<%BANNER%>

Optical and transport properties of conjugated polymers and their application to devices

University of Florida Institutional Repository

PAGE 1

OPTICAL AND TRANSPORT PROPERTIES OF CONJUGATED POLYMERS AND THEIR APPLICATION TO ELECTROCHROMIC DEVICES By IRINA SCHWENDEMAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2002

PAGE 2

Copyright 2002 by Irina Schwendeman

PAGE 3

To John, Laura and Ileana

PAGE 4

ACKNOWLEDGMENTS First, I would like to thank my advisor, Dr. John Reynolds, for sending me a nice contact letter in 1996 when I was deciding what university to join for graduate studies. Since then, his continuous support and advice, which were not limited to science, made this work possible. I greatly appreciate that Dr. Reynolds was there when I needed counseling, and seemed always interested and appreciative of my work. Learning new concepts is not hard and can be done independently, but becoming a good scientist requires a great deal of fine guidance and effort, and for this I thank Dr. Reynolds the most. Several professors from Polytechnical University of Bucharest had a major contribution to my interest in polymer science. Alexandru Stefan, my undergraduate advisor, Nicolae Cobianu, my masters advisor, and Mihai Dimonie, who was about to become my Ph. D. advisor, prepared me for the next step in my scientist life. For being outstanding coworkers and friends, and helping me to adjust to a new country, I thank Jennifer and David Irvin, Fatma Sotzing and Hiep Ly. I was fortunate to work with two wonderful undergraduate students, Jessica Hancock and Roberta Hickman, who made my work day twice as productive and ten times more fun. I do not attempt to point out which experiments they helped me with because their contributions are spread over each project presented in this thesis. Room 300 in Leigh Hall is a special environment, where many people come and go in a short period of time. Over the last two to three years, besides myself, there has been only one other long term resident of this lab, Avni Argun. I would like to thank Avni for taking over all the lab duties, for his iv

PAGE 5

immediate help when I needed it and for the comfortable thought that there is somebody to count on around. I thank Chris Thomas, who took the time to teach me the basis of electrochemistry and spectroelectrochemistry like nobody else, despite his approaching oral examination. I also thank Chris Thomas, Phillippe Schottland and P.H. Aubert for being so knowledgeable and answering any question that I might have. Kyukwan Zong, Dean Welsh and Carl Gaupp have significantly contributed to this work by providing the monomers and allowing me to take the fun part of the work, which is the characterization of new compounds and their application to devices. Thanks go to Mohamed Bouguettaya and Charlotte Cutler for fun and relaxing conversations over a coffee at Java. I would like to thank Barry Thompson, Nisha Ananthakrishnan and all of the other Reynolds group members for being helpful when I needed it and generally fun to be around. My work is characterized by having multiple collaborations, especially with physics departments (You are turning into a physicist! Frechet, Fall 2000). Dr. Arthur Epstein from Ohio State University made a great impact on my work. Data presented in Chapter 3 were collected by his students as follows: Alexey Saprigin did the reflectance and the X-ray measurements, Won-Pil Lee did the low temperature mw and dc studies, and Keith Brenneman performed EPR experiments. I would like to thank Dr. Epstein and his group, and especially Alexey, for being such nice hosts when I visited Columbus, OH. Dr. Tanner from the Physics Department at UF has been like a second advisor for me, and was always extremely nice and helpful. The collaboration with his group resulted in the reflectance/transmittance data from Chapter 4 and the variable reflectance mirror project presented in Chapter 5. Jungseek Hwang from Dr. Tanners group (my three year long collaborator), and recently Maria Nikolou and Matt Cornick, performed the above v

PAGE 6

experiments. Bert Groenendaal from AGFA provided the alkyl-substituted EDOTs and EDOT-F. The greatest acknowledgment goes to my family for their love and understanding that make my life complete. To John, thanks go for being such a loving husband and good team member, and for sharing with me every joy and every worry. My daughter, Laura, actively contributed to this thesis by being such an adorable baby and sleeping through the night when we needed to work, and loving us unconditionally even when we were tired and not in the mood to play. I thank my mom and dad, my aunt and grandmother for giving me the intellectual and emotional guidance that made me who I am. vi

