
Dodecylbenzosulfonate
500
125
Ionic liquids
1-Butyl-3methylimidazolium tetrafluoroborate
10
127
Salts
Copper(II) chloride
700
128
6911X.indb 150
10/1/10 9:00:06 PM
151
PEDOT:PSS
It should be noted that the pristine PEDOT:PSS dispersions quoted in the different references are not identical so that the reported enhancement factors give only a crude orientation rather than an exact measure of the quality of such an additive as a conductivity enhancement agent. The highest conductivities are currently obtained using dimethyl sulfoxide as an additive.41 The conductivity increase is typically observed when the secondary dopant is added to the aqueous PEDOT:PSS dispersions, and the film is subsequently formed.31 A heating step for the film forming is not obligatory. Drying at room temperature under reduced pressure also leads to the desired effect.126 However, heat treatment can lead to a further increase in conductivity.126 The heat treatment also has the effect that many of the conductivity enhancement agents are removed from the final film. However, there are certain secondary dopants such as sugars, ionic liquids, or salts that remain in the film.127,128 The conductivity enhancement agent can also be applied to the dried film. When a film of pristine PEDOT:PSS is immersed into ethylene glycol and subsequently dried at room temperature or elevated temperatures, a similar conductivity enhancement is achieved.93 Only if the additive is a solid at room temperature such as meso-erythritol, a heating step above melting point of the additive is essential for the conductivity enhancement to be obtained.93 Crispin et al. found that the conductivity increase is not proportional to the additive concentration.118 As shown in Figure 9.29, no effect is observed at 100
Conductivity [S/cm]
10
1
0.1
0.01
1E-3
0.01
0.1
1
10
wt% (DEG) Figure 9.29 Plot of the conductivity of a films based on PEDOT:PSS versus the amount of diethylene glycol introduced to the water emulsion. (Reprinted with permission from X. Crispin, F. L. E. Jakobsson, A. Crispin et al., The Origin of the High Conductivity of Poly(3,4-ethylenedioxythiophene)Poly(styrenesulfonate) (PEDOT:PSS) Plastic Electrodes, Chem Mater 18(18):4354–4360. Copyright 2006 American Chemical Society.)
6911X.indb 151
10/1/10 9:00:06 PM
152
PEDOT
concentrations below 0.2% diethylene glycol in the dispersion. For concentrations between 0.2% and 0.8% the doping effect becomes stronger and finally reaches a plateau for concentrations above 0.8%. 9.3.3 Properties of PEDOT:PSS Films Including Secondary Dopants 9.3.3.1 Conductivity as a Function of Temperature Kim et al. have measured the conductivity of pristine PEDOT:PSS films as a function of temperature and compared it to those of films prepared with THF (tetrahydrofuran), DMF (dimethylformamide), and DMSO (dimethyl sulfoxide) as additives. A plot of the reduced activation energy W(T) = d ln σ(T) / d lnT shows whether the resulting film is in its insulating, critical, or metallic state. This information can be obtained from the slope of the curve which is either negative, zero, or positive, respectively. The authors find that all plots have a negative slope indicating that all films are in the insulating regime (Figure 9.30). However, the slope is decreasing for additives with a stronger conductivity enhancement. In the case of DMSO the slope is approaching zero.101 9.3.3.2 X-Ray Diffraction Pristine PEDOT:PSS films were described as essentially amorphous by Kim et al. No changes in the x-ray diffraction pattern were found when
W= d(lnσ)/d(lnT)
PEDOT:PSS (H2O, pristine)
PEDOT:PSS (THF) 1
PEDOT:PSS (DMF)
PEDOT:PSS (DMSO)
50
100 T (K)
300
Figure 9.30 Plot of the reduced activation energy W(T) against the temperature T(K) for PEDOT:PSS samples with and without secondary dopants. (Reprinted from Synth Met 126(2-3):311–316, J. Y. Kim, J. H. Jung, D. E. Lee, and J. Joo, Enhancement of Electrical Conductivity of Poly(3,4ethylenedioxythiophene)/poly(4-styrenesulfonate) by a Change of Solvents. Copyright 2002, with permission from Elsevier.)
6911X.indb 152
10/1/10 9:00:07 PM
PEDOT:PSS
153
conductivity enhancement agents were used.101 Martin et al. also described the films as amorphous with only a small range structural order but reported small but nonnegligible changes in the structural order upon the addition of thiodiethanol or glycerol.58 Overall, no crystallinity is found in films prepared using secondary dopants, which is in contrast to the findings for polyaniline. 9.3.3.3 Optical Characterization of PEDOT:PSS Films Ouyang et al. found no differences in the UV-Vis spectrum between pristine PEDOT:PSS films and those treated with a secondary dopant.93 Petterson et al. have pointed out a slight reduction in the absorbance above 720 nm for PEDOT:PSS prepared using sorbitol as a secondary dopant.129 The authors also used variable-angle spectroscopic ellipsometry (VASE). This technique allows the measurement of absorption constants and refractory indices of anisotropic films parallel and perpendicular to the plane of the film. Using VASE it was found that pristine PEDOT:PSS films are highly anisotropic and show a higher absorption coefficient within the plane than perpendicular to it.130 In films prepared using sorbitol this anisotropy is reduced and the extinction coefficients within the plane are 15% to 20% lower than those found in pristine films.129 Small changes were observed in Raman spectra when a secondary dopant is used. Pristine PEDOT:PSS films show a strong band at 1422 cm–1 with a shoulder at 1445 cm–1. A film of PEDOT:PSS treated with meso-erythritol only shows the absorption at 1422 cm–1 without the shoulder.119 Transmission electron microscopy (TEM) was performed by Louwet et al.123 A homogeneous film was observed in the case of pristine PEDOT:PSS whereas a film treated with N-methyl-2-pyrrolidone (NMP) results in a porous structure with voids of 10 to 20 nm. The authors describe this phenomenon as phase separation. 9.3.3.4 Surface Analysis of PEDOT:PSS Films Jönsson et al. used synchrotron radiation to study the kinetic energy of photoelectrons from the S(2p) core level in PEDOT:PSS films in electron spectroscopy for chemical analysis (ESCA).126 Photoelectrons from the S(2p) core level can be used as a quantitative measure of sulfur atoms in the aromatic thiophene ring of PEDOT and sulfur atoms in the sulfonic acid unit of PSS in the top layer of the film. By variation of the energy of the photons, the escape depth of photoelectrons from the polymer film can be tuned and a depth profile of the film can be obtained. Using this technique the authors demonstrated that pristine PEDOT:PSS films have a PSS rich surface. With a photon energy of 270 eV, 95% of the signal stems from the outmost 15 Å and a clear dominance of PSS is observed (see Table 9.4). When NMP and sorbitol are used as additives and the film is dried at room temperature, no PSS rich
6911X.indb 153
10/1/10 9:00:07 PM
154
PEDOT
Table 9.4 PEDOT-to-PSS Ratios for PEDOT:PSS Films Based on S(2p) Core Level Spectra Recorded by Electron Spectroscopy for Chemical Analysis Based Photon Energy Depth Profiling at Photon Energies of 270 and 1486.6 eV Samples PEDOT:PSS-pristine PEDOT:PSS + additives Heated PEDOT-PSS + additives
PEDOT to PSS Ratio (hv = 270 eV)
PEDOT to PSS Ratio (hv = 1486 eV)
0.12 0.28 0.34
0.22 0.27 0.26
Source: Data from S. K. M. Jönsson, J. Birgerson, X. Crispin, G. Greczynski, W. Osikowicz, A. W. Denier van der Gon, W. R. Salaneck, and M. Fahlman, 2003, Synth Met 139(1):1–10.
surface region is observed and the composition resembles that expected for the bulk of the material. At elevated temperatures the surface appears to contain more PEDOT than the bulk. Photons with an energy of 1486 eV can penetrate deeper into the film. In this case a smaller PSS enrichment is found for the pristine sample. In the treated samples the composition corresponds to that expected for the bulk. Contact angle measurements show that the surface becomes more hydrophobic upon the addition of ethylene glycol. The contact angle of a pristine film is 10° to 12°, whereas that of an ethylene glycol treated film is 20° to 22°.93 9.3.3.5 Atomic Force Microscopy and Scanning Tunnel Microscopy Scanning Tunnel Microscopy (STM) images of PEDOT:PSS films show clusters of high conductivity surrounded by poorly conductive regions. Individual conductive particles in pristine PEDOT:PSS have a size of approximately 10 × 10 nm2. Upon the addition of sorbitol, the apparent size of the conductive particles increases and the particles congregate into larger clusters up to a scale of 100 nm.121 Atomic force microscopy (AFM) topography images of pristine PEDOT:PSS on gold show grains and elongated structures with dimensions of about 20 nm (Figure 9.31a). The corresponding phase image shows particles of 10 nm size (Figure 9.31b). For both the topography and phase image, the contrast between the described structures and their surroundings is weak (10° and less in the phase image). In films prepared with diethylene glycol a much stronger contrast is observed in the topography and the phase image (Figure 9.32a,b). Dark regions can be clearly identified in the topography image with dimensions of about 20 nm. The brighter regions form an extended network. A strong contrast at the border of the islands to the surrounding network is observed in the phase image (with contrast values up to 40°). The authors describe the change induced by ethylene glycol as phase separation.
6911X.indb 154
10/1/10 9:00:08 PM
155
PEDOT:PSS
a
6.2 nm
b 10°
∆z
∆φ
50 nm
50 nm
Figure 9.31 (a) Topography and (b) phase image of a pure PEDOT:PSS film obtained with tapping-mode AFM at a scale of 0.25 × 0.25 µm2. (Reprinted with permission from X. Crispin, F. L. E. Jakobsson, A. Crispin et al., The Origin of the High Conductivity of Poly(3,4-ethylenedioxythiophene)Poly(styrenesulfonate) (PEDOT:PSS) Plastic Electrodes, Chem Mater 18(18):4354–4360. Copyright 2006 American Chemical Society.)
9.3.3.6 Work Function and Electron Paramagnetic Resonance The work function of PEDOT:PSS films with and without additives was examined by Huang et al. using a Kelvin probe.131 It was found that the work function is reduced when sorbitol or glycerol are added.
a
24.6 nm
50 nm
∆z
b
52.6°
∆φ 50 nm
Figure 9.32 (a) Topography and (b) phase image of a PEDOT:PSS film with 5% added diethylene glycol obtained with tapping-mode AFM at a scale of 0.25 × 0.25 µm2. (Reprinted with permission from X. Crispin, F. L. E. Jakobsson, A. Crispin et al., The Origin of the High Conductivity of Poly(3,4-ethylenedioxythiophene)-Poly(styrenesulfonate) (PEDOT:PSS) Plastic Electrodes, Chem Mater 18(18):4354–4360. Copyright 2006 American Chemical Society.)
6911X.indb 155
10/1/10 9:00:10 PM
156
PEDOT
Contrary to these findings, Lai et al. observed no changes in the work function using ultraviolet photoelectron spectroscopy (UPS).132 Also Crispin et al. found the same work function for films prepared with and without diethylene glycol.118 In both cases the density of states drops to zero at 5.0 eV. However, in the range between 5 and 6 eV the density of states is three times larger for PEDOT:PSS films with diethylene glycol than for pristine films. Electron paramagnetic resonance (EPR) spectroscopy allows the measurement of spins in a sample. It was found that the treatment of PEDOT:PSS films with ethylene glycol leads to a reduction in the EPR signal of 50% and therefore a reduction in the spins in the sample.93 9.3.4 Discussion The aim of the following discussion is to take all findings described in Section 9.3.3 into account and to find a consistent explanation for the effect of secondary dopants in PEDOT:PSS. As Ouyang et al. observed, it is not sufficient for a secondary dopant of PEDOT:PSS to have a high dielectric constant.119 Although menthol, acetonitrile, and nitromethane have high dielectric constants (ε) of 38, 38, and 39, respectively, they have no effect as secondary dopants, whereas ethylene glycol with ε = 37 has a strong effect. Crispin et al. concluded that it is the combination of high solubility in water, high boiling point, and a high dielectric constant that leads to a good secondary dopant for PEDOT:PSS.118 Although the concentration of these additives is low in the initial mixture, their concentration increases in the film-forming process due to their low vapor pressure, as compared to the main solvent water. A minimum quantity is required for the initial mixture to achieve the effect.118 In the final film the presence of the conductivity enhancement agent is no longer necessary. Based on these findings it becomes clear that the film forming process is the critical step for the secondary dopants to come into effect. Changes are observed both for the surface and the bulk of the material. On the surface of a pristine PEDOT:PSS sample PSS is found in excess. When a secondary dopant is used, the surface composition resembles the bulk composition.126 Not only does the composition on the surface changes but also the morphology. STM pictures show regions of high conductivity that can be assigned to PEDOT-rich regions and those of low conductivity that can be assigned to PSSdominated regions. The PEDOT-rich regions are about 10 × 10 nm2 in size in pristine PEDOT:PSS films and they aggregate when secondary dopants are used.121 Also the AFM pictures (Figure 9.31 and Figure 9.32) clearly show a morphological change. Bright regions in the phase image can be assigned to hard regions, whereas dark regions are soft. The AFM pictures in Figure 9.32 show a network of hard regions that is formed upon the addition of the secondary dopant. Hard regions are assigned to ordered regions of high PEDOT content, whereas the soft regions are assigned to those of excess PSS. These surface effects are already hinting at an ordering process that can be described as a phase separation.118
6911X.indb 156
10/1/10 9:00:10 PM
PEDOT:PSS
157
Since the conductivity is a bulk property, the effect of the secondary dopant must also be a bulk effect and not limited to the surface. The anisotropy found in the ellipsometry measurements shows that PEDOT chromophores are orientated in the plane of the film. On one hand, this may be due to the processing of the film via spin coating, printing, or doctor blading. On the other hand it is reasonable to assume that the drying process orientates the chromophores parallel to the plane. The degree of this orientation is reduced when high boiling solvents are added. When PEDOT:PSS is deposited from aqueous dispersion, it does not reach an equilibrium state in the solid film. Instead it is “frozen” in a nonequilibrium state.133 The presence of the high-boiling solvent with a high dielectric constant allows the blend to rearrange, which means that the PEDOT chromophores find a new thermodynamically preferable position that is not identical to that obtained in the processing and drying from the aqueous system. This effect can be described as that of a plasticizer.134 Since the polymer chains are charged it is reasonable to assume that the high polarity of the secondary dopant is necessary to interact with the ionic charges of the polyanion and the polycation. This has also been described as a screening between the two oppositely charged polymer chains, which allows the chains to orientate. In many cases, but not all, this plasticizer is then thermally removed and leaves the films in a thermodynamically favorable state.133 The rearrangement is, however, not to such a degree that crystallinity can be observed in x-ray diffraction spectra. In contrast to polyaniline it should be noted that in the case of PEDOT:PSS the counterion PSS is of polymeric nature. Nonetheless, changes in the PEDOT segments are observed. This ordering may also be the reason for the reduced overall spin of the samples. Since polarons have spin ½ and bipolarons are spinless, the ordering effect may increase the number of polarons that can be formed.93 The transport of electric charge in PEDOT:PSS films is performed by a hopping mechanism. The charge is transported from one PEDOT-rich conductive region to the next, bridging an area of less conductivity, which consists of PSS or defects. The PEDOT particles have a much higher intrinsic conductivity than the PEDOT depleted boundary, which is essentially insulating. Consequently, the main obstacle is to transport electric current between the PEDOT-rich particles.103 The ordering of the PEDOT segments in the film results in pathways of very high conductivity, which are only created if the high boiling solvent is present long enough to allow the thermodynamically driven order to occur. The insulation PSS chains are at least in places removed between the PEDOT-rich areas and a higher degree of organization between the PEDOT-rich particles is obtained resulting in a macroscopic increase of conductivity. Ouyang et al. have interpreted the changes observed in the Raman spectra as a change from a benzoid structure toward a quinoid structure of the PEDOT molecules due to a move from a coil structure toward an extended coil or linear structure of PEDOT chains.119
6911X.indb 157
10/1/10 9:00:10 PM
158
PEDOT
In summary, the effect of secondary dopants in PEDOT:PSS on the molecular level appears in some ways to be similar to that of m-cresol in polyaniline. In both cases the plasticizer effect of a very polar additive with low volatility induces reorganization. In the case of PEDOT:PSS no crystallinity is found— possibly due to the polymeric nature of the counterion—but still an ordering effect can be shown. In both cases, the organization on the nanometer scale leads to a macroscopic conductivity increase.
References
6911X.indb 158
1. H. G. Bungenberg de Jong, and W. A. L. Decker. 1936. Zur Kenntnis der lyophilen Kolloide. XXVI. Mitteilung. Über Koazervation III: Komplexkoazervation des Systems Gummiarabikum-Gelatine. Kolloidchem Ber 43(143):213–271. 2. V. A. Kabanov and A. B. Zezin. 1984. Soluble interpolymeric complexes as a new class of synthetic polyelectrolytes. Pure Appl Chem 56(3):343–354. 3. V. A. Kabanov. 2005. Polyelectrolyte complexes in solution and in bulk. Russ Chem Rev 74(1):3–20. 4. K. N. Bakeev, V. A. Izumrudov, S. I. Kuchanov, A. B. Zezin, and V. A. Kabanov. 1992. Kinetics and mechanism of interpolyelectrolyte exchange and addition reactions. Macromolecules 25(17):4249–4254. 5. N. Karibyants, H. Dautzenberg, and H. Cölfen. 1997. Characterization of PSS/ PDADMAC-co-AA polyelectrolyte complexes and their stoichiometry using analytical ultracentrifugation. Macromolecules 30(25):7803–7809. 6. H. Dautzenberg. 1997. Polyelectrolyte complex formation in highly aggregating systems. 1. Effect of salt: Polyelectrolyte complex formation in the presence of NaCl. Macromolecules 30(25):7810–7815. 7. A. Zintchenko, G. Rother, and H. Dautzenberg. 2003. Transition highly aggregated complexes-soluble complexes via polyelectrolyte exchange reactions: Kinetics, structural changes, and mechanism. Langmuir 19(6):2507–2513. 8. A. F. Thünemann, M. Müller, H. Dautzenberg, J.-F. Joanny, and H. Löwen. 2004. Polyelectrolyte complexes. Adv Polym Sci 166:113–171. 9. K. Ogawa, S. Sato, and E. Kokufuta. 2005. Formation of intra- and interparticle polyelectrolyte complexes between cationic nanogel and strong polyanion. Langmuir 21(11):4830–4836. 10. B. Philipp, W. Dawydoff, and K.-J. Linow. 1982. Polyelectrolytkomplexe— Bildungsweise, Struktur und Anwendungsmöglichkeiten. Z Chem 22(1):1–13. 11. N. V. Pogodina and N. V. Tsvetkov. 1997. Structure and dynamics of the polyelectrolyte complex formation. Macromolecules 30(17):4897–4904. 12. H.-G. Elias. 2001. Makromoleküle Physikalische Strukturen und Eigenschaften (Vol. 2, 6th ed.). Weinheim: Wiley-VCH. 13. S. Förster and M. Schmidt. 1995. Polyelectrolytes in solution. In: Advances in Polymer Science, ed. G. Wegner, 120: 53–133, Berlin: Springer. 14. G. S. Nahor and J. Rabani. 1989. Systems containing two oppositely charged polyelectrolytes: behavior of ionic photosensitizers and quenchers. Macromolecules 22(5):2516–2519.
10/1/10 9:00:11 PM
PEDOT:PSS
159
15. D. G. Pfeiffer, R. A. Weiss, and R. D. Lundberg. 1982. Microphase separation in sulfonated polystyrene. J Polym Sci: Polym Phys Ed 20:1503–1509. 16. K. Petrak. 1992. Polyelectrolyte complexes. In: Polyelectrolytes, ed. M. Hara, 265–297, New York: Marcel Dekker. 17. Y. Mir, P. Auroy, and L. Auvray. 1995. Density profile of polyelectrolyte brushes. Phys Rev Lett 75(15):2863–2866. 18. S. S. Saltybaeva, L. A. Bimendina, and E. V. Bekturov. 1979. Properties of solutions of complexes of polyethylene glycol with statistical copolymers of methacrylic acid and methyl methacrylate. Izv Akad Nauk Kazakhskoi SSR Ser Khim 3:26–30. 19. F. Jonas and W. Krafft. EP 440957 (Bayer AG), Prior: December 20, 1990. 20. F. Jonas, W. Krafft, and B. Muys. 1995. Poly(3,4-ethylenedioxythiophene): Conductive coatings, technical applications and properties. Macromol Symp 100:169–173. 21. L. Groenendaal, F. Jonas, D. Freitag, H. Pielartzik, and J. R. Reynolds. 2000. Poly(3,4-ethylenedioxythiophene) and its derivatives: Past, present, and future. Adv Mater 12(7): 481–494. 22. R. C. Weast. 1985. CRC Handbook of Chemistry and Physics, 66th ed. Boca Raton, FL: CRC Press. 23. G. Zotti, S. Zecchin, G. Schiavon, F. Louwet, L. Groenendaal, X. Crispin, W. Osikowicz, W. Salaneck, and M. Fahlman. 2003. Electrochemical and XPS studies towards the role of monomeric and polymeric sulfonate counterions in the synthesis, composition and properties of poly(3,4-ethylenedioxythiophene). Macromolecules 36: 3337–3344. 24. K. E. Aasmundtveit, E. J. Samuelsen, L. A. A. Pettersson, O. Inganäs, T. Johansson, and R. Feidenhans’l. 1999. Structure of thin films of poly(3,4-ethylene dioxythiophene). Synth Met 101(1–3):561–564. 25. J. Hwang, D. B. Tanner, I. Schwendeman, and J. R. Reynolds. 2003. Optical properties of nondegenerate ground-state polymers: Three dioxythiophene-based conjugated polymers Phys Rev B 67(11):115205-1–115205-10. 26. W. Lövenich, H.C. Starck Clevios GmbH. 2008. Unpublished results. 27. M. Lefebvre, Z.Qi, D. Rana, and P. G. Pickup. 1999. Chemical synthesis, characterization, and electrochemical studies of poly(3,4-ethylenedioxythiophene)/ poly(styrene-4-sulfonate) composites. Chem Mater 11(2):262–268. 28. S. Kirchmeyer and K. Reuter. 2005. Scientific importance, properties and growing applications of poly(3,4-ethylenedioxythiophene). J Mater Chem 15(21):2077–2088. 29. F. Jonas and R. Dhein. DE 4229192 (Bayer AG) Prior: September 2, 1992. 30. F. Louwet, E. Van Thillo, and B. Groenendaal. EP1639607 (Agfa-Graevert), Prior: 6 June 2004. 31. F. Jonas, A. Karbach, B. Muys, E. van Thillo, R. Wehrmann, A. Elschner, and R. Dujardin. EP 686662 (Bayer AG), Prior: May 6, 1994. 32. S. Kirchmeyer, K. Wussow, T. Hosomi, and C. Yasuo. EP 1384739 (H.C. Starck GmbH), Prior: July 26, 2002. 33. M. Bandyopadhyay and R. S. Konar. 1974. Thermal decomposition of potassium persulphate in the aqueous media. J Indian Chem Soc LI:722–724. 34. I. M. Kolthoff and I. K. Miller. 1951. The chemistry of persulfate. I. The kinetics and mechanism of the decomposition of the persulfate ion in aqueous medium. J Am Chem Soc 74(7):3055–3059.
6911X.indb 159
10/1/10 9:00:11 PM
160
PEDOT
35. V. Rumbau, J. A. Pomposo, A. Eleta, J. Rodriguez, H. Grande, D. Mecerreyes, and E. Ochoteco. 2007. First enzymatic synthesis of water-soluble conducting poly(3,4-ethylenedioxythiophene). Biomacromolecules 8(2):315–317. 36. Y. Yoshioka and G. E. Jabbour. 2006. Inkjet printing of oxidants for patterning of nanometer-thick conducting polymer electrodes. Adv Mater 18(10):1307–1312. 37. H. G. Müller. 1989. Automated determination of particle-size distributions of dispersions by analytical ultracentrifugation. Colloid Polym Sci 267(12):1113–1116. 38. F. C. Krebs. 2009. Fabrication and processing of polymer solar cells: A review on of printing and coating techniques. Sol Energy Mater Sol Cells 93(4):394–412. 39. D. Hohnholz and A. G. MacDiarmid. 2001. Line patterning of conducting polymers: New horizons for inexpensive, disposable electronic devices. Synth Met 121(1):1327–1328. 40. T. Tachikawa, N. Itoh, S. Handa, D. Aoki, and T. Miyake. 2005. High-resolution full-color polymer light-emitting devices using photolithography. SID-Proceeding 34.3:1280–1283. 41. Manufacturer: H.C. Starck Clevios GmbH, Leverkusen, Germany, www.clevios. com. 42. A. Elschner. 2008. H.C. Starck Clevios GmbH. Unpublished results. 43. A. Elschner and S. Kirchmeyer. 2008. PEDOT-type materials in organic solar cells. In: Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies, ed. C. Brabec, V. Dyakonov, and U. Scherf. Weinheim: Wiley-VCH. 44. C. Jonda, A. B. R. Mayer, U. Stolz, A. Elschner, and A. Karbach. 2000. Surface roughness effects and their influence on the degradation of organic light emitting devices. J Mater Sci 35(22):5645–5651. 45. The TGA analysis has been conducted by Bayer Technology Services, Leverkusen, 2006. 46. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 47. I. S. Kolesov and H. Münstedt. 2003. Stabilität antielektrostatischer Beschichtungen auf der Basis von Polyethylendioxithiophenen. Materialwissenschaft und Werkstoff technik 34(6):542–548. 48. K. Jeuris, L. Groenendaal, H. Verheyen, F. Louwet, and F. C. De Schryver. 2003. Light stability of 3,4-ethylenedioxythiophene-based derivatives. Synth Met 132(3): 289–295. 49. I. Winter, C. Reese, J. Hormes, G. Heywang, and F. Jonas. 1995. The thermal aging of poly(3,4-ethylenedioxythiophene). An investigation by x-ray absorption and x-ray photoelectron spectroscopy. Chem Phys 194(1):207–213. 50. K. Norrman and F. C. Krebs. 2006. Lifetimes of organic photovoltaics: Using TOF-SIMS and 18O2 isotopic labelling to characterise chemical degradation mechanisms. Sol Energy Mater Sol Cells 90(2):213–227. 51. M. Vazquez, J. Bobacka, A. Ivaska, and A. Lewenstam. 2002. Influence of oxygen and carbon dioxide on the electrochemical stability of poly(3,4-ethylenedioxythiophene) used as ion-to-electron transducer in all-solid-state ion-selective electrodes. Sensors and Actuators, B: Chemical 82(1):7–13. 52. K. Kawano, R. Pacios, D. Poplavskyy, J. Nelson, D. D. C. Bradley, and J. R. Durrant. 2006. Degradation of organic solar cells due to air exposure. Sol Energy Mater Sol Cells 90(20):3520–3530. 53. H. S. Kang, H.-S. Kang, J. K. Lee, J. W. Lee, J. Joo, J. M. Ko, M. S. Kim, and J. Y. Lee. 2005. Humidity-dependent characteristics of thin film poly(3,4-ethylenedioxythiophene) field-effect transistor. Synth Met 155(1):176–179.
6911X.indb 160
10/1/10 9:00:11 PM
PEDOT:PSS
161
54. R. Paetzold, K. Heuser, D. Henseler, S. Roeger, G. Wittmann, and A. Winnacker. 2003. Performance of flexible polymeric light-emitting diodes under bending conditions. Appl Phys Let 82(19):3342–3344. 55. U. Lang, N. Naujoks, and J. Dual. 2009. Mechanical characterization of PEDOT:PSS thin films. Synth Met 159(5–6):473–479. 56. H. Okuzaki, H. Suzuki, and T. Ito. 2009. Electromechanical properties of poly(3,4-ethylenedioxythiophene)/poly(4-styrene sulfonate) films. J Phys Chem B 113(33):11378–11383 . 57. A. N. Aleshin, S. R. Williams, and A. J. Heeger. 1998. Transport properties of poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate). Synth Met 94(2):173–177. 58. B. D. Martin, N. Nikolov, S. K. Pollack, A. Saprigin, R. Shashidhar, F. Zhang, and P. A. Heiney. 2004. Hydroxylated secondary dopants for surface resistance enhancement in transparent poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) thin films. Synth Met 142(1–3):187–193. 59. K. E. Aasmundtveit, E. J. Samuelsen, L. A. A. Pettersson, O. Inganäs, T. Johansson, and R. Feidenhans’l. 1999. Structure of thin films of poly(3,4-ethylene dioxythiophene). Synth Met 101(1–3):561–564. 60. K. Su, N. Nuraje, L. Zhang, I.-W. Chu, R. M. Peetz, H. Matsui, and N.-L. Yang. 2007. Fast conductance switching in single-crystal organic nanoneedles prepared from an interfacial polymerization-crystallization of 3,4-ethylenedioxythiophene. Adv Mater 19(5):669–672. 61. P. Adriaensens, R. Carleer, L. Storme, D. Vanderzande, and J. Gelan. 2002. Morphological study of a poly(3,4-ethylenedioxythiophene)/polystyrenesulfonic acid mixture by solid state 13C-CP/MAS NMR relaxometry. Polymer 43(25):7003–7006. 62. G. Greczynski, T. Kugler, and W. R. Salaneck. 1999. Characterization of the PEDOT-PSS system by means of X-ray and ultraviolet photoelectron spectroscopy. Thin Solid Films 354(1,2):129–135. 63. G. Greczynski, T. Kugler, M. Keil, W. Osikowicz, M. Fahlman, and W. R. Salaneck. 2001. Photoelectron spectroscopy of thin films of PEDOT-PSS conjugated polymer blend: A mini-review and some new results. J Electr Spectr Rel Phenom 121(1–3):1–17. 64. P. C. Jukes, S. J. Martin, A. M. Higgins, M. Geoghegan, R. A. L. Jones, S. Langridge, A. Wehrum, and S. Kirchmeyer. 2004. Controlling the surface composition of poly(3,4-ethylene dioxythiophene)-poly(styrene sulfonate) blends by heat treatment. Adv Mater 16(9–10):807–811. 65. N. Koch, Humboldt University Berlin, 2009. Personal communication. 66. C. Ionescu-Zanetti, A. Mechler, S. A. Carter, and R. Lal. 2004. Semiconductive polymer blends: correlating structure with transport properties at the nanoscale. Adv Mater 16(5):385–389. 67. A. M. Nardes, M. Kemerink, R. A. J. Janssen, J. A. M. Bastiaansen, N. M. M. Kiggen, B. M. W. Langeveld, A. J. J. M. van Breemen, and M. M. de Kok. 2007. Microscopic understanding of the anisotropic conductivity of PEDOT:PSS thin films. Adv Mater 19(9):1196–1200. 68. X. Crispin, S. Marciniak, W. Osikowicz, G. Zotti, A. W. Denier van der Gon, F. Louwet, M. Fahlman, L. Groenendaal, F. De Schryver, and W. R. Salaneck. 2003. Conductivity, morphology, interfacial chemistry, and stability of poly(3,4ethylene dioxythiophene)-poly(styrene sulfonate): A photoelectron spectroscopy study. J Polym Sci B: Polym Phys 41(21):2561–2583.
6911X.indb 161
10/1/10 9:00:11 PM
162
PEDOT
69. A. M. Higgins, J. Simon, P. C. Jukes, M. Geoghegan, R. A. L. Jones, S. Langridge, R. Cubitt, S. Kirchmeyer, A. Wehrum, and I. Grizzi. 2003. Interfacial structure in semiconducting polymer devices. J Mater Chem 13(11):2814–2818. 70. A. M. Nardes, M. Kemerink, and R. A. J. Janssen. 2007. Anisotropic hopping conduction in spin-coated PEDOT:PSS thin films. Phys Rev B 76: 085208/1–085208/7. 71. U. Lang, E. Müller, N. Naujoks, and J. Dual. 2009. Microscopical investigations of PEDOT:PSS thin films. Adv Funct Mater 19(8):1215–1220. 72. B. M. Medina, D. Wasserberg, S. C. J. Meskers, E. Mena-Osteritz, P. Bäuerle, and J. Gierschner. 2008. EDOT-type materials: Planar but not rigid. J Phys Chem 112(51):13282–13286. 73. J. J. Apperloo, L. Groenendaal, H. Verheyen, M. Jayakannan, R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni, and J.-L. Bredas. 2002. Optical and redox properties of a series of 3,4-ethylenedioxythiophene oligomers. Chemistry 8(10):2384–2396. 74. E. Reinoldi and P. Bäuerle, March 2005. Personal communication. 75. V. Wintgens, P. Valat, and F. Garnier. 1994. Photochemical generation of radical cations from thiophene oligomers. J Phys Chem 98(1):228–232. 76. H. Bässler. 1998. Electronic Exitations. In: Electronic Materials: The Oligomer Approach, ed. K. Müllen, and G. Wegner. Weinheim: Wiley-VCH. 77. C. Ziegler. 1997. Thin film properties of oligothiophenes. In: Handbook of Organic Conductive Molecules and Polymers, ed. H. S. Nalva, Vol. 3, 678–737. New York: J. Wiley & Sons. 78. J. Gierschner, J. Cornil, and H. J. Egelhaaf. 2007. Optical bandgap of π-conjugated organic materials at the polymer limit: Experiment and theory. Adv Mater 19(2):173–191. 79. J. Cornil, D. Beljonne, and J. L. Brédas. 1998. A quantum chemical approach to conjugated oligomers: The case of oligothiophens. In: Electronic Materials: The Oligomer Approach, ed. K. Müllen and G. Wegner. Weinheim: Wiley-VCH. 80. A. Meyer. 2000. Modifizierte Elektrodenoberflächen aus funktionalisierten Oligo- und Poly(ethylendioxythiophenen). Ph.D. Thesis, University of Ulm. 81. M. Turbiez, P. Frère, and J. Roncali. 2003. Stable and soluble oligo(3,4-ethlenedi oxythiophene)s end-capped with alkyl chains. J Org Chem 68(13):5357–5360. 82. L. A. A. Pettersson, S. Ghosh, and O. Inganäs. 2002. Optical anisotropy in thin films of poly(3,4-ethylenedioxythiophene)-poly(4-styrenesulfonate). Org Electron 3(3–4):143–148. 83. J. C. Gustafsson, B. Liedberg, and O. Inganäs. 1994. In situ spectroscopic investigations of electrochromism and ion transport in a poly(3,4-ethylenedioxythiophene) electrode in a solid state electrochemical cell. Solid State Ionics 69(2):145–152. 84. M. Dietrich, J. Heinze, G. Heywang, and F. Jonas. 1994. Electrochemical and spectroscopic characterisation of polyalkylenedioxythiophenes. J Electroanal Chem 369(1–2):87–92. 85. H. W. Heuer, R. Wehrmann, and S. Kirchmeyer. 2002. Electrochromic window based on conducting poly(3,4-ethylenedioxythiophene)-poly(styrene sulfonate). Adv Funct Mater 12(2):89–94. 86. A. O. Patil, A. J. Heeger, and F. Wudl. 1988. Optical properties of conjugated polymers. Chem Rev 88(1):183. 87. J. Cornil and J. L. Brédas. 1995. Nature of the optical transitions in charged oligothiophenes. Adv Mater 7(3):295–297.
6911X.indb 162
10/1/10 9:00:11 PM
PEDOT:PSS
163
88. L. A. A. Pettersson, F. Carlsson, O. Inganäs, and H. Arwin. 1998. Spectroscopic ellipsometry studies of the optical properties of doped poly(3,4-ethylenedioxythiophene): An anisotropic metal. Thin Solid Films 313–314:356–361. 89. S. Garreau, G. Louarn, S. Lefrant, J. P. Buisson, and G. Froyer. 1999. Optical study and vibrational analysis of the poly(3,4-ethylenedioxythiophene) (PEDT). Synth Met 101(1–3):312–313. 90. S. Garreau, G. Louarn, J. P. Buisson, G. Froyer, and S. Lefrant. 1999. In situ spectroelectrochemical Raman studies of poly(3,4-ethylendioxythiophene) (PEDT). Macromol 32(20):6807–6812. 91. C. Kvarnstrom, H. Neugebauer, N. S. Sariciftci, S. Blomquist, A. Ivaska, H. J. Ahonen, and J. Kankare. 1999. In situ FTIR spectroelectrochemical characterization of poly(3,4-ethylenedioxythiophene) films. Synth Met 101(1):66. 92. C. Kvarnstrom, H. Neugebauer, S. Blomquist, H. J. Ahonen, J. Kankare, and A. Ivaska. 1999. In situ spectroelectrochemical characterization of poly(3,4-ethylenedioxythiophene). Electrochim Acta 44(16):2739–2750. 93. J. Ouyang, Q. Xu, C.-W. Chu, Y. Yang, G. Li, and J. Shinar. 2004. On the mechanism of conductivity enhancement in poly(3,4-ethylenedioxythiophene):poly (styrene sulfonate) film through solvent treatment. Polymer 45(25):8443–8450. 94. S. Ghosh and O. Inganäs. 1999. Self-assembly of a conducting polymer nanostructure by physical crosslinking: Applications to conducting blends and modified electrodes. Synth Met 101(1–3):413–416. 95. L. Giraudet, S. Fauveaux, O. Simonetti, C. Petit, K. Blary, T. Maurel, and A. Belkhir. 2006. Spin-coated conductive polymer film resistivity measurement using the TLM method. Synth Met 156(11–13):838–842. 96. W. Bantikassegn and O. Inganäs. 1997. Electronic properties of junctions between aluminum and doped poly(3,4-ethylenedioxythiophene). Thin Solid Films 293(1–2):138–143. 97. S. K. M. Jönsson, W. R. Salaneck, and M. Fahlman. 2003. X-ray photoelectron spectroscopy study of the metal/polymer contacts involving aluminium and poly(3,4-ethylenedioxythiophene)-poly(styrenesulfunic acid) derivatives. J Mater Res 18(5):1219–1226. 98. J. P. Drummond, S. J. Clarson, J. S. Zetts, F. K. Hopkinsa, and S. J. Caracci. 1999. Enhanced electro-optic poling in guest-host systems using conductive polymerbased cladding layers. Appl Phys Lett 74(3):368–370. 99. P. M. Borsenberger and D. S. Weiss. 1993. Organic Photoreceptors for Imaging Systems. New York: Marcel Dekker. 100. N. F. Mott and E. A. Davis. 1979. Electronic Processes in Non-Crystalline Materials. Oxford: Clarendon Press. 101. J. Y. Kim, J. H. Jung, D. E. Lee, and J. Joo. 2002. Enhancement of electrical conductivity of poly(3,4-ethylenedioxythiophene)/poly(4-styrenesulfonate) by a change of solvents. Synth Met 126(2–3):311–316. 102. S. Ashizawa, Y. Shinohara, H. Shindo, Y. Watanabe, and H. Okuzaki. 2005. Polymer FET with a conducting channel. Synth Met 153(1–3):41–44. 103. A. M. Nardes, R. A. J. Janssen, and M. Kemerink. 2008. A morphological model for the solvent-enhanced conductivity of PEDOT:PSS thin films. Adv Funct Mater 18(6):865–871. 104. L. Zuppiroli, M. N. Bussac, S. Paschen, O. Chauvet, and L. Forro. 1994. Hopping in disordered conducting polymers. Phys Rev B 50(8):5196–5203.
6911X.indb 163
10/1/10 9:00:11 PM
164
PEDOT
105. B. Winther-Jensen, M. Forsyth, K. West, J. W. Andreasen, G. Wallace, and D. R. MacFarlane. 2008. High current density and drift velocity in templated conductive polymers. Org Electron 8(6):796–800. 106. V. N. Prigodin and A. J. Epstein. 2003. Quantum hopping in metallic polymers. Physica B 338 (1):310–317. 107. H. Okuzaki, M. Ishihara, and S. Ashizawa. 2003. Characteristics of conductive polymer transistors prepared by line patterning. Synth Met 137(1–3):947–948. 108. F. C. Hsu, V. N. Prigodin, and A. J. Epstein. 2006. Electric field controlled conductance of metallic polymers in transistor structure. Phys Rev B 74:235219/1–235219/12. 109. V. Podzorov, E. Menard, A. Borissov, V. Kiryukhin, J. A. Rogers, and M. E. Gershenson. 2004. Intrinsic charge transport on the surface of organic semiconductors. Phys Rev Lett 93:086602/1–086602/4. 110. S. Möller, C. Perlov, W. Jackson, C. Taussig, and S. R. Forrest. 2003. A polymer/semiconductor write-once read-many-times memory. Nature 426(6963): 166–169. 111. X. Xu, R. A. Register, and S. R. Forrest. 2006. Mechanisms for current-induced conductivity changes in a conducting polymer. Appl Phys Lett 89(14): 142109/1–142109/3. 112. D. M. Taylor, D. Morris, and J. A. Cambridge. 2004. Time evolution of the electric field at the electrode interfaces with conducting polymers. Appl Phys Lett 85(25):5266–5268. 113. P.-J. Chia, L.-L. Chua, S. Sivaramakrishnan, J.- M. Zhuo, L.-H. Zhao, W.-S. Sim, Y.-C. Yeo, and P. K.-H. Ho. 2007. Injection-induced de-doping in a conjugated polymer during device operation: Asymmetry in the hole injection and extraction rates. Adv Mater 19(23):4202–4207. 114. B. C. de Brito, E. C. P. Smits, P. A. van Hal, T. C. T. Geuns, B. de Boer, C. J. M. Lasance, H. L. Gomes, and D. M. de Leeuw. 2008. Ultralow power microfuses for write-once read-many organic memory elements. Adv Mater 20(19):3750–3753. 115. A. G. MacDiarmid and A. J. Epstein. 1994. The concept of secondary doping as applied to polyaniline. Synth Met 65(2–3):103–106. 116. W. W. Chiu, J. Travaš-Sejdić, R. P. Cooney, and G. A. Bowmaker. 2005. Spectroscopic and conductivity studies of doping in chemically synthesized poly(3,4-ethylenedioxythiophene). Synth Met 155(1):80–88. 117. Y. Cao, G. M. Treacy, P. Smith, and A. J. Heeger. 1992. Solution-cast films of polyaniline: Optical-quality transparent electrodes. Appl Phys Lett 60(22):2711–2713. 118. X. Crispin, F. L. E. Jakobsson, A. Crispin, P. C. M. Grim, P. Andersson, A. Volodin, C. van Haesendonck, M. Van der Auweraer, W. R. Salaneck, and M. Berggren. 2006. The origin of the high conductivity of poly(3,4-ethylenedioxythiophene)poly(styrenesulfonate) (PEDOT-PSS) plastic electrodes. Chem Mater 18(18): 4354–4360. 119. J. Ouyang, C.-W. Chu, F.-C. Chen, Q. Xu, and Y. Yang. 2005. High-conductivity poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) films and its application in polymer optoelectronic devices. Adv Funct Mater 15(2):203–208. 120. W. H. Kim, A. J. Makinen, N. Nikolov, R. Shashidhar, H. Kim, and Z. H. Kafafi. 2002. Molecular organic light-emitting diodes using highly conducting polymers as anodes. Appl Phys Lett 80(20):3844–3846. 121. S. Timpanaro, M. Kemerink, F. J. Touwslager, M. M. De Kok, and S. Schrader. 2004. Morphology and conductivity of PEDOT/PSS films studied by scanningtunneling microscopy. Chem Phys Lett 394(4–6):339–343.
6911X.indb 164
10/1/10 9:00:11 PM
PEDOT:PSS
165
122. M. Satoshi, Y. Kazuyoshi, N. Hiromichi, A. Toshika, and N. Mitsuaki. JP 2007204704 (Shin Etsu Polymer Co Ltd), Prior: February 6, 2006. 123. F. Louwet, L. Groenendaal, J. Dhaen, J. Manca, J. Van Luppen, E. Verdonck, and L. Leenders. 2003. PEDOT/PSS: Synthesis, characterisation, properties and applications. Synth Met 135–136(1–3):115–117. 124. H. Yasushi, M. Yasushi, Y. Kazuyoshi, N. Hiromichi, and A. Toshika. JP 2006 328276 (Shin Etsu Polymer Co Ltd), Prior: May 27, 2005. 125. B. Fan, B. X. Mei, and J. Ouyang. 2008. Significant conductivity enhancement of conductive poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) films by adding anionic surfactants into polymer solution. Macromolecules 41(16): 5971–5973. 126. S. K. M. Jönsson, J. Birgerson, X. Crispin, G. Greczynski, W. Osikowicz, A. W. Denier van der Gon, W. R. Salaneck, and M. Fahlman. 2003. The effects of solvents on the morphology and sheet resistance in poly(3, 4-ethylenedioxythiophene)polystyrenesulfonic acid (PEDOT-PSS) films. Synth Met 139(1):1–10. 127. M. Döbbelin, R. Marcilla, M. Salsamendi, C. Pozo-Gonzalo, P. M. Carrasco, J. A. Pomposo, and D. Mecerreyes. 2007. Influence of ionic liquids on the electrical conductivity and morphology of PEDOT: PSS films. Chem Mater 19(9): 2147–2149. 128. Y. Xia and J. Ouyang. 2009. Salt-induced charge screening and significant conductivity enhancement of conducting poly(3,4ethylenedioxythiophene):polystyrenesulfonate). Macromolecules 42(12): 4141–4147. 129. L. A. A. Petterson, S. Ghosh, and O. Inganäs. 2002. Optical anisotropy in thin films of poly(3,4-ethylenedioxythiophene)-poly(4-styrenesulfonate). Org Electron 3(3–4):143–148. 130. L. A. A. Pettersson, T. Johansson, F. Carlsson, H. Arwin and O. Inganäs. 1999. Anisotropic optical properties of doped poly(3,4-ethylenedioxythiophene). Synth Met 101(1–3):198–199. 131. J. Huang, R. F. Miller, J. S. Wilson, A. J. de Mello, J. C. de Mello, and D. D. C. Bradley. 2005. Investigation of the effects of doping and post-deposition treatments on the conductivity, morphology, work function of poly(3,4ethylendioxythiophene)/polystyrene sulfonate) films. Adv Funct Mater 15(2):290–296. 132. S. L. Lai, M. Y. Chan, M. K. Fung, C. S. Lee, and S. T. Lee. 2003. Concentration effect of glycerol on the conductivity of PEDOT film and the device performance. Mater Sc Engin B 104(1–2):26–30. 133. A. M. Nardes, R. A. J. Jansen, and M. Kemerink. 2008 A morphological model for the solvent-enhanced conductivity of PEDOT: PSS thin films. Adv Funct Mater 18(6):865–871. 134. S. Ghosh and O. Inganäs. 2001. Nano-structured conducting polymer network based on PEDOT-PSS. Synth Met 121(1–3):1321–1322.
6911X.indb 165
10/1/10 9:00:11 PM
6911X.indb 166
10/1/10 9:00:11 PM
10 Applications
10.1 Solid Electrolyte Capacitors 10.1.1 Introduction Since the discovery of the electrical conductivity of doped polyacetylene by Shirakawa, MacDiarmid, and Heeger in 1977, a variety of conducting polymers (Figure 10.1) and many applications for conducting polymers have been developed.1,2 Today, 30 years later, solid electrolyte capacitors are one of the major markets for conducting polymers. Conducting polymers form the cathode in solid electrolyte polymer capacitors. Due to their high electrical conductivity and self-healing property, conducting polymers are more often replacing other solid electrolytes like MnO2 or TCNQ as well as liquid electrolytes. The driving force for the replacement of traditional electrolytes has been the significant reduction of the internal electrical resistance (equivalent series resistance or ESR) of the electrolyte capacitors by conducting polymers.3 General requirements for conductive polymers in electrolytic capacitors are high. Especially temperature stability is a serious problem: Most conductive polymers are not stable under atmospheric conditions even at a relatively low temperature. The performance of conducting polymers has been significantly enhanced within the last three decades to meet the requirements of modern solid electrolyte capacitors: While the initially developed conducting polymer polyacetylene was not sufficiently stable, conducting polypyrroles, polyanilines, and polythiophenes were evaluated to manufacture solid electrolyte capacitors.4–8 In the early 1990s the first tantalum and aluminum polymer capacitors were introduced into the market.9–11 At that time polypyrroles outperformed polyanilines and polythiophenes. Meanwhile the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) and its application to electrolytic capacitors had been developed by Bayer.12–14 In the late 1990s the first solid electrolyte capacitors based on PEDOT were commercialized.15 Soon after, PEDOT became the material of first choice. Processing of 3,4-ethylenedioxythiophene (EDOT) is much
167
6911X.indb 167
10/1/10 9:00:12 PM
168
PEDOT
Polyacetylene
N
N
N
S
S
S
O
O
O
O
O S
S
Poly [3, 4-ethylenedioxythiophene] (PEDOT)
S
S O
Polythiophene
S
S S
Polypyrrole (PPY)
N
S S
O
Polyaniline (PANI)
N N
N
O
N
N
N
O
O
O
O
Figure 10.1 Most common conducting polymers.
simpler than the processing of other conducting polymers: The polymerization of EDOT in the presence of oxidants does not require cryogenic cooling to slow down the reaction speed for an efficient application to electrolytic capacitors. Furthermore, EDOT is not classified as a toxic chemical like pyrrole, thus its handling is much safer. In 1999 serious health risks for Japanese workers handling pyrrole in capacitor manufacturing occurred due to the relatively high vapor pressure of the toxic pyrrole.16 During the last decade, the electronic industry has introduced lead-free soldering for more electronic components.17 Lead-free soldering conducting polymers in electrolytic capacitor materials have to withstand a peak temperature of about 260°C. The high temperature stability of PEDOT was a major argument to substitute the much less stable polypyrrole and polyaniline in electrolytic capacitors.15,18 Figure 10.2 shows an example of the surface resistance of a PEDOT film, which was polymerized on a glass substrate and heated for 60 s in air in cycles of increasing temperature stress. The resistance is stable up to 280°C. PEDOT capacitors can withstand 125°C for several thousand hours if they are encapsulated appropriately. Even an application at 150°C seems to be feasible.19 Today, PEDOT has by far the highest market share for conducting polymers in the capacitor market.
6911X.indb 168
10/1/10 9:00:12 PM
169
Applications
Surface Resistance [ohms/square]
220 200 180 160 220
240
260
280
300
Temperature [°C] Figure 10.2 Change in surface resistance of a chemical polymerized PEDOT film on glass when heated for 60 s in air in cycles of increasing temperature.
10.1.2 Capacitor Basics In principle a capacitor consists of an insulating dielectric between two electrodes. The capacitance (C) of a capacitor is given by
C = ε 0ε r A/d ,
with ε0 the permittivity of the vacuum (8.85 pF/m), εr the relative permittivity of the dielectric material, A the surface of the electrode, and d the thickness of the dielectric. The ongoing trends of miniaturization and cost reduction require an increase of capacitance per volume of a capacitor. According to the previous equation above this can be realized by increasing the surface (A) of the electrode, by decreasing the thickness (d) of the dielectric, and by choosing a dielectric material having a higher relative permittivity (εr). While an ideal capacitor can be characterized solely by its capacitance, a real capacitor is by far more complex. Besides a capacitance (C) the equivalent circuit of a real capacitor consists at least of a resistance in series with the capacitance (Rs), a resistance parallel to the capacitance (Rp), and an inductance (L) (Figure 10.3a). In Rs all ohmic losses in series with the capacitance are added. Rp is associated with the resistance of the dielectric, which typically is very high and responsible for the leakage current of the capacitor. The equivalent circuit of electrolytic capacitors is often simplified to a circuit consisting of the series resistance and the capacitance (Figure 10.3b). Rs is usually referred to as equivalent series resistance (ESR). Since the ESR is typically measured at a high frequency (100 kHz) the contribution of Rp can be neglected because most of the current flow at high frequencies is bypassed through the capacitance. Furthermore the capacitance is measured at a low frequency (120 Hz) where the inductance can be neglected. The ESR of an electrolytic capacitor
6911X.indb 169
10/1/10 9:00:13 PM
170
PEDOT
C C
Rs
Rs
L
ESR
Rp (b)
(a)
Figure 10.3 (a) Equivalent circuit and (b) simplified equivalent circuit of an electrolytic capacitor.
is responsible for how fast the capacitor can be charged or discharged. In modern electronic circuits very low ESR values are required, for example, to guarantee high-speed performance of the integrated circuits. 10.1.3 Design of Solid Electrolyte Capacitors Electrolytic capacitors consist of highly porous anodes, which are made either by sintering a high surface metal powder (tantalum and niobium capacitors) or by etching a metal foil (aluminum capacitors). In Figure 10.4 scanning electron microscope (SEM) pictures of tantalum powders of different surface area are shown. The larger the surface area the higher the specific electrical charge (given in μC/g) and the smaller the average pore size.20 A thin dielectric of tantalum pentoxide (Ta2O5), niobium pentoxide (Nb2O5), or aluminum oxide (Al2O3) is formed by electrochemical oxidation of the anode surface. A very high capacitance per volume is achieved by the high surface area in combination with the thin dielectric, which is in the range of 10 nm to a few hundred nm depending on the required voltage range of the capacitor. The porous structure of the oxidized anode has to be penetrated with a conducting material to form the cathode of the capacitor. Traditional cathode materials like liquid electrolytes, manganese dioxide (MnO2), or
00069350
1 µm
HCST
00102842
1 µm
HCST
00090725
1 µm
HCST
Figure 10.4 Tantalum powders with specific electrical charge of 18,000 μC/g and average pore size of 1500 nm (left), with 50,000 μC/g and 600 nm (middle), and with 150,000 μC/g and 200 nm (right).
6911X.indb 170
10/1/10 9:00:15 PM
171
Applications
Conductivity [S/cm] PEDOT 100 10 1 0.1 0.01
MnO2 Liquid electrolyte
Figure 10.5 Conductivity of electrolytes used for electrolytic capacitors.
conducting salts (TCNQ complexes) have a conductivity of less than 1 S/cm, which limits the ESR performance (Figure 10.5). Since conductive polymers have a 100 to 1000 times higher conductivity, they can reduce the ESR of electrolytic capacitors drastically. Figure 10.6 shows the manufacturing process of a polymer tantalum capacitor. Highly porous tantalum powder is pressed and sintered to an anode pellet. A Ta2O5 dielectric is formed on the surface of the anode pellet by applying a positive voltage to the Ta pellets within an aqueous electrolyte
Press, sinter
Anodization
Powder + lead
Impregnation
Monomer/oxidizer
Anode with oxide layer Chemical in-situ polymerization
Contacts, encapsulation
Anode with polymer cathode
Low ESR polymer electrolyte capacitor
Figure 10.6 Principals of the manufacturing process of conducting polymer tantalum capacitors.
6911X.indb 171
10/1/10 9:00:16 PM
172
PEDOT
like phosphoric acid. The Ta2O5 dielectric grows at a rate of about 1.8 nm/V.21 Thus the thickness of the dielectric can be easily controlled by the applied anodic voltage. The thicker the dielectric the higher is the voltage the finished capacitor is able to withstand. The working or rated voltage of the capacitor is typically two to four times lower than the voltage used for the formation of the dielectric to ensure a high reliability. The conducting polymer cathode is made on the dielectric, for example, by chemical in situ polymerization of a monomer with an oxidizer. A conducting carbon and silver paint is applied to the outside of the anode to realize a highly conductive current collector. Finally the capacitor is contacted to a metal frame and encapsulated. Figure 10.7 shows SEM pictures of a broken tantalum polymer capacitor. A thin PEDOT layer on top of the Ta2O5 dielectric, which covers a tantalum core, is clearly seen. A schematic setup of a polymer tantalum or niobium capacitor is shown in Figure 10.8a. In aluminum stacked-type capacitors the conductive polymer cathode is formed on aluminum anodes made from porous foils. After the application of a carbon and silver coat the anode foils are stacked and encapsulated (see Figure 10.8b). In aluminum winding-type capacitors the porous aluminum anode foil is wound together with a separator and a second aluminum foil (cathode foil) that serves as a current collector similar to the silver coat in tantalum capacitors (Figure 10.8c). After the winding the conductive polymer cathode is deposited in the wound capacitor element and the element is encapsulated into a metal housing (not shown in Figure 10.8c). 10.1.4 Deposition Methods for PEDOT Cathode Ideally the polymer would fill the open pore structure of the porous anode material, which is covered with the dielectric. However, typically only a few
Ta Polymer cathode
1989 : 1
20 µm
49680 : 1
Ta2O5 500 nm
Figure 10.7 SEM pictures of the porous structure of a tantalum anode coated with a PEDOT cathode film at low (left) and high resolution (right).
6911X.indb 172
10/1/10 9:00:18 PM
173
Applications
Porous Ta/Nb anode with polymer cathode
Epoxy encapsulant
Lead frame
Lead frame
Lead
Carbon and silver layer (a)
Porous Al anode foil with polymer cathode
Epoxy encapsulant
Lead frame
Lead frame
Carbon and silver layer (b) Anode lead Cathode lead Separator
Anode foil Cathode foil (c)
Figure 10.8 (a) Schematic setup of a polymer tantalum/niobium, (b) an aluminum-stacked type, and (c) an aluminum-winding type capacitor (housing not shown).
6911X.indb 173
10/1/10 9:00:19 PM
174
PEDOT
tens of a nanometer thick film is realized in today’s processes (see Figure 10.7). The conductivity of the polymer has to be very high to ensure a low ESR of the capacitor. Only the fraction of the surface of the porous anodes body that is covered by conducting polymer will contribute to the overall capacitance. Besides their function as solid electrolytes, conducting polymers are used as barrier and buffer layers on the outside of the porous capacitor body in tantalum, niobium, or aluminum stacked capacitors. The importance of this outer polymer layer is often underestimated. First, the outer polymer layer has to prevent the penetration of carbon or silver particles to the dielectric, which are used as current collector layers on the outside of the capacitor pellet. Direct contact of carbon or silver with the dielectric can induce a very high leakage current. Second, the outer polymer layer has to recover the fragile capacitor body from the damages arising from the mechanical stress during the encapsulation process, which typically is done by pressure molding. Although the polymer layer is not strong enough to prevent damage to the capacitor body, its self-healing ability (see Section 10.1.6) cures defect sites during an aging process after encapsulation. To fulfill these barrier and buffer functions the outer polymer layer has to be a dense layer covering the total outside of the capacitor body. The formation of a conductive polymer cathode in electrolytic capacitors is hindered by the porous structure of the anode, especially for modern materials with very fine pores. In situ polymerization or the deposition of conducting polymer dispersions have to be applied for the formation of the polymer cathode since high performance conducting polymers are not soluble. Until a few years ago it was not possible to realize high capacitance and low ESR solid electrolyte capacitors with conductive polymer dispersions. Therefore, today, in situ polymerization is mainly used for conductive polymer processing in electrolytic capacitors. In situ polymerization can either be conducted by electrolytic oxidation or by chemical oxidation. 10.1.4.1 Electrochemical Oxidative Polymerization In the electrochemical oxidative polymerization a monomeric precursor of the conductive polymer is polymerized at an electrode. During the polymerization, ionic dopants from the electrolyte are incorporated into the polymer. For capacitor application, first, an auxiliary electrode layer has to be deposited on the surface of the insulating dielectric. Such auxiliary electrodes can be made, for example, by chemical oxidative polymerization as described in Section 10.1.4.2 or by deposition of a manganese dioxide layer through pyrolysis.22,23 For electrolytic polymerization the auxiliary electrode is contacted with an external electrode or it is connected with a bridging layer to the anode lead. Polymer built up in the porous anode body by electrolytic polymerization is quite difficult because of the large inner surface and the high aspect ratio of the small pores. The current density on the outer surface of the anode is always orders of magnitudes higher than on the inner surface. Therefore polymer growth is
6911X.indb 174
10/1/10 9:00:19 PM
175
Applications
favored on the outside and is suppressed in the inside of the anode. Especially for highly porous materials, an inner impregnation is very difficult. Moreover, investment in equipment is much higher than for traditional dipping processes and reliable contacting of each individual anode is challenging for large volume mass production. Major advantages of electrolytic oxidative polymerization are the high efficiency of material usage and the formation of a homogeneous outer polymer layer, which show excellent edge and corner coverage. A key factor for the formation of dense outer polymer layers is the concentration of the monomer. At low concentrations the polymer formation is limited by diffusion, which results in inhomogeneous porous layers. Since the solubility of EDOT in water, which is the favored solvent for industrial applications, is very low, outer layer formation with EDOT is hindered. Sufficiently high concentrations of EDOT in water to grow a dense polymer layer can be realized by forming microemulsions through the use of surfactants.24,25 10.1.4.2 Chemical Oxidative Polymerization In the chemical oxidative polymerization a monomeric precursor of the conducting polymer is polymerized by an oxidizer. Ions of the oxidizer or additional ions act as dopants. The monomer and oxidizer can be brought into the porous anode structure either sequentially or as a premixed reactive solution. The polymerization reaction for PEDOT is shown in Figure 10.9. Iron(III) salts like Fe(III) toluenesulfonate are commonly used as oxidizers for the polymerization. p-Toluenesulfonate counterions are incorporated into the polymer to stabilize the positive charges in the PEDOT backbone. S
O 2n Fe(III)
n O
O
O
O
O
S
O
– 2n Fe(II)
O
O
S
S
S O
O
S
S O
n
S O
O
O
O
EDOT
O
O
O S
+ α, n Fe(III) – α, n Fe(II)
O
O–
O
S
+
O S O O– O S
S O
O
S
S O
O
O
O
O
n
O
O
+
S O
– Fe3+ O S O
S
3
Iron(III) p-toluene sulfonate
O
O S O PEDOT
Figure 10.9 Chemical polymerization reaction of EDOT to PEDOT (left) using Fe(III) toluenesulfonate (right) as oxidizer.
