Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. Functionalised Polythiophenes: Synthesis, Characterisation and Applications Amy Marisa Ballantyne 2005 Functionalised Polythiophenes: Synthesis, Characterisation and Applications A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy ID Chemistry at Massey University, Palmerston North New Zealand Amy Marisa Ballantyne 2005 Abstract Conducting polymers display properties such as high conductivity, light weight and redox activity giving them great potential for use in many applications. Polythiophenes have proved to be particularly useful because they are readily functionalised and have good chemical stability. The purpose of this work was to investigate the effect of electron-withdrawing and electron-donating substituents on the synthesis and properties of polythiophenes. Initial work entailed the synthesis of a series of styryl-substituted terthiophenes. Polymerisation of these materials using both chemical and electrochemical methods was found to produce predominantly short chain oligomers (n < 4) and insoluble material that could not be further processed. An analogous senes of styryl-substituted terthienylenevinylene materials were electrochemically oxidised for comparison to the terthiophene series. These materials were also found to produce predominantly dim er and short oligomers, but with the expected higher conjugation length than the corresponding terthiophene oligomers. To enhance polymerisation and increase the solubility of the resulting materials, the polymerisation of styryl-terthiophenes with alkyl and alkoxy functionalities was investigated. The properties of the resulting polymeric materials were determined using electrochemistry, mass spectrometry, spectroscopy and microscopy. The alkoxy substituted polymer was found to have a longer average polymer length than the corresponding alkyl derivative (-n = 1 1 compared to -n = 6), but was less soluble (78% compared to 100%). It was found, however, that by increasing the alkoxy chain length from 6 carbons to 10 carbons, the solubility of the polymer could be increased to 97% without affecting the average polymer length. The alkoxy-substituted polymers were observed to be very stable in the oxidised, conducting state compared to the alkyl-substituted polymer, which appeared to be more stable in the neutral, non­ conducting state. It was found that these soluble materials could be separated into fractions of different length polymers by usmg sequential soxhlet extractions m different solvents. Preliminary investigations were made into the suitability of these soluble oligomeric and polymeric materials for use in photovoltaic, actuator and organic battery applications and promising results were achieved for actuator and battery functions. In addition, the solubility of these materials allowed nano- and micro-structured fibre and fibril surfaces to be prepared for use in high surface area electrodes. ii Acknowledgements The last three years has been a journey on which I have met and worked with so many fantastic people and friends. I would like to take this opportunity to acknowledge these people who have generously given me their time, moral support and financial assistance . Firstly, I would like to thank my supervisors Professor David Officer and Associate Professor Simon Hall. David, you have always been supportive and encouraging to me. Even with your hectic schedule, you have made time to see me and I really appreciate this. Simon, I am very grateful for your time and assistance, particularly during the arduous writing process. I would also like to acknowledge my '3rd, (lab) supervisor, Dr. Warwick Belcher, who started me off on this project and familiarised me in the lab. I would like thank the Institute of Fundamental Sciences, Massey University, the Nanomaterials Research Centre, Massey University and the MacDiarmid Institute for Advanced Materials and Nanotechnology for my scholarships and funding for this project. To the members of the NRC, past and present: Shannon Bullock, Wayne Campbell, Beatrice Eccles, Sonja Ensink, Sanjeev Gambhir, Daina Grant, Susan Habas, Ken lolley, Fabio Lodato, Donna Macpherson, Giovanna Moretto, Yvonne Ting, Mark Vigneswaren, Klaudia Wagner, Pawel Wagner and Amy Watson. You have all helped me in some way, from synthesising materials used in this project, to advice, support and encouragement. Thank you. I would like to thank the members of the IPR! at the UOW for their time and assistance, namely Professor Gordon Wallace, Chee Too, Caiyun Wang, George Tsekouras, Yanzhe Wu (Richard), Violeta Misoska, Phil Smugreski, lun Chen and Peter Innis. I was always made to feel welcome and 'one of the bunch'. vi I would to like thank Dr. Paul Dastoor and Chris McNeill at the University of Newcastle for their time, expertise and hospitality. I would like to thank Doug Hopcroft at HortResearch for taking SEM images of my samples, Keith Gordon and Tracey Clarke from the University of Otago for useful discussions on computational chemistry of the compounds used in this study, and all the Massey University IFS staff for their help during the last few years. Thank you to my friends, flatmates and family who have graciously accepted the emotional effects of the ups and downs of research and have supported me throughout. In particular, thank you to Graeme Ballantyne, Ingrid Dunckley and Giovanna Moretto for proof reading. A special thank you also to David Sherwin, who picked me up at the end of my first year, got me on track and supported me right to the end. Finally, I would like to thank my parents, Graeme and Deborah Ballantyne, for the inspiration to start this project and the love and support to finish it. I love you both very much. © vii Table of Contents ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i DECLARATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. .. . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . iii ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . vi TABLE OF CONTENTS . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii LIST OF FIGlJRES . .. ... .. .... . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . .. . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . .. . . . . . xii LIST OF TABLES . . . . . . . . . . . ... . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxv LIST OF SYMBOLS . . . . . . . . . . .. . . . . . . . .. .... . . . . . . . .. . .. .. . . . . .. . .... .... .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvi LIST OF ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . ... . . . . . . . .. . . . . . . . . . .. . . . . . . . . . . . . .. . . . xxviii MONOMER ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . ... . .. . . .. .. . . ... .. . . ... . . . . . . . . . . . . . . . . . . . . . . . . xxxi CHAPTERl INTRODUCTION . . . . . . . . . .. . . . . . . . . . .. . . ... .. . .. . . . . ... . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 . 1 INTRODUCTION . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . ..... . . . . . . . ... .. . . . .... . ..... . . ... . ..... .. . . . .. . . . 1 1 . 2 CONDUCTING POLYMERS . . . . . . . ... . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 .3 TIllOPHENE BASED POLYMERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3.1 Polymerisation ................................................................................................. 7 1.3.2 Polymerisation methods . .... .. . . . . .... . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. . .... .. . . . . . .. 11 1.3.3 Effect of substituents on polythiophenes properties .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.3.4 Styryl-substituted polythiophenes . . . . . . . . . . . .. . .. . . . ... . . . . . .. . .... . . . ..... . . . .. . . . . ... .. . .. . .. . .. . .. 15 1.3.5 Alkyl and alkoxy substituents.. ......................................................................... 18 1.3.6 Thienylenevinylene based conducting polymers . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 l A SCOPE OF WORK . . . . . . . . .. . . . . . . . .. . . . . . . . . . . . .. . ... . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 CHAPTER 2 EXPERIMENTAL METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 23 2. 1 ELECTROCHEMICAL TECHNIQUES ..... ........... . . .. .. . . . . ....... . . . . . . . . . . . . . . . . .... . . . . .. . . . . . 23 2.1.1 Introduction . . . . . . . . . . . . . . . . . . . . .. . . . . . . . .. . . . . . . . . . . . . . . . .. . ... . .. . . . . .. . . . . .. . . . . . .... ... . . . . .. .. . . . . . .. .. . ... 23 2.1.2 Cell design . . . . . . ... . . . . . . . . . . . . . .. ... . .. .. . . . . .. . . ... . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.1.3 Electrochemical techniques ... . . . . . . . . . . ... .. . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.2 MALDI-TOF MASS SPECTROMETRY ... . . . . . . .. . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 2.2.1 Introduction . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . ... . . . .... . . ..... . . . . . . . . . . . . . . . . . . ... . .. . . . ..... . . . . . . . . . . . .. .. . 33 2.2.2 Instrumentation . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. 33 2.2.3 Data acquisition and sample preparation . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 viii 2.2.4 Calculation of polydispersity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .. .. . 39 2.3 IH NMR SPECTROSCOPY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . .. . .. . . . . .. . . . .40 2.4 UV-VIS-NIR SPECTROSCOPY . . . ........ . . . . . .. . . . . . . . . . . ..... . .. ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42 2.5 IMAGING TECfIN1QUES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . ....... ... . . . . . . . . . . 44 2.6 DEPOSITION TECHNIQUES . . . .. ..... . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . .45 2.7 ELECTROSPINNING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... . . . . . . ... . .. . .. . . . . . . .47 2.8 PHOTOVOL TAlCS DEVICES .. . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . ... ..... .48 CHAPTER 3 SYNTHESIS AND POLYMERISATION OF A SERIES OF STYRYL-SUBSTITUTED TERTIllOPHENES . . . .. . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 50 3 .1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. .. . . . . . . . 50 3 .2 SYNTHESIS OF MONOMERS . . .... ....... ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3 . 3 POLYMERISATION USING CHEMICAL OXlDATION . . . . . . . . ... . . . . .. ... .. . . . . . . . . . . . . . . . 53 3.3.1 Polymerisation and reduction methods .. ... .. .. . . .. . . . .. .. ... . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.3.2 Characterisation of soluble fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . ... . ... . . . ... . . . . . -.. -.. ?r---_ 3.3.3 Attempted purification of dimer . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . 58 3.3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 3 .4 POLYMERISATION USING ELECTROCHEMICAL METHODS . . . . . . . . . . . . . . . . . . . . . . 63 3.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... . . . . . . . . . . . . . . . .. . . 63 3.4.2 Electrochemical growth. . . . .. . . . . . .. .. . . . . . .. . . . . .. . . . . .... .. . . . . .. . .. .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 63 3.4.3 Characterisation of electrochemically deposited materials .............................. 71 3.4.3.1 Electrochemical characterisation ... . . . . .. . . . . . . . . ..... . ... . ... ... . .. . . . . . . . . . . . . . . . . . . .. .. . . ... . . ..... . 71 3.4.3.2 Analysis by UV-VIS-NIR spectroscopy ... . . . ... . ........ . . . ... . . . . .. . ... . .. . . . . . .. . . . . ... ...... .. . . . 79 3.4.3.3 Scanning electron microscopy (SEM) . . . . .... . . . . ... . . . .. . . . .. . . .. . . . . . . . ... . ..... .. ..... . ... . ... . . . . 85 3 .5 CONCLUSIONS . . .. . . . . . . . . . ... . . .. .. . . .. . . . . . .. . . .. . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3.6 EXPERIMENT AL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . .. . . . . . . . . . . . . . . 91 3.6.1 Reagents and materials . . . . . . . . . . . . . . . .... . . . . . . . . .. . . . . . ...... . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .... . 91 3.6.2 SyntheSiS of styryl-substituted terthiophene ..................................................... 91 3.6.3 Polymerisation of styryl-substituted terthiophene monomers ........................... 95 3.6.4 Electrochemical synthesis ............................................................................... 97 3.6.5 MALDI-TOFMS ............................................................................................. 98 3.6.6 UV-VIS-NIR spectroscopy ............................................................................... 99 3.6. 7 I H NMR spectroscopy .... . . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ . .... 99 3.6.8 SEMimaging .................................................................................................. 99 CHAPTER 4 ELECTROCHEMICAL POLYMERISATION OF A SERIES OF STYRYL-SUBSTITUTED TERTlllENYLENEVINYLENES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 4.1 INTRODUCTION . . ... . . . . . . ... . ........ . . ...... . . . . . . .. . .. ... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . .. . .. 100 4.2 ELECTROCHEMICAL DEPOSITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . .. . . . . . . . . . . . . . . . . . . . . 102 4.3 CHARACTERISATION OF FILMS ... . . . . . .. . .. . . ... .... . . . . . . . . ... . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 4.3.1 MALDI-TOF MS characterisation ................................................................. 1 06 4.3.2 Cyclic voltammetry ....................................................................................... 106 3.3.3 UV-VIS-NIR spectroscopy ............................................................................. 11 0 ix 4.4 CONCLUSIONS . . .. .. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 4.5 EXPERIMENTAL . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . ..... . .... .. . ... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 4.5.1 Reagents and materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 4.5.2 Synthesis of materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 4.5.3 UV-VlS-NIR spectroscopy ............................................................................. 119 4.5.4 MALDI-TOFMS .. . . . ... . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 CHAPTERS POLYMERISATION OF ALKYL- AND ALKOXY-SUBSTITUTED STYRYLTERTHIOPHENES . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 120 5.1 INTRODUCTION . . .. .... ... . . . .. . . . . . .. . ... . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . 120 5.2 CHEMICAL POLYMERISATION OF OCJ)ASTT . . . . .. . . ... .. ... . . .... . ....... . . . . . . . . . . .... 122 5.2.1 Polymerisation and reduction procedure . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . ... . . . . . . . . . ... . .. 122 5.2.2 Characterisation of the soluble fraction of polyOCJ)ASTT. . . . . . . . . . . . . . . . . . . . . . . . . . . 124 5.2.3 Improvement of the polyOCJ)ASTT soluble fraction . . . . . . . . . . . . . . . . . . . . . . . . . .. .... . ... . . 128 5.2.4 Polymer separation according to chain length .. ...... . ... . ... . . .... . . . . . ...... . . . . . . . . . . . . . 129 5.2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 5.3 CHEMICAL POLYMERISATION OF C7DASTT. . . . . . .... . .... . ...... ... . . ....... ..... . .. ..... 137 5.3.1 Introduction . .. . . . . . . . . . . . . . .... ...... . . . . . . . .. . . . .. . . .. ... .... . . . . .. . . . . . . . . . . . . . .. . . . . . ... . . . ...... . ..... . . . 13 7 5.3.2 Polymerisation and reduction. ....................................................................... 1 3 7 5.3.3 Polymer separation according to oligomer length . . ... . . . . . . .... .... . .. .. . . . ... . . . . . ...... 140 5.3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . .. . . . . ........ . ... . . .. .. ..... . . . ...... . .... . ...... . ... . . .......... . . ....... 144 5.4 CHEMICAL POLYMERISATION OF OCIODASTT ........................................... 146 5.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. .. . . . . . . . ... . . . ... . . ... . . . . .. . . .. . . . .. . .. . . . . . . . . . . . . . . . 146 5.4.2 Polymerisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 5.4.3 Polymer separation according to oligomer length . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 151 5.4.4 Conclusions . . . . . . . . . . . . .. .. . ... . ... . .. . . . . . . . . . .. . . . . . .. . . . . . . . . . . . . . ... . . . . . . . ... . . . . . . . . .. ....... . . .... . ... 157 ).) ELECTROCHEMICAL POLYMERISATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . .. . . . . . . 158 5.5.1 Introduction . . . . . . . . . . .. . ... . .. . .. ... . .. . . ... .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . ... ... . . .... . . . . ..... . . . . ..... 158 5.5.2 Electrochemical growth and post-growth analysis ofpolyC7DASTT . ... . . . . . . . . . . 159 5.5.3 Electrochemical growth and post-growth analysis of polyOCJ)ASTT . . . . . . . . . . . 164 5.5.4 Electrochemical growth and post-growth analysis of polyOCJ(fJASTT . . . ....... 171 5.5.5 Summary of electrochemically polymerised material . . . . . . . . . . . ... . . .. . . . . . . . . . . . . . . . . . . . 181 5.6 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 5.7 EXPERIMENTAL . . . . . . . . .. . . . . .... ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ... . . . . ... . . ... . 185 5.7.1 Reagents and materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 185 5.7.2 Chemical synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 5.7.3 Electrochemical synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 5.7.4 MALDI-TOFMS . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 190 5.7.5 UV-VlS-NIR instrumentation and spectroelectrochemistry . . . . . . . . . .. . .. . ... . .. . .. . . . . 190 5.7.6 SEM imaging . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 CHAPTER 6 DEVICE FABRICATION AND ANALySIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 192 6.1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . ... . .. . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 x 6.2 PHOTOELECTROCHEMICAL CELLS . . . . . . . . . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 6.2.1 Introduction . . ... . . . . . ... ... . . . . . . . . . . . . ... . . ......... . . .. . . . . . . . .. . . . ... .. . . .. . . ... . . . . . . . . . . . . . ...... . . . . . . . 194 6.2.2 Device assembly and materials . . . . . . . . . . . . . . . . . . . .. .. .. . . . . . . . . . ... . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . 196 6.2.3 NMe�TT oligomer films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 6.2.4 PolyOCJ()DASTT films ................................................................................... 198 6.2.5 Styryl thienylenevinylene derivatives . ... . . .. . . . . . .. . . . . . . . .. . . . .. . . . . . .... . ... ........... . . . ... . . 199 6. 2. 6 Conclusions . . . .. . . ... . . . . . . . . . . . . . . .. . .. . . . . . . .. . .. . . . . . . . . . . .. . .. . . . . . . . . . .. . . . . . ... . . . . . .. . . . . . . . . . . . . . . .. . 203 6.3 ELECTROMECHANICAL ACTUATORS . . . . . . . . . . . . . .. . .. . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 6. 3.1 Introduction .. . . . . .. . . . . . . . . . . . . .. . ....... . . . . . . .. . . . . . . . . . . .. ... ... . . . . . . . . . . .. . . . . . .... . . . . .. .. . .. . . . . .. . ... 204 6.3.2 Testing procedures and terminology .. . ........ . . . . . . . ... . . ... . . . .. . . ...... . .... ........ . ..... . . . . 205 6.3.3 Bilayer benders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . .. . . . . . . . . ... . . . .. . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . ... 208 6.3.4 Fibres . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 6.3.5 Free-standingfilm with incorporated wire ..................................................... 219 6.3.6 Conclusions .. . . . ... . . . . . . . . . . ... . . . . . . . . ... . . . . . .. . ....... . . . . . . . . . . . . . . . . . . . . . .. . . . . . ... . .. . . . . . . . . . . . . . . . . . 224 6.4 BATTERIES ....................................................................................................... 226 6.4.1 Introduction . . . . . . . ... . ... . .... . . . . . . . . . . . .. . . . . . . . . . ... . . . . . . . . . . . . . . . . .. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 6.4.2 Cell construction, testing procedures and terminology . . . .. ... .. . . . . . .. . . ..... . . . ...... . 228 6.4.3 Analysis of electrodes .................................................................................... 229 6.4.4 Charge/discharge characteristics . . . . . . . . .. . . . . . . . .. .. .. .. . . . . . . . . . . . . . ... . . ... . . . . . . . . . . .. . . . . . . . . 231 6.4.5 Cycle life . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . .. . . . ... ... . ... ... . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . 233 6. 4. 6 Conclusions . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . .. . .... . . . . . . . . .. . . . . . . . .. . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 234 6.5 MICRO- AND NANO-STRUCTURED SURFACES .......................................... 225 6.5.1 Introduction .................................................................................................. 235 6.5.2 Fibrils . . . . . . . . . . . . .. . . . . . . . . .. . . . .. . . . . . . . . . . .. . . . . . . . . . . .. . . .. . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . .. 236 6.5.3 Inverse opal and opal structures . . . . . . . . . . . . . . . . . . . .. . . . . . . ..... . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . 238 6.5.3.1 Introduction . . . .. . . .... . . . .. . . . .. . . .. .. . . . . . . ... . . . . . .. . . .. . ... . . . . . . ... . . . . .... . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . .. 238 6.5.3.2 Platinum inverse opals (honeycomb structure) . . . . . . . . . . ... .. . . .. . . . .. . .. . . . . . ... . .. . .. . . .. . . . . 239 6.5.3.3 Platinum, gold and ITO opals (bead structure) . . . ... . . . . . .. . . ... . . . . . . . . . . . . . . . . . . . . . .. .... .. . . 239 6.5.3.4 Summary . . . . . . . . . . . . . . . . . . . . . . .. . .. .. .. . . . . . . . . . . . .. .. . . . ... . . . . . . . . . . . . . . . . . . . ... . .. . . . . . . . . . . . ... . . . . . . . . . . . . . . 244 6. 5. 4 Electrospinning . . . . . . . . .. . . . . . . . . .... . ... . . . .. . . . . . . . . . . . . . . ... .. . . ..... . ... . . .. ... . . . .. . . . . . . . . . . . . . ... . .. . 245 6. 5. 5 Conclusions .................................................................................................. 247 6.6 CONCLUSIONS . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 6.7 EXPERIMENTAL PROCEDURES . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 260 6. 7.1 PEC devices .................................................................................................. 260 6. 7.2 Actuator fabrication and experimental procedures . .. . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . 261 6. 7.3 Details of battery fabrication and testing procedures .................................... 263 6. 7.4 Fabrication and evaluation of nanostructures ............................................... 264 CHAPTER 7 CONCLUSIONS .................................................................................................. 259 BIDLIOGRAPHY ................................................................................................ 264 xi List of Figures Number Description Page 1.1 A selection of polymers capable of exhibiting conductivity. 2 1.2 Relative conductivities of polythiophene and trans-polyacetylene ID doped and undoped states. 3 1.3 Scheme showing the reversible p-doping (oxidation) and n-doping (reduction) redox processes of polythiophene, with the incorporation of required cation/anion species ex'" 1 A 1 to balance the charge of the polymer matrix. 3 1.4 (a) The semi-conducting, aromatic, neutral chain. (b) Formation of a polaron species. (c) The conducting, quinoidal, bipolaron species formed on further oxidation. X = NH or S, A - = anion (e.g. cr, CI04} 4 1.5 Energy level diagram showing the difference in band structure of a conducting polymer in the neutral state, and oxidised state with formation of polaron and bipolaron species. 5 1.6 (a) al13 positions on a thiophene ring, (b) a - a coupling, (c) a -13 coupling and (d) 13 -13 coupling. 7 1.7 Proposed mechanism for the polymerisation of thiophene. 8 1.8 Functionalisation of thiophene through the 13 positions. 8 1.9 Possible regioisomers formed by coupling of two, non-centrosymmetric monomer units: head-to-head (llli), head-to-tail (Hr) and tail-to-tail (TT). 10 1.10 Examples of (a) regioregular and (b) regiorandom functionalised polythiophenes. 10 1.11 Potentiodynarnic grmvth of terthiophene on a platinum disc electrode (SA = 1.8 mm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 mV S·I. 12 1.12 Synthetic polymerisation of functionalised thiophene using the grignard method. 12 1.13 Oxidative polymerisation of functionalised thiophene using iron(ill) chloride. 13 1.14 Examples of (a) alkyl and (b) alkoxy spacers used to reduce steric hindrance of functional groups between thiophene units. 15 1.15 3 -Aryl substituted polythiophene. 15 1.16 Styryl-substituted thiophene monomers investigated by Cutler. 16 1.17 Polyterthiophene functionalised through a vinyl linker. 17 1.18 Synthesis of vinyl-substituted terthiophene using Wittig reactions. R = Ph-R. X = Br, Cl or I. 17 xii 1.19 1.20 1.21 1.22 1.23 1.24 2.1 2.2 2.3 2.4 Examples of (a) crown ether and (b) polyether substituted styryl­ terthiophene derivatives investigated by Grant. (a) Terthiophene substituted with a tetraphenyl phorphyrin (TPP) via a vinyl linker and (b) nitro-substituted styrylterthiophene synthesised and investigated by Cutler. Supramolecular assembly of 3-hexylthiophene by interdigitation of alkyl chains. Alkoxy substitution in the (a) 3,3"-positions and (b) 4,4"-positions on terthiophene monomers. Thienylenevinylene derivatives. (a) poly(thienylenevinylene) and (b) poly( terthienylene-vinylene). Styryl-substituted terthiophene (m = 0) and terthienylenevinylene (m = 1) materials investigated in this study. Schematic of a three-electrode one compartment cell. Illustration showing the linear and radial diffusion fields produced at standard electrodes (SA: > 100 f..Ul12) and disc rnicroelectrodes. Schematic of the Ag/O. 0 1 M AgN 03 reference electrode and TBAP 1 AN salt bridge used in this study. Variation of potential with time for cyclic voltammetry. 2.5 A CV of ferrocene, demonstrating a single reversible electron transfer process 2.6 2.7 Potentiodynamic growth of terthiophene on a platinum disc electrode (SA = 1.8 mm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 mV S-I. Post-growth CV of OCIODASTT on a Pt mIcro electrode (SA: lOI..lm2). Electrolyte solution: 0.1 M TBAP 1:1 AN:DCM. Potential limits: -5001+800 m V. Scan rate: 100 m VS-I. 2.8 (a) A chronoamperogram with (b) the corresponding function of potential 2.9 2.10 2.11 2.12 2.13 versus time. Potentiostatic growth of polySTT on a Pt micro electrode (SA: 10 11m2). Solvent: 1:1 AN:DCM. Potential held at -500 mV for 1 s, then stepped to 900 mV for 19 seconds. Schematic of MALDI-TOF MS instrumentation. Schematic illustrating the difference between the (a) linear and (b) reflection modes used in MALDI-TOF MS. MALDI-TOF mass spectra of a poly(OCIODASTT) sample (monomer Mr = 662.48 g morl). (a) Measured using reflectron mode. (b) Measured using linear mode. (c) Enlargement of a peak generated by the linear mode. Dimerisation of terthiophene showing the loss in terminal protons. 17 18 19 20 21 22 24 25 26 28 28 29 31 32 32 34 35 36 40 xiii 2.14 2.15 2.16 2.17 2.18 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 Dimerisation of STT to form three different possible regioisomers: the HH isomer (generated by coupling through the 5 positions), HT isomer (coupling through the 5 and 5" positions) and TT isomer (coupling through the 5" positions). Spectroelectrochemistry of 4,4"-bis(decyloxy-3' -nitro-2,2':5' ,2")ter­ thiophene obtained by Gambhir et al.31 Reprinted by permission. A free­ carrier tail is displayed at wavelengths above 700 run. Experimental setup for electrospinning. Schematic diagram of a typical polymer photoelectrochemical cell structure consisting of an ITO/polymer/electrolytelPt sandwich structure. The interface between the polymer and liquid electrolyte are explicitly included to illustrate that the photovoltage is generated by the difference in standard electrode potential (work function) between the ITO-polymer interface and the polymer-electrolyte interface. Irradiation of the semiconducting polymer results in the formation of an exciton, which is separated at an interface. Reprinted with permission Typical I - V curve of a photovoltaic device.Ise is the short circuit current and Voc is the open circuit voltage.Ipp is the current at peak power and Vpp is the voltage at peak power. Substituents linked by (a) phenyl and (b) styryl moieties to polythiophene. Structures of the five styryl-substituted terthiophene monomer derivatives investigated in this study. Wittig reaction between a terthiophene aldehyde and phenyl substituted phosphonium salt for the synthesis of the substituted styrylterthiophene monomers used in this study. Reaction procedure employed for the chemical polymerisation of styryl­ substituted terthiophene derivatives. The reversible doping (by iron (Ill) chloride or copper perchlorate) and dedoping (by washing with water or hydrazine) process of the polymer. MALDI-TOF MS of crude, soluble fractions of chemically polymerised NMe2STT, OMeSTT, STT, CNSTT and N02STT. Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit: 320 Da. IH NMR spectra (aromatic region) of NMe2STT and the dimeric HH regioisomer produced by chemical oxidation. IH NMR spectra (aromatic region) of styryl-15-crown-5 terthiophene, and the dimeric HH regioisomer that is produced by oxidation. 'Charge-trapping' over the two central thiophene nngs and styryl substituents as exhibited by theoretical calculations. CV of terthiophene monomer on a platinum disc electrode (SA = 1.8 mm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: - 10001+1500 mV. Scan rate: 100 mV S-I. A possible mechanism to explain redox peaks (using thiophene for clarity) observed in the first scan of terthiophene derivatives. 41 43 47 48 49 50 51 52 53 54 57 59 60 61 64 65 xiv 3.12 3.13 3.14 3.15 3.16 3.17 3.18 3.19 3.20 3.21 3.22 3.23 3.24 3.25 CV scans of CNSTI, STT, OMeSTT and NMe2STI on platinum disc electrodes (SA = l.8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: -1000/+1500 mY. Scan rate: 100 mV S·I. CV of N02STI monomer on a platinum disc electrode (SA = 1.8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: -10001+1500 mY. Scan rate: 100 mV S·I. Potentiodynamic growth of NMe2STI on a platinum disc electrode (SA = l .8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 m V S·I. Potentiodynamic growth of OMeSTI on a platinum disc electrode (SA = l.8 rnm2). Supporting electrolyte: 0.1 M TBAP IAN. Potential limits: +3001+750 mY. 9 cycles. Scan rate: 100 mV S·I. Potentiodynamic growth of STT on a platinum disc electrode (SA = 1.8 rnm2). Supporting electrolyte: 0.1 M TBAP IAN. Potential limits: + 375/+900 mY. 9 cycles. Scan rate: 100 m V S·I. Potentiodynamic growth of CNSTI on a platinum disc electrode (SA = l.8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 m V S·I . Potentiodynamic growth of N02STI on a platinum disc electrode (SA = l.8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 mV S·I. Potentiostatic growth of STT on a Pt disc electrode (SA = l.8 rnm2). Supporting electrolyte: 0.1 M TBAP/AN. Potential held at 0 mY for 10 s, then stepped to 900 mV for 290 s. Post-polymerisation CV of NMe2STT films polymerised (a) potentiodynamically and (b) potentiostatically. Electrode: platinum disc (l. 8 rnm2). Electrolyte solution: 0.1 M TBAP 1 AN. Potential limits: 01+900 mY. 15 cycles. Scan rate: 100 mV S·I. Post-polymerisation CYs of (a) potentiodynamically and (b) potentiostatically deposited oligoOMeSTI films. Electrode: platinum disc (l.8 rnm\ Electrolyte solution: 0.1 M TBAP/AN. Scan rate: 100 mV S·I. Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoSTI films. Electrode: platinum disc ( l .8 rnm2). Electrolyte solution: 0.1 M TBAPI AN. Potential limits: +375/+1200 mY. 5 cycles. Scan rate: 100 mY S·I. Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoCNSTI films. Electrode: platinum disc ( l .8 rnm2). Electrolyte solution: 0.1 M TBAP/AN. 5 cycles. Scan rate: 100 mY S·I. Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited 0ligoN02STT films. Electrode: platinum disc (l.8 rnm2). Electrolyte solution: 0.1 M TBAP/AN. 5 cycles. Scan rate: 100 mV S·I. Chromophores produced by styryl-substituted terthiophene derivatives. 66 66 68 69 69 70 70 71 73 75 76 77 78 81 xv 3.26 3.27 3.28 3.29 3.30 3.31 3.32 3.33 3.34 3.35 3.36 3.37 3.38 3.39 4.1 4.2 4.3 UV-VIS-NIR spectra of a film of electrodeposited 0ligoN02STT in the oxidised (solid line) and neutral (dashed line) states. UV-VIS-NIR spectra of a film of electrodeposited oligoCNSTT in the oxidised (solid line) and neutral (dashed line) states. UV-VIS-NIR spectra of a film of electrodeposited oligoSTT in the oxidised (solid line) and neutral (dashed line) states. UV -VIS-NIR spectra of oxidised films (solid lines) and neutral films (dashed lines) of OMeSTT deposited usmg potentiodynarnic and potentiostatic methods. UV-VIS-NIR spectra of films that were previously oxidised (solid lines) and neutral films (dashed lines) of NMe2STT deposited usmg potentiodynarnic and potentiostatic methods. Spectroelectrochemistry of an e1ectrochemically grown film of oligoSTT on ITO-coated glass. Potentials between 0.4 V and 1.1 V were applied in steps of 0.1 V. SEM images of neutral and oxidised films of oligoNMe2STT deposited onto ITO-coated glass using potentiodynarnic methods. Magnification: xIOOOO. SEM image of an oxidised film of oligoOMeSTT deposited using potentiodynarnic methods. Magnification: xIOOOO. Potentiostatically deposited films of STT oligomers. (a) Neutral film: x1400, inset: x70. (b) Oxidised film: x1400. SEM images of oligoSTT films deposited using cyclic voltarnrnetry. (a) Neutral film: x l 000, inset: x5000. (b) Oxidised state: x2000 magnification. SEM images of oligoCNSTT films deposited using a constant potential. (a) oxidised state, x l 0000 magnification, (b) neutral state, x5 000 magnification. SEM images of oligoCNSTT films deposited using cyclic voltarnrnetry. (a) oxidised state, x l 0000 magnification, (b) neutral state, x l 0000 magnification. SEM image of a neutral 0ligoN02STT film deposited using a constant potential, x 10000 magnification. SEM unages of oxidised 0ligoN02STT films deposited by (a) potentiostatic and (b) potentiodynamic methods, x l 0000 magnification. A polythienylenevinylene derivative Terthienylenevinylene monomer derivatives. Potentiodynarnic growth of N02STV on a platinum micro electrode (SA = 10 11m2). Monomer concentration: 5 mM. Supporting electrolyte: 0.1 TBAP/1: l AN:DCM. Potential limits: -500/+800 mY. 15 cycles. Scan rate: 100 m V S·I. Inset: first three cycles. 82 82 83 83 84 84 86 86 87 87 88 88 89 89 10 0 101 103 xvi 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4. l 4 4.15 4.16 4.17 Potentiodynamic growth of CNSTV on a platinum micro electrode (SA = lO,..lm2). Monomer concentration: 5 mM. Supporting electrolyte: 0.1 TBAPI1:1 AN:DCM. Potential limits: -500/+800 mY. 15 cycles. Scan rate: 100 mV S·l. Inset: first three cycles. Potentiodynamic growth of STV on a platinum micro electrode (SA = 10 ,....m2). Monomer concentration: 5 mM. Supporting electrolyte: 0.1 TBAPI1:1 AN:DCM. Potential limits: -500/+800 mY. 15 cycles. Scan rate: 100 m V S·l. Inset: first six cycles. Potentiodynamic growth of OMeSTV on a platinum micro electrode (SA = 10 ,....m2). Monomer concentration: 5 mM. Supporting electrolyte: 0.1 TBAP/1:1 AN:DCM. Potential limits: -500/+800 mY. 15 cycles. Scan rate: 100 mV S·l. Inset: first three cycles. Potentiodynamic growth of NMe2STV on a platinum disc electrode (SA = 1.8 mrn2). Monomer concentration: 5 mM. Supporting electrolyte: 0.1 TBAP /U AN:DCM. Potential limits: 0/+800 mY. 15 cycles. Scan rate: 100 mV S·l. Inset: first three cycles. Post growth cycling of oligoN02STV deposited on a platinum micro electrode (SA = 10 ,....m2). Supporting electrolyte: 0.1 TBAP/AN. Potential limits: -500/+800 m V. 10 cycles. Scan rate: 100 m V S·l. Post growth cycling of oligoCNSTV deposited on a platinum micro electrode (SA = 10,....m2). Supporting electrolyte: 0.1 TBAP/AN. Potential limits: -500/+800 mY. 10 cycles. Scan rate: 100 mV S·l. Post growth cycling of oligoSTV deposited on a platinum micro electrode (SA = 10 ,....m2). Supporting electrolyte: 0.1 TBAP/AN. Potential limits: - 500/+800 mY. 10 cycles. Scan rate: 100 mV S·l. Post growth cycling of oligoOMeSTV deposited on a platinum micro electrode (SA = 10,....m2). Supporting electrolyte: 0.1 TBAP/AN. Potential limits: -500/+800 mY. 10 cycles. Scan rate: 100 mV S·l. Postgrowth cycling of oligoNMe2STV deposited on a platinum disc electrode (SA = 1.8 mrn2). Supporting electrolyte: 0.1 TBAP/AN. Potential limits: -5 00/+800 mY. 10 cycles. Scan rate: 100 mV S·l. UV-VlS-NIR spectrum of oligoN02STV electrodeposited onto ITO­ coated glass and electrochemically oxidised (solid line) and reduced (dashed line). UV-VlS-NIR spectrum of oligoCNSTV electrodeposited onto ITO­ coated glass and electrochemically oxidised (solid line) and reduced (dashed line). UV-VlS-NIR spectrum of oligoSTV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). UV-VlS-NIR spectrum of oligoOMeSTV electrodeposited onto ITO­ coated glass and electrochemically oxidised (solid line) and reduced (dashed line). UV-VlS-NIR spectrum of NMe2STV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). 104 104 105 105 113 108 108 109 110 112 113 113 114 114 xvii 4. 1 8 5 . 1 5 .2 5 .3 5 .4 5 .5 5 .6 5 . 7 5 . 8 5 .9 5 . 1 0 5 . 1 1 5 . 1 2 5 . 1 3 Relationship between electron withdrawing/donating effect of the substituent and the wavelength maxima due to the 1t -+ 1t* transition of the oligomer-chain chromophore for both terthiophene and thienylenevinylene derivatives. Alkyl- and alkoxy-substituted styrylterthiophene monomers that were initially polymerised and investigated in this study. Polymerisation of OC6DASTT. Possible structure created by the mesomeric effect of the oxygen on alkoxy substituted thiophene polymers . MALDI-TOF mass spectrum of polyOC�ASTT. Signals are labelled with the assigned oligomer length in terms of mono mer units (n). UV-VIS-NIR spectrum of the soluble fraction of polyOC�ASTT in solution in the oxidised state (excess Cu(Cl04)2, solid line) and neutral state (dashed line). The 1t-1t* band in the neutral state is labelled with its wavelength maximum. CV of chemically polymerised OC6DASTT, which has been cast as a film onto an ITO-coated glass electrode (- 1 cm\ Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits : -500/+ 1200 mY. 16th to 20th cycles . Scan rate: 1 00 mV S·I . • Photograph of the monomer and solvent fractions as solutions ID chloroform. (a) Monomer, (b) hexane, (c) acetone, (d) dichloromethane and (e) chloroform. MALDI-TOF mass spectra of OC6DASTT oligomer fractions extracted using hexane followed by acetone, dichloromethane and chloroform. Signals are labeled with the assigned oligomer length in terms of monomer units (n). Mr of mono mer: 550.5 g mOrl . UV -VIS-NIR spectra of OC6DASTT oligomer fractions separated by (a) hexane, (b) acetone, (c) DCM and (d) chloroform. Samples were measured in the oxidised state (excess Cu(Cl04h solid line) and neutral state (dashed line) from solutions in chloroform. UV-VIS-NIR spectra of neutral samples of OC6DASTT oligomer fractions which have been separated using hexane, acetone, DCM and chloroform. The A-max of the major 1t-+1t* transition for each fraction is shown. MALDI-TOF mass spectrum of poly(C7DASTT). Signals are labelled with the assigned oligomer length in terms of mono mer units (n). UV-VIS-NIR spectra of polyC7DASTT in the oxidised state (excess Cu(CI04)2, solid line) and neutral state (dashed line). The 1t-1t * bands in the neutral state are labelled with their wavelength maxima. CV of a chemically polymerised C7DASTT film, which has been cast onto an ITO-coated glass electrode (SA: -1 cm2) . Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -500/+1 200 mY. 5 cycles . Scan rate: 1 00 mV s·l . 1 1 5 1 2 1 1 22 1 24 1 25 1 26 127 1 3 0 1 32 1 34 1 35 1 3 8 1 39 1 39 xviii 5 . 1 4 5 . 1 5 5 . 1 6 5 . 1 7 5 . 1 8 5 . 1 9 5 .20 5 .2 1 5 .22 5 .23 5 .24 5 .25 5 .26 5 .27 MALDI-TOF MS of C7DASTT oligomer fractions extracted using methanol followed by hexane, acetone and dichloromethane. Signals are labelled with the assigned oligomer length in terms of mono mer units (n). Mr of mono mer: 546.5 g mOrl . UV-VIS-NIR spectra of C7DASTT oligomer fractions separated by (a) methanol, (b) hexane, (c) acetone and (d) DCM. Samples were measured in the oxidised state (excess Cu(CI04)2, solid line) and neutral state (dashed line) from solutions in chloroform. UV -VIS-NIR spectra of neutral samples of C7DASTT oligomer fractions which have been separated using hexane, acetone and DCM. The Amax of the 1t--+1t* transition for each fraction is shown. C IO-dialkoxy styrylterthiophene (OCIODASTT) Reaction procedure employed for the polymerisation of OCIODASTT MALDI-TOF MS of POlyOCIODASTT. Signals are labelled with the assigned oligomer length in terms of monomer units (n). UV-VIS-NIR spectra of polyOCIODASTT in chloroform in the oxidised state (excess Cu(CI04h, solid line) and neutral state (dashed line). The major "-max for the 1t--+1t* absorbance is labelled. CV of chemically polymerised OCIODASTT, which has been cast as a film onto an ITO-coated glass electrode (-1 cm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -5001+ 1200 mY. 1 0 cycles. Scan rate: 100 mV S-I. MALDI-TOF mass spectra of OCIODASTT polymer fractions extracted using hexane followed by acetone, dichloromethane and chloroform. Signals are labeled with the assigned oligomer length in terms of monomer units (n). Mr of monomer: 662.4 g mOrl . UV-VIS-NIR spectra of OCIODASTT oligomer fractions separated by (a) hexane, (b) acetone, (c) DCM and (d) chloroform. Samples were measured ill the oxidised state (solid line, oxidised using excess Cu(CI04)2,) and neutral state (dashed line) from solutions in chloroform. UV -VIS-NIR spectra of OCIODASTT oligomer fractions which have been separated using hexane, acetone, DCM and chloroform. The "-max of the 1t--+1t* transition for each fraction is listed. Cyclic voltammetry of OCIODASTT oligomer fractions that have been cast onto a glassy carbon electrode (SA: 7 mm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: - 1000/+800 mY. 1 0 cycles. Scan rate: 1 00 mV S-I . The average oligomer length in terms of mono mer units (nav) is displayed. Growth CV of oligoSTT (5 mM) on a platinum microe1ectrode (SA: 1 0 �m2). Electrolyte solution: 0 . 1 M TBAP/ 1 : 1 AN:DCM. Potential limits: - 5001+900 mY. 1 5 cycles . Scan rate: 1 00 mV S-I. Growth CV of C7DASTT (5 mM) on a platinum microelectrode ( 1 0 �m\ Electrolyte solution: 0 . 1 M TBAPI l : 1 AN:DCM. Potential limits: 01+800 mY. 15 cycles. Scan rate: 1 00 mV S·I. 14 1 143 144 146 147 147 148 149 1 52 1 54 1 55 1 56 1 59 1 60 xix 5 .28 5 .29 5 .30 5 .3 1 Relationship between the current produced during the growth of C7DASTI (measured at 0 .8 V) and the cycle number. Post gro'wth cycling of polyC7DASTI which has been deposited using cyclic voltammetry on a platinum micro electrode (SA: 1 0 �m2) . Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits : -5001+800 mY. 1 0 cycles. Scan rate: 1 00 mV S-I . Potentiostatic growth of C7DASTI (5 mM) on a platinum micro electrode (SA: 1 0 �2). Solvent: 1 : 1 AN:DCM. Potential held at 0 mV for 1 s, then stepped to 700 mV for 29 seconds . Post growth cycling of polyC7DASTI, which has been deposited potentiostatically on a platinum microelectrode (SA: 1 0 �m2) . Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: 01+800 mY. 1 0 cycles . Scan rate: 100 m V S-I . 5 .32 Growth CV of OCJ)ASTT (5 mM) of a platinum micro electrode (SA: 10 �m2) . Electrolyte solution: 0 . 1 M TBAP/1 : 1 AN:DCM. Potential limits: -5001+800 mY. Scan rate: 1 00 mV S-I . The first scan is shown as 5 . 33 5 .34 5 .35 5 .36 5 .37 5 .38 5 .39 5.40 an inset. Post growth cycling of polyOCJ)ASTT which has been deposited using cyclic voltammetry on a platinum microelectrode (SA: 1 0 �m2) . Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits : -5001+800 mY. 1 0 cycles . Scan rate: 1 00 mV S-I . Potentiostatic growth of OC6DASTI (5 mM) on a platinum microelectrode (SA: 1 0 �m2) . Electrolyte solution: 0. 1 M TBAPI1 : 1 AN:DCM. Potential held at -500 mV for 1 s, then stepped to 800 mV for 1 9 s . Post growth cycling of polyOC6DASTI which has been deposited potentiostatically on a platinum microelectrode (SA: 1 0 �m2) . Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -5001+800 m V. 1 0 cycles . Scan rate: 1 00 m V S-I . Growth CV of OCJ)ASTT (5 mM) on an ITO-coated glass electrode (SA: -1 cm2) . Electrol}te solution: 0 . 1 M TBAP/1 : 1 AN:DCM. Potential limits: -500/+800 mY. 5 cycles . Scan rate: 1 00 mVs-l . Post growth cycling of polyOCJ)ASTI that has been deposited potentiodynamically on an ITO-coated glass electrode (SA: - 1 cm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -5001+800 m V. 1 0 cycles. Scan rate: 1 00 m V S-I . Potentiodynamically deposited films of OC6DASTI oligomers on ITO­ coated glass. (a) Neutral film: x 1400, (b) Oxidised film: x 1400 inset: x350. MALDI-TOF MS of polyOCJ)ASTI, which was polymerised by potentiodynamic deposition onto an ITO-coated glass electrode. Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit: 1000 Da. Growth CV of OCIODASTI (5 mM) on a platinum microelectrode (SA: 1 0 �m2). Electrolyte solution: 0. 1 M TBAP 1 : 1 AN:DCM. Potential limits: -5001+800 mY. Scan rate: 1 00 mV S-I . 1 6 1 1 62 1 63 164 1 65 1 66 1 67 1 67 1 68 1 69 1 70 1 70 1 7 1 xx 5 .4 1 5 .42 5 .43 5.44 5 .45 5 .46 5.47 5 .48 5 .49 5 .50 5.5 1 5 .52 Post growth cycling of polyOCIODASTT which has been deposited potentiodynamicaUy on a platinum microelectrode (SA: 1 0 Ilm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -5001+800 mY. 10 cycles . Scan rate: 1 00 mV S·I . Potentiostatic growth of polyOCIODASTT (5 mM) on a platinum microelectrode (SA: 1 0 1-I.ffi2) . Electrolyte solution: 0. 1 M TBAP/l : 1 AN:DCM. Potential held at -500 mV for 1 s, then stepped to 800 mV for 1 9 s . Post gwwth cycling of polyOCIODASTT which has been deposited potentiostaticaUy on a platinum microelectrode (SA: 1 0 Ilm2) . Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -5001+800 mY. 1 0 cycles . Scan rate: 1 00 mV S·I . Spectroelectrochemistry of a potentiostatically deposited poly(OCIODASTT ) film on ITO-coated glass. The film was initiaUy in a reduced state and was oxidised in steps of 0. 1 V. Supporting electrolyte: 0 . 1 M TBAP/AN. Potentials are reported vs Ag/Ag+. Growth CV of OCIODASTT (5 mM) on an ITO-coated glass electrode (SA: - 1 cm2) . Electrolyte solution: 0 . 1 M TBAPIl : 1 AN:DCM. Potential limits: -5001+800 m V. 5 cycles. Scan rate: 1 00 m Vs·l . Post growth cycling of pOly(OCIODASTT) that has been deposited potentiodynamically on an ITO-coated glass electrode (SA: - 1 cm2) . Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -5001+800 mY. 1 0 cycles . Scan rate: 1 00 mV S·I . Potentiostatic gro\\th of polyOCIODASTT (5 mM) on an ITO-coated glass electrode (SA: - 1 cm2). Electrolyte solution: 0. 1 M TBAP/l : 1 AN:DCM. Potential held at -500 mV for 1 s, then stepped to 800 mV for 1 9 s . Post grm\1h cycling of polyOCIODASTT that has been deposited potentiostatically on an ITO-coated glass electrode (SA: - 1 cm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits : -5001+800 mY. 10 cycles. Scan rate: 1 00 mV S·I . SEM images of potentiodynamicaUy deposited films of OCIODASTT oligomers. (a) Neutral film: x1400, (b) Oxidised film: x 1400. SEM unages of potentiostatically deposited films of OCIODASTT oligomers. (a) Neutral film: x 1400, (b) Oxidised film: x 1400. SEM images of (a/b) potentiostatically and (cId) galvanostatically deposited polyOCIODASTT films on ITO-coated glass. Images (b) and (d) are of the film edges formed at the solution/air interface. All images are displayed at x7000 magnification. MALDI-TOF MS of polyOCIODASTT, which was polymerised by potentiodynamic deposition onto an ITO-coated glass electrode. Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit: 1 000 Da. 1 72 1 73 1 73 1 74 1 75 1 76 1 76 1 77 1 78 1 78 1 79 1 80 xxi 5.53 5.54 6. 1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6. 10 6. 1 1 6. 12 6. 13 MALDI-TOF MS of polyOCIODASTI, which was polymerised by potentiostatic deposition onto an ITO-coated glass electrode . Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit 1000 Da. MALDI-TOF MS of polyOCIODASTI, which was polymerised by galvanostatic deposition onto an ITO-coated glass electrode. Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit: 1000 Da. Construction of a PEC cell. I - V curve of a cast film of chemically dimerised NM�STI in the dark (thin line) and under illumination (thick line). I-V curves of electrochemically grown films of styryl thienylenevinylene derivates. (a) NMe2STV, (b) OMeSTV, (c) CNSTV and (d) N02STY. The relationship between the photovoltaic activity at lse (Msc, where Msc == hgJ1t.,sc - Idark.sc) and the electron-withdrawing capability of substituents on films of thienylenevinylene oligomers. I - V curve of a PEC device comprising a cast film of STY dimer. Experimental setup using a Dual Mode Lever System for measuring electromechanical properties of a sample. Schematic of a bender made by casting polymer as a solution in chloroform on a PVDF memebrane strip. CV of a cast polyOCIODASTI film on a PVDF membrane substrate. Surface area of substrate: 7 mm2. Amount of polymer cast 15 Ilg. Supporting electrolyte: 0. 1 M TPAP/AN. Ten cycles. Scan rate: 100 mV S·I. Absorbance of a polyOCIODASTI film on PVDF (a) immediately after being cast from solution, (b) after electrochemical oxidation (+ 1 V vs Ag/ Ag +) and (c) after oxidation with iodine. UV -VlS-NlR spectra of a cast polymer/PVDF strip which has been was reduced and oxidised potentiostatically at -0.6V and + 1.0 V. Supporting electrolyte: 0.25 M TBAPFJPC. A piece of platinum coil was used to wrap around the polymer/PVDF film to increase charge accessability, and the potential was applied for 30 minutes. Experimental setup for testing a polymerlPVDF strip. Electrolyte: 0.25 M TBAPFJPC electrolyte. Counter electrode: stainless steel mesh. Reference electrode: Ag/Ag+. The strip is shown in both an (a) reduced and (b) oxidised state. Schematic showing the expansion of the polymer film on oxidation by incorporation of the anion, and contraction on reduction due to expulsion of the anion. Chronoamperometry of polyOCIODASTTIPVDF strip ID 0.25 M TBAPFJPC. Oxidation potentials : switching from -0.6 V to 0.8 V. Reduction potentials: switching from 0.8 V to -0.6 V. 180 180 196 197 200 20 1 202 206 207 209 209 2 1 1 2 12 2 12 2 13 xxii 6. 14 6. 1 5 6. 1 6 Correlation between the injected charge and the displacement of the free end of the film. The charge due to non-Faradaic process can be determined by extrapolation of the linear trend. Schematic of fibres on which polyOCIODASTI was deposited. (a) Hollow PVDF membrane fibre wrapped with a 50 J..Ull platinum wire. (b) Hollow PVDF membrane fibre which has been sputter-coated with platinum and wrapped with a 50 J..Ull platinum wire. (c) 250 J..I.m wire wrapped in a 50 J..Ull wire. Actuation of a hollow PVDF membrane fibre (49 mm long) wrapped in a 50 J..Ull wire and coated with polymer (39 J..I.g mm-I) . The force was held constant at 6 mN while the potential was alternated between -0.6 V and +0.8 V. The distance the fibre stretches and the current produced were measured. 6. 1 7 Relationship between displacement and charge injected into a PVDF membrane fibre coated with polyOCIODASTI during (a) reduction and (b) oxidation of one typical pulse. A negative charge is used to indicate reduction, a negative displacement to indicate contraction, a positive charge to indicate oxidation and positive displacement to indicate expanSIOn. 6. 1 8 6. 1 9 6.20 6.2 1 6.22 6.23 6.24 6.25 6.26 Comparison of the rate of contraction/expansion during oxidation and reduction of one typical pulse. The slope represents the rate at which the fibre is expanding or contracting. Free-standing film of polyOCIODASTI incorporating a zigzagged (50 J..I.m diameter) wire. Dimensions of film: 6 mm x 21 mm with a thickness of 8 1 J..Ull. Mass = 16 . 1 mg. Strain created by the polyOCIODASTI film over time as it is doped (oxidised) and then dedoped (reduced at -0.6 V) to and from various oxidation potentials. The effect of the oxidation potential on the strain measured after 5000 seconds of oxidation. Relationship between the strain generated at different oxidation potentials and the electrochemical efficiency (EE). Relationship between the isometric stress generated on the film and the charge density passed as the film is reduced. The film was pre­ conditioned at + 1 V to obtain the fully expanded state and then isometric measurements were performed as the potential was switched to -0.6 V. Schematic of the test cell used in this study. SEM unages of the electrodes . Anodes prepared by casting poly(OCIODASTI) onto (a) Ni/Cu substrate (b) carbon fibre substrate. The blank substrates before polymer deposition are given for comparison as insets in the top right hand corner. (c) Cathode prepared by electrodeposition of polypyrrole on stainless steel mesh. Images are displayed at x400 magnification with insets at x l 00 magnification. Cyclic voltammograms of polyOCIODASTI cast on carbon fibre substrate and Ni/Cu substrate. Electrolyte solution: 60 mM TBAPFJPC. Scan rate: 1 0 mV S-I . 2 14 2 16 2 1 7 2 1 8 2 1 9 220 221 223 223 224 228 230 23 1 xxiii 6.27 6.28 6.29 6 .30 6.3 1 6 .32 6.33 6.34 6.3 5 6 .36 6 .37 6 .3 8 Charge (A,C) and discharge (B,D) curves of cells with anodes comprising Ni/Cu substrate (A,B) and carbon fibre (C,D). The charge/discharge current density is 0.02 mA cm-2. Discharge capacities at different discharge current densities obtained for the Teflon cell using Ni/Cu and carbon fibre anode substrates. Variation of discharge capacity with cycle number for cells with an anode comprising polyOCIODASTT on Ni/Cu substrate and carbon fibre substrate. Current density: 0 .02 mA cm-2 . SEM images showing polyOCIODASTT fibrils with a platinum film as a support. (a) Partially dissolved template and viewed from platinum coated side, and (b) cross-section of the fibrils with a polymer film on left and platinum film on the right. CV of fibrils in a partially dissolved template with a platinum backing stuck on ITO coated glass. Supporting electrolyte: 0 . 1 M TBAP/AN. Scan rate: 100 mV S-I . Platinum inverse opals showing the honey comb structure. (a) x7500 and (b) x3000 magnification. Platinum opal structures produced by sputter-coating the polystyrene opal with platinum. (b) opal showing predominantly body-centred cubic square-packing. (c) showing predominantly hexagonal close-packing with polymer coating. AFM images of the underside of a platinum opal structure. (a) 3 - dimensional image, (b) Height data and (c) deflection data. Scan size: 1 0.00 I--lffi. (a) Gold opal surface on ITO-coated glass and (b) polyOC,oDASTT cast on a gold opal surface. Edge of ITO opal structure on ITO-coated glass. SEM images electrospun polyOCIODASTT fibres. Growth of platinum on opal coated ITO coated glass . 1 0 cycles. Scan rate: 1 00 mV S-I . 232 233 234 237 238 239 241 242 243 244 246 257 xxiv List of Tables Number Description Page 3 . 1 Soluble fractions of styryl-substituted terthiophene polymer derivatives. 59 3 .2 R£ values of monomers and dimers in 20% ethylacetate!hexane on silica TLC plates. 60 3 . 3 Anodic peaks shown by stryl-substituted terthiophene derivatives. 69 3 .4 Oxidation onset potentials of styryl terthiophene derivatives . 7 1 3 .5 Conditions of electrochemical deposition for films prepared on ITO- coated glass for characterisation by UV-VIS-NIR and microscopy. 1 02 4. 1 Polymer oxidation and reduction potentia1s on the second cycle. 1 12 4.2 Wavelength maxima shown by N02STV, CNSTV, STY, OMeSTV and NMe2STV. 1 16 5 . 1 Mass percentages of OC6DASTT oligomers separated by different solvents. 1 36 5 . 2 Mass percentages ofpolyC7DASTT separated by different solvents . 145 5 .3 Mass percentages of OCIODASTT oligomers/polymers separated by different solvents . 1 56 6. 1 Photovoltaic characteristics of a cast film ofNMe2STT dimer. 203 6 .2 Photovoltaic characteristics of PEC devices constituting nitro-styryl- substitutedd thiophene and thienylenevinylene oligomer films. 206 6.3 Photovoltaic characteristics of a cast film of chemically synthesised STY dimer. 207 6.4 Maximum strain rate on oxidation and reduction of films which are oxidised at various potentials . The reduction potential is kept constant at -0.6 V. 227 xxv A A A b oh Use E E> e EE Eoxidation onset Ered F Fc HH HT Idark,sc hght,sc Ipp Isc Amax m m M List of Sym bols Ampere Ammeter Anion Variable number Photovoltaic activity Photovoltaic activity under short circuit Potential Standard electrode potential Electron Electrochemical efficiency Electrochemical efficiency on oxidation Electrochemical efficiency on reduction Kinetic energy Peak oxidation potential Oxidation onset potential Peak reduction potential Faraday's constant Ferrocene Proton at head position (S-position) Proton at tail position (5" -position) Short circuit current, dark conditions Short circuit current produced under illumination Current at peak power Short circuit current (Current at zero potential) Wavelength at maximum absorbance in spectroscopy Variable number Mass Molecular weight Number average molecular weight xxvi n n N R Rf t v z Molar mass Weight average molecular weight Oligomer length in terms of monomer units Variable number Number of electrons per monomer unit Number of molecules Average oligomer length in terms of mono mer units Variable functional group Retention factor Time Velocity Volt Voltammeter Open circuit voltage (Potential at zero current) Voltage at peak power. Variable cation Charge xxvii ACTH AFM AN Anhyd. CE CP CV DB TT DBU DCM dppp ECE EDG EDOT EE El EQCM Equiv. EWG FAB Fc FF GPC GPES GRIM HR HOMO Hrs HT List of Abbreviations Adrenocorticotropic hormone Atomic force microscopy Acetonitrile Anhydrous Counter electrode Conducting polymer Cyclic voltammetry 3' ,4' -dibutyl-2,2' :5'2" -terthiophene 1,8-diazabicyc1o[5.4.0]undec-7-en Dichloromethane 1,3 -diphenylphosphinopropane Energy conversion efficiency Electron-donating group 3,4-ethylenedioxythiophene Electrochemical efficiency Electrospray ionisation Electrochemical quartz crystal microbalance Equivalents Electron-withdrawing group Fast atom bombardment Ferrocene Fill factor Gel permeation chromatography General purpose electrochemical system Grignard method Head-to-head Highest occupied molecular orbital Hours Head-to-tail xxviii ICP IPRI ITO LUMO MALDI-TOF MS Me MCP N/A NMR OFET OLED PCBM PD PEC PEDOT PEDOT-PSS PEO Ph PMMA PPV PV PVDF RE RT SA SCE SEM SHE TBAP TBAPF6 TLC Inherently conducting polymer Intelligent Polymer Research Institute Indium tin oxide Lowest unoccupied molecular orbital Matrix assisted laser desorptionlionisation time-of-flight mass spectrometry Methyl Micro-channel plate Not applicable Nuclear magnetic resonance Organic field effect transistor Organic light emitting diode 3 ' -phenyl-3 'H -cyclopropa[ 1 ,9] [5,6 ]fullerene-C60-h-3' - butanoic acid methyl ester Polydispersity Photoelectrochemical cell Poly(3, 4-ethylenedioxythiophene) Poly(3, 4-ethylenedioxythiophene )poly( styrenesulfonate) Polyethylene oxide Phenyl Polymethylmethacrylate Polyphenylenevinylene Photovoltaic Polyvinylidene fluoride Reference electrode Room temperature Surface area Standard calomel electrode Scanning electron microscopy Standard hydrogen electrode Tetrabutylammonium perchlorate Tetrabutylammonium hexafluorophosphate Thin layer chromatography xxix THF TOF TPP TT UV-VIS-NIR WE w.r.t . Tetrahydrofuran Time-of-flight Tetraphenyl porphyrin Tail-to-tail Ultraviolet-visible-near infrared Working electrode With respect to xxx Monomer Abbreviations eN Terthiophene CNSTT OMe STT OMeSTT xxxi OCJ)ASTT STY eN CNSTV OClODASTT OMe OMeSTV STT: Styryl terthiophene DASTT: Dialkoxy styryl terthiophene or Dialkyl styryl terthiophene STV: Styryl terthienylenevinylene xxxii Chapter 1 Introduction 1 . 1 I ntroduction Organic polymers have properties such as flexibility, light weight, strength and resistance to corrosion. In addition, they can be made from readily available and inexpensive starting materials. 1 These properties are valuable for components of clothing, kitchenware, vehicles, and appliances.2 Polymers are also renowned and utilised for their high resistance to electricity. However, in 1977, Alan McDiarmid, Alan Heeger and Hideki Shirakawa reported conductivity in polyacetylene (CH)n. Since then, interest in polymers that intrinsically conduct electricity has been escalating.3,4 There were more than 2900 papers published in 2004 relating to conducting polymers. The discovery of conducting polymers merited McDiarmid, Heeger and Shirakawa a joint Nobel prize in 2000. Conducting polymers combine the many benefits of plastics with the ability to conduct electricity at a level comparable to metals. In addition, conducting polymers possess the ability to act as a switch by reversibly alternating between the conducting state, and a semi-conducting or non-conducting state. This conversion is usually associated with a change in colour. These properties of conducting polymers provide great potential for their use in a variety of applications. Organic batteries 5,6 , electrooptical display devices (electrochrornic devices, smart windows and organic light emitting diodes (OLEDs)),7-9 sensors, I O, 1 1 molecular wires, 1 2, 1 3 modified electrodes, 14, 1 5 diodes, 1 6 capacitors,1 7- 19 organic field-effect transistors (OFETs),4,2o actuators2 1 and solar ceUs22 are examples of applications employing conducting polymers that are under active research. The broad range of functions demonstrates the versatility of these materials. 1 .2 Conducting polymers There is now a wide variety of conducting polymers and their derivatives. Examples of some that are commonly investigated are given in Fig. 1 . 1 . The fundamental structural feature of all conducting polymers is their extended conjugated 1t system (alternating single and double bonds), which is delocalised over a number o f recurring mono mer units along the polymer backbone. 23,24 lrans-Polyacetylene Polythiophene Polypyrrole Polyfuran n n Pol yparaphen ylene Polyheptadiyne Polyparaphenylene vinylene Polyaniline Fig. 1. 1. A selection of polymers capable of exhibiting conductivity. In their neutral state, conjugated polymers are either insulators or semiconductors with conductivity typically ranging from 1 0-10 to 1 0-5 S cm-I , as shown in Fig. 1 . 2 . They are unique, however, in that they can be redox-doped to possess metal-like electronic, magnetic, optical and electrical properties, producing conductivities in the range of 1 - 1 05 S cm-I . 4 2 Ag, Cu, Fe METALS Mg In, Sn Ge SEMICONDUCTORS Si AgBr Glass Diamond INSULATORS Nylon Quartz 1 06 S cm-1 104 S cm-1 Doped trans-po1yacety1ene 1 02 S cm-1 I OU S cm- Doped polythiophene 10-2 S cm-1 10-4 S cm-1 10-6 S cm-1 Trans-polyacetylene 1 0-8 S cm-1 1 0-10 S cm-1 1 0-12 S cm-1 Polythiophene 1 0-14 S cm- 1 O-1() S cm-1 Fig. 1. 2. Relative conductivities of po1ythiophene and trans-polyacetylene in doped and undoped states 4,25 Redox-doping involves the reduction (n-doping) or oxidation (p-doping) of the 1t­ electronic system,26 and can be achieved by chemicaf7 or electrochemical methods. 2 8 Associated with the redox processes is the incorporation or removal of an ionic species (X: / A), to balance the charge of the polymer matrix. 29 The n-doping and p-doping of polythiophene is shown in Fig. l . 3 as an example. Undoping Oxidation/p-doping ( n \ b- bY! -\'-s/f; n-doping Reduction/undoping Doped/Conducting state NeutraVsemi-conducting state Doped/Conducting state Fig. 1.3. Scheme showing the reversible p-doping (oxidation) and n-doping (reduction) redox processes ofpolythiophene, with the incorporation of required cation/anion species (X+/A-) to balance the charge of the polymer matrix. Oxidative doping of polyheterocycles results in a radical cation and ensuing structural change, known as a polaron (Fig. l . 4). 30 On further oxidation, the free radical is removed creating a new spinless defect called a bipolaron.3 1 The polaronlbipolaron charges are able to move freely along the polymer backbone and even 'hop' between polymer chains, acting as charge carriers and enabling electrical conduction. 32 3 x x x (a) Neutral Chain x x (b) Polaron x (c) Bipolaron Fig. 1 .4. (a) The semi-conducting, aromatic, neutral chain. (b) Formation of a polaron species. (c) The conducting, quinoidal, bipolaron species formed on further oxidation 3!,33 X = NH or S, A- = anion (e.g. cr, CI04} An energy level diagram representing the monomer, neutral polymer, and oxidised polymer with fonnation of polaronlbipolaron species, is shown in Fig. 1 . 5 . 34 On polymerisation, the HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) of the monomer are broadened. 34 The resulting n and n* molecular orbitals can be described as a fully occupied valence band and an empty conduction band respectively. 35 The energy difference between these two orbitals is known as the band gap, and determines important properties of the polymer such as the ability to conduct, and the light energy (wavelength) at which the polymer absorbs or is capable of emitting. The oxidised speCIes display energy levels that appear between the valence and conduction bands, which effectively narrow the band gap. 34 This results in enhanced conductivity and a bathochromic shift in wavelength absorbed by the polymer. In highly doped polymers, these energy levels can overlap with the valance and conduction bands to produce quasi-metallic behaviour. 25 Generally, polymers with a longer effective conjugation length have broader conduction and valence bands, resulting in a smaller band gap and consequentially are capable of higher conductivities. 36,37 4 Energy LUMO , ­ ': , - - HOMO.:' ' -' , ... ... ... .. ... Monomer Conduction Band II' 7t ---> 7t* Valence Band Neutral Polymer Ba nd gap Conduction Conduction Band Band Conduction Band • Valence Valence c;[] Band Band Band Polaron Bipolaron Quasi- metaUic state Fig. 1 .5. Energy level diagram showing the difference in band structure of a conducting polymer in the neutral state, and oxidised state ... vith formation of polar on and bipolaron species.J4 Doping is usually associated with a change in morphology of the polymer matrix, due to incorporation of counterions (and possibly associated solvent molecules) required to balance the charge. 29 Oxidative doping usually results in expansion by incorporation of anionic species, and dedoping is associated with the expulsion o f anion species causing contraction. 29 This ability of the polymer to expand and contract on oxidation and reduction has been utilised for applications in actuator materials. 38 A v aluable property of some conducting polymers is the ability to be readily modified by the attachment of substituents . 1 4,23 Functionalisation by covalent bonding of specific groups to the polymer backbone allows the properties of the polymer to be modified?9 For example, the conductivity and colour absorbed by a polymer can be varied by the attachment of electron withdrawing/donating substituents,40 which alter the polymer bandgap,28 and solubility in organic solvents can be increased by attachment of alkyl and/or alkoxy chains.41 Prosthetic groups, which interact with the environment, can also be attached to give additional functions. For example, porphyrins can be attached to form light harvesting polymers42 and polyether groups to create cation sensors. 28 The conjugated polymer chains are able to function as paths, or molecular wires, for the reversible transport of electrons between the functional groups and an electrode. 14 5 Thiophene is one of the most readily fimctionalised monomers, allowing a wide range of functionalised polythiophenes to be easily prepared. 8,43 In addition, polythiophenes connnonly have relatively high conductivities,44 excellent processing possibilities,45 and form some of the most chemically stable materials.41 This makes them one of the most widely investigated linear conjugated systems from both fundamental research and industrial viewpoints. 46 6 1 .3 Th iophene based polymers 1.3. 1 Polymerisation Polymerisation o f thiophene precursors can occur through either the a or � positions (Fig. 1 . 6) although po lymerisation through the more reactive a position (Fig. 1 . 6b) is usually more favorable. 44 P olymerisation through the � position (a-� or �-�, Fig. 1 . 6 c & d respectively) produces undesirable defects in the polymer backbone, which may lead to reduced overlap between 1t orbitals and consequentially lower conductivities. 23 (a) :a� s s (d) S Fig. 1.6. (a) aJ� positions on a thiophene ring, (b) a - a coupling, (c) a - � coupling and (d) � - � coupling. The currently accepted mechanism for the polymerisation of polythiophenes is d · 1 d ' F' 1 7 8 27 47 48 lSp aye ID Ig . . . ' . . Polymerisation is initiated by the oxidation o f two monomer units to produce radical cations, which subsequently couple to form a dirner. Chain propagation o ccurs by coupling of monomer and/or oligo meric radical cation species. 7 o 5 - e 10 • 5 H 0---0 O--J:J • • + 2H+ � # +5 H 5+ 5 5 O--J:J 5 5 + 10 5 • 5 • • • 5 5 5 Fig. 1.7. Proposed mechanism for the polymerisation of thiophene 8 Functionalisation of thiophene at one or both of the � positions ( Fig 1 . 8) is o ft en used to modify the properties of the resulting polymer. P o lymerisation of thiophene monomers, however, o ften requires oxidation at potentials (thiophene: + l . 65 V vs S CE)47 that cause over-oxidation (irreversible oxidation) o f polymer functional groups, or coupling through the �-position (seen by NMR).49-5 1 This is known as the polythiophene paradox. Fig. 1.8. Functionalisation of thiophene through the p positions. One method of alleviating this pro blem is by usmg bithiophene or terthiophene monomers, which require lower oxidation potentials ( l . 2 0 V and l . 05 V vs SCE 8 respectively) . 47,52 Howev er, the resulting polymer length has also been found to decrease with increasing monomer length53 due to a lower reactivity of the longer o ligomeric cation radicals. 54 The attachment of appropriate substituents may also be used to decrease the polymerisation potential by influencing the charge distribution and increasing the reactivity of the rnonomer and cation intermediate. 25 It has been observed, however, that steric effects between bulky substituents can hinder p o lymerisation by physically obstructing two species from co upling, thus increasing the p otential required for polymerisation.43 S ev eral groups54,55 have reported that electronic effects induced by substituents strongly affect the reactivity of the monomer and hence the resulting polymer length. Electron-donating substituents have been reported to decrease the oxidation potential o f the mono mer by increasing the stability of the cation radical formed during the p o lymerisation process. The decreased reactivity of the resulting radicals leads to a lower degree of polymerisation. Electron-withdrawing substituents appear to exert the opposite effect resulting in more reactive cation radicals and enhanced polymerisation. Howev er, Randriamahazaka et al. 56 found that thio phene monomers substituted with strongly electron-withdrawing substituents (e.g. 3-cyanothiophene and 3 - nitrothiophene) that have oxidation potentials approximately 0. 5-0. 7 V higher than thiophene, do not polymerise at all. This is thought to be due to the highly reactive radical cations initially formed on monomer oxidation undergoing competing reactions with solvent or anion molecules rather than polymerisation. 56 M onomer units that are not centrosymmetric, such as monosubstituted thiophene, can c o uple in three orientations: head-to-head (HH), head-to-tail (HT) and t ail-to-tail (TT) as sho\VT1 in Fig. 1 . 9. P o lymerisation can consequentially result in regioregular (repeated sequence of coupling o rientations) or regiorandom (arbitrary sequence of coupling orientations) polymers (Fig. l . 1O). If there are significant steric interactions between functional groups, these polymers can exhibit significantly different properties. 57 9 R R 2 d • HR S R R S HT R R s TT R Fig. 1 .9. Possible regioisomers formed by coupling of two, non-centrosymmetric monomer units: head-to-head (HH), head-to-tail (HT) and tail-to-tail (TT). (a) (b) Fig. 1. 10. Examples of (a) regioregular and (b) regiorandom fimctionalised polythiophenes. The regioregularity of substituted polythiophenes, as well as o ther critical properties such as polymer length and a.-� linkages, may be influenced by the polymerisation method. 58 1 0 1.3.2 Polymerisation methods Polymerisation of monomers is usually perfonned usmg either chemical27,58,59 or electrochemical methods. 8,28,60 Electrochemical synthesis allows deposition of polymer films directly onto a wide range of conductive substrates, including glassy carbon, indium-tin-oxide (ITO) and metals. It provides the opportunity to control film thickness by limiting charge passed, can be used to produce films of insoluble polymers, and allows an in situ electrochemical study of the polymerisation process. In addition, electrochemical polymerisation allows a large range of counter-ions to be easily incorporated into the polymer film that can determine its electronic and mechanical properties. Electrochemical polymerisation can be achieved usmg potentiodynamic (cyclic voltammetry), potentiostatic (constant potential) or galvanostatic (constant current) methods, although the latter method has been shown to produce films of poorer quality than the former metho ds.6 1 The morphology and properties of polymer films have been found to depend upon many parameters such as the electrode surface, the electrolyte solution, temperature and potential 1imits applied 61 Polymerisation of alkylthiophenes using electrochemical techniques has been shown to produce polymers with a regiospecificity of about 70% head-to-tail linkages. 62 A cyclic voltammogram for the growth and deposition of polyterthiophene is shown in Fig. 1 . 1 1 . An increase in current is 0 bserved with each cycle. This is typical of conducting polymer growth and is due to an increase in electro active surface area and/or more facile electrode kinetics. An oxidation onset potential of 0. 7 1 V vs Ag/Ag+ is evident. 1 1 400 300 200 � - 1 00 1: � 0 :::J 0 -1 00 -200 -300 0.0 1 �d.tioo 1 Reduction 0.1 0.2 03 Q4 Q5 Q6 Potential I V VS Ag/ Ag + 0.7 0.8 0.9 Fig. 1 . 1 1. Potentiooynamic growth of terthiophene on a platinwn disc electrode (surface area (SA) = l . 8 mm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: 01+900 mY. 9 cycles. Scan rate: 100 mV S·l Chemical synthesis allows the production of large amount s of material and improved control over the ass embly of monomer units by certain methods. Reproducibility of chemical polymerisations to produce identical polymer properties, however, is knO\vn to be difficult 6 I There are many different chemical polymerisation methods employed. 27 One commonly employed method for the synthesis o f polythiophenes is the Grignard method (GRIM), which uses a Ni catalyst (Nio or Ni(dppp)Ch) as shown in Fig. l . 1 2. 45 This method typically gives regioregularity in poly(alkylthiophenes) of �90%. 27,62 R n MeMgBr Br�s�Br THF R .. n [Ni] .. Br�s�MgBr R ~ Fig. 1. 12. Synthetic polymerisation of functionalised thiophene using the Grignard method. m+ m Br" An alternative to p olymerisation by directed chemical synthesis is polymerisation by chemical oxidation. This is a more direct method than chemical synthesis since the monomer does not require mo dification at the a-positions. The oxidation of thiophene 1 2 derivatives using iron(II I) chloride is commonly used to prepare polythiophenes from thiophenes,63-66 bithiophenes27,67,68 and terthiophenes. 44,63,69-71 Iron(III) chloride has been shown to produce polymers of high molecular weigheD and regioselectivity of -80% in alkyl thiophenes.62,63,68 The oxidation potential of iron(III) chloride and quality of the resulting polymer, such as conductivity and solubility, has been found to depend strongly upon oxidant concentration,61 and impurities such as water or ethano1.70 A typical reaction using iron(III) chloride is given in Fig. l . 1 3 . The polymer is initially produced in the oxidised state, which is insoluble and difficult to process. The polymers are generally made more processable by reduction to the more soluble, neutral state. 62 R d s 4 equiv. FeCl3 .. DCM or CHCl3 R m+ Fig. 1 . 13. Oxidative polymerisation offunctionalised thiophene using iron(III) chloride. 1.3.3 Effect of substituents on polythiophenes properties R Thiophenes can be functionalised by a wide range of substituents (varying in properties such as electron donating/withdrawing effects, size and shape), which may alter the electronic, optical and physical properties of the resulting polymer.36 These properties can be tuned by modification o f the substituents to give desired polymer properties. Substituent effects can be categorized into ( 1 ) electronic effects8 and (2) steric effects 72 In theory, polymers substituted with electron-withdrawing groups (EWGs) should more readily undergo n-type doping incorporating a cation into the polymer structure, and polymers substituted with electron-donating groups (EDGs) should more readily undergo p-type doping incorporating anions.69 EDGs are known to stabilise the oxidised (conducting) state of the polymer, which is useful for many applications. 8,73 I f 1 3 the oxidised state is too stable, however, difficulties may arise in reduction to the more soluble and o ft en required neutral state.62,74 Polythiophene derivatives that are stable in a reduced state would be useful for some applications such as charge storage (supercapacitors and organic batteries) and so lar cells . 75 Altho ugh many p olymers show satisfactory performances in tenns of p-doping, polymers that show a high level of n-doping and are stable in the reduced state are less common. One way to improv e n-doping is to attach electron-withdrawing substituents to the po lymer backbone. 2 8 The HOMO and LUMO energies (and hence bandgap) of polythiophene derivatives can be tuned by adj usting the electron density in the polymer thro ugh the use of electron-withdrawing and/or donating s ubstituents . This feature may be utilised for many applications, including OLEDs for display devices, so lar cells and OFETs. 8 The introduction of substituents, however, may lead to steric interactions between the substituents, causing twisting between thiophene ringS . 76 This results in reduced overlap between 7t orbitals, which increases the band gap and consequently lowers the conductivity of the polymer. 5 8,77 Steric effects can be reduced by two methods. Firstly, unsubstituted thiophene rings can be added t o the polymer chain to move bulky substituents apart from each other. This can be achieved by employing substituted thiophene oligomers (bithiophene, t erthiophene, tetrathiophene) rather than thiophene/4,69,78 or by copolymerising substituted thiophene with unsubstituted bithiophene. 1 5,79 The use of a ' spacer' between the polymer backbone and bulky substituents may also be used to reduce steric hindrance. 8o Alkyl 1 5 or alkOxy8 1 sp acers are co mmonly used as shown in Fig. 1 . 1 4a and b respectively. 1 4 Fig 1 . 14. Examples of (a) alkyl15 and (b) alkoxy81 spacers used to reduce steric hindrance of functional groups between thiophene units. To allow electronic communication between the polymer backbone and substituent, coQjugated linkers, such as phenylene or vinylene, can be employed. 25 1.3.4 Styryl-substituted polythiophenes Functionalisation of thiophene by substituted aromatics (Fig. l.15) is commonly used to influence the electronic properties of the resulting polymer. 70,82-84 The ability of aryl­ substituents to improve the charge capacity and n-doping ability of polythiophenes85 has attracted interest in these materials for applications such as rechargeable battery electrodes86 and electro chemical capacitors.87 However, Visy et al. have shown that aryl groups directly bonded to terthiophene force the monomers into a non-planar conformation, thus preventing the extension of the conjugation to the substituents and reducing the conductivity of the resulting polyrner. 54 R Fig. 1 . 15. 3-Aryl substituted polythiophene. I S In 1 994, Smith et al. 88 reported an attempt to improve the planarity and therefore conductivity of aryl-substituted polythiophenes ( 1 40 cm- I) by introducing an unsaturated ethylene spacer linkage between the substituted phenyl rings and the polythiophene backbone. The monomer units that were investigated are displayed in Fig. 1 . 1 6. It was found, however, that polymerisation of these monomers did not produce electroactive polymers. This was explained by Smith et al. ,48 who revealed by using theoretical calculations, that the 5-position has a very low spin density (reactivity). Although dimerisation through the 2-position was possible, further polymerisation could only be achieved through the more reactive vinylene linker (7- position), causing undesirable defects. 4� � 3 5 2 S 6 R R H N02 NH2 OMe Fig. 1. 16. Styry1-substituted thiophene monomers investigated by Cutler. 25 It is possible, however, to synthesise electroactive styryl-substituted polythiophenes by copolymerisation: Welzel et al. 89 reported the successful electrochemical copolymerisation of styryl-substituted thiophene with methylthiophene, and Cutler25 with bithiophene. Greenwald et al. also achieved chemical co-polymerisation of nitro­ substituted styrylthiophene with butylthiophene.90 The preparation of an electroactive styryl-substituted poly thiophene derivative has also been reported by the electrochemical polymerisation of a substituted styrylterthiophene.91 The synthesis of styryl-substituted terthiophene (Fig. 1 . 1 7) was reported by Collis et al. in 2003.92 These workers used Wittig chemistry via both the terthiophene aldehyde and terthiophene phosphorous species as shown in Fig. 1 . 1 8. 78 This procedure allows a wide variety of substituents to be attached. 1 6 R s s Fig. 1. 17. Terthiophene functionalised through a vinyl linker. R Base o �H CHO S ,-{ . · _,s \J'� �PPh:Y( • Base (2) ( l ) (3) Fig. 1. 18. Synthesis of vinyl-substituted terthiophene using Wittig reactions. R = Ph-R. X = Br, Cl or 1 . Grant ) ) has attached polyether chains and crown ethers to styryl-terthiophene units using this method (Fig. 1 . 1 9), with the purpose of producing ion responsive systems. As expected, an interaction of the ether functionality with cations was found to produce a change in the spectroscopic properties of the terthiophene unit. \ ) ? Fig. 1 . 19. Examples of (a) crown ether and (b) polyether-substituted styrylterthiophene derivatives investigated by Grant. ) ) 1 7 Other workers have investigated variants of vinyl-substituted terthiophene-based polymers for photovoltaic applications. 25•78 Burrell et aC8 attached a porphyrin dye as shown in Fig. 1 .20a in an attempt to further increase the extended conjugated structure and enhance the light harvesting capabilities of the material.42 Also using the Wittig reaction, Cutler25 attached an electron-withdrawing nitro-substituent (Fig. 1 .20b) to enhance charge separation and increase the photovoltaic performance of polyterthiophene photoelectrochemical (PE C) solar cells. This nitro derivative was electropolymerised and observed to produce a photovoltaic response in these cells. Ca) Cb) Fig. 1. 20. Ca) Terthiophene substituted 'vvith a tetraphenyl phorphyrin (TPP) via a vinyl linker78 and Cb) nitro-substituted styryiterthiophene synthesised and investigated by Cutler 25 Polymerisation of these materials, however, generally produces insoluble materials that can not be further processed.93 Attachment of alkyl and/or alkoxy chains is one method commonly used to increase the solubility of polythiophenes. 94,95 1.3.5 Alkyl and alkoxy substituents Extensive research has been undertaken into the effect of alkyl substituents on polythiophenes.57•73 If sufficiently long, the saturated chains are able to solubilise the polymer in common organic solvents such as DCM, THF and chloroform, which provides considerable advantage for further processing of the polymer.96 The chains can also reduce the melting points of alkylthiophenes to allow processing from a melt. 96 Furthermore, regioregular alkyl chains have been shown by X-ray and light- 1 8 scattering studies to improve supramolecular organisation through interdigitation (Fig. l . 2 1 ).62.97 This can decrease the resulting band-gap and improve conductivity. 73 Fig. 1 .21 . Supramolecular assembly of 3-hexylthiophene by interdigitation of alkyl chains. 97 Alkoxy chains have also been found to improve polymer solubility 62 However, less work has been focused on alkoxy-substituted polythiophenes than alkyl-substituted polythiophenes, presumably due to the high stability of the alkoxy derivatives in the less soluble oxidised state.98 Their high stability in the oxidised state means alkoxy derivatives are difficult to reduce, and hence are not very soluble or processable.98 The high stability of resulting cationic species in some alkoxythiophene derivatives, has also been found to impede polymerisation and reduce the length of resulting alkoxythiophene polymers. 8 The use of alkoxybithiophene as a starting material by Zotti et al. 99 was found to moderate this problem. Alkoxy derivatives do, however, present many advantages over their alkyl counterparts. The stronger electron-donating effect of the alkoxy substituents 1 9 decreases the oxidation potential of the monomer,32 although Gallazzi et al.62 reported that the reactivity of terthiophene monomers depends upon the substitution position of the alkoxy group. Substitution at the 4,4"-position (Fig. 1 . 22a) was noted to increase monomer reactivity (decrease oxidation potential) more than substitution at the 3 , 3 "­ position (Fig. 1 . 22b). (a) 4" RO OR Fig. 1. 22. Alkoxy substitution in the (a) 4,4"-positions and (b) 3,3"-positions on terthiophene monomers. Alkoxy substitution has also been found to generate smaller polymer band-gaps due to increased electron density on the thiophene ringS.99 This results in lower polymer oxidation potentials and a higher polymer stability in the doped state,32 which is useful for many applications. It has also been suggested by Gallazzi et al. 62 that the higher doping stability may be related to better interactions between alkoxy chains and dopant molecules. In addition, the presence of an oxygen atom in a 4-position has been shown to allow a more planar polymer chain by reducing steric hindrance between alkoxy substituents/9,99 and possibly by producing a favourable interaction "vith the sulphur ato m 1 00 This increase in planarity improves 1t overlap and the effective conjugation length, which in turn leads to enhanced o ptical and electrical properties. 99 No detailed study is published on the relationship between the chain length of alkoxy substituents on polythiophenes and polymer solubility, but it has been well established that an increase in chain length of alkyl substituents produces an increase in polymer solubility.6 1 Although an increase in alkyl chain length increases the interchain distance,62 it has also been found to increase the degree of order and planarity of the 20 polymer backbone since this is mainly governed by intermolecular interactions of the alkyl side chains .20 1.3.6 Thienylenevinylene based conducting polymers To date, thienylenevinylene-based polymers (Fig. l . 23) have been found to generate the smallest HOMO-LUMO band gap o f all 1t-conjugated polymers of comp arable chain length. 46 Po ly(thienylenevinylene) has been found to produce a band gap that is 0.20 - 0.3 eV lower than that for poly thiophene (2. 67 eVI O I ). 1 02 This is caused by an increase in the effective conjugation length due to a suppression of the rotational diso rder of the polymer chain created by the rigid ethylene linkages. 46 The exhibition of low band gaps by thienylenevinylene derivatives has prompted interest in these materials fOT use in applications such as photovoltaics l 03, 1 04 and field-effect transistors. 1 05 (a) (b) Fig. 1. 23. (a) Po1y(thieny1eneviny1ene) and (b) po1y(terthieny1ene-viny1ene). As o bserved for polythiophenes, Ono et al. has fo und that chain extension of thienylenevinylene polymers results in a bathochromic (red) shift of the absorption maximum, and a reduction in oxidation potential. 100 These workers also reported that electron-donating substituents, such as alkoxy groups, reduce the oxidation potential o f thienylenevinylene monomers . I OO Other workers have achieved solubilisation of the resulting poly(thienylenevinylene)s by attaching alkyl substituents . , 02, , 06, , 07 21 1 .4 Scope of work The aim of this study was to investigate the effect of electron-withdrawing and electron-donating substituents on the polymerisation of styryl-substituted terthiophene and terthienylenevinylene derivatives, and the influence of these substituents on the electronic and spectroscopic properties of the resulting polymer. As previously discussed, the use of functionalised terthiophene and terthienylenevinylene oligomers as mono mer units reduces the oxidation potential and decreases steric interactions between bulky substituents. The conjugated styryl linker also alleviates steric hindrance between substituents, and allows electronic communication between the functional group and polymer backbone. 25 The chemical and electrochemical polymerisation of a senes of styryl-substituted terthiophenes (Fig. 1 . 24, m = 0) was investigated in this study, and the resulting materials characterised usmg mass spectrometry, UV-VIS-NIR and NMR spectroscopy, and microscopy (Chapter 3). For comparison, the electrochemical polymerisation of a corresponding series of styryl-substituted thienylenevinylenes (Fig. 1 .2.4, m = 1 ) was also examined (Chapter 4). In order to develop processable polymers for a range of applications, the chemical and electrochemical polymerisation of alkyl and alkoxy-substituted styrylterthiophene was performed (Chapter 5), and preliminary studies of the resulting soluble materials were conducted in applications of solar cells, actuators and batteries (Chapter 6). Attempts were also made to develop high surface area electrode materials using one of these polymers. R Fig. 1 .24. Styryl-substituted terthiophene (m = 0) and terthienylenevinylene (m = 1 ) materials investigated in this study. 22 Chapter 2 Experimental Methods 2. 1 E lectrochemical techn iques 2. 