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT.....................................................................................................................xiv CHAPTERS 1 INTRODUCTION............................................................................................................1 1.1 Brief History of Conducting Polymers....................................................................1 1.2 Metals, Semiconductors and Insulators...................................................................5 1.3 Electrochemical Polymerization............................................................................13 1.4 Doping Induced Property Changes........................................................................16 1.5 Applications...........................................................................................................20 1.6 Structure of this Thesis..........................................................................................26 2 EXPERIMENTAL TECHNIQUES................................................................................29 2.1 Conductivity Measurements..................................................................................29 2.2 Electrochemical Methods.......................................................................................36 2.3 Spectroelectrochemistry and Other Optical Measurements...................................41 2.4 Colorimetry............................................................................................................43 2.5 Surface Analysis....................................................................................................45 2.6 Device Construction...............................................................................................47 2.7 Purification of Laboratory Chemicals and Materials.............................................51 3 ELECTRONIC TRANSPORT AND OPTICAL PROPERTIES OF PXDOT AND PXDOP FREE-STANDING FILMS...........................................................................53 3.1 Introduction............................................................................................................53 3.2 Mechanisms for ElectronicTransport.....................................................................54 3.3 Elementary Excitations in Conducting Polymers..................................................60 3.4 Metal-Insulator Transition in Conducting Polymers.............................................62 3.5 Electrochemical Synthesis of Free-Standing Films...............................................64 vii

PAGE 8

3.6 Temperature Dependence of the Conductivity......................................................67 3.7 Microwave Experiments........................................................................................71 3.8 Magnetic Susceptibility.........................................................................................73 3.9 X-ray Measurements..............................................................................................75 3.10 Optical Conductivity............................................................................................75 3.11 Room Temperature Conductivities of Alkyl Substituted PEDOTs.....................78 4 POLYMER THIN FILM OPTICS..................................................................................81 4.1 Introduction............................................................................................................81 4.2 Surface Analysis....................................................................................................82 4.3 Fundamental Spectroelectrochemistry...................................................................86 4.4 Reflectance and Transmittance Spectra.................................................................89 4.5 The Origin of the Fine Structure in Neutral Polymers...........................................97 4.6 Spectroelectrochemical Data of PXDOT Series..................................................103 5 VARIABLE REFLECTANCE MIRRORS..................................................................117 5.1 Introduction..........................................................................................................117 5.2 Devices as Platforms to Study Electronic Spectra...............................................118 5.3 In-situ Reflectivity Measurements.......................................................................121 5.4 Active Polymer Layer Thickness Optimization...................................................124 5.5 Switching Time....................................................................................................127 5.6 Long-Term Switching Stability...........................................................................129 5.7 O-H and C-H Removal........................................................................................130 6 ABSORPTION-TRANSMISSIVE WINDOWS..........................................................134 6.1 Introduction..........................................................................................................134 6.2 Absorption/Transmission Device Construction...................................................135 6.3 Cathodically Coloring Polymers..........................................................................137 6.4 Anodically Coloring Polymers............................................................................139 6.5 Matching Two Complementary Coloring Polymers............................................142 6.6 High Contrast ECDs............................................................................................146 6.7 Broadband Absorption Devices: Laminate Alternative.......................................153 6.8 Broadband Absorption Devices: PEDOT-F Alternative......................................159 6.9 Conclusions..........................................................................................................164 LIST OF REFERENCES.................................................................................................166 BIOGRAPHICAL SKETCH...........................................................................................178 viii

PAGE 9

LIST OF TABLES Table page 1.1 Qualitative properties of conducting polymers in two extreme redox states...............17 1.2 Applications of conducting polymers..........................................................................21 3.1 Electrochemical synthesis of free-standing PXDOT films..........................................65 3.2 Transport properties of PXDOT and PXDOP free-standing films..............................68 4.1 The electronic structure of PEDOT, PProDOT, and PProDOT-Me 2 in their neutral, lightly doped and fully doped states.........................................................................93 4.2 Spectroelectrochemical results for PXDOT series....................................................106 6.1 Band gap and monomer oxidation potential values for a series of propylenedioxy-derivatized pyrroles and thiophenes.......................................................................142 ix