6911X.indb 175
10/1/10 9:00:21 PM
176
PEDOT
In the sequential process the anode pellet is, for example, first dipped into an oxidizer solution, then the solvent is evaporated and the pellet is dipped into EDOT. If further impregnation cycles are necessary, residual precursor material and reduced oxidizer salts can be washed out after polymerization to open the pore structure of the anode pellet for new material. There are major disadvantages of the sequential process: Processing time is long because for every cycle two dips are necessary. Moreover, in a sequential process, EDOT and oxidizer cannot really be applied in a stoichiometric ratio. Typically a huge excess of EDOT is used and thus the EDOT usage can be as low as 10% to 20%. If multiple dipping cycles are applied the remaining EDOT has to be washed out with organic solvent after polymerization to open the pore structure again since solubility of EDOT in water is low. Dilution of EDOT with a solvent can help to increase yield of monomer usage but has other drawbacks like the cross contamination of the EDOT bath with oxidizer. Nevertheless the sequential process is widely applied in industry. The main reason is that premixed reactive solutions of EDOT and oxidizer are not sufficiently stable. Moreover typically a better electrical performance of the capacitor is achieved by the excess of EDOT, which favors the formation of nonconjugated EDOT oligomers as intermediates (see Chapter 8). For the premixed solution process, monomer and oxidizer are mixed in a solvent prior to the application to the anode pellet.26,27 In such mixtures EDOT and oxidizer can be used in a stoichiometric ratio to ensure 100% EDOT usage. Then the anode pellet is dipped into the solution and dried. If further impregnation cycles are necessary, residual salt of the oxidizer is washed out after the polymerization. Advantageous for the premixed solution process is that even smallest pores are filled with a reactive mixture by capillary forces and only one dip is necessary for each cycle. Thus EDOT and oxidizer do not have to mix by diffusion inside the anode pellet like in the sequential process. The major drawback of premixed solutions is their limited pot life in the dipping station. For industrial processes a pot life of about one day is desirable to obtain a high yield of material usage. Cooling to cryogenic temperatures can slow down polymerization and thus increase pot life but processing and handling gets difficult and expensive. Addition of a base or less reactive oxidizers can overcome these disadvantages.28,29 It is very difficult to form dense outer polymer layers, which are required on tantalum, niobium, or aluminum stacked capacitors with chemical in situ polymerization processes. In a chemical in situ polymerization, the oxidizer, for example, an iron(III) salt, is the major component. For low ESR performance the chemically reduced oxidizer (iron(II) salt) has to be washed out after polymerization, which results in voids in the conducting outer layer. Moreover a conducting outer layer made by chemical in situ polymerization does not adhere very well to the anode body. Mechanical or heat stress
6911X.indb 176
10/1/10 9:00:21 PM
Applications
177
to the capacitor can easily result in a delamination between the capacitor body and the polymer layer leading to a significant increase in ESR. A faster market penetration of polymer electrolyte capacitors is hindered by sophisticated manufacturing processes of in situ polymerization: The conducting polymer cathode is polymerized in situ within each single capacitor. The control of chemical reactions within millions of small capacitor anodes is not an easy task, especially when many process steps are involved. Side products of the polymerization have to be removed to guarantee a stable performance over the lifetime of the capacitor. Usually the process window for polymerization is very tight to ensure a high yield and constant quality. Thus the scale up of production lines for polymer electrolyte capacitors can be very challenging and expensive. Conducting polymer dispersions have the potential to overcome these issues of in situ polymerization. 10.1.4.3 Conducting Polymer Dispersions The continuous improvement of PEDOT:PSS, or poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), polymer dispersions over the last decade has made the application of these dispersions for polymer capacitors feasible. Waterborne PEDOT:PSS dispersions were developed for the formation of the outer polymer layer first.30 The requirement on conductivity is much lower for this application than for the inner solid electrolyte because the electrical current passes perpendicular to the 5 to 50 microns thick outer polymer layer. Filmforming properties, adhesion to the anode body and edge, and corner coverage, which are critical to guarantee good barrier layer properties, are adjusted by appropriate formulations of PEDOT:PSS. In Figure 10.10 a dense outer layer made of a PEDOT:PSS dispersion on a tantalum capacitor is shown. To further simplify the manufacturing process and to replace the electrolytic or in situ polymerization completely, a nanoscale conducting polymer PEDOT:PSS dispersions for the formation of the cathode layer within the porous structure of electrolytic capacitors was developed.31,32 This cathode layer formation is much more challenging than the build up of the outer polymer layer: The electrical conductivity of the dispersion has to be very high and the dispersion must be capable to impregnate highly porous anode structures. PEDOT:PSS dispersions with mean particle sizes of about 30 nm (Figure 10.11) and conductivity of up to 500 S/cm were developed to realize the formation of a solid electrolyte. The manufacturing of polymer electrolyte capacitors is facilitated significantly by PEDOT:PSS dispersions because all process steps of the chemical or electrochemical polymerization can be substituted by simple coating steps. Since no chemical polymerization takes place during the manufacturing process of the capacitor, the process is easier to control and there are no side products like iron salts, which have to be washed out or which could deteriorate the performance in the finished product. Furthermore, waterborne
6911X.indb 177
10/1/10 9:00:21 PM
178
PEDOT
Outer polymer layer
Porous tantalum capacitor body
5 µm Figure 10.10 Outer polymer layer made by a PEDOT:PSS formulation on a tantalum capacitor body (layer position is indicated by the arrow).
100 4.0
80 70
3.0
60 50
2.0
40 30
1.0
20
Integral Mass Distribution [%]
Differential Mass Distribution [%]
90
10 0.0 0.0
0.020
0.040
0.060
0.080
0 0.10
Particle Diameter [µm] Figure 10.11 Particle size distribution of PEDOT:PSS dispersion (mean particle size 30 nm) used for electrolytic capacitors.
6911X.indb 178
10/1/10 9:00:24 PM
Applications
179
dispersions allow for an environmental friendly production by avoiding the use of hazardous solvents. 10.1.5 Reformation and High Voltage Application Polymer electrolyte capacitors show a very high electrical DC leakage current after deposition of the conducting polymer by in situ polymerization. The high DC leakage is attributed to a deterioration of the insulating properties of the dielectric layer. This deterioration by in situ polymerization happens for all relevant dielectrics like Ta2O5, Nb2O5, and Al2O3. It is independent from the type of polymer (PEDOT, polypyrrole) and takes place for both chemical and electrolytic oxidative polymerization. A reformation or aging process is applied after the polymerization process to overcome this deterioration. In a reformation process, which is typically applied for tantalum capacitors and aluminum stacked-type capacitors, the capacitor element is put into a liquid electrolyte and positive voltage is applied to the anode similar as for the formation of the dielectric. The electrical current, which passes through the deteriorated sites of the dielectric, destroys or overoxidizes the conducting polymer at these sites. Thus the electrical defect sites are isolated. An aging process works similar except that the electrical current is applied to the finished capacitor after its encapsulation without any liquid electrolyte. For thin dielectrics, which are used for low voltage polymer capacitors, the quality of the dielectric oxide layer can be restored by the reformation or aging process; for thicker dielectrics (capacitors with voltage rating of 25 V or higher) the restoration typically fails. The application of an insulating protection layer on the dielectric can help to reduce the degradation of the dielectric by chemical polymerization.33 Although there has been a strong market request to lower the ESR of high voltage solid electrolyte capacitors, the application of polymer electrolyte capacitors has been limited to lower voltages until recently. Remarkable properties were found for cathode layers made by PEDOT:PSS dispersions. In contrast to chemical or electrolytic oxidative polymerization processes, the deposition of PEDOT:PSS dispersions does not degrade the insulating properties of the dielectric oxide layer. Thus a reformation or aging process is not required to ensure low leakage levels. Furthermore the application voltage is not limited any more to voltage below about 25 V but can be extended to high voltages.31 In Table 10.1 experimental results are shown for anodes made of a high voltage grade tantalum powder having a specific capacitance of 30,000 μC/g. The dielectric oxide layer was formed by anodization of the tantalum anodes at 100 V. Capacitors made with conducting polymer dispersion showed a low ESR and highly stable oxide layer. The breakdown voltage of the capacitor, that is, the maximum voltage the oxide layer within the capacitor can withstand, is a measure of the stability or quality of the oxide layer. The closer the breakdown voltage is to the anodization voltage, the better the
6911X.indb 179
10/1/10 9:00:24 PM
180
PEDOT
Table 10.1 Electrical Properties of Capacitors Made by a Polymer Dispersion and by In Situ Polymerizations Conducting Polymer Dispersion Capacitance [µF] ESR [mΩ] Breakdown voltage [V]
In Situ Polymerization without Reformation
15 39 96
In Situ Polymerization with 50 V Reformation
In Situ Polymerization with 75 V Reformation
15 32 62
14 500 89
15 38 0
quality of the oxide layer is after polymer deposition. Capacitors whose cathode layer was formed by in situ polymerization showed a similar low ESR but they were short-circuited. A reformation process during manufacturing increased the breakdown voltage of the capacitors made by in situ polymerization. The higher the reformation voltage, the more the breakdown voltage could be increased. But at a reformation voltage of 75 V, ESR performance was completely destroyed. Even for high voltages the breakdown voltage of a polymer electrolyte capacitor made by the deposition of PEDOT:PSS dispersions can be close to the anodization voltage of the dielectric (see Figure 10.12). Thus the
100 90
BDV chemical in situ
Percentile [%]
80 70
BDV data dispersion
60 50 40
Anodization voltage
30 20 10 0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
Break-down Voltage [V] Figure 10.12 Comparison of breakdown voltage (BDV) of polymer tantalum capacitors having polymer cathodes made by chemical in situ polymerization (typical range) and by polymer dispersions (experimental data points). Tantalum anodes were anodized at 100 V.
6911X.indb 180
10/1/10 9:00:25 PM
181
Applications
Molecules for in-situ polymerization O
O
Ta2O5
Ta2O5
Ta
S Ta
Crack/ pinhole Particles in dispersion
500 nm
Figure 10.13 Model for degradation of high-voltage dielectric oxide layer. Left: scheme of model; right: pinhole painted into SEM picture of broken oxidized Ta particle.
limitation in breakdown voltage of in situ polymerization is overcome by polymer dispersions. In a simple model, cracks or pinholes in the dielectric are assumed to be responsible for this distinct behavior of polymer dispersions (Figure 10.13).31 When in situ polymerization is applied to form a cathode layer on top of the dielectric oxide layer, the monomeric precursors of the conducting polymer can enter these cracks or pinholes and form a conducting path within the dielectric. So the breakdown voltage of the dielectric is reduced. The quality of the oxide can be recovered when a reformation voltage is applied to the dielectric: The resulting electrical current destroys the conducting polymer within the crack or pinhole. For very thick oxide layers, that is, for high voltage applications, the cracks or pinholes get large and thus the resistivity of the conducting film in the fault site lower. Now the total current through the fault site has to be very high to destroy the conducting polymer film. At this high current the conductivity of the polymer cathode layer suffers significantly and ESR increases. Defects in the dielectric are not addressed and the quality of the dielectric is not degraded when polymer dispersions are used because the particles are too large to enter the cracks or pinholes. However, so far cracks or pinholes, which are responsible for the deterioration of the dielectric, have never been observed directly. Defects in the Ta2O5 dielectric induced by the chemical in situ polymerization were proposed to explain the deterioration of the oxide layer despite the fact that Ta2O5 is extremely resistant against almost any chemical attack.34 Another group observed a metal–insulator–semiconductor (MIS) characteristic at the tantalum–Ta2O5–polymer interface.35 Residuals of EDOT or oxidizer could cause surface charges at the Ta2O5–polymer interface. These charges could change the energy barrier at the interface and facilitate the current
6911X.indb 181
10/1/10 9:00:25 PM
182
PEDOT
flow through the MIS structure. So far this MIS model does not explain the different reformation behavior for low and high voltage polymer capacitors. Whatever the real reason for the degradation of the dielectric during chemical in situ polymerization is, the application of PEDOT:PSS polymer dispersions does not only simplify the manufacturing process of polymer electrolyte capacitors but preserves the quality of the dielectric layer as well. PEDOT:PSS dispersions enable the manufacturing of high voltage polymer electrolyte capacitors for the first time and thus new markets like industrial and automotive applications can be accessed with polymer capacitors in future. 10.1.6 Self-Healing and Thermal Runaway Solid electrolyte capacitors require a self-healing ability due to their large internal surface area. The larger the surface area, the more probable the generation of fault sites in the dielectric, for example, by mechanical or heat stress during the soldering process or by field crystallization during the life time of the capacitor. These fault sites have to be repaired (self-healed) intrinsically to guarantee low leakage current. Similar to reformation or aging, a high electrical current through the fault site destroys the conductivity of the solid electrolyte close to the fault site and thus the fault site is isolated. Self-healing of manganese (MnO2), which is used in traditional tantalum capacitors as solid electrolyte, is well known.36 Conducting polymers have a similar selfhealing ability. The thin layer of conducting polymer on the dielectric can easily be destroyed or overoxidized locally at the fault site.36 If self-healing is not able to isolate the fault site, a thermal runaway can happen. Thermal runaway or ignition is a severe issue for tantalum capacitors having a solid electrolyte made of MnO2. MnO2 tantalum capacitors made of high charge powders in large case sizes could ignite under certain circumstances if they are not properly manufactured. Thermal runaway in tantalum capacitors occurs when the electrical energy stored in the capacitor is set free by an electrical breakdown of the Ta2O5 oxide layer and the resulting Joule heating is sufficient to ignite the porous tantalum body (see Figure 10.14).37 Small fault sites in the oxide layer are self-healed: If the heat generated by the leakage current near the fault site exceeds about 400°C, the semiconducting MnO2 is reduced to the insulating Mn2O3. Thus the fault site is surrounded by an insulating material and the flow of current through the defect is stopped. If the fault site is too large or a breakdown of the dielectric is induced by harsh electrical conditions, the self-healing mechanism is not sufficient to stop the electrical current. As a consequence the temperature reaches a level where the oxide layer is damaged further and the capacitor is shorted. Thermal runaway occurs if the heat generated during the short is sufficient to ignite the porous tantalum body. Oxygen, which is set free by the reduction of MnO2, maintains the ignition. In case the Joule heating by electrical energy is not sufficient, the capacitor will go defect but no ignition will occur.
6911X.indb 182
10/1/10 9:00:26 PM
183
Applications
High-leakage current Ta Ta2O5
MnO2
2 MnO2
Defects in Ta2O5 layer
Mn2O3+1/2 O2
Self-healing High temperature Ta2O5 Short
Defect
TaxOy
Ignition
Figure 10.14 Principal mechanism of thermal runaway of tantalum capacitors.
MnO2 cathodes in tantalum capacitors are replaced by conducting polymers to prevent thermal runaway. Contrary to MnO2, conducting polymers do not release oxygen in case of a breakdown and an ignition process is not supported. Therefore conducting polymer tantalum capacitors are much safer than traditional MnO2 tantalum capacitors. 10.1.7 Conclusions The introduction of conductive polymers into electrolytic capacitors was a major breakthrough for the ongoing ESR reduction of aluminum and tantalum capacitors. The performance of polymer electrolyte capacitors has been improved significantly over the last 10 years. Today, their ESR performance outreaches that of traditional solid electrolytic capacitors by far. Conducting polymers allow tantalum capacitors to compete against ceramic capacitors in the market, which requires high capacitance and low ESR. Whereas for tantalum capacitors a reduction of ESR is necessary to meet market requirements, for ceramic capacitors capacitance has to be increased (Figure 10.15). The demand for polymer electrolyte capacitors with low ESR and high capacitance is quite strong although the total market for solid electrolytic capacitors is not growing. A shift is taken from traditional MnO2 tantalum capacitors and wet aluminum capacitors toward polymer tantalum and polymer aluminum capacitors. This trend is due to the ongoing development in microelectronics, which results, for example, in continuous miniaturization, rising clock frequencies and gate counts, and falling logic voltages. Low ESR capacitors are necessary to meet the requirements of such a technology. Furthermore low ESR capacitors increase the power conversions efficiency of portable, battery-powered electronics. Due to its outstanding thermal
6911X.indb 183
10/1/10 9:00:26 PM
184
PEDOT
ESR at 100 kHz [Ω]
1.000 Solid
Elec
0.100
0.010
0.001
Cera
1
mics
troly
te
Polymer Electrolyte
10 100 Capacitance at 120 Hz [µF]
High capacitance and low ESR 1000
Figure 10.15 Performance of ceramic and solid electrolyte capacitors with respect to capacitance and ESR (market trend is indicated by a star).
stability, high conductivity and easy processing PEDOT is the dominant conducting polymer in today’s solid electrolytic capacitor market.
10.2 Through Hole Plating for Printed Wiring Boards Printed wiring boards consist of a substrate, typically a glass fiber-reinforced epoxy laminate, copper foils as conducting layers, and electronic components mounted on the laminate surface. For the majority of printed circuit boards the manufacturing process comprises these process steps: • • • • • • •
Drilling Through hole plating Lamination of the photoresist Exposition Development Etching Rinsing and drying
Holes are drilled into the board to take up leads of electronic components passed through the holes and soldered to the copper laminate wiring. Holes are usually plated with a thin layer of copper, which connect the copper layers on the top and bottom sides of the board, and embedded copper layers in multilayer boards.38
6911X.indb 184
10/1/10 9:00:27 PM
185
Applications
Table 10.2 Processes for Copper Plating Process
Activation
Chemical copper
Palladium salts
Palladium activated direct copper Carbon black activated direct copper Conductive polymer direct copper
Palladium salts Carbon black PEDOT
Copper Deposition Copper sulfate, formaldehyde (reducing agent, EDTA (complexing agent) Copper salts, electrodeposition Copper salts, electrodeposition Copper salts, electrodeposition
In the mid 1990s, Blasberg Oberflächentechnik GmbH, now Enthone GmbH, a company of Cookson Electronics,39 developed a process for the direct metallization of copper through holes in printed wiring boards (PWB) and multilayer structures called DMS-E®. This process employs conductive polymers as template for the galvanic copper deposition and competes with chemical (electroless) copper deposition processes that employ strong complexing agents (like ethylenediamineN,N,N′,N′-tetraacetic acid) and formaldehyde in presence of copper salts and other direct metallization processes. An overview on copper deposition processes is given in Table 10.2. The direct metallization process selectively activates organic surfaces during the permanganate etching step while leaving copper areas unmodified. It avoids formaldehyde and other hazardous or environmentally unfriendly materials and is able to plate highly complex structures. In a 1988 patent application, Blasberg Oberflächentechnik GmbH claimed a process for the through hole plating of printed wiring boards that employs the oxidation of exposed areas in a pretreatment step, the selective formation of a conductive polymer in the pretreated areas, and the use of the conductive polymer template to deposit copper by electroplating.40 Initially, polypyrrole was employed in this process as a conductive polymer. The monomer pyrrole has a high vapor pressure at ambient temperature and readily evaporates from the plating baths, which consequently caused contamination in the bath vicinity and health concerns. EDOT was found to have a significantly lower vapor pressure than pyrrole and subsequently replaced pyrrole as the monomer. However, the low solubility of EDOT in water of approximately 0.1% was an obstacle that was overcome by the development of microemulsions of EDOT, formulations of EDOT with surfactants.41 Acids, typically sulfonic acids, promote the process and simultaneously act as charge-balancing counterions during polymerization.42,43 Preferred process conditions employ a scheme of several rinsing steps.44 Phosphoric acid as an additive efficiently nucleates the polymer film formation.45 In parallel some attempts were made to extend the direct copper metallization process to the direct structuring of printed wiring boards.46,47
6911X.indb 185
10/1/10 9:00:27 PM
186
PEDOT
Triggered by the technical progress and significantly driven by environmental arguments, electroless copper deposition was partially replaced by direct metallization, which gained a market share of 5% to 8% worldwide in 2004.48,49 The limited market penetration has been attributed to the necessity to install new equipment and limited cost savings when compared to the chemical electroless plating process. The direct metallization by PEDOT has technical advantages in the plating of complex structures like microvias and vias with high aspect ratios, which are difficult to plate with other metallization processes. With an increasing demand for such structures the market share of the direct metallization process may also increase. The principal processing steps of the direct metallization process are illustrated in Figure 10.16. The polymer based direct metallization involves the following main steps: • • • •
Conditioning and oxidative etching using potassium permanganate Dipping the board into a monomer microemulsion Polymerization of the monomer in an acid containing fixing bath Galvanic copper plating
The conditioning step serves the purpose to provide a clean polymer surface that allows subsequent permanganate etching for a continuously adhering manganese dioxide layer. In practice the manganese dioxide formation is complicated by the fact that glass fibers contained in the epoxy laminate will not be chemically attacked by permanganate. Drilling hole
Copper
Epoxy resin (a)
(b)
(c)
Manganese dioxide
Conductive polymer (PEDOT)
Copper
Figure 10.16 Principal process steps of the DMS-E direct metallization process. (a) Swelling and etching with aqueous potassium permanganate, (b) dipping in 3,4-ethylenedioxythiophene microemulsion, and fixing with sulfonic acid (c) galvanic copper.
6911X.indb 186
10/1/10 9:00:28 PM
187
Applications
The immersion to permanganate induces a redox reaction in which a thin layer of the organic surface is oxidized and permanganate is reduced to manganese dioxide. The manganese dioxide is subsequently deposited on top of the organic surface and serves as an oxidizing agent in the subsequent conductive polymer deposition steps. During the conductive polymer (PEDOT) deposition the etched board is dipped into the monomer microemulsion. An amount of monomer is deposited in the etched via or hole to be plated and subsequently polymerized. In the absence of acid the precipitated manganese dioxide and the monomer do not react. The polymerization reaction is therefore initiated during dipping into the acid containing fixing bath. In the presence of acid the manganese dioxide reduces to soluble manganese(II) and simultaneously polymerizes the monomer EDOT to the conductive polymer PEDOT. The acid anions serve as charge-balancing counterions and get incorporated into the polymer. The chemical nature of the acid employed will influence the nucleation and the crystallization of the polymer. Morphology and achievable conductivity of the resulting polymer will therefore be critically influenced by the acid involved.50 The mechanism of the electrochemical polymerization employing emulsified EDOT monomers has been studied in detail51–53 and is assumed to proceed similarly during chemical polymerization. The microemulsions containing surfactant and monomer show a complex phase behavior of lamellar liquid crystalline phases in intermediate regimes between oil in water and water in oil phases. A model of the phenomena during polymerization from the micellar medium is shown in Figure 10.17. EDOT Surfactant
PEDOT Figure 10.17 Model of the phenomena during polymerization of EDOT from the micellar medium. (Adapted from V. Tsakovka, S. Winkel, and J. W. Schultze, 2001, Electrochim Acta 46(5):759–768.)
6911X.indb 187
10/1/10 9:00:30 PM
188
PEDOT
KMnO4 + 2 H2O + epoxy resin KMnO4 + 4 H+ + 3 e– MnO2 + 4 H+ + 2 e– EDOT + m(z/y) HyA Cu2+ + 2 e–
MnO4 +
KOH +
MnO2 + 2 H2O + Mn2+ + 2 H2O
Oxidation products from epoxy resin
(a)
K+
(b) (c)
mz+
(EDOT)n + m(z/y) Ay– + 2n H+ + (mz + 2n) e– (d) Cu
(e)
Figure 10.18 Chemical reactions involved in the through hole metallization process (a) swelling and etching with aqueous potassium permanganate, (b) deposition of manganese dioxide, (c) reduction of manganese dioxide during fixing, (d) polymerization of EDOT during fixing, and (e) galvanic copper.
During the early reaction stage protons and electrons migrate through the phase boundary between EDOT and the manganese dioxide, and start the polymerization reaction. Following the initial reaction water and manganese(II) ions will move through the formed polymer film while anions get incorporated into the polymer. The polymerization was estimated to proceed with approximately 10 nm polymer per second. The speed of layer formation decreases with increasing film thickness.54 The electrochemical reactions occurring during the polymerization reaction are depicted in Figure 10.18.
10.3 ITO Substitution Transparent conductive oxides (TCOs) are widely used as conductive electrodes, for example, in solar cells, flat panel displays, touch-sensitive control panels, to shield radiation (i.e., for thermally insulating architectural glass), electromagnetic shielding windows, low-emissivity windows in refrigerators, or antistatic coatings to name the most important applications.55 All these devices have in common the need for materials that combine optical transparency with electrical conductivity or the ability to shield from electromagnetic radiation. These requirements limit the material’s choice. Thin evaporated metal films are excellent electrical conductors, however, they suffer from high absorption coefficients in the visual spectral range. Metal oxides as indium tin oxide (ITO), tin oxide doped with antimony (ATO) or fluorine (K-glasses), or zinc oxide doped with Al (AZO) have turned out to be the most prominent candidates for transparent conductive coatings.56 These TCOs comprise properties as wide bandgap with ∆E > 3.1 eV, metallic cations with filled d-shells to avoid interband transition, high doping levels in the
6911X.indb 188
10/1/10 9:00:33 PM
189
Applications
order of 1020 to 1021 cm–3, and high electron or hole mobilities (50–100 cm²/Vs) necessary to fulfill the demands as transparent electrode.57 The overall acceptance of TCO films is based on its good mechanical, chemical, and thermal stability as well as the fact that TCOs can be structured to fine lines by employing standard photolithography, compatible with semiconductor processing lines. Some limitations for TCOs drive the demand for alternative materials. Especially the deposition process either by plasma sputtering or by chemical vapor deposition, which are necessary to obtain high quality films, will add to the overall costs for films significantly. Other deposition processes such as magnetron sputtering, which is employed, for example, for the deposition of AZO, proceed faster but costs for obtaining high vacuum and elaborate equipment will remain.56 The use of materials that can be directly coated from solution like intrinsic conductive polymers,58 single-wall carbon nanotubes,59 or solution-processable nanoparticle dispersions60 are considered as an alternative approach to reduce production costs. Besides the opportunity to employ continuous production processes such as roll-to-roll coating, especially wet processable intrinsic conducting polymers have the advantage of being flexible and forming very smooth film surfaces. Thin film coatings need to be highly flexible to meet the mechanical requirement when bending or rolling devices, for example, in flexible displays or touch-sensitive control panels. In such cases, TCO layers are disadvantageous since they tend to decrease and ultimately lose conductivity over time upon continuous bending. Microcracks will be formed due to the inherent brittleness of crystalline oxides.61 Paetzold et al. have addressed this issue by comparing the conductivity of ITO and PEDOT:PSS coatings on polyethylene terephthalate (PET) substrates under oscillating deformation (see Figure 10.19).62 Cairns and Crawford applied mechanical and thermal stress on ITO- and PEDOT:PSS-coated PET and observed poor mechanical but excellent thermal, electrical and optical properties for ITO on PET. For PEDOT:PSS on PET the result was vice versa: Good mechanical properties but severe degradation at elevated temperature and inferior electrical and optical properties.63 For transparent conductive electrodes two materials properties are most relevant: Optical transparency in the visible spectral range and electrical conductivity. The optical transmission of a layer depends on the absorption coefficient (α) and the layer thickness (d) according to
T = T0 · e−αd,
whereas the absorption coefficient (α) and extinction coefficient (k) (see Chapter 9) are related: α = 4πk/λ. The resistivity and its inverse the conductivity are determined by the product of sheet resistance (Rsq) times the layer thickness (d):
6911X.indb 189
ρ = σ –1 = Rsq · d.
10/1/10 9:00:33 PM
190
PEDOT
Sheet Resistance [Ohm/sq]
PET//ITO PET//PEDOT/PSS 104
103
102
5
10
15
20
25
30
35
40
45
50
Radius of Curvation [mm] (a) 102
Relative Resistance
PET//ITO (100 nm) PET//PEDOT/PSS (100 nm)
101
Bending radius: 8 mm 100 100
101
102
103
104
105
Number of Cycles (b) Figure 10.19 (a) The sheet resistance of PEDOT:PSS spin coated on PET and ITO-coated PET has been compared as a function of bending radius. At a radius or r = 8 mm the resistance of ITO-coated PET film shoots up due to crack formation within the layer. The resistivity partly recovers when the film is stretched again. (b) The change of relative resistance is monitored when the two films are periodically bent to a bending radius of r = 8 mm. The resistivity of PEDOT:PSS does not change, whereas ITO coated PET exhibits a continuous increase. (Figures adapted from R. Paetzold, K. Heuser, D. Henseler et al., 2003, Appl Phys Lett 82(19):3342–3344.)
6911X.indb 190
10/1/10 9:00:34 PM
191
Applications
1.0
100 200
Y/YSubstrate
0.8
50
500
0.6 0.4
Clevios PH1000 σ = 1000 S/cm In situ PEDOT σ = 1200 S/cm ITO σ = 2000 S/cm
1000
0.2 0.0
10–1
100
101
d/nm
102
103
104
Rsq\(91)Ωsq\(93) Figure 10.20 Calculated relative luminous transmission Y/Ysubstrate of ITO, PEDOT:PSS and in situ PEDOT versus sheet resistance Rsq. The layer thickness (d) is the parameter that determines Rsq and Y, d has been assigned by asterisks for PEDOT:PSS. (Data from A. Elschner, 2009. Unpublished results.)
By increasing the layer thickness, the sheet resistance and transparency will drop both. In Figure 10.20, the transmission of ITO, PEDOT:PSS, and in situ PEDOT have been calculated in terms of luminous transmission (Y) for increasing layer thickness and plotted against sheet resistance.64 Although this plot does not take into account thin film interferences, it allows to estimate the level of transmission for a given sheet resistance. For ITO on PET the conductivities were assumed to be 2000 S/cm. This value is significantly lower than conductivities reported for ITO deposited on glass, which reaches values as high as 6000 S/cm due to the persistence of glass toward higher processing temperatures. Compared to PEDOT:PSS and in situ PEDOT the transmission curve of ITO (see Figure 10.20) is shifted to lower sheet resistances mainly due to its lower absorption constant. The k values at λ = 550nm for ITO, PEDOT:PSS, and in situ PEDOT have been determined to be 0.016, 0.038, and 0.197, respectively. To compare various materials independent of their layer thickness, R. G. Gordon proposed a classification of coatings by a figure of merit, defined as the ratio of its conductivity and its visible absorption coefficient: σ/α = – {Rsq · ln(T + R)}–1.65 For high quality ITO and for AZO values of 4 to 5 Ω–1 were determined. From Figure 10.20 the values for ITO on PET, PEDOT:PSS, and in situ PEDOT were extracted to be 0.5, 0.1, and 0.03 Ω–1, respectively. Depending on the application either conductivity or transparency may be maximized. For current driven devices, for example, solar cells, electrochromic windows, organic electroluminescent displays, and lamps, where the voltage drop across the conductive electrode needs to be as low as possible,
6911X.indb 191
10/1/10 9:00:35 PM
192
PEDOT
the demand on high-performing low-light-absorbing conductors is still highest. Even most efficient TCO layers do not fulfill these demands completely and therefore need to be combined with metallic support lines (bus bars). These nontransparent, metallic lines contact the TCO layer and distribute the current laterally on large substrate areas to decrease the overall resistance. Unavoidably, such lines will cover some of the active area. To minimize this effect the bus bar grid needs to be optimized when combined with transparent conductive polymers. Compared to TCOs the distance between the individual metal lines has to be chosen smaller to compensate for the lower σ/α. Carter et al. combined thin Au stripes with PEDOT:PSS, which allowed the construction of ITO-free PLEDs.66 Neyts et al. calculated the voltage drop across the transparent anode and compared their results with the actual luminance drop observed in OLEDs.67 Due to the excellent agreement of calculated and actual values, modeling of various bus-bar designs was used to further optimize the performance by balancing between voltage drop across the transparent, conductive layer and shadowing due to the bus bar lines.68 The substitution of ITO by PEDOT:PSS already has been successfully demonstrated in a number of publications. By employing PEDOT:PSS or polyaniline as transparent anode in PLEDs the oxidative degradation rate of the conjugated emitting polymer resulting from the transfer of oxygen atoms stemming from ITO was reduced by several orders of magnitude.69 By adding the high-boiling solvent glycerole to PEDOT:PSS the voltage drop across the anode of small molecule OLEDs was significantly reduced.70 Highconductive PEDOT:PSS having a conductivity of 500 S/cm has been directly compared to ITO anodes in high-efficient pin OLEDs.71 With polymeric anodes even slightly higher device efficiencies were obtained. The lifetime of these devices was found to be independent of the anode material,72 or in other words, the PEDOT:PSS anode did not accelerate OLED degradation relative to commonly employed ITO. PEDOT:PSS on PET has also been successfully patterned by lithographically to define the anode structure73 and PLED efficiencies were found to be similar compared with devices processed on patterned ITO-coated glass substrates. PEDOT:PSS electrodes have also been successfully introduced in organic solar cells (OSCs) as transparent bottom electrodes deposited directly onto the substrate or as transparent top electrodes deposited onto the semiconductor layer as outlined in more detail in Section 10.7. In less sensitive applications like electroluminescent lamps and touch panels that do not require high current densities, the implementation of PEDOT:PSS as a transparent, conductive layer to substitute ITO is more straight forward. Another application for PEDOT:PSS as ITO substitution are touch panels. Touch panels are devices that detect the presence and location of a contact within the screen area.74 Usually touch panels are combined with displays (i.e., LCD displays) to make the display responsive to touches or contacts
6911X.indb 192
10/1/10 9:00:35 PM
193
Applications
with fingers or pens and therefore have to be transparent. Device combinations of displays and touch panels are commonly called touch screens. Two construction technologies for touch panels have been most successful so far: resistive and capacitive detection. Both types of touch panels need transparent conductive layers to fulfill their function. Resistive touch panels are composed of several layers, the essential layers being two conductive layers separated by a narrow gap. Upon contact pressure the two conductive layers will be shortened creating two pairs of voltage dividers. As a response the electrical current will change, which can be detected and processed to determine the contact location. Capacitive touch panels consist of an insulating glass or polymer substrate coated with a conductive film. When a capacitance, for example, a human body, is approaching the surface, the electrical field within the conducting layer is distorted causing a change locally in capacitance. This change in capacitance is detected and processed to calculate the contact location (see Figure 10.21). The main advantages of resistive touch panels over capacitive touch panels is their robustness with encapsulated conductive layers, which are against the environmental influences. With the increasing use of pens to drive touch panels this advantage appears to decrease since pens stress the upper layer by bending with small radius especially at the edges, an area that most software uses for scroll bars. The major advantage of capacitive touch panels is their ability to detect multiple touches, which, for example, may be used as a mean to control a zoom function. High requirements on the electrical and
Flexible film Spacer Conductive coating (a)
Substrate
Conductive coating (b)
Substrate
Figure 10.21 Structure of (a) resistive and (b) capacitive touch panels.
6911X.indb 193
10/1/10 9:00:36 PM
194
PEDOT
mechanical stability have to be met by the conductive layer, however, as the material is exposed to the environment without any protecting layer. The technical requirements on conductive layers for touch panels are quite demanding. Coated films have to exhibit a transparency of minimally 85% up to 90% depending on the end application. To achieve a high accuracy for the detected location the surface resistivity across the coated area has to be highly constant. The projected lifetime of touch panels is from two to five years during which the surface resistance of the coating has to be stable within the specified values. The majority of touch panels fulfill their function with glasses coated with ITO, the flexible films for resistive touch panels are usually made from ITO-coated PET films. The trends driving the motivation to replace ITO in resistive75 and capacitive76 touch panels by polymers have been the costs that ITO sputtered substrates contribute to the overall costs of the touch panels and the resistance of polymers against mechanical stress.
10.4 Antistatic Coatings The generation of electrostatic charges is an undesired event and a major problem during the production of electronic components. Electrostatic charging is a consequence of the triboelectric effect, which charges nonconducting surfaces after contacting a different material.77 Such charges may be generated, for example, when polymeric parts pass machinery. Charges located on a nonconducting surface will produce an electrical field and collect dust. As a consequence the surface will be contaminated and particles may cause shorts, for example, between pins. Electrostatic charges need to be discharged in a controlled way to prevent electrical damages. An uncontrolled electrical discharge may generate high currents and may damage functional layers. As an example electrostatic charges may get in contact with a pin of an integrated circuit containing transistors with MOSFET (metal-oxide-semiconductor field-effect transistor) structures. As the charge generates a voltage exceeding the breakdown voltage of the transistor’s dielectric layer, the charge will discharge through the transistor’s dielectric layer, heat may be generated locally, and the dielectric will be irreversibly damaged (see Figure 10.22). There is no reliable method to predict whether a combination of two materials will generate electrostatic charges. It will therefore always be necessary to take actions to control electrostatic charges. A common measure to control electrostatic charges is the use of antistatic agents either in the bulk or as surface coating. Common antistatic agents are
6911X.indb 194
10/1/10 9:00:36 PM
195
Applications
Polysilicon melt filaments in nMOS output device
(a)
(b) Figure 10.22 (a) Typical ESD failure mode in an nMOS transistor showing gate to drain melt filaments between polysilicon gate and silicon surface, (b) SEM photograph of silicon melting due to current filamentation in an nMOS output transistor, and (c) SEM photograph of an nMOS transistor showing gate oxide damage. (From A. Amerasekera and C. Duvvury: ESD in Silicon Integrated Circuits, 2002. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
6911X.indb 195
10/1/10 9:00:39 PM
196
PEDOT
Gate Source
Drain
(c) Figure 10.22 (Continued)
surfactants like long-chain aliphatic amines and amides, quaternary ammonium salts, esters of phosphoric acid, polysulfonic acids, or polyethylene glycols and their esters. Discharge currents are controlled by the surface resistance measured in the unit Ω/sq (Ohms per square) according to IEC Standard 93 (VDE 0303, Part 30) or ASTM D 257. A common range to discharge electrostatic charges is 105 to 1011 Ω/sq, which allows easy charge migration but avoids high currents. Salt-like and nonionic surfactants have the disadvantage that the mechanism to conduct charges is based on moisture that is absorbed from the surrounding atmosphere. Therefore the surface resistance of protected parts will change with humidity. In dry atmospheres the surface resistance may increase to an unacceptable value and a reliable discharge of electrostatic charges will not be possible. However, in many cases salt-like and nonionic surfactants are readily available at a low price and are therefore widely used for the majority of less critical applications. Another common antistatic agent is carbon black, a form of amorphous carbon with intrinsic conductivity. The conductivity of carbon black, therefore, does not change with humidity. Carbon black is used in coating formulations and as a conductive filler in polymeric materials, especially in the electronics industry. The main drawback of carbon black is its intrinsic black color and the tendency to generate particles that dissipate in the clean room air and may cause undesired shorts.
6911X.indb 196
10/1/10 9:00:40 PM
197
Spectral Internal Transmission T/TSubstrate
Applications
100 95 90 85 80 75 PEDOT:PSS (d = 230 nm) Polyaniline (d = 200 nm)
70 65 60 400
500
600
700 800 Wavelength (nm)
900
1000
Figure 10.23 Comparison of the transmission of PEDOT:PSS and polyaniline.
Following the trend to decrease the size and structures of electronic components like chips and display panels, highly reliable and highly transparent packaging materials were needed. Starting in the late 1980s intrinsically conductive polymers were therefore evaluated as antistatic coatings. The availability of polyaniline and poly(3,4-ethylenedioxythiophene) in organic solution or in aqueous dispersion triggered the formulation of antistatic coatings. Due to the eye’s sensibility to the greenish hue of polyaniline, such coatings are clearly visible, whereas the light blue hue of PEDOT coatings offer the potential of almost invisible transparent coatings (see Figure 10.23). Besides having a surface resistance of approximately 105 to 1011 Ω/sq and being highly transparent, coatings need to be wear-resistant and should have a good adhesion. Coating formulations need to fulfill the requirements of the coating process such as suitable viscosity and good surface wetting. The surface resistance can be adjusted to some extent by varying the layer thickness, but a thicker layer will also decrease the layer transmission (see Figure 10.24). At a fixed transmission it will therefore be necessary to adjust the specific conductivity of the coating formulation. In 1990, the PEDOT:PSS-complex was proposed for antistatic coatings.78 Subsequently a number of formulation components have been developed such as solvents, surfactants, and wetting agents to modify the surface tension and substrate wetting by the coating formulation, binders to increase film forming, promoters to increase adhesion and highly polar additives, so-called conductivity enhancement additives. Nature and function of formulation components suitable for the formulation of PEDOT:PSS will be discussed in more detail in the following sections.
6911X.indb 197
10/1/10 9:00:41 PM
Transmission [%]
198
PEDOT
100 90 80 70 60 50 40 30 20 10 0 300
Layer Thickness: 50 nm 100 nm 200 nm 500 nm 1000 nm
400
500
600
700
800
Wavelength [nm] Figure 10.24 Transmission of PEDOT:PSS at various layer thicknesses.
10.4.1 Solvents The aqueous dispersion of PEDOT:PSS has a surface tension of approximately 65 to 71 mN/m depending on the ratio of PEDOT to PSS. Such a surface tension is too high to wet hydrophobic surfaces such as plastic substrates like polyethylene terephthalate (PET), polycarbonate (PC), or polyethylene (PE) with aqueous PEDOT:PSS dispersion. The addition of solvents will significantly lower the surface tension. Suitable solvents in general are low-boiling solvents miscible with water. Some examples are methanol, ethanol, 2-propanol, and 1-butanol as well as volatile water miscible ketones like methyl ethyl ketone. An example for a formulation containing 2-propanol is given in Table 10.3. As a drawback the addition of solvent may decrease the dispersion stability of the PEDOT:PSS-dispersion. A special conductivity enhancing effect of low volatile polar solvents will be discussed in Section 10.4.5 “Conductivity-Enhancing Additives.” Table 10.3 Effect of the Solvent 2-Propanol on the Surface Tension Coating Solution Aqueous PEDOT:PSSa Aqueous PEDOT:PSS N-Methylpyrrolidone Glycidyloxypropyltrimethoxysilane 2-Propanol
Component (% by Weight) 100.0 42.9 2.6 0.9 53.6
Surface Tension (mN/m) 65.0 25.5
Source: A. Elschner, 2007. Unpublished results. a Clevios P (H.C. Starck Clevios GmbH, Leverkusen, Germany), weight ratio PEDOT:PSS = 2.5:1.
6911X.indb 198
10/1/10 9:00:42 PM
Applications
199
10.4.2 Surfactants Another option to modify the surface tension may be the addition of surfaceactive components. Surfactants reduce the surface tension of the aqueous dispersion by accumulation at the liquid–gas interface. Surfactants are usually nonvolatile and will accumulate in the dried films. Surfactants may be ionic or nonionic compounds containing ethylene oxide groups. Because the solid content of aqueous PEDOT:PSS dispersions is usually rather low (between 1.5% and 4%) surfactants need to be highly effective at extremely low concentrations. For example, a surfactant that has been added to a PEDOT:PSS coating formulation at a concentration of 1% total of the formulation will yield as much as 40% of the solids of the final, dried film. 10.4.3 Binders Polymeric binders are often used to improve the films’ adhesive and mechanical properties. In some cases additional polystyrene sulfonic acid can be used to increase film properties, in most cases polyvinyl alcohol or waterborne polyesters and polyurethanes are more desirable. Special polymers like silicone emulsions79 have been used for antistatic releasing agent compositions. Pendant vinyl groups serve to form a network by addition of Si-H using platinum catalysis. Acrylic binders have been claimed to offer fast drying80 presumably due to their high glass temperature. When plastic substrates are coated by PEDOT:PSS and subsequently mechanically treated after coating, for example, by thermoforming, special binders are needed to prevent cracks in the coating during mechanical treatment and to maintain the overall conductivity.81 In such a case polymeric binders are preferred with a softening point that is below the softening point of the underlying polymer substrate. Especially some polyurethane dispersions appear to fulfill this requirement. The choice of binder, therefore, varies with the targeted properties of the final film. If for example, an end user desires a final film that is resistant to a particular solvent and well-adherent to PC, then a binder designed for solvent-resistant coatings on PC should be chosen. Binders will remain in the film as a nonconductive filler and in general will decrease the film conductivity.82 An example of a PEDOT:PSS polyurethane formulation is shown in Figure 10.25 in which the surface resistance has been recorded for several compositions of PEDOT:PSS and polyurethane binder. It should be noted that the total amount of nonconductive material in the film is higher than the value listed in the X-axis due to the nonconductive polystyrene sulfonic acid contained in the PEDOT:PSS. For example, a value of 50% PEDOT:PSS will result in 86% overall nonconductive material in the film. In some cases small amounts of highly polar binders might increase the conductivity similar to conductivity enhancing additives.83
6911X.indb 199
10/1/10 9:00:42 PM
200
PEDOT
Surface Resistance in MOhm per Square
14 12 10 8 6 4 2 0
0
20
40
60
80
100
% PEDOT:PSS in Dry Film Figure 10.25 Surface resistance of PEDOT:PSS/polyurethane hybrid-film in dependence of film composition.
10.4.4 Hardness and Abrasion Various components have been proposed to increase hardness, solvent resistance, and adhesion to substrates of PEDOT:PSS films. Cross-linking agents that can be cured by UV light84 increase hardness and wear resistance. Water-soluble melamine resins can be used to efficiently increase solvent resistance.85 Melamine resins efficiently cross-link acid catalyzed when heated. It is therefore presumed that for melamine the polystyrene sulfonic acid in PEDOT:PSS acts as a cross-linking catalyst. Silanes like glycidoxypropyltrimethoxysilane, tetraethoxysilane, and methyltrimethoxysilane86 (see Figure 10.26), have been used in formulations to increase the adhesion of PEDOT:PSS films to glass87 and polymeric substrates. During formulation Si-O-C-bonds of the agent hydrolyze in water C2H5
CH3 O
O O
Si(OCH3)3
O
(a)
H5C2 – O – Si – O – C2H5
H3C
Si O CH3
O
O
C2H5
CH3
(b)
(c)
Figure 10.26 Chemical structures of (a) glycidoxypropyltrimethoxysilane, (b) tetraethoxysilane, and (c) methyltrimethoxysilane.
6911X.indb 200
10/1/10 9:00:43 PM
Applications
201
catalyzed by the acidic conditions to form silanoles (Si-OH-groups) that are essentially stable at a moderately low pH. During drying silanol groups will condensate to form a cross-linked network. Hydrolyzed tetraethoxysilane will condensate to silicon dioxide after drying and is therefore an efficient tool to increase the hardness of films88 up to a pencil hardness* of 9H. Using a combination of oligomomeric silanes89,90 to increase flexibility and highly cross-linking low molecular weight silanes, suitable formulations may be obtained for antistatic and antireflective coatings for glass, especially for cathode ray tubes. Silanes and tetraalkylorthosilicates also have been proposed to increase the wear and solvent91 resistance of the conducting film. Since silanes are easily hydrolyzed in acidic solution, care has to be taken to prevent early cross-linking of the resulting silanols before film drying. Uncontrolled condensation might result in larger aggregates that scatter light and decrease transparency. During formulation silanes are therefore frequently hydrolyzed in slightly acidic pure water before further components are added. In a special approach EDOT has been polymerized in the presence of tetraalkylorthosilicates to make antistatic hard coatings with a pencil hardness > 7H.92,93 10.4.5 Conductivity-Enhancing Additives The conductivity of layers made from nonformulated pristine PEDOT:PSS is rather low. With the addition of specific compounds, the layer conductivity can usually be increased by one to two orders of magnitude. This is clearly in contrast to expectation: the addition of nonconductive components is expected at best to not lower the overall conductivity. An enhancement of conductivity is usually observed when a low volatile and highly polar component is present during drying. Due to their effect, the term conductivityenhancing additives has become commonly used. In general, such compounds are highly polar. It has been found that compounds containing hydroxyl94 or amide95 groups are very effective. Particularly useful are amides such as N-methylpyrrolidone and dimethylformamide, polyhydroxy compounds like ethylene glycol and sugar alcohols, and sulfoxides like dimethylsulfoxide. These solvents, often called “secondary dopants,”96 are used in amounts of 1% to 10 % based on the overall composition to increase the conductivity of the final, dried film. The effects of these additives seem to be independent of whether they remain in the film after drying. The mechanism of this conductivity enhancement has been discussed in depth.97,98 The interpretation currently favored is that polar solvents act as plasticizers for PEDOT:PSS after water evaporation from the formulation, thereby creating an opportunity for a more favorable morphological rearrangement of PEDOT:PSS.99,100 This rearrangement leads to a decreased resistance between the dry gel particles, thus increasing the overall conductivity of the film. * DIN EN 13523-4:2001, ASTM D3363.
6911X.indb 201
10/1/10 9:00:43 PM
202
PEDOT
Protection layer Blue light high sensibility Blue light low sensibility Yellow filter Light sensitive front side
Separation layer Green light high sensibility Green light low sensibility Red light high sensibility
Substrate (cellulose triacetate)
Red light low sensibility Mikrat interlayer
Antistatic coating (PEDOT:PSS)
Anti halo layer
Figure 10.27 Antistatic PEDOT:PSS back side coating of a photographic film.
10.4.6 Use of PEDOT in Antistatic Coatings One of the first industrial antistatic uses of PEDOT:PSS was as an antistatic layer in silver halide-based photographic films. Photographic silver halide films consist of a polymeric substrate, usually cellulose acetate, having a light sensitive layer structure on the front side (see Figure 10.27). Electrostatic charges located at the back side of the film that may be generated during its production process may discharge into the photographic layer during winding. Such discharges will result in visible flashes after film development. Such flashes can be avoided when the back side of the photographic film is coated with an antistatic layer. In the 1990s Agfa Gevaert in Belgium (at this time a subsidiary of Bayer AG) introduced PEDOT:PSS-based coatings as antistatic back side layers into industrial photographic films.101 The success of PEDOT:PSS in this application has its origin in its high conductivity, low color, stability, easiness to process, and it is inherently to conductive polymers, a moisture-independent antistatic effect.102–104 While the development of PEDOT-based antistatic layers started in the early 1990s, a few years later PEDOT:PSS hard coatings were developed for cathode ray tubes (CRT).105,106 The outer surface antistatic layer was used to avoid electrostatic shocks and dust contamination during manufacture and use (Figure 10.28). Such antistatic layers were previously made by sputtering a thin layer of indium-doped tin oxide (ITO). It therefore appeared to be highly
6911X.indb 202
10/1/10 9:00:45 PM
203
Applications
CRT glass plate
Antistatic layer PEDOT:PSS-SiO2 hybrid layer SiO2-overcoat
Figure 10.28 PEDOT:PSS as an antistatic coating for cathode ray tubes according to Yoon et al. (KR 2000 009403/US 6630196, Prior: July 24, 1998.107)
attractive in terms of costs to replace this vacuum deposition process by a simple coating process. The availability of coatings incorporating silanes to increase hardness turned out to be a prerequisite for this development.107–109 In addition to the antistatic effect, the PEDOT:PSS layer was found to enhance the optical contrast of the displays. Antistatic layers for optical applications require a low content of large particles, sufficient hardness, high adhesion to glass, and a surface resistance of approximately 106 Ω/sq. During recent years, PEDOT:PSS has gained additional applications in the manufacturing process of flat panel displays. Display components such as the polarizer or the backlight are manufactured separately and finally assembled to fully functional flat panel display. To avoid dust contamination, the components are covered with antistatic protective films that are removed during assembly.79,110,111 Although such films can be also made by in situ polymerization of EDOT,112 coating a suitable PEDOT:PSS formulation clearly is the preferred solution. In addition, antistatic adhesive tapes are used for the assembly of the display.113,114 Static charges from assembly would generate electrical fields that cause optical defects in liquid crystal displays. The layer structure of antistatic protection of the polarizer and adhesive films is depicted in Figure 10.29. Other uses of PEDOT:PSS for antistatic layers are numerous102 and mainly described in patent applications. A few examples are highlighted in the following list: scratch resistant antistatic hybrid coatings polycarbonate sheets,115 antistatic latex gloves coated with PEDOT:PSS fulfilling the tests for surface resistivity, puncture resistance and charge decay,116 antistatic textiles,117 floor coverings,118 and antistatic self-adhesive tapes.119
6911X.indb 203
10/1/10 9:00:45 PM
204
PEDOT
Antiglare coating Polarizer Protection film TAC PVA TAC PSA Color filter glass substrate TFT Glass substrate
Polarizer
prot. film TAC PSA
PEDOT:PSS antistatic coating PEDOT:PSS antistatic coating
Separator release film (removed during assembly) LC LC: TAC: PVA: PSA:
Liquid crystal Cellulosetriacetate Polyvinylacetate Pressure sensitive adhesive
Figure 10.29 PEDOT:PSS antistatic layers in liquid crystal displays.
10.5 Electroluminescent Lamps The use of PEDOT:PSS in electroluminescent lamps may be considered as a specific case of PEDOT:PSS substituting ITO in its function as transparent electrode. Inorganic electroluminescent (EL) devices comprise a composite active layer of a zinc sulfide (ZnS) emitter and a dielectric such as barium titanate sandwiched between two conducting layers, one of which must be transparent.120 When an AC voltage of approximately 100 V/400 Hz is applied electrons are excited in the electrical field within the ZnS particles and recombine by the emission of light. The color of emission can be tuned by the addition of appropriate doping agents to ZnS. Figure 10.30 shows a schematic of the layered structure of a typical EL device in which the transparent electrode is made of PEDOT:PSS. The sheet resistance of the transparent conductors in EL devices can be relatively low, typically 103 Ω/sq, due to the high applied voltage and the low current densities. Therefore it is possible to replace the normally used ITO with conducting polymers such as PEDOT:PSS.121–123 Even though the polymer has a lower conductance compared to ITO, all of the layers in the polymer-based device can be applied by printing techniques such as silkscreen printing. ITO, on the other hand, must usually be applied by costly sputtering deposition techniques. In case the conductivity of the PEDOT:PSS layer is considered to be insufficient Hüppauff et al. proposed to combine the layer with a silkscreen printed metallic bus bar grid.121 Besides lowered process costs, an additional technical advantage is the flexibility of the contact layer. ITO is a brittle, inorganic material not ideally suited to destruction-free thermal deformation. In contrast, devices fabricated
6911X.indb 204
10/1/10 9:00:47 PM
205
Applications
Silver paste electrode Dielectric (BaTiO3) Silver paste bus bar Carrier film
Emitter (ZnS)
PEDOT silk screen paste
Figure 10.30 Schematic structure of a inorganic electroluminesent device; ITO layer substituted by PEDOT:PSS layer from a commercial printing paste formulation (screen printing technique).
with transparent, conductive PEDOT:PSS layers can be thermoformed after construction of the EL elements. The possibility to deep-draw films coated with PEDOT:PSS without increasing the resistance has made this material unique as an transparent electrode for 3D-electroluminescent panels.124,125
10.6 Organic Light Emitting Diodes (OLEDs) 10.6.1 Introduction The observation of electroluminescence in organic solids was first made by Pope et al.126 and Helfrich and Schneider.127 almost half a century ago. It was shown that crystalline anthracene flakes emitted light when an electrical field was applied. This discovery was left almost unrecognized as only low luminous intensities could be achieved, not mentioning the difficulties to grow and handle organic crystals in general. The Kodak researchers Tang and VanSlyke in the late 1980s followed a more reliable approach.128 They fabricated organic light emitting diodes (OLEDs) by depositing thin vacuum sublimed layers of small molecules sandwiched between two electrodes. Their work has paved the way to a new field of solid-state physics and chemistry.129–132 More than 20 years later, a broad understanding on OLEDs has been obtained and OLEDs have become an established technology within the display industry generating a turnover of about $1 billion in 2009 with an anticipated growth to $5.5 billion in 2015.133 The principle of organic electroluminescence is described next. One of its major advantages is the simplicity of the setup, schematically depicted in Figure 10.31a,b. The most simple OLED layer stack shown in Figure 10.31a consists of three layers: a transparent anode, a thin organic light emitting layer of approximate 100 nm thickness, and finally a vacuum deposited metallic cathode.
6911X.indb 205
10/1/10 9:00:47 PM
206
PEDOT
Cathode –
_
_
+
_
+
+
U
+
Transparent anode Substrate (a)
_
_ +
_
+
_
+
+ eU (b)
Figure 10.31 (a) Cross-section of an OLED illustrating the origin of organic electroluminescence, and (b) energy levels of the organic emitting layer tilted by the applied bias. Depicted are hole and electron injection from the two electrodes, charge transport and electron-hole pair formation.