1.1 Introduction Electrochemistry can be used to produce a redox reaction by applying a potential between two electrodes. In the case of conducting polymers, it can be used to oxidise monomeric species in solution to form a polymeric film on an electrode surface. TIlls film can then be repeatedly oxidised and reduced by applying appropriate potentials, as evidenced by the flow of electric current . When compared to other methods used for synthesising conducting polymers such as electrospinning7 and chemical metho ds,27 electropolymerisation presents several distinct advantages. These include absence of a catalyst, direct grafting o f the doped conducting po lymer onto the electrode surface, and ready control of film thickness by the deposition charge. Electropolymerisation also offers the prospect of an in-situ study of the growing process by electrochemical and/or spectroscopic techniques. 2. 1.2 Cell design Electrochemistry was performed in this study usmg a standard one compartment, three-electrode cell that consisted of a working electrode (WE), counter electrode (CE) and a reference electrode (RE). A schematic of the set-up is shown in Fig. 2. 1 . In three-electrode circuits, the working electrode is where the electrochemistry of interest occurs. The potential of the WE is monitored by a reference electrode and controlled by the potentiostat. No current flows through this reference electrode; instead, a counter electrode is introduced to complete the circuit. The potentiostat controls the potential difference between the WE and RE by altering the current flowing though the WE - CE circuit. 23 Potentiostat 0/ F0 RE CE \l iVE , I Fig. 2. 1. Schematic of a three-electrode one compartment cell. At the WE, the generation of Faradaic current takes place due to an electron transfer process.9 1 A wide variety of types and sizes of substrates may be used for the WE so long as they are conductive. A platinum microelectrode or disc electrode was used in this study for the electrochernical analysis of polymer growth and the resulting polymer. Microelectrodes have very small surface areas, typically o f ca. 1 0 ,..1m2 The use of electrodes of micron dimensions provides several advantages over standard electrodes with larger surfaces areas (>1 00 J..I.m2). These include a decrease in the ohmic potential (iR) drop at the electrode/solution interface (as very low currents are produced), a decrease in the production of an electrode/solution capacitance, and a change in the diffusion field from the bulk solution to the electrode surface from linear to radial (Fig. 2 . 2). 1 08 The formation of a radial diffusion field allows the effects of diffusion of species to the electrode surface, such as mo no mer and/or electrolyte, to be minimised. 1 08 24 I I I I I [ [ Semi-infinite planar (standard) electrode Microelectrode (disc) Fig. 2.2. Illustration showing the linear and radial diffusion fields produced at standard electrodes (SA: > 100 �2) and disc microelectrodes.108 A platinum disc electrode, which has a higher surface area (1 . 8 mm2 ) was employed for the deposition and analysis of more soluble materials. Other working electrodes employed in this study include conductive fabrics, polymer-coated membranes and transparent ITO-coated glass for characterisation of thin polymer films by spectroscopic and microscopic techniques. The role of the CE is to balance the Faradaic process of the WE with an equal but opposite redox reaction, but has no effect on the Faradaic processes. The processes occurring at the CE are therefore of no interest in this experiment and no attempt is made to monitor this electrodes potential. Platinum or stainless steel mesh was used to ensure the CE provided a larger surface area than the working electrode, so the current flow would not be limited. A AgI AgN03 reference electrode \vith a 0. 1 M tetrabutylammonium perchlorate (TBAP)/acetonitrile (AN) salt bridge, as shown in Fig. 2.3, was generally used in this study. This electrode has an If = 0.098 V vs the Fc+IFc (ferrocene) couple and an If = -0. 300 V vs SHE. In this study, potentials are measured and quoted with respect to this Ag/ Ag+ electrode, except for the spectroelectrochemical experiments in which a silver wire was used as a pseudo reference electrode (If = +0.380 V vs Ag/ AgN03). 25 0. 1 M TBAP/AN 0.0 1 M AgN03 and +--+-- 0 . 1 M TBAP in AN �----1r--- Ag wire Vycor tips Fig. 2.3. Schematic of the Ag/O . O l M AgN03 reference electrode and TBAP/AN salt bridge used in this study. Monomer concentrations of 5 mM were generally employed for electrochemical analysis. The solvent used for electropolymerisation is an important factor influencing not only the quality of the material obtained, but also its conductivity, morphology and subsequent electrochemical and chemical behaviour.6 1 , 1 09 A 1 : 1 mix of acetonitrile (AN) and dich1oromethane (DCM) was most commonly used in this study. Acetonitrile is a versatile and popular polymerisation solvent for several reasons. Firstly, it provides poor solubility or insolubility of most oligomers (materials with chain lengths comprising less than ten monomer units) and conducting polymers, which allows nucleation and growth on an electrode surface. I I O It is also slightly basic, assisting proton removal for the polymerisation process. 109 It was found, however, that some monomers investigated in this study have poor solubility in acetonitrile. For this reason, addition of DCM to the solvent mix was sometimes required. A I : 1 AN :DCM mix was found to dissolve all monomers involved in this study at a concentration of at least 5 mM, but provided a sufficiently low solubility of longer oligomers as they were formed so they would adhere to the electrode. 26 Tetrabutylammonium perchlorate (TBAP) is a commonly employed electrolyte and was used in this study in large excess ( 1 00 mM, electrolyte:monomer ratio of minimum 20: 1 ) to ensure the current flow was not limited. All solutions were freshly prepared and degassed by sonication or nitrogen purging immediately prior to experimentation. 2. 1.3 Electrochemical techniques Introduction Cyclic voltarrunetry (potentiodynarnic grow1h) and constant potential (potentiostatic growth) have been used to deposit the materials investigated in this study. Although galvanostatic growth (constant current) procedures can be used when control of charge is desired, this method has a propensity to yield polymer of poorer morphology and conductivity. 1 09, 1 1 1 Cyclic Voltammetry Voltarrunetry is an electroanalytical technique in which redox infonnation about the analyte (monomer, oligomers or polymer in this study) can be obtained by measuring the current as a function of applied potential. In cyclic voltarrunetry, the potential is applied in a saw tooth mode to the working electrode as demonstrated in Fig. 2.4. The cycle commences at a predetermined potential. The potential is increased or decreased in a linear fashion with a fixed scan rate (usually expressed in mV S-I ) to the maximum/minimum limit. The direction of the linear scan is immediately reversed (at the same rate) until the opposite potential limit is attained. The scan direction of the potential scan is again immediately reversed until the starting potential is reached. This cycling may be repeated many times. 27 Time Fig. 2.4. Variation of potential with time for cyclic voltammetry. A typical cyclic voltammogram (CV), usmg ferrocene as an example, is shown in Fig. 2.5 . Two peaks are displayed due to a single reversible electron transfer process. 1 1 2 The peak on the positive scan (forward scan) is due to oxidation of ferrocene to a positively charged state, and the peak on the negative scan (reverse scan) is due to reduction back to a neutral state. From cyclic voltammetry, the oxidation potential and reduction potentials of the FclFc+ couple can be estimated as shown. The total charge passed during the oxidation/reduction process can also be determined by measuring the area under the curve. et Forward scan - ... c e � :::J o > < Reverse scan i Oxidation > Potential I V I t Reduction Fig. 2.S. A CV of ferrocene, demonstrating a single reversible electron transfer process on a platinum disc electrode (SA = 1 .8 mm2). 28 Although cyclic voltammetry is often used as a characterisation technique, oxidation of some species may result in insoluble compounds, which are deposited onto the electrode surface. In this way, monomers can be oxidised to form films of oxidised polymer attached to the electrode surface. The growth CV of terthiophene on a platinum disc electrode is displayed in Fig. 2 .6. The initial cycles differ from subsequent scans with respect to charge passed, oxidation threshold potential, and the shape of the current-response curve. As each scan represents further growth of polymer on the electrode, this response is most likely due to deposition onto different substrates with each scan as the platinum surface is modified with the deposited material. The observed increase in current with successive scans is consistent with an increase in electroactive surface area and/or more facile electrode kinetics due to the change in electrode substrate (platinum to polymer) as a polymer film is deposited. A higher electroactive surface area allows more polymer to be deposited and hence more current to flow. In contrast, deposition of a non­ conductive polymer film would cause a decrease in current as the electrode becomes insulated from the solution. Also evident is a decrease of the oxidation onset potential after the first cycle. This suggests that polymer growth on the deposited material is more readily achieved than growth on the polished platinum electrode, possibly due to the formation of nucleation sites. 400 300 200 � - 1 00 C � 0 a - 100 -200 -300 0.0 0. 1 0.2 0.3 0.4 0.5 0.6 Potential I V vs Ag/Ag+ 0.7 0.8 0 .9 Fig. 2.6. Potentiodynamic grQ\.vth of terthiophene on a platinum disc electrode (SA = 1 . 8 mm2). Supporting electrolyte: 0. 1 M TBAPI AN. Potential limits: 01+900 m V. 9 cycles. Scan rate: 100 mV so ' . 29 One problem with usmg cyclic voltammetry as a growth method of conducting polymer films, is that over-oxidation can easily occur if scan limits are not set correctly. 1 1 2 An electrochemical 'window' exists for most conducting polymers within which the doping/dedoping process is mostly reproducible (although not necessarily reversible). Anodic potentials which are set too high have been shown to cause irregularly a,p-linked rings 1 13, oxidation of functional groups,1 I 3 or reactions of the polymer with solvent molecules or impurities (e.g. water or gas).70 Each mono mer was initially analysed by scanning between potentials of -1 V and + l . 5 V to determine appropriate scan limits for electrochemical growth. The anodic limit was set at a potential slightly higher than the potential at which material was observed to adhere to the electrode surface. Films of conducting polymer deposited on an electroactive surface (by methods such as electrochemical growth, or casting or spin-coating of a polymer solution) may be investigated by cyclic voltarnmetry. This is usually achieved by cycling the film, as the working electrode, in a mono mer-free electrolyte solution The post-polymerisation CV of an electrochemically polymerised dialkoxystyryl-substituted terthiophene (OCIODASTT), deposited on a platinum micro electrode is shown in Fig. 2.7 . In comparison to the narrow peaks produced by ferrocene (Fig. 2.4), this CV displays broad oxidation and reduction peaks, typical of redox processes for a number of species and indicating a wide range in polymer lengths. From this CV, the oxidation onset potential, Eoxidation onset, of the polymer can be determined, and the total charge passed during the cycle calculated from the area inside the curve. 30 8 � 4 -4 -8+-------,--------.-------.-------.,-------.-------.---� -0.5 -0.3 -0. 1 0. 1 0. 3 0.5 Potential ! V vs AgI Ag + 0.7 Fig. 2.7. Post-growth CV of OClODASTT on a platinwn micro electrode (SA: 10 1ffil2). Electrolyte solution: 0. 1 M TBAP 1 : 1 AN:DCM Potential limits: -500/+800 m V. Scan rate: 100 mV S·l Potentiostatic growth Potentiostatic growth involves ' stepping' the potential from a potential that produces no reaction, to one that does. The potential is then held steady as the current pro duced is measured as a function of time. A typical potential vs time profile and corresponding chronoamperogram are shown in Fig. 2. 8. A sharp current transient corresponding to the potential step is observed. This is due to charging of the double layer (layer between the electrode and solution), which produces a non-Faradaic current. Oxidation of the monomer then produces a Faradaic current. If a conducting polymer is formed, the current should increase due to an increase in the electroactive (conducting polymer) surface area.6 J This is o bserved in the chronoamperogram of the growth of styryl terthiophene (SIT) polymer, shown in Fig. 2.9. Polymer growth, as observed by the current produced, may be limited by diffusion of monomer from the bulk solution to the vicinity of the electrode.47 Potentiostatic growth is useful as the thickness of the film can be easily estimated and controlled by measuring the charge which has passed. 3 1 (a) (b) - s:: ! ... :::J o Time Time Fig. 2.8. (a) The function of potential versus time with (b) the corresponding chronoamperogram. 300 250 � 200 - c 1� e ... :::J 0 100 � 0 0 2 4 6 8 10 12 Time I s 14 16 18 20 Fig. 2.9. Potentiostatic grm.vth of polySTT on a Pt micro electrode (SA: 10 j.UIl2). Solvent: 1 : 1 AN:OCM Potential held at -500 m V for 1 s, then stepped to 900 m V for 1 9 seconds. 32 2.2 MALDI-TOF mass spectrometry 2.2. 1 Introduction The average molecular weight and polydispersity (range of weights) of conducting polymers determine fundament al properties such as conductivity, colour and solubility. 1 1 4 Matrix Assisted Laser Desorptionllonsition Time-Of-Flight Mass Spectrometry (MALDI-TOF MS) is a well established and frequently used technique for studying the chain length of conducting polymers,64.73.96. 1 1 5 and has been used in this study to investigate soluble fractions o f chemically and electrochemically synthesised polymers. Compared to other mass spectrometry techniques such as Fast Atom Bombardment (F AB) and Electrospray Ionisation (El), MALDI -TOF MS is a relatively 'soft' technique in that it results in little molecular fragmentation. This is desired for polymer analysis since a relatively simple spectra displaying the mass of p arent ions is produced, and a more accurate distribution of polymer masses can be obtained. MALDI-TOF MS also has a \vide mass range of up to 300000 Da. 1 1 6 Like most mass spectrometry techniques however, MALDI-TOF MS is not quantitative and the intensity of peaks depends on many v ariables including the type of sample, sample preparation and instrumental pararneters. 1 1 7 2.2.2 Instrumentation In this study, mass spectrometry was performed using a Micromass MALDI LR mass spectrometer (Micromass Ltd . , M anchester, UK), with a 337 nm nitrogen laser with a pulse duration of 2 nm, an acceleration voltage of 15 kV, and MassLynx control and processing software. A fast dual micro-channel plate (MCP) detector was used, with signal gating for the rejection of low mass matrix ions. The instrument was used in linear mode to maximise sensitivity. A pulsed laser beam is directed at a sample disso lved or suspended in a matrix as shown in Fig. 2. 1 0. The matrix absorbs the laser light energy causing it to vaporise and 33 carry some of the sample with it. The matrix may subsequently help to ionise the analyte molecules, which are then carried into a mass analyser. Laser light 3000 V Timt>-of-tlight mass analyser . . . . . . . . . . ------�-------------------.--------------------. t Beam of ionised analyte molecules Sample and matrix on target plate Fig. 2.10. Schematic ofMALDI-TOF MS instrumentation. Detector TOF mass analysers are relatively simple. They are based on accelerating a set of ions to a detector with the same amount of energy. Since the ions have the same amount of kinetic energy (EK) yet different mass (m), the ions have different velocities (v) and therefore reach the detector at different times. By calibration of the instrument with compounds of known masses, the mass of a sample may be determined from the time of flight. v =Jif v - Velocity EK-Kinetic energy m -mass The time-of-flight mass analyser can be used in two different modes: linear or reflectron. The fundamental difference between these modes is that the reflectron mode utilises an ion mirror which reflects the ions and creates a longer flight path (Fig. 2. 1 1) . By increasing the flight path and slowing down the ions, the ion mirror reduces the ion distribution and effectively increases the resolution. However, it also reduces the sensitivity and lowers the mass limit to below 1 0000 Da since heavier ions are not able to be reflected. 34 Pulsed laser light �, , , , , , � , , , I Detector Sample Probe Ca) Linear mode Pulsed laser� light " , , , , � , , , Ion mirror Sample Probe (b) Reflectron mode Fig. 2 1 1 . Schematic illustrating the difference betw-een the Ca) linear and (b) reflection modes used in MALDI-TOF MS. The linear method is much more sensitive than the reflectron method and is better suited for mass analysis between 5000 and 300000 Da However, the linear mode exhibits poor resolution when compared to the reflectron method and other mass analysis techniques. This is because the ions do not travel as far in the mass analyser and are therefore not separated as well before reaching the detector. It was also found in this study that the linear mode can be difficult to calibrate accurately. Spectra of a terthiophene polymer derivative which has been analysed by the linear and reflectron modes, are shown in Fig. 2. 12a and b, respectively, for comparison. Although the longest oligomer detected in the reflectron mode has a mass of �5300 Da, oligomers with molecular weights of up to �8300 Da can be detected in the same sample using linear mode. This demonstrates the enhanced sensitivity in detection of oligomers, particularly long oligomers, with the linear method. The reproducibility of spectra obtained from a sample when using the same mode (linear or reflection) is generally very good with peak intensities showing comparable distributions. Fig. 2 . 1 2c is an enlargement of one of the peaks produced by the linear method, most likely corresponding to an oligomer with a mass of 5946.32 Da, and demonstrates the poor resolution of the linear mode. As observed, the band is very broad and ranges over approximately 36 units from approximately 5939 to 5975 Da. 35 1407c reflectron am)I407c�f 50 (0 932) Sb (15 1407c linear amVI407chn 68 , 1 234. 3b , 15. 10 00 , Cm ,5 14�, 1 .7 1407c linear 3314.2 amyU07c1in 68 ( 1 �34 Sb �15,10 00 Cm (5 144 1 MCP detector: 1800V Laser energy: 790;' Mep detector: 2OO0V Laser anergy: 690Al 10000 12000 MCP detector: 2000V Laser energy: 69% 5954.5 59509 S!M9.lii 5960.8 59406.04 5965.7 5944. 5968.4 (a) O5-Oct-2004 Well: E6 TOf lO. 553 ___ -mIz (b) 16000 19000 (c) 05-O 1. 2) are often reported not to be representative of the mo lecular weight distribution, with the high-mass component underrepresented with respect to the low-mass component, resulting in significantly lower average molecular weight values.96. 1 1 7. 1 20.1 2 1 This mass-discrimination effect i s thought to be caused by several factors including mass-dependent desorption/ionisation, sample preparation, ion focussingltransmission and mass-dependent ion detection. 1 1 7. 122 39 2.3 1 H NMR Spectroscopy IH NMR spectroscopy can be a useful technique for the characterisation of soluble polythiophene derivatives. Defects in polymerisation such as a.-� coupling or substitution by nucleophiles derived from the oxidant (such as choride ions) may be determined. 62 In addition, if oligomers have equivalent protons at the terminal positions, the chain length of a polymer can sometimes be determined by comparison of the integration of signals due to the terminal protons with signals from protons at other sites. 1 14 For example, polymerisation of terthiophene results in a loss of equivalent protons at the terminal 'a.' positions as seen in Fig. 2. 1 3 . This is observed in IH NMR spectra as a decrease in integrated area of the signal due to this proton in relation to other signals. The length of the polymer can be determined by the ratio of the terminal protons to protons at the 3' and 4' position. 2 (H- /S" � /S" -H) -----.. H �'S'� Fig. 2. 13. Dimerisation of terthiophene ShO'vv1ng the loss in terminal protons. H Terthiophene monomers that are not centrosymmetric, such as the styrylterthiophene derivatives investigated in this study (Fig. l . 24, Chapter 1 ), show two signals due to inequivalent protons at the terminal positions, HH (at the 5 position) and HT (at the 5" position). These unsymmetric monomers can join in three different orientations: head-to-head (HH), tail-to-tail (TT) and head-to-tail (HT) (Fig. 2. 1 4). IH NMR spectra of pure samples of each of these isomers would be expected to show, respectively, a loss in signal due to the HH protons, equal signals due to the HT and HH protons, and a loss in signal due to the HT protons. Further polymerisation to produce non-regioregular polymer would be expected to show broad bands due to a range of different protons, but regioregular polymers may show less complicated spectra due to very similar protons.58 40 2 HH lIT IT Fig. 2. 14. Dimerisation of SIT to form three different possible regioisomers: the HH isomer (generated by coupling through the 5 positions), HT isomer (coupling through the 5 and 5" positions) and TT isomer (coupling through the 5" positions). 4 1 2.4 UV-VIS-NIR Spectroscopy Ultraviolet-visible-near infrared (UV-VIS -NIR) spectroscopy is a powerful t echnique for investigating the electronic properties of conducting polymers. As well as measuring the spectral absorbance of the polymer, properties such as the effective conj ugation length, the electronic effect of substituents, the p resence of chromophores and the degree to which the polymer is doped can be determined. All polymers in this study are electrochromic. In the neutral state, thin films usually appear either red or blue, but in the o xidised state they can appear green, or even translucent. This ability of the po lymer to reversibly change colour with potential provides promise for applications such as electrochromic display devices and 'smart windows' . 123 Monomeric units possess a HOMO (highest unoccupied molecular orbital) and a LUMO (lowest unoccupied molecular orbital) energy level as shown in Fig. 1 . 5. P olymerisation o f these monomers to give a highly conj ugated system results in broadening o f these bands to give a corresponding valence band and conduction band. In the neutral state, absorbance bands appear between 200 nm and 900 nm due to n � n* transitions, 1 24 as shown by spectra of 4,4 "-bis(decyloxy-3 ' -nitro-2 , 2 ' : 5 ' ,2"­ t erthiophene) (Fig. 2. 1 5). An increase in the effective conj ugation of the polymer leads to broadening of the valence and conduction bands, creating a decrease in the bandgap and a bathochromic shift in this spectroscopic band. 1 14 Oxidation of the polymer may result in the formation of polaronlbipolaron species. These oxidised species show bands which appear between the valence and conduction bands, effectively narrowing the bandgap (Fig. 1 . 5, Chapter 1 ), and are typically shown by spectroscopy as an absorbance between about 600 and 1 1 00 nm. 1 25 F urther oxidation of some conducting polymers may result in a quasi-metallic state being formed. This is shown by UV-VIS -NIR spectroscopy as a broad band extending beyond 1 000 run, known as a ' free-carrier tail' (Fig. 2 . 1 5) . 1 25-1 2 7 42 1 �--------------------------� 0.8 0.6 <{ 0.4 0.2 Free-carrier tail oL----���� 300 600 900 1200 1500 �_ 1 11111 0 .9 V Oxidised Polymer Neutral Polymer -0 .4 V Fig. 2.1S . Spectroelectrochemistry of 4,4"-bis( decyloxy-3' -nitro-2,2' : 5' ,2"-terthiophene) obtained by Gambhir et al. 1 28 Reprinted by permission. A free-carrier tail is displayed at wavelengths above 700 nm 126 Polymers in this study have been investigated both as films on ITO-coated glass (either cast or electrochemicaUy deposited) and, if soluble, in solution. 43 2.5 Imag ing techn iques The morphology of conducting polymer films may affect their physical properties such as conductivity and tensile strength. Different morphologies are desirable for different applications. For example, a porous film with a high surface area may be advantageous for applications such as photovoltaics, but a more compact, tough and highly conductive film may be required for actuator materials. Surface nanostructures or defects may also affect the properties of a film Surface morphology can be investigated by several techniques. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) have been employed in this study. Scanning electron microscopy (SEM) allows visual analysis of surface morphology. SEM operates by scanning the surface with a beam of electrons. Detection of the deflection of the electron beam allows the surface to be irnaged. SEM is a useful technique for obtaining a visual image of surfaces. However, the sample must be able to survive being placed under vacuum, and being exposed to a beam of electrons. The resolution of an SEM is limited by lens aberrations to 0. 1 8 nm and magnifications typically range between 2x and 50000x. Atomic force microscopy (AFM) uses a sharp tip to probe the surface, either by using a dragging motion (contact mode) or tapping motion (tapping mode). Van der Waal interactions between the sample and the tip, which is situated at the end of a cantilever, causes the cantilever to be deflected. Computer generated mapping of these deflections allows production of a 3-dirnensional image of the surface. The resolution of the technique is governed by the size of the probe tip, and is typically about 2 nm in the x ­ y axis and 1 nm in the z axis. The main advantage of AFM is that the surface morphology can be measured and information can be obtained on surface roughness. AFM is carried out in atmospheric conditions and is non-destructive. 44 2.6 Deposition Techn iques There are several methods to deposit materials on substrates. For device fabrication, a unifonn polymer layer is typically preferred. Three techniques which have been employed in this study are casting and spin-coating, (which require soluble materials) and electrochemical deposition (previously discussed). These techniques are effective in different applications because they fonn films which vary in smoothness, thickness and the state in which the polymer is deposited in (oxidised or reduced). The simplest method to deposit polymer onto a substrate is to cast it. This involves dissolving the polymer in a solvent, applying the solution to the substrate, and then evaporating the solvent to leave a film Application of the solution to the substrate can be achieved by drop-casting (dropping solution onto the substrate) or dip-coating (dipping the substrate into the solution). Thickness of the film can be varied by adjusting the concentration of the solution and by applying multiple layers. Evaporation can be achieved simply by exposure to air (which can be slowed by leaving the sample in a closed environment) or by subjecting the sample to a vacuum Whereas other film deposition techniques (such as electrodeposition and spin-coating) require substrates that are conductive or flat, films can be cast onto a wide range of substrates. Casting also allows a large thickness range to be obtained including films thick enough to be 'free-standing. ' However, it is often difficult to control the overall film thickness and homogeneity. In contrast, spin-coating is a widely used technique to produce very thin and unifonn films of soluble materials on flat substrates. It is employed to manufacture film; for a variety of applications such as compact disks and flat screen display antireflection coatings. Spin-coating uses simple fluid flow and evaporation processes to form a thin, unifonn layer. The substrate is first completely covered with a substantial excess of polymer solution compared to the amount of material required in the final coating. The substrate is then rapidly spun to remove excess solvent and material to leave a thin film. A major advantage of using spin-coating is that it allows the production of films of reproducible thicknesses. The :film thickness and morphology can be altered by varying parameters such as solvent, polymer concentration, applying multiple layers, 45 acceleration rate, spin-rate, time of spinning and solution viscosity. However, the major drawback is that it requires large amounts of material. It was also found in this study that some polymer solutions have a tendency for the polymer to aggregate as it is spun, causing defects. 46 2.7 E lectrospinn ing Electrospinning is used to form polymer fibres with a very small diameter (typically <500 run), which may be useful for applications such as batteries and solar ceUs where a very high surface area is advantageous. Electrospinning employs an electric field to create a charged jet of polymer solution, while evaporating solvent molecules to form very fine fibres. 1 29 A schematic of the experimental setup used in this study is shown in Fig. 2. 1 6. The solution to be electrospun is placed in a hypodermic syringe -30 cm from a galvanized steel target plate covered with aluminium foil. The positive electrode of a variable high voltage transformer is attached to the metal needle of the syringe, the negative terminal attached to the target plate, and a potential difference (-30 kV) applied across the electrodes. As the solution drips out of the syringe, the positively charged particles overcome surface tension and fly towards the negatively charged target plate.4 Electrostatic repulsion of the charged particles assists solvent evaporation to generate fine polymer fibres which coUect on the target plate. Polymer and Solvent in syringe � Target /' ---"..:-�-: ____ -_-_-_-_-__ :_-_-_-_-_-_-_-_-::_-_-_-::_-_-:�A - - - - - - - - - - - - - - - - - - - - - - - - - - - - � � Negative Terminal Fig. 2. 16. Experimental setup for electrospinning. The vis co elasticity and surface tension of the solution are known to significantly affect the formation of fibres. 1 30 In this study a solution of polymer in chloroform (typically -40 mg mL-1 ) was used, with addition of small amounts of polyethylene oxide (PED) to increase solution viscosity if required. 47 2.8 Photovoltaic devices An orgaruc photovoltaic ceU consists of a cathode and a polymer coated anode. Irradiation by light of a higher energy than the polymer bandgap results in the formation of an exciton (electron/positive-hole pair, Fig. 2. 1 7). Dissociation of the excitons to produce an excited electron and positive hole (polaron or bipolaron) is most efficient at interfaces between two dissimilar materials, one with a low ionisation potential and good hole-transport properties, and the other with a high electron affinity that can act as an electron-transport material. 1 3 1 . 132 In polymer-based photoelectrohemical (PEe) cells, the excited electron is passed from the polymer to the electrolyte, where the oxidised component of the electrolyte, e.g. tri-iodide ions, are reduced, e.g. to iodide ions. These effectively transport the electrons to the cathode, where they pass through an external circuit back to the anode. 1 32 The photocurrent produced by a cell can be controUed and measured by connecting the ceU in parallel with a potentiostat and scanning a range of potentials. A typical l - V curve is given in Fig. 2. 1 8 . E Net hole current .. Net electron current Conducting polymer Liquid electrolyte Interface - -- - -f Voc Oxidize _ Reduce 1 Mediator - - - - - - - - - - - - - - - - - - - _ . ITO Semiconducting polymer Pt - Fig. 2. 17. Schematic diagram of a typical polymer photoelectrochemical cell structure consisting of an ITO/polymer/electrolyte/Pt sand"vich structure. The interface between the polymer and liquid electrolyte are explicitly included to illustrate that the photovoltage is generated by the difference in standard electrode potential (work function) between the ITO-polymer interface and the polymer­ electrolyte interface. Irradiation of the semiconducting polymer results in the formation of an exciton, which is separated at an interface. Redrawn from reference [ 1 32] . 132 48 I"" CWTent / I o Voltage / V Fig. 2. 18. Typical I - V curve of a photovoltaic device. Is< is the short circuit clUTent and Voc is the open circuit voltage. Ipp is the CWTent at peak power and Vpp is the voltage at peak power. Data collected from the I - V curve is used to calculate parameters such as the fill factor (FF) and energy conversion efficiency (ECE) to determine the photovoltaic performance. These are defined below: The open circuit voltage, Voc, is the potential measured at zero current. The short circuit current, Isc is the current on the I - V curve at zero voltage. The fill factor (FF) is given by: FF = Voltage at peak power (Vpp) x ClUTent at peak power (lpp) Open circuit voltage (Vex::) x Short circuit ClUTent (Isc) The energy conversion efficiency (ECE) is given by: ECE (0/0) = Voc x Isc x 100 Total power of light radiating on the cell area 49 Chapter 3 Synthesis and Polymerisation of a Series of Styryl-Substituted Terthiophenes 3.1 I ntroduction The objective of the work presented in this chapter was to investigate the effect of electron-withdrawing and electron-donating substituents on the properties of polythiophene. Due to their simple attachment by straightforward chemical synthesis, substituted aryl functionalities have been attached to thiophene monomers (Fig. 3 . 1 a) . 54. 1 33 However, as several workers have demonstrated that the resulting phenylthiophene polymers display low conductivity due to twisting of the rings,54. 133 a styryl functionality was used in this study to link the substituent to the polythiophene backbone (Fig. 3 .1 b). The vinylene link er allows conjugation to be preserved while increasing the planarity of the polymer system Altho ugh po lymerisation of styryl­ substituted thiophene monomers has proven difficult due to susceptibility of the vinylene linker to polymerisation,88 copolymerisation o f styryl thiophene with bithiophene was achieved by Cutler. 25 The separation o f the vinyl substituents by bithiophene suggested the possibility of polymerisation of styryl terthiophene derivatives. R (a) (b) R Fig. 3. 1. Substituents linked by (a) phenyl and (b) styryl moieties to polythiophene. so In 2003, Collis et al. reported the synthesis of a series of terthiophene derivatives consisting of a terthiophene unit linked to a para-substituted aryl ring by an alkene moiety, as shown in Fig. 3.2 . 80 In order of increasing electron-withdrawing ability, the five substituents were: dimethylamino (NM�), methoxy (OMe), hydrogen (H), cyano (CN), and nitro (N02). The facile synthesis of these five styryl monomers presented the opportunity to investigate both their chemical and electrochemical polymerisation, and the properties of the resulting polymers. R 5" 5 R NM� (NM�STI) OMe (OMeSTT) H (STT) CN (CNSTT) N02 (N�STI) Fig. 3.2. Structures of the five styryl-substituted terthiophene monomer derivatives investigated in this study. Some initial electrochemical polymerisation studies on N02STT has been reported by Cutler et al. 9 1 They showed by post-growth electrochemical analysis and spectroscopic characterisation of the polymer films, that the nitro-styryl substituent has a strong influence on the oxidation potential of the polymer (744 mY), when compared to polyterthiophene (1 054 mY). This work confirmed the potential of styryl terthiophenes as precursors to aryl-functionalised polythiophenes. Therefore, the detailed chemical and electrochemical polymerisation of the five functionalised styryl terthiophenes shown in Fig. 3.2 was undertaken in this study. For convenience, in this study, these monorners will be referred to as NM�STT, OMeSTT, STT, CNSTT and N02STT. 5 1 3.2 Synthesis of monomers The monomers were prepared according to the method of Collis e t a l. 80 lbis procedure employs a Wittig reaction between the terthiophene aldehyde and the appropriate phosphonium salt as shown in Fig. 3.3. The yield obtained depended on the substituent and ranged from 68 to 90%, similar to those reported by Collis et al. CHO R DBUIDCMor KOBu1rrHF Reflux R R == N�, 70% yield R == eN, 68% yield R == H, 90% yield R = OMe, 79% yield R == NM�, 7 1 % yield Fig. 3.3. Wittig reaction between a terthiophene aldehyde and phenyl substituted phosphonium salt for the synthesis of the substituted styrylterthiophene monomers used in this study. 52 3.3 Polymerisation usi ng chemical oxidation 3.3. 1 Polymerisation and reduction methods The general reaction procedure used in this study for the polymerisation of styryl­ substituted terthiophenes is given in Fig. 3 .4 . Four equivalents of iron(ill) chloride were added dropwise as a suspension in dichloromethane (DCM) to a solution of mono mer in the same solvent. This is a typical process used to polymerise ftmctionalised thiophenes.44,63,68,70,98, 1 34, 135 The use of iron(III) chloride as an oxidant provides a simple polymerisation method to produce large amounts of high molecular weight polymer. 70 It was suggested by Laakso et al. that the additional equivalents of iron(II1) chloride are required as this oxidant is only partially soluble in chloroform, forming an inert solution. 1 36 Laakso et al. also explained the requirement for extra iron(III) chloride through consumption o f this oxidant in the formation o f HCI gas. 1 36 The slow addition of the oxidant as a slurry was intended to keep the Fe3+/Fe2+ ratio (and hence oxidation potential) low and more constant throughout the polymerisation procedure to reduce the occurrence of overoxidation. 70 Several authors have reported that the slow addition of iron chloride as a slurry rather than the rapid addition of solid iron chloride results in a higher yield of soluble polymer. 7o,98 Gallazzi et al. 98 found that adding iron(III) chloride in a single addition to 3,3"-didodecyl-2,2 ' : 5 '5"-terthiophene resulted in a polymeric yield that was 60% soluble, compared to pro duction of polymeric material, which was 1 00% soluble, from dropwise addition. It was suggested that this may be related to a sudden accumulation o f HCI gas in the reaction mixture, which is produced on polymerisation.98 R Anhyd. FeCI3 (4 equiv. ) • DCM RT, 4-24 hrs R m+ mcr Fig. 3.4. Reaction procedure employed for the chemical polymerisation of styryl-substituted terthiophene derivatives. S3 After addition of the oxidant, the reaction mixture was stirred for several hours to allow polymerisation. As black congealed precipitates were immediately produced, the mixtures were sonicated at intervals to redisperse the material and allow further reaction to o ccur. The reactions were stopped by quenching with water when all monomer had reacted as determined by TLC . The crude products were black, insoluble materials, which were assumed to be the o ligomeric and/or polymeric material in the oxidised state. This observation is similar to work done by GaIlazzi et al. 62 and Wang et ai. ,44 on the polymerisation o f alkyl- and alkoxy-substituted terthiophenes by iron(III) chloride. The oxidised polymers were reduced (de-doped) to produce the neutral state (Fig. 3 . 5), which is generally more so luble in organic solvents than the oxidised form, and more amenable to characterisation. 62,74 The crude oxidised material was first rinsed with water, and then thoroughly washed with methano l using soxhlet extraction to remove residual oxidant and monomer. A number of s ubstituted p o lythiophenes appear to readily reduce to the neutral state when the residual oxidant is removed by washing the crude material with water or methanol,44,62,98 although GaIlazzi has found that a strong reducing agent, such as hydrazine, is required fo r the reduction of poly(4,4"­ dipentoxyterthiophene ) . R m+ R f , Water or Hydrazine > , , FeCl3 or mcr Cu(ClO4)2 Oxidised Neutral Fig. 3.5. The reversible doping (by iron(lII) chloride or copper perchlorate) and dedoping (by washing "vith water or hydrazine) process of the polymer. Washing the oxidised polymer with methanol appeared to reduce the material to the neutral state as observed by a change in colour from black to red. S o xhlet extraction of this neutral p o lymeric material with DCM gave an orangelred powder that is soluble in DCM and chloroform (soluble fraction). A dark red powder remained that was not 54 soluble in any common solvent (insoluble fraction). The percentage yields of the soluble fractions obtained for each derivative are given in Table 3. 1 . The mass of the monomer used in the reaction has been used as an approximation of the expected polymer yield since the length of the oligomers can not be accurately measured. At most, this leads to an uncertainty of 0.3% in the calculated polymer yields (due to not taking into account the loss of a-hydrogen on polymerisation). Table 3. 1. Soluble fractions of styryl-substituted terthiophene polymer derivatives. Polymerised monomer Total soluble yield (%) <5% 63% SIT 3 3% OMeSTT NM�SIT 20% 20% The remammg insoluble material is unlikely to be oxidised oligomer/polymer as attempts to reduce this material with hydrazine did not produce further soluble material. The insoluble component was found to be difficult to characterise and could not be easily further processed, thus further discussion focuses solely on the soluble fraction. 3.3.2 Characterisation of soluble fractions The soluble fractions were investigated by mass spectrometry and thin layer chromatography (TLC). MALDI-TOF MS (Matrix-Assisted Laser Desorptionllonisation Time-Of-Flight Mass Spectrometry), described in Chapter 2, is a frequently used method for investigating the chain length of polymer samples. I 1 8. 1 37. 138 MALDI-TOF MS of the neutral, soluble fractions of NMe2STT, OMeSTT, STT, CNSTT and N02STT oligomers are displayed in Fig. 3 .6. Dimer is found to be the primary species present in all the samples, with small amounts of trimer and tetramer also detected in NM�STT, STT and CNSTT samples. The significant signal at 354 Da observed in the spectrum of N02STT may be due to decomposition of the monomer (Mr = 395.4 g mOrl) or a reaction with the matrix (U = 226.2 g mOrl). Chromatography was used to investigate the chain length of oligomeric materials. The soluble styryl-substituted terthiophene materials were investigated by thin layer 55 chromatography (TLC) using silica plates. A significant spot with a polarity less than that of the monomer was observed for all soluble fractions, which may be due to dimer (Table 3.2). Additional spots were observed at even lower retention factor values (Rr) for the NM�STT oligomer fraction (one spot) and the STT oligomer fraction (two spots), and these may be due to the presence of longer oligomers, consistent with results from MALDI-TOF mass spectrum The OMeSTT soluble oligomer fraction also displayed a spot at an Rr value lower than the assumed dirner, which may be due to a regioisomer of the dimer, or an impurity. The CNSTT and N02STT oligomer fractions displayed only one spot, consistent with low amounts of oligomeric impurities as indicated by MALDI-TOF mass spectra. Table 3.2. Revalues ofmonomers and dimers in 20% ethylacetate/hexane on silica TLC plates. Monomer Re Dimer Re NM�STT 0.60 0. 38 OMeSTT 0.68 0.44 STT 0.85 0.7 CNSTT 0.54 0 .21 N(hSTT 0.6 1 0.33 56 I. I 1 J 0 500 2 2 2 2 2 .... 3 I 3 1000 3 I NM�STT Mr = 393.6 g mol-1 4 J OMeSTT Mr = 380.5 g mor1 STT Mr = 350.5 g mor1 4 l CNSTT Mr = 375 .5 g mor1 N02STT Mr = 395.5 g mor1 1500 2500 mlz Fig. 3.6. MALDI-TOF MS of crude, soluble fractions of chemically polymerised NM�STT, OMeSTT, STT, CNSTT and N�STT. Significant signals are labelled with oligomer length in terms of monomer units (n). Detection suppression limit: 320 Da. 57 3.3.3 Attempted purification of dimer The results for MALDI-TOF MS indicated that dimers were the major components of the soluble fractions. This is consistent with work by Grant et a l . who, during the course of this study, reported that attempts at polymerising polyether-substituted styrylterthiophenes, by both iron(III) chloride and electrochemical techniques, resulted in high yields (55-87%) of soluble dimer. 1 1 Attempts were made to isolate and purify the dimers produced in this study. Impurities of trimer and longer oligomers were removed by recrystallisation of the crude soluble materials to provide materials that contained no detectable oligomer impurities as indicated by MALDI-TOF MS and TLC. The purified dimers were reasonably soluble in organic solvents and this provided the opportunity to characterise them by IH NMR spectroscopy. The simple spectra that were obtained by the dimer samples indicate that the samples consist predominantly of a single regioisomer. The mono mer and dimer spectra of the dimethylarnino derivative are given in Fig. 3 . 7 as an example, and show the loss of the signal due to H I , and consequential simplification of the H2 quartet to a doublet on dimerisation indicating head-to-head (llli) coupling. Grant et al. 1 39 also reported that dimer formation by ether-substituted styryl terthiophenes was regioselective, giving predominantly HH isomers as shown by IH NMR spectroscopy. Spectra of the diether-substituted styrylterthiophene monomer and dimer (Fig. 3 . 8) clearly shows loss of the HI proton. 58 1 0 ,......., 4 1 ,......., ..::' U ::r: U 6 1 1 ,........,..., 1 1 3 2 -'-r-rTI 'I,'rr-TT' 1"-'--'" TI"rr-r-rI-,-"rr-1 'I�-I , -,---,, , TI 'I,,-,--rr-r--,-rl -rl ,I T, -'-"-'--1 '-'-"'---'--1 01 ,-, I I I 1 I ppm 7.6 7.5 7.4 7.3 7.2 7.1 7.0 4 1 0 3 ,...., 8 7 ,........,..., 6.9 6.8 \ 8 6.7 6.6 8� 9 � j 10 1 1 ,Irr-I T'"I "T", -, ,-,r-r-, J l----r I I " T, --r,-,r-,-,-""rr-I 01 ,-" I r T"ITI ' i i i I ! i i j I I I i I i i i i I i 7.6 7 .5 7.4 7.3 7.2 ppm 7.1 7.0 6.9 6.8 6.7 6.6 4 6. 5 i i i i 6.5 Fig. 3.7. IH NMR spectra (aromatic region) ofNMe2STT and the dimeric HH regioisomer produced by chemical oxidation. 5 9 Monomer 4 Head-a 1 ,-A-.. 8 Tail-a ,,-----'---. 2 ,-A-.. 10, 12 � r-A:-. 9 � 1 1 ,-A-.. I 7 . 4 I 7 . 3 I 6 . 9 ppm Dimer IllI regioisomer I 7 . 4 Fig. 3.8. 1 H NMR spectra (aromatic region) of styryl-1 5-crown-5 terthiophene, and the dimeric HH regioisomer that is produced by oxidation. 