PAGE 10

LIST OF FIGURES Figure page 1.1 Common conjugated polymers......................................................................................2 1.2 Energy bands developing in metallic sodium................................................................5 1.3 Description of materials according to their band structure............................................6 1.4 Conducting properties of metals, semiconductors and insulators..................................8 1.5 Evolution of the band gap in PAc................................................................................10 1.6 Magnitude of room temperature conductivity for different types of conducting polymers ..................................................................................................................12 1.7 Electropolymerization mechanism................................... ...........................................14 1.8 Doping mechanisms in conjugated polymers and their applications...........................18 1.9 Doping methods in conjugated polymers ...................................................................19 1.10 General designs of conducting polymer devices.......................................................23 2.1 Four-probe conductivity methods................................................................................32 2.2 Cyclic voltammogram of BEDOT-B(OG) 2 .................................................................37 2.3 Electrochemical cell setup...........................................................................................39 2.4 Electrode potential relationship between common reference electrodes.....................40 2.5 x-y chromaticity diagram.............................................................................................44 3.1 Schematic drawing of density of states........................................................................55 3.2 Scanning force microscopy image of polymer chains.................................................58 3.3 Schematic drawing on the conducting polymer structure............................................59 3.4 Band diagrams for neutral and positive solitons..........................................................60 x

PAGE 11

3.5 Polaron and bipolarons in non-degenerate ground state polymers..............................61 3.6 Temperature dependence of the conductivity..............................................................70 3.7 The conductivity dependence of T -1/4 in PXDOT films...............................................71 3.8 Temperature dependence of the dielectric constant in PXDOT films.........................72 3.9 Temperature dependence of the mw and dc conductivity............................................73 3.10 EPR Susceptibility of Doped PXDOT Films.............................................................74 3.11 X-ray diffraction........................................................................................................76 3.12 Optical properties of PXDOT films...........................................................................77 3.13 Conductivity of alkyl substituted PEDOTs................................................................80 4.1 The model of film formation........................................................................................84 4.2 Surface morphology of PANI films ...........................................................................84 4.3 Surface morphology.....................................................................................................85 4.4 Doping induced electronic transitions in conducting polymers...................................87 4.5 Neutral PProDOT stability studies...............................................................................90 4.6 Schematic with polymer/ITO/glass layers...................................................................91 4.7 Absorption coefficients of: A) PEDOT, B) PProDOT, C) PProDOT-Me 2 ................92 4.8 Infrared active vibrational modes................................................................................95 4.9 Solid-State UV-Vis spectra of regioregular poly(3-dodecylthiophene)......................98 4.10 Schematic representation of the vibronic transition.................................................100 4.11 Absorption spectra of PPV oligomers with different molecular weights................101 4.12 Visible absorption spectra........................................................................................105 4.13 Band gap dependence of the side chain length in alkyl-substituted PEDOTs.........108 4.14 Four electrochemical polymerization routes to PProDOT.......................................110 4.15 Potential sweep growth of PProDOT.......................................................................111 4.16 Regioregularity of alkyl-substituted PEDOTs.........................................................113 xi