The material forming the transparent anode needs to be electrically conductive and transparent. Typically materials employed here are transparent conductive oxides as indium–tin–oxide or conductive polymers (see Section 10.3). The work function or the oxidation potential of these materials is preferentially high to enable hole injection into the highest occupied molecular orbitals (HOMOs) of the adjacent organic light emitting layer. In contrast the work function of the cathode material should be low to reduce the energy barrier for electron injection into the lowest unoccupied molecular orbitals (LUMOs). Typical cathode materials are vacuum-deposited calcium, barium, or lithium fluoride. These highly corrosive metals make it necessary to encapsulate the finished device thoroughly against oxidation. An external bias has to be applied to the electrodes to generate electroluminescence. In the case of an ideal OLED, only 2 to 4 V are necessary for light
6911X.indb 206
10/1/10 9:00:49 PM
Applications
207
emission, depending on the energy difference between the HOMO and the LUMO levels of the organic material. Electrons and holes are injected simultaneously from the cathode and the anode, respectively, and migrate under the electrical field to their counter electrode as illustrated in Figure 10.31b. When electron and hole densities are sufficient to enable interaction, electron–hole pairs are formed creating typically Frenkel excitons. These excited states localized within a molecular unit can recombine radiation less or preferentially under the emission of fluorescent or phosphorescent light, determined by the emitting material’s properties. The materials employed as light emitting layers need to fulfill certain requirements: They have to be fluorescent or phosphorescent, enable the transport of electrons and holes, and must have the ability to form excitons. Prominent candidates employed as emitting materials are fluorescent dyes like substituted quinacridone134 or phosphorescent dyes like tris(2-phenylpyridine) iridium (Ir(ppy)3)135 doped into a matrix or wide band gap organic semiconductors. In the meanwhile elaborated layer stacks of multiple layers often containing more than just a single component have been designed. Efficiencies of 90 lm/W for white emitting OLEDs have been reached.136 For red emitting OLEDs extrapolated lifetimes of over 100,000 h at 500 cd/m² initial luminance have been reported137 and further progress is anticipated. Parallel to the development of vapor deposited small molecule devices, light emitting materials that can be deposited from solution as first outlined by Burroughes et al.138 are of high interest. Although these devices do not reach the performance of their vacuum deposited counterparts yet, the strategy to employ printing as deposition techniques, like inkjet, nozzle coating, or curtain coating is seen as a possibility to reduce OLED production costs, caused by slow vacuum deposition. Additionally, solution-based deposition techniques do not require shadow masks. Shadow masks so far are limiting the device size of evaporated small molecule OLEDs in a production process. 10.6.2 PEDOT:PSS as a Hole-Injection Layer Waterborne dispersions of PEDOT:PSS have gained considerable interest in OLEDs in two respects. First, as a high-conductive material to substitute ITO as outlined in Section 10.3; and second as an interlayer to improve hole injection and to smooth the rough surface of the anode.139 The role of PEDOT:PSS as an interlayer in OLEDs will be discussed in the following. Soon after the discovery of poly(phenylenvinylene) (PPV) as light emitting material in OLEDs138 it was found by Heeger et al. that a thin interlayer of polyaniline (PANI) placed between the ITO anode and the PPV layer will increase device efficiency.140 This phenomenon was discussed in terms of improved hole injection due to lowering the work function of the anode. The energy barrier between the Fermi level of the ITO anode and the HOMO level of the conjugated light emitting polymer was considered to be reduced by the PANI interlayer.
6911X.indb 207
10/1/10 9:00:49 PM
208
PEDOT
Researchers from Bayer AG discovered that PEDOT:PSS exhibits similar properties when employed as interlayer in polymeric OLEDs.94 Carter et al. compared PEDOT:PSS with PANI as an interlayer between ITO and poly [2- (2-ethylhexyloxy)-5-methoxy-1, 4-phenylenevinylene] (MEH-PPV) directly and found no significant difference regarding device efficiency.66 A significant improvement to devices without a polymeric buffer-layer was observed however. The polymeric interlayer was considered to block oxygen being released from ITO under operation and slow down device degradation.69 Additionally, the polymer will planarize the surface and reduce the probability for electrical shorts.141 Possible candidates for electrical shorts in OLEDs are imperfections in the surface of the ITO anode as suggested by Pichler leading to hot spots or black spots.142 Depending on the ITO quality, local spikes are present owing to the sputtering process. These spikes form electrical shorts to the cathode as they will not be completely covered by the electrically insulating emitting layer. Berntsen et al. demonstrated that a PEDOT:PSS interlayer has a beneficial impact on lifetime stability in passive matrix polymeric OLEDs.143 Figure 10.32 compares the surface roughness of ITO and ITO coated with PEDOT:PSS. The polymer was deposited by spin-coating and dried 1.00
100.0 deg
1.00
100.0 deg
0.75
50.0 deg
0.75
50.0 deg
0.0 deg
0.50
0.0 deg
0.50
0.25
0.25
0
0.25
0.50
0.75
(a)
0 1.00 µm
0
0.25
0.50
0.75
(b)
10.0
10.0
0
0
–10.0
0 1.00 µm
–10.0 0
0.20
0.40
µm (c)
0.60
0.80
0
0.20
0.40
µm
0.60
0.80
(d)
Figure 10.32 Atomic-force microscope contrast images (top) and surface profiles (bottom) of ITO (a),(c) and ITO + PEDOT:PSS (b),(d). The arithmetical mean roughness Ra of ITO was determined to be 1.8 nm, whereas the 50 nm thick polymer layer spin coated on ITO smoothens the surface, resulting R a = 0.8nm. (Adapted from C. Jonda, A. B. R. Mayer, U. Stolz, A. Elschner and A. Karbach, 2000, J Mater Sci 35:5645–5651.)
6911X.indb 208
10/1/10 9:00:51 PM
Applications
209
successively at 200°C for 5 min resulting in a 50-nm thick layer. The atomic force microscope contrast images and the surface profiles show that the surface becomes much smoother with the polymer layer. Elschner et al. compared hybrid OLED devices with and without a PEDOT: PSS interlayer.144 The layer stack consisted of solution deposited methoxysubstituted 1,3,5-tris[4-(diphenyl-amino)phenyl]-benzene (TDAPB) as the hole transport layer and an evaporated tris-(8-hydroxyquinolinato)aluminum (ALQ3) as the emitting layer. The device current was found to increase by a factor of 15 when a PEDOT:PSS layer (commercial grade Clevios P AI 4083) is introduced (Figure 10.33a). This is discussed in terms of a reduction of energy barrier height at the interface PEDOT:PSS to TDAPB due to doping145,146 and the formation of a dipolar interfacial layer. Lifetime measurements conducted at a constant current of 8 mA/cm² show a significant decrease of luminance decay and voltage increase when the polythiophene interlayer was introduced (Figure 10.33a). The improved hole injection leads to a reduction of bias and consequently to a reduction of dissipated energy within the device. The smoothing properties of the PEDOT:PSS planarizing the ITO anode are illustrated by the reduction of leakage current (Figure 10.33a) and by the reduction of electrical noise on the lifetime traces (Figure 10.33b). The leakage current was found to decrease with increasing PEDOT:PSS layer thickness. The surface treatment of ITO has an effect on the performance of polymeric OLEDs even when a PEDOT:PSS buffer layer is incorporated. Especially an oxygen plasma treatment of ITO prior to PEDOT:PSS deposition will increase efficiency and lifetime.147,148 This is surprising as the interface between ITO and PEDOT:PSS is found to be always ohmic. Because oxygen plasma treatment changes the ITO morphology and reduces the contact angle of water, it is assumed that this drives the phase separation of PEDOT and PSS in ways more favorable for electroluminescence. 10.6.3 The PEDOT:PSS–Semiconductor Interface The interface between PEDOT:PSS and the adjacent hole-transporting or light-emitting layers is of special interest for the understanding of electronic processes in polymeric OLEDs. The alignment of energy levels needs to be unraveled to gain a better knowledge on interface properties, for example, charge injection, charge blocking and quenching of excited states. For the sake of simplicity, the energy alignment between a metal and an organic semiconductor has been traditionally discussed within the Schottky–Mott model. This model assumes vacuum level alignment at the interface and band bending with Fermi level alignment. The energy barrier for injecting or extracting holes from a metal into the organic layer or vice versa, respectively, is defined within this model by the difference between the metal work function and the ionization potential of the organic layer. By employing experimental techniques like ultraviolet photoelectron spectroscopy (UPS) or Kelvin probe to investigate energy level alignment of
6911X.indb 209
10/1/10 9:00:51 PM
210
PEDOT
100
With PEDOT:PSS Without
10–1
105
10–2
104
10–3
103
10–4
102
10–5
101
10–6
100
10–7
10–1
10–8
–12
–8
–4
0
4
8
12
Luminance [cd/m2]
Current [A/cm2]
106
10–2
Bias [V] (a)
Bias [V]
10
150
Without PEDOT:PSS
8 6
With PEDOT:PSS
4
50
2 0
100
0
100
200 300 Time [h]
400
Luminance [cd/m2]
200
12
0 500
(b) Figure 10.33 (a) I-V curves (solid symbols) and L-V curves (open symbols) of the three-layer device (PEDOT:PSS–TDAPB–ALQ3) (squares) compared to the two-layer device (TDAPB–ALQ3) (triangles). Introducing PEDOT:PSS leads to higher efficiencies and improves the rectification ratio. (b) Lifetime data of the three-layer and the two-layer device as in (a) driven by a constant current (I = 8 mA/cm²). The bias and the luminance are monitored simultaneously. In the three-layer device, with the PEDOT:PSS layer the lifetime is prolonged and the electrical noise due to microshorts is reduced. (Figures adapted from A. Elschner, F. Bruder, H. W. Heuer, F. Jonas, A. Karbach, S. Kirchmeyer, and S. Thurm, 2000, Synth Met 111–112:139–143.)
6911X.indb 210
10/1/10 9:00:53 PM
211
Applications
metal–organic interfaces it has been shown that the Schottky–Mott model can seldom be applied.149,150 The work function of PEDOT:PSS has been determined first by UPS151 and Kelvin probe152 to be 5.0 ± 0.1 eV and 5.2 ± 0.1 eV, respectively. In the meantime several laboratories have measured the work function of PEDOT:PSS reporting values within the range of 4.7 eV153 to 5.6 eV.154 This confusing spread of reported data is attributed to different preparation conditions and different PEDOT:PSS types being investigated. As PEDOT:PSS film deposition is normally done in ambient air, it is unavoidable that surface contaminations will modify the work function. Leaving PEDOT:PSS layers unprotected in ambient air will reduce the work function over time.155 Especially traces of water in the film depending on postbaking conditions alter the work function significantly.154 Additionally different types of commercial and self-made PEDOT:PSS have been compared without noticing that the ratio of PEDOT:PSS might be different. Koch et al. have determined the work function of various PEDOT:PSS types in a comparative experiment finding higher values for the PSS-rich types (see Table 10.4).156 The interface between the quasi-metal PEDOT:PSS and organic semiconductors has been investigated in numerous experiments. As illustrated in Figure 10.34 the energy barrier for hole injection (∆E) is not simply determined by the difference between the PEDOT:PSS work function (Φ) and the ionization potential (IP) of the semiconductor as predicted by the Schottky– Mott model. If a dipole layer is formed at the interface of two adjacent layers,
Table 10.4 Summary of Sample Parameters: Pristine Conducting Polymer Work Function ϕ, and Hole Injection Barrier ∆E and Work Function ϕP after Deposition of 2.5-nm Pentacene, Interface Dipole ID Polymer (Ratio of PEDOT to PSS by Weight) In situ PEDOT PEDOT-Sa PEDOT:PSS (1:2.5) PEDOT:PSS (1:6) PEDOT:PSS (1:20)
(eV)
∆E (eV)
_P (eV)
ID (eV)
4.25 4.55 4.75
0.40 0.30 0.30
4.20 4.30 4.25
0.05 0.25 0.50
4.85 5.15
0.35 0.40
4.20 4.20
0.65 0.95
Source: Adapted from N. Koch, A. Elschner, and R. L. Johnson, 2006, J Appl Phys 100, 024512-1–024512-5. a PEDOT-S is a modification of PEDOT with sulfonate groups attached to alkoxy chains to introduce solubility in water (see Chapter 12).
6911X.indb 211
10/1/10 9:00:53 PM
212
PEDOT
ID
IP LUMO
Φ
Semiconductor
LUMO ∆E
HOMO PEDOT:PSS
IP
Φ
EF HOMO
PEDOT:PSS Semiconductor
(a)
(b)
Figure 10.34 Energy level alignment at the PEDOT:PSS–semiconductor interface. The two layers are (a) separated and (b) in contact. A formation of an interface dipole (ID) might significantly determine the energy barrier for hole injection ∆E.
their vacuum levels are shifted by the energy ID when they come into contact.156–159 ∆E has to be calculated according to ∆E = Φ – IP – ID. For various PEDOT types the hole-injection barrier to evaporated pentacene was measured by UPS.156 Although the work function varies significantly between 4.25 eV for in situ PEDOT and 5.15 eV for PEDOT:PSS with a ratio of 1 to 20, the hole-injection barrier remains always unchanged (Table 10.4). The dipole layer formation at the PEDOT:PSS to semiconductor interface is believed to be triggered by sulfonate moieties within the PSS. Cationic semiconductor species generated within the interface formation are being counterbalanced by sulfonate (SO3–) groups to provide a stable saltlike configuration.156 The strength of the dipole determining ID scales linearly with Φ. Therefore ∆E is effectively independent of Φ. This has been addressed as Fermi level pinning.158,160 Tengstedt et al. have investigated poly(3-hexylthi ophene), a commonly used light absorbing semiconductor in organic solar cells and other organic semiconductors on various anodes and found that Fermi level pinning occurs if Φ exceeds a threshold defined by IP and the semiconductor’s polaronic relaxation energy.158 However, it has to be kept in mind that energy barriers determined by UPS and Kelvin probe mimic only steady-state conditions. The conditions change in real bipolar OLED devices significantly when free charge carriers trapped at the interface will alter the energy level alignment by generating locally high electric fields.159,161–165 The surface of a PEDOT:PSS layer was found to be PSS enriched166 (see Chapter 9, Section 9.2.2.3). This enrichment of an electrically insulating wide band gap polymer has consequences on OLED properties in those cases
6911X.indb 212
10/1/10 9:00:54 PM
213
Applications
PSS rich phase + PEDOT:PSS
– –
+ LEP
Figure 10.35 Schematic representation of the electron-blocking mechanism with the light-emitting polymer by interfacial wide energy-gap PSS. (Adapted from N. Koch, A. Elschner, and R. L. Johnson, 2006, J Appl Phys 100, 024512-1–024512/-5.)
where it comes to direct contact with the light emitting layer.159,161,164 This PSS-rich layer blocks electrons at the interface and prevents the recombination of electrons at the anode (Figure 10.35). The trapped electrons themselves promote hole injection due to coulombic attraction.162,164 Additionally, the PSS interlayer acts as a barrier for excitons being formed close to the anode’s interface.164 Because of this wide band gap polymer the electron–hole pairs will not quench at the metal-like PEDOT:PSS layer but will preferentially recombine under the emission of light. Hwang et al. removed the PSS enriched layer by sputtering the surface with 250 eV Ar+ ions. The work function of PEDOT:PSS could be reduced from 5.13 eV to 4.75 eV.167 The idea to block electrons and prevent excitons from quenching at the PEDOT:PSS-interface has also been pursued by the following concept: A thin wide band gap polymer interlayer of 10–20 nm thickness was deposited on top of PEDOT:PSS and baked at elevated temperature to achieve good adherence.168,169 The light emitting polymer was subsequently deposited by spin coating. The interlayer polymer comprised triarylamin segments to enable hole transport and hole injection leads to higher efficiencies and lifetimes, but it requires the deposition of an extra layer, which makes production more difficult. 10.6.4 Lifetime Restraints by PEDOT:PSS Hole Injection Layers As noted, PEDOT:PSS interlayers are widely employed at least in solutionbased OLEDs because of two reasons: Improvement of the hole injection and smoothening of the anode’s surface. Additionally, the easy deposition by spin coating together with the persistence against dissolution when a second layer is deposited on top from an organic solvents has made PEDOT:PSS a standard in OLED processing. In addition to these acknowledged advantages, there are also reports on restraints when PEDOT:PSS is introduced as an interlayer in OLEDs that have to be addressed.
6911X.indb 213
10/1/10 9:00:54 PM
214
PEDOT
TCO layers will be etched by PEDOT:PSS due to the low pH being in the order of 1.5 to 2. De Jong et al. found that traces of In released after PEDOT:PSS deposition on ITO migrate into the emitting layer under device operation.170 Wong et al. confirmed this result and found that the deposition of a self-assembled monolayer of alkylsiloxanes on ITO prior to spin-casting PEDOT:PSS was effective and practical in blocking the reactions between ITO and PEDOT:PSS.171 In contrast to that, Lee and Chung did not recognize a correlation between In concentration in the PEDOT:PSS layer and OLED lifetime.172 It should be noted that it is possible to increase the pH to a level where ITO is no longer etched by simply adding a base like ammonia, NaOH, or others as proposed by M. M. de Kok et al.173 Kim et al. investigated polymeric devices under high current densities and observed by Raman spectroscopy sulfonate moities, most probably stemming from PSS, in the emitting polymer.174 Chung et al. detected on the surface of the emitting layer islands of sulfur when the device had been stressed and suggested the diffusion of HSO4– or SO42– from the PEDOT:PSS layer.175 J. Halls reported on progressive cross-linking of the polymeric emitting layer under operation beginning at the PEDOT:PSS interface.176 The cause for this degradation is assumed to be initiated by undetermined residues stemming from the PEDOT:PSS layer after electron impact. Png et al. made the ionic drift of PEDOT+ occurring at low electrical fields responsible for device degradation and claimed that cross-linking the PEDOT:PSS layer will slow the migration of the cation.177 Nowy et al. observed interface degradation between PEDOT:PSS and NPB of unknown origin by employing dielectric spectroscopy on stressed and pristine devices.178 10.6.5 Modified PEDOT-Based Materials for HILs The worldwide attempt to improve lifetime stability of OLEDs has motivated many research groups to investigate PEDOT-based hole injection layers (HILs) further. The following sections will give a brief overview on these activities. It should be kept in mind that the evaluation of HILs in OLEDs is not straightforward as the complete layer stack has to be optimized when a new material is introduced. Properties like balanced charge injection and optical light out-coupling have to be fine-tuned again. Ho et al. improved the photometric efficiency of polymeric OLEDs by increasing the work function of PEDOT:PSS with layer thickness to form a graded layer.179 The authors deposited PEDOT:PSS together with cationic poly(p-xylylene-α-tetrahydrothiophenium) (PXT) to dedope PEDOT. The concentration of PXT was increased from one to the next successively deposited layer to achieve a continuous increase of the work function and the electron blocking properties over the HIL. Lifetime improvements of a factor of two were reported by Lee when a PEDOT:PSS layer was cross-linked with glycidoxypropyltrimethoxysilane.180
6911X.indb 214
10/1/10 9:00:55 PM
Applications
215
This result was explained by improving the adhesion between PEDOT:PSS and the light emitting polymer together with the assumption that released In ions from the ITO anode are trapped in the cross-linked layer. The influence of the ratio of PEDOT to PSS on device performance was studied on polymeric OLEDs.181 A significant increase in device efficiency was found when a layer of PEDOT:PSS with a ratio of PEDOT to PSS of 1:20 was coated on a high conducting polymeric anode with a ratio of 1:2.5. The high PSS content is necessary to form a PSS-enriched surface to block electrons and to prevent excitons from being quenched, as discussed earlier. A correlation between PSS enrichment at the surface, work function, and device efficiency was found by Lee and Chung investigating various types of PEDOT:PSS.172 The segregation of PSS to the surface was enhanced when low molecular weight PSS was employed. The addition of multivalent cations like Mg++ or Ca++ will ionically cross-link PEDOT:PSS.182 Seo et al. added CuSO4 to form a coordination-complex polymer and reported a 60% increase of device efficiency over standard PEDOT:PSS.183 The surface of a PEDOT:PSS buffer layer was exposed to a mild oxygen plasma enhancing OLED efficiency and lifetime.184 The plasma treatment changed the surface morphology by roughening the surface and presumably changing the ratio of PEDOT to PSS. The work function of PEDOT:PSS layers were changed by Zhang et al. through electrochemically doping PEDOT.185 The work function was increased by up to 0.2 eV by applying an oxidizing potential and in parallel the electroluminescent efficiency increased from 2.4 to 3.25 cd/A at 100 cd/m² in small molecule devices. Several groups have investigated the combination of PEDOT with fluorinated sulfonic acid polymers such as Nafion™.186–190 Such PEDOT-containing dispersions were employed in polymeric OLEDs comprising blue, green, and red emitting polyfluorenes. A significant improvement in device properties was observed relative to PEDOT:PSS. The improvement was attributed to the increase of work function up to 6.0 eV at pH = 1.8 and 5.4 eV at pH = 7.189 Lee et al. reported a work function of 5.9 eV for the combination of perfluorinated ionomer and PEDOT:PSS.191 An enrichment of CF2 segments was found at the surface by depth profiling the layer.192 The surface of PEDOT:PSS was modified by Zhang et al.193 by spin coating an ultra-thin layer of fluorinated polyimide (F-PI) on top in order to form an electronic blocking layer as discussed earlier.168 A mechanism was proposed to explain the improved device performance: (a) an electron blocking effect at the surface of the PEDOT:PSS owing to its intrinsic insulating property, (b) the lowered hole-injection barrier via partial dedoping of PEDOT by withdrawing electrons from the adjacent F-PI and the formation of a stepped increased work function, similar to the concept proposed by Ho et al.179 Kim et al. polymerized EDOT in the presence of imidazolium-based poly(ionic liquid)s (PILs) and employed the dispersion as a hole-injecting
6911X.indb 215
10/1/10 9:00:55 PM
216
PEDOT
layer in small molecule OLEDs.194 The improvement of lifetime and efficiency was attributed to hindrance of Indium dissolution owing to the pH of 7 and the increase of work function.
10.7 PEDOT:PSS in Organic Solar Cells 10.7.1 Introduction Organic solar cells (OSCs) have gained considerable interest within the last decade as forward-looking technology enabling the production of photovoltaic elements by roll-to-roll deposition, for example, by printing195,196 or vacuum sublimation.197 The principles of OSCs have been outlined in detail elsewhere.198,199 Various types of cell architectures have been proposed of which three are of main interest: (1) bulk heterojunction solar cells in which electron donors and acceptor are homogeneously mixed to jointly form a functional layer, for example, poly(3hexylthiophene) (P3HT) as the donor and substituted fullerenes like [6,6]-phenyl-C61-butyric acid methylester (PCBM) as the acceptor; (2) multilayer devices of vacuum-deposited small molecules; and (3) dye-sensitized nanostructured oxide cells, for example, dyes attached to the surface of nano-TiO2. In Figure 10.36 the cross-section of an OSC with an active layer of P3HT:PCBM and a buffer layer of PEDOT:PSS is shown together with the corresponding energy levels. P3HT and PCBM form a phase-separated interpenetrating network constituting the bulk heterojunction. When light is absorbed by the P3HT, transient excited states are formed. These electron– hole pairs dissociate at the P3HT/PCBM interface as electrons transfer to the Cathode
eV
eV
PEDOT:PSS 5.0eV
hν +
_
Substrate
O 3.2
O 4.3
+
Al 4.3eV
+
P3HT
hν
(a)
_
_
PEDOT:PSS Transparent anode
LUM LUM
_
PCBM
_
P3HT
+
(b)
HOM O 5.1 eV PCBM HOM O 6.1 eV
Figure 10.36 Bulk heterojunction OSC of P3HT:PCBM with a buffer layer of PEDOT:PSS. (a) Cross-section of the OSC, (b) energy levels according to K. Lee, J. Y. Kim, and A. J. Heeger. (In: Organic Photovoltaics: Materials, Device Physics, and Monufacturing Technologies, ed. C. Brabec, V. Dyakonov, and U. Scherf, 2008. Weinheim: Wiley-VCH.)
6911X.indb 216
10/1/10 9:00:56 PM
217
Applications
Aluminium Active layer PEDOT:PSS
PEDOT:PSS Active layer
Substrate
Aluminum Substrate
(a)
(b)
Aluminium Active layer PEDOT:PSS Transparent electrode Substrate (c)
Figure 10.37 PEDOT:PSS as transparent conductive layer deposited (a) on the substrate, (b) on the active layer, and (c) as the buffer layer between the transparent electrode and the active layer.
LUMO of PCBM, which is about 1.1 eV lower in energy than the LUMO of P3HT.200 Electrons and holes advance within the internal field of the device along the PCBM sites to the cathode and along the P3HT domains to the anode, respectively. PEDOT:PSS layers have been employed in OSCs to fulfill various functions: as transparent conductive layers deposited directly on the substrate (Figure10.37a); as transparent conductive layers deposited on the active layer (Figure 10.37b); and as buffer layers mainly deposited between the transparent anode, typically a transparent conductive oxide (TCO) layer, and the active layer (Figure 10.37c). PEDOT:PSS has also been employed as a coating on the counter electrode and as a solid-state hole conductor in dye-sensitized solar cells (DSSCs)201 forming its own class of organic photovoltaic devices. 10.7.2 PEDOT:PSS as a Transparent Anode in OSCs A prerequisite for the commercial success of organic photovoltaic devices is a significant reduction of device processing costs compared to conventional, silicon-based solar cells. In OSCs the employment of TCOs as the transparent electrode contributes significantly to the overall costs, as TCOs have to be deposited by slow process techniques, for example, sputtering, and cannot be deposited by cost efficient printing methods. This has motivated approaches that follow alternative materials to substitute TCOs. For a detailed discussion on this topic see Section 10.3. High conductive PEDOT:PSS is considered as the most relevant polymer to replace TCOs and has been successfully introduced in OSCs as transparent bottom electrode, located directly on the substrate (Figure 10.37a), or as transparent top electrode (Figure 10.37b). In an experiment to prove the principle Arias et al. have shown that a poly(p-phenylene vinylene) (PPV) layer sandwiched between PEDOT:PSS and Al forms a photovoltaic device independent of whether a polymer or a metal is deposited as the final layer.202
6911X.indb 217
10/1/10 9:00:56 PM
218
PEDOT
To overcome resistive losses across the anode the conducting polymer has been deposited by spin coating or screen printing on an underlying metal grid of evaporated gold203 or silver deposited by contact printing.204 To achieve a suitable conductivity high-boiling solvents have to be added to the PEDOT:PSS dispersion prior to deposition as outlined in Chapter 9. Zhang et al. showed the feasibility to substitute ITO (indium tin oxide) as the most widely used TCO by PEDOT:PSS formulated with glycerol or sorbitol.205 The influence of these so-called secondary dopants has been investigated by Hsiao et al. reporting that the efficiency of solar cells is mainly correlated with the sheet resistance of the transparent anode.206 Glatthaar et al. deposited a commercially available PEDOT:PSS formulation (Clevios CPP 105D*) on top of the photoactive layer to form inverted solar cells.207 In contrast TCO layers when used as top electrode will damage the sensitive organic active layer, for example, P3HT:PCBM, owing to the sputter-deposition process, and as a consequence the devices will not function. The concept of polymeric top electrodes allows also the fabrication of semitransparent solar cells.208 As a polymer the mechanical properties of PEDOT:PSS, especially during bending, clearly outperform the properties of brittle TCO films. As an impressive demonstration flexible OSCs with PEDOT:PSS and ITO anodes on PET substrates were compared under bending stress.209 Heterojunction solar cells with polymeric anodes were found to last significantly longer than anodes made from ITO. When being bent periodically the sheet resistance of the ITO layer increased due to microcrack formation. The efficiencies of devices with PET substrates were found to be independent of the anode material. Especially with respect to roll-to-roll coating PEDOT:PSS has gained additional interest since it is offering an opportunity for a cost-efficient wet deposition. Ahlswede et al. have deposited the polymer by air brush technique and achieved efficiencies of up to 3%, 210 whereas Lim et al. employed spray deposition in an inverted cell setup achieving 2%.211 10.7.3 PEDOT:PSS as a Buffer Layer in OSCs Like in OLEDs, PEDOT, and PEDOT-like polymers, and in most cases PEDOT:PSS, are widely used in organic photovoltaic cells as a buffer layers between the anode, typically ITO, and the photoactive layer as depicted in Figure 10.37c. The role of PEDOT buffer layers have been studied in detail in various device structures. This includes polymeric photovoltaic cells212–214 also in combination with titanium dioxide200,215; dye sensitized photovoltaic cells216,217
* Manufacturer: H.C. Starck Clevios GmbH, Leverkusen, Germany.
6911X.indb 218
10/1/10 9:00:56 PM
Applications
219
also in combination zinc compounds218; as well as silicon hybrid solar cells219 and hybrid organic-nanocrystal solar cells.220 The advantage to incorporate PEDOT:PSS buffer layers in OSCs is based on several properties being closely related to its beneficial use in OLEDs (see Section 10.6). First, PEDOT:PSS planarizes the anode surface by smoothening surface imperfections. In particular local spikes on the ITO surface that cause electrical shorts in thin film devices142 will be partly covered by the polymer. These spikes have a size of several tens of nanometers in size and originate from the ITO deposition. Therefore, a PEDOT:PSS buffer layer has been typically found to increase the yield of functional devices as the probability for electrical shortages within active layer is reduced.221 Second, PEDOT:PSS buffer layers make processing more reproducible. The work function of the hole transport layer in contact with the functional layer will become independent of ITO precleaning steps and of the resulting ITO work function. This is a consequence of the high density of free charge carriers in PEDOT:PSS, which allows the Fermi levels of ITO and PEDOT:PSS to equilibrate as perfect metals would do. UV/Ozon or plasma treatment of ITO surfaces is commonly used to improve surface wetting and the quality of subsequently deposited layers. The cleaning steps of ITO may have additional consequences that have not been fully understood. In contrast to expectations it was found that an oxygen plasma treatment improves OLED performance despite an intermediate PEDOT:PSS buffer layer.148 It is assumed that ITO treatment might have a beneficial influence on PEDOT:PSS film formation and morphology in that specific device configuration. Third, the voltage drop across the buffer layer can be neglected. For most PEDOT:PSS grades used as buffer layers in OSCs with a conductivity of 1 mS/cm this voltage drop can be estimated to be 0.1 mV for a 100 nm thick layer at a current density of 10 mA/cm2. Fourth, PEDOT:PSS buffer layers will improve device performance. The work function Φ of PEDOT:PSS is in the order of 5.0 to 5.2 eV200,151 leading to a build in potential of 0.8 to 1.0 V in combination with Al cathodes ((Φ)Al = 4.2 eV) according to the simplifying metal–insulator–metal (MIM) model.222,223 It is common understanding that the open circuit voltage VOC in OSCs is determined by the energy difference between the LUMO of the acceptor and the HOMO of the donor as long as ohmic contacts between the electrodes and the active layer exist. In the case of nonohmic contacts, VOC is determined by the difference in the electrodes work function.224 The dipole formation at the interface between PEDOT:PSS and the semiconducting layer additionally defines the contact properties and the built-in potential. This will be discussed in more detail in Section 10.6. Despite its advantages, the introduction of PEDOT:PSS as a buffer layer in OSCs also faces obstacles. Reports on obstacles focus on the acidity of
6911X.indb 219
10/1/10 9:00:57 PM
220
PEDOT
the aqueous PEDOT:PSS dispersion and the inherent moisture sensitivity of PEDOT:PSS, which results in a rapid water absorption of solid films. The acidity of PEDOT:PSS being in the pH range of 1 to 2 is suspected to dissolve cations like indium ions from the ITO layer that migrate from the anode into the buffer layer and even may contaminate the photoactive layer.170,225 In contrast to this concern the acidity of PEDOT:PSS is considered to be advantageous to enable good contact formation at the interface by etching off all contaminations from the TCO surface during deposition. PEDOT:PSS layers will absorb water from the environment if left unprotected in air even after complete drying (see Chapter 9). The release of water in a functional device may lead to severe degradation as water might corrode the metallic contacts or might oxidize the adjacent semiconducting organic layer. Additionally it was found that the resistivity of the PEDOT:PSS–semiconductor interface will increase when the layer is exposed to air and this may cause OSC degradation.226 To account for these concerns PEDOT:PSS films are commonly baked at elevated temperatures in inert atmosphere and further kept inert during further processing to avoid any water absorption. Finished devices need to be encapsulated to ensure prolonged lifetimes. Kawano and Adachi claimed that even in encapsulated heterojunction OSCs the degradation is caused by instabilities of the interfaces of the PEDOT:PSS–semiconductor and semiconductor–cathode.227 The formation of deep trap states at the interfaces under continuous illumination has accounted for the observed drop in efficiency. Charges will accumulate at the interfaces and form nonohmic contacts. Consequently VOC will drop. Ramuz et al. discussed the influence of a PEDOT:PSS interlayer in P3HT:PCBM-based photodetectors.228 Higher lifetimes and higher on/off ratios were reported for detectors omitting the interlayer, although the open circuit voltage dropped in parallel by 0.4 V.229 Various modifications of PEDOT:PSS have been discussed as alternative buffer layers in OSCs. Some of these results are summarized next. Zhang et al. have compared various PEDOT:PSS grades as thin layers in between ITO and an active layer comprising a polyfluorene copolymer blended in proportion (1:4) with (PCBM) in OSCs.213 In a study of buffer layers made from PEDOT:PSS with a weight ratio of 1:2.5 respectively 1:6 with and without the addition of conductivity enhancing sorbitol the authors suggested a segregation of PSS to the top of a PEDOT:PSS film. These surface modifications have been made responsible for the observed differences in short circuit current, fill factor, open circuit voltage, and power conversion efficiency. Single wall carbon nanotube (SWCNT) films have been invoked to replace ITO as printable and flexible anodes in OSCs. Again, the introduction of a PEDOT:PSS was found to significantly improve device performance, independent on whether the PEDOT:PSS was introduced as a buffer layer229,230 or as a blend together with CNTs.231
6911X.indb 220
10/1/10 9:00:57 PM
Applications
221
Williams et al. have investigated spin-casted PEDOT:PSS as a p-layer on ITO combined with amorphous silicon and microcrystalline silicon in an organic–inorganic p-i-n stack. A power efficiency (η) of 2.1% and an open circuit voltage of 0.883 V has been achieved, which dropped to 0.21% and 0.176 V when the PEDOT:PSS layer was omitted.232 Peumans et al. found that treatment of PEDOT:PSS surfaces with mild Ar or O2 plasma improves the carrier collection properties in OSCs.221 Apparently the impact of gas ions with the polymer surface leads to layer thinning and increased layer microroughness184 and will most probably modify the chemical composition of PEDOT:PSS too. Frohne et al. modified electrochemically the work function (Φ) of PEDOT:PSS anodes and claimed an improved overall OSC performance by matching Φ with the oxidation potential of the semiconductor.233 PEDOT:PSS, CuPC, and thin evaporated Au films were compared as buffer layers in bulk-heterojunction OSCs by Yoo et al.214 A significant increase of VOC and η was found-for the two organic layers in contrast to bare ITO or Au. 10.7.4 PEDOT:PSS in Dye-Sensitized Solar Cells In the early 1990s a new type of photovoltaic device was proposed by O’Regan et al. exploiting redox chemistry to transform visible light into electrical power.201 The anode of the so-called Grätzel cell consists of a mesoporous TiO2 covered by a monolayer of dye molecules, for example, Ru complexes. When light is absorbed by the sensitizer the excited states dissociate, whereas the electrons are transferred to the n-type TiO2 and the holes are carried off by an I3–/I–-redox electrolyte penetrating the pores of sintered TiO2 body. As iodide-based liquid electrolytes are corrosive and difficult to confine within the device, researchers have strived for alternatives and proposed solid-state hole conductors as an alternative. With the hole-transporting spiro-MeOTAD a conversion efficiency of 4% could be achieved in comparison to 11% for cells filled with the iodide electrolyte.234 PEDOT-based hole conductors were first introduced by Saito et al. in DSSCs.235 The monomer EDOT or the dimer bis-EDOT was polymerized in the cell to achieve close electrical contact to the sensitizing dye. The polymerization was conducted chemically with Fe(III) as the oxidant or photoelectrochemically by light excitation of the dye together with an applied electrical field.236 Limited power efficiencies of about 2% to 3% have been achieved.237 Johansson et al. deposited a layer of the aqueous PEDOT:PSS dispersion as a hole conductor in DSSCs and avoided the polymerization of EDOT within the device. Although only a planar, nonporous TiO2 electrode had been employed leading to a very limited light absorption the incident photon to current conversion efficiency was found to be similar to cells with the standard electrolyte as hole-transport material.238 The counter electrode in DSSCs is typically formed by a transparent conductive oxide, for example, fluorine-doped tin oxide covered with adhered
6911X.indb 221
10/1/10 9:00:57 PM
222
PEDOT
platinum particles. The platinum particles catalyze the reaction of the I3–/I–redox couple at the cathode’s surface. A layer of PEDOT deposited onto the TCO counterelectrode was found to have a similar catalytic effect and was proposed to replace Pt.236 A similar observation was made by Chen et al. coating the counter electrode with PEDOT:PSS.239 When high-boiling solvents were added to the dispersion an increase of device efficiency was observed. This effect was attributed to an increase of the surface roughness enlarging the active surface and an increase of electrode’s conductivity. Cell efficiencies of 5.81% were reached in comparison to 5.66% for reference cells using Pt as counter electrodes. Biancardo et al. confirmed this result by employing high-conductive PEDOT:PSS-coated counter electrodes in combination with gel electrolytes.240
10.8 Electrochromic Behavior 10.8.1 Introduction Some materials change color corresponding to the applied electrochemical potential, a phenomenon called electrochromism. Electrochromic materials have been known for a long time.241,242 Common to all electrochromic materials is the fact that a redox process is triggered electrochemically, which changes the material’s optical absorption. The glass industry has exploited the electrochromic effect in the development of multilayer glasses that darken electrically (“smart windows”). However, until now these windows have had little success because of harsh lifetime requirements, their complicated and cost-intensive production process, and their high price that appears to be prohibitive for a major market breakthrough. Sage Electrochromics, Inc., a U.S.-based company, commercialized an electrochromic roof window in an alliance with Velux/USA that has been withdrawn from the market. Major glass manufacturers, for example, Pilkington/ FLABEG (Germany) and Saint Gobain (France), have attempted to commercialize electrochromic windows in architectural and automotive glazings but without commercial success. Automotive sunroof glazings have been demonstrated by Schott-Donnelly (United States), Saint-Gobain (France), and Central Glass (Japan) and serve a niche market as sunroofs in luxury cars. All commercial electrochromic glazings were based on tungsten trioxide as the electrochromic material, which has to be applied on the glass surface in a sputtering process under high vacuum. All efforts to develop a window system using organic viologen chemistry have failed so far most probably due to lack of long-term stability and the low acceptance of liquid electrolytes. Nevertheless, established technical applications are a liquid and gel-type automotive rear-view mirror system (Gentex, United States) and electrochromic displays in cameras (Nikon, Japan).243
6911X.indb 222
10/1/10 9:00:57 PM
223
Applications
RED1
+
OX1
OX2
(colorless)
+
RED2
(dyed)
Figure 10.38 Redox reactions of viologens in an electrochromic cell.
More recently, electrochromic displays especially based on PEDOT have attained attraction in the field of printed electronics. Here the main idea is to develop a paperlike display by printing techniques. Typical electrochromic materials are bipyridinium salts (viologens).244,245 Dyes like viologens often consist of different chemical species in an electrolyte which are either oxidized or reduced at the electrodes to form radical cations and anions that recombine charges after diffusion (Figure 10.38 and Figure 10.39). In such systems a small but significant current will be necessary to maintain the colored state. Electrochromism with viologens as electrochromics and liquid electrolytes may be combined with PEDOT as a transparent electrode material.246 In contrast to the foregoing electrochromic solids, polymers247 especially have attracted interest since they eliminate the need for liquid electrolytes and offer bistable switching states. The electrochromic layer switches reversibly between an oxidized and a reduced state. This redox reaction is charge balanced by the release and incorporation of counter anions, which are either ionically bound to the electrochromic solid or exhibit a low mobility. Cations, usually lithium ions, therefore, have to move through the polymeric electrolyte to counteract the electron current. A general scheme of the redox reaction and layers employed is depicted in Figure 10.40 and Figure 10.41. Inorganic oxides241 as well as electroactive polymers are known for their electrochromic behavior. Tungsten trioxide248,249 has been utilized in electrochromic windows.250,251 Among electroactive polymers polyaniline252,253 and polythiophenes254,255 have been studied most extensively as electrochromic solids. Glass TCO* Liquid electrolyte TCO*
Dye 1ox
+e–
Dye 1red Recombination
Dye 2red
–e–
Dye 2ox
– +
Glass Figure 10.39 Layer structure of a viologen electrochromic cell (*TCO: Transparent-conducting oxide [e.g., Indium Tin Oxide].)
6911X.indb 223
10/1/10 9:00:59 PM
224
PEDOT
Electrochromic Layer: OX1+An–
e–
+
+
Li+
RED1
(colorless)
Li+An–
+
(dyed)
Ion Storage Layer: RED2
+
Li+An–
OX2+An–
(colorless)
+
e–
+
Li+
(colorless)
Figure 10.40 Redox reactions of electrochromic solids in an electrochromic cell.
PEDOT has been found to have useful electrochromic properties256,257 and for a long time fed the hope of inexpensive electrochromic glazings made by simple coating techniques for architectural and automobile applications.258 At present it seems unlikely that PEDOT-based electrochromic glazings will appear on the market in the near future. However its potential has inspired numerous and continuing studies that intend to improve the properties of the electrochromic solid, the electrolyte, and the ion storage material and to assemble innovative electrochromic devices. In thin layers PEDOT and PEDOT:PSS switch from the oxidized light blue to the reduced deep blue form exhibiting similar colors to tungsten trioxide. This switching behavior, however, with a reversed coloring since tungsten trioxide exhibits a deep blue color in the oxidized state, made PEDOT
e– Li+ Glass
TCO* Electrochromic layer Gel electrolyte
TCO* Electrochromic layer Ion storage layer
(polymeric) Gel Electrolyte
2.5 V
– +
TCO* Glass Li+ e–
Gel electrolyte Ion storage layer TCO*
Figure 10.41 Layer structure of a solids electrochromic cell. (*TCO: Transparent-Conducting Oxide [e.g., Indium Tin Oxide].)
6911X.indb 224
10/1/10 9:01:00 PM
Applications
225
attractive as a material to drop into a layer structure originally developed with tungsten trioxide. Electrochromic layers with PEDOT are either prepared by electrochemical or chemical polymerization of the corresponding monomer. Both the electrochemical and the chemical polymerization can be well controlled and this makes it easy to test a larger number of EDOT derivatives. However, the aqueous dispersion of PEDOT:PSS is much easier to process by simple coating and therefore has been a preferred material in technical developments. Electrochromic cells based on PEDOT employ three functional layers:
1. A PEDOT or PEDOT:PSS layer as the electrochromic solid, which reversibly switches color corresponding to its redox state. 2. An electrically insulating but ion conducting ion-transport layer consisting of a solid electrolyte formulation of polymers, liquid electrolytes, and lithium salts. 3. An ion storage layer, a material undergoing redox processes without changing the optical absorption. The main requirements for electrochromic elements are: • Color—The ideal is a free choice of color in the colored state and a highly transparent (neutral) color in the transparent state. However, in reality, the choice of colors is limited. The color of electrochromic elements is determined by the redox-active chromophore in the oxidized and reduced state. Based on its chemical structure PEDOT exhibits a blue absorption in the transparent state, which cannot be changed by simple means, therefore making it necessary to modify the chemical structure of the chromophore. As a consequence one strategy has been to synthesize derivatives of EDOT that were polymerizable to the corresponding polymers and studied for their electrochemical and electrochromic properties. To find a more effective leverage to adjust the chromophore absorption the EDOT, alkylenedioxythiophenes, especially ProDOT, and other chromophores were combined to yield copolymers or blends. Much work has been published on these strategies driven by various groups, in particular by the group of J. R. Reynolds. Details will be discussed in Section 10.8.2 to Section 10.8.6. • Contrast—The contrast ratio is supposed to be as high as possible to block light in the dyed state effectively. Its value is determined by the chromophore’s absorption coefficient in the bleached and in the dyed state and usually this ratio does not exceed a value of 10. Various strategies have been followed to enhance the contrast ratio by either selectively modifying the electrochromic layer chemically or by modifying the layer structure, for example, by the introduction of a complimentary electrochromic layer as the ion storage layer.
6911X.indb 225
10/1/10 9:01:01 PM
226
PEDOT
• Switching speed—The switching of electrochromic devices is limited by various factors, comprising the redox kinetics of the active layer, electrical resistors limiting electrical currents, and the limited speed of ion movements. The conductivity of PEDOT depends on its redox state. In the reduced state the conductivity of PEDOT is several orders of magnitude lower than in the oxidized state.259 When PEDOT has been fully switched to the reduced state, injecting charges into the nonconducting layer of PEDOT might become the speed-limiting factor. The ion movement in a given electrolyte decreases with an increasing ion radius. Therefore lithium salts are commonly preferred to formulate electrolytes. Lithium ions in solid or gel-like electrolytes usually exhibit an ion conductivity of around 10 –3 S/cm which is approximately one order of magnitude lower than their conductivity in liquids. The switching speed increases with an increasing driving voltage and increasing temperature, and decreases with an increasing thickness of the electrolyte layer. In practice switching might occur in the subsecond range for very small areas while larger areas like windows may need more than 10 seconds to change its state. Although the switching speed of electrochromic cells appears to be a significant obstacle for their technical realization, limited work has been published to solve this problem. • Long-term stability—Electrochromic glazings need to be stable for an extended time, for example, 20 years, without losing the electrochromic properties or showing other long-term degradation effects like bleaching or yellowing. Organic materials are in general more sensitive to thermal and UV degradation than inorganic materials. During their lifetime, glazings—since they are deliberately submitted to the sunlight—need proper measures to ensure UV-light protection. • Flexibility—Flexibility has dominantly been discussed for display applications. Unlike liquid systems electrochromic polymers offer the potential to make true all-polymeric flexible electrochromic cells that employ not only polymeric functional layers but also polymeric substrates. • Cost—Polymers may offer lower cost for electrochromic devices in case vacuum sputtering processes can be replaced by wet processing of polymers. High costs might be a significant limitation for commercialization. 10.8.2 Control of Optical Properties The primary optical properties of an electrochromic device depend on the electrochromic chromophore. The control of color in π-conjugated organic polymers for electrochromic applications has recently been documented in
6911X.indb 226
10/1/10 9:01:01 PM
227
Applications
(a)
(b)
Figure 10.42 PEDOT:PSS in the (a) oxidized and (b) reduced form.
a comprehensive review.260 PEDOT is limited to a single color: blue. PEDOT appears to be a suitable electrochromic chromophore of choice due to the fact that its optical absorption is similar to blue tungsten trioxide, and PEDOT:PSS is available in the industrial scale in form of an aqueous dispersion, which allows convenient film formation by various deposition methods. In the doped state the intrinsic color of PEDOT is light blue, which changes to deep blue during reduction (Figure 10.42). For thin films of PEDOT:PSS the optical absorption has been studied as a function of doping level in electrochemical cells.257 Figure 10.43i depicts the absorption spectra taken at different applied voltages oxidizing or reducing PEDOT:PSS. The main absorption peak at 2.2 eV can be assigned to partly neutralized PEDOT corresponding to the LUMO–HOMO transition schematically shown in Figure 10.43ii. By increasing the electrode potential this peak disappears and the deep blue color of the film vanishes to yield almost full transparency. The absorption in the IR region increases accordingly as new electronic states are generated within the band gap due to cationic PEDOT at high oxidation levels (see Figure 10.43iii).261 During electrochemical reduction electrons migrate into the polymer. It has been proven that the reduced species differs significantly from the fully undoped neutral form of PEDOT.262,263 It is therefore likely that the reduced and oxidized PEDOT species are represented by the radical cationic (“polaron”) and cationic (“bipolaron”) polymer. However it remains unclear whether protonated neutral species also contribute to a deep blue color. Several concepts have been proposed to modify the optical absorption and achieve a certain freedom in switching between several colors. The optical absorption is a function of the electronic structure of polymer, which can be predicted only to a limited extend. Intermolecular interactions like aggregation of polymer chains and solvent effects impact energetic levels and the optical absorption. Therefore, predictions often fail and force to follow a more heuristic trial-and-error approach.
6911X.indb 227
10/1/10 9:01:02 PM
228
PEDOT
2.0
a
1.5 Abs [A]
U a –1.5 V b –1.0 V c –0.5 V d 0V e 0.5 V
(i)
b
1.0
c
0.5
d e
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Energy [eV] (i)
LUMO
LUMO
HOMO
HOMO
(ii)
(iii)
Figure 10.43 (i) Optical absorption spectra of a PEDOT electrochemical cell for different applied voltages: (a) –1.5 V, (b) –1.0 V, (c) –0.5 V, (d) 0 V, (e) +0.5 V. Schematic representation of energy levels and optical transitions of polythiophenes in the (ii) neutral and the (iii) p-doped state. (Data from J. C. Gustafsson, B. Liedberg, and O. Inganäs, 1994, Solid State Ionics 69(2):145–152; X. M. Jiang, R. Österbacka, O. Korovyando, C. P. An, B. Horovitz, R. A. J. Janssen, and Z. V. Vardeny, 2002, Adv Funct Mater 12(9):587–597.)
This was demonstrated in a pioneering work of Ferraris and colleagues in 1999, who correlated the observed color coordinates of polymer blends, copolymers, laminates, and patterns with the colors expected from colorimetric theory.264 The goal was to evaluate the direct steady relationship between color and chemical composition. Color coordinates of laminates and patterns were predicted quite accurately; predicted colors deviated significantly from the observed color coordinates in case of blends and copolymers. The most important concepts to modify the optical properties of PEDOT comprise the following:
6911X.indb 228
10/1/10 9:01:03 PM
229
Applications
• Polymers of modified EDOT-structures to change the optical absorption. • Copolymers of alkylenedioxythiophenes with other chromophore units to introduce modified and additional redox and optical states. • Attachment of electrochromic dyes. • Complementary dying ion storage layers to enhance the optical contrast. • Use of blends. • Use of structured layers. As a result of the tremendous scientific work that has been published, a set of electrochromic polymers and copolymers is now available besides the blue PEDOT that gain access to the colors green265–267 (copolymers with EDOT structures, ProDOT structures, and benzothiazole structures), red268,269 (polymers with ProDOT structures) as well as high band gap polypyrroles with transparent oxidized and transparent reduced states that serve as optical neutral ion storage layers.270 10.8.3 EDOT Derivatives The electrochemistry and optical properties of poly(alkylenedioxy)thiophenes have been reviewed by Groenendaal et al.271 With increasing ring size, λmax (as well as the contrast ratio) of electrochemically polymerized poly(alkylenedioxy)thiophenes increases in the oxidized polymer from 499 nm (MDOT) to 619 nm (BuDOT; for structures, see Figure 10.44).272 This effect has been attributed to the electronic overlap between thiophene π-orbitals and oxygen σ-orbitals in the monomeric units, which may help to dissipate charges carried by the polymer to the oxygen atoms in dependence of the exact ring geometry.273 However it remains unclear whether variances in the molecular weight may also have contributed to this effect. Attached groups may also impact the optical absorption of EDOT, however, significant changes of the optical absorption may not be expected as OH O
O
O
O
O
O
S
S
S
MDOT
EDOT
ProDot
O
R O
S ProDot-OH
O
O S BuDOT
O
O S
n = 3, R = C3H7 n = 8, R = C8H17 n = 14, R = C14H29 n = 16, R = C16H33
EDOT-Cn
Figure 10.44 Chemical structures of the alkylene dioxythiophenes MDOT, EDOT, ProDOT, ProDOT-OH, BuDOT, and EDOT-Cn.
6911X.indb 229
10/1/10 9:01:03 PM
230
PEDOT
long as the substituents do not interact with the electronic system of the polythiophene chromophore. In EDOT-Cn, linear alkyl groups with a chain length from 1 to 8 carbon atoms lower the conductivity; however, they seem to have a low effect on the optical absorption spectrum.272,274 Most of the corresponding polymers appear light blue in the oxidized state and deep blue to purple in the reduced state. The tetradecyl and hexadecyl derivatives EDOT-C14 and EDOT-C16 were reported to have a high contrast ratio in the visible region274–276 which was attributed to the known tendency of long linear alkyl groups to crystallize. C8-PEDOT was found to switch faster than PEDOT, which was attributed to a “loosened morphology” allowing enhanced ion movement.277 The hydroxymethyl group in PEDOT-CH2OH as well as its derivatives PEDOT-S271 (which has been extensively studied as polymer in layer-by-layer deposition278–280) and esters like the perfluoro-derivative PEDOT-F281 exhibit similar optical behavior as alkyl derivatives. It appears to be a general trend that long alkyl chains with a higher tendency to crystallize yield a high contrast ratio as well as bulky alkyl groups.282 Attached groups or modifications that significantly alter the ring geometry may impact the optical absorption of the electrochromic polymer. For example, the polymers of VDOT and benzo-EDOT exhibit a purple color instead of blue (Figure 10.45). Due to the convenient access to functional groups in the 2-position of the propylene bridge ProDOTs have gained special interest.271 A multiplicity of ProDOT derivatives, for example, alkyl derivatives like the dimethyl,283 dibutyl,268 and dibenzyl284 derivatives have been synthesized, electrochemically polymerized, and studied for their electrochromic properties. Polymers from ProDOT alkyl derivatives were found to exhibit a high contrast ratio due to a sky blue oxidized state with a low optical absorption and a red to purple reduced state. ProDOT-OH is convenient to functionalize and various derivatives have studied for their electrochromic behavior.285 The easy access to alkyl hydroxyl ProDOTs (I) motivated the synthesis of more complex functionalized ProDOT derivatives including
O OH O
O
O O
O
S
S
EDOT-CH2OH
EDOT-F
C
C8H17
O
O
O S
EDOT-S
SO3H O
O
O
O
S
S
Benzo-EDOT
VDOT
Figure 10.45 Chemical structures of substituted ethylenedioxythiophene (EDOT-OH, EDOT-S, EDOT-F) and unsaturated EDOT derivatives (Benzo-EDOT, VDOT).
6911X.indb 230
10/1/10 9:01:04 PM
231
Applications
S HO
n
R
O
O
O
O
O
n
O
O
O
O
O O
S R: a) H, b) OH
S
I
II
n
*
O
O S
4–12
III
O
O S
n
**
O
O S
o
*
*
O S
p
*
IV
Figure 10.46 Chemical structures of conformationally locked alkylenedioxythiophenes I–IV.
Poly(SpiroBiProDOT)286 (II) and so-called conformational locked or tethered ProDOT derivatives like (III).287,288 The synthesis of conformationally locked structures was motivated by the idea to restrict the conformational freedom during polymerization and hence control the optical absorption of the obtained polymer in the reduced and the oxidized state more tightly. While (II) should polymerize to a close-meshed cross-linked polymer, polymer (III) will be linear and the polymerization of monomeric units of (IV) might either result in linear structures by in situ intramolecular dimerization or cross-linked polymer units during intermolecular reaction. Indeed polymer (III) exhibits a hypsochromic shift in the neutral state when compared to nontethered derivatives of similar structure and shows an orange color. The chain length in (III) obviously impacts the twist angle between the two thiophene units and therefore the effective conjugation length in the resulting polymer (for structures, see Figure 10.46). Long chain and branched alkyl289 as well as phenylene290 derivatives of (Ib) change color from deep blue to sky and have been found to be highly soluble in common solvents like toluene. 10.8.4 Copolymers To control the optical absorption spectrum it was anticipated that comonomers will introduce additional electronic states in the corresponding copolymers that will tune the band gap and allow multicoloring.291 EDOT and alkylenedioxythiophene derivatives, especially propylenedioxythiophenes, have been used as structural units in diade and triade polymers in various approaches to A-B-A-type poly(BEDOT-arylenes) (V) and their wide range of electrochemical properties (Figure 10.47).271 The incorporation of electron accepting groups like the fluoro groups of 2,5-difluorobenzene results only in a small net effect on the overall polymer redox properties of Va, which can be attributed to slightly more extended
6911X.indb 231
10/1/10 9:01:04 PM
232
PEDOT
F
O *
a
O S
X
S
F n
O
*
b
N
O c N CH3
V
CH
d X= S
e
C9H12
N N
S
H20C9
O
O
C8H17
f X
S VI
n
* H17C8
g
H17C8
C8H17
Figure 10.47 Chemical structures of A-B-A-type V and A-B-type VI poly(BEDOT-arylenes).
delocalization of electrons charges in the polymers.292 The incorporation of strong electron acceptors like 2-5-pyridinediyl293 in Vb, carbazoles294,295 in Vc, cyanovinylidene296 in Vd, and more recently 4,4′-dinonyl-2,2′-bithiazole297 Ve allows an additional n-doped charged transparent reduced state and three distinct colors. While A-B-A with composed alkylene dioxythiophene monomeric A units can be polymerized oxidatively to the corresponding polymer, linear (AB)n and star branched polymers have to be made by organometallic reactions due to a low reactivity of the B unit. (AB)n polymers with EDOT as the A unit and phenylene (VIf) as well as fluorene (VIg) structures as the B unit were synthesized by Suzuki polycondensation298 and show electrochromic behavior without any conspicuous properties. On the other hand, hyperbranched polymers with an additional triarylaminic core unit and similar composed branches made from 3,4-ethylenedioxythiophene and didodecyloxybenzene structures299 exhibited multicolor electrochromism to produce the three basic colors in the RGB system: red, green, and blue.
6911X.indb 232
10/1/10 9:01:05 PM
233
Applications
O *
O
S
S
X
n
O
*
X = -S-S-, -Se-Se-, -Te-Te-, -Se-, -Te-
O
VII Figure 10.48 Chemical structures of A-B-A-type organic–inorganic hybrid PEDOT polymers VII.
10.8.5 Organic–Inorganic Hybrid Polymers A relatively new approach to modify the color of EDOT has been the incorporation of electron-rich chalcogenides between two EDOT structural units to lower the oxidation potential of the monomer as compared to that of EDOT.300 This approach yields organic–inorganic hybrid polymers VII that were found to have transparent yellow color in the reduced state and are opaque blue in the oxidized state, which made them attractive as anodically coloring electrochromic layers. By modification with gold or silver nanoparticles, the color of PEDOT:PSS can be tuned associated with the surface plasmon absorption resonance of the metal nanoparticles and the excitation of the bipolaron band of the conducting polymer to green or violet.301 The mixtures can be used as hybrid electrochromic layers (Figure 10.48).302 10.8.6 PEDOT with Pendant Electrochromic Dyes Ko et al. employed the functionality of EDOT-CH2OH to attach additional electrochromic chromophores as side chains303 of the PEDOT polymer in which both moieties, the viologen and the PEDOT main chain may contribute to the electrochromic effect to allow color tuning and multiple electrochromism. In the polymer PolyViolEDOT the cathodically dyeing viologen group is highly transparent as a dication and matches PEDOT in terms of color change. Upon reduction it switches from the sky blue dicationic state to the deep blue radical cation in parallel to PEDOT. However, both electrochromic functionalities, the polymer and the attached chromophore, apparently do not interact electrochemically and show clearly separated electrochemical oxidation and reduction peaks, thus switching in two distinct steps. Upon reduction the optical absorption increase it to a maximum after which it decreases to some extent at higher potentials. Besides PEDOT other conducting polymers have been used to attach electrochromic dyes such as polypyrrole, polythiophene, poly(cyclopentadithiophene), and poly{1.4-bis[2-(3.4-ethylenedioxy)thienyl]benzene}perylenetetracarboxylic diimides besides viologens as chromophores. The same
6911X.indb 233
10/1/10 9:01:06 PM
234
PEDOT
O
(CH2)6 N
+
N
–
O
(CH2)3 CH4
PF6–
PF6
O
+
m+ n
S
PolyViolEDOT
O
O m+ S
n
+
*
(CH2)6 N
+
N
Br –
PEDOT
+ n
*
Br – PolyViol
Figure 10.49 Chemical structures of viologen side group modified PEDOT (PolyViolEDOT), polyethylenedioxythiophene (PEDOT), and polymeric viologens (PolyViol).
principle has been applied by combining PEDOT with viologen polymers in a layer-by-layer assembly304 (see Figure 10.49 and Section 10.8.7). 10.8.7 Blends and Layer-by-Layer Deposition Electrochromic layers may be tuned by blending two polymeric systems to exhibit either intermediate or synergistic properties. To prepare defined layer structures elaborated preparation techniques have been developed starting from initial spin coating of blend films, electropolymerization on top of spincoated films of one component, spin coating on top of electropolymerized films, and the highly sophisticated layer-by-layer deposition technique. Ferraris and colleagues studied laminate films of PEDOT:PSS spin coated on electropolymerized N-methylpolypyrrole, polyaniline spin coated on PEDOT: PSS, and polyaniline deposited on electropolymerized N-methylpolypyr role.305 The films were analyzed using CIE (x, y)-chromaticity coordinates. The observed colors in their fully doped and reduced states were found to be linearly dependent on the color coordinates of the two individual polymers of the laminate, which allows for adjusting the observed color by variation of layer thickness of an electrochromic device in a viable and predictable way. Inganäs306 studied a laminate structure made from polypyrrole electrochemically deposited on top of a PEDOT:PSS layer formed by spin coating. In contrast to Ferraris et al. he found that the spectra of the laminate electrode matches well with spectra of pure polypyrrole. The application of the layer-by-layer deposition technique originally developed by Decher et al.307–309 intended to provide a simple and reproducible method that allows the formation of homogeneous blend layers.