1 1 Reproduced by permission. ppm 60 The high yield of regio specific dimer produced by these styryl terthiophene derivatives was subsequently supported by theoretical calculations performed by Clarke et al. 140 on the reactivity of the terminal '0.' positions of styryl-substituted terthiophene and sexithiophene. With styryl substitution, the reactivity of these positions appears to decrease as the oligomer length increases, accounting for the low degree of polymerisation and high yield of dimer. It was suggested by Clarke et al. 1 40 that this was due to a ' trapping' of the charge required for further polymerisation by the styryl substituent as shown in Fig. 3 .9. Clarke et al. 140 was also able to account for the regiospecificity and tendency of styryl-substituted terthiophene monomers to produce HH regioisomers by performing theoretical calculations on the SIT monomer to show that the 5-position is more reactive than the 5" position. R R Fig. 3.9. 'Charge-trapping' over the two central thiophene rings and styryl substituents as exhibited by theoretical ca1culations. 1 40 Although the IH NMR spectra confirm results by MALDI-TOF MS and TLC that oligomer impurities have been removed from the dimer samples in this study, additional peaks generated by the dimer samples (the set of low field peaks between 7.46 and 7.70 ppm in the case of NM�SIT, Fig. 3 . 7) indicates that impurities are still present. Efforts to further purify the materials, including additional recrystallisations, columns 6 1 on silica and alumina, radial chromatography and soxhlet extractions, were not successful. 3.3.4 Conclusions Chemical oxidation of the styryl-substituted terthiophenes investigated in this study resulted in the production of significant amounts of insoluble material (>36%), which were difficult to characterise. Investigation of the soluble fractions revealed that they consisted primarily of dimer and short chain (n < 4) oligomers. The dimer could be further isolated, but not purified, and was shovvTI by 1H NMR spectroscopy to be regioregular. This regioregularity is consistent with work reported by Grant et aI. , 1 39 and predictions made by Clarke et al. 140 62 3.4 Polymerisation using electrochemical methods 3.4. 1 Introduction Although characterisation of the insoluble fractions of chemically oxidised NMe2STT, OMeSTT, STT, CNSTT and N02STT proved difficult, the soluble yields revealed the production o f mostly short o ligo mers , with chain lengths comprising four monomer units or less. In an attempt to produce longer oligomers, the electrochemical polymerisation of these materials was investigated. Electrochemical polymerisation is a commonly employed method to grow films of conducting polymers8,83 and, unlike chemical polymerisation, allows control over the oxidation potential. Electrochernical polymerisation also allows the straightforward growth of insoluble polymer films, which are amenable to characterisation by cyclic voltarrnnetry, spectroscopy and microscopy. Both cyclic voltammetry and potentiostatic methods have been used in this study to grow oligomeric/polymeric films onto platinum and ITO-coated surfaces. 3.4.2 Electrochemical growth To determine appropriate scan limits and oxidation potentials, each monomer was initially analysed by scanning between potentials of -1 V and + 1 . 5 V. The CV scan of terthiophene (Fig. 3. 1 0) shows several oxidation peaks as is typically observed for short oligothiophenes. 1 41 , 142 The same mechanism is reported to o ccur for the electrochemical oxidation of thiophene as for thiophene oligomers,8, 143 and is shown in Fig. 3. 1 1 using thiophene for simplicity. From work reported by Audebert et al. 141 and Roncali et al. 144 on the electrochemical oxidation of terthiophene, the initial anodic peak, Eox, J , is assigned to the oxidation of the terthiophene monomer to its radical cation (Fig. 3 . 1 1 a). As described by Waltman et aI. , 72 Audebert et aI., 141 and Roncali,8 the assignment of further anodic peaks depends upon the reactivity of the terthiophene mono mer cation. Audebert et al. has conclusively shown that terthiophene mono mer cations, which are not very reactive, may further oxidise to generate multiply charged cations (Fig. 3 . 1 1 b). 141 On the other hand, very reactive cation species may undergo oxidative degradation reactions with either solvent molecules, electrolyte or impurities (Fig. 3 . l1 C). 72, 145 Between these two stability ranges, which vary depending on 63 substituents and chain length of the monomer, the radical cations can dimerise by coupling (Fig. 3 . 1 1 d). 72, 141 , 143 These resulting oligomers may then oxidise to form cation species (Fig. 3 . 1 1 e). Further oxidation of these species at higher oxidation potentials may lead to the formation of multiply charged species, polymer or overoxidation (Fig. 3. 1 I f). 8, 145 No detailed study has been reported on the characterisation and assignment of anodic peaks produced during the growth of terthiophene. However, in a report by XU et al. 143 on the electropolymerisation of tetrathiophene, after the initial anodic peak due to the oxidation of the monomer to its radical cation, this species was observed to immediately couple to form octathiophene, which deposited on the electrode. The second ano dic peak was assigned to oxidation of octathiophene. A similar process is likely to occur in terthiophenes. Audebert et al. reports that Eox, I is non-reversible even at very fast scan rates (up to 1 5 000 V S-I ) demonstrating that terthiophene cation radicals are highly reactive, I 41 and suggesting they are likely to polymerise and/or overoxidise. As Cutler et al. reported the overoxidation of N02STT at potentials above l . I V,91 the third oxidation peak, Eox.,3, is most likely due to overoxidation of the terthiophene monomer or oligomer cations. Little can be said about the cathodic peaks in these scans as they may be due to the reduction of cation radicals and/or overoxidised species. c:( ::1. - .. c � :::J 0 1��----------------------------------------------, Eox,3 1 00 � 0 -�+---------�-------'---------'---------r--------� - 1 .0 -0.5 0.0 0.5 Potential I V vs. Agll4fl 1 .0 1 .5 Fig. 3. 10. CV of terthiophene monomer on a platinum disc electrode (SA = 1 .8 mm\ Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: - 10001+ 1 500 mV. Scan rate: 100 mV S-I . 64 (cl) )-211' (e) G-D s s - e - ... + e - Oxidative degradation (t) - e .. MUltiply charged species? Polymerisation? Overoxidation? Fig. 3. 1 1. A possible mechanism to explain redox peaks (using thiophene for clarity) observed in the first scan of terthiophene derivatives. The CV scans of CNSTT, SIT, OMeSTT and NMe2STT are displayed in Fig. 3 . 1 2. As for terthiophene, three anodic peaks are observed for each species (listed in Table 3 . 3). The first peak is assigned to the oxidation of the monomer to the radical cation, and the second to the oxidation of longer oligomers as, for example, in process (t) in Fig. 3 . 1 1 . Scanning to potentials over the third peak leads to deposition of a black precipitate, most likely due to overoxidised species. Table 3.3. Anodic peaks shown by styryl-substituted terthiophene derivatives. Monomer 1 st anodic peak 2nd anodic peak 3rd anodic peak CNSTT 0.75 V 0.81 V l .36 V STT OMeSIT 0 .72 V 0.66 V 0.75 V 1 .06 V 1 . 34 V l .30 V NM�STT 0.28 V 0. 98 V l . 30 V 65 � - - C ! ... ::s 0 300 -Cyano -- Phenyl 200 - - - - - tv'ethoxy . . . . . . . . . Dimethylamino 1 00 : 0 -.=.,-.,;.-,.;:.: :-:.- - - - - - -100 +------.-------r-----,---------,,--------i -1 -0.5 o 0.5 Potential I V VS. Ag/Ag + 1 .5 Fig. 3. 1 2. CV scans of CNSTT, STT, OMeSTT and NM�STT on platinum disc electrodes (SA = 1 . 8 mm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: - 10001+ 1 500 mY. Scan rate: 100 mV S·l In contrast, a CV ofN02STI reveals multiple oxidation peaks (Fig. 3 . 1 3). As nitro is the most electron-withdrawing substituent of the series, this may be due to a low stability of the corresponding cations.72 Instead of polymerising, the cations may be undergoing oxidative degradation by reacting with nucleophiles such as water impurities. 72, 145 This was observed by Waltman et al. during the attempted electrochemical polymerisation of 3-cyano- and 3-nitrothiophene. 72 250 200 � 150 - 100 - C e 50 ... ::s 0 0 -50 -100 -1 -0.5 o 0.5 PotentialN VS Ag/Ag+ 1 .5 Fig. 3.13 CV of N02STT monomer on a platinum disc electrode (SA = 1 .8 mm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: - 1 0001+ 1 500 mY. Scan rate: 1 00 mV S·l 66 The mono mer oxidation onset potentials of the terthiophene derivatives were determined from these scans, and are listed in Table 3.4. Oxidation of terthiophene commences at 0.7 1 V. The higher potential when compared to that for STT (0. 62 V) suggests that the styryl group is activating the monomer. The increasing oxidation potential with increasing electron withdrawing ability of the substituents, except for the nitro derivative, reflects the relative stability of the cation radicals. 54 Thiophene cation radicals that support electron-donating substituents have been suggested to be stable and therefore readily formed. On the other hand, cation radicals containing electron-withdrawing groups are less stable and therefore harder to form. Similar results have been reported by Michalitsch et al. , 109 MacDiarmid et al. l46, Collis et al.80 and Cutlers for thiophene derivatives. This suggests that cations containing EWGs are more reactive and may form longer polymer chains, whereas cations of short-chain oligomers containing EDGs may not be sufficiently reactive to promote further polymerisation. The nitro derivative does not follow the same relationship, most likely due to facile reactions ofthe highly reactive cation radical with electrolyte anions and/or impurities. Table 3.4. Oxidation onset potentials of styryl terthiophene derivatives (vs Ag/ Ag +). Monomer Terthiophen NM�STT OMeSTT STT CNSTT N02STT e Oxidation onset potential 0. 7 1 V 0 . 1 5 0 . 5 2 0.60 0 . 6 2 0 . 2 1 Growth CVs of NMe2STT, OMeSTT, STT, CNSTT and N02STT on platinum disc electrodes are shown in Figs. 3. 14 to 3. 1 8 respectively. Anodic scan limits were set at a potential about 100 mV beyond the potential at which material was observed to deposit. The CVs show an increase in current with successive scans, characteristic of conducting polymer growth as discussed in Chapter 2, Section 2. 1 .3 . Except for the NM�STT sample, a decrease in oxidation onset potential is observed after the first scan. This is most likely due to oxidation of deposited oligomers, which are expected to have a lower oxidation potential than the monomer. Long oligothiophenes were shO\vn by Sumi et al. to have a lower oxidation potential than short oligomers. 1 14 The 67 NM�STI sample is distinctive, however, in that the oxidation potential increases with scan number. This may be due to a number of reasons such as ( 1 ) rapid depletion of the mono mer in the vicinity of the electrode,61 (2) very low conductivity o f the film in neutral state, effectively insulating the electrode from the monomer, 142 or (3) hindrance of the interchange of anionic species due to growth of a very compact film. 72, 142 Multiple oxidation and/or reduction peaks are observed during the growth of CNSTT and N02STT. Although multiple peaks are commonly observed during the growth and post-growth cycling of polythiophene derivatives,23,99, 147 their origin is not well understood. It has been suggested by several authors that the peaks are due to transitions between the neutral, polaron, bipolaron and metallic states of the polymer. 148.1 50 As discussed by Pringle et al. 1 47 the potential at which these transitions occur may be influenced by factors such as reduction of different areas of the polymer fIlm147, different length polymer chains, 1 1 4 the effect of 'charge-trapping', 1 5 1 and conformational changes of the polymer accompanying radical cation formation. 1 52 The growth CV ofN02STT is consistent with work reported by Cutler et al?5 250 200 150 � - 100 1: � 50 :::J 0 0 -50 -100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Potential I V VS AgI Ag + 0.7 0.8 0.9 Fig. 3. 14. Potentiodynamic growth of NM�STT on a platimun disc electrode (SA = 1 . 8 mm2). Supporting electrolyte: 0. 1 M TBAPI AN. Potential limits : 01+900 m V. 9 cycles. Scan rate: 100 mV S·l 68 3OO �----------------------------------------------------� 200 � 1 oo C Scan number � d O � -1 oo� -2oo +---------�----------.---------�-----------r--------� 0.3 0.4 0.5 0.6 Potential I V VS Ag/Ag+ 0.7 0.8 Fig. 3. 15. Potentiodynamic grO'vvth of OMeSTT on a platinum disc electrode (SA == 1 .8 mm2). Supporting electrolyte: 0. 1 M TBAP IAN. Potential limits: +3001+750 mY. 9 cycles. Scan rate: 1 00 mY S-l 400 300 200 � - 1 00 C � 0 :::J 0 -100 -200 -300 0.3 0.4 Scan number 0.5 0.6 0.7 Potential I V VS Ag/Ag+ 0.8 0.9 Fig. 3. 16. Potentiodynamic growth of SIT on a platinum disc electrode (SA == 1 .8 mm2). Supporting electrolyte: 0. 1 M TBAP IAN. Potential limits: +375/+900 mY. 9 cycles. Scan rate: 100 mY S- l 69 ��---------------------------------------------------, 400 -200 �+-----�----�----�----�----�----r-----r-----r---� 0.0 0.1 0.2 Q3 Q4 o� Q6 Pote ntial I V vs Ag/Ag+ 0.7 0.8 0.9 Fig. 3. 17. Potentiodynamic growth of CNSTT on a platinum disc electrode (SA = 1 . 8 mm\ Supporting electrolyte: 0. 1 M TBAP/ AN. Potential limits: 0/+900 m V. 9 cycles. Scan rate: 100 mY S·l . 160 120 �80 - C 40 � ::J 0 0 -40 -80 0.3 0.4 0.5 0.6 0.7 Potential I V VS Ag/ Ag + 0.8 0.9 Fig. 3. 18. Potentiodynamic growth of N02STT on a platinum disc electrode (SA = 1 .8 mm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: 0/+900 mY. 9 cycles. Scan rate: 100 mY S·l As conducting polymer films are sometimes observed to be different depending on their method of growth, 6 1 films of styryl terthiophene derivatives were grown using potentiostatic methods for comparison to the potentiodynamically grown films. The 70 chronoamperograms produced by these materials were observed to be similar. A representative chronoamperogram using that of STT is displayed in Fig. 3 . 19, and reveals characteristics typical of conducting polymer growth. 25,47 After the initially high but rapidly tailing current transient (t < 30 seconds), the current increases steadily indicating an increase in electroactive surface area (by polymer deposition) and/or more facile electrode kinetics. 100,--------------------------------------------------, 80 � 60 C � ::::I 40 o 20 O+-L------r------�--------r_------,_------�------� o 50 100 150 Time I s 200 250 300 Fig. 3. 19. Potentiostatic growth of STT on a platimnn disc electrode (SA = 1 . 8 mm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential held at 0 mY for 1 0 s, then stepped to 900 mY for 290 s. Both the potentiostatically and potentiodynarnicaUy deposited oligomeric/polymeric films of these styryl terthiophenes were compared using cyclic voltarrnnetry, UV -VIS­ NIR spectroscopy and scanning electron microscopy. 3.4.3 Characterisation of electrochemically deposited materials 3.4.3.1 Electrochemical characterisation The oligomeric/polymeric films of NMe2STT, OMeSTT, STT, CNSTT and N02STT grown using potentiodynarnic and potentiostatic methods were compared using cyclic voltarrnnetry. All CVs are completed in monomer-free electrolyte solution. 7 1 The post-growth cycles of polyNMe2STT deposited potentiodynamically and potentiostatically, shown in Fig. 3.20, have different peak potentials, suggesting differences in the films . In both CVs, the peak currents are observed to decrease with increasing cycle number indicating that the films are unstable. In addition, the first scan of the potentiodynamically deposited material shows a current response on the initial anodic scan which is 2 .7 times larger than the current response on the cathodic scan. This suggests a modification of the film and may be a result of one of the foUowing processes. First, short oligomers in the film may further polymerise. Eales et al. 145 also observed a difference between the first scan and subsequent scans with poly(terthiophene) films and suggested that the similarity of monomer and polymer oxidation potentials allows deposition of incompletely polymerised species. These incompletely polymerised species (which may even include monomer cations) are thought to become trapped in the film during growth and subsequently polymerised during the initial anodic post­ polymerisation scan causing the initial increase in current. A second possibility is that the film may be undergoing degradation. Loss in electroactivity of thiophene films is often caused by nucleophilic attack of overoxidised cation radicals by nucleophiles such as water contarninants. 72, 145 This is unlikely to o ccur during polymerisation of NMe2STT however since cations with electron­ donating substituents are likely to be relatively stable. A more likely reaction is the protonation of the dimethylarnino substituent by protons pro duced during growth. Proton production is required by the accepted polymerisation mechanism and has been proved experimentally. 1 53 A third prospect is that the morphology of the deposited material may be changing or compacting during the potential scans (i. e. an irreversible electroactuator effect), causing a reduction in the electroactive surface area. 72 200 .----------------------------------------------------, (a) 1 SO -so � Scan 1 'V Scans 2-5 � Scan 10 � Scan 20 -1 oo+-----�----�----�----r_----r_--�r_--_,----_,----� 0.0 0 .1 0 .2 120 100 80 �60 - -i 40 t: :::J 20 0 0 -20 -40 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Potential I V VS Ag/ Ag + Q3 OA 0.5 O� Potential I V VS Ag/Ag 0.7 0.7 0.8 0.9 (b) 0.8 0.9 Fig. 3.20. Post-polymerisation CV of NM�STT fihns polymerised (a) potentiodynamically and (b) potentiostatically. Electrode: platinum disc ( 1 . 8 mm2). Electrolyte solution: 0 . 1 M TBAP/AN. Potential limits: 01+900 m V. 1 5 cycles. Scan rate: 100 m V S·l 73 I I I I Post-growth CVs for potentiodynamically and potentiostatically deposited polyOMeSTT films are shown in Fig. 3 .2l. The post growth CV of the potentiodynamically deposited material is a simple scan, which shows one broad oxidation peak (0. 70 V) and one broad reduction band (0.48 V). The first scan is again distinct from subsequent scans and has a higher oxidation current. Subsequent scans are almost identical and indicate stable, reversible oxidation/reduction properties of the film. The post-growth CV of the potentiostatically deposited material is very different to that shown by potentiodynamicaUy deposited material and displays a complicated voltarnmogram with a number of distinct features. The first scan shows two sharp oxidation peaks at 0. 32 V and 0.65 V, and two reduction peaks at 0.43 V and 0.32 V. These may be due to the oxidation of short oligomers as the first oxidation peak (0. 32 V) completely disappears on the second scan and a large sharp oxidation peak at 0 .50 V emerges. This peak and the oxidation peak at 0 .65 V are observed to reduce in intensity with increasing scan number to eventually produce a broad band. Interestingly, the reduction peak at 0.42 V remains almost the same in shape and current intensity, but the peak at 0. 15 V disappears. The additional peaks observed by the potentiostaticaUy grown films compared to the potentiodynamically grown film may be due to the different scan limits, and/or the presence of monornerlshort oligomers trapped in the potentiostatic film. Multiple oxidation and reduction peaks were reported by Zotti et al. during initial post grow1h cycling of poly(3,3"-dipentoxyterthiophene), and were explained by further polymerisation of deposited oligorners.99 74 �.----------------------------------------------------, 1st scan (a) 1 50 1 00 � - 50 1: � 0 :::J o -50 -1 00 _ ----....... It -1 5O +----------.---------.----------,---------,---------� 0.3 0.4 0.5 0.6 Potential ! V vs Ag/Ag+ 0.7 0.8 1 50 �--------------------------------------------------� 1 00 � 50 - 1: � :::J o o -50 -1 00 +-----,-----,-----,-----,-----,-----,-----,-----,-----; 0.0 0.1 0.2 0.3 0.4 0.5 O.� Potential ! V vs Ag!Ag 0.7 0.8 0.9 (b) Fig. 3.21 . Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoOMeSIT films. Electrode: platinum disc (1 . 8 mm2). Electrolyte solution: 0. 1 M TBAP/AN. Scan rate: 1 00 mV S·l 75 The post-growth CVs of potentiodynarnically and potentiostatically grown STT material are shown in Fig. 3 .22, and differ significantly. As for the OMeSTT films, the potentiostatic film displays a number of sharp peaks, which may be due to oxidation of trapped monomer/oligomers. The difference in the scans could also be due to the different scan limits. An additional reduction peak at 0.27 V is included during the post growth cycling of the potentiostatically deposited material, which may be partially irreversible. 300 ._-------------------------------------------------, 200 � 1 00 - 1: 0 � ::s o -1 00 -200 �OO +-----r_--_,----_.----_r----._----._----._--_,----� 0 .3 0.4 0.5 0.6 0.7 0.8 0.9 Potential ! V vs Ag/Ag+ 1 .0 1 . 1 1 .2 oo._------------------------------------------------� 40 -40 ��----._----._--_,----_.----_.----_r----�----�--� 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Potential ! V vs Ag! Ag + 0.7 0.8 0.9 (a) (b) Fig. 3.22. Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoSTI films. Electrode: platinum disc (1 .8 mm2). Electrolyte solution: 0. 1 M TBAP/AN. Potential limits: +3751+ 1 200 m V. 5 cycles. Scan rate: lOO m V S·l 76 The post-polymerisation CVs for potentiodynamically and potentiostatically grown CNSTT films are shown in Fig. 3 .23. As previously observed, the first cycle of the potentiodynarnically grown material is noticeably different to subsequent scans. The potentiostatically deposited material shows a large decrease in current in the first two scans, which, as mentioned for polymerisation of NMe2STT, is most likely due to either a change in morphology or further polymerisation of short oligomers. 600 400 � 200 - - c e '- 0 :::J 0 -200 -400 0.0 0.1 250 200 � 1 50 - 100 -C CD t: 50 :::J 0 0 -50 -100 -1 50 0 0. 1 0 .2 0.2 0.3 0.4 0.5 o.� Potential I V VS Al;J/AI;J 0.3 0.4 0.5 Potential I V VS Al;J/AI;J+ (a) 1 st scan 0.7 0.8 0.9 0.6 0.7 0.8 Fig. 3.23. Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoCNSTT fllms. Electrode: platinwn disc ( 1 . 8 mm2). Electrolyte solution : 0 . 1 M TBAP/AN. 5 cycles. Scan rate: 1 00 mV S·l 7 7 Post growth CVs of N02STT produced using potentiodynamic and potentiostatic methods are shown in Fig. 3.24. The post-growth CVs displayed by N02STT deposited by potentiodynamic and methods are very similar, indicating similar films. A sharp peak is observed on the first scan at 0.56 V, which decreases in current and shifts to 0.49 V on subsequent scans. Further cycling reveals a material that is stable with respect to current produced between the potentials of 0 and 0.9 V, and produces broad oxidation and reduction bands. These CV s are very similar to those measured by Cutler et aC' 1 20 .--------------------------------------------------. 80 � 40 - -c � a 0 -40 �O+-----�----�--�----�-----r----�----�----�--� 0.0 0.1 0.2 0.3 0.4 0.5 O.� Potential ! V vs Ag! Ag 0.7 0.8 0.9 1 � ._------------------------------------------------_. 80 -40 �+_----�---,,---�----_,----_r----_r----,_----._--� 0.0 0.1 0.2 0.3 0.4 0.5 O.E? Potential ! V vs Ag/ Ag 0.7 0.8 0.9 (a) (b) Fig. 3.24 Post-polymerisation CVs of (a) potentiodynamically and (b) potentiostatically deposited oligoN�STT films. Electrode: platinum disc ( 1 . 8 mm2). Electrolyte solution: 0. 1 M TBAP/AN. 5 cycles. Scan rate: 1 00 m V S·l . 78 In summary, the post-polymerisation CVs of the deposited materials showed significant changes between the initial scan and subsequent scans. Sharp peaks which are observed in the initial scan may be due to further polymerisation of trapped monomer and/or short oligomers. These are replaced by a broad oxidation band in subsequent cycles, which is consistent with the oxidation of a wider range of oligomer lengths. Although post-growth CVs of potentiodynamically and potentiostatically deposited N02STT and NMe2STT materials were similar in terms of shape and redox potentiais, the OMeSTT, STT and CNSTT showed very different, more complicated post-growth spectra for materials deposited using potentiostatic methods. This may suggest that potentiostatically deposited materials are less stable than potentiodynamically deposited materials and undergo further processes (morphological or chemical) during post-growth potential cycling. Cyclic voltammetry appears to be a better method for the production of stable oligo(styryl terthiophene) films. 3.4.3.2 Analysis by UV-VJS-NIR spectro sco py The films prepared in this study were observed to be electrochromic, since they changed colour with a change in potential. Red-brown colours were observed for the neutral films of the styrylterthiophene derivatives, and green/blue or purple colours generally observed for the oxidised films. In order to measure the spectroscopic properties of the materials, films were electrosynthesised on ITO-coated glass by both potentio dynamic and potentiostatic methods. UV -VIS-NIR spectra were obtained for each film in both the oxidised (doped) and neutral (dedoped) state. UV -VIS-NIR spectra for films produced from N02STT, CNSTT, STT, OMeSTT and NM�STT derivatives are shown in Figs. 3 .26 to 3 .30 respectively. Films of N02STT, CNSIT, SIT, OMeSTT derivatives appeared to be stable with respect to colour in both the oxidised and neutral states. In contrast, films of the NM�STT derivative were purple in colour when subjected to an oxidised potential, 79 but were observed to spontaneously reduce (perceived by a change to an orangelbrown colour) when the applied potential was removed. Spectra of the N02STT, CNSTT and STT terthiophene films revealed no variation between potentiodynamic and potentiostatic deposition, although small differences were found in the UV -VIS-NIR spectra of OMeSTT and NM�STT films. The spectrum of the N02STT film is comparable with results obtained by Cutler et al. 9 1 who also studied electrochernically deposited films ofthis polymer. Neutral films of CNSTT, STT and OMeSTT oligomers are observed to generate two absorption maximum (Amax) below 550 run. The oligoN02STT film also shows a Amax at 4 1 1 nm, with a shoulder at 496 nm It is possible that the NMe2STT film produces a second peak as well, but it is obscured by the peak at 40 1 nm. It has been postulated by Wagner and Officer, 1 54 that the two absorptions observed in the spectra for styryl terthiophene monomers can be attributed to a styryl-substituted thiophene chromophore and a terthiophene chromophore (Fig. 3. 25). It is anticipated that analogous bands would be observed for styryl-functionalised oligo(terthiophene)s, and indeed Grant confirms this for polyether-substituted styryl-sexithiophenes; the styryl chromophore is observed at ca. 340 nm and the sexithiophene chromophore at ca. 500 nm l I On oxidation these bands are observed to diminish while bands at about 800 nm and > 1 000 nm emerge due to the formation of polaronic and/or dipolaronic species. 1 25, 1 27 80 Terthiophene chromophore Styryl-substituted thiophene chromophore Fig. 3.25. Chromophores produced by styryl-substituted terthiophene derivatives. 154 The absorption maxima measured by UV -V I S -NIR spectroscopy, provide an indication of the band gap of the polymer/3 and hence the effective conjugation length. 1 1 4 In the reduced state, the Amax due to the oligothiophene chromophore for the materials in this study are observed to be similar (ranging between 484 and 5 1 8 run), and are comparable to the Amax reported by Grant for a series of styryl-substituted sexithiophene derivatives (ca. 500 nrn), 1 1 providing further evidence that these materials are producing predominantly dimer. When oxidised, the effective conjugation length is increased due to delocalisation of charge. Polaron and bipolaron energy levels are produced, which decrease the band gap between the valence and conduction bands; hence the appearance of bands at higher wavelengths (lower energy). The decline in the n---+n* band and growth in the polaron band (-750 nrn) due to the oxidation of the neutral species as the applied potential is increased, is clearly seen by spectroelectrochemistry of oligoSTI (Fig. 3 .3 1 ). The spectra of the polyNMe2STI films (Fig. 3. 30) display significantly smaller polaronlbipolaron bands than shown by the other samples. This is most likely due t o reduction of the film, which was observed o n removal of the oxidising potential. Compared to other styryl-substituted terthiophene derivatives, C larke et al. also 8 1 observed the production of vel)' small polaronlbipolaron bands during the chemical oxidation ofpolyNM�STI. 155 G o c " of o B et: 4 1 1 run 300 500 \ \ \ \ \ , , " '-, ----.. �--.... - .... - - ----- 700 900 1 100 Wavelength I nm - - - ---------- 1300 1 500 Fig. 3.26. UV -VIS-NIR spectra of a film of electrodeposited oligoN02STT in the oxidised (solid line) and neutral (dashed line) states. G o c .e o B et: 349 run 300 5 1 8 nm 500 \ \ \ \ , ... ... , '­ -- ... -.... _------- -- .. _- -- - ---- - ---- -- 700 900 1 100 Wavelength I nm 1 300 1500 Fig. 3.27. UV-VIS-NIR spectra of a film of electrodeposited oligoCNSTT in the oxidised (solid line) and neutral (dashed line) states. 82 CD o c � o .a c( 300 500 � .... - --- - - --... "--- - ---- - -- - ... - ..... - 700 900 1 1 00 Wavelength I nm -- --... _- 1 300 1 500 Fig. 3.28. UV -VIS-NIR spectra of a filin of electrodeposited oligoSTT in the oxidised (solid line) and neutral (dashed line) states. CD o c € o ! c( ---cv growth, Oxidised --- - - CV growth, Reduced --- CA growth, Oxidised . . . . . . . . . CA growth, Reduced \ , �-52 nnt" " 300 484 �', _ _ � - --- , .. . . . .... _-- -" 500 ---- - - - 700 900 1 1 00 Wavelength I nm 1 300 1 500 Fig. 3.29. UV-VIS-NIR spectra of oxidised films (solid lines) and neutral films (dashed lines) of OMeSTT deposited using potentiodynamic and potentiostatic methods. 83 401 run 300 500 -cv growth, 'Oxidised' -- -- - CV growth, Reduced ---CA growth, 'Oxidised' - - - - - - - - - CA growth, Reduced 700 900 1 100 Wavelength I nm 1 300 1 500 Fig. 3.30. UV-VIS-NIR spectra of films that were previously oxidised (solid lines) and neutral films (dashed lines) ofNM�STT deposited using potentiodynamic and potentiostatic methods. u u C et -e o en ..c c( 300 500 700 900 1 100 Wavelength I nm 1 . 1 V vs. Ag wire I __ ----, 0.4 V vs. Ag wire 1300 1500 Fig. 3.31. Spectroelectrochemistry of an electrochemically grown film of oligoSTT on ITO-coated glass. Potentials between 0.4 V and 1 . 1 V \here applied in steps of 0. 1 V. In summary, the low Amax due to the n � n* transition, and the poor free carrier tails displayed by these films when compared to polymeric materials such as polypyr ole, 1 26 indicate low conjugation. This in turn, is suggestive of short oligomers rather than polymeric material. 84 3.4.3.3 Scanning electron microscopy (SEM) The morphology of films is useful for determining properties such as conductivity and surface area. Generally, more compact films lead to a higher conductivity as polymer molecules have better connections. 57 On the other hand, porous films or films with rough surfaces can give a higher surface area, which is desirable in many applications to increase reaction ef'ficiencies. The morphology of the electrochernically deposited films of conducting polymers is dependent on the kinetics of nucleation and growth, which, in turn, generally depends on three parameters: 1 09 (I) The structure of the mono mer/p 0 lymer. (I1) The nature of the dopant. (Ill) The thickness ofthe film grafted on the electrode. SEM images were taken of NM�STT, OMeSTT, STT, CNSTT and N02STT materials grown on ITO-coated glass. The films were deposited using both potentiostatic and potentiodynarnic methods and were compared in the oxidised and neutral state. SEM images for oligoNMe2STT films deposited using potentiostatic methods, were found to have an extremely smooth, homogenous morphology with very few features in both the 'oxidised' and neutral forms. The neutral form of the film synthesised using cyclic voltarnmetry was also very smooth (Fig. 3. 32a). The compactness is not inconsistent with the idea that the increase in oxidation potential during potentiodynarnic growth of the film, may be due to difficulty in anion exchange during the oxidation process, as discussed in Section 3.4.2. In contrast to the film grown potentiostatically, the 'oxidised' potentiodynarnically grown films displayed a ' cauliflower'-like structure (Fig. 3 . 32b). The significant difference between the neutral and oxidised films is surprising since the UV -VIS-NIR spectra showed little difference between the films. However, as they are two different films (although deposited under 85 identical conditions), it cannot be ruled out that the cauliflower structure shown by the 'oxidised' film is due to a slightly faster nucleation and growth process. (a) neutral Cb) oxidised Fig. 3.32. SEM images of neutral and oxidised films of oligoNM� STT deposited onto ITO-coated glass using potentiodynamic methods. Magnification : x l OOOO. Oxidised oligoOMeSTT films deposited by potentiostatic (Fig. 3.33) methods and potentiodynamic methods appear almost identical in structure. Neutral films were observed to poorly adhere to the ITO-coated glass. Fig. 3.33. SEM image of a n oxidised film o f OMeSTT deposited using potentiodynam ic methods. Magnification: x 1 0000. Potentiostatically and potentiodynamically deposited oligoSTI films display common features as shown in Fig. 3.34 and Fig. 3. 35. The neutral films reveal crystalline­ looking structures; very similar structures were also observed by Grant et al. of electrochemically deposited dimeric SIT derivatives. ! ! On oxidation, these structures disappear to leave a porous film. The cracks observed in the oxidised 86 potentiodynamically deposited film (Fig. 3 . 35a) may be due to evaporation of the solvent causing a contraction of the film and suggest a brittle texture when dry. Ca) Neutral (b) Oxidised Fig. 3.34. Potentiostatically deposited films of STT oligomers. Ca) Neutral film: x I400, inset: x70. (b) Oxidised film: x1 400. (a) neutral (b) oxidised Fig. 3.35. SEM images of oligoSTT films deposited using cyclic voltammetry. Ca) Neutral [illn : xIOOO, inset: x5000. (b) Oxidised state: x2000 magnification. The oxidised and neutral oligoCNSTT films deposited using constant potential and cyclic voltammetry are shown in Figs. 3 .36 and 3 .37 respectively. The films show distinct fibre growth, with particularly prominent fibres in both the oxidised and neutral films deposited using potentiodynarnic methods. Fibres are an interesting feature because they increase the surface area of the polymer. 87 Fig. 3.36. SEM images of oligoCNSTT fihns deposited using a constant potential. (a) oxidised state, x I 0000 magnification, (b) neutral state, x5 000 magnification. Fig. 3.37. SEM images of oligoCNSTT fihns deposited using cyclic voltammetry. (a) oxidised state, x I 0000 magnification, (b) neutral state, x 1 0000 magnification. Films of N02STT deposited by potentiostatic and potentiodynamic methods were observed to be very similar in the neutral state. The film deposited using a constant potential is shown in Fig. 3 . 38 and reveals a surface consisting of short, fibre-like structures. Oxidised films vary slightly depending on whether they have been deposited using potentiostatic or potentiodynamic methods as shown in Fig. 3. 39. 88 Fig. 3.38. SEM image of a neutral oligoN02STT film deposited using a constant potential, x 10000 magnification. Fig. 3.39. SEM images of oxidised oligoN�STT films deposited by (a) potentiostatic and (b) potentiodynamic methods, xl 0000 magnification. In summary, SEM images of electrodeposited styryl-substituted terthiophene derivatives show crystalline structures. Similar structures observed by Grant et al. in dimeric films of styryl-substituted terthiophene derivatives, suggest these materials also consist primarily of short oligomers. 1 1 89 3.5 Conclusions A series of styryl-functionalised terthiophene derivatives that supported substituents of varying electron withdrawing/donating ability were synthesised. These materials were oxidised using both chemical and electrochemical techniques and the resulting materials investigated using MALDI-TOF MS, IH NMR spectroscopy, cyclic voltarrrrnetry, UV -VIS-NIR spectroscopy and microscopy. The materials comprise significant amounts of insoluble material (37-95%), which was difficult to further characterise and process. The soluble fractions were found to consist predominantly of dimer and short oligomers rather than polymer. It was also shown by 'H NMR spectroscopy that dimerisation was regioselective, to form primarily the head-to-head isomer. The high yields of head-to-head dimer is consistent with previous work accomplished by Grant" on ether-substituted styrylterthiophene compounds and with predictions made from theoretical calculations of the polymerisation of these materials. 90 3.6 Experimental 3.6. 1 Reagents and materials All reagents were used as received from suppliers unless specified otherwise. Chloroform obtained from BDH Laboratory Supplies (London) was used in all reactions unless specified otherwise and contains 0.5 to 1 . 0% ethanol as a stabiliser. TBAP (Fluka, Purum) was partially dried under vacuum at 70°C for 48 hours, followed by further drying under vacuum with potassium hydroxide. Note: as the length of the oligomers yielded can not be accurately measured, the theoretical yield has been calculated without taking into account the loss of hydrogen on polymerisation. At most, this leads to an uncertainty of 0.3% in the calculated polymer yields. 3.6.2 Synthesis of styryl-substituted terthiophene monomers The following monomers were synthesised according to the procedure reported by Collis et al. 80 The NMR spectra obtained matched those previously reported. Synthesis of N02STT trans-I-« 2 ' ,2" :5" ,2 ' " -terthiophen)-3" -yl)-2-( 4" " -nitrophenyl)ethane A mixture of 3 '-formyl-2,2' : 5 '2"-terthiophene (3. 3 1 g, 1 2 mmol), 4-nitrobenzyl triphenylphosphonium bromide (6.67 g, 1 3 . 87 rnmol) and DBU (2.07 mL, 1 3 . 86 mmol, 2 . 1 1 g) in dry DCM ( 1 65 mL) was heated under reflux. After 8 hrs, the reaction mixture was diluted with dichloromethane (300 mL) and washed \vith 1 M solution of HCI (2 x 1 00 mL), 1 0% sodium bicarbonate solution ( lOO mL) and water ( l OO rnL). The organic layer was dried and concentrated to give a crude orange solid. The solid was dissolved in a small quantity of DCMlhexane ( l : 1 ) and passed through a column of silica with continued elution using this solvent system until all the yellow/orange material had been collected. Analysis of the material by 1 H NMR spectroscopy indicated it consisted of a mixture of the cis and trans isomers in a ratio of 3 :2. This material was dissolved in dry chloroform (500 rnL) and irradiated for 3 hours using a 9 1 250 W flood lamp. After removal of the solvent, analysis of the sample by ' H NMR spectroscopy indicated it consisted of essentially the trans product. The material was recrystallised from dichloromethane/ ether to give the product as orange crystals (3 . 30 g, 70%). ' H NMR (27 0 . 2 MHz, CDCh) 5 8.22-8. 1 7 (2H, AA' part of AA'XX'); 7 . 60-7 . 55 (2H, XX' part of AA'XX;); 7 . 49 (d, I H, J = 1 6. 2 Hz), H I ), 7 . 44 (dd, I H, J = 5 . 1 , 1 . 2 Hz, H5 '), 7 . 4 1 (s, I H, H 4 "); 7 . 28 (dd, I H, J = 5 . 1 , 1 . 1 Hz, H 5 " ') ; 7 . 22 (dd, I H , J = 3 . 6, 1 . 1 Hz, H 3 " ' ) ; 7 . 20 (dd, I H , J = 3 . 6, 1 . 2 Hz, H 3 ' ); 7 . 1 5 (dd, I H, J = 5 . 1 , 3 . 6 Hz, H 4'); 7 . 05 (dd, I H, J = 5 . 1 , 3 . 6 Hz, H 4 " '); 7 . 05 (d, I H, J = 1 6. 2 Hz, H 2). Synthesis of CNSTT t" ans-l-« 2 ' ,2 " :5" ,2' " -terthiophen)-3 " -yl)-2-( 4"" -cyan ophenyl)ethane A mixture of 3' -formyl-2,2 ' : 5 ' 2"-terthiophene (3 . 3 g, 1 1 . 7 mmol), 4-(cyanobenzyl) triphenylphosphonium bromide (6.45 g, 14. 1 mmol) and DBU (2. 1 1 rnI, 1 4 . 1 3 mmol, 2. 1 5 g) in dry DCM (200 mL) was heated under reflu.x. After 3 hrs the reaction mixture was diluted with DCM (300 mL) and washed with 1 M so lution of HCI ( l OO mL), 1 0 % sodium bicarbonate solution ( 1 00 mL) and water ( l OO mL). The organic layer was dried and concentrated to give a crude yellow solid. Analysis of this crude material by 'H NMR spectroscopy indicated it contained the triphenylphosphine oxide byproduct and only the trans product . A silica column was used to separate the yellow solid product. This was recrystallised from ether/pentane to give yellow crystals (3. 006 g, 68%). ' H NMR (270 . 2 MHz, CDCh) 57.64-7 . 6 1 (2H, AA' part of AA'XX'); 7 . 55-7. 52 (2H, XX' part of AA'XX'), 7 . 45 (d, 1 H, J = 1 6 . 2 Hz, H I ), 7 . 43 (dd, I H, J = 5 . 1 , 1 . 2 Hz, H 5 '), 7 . 4 1 (s, I H, H 4"), 7 . 2 7 (dd, I H, J = 5 . 1 H, 1 . 1 Hz, H 5 "); 7. 2 (dd, I H , J = 3 . 6, 1 . 1 Hz, H3 " ' ); 7 . 1 9 (dd, I H, J = 3 . 6, 1 . 2 Hz, H3 '). 7 . 1 4 (dd, I H, J = 5 . 1 , 3 . 6 Hz, H 4'); 7 . 05 (dd, J = 5 . 1 , 3 . 6 Hz, H 4 " '); 7 . 00 (d, I H, J = 1 6. 2 , Hz, H 2). 92 Synthesis of STT trans-l-« 2' ,2" :5" ,2' " -terthiophen)-3" -yl)-2-(phenyl)ethane A mixture of 3' -formyl-2,2 ' : 5 '2"-terthiophene (3. 3 g, 1 1 .9 rnmol), benzyl triphenylphosphonium bromide (6. 2 g, 14 mmol), DBU (2. 2 ml, 14 mmol, 2 . 1 5 g) in dry DCM (330 mL) was heated under reflux. After 3 hrs the reaction mixture was diluted with DCM (300 mL) and washed with 1 M solution ofHCI (2 x 100 mL), 1 0% sodium bicarbonate so lution ( l OO mL) and water ( l OO mL). The organic layer was dried and concentrated to give a crude yellow solid. Analysis of this crude material by IH NMR spectroscopy indicated it consisted primarily of the desired trans product. A silica column ( 1 : 5 ethylacetatelhexane) was used to separate the product as a yellow oil, which crystallized on standing. RecrystaUisation from ether/pentane gave the product as yellow crystals (3. 778 g, 90%). IH NMR (270.2 MHz, CDCh) 87 .5 1 -7 .46 (m, 2H, aryl H); 7.43 (s, 1 H, H 4"); 7. 39 (dd, 1H, J = 5.2, 1 .2 Hz, H5 '); 7 .37 (d, 1H, J = 1 6. 1 Hz, H I ); 7.40-7 .34 (m, 2H, aryl H); 7.27 (dd, 1H, J = 5. 1 , 1 . 1 Hz, H 5" '); 7 . 30-7.24 (m, 1H, aryl H); 7 .22 (dd, 1 H, J = 3 .6, 1 . 1 Hz, H 3" '); 7 .2 1 (dd, 1 H, J = 3 .6, 1 . 2 Hz, H3'); 7. 1 2 (dd, 1H, J = 5 .2, 3 .6 Hz, H 4'); 7 .04 (dd, 1 H, J = 5 . 1 , 3 .6 Hz, H 4" '); 7.04 (d, 1 H, J = 1 6. 1 Hz, H 2). Synthesis of OMeSTT trans-l-« 2' ,2" :5" ,2' " -terthiophen)-3" -yl)-2-( 4"" -methoxyphenyl)ethane A mixture of (4-methoxybenzyl) triphenylphosphonium chloride (9. 97 g, 24 mmol, 2 equiv. ) and KOBul (2. 68 g, 24 mmol) in dry THF ( l OO mL) was heated under reflux for 1 5 min. To this was added a solution of3 ' -formyl-2,2 ' :5 '2"-terthiophene (3 .3 g, 1 2 mmol) in dry THF ( 100 mL). After 8 hrs the reaction was halted, diluted with dichloromethane (200 mL) and washed with 1 M solution of HCI (2 x 1 00 mL), 1 0% sodium bicarbonate solution ( l OO mL) and water ( l OO mL). The organic layer was dried (MgS04) and concentrated to give a crude yellow solid. This material was adsorbed onto silica before being passed through a column of silica with 1 0% ethylacetatelhexane as solvent. The yellow solid was recrystallised from ether/pentane to afford the product as yellow crystals. (3.6 1 g, 79%). 93 IH NMR (270.2 MHz, CDCb) 87.45-7 .4 1 (2H, AA' part of AA'XX'); 7.40 (s IH, H4"); 7 .37 (dd, IH, J = 5. 1 , 1 .2 Hz, H 5'); 7 . 25(dd, IH, J = 5. 1 , 1 . 2 Hz, H 5' ' '); 7 . 22 (d, IH, J = 1 6.2 Hz, H I) ; 7 .2 1 , (dd, IH, J = 3 .6, 1 . 2 Hz, H 3 ' ' '); 7 . 19 (dd, IH, J = 3 .6, 1 . 2 Hz, H 3') ; 7 . 1 2 (dd, IH, J = 5 . 1 , 3 . 6 Hz, H 4'); 7 .04 (dd, I H, J = 5. 1 , 3 .6 Hz, H 4' ' '); 6.99 (d, IH, J = 1 6. 2 Hz, H2); 6.92-6. 86 (2H, XX' part of AA'XX'); 3.83 (s, 3H, OMe). Synthesis of NMezSTT trans-l-« 2 ' ,2" :5" ,2" '-terthiophen)-3" -yl)-2-(4 " "-N,N­ dimethylami n o phenyl)ethane A mixture of (4-N,N-dimethylarninobenzyl) triphenylphosphonium iodide ( 1 3 . 1 g, 2S . 1 mmol, 2. 1 equiv. ) and KOBut (2. 82 g, 25. 1 mmol) in dry lHF (lOO mL) was heated under reflu.x for 15 min then cooled. To this was added a solution of 3 '-formyl- 2,2' :5 '2"-terthiophene (3 . 3 g, 1 1 . 9 mmol) in dry THF ( l OO mL) and heated under reflux. After 8 hrs the reaction was diluted with dichloromethane (200 mL) and washed with 1 M solution of HCI ( lOO mL), 1 0% sodium bicarbonate solution ( lOO mL) and water ( lOO mL). The organic layer was dried (MgS04) and concentrated to give a crude yellow solid. The solid was dissolved in a small quantity of dichloromethanelhexane ( l : 1 ) and passed through a column of silica with continued elution with the same solvent system until the yellow material had been collected. This was recrystallised from ether to afford the product as yellow crystals. (3 .33 g, 7 1 %). I H NMR (270.2 MHz, CDCh) 87.42 (s, I H, H 4"); 7 .43-7 .49 (2H, AA' part of AA'XX'); 7 .36 (dd, IH, J = 5.2, 1 .2 Hz, H 5 '); 7 .23 (dd, IH, J = 5. 1 , 1 . 2 Hz, H 5 ' ' '), 7 .20 (d, I H, J = 16. 1 Hz, H I ); 7. 1 9 (dd, IH, J = 3 .6, 1 .2 Hz, H 3" '); 7. 1 5 (dd, I H, J = 3 .6, 1 . 2 Hz, H3 '); 7 . 1 1 (dd, IH, J = 5 .2, 3 .6 Hz, H 4')7 .04 (dd, I H, J = 5. 1 , 3 .6 Hz, H 4"'); 6 .99 (d, IH, J = 1 6. 1 Hz, H 2); 6. 76-6 .70 (2H, XX' part of AA'XX'); 3 .00 (s, 6H, NM�). 3.6.3 Polymerisation of styryl-substituted terthiophene monomers Polymerisation NMe2STT 94 A suspension of anhydrous FeCh ( 1 80 mg, l . 078 mmol, 4.23 equiv. ) in dry DCM ( 1 5 mL) was added dropwise over 1 hr to a stirred solution of NM�STT ( 100 mg, 2 .55 x 1 0-4 mo� 1 equiv. ) dissolved in dry DCM (4 mL). The mixture was stirred at room temperature for 4 hrs and sonicated every 30 min The reaction mixture was then filtered to give a black powder, which was washed with H20 (300 ml), and then washed with methanol for 24 hrs using a soxhlet extractor. The resulting red powder (0. 776 g) was dried in vacuo for 24 hrs and then exhaustively extracted with DCM for 24 hrs in a soxhlet extractor. The eluent gave a red powder which was precipitated in chloroformlhexane to concentrate the most soluble dimers, giving a yield of 20 mg (20%). NMR, MALDI and TLC showed that it consisted mostly of HH isomers of dimer (-80%), but also a little trimer, tetramer and/or HTfTT isomers of dimer (-20%). Polymelisation of OMeSTT A suspension of anhydrous FeCh ( 1 80 mg, 1 . 078 mmol, 4. 10 equiv. ) suspended in dry DCM ( 1 5 mL) was added dropwise over 1 hr to a stirred solution of OMeSTT (of 1 00 mg, 2 .63 x 10-4 mo� 1 equiv. ) dissolved in dry DCM (4 mL). The mix1ure was stirred at room temperature for 4 hrs, sonicating briefly every 30 min . The reaction mixture was then filtered to give a black powder, which was rinsed with H20 (300 ml) and then washed in methanol using a soxhlet extractor for 24 hrs. The resulting red powder was dried in vacuo for 24 hrs and then exhaustively extracted in DCM using a soxhlet extractor for 24 hrs. The orange eluent gave a red powder, which was precipitated in chloroformlhexane, the eluent giving a yield of l9.5 mg ( 1 9. 5%). NMR, MALDI and TLC showed that it consisted mostly ofHH isomers of dimer (-80%), but also a little trimer, tetramer and/or HTfTT isomers of dimer (-20%). Polymelisation of STT A suspension of anhydrous FeCh ( 1 80 mg, 1 . 078 mmol, 3 .77 equiv. ) in dry DCM ( 1 5 mL) was added dropwise (over 1 hr) to a stirred solution of STT ( 1 00 mg, 2 .86 x 1 0-4 mo� 1 equiv. ) dissolved in dry DCM (4 mL). The mixture was stirred at room temperature for 4 hrs and sonicated briefly every 30 min. The reaction mixture was then filtered to give a black powder, which was rinsed with H20 (300 ml), and then washed with methanol using a soxhlet extractor. The resulting red powder was 9S dried in vacuo for 24 hrs before being exhaustively extracted in DCM using a soxhlet extractor to give a bright orange eluent (33 mg, 3 3%). The eluent was dried to give a redlblack powder. It was then precipitated in chloroformlhexane, the eluent giving a yield of 4 mg (4%). NMR, MALD! and TLC showed that this fraction consisted mostly ofHH isomers of dimer (�70%), but also a little trimer, tetramer and/or HTnT isomers of dimer (�30%). Polymerisation of CNSTT A suspension of anhydrous FeCh ( 1 80 mg, 1 . 0 7 8 mmol, 4 . 04 equiv. ) in dry DCM ( 1 5 mL) was added dropwise over 1 hr to a stirred solution of CNSTT ( 1 00 mg, 2 . 67 x 1 0-4 mol, 1 equiv. ) dissolved in dry DCM (4 mL). The mixture was stirred at room temperature for 24 hrs, sonicating briefly every 30 min for the first 4 hrs. The reaction mixture was then filtered to give a black powder, which was rinsed with H20 (300 mL), and then washed for 24 hrs with methanol using a soxhlet extractor. The resulting red powder was dried in vacuo for 24 hrs and then soxhletted for 24 hrs with DCM. The orange eluent gave a red powder, which was then precipitated in chloroformlhexane, the eluent giving a yield of 63 mg (63%). NMR, MALD! and TLC showed that it consisted mostly of HH isomers of dimer (�70%), but also a little monomer, trimer, tetramer and/or HTffT isomers of dimer (�30%). Polymerisation of N02STT A suspension of anhydrous FeCh ( 1 80 mg, 1 . 078 mmol, 4 . 26 equiv . ) in dry DCM ( 1 5 mL) was added dropwise (over 1 0 min.) to a stirred solution ofN02STT ( 1 00 mg, 2 . 5 32 x 1 0-4 mol, I equiv) in dry DCM (4 mL). The mixture was stirred for 1 hr under reflux. The reaction mixture was then filtered to give a black powder which was washed with methanol using a soxhlet extractor for 24 hrs. The resulting black, dumpy material (92 mg) was dried in vacuo for 24 hrs and then soxhletted for 24 hrs with DCM. The red eluent « 5 mg) was adsorbed onto silica and passed through a column of silica with continued elution using DCMlhexane ( 1 : 1 ) as a solvent until the orange material had been collected « 1 mg). 96 3.6.4 Electrochemical synthesis The electrochemical synthesis of the polymers and their electrochemical analysis were performed with an Ecochemie Autolab system PGSTAT30 potentiostatlgalvanostat service, with the associated General Purpose Electrochemical System (GPES) software. If possible, polymers were electrochemically deposited using both potentiostatic and potentiodynarnic methods. However, good solubility of some films in the neutral state prevented their potentiodynamic growth. Anodic and cathodic limits were chosen based on oxidation and reduction potentials of the monomer. A scan rate of 1 00 mV S-I was used for all growth cyclic voltammetry. Fifteen cyclic voltammograrns were used to grow the polymer. The cathodic limit was also used for potentiostatic growth where the chosen potential was held for 20 seconds. All electrochemical experimentation was performed at a controlled temperature of 22°C. Films were grO\VI1 onto either a platinum micro electrode or disc electrode (surface area = 1 . 8 mm2). The platinum micro electrode reduces the effects of diffusion and allows the monomer to be recovered as an electrolyte is not required. As not all polymers grew well on the microelectrode, a platinum disc electrode was occasionally employed. Films were also grO\VI1 onto ITO-coated glass for characterisation of the film by spectroscopic and microscopic techniques as well as analysis by mass spectrometry. Platinum mesh ( 1 cm2) was used as a counter electrode and Ag! AgN03 electrode with a 0. 1 M TBAPI AN salt bridge used as the reference electrode. The concentration of monomer for all solutions was 5 mM and an electrolyte concentration (if required) of 1 00 mM. A solvent that allows solubility of the monomer without solvation of the deposited polymerised material is needed to deposit polymer on the electrode surface. The styryl terthiophene monomer derivatives were dissolved in acetonitrile, and a 1 : 1 mixture of AN :DCM was found to have the required intermediate relative permittivity for the alkyl and alkoxy derivatives. All solutions were degassed by sonication prior to measurements. After deposition, the modified 97 electrode was rinsed in acetonitrile and transferred to mono mer-free electrolyte solution to be electrochernically analysed. Films prepared for UV-VIS-NIR analysis were electrodeposited on ITO-coated glass using potentiodynarnic and potentiostatic growth conditions given in Table 3 . 5. The films were held at an oxidising potential (using the upper limit) in monomer-free 0. 1 M TBAP/AN for 1 80 seconds to polymerise any trapped oligomers and to fully oxidise the films. The UV -VIS-NIR of the oxidised films was measured. The films were then reduced at 0 V for 1 80 seconds (using the lower limit) and the UV-VIS-NIR of the neutral film measured. SEM images were taken of this film, and another film prepared under identical conditions but left in the oxidised state. Table 3.5. Conditions of electrochemical deposition for films prepared on ITO-coated glass for characterisation by UV-VIS-NIR and microscopy. Potentiodynamic Growth Potentiostatic Growth Potential limits Cycles Potential Time NMe2STT 0 - 0. 9 5 0 . 9 20 OMeSTT 0.3 - 0. 75 3 0.75 10 STT 0.375 - 0 . 9 5 0 . 9 1 0 CNSTT 0 - 0. 9 5 0 . 9 5 N02STT 0 . 3 - 0 . 9 5 0 . 9 1 5 3.6.5 MALDI-TOF MS Spectra were recorded in the linear mode in this study due the higher sensitivity of this technique. Dithranol was used as the matrix. Chloroform was used as the solvent for both the sample and the matrix. The sample was prepared by mix.ing a sample solution (-500 Ilg mL'l) rnonomer solution with an equal volume of a 10 mg mL'l matrix solution. 1 00 ilL of this mixture was spotted onto a target plate. Detection has been suppressed below 320 mlz units to prevent overload of the detector by matrix ions. 9 8 3.6.6 UV-VlS-NIR spectroscopy UV-VIS-NIR spectra were obtained usmg a Shimadzu UV 160 1 spectrometer, scanning over the range 300- 1 500 run. Samples that were measured in solution were dissolved in chloroform, and were oxidised with Cu(CI04)2 .H20 and reduced with hydrazine.H20. Electrochemically deposited films were doped or dedoped by being held at either an oxidising or reducing potential respectively, in monomer-free solution. 3.6.7 1 H NMR spectroscopy IH NMR spectra were obtained at 270. 1 9 MHz using a JMN-GX270 FT NMR Spectrometer with Tecmag Libra upgrade. The chemical shifts are relative to the residual proton signal in deuterated solvents (CDCb 8 7 .27). Chemical shifts are reported as position (8), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, dd = doublet of doublets, m = multiplet), coupling constant (J Hz), relative integral and assignment. 3.6.8 SEM imaging SEM analysis was performed by Mr. Doug Hopcroft at HortResearch, Palmerston North, New Zealand. Samples were mounted on an aluminium specimen stub, sputter coated with approximately 1 00- 1 50 run of gold and studied using a Cambridge 250 Mk3 Scanning Electron Microscope in the secondary electron mode. Images were recorded on Ilford FP4 black and white film at the chosen magnifications and printed on Kodak multigrade paper 99 Chapter 4 Electrochemical Polymerisation of a Series of Styryl-Substituted Terthienylenevinylenes 4. 1 I ntroduction The last 1 0 years has seen a growing interest in poly(thienylenevinylene) and its derivatives (Fig. 4. 1 ), which have been found to produce the largest effective conjugation and hence the lowest HOMO-LUMO bandgap of all conducting polymers of comparable lengths.46 These materials differ from polythiophenes by having ethylene linkers between the thiophene rings. The rigid ethylene tinkers suppress rotational disorder of the polymer chain to allow enhanced n-orbital overiap,46 and reduce the overall aromatic structure of the polymer, resulting in an increase in n-electron delocalisation. The ensuing low bandgaps give poly(thienylenevinylene) and its derivatives great potential for use in applications such as solar cellsl03 and field-effect transistors. 1 05, 1 06 Fig. 4. 1. A polythienylenevinylene derivative A series of styryl-substituted terthienylenevinylene (STV) derivatives (Fig. 4.2) have been synthesised by Wagner et al. 1 56 Their chemical polymerisation, also investigated by Wagner et al. , was found to produce predominantly dirner. This absence of polymerisation was not unexpected as i t was previously shown by Levillain et al. that tetrathienylenevinylene oligomers readily dirnerise, 107 which may be due to the low reactivity of the terminal 'a' positions of the monomers as a result of their extended n- 1 00 conjugation. 23 NMR spectroscopic studies revealed that the dimers produced by Wagner et al. were not regiospecific and likely consisted of mixtures of Iffi, HT and TT isomers. 1 57 This is in contrast to the dimerisation of styryl-substituted terthiophenes where these materials were found to exhibit regioselectivity giving the HH isomer (as discussed in Chapter 3), and may be caused by loss of the polaron trap due to an increased planarity ofthe molecule produced by the addition of the vinyl linkers. R Fig. 4.2. Terthienylenevinylene monomer derivatives. R NM� (NM�STV) OMe (OMeSTV) H (STV) CN (CNSTV) N� (N�STV) In this study, the electrochemical polymerisation of these materials was investigated and compared to the styryl-substituted terthiophenes discussed in Chapter 3. 101 4.2 E lectrochemical deposition Electrochemical analysis o f polymer growth and polymer characteristics was performed on platinum micro electrodes. As described in Chapter 2, the small electroactive surface area of the microelectrode minimises distortion of experimental data by ohmic potential drop and capacitance effects, and eliminates diffusion effects of the monorner species to the electrode. The NMe2STV mono mer, however, was observed to grow very poorly on the platinum micro electrode « 0.5 nA after 1 5 cycles) so a platinum disc electrode (SA - 1 . 8 mm2) was used for the electrochemical growth and analysis of this material. An anodic limit 200 - 300 mV more positive than the potential at which polymerisation was observed to commence was chosen for polymer growth to reduce the possibility of oxidation of the vinyl groups. The electrochemical growth cyclic voltarnrnograrns of N02STV, CNSTV, STV and OMeSTV are given in Figs. 4. 3 to 4 .6 respectively, with the initial cycles given as insets for clarity. The N02STV, CNSTV and STV voltammograrns are similar and show an oxidation onset potential of 0.56 V, indicating that the reactivity of the 5 and 5" positions are not significantly affected by electron-withdrawing substituents at the para-position on the phenyl ring. This is in contrast with the observations made of analogous terthiophene derivatives where the substituents were found to strongly affect the oxidation potential. The OMeSTV derivative oxidises at a lower potential (0.44 V), possibly due to stabilisation of the resulting radical cation by the electron­ donating substituent as described in Chapter 3 . The oxidation onset potential of STY (0. 56 V) i s much lower than that measured for SIT (0. 74 V on a micro electrode), indicating that the vinyl linkers between the thiophene rings have a significant effect on the oxidation potential of the styryl derivatives. The lower oxidation potential may be caused by improved stabilisation of the more conjugated terthienylenevinylene radical cation. Subsequent cycles show two oxidation peaks. The oxidation peak at potentials above 0.6 V commences at potentials similar to the oxidation potential of mono mer (0. 56 V) on the first few cycles, indicating that this peak may be due to further oxidation of 1 02 monomeric species. The oxidation peak at lower oxidation potentials « 0 .5 V) is due to the oxidation of the deposited electroactive film. The cathodic peak is most likely due to reduction of the oxidised film back to neutral material. A positive shift of the anodic peaks, and negative shift of the cathodic peaks is observed during the growth of the N02STV and CNSTV :films as these :films thicken. Similar shifts were observed by Pringle et al. during the growth of thiophene, and were suggested to be caused by heterogeneous electron-transfer kinetics, and a decrease in film conductivity, counter­ ion mobility and possibly conjugation length. 1 47 The growth cyclic voltarnrnogram ofNMe2STV (Fig. 4 .7) appears different to growth of the other samples. A much lower oxidation onset potential of 0.22 V is observed, and four oxidation and four reduction peaks are produced on the first scan. Subsequent scans show an increase in oxidation onset potential, and a decreasing growth in current with each cycle, indicating the deposition of poorly conducting material. These observations suggest that processes other than polymerisation through the 5 and 5" positions are occumng, possibly involving the dimethylarnino substituent. Polymerisation through the dimethylarnino substituent has been suggested by Kitani et al. in the electrochemical oxidation of N,N-dimethylaniline. 1 58 15 T�==============�--------------------�� 0 4 �02 10 1 � U o o - C 5 � "2 +-�-���-..------...--.J ·05 �3 -0 1 O t 0 3 05 0 7 PoIentiol ,V .. A�g' :::I 0 0 1'--___ ._ -5 -10 +----,-----,-----�-----,------�---�� -0.5 -0.3 -0. 1 0.1 0.3 Potential I V vs Ag/Ag + 0.5 0.7 Fig. 4.3. Potentiodynamic growth of N02STV on a platinwn micro electrode (SA = 10 1lID2). Monomer concentration: 5 mM. Supporting electrolyte: 0. 1 TBAPI1 : 1 AN:DCM. Potential limits: -500/+800 m V. 1 5 cycles. Scan rate: 100 m V S· I . Inset: frrst three cycles. 1 03 8r-�D;�;===================�------------------------------� 6 �D2 �4 LDt--��r - -<1.2 +--�--______ �--��--' -0.5 -0.3 ..0.1 0.' 0.3 0.5 0.7 '",..,.01 IV VS /VAg' o � __ ............ . -2 4+-------�--------�------�--------�-------r--------�� -0.5 -0.3 41 Q 1 Q3 Q5 Potential I V VS AQlAg+ 0.7 Fig. 4.4. Potentiodynamic growth of CNSTV on a platinum micro electrode (SA = 10 1illl2). Monomer concentration: 5 mM. Supporting electrolyte: 0. 1 TBAPIl : 1 AN:DCM. Potential limits: -5001+800 m V. 1 5 cycles. Scan rate: 100 m V so l . Inset: first three cycles. 1 � ���==============�----------------------� �'D 100 ; � U Dr---__ ���� -- � ! ... -'0 +-�--�--__ �--__ --..---J -0.5 ..Q.J 0.1 :::J o 01--______ _ -100 +-------.--------.-------r-------r------�------_,--� -0.5 -0.3 -0.1 0 .1 0 .3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 4.5. Potentiodynamic growth of STY on a platinum micro electrode (SA = 10 1illl2). Monomer concentration: 5 mM. Supporting electrolyte: 0. 1 TBAP/l : l AN:OCM. Potential limits: - 5001+800 mY. 1 5 cycles. Scan rate: 100 mV so l Inset: first six cycles. 104 � r-.�o;===================�-----------------------------' - �o��------��--__ --� - 1 00 � -0.5 -0.3 -�enti:;1/VvsOifiAQ .. O,5 0.1 � o 0 __ -100 -�+-------�------�------�------�------�------�� -{).5 -{).1 0.1 0.3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 4.6. Potentiodynamic growth of OMeSTV on a platinum micro electrode (SA = 10 1J.I1l2). Monomer concentration: 5 mM. Supporting electrolyte: 0. 1 TBAP /1 : 1 AN:OCM. Potential limits: -500/+800 m V. 1 5 cycles. Scan rate: 1 00 m V S· l Inset: first three cycles. 80 60 -; � 0 ,\40 .:lD - 0.0 - C 20 ! ... � 0 0 -20 -40 0.0 0.1 02 0.4 0.6 0.8 PolenUal lVvs Ag/Ag' 1 sI cycle 0.2 0.3 0.4 0.5 0.6 Potential I V VS Ag/ Ag + 0.7 0.8 Fig. 4.7. Potentiodynamic growth of NM�STV on a platinum disc electrode (SA = 1 . 8 mm\ Monomer concentration: 5 mM. Supporting electrolyte: 0. 1 TBAP /1 : 1 AN:OCM. Potential limits: 0/+800 m V. 1 5 cycles. Scan rate: 100 m V S·l . Inset: first three cycles. 105 4.3 Characterisation of fi lms 4.3. 1 MALDI-TOF MS characterisation Except for the NMe2STV derivative, the neutral terthienylenevinylene films grown on ITO-coated glass were found to be soluble in chloroform to give solutions of various shades of purple. Analysis of the dissolved materials using MALDI -TOF MS revealed that the films comprised mostly dimer, with traces of trimer and tetramer. This is consistent with MALDI-TOF MS of the chemically oxidised materials. 1 56 In contrast, the NMe2STV derivative was found to be only partially soluble in chloroform to give a brown solution. MALDI-TOF MS analysis of this material gave a major peak at 888 Da, which corresponds to the dimer, but also major peaks at 773 and 1200 Da. Although these peaks do not correspond to any obvious product, it appears that NMe2STV is undergoing secondary reactions, consistent with the observations from the electrochemical growth data. 4.3.2 Cyclic voltammetry Post-growth CVs of electrochemically deposited N02STV, CNSTV, STY, OMeSTV and NMe2STV materials in monomer-free electrolyte are shown in Fig. 4 .8 to Fig. 4. 12, and reveal a considerable variation in the stability of the depositions. The deposited N02STV derivative appears to be very stable, the deposited CNSTV derivative slightly degrades, and a substantial reduction in current on post growth cycling of the deposited STY derivative indicates a very unstable material was produced. The OMeSTV appears stable after the first cycle, and the NMe2STV deposition is very unstable and completely degrades after 1 0 cycles. The variation in stability of the films may be due to further polymerisation of short oligomers or film degradation, as discussed for films of terthiophene oligomers (Chapter 3, Section 3.4. 3 . 1 ). 106 The peak oxidation and reduction potentials of the deposited materials after ten cycles in monomer-free electrolyte are listed in Table 4. l . The N02STV and OMeSTV depositions each show one oxidation and reduction peak, which are assigned to oxidation and reduction of oligomeric material. The peaks are narrow, which is consistent with a low polydispersity l 59 and the presence of predominantly dimer as indicated by MALDI-TOF MS results. The CNSTV and STV depositions generate two distinct oxidation peaks. As discussed in Chapter 3, Section 3 .4 .2, the presence of multiple peaks may be due to a number of reasons, including the presence of different length oligomers. 1 14 The two anodic peaks may be due to the oxidation of dimeric and trimeric material. The low oxidation onset potentials (0. 14 and 0.21 V respectively) of these derivatives support the presence of oligomers longer than dimer. 1 oo Another possible explanation for two oxidation peaks is the formation of multiple oxidation states. Elandaloussi et al. reported the formation of multiple oxidation and reduction peaks during the electrochernical cycling of butyl-substituted octathieneylenevinylene due to the formation of cation, dication and trication states.37 Electron­ spin resonance (ESR) spectroscopic characterisation could provide clarification of these hypotheses. No obvious trend is shown between the oxidation potential of the peaks and the electron withdrawing or donating properties of the substituents. Table 4. 1. Oligomer oxidation and reduction potentials on the second cycle. N02STV CNSTV STY OMeSTV Eoxidation onset / V 0.27 0. 14 0 .2 1 0.30 Eox / V 0.59 0. 39 0.32 0.52 0.63 0 .54 Ered / V 0.43 0.54 0.32 0.29 107 1 S�--------------------------------------------------� 10 '\s - c � Or----========:====---- d -s -1 0 -1 S+_-------r-------.--------.-------.--------r-------,r-� -O.s -0.3 -0.1 0. 1 0.3 O.S Potential I V VS Ag/Ag+ 0.7 Fig. 4.8. Post gro-.vth cycling of oligoN02STV deposited on a platinum micro electrode (SA = 10 !illl2). Supporting electrolyte: 0. 1 TBAP/ AN. Potential limits: -500/+800 m V. 10 cycles. Scan rate: 1 00 mV S·l. 4,-------------------------------------------------------, 2 '\ - c � 0 r----===::::::--- o -2 �+--------.-------,--------.--------r-------,--------.-� -O.S -0.3 -0.1 0.1 0.3 + O.S Potential I V VS AglAIJ 0.7 Fig. 4.9. Post grmvth cycling of oligoCNSTV deposited on a platimun micro electrode (SA = 10 !illl\ Supporting electrolyte: 0. 1 TBAP/AN. Potential limits: -500/+800 mY. 10 cycles. Scan rate: 1 00 mV S·l 1 08 ��------------------------------------------------------, �1 0 - -c � 0F-"""""";;;:-d -1 0 -�+--------r-------'--------�-------r-------'--------�� -0.5 -0.3 -0.1 0. 1 0.3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 4. 10. Post growth cycling of oligoSTV deposited on a platinum micro electrode (SA = 10 1ffil2). Supporting electrolyte: 0 . 1 TBAP/AN. Potential limits: -5001+800 m V. 10 cycles. Scan rate: 100 mV S" l 400�----------------------------------------------------� 300 - C 100 � d 0 -100 -300+-------.--------.-------,-------,.-------.-------.---� -0.5 -0.3 -0.1 0.1 0.3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 4. 1 1. Post growth cycling of oligoOMeSTV deposited on a platinum micro electrode (SA = 10 Iffil\ Supporting electrolyte: 0. 1 TBAP/AN. Potential limits: -5001+800 m V. 10 cycles. Scan rate: 100 mV S"l . 1 09 1� �------------------------------------------------� 100 � Cycle nwn"'" 1 - o -� +-------.-------r------.-------.-------.-------.--� -0.5 -0.3 -0.1 0 . 1 0.3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 4. 12. Postgrowth cycling of oligoNM�STV deposited on a platinum disc electrode (SA = 1 . 8 mm2). Supporting electrolyte: 0 . 1 TBAP/AN. Potential limits: -500/+800 mY. 1 0 cycles. Scan rate: 100 mV so l 4.3.3 UV-v/S-NIR spectroscopy Electrochemically deposited films of N02STV, CNSTV, STY, OMeSTV and NM�STV oligomers on ITO-coated glass were investigated using UV-VIS-NIR spectroscopy (Figs. 4. 1 3 to 4. 1 7 respectively). The films were first oxidised by applying a potential of 800 mV for 3 minutes. They were then removed from the electrochemical apparatus and immediately scanned. The films were then reduced (to a neutral state) by applying a potential of -500 mY, and were again spectroscopically scanned. The films were generally observed to be green/grey in both states, except for the NM�STT film, which appeared brown in both the oxidised and reduced states. The CNSTV film was the only material to show significant electrochrornic properties by turning purple when reduced. The spectra of the N02STV, CNSTV, STV and OMeSTV deposits have several features in common. In the reduced state, two peaks are observed below 800 nm most likely due to 1t � 1t* transitions. 1 24 The band at lower wavelengths (350 - 385 nm) is present in the monomers, but is absent in thienylenevinylene oligomers without styryl 1 1 0 substituents, indicating that it is due to the styryl substituents. 1 56 The band at higher wavelengths (520 - 660 run) is likely due to the oligo(terthienylenevinylene) chromophore, as observed and discussed for the terthiophene materials in Chapter 3 (Section 3.4 . 3 .2). On oxidation, these peaks greatly reduce in intensity (although are still present) and broad bands at about 900 nm and 1 400 nm appear due to the formation of radical cation species. Levillian et aI. , 1 07 reported similar peaks at 860 run and 1400 nm in the spectra of oxidised hexyl-substituted tetrathienylenevinylene due to formation of the radical cation, and Casado et al. , 1 60 reported bands at 858 nm and 1 555 nm in the spectra of an oxidised vinylene-bridged sexithiophene co-oligomer, which were also assigned to radical cation species. The NMezSTV film is different from the other spectra in that it displays two broad bands at approximately 450 nm and 860 nm rather than prominent peaks. This film does not appear to be electrochrornic. A summary of peak positions for the thienylenevinylene derivative films in the oxidised and reduced states are given in Table 4.2. Table 4.2. Wavelength maxima shown by N02STV, CNSTV, STY, OMeSTV and NM�STV oligomer ftlms. Assignment N�STV CNSTV STY OMeSTV Reduced film 1t - 1t* 375 nm 385 nm 350 nm 375 nm 520 nm 660 nm 600 nm 545 nm Oxidised 1t - 1t* 385 nm 400 nm 350 run 345 nm ftlm 535 nm 585 nm 530 nm 530 nm Polaronlbipo1aron 825 nm 855 run 890 nm 910 nm peaks 1 380 nm 1440 run 1 345 run 1 380 nm As found for the oligo(terthiophene)s, the Amax generated by the oligo(terthienylenevinylene) derivatives gives an indication of the effective conjugation length, and hence the oligomer length. 1 00 The relationship between the electron withdrawing/donating effect of the substituent and the A.nax due to the 7t�7t* transition (in the neutral state) for both terthiophene and thienylenevinylene derivatives is given 1 1 1 in Fig. 4. 1 8 . The Amax of the o ligo(terthienylenevinylene)s are consistently higher than for the corresponding oligo(terthiophene) derivatives, due to an increase in the conjugation length caused by the addition of the ethylene linker between the thiophene rings. For example, dirners have chains consisting of 1 6 double bonds, compared to sexithiophenes that have 1 2 double bonds. The increased effective conjugation of the terthienylenevinylene materials may also be due to an increase in 1t-orbital overlap created by a more planar oligomer chain due to the ethylene linkages or reduced intermolecular steric interactions as discussed previously (Section 4. 1 ). A correlation of the Amax with the substituent for both the oligo(terthienylenevinylene) and the oligo(terthiophene)s is also evident. The cyano derivatives produce the highest Amax suggesting these derivatives produce materials with the longest mean effective conjugation length. The higher Amax may also be due to a lower HOMO-LUMO bandgap caused by a lowering of the LUMO level, which is known to occur with the attachment of electron-withdrawing substituents.23 8 c ! ... o 11 « " / " '''- --' .. - , I , I , , , 300 soo , \ \ , , , \ , , '\ , .. ..... - --, -----... ... - - .. ... _ - ... - 700 900 1 1 00 Wavelength I nm - ... _- -... ----- - 1 300 1 500 Fig 4. 13. UV-VIS-NIR spectrum of oligoN02STV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). 1 1 2 700 � 1 100 Wavelength I n m Fig. 4. 14. UV-VIS-NIR spectrum of oligoCNSTV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). 300 500 700 900 1 1 00 Wavelength I nm 1 300 1 500 Fig. 4. 15. UV -VIS-NIR spectrum of oligoSTV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). 1 1 3 [�\ I r : I , �r� r l If' r J , ( \ CD , ( \ () , I \ C , f \ Cl , I \ of \ / \ o - ; � \ ---.;,.--� � \ \ " � � � -----. - - - - .. _- - - - - - ... - -- - -- - ---- -- -- - - -- 300 500 700 900 1 1 00 Wavelength I nm 1 300 1500 Fig. 4. 16. UV-VIS-NIR spectnn of oligoOMeSTV electrodeposited onto ITO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). CD () c Cl -e o ! c:( 300 500 700 900 1 100 Wavelength I nm 1300 1500 Fig. 4. 17. UV-VIS-NIR spectnnn of NM�STV e1ectrodeposited onto nO-coated glass and electrochemically oxidised (solid line) and reduced (dashed line). 1 1 4 700�--------------------------------------------� o Terthiophene • Terthienylenevinylene E 600 c - J: a. c ID 'ii � 500 400 300 -t----'-- Nitro-styryl Cyano-styryl Styryl Methoxy-Styryl ) Decreasing electron withdravving ability of substituent Fig. 4. 18. Relationship between electron withdrawing/donating effect of the substituent and the wavelength maxima due to the 7t -> 7t* transition of the oligomer-chain chromophore for both terthiophene and thienylenevinylene derivatives. As for the terthiophene derivatives discussed in Chapter 3, films of these terthienylenevinylene derivatives do not display free-carrier tails in the oxidised state, indicating a quasi-metallic state is not obtained by these materials. 1 1 5 4.4 Conclusions Electrochemical deposition of styryl-ftmctionalised terthienylenevinylenes was fOlIDd (by MALDI-TOF MS) to produce predominantly dirner with traces of short oligomers, rather than polymer. Cyclic voltammetry and spectroscopic analysis supported the presence of oligomers with short conjugation lengths in these films, consistent with analysis of chemically oxidised materialsl 56 and with analogous styryl-substituted terthiophene derivatives. The low degree of polymerisation may be due to low reactivity of the terminal ' a. ' positions of the monomers due to their extended n- . . 23 conJugatIOn. Bathochromic shifts of the n--+n* absorbance band, suggesting smaller band gaps, were consistently observed for electrochemically deposited styryl-substituted terthienylenevinylene derivatives compared to the analogous terthiophene derivatives discussed in Chapter 3. This is consistent with reports in the literature of smaller band gaps produced by poly(thienylenevinylene)s than polythiophenes, 1 02 and is most likely due to an increase in the effective conjugation length. This may be a result of ( 1 ) an increase in the number of conjugated double bonds along the polymer chain and/or (2) a more rigid polythiophene backbone caused by the inflexible ethylene linkers and/or (3) enhanced orbital overlap due to less twisting caused by interactions between the vinyl link er of the styryl ftmctionality and the thiophene rings. Electrochemical oxidation of the cyano-substituted monomer derivative appears to produce a film with the lowest Amax suggesting that this derivative produces the material with the lowest bandgap.23 This premise was supported by this film producing the lowest oxidation onset potential of the terthienylenevinylene series. Electrochemical polymerisation of NMe2STV gave unusual results including unexpected MALDI-TOF MS signals, multiple oxidation and reduction peaks by cyclic voltammetry and broad peaks by UV -VIS-NIR spectroscopy, suggesting that secondary reactions involving the dimethylarnino substituents may be occurring. As for the terthiophene derivatives discussed in Chapter 3, these terthienylenevinylene materials produce small free-carrier tails, indicating poorly conducting materials. An 1 1 6 attempt was therefore made to enhance polymerisation, and increase the solubility of these materials by attachment of alkyl or alkoxy chains to the terthiophene monomers. 1 1 7 4.5 Experimental 4.5. 1 Reagents and materials All reagents were used as received from suppliers unless specified otherwise. Chloroform obtained from BDH Laboratory Supplies (London) was used in all reactions and contains 0. 5 to 1 . 0% ethanol as a stabiliser. TBAP (Fluka, Purum) was dried under vacuum at 70°C for 48 hours and then under vacuum with potassium hydroxide. 4.5.2 Synthesis of materials Monomer synthesis was completed by Dr. Pawel Wagner at Massey University and are reported by Wagner et al. 1 56 The electrochemical synthesis of the polymers and their electrochemical analysis were performed with an Ecochemie Autolab system PGST AT30 potentiostat/galvanostat service, with the associated General Purpose Electrochemical System (GPES) software. Films were electrochemically deposited onto a platinum micro electrode (SA = 10 Ilm2) or disc electrode (SA = 1 . 8 mm2) for growth and post-growth analysis, and onto ITO-coated glass (ca. I cm2) for characterisation of the film by spectroscopy and mass spectrometry. Platinum mesh ( I cm2) was used as a counter electrode and a Ag/AgN03 electrode with a 0. 1 M TBAP/AN salt bridge used as the reference electrode. The concentration of monomer for all solutions was 5 mM in an electrolyte solution consisting of 0. 1 M TBAP in AN. All solutions were degassed by sonication prior to measurements. Polymers were grown using both potentiostatic and potentiodynarnic methods. An anodic limit 200 - 300 mV beyond the potential at which polymerisation was observed to commence was chosen for polymer growth. Fifteen cycles were used to deposit onto the microelectrode or disc electrode, and 2 to 5 cycles used to deposit onto ITO-coated glass to obtain a film thickness suitable for UV -VIS-NIR spectroscopy. A scan rate of 100 mV S-l was used for all growth cyclic voltammetry and post-growth analysis. The anodic limit was also used for potentiostatic growth where the chosen potential was held for 20 seconds. All 1 1 8 electrochemical experimentation was performed at a controlled temperature of 22°C. After deposition, the modified electrode was rinsed in acetonitrile and transferred to monomer-free electrolyte solution to be electrochemically analysed. Films prepared for UV -VIS-NIR analysis on ITO-coated glass were held at an oxidising potential (anodic limit) in mono mer-free 0. 1 M TBAP/AN for 1 80 seconds to polymerise any trapped oligomers and to fully oxidise the films . The UV-VlS-NIR spectra of the oxidised films were measured. The films were then reduced at -0. 5 V for 1 80 seconds (using the cathodic limit used for cyclic voltammetry) and the spectra again measured. The reduced films were dissolved in chloroform for MALDI -TOF MS. 4.5.3 UV-V1S-NIR spectroscopy UV-VIS-NIR spectra were obtained usmg a Shimadzu UV1601 spectrometer, scanning over the range 300- 1 500 om. 4.5.4 MALDI-TOF MS Spectra have been recorded in the linear mode in this study due the higher sensitivity of this technique. Dithranol was used as the matrix. Chloroform was used as the solvent for both the sample and the matrix. The sample was prepared by mixing a sample solution (�500 �g rnL-1) monomer solution with an equal volume of a 1 0 mg rnL-1 matrix solution. 1 00 �L of this mixture was spotted onto a target plate. Detection has been suppressed below 320 m1z units to prevent overload of the detector by matrix IOns. 1 1 9 Chapter 5 Polymerisation of Alkyl and Alkoxy Substituted Styrylterthiophenes 5. 1 I ntroduction In Chapters 3 and 4 it was shown that oxidation of styryl-substituted terthiophenes and terthienylenevinylenes produced predominantly short oligomers (n < 4) with limited conjugation. The styryl-substituted terthiophenes also produced large amounts of insoluble material (37 - 95% of total yield), making them unsuitable for device fabrication. The most common method for increasing poly thiophene processability is by solubilisation through the attachment of long alkyl or alkoxy chains.62 Alkoxy chain lengths of between 4 and 7 carbons, and alkyl chains of between 6 and 1 2 carbons have been typically attached. 1 7,59,73, 1 1 1 It has been shown that alkyl chains with a length less than 12 carbons do not significantly affect the electrical properties of polythiophenes. 59,65 Chains are often attached directly to thiophene, bithiophene or terthiophene, or to a phenyl ring that is attached to the thiophene ring.32,16 1 In addition to improving polymer solubility, alkoxy chains have been sho\\-TI to decrease the oxidation potential of the polymer, and, if located at a position a to where polymerisation occurs, may promote electropolymerisation kinetics. 1 1 1 The chemical and electrochernical polymerisation of alkyl- and alkoxy-substituted styryl terthiophenes is discussed in this chapter. Two styryl terthiophene derivatives, C7DASTT and OC6DASTT (Fig. 5 . 1 ), were initially considered as their chains are approximately the same length. The unsubstituted styryl functionality was chosen for an initial investigation of the effect of alkyl and alkoxy substituents on the resulting styryl-substituted polyterthiophene. The alkyl and alkoxy chains were substituted at the 4 and 4" positions, as Gallazzi et al. reported that terthiophenes substituted by alkyl or alkoxy substituents at the 4 and 4" positions are much more reactive and form 1 20 polymers of higher molecular weight than terthiophenes substituted at the 3 and 3" positions.62, 1 62 Both the OC6DASTT and C7DASTI monomers were available in our laboratories, having been previously made by Dr. Pawel Wagner, Dr. Sanjeev Gambhir or Ms. Amy Watson. o Fig. 5. 1 . Alkyl- and alkoxy-substituted styrylterthiophene monomers that were polymeri sed and investigated in this study. 1 2 1 5.2 Chemical polymerisation of OC6DASTT 5.2. 1 Polymerisation and reduction procedure For reasons previously discussed in Chapter 1 , Section l . 3 .2, the iron(III) chloride oxidation method was investigated for the polymerisation of OC6DASTT. A similar method to that used in Chapter 3 was employed, slowly adding 4 equivalents of oxidant as a slurry in chloroform to a solution of the mono mer. However, since chloroform was observed to be a better solvent than DCM for the styrylterthiophene derivatives discussed previously (dissolving a larger portion of the oligomeric material), chloroform was used as a solvent for the polymerisation reaction. It was hypothesised that a better solvent may lead to the production of longer polymer due to oligomers staying in solution and allowing further polymerisation to occur. The reaction procedure is outlined in Fig. 5 .2. 4 equiv. anhyd. FeCI3 CHCI3 Stirred 4 hrs RI Fig. 5.2. Polymerisation ofOC6DASTT. m+ mCr Initial addition of small amounts of iron(III) chloride (as a slurry in chloroform) to the OC6DASTT monomer solution resulted in a change in the colour of the reaction mixture from yellow to red/purple. This is possibly due to the formation of short oligomers in the neutral state. Further addition of oxidant led to the formation of a black precipitate, similar to that observed by chemical oxidation of the styryl­ substituted terthiophene derivatives in Chapter 3 . This was assumed to be the oligomers/polymers in the oxidised state. 1 22 Many substituted polythiophenes are reported to easily reduce to the neutral state by washing the oxidant from the crude material with water or methanol.44,62,7o,98 TIlls process is not well understood. However, several workers have reported that alkoxy­ substituted polythiophenes are difficult to reduce to the neutral state, and require washing with strong reducing agents, usually hydrazine or aqueous arnmonia.62,99, 1 63 An attempt was made in this study to reduce the resulting black, insoluble polymer by stirring with an aqueous hydrazine solution PolyOC6DASTT was found to be very difficult to reduce. Stirring with hydrazine for 24 hours resulted in the production of only 26% soluble material, which was later confirmed to be in the neutral state by UV-VIS-NIR spectroscopy. Further reduction of the insoluble material by stirring for several days with hydrazine provided an additional 43% of soluble material, to give a total of 67% soluble, neutral material. The soluble fraction was observed to be an intense purple colour in chloroform solution, and, when dried, appeared as a shiny, flexible, bronze-coloured film. The insoluble fraction appeared as a black, clumpy material. The difficulty in reducing the polyOC6DASTT material was partially attributed to the poor ability of the aqueous hydrazine solution to wet and penetrate the polymer particles. It was later found that the reduction process could be improved by grinding the polymer into a powder to increase the surface area available to the hydrazine. In addition, it was found that the aqueous hydrazine solution would better wet the hydrophobic polymer if the organic solvent was first completely removed. After such treatment and then stirring for several hours with hydrazine, chloroform was added to the mixture to dissolve the reduced surface layer of the particles so that the underlying oxidised polymer could be exposed, and reduced by the hydrazine. The organic layer containing the reduced polymer was then washed and dried, while the insoluble polymer was separated, dried, and the reduction process repeated until no increase in soluble material was obtained. An attempt was also made to improve the wettability of the polymeric material by adding the hydrazine in methanol, but this did not aid the reduction process. 1 23 This difficulty in reducing alkoxy substituted polythiophenes was also observed by Gallazzi et al., 62 who found they could only reduce very low molecular weight fractions of poly(4,4"-dipentoxyterthiophene). Gallazzi et al. proposed that the less soluble character of the alkoxy-substituted terthiophenes when compared to analogous alkylthiophenes, may be due to a mesomeric effect of the oxygen creating a partial double bond character between the thiophene units (Fig. 5. 3). This would increase the rigidity of the polymer backbone that would, in turn, likely decrease the solubility of the polymer. Fig. 5.3. Possible structure created by the mesomeric effect of the oxygen on alkoxy substituted thiophene polymers 62 It is also likely that the higher electron-donating ability of the alkoxy substituents than alkyl substituents stabilises the positive charge and increases the stability of alkoxy­ substituted polythiophenes in the oxidised state. 62.1 62 5.2.2 Characterisation of the soluble fraction of polyOC6DASTT The MALDI-TOF mass spectrum of the soluble fraction of polyOC6DASTT is shown in Fig. 5 .4, with peaks labelled according to the number of monomer units (n) in the oligomer. Although MALDI-TOF MS is a well established and frequently used technique for studying the chain length of conducting polymers,64.73.96. 1 1 5 spectra of polymer samples with a high polydispersity (>1 .2) are often observed to be skewed slightly towards a lower molecular weight, as described in Chapter 2, Section 2 .2 .96. 1 1 7. 1 20 The polydispersity of the crude polyOC6DASTT sample was estimated as ca. 1 . 39 by calculating approximate values of Mn and Mw using the ion-count measured for each oligomer/polymer of molecular weight, M;, as the number of molecules, "M, of 1 24 that particular molecular weight. Due to the high polydispersity, it was thought possible (and later confirmed in Section 5 .2 .4) that significant skewing of the data towards shorter oligomers was occurring. Although the spectrum of polyOC6DASTT may not be representative of the actual oligomer distribution, and peak intensities have been reported to vary for different materials, 1 l 7 the presence of signals at higher m/z may indicate a higher molecular weight for polyOC6DASTT when compared to the chemically oxidised materials in Chapter 3 . 3 Ion count 4 5 o PolyOCJ)ASTT Mrofmonomer = 550.5 g mor! 10 6000 1 0000 12000 14CXXJ mlz Fig. S.4. MALDI-TOF mass spectrum of polyOC6DASTT. Signals are labelled with the assigned oligomer length in terms ofmonomer units (n). An increase in oligomer length of polyOC6DASTT, when compared to the chemically oxidised materials studied in Chapter 3, is also evident from spectroscopic analysis. The absorption spectra of this soluble polyOC6DASTT fraction was recorded, and then the polymer oxidised using an excess of Cu(CI04hH20 and the spectra again recorded (Fig. 5 .5). Three bands appear in the neutral state. The major peak at 532 nm is assigned to the 1t--+-1t* transition of the oligo(terthiophene) chromophore as described in Chapter 3, Section 3 .4 .3 .2 . The broad band at about 950 nm is due to small amounts of oxidised material. The shift in the 1t--+-1t* transition (from 497 nrn in oligoSTT to 532 nm in polyOC6DASTT) is evidence of an increase in the mean conjugation length of these materials compared to the materials investigated in Chapter 3. 1 14, 127 The oxidised state displays a broad band between 500 and 900 run, which, by comparison to UV -VIS-NIR spectra of poly(3-hexylthiophene) reported by 1 25 Skompska et al. 1 25 is attributed to polaronlbipolaron speCIes. The significant free­ carrier tail generated by polyOC6DASTI in the oxidised state at a wavelength greater than 900 nm shows quasi-metallic behaviour, and is further evidence of extended conjugation. 1 27 � ... c == .c ... Si .c 00( 300 500 "" � -- ...... ---- -.. - - - 700 900 1 1 00 1300 1 500 Wavelength I nm Fig. 5.5. UV-VIS-NIR spectrwn of the soluble fraction of polyOCJ)ASTT in solution in the oxidised state (excess Cu(CI04h, solid line) and neutral state (dashed line). The 1t-1t* band in the neutral state is labelled with its wavelength maximum. A cyclic voltamrnogram of a cast film of polyOC6DASTT on an ITO-coated glass electrode (Fig. 5 .6) displays very broad oxidation and reduction peaks. Broad oxidation and reduction peaks may be due to a number of factors. Yasser et al. suggested that the broadness of the anodic peak observed while cycling polythiophenes is indicative of the length of conjugated segments, with broad peaks consistent with a wide distribution of the various lengths of conjugated segments, each with a particular Eox. 1 59 Alternatively, Pringle et al. 147 proposed that the broad bands observed while cycling thiophene in an ionic liquid may be related to the reorganisation of the polymer chain that occurs \vith ion movement int% ut of the film matrix, or due to a range of oxidation potentials due to defects in the polymer chain. The very low oxidation onset potential shown by polyOC6DASTI (-0.34 V) implies the presence of highly conjugated oligorners. It is well established theoreticallyl64 and experimentallyl 1 4, 1 27, 1 65, 1 66 that an increase in conjugation length of conducting polymers results in a reduction of oxidation potential. 