6911X.indb 234
10/1/10 9:01:06 PM
235
Applications
Dipping
Dipping
Substrate Final layer structure
Solution of polycation
Solution of polyanion
Figure 10.50 Assembly of layer-by-layer laminates.
The layer-by-layer technique utilizes the fact that polymers with basic (e.g., aminic) groups may form salts upon contact with acidic PEDOT:PSS. Thus disposing a surface to dilute solutions of polycation and polyanion in an alternating manner followed by a copious water rinse in between results in alternating self-assembled monolayers. Each monolayer is fixed on top of the preceding layer by a salt formation reaction between the polymers (Figure 10.50). This deposition method allows the assembly of materials of different functionalities into a single film without phase separation issues. Layer-bylayer assembly works well with PEDOT-S sodium salt as the polyanion combined polyallylamine hydrochloride (PAI)278–280 as well as PEDOT:PSS as the polyanion combined with linear poly(ethyleneimine) (LPEI), polyaniline310 and viologen polymers304 as polycations. In layer-by-layer assemblies of PEDOT:PSS with a viologen polymer it was clearly shown that both redox steps, the redox step from the PEDOT main chain and the redox step from the viologen, occur successively and not simultaneously. 10.8.8 Electrolytes During switching ions will move from the chromophore and the ion storage layer to counterbalance the charges injected into both layers due to the redox reactions. Electrolytes provide a suitable ion-conducting medium for ions to move between the layers. Electrolytes need to be optimized for a maximum of ion conductivity combined with maximum mechanical stability. Ions with a smaller van der Waals radius move faster in a given electrolyte, therefore usually lithium ions are employed due to their small size and stability toward reduction. Although protons are even smaller and exhibit an even higher ion conductivity311
6911X.indb 235
10/1/10 9:01:07 PM
236
PEDOT
they are in general less suitable since protons can be irreversibly reduced to hydrogen and therefore degrade. The overall ion conductivity is dominated by movement of cations since the larger anions have been found to be much slower or even immobile.312 Liquid electrolytes are usually formulated from liquid compounds that effectively chelate lithium ions like ethylene glycol, ethylene, and propylene carbonate. Solid electrolytes are composed from liquid electrolytes, lithium salts that are incorporated into a solid polymer matrix.313,314 The name “gel electrolyte” has been connected with these formulations due to their soft gel-like nature. The polymeric matrix in general contains polyethylene oxide units which chelate lithium ions like the liquid electrolytes. A simple formulation of a gel electrolyte may be composed of ethylene carbonate, lithium triflate, and poly(ethylene glycol)s as the polymeric component. To increase the mechanical strength cross-linking the polymeric component may contain acrylate groups that can be polymerized photochemically after the cell assembly and so modify the electrolyte to become an adhesive for the cell.258 The mechanism of the lithium ion movement and their incorporation into the electrochromic polymer315,316 as well as the ion storage material317 during the redox process has been studied in detail. Imidazolium salt base ionic liquids, which had been formulated into a polymer matrix, increase the ion conductivities up to 10–2 S/cm and were found to be more stable than conventional poly(ethylene oxide) electrolytes.318 10.8.9 Ion Storage Materials The ion storage layer acts in a similar active redox manner as the electrochromic layer. During switching this layer incorporates and releases lithium ions, but in the opposite direction of the electrochromic layer. Usually this layer does not change its optical absorption during switching. Metal oxides or metal oxide combinations have been proposed as ion storage layers such as oxides based on cerium and titanium oxide319 as well as nickel oxides320 or vanadium oxides.321 Metal oxides suffer from their brittle nature, which limits their use in flexible devices. A special case are ion storage layers with complementary electrochromic activity, which may be assembled with a PEDOT layer. In a complementary electrochromic cell, both layers—dye and bleach—simultaneously increase to the overall optical contrast. Tungsten trioxide as well as Prussian Blue, iron(III) hexacyanoferrate(II/III) bleach upon reduction and were used in combination with PEDOT. Cells made from PEDOT in combination with Prussian Blue exhibited a deep blue-violet color at a potential of –2.1 V and become light blue at 0.6 V.322 Combination cells of alkylenedioxypyrrols or -thiophenes with ferrocene323 have been proposed as variable optical attenuator devices due to a large dynamic range of optical attenuation at the telecommunication wavelength of 1550 nm.324
6911X.indb 236
10/1/10 9:01:07 PM
Applications
237
Besides inorganic metal oxides, redox active polymers can be employed as ion storage and complementary electrochromic materials. Cells with two electrochromic polymers with complementary dyeing characteristics have been called “dual polymer cells.” Surprisingly, it is possible even to use a second PEDOT layer for ion storage.320 In each switching state one layer should be bleached and one should be dyed. Therefore, the optical contrast would be low. However, an optical contrast of 1:2 has been obtained if the switching endpoints are tightly controlled. This may be explained by the fact that for PEDOT dyeing does occur linearly with the applied electrochemical potential, and points can be found with a minimum and a maximum transparency. However, the optical contrast of such a cell is inherently limited. 10.8.10 Dual Polymer Cells Cells with two electrochromic polymers with complementary dyeing characteristics, dual polymer cells, like cells in which PEDOT:PSS was combined with polypyrrole325 gained significant attraction as early as 1997 since besides flexibility they offer potentially high optical contrasts (>50% transparency) as well as multicolor capabilities326 and gave rise to an optimistic view of the future of such systems.327 Transition metal ions, electronically coupled to the conjugated backbone, allowed further tuning of the polymer’s redox properties, which behaved complementary to PEDOT. A-B-type polymers with EDOT and carbazole structural units turned out to be another class of polymers anodically coloring that can successfully be combined with alkylenedioxythiophenes.328,329 Dual cells made from PEDOT and polyaniline as the complementary layer330,331 seem to be less suitable for technical applications since they exhibit a pale yellow bleached state that switches to dark blue. However, a combination with a polyaniline derivative, poly(o-methoxy aniline) a device with high optical contrast of 75% and a transition from transparent to dark blue was obtained, which exhibited significant degradation in short time. 10.8.11 Substrates and Patterning Despite the fact that flexible all-polymeric electrochromic cells have gained significant attraction, in many cases glass has been employed as a substrate for electrochromic cells. Conductive glasses such as ITO-coated glass or other less commonly used glasses like fluorine-doped tin-dioxide-coated (heat shielding) automobile glass are readily available in technical quantities and can easily be employed as transparent electrodes especially for smart windows. The use of polymeric ion storage layers opened a way to flexible “all polymer cells.” The next step was to replace the metal oxides used as transparent electrodes by conductive polymers, which due to their brittleness represent the Achilles heel of a flexible device. Reynolds and coworkers coated PEDOT:PSS as electrodes on plastic substrates to yield “true all polymeric”
6911X.indb 237
10/1/10 9:01:07 PM
238
PEDOT
electrochromic cells, which employed PEDOT:PSS as the transparent electrode, electrochemically deposited poly(alkylenedioxythiophene) as the electrochromic polymer, a carbazole copolymer as the complementary electrochromic polymer, and a polymeric electrolyte composition.332 Recently electrochemical cells even have been printed on textiles.333 For smart windows structuring is not an issue, but in electrochromic displays either the electrode material or the electrochromic polymer needs to be structured to obtain suitable patterns. In addition to classical subtractive methods of removing material from undesired areas, PEDOT:PSS may be structured by printing methods. Berggren’s group. demonstrated the usefulness of printing in combination with the substrate paper by printing an electrochromic active matrix displays.334 Employing a printing process they succeed to print the driving transistors and the electrochromic display in one step. Both the electrochromic display and the electrochemical transistors exploit the electrochemical switching of PEDOT either as a change in optical absorption or as a change in conductivity of an electrochemical transistor.335 Patterning of electrochromic films by soft lithography may increase the coloring efficiency by diffraction.336 Laterally configured polymers and metallic interdigitated electrodes337 yielded a device that switched by stepping the applied voltage between –1.2 V to +1.2 V from a highly reflective gold state to an absorptive blue state.
10.9 Organic Field-Effect Transistors 10.9.1 Introduction The prospect of electronic circuits on flexible substrates that can be fabricated by large area and low-cost printing techniques has motivated researchers to provide appropriate materials as well as device engineers to establish new deposition techniques. Predicted applications for organic field-effect transistors (OFETs) are radio frequency identification tags, active matrix back-planes for electrophoretic displays, and low-cost sensors. The working principle of OFETs has been outlined by others in detail.338–340 Figure 10.51 depicts the cross-section of an OFET setup exploiting bottom electrode geometry. In case of an organic p-type semiconductor as illustrated in Figure 10.51, in which the mobility of holes is much larger than the mobility of electrons, the organic layer becomes conductive at the interface to the dielectric layer when a negative bias is applied at the gate electrode. As in Si-based MOS-FETs the drain current (ID) can be controlled by the gate potential (UG) when a bias is applied between source and drain (UD).
6911X.indb 238
10/1/10 9:01:08 PM
239
Applications
ID
UD –
Organic semiconductor UG
Source + –
+ + + + + Dielectric layer
Drain
Gate electrode Substrate
Figure 10.51 Cross-section of an OFET with bottom electrode geometry. A conductive channel is created at the interface between the p-type semiconductor and the dielectric layer when a negative bias is applied at the gate electrode.
Prominent materials for p-type semiconductors are pentacene341 or oligothiophenes342; both are commonly deposited by vacuum sublimation. To reduce production costs materials that can be processed from solution like poly(3-alkyl-thiophene)343 or TIPS-pentacene344 have gained considerable attention. A semiconductor with high-charge carrier mobility is necessary to achieve low driving voltages, high switching speeds, and high rectification ratio in the device. In parallel, the dielectric layer and appropriate electrodes determine the OFET properties as well. The electrode’s geometry, especially the channel length (L) between source and drain has direct impact on the drive voltage and the switching speed and should be as narrow as possible. In addition, the injection from the source electrode into the semiconductor is of importance and the contact resistance should be low compared with the channel resistance to avoid voltage drop.345 PEDOT:PSS has been applied in OFETs dominantly as printable electrode. There are, however, also reports employing the conducting polymer as interlayer to improve charge injection and as active layer. 10.9.2 PEDOT:PSS as Electrodes Metal electrodes for source, drain, and gate in OFETs are most commonly deposited by photolithographic etching or by evaporating metals through a shadow mask to achieve small patterns with dimensions in the order of microns. To reduce manufacturing costs it is however necessary to find simpler means of deposition. Printing the materials directly onto a substrate is considered as a reliable approach. The conductive polymer PEDOT:PSS is solution based and can be deposited as an electrode in OFETs by accessing various printing techniques as
6911X.indb 239
10/1/10 9:01:08 PM
240
PEDOT
a
n
O
S
SO3–
HSO3 O O O S S S+ nS O O O O H
O S
c
m
PVP
G
PI
d
S
d
b 200 nm
D
S
G
50 µm
D
µm
PEDOT 10
n
S
O
PEDOT
PI 20
30
40
µm
20 10
Channel
Figure 10.52 (a) Schematic diagram of high-resolution inkjet printing (IJP) onto a prepatterned substrate. (b) AFM showing accurate alignment of inkjet printed PEDOT:PSS source and drain electrodes separated by a repelling polyimide (PI) line with L = 5 µm. (c) Schematic diagram of the topgate IJP TFT configuration with an F8T2 semiconducting layer (S, source; D, drain; and G, gate). (d) Optical micrograph of an IJP TFT (L = 5 µm). The arrow indicates pronounced roughness of the unconfined PEDOT boundary. (Adapted from H. Sirringhaus, T. Kawase, R. H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, and W. P. Woo, 2000, Science 290:2123–2126.)
summarized next. PEDOT:PSS-electrodes for gate were deposited by the screen printing technique.346 The resolution of the pattern is limited to approximate 100 µm defined by the coarseness of the screen. Sirringhaus et al. proposed a concept of an all-polymer transistor with PEDOT:PSS electrodes deposited by inkjet printing. A channel length of L = 5 µm could be obtained as illustrated in Figure 10.52.347 To keep the inkjetprinted PEDOT:PSS lines separated at this low dimension polyimide banks structured by photolithography were deposited on the substrate first. Another all-polymer-FET with inkjet printed PEDOT:PSS electrodes was demonstrated by Liu et al. by choosing a different setup with the gate electrode printed first followed by the inkjet printed dielectric layer, semiconducting layer, and source-drain electrodes.348 Self-aligned inkjet deposited electrodes of PEDOT:PSS in OFETs were realized by Li et al.349 The alignment was achieved by contact printing of a self-assembled monolayer (SAM) and hence turning the surface hydrophobic at those parts where PEDOT:PSS wetting should be inhibited. The interfacial contact between inkjet printed PEDOT:PSS and poly(3-hexyl-thiophene) was improved by adding DMSO
6911X.indb 240
10/1/10 9:01:10 PM
241
Applications
to the dispersion.350 Simultaneously, the resistive loss of the electrode was reduced by increasing its conductivity. Electrodes composed of PEDOT:PSS in all-polymer integrated circuits were structured down to dimensions of 2.5 µm channel length by I-line lithography.351 An photosensitive cross-linker was added to the PEDOT:PSS dispersion to make the irradiated parts become insoluble. The non-illuminated parts of the spin-coated layer were removed by purging with water. To realize integrated circuits it is necessary to connect gain, source, and drain electrodes of adjacent transistors. This was accomplished by creating contact holes with standard photolithography subsequently filled by PEDOT:PSS to form the vertical interconnects (vias).352 Spin-coated films of PEDOT:PSS were lithographically structured by Rost et al. achieving electrode separations of 1 to 50 µm.352 Becker et al. followed a concept based on selective area electropolymerization of PEDOT:PSS on a prepatterned anode. The electrochemical deposited conductive polymer was transferred on a flexible substrate and a channel length of 5 µm was achieved.346 Soft lithography (microcontact printing) was employed to pattern source and drain contacts. A poly(dimethylsiloxane) (PDMS) stamp, reproducing the electrode’s layout, was used for transferring the PEDOT:PSS pattern on the substrate.353 It was possible to realize final electrode structures with a channel length of down to 2 µm.354,355 10.9.3 PEDOT:PSS as an Interlayer To operate an organic field effect transistor at low bias it is necessary to avoid large contact resistance within the source-drain circuit.345 It has been reported that an interlayer of PEDOT:PSS as illustrated in Figure 10.53 will promote hole injection.356,357 This is in accordance with the observations made in OLEDs and discussed in detail in Section 10.6. PEDOT:PSS Pentacene Au
Au Dielectric layer Gate electrode Substrate
Figure 10.53 A thin coating of PEDOT:PSS on the Au contact will increase charge carrier injection from Au source contact into the pentacene layer.
6911X.indb 241
10/1/10 9:01:10 PM
242
PEDOT
Schroeder et al. deposited a 45 nm thin layer of PEDOT:PSS electrochemically onto the Au source electrode356 and observed an increase of sourcedrain current no matter if pentacene or poly(triarylamine) (PTAA) was employed as the semiconductor. The increase was found to be 2.5-fold for pentacene and 6-fold for PTAA where additionally a decrease of threshold voltage was observed. The increase of hole injection by the PEDOT:PSS modified Au contact was discussed in terms of locally confined contact doping at the interface between source and pentacene. The interface between pentacene and a layer of PEDOT:PSS being spin coated onto Au source and drain contacts was analyzed by AFM, grazing incident angle x-ray diffraction and UPS.357 It was found that the PEDOT:PSS interlayer reduced the hole-injection barrier and simultaneously led to increased crystalline domains of the semiconductor. Both changes improved the OFET characteristics. 10.9.4 PEDOT:PSS as an Active Layer In OFETs, the active layer is composed of an undoped semiconductor that forms a conductive channel at the interface with the dielectric when an electrical field is applied. It was therefore unexpected to observe a field effect in devices comprising an intrinsic conducting polymer like PEDOT:PSS as the active layer (Figure10.54a).358 ID was modulated by UG as depicted in Figure 10.54b.359 Others reported that the current was suppressed or enhanced by applying a positive or negative gate potential respectively.360 The change of resistance of the conducting channel on variations of the external field was different from semiconductor-based OFETs. As illustrated in Figure 10.54b the current change on the applied electrical field was slow, in the order of minutes. Additionally Epstein et al. observed that the field effect disappeared completely when the temperature was decreased from room temperature by only 10°C.359 This has led to the assumption that ion diffusion motion within the PEDOT:PSS channel is responsible for the observed ID modulation. Others observed a pronounced dependence of the field effect on the humidity level. In dry atmosphere almost no change of ID on the gate voltage was found.361,362 An electrochemical mechanism was proposed to explain the electric field dependence including the dedoping of PEDOT in the presence of water. A different explanation was proposed by Hsu et al.363 Dedoping of PEDOT was ruled out as the total number of injected ions from out of the gate into the PEDOT:PSS channel was too low to explain the observed current decrease. Instead a model was favored assuming a change of percolation paths caused by rendered ion positions. A small fraction removal of mediated hopping states near the Fermi level on charge transport paths causes carriers to hop over longer distance to conduct current and therefore ID is reduced.
6911X.indb 242
10/1/10 9:01:10 PM
243
Applications
Gate Dielectric layer PEDOT:PSS
Source
Drain
Substrate
20
IDS [µm]
200
0
Vg [V]
(a)
160 120 80 0
10
20
30
40
50
60
Time [min] (b) Figure 10.54 (a) Transistor structure with PEDOT:PSS as active layer, (b) source–drain–current IDS as a function of time (lower trace). A positive gate voltage was applied periodically (upper trace) to modulate IDS.
Nilsson et al. created an organic transistor based on a PEDOT:PSS film as the active layer that was covered by an electrolyte comprising metal ions (M+).364 The conductivity of the p-type channel was modulated by reducing or oxidizing PEDOT according to:
PEDOT+PSS – + M+ + e– ↔ PEDOT0 + M+PSS – .
By applying a negative potential of only Ug = 0.1 to 1 V to the electrolyte it was possible to transfer the conductive PEDOT:PSS film in its almost nonconductive state. The redox process was fully reversible and switching speeds in the order of seconds were achieved.
6911X.indb 243
10/1/10 9:01:11 PM
244
PEDOT
References
6911X.indb 244
1. H. Shirakawa, E. J. Louis, A. G. MacDiarmid, C. K. Chiang, and A. J. Heeger. 1977. Synthesis of electrically conducting organic polymers: Halogen derivatives of polyacetylene, (CH)x. J Chem Soc, Chem Commun 578–580. 2. T. A. Skotheim, R. L. Elsenbaumer, and J. R. Reynolds. 1998. Handbook of Conducting Polymers, 2nd ed., New York: Marcel Dekker. 3. E. Reed, J. Marshall, and R. Hahn. 2002. How low can you go—tantalum polymer capacitors with ESR under 7 mΩ. In: Proceedings of the 16th Passive Components Symposium CARTS Europe, 60–67. Port St. Laurent, Electronic Components Institute Internationale Ltd. 4. J. S. Shaffer. EP 0135223 (North American Philips Corporation), Prior: August 11, 1983. 5. M. Fukuda, H. Yamamoto, and I. Isao. EP 274 755 (Japan Carlit CO. Ltd), Prior: January 13, 1987. 6. S. Tsuchiya, Y. Kudoh, T. Kojima, M. Fukuyama, and S. Yoshimura. EP 336 229 (Matsushita Electric Industrial Co., Ltd.), Prior: March 31, 1988. 7. H. Ikeda, N. Furukawa, K. Nishio, M. Fujimoto, H. Ono, and O. Ando. EP 342 998 (Sanyo Electric Co Ltd, Mitsubishi Chemical Corporation), Prior: May 20, 1988. 8. Y. Salki, E. Tsuchida, K. Yamamoto, and S. Takeoka. EP 477 584 (NEC Corporation), Prior: August 31, 1990. 9. Y. Kudoh, S. Tsuchiya, T. Kojima, M. Fukuyama, and S. Yoshimura. 1991. An aluminum solid electrolytic capacitor with an electroconducting-polymer electrolyte. Synth Met 41–43(3):1133–1136. 10. M. Fukuyama, Y. Kudoh, N. Nanai, and S. Yoshimura. 1993. Materials science of conducting polymers: An approach to solid electrolytic capacitors with a highlystable polypyrrole thin film. Mol Cryst Liq Cryst Sci Technol A 224:61–67. 11. M. Ohi, T. Fukami, A. Kobayashi, and H. Taniguchi. 1995. NEC Technical J 48(10):77. 12. F. Jonas, G. Heywang, W. Schmidtberg, J. Heinze, and M. Dietrich. EP 339 340 (Bayer AG), Prior: April 22, 1988. 13. F. Jonas and G. Heywang. EP 340 512 (Bayer AG), Prior: April 30, 1988. 14. F. Jonas and G. Heywang. 1994. Technical applications for conductive polymers. Electrochim Acta 39(8–9):1345–1347. 15. Y. Kudoh, A. Akami, and Y. Matsuya. 1999. Solid electrolytic capacitor with highly stable conducting polymer as a counter electrode. Synth Met 102(1–3):973–974. 16. Saga Newspaper, Japan, June 17, 1999. 17. T. Zednicek, P. Vasina, Z. Sita, and B. Vrana. 2003. Lead-free soldering effect to tantalum capacitors. In: Conference Proceedings of 26th International Spring Seminar on Electronics Technology: Integrated Management of Electronic Materials Production, 85–89. Slovakia, Institute of Electrical and Electronics Engineers. 18. S. Otani. 1998. Investigation of electroconductive polymers for solid electrolytic capacitors. Denkai Chikudenki Hyoron 49(2):107–118 . 19. E. Reed, J. Chen, J. Marshall, J. Paulsen, and R. Weisenborn. 2003. New 125°C capable tantalum polymer capacitors. In: Proceedings of the 17th Passive Components Symposium CARTS Europe, 24–31. Stuttgart, Electronic Components Institute Internationale Ltd.
10/1/10 9:01:11 PM
Applications
245
20. A. Michaelis, C. Schnitter, K. H. Reichert, R. Wolf, and U. Merker. 2002. New tantalum metal powder quality for solid electrolyte capacitors. In: Proceedings of the 16th Passive Components Symposium CARTS Europe, 209–212. Port St. Laurent, Electronic Components Institute Internationale Ltd. 21. M. M. Lohrengel. 1993. Thin anodic oxide layers on aluminum and other valve metals: High-filed regime. Mater Sci Eng R 11(6):243–294. 22. H. Yamamoto, M. Oshima, M. Fukuda, I. Isa, and K. Yoshino. 1996. Characteristics of aluminum solid electrolytic capacitors using a conducting polymer. J Power Sources 60(2):173–177. 23. Y. Kudoh, T. Kojima, M. Fukuyama, S. Tsuchiya, and S. Yoshimura. 1996. Covering anodized aluminum with electropolymerized polypyrrole via manganese oxide layer and application to solid electrolytic capacitor. J Power Sources 60(2):157–163. 24. Y. Kudoh, K. Akami, and Y. Matsuya. 1998. Chemical polymerization of 3,4ethylenedioxythiophene using an aqueous medium containing an anionic surfactant. Synth Met 98(1):65–70. 25. H. Karabulut, U. Merker, and J. W. Schultze. 2003. Poly-(3,4-ethylenedioxythio phene) as cathode material in tantalum capacitors. GDCh Monographie 29:181. 26. U. Merker, K. Reuter, K. Wussow, S. Kirchmeyer, and U. Tracht. 2002. PEDT as conductive polymer cathode in electrolytic capacitors. In: Proceedings of the 16th Passive Components Symposium CARTS Europe, 71–76. Port St. Laurent, Electronic Components Institute Internationale Ltd. 27. U. Merker, K. Wussow, S. Kirchmeyer, C. Schnitter, and K. Lerch. 2003. Manufacturing process for low ESR polymer electrolyte capacitors. In: Proceedings of the 17th Passive Components Symposium CARTS Europe: 79–82. Stuttgart, Electronic Components Institute Internationale Ltd. 28. D. M. De Leeuw, C. M. J. Mutsaers, and M. M. J. Simenon. EP 615 257 (Philips Electronics N.V.), Prior: March 9, 1993. 29. U. Merker, S. Kirchmeyer, and K. Wussow. WO 2004/088672 (H.C. Starck), Prior: April 2, 2003. 30. U. Merker, K. Wussow, and W. Lövenich. 2005. New conducting polymer dispersions for solid electrolyte capacitors. In: Proceedings of the 19th Passive Components Symposium CARTS Europe, 30–35. Prague, Electronic Components, Assemblies & Materials Association. 31. U. Merker, W. Lövenich, and K. Wussow. 2006. Conducting polymer dispersions for high-capacitance tantalum capacitors. In: Proceedings of the 20th Passive Components Symposium CARTS Europe, 21–26. Bad Homburg, Electronic Components, Assemblies & Materials Association. 32. U. Merker, W. Lövenich, and K. Wussow. 2007. Tuning conducting polymer dispersions for high-CV tantalum capacitors. In: Proceedings of the 21st Passive Components Symposium CARTS Europe, 293–298. Barcelona, Electronic Components Association. 33. K. Nogami, K. Sakamoto, T. Hayakawa, and M. Kakimoto. 2007. The effects of hyperbranched poly(siloxysilane)s on conductive polymer aluminum solid electrolytic capacitors. J Power Sources 166(2):584–589. 34. J. Young, J. Qiu, and R. Hahn. 2008. High voltage tantalum polymer capacitors. In: Proceedings of the 28th Passive Components Symposium CARTS US, 1–12. Newport Beach, Electronic Components Association.
6911X.indb 245
10/1/10 9:01:12 PM
246
PEDOT
35. Y. Freeman, W. R. Harrell, I. Luzinov, B. Holman, and P. Lessner. 2008. Electrical characterization of polymer tantalum capacitors with poly (3,4-ethylenedioxythiophene) cathode. In: Proceedings of the 22nd Passive Components Symposium CARTS Europe: 25–33. Helsinki, Electronic Components Association. 36. J. D. Prymak. 1999. Replacing MnO2 with Conductive Polymer in Tantalum Capacitors. In: Proceedings of the 19th Passive Components Symposium CARTS US, 1–6. New Orleans, Electronic Components Institute Internationale Ltd. 37. U. Merker, E. Droste, and A. Michaelis. 2002. Thermal runaway of tantalum capacitors. In: Proceedings of the 16th Passive Components Symposium CARTS Europe, 102–106. Port St. Laurent, Electronic Components Institute Internationale Ltd. 38. C. F. Coombs. 2007. Printed Circuits Handbook. Columbus, OH: McGraw Hill. 39. Cookson Electronics. www.cooksonelectronics.com (accessed August 2010). 40. J. Hupe and W. Kronenberg. EP 402381 (Blasberg Oberflächentechnik GmbH), Prior: March 3, 1988. 41. F. Jonas and G.-D. Wolf. EP 553671 (Bayer AG). Prior: January 29, 1992. 42. G.-D. Wolf, F. Jonas, and R. Schomaecker. EP 707440 (Bayer AG), Prior: October 12, 1994. 43. R. Czeczka and L. Stamp. DE 10124631 (Atotech Deutschland GmbH), Prior: May 18, 2001. 44. J. Hupe and S. Fix. DE 19527065 (Blasberg Oberflächentechnik GmbH), Prior: July 25, 1995. 45. S. Kirchmeyer and F. Jonas. EP 1234487 (Bayer AG), Prior: January 14, 2000. 46. G.-D. Wolf and F. Jonas. DE 19637018 (Bayer AG), Prior: September 12, 1996. 47. J. Hupe, W. Kronenberg, J. Kickelhain, and D. Maier, EP 1191127 (Enthone-OMI, LPKF Laser & Electronics AG), Prior: September 26, 2000. 48. E. Stafstrom. 2006. Utilizing conductive polymers to improve HDI yields. CircuiTree 10:10,12,14. 49. Utilizing conductive polymers to improve HDI yields. www.enthone-omi.com/ de/docs/StafstromENVHDI1004CircuiTree.pdf (accessed August 2010). 50. J. Hupe, G. D. Wolf, and F. Jonas. 1995. DMS-E—Bekanntes Prinzip mit neuer Basis. Galvanotechnik 86(10):3404–3411. 51. V. Tsakovka, S. Winkel, and J.W. Schultze. 2001. Anodic polymerization of 3,4-ethylenedioxythiophene from aqueous microemulsions. Electrochim Acta 46(5):759–768. 52. F.-H. Haegel, J. Schlüpen, J. W. Schultze, S. Winkels, and C. Stromberg. 2001. Anodic polymerization of thiophene derivatives from microemulsions and liquid crystals. Electrochim Acta 46(26–27):3973–3994. 53. C. Stromberg, V. Tsakova, and J. W. Schultze. 2003. Composition of the micro emulsion and its influence on the polymerisation and redox activation of PEDOT. J Electroanalyt Chem 547(2):125–133. 54. D. Schattka, S. Winkels, and J. W. Schultze. 1997. Through hole plating with conducting polymers. Analytical surface investigations of micro-electroplating. Metalloberfläche 51(11):823–824,826–828. 55. D. S. Ginley and C. Bright. 2000. Transparent conductive oxides. MRS Bulletin, 5(8):15–18. 56. B. Szyszka. 2001. Transparent and conductive oxide layers. Vakuum in Forschung und Praxis 13(1):38–45. 57. B. J. Ingram, G. B. Gonzales, D. R. Kammler, M. I. Bertoni, and T. O. Mason. 2004. Chemical and structural factors governing transparent conductivity in oxides. J Electroceram 13(1–3):167–175.
6911X.indb 246
10/1/10 9:01:12 PM
Applications
247
58. G. Gustafsson, Y. Cao, G. M. Treacy, F. Klavetter, N. Colaneri, and A. J. Heeger. 1992, Flexible light-emitting diodes made from soluble conductive polymers. Nature 357(6378): 477–479. 59. J. Peltola, C. Weeks, I. A. Levitsky, D. A. Britz, P. Glatkowski, M. Trottier, and T. Huang 2007. Carbon-nanotube transparent electrodes for flexible displays. Information Display 23(2):20–23. 60. R. A. Gilstrap Jr., C. J., Capozzi, C. G. Carson, R. A. Gerhardt, and C. J. Summers. 2008. Synthesis of nonagglomerated indium tin oxide nanoparticle dispersions. Adv Mater 20(21):4163–4166. 61. Y. Qiu, L. Duan, and L. Wang. 2002. Flexible organic light-emitting diodes with poly-3,4-ethylenedioxythiophene as transparent anode. Chin Sci Bull 47(23):1979–1982. 62. R. Paetzold, K. Heuser, D. Henseler, S. Roeger, G. Wittmann, and A. Winnacker. 2003, Performance of flexible light-emitting diodes under bending conditions. Appl Phys Lett 82(19):3342–3344. 63. D. R. Cairns and G. P. Crawford. 2005. Electromechanical properties of transparent conducting substrates for flexible electronic displays. Proceedings of the IEEE 93(8):1451–1458. 64. A. Elschner, F. Jonas, S. Kirchmeyer, and W. Lövenich. 2008. PEDOT-based layers for TCO-substitution and hole-injection. SID 08 Digest 29.2:407–410. 65. R. G. Gordon. 2000. Criteria for choosing transparent conductors, ed. D. S. Ginley, C. Bright, MRS Bulletin, Materials Research Society, 506 Keystone Drive, Warrendale, PA 15086–7573 USA, 52–57. 66. S. A. Carter, M. Angelopoulos, S. Karg, P. J. Brock, and J. C. Scott. 1997. Polymeric anodes for improved polymer light-emitting diode performance. Appl Phys Lett 70(16):2067–2069. 67. K. Neyts, M. Marescaux, A. U. Nieto, A. Elschner, W. Lövenich, K. Fehse, Q. Huang, K. Walzer, and K. Leo. 2006. Inhomogeneous luminance in organic light emitting diodes related to electrode resistivity. J Appl Phys 100(11):1145131–114513-4. 68. K. Neyts, A. Real, M. Marescaux, S. Mladenovski, J. Beeckman. 2008. Conductor grid optimization for luminance loss reduction in organic light emitting diodes. J Appl Phys 103(9):093113-1–093113-5. 69. J. C. Scott, S. A. Carter, S. Karg, and M. Angelopoulos. 1997. Polymeric anodes for organic light-emitting diodes. Synth Met 85(1–3):1197–1200. 70. W. H. Kim, A. J. Mäkinen, N. Nikolov, R. Shashidhar, H. Kim, and Z. H. Kafafi. 2002. Molecular organic light-emitting diodes using highly conductive polymers as anodes, Appl Phys Lett 80(20):3844–3846. 71. K. Fehse, K. Walzer, K. Leo, W. Lövenich, and A. Elschner. 2007. Highly conductive polymer anodes as replacement of inorganic materials for high efficiency organic light-emitting diodes. Adv Mater 19(3):441–444. 72. K. Fehse, R. Meerheim, K. Walzer, K. Leo, W. Lövenich, and A. Elschner. 2008. Lifetime of organic light-emitting diodes on polymer anodes. Appl Phys Lett 93(8):083303/1–083303/3. 73. J. Huang, X. Wang, A. J. deMello, J. C. deMello, and D. D. C. Bradley. 2007. Efficient flexible polymer light emitting diodes with conducting polymer anodes. J Mater Chem 17(33):3551–3554. 74. H. Mori, Y. Sudou, and T. Yokoyama, Tooru. JP 5166781 (Fujitsu Ltd), Prior: June 14, 1979.
6911X.indb 247
10/1/10 9:01:12 PM
248
PEDOT
75. L. M. Welsh, L. H. Radzilowski, T. A. M. Cloots, B. A. A. Muys, and P. J. M Willaert. US 61429000 (Elo Touchsystems Inc), Prior: July 12, 2000. 76. E. M. Cross, R. S. Moshrefzadeh, F. J. Bottari, D. R. Cairns, A. F. Chernefsky, and P. J. Richter. US 2003/0213624 (3M Innovative Properties company), Prior: May 20, 2002. 77. K. C. Frisch and A. Patsis. 1982. Triboelectrification of polymers. In: Electrical Properties of Polymers, ed. D. A. Seanor, 37–58. New York: Academic Press. 78. F. Jonas and W. Krafft. EP 440 957 (Bayer AG), Prior: February 8, 1990. 79. T. Kawashima and T. Sekiya. JP 2002241613 (Sony Chemical Corp), Prior: February 20, 2001. 80. J. S. Bae, N. H. Kim, and Y. S. Kim. KR 2005070534 (Saehan Ind Inc, Daeha Mantech Co. Ltd.), Prior: December 30, 2003. 81. H. Müller, D. Schultze, F. Jonas, K. Lerch, and H. Weber. EP 821025 (Bayer AG), Prior: July 11, 1996. 82. X. Ding, H. Yang, C. Wu, and L. Chen. 2004. Preparation of antistatic thin film for electronic component packing. Dianzi Yuanjian Yu Cailiao 23(7), 38–40 (CA: 143:471000). 83. T. Wang, Y. Qi, J. Xu, X. Hu, and P. Chen. 2003. Effect of addition of poly(ethylene glycol) on electrical conductivity of poly(3,4-ethylenedioxythiophene)poly(styrenesulfonate) hybrid. Chinese Science Bulletin 48(22):2444 (CA 141: 97175). 84. F. Jonas, K. Lerch, and W. Fischer. EP 752454 (Bayer AG), Prior: July 3, 1995. 85. F. Jonas and U. Guntermann. EP1785450 (H.C. Starck GmbH & Co. KG), Prior: November 10, 2005. 86. F. Jonas, K. Lerch, and U. Guntermann, EP 825219 (Bayer AG), Prior: June 19, 1996. 87. F. Jonas and R. Dhein. DE 4229192 (Bayer AG), Prior: September 2, 1992. 88. Y. Lee and J. Kim. 1999. Transparent and conductive composite of poly(3,4ethylenedioxythiophene) and silica sol-gel materials. Mol Cryst Liq Cryst Sci Technol, Sect. A 337:213–216. 89. M. Mager, F. Jonas, A. Eiling, and U. Guntermann. EP 979514 (H.C. Starck GmbH), Prior: December 4, 1996. 90. H. L. Chang. KR 2000 07535/US 6638566 (Orion Electric Co.), Prior: May 31, 1999. 91. M. Asai, JP 2005313342 (Teijin-Du Pont Film Inc.), Prior: April 27, 2004. 92. Y. Lee and J. Lee. 2000. An organic/inorganic conductive composite: PEDOT/ silicate. Polym Prepr 41(1):595–595 . 93. Y. Lee, J. Kim, and Y. Son. 1999. The preparation of organic/inorganic conducting polymer composite films using in-situ sol-gel process. Polymer (Korea) 23(3):443–449. 94. F. Jonas, A. Karbach, B. Muys, E. van Thillo, R. Wehrmann, A. Elschner, and R. Dujardin, EP 1162238 (Bayer AG), Prior: May 6, 1994/March 3, 1995. 95. F. Jonas, A. Karbach, B. Muys, E. van Thillo, R. Wehrmann, A. Elschner, and R. Dujardin. EP 686 662 (Bayer AG), Prior: May 6, 1994/March 3, 1995. 96. B. D. Martin, N. Nikolov, S. K. Pollack, A. Saprigin, R. Shashidar, F. Zhang, and P. A. Heiney. 2004. Hydroxylated secondary dopants for surface resistance enhancement in transparent poly(3,4-ethylenedioxythiophene)–poly(styrenesulfonate) thin films. Synth Met 142(1–3):187–193. 97. L. A. A. Pettersson, S. Ghosh, and O. Inganäs. 2002. Optical anisotropy in thin films of poly(3,4-ethylenedioxythiophene)–poly(4-styrenesulfonate). Org Electronics 3(3–4):143–148.
6911X.indb 248
10/1/10 9:01:12 PM
Applications
249
98. S. K. M. Jönsson, J. Birgerson, X. Crispin, G. Greczynski, W. Osikowicz, A. W. Denier van der Gon, W. R. Salaneck, and M. Fahlman. 2003. The effects of solvents on the morphology and sheet resistance in poly(3,4-ethylenedioxythiophene)– polystyrenesulfonic acid (PEDOT–PSS) films. Synth Met 139(1):1–10. 99. J. Ouyang, Q. Xu, C.-W. Chu, Y. Yang, G. Li, and J. Shinar. 2004. On the mechanism of conductivity enhancement in poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) film through solvent treatment. Polymer 45(25):8443–8450. 100. S. Timpanaro, M. Kemerink, F. J. Touwslager, M. M. De Kok, and S. Schrader. 2004. Morphology and conductivity of PEDOT:PSS films studied by scanning– tunneling microscopy. Chem Phys Lett 394(4–6):339–343. 101. W. Krafft, F. Jonas, P. Lalvani, H. Andries, and W. Mües. US 5443944 (AgfaGaevert AG), Prior: November 16, 1992. 102. F. Jonas, W. Krafft, and B. Muys. 1995. Macromol Symp 100:169. 103. F. Jonas and J. T. Morrison. 1997. 3,4-polyethylenedioxythiophene (PEDT): Conductive coatings technical applications and properties. Synth Met 85 (1–3): 1397–1398. 104. F. Jonas and K. Lerch. 1997. Leitfähige Kunststoffe: Antistatische Beschichtung mit attraktiven Begleiteigenschaften. Kunststoffe 87(10):1401–1402. 105. G. V. A. Aben and L. H. M. Krings. EP 727094 (Philips Electronics N.V.), Prior: August 6, 1994. 106. S. Y. Yoon, US 6630196 (Orion Electric Co. Ltd), Prior: July 24, 1998. 107. S. Y. Yoon, C. H. Lee, and H. S. Son. KR 2000 009403/ US 6630196 (Orion Electric Co. Ltd), Prior: July 24, 1998. 108. S. Y. Yoon, C. H. Lee, and H. S. Son KR 2000 022633/ US 6391227 (Orion Electric Co. Ltd), Prior: September 22, 1998. 109. H. Tong and J. Hu. CN 1 171 617 A (Zhonghua Kinescope Co.), Prior: July 23, 1996. 110. Y. Morimoto. JP 2000 026817 A2 (Teijin Ltd.), Prior: July 14, 1998. 111. T. Kawashima and T. Sekiya. JP 2003 251756 (Sony Chemical Corp.), Prior: February 28, 2002. 112. K. Ishikawa and H. Yoshino. JP 2006169455 (JSR Ltd.), Prior: December 20, 2004. 113. K. S. Suh, J. E. Kim, T. Y. Kim, and Y. S. Kim. WO 2006/057465 (K. S. Suh), Prior: October 15, 2003. 114. H. Shinohara. JP 2003 246874 A2 (HS Planning Y. K.), Prior: February 26, 2002. 115. M. E. L.Wouters, D. P. Wolfs, M. C. van der Linde, and J. H. P. Hovens. 2004. Transparent UV curable antistatic hybrid coatings on polycarbonate prepared by the sol-gel method. Progress in Organic Coatings 51(4):312–320. 116. A. Thiess. DE 20021226 (A. Thiess), Prior: December 25, 2000. 117. G. Hardtke and H. Fuchs 1999. Electrically conductive polymers in textile coatings. TECHNOMER ’99, 16th Fachtagung über Verarbeitung und Anwendung von Polymeren, Volume 1, ed. G. Mennig, F. Meyer. Chemnitz: Technische Universität Chemnitz, Institut für Allgemeinen Maschinenbau und Kunststofftechnik (CA 137:80211). 118. R. C. Foitzik, A. Kaynak, T. Lin, L. Wang, and X. Wang. 2006. WO2006081622 (Australian Wool Innovation Ltd, University of Deakin), Prior: February 3, 2006. 119. M. Husemann and S. Zöllner. EP 1570017 (Tesa AG). Prior: December 4, 2002. 120. S. S. Chadha. 1993. Powder electroluminescence. In: Solid State Luminescence: Theory, Materials and Devices, ed. A. H. Kitai. New York: Chapman & Hall.
6911X.indb 249
10/1/10 9:01:12 PM
250
PEDOT
121. M. Hüppauff, R. Sybrichs, and A. Gehring. DE 19757874, (Bosch GmbH) Prior: December 12, 1997. 122. H. Andriessen. EP 1 231 251 A1 (Agfa-Gevaert), Prior: August 14, 2002. 123. A. Saito. JP 2002 124391 A2 (Seiko Precision Inc.), Prior: August 8, 2000. 124. E. Enz. WO 2003 037039 A1 (Lumitec AG), Prior: October 24, 2001. 125. U. Guntermann and F. Jonas. EP 1568750 (H.C. Starck GmbH), Prior: February 27, 2004. 126. M. Pope, H. P. Kallmann, and P. Magnante. 1963. Electroluminescence in organic crystals. J Chem Phys 38:2042–2043. 127. W. Helfrich and W. G. Schneider. 1965. Recombination radiation in anthracene crystals. Phys Rev Lett 14:229–231. 128. C. W. Tang and S. A. VanSlyke. 1987. Organic electroluminescent diodes. Appl Phys Lett 51: 913–915. 129. J. R. Sheats, H. Antoniadis, M. Hueschen, W. Leonard, J. Miller, R. Moon, D. Roitman, and A. Stocking. 1996. Organic electroluminescent devices. Science 273:884–888. 130. 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. 1999. Electroluminescence in conjugated polymers. Nature 397: 121–128. 131. Joseph Shinar, ed. 2004. Organic Light-Emitting Devices: A Survey. New York: Springer. 132. D. Hertel, C. D. Müller, and K. Meerholz. 2005. Chemie in unserer Zeit 39:336–347. 133. Market study published by DisplaySearch. 2009. http://www.displaysearch. com. 134. J. Shi and C. W. Tang. 1997. Doped organic electroluminescent devices with improved stability. Appl Phys Lett 70(13):1665–1667. 135. M. A. Baldo, D. F. O’Brian, Y. You, A. Shoustikov, S. Silbey, M. E. Thompson, and S. R. Forrest. 1998. Highly efficient phosphorescent emission from organic electroluminescent devices. Nature 395:151–154. 136. S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lüssem, and K. Leo. 2009. White organic light-emitting diodes with fluorescent tube efficiency. Nature 459:234–238. 137. R. Meerheim, K. Walzer, M. Pfeiffer, and K. Leo. 2006. Ultrastable and efficient red organic light emitting diodes with doped transport layers. Appl Phys Lett 89:061111-1–061111-3. 138. 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. 1990. Light-emitting diodes based on conjugated polymers. Nature 347:539–541. 139. C. Jonda, A. B. R. Mayer, U. Stolz, A. Elschner, and A. Karbach. 2000. Surface roughness effects and their influence on the degradation of organic light emitting devices. J Mater Sci 35:5645–5651. 140. Y. Yang and A. J. Heeger. 1994. Polyaniline as a transparent electrode for polymer light-emitting diodes: Lower operating voltage and higher efficiency. Appl Phys Lett 64(10):1245–1247. 141. S. Karg, J. C. Scott, J. R. Salem, and M. Angelopoulos. 1996. Increased brightness and lifetime of polymer light-emitting diodes with polyaniline anodes. Synth Met 80:111–117. 142. K. Pichler. 1997. Conjugated polymer electroluminescence: Technical aspects from basic devices to commercial products. Phil Trans R Soc Lond A 355:829–842.
6911X.indb 250
10/1/10 9:01:12 PM
Applications
251
143. A. Berntsen, Y. Croonen, C. Liedenbaum, H. Schoo, R.-J. Visser, J. Vleggaar, and P. v. d. Weijer. 1998. Stability of polymer LEDs. Op Mater 9:125–133. 144. A. Elschner, F. Bruder, H. W. Heuer, F. Jonas, A. Karbach, S. Kirchmeyer, and S. Thurm. 2000. PEDT/PSS for efficient hole-injection in hybrid organic lightemitting diodes. Synth Met 111–112:139–143. 145. J. M. Bharathan and Y. Yang. 1998. Polymer/metal interfaces and the performance of polymer light-emitting diodes. J Appl Phys 84(6):3207–3211. 146. K. Book, H. Bässler, A. Elschner, and S. Kirchmeyer. 2003. Hole injection from an ITO/PEDT anode into the hole transporting layer of an OLED probed by bias induced absorption. Org Electron 4(4):227–232. 147. J. S. Kim, M. Granström, R. H. Friend, N. Johansson, W. R. Salaneck, R. Daik, and W. J. Feast. 1998. Indium–tin oxide treatments for single- and double-layer polymeric light-emitting diodes: The relation between the anode physical, chemical, and morphological properties and the device performance. J Appl Phys 84(12):6859–6870. 148. J. S. Kim, R. H. Friend, and F. Cacialli. 1999. Improved operational stability of polyfluorene-based organic light-emitting diodes with plasma-treated indium–tin– oxide anodes. Appl Phys Lett 74(21):3084–3086. 149. H. Ishii, K. Sugiyama, E. Ito, and K. Seki. 1999. Energy level alignment and interfacial electronic structures at organic/metal and organic/organic interfaces. Adv Mater 11: 605–625. 150. N. Koch. 2007. Organic electronic devices and their functional interfaces. ChemPhysChem 8:1438–1455. 151. T. Kugler, W. R. Salaneck, H. Rost, and A. B. Holmes. 1999. Polymer band alignment at the interface with indium tin oxide: consequences for light emitting devices. Chem Phys Lett 310(5,6):391–396. 152. T. M. Brown, J. S. Kim, R. H. Friend, F. Cacialli, R. Daik, and W. J. Feast. 1999. Built-in field electroabsorption spectroscopy of polymer light-emitting diodes incorporating a doped poly(3,4-ethylene dioxythiophene) hole injection layer. Appl Phys Lett 75:1679–1681. 153. G. Greczynski, Th. Kugler, and W. R. Salaneck. 2000. Energy level alignment in organic-based three-layer structures studied by photoelectron spectroscopy. J Appl Phys 88:7187–7191. 154. N. Koch, A. Vollmer, and A. Elschner. 2007. The influence of water on the work function of conducting poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate). Appl Phys Lett 90:043512-1–043512-3. 155. J. Huang, P. F. Miller, J. S. Wilson, A. J. deMello, J. C. deMello, and D. D. C. Bradley. 2005. Investigation of the effects of doping and post-deposition treatments on the conductivity, morphology and work function of poly(3,4-ethylenedioxythiophene)/poly(styrene sulfonate) films. Adv Funct Mater 15:290–296. 156. N. Koch, A. Elschner, J. P. Rabe, and R. L. Johnson. 2005. Work function independent hole-injection barriers between pentacene and conducting polymers. Adv Mater 17:330–335. 157. A. J. Mäkinen, I. G. Hill, R. Shashidhar, N. Nikolov, and Z. H. Kafafi. 2001. Hole injection barriers at polymer anode/small molecule interfaces. Appl Phys Lett 79:557–559. 158. C. Tengstedt, W. Osikowicz, W. R. Salaneck, I. D. Parker, C. H. Hsu, and M. Fahlman. 2006. Fermi-level pinning at conjugated polymer interfaces. Appl Phys Lett 88: 053502-1–053502-3.
6911X.indb 251
10/1/10 9:01:12 PM
252
PEDOT
159. N. Koch, A. Elschner, and R. L. Johnson. 2006. Green polyfluorene-conducting polymer interfaces: Energy-level alignment and device performance. J Appl Phys 100:024512/1–024512/5. 160. N. Koch and A. Vollmer. 2006. Electrode-molecular semiconductor contacts. Work-function-dependent hole injection barriers versus Fermi-level pinning. Appl Phys Lett 89:162107-1–162107-3. 161. A. van Dijken, A. Perro, E. A. Meulenkamp, and K. Brunner. 2003. The influence of a PEDOT:PSS layer on the efficiency of a polymer light-emitting diode. Org Electron 4:131–141. 162. P. J. Brewer, P. A. Lane, A. J. deMello, D. D. C. Bradley, and J. C. deMello. 2004. Internal Field screening in Polymer Light-Emitting Diodes. Adv Funct Mater 14(6):562–570. 163. P. J. Brewer, P. A. Lane, J. Huang, A. J. deMello, D. D. C. Bradley, and J. C. deMello. 2005. Role of electron injection in polyfluorene-based light emitting diodes containing PEDOT:PSS. Phys Rev B 71:205209-1–205209-6. 164. A. J. A. B. Seeley, R. H. Friend, J. S. Kim, and J. H. Burroughes. 2004. Trapassisted hole injection and quantum efficiency enhancement in poly (9,9′ dioctylfluorene-alt-benzothiadiazole) polymer light-emitting diodes. J Appl Phys 96(12):7643–7649. 165. T. van Wondenbergh, J. Wildeman, P. W. M. Blom, J. J. A. M. Bastiaansen, and B. M. W. Langeveld-Voss. 2004. Electron-enhanced hole injection in blue polyfluorenebased polymer light-emitting diodes. Adv Funct Mater 14(7):677–683. 166. G. Greczynski, Th. Kugler, and W. R. Salaneck. 1999. Characterization of the PEDOT-PSS system by means of X-ray and ultraviolet photoelectron spectroscopy. Thin Solid Films 354(1,2):129–135. 167. J. Hwang, F. Amy, and A. Kahn. 2006. Spectroscopic study on sputtered PEDOT·PSS: Role of surface PSS layer. Org Electron 7: 387–396. 168. J. S. Kim, R. H. Friend, I. Grizzi, and J. H. Burroughes. 2005. Spin-cast thin semiconducting polymer interlayer for improving device efficiency of polymer lightemitting diodes. Appl Phys Lett 87:023506-1–023506-3. 169. S. A. Choulis, V. E. Choong, M. K. Mathai, and F. So. 2005. The effect of interfacial layer on the performance of organic light-emitting diodes. Appl Phys Lett 87:113503-1–113503-3. 170. M. P. de Jong, L. J. van IJzendoorn, and M. J. A. de Voigt. 2000. Stability of the interface between indium-tin-oxide and poly(3,4-ethylenedioxythiophene)/ poly(styrenesulfonate) in polymer light-emitting diodes. Appl Phys Lett 77(14):2255–2257. 171. K. W. Wong, H. L. Yip, Y. Luo, K. Y. Wong, W. M. Lau, K. H. Low, H. F. Chow, Z. Q. Gao, W. L. Yeung, and C. C. Chang. 2002. Blocking reactions between indiumtin oxide and poly (3,4-ethylene dioxythiophene):poly(styrene sulphonate) with a self-assembly monolayer. Appl Phys Lett 80(15):2788–2791. 172. T. W. Lee and Y. Chung. 2008. Control of the surface composition of a conducting-polymer complex film to tune the work function. Adv Funct Mat 18(15):2246–2252. 173. M. M. de Kok, M. Buechel, S. I. E. Vulto, P. van de Weijer, E. A. Meulenkamp, S. H. P. M. de Winter, A. J. G. Mank, H. J. M. Vorstenbosch, C. H. L. Weijtens, and V. van Elsbergen. 2004. Modification of PEDOT:PSS as hole injection layer in polymer LEDs, Phys Stat Sol (a) 201(6):1342–1359.
6911X.indb 252
10/1/10 9:01:12 PM
Applications
253
174. J. S. Kim, P. K. H. Ho, C. E. Murphy, A. J. A. B. Seeley, I. Grizzi, J. H. Burroughes, and R. H. Friend. 2004. Electrical degradation of triarylamine-based light-emitting polymer diodes monitored by micro-Raman spectroscopy. Chem Phys Lett 362:2–7. 175. J. Chung, K. H. Kim, J. C. Lee, M. K. Kim, and H. J. Shin. 2008. Spectromicroscopic investigation of polymer light-emitting device degradation. Org Electron 9(5):869–872. 176. J. Halls. 2007. Recent Advances in Polymer OLED-Technology, SID Seminar May 2007. 177. R. Q. Png, P. J. Chia, S. Sivaramakrishnan, L. Y. Wong, M. Zhou, L. L. Chua, and P. K. H. Ho. 2007. Electromigration of the conducting polymer in organic semiconductor devices and its stabilization by cross-linking. Appl Phys Lett 91(1):013511-1–013511-3. 178. S. Nowy, W. Ren, A. Elschner, W. Lövenich, and W. Brütting. 2010. Impedance spectroscopy as a probe for the degradation of organic light-emitting diodes: J Appl Phys 107:054501-1–054501-9. 179. P. K. H. Ho, J. S. Kim, J. H. Burroughes, H. Becker, S. F. Y. Li, T. M. Brown, F. Cacialli, and R. H. Friend. 2000. Molecular-scale interface engineering for polymer light-emitting diodes. Nature 404:481–484. 180. J. Y. Lee. 2004. Lifetime improvements of polymer light emitting diodes through a poly(3,4-ethylenedioxythiophene): Silane hole transport layer. Chem Phys Lett 393:260–263. 181. A. Elschner, F. Jonas, S. Kirchmeyer, W. Lövenich, N. Koch, K. Fehse, M. Pfeiffer, K. Walzer, and K. Leo. 2006. High-conductive PEDOT/PSS for ITO-substitution in OLEDs. IDW06, OLED3-4L. 182. S. Ghosh, J. Rasmusson, and O. Inganäs. 1998. Supramolecular self-assembly for enhanced conductivity in conjugated polymer blends: Ionic crosslinking in blends of poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) and poly (vinylpyrrolidone). Adv Mater 10(14):1097–1099. 183. S. M. Seo, J. H. Kim, J. Y. Park, and H. H. Lee. 2005. Coordination-complex polymer as an organic conductor for organic light-emitting diodes. Appl Phys Lett 87:183503/1–183503/3. 184. Y. Zhou, Y. Yuan, J. Lian, J. Zhang, H. Pang, L. Cao, and X. Zhou. 2006. Mild oxygen plasma treated PEDOT:PSS as anode buffer layer for vacuum deposited organic light-emitting diodes. Chem Phys Lett 427(4–6): 394–398. 185. F. Zhang, A. Petr, and L. Dunsch. 2003. Electrochemically doped polymeric anode for improving the performance of molecular organic light-emitting diodes. Appl Phys Lett 82(25):4587–4589. 186. P. G. Pickup, C. L. Kean, M. C. Lefebvre, G. Li, Z. Qi, and J. Shan. 2000. Electronically conducting cation-exchange polymer powders: synthesis, characterization and applications in PEM fuel cells and supercapacitors. J New Mat Electrochem Systems 3(1):21–26. 187. T. Ohnishi, WO 2003065770 (Sumitomo Chemical Company) Prior: January 31, 2002. 188. C. H. Hsu, Y. Cao, S. Kim, D. D. LeCloux, H. Li, C. D. Macpherson, C. Zhang, and H. Skulason, EP 1612971 (Lucent Technologies Inc), Prior: April 22, 2003. 189. C. H. Hsu, Y. Shen, E. M. Smith, D. D. LeCloux, H. Skulason, S. Kim, and I. D. Parker. 2006. A high work function hole-injection material for enhanced OLED performance, SID 06 Digest 5.4:49–51.
6911X.indb 253
10/1/10 9:01:12 PM
254
PEDOT
190. A. Elschner, F. Jonas, K. Reuter, and W. Lövenich. EP 1564251 (H.C. Starck GmbH), Prior: February 1, 2005. 191. T. W. Lee, Y. Chung, O. Kwon, and J. J. Park. 2007. Self-organized gradient hole injection to improve the performance of polymer electroluminescent devices. Adv Funct Mat 17(3):390–396. 192. T. W. Lee, O. Kwon, M. G. Kim, S. H. Park, J. Chung, S. Y. Kim, Y. Chung, J. Y. Park, E. Han, D. H. Huh, J. J. Park, and L. Pu. 2005. Hole-injecting conductingpolymer compositions for highly efficient and stable organic light-emitting diodes. Appl Phys Lett 87: 231106-1–231106-3. 193. B. Zhang, W. Li, J. Yang, Y. Fu, Z. Xie, and S. Zhang. 2009. Performance enhancement of polymer light-emitting diodes by using ultrathin fluorinated polyimide modifying the surface of poly(3,4-ethylene dioxythiophene):poly(styrenesulfo nate. J Phys Chem C 113 (18):7898–7903. 194. T. Y. Kim, M. Suh, S. J. Kwon, T. H. Lee, J. E. Kim, Y. J. Lee, J. H. Kim, M. P. Hong, and K. S. Suh. 2009. Poly(3,4-ethylenedioxythiophene) derived from poly(ionic liquid) for the use as hole-injecting material in organic light-emitting diodes. Macromol Rapid Commun 30(17):1477–1482. 195. C. J. Brabec, J. A. Hauch, P. Schilinsky, and C. Waldauf. 2005. Production aspects of organic photovoltaics and their impact on the commercialization of devices. MRS Bull 30:50–52. 196. F. C. Krebs. 2009. Fabrication and processing of polymer solar cells: A review on of printing and coating techniques. Sol Energy Mater Sol Cells 93(4):394–412. 197. P. Peumans, A. Yakimov, and S. R. Forrest. 2003. Small molecular weight organic thin-film photodetectors and solar cells. J Appl Phys 93(7):3693–3722. 198. C. Brabec, V. Dyakonov, and U. Scherf, ed. 2008. Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies. Weinheim: Wiley-VCH. 199. S. E. Shaheen, D. S. Ginley, and G. E. Jabbour. 2005. Organic-based photovoltaics: Toward low-cost power generation. MRS Bull 30:10–19. 200. K. Lee, J. Y. Kim, and A. J. Heeger. 2008. Titanium oxide films as multifunctional components in bulk heterojunction “plastic” solar cells. In: Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies, ed. C. Brabec, V. Dyakonov, and U. Scherf. Weinheim: Wiley-VCH. 201. B. O’Regan and M. Grätzel. 1991. A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films. Nature 335:737–740. 202. A. C. Arias, M. Granström, D. S. Thomas, K. Petritsch, and R. H. Friend. 1999. Doped conducting-polymer-semiconducting-polymer interfaces: Their use in organic photovoltaic devices. Phys Rev B 60(3):1854–1860. 203. T. Aernouts, P. Vanlaeke, W. Geens, J. Poortmans, P. Heremans, S. Borghs, R. Mertens, R. Andriessen, and L. Leenders. 2004. Printable anodes for flexible organic solar cell modules. Thin Solid Films 451–452:22–25. 204. K. Tvingstedt and O. Inganäs. 2007. Electrode grids for ITO-free organic photovoltaic devices. Adv Mater 19(19):2893–2897. 205. F. Zhang, M. Johansson, M. R. Andersson, J. C. Hummelen, and O. Inganäs. 2002. Polymer photovoltaic cells with conducting polymer anodes. Adv Mater 14(9):662–665. 206. Y.-S. Hsiao, W.-T. Whang, C.-P. Chen, and Y.-C. Chen. 2008. High-conductivity poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) film for use in ITOfree polymer solar cells. J Mater Chem 18(48):5948–5955.