1 26 2 .---------------------------------------------------� � 1 - -1 -2 +-----.-----.------.-----,-----,----�----�------r_� -0.5 -0.3 -0.1 Q1 Q3 Q5 Q7 Potential I V VS Ag/Ag+ 0.9 1 . 1 Fig. 5.6. CV of chemically polymerised OC6DASTT, which has been cast as a film onto an ITO­ coated glass electrode (ca. 1 cm2). Supporting electrolyte: 0 . 1 M TBAPI AN. Potential limits: -500/+ 1200 m V. 16th to 20th cycles. Scan rate: 100 m V S-I . The remaining insoluble material (33%) is likely to be made up of longer polymer chains, which are either in the reduced or oxidised state, since long oligomers are less soluble than short oligomers. 96. 1 27 It has been also been shown by Bredas et al. 1 64 and Sumi et al. 1 I 4 that long oligomers are more difficult to reduce, and therefore solubilise, than short oligomers. The difficulty of reducing long oligomers compared to short oligomers was also observed in this study, as oligomers from later reductions were found to be longer than oligomers that were obtained by previous reductions. It is also possible that the insoluble material is a result of cross linking between oligomers, or oligomer defects. However, this is unlikely since Gallazzi et al.62 reported that the NMR spectrum of remaining insoluble chemically polymerised poly(4,4"­ dipentoxyterthiophene) confirmed that the insolubility was not due to polymerisation structural defects such as branching or p-coupling. 5.2.3 Improvement of the polyOCsDASTT soluble fraction As the insoluble material was believed to consist of polymers too long to be soluble or reduced, it was thought the soluble fraction could be increased by reducing the 1 27 polymer length. The degree of polymerisation can be moderated either by reducing the rate of reaction or by reducing the reaction time. In these investigations, the reaction time was kept constant while the rate of reaction was reduced by either decreasing the oxidantmonomer ratio or lowering the temperature. Effect of the oxidant concentration. An iron (Ill) chloride oxidant-to-monomer ratio of about 4: 1 was used for initial reactions in this study since this ratio is commonly reported in the literature for the polymerisation of thiophene derivatives.44.63.7o.98.1 34.1 35 An attempt was made to decrease the resulting oligomer length by employing a lower oxidantmonomer ratio. Kurnar et al. 1 67 found that decreasing the oxidant monomer ratio during polymerisation of a CI4 alkyl derivative of EDOT produced a higher yield of soluble polymer. GPC characterisation of the soluble product revealed this was most likely caused by a significant decrease in the average molecular weight of the resulting polymer. The polymerisation of OC6DASTT usmg a 1 : 1 oxidant to rnonomer ratio was investigated. Since investigation of the reaction mixture by TLC showed the presence of large amounts of monomer after 4 hours, the reaction mixture was stirred for a further 92 hours to allow time for the polymerisation to complete. Other conditions were not changed from the initial polymerisation (Fig. 5 . 2). A black precipitate was produced, which was reduced with hydrazine to give a chloroform-soluble fraction of 5 5% . A soxhlet extraction of this soluble material with methanol resulted in a yellow product, which was shown by MALDI-TOF MS to consist of unreacted monomer that was 3 1 % of the total expected yield. It appears that using only one equivalent of oxidant results in the production of mostly long, insoluble polymeric material (45%), and unreacted monomer (3 1 %). Effect of temperature. An attempt was made to increase the soluble fraction of polyOC6DASTT by reducing the reaction temperature to slow the rate of polymerisation and reduce the average oligomer length. Gallazzi et al. 62 investigated the effect of temperature during 128 polymerisation of 3,3"-dihexylterthiophene and reported that polymerisation at 30°C produces shorter oligomers than polymerisation at 50°C. By keeping the conditions the same as for the first polymerisation of polyOC6DASTT (Fig. 5 . 2), but reducing the temperature during the reaction from room temperature to - l OoC (difference of ca. 30°C), the total soluble material obtained was increased from 67% to 78% of the expected yield. The increase in the soluble fraction with decrease in temperature, is most likely due to a reduction in rate of the polymerisation reaction and hence the production of shorter (more soluble) oligomers, as observed by Gallazzi et al. 62 5.2.4 Polymer separation according to chain length An attempt was made to separate the soluble polyOC6DASTT product (in the neutral state) into fractions according to chain length. This was expected to provide materials with lower polydispersities (range of oligomer lengths), and different properties than the crude material. It has been established that an increase in chain length of polythiophenes results in a bathochrornic shift of the 1t�1t* transition, and an increase in the doped conductivity of the material. 1 1 4.1 27 Separation of the crude neutral polyOC6DASTT product mixture was accomplished by subjecting the total amount of neutral polymeric material to sequential soxhlet extractions using a series of solvents. As reported by other researchers,96. 1 27 different length oligomers were found to have different solubilities, with longer oligomers being less soluble. Methanol and hexane were used first to remove any residual monomer and very short oligomers such as dimer and trimer. Acetone followed by dichloromethane and chloroform were then used to extract increasingly longer oligomers. Each solvent was found to be able to dissolve the oligomers extracted by the preceding solvents. For example, acetone dissolved all material extracted from the methanol and hexane fractions, and DCM dissolved all material extracted from the methanol, hexane and acetone fractions. Chloroform was found to dissolve all soluble material. A photograph of chloroform solutions of the monomer and these oligomer fractions is displayed in 1 29 Fig. 5 . 7. The more intense colour of the latter fractions suggests these contain polymers with longer conjugation lengths. (a) (b) (c) (d) (e) Fig. 5.7. Photograph of the monomer and solvent fractions as solutions in chloroform. (a) Monomer, (b) hexane, (c) acetone, (d) dichloromethane and (e) chloroform. The amount of polyOC6DASTT (polymerised using 4 equivalents of iron chloride at - 1 0°C) that was extracted by each solvent, and the total soluble fraction obtained, are listed in Table 5 . 1 as mass percentages of the total expected polymer yield. The expected yield is based on the mass of the monomer given that the length of the polymer products is not accurately known. At most, this leads to an uncertainty of 0 .3% in the calculated polymer yields due to not taking into account the loss of (l­ hydrogen on polymerisation. No material was extracted by methanol. As both the oxidant and monorner dissolve in methanol, this confirmed that all of the oxidant had been removed, and that no unreacted monomer remained. The majority of the soluble material (66%) was extracted into dicWoromethane and chloroform. 1 30 Table 5. 1. Mass percentages of OCJ)ASTT oligomers separated by different solvents. Methanol (%) Hexane (%) Acetone (%) OCM (%) o 7 5 24 42 Total soluble material 78% MALDI-TOF mass spectra of the polyOC6DASTT oligomer fractions are gtven ID Fig. 5 . 8. Signals have been labelled with the assigned oligomer length in terms of monomer tm.its (n) . As expected, the average oligomer length increases with each successive solvent extraction. Short oligomers (n = 1 - 4) have been for the most part extracted by hexane and acetone. The DCM fraction is shown to consist of oligomers of length n = 4 - 1 3, with an average oligomer length of n = 7 . The chloroform fraction consists of even longer oligomers with an average length of approximately n = 14 and displaying signals corresponding to oligomers of length up to n = 35. The initial cluster of peaks observed in this spectrum (less than ca. 3000 Da) is likely due to multiple charging of oxidised polymer or fragmentation of longer polymer chains, given that all of the polymeric material of mass < 4000 Da should have been extracted into one of the other solvents. In Section 5 .2 .2 it was noted that it was difficult to determine the number average molecular weight (Mn) of the crude polyOCIODASTT soluble fraction from the MALDI-TOF MS data due to the high sample polydispersity (Section 2.2.4). By estimating the number average molecular weight (Mn) of each solvent fraction and considering the yields of the solvent fractions, an average length of about n = 1 1 was estimated for the total polymer sample. This estimate of the average polymer length of polyOC6DASTT (n - 1 1 ) is much higher than what appears to be the case from the spectrum of the crude material (n - 3, Fig. 5 . 4), and supports the notion of significant skewing of the crude MALDI-TOF MS data to smaller oligomer lengths. As indicated previously, Byrd et al. showed that degree of skew is more likely to occur for \vide polydisperse polymer samples (PD > 1 .2) than for narrow polydisperse polymer samples (PD < l . 2). 1 1 7 1 3 1 3 2 3 4 5 7 4 1 2 1 5 Solvent: Hexane Yield: 7% Solvent: Acetone Yield: 5% Solvent: [)CM Yield: 24% Solvent: Chloroform Yield: 42% 35 o 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 mlz Fig. 5.8. MALDI-TOF mass spectra of OCJ)ASTT oligomer fractions extracted using hexane followed by acetone, dichloromethane and chloroform. Signals are labelled with the assigned oligomer length in terms of mono mer units (n). Mr of mono mer: 550.5 g mOrl . The absorption spectra of the OC6DASTT oligomer fractions, in both the neutral and oxidised states are shown in Fig. 5 . 9. A considerable increase in the intensity and Amax 1 32 of the free-carrier tail produced by the materials in the oxidised state is observed with subsequent solvent extractions, (assuming a A.n.x higher than 1 500 nm for the chloroform fraction as estimated by extrapolation). In addition, an overlay of the spectra of the materials in the neutral state (Fig. 5 . 1 0) clearly shows a significant bathochrornic shift of the A.n.x due to the 1t -+ 1t* transition with each extraction, in accordance with the range of colours displayed by these fractions (Fig. 5 . 7). The bathochrornic shift of both the free-carrier tail and the A.n.x(1t-+1t*) indicate an increase in the mean conjugation length of the materials with subsequent soxhlet extractions. 1 14 This is consistent with an increase in the mean oligomer length as indicated by the MALDI-TOF MS results (Fig. 5 . 8). It is interesting to note the increase in two absorbances at ca. 660 and 900 nm with increasing oligomer length (Fig. 5 . 1 0). While the shoulder at 660 nm may be indicative of longer oligomers, the broad absorbance at 900 nm suggests that the "neutral' oligomer fraction is partially oxidised. This may be due to polymer that was not fully reduced by hydrazine. As mentioned previously, it has been shown by Bredas et al. l 64 and Surni e t al. 1 14 that long thiophene polymers are more difficult to reduce than short oligomers. Partially oxidised material may also result from an increasing tendency for the longer polymers to oxidise on standing. This could be attributed to an increased polaron stability in these styryl-substituted polymers as a result of the "defect confinement" postulated for styryl-substituted sexithiophenes. l 40 1 33 (a) � - - ,.... - -- - - - -� - - _ ... .",- ... 300 500 700 900 1 1 00 1300 1 500 (b) , ...... - - -- - - - -- -et - - - _w 300 500 700 900 1 1 00 1 300 1 500 (c) ..... - -,,- -. .... - - .- - - -.. � 300 500 700 900 1 1 00 1300 1 500 - I , I " (d) "\ I \ I \ :\ / - - - .... '-... - -... � - 300 500 700 900 1 1 00 1300 1 500 Wavelength I run Fig. 5.9. UV-VIS-NIR spectra ofOC6DASTT oligomer fractions separated by (a) hexane, (b) acetone, (c) OCM and (d) chloroform. Samples were measured in the oxidised state (excess Cu(CI04:n, solid line) and neutral state (dashed line) from solutions in chloroform. 1 34 � u = = .. .. Q '" .. -< 300 500 700 900 Wavelength / nm 1 1 00 1300 A.uax of 1t-+1tW transition -Chloroform 564 run --OCM 539 run --Acetone 5 lO run - - - - - Hexane 500 run Fig. 5. 10. UV-VIS-NIR spectra of neutral samples of OC6DASTT oligomer fractions which have been separated using hexane, acetone, OCM and chloroform. The A.nax of the major 7t-+7t* transition for each fraction is shown. IH NMR spectra of these fractions showed very broad peaks, indicating that polymerisation is not regioselective. 5.2.5 Conclusions OC6DASTT was successfully polymerised by iron(III) chloride. As anticipated from previous work by others,62 the attachment of alkoxy chains to the polymer backbone was found to greatly increase the solubility of the polymeric product in organic solvents, compared to oligoSTT investigated in Chapter 3. PolyOC6DASTI was found to be very stable (w. r.t. charge) in the oxidised state, and hence difficult to reduce. A soluble fraction of 67% was initially obtained by reduction with hydrazine, but was increased to 78% by decreasing the polymerisation temperature from room temperature to -1 QOC, probably due to a decrease in the average oligomer length. The insoluble material most likely consists of oligomers that are too long to be soluble and/or too long to be reduced (and therefore solubilised). UV -VIS-NIR spectroscopy of the soluble fraction revealed a significant free-carrier tail generated by this material in the oxidised state, consistent with the production of highly conductive materials. The soluble material was successfully separated into fractions of different length oligomers/polymers by soxhlet extractions in a series of solvents. 1 35 MALDI-TOF MS of these fractions showed that the total soluble product consists of polymers with an average length of ca. n = 1 1 . The effective conjugation length of the polymer fractions was observed (by UV -VIS-NIR spectra) to increase with an increase in polymer length (determined by MALDI-TOF MS). 1 36 5.3 Chemical polymerisation of C7DASTT 5.3. 1 Introduction The polymerisation of C7DASTT (Fig. 5 . 1 ) was investigated. Gallazzi et al. have shown that alkyl-substituted polythiophenes are more easily reduced than alkoxy derivatives, and are consequently easier to solubilise.62 Cralkyl chains, which are approximately the same length as the OC6-alkoxy chains, were chosen to allow a more direct comparison to OC6DASTT. 5.3.2 Polymerisation and reduction C7DASTT was polymerised using a similar method to that used for the polymerisation of OC6DASTT (Fig. 5 .2). As for OC6DASTT, chloroform was found to be a good solvent for the C7DASTT polymerisation. The reaction was completed at room temperature, rather than at - 10°C, to allow the formation of longer polymer chains. After four hours, the reaction mixture manifested a black precipitate, similar to that observed for the polymerisation of OC6DASTT, and the styryl-substituted terthiophenes discussed in Chapter 3. As for those materials, the insoluble material was assumed to be oxidised polymer. Removal of the unreacted oxidant from the chloroform reaction mixture by washing with water resulted in an immediate change to a bright orange solution, most likely due to reduction of polyC7DASTT to the neutral state. Andersson et al. suggested that the high stability of alkyl substituted polythiophene derivatives (including 4,4"-alkyl-substituted poly(terthiophene» in the neutral state may be due to low accessibility of the counter ion to the polymer backbone due to steric hindrance by the alkyl side chains. 168 Upon removal of the chloroform solvent, a red gum was obtained that was 1 00% soluble in chloroform. The MALDI-TOF mass spectrum of this crude material (Fig. 5 . 1 1 ) showed a peak distribution similar to that observed by polyOC6DASTT (Fig. 5 .4). 1 37 4 2 3 5 o 2000 Poly(C7DASTT) Mrof monomer = 546. 5 g mor1 10 4000 6000 8000 1000 1 2000 1 4000 mlz Fig. 5. 11 . MALDI-TOF mass spectrum of poly(C7DASTT). Signals are labelled with the assigned oligomer length in terms of monomer lDlits (n). The UV -VIS-NIR spectra of polyC7DASTT, measured in both the neutral and oxidised states are given in Fig. 5 . 1 2 . Neutral C7DASTT oligomer/polymer affords two peaks at 3 2 3 run and 407 run due to the styrylthiophene and oligoterthiophene chromophores respectively, as described for the materials in Chapter 3. The much lower Amax(1t-1t*) of the polyC7DASTT oligoterthiophene chromophore (407 run) than that observed for the polyOC6DASTT oligoterthiophene chromophore (532 nm) may be indicative of a shorter mean effective conjugation length. 1 14. 1 27 The hypsochromic shift may also be partially attributed to the alkyl substituents . While non­ bonding pair substituents, such as alkoxy groups, are reported to shift absorptions substantially to longer wavelengths, alkyl substituents are known to shift absorptions less significantly to longer wavelengths. 1 69 As is typical for polythiophenes, 1 7o two bands are observed in the oxidised state, assigned to the polaronlbipolaron species (788 nm) and polymeric chains displaying quasi-metallic behaviour (free-carrier tail, 1 36 3 nm). However, the low A..n.x of the free­ carrier tail in comparison to that of polyOC6DASTT (Fig. 5 . 5 , peak > 1 5 00 nm) also suggests that the polyC7DASTT sample may consist of shorter oligomers. 1 38 r I 407 � , ,. u I:: \, \ � � ... , Q '" , � -< \ \ 300 500 700 900 1 1 00 1 300 1 500 Wavelength I nm Fig. 5. 12. UV-VIS-NIR spectra of polyC7DASTT in the oxidised state (excess Cu(CI04:n, solid line) and neutral state ( dashed line). The CV of a polyC7DASTT film cast on ITO-coated glass (Fig. 5. 13) displays an oxidation peak at 0 . 98 V and a reduction peak at 0 . 3 2 V, consistent with a one­ electron reversible redox process. In comparison to polyOC6DASTT (Fig. 5 . 6), which generated very broad oxidation and reduction bands suggesting a wide distribution of oligomer lengths, the polyC7DASTT peaks are well-defined, not inconsistent with a narrow distribution of oligomer lengths. 1 59 4.----------------------------------------------------, 3 1 2 - c � 0 1_�==;;��------� � -1 -2 � +-----,_----_r----_r----�----�r_----r_----._----,_� -0.5 -0.3 -0. 1 0.1 0.3 0.5 0.7 Potential I V VS Ag/Ag+ 0.9 1 . 1 Fig. 5. 13. CV of a chemically polymerised C7DASTT film, which has been cast onto an ITO-coated glass electrode (SA: -1 cm2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -500/+l200 mV. 5 cycles. Scan rate: 100 mV s·1 1 39 5.3.3 Polymer separation according to oligomer length PolyC7DASTI was separated into fractions of different length polymers usmg sequential soxhlet extractions in a series of solvents as for polyOC6DASTT. Unlike the neutral polyOC6DASTI fractions, which were observed to be a range of colours in chloroform solution, the neutral polyC7DASlTI polymer fractions were all orange. The mass percentages obtained in each solvent are listed in Table 5 .2, and the MALDI­ TOF mass spectra of each fraction given in Fig. 5 . 14 . As 63% of the crude product is extracted by methanol and hexane, which is shown to consist of mostly short polymers (n = 1 - 8), the overall molecular weight appears to be low with an average polymer length of about n = 6 (estimated as for polyOC6DASTI, Section 5 .2.4). Inspection of the DCM fraction reveals the presence of polymers up to n = 12. It appears that polyC7DASTT consists of oligomers/polymers that are significantly shorter than those measured for polyOC6DASTI, where an average polymer length of n = 1 1 was estimated, and polymer with a length of up to n = 35 was detected (Fig. 5 . 8). Table 5.2. Mass percentages ofpolyC7DASTT separated by different solvents. Methanol (%) Hexane (%) Acetone (%) OCM (%) 6 57 1 3 23 CHCh (%) Total soluble material 100% 1 40 r-­ o �1 2 2 4 3 5 j I 1 4 5 1 .... .J. 200 ... 8 10 ,l ....l M, l .... 4000 6000 Solvent: Methanol Yield: 6% Solvent: Hexane Yield: 57% Solvent: Acetone Yield: 13% Solvent: DCM Yield: 23% 0000 10000 Fig. 5. 14. MALDI-TOF MS of C7DASTT oligomer fractions extracted using methanol followed by hexane, acetone and dichloromethane, Signals are labelled with the assigned oligomer length in terms of monomer units (n), Mr of monomer: 546,5 g mor '! , 1 4 1 Absorption spectra of C7DASTT polymer fractions, in both the neutral and oxidised states, are displayed in Fig. 5 . 1 5 , with an overlay of the spectra of the neutral fractions in Fig. 5 . 1 6 . The Amax due to the 1t-1t* transition of the oligo(terthiophene) chromophore (ca . 4 10 nm) is observed to slightly increase with subsequent extractions, indicating an increase in effective conjugation length. 1 1 4 There is also a significant bathochromic shift in the free-carrier tail (> 1 000 nm) with subsequent extractions, which also indicates the presence of oligomers with a longer conjugation length in later fractions. 1 1 4 The increase in polymer length with subsequent extractions is consistent with conclusions from MALDI-TOF MS data for these fractions. Comparison of the positions of the Amax(1t-1t*) and the free carner tail of the C7DASTT polymer fractions to those of the OC6DASTT polymer fractions (Fig. 5 . 8) suggests that the C7DASTT solvent fractions contain oligomers/polymers with much shorter effective conjugation lengths. This is consistent with the shorter C7DASTT oligomer/polymer lengths detected by MALDI-TOF MS. In addition, it is clear that the reduced (neutral) form of polyC7DASTT is not as prone to partial oxidation as polyOC6DASTT is, as all the C7DASTT oligomer fractions appear fully reduced. This supports the proposal that the long chain soluble fractions of polyOC6DASTT are partially oxidised as a result of polaron stabilisation by the alkoxy groups. 1 42 (a) 300 500 700 900 1 100 1300 1 500 (b) , 300 500 700 900 1 1 00 1300 1 500 (c) 300 500 700 900 1 100 1300 1 500 (d) 300 500 700 900 1 1 00 1300 1 500 Wavelength I nm Fig. 5. 15. UV-VIS-NIR spectra of C7DASTT oligomer fractions separated by (a) methanol, (b) hexane, (c) acetone and (d) DCM. Samples were measured in the oxidised state (excess CU(CI04h, solid line) and neutral state (dashed line) from solutions in chloroform. 1 43 300 350 400 450 500 Wavelength I run 550 600 DCM Acetone Hexane A....x of 1t-+1t� transition 41 3 run 407 run 404 run Fig. 5. 16. UV-VIS-NIR spectra of neutral samples of C7DASTT oligomer fractions \,vhich have been separated using hexane, acetone and DCM. The Am.x of the 1t-+1t* transition for each fraction is shown. 5.3.4 Conclusions C7DASTT was successfully polymerised with iron(III) chloride to give a polymeric product that was 1 00% soluble in chloroform. The increase in the soluble fraction of polyC7DASTT ( 1 00%) compared to that obtained for polyOC6DASTT ( 7 8 %) may be due to two reasons. First, an average polymer length of n = 6 was estimated for this product (from MALDI-TOF MS of oligomer fractions), which is much shorter than the average length of polyOC6DASTT, which was estimated as ca . n = 1 1 . It has been shown by several workers that shorter oligothiophenes are more soluble than long oligothiophenes.44, 1 1 4,1 27 Secondly, polyC7DASTT was found to be very easy to reduce to the neutral, more soluble state than polyOC6DASTT. UV -VIS-NIR spectra of polyC7DASTT showed a substantial polaron band and less significant free-carrier tail than polyOC6DASTT, indicative of a shorter effective conjugation length. This could be due to the presence of shorter oligomers as revealed by MALDI-TOF MS. The o:x.'}'gen atom of alkoxy substituents appears to be activating polymerisation in some way. This is consistent with observations by 1 44 Girotto et al. I I I and Gallazzi et al. 62 on the polymerisation of 4,4"-pentoxy-substituted terthiophene derivatives. Since C7DASTT gave shorter oligomers than OC6DASTT, further attention was focussed on alkoxy derivatives, and increasing the soluble fraction. A terthiophene substituted with longer alkoxy chains was investigated. 145 5.4 Chemica l polymerisation of OC10DASTT 5.4. 1 Introduction OC6DASTT gave longer oligomers than C7DASTT, but only 78% of the oligomeric/polymeric product was soluble. An attempt was made to further increase the soluble fraction by investigating styrylterthiophene with longer, CIO-alkOxy chains attached (Fig. 5 . 1 7). This compound is referred to as OCIODASTT. OC1oDASTT Fig. 5. 17. Cwdialkoxy styrylterthiophene (OC1oDASTT). 5.4.2 Polymerisation The reaction conditions used to polymerise OCIODASTT (Fig. 5 . 1 8) were similar to those used for the polymerisation of OC6DASTT. Since a higher soluble yield of polyOC6DASTT was obtained when polymerising at temperatures of -1 0°C rather than RT, this temperature was used for the polymerisation ofOCIODASTT. 1 46 4 equiv. anhyd. FeCI3 CHCI3 Stirred 4 hrs - 10DC Fig. 5. IS. Reaction procedure employed for the polymerisation of OC10DASTT. m+ mCr Polymerisation using these conditions gave a soluble fraction of 97%, considerably higher than that obtained for OC6DASTT (78%), and the MALDI-TOF MS of the soluble product (Fig. 5 . 1 9) shows peaks due to oligomers of lengths up to n = 1 5 . 3 5 4 o 200 PolyOC1 oDASTT Mrofmonomer = 662.6 g mor 1 1 5 4000 600 8000 10000 12000 mlz 14000 Fig. 5. 19. MALDI-TOF MS of polyOC lODASTT. Signals are labelled with the assigned oligomer length in terms of mono mer units (n). The UV-VIS-NIR spectra of polyOC lODASTT (Fig. 5 .20) in the neutral and oxidised states offers further evidence of long polymer chains. The Amax for the 1t-1t* transition of the material in the neutral state is 537 nm, which is slightly higher than that observed for polyOC6DASTT (532 nm). This is consistent with reports by Casalbore-Miceli et al. 1 62 of a bathochromic shift in absorbance maxima with an increase in alkyl chain length on poly(terthiophene)s. It was suggested by Casalbore-Miceli et al. that this shift may be due to an increasing planarity of the polythiophene chains. 162 As seen for polyOC6DASTI, a 'shoulder' has appeared at about 700 nm and a second absorbance at about 900 nm indicative of partial oxidation. 1 47 ..... _ - -- - ..... - .......... ""-. 300 500 700 900 1 1 00 1 300 1 500 Wavelength I nm Fig. 5.20. UV-VIS-NIR spectra of POlyOClODASTT in chloroform in the oxidised state (excess Cu(CI04n, solid line) and neutral state (dashed line). The major A.nax for the 7t-->7t* absorbance is labelled. A cyclic voltammogram of a cast film of polyOCIODASTT on an ITO-coated glass electrode (Fig. 5 . 2 1 ) displays very broad oxidation and reduction peaks, as for the cast film of polyOC6DASTT (Fig. 5 . 6 , Section 5 . 2 . 2). The first cycle is significantly different to subsequent cycles as is commonly observed in post-growth CVs of polythiophenes. 147 Although the origin of this is not well understood, possible causes may include degradation of the polymer film, 72,1 45 a change in the organisation of the polymer chains and morphology of the film affecting ion movement int% ut of the film, and further polymerisation of shorter oligomers, J 7 J . A higher oxidation onset potential is observed (-0 . 2 3 V) than for the polyOC6DASTI film (-0 . 34 V). This may indicate that the polyOCIODASTI film consists of polymer with a lower mean conjugation length. 1 14, 164 1 48 0.8 ,----------------------------, 0.6 �0.4 -1 0.2 t::: :::J 0.0 }--::::; __ � o �.2 �.4 �.6 +----.----.----r----r----r---�r-� �.5 '{).3 �.1 0 .1 0 .3 0.5 Potential I V VS Ag/Ag+ 0.7 Fig. 5.21 . CV of chemically polymerised OC,oDASTT, which has been cast as a film onto an no. coated glass electrode (SA: ca. I cm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -5001+1 200 mY. 1 0 cycles. Scan rate: lOO mV s· 1 Attempts were made to further increase the average polymer length of polyOCIODASTT without decreasing the soluble fraction. The effect of the order of addition of reagents on the soluble fraction of polyOCIODASTT was investigated. While polymerising 3, 3"-didodecyl-2,2 ' : 5 ' 5 "-terthiophene with iron(III) chloride, GaIlazzi et al. 98 found that the molecular weight was strongly dependent on the addition sequence of the reagents, which affected the initial oxidant concentration. They compared the difference between adding iron(III) chloride to a solution of monomer, and adding a solution of the monomer to the iron(III) chloride. Although they observed 1 00% solubility for both resulting polymers, a product consisting of longer oligomers was obtained when the rnonomer was added to the iron(III) chloride. In all previous reactions mentioned in this study, the iron(III) chloride oxidant was added dropwise (as a suspension in chloroform) to the monomer over about thirty minutes. Polymerisation o f OC I ODASTT by addition of the oxidant to the monomer resulted in a soluble fraction of 97%, but it was found that by adding the monomer to the oxidant, a soluble fraction of only 44% was obtained. Assuming the insoluble polymer is due to oligomers too long to be soluble, this result is consistent with 149 Gallazzi's result of longer oligomer formation by removing impurities. This increase in o ligomer length when the monomer is added to the oxidant is most likely due to the initial high ratio of oxidant to rno no mer, as a similar increase in polyOC6DASTT oligomer length was previously observed when the oxidantmonomer ratio was increased from 1 to 4 equivalents (Section 5 . 2 . 4). The effect o f ethanol on the soluble fraction polymer length was also investigated. While polymerising a C I 2 alkyl-substituted terthiophene with iron(III) chloride, Gallazzi et al.98 found that the molecular weight and yield of soluble product was strongly dependent on the quality of the reagents. They found that washing and drying the reagents produced a higher soluble yield, presumably as a result of a shorter oligomer length. It was unclear, however, which impurity was causing the decrease in solubility and no explanation was offered. Although poly thiophene reactions are reported to be generally less sensitive to water than other conducting polymers such as polypyrrole and polyaniline,41,1 72 both ethanol and water have been shown to influence the oxidation potential of iron(III) chloride. 10 The chloroform solvent used in the reactions to this point contained 0 . 5 - 1 . 0% ethanol. The effect of this ethanol was investigated by comparing the product of an ethanol-free reaction with one containing 0. 75% ethanol. It was found that \vith ethanol present, a soluble fraction of 1 00% was achieved, but when the ethanol was removed, a soluble fraction o f only 6 1 % was obtained. Since iron(III) has its greatest affinity for ligands that coordinate via oxygen, 1 13 it is likely that iron-ethanol complexes form, impairing the iron(III) oxidation potential and slowing polymerisation to produce shorter oligomers. These results for the polymerisation o f an alkoxy terthiophene contradict the findings of Gallazzi et al.98 This suggests that either the effect of ethanol is different for the polymeris ation of alkyl-substituted terthiophenes or Gallazzi et al. had removed another impurity which was causing the production of insoluble polymer. Since both altering the addition sequence o f the reagents, and removing water and/or ethanol impurities resulted in a much lower soluble fraction, the products from these reactions were not further investigated, and the initial product mixture (S ection 5 . 4 . 2) was fractionated. 1 50 5.4.3 Polymer separation according to length As for polyOC6DASTT and polyC7DASTT, the crude polyOCI ODASTT was separated into fractions of different length o ligomers/polymers. The arnotmt extracted in each solvent is given as mass percentages of the total expected polymer yield in Table 5 .3 . The corresponding MALDI-TOF mass spectra for each solvent fraction are displayed in Fig. 5 .22. MALDI-TOF mass spectra of the polyOCIODASTT solvent extractions display very similar oligomer/polymer distributions to that obtained for polyOC6DASTT. The major difference in the two polymerisations is the higher amotmt of polyOCIODASTT extracted into chloroform (54% compared to 42%). As fotmd for polyOC6DASTT, an average estimated length of about n = 1 1 is obtained for the crude sample of polyOC)oDASTT. Although addition of a longer alko",-'Y chain increases the total yield of soluble polymer, it does not appear to significantly affect the degree of polymerisation. Table 5.3. Mass percentages ofOCIODASTT oligomerslpolymers separated by different solvents. Methanol (%) Hexane (%) Acetone (%) [)CM (%) o 1 3 4 26 CHCb (%) 54 Total soluble material 97% 1 5 1 3 lOo �1I1 J 3 5 1 4 Solvent: Hexane Yield: 1 3% Solvent: Acetone Yield: 4% Solvent: DCM Yield: 26% Solvent: Chloroform Yield: 54% 29 o 2000 4000 6000 8000 10000 1200 1400 16000 18000 20000 mlz Fig. 5.22. MALDI-TOF mass spectra of OC1oDASTT polymer fractions extracted using hexane followed by acetone, dichloromethane and chloroform. Signals are labeled "vith the assigned oligomer length in terms of mono mer units (n). Mr of mono mer: 662.4 g mOrl . 1 52 The UV -VIS-NIR spectra of the OCIODASTT oligomer/polymer fractions are displayed in Fig. 5 . 23, and an overlay of the spectra of the neutral materials in Fig. 5 . 24. As for the OC6DASTT polymer fractions (Fig. 5 . 8), a considerable increase in the intensity of the free carrier tail, and a significant bathochromic shift of the A.nax (1t-1t*) is observed with subsequent extractions. It is interesting to note that each of the fractions of polyOC I ODASTT consistently display a Amax (1t-1t*) at higher wavelengths than those measured for polyOC6DASTT (Fig. 5 . 8). This is consistent with observations by Casalbore-Miceli et al. 1 62 on the effect of increasing alkyl chain lengths, and may be due to better supramolecular ordering of polymer chains. Significantly, the major absorbance of the chloroform fraction of polyOC I ODASTT now appears to be due to the partially oxidised polymer (65 1 nm) rather than the neutral polymer (ca. 560 nm). This suggests an increasing ease of oxidation of polyOCIODASTT due to increased charge stabilisation by the longer alkoxy substituents compared to polyOC6DASTT. 1 53 , (a) 300 500 700 900 1 1 00 1 300 1 500 (b) , - - - - -""- - - � - - - � 300 500 700 900 1 1 00 1 300 1500 "\ (c) I , I \ I \ \ ..... .... - - - - - - -- 300 500 700 900 1 1 00 1 300 1500 .,... .... " (d) I \ I \ \ .... _-- - - - -... .... - -- 300 500 700 900 1 1 00 1 300 1 500 Wavelength I run Fig. 5.23. UV-VI S-NIR spectra of OC1oDASTT oligomer fractions separated by (a) hexane, (b) acetone, (c) OCM and (d) chloroform. Samples were measured in the oxidised state (solid line, oxidised using excess Cu(CI04�') and neutral state (dashed line) from solutions in chloroform. 1 54 300 500 700 Wavelength I nm 900 1 100 A...ax of 7t---+7t* transition - Chlorofonn 65 1 run -- OCM 541 run -- Acetone 5 16 run - - - - - . Hexane 502 run 1300 Fig. 5.24. UV -VI S-NIR spectra of QC lODASTT oligomer fractions which have been separated using hexane, acetone, OCM and chlorofonn. The Am.x of the 1t---+7t* transition for each fraction is listed. Cyclic voltarnmetry of fractions of different length OCIODASTI oligomers, which were drop-cast onto a GC electrode, are displayed in Fig. 5 . 2 5 . The oxidation onset potential is observed to decrease with each oligomer extraction (the acetone, DCM and chloroform fractions commence oxidation at 0. 28 V, 0. 1 9 V and 0 . 09 V respectively), consistent with an increasing oligomer length 1 14 as indicated by MALDI -TOF MS and UV -VIS-NIR spectroscopy. The reduction peak also shows a decrease in potential with increasing oligomer length: the acetone, DCM and chloroform fractions show peaks at 0.49 V, 0.20 V and -0. 1 6 V respectively. The lower oxidation and reduction potentials suggest that longer oligomers are more stable in the oxidised state. As each of the films consist of the same amount of material ( 1 5 f..lg mm·2) and covered the same area of electrode (7 mm2), the significant increase in current produced by the films as the polymer length increases may be caused by increasing conductivity of the films. 1 55 5 4 � 3 - 2 (a) acetone (nav - 4) - c:: G) ... ... :::J 0 0 -1 -2 -3 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 Potential I V vs AlJ/AIJ+ 80 60 � 4O - (b) DCM (nav - 7) - c:: G) 20 ... ... :::J 0 0 -20 -40 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 Potential I V vs PGlAg+ 120 80 � (c) Chloroform (nav - I S) - 40 -c:: G) ... ... :::J 0 0 -40 ..8Q -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 Potential I V vs PGlAg+ Fig. 5.25. Cyclic voltammetry of OCloDASTT oligomer fractions that have been cast onto a glassy carbon electrode (SA: 7 mm2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: - 1000/+800 mY. 10 cycles. Scan rate: l OO mV S· l . The average oligomer length as estimated from MADLI-TOF MS data is displayed in terms of monomer units (nav). 1 56 The reaction conditions shown in Fig. 5 . 1 8 were successfully used to polymerise OCIODASTT to produce soluble polyOCI ODASTT in gram quantities. This allowed the material to be tested for use in devices such as actuators, solar cells and batteries as described in Chapter 6 . 5.4.4 Conclusions OCIODASTT was successfully polymerised using iron(III) chloride to give a soluble fraction of 97% with an average polymer length of about n = 1 1 . Although no enhancement in polymerisation was achieved by the attachment of longer alkoxy chains to the terthiophene monomer (polyOCIODASTT and polyOC6DASTT show a similar range of oligomer/polymer lengths), the amount of soluble polymer was significantly increased by the attachment of longer alkoxy chains. 1 57 5.5 E lectrochemical polymerisation 5.5. 1 Introduction Polymerisation by cyclic voltammetry allows an in-situ investigation into the growth process of polythiophenes. Films of the alkyl and alkoxy styrylterthiophene derivatives were prepared by cyclic voltarrunetry, as well as deposited by potentiostatic (constant potential) methods to determine the difference in film growth by these techniques. Post-growth cyclic voltammetry of the resulting polymer films (in mono mer-free solution) was performed to investigate the redox processes. However, an extensive mechanistic study of the electrochemical growth and redox processes of the polymer films was beyond the scope of this study. A platinum micro electrode (surface a,rea: 1 0 �m2) was used for the electrochemical growth and analysis of the alkyl and alkoxy substituted derivatives. The small electroactive surface area of the micro electrode eliminates diffusion effects of the monomer species to the electrode, and minimises distortion of experimental data by ohmic potential drop and capacitance effects as described in Chapter 2. In addition, films were electrochemically deposited onto ITO-coated glass (surface area: � 1 cm2). Post-growth cyclic voltammetry of these films allowed comparison to the cyclic voltarrunetry of the cast films of the chemically oxidised materials (Sections 5 . 2 . 2, 5 . 3 . 2 and 5 . 4 . 2). The morphology of these films, in both the neutral and oxidised states, was investigated by scanning electron microscopy (SEM). For direct comparison between the polymerisation of STT and the alkyl and alkoxy (DASTT) derivatives, a growth CV of oligoSTT on a platinum micro electrode was first determined, and is shown in Fig. 5 . 26. The lower current compared to that for the polymerisation of STT on the platinum disc electrode (Fig. 3 . 1 6, Chapter 3 , Section 3 . 4 . 2) is due to the much smaller surface area of the micro electrode ( 1 0 �m2 compared to 1 . 8 mrn2). A higher oxidation onset potential of 0 . 74 V is also observed on the platinum micro electrode, compared to the platinum disc electrode ( 0 . 6 0 V). This indicates that nucleation and growth on the smaller surface area micro electrode is 1 58 more difficult, likely due to ready diffusion of insoluble oligomers away from the electrode surface by hemispherical diffusion, rather than their deposition. The increase in current with successive cycles indicates the growth of an electroactive film. The positive shift of the anodic peak as the film thickens during polymerisation may indicate that the film is not very conductive or that counter-ion mobility is hindered. 0.20 0. 1 5 �0. 10 - .. c � 0.05 a 0.00 -0.05 -0. 10 -0.5 -0.3 Forward scan • Reverse scan -0.1 0.1 0.3 �.5 Potential I V VS Agl Ag 0.7 0.9 Fig. 5.26. GrO\.vth CV of oligoSTT (5 mM) on a platinum microelectrode (SA: I D J.1II12). Electrolyte solution: 0. 1 M TBAPIl : I AN:DCM Potential limits: -500/+900 m V . 1 5 cycles. Scan rate: 100 m V · 1 S . 5.5.2 Electrochemical growth and post-growth analysis of polyC7DASTT The potentiodynamic growth of C7DASTT on a platinum micro electrode (Fig. 5 . 27) shows an increase in current with scan number, indicating growth of an electroactive film. A plot of the current produced at the upper vertex (0. 8 V) for each cycle (Fig. 5 . 28) displays an exponential increase. This indicates a very conductive film, and fast growth kinetics. Oxidation commences at 0 . 64 V. The lower oxidation onset potential when compared to that of STT (0. 74 V) may be due to stabilisation of the resulting radical cation species by the electron-donating effect of the alkyl substituents.8 This lower oxidation potential may be contributing to faster growth kinetics by the alkyl derivative, as 1 59 indicated by double the amount of current produced after fifteen cycles (0.4 IlA) compared to that generated by the growth of STT after the same number of cycles (0.2 IlA). On further cycling, the onset of this anodic peak, which is possibly due to monomer oxidation, shifts towards lower oxidation potentials (0.60 V on the 1 5th scan) suggesting that oxidation onto the polymer film is more readily achieved than onto the platinum surface. In addition, a small peak at 0.4 1 V emerges on subsequent scans, possibly due to the oxidation of the resulting polymer. Two reduction peaks are shown: a sharp peak at 0. 59 V and broader peak at 0.43 V. As discussed in Chapter 3 , Section 3 .4. 2, it has been suggested that the presence of multiple peaks during the growth and post-growth cyclic voltamrnetry of polythiophene derivatives may be due to transitions between the neutral, polaron, bipolaron and metallic states, which are affected by factors such as reduction of different areas of the polymer film147, different length polymer chains, 1 1 4 the effect of 'charge-trapping, 1 51 and conformational changes accompanying radical cation formation. 1 52 0.5 0.4 �0.3 - 1: 0.2 � ::J 0.1 0 0.0 '{).1 '{).2 0 0.1 0.2 0.3 0.4 0.5 0.6 Potential I V VS AgI Ag + 0.7 0.8 Fig. 5.27. Growth CV of C7DASTT (5 mM) on a platinwn microe1ectrode ( 10 f..LIIl2). Electrolyte solution: 0 . 1 M TBAPII : 1 AN:DCM. Potential limits: 0/+800 m V. 1 5 cycles. Scan rate: lOO m V S·I . 1 60 0.5 • 0.4 � • _ 0.3 1: • � :::s 0.2 0 • • 0.1 • • • • • • • • • • 0.0 0 2 4 6 8 10 1 2 1 4 1 6 Cycle number Fig. 5.28. Relationship bet .. veen the current produced during the growth of C7DASTT (measured at 0. 8 V) and the cycle number. The post-growth cycling of the resulting polyC7DASTT film in monomer-free solution is displayed in Fig. 5 . 29. One anodic peak and one broad cathodic peak are observed. The anodic scan of the first cycle is significantly different from subsequent scans, displaying a much larger peak current. As mentioned previously, possible causes of the different first cycle may include degradation of the polymer film, 72, 145 a change in the organisation of the polymer chains and/or morphology of the film affecting ion movement int% ut of the film, and further polymerisation of short oligomers. I I I , 1 71 The higher anodic current when compared to cathodic current further supports the inclusion of monomer or short oligomers in the film. I I I Solvation of short oligomers resulting in their removal from the film is also possible.99 Subsequent scans show a much smaller decrease in current of the anodic peak (indicating lesser changes to the film), and a shift of this peak to higher potentials. This may be due to a decrease in conductivity of the film, or to a change in morphology resulting in a decrease in counter-ion mobility. 147 A high oxidation onset potential for polyC7DASTT (0. 4 8 V) is observed. This is consistent with a high stability in the neutral state and correlates with the observation in Section 5 . 3 . 2 of the ease of reduction of the chemically polymerised C7DASTT. 16 1 0 . 12 0 . 10 �0.08 -'i 0.06 ... $ 0.04 0 0.02 0.00 -0.02 0 0.1 0.2 l ·t cycle Cycle nwnber 1 0.3 0.4 0.5 0.6 Potential I V VS Ag/Ag+ 0.7 0.8 Fig. 5. 29. Post gro'wth cycling of polyC7DASTT which has been deposited using cyclic voltammetry on a platinwn micro electrode (SA: 1 0 1ID1\ Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -500/+800 m V. 10 cycles. Scan rate: 1 00 m V S· I . The chronoamperogram of the growth of C7DASTT onto a platinum micro electrode is shown in Fig. 5 . 3 0. The potential used for the potentiostatic growth was chosen as 700 mY. Use of this potential resulted in polymerisation and film deposition. When the potential is stepped from 0 mV to 700 mY, a small, sharp transient is produced due to non-Faradaic charging of the electrode/solution interface. The current then increases steadily indicating an increase in electroactive surface area (by polymer deposition) and/or more facile electrode kinetics. The change in the current-time transient after approximately 1 0 seconds of growth o f C7DASTT may be due to a change in the deposition morphology. It has been shown by Pringle et al. that the growth of some polythiophenes initially results in a smooth compact film, but further grow1h may result in the deposition of material of porous morphology. 1 47 The increased surface area of the pol)wer-modified electrode caused by porous material (compared to the smooth film) may affect the rate at which more material is deposited and consequently the current produced. 1 62 O.�.--------------------------------------------------. 0.30 - 1: � 0.20 d 0. 1 0 O.OO+-�----�------�--------r-------�-------r------� o 5 10 15 Time I s 20 25 30 Fig. 5.30. Potentiostatic growth of C7DASTI (S mM) on a platinum micro electrode (SA: 1 0 J.lIIl2). Solvent: 1 : I AN: DC M. Potential held at 0 m V for I s, then stepped to 700 m V for 29 seconds. The post-growth CV (in monomer-free solution) of this film is displayed in Fig. 5 .3 1 . As observed for the potentiodynamically grO\W1 polyC7DASTT film (Fig. 5 . 29), the first scan is distinct to subsequent scans. These subsequent scans are significantly different to the post-growth CV of the potentiodynarnically gro\vn polyC7DASTT film in terms of the number, position and shape of the oxidation and reduction peaks. Subsequent scans show two oxidation peaks (0 . 5 8 and 0. 76 V) and two reduction peaks (0. 5 8 and 0. 4 1 V). As previously discussed (Chapter 3, Section 3 . 4 . 2), the presence of multiple oxidation peaks may be due to transitions between the neutral, polaron, bipolaron and metallic states. An attempt was made to polymerise C7DASTT onto an ITO-coated glass slide (- 1 cm2) using both cyclic voltarnmetry and potentiostatic methods. Although small amounts of black material were produced (assumed to be oxidised polyC7DASTT), the material did not adhere to the electrode. Consequentially, no further studies were perfonned on this material. 1 63 0.6 � 0.4 - C 0.2 � d 0.0 -0.2 -0.4 0 0.1 0.2 0.3 0.4 0.5 0.6 Potential I V VS Ag/Ag+ 0.7 0.8 Fig. 5.31. Post growth cycling of polyC7DASTT, which has been deposited potentiostatically on a platinum microelectrode (SA: 1 0 1lll2). Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: 01+800 m V. 1 0 cycles. Scan rate: 1 00 m V S·l . 5.5.3 Electrochemical growth and post-growth analysis of polyOC6DASTT The growth CV of OC6DASTT is in Fig. 5 .32. The decrease in oxidation onset potential (0.6 1 V) compared to both STT and C7DASTT (0. 74 and 0.64 V respectively) may be due to enhanced stabilisation of the resulting monomer cation radicals caused by the strong electron-donating effect of the alkoxy substituents. This CV shows a significant increase in current with each scan indicating rapid growth of an electroactive polymer film. A maximum current of 2 .3 IlA is generated after 1 5 cycles, which is more than 5 times higher than produced during the growth of C7DASTT (0.4 IlA) and 1 0 times higher than that produced during the growth of STT (0.2 IlA) after the same number of cycles. This indicates very fast growth kinetics, and may be a result, in part, of the reduced oxidation onset potential. After the first cycle, an oxidation peak is produced at ca. -0. 06 V (Eo><.I )' which increases to 0 .05 V after 1 5 cycles. This is followed by a broad oxidation peak at 1 64 ca. 0 .2 V (£0><,2)" and another peak at 0. 65 V (£0><,3). A broad reduction band is generated with peaks at 0 .44 V (£red.l ) and 0. 1 4 V (£red.2). The post-growth CV of this film in mono mer-free solution (Fig. 5. 33) displays similar peaks to the growth CV in terms of shape and position, but with the loss of the oxidation peak, £0><.,3 , which is probably due to oxidation of the mono mer species. The other redox peaks, including small peaks appearing at about 0 .55 V on oxidation (£0><,4) and -0. 5 V on reduction (£red.3), are most likely due to transitions between the neutral, polaron, bipolaron and metallic states of the deposited polymeric material. The oxidation onset potential ofthe polymeric material (0. 1 1 V), is much lower than that of both the potentiodynarnically deposited STT and C7DASTT derivatives (0. 30 and 0.44 V respectively). A decrease in oxidation potential of polythiophenes by attachment of alkoxy substituents has been reported by Girrotto et al. III and may be due to a stabilisation of the oxidised species and/or a result of an increase in oligomer length. 1 14 The low oxidation potential of electropolymerised polyOC6DASTT is also consistent with the apparent stability of this material in the oxidised state, as observed by the chemically polymerised materials (Section 5.4 .2). 2.5 roO;; ..;;===============::::;------------�:___:_-___, 2.0 �O .. i 0 02 v 1 .5 � ooot--=-'---� . 002l--__ ��-_--.--J _ 1 .0 ·05 C ! 0.5 ... :::J 0 0.01-__ ._ -0.5 -1 .0 -1 .5 +-----r-----..---------,r--------.---�---____r---' -0.5 '{).3 -0. 1 0.1 0.3 + 0.5 Potential I V VS AglAg 0.7 Increasing scan number Fig. 5.32. Growth CV of OCJ)ASTT (5 mM) of a platinum micro electrode (SA 1 0 1lITl2). Electrolyte solution: 0. 1 M TBAPIl : 1 AN:DCM. Potential limits: -500/+800 m V. Scan rate: 100 mV S·l The first scan is shown as an inset. 1 65 2.0,-----------------------------, 1 .5 � 1 .0 - 1: 0.5 � 6 0.0 l--IIIIC:::::::'--- -0.5 -1 .0 Ered.2 -1 .5 +-----r-----..-------,..-------,------y------.----' -0.5 -0.3 -0.1 0.1 0.3 + 0.5 Potential I V VS AglAg 0.7 Fig. 5.33. Post grm.vth cycling ofpolyOC6DASTI which has been deposited using cyclic voltammetry on a platinum microelectrode (SA 10 j.1IIl2). Supporting electrolyte: 0 . 1 M TBAP/AN. Potential limits: -5001+800 ID V. 10 cycles. Scan rate: 100 ID V S· l . Small peaks may be present at about 0.55 V on oxidation (Eo><,4) and -0.5 V on reduction (Ered,3), The chronoamperogram of the potentiostatic growth of polyOC6DASTT is displayed in Fig. 5 . 34. A potential of 0.8 V was chosen for polymerisation, as this potential generates polymerisation, and does not appear to cause overoxidation. As for the potentiostatic growth of polyC7DASTT, a small, sharp current transient is initially observed due to non-Faradaic charging of the electrode/solution interface, and is followed by a steadily increasing current with time. The change in the current/time transient at ca. 1 6 seconds may be due to deposition of polymer with a different (possibly more porous) morphology, as discussed for the growth ofpolyC7DASTT. The post-growth CV of this potentiostatically grown film is displayed in Fig. 5 . 35, and is very similar to the post-growth CV of the potentiodynarnically deposited material (Fig. 5 . 33), in terms of shape and position of the peaks. The major difference is a smaller Eox•2 shoulder, and a more prominent Ered,4 peak. 1 66 0.4._---------------------------------------------------. 0.3 - - i 0.2 t::: :::J (J 0.1 O.O+--L-,-----r----._---,r----r----.----,r----r----,---� o 2 4 6 8 10 12 Time I S 14 1 6 18 20 Fig. 5.34. Potentiostatic grO\,vth ofOC6DASTT (5 mM) on a platimnn microelectrode (SA: 10 J..Ull2). Electrolyte solution: 0. 1 M TBAPII : 1 AN:DCM. Potential held at -500 mV for 1 s, then stepped to 800 mV for 1 9 s. 0. 15 .----------------------------------------------------, Eo"-l �.10+_------._------._------._------._------._------._� �.5 �.3 �. 1 0.1 0.3 0.5 Potential I V VS AQllvl 0.7 Fig. 5.35. Post growth cycling of polyOC6DASTT ... vhich has been deposited potentiostatica11y on a platinum microelectrode (SA: 1 0 J..Ull2). . Supporting electrolyte: 0. 1 M TBAP/AN. Potential limits: -500/+800 m V. 1 0 cycles. Scan rate: 1 00 m V S-1 . Films of polyOC6DASTT were also grown onto ITO-coated glass for SEM analysis and to obtain enough material for MALDI-TOF mass spectrometry. The growth CV is shown in Fig. 5 . 36, with an oxidation onset potential of 0 .67 V. This higher oxidation 1 67 onset potential when compared to the film grown on the platinum micro electrode (0. 6 1 V) suggests that nucleation onto the !TO surface is more difficult. In addition, the shape of the CV is significantly different to that generated by growth on the platinum microelectrode, possibly due to inhibition of the growth by formation of a monomer-depleted area around the large electrode. The post-growth CV this film (Fig. 5 .37) displays one broad oxidation peak at 0.30 V and one broad reduction peak at 0.04 V. The peaks are not quite as broad as observed in the CV generated by the cast film of chemically oxidised polyOCIODASTT (Fig. 5 .6), possibly due to a narrower polymer polydispersity, 1 59 or a more regular film morphology. 1 .2 1 .0 0.8 �0.6 - 1: 0.4 � :::J 0.2 0 0.0 -0.2 -0.4 -0.6 -0.5 -0.3 -0.1 0.1 0.3 0.5 Potential I V VS Ag/ Ag + 0.7 Fig. 5.36. Growth CV of OC6DASTI (5 mM) on an ITO-coated glass electrode (SA: -I cm2). Electrolyte solution: 0. 1 M TBAPIl : I AN:DCM. Potential limits: -500/+800 m V. 5 cycles. Scan rate: 1 00 mVs·1 1 68 0.8 0 .6 �0.4 - 1: 0.2 e ... ;:, 0.0 0 - 1 MQ cm-I ). 6.3.4 Fibres Spinks et al. 1 9 1 demonstrated that polypyrrole actuators in the form of hollow polymer fibres, have better strain and strain-rate properties than flat films, attributed to a higher electrical conductivity of the tubes. Accordingly, an attempt was made to construct three different types of fibre structures coated with polyOCIODASTT. Schematics of the fibre structures prior to coating are shown in Fig. 6. 1 S. Fig. 6. 1 S(a) displays a schematic of a hollow PVDF fibre wrapped with a SO Ilm platinum wire. The purpose of the wire is to increase the electrical connectivity through the entire length of the polymer tube when it is in the low conducting, neutral state to increase the rate of actuation. Platinum coating of the PVDF fibre (Fig. 6. I Sb) was also intended to increase the charge distribution but unfortunately the polymer did not adhere well to the platinum coating and peeled off when the thickness was increased. As the PVDF membrane is not very pliable and restricts expansion and contraction, it was attempted to create a hollow polymer fibre wrapped in a platinum coil without the PVDF membrane. With polypyrrole, this is normally achieved by electrochemically growing a thick film onto a thick (2S0 f..1m diameter) platinum wire that is coiled in a fine platinum wire (Fig. 6. 1Sc). The inner wire is then removed to leave a hollow polymer fibre incorporating a platinum coil. Unfortunately, poor adhesion of the polymer to the platinum wire meant the coating could not be made thick enough by casting or electrochemical deposition methods to produce a free-standing polymer fibre. 2 1 5 b Hollow PVDF membrane fibre 50 I.lID Pt wire coil with Pt coatink rcccct:cC ' \ C rcccctcc Fig. 6. 15. Schematic of fibres on which POlyOCloDASTT was deposited. (a) HollowPVDF membrane fibre wrapped with a 50 J.1IYl platimnn wire. (b) Hollow PVDF membrane fibre which has been sputter­ coated with platinum and wrapped with a 50 JlIll platinum wire. (c) 250 J.1IYl wire "\Tapped in a 50 J.1IYl wue. Since polyOCIODASTT could be successfully deposited only on the PVDF fibre coiled in platinum (Fig. 6. 1 Sa), this type of fibre was selected for further experimentation. Isotonic electromechanical measurements of a fibre that was 49 mm long and supported a polymer coating of 39 Ilg mm·1 length of fibre, are given in Fig. 6. 1 6. The force stretching the fibre (stress) was held constant while the potential was alternated between -0.6 V and +0. 8 V. The current, and the distance the fibre stretched/contracted (displacement) were measured. In this study, a more negative value for displacement is used to indicate that the fibre contracted and a less negative value indicates it expanded. Total expansion of the sample between the oxidised and neutral state was 10 .8 Ilm, giving a strain of 0.0002%. 2 16 8 Z E 7 - � 6 0 U- 5 4 :> 0.8 h is 0.3 � .Q.2 ! Q. .Q.7 -1 .2 25 E ::1. 20 - .. c � 15 Cl) 10 � Q. 5 .!!l c 0 30 � 10 E - 1: -10 I! ... -30 :::J 0 -50 0 50 100 150 200 250 300 350 Time I s Fig. 6. 16. Actuation of a hollow PVDF membrane fibre (49 mm long) wrapped in a 50 J.lIIl wire and coated with polymer (39 � mm-\ The force \-vas held constant at 6 mN while the potential was alternated between -0.6 V and +0.8 V. The distance the fibre stretches and the current produced were measured. Fig. 6. 1 7 shows the relationship between the charge injected into the sample and the change in length of the sample (displacement). Reduction and contraction are represented by negative values, and oxidation and expansion by positive values. Although an ideal actuation system would display a linear relationship to indicate efficient power consumption, in this case, the response is clearly not linear. The non­ linear relationship may be due to an impedance of the expansion and contraction by the non-pliable PVDF membrane substrate. 2 1 7 -16�--------------------------------------------� (a) Reduction process §. -12 - C Cl) E -8 8 Cl Q. en -4 is 0 0 • • •• • • -20 -40 ____ 'flJRllldl eR �o -80 Charge / mC -100 1 6�--------------------------------------------� §. 1 2 - 1: Cl) E 8 8 Cl Q. en is 4 .... V-wlUW.�·. /' • I • • • (b) Oxidation process O+---------r-------�--------_r--------,_------� o 20 40 60 Charge / mC 80 100 Fig. 6. 1 7. Relationship between displacement and charge injected into a PVDF membrane fibre coated with POlyOC lODASTI during (a) reduction and (b) oxidation of one typical pulse. A negative charge is used to indicate reduction, a negative displacement to indicate contraction , a positive charge to indicate oxidation and positive displacement to indicate expansion. Fig. 6. 1 8 shows the displacement with time that is displayed by the fibre during oxidation and reduction. The slope represents the rate at which the fibre is expanding or contracting. This diagram reveals that the rate at which the fibre is contracting on reduction is much quicker than the rate of expansion on oxidation. Again, this is most likely due to the higher conductivity of the polymer in the oxidised state. 2 1 8 14 12 �I"'!"'-��""'IIIV"\;PI�_ Oxidation E 1 0 ::1. Reduction - C 8 CD i 6 U Cl • Q. · • fIJ 4 · C 2 0 0 1 0 20 30 40 50 60 Time I s Fig. 6. 18. Comparison of the rate of contraction/expansion during oxidation and reduction of one typical pulse. The slope represents the rate at which the fibre is expanding or contracting. In comparison to polypyrrole actuators that typically display strains of 3-30% and strain-rates up to 3.2% s-' , this polyOClODASTT/PVDF fibre displays a very low strain (0.0002%) and strain-rate (ca. 0 .00001 % sol ) . 6.3.5 Free-standing film with incorporated wire The non-linear relationship between time and displacement shown by the PVDF membrane fibre suggests that the actuation performance is hindered by the relatively inflexible PVDF membrane. To prevent this problem, a free-standing polymer film was cast and tested. A 50 Jlm wire was zigzagged through the film in an attempt to increase the actuation response by improving the accessibility of charge to the polymer when in its semi-conducting undoped state « 1 . 1 x 1 0-4 S cm-I ) as shown in Fig. 6. 1 9. The conductivity of this film when doped with iodine was l . 1 S cm-' , i .e. a 1 0000 fold change. 2 1 9 t 6 mm P01yOClODASTI film P1atimnn wire (SO 1JIIl) Fig. 6. 19. Fr�standing film of po1yOClODASTT incorporating a zigzagged (SO IJIIl diameter) wire. Dimensions of film: 6 mm x 2 1 mm with a thickness of 8 1 1JIIl. Mass = 16 . 1 mg. The film was tested isotonically using the dual-mode lever system with a constant tensile stress of 44 KPa (40 mN tensile loading). Before perfonning electromechanical measurements, the film was pre-conditioned. The constant stretching force (40 mN) was applied to the film in order to keep it straight, and 4 hour square wave potential function which pulsed between +1 V and -0. 6 V (vs Ag/Ag+) was applied from the electrical contact at the bottom of the film. This process was conducted for 24 hours until consistent displacement and current measurements were obtained. After pre-conditioning, the film was oxidised and reduced under isotonic conditions to and from various oxidation potentials as shown in Fig. 6. 20. In each case, the initiallfinal reduction potential was -0.6 V. As the potential was switched from a reducing potential to an oxidising potential, a rapid initial strain-rate (rate of expansion) was observed as the polymer became positively charged and anions were inserted into the polymer matrix. This initially relatively rapid expansion was followed by a slower rate of expansion as the film approached a neutral state, possibly due to further chain relaxation. A similar process seemed to happen on reduction as the change in potential of the polymer forced out anions to give a relatively fast initial contraction. This was again followed by a slower rate of contraction as the film approached a neutral state. An exception is the oxidation at +0.2 V, which shows an initial slight contraction on application of this potential, possibly due to an initial reduction of the film which may have been partially oxidised. 220 4 3 � 0 2 - ( c: 'ii � en I V- 0 -1 0 -+{j.8 V -+{j.6 V -+{j.4 V ,i 2 +1 V -+{j.2 v 3 4 Time I h - 5 6 7 8 Fig. 6.20. Strain created by the po1yOC\ODASTT film over time as it is doped (oxidised) and then dedoped (reduced at -0.6 V) to and from various oxidation potentia1s. The maxunum strain rates for the oxidation and reduction of films at different oxidation potentials are given in T able 6.4. The strain rate was observed to increase with applied oxidation potential. This was expected as the higher potential differences at the int erface between the film and solution should induce faster anion absorption/expulsion and therefore expansion/contraction of the film. The strain-rate of the film was measured to be about three times higher during reduction than during expansion. This may be due to two possible reasons. The low conductivity of the film in the neutral state may be impeding the accessibility of charge thro ugh the film, thus hindering o xidation and expansion of the film. A second possibility is a difference in anion accessibility int o the neutral and oxidised polymer matrix. A higher strain rate during oxidation is often reported in literaturel88,199 due to a higher energy requirement to open the molecular entanglement and allow the penetration of cotmterions. During reduction, however, counterions are able to diffuse along the opened structure to the solution without any resistance, giving rise to a faster contraction behaviour. 221 Table 6.4. Maximum strain rate on oxidation and reduction of fihns which are oxidised at various potentials. The reduction potential is kept constant at -0.6 V. Upper potentia1 limit +0.2 V +0.4 V +0.6 V +0.8 V + 1 .0 V Max.imwn strain rate on oxidation (% S-l ) 0.001 0.00 1 3 0.0040 0.0033 0.0042 Maximum strain rate on N/A 0.0023 0 .0056 0.010 0.01 1 Reduction (% S-l) A linear correlation is revealed between the oxidation potential and the strain measured after 1 .4 hours of oxidation (Fig. 6.2 1 ). This can be explained by a higher level of doping (and hence film expansion) at more positive oxidation potentials. The doping level is measured by the charge passed, and is used to calculated the electrochemical efficiency. The increase in strain generated by the film with the increase in electrochemical efficiency and charge passed is shown in Fig. 6 .22 . The maximum strain obtained from the polyOClODASTT film (at an oxidation state of + 1 V after 4.2 hours as determined by Fig. 6 .22) is about 3 .2%. This measurement is twice that displayed by another example of solution process able ICP actuator film based on a polyaniline film plasticized with di-2-butoxy-ethyl ester of sulfosuccinic acid (about 1 . 5% at 0 tensile loading). 200 At a constant tensile stress of 44 KPa, reversible strain was observed in the potential range from + 1 V to 0.6 V without significant creeping during a 48 hour experimental period. One possibility for this result is favourable interactions between the polymer substituents (involving either interdigitation of the alkoxy substituents, or 1t-1t interactions between the styryl substituents), which may prevent the polymer chains from slipping and causing irreversible chain relaxation during the electrochemical doping process. 222 3.5.------------------------------------------------------, 3.0 '#. - 2.5 � .s:. 2.0 � - 1U 1 .5 c 'i 1 .0 "- en 0.5 O.O+-------�--------�--------r--------.--------.-------� 155 100 165 170 Potential I mV 175 180 185 Fig, 6. 21. The effect of the oxidation potential on the strain measured after 5000 seconds o f oxidation. Charge I me mg-1 0 20 40 60 80 100 120 3.0 • .{).6 V to +1 .0 V ",I W" 2.5 o .{).6 V to +0.8 V . ' • .{).6 V to 0.6 V •• o .{).6 V to 0.4 V • • '#. 2.0 /-.. -c 'd 1 .5 ... / . . 0 1 .0 • • 0.5 • 0.0 0 10 20 30 40 50 60 70 80 EE / % Fig. 6.22 Relationship between the strain generated at different oxidation potentials and the electrochemical efficiency (EE). The electromechanical tensile performance was evaluated under isometric conditions. The polymer film was oxidised and pre-conditioned at + 1 V to allow full expansion before the lever arm was fixed. The force applied on the lever by the film due to contraction was then measured as the film was reduced from the oxidised state to the 223 neutral state (+ 1 V to -0. 6 V). The stress generated by the film in relation to the charge density passed into the film was calculated and is shown in Fig. 6.23. The contraction force (stress) is observed to increase almost linearly to the reduction charge, as expected if the charge is balanced by removal of anions rather than insertion of cations. Although the polyOCIODASTT film broke at I I I KPa, extrapolation of this linear trend line to the maximum redox charge density at 1 05 mC mg-1 yields a tensile stress of about 260 KPa. This lies in the range shown by natural skeletal muscles, which usually display tensile stresses of between 0. 1 MPa and 0. 5 MPa. 1 90 However, failure of the film at I I I KPa reveals that the largest tensile loading must be less than this value. Further structural strengthening is required to make a tougher film with a higher tensile modulus (ability to elongate) when submerged in an electrolytic medium. - �.---------------------------------------------------, 160 - - -.. - - - -. - - -. -- - - -. 40 o -10 - -.. - -.-.. -. - -. - -.- -'" -.. -20 Charge I mC mg-1 - - -- - - -.-... . . . • -30 - - - • • - - - • -40 Fig. 6.23. Relationship between the isometric stress generated on the fllm and the charge density passed as the film is reduced. The film was pre-conditioned at + 1 V to obtain the fully expanded state and then isometric measurements were performed as the potential was switched to -D.6 V. 6.3.6 Conclusions PVDF membrane benders and fibres, as well as a stand-alone film have been prepared by solution casting of polyOCIODASTT. These structures have been shown to possess electromechanical properties by undergoing a reversible volume change in response to a repetitive potential pulse in an electrolytic media. 224 Although the PVDF substrate appeared to hinder the expansion/contraction of the polyOCIODASTTIPVDF fibre giving a very low strain, it was shown that the stand alone polyOCI ODASTT film actuator could achieve a strain of up to 3 .2% under a stress level of 44 kPa. However, a relatively low strain rate of about 0.0 1 1 % S-I was observed for this film. This modest result may be due to the electrolyte struggling to penetrate the film and/or poor accessibility of charge due to the low conductivity of the material. Although this strain rate may be improved by using a metallic substrate, casting or electrochernical growth ofa polyOC1oDASTT film of suitable thickness onto a platinum surface proved difficult. Although some aspects, including the electrochernical efficiencies, mechanical toughness and long-term durability of polymer actuators still need to be improved, the versatility of polythiophene materials, created by easy functionalisation of the polymer backbone, provides possibilities for overcoming these problems. Although PVDF membrane was suitable as a substrate for the preliminary investigations discussed in this report, further work would require substrates that show a higher elasticity as the PVDF membrane does not appear to allow full contraction/expansion of the polymer. 225 6.4 Batteries 6.4. 1 Introduction Batteries incorporating polymeric materials are potentially a form of energy storage, which is low in weight, flexible and environmentally safe. Although polymer batteries currently show modest voltages compared to metal-based secondary batteries, they have high specific capacities due to possible fabrication of high surface area electrodes. Gofer et al. 20 1 reported a poly thiophene based battery, where both cathode and anode were made from fluoro-substituted phenyl thiophene polymers, which produced a discharge voltage of about 2.4 V and capacities of 9. 5 to 1 1 .5 mAh g-l Batteries consisting of these materials were reported by Ferraris et al. to exhibit a cycle efficiency (ratio of the recovered charge:injected charge) of96±1 %. 86 An all-polymer battery reported by Rehan, 202 where the anode and cathode consisted of poly-l -naphtol and polyaniline respectively, was reported to have a specific capacity of 150 mAh g-1 with an open-circuit potential of 1 .4 V . However, since the reported specific capacity is higher than the theoretical capacity for the two electrodes ( 1 08 mAh g-l , calculation shown below, the theoretical capacities of the polyaniline and poly(l -napthol) electrodes were reported as 298 and 1 7 1 Ab kg-1 respectively), there must be an error with their Ab or mass measurements, or additional reactions are taking place possibly involving impurities. This paper also contains an error in the calculation for the theoretical capacity of poly( l -napthanol). From Fig. 2 of the paper, which shows one electron transferring for every two napthanol units (280 g mor\ a capacity for polynapthol of 96 mAh g-1 is calculated (using the equation for calculating theoretical charge given in Section 6. 3 .2 and dividing by 3600 to convert units to hours) rather than 1 7 1 mAh g-1 reported by Rehan for this material. Using this corrected value of 96 mAh g-1 for the capacity of poly( l -napthanol), an even lower value of 74 mAh g-1 is calculated for the specific capacity. 226 Theoretical capacity of two battery electrodes Theoretical capacity of first electrode + Theoretical capacity of second electrode The development of electrochemical devices for applications involving charge storage usually requires polymers with a high doping level and good redox reversibility. 28 Many polymers including polypyrrole and polyacetylene203 have been found to perform as promising cathodes. However, a suitable material for an organic anode has not been reported. In this study, polyOCIODASTT was investigated for use as an anode-active material. The low HOMO level shown by polyOCIODASTT (oxidation onset potential of approximately -0.34 V, Chapter 5, Section 5 .4 .2), and the high redox stability of this material, indicated that polyOCIODASTT might be a useful material as an anode-active material. Chemically polymerised polyOC 1oDASTT was cast onto two different substrates for comparison: carbon fibre mat (carbon fibre substrate) and Ni/Cu coated nonwoven polyester (Ni/Cu substrate). These conductive materials were chosen for their high surface area and the ability of the polymer to adhere to the fibres. Polypyrrole electrodeposited on stainless steel mesh was employed as cathodes. 6.4.2 Cell construction, testing procedures and terminology This work was carried out in collaboration with Dr. Caiyun Wang at IPRI, University of Wo lion gong using their well established battery testing facilities. A schematic of the test cell used in this study is given in Fig. 6 .24. The anode consists of polyOCIODASTT, which was drop-cast onto either carbon fibre mat or Ni/Cu­ coated nonwoven polyester fabric to give a polymer coating of about I mg cm-2. The cathode was prepared by electrodeposition of polypyrrole onto stainless steel mesh and is separated from the anode by a microporous polypropylene membrane. The 227 electrodes were submerged ill a LiPF6 ( 1 . 0 M), 1 : 1 ethylene carbonate:dimethylcarbonate solution, and are kept adjacent through use of a spring. The cells were designed to be anode-limited (the theoretical capacity of the cathode was 6 .7 times that of the anode) to allow the features of this electrode to be investigated since polyOCIODASTT is the material of interest. ___ E-- - Metal lid '(1) 4 0 � c( U c CG 0 3 5 a. ... CG O 0 0 3 0 Q) � Cl O � a. 2 5 ,c o;- 2 0 U Cl .!! ,c C c( 1 5 E 1 0 5 0 � � � � I- t- - - � 0 PolyOC lODASTI on Ni/Cu substrate PolyOClODASTT on carbon fibre substrate •••••••••••••••• •••••••••••••••••••••••••••••••••• 1 0 2 0 3 0 4 0 5 0 Cycle number Fig 6.29. Variation of discharge capacity with cycle number for cells with an anode comprising polyOClODASTI on Ni/Cu substrate and carbon fibre substrate. Current density: 0.02 mA cm-2 6.4.6 Conclusions The good discharge capacity and excellent cycle life shown by the Teflon batteries described in this study, reveals potential for polyOC1 oDASTT as an active anode material for all-polymer batteries. The substrate that the polyOCIODASTT was cast onto was shown to have a significant effect upon the performance of the cell. Ni/Cu coated nonwoven polyester fabric was found to produce significantly improved discharge capacity compared to anodes comprising carbon fibre mat as a substrate. A high coulometric efficiency of >94% was generated by both types of cells with anodes employing polyOCIODASTT. 234 6.5 M icro- and nano-structured surfaces 6.5. 1 Introduction In this section, it is described how polyOCIODASTI may be processed into a variety of structures with a high surface area. High surface areas are desirable in some applications, such as solar cells, molecular sensors and batteries. This is because they provide a high interfacial area for reactions to occur (for example charge separation at solar cell electrode interfaces or interactions of the polymer with atmospheric molecules), and hence allow the production of more efficient devices. Since most conducting polymers are insoluble, high surface area structuring is often achieved by electrochemical growth and deposition onto a template. 204 For example, Misoska204 prepared a polypyrrole inverse-opal by electrochemical deposition through the void spaces in polystyrene synthetic opal crystal on an electrode. One limitation of this technique is that an electroactive surface is required for electrochemical growth. The solubility of the materials produced in this study allows them to be processed by techniques such as casting, spin-coating and electrospinning. This means they can be deposited onto a variety of conductive and non-conductive substrates. In this study, polyOCIODASTT has been cast onto templates, which are then removed to produce fibrillar, opal and inverse-opal polymer structures. This work was undertaken in the IPRI at the University ofWollongong with the assistance of Dr. Violeta Misoska. In addition, polyOCIODASTI fibres have been produced by electrospinning. 235 6.5.2 Fibrils Fibrils are fine, thread-like structures. Their small diameter (typically <500 run) means they have a very high specific surface area. Fibrils of conducting polymers, such as polypyrrole, can be made by electrochemical deposition onto a porous alurnina template which has one side sputter coated with platinum. 205 The polymer grows from the platinum electrode through the pores in the template. The template is then dissolved to leave polymer fibrils adhered to the platinum film. In this study, however, fibrils have been prepared by casting dissolved OCIODASTT polymer directly onto the template without any platinum backing. The advantage of casting the polymer, rather than by electrochemical growth, is that it is a much simpler technique, and eliminates the possibility of mono mer or short oligomers becoming merely trapped in the template pores and not contributing to a polymeric structure. The template used in this study was composed of alumina (Anodic Aluminium Oxide, AAO), and had honeycomb shaped pores 200 run in diameter. The template was soaked in a solution of polymer (about 2 mg mL-' ) for one hour. The template containing the fibrils was then removed from the solution and drip-dried. Templates are typically dissolved by soaking in I M NaOH for 20 minutes. 206,207 However, electrochemical analysis of a poly(OCIODASTT) cast film, which had been exposed to these conditions, showed that the film was severely damaged. Soaking in 0 .5 M NaOH for 50 minutes at room temperature, however, completely dissolved the template while having no effect on the electrochemical properties of the polymer. The template was either fully dissolved (by soaking for 50 minutes) to leave a very fragile film of polymer fibrils, or partially dissolved (by soaking for 25 minutes) so the remaining template structure could operate as support. To further increase the robustness of the fragile fibrils, templates could be sputter-coated on one side with platinum prior to soaking in polymer. The resulting conducting platinum film (with fibrils attached) could be used as an electrode to investigate the properties of the polymer fibrils. 236 SEM images (Fig. 6 .30) show polyOCJQDASTT fibrils, held together by (a) a thin polymer film and (b/c) a platinum film and partially remaining template as support. The fibrils appear to be more structured near the platinum-coated side, and tangled at the unsupported end. They appear to be about 200 nm in diameter, which is expected from the 200 nm pore size in the template. Polymer film ----:;*"''''' Pt film Fibrils Fig. 6.30. SEM images showing polyOC1oDASTT fibrils with a platimun fIlm as a support. (a) Partially dissolved template and viewed from platinum coated side, and (b) cross-section of the fIbrils with a polymer film on left and platinum film on the right. 237 By drying the fibrils on ITO-coated glass (with the platinum side contacting the ITO), a satisfactory contact could be made. Electrochemical analysis of fibrils on ITO-coated glass is given in Fig. 6. 3 1 . The polymer fibrils appear to be very stable, showing negligible change in current on cycling. -.. c 3,----------------------------------------. 2 � 0)e======:::::------ u -1 -2 -3�----�----�----�----�------�----�� �.5 �.3 �. 1 0. 1 0.3 0.5 Potential I V vs PtJIPtJ+ 0.7 Fig. 6.31. CV of fibrils in a partially dissolved template with a platinum backing stuck on ITO coated glass. Supporting electrolyte: 0 . 1 M TBAP/AN. Scan rate: lOO mV S-l In summary, fibrils of polyOClODASTT were successfully made by using an alumina template. The fibrils appeared to be stable during repetitive oxidation, and reduction back to a neutral state. 6.5.3 Inverse opal and opal structures 6.5.3 . 1 Introduction Films of synthetic opal crystals can be made by sedimentation of colloidal polystyrene (0.2 - 1 �m diameter) onto a substrate. This results in a highly ordered, self-assembled beaded structure with a high surface area Using this template, opal (bead morphology) and inverse opal (honeycomb morphology) polymer structures can be made. The opal template can be removed by either soaking in an organic solvent such as acetone, or heating in a furnace (450°C). As both these methods may damage a fine layer of polyOClODASTT, platinum opal and inverse opal structures were made. After removal of the polystyrene, the remaining platinum structure was coated with polymer. 238 The polystyrene opal was deposited onto ITO-coated glass by evaporation of a emulsion droplet, containing 200 nm or 920 nm polystyrene micro spheres in water, at room temperature. Films were also made by spin-coating methods using the same polystyrene emulsion. Although drop-casting resulted in uneven films (even when using gaskets), the films generally resulted in a high degree of ordering. Spin-coating of the polystyrene bead solution onto ITO-coated glass allowed the production of thin films with a larger area than cast films (�2 cm2), but with an inferior degree of ordering. 6.5.3.2 Platinum inverse o pals (ho neycomb structure). Platinum opals are made by electrochemical deposition of platinum on the ITO-coated glass through the spaces between the opal beads. The opal is then removed (by heating in a furnace at 450°C) to leave a platinum honey-comb inverse-opal structure as shown by SEM in Fig. 6.32. Fig. 6.32. Platinum inverse opals showing the honey comb structure. (a) x7500 and (b) x3000 magnification. 6.5.3.3 Platinum, gold and ITO o pal s (bead structure). The polystyrene opal (cast onto ITO-coated glass) was sputter-coated with platinum, to give platinum films which retained the iridescent sheen of the tmderlying opal surface. These films were then heated in a furnace 450°C for 2 hours in an attempt to remove the polystyrene support. SEM images of the resulting platinum opal structures are shown in Figs. 6. 33. The metal coating appears to have completely encapsulated 239 the polystyrene beads, retaining the bead-like morphology and forming a platinum opal structure. This explains the enduring opalescent appearance of these films. Samples showed close-packed (hexagonal) and/or square-packed structuring. Although these materials were heated in a furnace at 450°C to burn out the polystyrene (melting point: �240°C), it is lIDclear whether the polystyrene was completely removed. The platinum opal films (after heating in the furnace) were removed from the glass slides using adhesive tape. Inspection of the under-side (glass side) of the opal by AFM reveals 'pits', which appear to be left after the polystyrene has been removed (Fig 6. 34). The six prominent protrusions observed arolIDd each pit is most likely due to filling of the spaces left between opals colloids that are hexagonally ( close) packed. The pits indicate that the polystyrene colloids may have been removed. The platinum opal films were coated with polymer by drop-casting from a chloroform solution. However, poor adhesion of the polymer to the platinum surface created a very uneven film. An attempt was made to improve the quality of the film by synthesising gold and ITO sputter-coated opals. These opals were produced in the same way that the platinum opals were made. SEM images of the resulting gold and ITO opals (after heating in the furnace at 450°C) are displayed in Figs. 6 .35 and 6. 36 respectively. The gold opal structure (Fig. 6. 35) shows uneven bead sizes and non­ spherical shapes, possibly due to partial collapsing of the beads by removal of the polystyrene support, and is further evidence that the polystyrene has been removed. As for the platinum opal film, it was fOlIDd that polyOCIODASTT did not adhere well to these ITO and gold opal structures. 240 • _ _ a., 4IIt'W· It- an,.. I;- -. - '- e - •• " ... . , " •• _· ... e ... . ·.-;" lit e ........ .. . - • " I •• •• :: ... & ... ". ... :,. ..... :.,. •• : •• • � _ •• Cl> ... .. ,. ••• e. .. ' . to • . . .. .. . ., ... . . " . . ...... ...... . .. . :. . . . - ...••... . .. ':. . . . .. . . .. . . � . . . .. . , . - . . .. . . . . . . . .. . . . . ... . . . . .. ...... • oe .. � • • • � • • • • •• " • ••• ... · . 18 .. • . . . ..... . . . . . . . .. . . . .. . . .. ... � . ..•.. . ... � ... � . .. � . .. ......... , s . . .. .. ft .. . . . .. . . ... .. . . . _ ••• • • " . . .. . ft •••• • • ,: •• ",· • • • • .. . .. .... . . . . ... .. .... . . ..... ......... " ..... , - . . .. . . . . . .. "' .... .. ,. ... . . ... .. . ", . . .. . .. . .. .. . ... e •• • ... . • • . .. ·Of .. . . . .... . •• .,� • . . ....... "!. •• ':. ••• :.." ... . ... . . • • • -:. ... . .. -:, .. . ... . .. ..... . :.. e ""e" ••• • • • • • - . .. ". e . ..... " •• 0 . . .. . . . .... . . . .. . .. ft. ... .. ...... 8 " . . .... e . •• "' .11 " . ..t ....... .... -... ':.� ... � ... " . ... ,.:: .... .... • ... .... . -oil •••• oil.· " oil •• ".oe •••••• � ••• � .......... _._ •• I. Fig. 6.33. (a) Platinwn opal structure produced by sputter-coating the polystyrene opal with platimun, (b) opal showing predominantly body-centred cubic square-packing and (c) opal showing predominantly hexagonal close-packing with polymer coating. 24 1 o 10 . 0 �m 0 10 . 0 �m Fig. 6.34. AFM images of the underside of a platinum opal structure. Ca) 3-Dirnensional image, Cb) height data and Cc) deflection data. Scan size: 1 0.00 IJIII. 242 Fig. 6.35. (a) Gold opal surface on ITO-coated glass and (b) poly(OCJoDASTT) cast on a gold opal surface. 243 Fig. 6.36. Edge ofITO opal structure on ITO-coated glass. 6.5.3.4 Summary Platinum opal and mverse opal structured films were successfully made by electrodeposition and sputter-coating of the platinum metal. It was found, however, that polyOCIODASTI did not adhere well to the platinum surfaces. Gold and ITO opal films were made in an attempt to improve the polymer film quality, but this did not work. 244 6.5.4 Electrospinning Fibres are fibrils that are entwined to form rope-like structures. Fibres of conducting polymers show potential for applications such as nanowires, 1 3 high surface area conducting materials/08 electrodes and sensors. Electrospinning is one technique used for the formation of polymer fibres. 1 3,208,209 By applying a potential difference (typically 30 kV) between a polymer-containing droplet and a target plate, the polymer particles are induced to fly towards the target. Fine polymer fibres are generated as the solvent evaporates. The electrospinning of chemically synthesised polyOCIODASTT was investigated in this study. Although attempts at electrospinning a pure solution of polyOCIODASTT (50 mg mL-1 ) in chloroform resulted only in a spray of fine droplets, addition of 5% polyethyleneoxide (PEO) allowed the production of bluelblack fibres. SEM images of these fibres (Fig. 6.37) show that they range between 400 and 600 om in diameter. Pure PEO fibres were established to be white in colour, and soluble in methanol. The polyOCIODASTTIPEO fibres produced in this study were bluelblack and not soluble in methanol. This suggests that the fibres contain polyOCIODASTT, although further analysis would be required to reveal the actual polyOCIODASTT to PEO proportion. PEO does, however, appear necessary for the formation of polyOCIODASTT fibres. Solution viscosity has been found to be a significant parameter in fibre forrnation. 1 29 Since it was observed that addition ofPEO to the solutions (in amounts as low as 5%) dramatically increased the viscosity of the solution, this may be a significant role of PEO. 245 Fig. 6.37. SEM images at different magnifications of electrosplUl POlyOCloDASTT fibres. 246 6.5.5 Conclusions The solubility of polyOCI ODASTI allowed preparation of nanostructured surfaces by casting and electrospinning methods. This material was successfully made into fibrillar structures by casting onto a template, and fibres by electrospinning. Both of these structures display a high surface area that should be useful for improving the efficiency of devices. Platinum, gold and ITO opal structured films were successfully made by sputter­ coating polystyrene opal surfaces. Platinum inverse-opal surfaces on ITO-coated glass could also be made by electrodeposition through the polystyrene opal beads, followed by removal of the polystyrene. Unfortunately, it was found that polyOCIODASTT did not adhere well to these surfaces. 247 6.6 Conclusions A selection of the styryl-substituted oligo- and polythiophenes reported in Chapter 3-5 were incorporated and tested in PEC cells. A cell incorporating a cast film of chemically dimerised NMezSTI produced a FF of 35%, the same as that reported by Cutler et al. for a similar cell incorporating a dimethylarninostyryl-substituted thiopheneibithiophene copolymer. Photoelectrochernical devices made from polyOCIODASTT did not give any detectable photovoltaic activity, possibly due t o the high stability of this polymer in the oxidised state. PEC cells incorporating films of styryl-thienylenevinylene derivatives showed an increase in photovoltaic activity with an increasing electron-donating effect of the p-phenyl substituent. PVDF membrane benders and fibres, as well as a stand-alone film were synthesised from simple solution casting of polyOCIODASTT. These structures have been shown to possess electromechanical properties by undergoing a reversible volume change in response to a repetitive potential pulse in an electrolytic media. It was shown that a stand alone polyOCI ODASTI film actuator could achieve a strain of up to 3 .3% under a stress level of 44 kPa. A relatively low strain rate of about 0 .01 1 % was observed for a stand-alone film. This modest result may be due to the electrolyte struggling to penetrate the film and/or poor accessibility of charge due to the low conductivity of the material. Although this strain rate may be improved by using a conductive substrate, casting or electrochemical growth of a polyOCI ODASTT film of suitable thickness onto a platinum surface proved difficult. The good discharge capacity and excellent cycle life shown by the Teflon test batteries described in this study, reveals potential for polyOCI ODASTI as an active anode material for all-polymer batteries. The substrate that the polyOClODASTI was cast onto was shown to have a significant effect upon the performance of the cell. NilCu coated nonwoven polyester fabric was found to produce significantly improved discharge capacity compared to anodes comprising carbon fibre mat as a substrate. A high discharge efficiency of >94% was generated by both types of cells with anodes employing polyOClODASTT. 248 Soluble polyOCIODASTT was successfully fabricated into fibrillar structures by casting onto templates, and fine fibres by electrospinning. These materials were observed by microscopy to have high surface areas. 249 6.7 Experimental Procedu res 6.7. 1 PEC devices Tetrabutylammonium perchlorate (TBAP, Fluka), iodine (Univar, Ajax or Aldrich 99. 8%), methanol (Univar, Ajax), acetonitrile (AN, Univar, Ajax), dichloromethane (DCM, Univar, Ajax), isopropanol (Univar, Ajax), tetrapropylamrnonium iodide (Sigma-Aldrich, � 98%), ethylene carbonate (Sigma-Aldrich 99%), propylene carbonate (Sigma-Aldrich 99%) and chloroform (Sigma-Aldrich) were used without further purification. ITO (indium tin oxide) coated glass (S 1 0 n cm-2) was purchased from Delta Technologies Limited (USA), cut into required sizes, washed with liquid detergent, rinsed thoroughly with Milli-Q water followed by acetone, and allowed to dry. Before coating with polymers, the ITO coated glass was treated in an UVO­ cleaner (Model No. 42-220, Jelight Co. Inc. , USA). Electrochemical synthesis of films , and photovoltaic testing was performed using an EG&G PAR 363 Potentiostat/Galvanostat, a MacLab 400, and EChem v l . 