6911X.indb 254
10/1/10 9:01:12 PM
Applications
255
207. M. Glatthaar, M. Niggemann, B. Zimmermann, P. Lewer, M. Riede, A. Hinsch, and J. Luther. 2005. Organic solar cells using inverted sequence. Thin Solid Films 491(1–2):298–300. 208. F. Nickel, A. Puetz, M. Reinhard, H. Do, C. Kayser, A. Colsmann, and U. Lemmer. 2010. Cathodes comprising highly conductive poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) for semi-transparent polymer solar cells. Org Electronics doi:10.1016/j.orgel.2009.12.009. 209. S.-I. Na, S.-S. Kim, J. Jo, and D. Y. Kim. 2008. Efficient and flexible ITO-free organic solar cells using highly conductive polymer anodes. Adv Mater 20(21):4061–4067. 210. E. Ahlswede, W. Mühleisen, M. W. bin Moh Wahi, J. Hanisch, and M. Powalla. 2008. Highly efficient organic solar cells with printable low-cost transparent contacts. Appl Phys Lett 92(14):143307-1–143307-3. 211. Y. F. Lim, S. Lee, D. J. Herman, M. T. Lloyd, J. E. Anthony, and G. G. Malliaras. 2008. Spray-deposited poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) top electrode for solar cells. Appl Phys Lett 93(19):193301-1–193301-3. 212. T. Aernouts, W. Geens, J. Poortmans, P. Heremans, S. Borghs, and R. Mertens. 2002. Extraction of bulk and contact components of the series resistance in organic bulk donor-acceptor-heterojunctions. Thin Solid Films 403:297–301. 213. F. L. Zhang, A. Gadisa, O. Inganäs, M. Svensson, and M. R. Andersson. 2004. Influence of buffer layers on the performance of polymer solar cells. Appl Phys Lett 84:3906–3908. 214. I. Yoo, M. Lee, C. Lee, D. W. Kim, I. S. Moon, and D. H. Hwang. 2005. The effect of a buffer layer on the photovoltaic properties of solar cells with P3OT:fullerene composites. Synth Met 153:97–100. 215. M. Y. Song, K. J. Kim, and D. Y. Kim. 2005. Enhancement of photovoltaic characteristics using a PEDOT interlayer in TiO2/MEHPPV heterojunction devices. Sol Energy Mater Sol Cells 85:31–39. 216. Y. Saito, T. Azechi, T. Kitamura, Y. Hasegawa, Y. Wada, and S. Yanagida. 2004. Photo-sensitizing ruthenium complexes for solid state dye solar cells in combination with conducting polymers as hole conductors. Coordination Chemistry Reviews 248:1469–1478. 217. N. Fukuri, Y. Saito, W. Kubo, G. K. R. Senadeera, T. Kitamura, Y. Wada, and S. Yanagida. 2004. Performance improvement of solid-state dye-sensitized solar cells fabricated using poly(3,4-ethylenedioxythiophene) and amphiphilic sensitizing dye. J Electrochem Soc 151(10):A1745–A1748. 218. J.-Y. Liao and K.-C. Ho. 2005. A photovoltaic cell incorporating a dye-sensitized ZnS/ZnO composite thin film and a hole-injecting PEDOT layer. Sol Energy Mater Sol Cells 86:229–241. 219. E. L. Williams, Q. Wang, S. E. Shaheen, D. S. Ginley, and G. E. Jabbour. 2004. Conducting polymer and thin-film silicon hybrid solar cells. Polym Mat Sci Eng 91:802–804. 220. W. U. Huynh, J. J. Dittmer, and A. P. Alivisatos. 2002. Hybrid nanorod-polymer solar cells. Science 295:2425–2427. 221. P. Peumans and S. R. Forrest. 2001.Very-high-efficiency double-heterostructure copper phthalocyanine/C60 photovoltaic cells. Appl Phys Lett 79:126–128; Erratum, Appl Phys Lett 2002, 80:338. 222. I. D. Parker. 1994. Carrier tunneling and device characteristics in polymer light emitting diodes. J Appl Phys 75(3):1656–1666.
6911X.indb 255
10/1/10 9:01:13 PM
256
PEDOT
223. G. G. Malliaras, J. R. Salem, P. J. Brock, and J. C. Scott. 1998. Photovoltaic measurement of the built-in potential in organic light emitting diodes and photodiodes. J Appl Phys 84(3):1583. 224. K. Kawano, N. Ito, T. Nishimori, and J. Sakai. 2006. Open circuit voltage of stacked bulk heterojunction solar cells. Appl Phys Lett 88:073514-1–073514-3. 225. J. K. J. v. Duren, J. Loos, F. Morrissey, C. M. Leewis, K. P. H. Kivits, L. J. v. IJzendoorn, M. T. Rispens, J. C. Hummelen, and R. A. J. Janssen. 2002. In-situ compositional and structural analysis of plastic solar cells. Adv Funct Mater 12(10):665–669. 226. K. Kawano, R. Pacios, D. Poplavskyy, J. Nelson, D. D. C. Bradley, and J. R. Durrant. 2006. Degradation of organic solar cells due to air exposure. Sol Energy Mater Sol Cells 90:520–3530. 227. K. Kawano and C. Adachi. 2009. Evaluating carrier accumulation in degraded bulk heterojunction organic soalr cells by a thermally stimulated current technique. Adv Funct Mater 19(24):3934–3940. 228. M. Ramuz, L. Bürgi, C. Winnewisser, and P. Seitz. 2008. High sensitivity organic photodiodes with low dark currents and increased lifetimes. Org Electronics 9(3):369–376. 229. J. v. d. Lagemaat, T. M. Barnes, G. Rumbles, S. E. Shaheen, T. J. Coutts, C. Weeks, I. Levitsky, J. Peltola, and P. Glatkowski. 2006. Organic solar cells with carbon nanotubes replacing In2O3:Sn as the transparent electrode. Appl Phys Lett 88:233503-1–233503-3. 230. M. W. Rowell, M. A. Topinka, M. D. McGehee, H.-J. Prall, G. Dennler, N. S. Sariciftci, L. Hu, and G. Gruner. 2006. Organic solar cells with carbon nanotube network electrodes. Appl Phys Lett 88:233506/1–233506/3. 231. A. Reale, T. M. Brown, A. Di Carlo, F. Giannini, F. Brunetti, E. Leonardi, M. Lucci, M. L. Terranova, S. Orlanducci, E. Tamburri, F. Toschi, and V. Sessa. 2006. Nanocomposites for organic and hybrid organic-inorganic solar cells. Proceedings of SPIE 6334:63340Y-1–63340Y-8. 232. E. L. Williams, G. E. Jabbour, Q. Wang, S. E. Shaheen, D. S. Ginley, and E. A. Schiff. 2005. Conducting polymer and hydrogenated amorphous silicon hybrid solar cells. Appl Phys Lett 87:223504-1–223504-3. 233. H. Frohne, S. E. Shaheen, C. J. Brabec, D. C. Müller, N. S. Sariciftci, and K. Meerholz. 2002. Influence of the anodic work function on the performance of organic solar cells. ChemPhysChem 3:795–799. 234. M. Grätzel. 2005. Dye-sensitized solid-state heterojunction solar cells. MRS Bull 30:23-27. 235. Y. Saito, T. Kitamura, Y. Wada, and S. Yanagida, 2002. Poly(3,4-ethylenedio xythiophene) as a hole conductor in solid state dye sensitized solar cells. Synth Met 131(1–3):185–187. 236. J. Xia and S. Yanagida. 2009. Strategy to improve the performance of dye-sensitized solar cells: Interface engineering principle. Solar Energy, doi:10.1016/j. solener.2009.10.005. 237. Y. Kim, Y.-E. Sung, J.-B. Xia, M. Lira-Cantu, N. Masaki, and S. Yanida. 2008. Solid-state dye-sensitized TiO2 solar cells using poly(3,4-ethylenedioxythiophene) as substitute of iodine/iodide electrolytes and noble metal catalysts on FTO counter electrodes. J Photochem Photobiol A 193:77–80.
6911X.indb 256
10/1/10 9:01:13 PM
Applications
257
238. E. M. J. Johansson, A. Sandell, H. Siegbahn, H. Rensmo, B. Mahrov, G. Boschloo, E. Figgemeier, A. Hagfeldt, S. K. M. Jönsson, and M. Fahlman. 2005. Interfacial properties of photovoltaic TiO2/dye/PEDOT-PSS heterojunctions. Synth Met 149(2–3):157–167. 239. J.-G. Chen, H.-Y. Wie, and K.-C. Ho. 2007. Using modified poly(3,4-ethylenedioxy thiophene):Poly(styrene sulfonate) film as a counter electrode in dye-sensitized solar cells. Sol Energy Mater Sol Cells 91:1472–1477. 240. M. Biancardo, K. West, and F. C. Krebs. 2007. Quasi-solid-state dye-sensitized solar cells: Pt and PEDOT:PSS counter electrodes applied to gel electrolyte assemblies. J Photochem Photobiol A 187:395–401. 241. J. R. Platt. 1961. Electrochromism, a possible change of color producible in dyes by an electric field. J Chem Phys 34:862–864. 242. P. M. S. Monk, R. J. Mortimer, and D. R. Rosseinsky. 1995. Electrochromism, Fundamentals and Applications. Weinheim: VCH Verlagsgesellschaft mbH. 243. R. J. Mortimer, A. L. Dyer, and J. R. Reynolds. 2006. Electrochromic organic and polymeric materials for display applications. Displays 27:2. 244. L. Michaelis and E. S. Hill. 1933. On the absorbancy of reduced methyl viologen. J Gen Physiol 16:859–873. 245. C. L. Bird and A. T. Kühn. 1981. Electrochemistry of the viologens. Chem Soc Rev 10:49–82. 246. H. Berneth, F. Jonas, S. Kirchmeyer, U. Merker, A. Michaelis, and R. Neigl. DE 10148438 (Bayer AG), Prior: October 1, 2001. 247. A. L. Dyer and J. R. Reynolds. 2007. Electrochromism and conjugated conducting polymers. In: Handbook of conducting polymer, ed. T. A. Skotheim and J. R. Reynolds, 20-1–20-63. Boca Raton, FL: CRC Press. 248. S. K. Deb. 1969. A novel electrophotographic system. Appl Optics, Suppl 3:192–195. 249. C. G. Granqvist. 1993. Electrochromic tungsten-oxide-based thin films: Physics, chemistry and technology. In: Physics of Thin Films, Vol. 17, ed. M. H. Francombe and J. L. Vossen. 301–370. San Diego: Academic. 250. M. Green. 1996. The promise of electrochromic systems. Chem Ind 641–644. 251. R. A. Batchelor, M. S. Burdis, and J. R. Siddle. 1996. Electrochromism in sputtered WO thin films. J Electrochem Soc 143(3):1050–1055. 252. T. Kobayashi, H. Yoneyama, and T. Tamura. 1984. Polyaniline film-coated electrodes as electrochromic display devices. J Electroanal Chem 161(22):419–423. 253. T. Kobayashi, H. Yoneyama, and T. Tamura. 1984. Electrochemical reactions concerned with electrochromism of polyaniline film-coated electrodes. J Electroanal Chem 177(1–2):281–291. 254. J. E. Dubois, F. Garnier, G. Tourillon, and M. Gazard. 1983. Organic conducting polymers derived from substituted thiophenes as electrochromic material. J Electroanal Chem 148(2):299–303. 255. E. W. Tsai, S. Basok, J. P. Ruiz, J. R. Reynolds, and K. Rajeshwar. 1989. Electrochemistry of some β-substituted polythiophenes. J Electrochem Soc 136(12):3683–3689. 256. Q. Pei, G. Zuccarello, M. Ahlskog, and O. Inganäs. 1994. Electrochromic and highly stable poly(3,4-ethylenedioxythiophene) switches between opaque blueblack and transparent sky blue. Polymer 35(7):1347–1951.
6911X.indb 257
10/1/10 9:01:13 PM
258
PEDOT
257. J. C. Gustafsson, B. Liedberg, and O. Inganäs. 1994. In situ spectroscopic investigations of electrochromism and ion transport in a poly(3,4-ethylenedioxythiophene) electrode in a solid state electrochemical cell. Solid State Ionics 69(2):145–152. 258. H. W. Heuer, R. Wehrmann, and S. Kirchmeyer. 2002. Electrochromic window based on conducting poly(3,4-ethylenedioxythiophene)-poly(styrene sulfonate). Adv Funct Mater 12(2):89–94. 259. T. Johansson, L. A. Petterson, and O. Inganäs. 2002. Conductivity of de-doped poly(3,4-ethylenedioxythiophene). Synth Met 129(3):269–274. 260. P. M. Beaujuge and J. R. Reynolds. 2010. Color control in π-conjugated organic polymers for use in electrochromic devices. Chem Rev 110(1):268–320. 261. C. Ziegler. 1997. Thin film properties of oligothiophenes. In: Handbook of Organic Conductive Molecules and Polymers, Vol 3., ed. H. S. Nalva, 678–737. New York: J. Wiley & Sons. 262. F. Tran-Van, S. Garreau, G. Louarn, G. Froyer, and C. Chevrot. 2001. A fully undoped oligo(3,4-ethylenedioxythiophene): Spectroscopic properties. Synth Met 119(1–3):381. 263. F. Tran-Van, S. Garreau, G. Louarn, G. Froyer, and C. Chevrot. 2001. Fully undoped and soluble oligo(3,4-ethylenedioxythiophene)s: Spectroscopic study and electrochemical characterization. J Mater Chem 11(5):1378–1382. 264. I. D. Brotherston, D. S. K. Mudigonda, J. M. Osborn, J. Belk, J. Chen, D. C. Loveday, J. L. Boehme, J. P. Ferraris, and D. L. Meeker. 1999. Tailoring the electrochromic properties of devices via polymer blends, copolymers, laminates and patterns. Electrochim Acta 44(18):2993–3004. 265. P. M. Beaujuge, S. Ellinger, and J. R. Reynolds. 2008. Spray processable green to highly transmissive electrochromics via chemically polymerizable donoracceptor heterocyclic pentamers. Adv Mater 20(14):2772–2776. 266. G. Sonmez, H. B. Sonmez, C. K. F. Shen, and F. Wudl, 2004. Red, green, and blue colors in polymeric electrochromics. Adv Mater 16(21):1905–1908. 267. A. Durmus, G. E. Gunbas, P. Camurlu, and L.K. Toppare. 2007. A neutral state green polymer with a superior transmissive light blue oxidized state. Chem Commun (31):3246–3248. 268. D. M. Welsh, L. J. Kloeppner, L. Madrigal, M. R. Pinto, B. C. Thompson, K. S. Schanze, K. A. Abboud, D. Powell, and J. R. Reynolds. 2002. Regiosymmetric dibutyl-substituted poly(3,4-propylenedioxythiophene)s as highly electron-rich electroactive and luminescent polymers. Macromolecules 35(17):6517–6525. 269. A. Cirpan, A. A. Argun, C. R. G. Grenier, B. D. Reeves, and J. R. Reynolds. 2003. Electrochromic devices based on soluble and processable dioxythiophene polymers. J Mater Chem 13(10):2422–2428. 270. P. Schottland, K. Zong, C. L. Gaupp, B. C. Thompson, C. A. Thomas, I. Giurgiu, R. Hickman, K. A. Abboud, and J. R. Reynolds. 2000. Poly(3,4alkylenedioxypyrrole)s: Highly stable electronically conducting and electrochromic polymers. Macromolecules 33(1):7051–7061. 271. L. Groenendaal, G. Zotti, P.-H. Aubert, S. M. Waybright, and J. R. Reynolds. 2003. Electrochemistry of poly(3,4-alkylenedioxythiophene) derivatives. Adv Mater 15(11):855–879. 272. A. Kumar, D. M. Welsh, M. C. Morvant, F. Piroux, K. A. Abboud, and J. R. Reynolds. 1998. Conducting poly(3,4-alkylenedioxythiophene) derivatives as fast electrochromics with high-contrast ratios. Chem Mater 10(3):896–902.
6911X.indb 258
10/1/10 9:01:13 PM
Applications
259
273. A. Elschner and S. Kirchmeyer. 2008. PEDOT-type materials in organic solar cells. In: Organic Photovoltaics; Materials, Device Physics, and Manufacturing Technologies, ed. C. Brabec, U. Scherf, and V. Dyakonov, 213–242. Weinheim: Wiley-VCH. 274. B. Sankaran and J. R. Reynolds. 1997. High-contrast electrochromic polymers from alkyl-derivatized poly(3,4-ethylenedioxythiophenes). Macromolecules 30(9): 2582–2588. 275. A. Kumar and J. R. Reynolds. 1996. Soluble alkyl-substituted poly(ethyl enedioxythiophenes) as electrochromic materials. Macromolecules 29 (23): 7629–7630. 276. S. A. Sapp, G. A. Sotzing, and J. R. Reynolds. 1998. High contrast ratio and fastswitching dual polymer electrochromic devices. Chem Mater 10(8):2101–2108. 277. J.-D. Stenger-Smith, J. R. Reynolds, B. Sankaran, L. H. Warren, H. O. Marcy, and G. Bäuerle. 1995. Solid-state electroactive devices based on redox switching polymers. Proceedings of SPIE-The International Society for Optical Engineering (Optical and Photonic Applications of Electroactive and Conducting Polymers) 2528:162–168. 278. C. A. Cutler, M. Bouguettaya, and J. R. Reynolds. 2002. Electrostatic adsorption of PEDOT based electrochromic films. Polym Mater Sci Eng 86:256–257. 279. M. Valle, M. Bouguettaya, C. A. Cutler, and J. R. Reynolds. 2002. Sulfonatoalkoxy derivatized PEDOT via chemical polymerization. Polym Mater Sci Eng 86:187–188. 280. C. A Cutler, M. Bouguettaya, and J. R. Reynolds. 2002. PEDOT polyelectrolyte based electrochromic films via electrostatic adsorption. Adv Mater 14(9):684–688. 281. I. Schwendeman, C. L. Gaupp, C. J. M. Hancock, L. Groenendaal, and J. R. Reynolds. 2003. Perfluoroalkanoate-substituted PEDOT for electrochromic device applications. Adv Funct Mater 13(7):541–547. 282. K. Krishnamoorthy, M. Kanungo, A. Q. Contractor, and A. Kumar. 2001. Electrochromic polymer based on a rigid cyanobiphenyl substituted 3,4ethylenedioxythiophene. Synth Met 124(2–3):471–475. 283. D. M. Welsh, A. Kumar, E. W. Meijer, and J. R. Reynolds. 1999. Enhanced contrast ratios and rapid switching in electrochromics based on poly(3,4-propylenedioxythiophene) derivatives. Adv Mater 11(16):1379–1382. 284. V. Seshadri, J. Filloramo, W. K. Mino, and G. A. Sotzing. 2005. Electrochromic windows using a low band gap conjugated polymer as the ion storage layer. Polym Prepr 46(2):878–879. 285. S. P. Mishra, R. Sahoo, K. Krishnamoorthy, A.V. Ambade, A. Q. Contractor, and A. Kumar. 2004. Synthesis and characterization of electrochromic polymers based on 3,4-propylenedioxythiophenes. MACRO 2004, International Conference on Polymers for Advanced Technologies, Thiruvananthapuram, India, Dec. 15–17, (2004). Thiruvananthapuram: Society for Polymer Science. 286. B. D. Reeves, B. C. Thompson, K. A. Abboud, B. E. Smart, and J. R. Reynolds. 2002. Poly(SpiroBiProDOT): A new crosslinkable polymer network exhibiting multi-color and high contrast electrochromism. Polym Mater Sc Eng 86:204. 287. R. M. Walczak, J. S. Cowart, Jr., K. A. Abboud, and J. R. Reynolds. 2006. Conformational locking for band gap control in 3,4-propylenedioxythiophene based electrochromic polymers. Chem Commun (15):1604–1606. 288. R. M. Walczak, J. S. Cowart, and J. R. Reynolds. 2007. Tethered PProDOTs: Conformationally restricted 3,4-propylenedioxythiophene based electroactive polymers. J Mater Chem 17(3):254–260.
6911X.indb 259
10/1/10 9:01:13 PM
260
PEDOT
289. B. D. Reeves, C. R. G. Grenier, A. Argun, A. Cirpan, T. D. McCarley, and J. R. Reynolds. 2004. Spray coatable electrochromic dioxythiophene polymers with high coloration efficiencies. Macromolecules 37(20):7559–7569. 290. E. M. Galand, J. K. Mwaura, A. A. Argun, K. A. Abboud, T. D. McCarley, and J. R. Reynolds. 2006. Spray processable hybrid 3,4-propylenedioxythiophene: phenylene electrochromic polymers. Macromolecules 39(21):7286–7294. 291. Y.-G. Kim, B. C. Thompson, N. Ananthakrishnan, G. Padmanaban, S. Ramakrishnan, J. R. Reynolds. 2005. Variable band gap conjugated polymers for optoelectronic and redox applications. J Mater Res 20(12):3188–3198. 292. D. J. Irvin and J. R. Reynolds. 1998. Tuning the band gap of easily oxidized bis(2thienyl)- and bis(2-(3,4-ethylenedioxythiophene))phenylene polymers. Polym Adv Technol 9(4):260–265. 293. D. J. Irvin, C. J. DuBois, Jr., and J. R. Reynolds. 1999. Dual p- and n-type doping in an acid sensitive alternating bi(ethylenedioxythiophene) and pyridine polymer. Chem Commun (20):2121–2122. 294. S. A. Sapp, G. A. Sotzing, J. L. Reddinger, and J. R. Reynolds. 1996. Rapid switching solid-state electrochromic devices based on complementary conducting polymer films. Adv Mater 8(10):808–811. 295. G. A. Sotzing, J. L. Reddinger, A. R. Katritzky, J. Soloducho, R. Musgrave, J. R. Reynolds, and P. J. Steel. 1997. Multiply colored electrochromic carbazole-based polymers. Chem Mater 9(7):1578–1587. 296. G. A. Sotzing, C. A. Thomas, J. R. Reynolds, and P. J. Steel. 1998. Low band gap cyanovinylene polymers based on ethylenedioxythiophene. Macromolecules 31(11):3750–3752. 297. F. C. Cebeci, E. Sezer, and A. S. Sarac. 2006-2007. Synthesis and electrochemical characterization of bis(3,4-ethylene-dioxythiophene)-(4,4’-dinonyl-2,2’bithiazole) comonomer. Electrochim Acta 52(5):2158–2165. 298. P.-H. Aubert, M. Knipper, L. Groenendaal, L. Lutsen, J. Manca, and D. Vanderzande. 2004. Copolymers of 3,4-ethylenedioxythiophene and of pyridine alternated with fluorene or phenylene units: Synthesis, optical properties. Macromolecules 37(11):4087–4098. 299. F. Wang, M. S. Wilson, R. D. Rauh, P. Schottland, B. C. Thompson, and J. R. Reynolds. 2000. Electrochromic linear and star branched poly(3,4-ethylenedioxythiophene-didodecyloxybenzene) polymers. Macromolecules 33(6):2083–2091. 300. S. P. Mishra, K. Krishnamoorthy, R. Sahoo, and A. Kumar. 2006. Organicinorganic hybrid polymers containing 3,4-ethylenedioxythiophene and chalcogens in the main chain. J Mater Chem 16(32):3297–3304. 301. R. Pacios, R. Marcilla, C. Pozo-Gonzalo, J. A. Pomposo, H. Grande, J. Aizpurua, and D. Mecerreyes. 2007. Combined electrochromic and plasmonic optical responses in conducting polymer/metal nanoparticle films. J Nanosci Nanotechnol 7(8):2938–2941. 302. M. A. G. Namboothiry, T. Zimmerman, F. M. Coldren, J. Liu, K. Kim, and D. L. Carroll. 2007. Electrochromic properties of conducting polymer metal nanoparticles composites. Synth Met 157(13-15): 580-584. 303. H. C. Ko, M. Kang, B. Moon, H. Lee. 2004. Enhancement of electrochromic contrast of poly(3,4-ethylenedioxythiophene) by incorporating a pendant viologen. Adv Mater 16(19):1712–1716. 304. D. M. DeLongchamp, M. Kastantin, and P. T. Hammond. 2003. High-contrast electrochromism from layer-by-layer polymer films. Chem Mater 15(8):1575–1586.
6911X.indb 260
10/1/10 9:01:13 PM
Applications
261
305. J. L. Boehme, D. S. K. Mudigonda, and J. P. Ferraris. 2001. Electrochromic properties of laminate devices fabricated from polyaniline, poly(ethylenedioxythiophene), and poly(N-methylpyrrole). Chem Mater 13(12):4469–4472. 306. O. Inganäs, T. Johansson, and S. Ghosh. 2001. Phase engineering for enhanced electrochromism in conjugated polymers. Electrochim Acta 46(13–14):2031–2034. 307. G. Decher, J. D. Hong, and J. Schmitt. 1992. Buildup of ultrathin multilayer films by a self-assembly process: III. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on charged surfaces. Thin Solid Films 210–211(1–2):831–835. 308. G. Decher, Y. Lvov, and J. Schmitt. 1994. Proof of multilayer structural organization in self-assembled polycation-polyanion molecular films. Thin Solid Films 244(1–2):772–777. 309. G. Decher. 1997. Fuzzy nanoassemblies: Toward layered polymeric multicomposites. Science 277(5330):1232–1237. 310. D. DeLongchamp and P. T. Hammond. 2001. Layer-by-layer assembly of PEDOT/polyaniline electrochromic devices. Adv Mater 13(19):1455–1459. 311. H.-W. Heuer and R. Wehrmann. EP 1129150 (Bayer AG), Prior: October 6, 1998. 312. C. Li and T. Imae. 2004. Electrochemical and optical properties of the poly(3,4ethylenedioxythiophene) film electropolymerized in an aqueous sodium dodecyl sulfate and lithium tetrafluoroborate medium. Macromolecules 37(7):2411–2416. 313. W. H. Meyer. 1998. Polymer electrolytes for lithium-ion batteries. Adv Mater 10(6):439–448. 314. A. Reiche, T. Steurich, B. Sandner, P. Lobitz, and G. Fleischer. 1995. Ion transport in gel electrolytes. Electrochim Acta 40(13–14):2153–2157. 315. C. Carlberg, X. Chen, and O. Inganäs. 1996. Ionic transport and electronic structure in poly(3,4-ethylenedioxythiophene). Solid State Ionics 85(1–4):73–78. 316. Q. X. Zhou, C. J. Kolaskie, and L. L. Miller. 1987. The incorporation of electrolyte cations into polypyrrole and poly-3-methylthiophene during electrochemical reduction. J Electroanal Chem 223(1–2):283–286. 317. I. C. Faria, M. Kleinke, A. Gorenstein, M. C. A. Fantini, and M. H. Tabecniks. 1998. Toward efficient electrochromic NiOx films. J Electrochem Soc 145(1):235–240. 318. M. Döbbelin, R. Marcilla, M. Salsamendi, C. Pozo-Gonzalo, P. M. Carrasco, J. A. Pomposo, and D. Mecerreyes. 2007. Influence of ionic liquids on the electrical conductivity and morphology of PEDOT:PSS films. Chem Mater 19(9):2147–2149. 319. H. W. Heuer, R. Wehrmann, F. Jonas, and F. Osenberg. EP 942061 (Bayer AG), Prior: March 13, 1998. 320. H. W. Heuer and R. Wehrmann. EP 961158 (Bayer AG), Prior: May 25, 1998. 321. H. W. Heuer and R. Wehrmann. EP 961157 (Bayer AG), Prior: May 29, 1998. 322. T. Tung and K. Ho. 2003. A complementary electrochromic device containing poly(3,4-ethylenedioxythiophene) and Prussian blue. Proceedings Electrochemical Society 2003-17(Electrochromic Materials and Applications):254–265. 323. A. A. Argun, P. H. Aubert, J. R. Reynolds, P. Schottland, I. Schwendeman, G. Sonmez, and K. Zong. WO 2003/046106 (University of Florida), November 21, 2002. 324. J. D. Zhang, H. A. Yu, X. G. Wu, and Z. Y. Wang. 2004. Towards the improvement of attenuation range and response time of electrochromic polymer-based variable optical attenuators. Opt Mater 27(2):265–268. 325. M.-A. De Paoli, G. Casalbore-Miceli, E. M. Girotto, and W. A. Gazotti. 1999. All polymeric solid state electrochromic devices. Electrochim Acta 44(18):2983–2991.
6911X.indb 261
10/1/10 9:01:13 PM
262
PEDOT
326. J. R. Reynolds, A. Kumar, J. L. Reddinger, B. Sankaran, S. A. Sapp, and G. A. Sotzing. 1997. Unique variable-gap polyheterocycles for high-contrast dual polymer electrochromic devices. Synth Met 85(1–3):1295–1298. 327. E. M. Girotto and M. A. De Paoli. 1999. Electrochemistry, polymers and optoelectronic devices: A combination with a future. J Braz Chem Soc 10:394–400. 328. S.A. Sapp, G. A. Sotzing, J. L. Reddinger, and J. R. Reynolds. 1996. Rapid switching solid-state electrochromic devices based on complementary conducting polymer films. Adv Mater 8(10):808–811. 329. I. Schwendeman, R. Hickman, G. Sönmez, P. Schottland, K. Zong, D. M. Welsh, and J. R. Reynolds. 2002. Enhanced contrast dual polymer electrochromic devices. Chem. Mater. 14(7):3118–3122. 330. T.-H. Lin and K. C. Ho. 2006. A complementary electrochromic device based on polyaniline and poly(3,4-ethylenedioxythiophene). Solar Energy Mater Solar Cells 90(4):506–520. 331. L.-M. Huang, C.-H. Chen, and T.-C. Wen. 2006. Development and characterization of flexible electrochromic devices based on polyaniline and poly(3,4ethylenedioxythiophene)-poly(styrene sulfonic acid). Electrochim Acta 51(26): 5858–5863. 332. A. A. Argun, A. Cirpan, and J. R. Reynolds. 2003. The first truly all-polymer electrochromic devices. Adv Mater 15(16):1338–1341. 333. X. Li, G.-L. Zhao, J. Qian, and Z.-Y. Fu 2009. Preparation, structure and electrochromic properties of poly(3,4-ethylenedioxythiophene) based conducting textile. Gaodeng Xuexiao Huaxue Xuebao 30(5):1052–1054 (CA 2009:681783). 334. P. Andersson, D. Nilsson, P.-O. Svensson, M. Chen, A. Malmstrom, and M. Berggren. 2002. Active matrix displays based on all-organic electrochemical smart pixels printed on paper. Adv Mater 14(20):1460–1464. 335. D. Nilsson, M. Chen, T. Kugler, T. Remonen, M. Armgarth, T. Remonen, M. Armgarth, and M. Berggren. 2002. Bi-stable and dynamic current modulation in electrochemical organic transistors. Adv Mater 14(1):51–54. 336. S. Admassie and O. Inganäs. 2004. Electrochromism in diffractive conducting polymer gratings. J Electrochem Soc 151(6):H153–H157. 337. A. A. Argun and J. R. Reynolds. 2005. Line patterning for flexible and laterally configured electrochromic devices. J Mater Chem 15(18):1793–1800. 338. D. D. Dimitrakopoulos, D. J. Mascaro. 2001. Organic thin-film transistors: A review of recent advances. IBM J Res & Dev 45(1):11–27. 339. Z. Bao and J. Locklin. 2007. Organic Field-Effect Transistors. Boca Raton, FL: CRC Press. 340. R. A. Street. 2009. Thin-film transistors. Adv Mater 21(20):2007–2022. 341. Y. Y. Lin, D. J. Gundlach, S. Nelson, and T. N. Jackson. 1997. Stacked pentacene layer organic thin-film transistors with improved characteristics. IEEE Electron Device Lett 18:606–608. 342. M. Halik, H. Klauk, U. Zschieschang, G. Schmid, S. Ponomarenko, S. Kirchmeyer, and W. Weber. 2003. Relationship between molecular structure and electrical performance of oligothiophene organic thin film transistors. Adv Mater 15(11):917–922. 343. Z. Bao, A. Dodabalapur, and A. J. Lovinger. 1996. Soluble and processable regioregular poly(3-hexylthiophene) for thin film field-effect transistor applications with high mobility. Appl Phys Lett 69:4108–4110.
6911X.indb 262
10/1/10 9:01:13 PM
Applications
263
344. J. E. Anthony, D. L. Eaton, and S. R. Parkin. 2002. A road map to stable, soluble, easily crystallized pentacene derivatives. Org Lett 4:15–18. 345. M. J. Panzer and C. D. Frisbie. 2007. Contact effects in organic field-effect transistors. In: Organic Field-Effect Transistors, ed. Z. Bao and J. Locklin. Boca Raton, FL: CRC Press. 346. E. Becker, R. Parashkov, G. Ginev, D. Schneider, S. Hartmann, F. Brunetti, T. Dobbertin, D. Metzdorf, T. Riedl, H.-H. Johannes, and W. Kowalsky. 2003. Allorganic thin-film transistors patterned by means of selective electropolymerization. Appl Phys Lett 83(19):4044–4046. 347. H. Sirringhaus, T. Kawase, R. H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, and W. P. Woo. 2000. High-resolution inkjet printing of all-transistor circuits. Science 290:2123–2126. 348. Y. Liu, K. Varahramyan, and T. Cui. 2005. Low-voltage all-polymer field-effect transistor fabricated using an inkjet printing technique. Macromol Rapid Commun 26:1955–1959. 349. S. P. Li, C. J. Newsome, T. Kugler, M. Ishida, and S. Inoue. 2007. Polymer thin film transistors with self-aligned gates fabricated using ink-jet printing. Appl Phys Lett 90:172103-1–172103-3. 350. J. A. Lim, J. H. Cho, Y. D. Park, D. H. Kim, M. Hwang, and K. Cho. 2006. Solvent effect of inkjet printed source/drain electrodes on electrical properties of polymer thin-film transistors. Appl Phys Lett 88:082102-1–88:082102-3. 351. F. J. Touwslager, N. P. Willard, and D. M de Leeuw. 2002. I-Line lithography of poly(ethylenedioxythiophene) electrodes and application in all-polymer integrated circuits. Appl Phys Lett 81:4556–4558. 352. H. Rost, J. Ficker, J. S. Alonso, L. Leenders, and I. McCulloch. 2004. Air-stable allpolymer field-effect transistors with organic electrodes. Synth Met 145:83–85. 353. T. Granlund, T. Nyberg, L. S. Roman, M. Svensson, and O. Inganäs. 2000. Patterning of polymer light-emitting diodes with soft lithography. Adv Mater 12(4):269–273. 354. D. Li and L. J. Guo. 2006. Micron-scale organic thin film transistors with conducting polymer electrodes patterned by polymer inking and stamping. Appl Phys Lett 88:063513-1–063513-3. 355. P. Cosseddu and A. Bonfiglio. 2007. A comparison between bottom contact and top contact all organic field effect transistors assembled by soft lithography. Thin Sold Films 515:7551–7555. 356. R. Schroeder, L. A. Majewski, M. Grell, J. Maunoury, J. Gautrot, P. Hodge, and M. Turner. 2005. Electrode specific electropolymerization of ethylenedioxythiophene: Injection enhancement in organic transistors. Appl Phys Lett 87:113501-1–113501-3. 357. K. Hong, S. Y. Yang, C. Yang, S. H. Kim, D. Choi, and C. E. Park. 2008. Reducing the contact resistance in organic thin-film transistors by introducing a PEDOT:PSS hole-injection layer. Org Electron 9:864–868. 358. J. Lu, N. J. Pinto, and A. G. MacDiarmid. 2002. Apparent dependence of conductivity of a conducting polymer on an electric field in a field effect transistor configuration. J Appl Phys 92(10):6033–6038. 359. A. J. Epstein, F. C. Hsu, N. R. Chiou, and V. N. Prigodin. 2003. Doped conducting polymer-based field effect devices. Synth Met 137(1–3):859–861. 360. H. Okuzaki, M. Ishihara, and S. Ashizawa. 2003. Characteristics of conductive polymer transistors prepared by line patterning. Synth Met 137(1–3):947–948.
6911X.indb 263
10/1/10 9:01:13 PM
264
PEDOT
361. H. S. Kang, H.-S. Kang, J. K. Lee, J. W. Lee, J. Joo, J. M. Ko, M. S Kim, and J. Y. Lee. 2005. Humidity-dependent characteristics of thin film poly(3,4-ethylenedioxythiophene)field-effect transistors. Synth Met 155(11):176–179. 362. J. T. Stricker, A. D. Gudmundsdóttir, A. P. Smith, B. E. Taylor, and M. F. Durstock. 2007. Fabrication of organic thin-film transistors using layer-by-layer assembly. J Phys Chem B 111(23):6322–6326. 363. F. C. Hsu, V. N. Prigodin, and A. J. Epstein. 2006. Electric field controlled conductance of metallic polymers in transistor structure. Phys Rev B 74(23): 235219/1–235219/12. 364. D. Nilsson, N. Robinson, M. Berggren, and R. Forchheimer. 2005. Electrochemical Logic Circuits. Adv Mater 17(3):353–358. 365. A. Elschner. 2007. Unpublished results. 366. A. Amerasekera and C. Duvvury. 2002. ESD in Silicon Integrated Circuits. New York: J. Wiley & Sons. 367. X . M. Jiang, R. Österbacka, O. Korovyanko, C. P. An, B. Horovitz, R. A. J. Janssen, and Z. V. Vardeny. 2002. Spectroscopic studies of photoexcitations in regioregular and regiorandom polythiophene films. Adv Funct Mater 12(9):587–597.
6911X.indb 264
10/1/10 9:01:13 PM
11 Technical Use and Commercial Aspects An overview of technical applications of poly(3,4-ethylenedioxythiophene) (PEDOT) is given in Table 11.1. The largest quantity of the worldwide PEDOT demand is consumed in the form of the monomer 3,4-ethylenedioxythiophene (EDOT), which is subsequently polymerized to PEDOT by its users. However, the largest number of technical applications prefer poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), or PEDOT:PSS, as a prefabricated polymer, which allows the fabrication of conductive films quite conveniently by coating or printing the aqueous dispersion (see Figure 11.1). The film properties and processing conditions can be adjusted by adding suitable formulation components. A significant market for electrochemically and chemically polymerized PEDOT originates from capacitor manufacturing, which uses PEDOT as solid electrolyte in aluminum and tantalum polymer capacitors. The chemical polymerization or in situ polymerization of EDOT demands tight process control during the polymer formation. Concentration of reactants, temperature, and other factors have a complex influence on the polymerization speed and resulting morphology.1 The major technical disadvantage of in situ PEDOT is the fact that after deposition it forms an insoluble and brittle polymer that does not allow postprocessing. The strongest argument for chemically and electrochemically generated PEDOT has been its high conductivity easily reaching up to 1000 S/cm. However, since high-conducting PEDOT:PSS dispersions with similar conductivities have been developed, this argument has vanished.2 In the early 1990s, the capacitor industry started to exploit conductive polymers as a high conductive alternative to manganese dioxide and liquid electrolytes. EDOT was found to penetrate well into porous structures, build highly conducting and stable polymers and was subsequently used for the impregnation of anodes for polymer electrolyte capacitors. The electrochemical polymerization of conductive polymers is difficult to translate into the technical scale and has been implemented only for pyrrole in the industry. In this application electrochemically polymerized polypyrrole competes with chemically polymerized PEDOT. Clear advantages of the electrochemical process are the quantitative monomer consumption without significant byproducts and an excellent corner coverage of anodes, while the main obstacle results from the more sophisticated technical setup and the difficult electrochemical deposition of polymer in small pores. 265
6911X.indb 265
10/1/10 9:01:14 PM
6911X.indb 266
Cathode Cathode
Through hole plating
Antistatic layer Antistatic layer Antistatic layer Antistatic layer Transparent electrode Transparent electrode Hole injection layer Transparent electrode
Hole transport layer Transparent electrode
Electrochromic layer Transparent conductor
Printed wiring boards
Photographic films Packaging films Electronic films (LCD) CRT Electroluminescent lamps Touch screen OLED OLED
Organic solar cell Organic Solar Cell
Smart window Printed displays
Function
Solid electrolyte capacitors Solid electrolyte capacitors
Application
Technical Applications of PEDOT
Table 11.1
Coating Coating Coating Coating Screen printing Coating, printing (screen) Coating, inkjet printing Coating, printing (screen, inkjet, flexo) Coating, printing Coating, printing (screen, inkjet, flexo) Coating, printing Printing
Chemical polymerization Impregnation with polymer dispersion Chemical polymerization
Process
PEDOT:PSS PEDOT:PSS
PEDOT:PSS, PEDOT:polymer PEDOT:PSS
EDOT (microemulsion); Manganese dioxide (via KMnO4 as etching agent) PEDOT:PSS PEDOT:PSS PEDOT:PSS PEDOT:PSS PEDOT:PSS PEDOT:PSS PEDOT:PSS, PEDOT:polymer PEDOT:PSS
EDOT; Oxidant PEDOT:PSS
Materials
R&D Introduction
R&D R&D
Established Established Established Terminated Established Introduction Introduction R&D
Established
Established Established
Status in 2009
266 PEDOT
10/1/10 9:01:14 PM
267
Technical Use and Commercial Aspects
(a)
(b)
(c)
Figure 11.1 (a) EDOT, (b) PEDOT:PSS dispersion, and (c) PEDOT:PSS coated foils.
In 2004, PEDOT:PSS dispersions started to be used for the outer polymer layers in polymer electrolyte capacitors, which simplified the production of polymer capacitors significantly. Two years later, in 2006, nanosized PEDOT:PSS dispersions became available that penetrate into the microscopic pores of sintered or etched anodes. Since the chemical polymerization adds a significant amount of process costs due to repetitive polymerization cycles and waste of costly monomer by uncontrolled polymerization, the inner impregnation by a prefabricated polymer was recognized as an attractive cost alternative to the chemical polymerization. The technology change from chemical and electrochemical polymerization to polymer impregnation is still ongoing.
Figure 11.2 Tantalum polymer capacitors. (H.C. Starck GmbH.)
6911X.indb 267
10/1/10 9:01:16 PM
268
PEDOT
Today, almost all major capacitor manufacturers offer tantalum or aluminum polymer capacitors. The main feature of polymer capacitors is a low equivalent serial resistance (ESR) allowing operation at high frequency. A second market for EDOT is its use in a specific manufacturing step to produce printed wiring boards. System suppliers like Enthone-OMI (Germany) and Atotech (United States) offer direct through hole-plating processes that can be applied to vias and high aspect through holes.3 Limited approaches have been made to develop continuous coating processes based on in situ PEDOT.4 Around 1995, Agfa-Gevaert (Belgium) introduced PEDOT:PSS in an industrial product in the form of a roll-to-roll deposited antistatic layer on biaxially oriented poly(ethylenterephthalate) (PET). The antistatic films are used as substrates for the production of photographic films. The antistatic layer avoids undesired electrostatic discharges into the photoactive layers during film processing. Subsequently a market developed for high-performance antistatic materials, like transparent blister tapes, for the packaging of electronic components and as protection film for flat panel displays (see Figure 11.3). A key for coating or printing PEDOT:PSS is the availability of suitable formulations, or in other words, the addition of components that modify film and process properties specifically for the intended application and process. In a recent study, PEDOT was highlighted as a potential replacement for ind ium tin oxide (ITO) as an transparent conductor especially for flexible devices, like touch screens, organic solar cells, and organic light emitting diodes.5 The Organic Electronics Association (OE-A, a working group within VDMA, see www.vdma.org) has released a white paper6 that describes PEDOT:PSS as an essential component for organic and printed electronics like organic solar cells, printed memories, printed radio frequency identification tags, organic sensors, and printed smart objects. Prototypes of electrochromic windows and displays employing PEDOT created some hope for new markets but still wait for a technical breakthrough.
(a)
(b)
(c)
Figure 11.3 PEDOT:PSS coated (a) modul trays, (b) blister tapes, and (c) cover tapes.
6911X.indb 268
10/1/10 9:01:18 PM
Technical Use and Commercial Aspects
269
References
6911X.indb 269
1. K. Levon, K. C. Park, and C. Cai. 1997. Structure formation in conducting polymers. Synth Met 84(1-3):335–338. 2. Product Information on CLEVIOS PH 1000. www.clevios.com (accessed August 2010). 3. J. Hupe, G. D. Wolf, and F. Jonas. 1995. DMS-E-Bekanntes Prinzip mit neuer Basis. Galvanotechnik 86(10):3404–3411. 4. M. Bergsmann, F. Kastner, and E. Wagner. EP 1562154 (Hueck Folien Ges.m.b.H), Prior: December 20, 2004. 5. NanoMarkets. 2008. The future of ITO: Transparent conductor and ITO replacement markets. Glen Allen: NanoMarkets. 6. Organic Electronics Association. 2008. OE-A Roadmap for Organic and Printed Electronics. Frankfurt: Organic Electronics Association.
10/1/10 9:01:18 PM
6911X.indb 270
10/1/10 9:01:18 PM
12 EDOT and PEDOT Derivatives with Covalently Attached Side Groups
12.1 EDOT-CH 2OH and Its Derivatives 3,4-Ethylenedioxythiophene, or EDOT, functionalization with moieties in side chains is easily achievable by the introduction of hydroxyl groups. So, a wide range of interesting EDOT and PEDOT [poly(3,4-ethylenedioxythiophene)] derivatives is accessible starting with the EDOT-methanol (2,3dihydrothieno[3,4-b][1,4]dioxin-2-yl)methanol = the formula in Figure 12.1a). For the sake of readability, the abbreviation EDOT-CH2OH is used in the following. A versatile synthetic route to EDOT-CH2OH (Figure 12.1a) is very similar to the Gogte route, described for EDOT in Chapter 5, via 3,4-dihydroxythiophene2,5-dicarboxylic acid esters.1 Due to the difunctional epoxide group in epibromohydrine, not only reacting in the desired 2-position, but also at C-atom 3 to a minor extent, the product is accompanied by the isomeric hydroxy-3,4propylenedioxythiophene (Figure 12.1b) equals ProDOT-OH in a ratio of about 4:1. The synthesis is depicted in the reaction scheme in Figure 12.2. Ng et al. described the use of less expensive and industrially far better available epichlorohydrine instead of the bromo derivative,2 but in the experience of the authors of this book the synthesis could not be reproduced in good yields. Another nucleophilic route to EDOT-CH2OH, very similar to the scheme in Figure 12.2, utilizes 1-acetoxy-2,3-dibromopropane instead of epibromohydrin (Figure 12.3).3 Although formation of the undesirable isomer ProDOT-OH is omitted, the synthesis suffers from low yields (about 25%) in the nucleophilic substitution step.3 The isomeric mixture of EDOT-CH2OH (Figure 12.1a) and ProDOT-OH (Figure 12.1b) is difficult to separate. More easily than in the end product stage, the two isomeric ester intermediates can be separated by flash chromatography on silica gel.1 Further hydrolysis and decarboxylation of the free acids by copper catalysts allow the isolation of pure EDOT-CH2OH and ProDOT-OH. Chevrot and coworkers later found that a specific solvent mixture of diethylether/cyclohexane (95:5) facilitates separation of EDOT-CH2OH
271
6911X.indb 271
10/1/10 9:01:19 PM
272
PEDOT
OH
OH O
O
O
O
S
(a)
(b)
S
Figure 12.1 EDOT-CH2OH and its isomer, ProDOT-OH.
and ProDOT-OH by column chromatography.4 Preparative HPLC is also sufficient for small amounts.5 Another alternative synthesis for EDOT-CH2OH, resulting in only minor amounts of the byproduct ProDOT-OH, was suggested in a patent.6 The transetherification of lower 3,4-dialkoxythiophenes with glycerol can be extended to higher triols, like 1,2,4-butanetriol or 1,2,6-hexanetriol, so several homologues of EDT-CH2OH were accessible (ω-hydroxyalkyl-EDOTs; Figure 12.4).6 OH O
O Na+ O–
O– Na+
ROOC
O Br
ROOC
S
OH
COOR
O
O ROOC
O
S 80%
+
S
COOR
OH
OH 1. Hydrolysis O 2. Decarboxylation
COOR
S
O
O
S 20%
Figure 12.2 EDOT-CH2OH synthesis.
6911X.indb 272
10/1/10 9:01:20 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
HO EtOOC
Br
OH
OOC-CH3
Br OOC-CH3
COOEt
S
O
O EtOOC
COOEt
S
OH
HOOC
OH
O
O
NaOH/H2O
273
CuCrO2/Quinoline
O
O
COOH
S
S
Figure 12.3 EDOT-CH2OH synthesis from 1-acetoxy-2,3-dibromopropane.
This transetherification route suffers from low yields, presumably due to the polyfunctionality of the alcohol, and so does not represent a technical alternative to the synthesis via epibromohydrin. A more efficient transetherification route—although including more steps—has been claimed in another patent, starting, for instance, with 3,4dimethoxythiophene and 3-chloro(or bromo)-1,2-propanediol.7 The resulting EDOT-CH2Cl(Br) is transferred to the corresponding acetic acid ester (or similar); then pure EDOT-CH2OH, of course free from ProDOT-OH, can be isolated after hydrolysis. Direct alkaline hydrolysis of the halogen compound also may be sufficient. Although purification of EDOT-CH2OH (and the mixture with ProDOT-OH, respectively) by distillation is possible at 115°C, 1 mbar or even at water pump vacuum (166°C, 16 mbar), careful prepurification of the raw material before distillation has to be performed. When raw EDOT-CH2OH—from whatever synthetic route—is contaminated with traces of acid, a remarkable tendency to a rapid and severe decomposition at elevated temperature is observed, turning distillation into total loss of the material. OH n
O
O
n = 1, 3
S Figure 12.4 ω-Hydroxyalkyl-EDOTs.
6911X.indb 273
10/1/10 9:01:21 PM
274
PEDOT
The chemical behavior of EDOT-CH2OH is very similar to EDOT regarding the oxidative polymerization to the conductive polythiophene. Some aspects are influenced beneficially by the better solubility in water. The 4:1 mixture of EDOT-CH2OH and ProDOT-OH behaves similar to pure EDOT-CH2OH. The in situ polymerization of EDOT-CH2OH yields highly conductive layers with a somewhat different performance in capacitors, compared to in situ PEDOT layers.8 The overall conductivity of PEDOT-CH2OH is similar, in the range of several hundred siemens per centimeter, and may exceed the in situ PEDOT conductivity under particular conditions.1,8–10 The electrochemical polymerization of EDOT-CH2OH was studied by Chevrot’s group in comparison to EDOT.11 Dodecylbenzenesulfonate was used as the counterion. EDOT-CH2OH was found to be less resistant to oxidation than EDOT—the polymer seemed to be irreversibly damaged at about 1.2 V (relative to saturated calomel electrode [SCE]), compared to about 1.4 V/ SCE for PEDOT. Scanning electron microscopy (SEM) and surface profiling revealed a higher compaction of the PEDOT-CH2OH films. Electropolymerized EDOT-CH2OH with a poly(styrenesulfonate) (PSS) counterion was used to modify the surface of gold microfabricated neural probes by Reynolds and coworkers.12 Cyclic voltammetry demonstrated the increased charge capacity of PEDOT-CH2OH:PSS coated electrodes relative to the bare gold electrodes. The impedance was also lower for PEDOT-CH2OH:PSS over the frequency range of 1 to 105 Hz than for uncoated gold electrodes. SEM images revealed the good uniformity of the PEDOT-CH2OH:PSS films, compared to electrochemically polymerized PEDOT:PSS. PEDOT-CH2OH doped with the biologically active nonapeptide CDPGYIGSR (a laminine fragment) could also be successfully deposited onto the electrodes.12 The oxidative polymerization of EDOT-CH2OH or its 4:1 mixture with ProDOT-OH is easily performed in aqueous solution with polystyrenesulfonic acid as the charge balancing counterion due to the improved (about sevenfold) water solubility, compared to EDOT. The product has similar conductivity as PEDOT:PSS. Considerable scientific interest has been put on EDOT-CH2OH because of its facile derivatization. A few simple alkyl ethers of EDOT-CH2OH–ProDOT-OH mixtures (n-hexyl and benzyl ether) had been synthesized first by Blohm et al.1 More systematic investigations with pure EDOT-CH2OH were started by the group of Chevrot.4 They synthesized medium- and long-chain alkyl ethers from n-C6H13 to n-C16H33 (Figure 12.5) and polymerized these ethers chemically with iron(III)-chloride. The fraction of undoped, CH2Cl2- soluble polymeric material, obtainable by this polymerization method and with optimized substoichiometric amounts of FeCl3, depends on the alkyl chain length: No soluble part was isolated for chain length smaller than n-C12H25. This observation parallels the observations made by Reynolds and Sankaran with alkyl-(P)EDOTs.13 The readily soluble long-chain ethers could be analyzed by GPC; molecular weights (Mw) of 12,900 and 14,300 were found for PEDOT-CH2O-n-C14H29 and PEDOTCH2O-n-C16H33, resp., with a polydispersity of 1.45 and 1.65.4
6911X.indb 274
10/1/10 9:01:21 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
O– Na+
OH O
O
O
NaH –H2
O
275
OAlkyl
Alkyl-Hal
O
O
–NaHal S
S
S
Figure 12.5 EDOT-CH2OH ether synthesis.
Isocyanates can be added to the hydroxyl groups of EDOT-CH2OH and its isomeric mixtures under mild conditions. Several examples of urethanes have been prepared, exhibiting an electrical conductivity similar to that of PEDOT or PEDOT-CH2OH.6 More information and trends in EDOTCH2OH derived ethers and urethanes with various substituents, regarding their surface resistances after in situ polymerization, can be found in Chapter 8.6,10 The smooth addition of cyclic dicarboxylic acid anhydrides to the EDOTCH2OH hydroxyl group was first utilized in the early General Electric Patent1 with succinic anhydride, resulting in an EDOT bearing a COOH group. Another approach to carboxylic acid functionalized (P)EDOT was realized in a patent application by Agfa-Gevaert. EDOT-CH2OH was etherified (after sodium salt formation with sodium hydride) with ethyl chloroacetate with subsequent saponification to the free acid (Figure 12.6).14 The carboxylic acid moiety is not sufficient to give—after neutralization— water soluble, so-called self-doping PEDOT derivatives. Copolymers of EDOT-CH2O-CH2-COOH (the carboxylic acid in Figure 12.6) with EDOT and
O– Na+
OH O
O
NaH
O
O
Br–CH2–COO–C2H5
–H2
–NaBr
S
S O–CH2–COO–C2H5 O
O
(KOH)
S
O–CH2–COOH O
O
S
Figure 12.6 Synthesis of COOH-functionalized EDOT.
6911X.indb 275
10/1/10 9:01:22 PM
276
PEDOT
O O
O
O
O
O
O
S Figure 12.7 Example for oligo(oxyethylene)-modified EDOT.
poly(styrenesulfonic acid) as the counterion have been prepared.14 The preparation and some properties of truly waterborne, self-doped EDOT derivatives bearing the more solubilizing sulfonate groups are described later in this chapter. Hydrophilic oligoethyleneethers have been prepared by the reaction of EDOT-CH2OH with, for example, tetraethyleneglycolmonomethylethermesylate (see the formula of the product in Figure 12.7).15 Analog structures with penta- and hexa(oxyethylene) chains with EDOT residues at both chain ends of the PEG oligomer have also been prepared.15 The analysis of the cyclovoltammetric behavior of these polyfunctional compounds in the presence of metal cations like Li+, Na+, Ca2+, Sr2+, and Ba2+ revealed the cation recognition properties of the cross-linked structures. Peak potential shifts of up to several hundred millivolts, compared to Bu4N+ were observed and assigned to conformational changes induced by cation complexation by the oligo(oxyethylene) chains. Similar structures based on EDT-CH2OH, but etherified with ω-iodopolyethers have also been described.16 EDOT-CH2OH, for example, was mono-etherified with 3,6,9,12,15-pentaoxa-heptadecane-1,7-ditosylate (pentaethyleneglycol ditosylate), followed by conversion of the monotosylate intermediate to the corresponding iodo compound with sodium iodide. This monomer was electropolymerized and then subjected to an interesting postpolymerization functionalization. For this purpose, the potentiostatically grown polymer films were immersed into a N,N-dimethylformamide (DMF) solution of cyanoethyl-functionalized tetrathiafulvalene (TTF) and cesium hydroxide. After (stoichiometric) functionalization (see Figure 12.8) an electroactive polymer combining PEDOT and a TTF moiety was obtained and verified by its typical cyclovoltammogram, exhibiting both redox processes for the PEDOT backbone and the attached TTF. Another more complex side chain was described by Brisset, Roncali, et al. who were able to introduce ferrocenyl residues (see Figure 12.9).17 Since the oxidation potential of the ferrocene moiety is considerably lower than that of EDOT, severe difficulties were observed for the electropolymerization
6911X.indb 276
10/1/10 9:01:22 PM
277
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
O
O O
O
CH3S
S
S
S
CH3S
S
S
SCH3
+
m=5 S
CN
I
m
n
CH3S O
O O
S
m
S
S
SCH3
S
S
SCH3
O
CsOH S
n
Figure 12.8 TTF-functionalized PEDOT.
of this compound. But the coelectropolymerization of this precursor with hydroxymethyl EDOT and a two-site precusor involving two EDOT groups linked by a polyether chain [(EDOT)2-polyether] could be successfully investigated in potentiodynamic and potentiostatic conditions. The analysis of the electrochemical behavior of the films shows that potentiostatic copolymerization of ferrocenyl-EDOT and (EDOT)2-polyether in nitrobenzene leads to a stable electroactive copolymer, which retains the electroactivity of the ferrocene group even in aqueous medium.17 Bulky mesogenic groups were first introduced without utilizing the specific liquid crystalline (LC) properties of the EDOT derivative. Kumar’s group
O Fe O O
O
S Figure 12.9 Ferrocenyl-functionalized EDOT.
6911X.indb 277
10/1/10 9:01:23 PM
278
PEDOT
O
CN
O
O
O
S Figure 12.10 4-Cyanobiphenyl-functionalized EDOT.
studied the electrochromism of the cyanobiphenyl substituted EDOT depicted in Figure 12.10, prepared from EDOT-CH2OH and 4-(6-bromohexyloxy)-4′-cyan obiphenyl.18 The rigid and bulky substituent lead to an improved contrast (67% at 660 nm) and better switching in electrochromic devices18—not unexpected in the light of earlier findings with long-chain-alkyl-substituted EDOT derivatives.13 Systematically varying LC derivatives of EDOT-CH2OH with respect to the specific LC character and the influence of the mesogenic side chains (including the variation of spacers between EDOT and mesogenic group) were extensively studied.19,20 A remarkable, irreversible positive effect of annealing at about 150°C on the conductivity of in situ polymerized layers of these compounds was observed, compared to aliphatic and cycloaliphatic derivatives of EDOT-CH2OH, which do not show this effect. This topic was discussed in detail in Chapter 8. Figure 12.11 presents a typical example of the liquid crystalline phase of a mesogenic monomer (see the formula in Figure 12.10) and the transition to the crystalline state.
a
b
c
Figure 12.11 Polarizing microscopy pictures of monomeric 4-cyanobiphenyl-functionalized EDOT (formula see Figure 11.10) during secondcooling (44°C): (a) 5 °C/min, formation of a LC phase; (b) 5 min annealing at 44°C, LC phase, start of spherulite formation; (c) completely crystalline after more than 1 h annealing at 44°C. (From N. Wrubbel, H. Ritter, K. Reuter, A. Karbach, and D. Drechsler, 2006, e-Polymers 2, http://www.e-polymers.org/papers/ritter_030206.pdf. With permission.)
6911X.indb 278
10/1/10 9:01:26 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
279
O C7F15 O
OH O
O
C7F15CO–Cl
O
O
(Et3N) S
S Figure 12.12 Perfluorooctanoate-functionalized EDOT.
The versatile derivatization of EDOT-CH2OH has been utilized to prepare electrochromic fluoroalkanoate substituted PEDOT.21 EDOT-CH2OH could easily be esterified by perfluorooctanoylchloride (2,2,3,3,4,4,5,5,6,6,7,7,8,8,8pentadecafluorooctanoyl chloride) to the corresponding EDOT-F ester (see Figure 12.12). PEDOT-F is a cathodically coloring polymer that exhibits rapid switching between a dark-blue reduced state and a transmissive gray-blue oxidized state with high visible contrast, similar to PEDOT. The conductivity of free-standing films is lower than for PEDOT, but the benefit of a considerably enhanced hydrophobicity observed by contact angle measurements. A better environmental stability is therefore discussed. The azidomethyl-EDOT synthesis also starts from EDOT-CH2OH. The mesylate obtained with methanesulfonyl chloride can be reacted with sodium azide to azidomethyl-EDOT (see Figure 12.13).22,23 Decomposition to the corresponding amine (aminomethyl-EDOT) under evolution of nitrogen can be performed successfully with triphenylphosphane.22,23 The amine forms cyclic amides with, for example, naphthalenetetracarboxylic anhydride.22,23 The azidomethyl-EDOT is a versatile synthon for 1,3-dipolar cycloadditions (click chemistry) with alkynes. Poly(azidomethyl-EDOT) films have been “click” functionalized with various alkynes,24–26 including ethinylferrocene.27
O
O
N3
O–SO2–CH3
OH CH3SO2Cl
O
O
NaN3
O
O
(NEt3) S
S
S
Figure 12.13 Azidomethyl-EDOT and its synthesis.
6911X.indb 279
10/1/10 9:01:27 PM
280
PEDOT
12.2 EDOT-CH 2Cl and Its Follow-Up Products Several EDOT derivatives that are not very easily accessible from EDOTCH2OH can be synthesized starting with EDOT-CH2Cl (chloromethyl-3,4ethylenedioxythiophene; 2-(chloromethyl)-2,3-dihydrothieno[3,4-b][1,4]dioxine). EDOT- CH2 Cl—and analogously the bromo analog EDOT-CH2Br—is prepared via transetherification of 3,4-dimethoxythiophene with 3-chloro(bromo)-1, 2-propandiol (see Figure 12.14).7,28,29 EDOT-CH2I has been prepared from EDOT-CH2Cl via Finkelstein reaction in acetone.23 EDOT-CH2Cl is a very versatile synthon for functionalized EDOT derivatives. Two advantageous aspects, compared to EDOT-CH2OH, have to be considered. First, the more straightforward reactivity for the synthesis of ethers and amines or amides. Second, the product is free from its ProDOT isomer, which contaminates EDOT-CH2OH dependent on synthetic route and method of purification. Several bioinspired EDOT derivatives have been prepared starting with EDOT-CH2Cl. In a patent,28 electrochemical sensors have been claimed, for example, utilizing PEDOT with DNA sequences in the side chain. A similar approach was described by Bazaco et al.30 The authors covalently linked PEDOT with the nucleobase uracil and performed successful recognition experiments with the complementary base adenine. The electrochemical polymerization of uracil-substituted EDOT (synthesis; see Figure 12.15) was studied by cyclic voltammetry (CV) in dichloromethane, using tetrabutylammonium hexafluorophosphate as the electrolyte. The homogeneous polymer films were further electrochemically characterized (CV) in the presence of various concentrations of adenine or the noncomplementary bases cytosine and uracil. The concentration-dependent electroactivity in the aqueous electrolyte (0.1 M LiClO4) exhibited by far the biggest changes with the complementary base adenine. So a specific recognition (see Figure 12.16) can be concluded. A convenient and straightforward procedure for the preparation of an EDOT system functionalized by a perylenetetracarboxylic diimide also started with EDOT-CH2Cl. EDOT-CH2Cl was reacted with a mono-(4-hydroxphenylimide) of perylenetetracarboxylic acid by Williamson ether synthesis.29 The electrochemical polymerization of the monomer to the polymer depicted in Figure 12.17
OCH3
CH3O
S
HO
Cl(Br)
OH Cl(Br)
O
O
p-TSA S
Figure 12.14 Synthesis of chloromethyl- and bromomethyl-EDOT. (Adapted from S. Yeisley, C. J. Dubois, Jr., C.-H. Hsu, S. W. Shuey, Y. Shen, and H. Skulason, WO 2006/073968. Prior: December 30, 2004; S. J. Higgins and F. Mouffok, WO 2006/018643. Prior: March 18, 2004; J. L. Segura, R. Gómez, E. Reinold, and P. Bäuerle, 2004, Org Lett 7(12):2345–2348.)
6911X.indb 280
10/1/10 9:01:28 PM
281
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
H
O Cl
N
H
O
O
+
N
O
N
O
N
K2CO3 O
DMSO
H
O
O
S
S
Figure 12.15 Synthesis of uracil-modified EDOT.
N
N H
N
N
H
N
H
H
O N N O
O
O
n
S
Figure 12.16 Proposed complexation of adenine at the polymer surface of uracil-modified PEDOT.
O
O
N
N
O
O
C9H19 O
C9H19 O
O
S
n
Figure 12.17 Perylenebisimide-functionalized PEDOT.
6911X.indb 281
10/1/10 9:01:31 PM
282
PEDOT
Cl O
O
+
N N
N
CH3
S
80°C, 24 h
O
O
+
N
CH3 Cl–
S
Figure 12.18 Imidazolinium-modified EDOT (EDOT-Im).
yielded a novel donor–acceptor system with an absorption covering a wide range of the solar spectrum. So this polymer was suggested as a good candidate for organic photovoltaic devices.29 In a subsequent publication the studies were extended to hybrid systems with 9,10-anthraquinone and 11,11,12,12-tetracyano9,10-anthraquinodimethane substituents.31 A novel PEDOT type bearing cationic side chains was synthesized following the scheme in Figure 12.18 from EDOT-CH2Cl and 1-methylimidazole.32 Chloride is exchanged to more typical ionic-liquid salt anions like hexafluorophosphate, tetrafluoroborate, or bis(trifluoromethane)sulfonimide (CF3SO2)2N– (triflimide), by precipitation in water with the corresponding potassium salts. The synthesis and characterization of the new functional poly(3,4-ethylenedioxythiophenes) bearing imidazolinium ionic-liquid moieties (PEDOT-Im) is reported in more detail by DÖbbelin et al.33 PEDOT-Im shows multiresponsive properties to a variety of stimuli, such as temperature, pH, oxidative doping, and presence of anions. These stimuli provoke different changes in PEDOT-Im, such as changes in color, oxidation state, and wetting behavior. In all cases, a reversible effect is observed, and the polymers reveal responsive properties in solution as well as in the form of thin films. Responsiveness to temperature and to anions are reported as a unique property of PEDOT-Im. Anion exchanges induce fast, adjustable, and reversible contact angle changes between 24° and 107°. As a potential application, surfaces with switchable wettability triggered by anion solutions are prepared by spin coating PEDOT-Im films.33
12.3 Alkyl EDOTs With the first reports on EDOT’s special polymerization behavior yielding conductive polymers, several simple derivatives also were published simultaneously.34 Short-chain alkyl substitution at the dioxane ring does not change the EDOT and PEDOT properties very much. As is possible with EDOT-CH2OH, EDOT-CH3 can be polymerized oxidatively with PSS as the counterion in water. Conductivity is decreased, compared to
6911X.indb 282
10/1/10 9:01:31 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
HO
Hal R
Hal
+
R´OOC
283
OH COOR´
S
(1)
R OH R
OH
HO OH
(2)
+ R´OOC
S OR´
+ R
O
COOR´
S
R´O
OH
O
(3)
OH S
(1): Williamson ether synthesis, followed by hydrolysis/decarboxylation (2): Mitsunobu reaction, followed by hydrolysis/decarboxylation (3): Acid-catalyzed transetherification Figure 12.19 Alkyl-EDOT synthesis.
PEDOT:PSS. Medium- or long-chain alkyl EDOTs fail due to their very low water solubility. The synthesis of alkyl EDOTs and several special problems especially with longer alkyl chain educts have been described in Chapter 5. To give an overview here, too, a summary covering the most important methods is depicted in Figure 12.19. Alkyl ether synthesis by Willamson ether synthesis is not very sufficient for longer chain alkylbromides like tetradecylbromide, due to strongly competing elimination reactions. For example, yields of the Williamson ether synthesis are reported to be as low as 24% (R = C14H29).13 Here the transetherification has to be preferred. See also the corresponding references in Chapter 5. The conductivity of PEDOT-alkyl in the form of the in situ polymerizates is dependent on the alkyl group length, but the experimental results reported do not give a very clear picture. A good overview could be obtained by the special in situ measurement method described in detail by Aubert et al.35 Table 12.1 shows the data compiled in by Groenendaal et al.36 The interesting effect of going through a conductivity minimum at medium chain length and subsequent reincrease with longer alkyl substituents, even exceeding PEDOT conductivity, has to be discussed. A logic explanation is given by the authors by claiming morphological influences induced by steric hindrance at medium chain length and ordering effects with long alkyl chains. The relatively high conductivity reported for PEDOT-C14H29 cannot be
6911X.indb 283
10/1/10 9:01:32 PM
284
PEDOT
Table 12.1 Properties of Alkyl-EDOTs (Acetonitrile + 0.1 M Bu4NClO4) PEDOT Derivative PEDOT PEDOT-CH3 PEDOT-C6H13 PEDOT-C10H21 PEDOT-C14H29
Conductivitya (S/cm)
Redox Potential (V)
Absorption Maximumb (nm)
650 300 200 550 850
–0.6 –0.8 –0.45 –0.4 –0.35
610 590 650 650 650
Source: Data from L. Groenendaal, G. Zotti and F. Jonas, 2001, Synth Met 118 (1–3):105–109. a Maximum in situ conductivity σ (oxidized form). b Neutral form.
reproduced with in situ experiments following the recipes in Chapter 8 (iron(III) tosylate/butanol on glass plates), where more detailed information is given. Alkyl-EDOTs and their polymers have broadly been investigated by Reynolds and colleagues with respect to their optoelectrochemical properties.13,37–43 The typical PEDOT electrochromism had first been reported by the collaborating Bayer and University of Freiburg groups44 and—independently and only shortly afterward—by Inganäs and coworkers.45,46 The fundamentals and all aspects regarding PEDOT and PEDOT:PSS electrochromism, especially potential applications, are discussed in Chapter 10. Here the particular properties or sometimes advantages of alkyl-EDOTs shall be summarized, mainly following the systematic study published by Reynolds et al.40 The aryl-EDOT “EDOT-phenyl,” see Figure 12.20, was included in this study. Several results regarding switching times between colored = reduced and transmissive = oxidized state at 590 nm are given in Table 12.2. The data of Table 12.2 were obtained at electrochemically deposited films from acetonitril solutions with tetra-n-butylammonium perchlorate electrolyte. Obviously alkyl or aryl substitution of PEDOT improves not only the switching speed, but can also enhance the optical contrast, measured as ΔT (= transmission change) in percent, switching from the reduced to the more
O
O
S Figure 12.20 EDOT-phenyl.
6911X.indb 284
10/1/10 9:01:32 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
285
Table 12.2 Electrochromic Properties of PEDOT Derivatives PEDOT PEDOT-C14 PEDOT-Ph
Switching Time
Optical Contrast
2.2 s 0.8 s 0.8 s
44% 63% 45%
60
% Transmittance
50 40 30 20 10 0
0
10
20
0
10
20
30
40
50
60
30
40
50
60
70
% Transmittance
60 50 40 30 20 10 0
Time (s) Figure 12.21 Electrochromic switching, optical absorbance change, monitored at 590 nm for PEDOT (top) and PEDOT-Ph (bottom): in 0.1 M TBAB/Acetonitrile. (Reprinted with permission from A. Kumar, D. M. Welsh, M. C. Morvant, F. Piroux, K. A. Abboud, and J. R. Reynolds, Derivatives as Fast Electrochromics with High Contrast Ratios, Chem Mater 10(3):896–902. Copyright 1998 American Chemical Society.)
6911X.indb 285
10/1/10 9:01:34 PM
286
PEDOT
transparent oxidized state. Figure 12.2141 illustrates the more rapid switching of PEDOT-Ph, compared to PEDOT.
12.4 Water Soluble, “Self-Doping” EDOT Derivatives Although processible from aqueous medium, PEDOT is not truly water soluble, noted earlier. The request for truly water soluble PEDOT materials as early as 1991 led to the discovery of sulfonated, so-called self-doped PEDOT derivatives in a patent application of General Electric Co. published in 1992.1 Starting material was the mixture of the novel EDOT-CH2OH (Figure 12.1a) and its isomer (Figure 12.1b). The smooth nucleophilic ring opening reaction of its alcoholate anion with sultones can be utilized to prepare functional ethers with sulfonic acid groups (only depicted with the isomer EDT-CH2OH in Figure 12.22). Blohm et al. of General Electric used 1,3-propanesultone (n = 1) in US Patent 5,111,327,1 which is highly toxic, strongly carcinogenic in animal tests, and therefore restricted in its technical utility. Only one but rather short further study with the propanesultone-derived material was published in 2002.47 Most of the more recent investigations are based on EDOT-S synthesized from EDOT-CH2OH and the less harmful 1,4-butanesultone (n = 2).48–50 As EDOT moieties are proton sensitive, the free acid is not very stable. Even more pronounced is the tendency to oxidative decomposition in air.50 Therefore, EDOT-S is stored preferably in the form of its sodium salt of (see Figure 12.22; n = 2). EDOT-SNa is readily water soluble (as are the lithium and the n-Bu4N+ salt)50 and can easily be polymerized by similar preparative methods as EDOT. Following the PEDOT:PSS process, sodium persulfate and catalytic amounts of Fe salts [Fe(III) sulfate or similar] are necessary ingredients for the oxidative polymerization in water, but polystyrene sulfonic acid (PSS) can be omitted, because PEDOT-S is self-doping or, better, contains the conducting bipolaron moiety and the negatively charged counterion in the same molecule. So PEDOT-S is also water soluble and can be synthesized and processed from aqueous solution in by far bigger concentrations than PEDOT
OH nO
O
O
O
+ S
S
O
NaH
n
SO3Na
O
O
n = 1; 2
O S
Figure 12.22 EDOT-S, sodium salt synthesis.
6911X.indb 286
10/1/10 9:01:34 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
287
80 60 40 20
hs NT
Norm. W(log M)
100
0 103
104 Molecular Weight
105
106 g/mc
Figure 12.23 PEDOT-S, molecular weight distribution (GPC).
(>10%); the deep blue solutions yield transparent, sky-blue films after spin coating or doctor knifing, which exhibit a relatively high electric conductivity (but lower than that achievable with PEDOT:PSS). Other suitable oxidants for the EDOT-S polymerization are iron(III)-chloride49 or iron(III) tosylate.50 In sharp contrast to PEDOT:PSS, the good solubility of PEDOT-S facilitates a molecular weight estimation by gel permeation chromatography (GPC). In Figure 12.23, a typical GPC diagram from water solution and with refraction index detection against the poly(styrenesulfonic acid) standard is depicted. Relatively high values of Mn = 28,000 and Mw = 123,000 are obtained for a typical PEDOT-S, synthesized with iron(III) chloride oxidant. The material has a slightly bimodal molecular weight distribution, which can be reproduced.51 The reason may be that this “PEDOT-S“ in fact is a copolymerization product of EDOT-S (about 80%) and ProDOT-S (about 20%). Surprisingly, MALDI-TOF mass spectra for PEDOT-SLi and PEDOT-S, both electrochemically prepared, show a distribution of oligomers in accordance with GPC measurements for PEDOT-S, which give Mw = 2500 g/mol. PEDOT-S prepared with Fe(Tos)3 exhibits a slightly higher Mw = 5000 g/mol, which corresponds with a degree of polymerization of about 15. This is consistent with a MALDI-TOF MS exhibiting the maximum intensity at a higher molecular weight (at DP = 8) than for electrochemically prepared material.50 Perhaps EDOT-S is more sensitive to the action of different oxidants. This is supported by the fact that some experiments with peroxodisulfate as the oxidant also yielded oligomers51 similar to the products of Zotti et al.50 Zotti et al. reported a specific conductivity for as-deposited electrochemically synthesized PEDOT-S of about 10 S/cm.50 The ultraviolet-visible (UV-Vis) spectra show characteristics very similar to PEDOT. For example, the oxidized, highly conductive form exhibits the typical low tail into the infrared (IR) region (see Figure 12.24).
6911X.indb 287
10/1/10 9:01:35 PM
288
PEDOT
0.5
Absorbance [a.u.]
0.4 0.3 0.2 0.1 0.0
400
500
600
700
800
900
800
900
Wavelength [nm] (a) 1.2
Absorbance [a.u.]
1.0 0.8 0.6 0.4 0.2 0.0
400
500
600
700
Wavelength [nm] (b) Figure 12.24 UV-Vis spectra of (a) PEDOT-SLi and(b) PEDOT-S; de-doped (____) and oxidized (- - -) films. (From G. Zotti, S. Zecchin, G. Schiavon, and L. Groenendaal, Electrochemical and Chemical Synthesis and Characterization of Sulfonated Poly(3,4-ethylenedioxythiophene): A Novel Water-Soluble and Highly Conductive Conjugated Oligomer, Macromol Chem Phys, 2002, 203(13):1958–1964. Copyright Wiley-VCH Verlag GmBH & Co. KGaA. Reproduced with permission.)
6911X.indb 288
10/1/10 9:01:36 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
289
One has to keep in mind that dedoped does not have the meaning “fully undoped” for PEDOT-S. Complete dedoping is extremely difficult for PEDOT moieties. A truly undoped material would exhibit a remarkable red shift, compared to the spectra (solid line) in Figure 12.24, especially with respect to the oligomeric nature of the material described in Zotti et al.50 On the other hand, α-protonation of oligo-EDOTs or PEDOT results in a pronounced blue shift (for a detailed discussion of these facts, see Chapter 5). Therefore the absorption maxima shown in Figure 12.14 (solid line) should be associated with slightly doped and α-protonated oligomers. The electropolymerization of 1:1 mixtures (mol/mol) of EDOT and EDOT-S is performed with less experimental challenges due to the lower solubility and fouling tendency of the deposited films, compared to those of the PEDOT-S homopolymer.48,50,52 Moreover, the 1:1 copolymer has been reported as a cation exchange material by Chevrot’s groups.48,52 The anodic polymerization of the EDOT–EDOT-S mixture yielded polymer films having permanent cation-exchange properties. The authors confirmed the ability to extract cations with the examples of hexammine ruthenium(III) [Ru(NH3)63+] and uranyl (UO22+) cations.48 A detailed investigation of PEDOT-S has been performed by Reynolds and coworkers.53,54 PEDOT-S—both chemically and electrochemically prepared— was characterized and studied regarding its electrochromic and its hole transport properties. The electrochromic contrast was found to be lower than in many other electrochromic polymers, but rapid switching could be performed.
References
6911X.indb 289
1. M. L. Blohm, J. E. Pickett, and P. C. VanDort. US Patent 5,111,327 (General Electric Co.). Prior: March 4, 1991. 2. S. C. Ng, H. S. O. Chan, and W.-L. Yu. 1997. Synthesis and characterization of electrically conducting copolymers of ethylenedioxythiophene and 1,3propylenedioxythiophene with ω-functional substituents. J Mater Sci Lett 16:809–811. 3. D. M. Welsh, A. Kumar, and J. R. Reynolds. 1997. Easily functionalized 3,4ethylenedioxythiophene. Polymer Preprints 38(2):320. 4. P. Schottland, O. Stéphan, P. Y. Le Gall, and C. Chevrot. 1998. Synthesis and polymerization of new monomers derived from 3,4-ethylenedioxythiophene. J Chim Phys 95:1258–1261. 5. D. Gaiser, K. Reuter, and S. Weigand. 2001. Unpublished results. 6. K. Reuter. EP 1 352 918 (H.C. Starck GmbH). Prior: April 10, 2002. 7. S. Yeisley, C. J. Dubois, Jr., C.-H. Hsu, S. W. Shuey, Y. Shen, and H. Skulason. WO 2006/073968 (E.I. Du Pont de Nemours and Co.). Prior: December 30, 2004. 8. U. Merker, K. Reuter, and K. Lerch. EP 1 391 474 (Bayer AG), Prior: August 16, 2002.
10/1/10 9:01:37 PM
290
6911X.indb 290
PEDOT
9. Y.-H. Ha, N. Nikolov, S. K. Pollack, J. Mastrangelo, B. D. Martin, and R. Shashidar. 2004. Towards a transparent, highly conductive poly(3,4-ethylenedioxythiophene). Adv Funct Mater 14(6):615–622. 10. K. Reuter. 2000. Unpublished results. 11. A. Lima, P. Schottland, S. Sadki, and C. Chevrot. 1998. Electropolymerization of 3,4-ethylenedioxythiophene and 3,4-ethylenedioxythiophene methanol in the presence of dodecylbenzenesulfonate. Synth Met 93(1):33–41. 12. Y. Xiao, X. Cui, J. M. Hancock, M. Bouguettaya, J. R. Reynolds, and D. C. Martin. 2004. Electrochemical polymerization of poly(hydroxymethylated-3,4-ethylenedioxythiophene) (PEDOT-MeOH) on multichannel neural probes. Sens Actuators B 99(2–3):437–443. 13. B. Sankaran and J. R. Reynolds. 1997. High-contrast electrochromic polymers from alkyl-derivatized poly(3,4-ethylenedioxythiophenes). Macromolecules 30(9): 2582–2588. 14. B. Groenendaal. EP 1 321 483 (Agfa-Gevaert), Prior: December 20, 2001. 15. I. F. Perepichka, M. Besbes, E. Levillain, M. Sallé, and J. Roncali. 2002. Hydrophilic oligo(oxyethylene)-derivatized poly(3,4-ethylenedioxythiophenes): Cation-res ponsive optoelectrochemical properties and solid-state chromism. Chem Mater 14(1):449–457. 16. M. Besbes, G. Rippé, E. Levillain, M. Mazari, F. Le Derf, I. F. Perepichka, A. Derdour, A. Gorgues, M. Sallé, and J. Roncali. 2001. Rapid and efficient postpolymerization functionalization of poly(3,4-ethylenedioxythiophene) (PEDOT) derivatives on an electrode surface. Adv Mater 13(16):1249–1252. 17. H. Brisset, A.-E. Navarro, C. Moustrou, I. F. Perepichka, and J Roncali. 2004. Electrogenerated conjugated polymers incorporating a ferrocene-derivatized(3,4-ethylenedioxythiophene). Electrochem Commun 6(3):249–253. 18. K. Krishnamoorthy, M. Kanungo, A. Q. Contractor, and A. Kumar. 2001. Electrochromic polymer based on a rigid cyanobiphenyl substituted 3,4ethylenedioxythiophene. Synth Met 124(2–3):471–475. 19. K. Reuter, A. Karbach, H. Ritter, and N. Wrubbel. EP 1 440 974 (Bayer Chemicals AG). Prior: January 21, 2003. 20. N. Wrubbel, H. Ritter, K. Reuter, A. Karbach, and D. Drechsler. 2006. Annealing effects on the electrical conductivity of polymerized liquid crystalline 3,4ethylenedioxythiophene derivatives. e-Polymers 2; http://www.e-polymers. org/papers/ritter_030206.pdf. 21. I. Schwendeman, C. L. Gaupp, J. M. Hancock, L. Groenendaal, and J. R. Reynolds. 2003. Perfluoroalkanoate-substituted PEDOT for electrochromic device applications. Adv Funct Mater 13(7):541–547. 22. N. C. Tansil, H.-H. Yu, E. A. B. Kantchev, and J. Y. Ying. 2007. Naphthalene tetracarboxylic diimide-grafted poly(3,4-ethylenedioxythiophene)s. Polym Prepr 48(1):74–75. 23. H.-H. Yu, J.-Y. Ying, N. C. Tansil, and E. A. B. Assen. WO 2008057054 (Agency for Science, Technology and Research, Singapore), Prior: November 10, 2006. 24. H.-B. Bu, G. Götz, E. Reinold, A. Vogt, S. Schmid, R. Blanco, J. L. Segura, and P. Bäuerle. 2008. “Click”-functionalization of conducting poly(3,4-ethylendioxythiophene) (PEDOT). Chem Commun 2008(11):1320–1322. 25. A. E. Daudegaard, S. Hvilsted, T. S. Hansen, and N. B. Larsen. 2008. Conductive polymer functionalization by click chemistry. Macromolecules 41(12):4321–4327.
10/1/10 9:01:37 PM
EDOT and PEDOT Derivatives with Covalently Attached Side Groups
291
26. T. S. Hansen, A. E. Daudegaard, S. Hvilsted, and N. B. Larsen. 2009. Spatially selective functionalization of conducting polymers by “electroclick” chemistry. Adv Mater 21(44):4483–4486. 27. J. Xu, Y. Tian, R. Peng, Y. Xian, Q. Ran, and L. Jin. 2009. Ferrocene clicked poly(3,4-ethylenedioxythiophene) conducting polymer: Characterization, electrochemical and electrochromic properties. Electrochem Commun 11(10):1972–1975. 28. S. J. Higgins and F. Mouffok. WO 2006/018643 (The University of Liverpool), Prior: March 18, 2004. 29. J. L. Segura, R. Gómez, E. Reinold, and P. Bäuerle. 2005. Synthesis and electropolymerization of a perylenebisimide-functionalized 3,4-ethylenedioxythiophene (EDOT) derivative. Org Lett 7(12): 2345–2348. 30. R. B. Bazaco, R. Gómez, C. Seoane, P. Bäuerle, and J. L. Segura. 2009. Specific recognition of a nucleobase-functionalized poly(3,4-ethylenedioxythiophene) (PEDOT) in aqueous media. Tetrahedron Lett 50(28):4154–4157. 31. J. L. Segura, R. Gómez, R. Blanco, E. Reinold, and P. Bäuerle. 2006. Synthesis and electronic properties of anthraquinone-, tetracyanoanthraquinodimethane-, and perylenetetracarboxylic diimide-functionalized poly(3,4-ethylenedioxythiophene). Chem Mater 18(12):2834–2847. 32. M. Döbbelin, C. Pozo-Gonzalo, R. Marcilla, R. Blanco, J. L. Segura, J. A. Pomposom and D. Mecerreyes. 2009. Electrochemical synthesis of PEDOT derivatives bearing imidazolium-ionic liquid moieties. J Polym Sci A: Polym Chem 47(12):3010–3021. 33. M. Döbbelin, R. Tena-Zaera, R. Marcilla, J. S. Moya, J. A. Pomposo, and Mecerreyes. 2009. Multiresponsive PEDOT-ionic liquid materials for the design of surfaces with switchable wettability. Adv Funct Mater 19(20):3326–3333. 34. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 35. P.-H. Aubert, L. Groenendaal, F. Louwet, L. Lutsen, D. Vanderzande, and G. Zotti. 2002. In situ conductivity measurements on polyethylenedioxythiophene derivatives with different counter ions. Synth Met 122(2):425–429. 36. L. Groenendaal, G. Zotti, and F. Jonas. 2001. Optical, conductive and magnetic properties of electrochemically prepared alkylated poly(3,4-ethylenedioxythiophene)s. Synth Met 118(1–3):105–109. 37. B. Sankaran and J. R. Reynolds. 1995. Synthesis and electrochemistry of polydioxy-ethylenethiophene and its alkylsubstituted derivatives. Polym Mater Sci Eng 72:319–320. 38. J. R. Reynolds, G. A. Sotzing, B. Sankaran, S. A. Sapp, D. J. Irvin, J. A. Irvin, and J. L. Reddinger. 1996. Electrochromic polymers and devices via electropolymerized low potential monomers. Polymer Preprints 37(1):135. 39. J. R. Reynolds, B. Sankaran, G. A. Sotzing, D. J. Irvin, J. A. Irvin, J. L. Reddinger and S. A. Sapp. 1996. Electrochromic and redox electroactive polymers based on ethylenedioxy-thiophene derivatives. Mat Res Soc Symp Proc 413:373–376. 40. J. R. Reynolds, A. Kumar, J. L. Reddinger, B. Sankaran, S. A. Sapp and G. A. Sotzing. 1997. Unique variable-gap polyheterocycles for high-contrast dual polymer electrochromic devices. Synth Met 85:1295–1208. 41. A. Kumar, D. M. Welsh, M. C. Morvant, F. Piroux, K. A. Abboud, and J. R. Reynolds. 1998. Conducting poly(3,4-alkylenedioxythiophene) derivatives as fast electrochromics with high contrast ratios. Chem Mater 10(3):896–902.
6911X.indb 291
10/1/10 9:01:37 PM
292
PEDOT
42. S. A. Sapp, G. A. Sotzing and J. R. Reynolds. 1998. High contrast ratio and fastswitching dual polymer electrochromic devices. Chem Mater 10(8):2101–2108. 43. D. M. Welsh, A. Kumar, M. C. Morvant, and J. R. Reynolds. 1999. Fast electrochromic polymers based on new poly(3,4-alkylenedioxythiophene) derivatives. Synth Met 102:967–968. 44. M. Dietrich, J. Heinze, G. Heywang, and F. Jonas. 1994. Electrochemical and spectroscopic characterization of polyalkylenedioxythiophenes. J Electroanal Chem 369:87–92. 45. Q. Pei, G. Zuccarello, M. Ahlskog, and O. Inganäs. 1994. Electrochromic and highly stable poly(3,4-ethylenedioxythiophene) switches between opaque blueblack and transparent sky blue. Polymer 35(7):1347–1351. 46. J. C. Gustafsson, B. Liedberg, and O. Inganäs. 1994. In situ spectroscopic investigations of electrochromism and ion transport in a poly(3,4-ethylenedioxythiophene) electrode in a solid state electrochemical cell. Solid State Ionics 69:145–152. 47. K. Krishnamoorthy, M. Kanungo, A. V. Ambade, A. Q. Contractor, and A. Kumar. 2002. Electrochemically polymerized electroactive poly(3,4-ethylenedioxythiophene) containing covalently bound dopant ions: poly[2-(3-sodiumsulfinopro pyl)-2,3-dihydrothieno[3,4-b][1,4]dioxin. Synth Met 125:441–444. 48. O. Stéphan, P. Schottland, P.-Y. Le Gall, C. Chevrot, C. Mariet, and M. Carrier. 1998. Electrochemical behaviour of 3,4-ethylenedioxythiophene functionalized by a sulphonate group. Application to the preparation of poly(3,4-ethylenedioxythiophene) having permanent cation-exchange properties. J Electroanal Chem 443(2):217–226. 49. L. Groenendaal, F. Jonas, T. Cloots, and F. Louwet. EP 1 122 274 (Bayer AG). Prior: March 3, 2000. 50. G. Zotti, S. Zecchin, G. Schiavon, and L. Groenendaal. 2002. Electrochemical and chemical synthesis and characterization of sulfonated poly(3,4-ethylenedioxythiophene): A novel water-soluble and highly conductive conjugated oligomer. Macromol Chem Phys 203(13):1958–1964. 51. K. Reuter and H. Locke. 2005. Unpublished results. 52. O. Stéphan, P. Schottland, P.-Y. Le Gall, and C. Chevrot. 1998. New cationexchange material based on a sulfonated 3,4-ethylenedioxythiophene monomer. J Chim Phys 95(6):1168–1171. 53. C. A. Cutler, M. Bouguettaya, and R. Reynolds. 2002. PEDOT polyelectrolyte based electrochromic films via electrostatic adsorption. Adv Mater 14(9): 884–688. 54. C. A. Cutler, M. Bouguettaya, T.-S. Kang, and J. R. Reynolds. 2005. Alko xysulfonate-functionalized PEDOT polyelectrolyte multilayer films: Electro chromic and hole transport materials. Macromolecules 38(8):3068–3074.
6911X.indb 292
10/1/10 9:01:37 PM
13 XDOTs, EDXTs, EDOXs, and 2(5)X(2)-EDOTs: Ring Size Variations, Heteroanalogs, and Derivatives of EDOT with Substituents at the Thiophene Ring
13.1 3,4-Methylenedioxythiophene (MDOT) MDOT derivatives have been known since the pioneering work of Dallacker and Mues from the year 1975.1,2 The synthesis of MDOT-dicarboxylic acid follows the versatile route also useful for 3,4-ethylenedioxythiophene (EDOT), using bromochloromethane instead of 1,2-dihalogenoethanes (Figure 13.1). The decarboxylation step to MDOT could not be performed by the authors. The reasons remained unknown, but typically in more recent publications MDOT synthesis looked problematic, too. Whereas Ahonen et al. described the electropolymerization of MDOT,3 the synthesis of the unsubstituted parent compound had not been disclosed and not published until Lomas and coworkers studied the syn–anti rotamerization of sterically hindered 2-thienyldialkyl-methanols (for example, Figure 13.2 with formulae I, II, and III).4 Both R together in I could also represent an α, ω-alkandiyl group. So, thiophenes with n = 1 – 5 in formula III, Figure 13.2, were studied. As pointed out by Lomas et al., MDOT was expected to be an extraordinarily useful monomer for conductive polythiophenes. Logically, by simple linear continuation of the behavior known from EDOT, 3,4-propylenedioxythiophene (ProDOT), and 3,4-butylenedioxythiophene (BuDOT), the polymer from MDOT should exhibit better electric properties than PEDOT. In other words, “EDOT is situated on a continuum between 3,4-methylenedioxythiophene and other 3,4-alkylenedioxy-, 3-alkoxy- and 3,4-dialkoxythiophenes.”4 But the optimum end of this continuum is not MDOT; it is EDOT. The aforementioned quote by Lomas is based on the rotameric behavior of the thienyl methanol derivatives depicted in Figure 13.2.4 This is not unreasonable as an outcome of the steric situation. The resulting hydrogen bond strength,
293
6911X.indb 293
10/1/10 9:01:37 PM
294
PEDOT
O
OH
HO O EtO
O S
O
+ CH2BrCl
S
O
O O
EtO
OEt
O (OH–)
OEt
O
O HO
S
OH
Figure 13.1 3,4-Methylenedioxythiophene-2,5-dicarboxylic acid synthesis. (Data from F. Dallacker and V. Mues, 1975, Chem Ber 108:569–575.)
lowest in MDOT derivatives due to the longer O–H distance, clearly suggests low steric hindrance between the neighbored monomer moieties in MDOT polymers. In this context, the real chemical behavior of MDOT and PMDOT was very disappointing. The contrast to poly(3,4-ethylenedioxythiophene), or PEDOT, is drastic in every possible direction. First results were published by Ahonen et al.3 Their rather preliminary observations regarding the electrochemical polymerization of MDOT did not give rise for optimism. Compared to electrochemically prepared PEDOT, a higher band gap of 2.0 eV was estimated from the spectrum of the neutral film, and the anodic peak maximum, which is at higher potentials, also proved that PMDOT has shorter conjugation length than PEDOT. The interpretation of Ahonen’s results is nevertheless difficult. We were not able to prepare any doped PMDOT chemically in our laboratory. The MDOT withstands all possible oxidants we ever tried. Using the techniques applicable for EDOT—for example, in situ polymerization by iron(III)-tosylate or persulfate oxidation in aqueous medium—no polymerization occurs. Characteristically, and in sharp contrast to EDOT, MDOT may be nitrated to obtain 2,5-dinitroMDOT (Figure 13.3) in good yield and without serious decomposition.5 MDOT may be lithiated by butyllithium forming 2-Li-MDOT. The 2-lithiated MDOT easily is oxidatively coupled by copper(II)-chloride to bis-MDOT (Figure 13.4).6 Bis-MDOT is more reactive than the monomer—at least this is an analogy to EDOT, where the same relation between EDOT and bis-EDOT is found. So
R
R
R
O
O
H
O
t-Bu I
t-Bu
O t-Bu t-Bu
O S
(CH2)n
R O
S
O t-Bu t-Bu
S
OH II
OH III
Figure 13.2 Sterically hindered 2-thienyldialkyl-methanols, including MDOT derivatives.
6911X.indb 294
10/1/10 9:01:38 PM
295
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
O O2N
O NO2
S
Figure 13.3 2,5-Dinitro-3,4-methylenedioxythiophene.
O O
O
O
O
+ BuLi –C4H10
S
S
O S
(CuCl2) S
Li
O
O
Figure 13.4 Synthesis of bis-MDOT.
this intermediate could be polymerized via in situ polymerization with iron(III) tosylate to a brownish-red oligomer layer, which was obviously undoped and not conductive. Via organometallic syntheses, undoped oligomers like 2,5′′′dihexylquater-MDOT are accessible from bis-MDOT and 2-hexyl-MDOT.6
13.2 ProDOT (Propylenedioxythiophene) Derivatives Compared to EDOT, ProDOT (see Figure 13.5) is a molecule with rather similar properties but without exciting features. ProDOT was first described in the literature as early as EDOT and a considerably lower conductivity of its
CH3
H3C CH3 CH3
O
O
O
O
S
S
ProDOT
ProDOT-Me
O
O
S ProDOT-Me'
O
O
S ProDOT-Me2
Figure 13.5 3,4-Propylenedioxythiophenes.
6911X.indb 295
10/1/10 9:01:39 PM
296
PEDOT
homopolymer was observed in those first experiments.7 This observation had to be revised to some extent in later experiments. Electrochemically deposited poly(propylenedioxythiophene) (PProDOT) exhibited a conductivity of around 10 S/cm, characterized as the same range of PEDOT conductivity by comparison experiments.8 But in the light of almost all publications concerning chemically polymerized EDOT, these values seem to be untypically low, resulting from the special reaction conditions. PEDOT conductivity normally equals one to two orders of magnitude higher values. But indeed tentative experiments regarding chemical in situ polymerization with iron(III)-tosylate yield conducting films with surface resistances near those of in situ PEDOT or slightly lower. Specific conductivities for PProDOT can roughly be estimated by these experiments and achieve 50% to 100 % of in situ PEDOT values.9,10 In spite of this, pronounced handling difficulties together with several other properties of ProDOT are reducing its practical value and prevented technical use so far. In contrast to EDOT, ProDOT is a solid at room temperature (melting point 79°C–83°C). Therefore the use of a pure liquid or of highly concentrated solutions, pivotal for technical in situ polymerization procedures like electrolyte capacitor manufacture, is prohibited. Additionally, due to the insolubility of ProDOT in water, a PProDOT:PSS complex cannot be prepared. Nevertheless, ProDOT has attained practical attention because of another typical property. The slightly enhanced bulkiness of the anellated dioxepane ring, compared to the dioxane ring in EDOT, obviously is responsible for an improved electrochromic behavior of ProDOT. The electrochromic contrast ΔT—defined as difference between percent transmittance for the reduced opaque, deep blue state and the highly transmissive, light blue oxidized (and conducting) state—is clearly dependent on the steric demand of the polythiophene side group. This has been demonstrated by systematic investigations of the Reynolds group with electrochemically deposited alkyl-substituted PEDOTs.8 PProDOT, also part of this study, showed an enhanced electrochromic contrast of ΔT = 54%, compared to 44% for PEDOT. Consequently, further substitution in ProDOT-Me improved the contrast to 77%.11 Dimethyl-3,4-dipropoxythiophene (ProDOT-Me2) has further increased properties in several respects, but an alternative synthetic strategy had been necessary compared to ProDOT. The synthesis of ProDOT-Me2 was first published in 1999.11 In the same publication, ProDOT synthesis was performed in the manner described before by Williamson ether synthesis with 1,3- dibromopropane and 3,4-dihydroxythiopene-2,5-diethylcarboxylate and subse quent ester hydrolysis and decarboxylation.12 For the preparation of the new ProDOT-Me, in principle the same procedure has been used, utilizing 2-methylpropane-1,3-ditosylate instead of the 1,3-dibromide.11 In contrast to the less substituted ProDOT or ProDOT-Me, Williamson ether synthesis with neopentylbromide or similar compounds was not sufficient for ProDOT-Me2—not unexpected for a neopentyl structure. So acid catalyzed
6911X.indb 296
10/1/10 9:01:40 PM
297
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
H3C CH3 CH3
OCH3
CH3O
+
HO
OH
(H+)
O
O
CH3
S
S Figure 13.6 Synthesis of ProDOT-Me2.
transetherification was the synthetic route of choice, starting with 3,4-dimethoxythiophene.11 Higher dialkoxythiophenes like 3,4-diethoxy- and 3,4-din-propoxythiophene, which exhibit improved handling and storage stability, also can function as starting materials (Figure 13.6).13 The di-hydroxyfunctional compound, shown in Figure 13.7, can be typically observed as a by-product, especially when an excess neopentyl glycol is used to ensure the complete reaction of the most valuable educt dimethoxythiophene.13 Reynolds, Meijer, and coworkers reported several intriguing properties for electrochemically prepared PProDOT-Me2, when used as an electrochromic layer. They compared the switching behavior between the nearly transparent, fully oxidized state, and the deeply colored, reduced state of PProDOT, PProDOT-Me, and PProDOT-Me2. One positive effect observed for PProDOTMe2 is a significant enhancement of the switching speed, which typically is one order of magnitude better than for polyalkylthiophenes, for example, poly(3methylthiophene), and about three to seven times higher than for PEDOT.11 Compared to PEDOT:PSS, electropolymerized layers of PEDOT, or the less substituted analog PProDOT,8 the transmittance change between the doped
OH
OH
O
O
S Figure 13.7 Byproduct of ProDOT-Me2 synthesis.
6911X.indb 297
10/1/10 9:01:41 PM
298
PEDOT
Table 13.1 Relative Conductivity of PProDOT-Me2, Compared to Similar Polythiophenes PEDOT PProDOT PProDOT-Me‘ PProDOT-Me2
Reference 10
Reference 9
600 300
700 350 180 1
1
and reduced state is substantially enhanced to about ΔT = 78%. The color of the reduced state is shifted to purple instead of deep blue (λmax = 578 nm). Obviously, the methyl groups have a strong steric influence on the relative order of the polythiophene moieties in the conductive layer. This is largely supported by the drastic conductivity decrease observed for PProDOTMe2, compared to the less substituted analogs or to PEDOT. Zotti’s special in situ method for electropolymerization and conductivity measurement on platinum electrodes14 found a remarkably low conductivity for PProDOTMe2.10 Under the special electrochemical conditions of the Zotti measurements at CNR in Padova the following specific conductivities relation of PEDOT:PProDOT:PProDOT-Me2 was found—600:300:1. These values approximately equal siemens per centimeter and are included in Table 13.1.10 Similar correlations were found in tentative in situ polymerization experiments on glass.9 The values in Table 13.19 are given in arbitrary units (PProDOT = 1) extracted from surface resistances, in satisfactory accordance to the results from the electrochemical experiments. ProDOT-Me′ is the isomer of ProDOT-Me depicted in Figure 13.5. The conductivity of more than two orders of magnitude lower compared to PProDOT (and, of course, compared to PEDOT) must have steric reasons. It is probably a result of the morphology of this polymer; the electrochemical in situ experiments revealed that a dendritic structure was observed, clearly different from that of the other EDOT- and ProDOT-derived polymers with a cloud-shaped appearance.10 Functionalized derivatives of ProDOT and ProDOT-Me2 were synthesized by Sommerdijk and coworkers with the objective to prepare biosensors.15 They also used the transetherification of 3,4-dimethoxythiophene with an excess of various diols, exhibiting a neopentyl-based structure (see Figure 13.8). These ProDOT derivatives were polymerized and also copolymerized with EDOT electrochemically. A 1:4 copolymer of ProDOT-Me(CH2OH)–R1 = CH2OH, R2 = CH3 in Figure 13.8—with EDOT was grown on a platinum electrode and reacted with a solution of 4,4′-diisothiocyanatostilbene-2, 2′- disulfonic acid. Ferrocene-modified glucose oxidase was then coupled via its free
6911X.indb 298
10/1/10 9:01:41 PM
299
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
R1 R2 R1
OCH3
CH3O
+
HO
OH
R1 = CH2OH, R2 = CH3 +)
(H
O
R1 = CH2OH, R2 = H
O
R2
S
R1 = CH2OBz, R2 = H
S
Figure 13.8 Derivatized ProDOT derivatives suggested for glucose sensors. (Data from A. Kros, R. J. M. Nolte, and N. A. J. M. Sommerdijk, 2001, J Polym Sci A: Polym Chem 40(6):738–747.)
lysine amine groups to the surface. The obtained materials were shown to be able to detect glucose amperometrically. The same could be realized with in situ polymerized EDOT (see corresponding Chapter 8).16
13.3 Vinylenedioxythiophene (VDOT) and Benzo-EDOT The long-sought VDOT (Figure 13.9) was synthesized by Roncali’s group using an intelligent approach with an olefin metathesis step.17 This was shown in detail in Chapter 5. At first glance this compound looks very similar to EDOT, but the electrochemical behavior completely differs. The six-membered ring is planarized by the steric influence of the olefinic double bond. This results in a very similar behavior to MDOT, where the annelated dioxolane ring also is planar and in the same plane as the thiophene ring. Like MDOT, all attempts to electropolymerize VDOT remained unsuccessful.17 The “dimeric” bisVDOT can be synthesized by oxidative coupling of the 2-lithiated VDOT with CuCl2.17 Bis-VDOT (BVDOT) is more reactive than VDOT and can be
O
O
O
O S
S
S O
O
Figure 13.9 3,4-Vinylenedioxythiophene (VDOT) and bis-VDOT.
6911X.indb 299
10/1/10 9:01:42 PM
300
PEDOT
O X
Benzo-EDOT: X = H
O
S
2, 5-Dibromobenzo-EDOT: X = Br X
Figure 13.10 Benzo-EDOTs (PheDOTs).
electropolymerized; the CV suggests a considerably higher stability of the neutral state of poly-BVDOT (= PVDOT) compared to neutral, undoped PEDOT. Benzo-EDOT (Figure 13.10) is synthesized via transetherification of lower 3,4-dialkoxythiophenes with catechol.18,19 Benzo-EDOT (PheDOT) can be in situ polymerized chemically with iron(III) tosylate to polymeric (perhaps oligomeric) films with rather low conductivity.18 In this molecule two effects are leading into the same direction, with the low conductivity of the poly(benzo-EDOT) (PPheDOT) as the consequence. First, the VDOTanalogue planarization of the dioxane-derived ring turns the oxygen orbitals out of the optimum direction for the mesomeric stabilization of cationic, bipolaronic structures. The same reason is hindering further delocalization into the benzene ring. Additionally, it can be assumed that the effective conjugation length of the polymer is decreased by the mutual steric influence of benzene rings of every second monomer moiety, which are directed to the same side of the polymer (oligomer) chain and so are hindering planar configuration. In contrast to the chemical polymerization, PheDOT could not be electropolymerized.20 The deviating behavior of benzo-EDOT, compared to EDOT, is paralleled by the low reactivity of 2,5-dibromobenzo-EDOT (see formula in Figure 13.10; prepared from benzo-EDOT and N-bromo succinimide). In considerable contrast to 2,5-di-bromo-EDOT, which slowly but continuously decomposes to PEDOT and bromine, the 2,5-dibromobenzo-EDOT has far better stability.20 It can be stored for years in a refrigerator without discoloration21 and also withstands storage above 100°C for one month.20 2,5-Dibromobenzo-EDOT is a useful intermediate in the preparation of “trimeric” PheDOT. The analog dimeric molecule di-PheDOT is also accessible from PheDOT.20 The synthetic routes to the oligomeric PheDOT derivatives are summarized in Figure 13.11.20 These oligomers can, in spite of their low solubility and in contrast to the monomeric PheDOT, be readily electropolymerized.20 Theoretical calculations show a low reactivity of the PheDOT+ cation radical. The extension of the conjugated chain leads to a distribution of the singly occupied molecular orbital (SOMO) more similar to that of EDOT oligomers.
6911X.indb 300
10/1/10 9:01:43 PM
301
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
O
BuLi
O
O
S
O Li
S ZnCl2
NBS
O
Br
O
O
O
Br
S
S
CuCl2
O
S
O
S ZnCl
O
O
[Pd(Ph3)4]
O
O
O
O
S S
S O
O
Figure 13.11 Oligomers from benzo-EDOT (PheDOT). (Data from I. F. Perepichka, S. Roquet, P. Leriche, J.-M. Raimundo, P. Frère, and J. Roncali, 2006, Chem Eur J 12(11):2960–2966.)
13.4 3,4-Ethyleneoxythiathiophene (EOTT) EOTT, or, denoted by its IUPAC name, thieno-[3,4-b]-1,4-oxathiane, is easily accessible via acid-catalyzed transetherification from 3,4-dimethoxy-thiophene and 2-mercaptoethanol, as demonstrated by Blanchard, Roncali, and colleagues (Figure 13.12).22 A few derivatives were also described, demonstrating the chemical behavior of EOTT. Bromination by N-bromosuccinimide (NBS) results in a 85:15 isomer mixture of the two possible monobromination products (Figure 13.13). Unlike EDOT, where monobromination is only possible by using an EDOT excess over NBS,23 less than 3% of the dibromo derivatives are formed from
6911X.indb 301
10/1/10 9:01:43 PM
302
PEDOT
H3CO
OCH3 +
HO
O
(H+)
SH
S
(60%) S
S Figure 13.12 3,4-Ethylene-oxythiathiophene synthesis.
EOTT with stoichiometric amounts of NBS.22 Iodination of EOTT by I2 in the presence of mercury diacetate is even more regioselective, with an isomeric ratio of 97:3 (Figure 13.14).22 Formylation of EOTT—by monolithiation with n-butyllithium and quenching the Li-intermediate with N,N-dimethylformamide—yields the monoaldehyde (Figure 13.15). The ratio of 2- to 5-isomer is 85:15, the same as observed after bromination.22 The monomeric EOTT can easily be electropolymerized. In Bu4NPF6/ acetonitril an irreversible oxidation peak at 1.38 V (vs. AgCl/Ag) is observed. This value is intermediate between those for EDOT (1.5 V) and 3,4-ethylenedithiathiophene (EDTT; 1.32 V; see Figure 13.16), as expected. The cyclic voltammogram of the EOTT polymer in a monomer-free electrolytic medium exhibits a main anodic wave peaking at 0.4 V, compared to 0.2 V for PEDOT and 0.9 V for PEDTT. The use in capacitors has been claimed for EOTT in a patent application.24 The in situ polymerization with iron(III)-tosylate necessary for this application works without major problems. The achievable conductivity is, roughly estimated, in a similar range as for in situ PEDOT. In contrast to in situ polymerization, the oxidative polymerization in aqueous dispersion by peroxodisulfate is accompanied by the oxidation of the oxathian-sulfur to the sulfone moiety. The compound in Figure 13.17 is formed.25 This new EDOT analog cannot at all be polymerized to a conductive polythiophene. Steric hindrance together with the electron withdrawing effect of the sulfone group are obstacles against oxidative polymerization.
O
S
(NBS)
O
S
(80%) S
S
O +
S 15%
Br
Br
S 85%
Figure 13.13 Bromination of EOTT.
6911X.indb 302
10/1/10 9:01:44 PM
303
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
O
S
O
[I2/Hg(OAc)2]
S
S
O +
S
I
S
I
3%
S 97%
Figure 13.14 Iodination of EOTT.
O
S
S
O
(BuLi/DMF)
S
O +
(65%) S
CHO
S 15%
OHC
S 85%
Figure 13.15 Formylation of EOTT.
S
S
S Figure 13.16 3,4-Ethylene-dithiathiophene.
S
O
O O
S Figure 13.17 Sulfone from EOTT via S-oxidation.
6911X.indb 303
10/1/10 9:01:46 PM
304
PEDOT
13.5 3,4-Ethylenedithiathiophene (EDTT) EDTT was first prepared by Kanatzidis and colleagues in 1995.26 This compound has attracted less attention than EDOT. A rather inconvenient synthesis (see Figure 13.18), the lower conductivity and the opacity of the doped polymer PEDTT are drawbacks of this molecule, compared to EDOT and PEDOT, respectively. Nevertheless, several interesting details have been reported in the publica tion of Kanatzidis et al. EDTT can easily be polymerized by iron(III) chloride to PEDTT tetrachloroferrate. This type of doped PEDTT exhibited a specific conductivity of about 0.1 S/cm. The conductivity of electrochemically generated PEDTT perchlorate is only slightly higher (0.4 S/cm). These materials are not transparent in the doped state, which is a major drawback compared to PEDOT. The dark-green tetrachloroferrate can be reduced by hydrazine to the brown undoped material, which is completely soluble, for example in NMP. It can be redoped by FeCl3 or elemental iodine. Films of solution-cast neutral PEDTT were redoped by the action of iodine vapor. Soon after the publication of the spontaneous polymerization reaction of 2,5-dibromo EDOT to PEDOT by Wudl and colleagues,27,28 this kind of polythiophene formation was transferred to EDTT and similar compounds by Skabara and coworkers.29 Three dibromothiophenes—DBMDTT, DBEDTT and DBPTT—were prepared from thieno[3,4-d][1,3]dithiol-2-one (see Figure 13.19). All three pale yellow, crystalline dibromo compounds tend to darken during standing over several days. This process, indicating the polymerization to oligoor polythiophenes, could be accelerated by heating up to temperatures about 10°C below the melting point. In all cases, coupling of the thiophene moieties is accompanied by the concomitant release of elemental bromine, which dopes SLi
Br
Br
Br
BuLi/S8 S
S
SLi
LiS
CS2
BuLi/S8 S
S S
S
S Only isolated intermediate
NaOMe
NaS
SNa
S
BrCH2CH2Br
S
S
S EDTT
Figure 13.18 Ethylene dithiathiophene synthesis. (Data from C. Wang, J. L. Schindler, C. R. Kannewurf, and M. G. Kanatzidis, 1995, Chem Mater 7:58–68.)
6911X.indb 304
10/1/10 9:01:48 PM
305
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
O S
O S
S
S
Br2 Br
S
Br
Br
S R
Br
S
n = 1; R = CH2 DBMDTT
S
S
Br(CH2)nBr
SNa
NaS
NaOEt
n = 2; R = (CH2)2 DBEDTT
n = 1, 2, or 3
Br
Br
S
n = 3; R = (CH2)3 DBPDTT
Figure 13.19 Dibromo-3,4-alkylenedithiathiophene synthesis. (Data from H. J. Spencer, R. Berridge, D. J. Crouch, S. P. Wright, M. Giles, I. McCulloch, S. J. Coles, M. B. Hursthouse, and P. J. Skabara, 2003, J Mater Chem 13(9):2075–2077.)
the polymer, resulting in turning black. The doped materials could be dedoped by hydrazine to the corresponding neutral species as orange powders. Investigations by elemental analysis, MALDI-TOF MS, and absorption spectroscopy uncovered the formation of oligomers. Whereas for the dibromo3,4-propylenedithiathiophene (DBPDTT) only dimeric material could be identified, DBMDTT gave a DP of about 5 (elemental analysis), but nothing above a quaterthiophene was found by MALDI-TOF mass spectroscopy. Best results were obtained for DBEDTT, where MALDI-TOF MS clearly showed the presence of polymers containing up to 14 monomer units. Elemental analysis of dedoped PEDTT indicated an average of six repeat units per chain. Two more recent papers presented a deeper insight into the remarkable differences between PEDOT and PEDTT.27,28 Polymerizing the terthiophenes SOS and OSO (see Figure 13.20) electrochemically, the corresponding poly(terthiophenes) POSO and PSOS were
S
S
S
S
O
S
S S
S O
O
O
O
O
S
S S
S
EDTT-EDOT-EDTT
EDOT-EDTT-EDOT
SOS
OSO
Figure 13.20 Alternating terthiophenes from EDOT and EDTT. (Data from H. J. Spencer, P. J. Skabara, M. Giles, I. McCulloch, S. J. Coles, and M. B. Hursthouse, 2005, J Mater Chem 15(45):4783–4792.)
6911X.indb 305
10/1/10 9:01:49 PM
306
PEDOT
obtained.30 The well-defined copolymers of EDOT and EDTT were compared spectroelectrochemically with the homo-polymers PEDOT and PEDTT. These studies revealed strong similarities between PEDTT and PSOS, and between PEDOT and POSO. The results indicated the strong influence of conformational effects (intrachain S-O-contacts, stabilizing coplanar thiophene ring configuration) on the properties of the homo- and copolymers. The unique properties of PEDOT are discussed as the result of these conformational aspects, whereas the influence of substituent effects is considered less important in this study. These findings are supported by spectroelectrochemical investigations (e. g., different infrared active vibrational band patterns) as well as by semiempirical molecular orbital (AM 1) calculations.31
13.6 3,4-Ethylenedioxypyrrole (EDOP) and Its Derivatives EDOP, or, more generally spoken, poly(3, 4-dioxypyrrole)s (PXDOPs) and their monomers are not in the focus of this book actually. Nevertheless, considering all heterocyclic compounds with another heteroatom than S, EDOP should be that one with properties closest to EDOT. So a lot of efforts were made to establish EDOP as an interesting alternative to EDOT. From a scientific standpoint this has been successful, especially demonstrated in a considerable amount of papers, predominantly from the Reynolds group. But besides all experiments with valuable and interesting results, Reynolds and Walczak stated “PXDOPs as … yet underutilized electroactive and conducting polymers” in a comprehensive review.32 This has not changed until now (November 2009). The synthesis and properties of these compounds shall briefly be summarized in the following. The first synthesis of a 3,4-dialkoxypyrrole (3,4-dimethoxypyrrole) was published by Merz et al.33 The EDOT analog 3,4-ethylenedioxypyrrole (EDOP) soon followed,34 and its poly(styrenesulfonic acid) complex PEDOP:PSS analog to PEDOT:PSS was claimed in a patent 1996.35 The synthetic routes to the monomeric EDOP and similar compounds are roughly outlined in Figure 13.21, following the Reynolds and Walczak review, where more detailed comments can be found.32 The first route in Figure 13.21, based on the pioneering work of Merz et al.33,34 and largely similar to EDOT synthesis has been further developed by the Reynolds group.36–39 A modification without the need to functionalize and deprotect the pyrrole-N by starting with different residues R instead of Bz has been reported by Kim et al.40 The second route, beginning with 2, 5- dimethoxy3,4-dihydrofuran and its oxidation to the corresponding dihydroxyfuran with KMnO4, has been claimed to have some advantages over the first.41
6911X.indb 306
10/1/10 9:01:49 PM
6911X.indb 307
O
HO
O
O
O
Bz
+ N
OH
O
O
O
OEt
OR´
R´OOC
O
Bz
N
O
O
O O
COOR´
X
OH Br-X-Br
X = Dialkyl or Alkylene
R = Alkyl, Benzyl, Alkanesulfonyl, Alkoxy, Allyl
R´ = Me or Et
Br-X-Br
NaOEt
HO
HCl
Bz
N
X O COOR´
O
R
N
X O
NaOH/H+ TEA/Δ
R´OOC
O
RHal
Na/NH3
Pd/H2
Figure 13.21 Synthetic routes to XDOPs. (Data from R. M. Walczak and J. R. Reynolds, 2006, Adv Mater 18(9):1121–1131.)