3 .2 software (ADInstruments). The counter electrode (cathode) was produced by sputter coating a thin layer of platinum onto ITO coated glass. The sputter coating was done using a Dynavac Magnetron Sputter Coater Model SC I 00MS at a current of SOmA and argon pressure of 2 x 1 0-3 mbar. A Pt thickness of l oA was sputter coated. Liquid electrolyte was prepared by dissolving Iodine (60 mM) and tetrapropylammonium iodide (500 mM) in a 1 : 1 (by weight) mixture solution of ethylene carbonate and propylene carbonate. This results in a h -/1" redox couple being formed in solution. The polymer was deposited onto ITO coated glass either electrochemically from the mo no mer, by casting, or by spin-coating. Electrochemically produced films were reduced at -0. 5 V vs. Ag/Ag+ in 0. 1 M TBAPIAN, rinsed with acetonitrile and allowed 250 to dry. The device was assembled by sandwiching the rlI3- liquid electrolyte between the platinum sputter coated ITO coated glass electrode (cathode) and the polymer coated ITO glass electrode (anode) as shown in Fig. 6. l . A gasket made of parafilm was used to separate the electrodes to avoid short circuits and to contain the liquid electrolyte. The devices were tested immediately after fabrication in a large dark box containing a broad spectrum halogen lamp (500 W m-2, SOLUX, 4700K, Wiko Ltd. ) The polymer coated side was faced to the lamp. Linear sweep voltammetry with a lower limit of -50 mY, an upper limit of 250 mV and a scan rate of 1 00 mV S- l was used to record the 1 - V curve. This potential range was scanned first in the dark then under illumination to record the change in current due to the absorption of light. Films of N02STV, CNSTV and STY oligomers were electrodeposited onto ITO­ coated glass using cyclic voltammetry (-0.5 V to 0 .8 V), and OMeSTV and NMezSTV oligomers by constant potential (0. 8 V) due to their high solubility in the neutral state. The films were then reduced at -0. 5 V in monomer free 0. 1 M TBAP/AN. A film of oligoNMezSTT was electrochemically deposited using a constant potential (0. 9 V for 30 seconds) and then reduced (-0. 5 V in monomer free solution) to a neutral state. 6.7.2 Actuator fabrication and experimental procedures Chloroform, tetrabutylammonium hexafluorophosphate (TB AP F6) and propylene carbonate (PC) were obtained from Sigma-Aldrich and used without further purification. Inert porous polyvinylidene fluoride (PVDF) membrane sheets and fibres were obtained from Millipore, MA01 730. Platinum sputter-coating of the sheets was performed using a Magnetron Sputter Coater, DYNAV AC Sc l OOMS using a constant current of 50 mA for 20 minutes. PolyOCIODASTT ( 12 Jlg mm-2) was cast onto both platinum coated, and non-coated PVDF membranes. These membranes were cut into 30 x 2 mm strips which were 1 1 5 Jlm thick. The strips were tested using a 3 electrode cell in a Petri-dish. An 251 electrolyte consisting of 0.25 M TBAPF6/PC was used with a stainless steel mesh counter electrode and a AgI Ag + reference electrode. Electrochemical deposition of OCI ODASTT onto platinum sputter-coated PVDF membrane was attempted using galvanostatic methods by applying constant current (0. 5 mAcm-2) for twelve hours in a 1 : 1 AN:DCM solution containing 1 0 mM monomer and 0. 1 M TBAP. This solvent mix is required to dissolve the mono mer without solvation of the resulting polymer. The deposition was carried out at 4°C to further slow polymerisation, and the reaction vessel sealed to reduce evaporation of the dichloromethane. Electrochemical deposition was also attempted from a 2: 1 PC:DCM solution containing 5 mM OCIQDASTT and 2% polyhydroxy ether /propylene carbonate. Both galvanostatic methods (0.05 mA cm-2 for four hours) and potentiostatic methods (0. 8 V for l . 5 hrs) were performed at 4°C . A free-standing film was produced by casting a chloroform solution of polyOCIQDASTT evenly over a glass surface on which a zigzagged wire (50 !lm diameter) was placed. The resulting film with incorporated wire was peeled off and cut to appropriate dimensions : 6 mm wide and 14 .5 mm in length. Spectra were measured using a Cary 500 UV-VIS-NIR spectrophotometer (Varian) in the reflectance mode and the data recalculated to show absorbance. Cyclic voltarnmetry of cast poly(OCIQDASTT) films on glassy carbon electrodes or PVDF membranes was performed using a 3-electrode, I -compartment cell setup. A platinum mesh counter electrode and AglAg+ reference electrode was used with an electrolyte solution of 0. 1 M TP AP/ AN. A scan-rate of 1 0 mV s-J or 1 00 mV s-J was used as indicated. Conductivity measurements were made using a multiheight four-point probe with a landel raising and lowering mechanism and a cylindrical probe. Samples were tested using a 305B Dual Mode Lever System (Aurora Scientific Inc.) . The bottom of the sample was held stationary with a platinum wire wound around a clamp which also functioned as the working electrode connection. The top of the sample was held by a silicon rubber clamp which was connected to the lever by a stiff 252 wire. The entire sample was submerged in a 0.25 M TBAPF6/PC electrolyte solution. A high surface area stainless steel mesh sheet and a Ag/Ag+ reference electrode were employed as the counter and reference electrodes respectively to complete a 3- electrode cell setup. The potential was controlled, and data measured by an EG&G PAR model 363 potentiostat/galvanostat with a MacLab 400 electrical interface. The data was recorded by Chart software CV3.3 . 7, ADlnstruments). All measurements were made at room temperature. Before performing electromechanical measurements, the free-standing film was pre­ conditioned. A small constant 40 mN stretching force was loaded to on the top of film in order to keep it straight and 4 hour square wave pulsed potential between + 1/-0 .6 V (vs Ag/Ag+) was applied from the electrical contact at the bottom of the film. This process was conducted for 24 hrs until a steady equilibration state was obtained. 6.7.3 Details of battery fabrication and testing procedures The anode was prepared by drop-casting a solution of polyOCIODASTT in chloroform onto either a carbon fibre mat substrate (resistance of � 1 0 . 5 n cm-I) or Ni/Cu coated nonwoven polyester substrate (resistance of <0.5 n cm- I ). Several coats were applied to give a polymer coating of about 1 mg cm-2. The cathode was prepared by galvanostatic electro-deposition of polypyrrole onto stainless steel mesh A current density of 1 . 0 mA cm-2 was employed with a propylene carbonate (PC) solution containing 0 .06 M monomer (pyrrole) and 0.05M tetrabutylarnmonium hexafluorophosphate (TBAPF6). Prior to the experiment, the solution was deaerated with N2. A total charge density of 3 .0 C cm-2 was consumed in the deposition of the polypyrrole film. Cyclic voitarnmetry of the polymer was performed usmg an EG&G PAR 363 potentiostat/galvanostat, a MacLab 400, and EChem v 1 . 3 .2 software (ADlnstruments). The electrochemistry was performed using a standard one 253 compartment, three-electrode cell with a stainless steel mesh counter electrode, Ag/Ag+ reference electrode and a 0.05 M TBAPF6/PC electrolyte. The surface morphologies of the electrodes were investigated using a scanning electron microscope (SEM, Leica Model Stereoscan 440) with a secondary electron detector. All materials were sputter coated with gold prior to scanning. The test cells were assembled in an argon-filled glove box (Unilab, Mbraun, USA). For fabrication of the test cell, the anode and cathode material were cut into 1 x 1 cm2 squares. These were separated by a Celgard 2500 (microporous polypropylene membrane) and submerged in an electro lyte consisting of 1 . 0 M LiPF 6 in 1 : 1 ethylene carbonate:dimethylcarbonate solution. In order to let the electrolyte penetrate into the inner part of the active material in the electrodes, the Teflon cells were kept at room temperature for 2 hours after the assembly. Cells were tested usmg a battery testing device (Neware Electronic Co. China). Different current densities were employed to investigate the properties of the cell. The cells were galvanostatically charged to a cell voltage of 1 . 65 V, and then discharged at the same current density to a cut-off voltage of 0.2 V. For the cycle life test, the cell was initially activated by cycling at a current density of 0.02 mA cm-2 for several cycles until a steady charge/discharge capacity was obtained. Cycle life testing was then performed under a charge/discharge current density of 0.05 mA cm-2. 6.7.4 Fabrication and evaluation of nanostructures General Tetrabutylammonium perchlorate (TBAP, Fluka), methanol (Univar, Ajax), acetonitrile (AN, Univar, Ajax), dichloromethane (DCM, Univar, Ajax), chloroform (Sigma­ Aldrich), propylene carbonate (Aldrich 99%) were used without further purification. ITO (indium tin oxide) coated glass (S 1 0 n cm-2) was purchased from Delta Technologies Limited (USA), cut into required sizes, washed with liquid detergent, 254 rinsed thoroughly with Milli-Q water followed by acetone, and allowed to dry. Before coating with polymer, the ITO-coated glass was treated in an UVO-cleaner (Model No. 42-220, Jelight Co. Inc. , USA) for 1 5 mins. Electrochemistry was generally performed a MacLab 400, usmg an and EChem EG&G PAR 363 v l . 3 .2 software potentiostat/galvanostat, (ADInstruments). Charge measurements were made using a BAS CV27 voltarnmograph. A standard one compartment, three-electrode ceU with a stainless steel mesh or platinum mesh counter electrode and a AgI Ag+ reference electrode with 0. 1 M TBAP/AN salt bridge was used. An electrolyte solution of 0. 1 M TBAP/AN was employed for analysis of polymeric samples using cyclic voltammetry. Imaging by scannmg electron microscopy (SEM) was carried out usmg a Leica­ stereoscan SS 440 Microscope. Non-conductive surfaces were sputter-coated with gold prior to imaging. Atomic force microscopy (AFM) was performed on a DI (Digital Instruments) 3 1 00 NanoMan. A tapping mode was used for all imaging. Sputter-coating of gold, platinum or ITO was performed using a Dynavac Magnetron Sputter Coater Model SC I00MS. Fabrication of fibrillar structures The template used in this study to fabricate fibriUar structures was a Whatman anodisc 1 3 membrane composed of alumina, with honeycomb shaped pores of 200 nm diameter. Sputter-coating of the template with platinum was achieved at a current of 30mA and argon pressure of 2x 1 0-3 mbar for 1 0 mins. A film thickness of 1 0 A was sputter coated The template was soaked in a solution of polymer (about 2 mg mL-1) for one hour. The template containing the fibrils was then removed from the solution and drip dried. The template was either fully dissolved by soaking in 0.5 M NaOH for 50 minutes to leave a very fragile film of polymer fibrils, or partially dissolved by soaking in 0. 5 M NaOH for 25 minutes to leave an alumina support. Finally, the fibrils were rinsed with deionised water. 255 Fabrication of opal and inverse opal surfaces P olystyrene opals were prepared by air drying a drop-cast film or spin-co ated film of polybead® dyed blue solution (0. 20 micron microspheres, 2.63% solids-latex, P olysciences Inc . , Warringoton P A) or polystyryene particles (0. 92 � ±0.07 � diameter, Microparticles, GmbH, Germany) on ITO-coated glass (-2 cm2 area). This was allowed to dry and pack for at least 24 hours to form an organized structure. P latinum inverse opals (honey comb structure) were created by electrochemically depositing platinum from a solution containing hexachloroplatinic (IV) acid, concentrated nitric acid and concentrated hydrochloric acid. This so lution was made by stirring platinum granuals ( 1 g) in 1 0 mL of aqua regia (2: 1 v/v HCI:HN03) for 48 hrs while heating at 70°C. Both cyclic voltammetry growth methods (five or ten cycles at scan rates of 50 or 1 00 mVs-1 between the limits of -400 t o 1 600 mV vs. Ag/Ag+) and potentiostatic growth methods (-0.4 V at 1 00 seconds) were investigat ed. It was fo und that inverse opal adhered v ery poorly to the ITO surface and peeled away when the slide was removed from solution In addition, platinum did not always deposit properly and there was a problem of 'beading ' which covered the opal structure. The best films were obtained using cyclic voltarnmetry (either 1 0 cycles at 1 00 mVs-1 or five cycles at 50 mVs- 1 ) . The growth CV for platinum deposition at 1 00 mVs-1 is shown below in Fig. 6. 38. The voltammograrns at 50 mVs-1 are similar but display a slightly higher current. 256 20 .------------------------------------------------. 1 5 1 0 c( - 5 1: � 0 � o -5 -1 0 -15 -20 +---�----�--�----�--�----�--�----�--�--� -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1 .00 1 .20 1 .40 1 .60 Potential I V Fig. 6.38. Growth of platinum on opal coated ITO coated glass. 1 0 cycles. Scan rate: 100 m V S-I . Platinum, gold or ITO opal structures (beads) were prepared by sputter-coating the polystyrene opal with platinum A film thickness of � 1 20 nm was deposited by applying a current of 50 mA and argon pressure of 2 x l 0-3 mbar for 1 0 mins. The polystyrene opal was then removed from the structure by heating in a furnace at 450°C for two hours, or by soaking in chloroform for 24 hrs. The remaining metal or ITO opal structures were subjected to an ozone atmosphere for 1 5 minutes. PolyOCIODASTT was then drop cast onto a 27 mm2 area of the remaining metal or ITO opal structure by using a circular shaped gasket. Electrospinning PolyOCIODASTT fibres were synthesised by electrospinning a solution of chemically synthesised polyOCIODASTT (50 mg mL-1) with 5% PEO in chloroform ( l mL)_ The solution to be electrospun was filtered through a 0.45 mm filter and placed in a hypodermic syringe 20 - 30 cm from a galvanized steel target place covered with aluminium foil. The positive (anode) electrode of a variable high voltage transformer (Gamma High Voltage Research, Ormond, Florida 32 1 74) was attached to the metal needle of the syringe and the negative terminal was attached to the target electrode. A 20 kV differential was applied across the electrodes. Fibres were collected by passing a 257 glass slide or silicon wafer between the electrodes and washed with methanol to remove any residue PEO. 258 Chapter 7 Conclusions This work demonstrates that, although oxidation of styryl-terthiophene results in insoluble and short-chain material, the attachment of alkoxy substituents allows the synthesis of soluble polymer. Although not suitable as an active material in photovoltaics, alkoxy-substituted styrylterthiophene (polyOCIODASTT) exhibits potential as an active material for use in actuators and battery anodes. Styryl-substi tuted terthiophenes (Chapter 3) The chemical oxidation of a series of styryl terthiophene derivatives, fimctionalised by electron-withdrawing or donating substituents, resulted in the production of mostly insoluble materials. Although the insoluble fraction was difficult to characterise and further process, analysis of the soluble fraction revealed predominantly dimer with traces of short oligomers (n < 4). It was shown by NMR spectroscopy that the dimeric materials produced are predominantly of the head-to-head isomer, which is consistent with reports by Grant et al. who studied ether-substituted styrylterthiophene derivatives, 1 1 and theoretical calculations of similar materials by Clarke et al. 140 However, it was found to be difficult to isolate the head-to-head isomers of these materials. It was then demonstrated that films of these materials could be grown both potentiostatically and electrodynarnically. DV -VIS-NIR spectra of these films indicated that they consisted predominantly of oligomers with a short effective conjugation length. SEMs of the films showed evidence of crystallinity, consistent with films of styrylsexithiophene derivatives reported by Grant. 1 1 Styryl-substituted terthienylenevinylene (Chapter 4) The electrochemical oxidation of a series of styryl-substituted terthienylenevinylene derivatives was also found to generate films consisting predominantly of dimer. This is consistent with the chemical oxidation of these materials that was reported by Wagner 259 et al. 1 56 The low degree of polymerisation of these mat erials may be due to a low reactivity of the terminal 'a' po sitions of the monomers due to their extended 1t­ conj ugation. 23 Films of t erthienylenevinylene oligomers were observed t o generate absorption maxima (due to the 1t-1t* transition), at higher wavelengths than those o f films o f t erthiophene o ligomers with the same substituent. This is consistent with reports in the literature of smaller bandgaps produced by polythienylenevinylenes than polythiophenes, 1 02 and is most likely due to an increase in the effective conj ugation length provided by the addition of the vinyl link er and/or increased planarity of the oligomer chain allowing enhanced 1t-orbital overlap. A lkyl and alkoxy substituted styryl terthiophenes (Chapter 5) The chemical and electrochemical polymerisation of alkyl- and alkoxy-substituted styrylterthiophenes was investigated. It was found that attachment of these substituents at the 4 and 4" p ositions enhanced polymerisation of styrylterthiophene and improved the solubility o f the resulting materials. The chemical p olymerisation of Crdialkyl-substituted styrylterthiophene ( C7DASTT) was fO lIDd to pro duce a po lymer with an average chain length of ca. n = 6 and maximum detectable length (by MALDI-TOF MS) of ca. n = 1 2. This material (polyC7DASTT) appeared to be very stable in the neutral state and readily reduced to the neutral state. PolyC7DASTT was found to be completely so luble in chloroform. In companson, polymerisation of C6-dialkoxy-substituted styrylterthiophene (OC6DASTT) generated longer o ligomers, with an average polymer length of ca. n = 1 1 and maximum detectable length of ca. n = 35 (indicated by MALDI-TOF). PolyOC6DASTT was shown to be very stable in the oxidised state, and chemical reduction proved difficult . The high stability of the po lymer in the o xidised state may be explained by the strong electron-donating properties of the alkoxy substituent, which stabilises the positive charge. Only 78% of polyOC6DASTT was soluble. The poorer solubility of polyOC6DASTT compared to polyC7DASTT may be due to a much higher stability of the alkoxy derivative in the oxidised (less soluble) state. It is 260 also possible that the lower solubility when compared to polyC7DASTI may be due to the presence of longer chain OC6DASTI polymer. Lengthening the alkoxy chain from 6 carbons (polyOC6DASTI) to 1 0 carbons (polyOCIODASTI) did not appear to affect the resulting length of the polymer (which also showed an average polymer length of ca. n = 1 1 by MALDI-TOF MS), but did increase the solubility to provide a chloroform soluble fraction of 97% of the total expected polymer yield. UV -VIS-NIR spectroscopy indicated that more of the soluble polyOCIODASTI material appeared to be in a partially oxidised state. Therefore, the higher fraction of soluble material compared to polyOC6DASTT may be due to the fact that partially oxidised material is soluble. The presence of the alkyl and alkoxy substituents was found to cause a significant reduction in the oxidation onset potential of the monomers compared to STT, which may account for their higher degree of polymerisation. The alkoxy substituents were also observed to significantly decrease the oxidation potential of the resulting polymer, consistent with the high stability observed by chemically polymerised polyOC6DASTI and polyOCIODASTI in the oxidised state. Electrochemically deposited films of these alkoxy-DASTI materials on ITO-coated glass were observed to show folding on oxidation, most likely due to the incorporation of dopant ions into the polymer matrix. Experimental conditions during the chemical polymerisation of OC6DASTT and OCIODASTT were found to have a significant effect on the resulting polymer length, and hence solubility. It was found that addition of the oxidant to monomer produced shorter oligomers, most likely due to a decrease in the initial oxidantmonomer ratio and hence rate of reaction It was also found that the presence of ethanol decreased the average oligomer length, which may be explained by an effect of the ethanol on the oxidation potential ofthe iron(III) chloride oxidant. Polymers of different lengths were successfully separated using sequential extractions in a series of solvents. The average polymer length was found to increase with each sequential extraction as determined by MALDI-TOF MS . A bathochromic shift of the absorbance maxima, and reduction in oxidation onset potential of cast films of the 261 OCIODASTT oligomer/polymer fractions was observed with increasing polymer length, indicating a corresponding increase in the mean conjugation length. 1 14, 1 27 The CVs of cast films of chemically polymerised polyOC6DASTT and polyOCIODASTT displayed oxidation and reduction peaks that were much broader than peaks generated by electrochemically polymerised and deposited polyOC6DASTT and polyOCIODASTT films . This may be due to a wider polydispersity of the chemically polymerised materials , 1 59 and/or a less consistent film morphology produced by the casting method compared to electrochemical deposition The use of substituted terthiophene and terthienylenevinylene oligomeric/polymeric materials in photovoltaic devices (Chapter 6) Photoelectrochemical cells with anodes comprising electrochemically deposited films of styryl-thienylenevinylene derivatives showed an increase in photovoltaic activity with an increasing electron-donating effect of the p-phenyl substituent. Photoelectrochemical devices made from polyOC IODASTT did not give any detectable photovoltaic activity. This may be caused by a high stability of this polymer in the oxidised state. Schottky devices using polyOC 1oDASTT also produced poor photovoltaic results, with low Isc and Voc values. However, polyOCIODASTT was found to produce a spectral response that extended to 750 nm. Addition of PP V to the polymer mixture appears to add to the SR and photovoltaic performance. Use of polyOC/ODASTT in actuator devices (Chapter 6) PolyOCIODASTT actuators were shown to generate a moderate strain, although a low strain rate. This is possibly due to difficulty in diffusion of the anion molecules through the compact polymer matrix.. It could also be due to poor charge accessibility throughout the film in the neutral state. However, the simple preparation of actuator materials using this polymer provides potential for some applications where a low strain rate is required. 262 Use of polyOClODASTT as an anode in all-polymer batteries (Chapter 6) PolyOCIODASTI showed promise as a good anode active material for an all-polymer battery. A battery comprised of polypyrrole (cathode-active material), polyOCIODASTT (anode-active material) and LiPF6 as an electrolyte displayed an excellent discharge capacity, discharge efficiency and cycle life over fifty cycles. The substrate on which polyOCIODASTT was cast was found to affect the performance of the cell, suggesting that more work is required to find an optimum anode substrate. Micro- and nanostructured surfaces (Chapter 6) The solubility of polyOCIODASTT allows this material to be prepared by simple casting or spin-coating methods to produce films and surfaces for applications such as photovoltaic cells, actuator materials and battery electrodes. PolyOCIODASTT can be moulded into nanostructured surfaces (fibrils, opals and fibres), which have high surface areas. These surfaces can now be used in device fabrication to increase device efficiency. The ease of functionalising terthiophene with an assortment of substituents via a styryl linker presents the opportunity to supply a large range of polymer properties. By attachment of the alkoxy substituents to allow the preparation of process able polymer, the door is opened for use of these polymers in a variety of applications. Continuation of the work presented in this thesis could include reducing the alkoxy chain length to generate anode active materials with higher charge/discharge capacities. The effect of electron-withdrawing/donating substituents on the ability of poly(dialkoxy-substituted styrylterthiophene) to perform as an anode active material could also be investigated. Finally, it would be interesting to see if the tensile strength and strain rate of actuators constructed using poly(OCIODASTI) could be improved, possibly by blending the polymer with a conducting and strengthening material. 263 Bibliography ( 1) Wallace, G. G. ; Spinks, G. M.; Kane-Maguire, L. A P. ; Teasdale, P . R. Conductive electroactive polymers, 2nd Ed; C RC Press: Boca Raton, 2003 . (2) Allcock, H. R. ; Lampe, F. W . ; Mark, 1. E. Contemporary polymer chemistry, 3rd Ed. ; Pearson Education Inc. : New Jersey, 2003 . (3) Shirakawa, H. Synth. Met. 2001, 125, 3 - 10. (4) MacDiarrnid, A G. Synth. Met. 2001 , 125, 1 1 -22. (5) Biserni, M.; Marinangeli, A; Mastragostino, M. J. Electrochem. Soc. 1985, 132, 1597- 1 60 l . (6) Kaufinan, 1. H.; Chung, T. c . ; Heeger, A 1 . ; Wudl, F. J. Electrochem. Soc. 1984, 131, 2092-2093 . (7) Kraft, A ; Grimsdale, A C . ; Holrnes, A B. Angew. Chem., Int. Ed. 1998, 37, 403-428. (8) Roncali, 1. Chem. Rev. 1992, 92, 7 1 1 -738. (9) Beaupre, S . ; Leclerc, M. Adv. Funct. Mater. 2002, 12, 192- 196. ( 10) Reddinger, 1 . L. ; Reynolds, 1. R. Polym. Prepr. (Am. Chem. Soc. , Div. Polym. Chem.) 1 997, 38, 32 1-322. ( 1 1 ) Grant, D. K. , The synthesis and properties of poly ether substituted oligothiophenes. Ph.D. Thesis, Institute of Fundamental Sciences, Massey University, 2003. ( 12) Gamier, F. Angew. Chem. 1989, 1 01, 529-533. ( 13) Jones, W. E. , Jr. ; Dong, H.; Nyame, V. ; Ochanda, F. Annual Technical Conference ­ Society of Plastics Engineers 2003, Vol. 2, 1948- 1950. ( 14) Higgins, S. 1. Chem. Soc. Rev. 1997, 26, 247-258. ( 15) Higgins, S . 1 . ; Pounds, T . 1 . ; Christensen, P . A J. Mater. Chem. 2001 , 1 1 , 2253-226 l . ( 16) Shi, G. ; Yu, B. ; Xue, G.; Jin, S . ; Li, C. J. Chem. Soc., Chem. Commun. 1994, 2549- 2550. ( 1 7) Arbizzani, c . ; Gallazzi, M. C. ; Mastragostino, M.; Rossi, M. ; Soavi, F. Electrochem. Commun. 20 1 , 3, 1 6- 19. ( 18) Fusalba, F. ; Ho, H. A; Breau, L. ; Belanger, D. Chem. Mater. 2000, 12, 258 1 -2589. ( 19) Soudan, P . ; Lucas, P . ; Ho, H. A; Jobin, D.; Breau, L. ; Belanger, D. J. Mater. Chem. 2001 , 1 1 , 773-782. 264 (20) Fichou, D. J. Mater. Chem. 2000, 10, 5 7 1 -588. (2 1) Anquetil, P. A; Hunter, I . ; Madden, J . D . ; Madden, P . G. ; Pullen, A; Swager, T. M. ; Xu, B . ; Vu, H . - H . ; (Massachusetts Institute of Technology, USA). Application: W02003 10 1955 , 2003 ; 135 pp. (22) Gadisa, A ; Workalemahu, B. Synth. Met. 2002, 129, 1 79- 185. (23) Roncali, 1. Chem. Rev. 1997, 97, 1 73-205. (24) Sato, M . ; Tanaka, S . ; Kaeriyama, K. Synth. Met. 1 986, 14, 279 - 288. (25) Cutler, C. A, Electrochemical and photoelectrochemical studies offunctionalised polythiophenes. PhD Thesis, Department of Chernistry, University of Wollongong, 2000. (26) McQuade, D. T. ; Pullen, A E . ; Swager, T. M. Chem. Rev. 2000, 100, 2537-2574. (27) McCullough, R. D. Adv. Mater. 1998, 10, 93- 1 16. (28) Roncali, 1. J. Mater. Chem. 1999, 9, 1875- 1893 . (29) Fuchiwaki, M . ; Takashima, W. ; Pandey, S. S . ; Kaneto, K. Synth. Met. 2003, 135-136, 135-136. (30) Epstein, A 1. ; Yang, Y MRSBuii. 1997, 22, 1 3- 14. (3 1 ) Pratt, C . Conducting polymers. Electronic source. 2002. (32) Child, A D . ; Sankaran, B . ; Larmat, F . ; Reynolds, 1. R. Macromolecules 1995, 28, 6571 -6578. (33) Pron, A; Rannou, P . Prog. Polym. Sci. 2001, 27, 135- 190. (34) P ickup, P . G. In Modern aspects ofelectrochemistry; White, R. E . , Ed. ; Kluwer AcaderniclPlenium P ublishers : New York, 1999; Vol . 33, pp 549-597. (35) Swager, T. M. Acc. Chem. Res. 1998, 31, 201 -207. (36) Roncali, 1. Annu. Rep. Prog. Chem. , Sect. C, Phys. Chem. 1999, 95, 47-88. (37) Elandaloussi, E . H . ; Frere, P . ; Richomme, P . ; Orduna, 1 . ; Garin, 1. ; Roncali, J. J. Am. Chem. Soc. 1997, 1 19, 10774- 10784. (38) Baughman, R . H. Synth. Met. 1 996, 78, 339-353. (39) Shirota, Y J. Mater. Chem. 2000, 10, 1 -25. (40) Gallazzi, M. c . ; Toscano, F . ; Paganuzzi, D . ; Bertarelli, c . ; Farina, A; Zotti, G. Macromoi. Chem. and Phys. 2001, 202, 2074-2085. (4 1 ) E lsenbaumer, R . L. ; Jen, K. Y . ; Miller, G . G . ; Shacklette, L . W . Synth. Met. 1987, 18, 277-282. 265 (42) Chen, J . ; Burrell, A K . ; Campbell, W. M . ; Officer, D. L . ; Too, C. 0. ; Wallace, G. G. Electrochim. Acta 2004, 49, 329-3 3 7. (43) Waltman, R. 1 . ; Bargon, J . ; Diaz, A F . J. Phys. Chem. 1 983, 8 7, 1 4 59- 1 463 . (44) Wang, c . ; Benz, M. E. ; LeGoff, E . ; Schindler, 1 . L . ; Allbritton-Thornas, 1 . ; Kannewurf, C . R . ; Kanatzidis, M . G . Chem. Mater. 1 994, 6, 40 1 -4 1 1 . (45) Loewe, R. S . ; Ewbank, P. c.; Liu, J . ; Zhai, L . ; McCullough, R. D. Macromolecules 2001 , 34, 4324-4333. (46) Roncali, 1. Acc. Chem. Res. 2000, 33, 1 47- 1 56. (47) Tourillon, G. Handbook of conducting polymers; Marcel Dekker, Inc . : New York, 1986; Vol. l . (48) Smith, J . R . ; Cox, P. A ; Campbell, S. A ; Ratcli ffe, N. M. J. Chem. Soc. , Faraday Trans. 1 995, 91 , 233 1 -2338. (49) Krische, B . ; Zagorska, M. Synth. Met. 1 989, 28, C263-C268. (50) Jerome, c . ; Maertens, c . ; Mertens, M . ; Jerome, R . ; Quattrocchi, c. ; Lazzaroni, R . ; Bredas, 1 . L. Synth. Met. 1996, 83, 1 03- 1 09. (5 1 ) Smith, J . R . ; Campbell, S . A; Ratcliffe, N . M. Bull. Electrochem. 1995, 1 1 , 3 78-386. (52) Higgins, T. B . ; Mirkin, C . A Chem. Mater. 1 998, 1 0, 1 5 89- 1 595. (53) Wei, Y . ; Tian, 1 . Macromolecules 1 993, 26, 45 7-463. (54) Visy, c. ; Lukkari , J . ; Kankare, J. Macromolecules 1 994, 2 7, 3322-3329. (55) Demanze, F. ; Yassar, A; Gamier, F. Macromolecules 1 996, 29, 4267-4273 . (56) Randriamahazaka, H.; Noel, V; Chevrot, C. J. Electroanal. Chem. 2002, 52 1 , 1 07- 1 1 6. (57) McCullough, R. D. ; Tristram-Nagle, S . ; Williams, S. P . ; Lowe, R. D . ; Jayaraman, M. J. Am. Chem. Soc. 1993, 1 1 5, 49 1 0-49 1 l . (58) McCullough, R. D. ; Williams, S. P . ; Tristram-Nagle, S. ; Jayaraman, M. ; Ewbank, P. C . ; Miller, L. Synth. Met. 1 995, 69, 279-282. (59) Daoust, G. ; Leclerc, M. Macromolecules 1 99 1 , 24, 45 5 -459. (60) Krische, B . ; Zagorska, M. Synth. Met. 1 989, 33, 257-267. (6 1 ) Chandrasekbar, P . Conducting polymers: fundamentals and applications; Kluwer Academic Publishers: Boston, 1 999. (62) Gallazzi, M. C . ; Castellani, L . ; Marin, R. A; Zerbi, G. J. Polym. Sci. , Part A: Polym. Chem. 1 993, 31, 3 339-3349. (63) Xu, 1 . M . ; Chan, H. S. 0. ; Ng, S. c . ; Chung, T. S. Synth. Met. 2002, 132, 63-69. 266 (64) McCarley, T. D . ; Noble, C. 0. ; DuBois, C . 1 . , Jr. ; McCarley, R. L. Macromolecules 2001 , 34, 7999-8004. (65) Leclerc, M. ; Martinez Diaz, F. ; Wegner, G. Makromol. Chem. 1989, 190, 3 1 05-3 1 1 6. (66) Johansson, T. ; Mammo, W . ; Andersson, M. R ; lnganaes, O. Chem. Mater. 1999, 1 1, 3 133-3 139. (67) Martinez, F . ; Neculqueo, G. Int. J Polym. Mater. 1999, 44, 265-274. (68) Kowalik, 1 . ; Tolbert, L. M. ; Narayan, S . ; Abhirarnan, A S. Macromolecules 2001, 34, 5471 -5479. (69) Watson, K. 1 . ; Wolfe, P . S . ; Nguyen, S. T. ; Zhu, 1. ; Mirkin, C. A Macromolecules 2000, 33, 4628-4633. (70) Andersson, M. R. ; SeIse, D. ; Berggren, M. ; Jaervinen, H . ; Hjertberg, T. ; Inganaes, 0. ; Wennerstroem, 0. ; Oesterholm, 1. E. Macromolecules 1994, 27, 6503-6506. (7 1) Stein, P . C . ; Botta, c . ; Bolognesi, A; Catellani, M. Synth. Met. 1995, 69, 305-306. (72) Waltrnan, R 1. ; Bargon, 1. Can. J Chem. 1986, 64, 76-95. (73) Zotti, G. ; Marin, R A ; Gallazzi, M. C. Chem. Mater. 1997, 9, 2945-2950. (74) Irvin, J. A ; Reynolds, 1. R Polymer 1998, 39, 2339-2347. (75) Casalbore-Miceli , G. ; Gallazzi, M. c . ; Zecchin, S . ; Carnaioni, N. ; Geri, A ; Bertarelli, C . Adv. Funcl. Mater. 2003, 13, 307-3 1 2. (76) Bredas, 1. L. ; Street, G. B . ; Themans, B . ; Andre, J. M. J Chem. Phys. 1 985, 83, 1 323- 1329. (77) Bredas, 1. L. J. Chem. Phys. 1985, 82, 3808-381 1 . (78) Burrell, A K. ; Chen, 1 . ; Collis, G. E . ; Grant, D. K. ; Officer, D. L . ; Too, C. 0. ; Wallace, G. G. Synth. Met. 2003, 135-136, 97-98. (79) Cutler, C. A ; Burrell, A K. ; Officer, D. L. ; Too, C. 0. ; Wallace, G. G. Synth. Met. 2002, 1 28, 35-42. (80) Collis, G. E . ; Burrell, A K.; Scott, S. M. ; Officer, D. L. J. Org. Chem. 2003, 68, 8974-8983 . (81 ) Roncali, 1 . J. Mater. Chem. 1997, 7, 2307-232 1 . (82) Giaccai, 1 . ; Killian, 1. G. ; Gofer, Y ; Poehler, T. 0.; Searson, P . C. Corrosion in AdvancedMaterials and Systems, Proceedings ofCorrosionl98 Research Topical Symposium, San Diego, Mar. 22-27, 1 998, 1 5-23 . (83) Martinez, F . ; Retuert, 1 . ; Neculqueo, G. Int. J Polym. Mater. 1995, 28, 5 1 -59. (84) Searson, P . ; Killian, 1. G. ; Sarker, H . ; Giaccai, J . ; Gofer, Y ; Poehler, T. 0.; (Johns Hopkins University, USA). Application: US5733683, 1998; 1 3 pp. 267 (85) Kinbara, E . ; KWlUgi, Y; Harima, Y ; Yamashita, K Synth. Met. 2000, 1 1 4, 295-303 . (86) Ferraris, 1. P . ; Eissa, M. M. ; Brotherston, I. D. ; Loveday, D. C. Chem. Mater. 1998, 1 0, 3528-3535. (87) Rudge, A; Raistrick, 1 . ; Gottesfeld, S . ; Ferraris, 1. P. Electrochim. Acta 1994, 39, 273-287. (88) Smith, 1 . R; Campbell, S . A ; Ratcliffe, N. M. ; Dunleavy, M. Synth. Met. 1994, 63, 233-243. (89) Welzel, H.-P.; Kossmehl, G. ; Engelmann, G.; Plieth, W. Eur. Polym. J 1997, 33, 299- 30 l . (90) Greenwald, Y ; Cohen, G. ; Poplawski, 1. ; Ehrenfreund, E . ; Speiser, S . ; Davidov, D. J. Am. Chem. Soc. 1996, 1 18, 2980-2984. (9 1) Cutler, C . A; Burrell, A K ; Collis, G. E . ; Dastoor, P . c . ; Officer, D. L. ; Too, C. 0. ; Wallace, G. G. Synth. Met. 2001, 123, 225-237. (92) Collis, G. E . ; Burrell, A K ; Officer, D. L. Tetrahedron Lelt. 2001 , 42, 8733-8735 . (93) Mukoyama, 1 . ; Aoki, K ; Chen, 1. J Electroanal. Chem. 2002, 531 , 133 - 1 39. (94) Feldhues, M.; Kaempf, G. ; Litterer, H . ; Mecklenburg, T. ; Wegener, P . Synth. Met. 1989, 28, C487-C493. (95) McCullough, R. D. ; Lowe, R. D. J Chem. Soc. , Chem. Commun. 1992, 70-72. (96) Liu, 1 . ; Loewe, R S . ; McCullough, R. D. Macromolecules 1999, 32, 5777-5785 . (97) McCullough, R. D . ; Lowe, R D. ; Jayaraman, M. ; Anderson, D. L. J Org. Chem. 1993, 58, 904-9 12. (98) Gallazzi, M. C. ; Bertarelli, c. ; Montoneri, E. Synth. Met. 2002, 128, 91 -95. (99) 2otti, G. ; Gallazzi, M. c . ; 2erbi, G. ; Meille, S. V. Synth. Met. 1995, 73, 2 1 7-225. ( 100) Ono, N.; Okumura, H. ; Murashima, T. Heteroat. Chem. 2001, 12, 4 14-4 1 7. ( 10 1) Luo, Y ; Ruud, K. ; Norman, P . ; Jonsson, D . ; Agren, H. J. Phys. Chem. B 1998, 102, 1 710- 1 7 12. ( 102) Elandaloussi, E. H. ; Frere, P . ; Roncali, 1. Chem. Commun. (Cambridge) 1997, 30 1 - 302. ( t03) Wagner, P . ; Aubert, P . -H. ; Lutsen, L.; Vanderzande, D. Electrochem. Commun. 2002, 4, 912-91 6. ( 104) Henckens, A ; Knipper, M. ; Polec, I . ; Manca, 1 . ; Lutsen, L . ; Vanderzande, D. Thin Solid Films 2004, 451-452, 572-579. ( 105) Videlot, c . ; Ackermann, 1 . ; Blanchard, P . ; Raimundo, 1 . -M. ; Frere, P . ; Allain, M. ; de Bettignies, R ; Levillain, E . ; Roncali, 1. Adv. Mater. 2003, 15, 306-3 1 0. 268 ( 1 06) Jestin, I . ; Frere, P . ; Blanchard, P . ; Roncali, 1. Angew. Chem. , Int. Ed. 1998, 37, 942- 945 . ( 107) Levillain, E. ; Roncali, 1. J. Am. Chem. Soc. 1999, 121, 8760-8765. ( 108) Montenegro, M. I . ; Queiros, M. A ; Daschbach, 1 . L. , Eds. Microelectrodes: theory and applications; Kluwer Academic Publishers: London, 199 1 ; Vol. 197. ( 1 09) Michalitsch, R. ; El Kassmi, A; Yassar, A; Lang, P . ; Gamier, F. J Electroanal. Chem. 1998, 457, 129-139. ( 1 1 0) Sezai Sarac, A; Evans, V.; Serantoni, M. ; Clohessy, 1.; Cunnane, V. 1. Surf Coat. Technol. 2004, 182, 7- 13 . ( 1 1 1) Girotto, E. M. ; Casalbore-Miceli, G. ; Carnaioni, N. ; de Paoli, M. A; Fichera, A M. ; Belobrzeckaja, L.; Gallazzi, M. C. J Mater. Chem. 2001 , 1 1 , 1 072- 1076. ( 1 12) Schenning, A P. H. 1.; Kilbinger, A F. M. ; B iscarini, F. ; Cavallini, M. ; Cooper, H. 1 . ; Derrick, P . 1.; Feast, W. 1. ; Lazzaroni, R. ; Leclere, P . ; McDonell, L. A ; Meijer, E . W. ; Meskers, S. C . 1. J. Am. Chem. Soc. 2002, 124, 1269- 1 275 . ( 1 1 3) Smith, 1. R. ; Ratcliffe, N. M. ; Carnpbell, S. A Synth. Met. 1995, 73, 1 7 1- 182. ( 1 1 4) Surni, N . ; Nakanishi, H . ; Veno, S . ; Takirniya, K.; Aso, Y ; Otsubo, T. Bull. Chem. Soc. Jpn. 20 1, 74, 979-988. ( 1 1 5) Folch, I . ; Borros, S . ; Arnabilino, D. B. ; Veciana, J. J Mass Spectrom. 2000, 35, 550- 555 . ( 1 1 6) S iuzdak, G. Mass spectroscopy for biotechnology; Academic Press: San Diego, 1996. ( 1 1 7) Byrd, H . C. M. ; McEwen, C. N . Anal. Chem. 2000, 72, 4568-45 76. ( 1 1 8) Chaudhary, A K.; Critchley, G. ; Diaf, A; Beckrnan, E. 1. ; Russell, A 1 . Macromolecules 1996, 29, 22 13-222 l . ( 1 19) Dey, M. ; Castoro, 1. A ; Wilkins, C. L. Anal. Chem. 1995, 67, 1 575- 1 579. ( 120) Montaudo, G. ; Garozzo, D. ; Montaudo, M. S . ; Puglisi, C . ; Sarnperi, F. Macromolecules 1995, 28, 7983-7989. ( 12 1 ) Nielen, M. W. F. Mass Spectrom. Rev. 1999, 18, 309-344. ( 122) Axelsson, 1 . ; Scrivener, E . ; Hacklleton, D. M. ; Derrick, P. 1. Macromolecules 1996, 29, 8875-8882. ( l23) De Paoli, M.-A ; Nogueira, A F . ; Machado, D. A; Longo, C. Electrochim. Acta 2001 , 46, 4243-4249. ( 124) Lec1erc, M. Can. Chem. News 2001 , 53, 26-27. ( 125) Skompska, M. ; Szkurlat, A Electrochim. Acta 2001 , 46, 4007-40 1 5 . 269 ( 1 26) Lee, W. J . ; Kim, Y 1 . ; Jung, M. 0.; Kim, D. H . ; Cho, D. L. ; Kaang, S. Synth. Met. 2001 , 123, 327-333. ( 127) Trznadel, M. ; Pron, A; Zagorska, M.; Chrzaszcz, R. ; Pielichowski, 1 . Macromolecules 1998, 31, 505 1 -5058. ( 1 28) Gambhir, S . ; Wagner, K; Officer, D. L. Synth. Met. 2005, Submittedfor Publication. ( 129) Doshi, 1 . ; Reneker, D. H. J. Electrostatics 1995, 35, 1 5 1- 1 60. ( 130) Fong, H . ; Chun, I . ; Reneker, D. H. Polymer 1999, 40, 4585-4592. ( 13 1) Granstrom, M.; Petritsch, K ; Arias, A c . ; Lux, A; Andersson, M. R.; Friend, R. H. Nature (London) 1 998, 3 95, 257-260. ( 132) Wallace, G. G.; Too, C. 0.; Officer, D. L. ; Dastoor, P . C . MRS Bull. 2005, 30, 46-49. ( 133) Lemaire, M. ; Garreau, R. ; Delabouglise, D . ; Roncali, 1 . ; Youssoufi, H. K ; Gamier, F. New J. Chem. 1990, 14, 3 59-364. ( 134) Andreani, F . ; Salatelli, E. ; Lanzi, M Polymer 1996, 3 7, 66 1 -66 5 . ( 135) Amou, S . ; Haba, O. ; Shirato, K ; Hayakawa, T.; Ueda, M.; Takeuchi, K; Asai, M. J Polym. Sci., Part A: Polym. Chem. 1999, 37, 1 943-1948. ( 136) Laakso, 1 . ; Jarvinen, H . ; Skagerberg, B. Synth. Met. 1993, 55, 1 204- 1208. ( 1 37) Bauer, B. J . ; Flynn, K M.; Vogt, B. D. Polym. Mater. Sci. Eng. 2004, 90, 457-458. ( 138) Bier, M. E. Polym. Mater. Sci. Eng. 2003, 88, 1 7- 18 . ( 139) Grant, D. K ; Jolley, K W.; Officer, D. L. ; Gordon, K C. ; Clarke, T . M . Org. Biomol. Chem. 2005, 3, 2008-201 5. ( 140) Clarke, T. M ; Gordon, K C . ; Officer, D. L. ; Grant, D. K Manuscript in preparation. ( 14 1) Audebert, P . ; Catel, 1 . M ; Le Coustumer, G. ; Duchenet, V. ; Hapiot, P . J Phys. Chem. B 1998, 1 02, 866 1 -8669. ( 142) Wei, Y ; Chan, C. c . ; Tian, J . ; Jang, G. W. ; Hsueh, K. F. Chem. Mater. 1991 , 3, 888- 897. ( 143) Xu, Z. G. ; Horowitz, G. J Electroanal. Chem. 1992, 335, 1 23- 134. ( 144) Ronca1i, 1 . ; Gorgues, A; Jubault, M. Chem. Mater. 1993, 5, 1456- 1464. ( 145) Harada, H . ; Fuchigarni, T. ; Nonaka, T. J Electroanal. Chem. Interfacial Electrochem. 1991 , 303, 1 39-150. ( 146) MacDiarmid, A G.; Zheng, W. MRS Bull. 1997, 22, 24-30. ( 147) Pringle, 1 . M. ; Forsyth, M. ; MacFarlane, D. R.; Wagner, K. ; Hall, S. B . ; Officer, D. L. Polymer 2005, 46, 2047-2058. 270 ( 1 48) Rapta, P.; Petr, A ; Dunsch, L . Synth. Met. 2001 , 1 1 9, 409-4 1 0. ( 1 49) Zotti, G. ; Schiavon, G. Synth. Met. 1 989, 3 1 , 3 4 7-3 57. ( 1 5 0) Yoshino, K; Onoda, M.; Manda, Y.; Yokoyama, M . Jpn. J Appl. Phys., Part 2. 1 988, 2 7, L l 606-L l 608. ( 1 5 1) Zotti, G. ; Schiavon, G. ; Zecchin, S . Synth. Met. 1 995, 72, 275-28 l . ( 1 52) Domagala, W. ; Lapkowski, M . ; Guillerez, S . ; Bidan, G. Electrochim. Acta 2003, 48, 2379-2388. ( 1 53) Sun, L. ; Goerner, H. J. Phys. Chem. 1 993, 9 7, 1 1 1 86- 1 1 193. ( 1 54) Wagner, P . ; Officer, D. L. Synth. Met. 20 5, Accepted for publication. ( 1 5 5) Clarke, T. M. 2004, Unpublished work. ( 1 5 6) Wagner, P . ; Ballantyne, A M. ; Jolley, K. W. ; Officer, D. L. Manuscript in preparation 2005. ( 1 5 7) Wagner, P . ; Officer, D. L. 2004, Unpublished ",.,.ork. ( 1 5 8) Kitani, A ; Yano, 1 . ; Sasaki, K Chem. Let!. 1 984, 1 56 5 - 1 5 66. ( 1 59) Yassar, A; Roncali, 1 . ; Gamier, F. Macromolecules 1989, 22, 804-809. ( 1 60) C asado, 1 . ; Miller, L. L. ; Mann, K R. ; Pappenfus, T. M . ; Kanemitsu, Y . ; Orti, E. ; Virue1a, P. M. ; Pou-Amerigo, R . ; Hemandez, Y. ; Lopez Navarrete, 1. T. J Phys. Chem. B 2002, 106, 3872-388 l . ( 1 6 1) Andersson, M . R . ; Mammo, W . ; Olinga, T. ; Svernsson, M . ; Theander, M . ; Inganas, O. Synth. Met. 1999, 101 , 1 1 - 1 2 . ( 1 62) Casalbore-Miceli, G. ; Camaioni, N. ; Gallazzi, M. c. ; Albertin, L . ; Fichera, A M . ; Geri , A; Girotto, E. M. Synth. Met. 2001 , 125, 307-3 1 l . ( 1 63) Yamamoto, T. ; Kashiwazaki, A ; Kato, K Makromol. Chem. 1 989, 190, 1 649- 1 654. ( 1 64) Bredas, 1 . L. ; Silbey, R.; Boudreaux, D. S.; Chance, R . R . J Am. Chem. Soc. 1983, 1 05, 655 5-65 59. ( 1 65) Casalbore-Miceli, G. ; Camaioni, N.; C atellani, M . ; Gallazzi, M . C. Synth. Met. 1998, 95, 2 1 1 -2 1 5 . ( 1 66) Rasch, B . ; Vie1stich, W . J. Electroanal. Chem. 1 994, 3 70, 1 09- 1 1 7. ( 1 67) Kumar, A ; Reynolds, 1. R. Macromolecules 1 996, 29, 7629-7630. ( 1 68) Andersson, M. R . ; Pei, Q. ; Hjertberg, T . ; Inganaes, O . ; Wennerstroem, 0.; Osterholm, J . E . Synth. Met. 1 993, 55, 1 227- 1 23 l . ( 1 69) Kemp, W . OrganiC spectroscopy, 3rd Ed ; Macrnillan Education Ltd. : London, 1 99 1 . 271 ( 1 70) Groenendaal, L. B . ; Zotti, G. ; Aubert, P . -H . ; Waybright, S. M . ; Reynolds, 1. R Adv. Mater. 2003, 1 5, 855-879. ( 1 7 1 ) Eales, R M. ; Hillman, A R. J. Mater. Sci. 1990, 25, 3 806-3 8 1 3 . ( 1 72) Tourillon, G. ; Gamier, F . J. Electroanal. Chem. Interfacial Electrochem. 1984, 1 61 , 407-4 1 4 . ( 1 73) Cotton, F . A; Wilkinson, G. Advanced inorganic chemistry, 5 th Ed. ; John Wiley and Sons: New York, 1988. ( 1 74) Onoda, M. ; Okamoto, T. ; Tada, K ; Nakayama, H. Jpn. J. Appl. Phys. , Part 2. 1 999, 38, L l 070-L l 072. ( 1 75) Shaheen, S . E. ; Ginley, D. S . ; Jabbour, G. E . MRS Bull. 1995, 30, 10- 1 5 . ( 1 76) B rabec, C . 1. Sol. Energy Mater. 2004, 83, 273-292. ( 1 77) Green, M . A; Emery, K ; King, D. L . ; I gari, S . ; Warta, W. Progress in Photovoltaics 2003, 1 1 , 347-3 5 2 . ( 1 78) Janssen, R. A 1 . ; Hummelen, 1. c.; Sariciftci, N . S. MRS Bull. 20OS, 3 0 , 3 3 -36. ( 1 79) Gratzel, M . J. Photochem. Photobiol. , A 2004, 1 64, 3- 14. ( 1 80) Too, C . 0.; Wallace, G. G. ; Burrell, A K.; Collis, G. E.; Officer, D. L . ; Boge, E. W.; Brodie, S . G. ; Evans, E . 1. Synth. Met. 2001 , 1 23, 53 -60. ( 1 8 1 ) Brabec, C. J . ; Sariciftci, S. N . Monatsh. Chem. 2001 , 132, 42 1 -43 1 . ( 1 82) Li, A D. Q . ; Li, L . S . 1. Phys. Chem. B 2004, 1 08, 1 2 842- 1 2850. ( 1 83) Ho, P. K H . ; Thomas, D. S . ; Friend, R H . ; Tessler, N. Science (Washington, D. C.) 1 999, 285, 233-236. ( 1 84) Skotheim, T. A; Feldberg, S. W.; Arrnand, M . B . Journal de Physique, Colloque 1 983, 6 1 5-620. ( 1 85) Wallace, G. G.; Dastoor, P . c. ; Officer, D. L . ; Too, C. 0. Chem. Innovation 2000, 30, 1 4-22. ( 1 86) Campbell, W. M . 20OS. Personal communication. ( 1 87) Spinks, G. M . ; Liu, L. ; Wallace, G. G. ; Zhou, D. Adv. Funct. Mater. 2002, 1 2, 43 7- 440. ( 1 88) Otero, T. F. ; Sansinena, 1. M. Bioelectrochemistry and Bioenergetics 1995, 38, 4 1 1 - 4 1 4. ( 1 89) Zhou, D. ; Wallace, G. G.; Spinks, G. M.; Liu, L. ; Cowan, R ; Saunders, E . ; Newbold, C . Synth. Met. 2003, 1 35- 1 3 6, 39-40. ( 1 90) Della Santa, A ; De Rossi, D.; Mazzoldi, A Synth. Met. 1997, 90, 93-100. 272 ( 19 1 ) Spinks, G. M . ; Wallace, G. G. ; Liu, L . ; Zhou, D. Macromoi. Symp. 2003, 192, 16 1 - 1 69. ( 1 92) Chen, X.; Xing, K -Z. ; Inganaes, O. Chem. Mater. 1 996, 8, 2439-2443 . ( 1 93) Xi, B . ; Truong, V.-T.; Whitten, P . ; Ding, 1 . ; Spinks, G. M . ; Wallace, G. G. Proceedings ofSPIE-The International Society for Optical Engineering 2005, 5649, 1 3 7- 1 44. ( 1 94) Fuchiwaki, M. ; Takashima, W.; Kaneto, K Molecular Crystals and Liquid Crystals Science and Technology, Section A : Molecular Crystals and liqUid Crystals 2002, 374, 5 13-520. ( 1 95) Smela, E . AdvancedMaterials (Weinheim, Germany) 2003, 1 5, 48 1 -494. ( 196) Ding, 1 . ; Liu, L . ; Sp inks , G. M . ; Zhou, D . ; Wallace, G. G. ; Gillespie, 1. Synth. Met. 2003, 138, 391 -398. ( 1 97) Wu, Y 2004. Personal communication. ( 1 98) Ding, 1. PhD Thesis, IP RI, University of Wo lIon gong, 2001 . ( 199) Otero, T. F. ; Angulo, E . ; Rodriguez, J . ; Santamaria, C. J. Electroanai. Chem. 1 992 , 341, 369-375. (200) Takashirna, W.; Dufour, B . ; Pandey, S . S . ; Kaneto, K . ; Pron, A Sensors and Actuators, B: Chemical 2004, B99, 60 1 -607. (20 1) Gofer, Y ; Sarker, H. ; KilIian, 1 . G.; Giaccai, J . ; Poebler, T. 0. ; Searson, P. C . Biomedical instrumentation & technology / Association for the Advancement of Medical Instrumentation 1998, 32, 33-38. (202) Rehan, H . H . J. Power Sources 2003, 1 13, 57-6 1 . (203) Maclnnes, D. , Jr. ; Druy, M. A; Nigrey, P . 1 . ; Nairns, D . P . ; MacDiannid, A G. ; Heeger, A. 1. J. Chem. Soc., Chem. Commun. 1981 , 3 1 7-3 1 9. (204) Misoska, Y. ; Price, W . ; Ralph, S . ; Wallace, G. Synth. Met. 2001 , 121, 1 5 0 1 - 1 5 02. (205) Misoska, V. 2004. Personal communication. (206) Rajesh, B . ; Ravindranathan Thampi, K ; Bonard, J. M . ; Viswanathan, B. J. Mater. Chem. 2000, 10, 1 757- 1 759. (207) Wilson, J. N . ; Bangcuyo, C . G. ; Erdogan, B . ; Myrick, M. L. ; Bunz, U. H. F. Macromolecules 2003, 36, 1 426- 1428. (208) MacDiannid, A G.; Jones, W. E . ; Norris, I. D.; Gao, J.; Johnson, A. T. ; Pinto, N. 1 . ; Hone, 1 . ; Han, B. ; Ko, F. K ; Okuzaki, H . ; Llaguno, M . Synth. Met. 2001 , 1 19, 27-30. (209) Kameoka, J . ; C raighead, H. G. Appi. Phys. Let!. 2003, 83, 3 7 1 -3 73 . 273