R´O
EtO
H
N
X O COOR´
O
H
N
X
O
NaOH/H+ TEA/Δ
R´OOC
O
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs 307
10/1/10 9:01:49 PM
308
PEDOT
The PXDOPs are unique in combining low oxidation potential, as a result from the alkylenedioxy moiety and similar to PEDOT, with middle to high band gap resulting from the electron-rich pyrrole moiety. The property of PEDOP (and also of poly(3,4-dimethoxypyrrole), PDMOP) first and most of all from a practical point of view, the electrical conductivity, does not equal the PEDOT level. A detailed and comparative electrochemical study was presented by the Zotti group.42 Whereas Merz reports a conductivity of up to 200 S/cm for PDMOP in ex situ free-standing films,33,43 in Zotti et al., the maximum in situ conductivities for both PDMOP and PEDOP perchlorate are given as about only 15 S/cm.42 The unusual behavior of PDMOP films regarding the solvent influence on their conductivity, indicated by the remarkable drop with acetonitril in the in situ experiments and the reason for the discrepancy observed, has led to the construction of solvent vapor-sensing devices.42 Interestingly, Reynolds and Walczak presume, based on the studies of Zotti et al.,42 that Merz et al.43 should have been able to achieve even higher conductivities than reported by optimizing the dopant level.32 By stretching freestanding films, a 40% increase in conductivity could be achieved.43 N-substitution in PEDOP causes a dramatic decrease in specific conductivity: chemically polymerized PEDOP-N-dodecyl exhibited a conductivity of only 0.028 S/cm.40 The introduction of an N-alkyl group obviously leads to a more twisted polymer backbone and reduced conjugation length of the π-system. Additionally, efficient π-stacking is prevented, interrupting charge percolation. The multicolor anodic and cathodic electrochromism of PXDOPs is an intriguing property broadly reviewed by Reynolds and Walczak.32 A striking variety of colors can be achieved by varying the structure of the PXDOP system. Color swatches for different PXDOPs are shown by Reynolds and Walczak.32 After the designation of PXDOPs as underutilized materials by Reynolds and Walczak in 2006,32 only a limited amount of additional scientific papers regarding these conducting polymers seem to have been published up to now (November 2009).44–49 Most of them are dealing with fluorinated derivatives with oleophobic properties.46–49 An amperometric biosensor for cholesterol with cholesterol oxidase entrapped in PEDOP46 and dye-sensitized solar cells45 based on PEDOP are other objectives. But the statement of “underutilization” by Reynolds and Walczak32 has not lost its validity, and the reasons for this are also unchanged: (a) The synthesis of the most interesting candidate EDOP is laborious and not very efficient,32 so EDOP cannot be commercialized to a competitive price; (b) there seems to be no superior property pushing EDOP or PEDOP into potential applications in the market for conductive polymers. Conductivity remains significantly behind PEDOT, and electrochromism—one of the most intriguing properties of PXDOPs—is not a forefront application for these types of compounds. Even with the broadly and intensely investigated EDOT derivatives, no true technical application has been realized up to now (see Chapter 10).
6911X.indb 308
10/1/10 9:01:49 PM
309
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
13.7 3,4-Ethylenedioxyselenophene (EDOS) EDOS was first synthesized by Cava et al. using the analog synthetic pathway to EDOT but starting with diethyl selenodiglycolate (from sodium selenide and ethylchloroacetate).50 An overview is given in Figure 13.22. The more polarizable selenium atom lowers the oxidation potential of EDOS, compared to EDOT. The electrochemical behavior of EDOS, studied by cyclic voltammetry, is characterized by an irreversible oxidation peak around 1.2 V (1.18 V50 and 1.22 V51 are consistently reported values), which is significantly lower than for EDOT. Dark blue PEDOS is formed by electrochemical polymerization during repetitive cyclovoltammetry. The redox behavior of PEDOS was studied by cyclic voltammetry in the potential range –1 to +1 V; a sky-blue, transparent film of doped PEDOS could be obtained.50 The chemical oxidation of EDOS to PEDOS could be performed with iron(III) chloride in acetonitrile. Dark blue, soluble PEDOS with an absorption band at λmax = 594 nm was observed. From 1H-NMR spectroscopy a reasonably high molecular weight was concluded, because protons of selenophene moieties at about 6.82 ppm, as visible in the monomer, could not be found, indicating the lack of noteworthy amounts of end groups.50 These results were debated later.51 Obviously, a nondoped (or sparsely doped) material was obtained by Aqad, Lakshmikantham, and Cava.50
ONa
NaO EtOOC
Se
HO
COOEt
EtOOC
NaOEt
EtOOC
Se
COOEt
HOOC
EtOOC
Melt
O
Se
COOH
HCl
1) KOH 2) HCl
Br K2CO3
O
Se
COOEt
O
O
OH Br
EtOOC
COOEt
COOEt
Se
O
O
Se
Figure 13.22 3,4-Ethylenedioxyselenophene (EDOS) synthesis.
6911X.indb 309
10/1/10 9:01:50 PM
310
PEDOT
Se + SOCl2 OCH3
SeCl2 (NaOAc)
OCH3
OCH3
H3CO
HOCH2CH2OH
O
O
(p-TSA) Se
Se
Figure 13.23 Synthesis of EDOS via dimethoxyselenophene. (Data from A. Patra, Y. H. Wijsboom, S. S. Zade, M. Li, Y. Sheynin, G. Leitus, and M. Bendikov, 2008, J Am Chem Soc 130(21):6734–6736.)
Several years later, Bendikov* et al. reinvestigated 3,4-ethylenedioxyselenophene and could adapt the Hellberg route to EDOT (see Chapter 5) for EDOS via 3,4-dimethoxyselenophene, starting with 2,3-dimethoxy-1,3-butadiene and SeCl2 (see Figure 13.23).51,52 The synthesis of 2,5-dibromo-3,4-ethylenedioxyselenophene could easily performed by bromination with N-bromosuccinimide (as well as the iodo analog with N-iodosuccinimide). Very similar to the solid-state polymerization of 2,5-dibromo-EDOT (see Chapter 6), 2,5-dibromo-EDOS and 2,5-diiodo-EDOS also spontaneously polymerize to the bromine- or iodine-doped conductive polymers at elevated temperatures (for example, 50°C–80°C). Slow polymerization is observed at room temperature, with a smaller half-life for dibromoEDOS (21 days) than for dibromo-EDOT (70 days).53 Br3– and I3– can be supposed as the counterions, analogously to the dibromo-EDOT chemistry. Bendikov et al. were able for the first time to demonstrate the relatively high conductivity of PEDOS. The highest value of 30 S/cm in a pressed pellet was obtained from 2,5-diiodo-EDOS via solid-state polymerization. In contrast to Cava’s group,50 the authors were also able to polymerize EDOS to an unsoluble, doped PEDOS with the aid of an excess of FeCl3. The rather low conductivity of 0.002 S/cm was explained by defect formation during the course of the oxidative polymerization, due to chemical modification of the selenophene rings.51 The transition-metal-mediated synthesis of neutral, undoped PEDOS from 2,5-dibromo-EDOS with a nickel catalyst [Ni(COD)2, COD, 2,2’-bipyridyl); COD = Cycloocta-1,5-diene] was also reported by Patra et al.51 As traced by conductivity measurements, neutral, solid PEDOS is not as stable as the solid PEDOT analog, but is easily doped when exposed to air. MALDI-MS analysis of the neutral polymer (obtained from doped, highly conductive polymer by reduction with hydrazin) revealed weight fragments up to the 20-mer. The dithia-analog of EDOS, 3,4-ethylenedithioselenophene (EDTS) could be prepared by transetherification of 3,4-dimethoxyselenophene with 1, 2ethanedithiol.54 The homopolymer, PEDTS, was synthesized electrochemically and by solid-state oxidative polymerization. PEDTS has a significantly * Note: In the meantime, dimethoxytellurophene has also been synthesized by the Bendikov group: A. Patra, Y. H. Wijsboom, G. Leitus, and M. Bendikov, 2009, Synthesis, Structure, and Electropolymerization of 3,4-Dimethoxytellurophene: Comparison with Selenium Analogue, Org. Lett. 11 (7):1487–1490.
6911X.indb 310
10/1/10 9:01:51 PM
311
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
narrower band gap (0.6–0.8 eV) than PEDTT (see earlier) and a lower work function than PEDOT.54
13.8 2,5-Disubstituted EDOT Derivatives [2(,5)-X(2) -EDOTs] A vast amount of 2-substituted and 2,5-disubstituted EDOT derivatives have been synthesized. Several examples like simple derivatives elucidating the typical EDOT chemistry have been mentioned in Chapter 5. The following part of this chapter focuses on derivatives that have been prepared and investigated with the intention to create interesting electro-optical properties. A comprehensive review was published in 2005 by Roncali et al.55 Due to the great number of publications, only selected summaries will be presented here. A plain and, from a preparative point of view, very efficient concept has been introduced by Reynolds and his coworkers, starting in the middle of the 1990s, but following the same concept from earlier thiophene-related papers, see, for example, Reynolds et al.56 and Ruiz et al.57 In the fundamental paper from 1996, the symmetric EDOT derivatives shown in Figure 13.24 have been prepared via NiCl2.dppp-catalyzed Grignard coupling and studied regarding their electrochemical polymerization and the electrochromic properties of the polymers.58 The work has been extended to the analogous carbazole and fluorine derivatives, see also Figure 13.24.59–62 Later, this approach was applied also O
O
X
S
O
O
S
X=
O
S
N R R = CH3, C20H41 R
R
R = C10H21
Figure 13.24 Symmetric EDOT derivatives as described by Sotzing et al., Sapp et al., Reddinger et al., Reynolds et al., and Larmat et al.58–62
6911X.indb 311
10/1/10 9:01:51 PM
312
PEDOT
CN Br
S
1. BuLi 2. S8
NBS
3. Br(CH2)2CN
S
O
S
Br
S
CN
O
R
S
+ S
CN
S
S
SnBu3
(Pd-Kat)
S
S + R-CH2-X
S O
O
(CsOH)
S O
O
Figure 13.25 Synthesis of bithiophenic precursors with low oxidation potential.
to pyridines63 and bipyridines64 as well as to more complicated systems like tetrathiafulvalenes65 and N,N′-ethylene-bis-(salicylaldimine) cobalt com plexes.66,67 All these precursors with their median block between two EDOT groups were electropolymerizable yielding polymers with alternating functional groups and bis-EDOT moieties. The range of easily electropolymerizable, EDOT-derived functional monomers was expanded by bithiophenic molecules consisting of one EDOT and one thiophene moiety. This combination results in electropolymerization potentials around 1.15 V, which is significantly lower than for other substituted EDOT monomers. With the aid of this concept sensitive poly(thiophenes) like tetrathiafulvalene derivatives could be electrosynthesized.68,69 A new synthetic concept for the functionalization of thiophene-EDOT bithiophenic structures with sulfide groups was also elaborated by Roncali’s group.70,71 This concept is outlined in Figure 13.25. The sulfide group further reduces the oxidation potential of the bithiophene precursor, and the cyanoethyl group can be utilized for further derivatization. Electropolymerization at potentials of 0.9 V or lower was achieved with these “EDOT-T precursors.” Bipyridines were synthesized by thiolate deprotection with cesium hydroxide and reaction with halogenides R-CH2-Br, performing the last step in Figure 13.25 with 5,5′-bis(bromomethyl)-2,2′-bipyridine and 4,4′-bis(bromobutyl)-2,2′bipyridine.71 The metal complexes depicted in Figure 13.26 were synthesized with [Ru(bipy)2Cl2], RuCl3, and FeSO4, respectively.71 The potentiodynamically electropolymerized metal complexes exhibit both the typical behavior of the polythiophene structure and the metal complex.71 The same synthetic strategy described by Figure 13.25 was used to prepare the fullerene derivatives shown in Figure 13.27, which are potentially useful for photovoltaic energy conversion.72
6911X.indb 312
10/1/10 9:01:52 PM
313
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
2+ O
O
O
O
O
O
O
S
S
S
S
S
S
S
S ( )3
N
( )3
( )3
N
N Ru
N
S
S
( )3
N
S
S – 2 PF6
R
N
N
N Ru
Cl
N
R
R
R
N
Cl
O
R=
2+
O
O
S N R
N
Fe N
R
N
S
–
R
N N
2 PF6
S ( )3
R
Figure 13.26 Iron and ruthenium complexes from bithiophenes with EDOT-T structure. (Data from B. Jousselme, P. Blanchard, M. Oçafrain, M. Allain, E. Levillain, and J. Roncali, 2004, J Mater Chem 14(3):421–427.)
Another type of EDOT C60-derivatives has been prepared (see Figure 13.28) with the surprising ability to form self-assembled, chemically adsorbed monolayers on gold surfaces.73 Electro-oxidation lead to desorption of the SAM, traced by the disappearance of the fullerene related peaks. A lot of 2- or 2,5-substituted EDOT derivatives can be summarized under the heading “molecular band gap engineering.” A wide variety of polymers on EDOT basis with “tuned band gap” and hence modified optical properties have been obtained from tricyclic precursors mentioned earlier (see Figure 13.24). Copolymerization of EDOT with electron-withdrawing thiophene comonomers was utilized to construct low band gap polymers via electro-oxidation74 and via Stille coupling75 (see Figure 13.29). A lot of 2-substituted EDOT derivatives, forming tricyclic compounds with terminal EDOT moieties and similar to the examples in Figure 13.25, have been synthesized with median electron acceptor groups. Several examples,
6911X.indb 313
10/1/10 9:01:53 PM
314
PEDOT
O
O O
O
n
S S S O
O
O
O O
S
O
n
S S O
O
O
O S S S
n
O
O O
O
Figure 13.27 Fullerene derivatives of bithiophenes with EDOT-T structure. (Data from B. Jousselme, P. Blanchard, E. Levillain, R. de Bettignies, and J. Roncali, 2003, Macromolecules 36(9): 3020–3025.)
O
O CH3 N
S S O
O
Figure 13.28 Fullerene derivative (fulleropyrrolidine) of bi-EDOT. (Data from S.-G. Liu, C. Martineau, J.-M. Raimundo, J. Roncali, and L. Echegoyen, 2001, Chem Commun 2001(10):913–914.)
6911X.indb 314
10/1/10 9:01:54 PM
315
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
CN
NC
O
O
Electro-oxidation
+ S
S
O
N
S
O
Bu3Sn S
O
O
+
Br
EDOT-copolymer
S
Pd(Ph3)2Cl2
EDOT-copolymer
SnBu3
Br
Figure 13.29 Synthesis of low band gap EDOT copolymers. (Data from H. Huang and P. G. Pickup, 1998, Chem Mater 10(8):2212–2216; H. Meng, D. Tucker, S. Chaffins, Y. Chen, R. Helgeson, B. Dunn, and F. Wudl, 2003, Adv Mater 15(2):146–149.)
illustrating the concept of band gap tuning by combining these groups with EDOT as side donor groups in precursors for electro-oxidation are shown in Figure 13.30. Band gaps of 1.1 to 1.3 eV have been reported for the polymeric electrooxidation products from thienopyrazine (1),76,77 benzothiophene (2),78 benzothiadiazole (3),78 thienothiadiazole (4),76 and the imide (6).79 A little bit higher values (1.3 to 1.4 eV) were estimated for the polymer from the silole (5).80 The quaterthiophenes (7) with pendant EDOT moieties at a cyclopentanodithiophene tricyclic median residue were synthesized by Berlin and Zotti’s group. The dicyanoethen-derivative (7; R = C(CN)2) displayed a very low electrochemical gap of 0.8 eV.81 Nevertheless, the “tricyclic” precursors do not result in an optimum band gap reduction. This effect is attributed to the polymer structure: A regular alternating sequence of EDOT and acceptor group looks necessary for small band gaps instead of the alternating bis-EDOT/acceptor group moieties (see the discussion by Roncali, Blanchard, and Frère55). Thienopyrazines with a di-block instead of a tri-block structure (Formula 1 in Figure 13.30 with one EDOT moiety removed; the synthesis is outlined in Figure 13.31)82 allow the formation of alternating structures of EDOT and acceptor moieties. Consequently, the polymer obtained by electropolymerization exhibits an extremely narrow band gap of about 0.4 eV. Additionally, an exceptional stability under n-doping cycling conditions is observed.82 Substitution of the EDOT moiety with a medium chain alkyl residue (−CH2−O−C10H21, stemming from EDOT-CH2OH/etherification with n-decyl
6911X.indb 315
10/1/10 9:01:54 PM
316
PEDOT
H13C6
O
O
O
O
N
(1) Ar =
N
S
Ar
S
C6H13
S N
(2) Ar =
S
N
(3) Ar = S
N
(4) Ar =
S
N
Ph
(5) Ar =
Si
S
R
R O
N
Ph
O
R
R
(7) Ar =
(6) Ar = S
S
R = = O, = C(CN)2 S R = 2-Ethylhexyl
Figure 13.30 Tricyclic EDOT derivatives with median electron-acceptor group.
group) was performed with the goal to solubilize the polymer.83 Compared to the unsubstituted compound, the decyloxymethyl- derivative exhibited striking changes of the electrochemical and optical properties. The polymer revealed a considerable decrease of stability under n-doping conditions and a 0.35 V negative shift of the peak potential. The band gap was doubled to 0.80 eV, corresponding to a hypsochromic λmax shift of about 400 nm. The unexpected results were discussed as a result of an increased interchain distance, equivalent to lowered π-stacking. Polymers based on the following tricyclic EDOT derivatives with a median arylene- or hetarylene-group have been suggested as photoluminescent materials: Terthiophenes (1),84 benzene derivatives (2),85 thiazole (3),86 and oxadiazole (4)86 (see Figure 13.32). Various extended conjugated π-systems with terminal or central EDOT groups and cyanovinylidene and dicyanovinylidene moieties, respectively, have been
6911X.indb 316
10/1/10 9:01:55 PM
317
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
O2N Br
NO2
S
O
O +
Br
S
Br
Bu3Sn
NO2
O2N
Pd(PPh3)4
S
S
O H2N
C6H13
NH2 S
SnCl2/HCl
N
N
(C6H13–CO)2
S
S
O
O
O
C6H13
S O
O
Figure 13.31 Synthesis of EDOT-pyrazine di-block precursor.
prepared (Figure 13.33). Most of them (molecules with push-pull chromophores: compounds 1 to 4 in Figure 13.33) have been discussed for second-order nonlinear optics (NLO). Several optical properties are summarized in Table 13.2.55,87,88 Comparison of EDOT derived compounds, especially those with a biEDOT spacer, with similar systems reveals an efficiency better than for analog bithiophenes and in the same range as for dithienylethylene compounds. O
S
R
(1) Ar =
O
O Ar
O
S R
(2) Ar =
S R = Bu, Oct
R = NO2, F, CO2Me, OBu
(4) Ar =
(3) Ar =
N
N S
N O
Figure 13.32 Tricyclic EDOT derivatives with median (het)arylene group.
6911X.indb 317
10/1/10 9:01:56 PM
318
PEDOT
H5C2
O
O
NC Don
n
S NC
CN
O
S O
O
O
NC S
R
CN
CN
S O
S
NC
O
(4)
Don =
R = H, OCH3
O
S
R
(5)
H3C
O
(3)
(1): n = 1 (2): n = 2
Don
N CH 2 5
S
Don
NC
O
S
N
O
O
O
O
O
O
CN
NC N
NC
O
CN
S
H 3C CN NC
NC CN (6)
Figure 13.33 EDOT derivatives with dicyanovinylidene groups.
O
O
S R
S S
R
n
S R
S
R = COOCH3 (n = 1); SCH3 (n = 2); SC5H11; C3H7
R
Figure 13.34 EDOT-derived tetrathiafulvalenes.
6911X.indb 318
10/1/10 9:01:57 PM
319
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
Table 13.2 Optical and NLO Properties of Push-Pull Molecules with EDOT as Spacer between Push and Pull Moiety Compound (Figure 12.35)
λmax (nm)
µ𝝱 (10–48 esu)
µ𝝱0 (10–48 esu)
µ(D)
1 2 3 4
768 588 649 830
4600 2120 2000 11600
1300 1200 950 2400
8.0
17.1
Nevertheless, the EDOT derivatives are inferior to a special (bridged) group of dithienylethylenes.55,88,89 The symmetrically substituted, EDOT terminated arylenes (5) (Figure 13.33) have been investigated regarding their photoluminescence properties, similarly to the examples presented in Figure 13.32.86 The cyano-free basic compound had been prepared before by Cava’s group.90 Compound (6) (Figure 13.33) has been discussed for third-order nonlinear optics.55,91 Another group of EDOTs, functionalized by terminal tetrathiafulvalene (TTF) moieties (see Figure 13.34) has been studied by Roncali et al.92,93 Extraordinarily low oxidation potentials have been observed (for example, 0.08 V vs. saturated colomel electrode, SCE, for R = n-C3H7 and n = 1). A further decrease of the oxidation potential to about 0 V was achieved by changing from EDOT to bis-EDOT derivatives (n = 2). All these EDOT-TTF compounds exhibit lower oxidation potentials than their thiophene-based analogs. A lot of ferrocenylethynyl derivatives of EDOT have been published by Zhu and Wolf, accompanied by several thienyl analogs without EDOT moiety.94 Examples 1 to 4 of EDOT derivatives are depicted in Figure 13.35. The cyclic voltammograms of these complexes contain a reversible ferrocene oxidation wave, followed by an irreversible oligothiophene-based wave. The potential of this oligothiophene-based wave varies, depending on the length
O
O Fe
S
Fe
S n
S
m
(1): n, m, p = 0
(2): n, m = 0, p = 1
(3): n, m = 1, p = 0
(4): n, m, p = 1
p
Figure 13.35 Ferrocenylethynyl-EDOTs. (Data from Y. Zhu and M. O. Wolf, 2000, J Am Chem Soc 122(41): 10121–10125.)
6911X.indb 319
10/1/10 9:01:58 PM
320
PEDOT
O
O
R N
O
S
R = Alk O
R = Ar
S N R
O
O
Figure 13.36 EDOT-substituted pyrrolopyrrole.
and substitution of the oligothiophene chain. The mono(ferrocenylethynyl) complexes 1 and 3 couple when oxidized, resulting in the deposition of a redoxactive film on the electrode surface. In solution, electrochemical oxidation of the ferrocenyl groups yields the corresponding mono- and dications, which exhibit oligothiophene-to-Fe(III) charge-transfer transitions in the near infrared.94 Recently, EDOT-substituted diketopyrrolopyrroles (Figure 13.36; with R = CH3, first described in a patent95) have been synthesized, starting from 2-cyano-EDOT, and claimed as chromophores in electroluminescent compositions. Pyrrolo[3,4-c] pyrrole-1,4-dione derivatives carrying 3,4-ethylenedioxy-thiophenylphenyl groups in the 3- and 6-position, or in the 2- and 5-position of the chromophore were also synthesized and electrochemically polymerized.96 The polymers were investigated by cyclic voltammetry and ultraviolet-visible (UV-Vis) absorption spectroscopy. EDOT-phenyl groups in the 3- and 6-positions resulted in conjugated polymers with a low oxidation potential and reversible electrochromic properties, whereas the polymer with EDOT-phenyl groups in the 2- and 5-positions exhibited a high oxidation potential and irreversible redox behavior.96 A very interesting utilization of a 2-substituted EDOT-derivative with technical relevance was established by TDA Research.97–99 Commercialization took place under the trade names AedotronTM and OligotronTM. Luebben et al. functionalized EDOT by 2-substitution with oligoethylene groups and obtain ed multiblock (Figure 13.37) and triblock PEDOT-PEG block copolymers
O
O
O
S
*
x
O
OH
S
O
O y
*
O
z
OH
O
Figure 13.37 PEDOT-PEG multiblock polymer (structural formula without doping).
6911X.indb 320
10/1/10 9:01:59 PM
321
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
OH
C12H25
O
O
O
y
O
O
O
O
O
x
S O
OH O
S
O
O
S
O
y
C12H25
O
Figure 13.38 PEDOT-PEG triblock polymer (structural formula without doping).
(Figure 13.38) by oxidative copolymerization with EDOT. In Figure 13.37 and Figure 13.38, the undoped structure is depicted for simplification. Counterions like perchlorate or toluenesulfonate proved to be useful. The poly(ethyleneglycol) modification results in an improved dispersibility in organic solvents, for example, acetonitrile, nitromethane, or propylenecarbonate. Particle sizes of typically 200 to 1000 nm were reported, triblock PEDOT-PEG tending to give smaller particles than the multiblock analogs.97 The conductivity was found to be in the order of about 1 S/cm (multiblock) or 10 to 60 S/cm (triblock). The 2-monosubstituted EDOT derivative forming the multiblock copolymer shown in Figure 13.37 is available via coupling of two equivalents of 2-lithiated EDOT with commercially available poly(ethyleneglycol) diglycidyl ether, yielding the EDOT-PEG-EDOT building block for the copolymerization with EDOT.98 Similarly, EDOT-Li yields EDOT-PEG by performing the same reaction with poly(ethyleneglycol) monoglycidyl ether. The product is copolymerized with EDOT to form the triblock shown in Figure 13.38. Another route using a 2-monofunctionalized EDOT is starting from EDOT2-carbaldehyde. Imine (Schiff base) formation of the aldehyde with mono- or diamino-PEG yields the EDOT blocks necessary for copolymerization with monomeric EDOT.98 A third concept, utilizing thiophene carboxylic acid esters instead of EDOT derivatives for the construction of PEG-EDOT block copolymers was disclosed: Esterification of 2-thiophene carbonyl chloride with PEG or PEG monoethers yields reactive, thiophene-functionalized PEG blocks, which also can be copolymerized with EDOT.98 A similar concept has been realized by the same group to synthesize methacrylate-terminated PEDOT.99 EDOT was lithiated in the 2-position and reacted with glycidol, yielding the corresponding dialkoxide compound (see Figure 13.39).
O
O
+ S
Li
O
O
O
2
O
O +
OH S
O– Li +
S
O– Li+
Figure 13.39 Formation of EDOT dialkoxide intermediate.
6911X.indb 321
10/1/10 9:01:59 PM
322
PEDOT
O
O O
O
O
O
O O
O
O
S x
S O
O
S
O
O
Figure 13.40 Methacrylate-terminated PEDOT (undoped structure).
Acylation with methacryloyl choride gives a difunctional EDOT methacrylate, which can be copolymerized to the tetrafunctional and therefore cross-linking-active molecule of Figure 13.40.99 These oligomers also can be dispersed in organic solvents like propylene carbonate or nitromethane. Toluenesulfonate counterions are preferred for the doped form.99
References
6911X.indb 322
1. F. Dallacker and V. Mues. 1975. Zur Darstellung von 3,4-Methylendioxythiophen-, furan- und -pyrrol-Abkömmlingen. (Methylenedioxyhetarenes. 1. Preparation of 3.4-methylenedioxythiophene, -furan, and -pyrrole derivatives). Chem Ber 108:569–575. 2. F. Dallacker and V. Mues. 1975. Reaktionen des 3,4-Methylendioxy-2,5-thiophendicarbonsäurediäthylesters. (Methylenedioxyhetarenes. 2. Reactions of diethyl 3,4-methylenedioxy-2,5-thiophenedicarboxylate). Chem Ber 108:576–581. 3. H. J. Ahonen, J. Kankare, J. Lukkari, and P. Pasanen. 1997. Electrochemical synthesis and spectroscopic study of poly(3,4-methylenedioxythiophene). Synth Met 84(1–3):215–216. 4. J. S. Lomas, A. Adenier, K. Gao, F. Maurel, and J. Vaissermann. 2002. Hydrogen bonding and steric effects on rotamerization in 3,4-alkylenedioxy, 3-alkoxy- and 3,4-dialkoxy-2-thienyldi(tert-butyl)-methanols: An NMR, IR and X-ray crystallographic study. J Chem Soc Perkin Trans 2 2002:216–224. 5. K. Reuter. 2003. Unpublished results. 6. L. Brassat, S. Kirchmeyer, and K. Reuter. EP 1 598 358 (H.C. Starck GmbH). Prior: May 15, 2004. 7. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116-118. 8. A. Kumar, D. M. Welsh, M. C. Morvant, F. Piroux, K. A. Abboud, and J. R. Reynolds. 1998. Conducting poly(3,4-alkylenedioxythiophene) derivatives as fast electrochromics with high contrast ratios. Chem Mater 10(3):896–902. 9. K. Reuter. 2001. Unpublished results.
10/1/10 9:02:00 PM
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
323
10. G. Zotti and L. Groenendaal. 1999. Unpublished results; personal communication by L. Groenendaal. 11. D. M. Welsh, A. Kumar, E. W. Meijer, and J. R. Reynolds. 1999. Enhanced contrast ratios and rapid switching in electrochromics based on poly(3,4propylenedioxythiophene) derivatives. Adv Mater 11(16):1379–1382. 12. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 13. K. Reuter. 2001. Unpublished results. 14. G. Zotti. 1998. Doping-level dependence of conductivity in polypyrroles and polythiophenes. Synth Met 97(3):267–272. 15. A. Kros, R. J. M. Nolte, and N. A. J. M. Sommerdijk. 2001. Poly(3,4ethylenedioxythiophene)-based copolymers for biosensor applications. J Polym Sci A: Polym Chem 40(6):738–747. 16. A. Kros, R. J. M. Nolte, and N. A. J. M. Sommerdijk. 2002. Conducting polymers with confined dimensions: Track-etch membranes for amperometric biosensor applications. Adv Mater 14(23):1779–1782. 17. P. Leriche, P. Blanchard, P. Frère, E. Levillain, G. Mabon, and J. Roncali. 2006. 3,4Vinylenedioxythiophene (VDOT): A new building block for thiophene-based π-conjugated systems. Chem Commun 2006:275–277. 18. G. Rauchschwalbe, A. Klausener, S. Kirchmeyer, and K. Reuter. EP 1 275 649 (Bayer AG), Prior: July 12, 2001. 19. S. Roquet, P. Leriche, I. Perepichka, B. Jousselme, E. Levillain, P. Frère, and J. Roncali. 2004. 3,4-Phenylenedioxythiophene (PheDOT): A novel platform for the synthesis of planar substituted π-donor conjugated systems. J Mater Chem 14:1396–1400. 20. I. F. Perepichka, S. Roquet, P. Leriche, J.-M. Raimundo, P. Frère, and J. Roncali. 2006. Electronic properties and reactivity of short-chain oligomers of 3,4-phenylenedioxy-thiophene (PheDOT). Chem Eur J 12(11):2960–2966. 21. S. Kirchmeyer and K. Reuter. 2001. Unpublished results. 22. P. Blanchard, A. Cappon, E. Levillain, Y. Nicolas, P. Frère, and J. Roncali. 2002. Thieno[3,4-b]-1,4-oxathiane: An unsymmetrical sulfur analogue of 3,4-ethylenedioxythiophene (EDOT) as a building block for linear π-conjugated systems. Org Lett 4(4):607–609. 23. M. Belletête, S. Beaupre, J. Bouchard, P. Blondin, M. Leclerc, and G. Durocher. 2000. Theoretical and experimental investigations of the spectroscopic and photophysical properties of fluorene-phenylene and fluorenethiophene derivatives: Precursors of light-emitting polymers. J Chem Phys B 104(39):9118–9125. 24. K. Reuter, U. Merker, and F. Jonas, EP 1 498 920 (H.C. Starck GmbH), Prior: July 14, 2003. 25. K. Reuter. 2003. Unpublished results. 26. C. Wang, J. L. Schindler, C. R. Kannewurf, and M. G. Kanatzidis. 1995. Poly(3,4ethylene-dithiathiophene): A new soluble conductive polythiophene derivative. Chem Mater 7:58–68. 27. H. Meng, D. F. Perepichka, and F. Wudl. 2003. Facile solid-state synthesis of highly conducting poly(ethylenedioxythiophene). Angew Chem Int Ed 42(6):658–661; H. Meng, D. F. Perepichka, and F. Wudl. 2003. Facile solid-state synthesis of highly conducting poly(ethylenedioxythiophene). Angew Chem 42(6):682–685.
6911X.indb 323
10/1/10 9:02:00 PM
324
PEDOT
28. H. Meng, D. F. Perepichka, M. Bendikov, F. Wudl, G. Z. Pan, W. Yu, W. Dong, and S. Brown. 2003. Solid-state synthesis of a conducting polythiophene via an unprecedented heterocyclic coupling reaction. J Am Chem Soc 125(49):15151–15162. 29. H. J. Spencer, R. Berridge, D. J. Crouch, S. P. Wright, M. Giles, I. McCulloch, S. J. Coles, M. B. Hursthouse, and P. J. Skabara. 2003. Further evidence for spontaneous solid-state polymerisation reactions in 2,5-dibromothiophene derivatives. J Mater Chem 13(9):2075–2077. 30. H. J. Spencer, P. J. Skabara, M. Giles, I. McCulloch, S. J. Coles, and M. B. Hursthouse. 2005. The first direct experimental comparison between the hugely contrasting properties of PEDOT and the all-sulfur analogue PEDTT by analogy with well defined EDTT-EDOT copolymers. J Mater Chem 15(45):4783–4792. 31. A. Cravino, H. Neugebauer, A. Petr, P. J. Skabara, H. J. Spencer, J. J. W. McDouall, L. Dunsch, and N. S. Sariciftci. 2006. Spectroelectrochemistry of poly(ethylenedi thiathiophene): The sulfur analogue of poly(ethylenedioxythiophene). J Phys Chem B 110(6):2662–2667. 32. R. M. Walczak and J. R. Reynolds. 2006. Poly(3,4-alkylenedioxypyrroles): The PXDOPs as versatile yet underutilized electroactive and conducting polymers. Adv Mater 18(9):1121–1131. 33. A. Merz, R. Schwarz, and R. Schropp. 1992. 3,4-Dimethoxypyrrole: Monomer synthesis and conducting polymer formation. Adv Mater 4(6):409–411. 34. A. Merz, R. Schropp, and E. Dötterl. 1995. 3,4-Dialkoxypyrroles and 2,3,7,8,12, 13,17,18-octaalkoxyporphyrins. Synthesis 1995(7):795–800. 35. D. J. Savage, B. A. Schell, and B. K. Brady, US 5,665,498 (Eastman Kodak Co.). Prior: November 22, 1996. 36. C. L. Gaupp, K. Zong, P. Schottland, B. C. Thompson, C. A. Thomas, and J. R. Reynolds. 2000. Poly(3,4-ethylenedioxypyrrole): Organic electrochemistry of a highly stable electrochromic polymer. Macromolecules 33(4):1032–1033. 37. C. A. Thomas, K. Zong, P. Schottland, and J. R. Reynolds. 2000. Poly(3,4alkylenedioxypyrrole)s as highly stable aqueous-compatible conducting polymers,vith biomedical implications. Adv Mater 12(3):222–225. 38. P. Schottland, K. Zong, C. L. Gaupp, B. C. Thompson, C. A. Thomas, I. Giurgiu, R. Hickman, K. A. Abboud, and J. R. Reynolds. 2000. Poly(3,4-alkylenedioxypyrrole)s: Highly stable electronically conducting and electrochromic polymers. Macromole cules 33(19):7051–7061. 39. K. Zong and J. R. Reynolds. 2001. 3,4-Alkylenedioxypyrroles: Functionalized derivatives as monomers for new electron-rich conducting and electroactive polymers. J Org Chem 66(21):6873–6882. 40. I. T. Kim, J. Y. Lee, and S. W. Lee. 2004. New conducting polymers based on poly(3,4-ethylenedioxypyrrole): Synthesis, characterization, and properties. Chem Lett 33(1):46–47. 41. A. Merz and T. Meyer. 1999. A short and efficient synthesis of 3,4-dialkoxypyrroles. Synthesis 1999(1):94–99. 42. G. Zotti, S. Zecchin, G. Schiavon, and L. Groenendaal. 2000. Conductive and magnetic properties of 3,4-dimethoxy- and 3,4-ethylenedioxy-capped polypyrrole and polythiophene. Chem Mater 12(10):2996–3005. 43. A. Merz and S. Graf. 1996. Highly conductive free-standing films by electrochemical polymerisation of 3,4-dimethoxypyrrole. J Electroanal Chem 412(1–2):11–17.
6911X.indb 324
10/1/10 9:02:00 PM
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
325
44. R. M. Walczak, J.-H. Jung, J. S. Cowart, Jr., and J. R. Reynolds. 2007. 3,4Alkylenedioxypyrrole-based conjugated polymers with finely tuned electronic and optical properties via a flexible and efficient N-functionalisation method. Macromolecules 40(22):7777–7785. 45. S. Caramori, S. Cazzanti, L. Marchini, R. Argazzi, C. A. Bignozzi, D. Martineau, P. C. Gros, and M. Beley. 2008. Dye-sensitized solar cells based on PEDOP as a hole conductive medium. Inorg Chim Acta 361(3):627–634. 46. O. Tuerkarslan, S. K. Kayahan, and L. Toppare. 2009. A new amperometric cholesterol biosensor based on poly(3,4-ethylenedioxypyrrole). Sensors and Actuators, B: Chemical B136(2):484–488. 47. T. Darmanin and F. Guittard. 2009. One methylene unit to control super oil-repellency properties of conducting polymers. Chem Commun 2009(16): 2210–2211. 48. T. Darmanin and F. Guittard. 2009. Molecular design of conductive polymers to modulate superoleophobic properties. J Am Chem Soc 131(22):7928–7933. 49. T. Darmanin and F. Guittard. 2009. One-pot method for build-up nanoporous super oil-repellent films. J Colloid Interface Sci 335(1):146–149. 50. E. Aqad, M. V. Lakshmikantham, and M. P. Cava. 2001. Synthesis of 3,4-ethylenedioxyselenophene (EDOS): A novel building block for electron-rich π-conjugated polymers. Org Lett 3(26):4283–4285. 51. A. Patra, Y. H. Wijsboom, S. S. Zade, M. Li, Y. Sheynin, G. Leitus, and M. Bendikov. 2008. Poly(3,4-ethylenedioxyselenophene). J Am Chem Soc 130(21):6734–6736. 52. M. Bendikov, A. Patra, Y.-H. Wijsboom, M. Li, Y. Sheynin, N. Zamoshchik, and S. S. Zade. WO 2009/034571 (Yeda Research and Development Co. Ltd), Prior: September 10, 2007. 53. H. Meng, D. F. Perepichka, M. Bendikov, F. Wudl, G. Z. Pan, W. Yu, W. Dong, and S. Brown. 2003. Solid-state synthesis of a conducting polythiophene via an unprecedented heterocyclic coupling reaction. J Am Chem Soc 125(49):15151–15162. 54. Y. H. Wijsboom, A. Patra, S. S. Zade, Y. Sheynin, M. Li, L. J. W. Shimon, and M. Bendikov. 2009. Controlling rigidity and planarity in conjugated polymers: Poly(3,4-ethylenedithioselenophene). Angew Chem Int Ed 48(30):5443–5447. 55. J. Roncali, P. Blanchard, and P. Frère. 2005. 3,4-Ethylenedioxythiophene (EDOT) as a versatile building block for advanced functional π-conjugated systems. J Mater Chem 15:1589–1610. 56. J. R. Reynolds, J. P. Ruiz, A. D. Child, K. Nayak, and D. S. Marynick. 1991. Electrically conducting polymers containing alternating substituted phenylene and bithiophene repeat units. Macromolecules 24(3):678–687. 57. J. P. Ruiz, J. R. Dharia, J. R. Reynolds, and L. J. Buckley. 1992. Repeat unit symmetry effects on the physical and electronic properties of processable, electrically conducting, substituted poly(1,4-bis(2-thienyl)phenylenes. Macromolecules 25(2):849–860. 58. G. A. Sotzing, J. R. Reynolds, and P. J. Steel. 1996. Electrochromic conducting polymers via electrochemical polymerization of bis(2-(3,4-ethylenedioxy)thienyl) monomers. Chem Mater 8(4):882–889. 59. S. A. Sapp, G. A. Sotzing, J. L. Reddinger, and J. R. Reynolds. 1996. Rapid switching solid state electrochromic devices based on complementary conducting polymer films. Adv Mater 8(10):808–811.
6911X.indb 325
10/1/10 9:02:00 PM
326
PEDOT
60. J. L. Reddinger, G. A. Sotzing, and J. R. Reynolds. 1996. Multicoloured electrochromic polymers derived from easily oxidized bis[2-(3,4-ethylenedioxy)thienyl]carbazoles. Chem Commun 1996:1977–1978. 61. J. R. Reynolds, B. Sankaran, G. A. Sotzing, D. J. Irvin, J. A. Irvin, J. L. Reddinger, and S. A. Sapp. 1996. Electrochromic and redox electroactive polymers based on ethylenedioxythiophene derivatives. Mat Res Soc Symp Proc 413:373–376. 62. F. Larmat, J. R. Reynolds, B. A. Reinhardt, L. L. Brott, and S. J. Clarson. 1997. Comparative reactivity of thiophene and 3,4-(ethylenedioxy)thiophene as terminal electropolymerizable units in bis-heterocycle arylenes. J Polym Sci A 35(17):3627–3636. 63. D. J. Irvin, C. J. Dubois, and J. R. Reynolds. 1999. Dual p- and n-type doping in an acid sensitive alternating bi(ethylenedioxythiophene) and pyridine polymer. Chem Commun 1999:2121–2122. 64. S. S. Zhu and T. M. Swager. 1997. Conducting polymetallorotaxanes: Metal ion mediated enhancements in conductivity and charge localization. J Am Chem Soc 119(51):12568–12577. 65. G. Zotti, S. Zecchin, G. Schiavon, A. Berlin, L. Huchet, and J. Roncali. 2001. Electrochemical synthesis and characterization of poly(4,4′(5′)-bis[(3,4(ethylenedioxy)thien-2-yl]tetrathiafulvalene) films: A new TTF–bithiophene regular copolymer. J Electroanal Chem 504(1):64–70. 66. R. P. Kingsborough and T. M. Swager. 1998. Electroactivity enhancement by redox matching in cobalt salen-based conducting polymers. Adv Mater 10(14):1100–1103. 67. R. P. Kingsborough and T. M. Swager. 1999. Polythiophene hybrids of transitionmetal bis(salicylidenimine)s: Correlation between structure and electronic properties. J Am Chem Soc 121(38):8825–8834. 68. L. Huchet, S. Akoudad, and J. Roncali. 1998. Electrosynthesis of highly electroactive tetrathiafulvalene-derivatized polythiophenes. Adv Mater 10(7):541–545. 69. L. Huchet, S. Akoudad, E. Levillain, J. Roncali, A. Emge, and P. Bäuerle. 1998. Spectroelectrochemistry of electrogenerated tetrathiafulvalene-derivatized poly(thiophenes): Toward a rational design of organic conductors with mixed conduction. J Phys Chem B 102 (40):7776–7781. 70. P. Blanchard, B. Jousselme, P. Frère, and J. Roncali. 2002. 3- and 3,4-Bis(2-cyanoethylsulfanyl)thiophenes as building blocks for functionalized thiophene-based π-conjugated systems. J Org Chem 67(11):3961–3964. 71. B. Jousselme, P. Blanchard, M. Oçafrain, M. Allain, E. Levillain, and J. Roncali. 2004. Electrogenerated poly(thiophenes) derivatized by bipyridine ligands and metal complexes. J Mater Chem 14(3):421–427. 72. B. Jousselme, P. Blanchard, E. Levillain, R. de Bettignies, and J. Roncali. 2003. Electrochemical synthesis of C60-derivatized poly(thiophene)s from tailored precursors. Macromolecules 36(9):3020–3025. 73. S.-G. Liu, C. Martineau, J.-M. Raimundo, J. Roncali, and L. Echegoyen. 2001. Formation and electrochemical desorption of stable and electroactive selfassembled monolayers (SAMs) of oligothiophene-fulleropyrrolidine dyads. Chem Commun 2001(10):913–914. 74. H. Huang and P. G. Pickup. 1998. A donor-acceptor conducting copolymer with a very low band gap and high intrinsic conductivity. Chem Mater 10(8):2212–2216.
6911X.indb 326
10/1/10 9:02:00 PM
XDOTs, EDXTs, EDOXs, and 2(5)-X(2)-EDOTs
327
75. H. Meng, D. Tucker, S. Chaffins, Y. Chen, R. Helgeson, B. Dunn, and F. Wudl. 2003. An unusual electrochromic device based on a new low-bandgap conjugated polymer. Adv Mater 15(2):146–149. 76. J.-M. Raimundo. 2000. Ph.D. Thesis, University of Angers. Cited in Y. H. Wijsboom, A. Patra, S. S. Zade, Y. Sheynin, M. Li, L. J. W. Shimon, and M. Bendikov. 2009. Controlling rigidity and planarity in conjugated polymers: Poly(3,4-ethylenedithioselenophene). Angew Chem Int Ed 48(30): 5443–5447. 77. J. Casado, R. Ponce Ortiz, M. C. Ruiz Delgado, V. Hernández, J. T. López Navarrete, J.-M. Raimundo, P. Blanchard, M. Allain, and J. Roncali. 2005. Alternated quinoid/aromatic units in terthiophenes building blocks for electroactive narrow band gap polymers. Extended spectroscopic, solid state electrochemical, and theoretical study. J Phys Chem B 109(35):16616–16627. 78. J.-M. Raimundo, P. Blanchard, H. Brisset, S. Akoudad, and J. Roncali. 2000. Proquinoid acceptors as building blocks for the design of efficient π-conjugated fluorophores with high electron affinity. Chem Commun 2000:939–940. 79. G. Sonmez, H. Meng, and F. Wudl. 2003. Very stable low band gap polymer for charge storage purposes and near-infrared applications. Chem Mater 15(26):4923–4929. 80. Y. Lee, S. Sadki, B. Tsuie, and J. R. Reynolds. 2001. A new narrow band gap electroactive polymer: Poly[2,5-bis{2-(3,4-ethylenedioxy)thienyl}silole]. Chem Mater 13(7):2234–2236. 81. A. Berlin, G. Zotti, S. Zecchin, G. Schiavon, and B. Vercelli. 2004. New low-gap polymers from 3,4-ethylenedioxythiophene-bis-substituted electron-poor thiophenes: The roles of thiophene, donor-acceptor alternation, and copolymerization in intrinsic conductivity. Chem Mater 16(19):3667–3676. 82. S. Akoudad and J. Roncali. 1998. Electrogenerated poly(thiophenes) with extremely narrow bandgap and high stability under n-doping cycling. Chem Commun 2081–2082. 83. I. F. Perepichka, E. Levillain, and J. Roncali. 2004. Effect of substitution of 3,4ethylenedioxythiophene (EDOT) on the electronic properties of the derived electrogenerated low band gap conjugated polymers. J Mater Chem 14:1679–1681. 84. M. F. Pepitone, K. Eaiprasertsak, S. S. Hardaker, and R. V. Gregory. 2003. Synthesis of 2,5-bis[(3,4-ethylenedioxy)thien-2-yl]- 3-substituted thiophenes. Org Lett 5(18):3229–3232. 85. M. F. Pepitone, K. Eaiprasertsak, S. S. Hardaker, and R. V. Gregory. 2004. Synthesis of bis[(3,4-ethylenedioxy)thien-2-yl]-substituted benzenes. Tetrahedron Lett 45(29):5637–5641. 86. M. F. Pepitone, S. S. Hardaker, and R. V. Gregory. 2003. Synthesis and characterization of photoluminescent 3,4-ethylenedioxythiophene derivatives. Chem Mater 15:557–563. 87. J.-M. Raimundo, P. Blanchard, P. Frère, N. Mercier, I. Ledoux-Rak, R. Hierle, and J. Roncali. 2001. Push-pull chromophores based on 2,2′-bi(3,4-ethylene dioxythiophene) (BEDOT) π-conjugating spacer. Tetrahedron Lett 42(8):1507–1510. 88. J.-M. Raimundo, P. Blanchard, N. Gallego-Planas, N. Mercier, I. Ledoux-Rak, R. Hierle, and J. Roncali. 2002. Design and synthesis of push-pull chromophores for second-order nonlinear optics derived from rigidified thiophene-based π-conjugating spacers. J Org Chem 67(1):205–218.
6911X.indb 327
10/1/10 9:02:00 PM
328
PEDOT
89. J.-M. Raimundo, P. Blanchard, I. Ledoux-Rak, R. Hierle, L. Michaux, and J. Roncali. 2000. Huge enhancement of the quadratic nonlinear optical susceptability in push-pull chromophores based on bridged dithienylethylene spacers. Chem Commun 2000:1597–1598. 90. A. K. Mohanakrishnan, A. Hucke, M. A. Lyon, M. V. Lakshmikantham, and M. P. Cava. 1999. Functionalization of 3,4-(ethylenedioxy)thiophene. Tetrahedron 55(40):11745–11754. 91. J.-M. Raimundo, S. Akoudad, H. Brisset, and J. Roncali. 1998. New conjugated systems with multiple redox states. J Chim Phys Phys-Chim Biol 95(6):1234–1237. 92. S. Akoudad, P. Frère, N. Mercier, and J. Roncali. 1999. Low oxidation potential tetrathiafulvalene analogues based on 3,4-dialkoxythiophene π-conjugating spacers. J Org Chem 64(12):4267–4272. 93. P. Leriche, M. Turbiez, V. Monroche, P. Frère, P. Blanchard, P. J. Skabara, and J. Roncali. 2003. Strong π-electron donors based on a self-rigidified 2,2′-bi(3,4ethylenedioxy)thiophene-tetrathiafulvalene hybrid π-conjugated system. Tetrahedron Lett 44(4): 649–652. 94. Y. Zhu and M. O. Wolf. 2000. Charge transfer and delocalization in conjugated (ferrocenylethynyl)oligothiophene complexes. J Am Chem Soc 122(41):10121–10125. 95. H. Yamamoto and N. Dan. WO 2004090046 (Ciba Specialty Chemicals Holding Inc.), Prior: April 10, 2003. 96. K. Zhang, B. Tieke, J. C. Forgie, and P. J. Skabara. 2009. Electrochemical polymerisation of N-arylated and N-alkylated EDOT-substituted pyrrolo[3,4-c]pyrrole1,4-dione (DPP) derivatives: Influence of substitution pattern on optical and electronic properties. Macromol Rapid Commun 30(21):1834–1840. 97. S. A. Sapp and S. Luebben. http://www.tda.com. 98. S. DeVito Luebben, B. Elliott, and C. Wilson, US 7,279,534 (TDA Research, Inc.). Prior: September 3, 2002. 99. B. J. Elliott, W. W. Ellis, S. D. Luebben, S. A. Sapp, C.-H. Chang, and R. A. D’Sa, US Patent 7,361,728 (TDA Research, Inc.). Prior: September 30, 2004.
6911X.indb 328
10/1/10 9:02:00 PM
14 The Electrochemical Behavior of EDOT and PEDOT The first anodical polymerization of 3,4-ethylenedioxythiophene (EDOT) and several derivatives was performed by Jonas, Heywang, Heinze et al. as early as in the second half of the 1980s.1 Some scientific papers were published during the ongoing development,2,3 laying the groundwork for a vast and steadily growing amount of publications. A first comprehensive review was published in 2003.4 Although not practically used in the sense of commercial realization, EDOT and PEDOT [poly(3,4-ethylenedioxythiophene)] electrochemistry provided a lot of information about EDOT, its relatives, and their oxidation/reduction and polymerization reactivity. One has to keep in mind that even more information often is available by combining electrochemical data and (wet) chemical behavior. Figure 14.1 demonstrates the irreversible oxidation of EDOT at 1.39 V versus Ag/AgCl (c = 1.7 × 10 –2 M in propylene carbonate + 0.1 M TBAPF6 (tetran-butylammonium hexafluoro-phosphate); 20°C, sweep rate ν = 100 mVs–1).3 In Figure 14.2, multisweep voltammograms are drawn, obtained under the same experimental conditions. Current increase with each cycle is observed, resulting in the formation of PEDOT:PF6 films on the platinum electrode. The high stability of doped, highly conductive PEDOT observed with the chemically synthesized polymer can also be extracted from multisweep voltammograms. Figure 14.3 shows the multisweep voltammogram of PEDOT in H2O + 0.1 M LiClO4 (20°C; ν = 10 mVs–1). The high, in the time of its invention revolutionary, stability of PEDOT in water is clearly demonstrated by the fact that this polymer (in the form of PEDOT:ClO4) can be electrochemically cycled in water. The differences of EDOT or PEDOT to similar (poly)thiophenes are visualized by several electrochemical results. But not all cyclovoltammetric (CV) data are parallel to the preeminent role that PEDOT plays, compared to other members of the XDOT family (X = methylene, ethylene, propylene, butylene; see Figure 14.4). A truly remarkable step is observed between MDOT and EDOT. The optical band gap of electropolymerized MDOT is estimated to 2.05 or 2.054 eV, far more than with PEDOT, which has a band gap of 1.6 eV. There has been a lot of difficulties in electropolymerizing MDOT, beginning with the choice of the solvent. Standard solvents like propylene carbonate (PC) or acetonitrile (AN) did not result in satisfactory polymerization reactions. Only in nitrobenzene (NB) were PMDOT films obtained potentiodynamically. Other authors could reproduce Ahonen’s results:5 no polymerization of MDOT took place on Pt 329
6911X.indb 329
10/1/10 9:02:01 PM
330
PEDOT
4 µA
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
E (vs. Ag/AgCl) [V] Figure 14.1 Cyclovoltammogram for the irreversible oxidation of EDOT.3 (Reprinted from J Electroanal Chem 369(1-2):87–92, M. Dietrich, J. Heinze, G. Heywang, and F. Jonas, Electrochemical and Spectroscopic Characterization of Polyalkylenedioxythiophenes. Copyright 1994, with permission from Elsevier.)
electrodes in AN or NB, nor on indium tin oxide (ITO) in AN; only polymerization in NB on ITO electrodes could be observed.6 But the nature of these films must remain a matter of debate: a chemical oxidation of MDOT could not be performed with typical oxidants for EDOT, like peroxodisulfates or iron-III tosylate. Nitric acid as a very strong oxidant—neither EDOT nor
10 µA
–1.4
–0.9
–0.4
0.1
0.6
1.1
1.6
E [vs. Ag/AgCl]/V Figure 14.2 Multisweep voltammograms of EDOT. (Reprinted from J Electroanal Chem 369(1-2):87–92, M. Dietrich, J. Heinze, G. Heywang, and F. Jonas, Electrochemical and Spectroscopic Characteri zation of Polyalkylenedioxythiophenes. Copyright 1994, with permission from Elsevier.)
6911X.indb 330
10/1/10 9:02:02 PM
331
The Electrochemical Behavior of EDOT and PEDOT
1 µA
–1.0
–0.5
0.0
0.5
E [vs. Ag/AgCl]/V Figure 14.3 Multisweep voltammogram of PEDOT in H2O/LiClO4. (Reprinted from J Electroanal Chem 369(1-2):87–92, M. Dietrich, J. Heinze, G. Heywang, and F. Jonas, Electrochemical and Spectroscopic Characterization of Polyalkylenedioxythiophenes. Copyright 1994, with permission from Elsevier.)
PEDOT can withstand it—only yields 2,5-dinitro-MDOT without decomposition.7 No chemical oxidant for the transformation of monomeric MDOT into a more or less conductive or even a neutral polymer was found (see also the discussion in Chapter 6). A series of EDOT oligomers, blocked by phenyl end groups (Figure 14.5; EDOT1–EDOT4), has been synthesized to check EDOT-typical optical and redox properties without further oxidative polymerization.8 The β, β′- unsubstituted thiophene analogs were also described (T1–T4) and compared to the EDOT derivatives. Blocking the end groups by phenyl was performed to prevent irreversible α-coupling during oxidative doping. So anodic peak potentials EPA1 and EPA2 for the formation of radical cation and dication, respectively, were measured by cyclic voltammetry. The results are summarized in Table 14.1 (all values in V vs. calomel electrode, solvent: CH2Cl2, electrolyte: 0.1 M TBAPF6.
O
O
O
O
O
O
O
O
S
S
S
S
MDOT
EDOT
ProDOT
BuDOT
Figure 14.4 The XDOT family.
6911X.indb 331
10/1/10 9:02:03 PM
332
PEDOT
O
O
S
n
n = 1 – 4 (EDOT1 – EDOT4) Figure 14.5 Phenyl-capped EDOT oligomers. (Data from J. J. Apperloo, L. Groenendaal, H. Verheyen, M. Jayakannan, R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni, and J.-L. Bredas, 2002, Chem Eur J 8(10):2384–2396.)
Comparison of the EDOT1–DOT4 oxidation potentials to the T1–T3 values (T4 solubility was too low for cyclovoltammetry) clearly shows the influence of the electron-rich dioxane ring, which strongly reduces the oxidation potential. Electron paramagnetic resonance (EPR) spectra of the radical cations of EDOT1–EDOT 3 were recorded in dichloromethane and are shown in Figure 14.6. The following hyperfine coupling parameters were extracted from the spectra and resulted in correct simulations: EDOT1.+: a(H) = 3.194 gauss (2H), a(H) = 2.362 gauss (2H), a(H) = 0.747 gauss (4H), a(H) = 0.756 gauss (2H). EDOT2.+: a(H) = 1.506 gauss (4H), a(H) = 1.770 gauss (4H), a(H) = 0.530 gauss (4H), a(H) = 0.257 gauss (2H). The spectrum of EDOT3.+ did not exhibit a hyperfine structure; all hyperfine coupling parameters are too small to be resolved in relation to the line width. The extended π-system is responsible for the decreasing spin densities and hyperfine coupling parameters. The g values were measured to 2.0023 (EDOT1.+) and 2.0022 (EDOT2.+ and EDOT3.+). These values are nearly identical to the free-electron g factor (2.0023) and
Table 14.1 Anodic Peak Potentials of Phenyl-Capped EDOT and T Oligomers EDOT1 EDOT2 EDOT3 EDOT4 T1 T2 T3
EPA1
EPA2
1.18 0.72 0.42 0.26a 1.42 1.14 0.96
about 2 1.32 0.69 0.39a >2 1.65 1.33
Source: Data from J. J. Apperloo et al., 2002, Chem Eur J 8(10):2384–2396. a Tentative values due to small impurities.
6911X.indb 332
10/1/10 9:02:03 PM
333
The Electrochemical Behavior of EDOT and PEDOT
(a)
Exp.
Sim.
3490 (b)
3495
3500
Exp.
ESR/a.u. Sim.
3480
3485
3490
(c)
Exp.
3475
3480
3485
3490
B/Gauss Figure 14.6 EPR spectra and simulation of (a) EDOT1, (b) EDOT2, and (c) EDOT3 radical cations. (From J. J. Apperloo, L. Groenendaal, H. Verheyen, M. Jayakannan, R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni, and J.-L. Bredas, Optical and Redox Properties of a Series of 3,4-Ethylenedioxythiophene, J Oligomers Chem Eur, 2002, 8(10):2384–2396. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
in accordance with other thiophene radical cations, for example, the oligoor polythiophene g values of 2.0023 to 2.0025.9–11 A lot of sulfur containing radical cations exhibit much higher g values.12 Despite the high spin-orbit coupling parameter for sulfur (382 cm–1 for S3p),13 several thiophene-derived radical cations indicate that the single electron occupies an orbital with a node at the sulfur atom.14-17 From infrared investigations it could also be concluded that the sulfur atoms do not strongly contribute to the electronic structure of a typical oligothiophene radical cation (α-sexithiophen.+).18
6911X.indb 333
10/1/10 9:02:04 PM
334
PEDOT
The cation radicals of EDOT2 and EDOT3 were found to dimerize to spinless π-dimers by cooling their solutions in dichloromethane to 250 to 190 K.8 First, in situ EPR investigations of PEDOT itself have been published by Zotti et al.19 During oxidation the intensity maximum of the EPR signal corresponded to 0.02 spins per thiophene ring, attained at a doping charge of 0.06 electrons per ring. The EPR signal disappeared at a doping charge of about 0.15 electrons per ring. The authors compared conductivity and spin concentration as a function of doping charge and again claimed—not for the first time, but on the basis of very clear results—that conductivity and free electrons are not directly correlated. Very detailed EPR investigations with electrochemically produced PEDOT layers have been performed by Lapkowski et al.20–23 Several PEDOT derivatives have also been studied by EPR spectroscopy.20,24–28 The following procedure is described by Lapkowski and coworkers.23 After potentiodynamic EDOT electropolymerization is performed from 0.01 M solution in acetonitrile with 0.1 M TBAPF6 electrolyte, following a procedure published earlier,22 residual monomer is removed and the polymer film subjected to several cycles, then the investigation is started with the dedoped state.23 A lot of very detailed information was gathered,23 but not all results seem to fit the findings of other groups very well, for example, the investigations of the Zotti group.19 So, for example, doping levels of up to 0.55 (more than one electron per two EDOT mers) are discussed,23 but from other investigations a doping level of one charge per three to four EDOT moieties, representing one bipolaron per six to eight monomer units, has been established for doped PEDOT (see later). Moreover, relatively high spin concentrations corresponding to 0.08 polarons per repeating unit were found in highly doped states besides spinless polarons. This seems to not be consistent with the spectroelectrochemical results with highly doped, nearly transparent and colorless (light blue) state of PEDOT. Because the paramagnetic, lightly doped state of PEDOT, achievable by chemical or electrochemical reduction of PEDOT, should be associated to one of the intensely blue forms of PEDOT (the other one is the once protonated, diamagnetic undoped form [see Chapter 6]), a contradiction has to be stated, since the addition of bipolaronic species cannot eliminate the intense color of the paramagnetic species in the film. An intriguing observation by Lapkowski’s group regarding the paramagnetic PEDOT species formed during the doping–dedoping cycles was the fact that two different EPR signal components can be identified.21,23 One intense signal with a broad Lorentzian line shape is observed besides a clearly smaller and narrow signal with a Gaussian shape. The g factors differ remarkably (2.0023 to 2.0034 for the narrow and 2.0012 to 2.0018 for the broad EPR signals).23 Both components could not be assigned to distinct, chemically defined PEDOT species, but the second value seems to be very low for a sulfur containing radical cation.
6911X.indb 334
10/1/10 9:02:04 PM
The Electrochemical Behavior of EDOT and PEDOT
335
A remarkable observation is the nonzero EPR signal at the dedoped state after several cycles. A quantitative estimation, using double integration of the EPR spectra, results in about 0.02 residual spins per repeating unit. This is in sharp contrast to similar polymers, like polythiophene29 or polypyrrole30 with far lower values, but not really surprising, although Zotti et al. could not observe this residual paramagnetic species.19 Johansson et al. had checked the conductivity of chemically synthesized PEDOT after electrochemical dedoping.31 Dedoping of two different samples resulted in a decrease of about five orders of magnitude, starting from the highly conductive state to 1.2 – 5 × 10 –4 S/cm at a bias of –2 V and 6 × 10 –5 S/cm at –3 V. A rather stable conductivity level at –1.2 to –5 V was reached without the tendency to reincrease the conductivity by n-doping, which had been reported before.32 The residual conductivity of electrochemically dedoped PEDOT is about two to six orders of magnitude higher than observed for other nondoped, semiconducting polythiophenes. Inganäs and coworkers31 concluded that they never obtained the true neutral state of PEDOT. The same was true for chemically dedoped films with hydrazine as the reducing agent.31 On the contrary, both chemically and electrochemically dedoped PEDOT was redoped, traced by sharply increasing conductivity, under ambient conditions by air oxygen. These observations are consistent with the following statements: (1) The enormous stability of cationic moieties in PEDOT does not allow absolutely complete reduction, and (2) the tendency of forming cationic (polaronic and bipolaronic) PEDOT structures is too high to keep residual radical-cationic PEDOT nonoxidized in the presence of oxygen. The remarkable stability of truly neutral PEDOT, prepared in an oxidative synthesis, does not contradict; and in solution, especially in the presence of protons, neutral PEDOT is very easily oxidized. Chevrot and colleagues investigated the electrodeposition processes of EDOT to PEDOT in a kinetic and mechanistic study under potentiostatic conditions.33 After an initial stage of simultaneous two- and three-dimensional instantaneous nucleation, a layer-by-layer deposition and growth in accordance with the so-called Stranski–Kastanov model was observed. The authors also determined the doping level of the PEDOT layer. A value of γ = 0.3 was obtained (γ = number of electrons per EDOT moiety, required for switching, and number of positive charges per EDOT in the fully doped state). This is confirmed by results of the Reynolds group. Electrochemical polymerization resulted in PEDOT with a doping level corresponding to one ClO4– anion per three to four EDOT moieties.34 The investigations of the properties of electrochemically prepared PEDOT have been considerably improved and extended by in situ methods, established, for example, by Zotti and coworkers.35,36 A lot of conductivity data have been collected by Zotti et al. regarding the influence of different counterions. Several details, including publications from other groups,37–41 are presented in Chapter 7.
6911X.indb 335
10/1/10 9:02:04 PM
336
PEDOT
Detailed investigations regarding the influence of the electropolymerization conditions on the properties of PEDOT were performed by the Abrantes group.42 The authors deposited PEDOT on platinum electrodes by consecutive potential scanning from acetonitrile solutions of EDOT and studied the effect of different supporting electrolytes. They used LiClO4, tetra-nbutylammonium perchlorate (TBAClO4) and TBAPF6. So the influence of a change of the cation could be investigated as well as the influence of different anions. The highest electropolymerization efficiency was obtained with LiClO4. The use of LiClO4 also yielded PEDOT layers with increased crystallinity and electroactivity. LiClO4 promotes a more compact morphology, formed by clusters of different size, probably because the Li+ cations can participate in the potentiodynamic electropolymerization resulting in a conformationally preferred situation for the growth of PEDOT in one plane (100). This is confirmed by the pronounced peak at low angle in the x-ray diffraction (XRD) pattern. The thicker the films (dependent on the number of growing cycles), the higher was their structural organization. This was attributed to an increase of crystallites. More porous PEDOT layers than with Li+ cations were obtained with the larger TBA+ cation. The same effect of increased film porosity is observed when PF6– counterions are used instead of perchlorate anions. The films resulting with TBAPF6 present a distinct morphology, exhibiting fibrillar structures. The morphological features were based on scanning electronic micrographs (SEM). The electrochemical synthesis of PEDOT is not limited to the unsubstituted EDOT (or EDOT oligomers) as the starting material. 2,5-Bis(trimethylsilyl)EDOT (1) and the corresponding bis-EDOT (2), depicted in Figure 14.7, have also efficiently been electropolymerized, undergoing desilylation.43 The desilylation takes place with the trimethylsilyl cation (Me3Si+) as the leaving group. The same had been demonstrated in silylated thiophene derivatives without dioxane moiety by trapping the trimethylsilyl group in the form of Me3SiF by fluoride ions or (Me3Si)2O by air oxygen.44 From visible spectroscopy, the authors concluded that EDOT-TMS2 (compound 1 in Figure 14.7) is polymerized 2.5 times faster than EDOT in parallel separate experiments under identical conditions. Interestingly, the redox switching rate of PEDOT from EDOT-TMS2 was found to be enhanced, compared to PEDOT from EDOT. The authors attribute a silylation of the ITO surface used in the
O (CH3)3Si
O
S
n
Si(CH3)3
(1): n = 1 (2): n = 2
Figure 14.7 Silylated EDOT and bis-EDOT, suitable for electrochemical polymerization.
6911X.indb 336
10/1/10 9:02:05 PM
The Electrochemical Behavior of EDOT and PEDOT
337
optoelectrochemical analysis, causing a better surface adherence of PEDOT from compound 1, as the reason for this observation.43 Higher oligomers of the same formula as in Figure 14.7, but with n = 3–5, have also been prepared, but investigated only photophysically.45 Similar oligomers with 1 to 4 EDOT units and phenyl endgroups are discussed earlier. Cyclovoltammogrammetric studies with mesitylthio instead of phenyl groups revealed the high stability of the corresponding radical cations and dications, suggesting the localization of the positive charges at the end of the short conjugated chain.46 This is a noticeable contrast to the also structurally analog EDOT oligomers (n = 1–4) with α,ω-dihexyl substitution: here the electron releasing effect of the alkyl residues results in a pronounced localization of the positive charges in the middle of the conjugated chain.47 The electrochemical studies on EDOT and PEDOT have been extended by Reynolds and colleagues to the electropolymerization in the presence of socalled ionic liquids.48 They built supercapacitors of PEDOT and PProDOT as electrode couples and investigated their switching and storage properties. This supercapacitor type was applied for a patent.49 The room-temperature ionic liquid material between the PEDOT and PProDOT electrodes consisted of 1-ethyl-3-methyl- 1-H-imidazolium bis(trifluoromethylsulfonyl)imide (EMIBTI) (see Figure 14.8) and was not used in pure form, but in propylene carbonate (PC) solution. The PProDOT/PEDOT/EMI-BTI/PC system was compared to the same configuration with lithium bis(trifluoromethylsulfonyl)imide (Li-BTI)/PC as the gel electrolyte. The CV responses at 100 mV/s sweep rate of PEDOT/Li-BTI, PEDOT/EMI-BTI, PProDOT/Li-BTI and PProDOT/EMI-BTI are similar, with the same E1/2 potentials for PEDOT/Li-BTI and PEDOT/EMI-BTI, and also the same E1/2 potentials for PProDOT/Li-BTI and PProDOT/EMI-BTI. The peaks in the CV response of PProDOT/EMI-BTI in the oxidative wave are slightly better defined than the same peaks for PProDOT/Li-BTI. A more pronounced improvement was observed for PEDOT/EMI-BTI, compared to the less defined PEDOT/Li-BTI CV peaks. The better definition of the peaks in the CV of the EMI-BTI systems suggests faster transport processes with the EMI cation. Switching speed and cycle lifetimes were improved using the organic liquids.48 Pure EMI-BTI as an electrolyte in cyclovoltammetric investigations of PEDOT was first used by Randriamahazaka, Chevrot, and coworkers.50,51
N + N
O F3 C
S
– N O O
O S
CF3
Figure 14.8 1-Ethyl-3-methyl-1-H-imidazolium bis(trifluoromethylsulfonyl)imide (EMI-BTI) ionic liquid.
6911X.indb 337
10/1/10 9:02:06 PM
338
PEDOT
The cyclic voltammetry responses and the redox switching dynamics of PEDOT in EMI-BTI were investigated.50 The shape of the CVs showed two anodic and two cathodic peaks. These peak currents varied linearly with the scan rate, indicating a thin-layer behavior. The redox switching dynamics of PEDOT were studied by potential step experiments: The time dependence of the charge transferred during the potential step showed two time constants. These results were consistent with a morphology of the PEDOT film, which contained two types of coexisting zones: a compact and an open structure.50 Electrochemical impedance spectroscopy was also used by Randria mahazaka, Chevrot, and colleagues to study the responses of PEDOT (electrodeposited from acetonitrile solution on platinum) in EMI-BTI.51 The results indicated two coexisting zones (compact and more open) in the PEDOT film. Consistently, two parallel diffusion paths with two time constants—for a slow and a fast process—are observed in accordance with chronocoulometric experiments.50 The results could be further confirmed by the same group, using detailed large amplitude potential step experiments.52 Relaxation kinetics revealed that the time constant of the slow process was virtually independent of the film thickness, whereas the fast process depended upon the PEDOT film thickness. The kinetics for the oxidation were found to be faster than that of reduction. In the same year, charging and discharging experiments with PEDOT under galvanostatic conditions in EMI-BTI were published by the CergyPontoise group.53 A linear variation of the voltage with respect to time was observed. Analyzing the electrochemical response in terms of a series combination of a resistor and a capacitor, a linear variation of the capacitance as a function of the amount of PEDOT was found. The electrochemistry of PEDOT in ionic liquids was also studied with more bulky anions than BTI; particularly bulky organic anions (in combination with 1-butyl-3-methylimidazolium = BMI cations instead of EMI) were investigated.54 The mixture of EDOT with BMI/diethyleneglycol monomethylether sulfate (BMI-MDEGSO4) did not yield any or only very limited amounts of PEDOT during attempted electrosynthesis. In contrast, BMI/ octylsulfate (BMI-OctSO4) in 1:1-mixture (wt./wt.) with EDOT resulted in a high yield of PEDOT on the electrode surface. The electrosynthesis in water– ionic liquid mixtures gave good yields of electroactive PEDOT layers for both BMI-MDEGSO4 and BMI-OctSO4. Ionic liquids (substituted imidazolium salts) were utilized in the electrochemical synthesis of PEDOT films with specific structural or morphological features: randomly oriented nanofibers and particles in submicrometersized domains with EMI-bis(pentafluoroethylsulfonyl)imide, 55 the structure of a solid polymer actuator with PEDOT layer on the surface of a solid nitrile rubber-BMI/BTI-mixture, 56 single strand nanowires with BMIhexafluorophosphate,57 and codeposited polypyrrole/PEDOT in BMI-BTI.58 A lot of work has been performed by Bund and Schneider concerning investigations of electrochemically prepared PEDOT by the electrochemical
6911X.indb 338
10/1/10 9:02:06 PM
The Electrochemical Behavior of EDOT and PEDOT
339
quartz crystal microbalance (EQCM) technique. Starting with PEDOT electrodeposited from organic solvents like acetonitrile or propylenecarbonate, the analysis of viscoelastic properties and surface roughness of the PEDOT film was calculated from the shift of the resonance curve of the quartz crystal,59 and the doping and dedoping behavior of PEDOT depending on solvents and anions was studied by EQCM in acetonitrile or aqueous electrolytes.60 For most of the anions, the exchanged molar mass during oxidation could be attributed to the reversible replacement of solvent molecules by anions (perchlorate, nitrate, citrate, or toluenesulfonate), whereas no substantial cation exchange was observed.60 A special behavior was found for sulfate, carrying water into the film (explained by the high charge density of SO42–). PEDOT layers were also prepared electrochemically from water containing different supporting electrolytes (sulfate, perchlorate and dihydrogenphosphate); the ratio of the incorporated mass and the passed electric charge during redox switching of the PEDOT film was suggested as a detection tool to identify the ionic species.61 The EQCM investigations were extended to the use of various ionic liquids later, including acoustic impedance measurements.62 Besides EMI-BTI, the analog triflate (EMI-TF), 1-butyl-1-methylpyrrolidinium/BTI (BMP-BTI; see Figure 14.9), BMP-TF and butyltrimethylammonium/BTI were examined for the polymerization of EDOT on the gold electrode of an EQCM. Elastic shear moduli—calculated from acoustic impedance data—were found to be approximately two orders of magnitude higher than for PEDOT from acetonitrile solution. This effect was explained by the complete absence of the plasticizing effect from incorporated solvent. The electropolymerization of EDOT in imidazolium and pyrrolidinium ionic liquids had also been investigated by Officer’s group.63 Film growth in EMI-BTI was found to be faster than in acetonitrile/tetrabutylammonium perchlorate (Bu4NClO4). The more viscous and less ion conductive BMP-BTI had a negative influence on film growth. The same tendency was observed regarding the electrochemical activity of the deposited films: Comparing growth rate in electrodeposition and electroactivity in postpolymerization measurements in different electrolytes, the order EMI-BTI > Bu4NClO4 > BMP-BTI was observed for both effects.
+ N
F3C
O
– N
O S O
S O
CF3
Figure 14.9 1-Butyl-1-methyl-pyrrolidiniumium bis(trifluoromethylsulfonyl)imide (BMP-BTI) ionic liquid.
6911X.indb 339
10/1/10 9:02:06 PM
340
PEDOT
References
6911X.indb 340
1. F. Jonas, G. Heywang, W. Schmidtberg, J. Heinze, and M. Dietrich. EP 339 340 (Bayer AG), Prior: April 22, 1988. 2. G. Heywang and F. Jonas. 1992. Poly(alkylenedioxythiophene)s: New, very stable conducting polymers. Adv Mater 4(2):116–118. 3. M. Dietrich, J. Heinze, G. Heywang, and F. Jonas. 1994. Electrochemical and spectroscopic characterisation of polyalkylenedioxythiophenes. J Electroanal Chem 369(1–2):87–92. 4. L. Groenendaal, G. Zotti, P.-H. Aubert, S. M. Waybright, and J. R. Reynolds. 2003. Electrochemistry of poly(3,4-alkylenedioxythiophene) derivatives. Adv Mater 15(11):855–879. 5. H. J. Ahonen, J. Kankare, J. Lukkari, and P. Pasanen. 1997. Electrochemical synthesis and spectroscopic study of poly(3,4-methylenedioxythiophene). Synth Met 84(1–3):215–216. 6. H.-B. Bu, G. Götz, P. Bäuerle, L. Brassat, and K. Reuter. 2004. Unpublished results. 7. K. Reuter. 2003. Unpublished results. 8. J. J. Apperloo, L. Groenendaal, H. Verheyen, M. Jayakannan, R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni, and J.-L. Bredas. 2002. Optical and redox properties of a series of 3,4-ethylenedioxythiophene oligomers. Chem Eur J 8(10):2384–2396. 9. K. Kaneto, S. Hayashi, S. Ura, and K. Yoshino. 1984. ESR and transport studies in electrochemically doped polythiophene film. J Phys Soc Jpn 54(3):1146–1153. 10. A. Alberti, L. Favaretto, G. Seconi, and G. F. Pedulli. 1990. Radical cations and anions from some simple polythiophenes: An ESR investigation. J Chem Soc Perkin Trans 2 1990(6):931–935. 11. P. Bäuerle, U. Segelbacher, K.-U. Gaudl, D. Huttenlocher, and M. Mehring. 1993. Thiophenes. 9. Didodecylsexithiophene, a model compound for the generation and characterization of charge carriers in conjugated chains. Angew Chem 105(1):125–127; Angew Chem Int Ed 32(1):119–121. 12. See, for example, K. Scheffler and H. B. Stegmann. 1970. Elektronenspinresonanz. Berlin: Springer-Verlag. 13. G. Tourillon, D. Gourier, F. Garnier, and D. Vivien. 1984. Electron spin resonance study of electrochemically generated polythiophene and derivatives. J Phys Chem 88(6):1049–1051. 14. D. Fichou and C. Ziegler. Structure and properties of oligothiophenes in the solid state: Single crystals and thin films. In: Handbook of Oligo- and Polythiophenes, ed. D. Fichou, 183–282. Weinheim: Wiley-VCH. 15. K. Tanaka, Y. Matsuura, Y. Oshima, T. Yamabe, and. S. Hotta. 1994. Electronic properties of p-type doped oligothiophenes. Synth Met 66(3):295–298. 16. D. N. R. Rao and M. C. R. Symons. 1983. Unstable intermediates. Part 205. Radical cations of pyrrole, furan, and thiophene derivatives: An electron spin resonance study. J Chem Soc Perkin Trans 2 1983(2):135–137. 17. M. Shiotani, Y. Nagata, M. Tasaki, J. Sohma, and T. Shida. 1983. Electron spin resonance studies on radical cations of five-membered heteroaromatics. Furan, thiophene, pyrrole, and related compounds. J Phys Chem 87(7):1170–1174.
10/1/10 9:02:07 PM
The Electrochemical Behavior of EDOT and PEDOT
341
18. D. Fichou, G. Horowitz, B. Xu, and F. Garnier. 1990. Stoichiometric control of the successive generation of the radical cation and dication of extended α-conjugated oligothiophenes: A quantitative model for doped polythiophene. Synth Met 39(2):243–259. 19. G. Zotti, S. Zecchin, G. Schiavon, and L. Groenendaal. 2000. Conductive and magnetic properties of 3,4-dimethoxy- and 3,4-ethylenedioxy-capped polypyrrole and polythiophene. Chem Mater 12(10):2996–3005. 20. A. Zykwinska, W. Domagala, A. Czardybon, B. Pilawa, and M. Lapkowski. 2003. In situ EPR spectroelectrochemical studies of paramagnetic centres in poly(3,4-ethylenedioxythiophene) (PEDOT) and poly(3,4-butylenedioxythiophene) (PBuDOT) films. Chem Phys 292(1):31–45. 21. A. Zykwinska, W. Domagala, and M. Lapkowski. 2003. ESR spectroelectrochemistry of poly(3,4-ethylenedioxythiophene) (PEDOT). Electrochem Commun 5(7):603–608. 22. A. Zykwinska, W. Domagala, B. Pilawa, and M. Lapkowski. 2005. Electrochemical overoxidation of poly(3,4-ethylenedioxythiophene)—PEDOT studied by means of in situ ESR spectroelectrochemistry. Electrochim Acta 50(7–8):1625–1633. 23. W. Domagala, B. Pilawa, and M. Lapkowski. 2008. Quantitative in-situ EPR spectroelectrochemical studies of doping processes in poly(3,4-alkylenedioxythiophene)s: Part 1: PEDOT. Electrochim Acta 53(13):4580–4590. 24. S. Garreau, G. Louarn, G. Froyer, M. Lapkowski, and O. Chauvet. 2001. Spectroelectro-chemical studies of the C14-alkyl derivative of poly(3,4-ethylenedioxythiophene) (PEDT). Electrochim Acta 46(8):1207–1214. 25. A. Zykwinska, W. Domagala, A. Czardybon, B. Pilawa, and M. Lapkowski. 2004. Investigation of charge carriers in poly(3,4-butylenedioxythiophene) (PBuDOT) by means of ESR spectroelectrochemistry. J Solid State Electrochem 8(6):369–375. 26. T. Moskwa, W. Domagala, A. Czardybon, B. Pilawa, and M. Lapkowski. 2005. ESR spectroelectrochemistry of functionalised long side chain derivatives of poly(3,4-ethylenedioxythiophene). Synth Met 152(1-3):189–192. 27. A. Zykwinska, W. Domagala, A. Czardybon, B. Pilawa, and M. Lapkowski. 2006. In-situ ESR spectroelectrochemical studies of overoxidation behaviour of poly(3,4-butylenedioxythiophene). Electrochim Acta 51(11):2135–2144. 28. P. Berdyczko, W. Domagala, A. Czardybon, and M. Lapkowski. 2009. Long alkyl chain bearing derivatives of poly(3,4-ethylenedioxythiophene) studied by in situ EPR spectroelectrochemistry. Synth Met 159(21–22):2240–2244. 29. K. Kaneto, S. Hayashi, S. Ura, and K. Yoshino. 1985. ESR and transport studies in electrochemically doped polythiophene film. J Phys Soc Jpn 54:1146–1153. 30. J. Scott, P. Pfluger, M. T. Krounbi, and G. B. Street. 1983. Electron-spin-resonance studies of pyrrole polymers: evidence for bipolarons. Phys Rev B 28(4): 2140–2145. 31. T. Johansson, L. A. A. Pettersson, and O. Inganäs. 2002. Conductivity of dedoped poly(3,4-ethylenedioxythiophene). Synth Met 129(3):269–274. 32. H. J. Ahonen, J. Lukkari, and J. Kankare. 2000. n- and p-Doped poly(3,4ethylenedioxythiophene): Two electronically conducting states of the polymer Macromolecules 33(18):6787–6793. 33. H. Randriamahazaka, V. Noël, and C. Chevrot. 1999. Nucleation and growth of poly(3,4-ethylenedioxythiophene) in acetonitrile on platinum under potentiostatic conditions. J Electroanal Chem 472(2):103–111.
6911X.indb 341
10/1/10 9:02:07 PM
342
PEDOT
34. J. Hwang, D. B. Tanner, I. Schwendeman, and J. R. Reynolds. 2003. Optical properties of non-degenerate ground-state polymers: Three dioxythiophene-based conjugated polymers. Phys Rev B 67:115205/1–115205/10. 35. G. Schiavon, S. Sitran, and G. Zotti. 1989. A simple two-band electrode for in situ conductivity measurements of polyconjugated conducting polymers. Synth Met 32(2):209–217. 36. P.-H. Aubert, L. Groenendaal, F. Louwet, L. Lutsen, D. Vanderzande, and G. Zotti. 2002. In situ conductivity measurements on polyethylenedioxythiophene derivatives with different counter ions. Synth Met 126(2–3):193–198. 37. A. Aleshin, R. Kiebooms, R. Menon, and A. J. Heeger. 1997. Electronic transport in doped poly(3,4-ethylenedioxythiophene) near the metal-insulator transition. Synth Met 90(1):61–68. 38. R. Kiebooms, A. Aleshin, K. Hutchison, and F. Wudl. 1997. Thermal and electromagnetic behavior of doped poly(3,4-ethylenedioxythiophene) films. J Phys Chem B 101(51):11037–11039. 39. A. N. Aleshin, R. Kiebooms, and A.J. Heeger. 1999. Metallic conductivity of highly doped poly(3,4-ethylenedioxythiophene). Synth Met 101(1–3):369–370. 40. R. Kiebooms, A. Aleshin, K. Hutchison, F. Wudl, and A. Heeger. 1999. Doped poly(3,4-ethylenedioxythiophene) films: Thermal, electromagnetical and morphological analysis. Synth Met 101(1–3):436–437. 41. I. Giurgiu, K. Zong, J. R. Reynolds, W.-P. Lee, K. R. Brenneman, A. V. Saprigin, A. J. Epstein, J. Hwang, and D. B. Tanner. 2001. Dioxypyrrole- and dioxythiophene-based conducting polymers: Properties and applications. Synth Met 119(1–3):405–406. 42. A. I. Melato, M. H. Mendonça, and L. M. Abrantes. 2009. Effect of the electropolymerisation conditions on the electrochemical, morphological and structural properties of PEDOTh films. J Solid State Electrochem 13(3):417–426. 43. J. R. Reynolds, G. A. Sotzing, and P. J. Steel. 1997. Poly(3,4-ethylenedioxythio phene) (PEDOT) prepared via electrochemical polymerization of EDOT, 2,2′bis(3,4-ethylenedioxythiophene) (BiEDOT), and their TMS derivatives. Adv Mater 9(10):795–798. 44. M. Bouachrine, J.-P. Lere-Porte, J. J. E. Moreau, and M. W. C. Man. 1995. Role of silicon in polymer synthesis: Activation of the oxidative coupling of thiophene by trimethylsilyl substituents; a route to high molecular weight poly(3-alkyl) thiophene. J Mater Chem 5(5):797–799. 45. D. Wasserberg, S. C. J. Meskers, R. A. J. Janssen, E. Mena-Osteritz, and P. Bäuerle. 2006. High-resolution electronic spectra of ethylenedioxythiophene oligomers. J Am Chem Soc 128(51):17007–17017. 46. R. G. Hicks and M. B. Nodwell. 2000. Synthesis and electronic structure investigations of α,ω-bis(arylthio)oligothiophenes: Toward understanding wirelinker interactions in molecular-scale electronic materials. J Am Chem Soc 122(28):6746–6753. 47. M. Turbiez, P. Frère, and J. Roncali. 2003. Stable and soluble oligo(3,4-ethylenedioxythiophene)s end-capped with alkyl chains. J Org Chem 68(13):5357–5360. 48. J. D. Stenger-Smith, C. K. Webber, N. Anderson, A. P. Chafin, K. Zong, and J. R. Reynolds. 2002. Poly(3,4-alkylenedioxythiophene-based supercapacitors using ionic liquids as supporting electrolytes. J Electrochem Soc 149(8):A973–A977.
6911X.indb 342
10/1/10 9:02:07 PM
The Electrochemical Behavior of EDOT and PEDOT
343
49. J. R. Reynolds, K. Zong, J. D. Stenger-Smith, N. Anderson, C. K. Webber, and A. P. Chafin. US Patent 6,965,509, Prior: December 2, 2002. 50. H. Randriamahazaka, C. Plesse, D. Teyssié, and C. Chevrot. 2003. Electrochemical behaviour of poly(3,4-ethylenedioxythiophene) in a room-temperature ionic liquid. Electrochem Commun 5(7):613–617. 51. H. Randriamahazaka, C. Plesse, D. Teyssié, and C. Chevrot. 2004. Ions transfer mechanisms during the electrochemical oxidation of poly(3,4-ethylenedioxythiophene) in 1-ethyl-3-methylimidazolium bis((trifluormethyl)sulfonyl)amide ionic liquid. Electrochem Commun 6(3):299–305. 52. H. Randriamahazaka, C. Plesse, D. Teyssié, and C. Chevrot. 2005. Relaxation kinetics of poly(3,4-ethylenedioxythiophene) in 1-ethyl-3-methylimidazolium bis((trifluoromethyl)sulfonyl)amide ionic liquid during potential step experiments. Electrochim Acta 50(7–8):1515–1522. 53. H. Randriamahazaka, C. Plesse, D. Teyssié, and C. Chevrot. 2005. Charging/ discharging kinetics of poly(3,4-ethylenedioxythiophene) in 1-ethyl-3-methylimidazolium bis((trifluoromethyl)sulfonyl)amide ionic liquid under galvanostatic conditions. Electrochim Acta 50(21):4222–4229. 54. P. Danielsson, J. Bobacka, and A. Ivaska. 2004. Electrochemical synthesis and characterisation of poly(3,4-ethylenedioxythiophene) in ionic liquids with bulky organic anions. J Solid State Electrochem 8(10):809–817. 55. S. Ahmad, M. Deepa, and S. Singh. 2007. Electrochemical synthesis and surface characterization of poly(3,4-ethylenedioxythiophene) films grown in an ionic liquid. Langmuir 23(23):11430–11433. 56. M. Cho, H. Seo, J. Nam, H. Choi, J. Koo, and Y. Lee. 2007. High ionic conductivity and mechanical strength of solid polymer electrolytes based on NBR/ionic liquid and its application to an electrochemical actuator. Sensors and Actuators B: Chemical B128(1):70–74. 57. D. H. Park, H. S. Kim, Y. B. Lee, J. M. Ko, J.-Y. Lee, H.-Y. Kim, D.-C. Kim, J. Kim, and J. Joo. 2008. Light emission of a single strand of poly(3,4-ethylenedioxythiophene) (PEDOT) nanowire. Synth Met 158(3–4):90–94. 58. G. A. Snook and A. S. Best. 2009. Co-deposition of conducting polymers in a room temperature ionic liquid. J Mater Chem 19(24):4248–4254. 59. A. Bund and M. Schneider. 2002. Characterization of the viscoelasticity and the surface roughness of electrochemically prepared conducting polymer films by impedance measurements at quartz crystals. J Electrochem Soc 149(9): E331–E339. 60. A. Bund and S. Neudeck. 2004. Effect of the solvent and the anion on the doping/dedoping behavior of poly(3,4-ethylenedioxythiophene) films studied with the electrochemical quartz microbalance. J Phys Chem B 108(46):17845–17850. 61. A. Bund and R. Peipmann. 2008. Application of PEDOT layers for the electrogravimetric detection of sulphate and phosphate in aqueous media. Electrochim Acta 53(11):3772–3778. 62. A. Ispas, R. Peipmann, A. Bund, and I. Efimov. 2009. On the p-doping of PEDOT layers in various ionic liquids studied by EQCM and acoustic impedance. Electrochim Acta 54(20):4668-4675. 63. K. Wagner, J. M. Pringle, S. B. Hall, M. Forsyth, D. R. MacFarlane, and D. L. Officer. 2005. Investigation of the electropolymerization of EDOT in ionic liquids. Synth Met 153(1–3):257–260.
6911X.indb 343
10/1/10 9:02:07 PM
6911X.indb 344
10/1/10 9:02:07 PM
Index a Aedotron•, 320 AFM; See Atomic force microscopy Agfa-Gevaert AG, 44, 202, 275 Aliphatic alcohols, 68 Aniline black, 6, 7 chloro hydrate, 6 oxidation products, 6 toluidine-contaminated, 5 Antistatic coatings, 194–203 binders, 199 conductivity-enhancing additives, 201 hardness and abrasion, 200–201 solvents, 198 surfactants, 199 use of PEDOT in antistatic coatings, 202–203 Applications, 167–264 antistatic coatings, 194–203 binders, 199 conductivity-enhancing additives, 201 hardness and abrasion, 200–201 solvents, 198 surfactants, 199 use of PEDOT in antistatic coatings, 202–203 auxiliary electrodes, 174 AZO deposition, 189 capacitor, breakdown voltage, 179 change of percolation paths, 242 conductivity enhancement additives, 197, 201 conformational locked ProDOT derivatives, 231 copper plating, processes for, 185 cryogenic temperatures, 176 degradation of high-voltage dielectric oxide layer, model of, 181 deposition methods for PEDOT cathode, 172–179
chemical oxidative polymerization, 175–177 conducting polymer dispersions, 177–179 electrochemical oxidative polymerization, 174–175 direct metallization process, 185 DMS-E®, 185 dual polymer cells, 237 electrochromic behavior, 222–238 blends and layer-by-layer deposition, 234–235 control of optical properties, 226–229 copolymers, 231–232 dual polymer cells, 237 EDOT derivatives, 229–231 electrolytes, 235–236 introduction, 222–226 ion storage materials, 236–237 organic–inorganic hybrid polymers, 233 PEDOT with pendant electrochromic dyes, 233–234 substrates and patterning, 237–238 electrochromism, 222 electroluminescent devices, 204 electroluminescent lamps, 204–205 equivalent series resistance, 169 Grätzel cell, 221 heterojunction solar cells with polymeric anodes, 219 inkjet printing, 240 ionization potential, organic layer, 209 ITO substitution, 188–194 LCD displays, 192 manganese, self-healing of, 182 metal–insulator–metal model, 219 miniaturization, ongoing trends, 169 MIS model, 182
345
6911X.indb 345
10/1/10 9:02:07 PM
346
most common conducting polymers, 168 oligomomeric silanes, 201 organic field-effect transistors, 238–243 introduction, 238–239 PEDOT:PSS as active layer, 242–243 PEDOT:PSS as electrodes, 239–241 PEDOT:PSS as interlayer, 241–242 organic light emitting diodes, 205–216 introduction, 205–207 lifetime restraints by PEDOT:PSS hole injection layers, 213–214 modified PEDOT-based materials for HILS, 214–216 PEDOT:PSS as hole-injection layer, 207–209 PEDOT:PSS–semiconductor interface, 209–213 organic solar cells, 192 PEDOT:PSS in organic solar cells, 216–222 introduction, 216–217 PEDOT:PSS as buffer layer in OSCs, 218–221 PEDOT:PSS in dye-sensitized solar cells, 221–222 PEDOT:PSS as transparent anode in OSCs, 217–218 polymerization of EDOT, model of, 187 printed wiring boards, 185 p-type semiconductor, 238 pyrrole, health risks for handling, 168 Schottky–Mott model, 209, 211 secondary dopants, 201, 218 semiconductor crystalline domains, 242 high-charge carrier mobility, 239 oxidation potential, 221 solid electrolyte capacitors, 167–184 capacitor basics, 169–170 conclusions, 183–184 deposition methods for PEDOT cathode, 172–179 design, 170–172 introduction, 167–168
6911X.indb 346
Index
reformation and high voltage application, 179–182 self-healing and thermal runaway, 182–183 tantalum anode, 172 through hole plating for printed wiring boards, 184–188 TIPS-pentacene, 239 touch screens, 193 transparent conductive oxides, 188 Atomic force microscopy (AFM), 133 inkjet printing, 240 phase-shift, 133 pristine PEDOT:PSS, 154 tapping-mode, 155 ATR; See Attenuated total reflection Attenuated total reflection (ATR), 76 b Batteries, rechargeable, 44 Bayer AG, 92, 202, 208 Beer’s law, 98 Bipolarons, 24, 69 Blasberg Oberflächentechnik GmbH, 185 c Capacitive touch panels, 193 Capacitor(s), tantalum polymer, 267 Charge carrier(s) Au contact and, 241 energy levels and, 140 free, 133, 137, 144, 147 in semiconducting solids, 22 localized to defect sites, 25 mobile, 26, 28 semiconducting solids, 22 transport, scattering processes, 21 Chemical doping, 23 Clevios•, 122 Commercial aspects; See Technical use and commercial aspects Conducting polymers, 2; See also Metals and insulators intrinsic, 189 most common, 168 self-healing ability, 182
10/1/10 9:02:08 PM
347
Index
Conducting polymers, discovery and development of, 1–20 aniline black, 5, 6, 7 oxidation products, 6 cuprene, 11 doped polyacetylene, 10–15 doping, 4 Eastman Kodak patents, 6 electropolymerization, landmark in, 5 emeraldine, 5 extrinsically conductive polymers, 3 indole precursor, 5 intrinsic semiconductor, 11 introduction, 2–4 language problems, 12 Lightfoot patent, 7 melanin, 5 molecular engineering, 3 polyacetylene, 11 polyaniline, 4–8 beginnings of, 5 iodine doping for, 8 oxidation states, 7 polypyrrole, 9–10 scope of historical overview, 1–2 toluidine-contaminated aniline, 5 Ziegler–Natta catalysts, 11 Conductivity enhancement additives, 197, 201 Conductivity enhancement agent, 149 Conjugated polymers charge carriers introduced into, 26 nitrogen atoms in, 23 simplest, 23 Cookson Electronics, 185 Copper plating, processes for, 185 Coulomb repulsion, 116 Crossover temperature, 25 Cuprene, 11 Cyclic voltammetry (CV), 280 d Degenerate polymers, 24 Dimethyl formamide (DMF), 59
6911X.indb 347
carbonylation procedure, 60 EDOT derivatives, 59 monomer synthesis, 48 potentiostatically grown polymer films, 276 pristine PEDOT:PSS films, 152 Dimethyl sulfoxide (DMSO), 126, 137, 152 DMF; See Dimethyl formamide DMSO; See Dimethyl sulfoxide Doping, 4, 149 charge, 334 chemical, 23 counterions, 53 electrochemical, 33 in situ polymerization, 45 iodine, 8, 9 n-, 315, 335 oxidative, 282, 331 paramagnetic polaron state, 93 perchlorate, 85 premature, 76 primary, 149 secondary, 149 Dual polymer cells, 237 e Eastman Kodak patents, 6 EDOP; See 3,4-Ethylenedioxypyrrole EDOS; See 3,4Ethylenedioxyselenophene EDOT; See 3,4-Ethylenedioxythiophene EDOT and PEDOT derivatives with covalently attached side groups, 271–292 alkyl EDOTs, 282–286 cyclic voltammetry, 280 EDOT-CH2Cl and its follow-up products, 280–282 EDOT-CH2OH and its derivatives, 271–279 ferrocene moiety, 276 flash chromatography, 271 General Electric Patent, 275 imidazolinium-modified EDOT, 282 oxidative doping, 282 propanesultone-derived material, 286
10/1/10 9:02:08 PM
348
self-doped PEDOT derivatives, 286 self-doping PEDOT derivatives, 275 water soluble, “self-doping” EDOT derivatives, 286–289 EDTT; See 3,4-Ethylene dithiathiophene EDX analysis; See Energy dispersive x-ray analysis EL devices; See Electroluminescent devices Electrochemical behavior of EDOT and PEDOT, 329–343 Electrochemical polymerization, 44 Electrochemical quartz crystal microbalance (EQCM) technique, 338–339 Electrochromism, 222 Electroluminescent (EL) devices, 204 Electroluminescent lamps, 192, 204–205 Electron paramagnetic resonance (EPR), 332 EDOT1–EDOT3 radical cations, 332 signal development, 73 signal reduction, 156 Electropolymerization, landmark in, 5 Emeraldine, 5, 6 Energy dispersive x-ray (EDX) analysis, 72, 134 Enthone GmbH, 185 EOTT; See 3,4Ethyleneoxythiathiophene EPR; See Electron paramagnetic resonance EQCM; See Electrochemical quartz crystal microbalance technique Equivalent series resistance (ESR), 167 electrolyte capacitors, 167 increase, 177 measurement, 169 performance limitation, 171 polymer capacitors, 268 ESR; See Equivalent series resistance 3,4-Ethylenedioxypyrrole (EDOP), 306–308 3,4-Ethylenedioxyselenophene (EDOS), 309–310
6911X.indb 348
Index
3,4-Ethylenedioxythiophene (EDOT), 14; See also EDOT and PEDOT derivatives with covalently attached side groups; In situ polymerization of EDOT to PEDOT absorption spectrum, 136 beginnings, 47 chemical oxidation, 77 derivatives of with substituents at thiophene ring, 293 dilithiation of, 59 dimer, 94 dimerization, 54 EI-mass spectrum, 52 electrochemical oxidation, 67 first anodical polymerization, 329 halogenation, 62 imidazolinium-modified, 282 oligo-, 137 oxidation, 92, 332 phenyl, 284 polymerization of EDOT in presence of oxidants, 168 structure, 14 synthesis of from oxalic acid ester, 48 tetra-n-butyl ammonium hexafluorophosphate, 85 trimer, 94 uracil-modified, 281 3,4-Ethylenedioxythiophene (EDOT), electrochemical behavior of, 329–339 doping charge, 334 electrochemical quartz crystal microbalance technique, 338–339 n-doping, 335 oxidative doping, 331 Stranski–Kastanov model, 335 supercapacitor, 337 3,4-Ethylenedioxythiophene (EDOT), monomer, synthesis of, 47–66 chemical properties, 53–63 di-halogenation, 62 dioxepane ring formation, 50 doping counterions, 53 EDOT dimerization, 54 Gogte synthesis, 47
10/1/10 9:02:08 PM
349
Index
Grubbs ruthenium catalyst, 49 Mitsunobu reaction, 48, 50 monomer synthesis, 47–50 Negishi coupling, 59, 60 physical properties, 50–52 Williamson ether synthesis, 50 3,4-Ethylenedioxythiophene (EDOT), oxidation of to PEDOT, 67–81 aliphatic alcohols, 68 bipolarons, 69 gravimetric experiments, 73 hypervalent iodine compounds, 70 in situ doping, 78 neutral, undoped PEDOT by oxidative polymerization, 76–79 organometallic route to PEDOT, 74–76 oxidative polymerization and doping, 67–72 polymerization mechanisms, 73 premature doping, 76 proton scavenging, 71 reductive pathway, 76 “self-oxidation” of EDOT halogen derivatives, 72–74 toluenesulfonic acid, 68 vapor phase polymerization, 71 3,4-Ethylenedithiathiophene (EDTT), 304–306 3,4-Ethyleneoxythiathiophene (EOTT), 301–303 Extrinsically conductive polymers, 3 f Fabric, printing on, 5 Fermi–Dirac distribution, 22 Fermi energy, 145 Fermi glass, 26 Fermi level, 21, 25, 209 FETs; See Field effect transistors Field effect transistors (FETs), 147 g Gel permeation chromatography (GPC), 76 General Electric Patent, 275, 286
6911X.indb 349
Glucose oxidase (GOx), 101 Gogte synthesis, 47 GOx; See Glucose oxidase GPC; See Gel permeation chromatography GPE; See Guest polyelectrolyte Grätzel cell, 221 Grignard metathesis (GRIM), 32 GRIM; See Grignard metathesis Grubbs ruthenium catalyst, 49 Guest polyelectrolyte (GPE), 115 h HAADF scanning transmission electron microscopy; See High angle annular dark field scanning transmission electron microscopy H.C. Starck Clevios GmbH, 122 H.C. Starck GmbH, 106 HDPE; See High density polyethylene High angle annular dark field (HAADF) scanning transmission electron microscopy, 133 High density polyethylene (HDPE), 10 Highest occupied molecular orbital (HOMO), 137 Hoechst AG patents, 35 HOMO; See Highest occupied molecular orbital Host polyelectrolyte (HPE), 115 HPE; See Host polyelectrolyte Hypervalent iodine compounds, 70, 97 i ICPs; See Intrinsically conductive polymers In situ polymerization of EDOT to PEDOT, 91–111 alternative cerium salt oxidants, 96 aromatic and analog nonaromatic PEDOT derivatives, 108 Beer’s law, 98
10/1/10 9:02:08 PM
350
EDOT derivatives with mesogenic side chains/compound types, 107 hypervalent iodine compounds, 97 in situ polymerization of EDOT derivatives and relatives, 102–109 liquid crystalline behavior, 106 metal salt oxidants, 91 paramagnetic polaron state, 93 properties of in situ PEDOT, 97–101 synthesis of in situ PEDOT, 91–97 Indium tin oxide (ITO), 192, 218 Inkjet printing, 240 Insulators; See Metals and insulators Intrinsically conductive polymers (ICPs), 3 Iodine doping, 8, 9 Iron-III, 91 ITO; See Indium tin oxide j Jahn–Teller effect, 23 k Kelvin probe, 209, 212 Kodachrome films, 45 Kodak, 44 Koser’s reagent, 97 l LCD displays, 192 Leucoemeraldine, 6 Lightfoot patent, 7 Localization length, 145 Lowest unoccupied molecular orbital (LUMO), 137 m MDOT; See 3,4-Methylenedioxythiophene Melanin, 5 Metal–insulator–metal (MIM) model, 219 Metal–insulator–semiconductor (MIS), 181
6911X.indb 350
Index
Metal-oxide-semiconductor fieldeffect transistor (MOSFET) structures, 194 Metal salt oxidants, 91 Metallic diffusion, typical behavior for, 25 Metals and insulators, 21–30 activation energy, 28 bipolaron formation, 24 carrier diffusion, 27 charge carriers, semiconducting solids, 22 charge carrier transport, scattering processes, 21 charge transport model, 26 chemical doping, 23 conjugated polymers, 22–24 critical energy, 28 crossover temperature, 25 dopants, 22 energy gap, 22, 24 Fermi–Dirac distribution, 22 Fermi glass, 26 Fermi level, 21, 25 heterogeneous disorder model, 27 Jahn–Teller effect, 23 low temperature semiconductor– metal transition, 25 metals, semiconductors, and insulators, 21–22 Mott’s temperature law, 28 order and disorder, 26–29 percolation model, 27 polaron formation, 24 polyacetylene, changes in conductivity and thermopower, 25 polyaniline, spin-unpairing, 24 polypyrrole crystallinity, 26 randomly disordered systems, 26 semiconductor, distinction between insulator and, 22 solitons, 24 temperature-dependent conductivity, 24–25 thermopower, 25 valence band, 22 valence electrons, 22 variable range hopping model, 28
10/1/10 9:02:08 PM
351
Index
3,4-Methylenedioxythiophene (MDOT), 42, 50, 293–295 chemical oxidation, 330 difficulties in electropolymerizing, 329 failure synthesizing, 44 structure, 229, 331 Williamson ether synthesis, 50 N-Methyl-2-pyrrolidone (NMP), 153, 304 MIM model; See Metal–insulator–metal model Miniaturization, ongoing trends, 169 MIS; See Metal–insulator– semiconductor MIS model, 182 Mitsunobu reaction, 48, 50 Model Bloch’s eigenstates, 145 change of percolation paths, 242 charge transport, 26 degradation of high-voltage dielectric oxide layer, 181 electron gas, 137 heterogeneous disorder, 27 lamellae, 132 metal–insulator–metal, 219 MIS, 182 orthorhombic unit cell, 86 PEDOT:PSS, 133 percolation, 27 polymerization of EDOT, 187 Schottky–Mott, 209 Stranski–Kastanov, 335 variable range hopping, 28, 145, 146 MOSFET structures; See Metal-oxidesemiconductor field-effect transistor structures Mott’s temperature law, 28 n Negishi cross-coupling, 60 NMP; See N-Methyl-2-pyrrolidone o OFETs; See Organic field-effect transistors OLEDs; See Organic light emitting diodes Oligo-EDOTs, 137
6911X.indb 351
Oligotron•, 320 Organic Electronics Association, 268 Organic field-effect transistors (OFETs), 238–243 introduction, 238–239 PEDOT:PSS as active layer, 242–243 PEDOT:PSS as electrodes, 239–241 PEDOT:PSS as interlayer, 241–242 semiconductor-based, 242 Organic light emitting diodes (OLEDs), 205–216 introduction, 205–207 lifetime restraints by PEDOT:PSS hole injection layers, 213–214 modified PEDOT-based materials for HILS, 214–216 PEDOT:PSS as hole-injection layer, 207–209 PEDOT:PSS–semiconductor interface, 209–213 Organic solar cells (OSCs), 192 Ormecon GmbH, 8 OSCs; See Organic solar cells Oxalic acid ester, synthesis of EDOT from, 48 Oxidative doping, 282, 331 p Panipol Oy, 8 Patent(s) alternative synthesis for EDOTCH2OH, 272 aniline black, 5 applications antistatic layers, 203 EOTT, 302 peroxides, 97 self-doped PEDOT derivatives, 286 special conducting polymers, 14 supercapacitor, 337 diketopyrrolopyrroles, 320 Eastman Kodak, 6 electrochemical sensors, 280 General Electric, 275, 286 German application, 44 Hoechst AG, 35 Lightfoot, 7 metal salt oxidants, 91
10/1/10 9:02:08 PM
352
numbers, 4 oxygen-bearing polythiophenes, 34 succinic anhydride, 275 PDMS; See Poly(dimethylsiloxane) PEC; See Polyelectrolyte complex PEDOT; See Poly(3,4ethylenedioxythiophene) PEDOT:PSS, 113–165 activation spectrum, 127, 128 Clevios dispersions, 122 conductivity enhancement agent, 149 Coulomb repulsion, 116 difference between charged complexes, 113 dispersions, 113–122 commercial PEDOT:PSS types and their properties, 122 introduction, 113 polyelectrolyte complexes, 113–117 synthesis of PEDOT:PSS complex, 117–122 dissociated sulfonate groups, 144 doping, 149 electrical properties, 144–149 conductivity, 144 free charge carrier mobility, 147 microscopic model for conductivity in PEDOT:PSS, 144–146 threshold currents, 148–149 electron gas model, 137 electronic states, 136–144 energy levels in PEDOT, 137–141 optical constants, 142–143 UV-VIS spectra, 136–137 vibrational spectra, 143–144 Fermi energy, 145 field effect transistors, 147 guest polyelectrolyte, 115 host polyelectrolyte, 115 lamellae model, 132 localization length, 145, 146 natural solvent for polyelectrolytes, 114 oligo-EDOTs, 137 in organic solar cells, 216–222 introduction, 216–217
6911X.indb 352
Index
PEDOT:PSS as buffer layer in OSCs, 218–221 PEDOT:PSS in dye-sensitized solar cells, 221–222 PEDOT:PSS as transparent anode in OSCs, 217–218 primary doping, 149 properties, 123–149 deposition of PEDOT:PSS, 123 electrical properties, 144–149 electronic states, 136–144 thin-film properties, 123–136 properties of films including secondary dopants, 152–156 atomic force microscopy and scanning tunnel microscopy, 154 conductivity as function of temperature, 152 optical characterization of PEDOT:PSS films, 153 surface analysis of PEDOT:PSS films, 153–154 work function and electron paramagnetic resonance, 155–156 x-ray diffraction, 152–153 secondary doping, 149–158 chemical nature of secondary dopants in PEDOT:PSS, 150–152 discussion, 156–158 introduction, 149 properties of PEDOT:PSS films including secondary dopants, 152–156 static light scattering, 115 thin-film properties, 123–136 mechanical properties, 130 morphology (surface and bulk), 131–136 thermal and lifetime stability, 123–126 UV stability, 126–128 water uptake, 128–130 variable range hopping model, 145, 146 wavelength of radiation, 127 write-once read-many-times electrically addressable memories, 148
10/1/10 9:02:08 PM
Index
Pernigraniline, 6 Polaron formation, 24 Polyacetylene, 11 changes in conductivity and thermopower, 25 dicationic form, 13 doped, 10 first recognition of, 11 Polyaniline acid–base chemistry, 24 beginnings of, 5 conductivity, 8 crystallinity, 26 drawback, 8 early research, 6 hue, 197 iodine doping for, 8 nitrogen atoms in, 23 oxidation states, 6, 7 protonation, 149 spin-unpairing, 24 structure, 168 as transparent anode in PLEDs, 192 Poly(dimethylsiloxane) (PDMS), 241 Polyelectrolyte complex (PEC), 113 arrangements, 114 exchange reactions, 117 gel particles in dispersion, 115, 121 ladder type, 114 precipitate, 120 salt concentrations, 116 scrambled egg type, 114 stable, 113 Poly(3,4-ethylenedioxythiophene) (PEDOT), 38; See also In situ polymerization of EDOT to PEDOT; PEDOT:PSS antistatic coating, 202 Bayer development, 176 cathode, deposition methods for, 172–179 chemical oxidative polymerization, 175–177 conducting polymer dispersions, 177–179 electrochemical oxidative polymerization, 174–175 counterions, 83–89
6911X.indb 353
353
chemically polymerized PEDOT, 86–87 conductivities obtained in electrochemical polymerization of EDOT, 84 electrochemically polymerized PEDOT, 83–86 redox-active counterion, 86 derivatives; See EDOT and PEDOT derivatives with covalently attached side groups electrochemical behavior of, 329–339 doping charge, 334 electrochemical quartz crystal microbalance technique, 338–339 n-doping, 335 oxidative doping, 331 Stranski–Kastanov model, 335 supercapacitor, 337 etched board, 187 imidazolinium ionic-liquid moieties, 282 invention, short history of, 41–46 antistatics, drawbacks, 44 electrochemical polymerization, 44 in situ polymerization, 44 Kodachrome films, 45 oxygen-bearing substituent, 41 photographic films, 45 rechargeable batteries, 44 neutral, 76 oxidizing agent, 120 polymerization reaction, 175 structure, 168 use of in antistatic coatings, 202–203 viologen side group modified, 234 worldwide demand, 265 Polymer(s); See also Conducting polymers cells, dual, 237 charge states, 24 definition of, 2 degenerate, 24 electrolyte capacitors, 177 extrinsically conductive, 3 intrinsically conductive, 3, 35
10/1/10 9:02:09 PM
354
oxidation potentials, 33 paramagnetic behavior, 9 saturated, 22 Polypyrrole, crystallinity, 26 Polystyrenesulfonic acid (PSS), 45, 85, 286 Polythiophenes, 31–40 akoxy-substituted thiophenes, 34 electrochemical doping, 33 Grignard metathesis, 32 Hoechst AG patents, 35 introduction, 31–33 oxidative coupling reaction, 31 oxygen-bearing polythiophenes, 34 oxygen-substituted polythiophenes, 33–38 polyheterocycles, conductivity of, 32 polymerization reaction, 32 tetrachloroferrates, 36 Primary doping, 149 Printed wiring boards (PWB), 184–188 ProDOT; See 3,4Propylenedioxythiophene 3,4-Propylenedioxythiophene (ProDOT), 50, 295–299 Proton scavenging, 71 PSS; See Polystyrenesulfonic acid PSS-Na; See Sodium salt of poly(styrenesulfonic acid) PWB; See Printed wiring boards Pyrrole, health risks for handling, 168 r Randomly disordered systems, 26 Rechargeable batteries, 44 Redox-active counterion, 86 Reductive pathway, 76 Resistive touch panels, 193 s SAM; See Self-assembled monolayer Saturated calomel electrode (SCE), 274 Scanning electron microscopy (SEM), 170, 274, 336 Scanning transmission electron microscopy (STEM), 72, 133
6911X.indb 354
Index
Scanning tunnel microscopy (STM), 134, 154 SCE; See Saturated calomel electrode Schottky–Mott model, 209, 211 Secondary dopants, 201, 218 Self-assembled monolayer (SAM), 240, 313 SEM; See Scanning electron microscopy Semiconductor(s) crystalline domains, 242 distinction between insulator and, 22 high-charge carrier mobility, 239 hopping of charges, 25 inorganic, 23 intrinsic, 11 light absorbing, 212 metal–insulator–, 181 n-type, 22 oxidation potential, 221 p-doped, 22 PEDOT:PSS–, 209, 220 poly(3-hexylthiophene) as, 32 p-type, 238 Single wall carbon nanotube (SWCNT) films, 220 Sodium salt of poly(styrenesulfonic acid) (PSS-Na), 45 Solar cells (organic), PEDOT:PSS in, 216–222 introduction, 216–217 PEDOT:PSS as buffer layer, 218–221 PEDOT:PSS in dye-sensitized solar cells, 221–222 PEDOT:PSS as transparent anode, 217–218 Solitons, 24 STEM; See Scanning transmission electron microscopy STM; See Scanning tunnel microscopy Stranski–Kastanov model, 335 Surfactants, 199 SWCNT films; See Single wall carbon nanotube films t Tantalum polymer capacitors, 267 TCNQ; See Tetracyanoquinodimethane
10/1/10 9:02:09 PM
355
Index
TCOs; See Transparent conductive oxides TDA Research, 320 Technical use and commercial aspects, 265–269 electrochemical process, advantages of, 265 PEDOT, 266 tantalum polymer capacitors, 267 Teflon, 10 TEM; See Transmission electron microscopy Tetracyanoquinodimethane (TCNQ), 10, 167 Tetrahydrofuran (THF), 152 Textile dye, 5 THF; See Tetrahydrofuran Touch panels capacitive, 193 conductive layers, 194 ITO substitution, 192 projected lifetime, 194 resistive, 193 Touch screens, 193 Transmission electron microscopy (TEM), 153 Transparent conductive oxides (TCOs), 188 u Ultraviolet photoelectron spectroscopy (UPS), 101, 209 UPS; See Ultraviolet photoelectron spectroscopy v Valence band, 22 Vapor phase polymerization (VPP), 71 Variable-angle spectroscopic ellipsometry (VASE), 153 Variable range hopping (VRH), 28, 145, 146 VASE; See Variable-angle spectroscopic ellipsometry VDOT; See Vinylenedioxythiophene
6911X.indb 355
Vinylenedioxythiophene (VDOT), 49, 299–300 VPP; See Vapor phase polymerization VRH; See Variable range hopping w WORM; See Write-once read-manytimes electrically addressable memories Write-once read-many-times electrically addressable memories (WORM), 148 x XDOTs, EDXTs, EDIXs, and 2(5)-X(2)EDOTs, 293–328 diketopyrrolopyrroles, 320 2,5-disubstituted EDOT derivatives, 311–322 3,4-ethylenedioxypyrrole and its derivatives, 306–308 3,4-ethylenedioxyselenophene, 309–310 3,4-ethylenedithiathiophene, 304–306 3,4-ethyleneoxythiathiophene, 301–303 3,4-methylenedioxythiophene, 293–295 n-doping cycling conditions, 315 ProDOT (propylenedioxythiophene) derivatives, 295–299 vinylenedioxythiophene and benzo-EDOT, 299–300 XPS; See X-ray photoelectron spectroscopy X-ray diffraction (XRD), 131 crystalline regimes, 131 electropolymerization, 336 X-ray photoelectron spectroscopy (XPS), 72, 101 XRD; See X-ray diffraction z Ziegler–Natta catalysts, 11
10/1/10 9:02:09 PM
6911X.indb 356
10/1/10 9:02:09 PM