PAGE 12

5.1 Side-view schematic diagram of a dual polymer electrochromic device..................119 5.2 Top-view photograph of a PProDOT-Me2/PBEDOT-NMeCz dual polymer electrochromic device ...........................................................................................120 5.3 Reflectance of the PProDOT-Me2/PBEDOT-NMeCz dual polym device .....................................................................................................................123 er electrochromic 5.4 Device contrast as a function of frequency................................................................125 5.5 Active layer thickness dependence of the electrochromic contrast in PEDOT based devices,...................................................................................................................126 5.6 Switching time of PProDOT-Me2 based device recorded at two IR frequencies......128 5.7 Reflectance as a function of number of deep double potential switches for PProDOT-Me2 based device....................................................................................................130 5.8 Transmittance spectra of two gel electrolyte batches................................................132 5.9 Reflectance spectra of a PProDOT-Me 2 based device...............................................133 6.1 Variable transmittance window schematic................................................................136 6.2 Cathodically and anodically coloring polymers .......................................................138 6.3 Trimer of A) PProDOP and B) PProDOP-NPrS Hyperchem illustrations................141 6.4 The superimposition of the UV-Vis-NIR spectra of the individual polymer films deposited on ITO/glass substrates..........................................................................143 6.5 Thickness as a function of charge density passed during electrodeposition..............145 6.6 AFM topographic image representing a scratch through a polymer film..................146 6.7 Transmittance spectra and photographs of devices using PProDOT-Me2 as the cathodically coloring polymer................................................................................147 6.8 Transmittance as a function of switching time..........................................................149 6.9 Hue and saturation (x-y track) for the device B.........................................................150 6.10 Luminance PProDOT-Me2/ PProDOP-NPrS device...............................................151 6.11 Lifetime of the PProDOT-Me2/PProDOP-NPrS device..........................................152 6.12 Laminate device schematic......................................................................................153 6.13 PEDOP/PProDOT-Me 2 bilayer electrode absorption spectra..................................155 xii

PAGE 13

6.14 Charge trapping mechanism in PEDOP/PProDOT-Me 2 bilayer electrode..............156 6.15 Transmittance spectra of the PEDOP/PProDOT-Me 2 /PBEDOT-NMeCz laminate device.....................................................................................................................157 6.16 Luminance analysis of the PEDOP/PProDOT-Me 2 /PBEDOT-NMeCz laminate device.....................................................................................................................158 6.17 Absorption spectra of PEDOT-F in 0.1 M TBAP/ACN..........................................159 6.18 Spectroelectrochemical analysis of PEDOT-F/PBEDOT-NMeCz device..............162 6.19 Colorimetric analysis of PEDOT-F/PBEDOT-NMeCz device...............................163 xiii

PAGE 14

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy OPTICAL AND TRANSPORT PROPERTIES OF CONJUGATED POLYMERS AND THEIR APPLICATION TO ELECTROCHROMIC DEVICES By Irina Schwendeman December 2002 Chair: Professor John R. Reynolds Major Department: Chemistry This work combines fundamental studies on the electrical and optical properties of conducting polymers from the poly(3,4-alkylenedioxythiophene) (PXDOT) and poly(3,4-alkylenedioxypyrrole) (PXDOP) class with more application-driven studies of electrochromic devices. Understanding the mechanism behind achieving a high conductivity allows for an accurate design of the next generation of organic materials with metal-like electrical properties. Optical reflectance, microwave and dc conductivity, and EPR and X-ray measurements are employed to gain insight into the electronic transport of doped, free-standing polymer films. These materials electronic properties can be tuned from metallic to insulator by varying the dopant nature and monomer unit planarity. For example, PEDOT-PF 6 shows a metallic-like transport behavior, PEDOP-CF 3 SO 3 and PEDOT-CF 3 SO 3 are in the critical regime, whereas PEDOT-(CF 3 SO 2 ) 2 N, PProDOT-CF 3 SO 3 and PProDOT-PF 6 are on the insulator side of metal to insulator transition and show semiconducting properties. xiv

PAGE 15

Since many promising applications revolve around dynamic changes in the optical properties of electroactive thin films, this work focuses on elucidating the doping-induced electronic structure and infrared active vibrational modes (IRAV) in PXDOT films. Transmission and reflectance spectra spanning the UV, visible, nearand mid-IR regions of the spectrum reveal polaron and bipolaron signatures as the doping level increases. Neutral polymers have unique visible absorption spectrum envelopes, such as a broad single peak or multiple sharp peaks, as a function of the nature of the monomer unit. A discussion on the existing theories for the origin of this fine structure supported by data taken for a basis set of sixteen PXDOT thin films suggests that vibronic coupling of excited electrons with a vibrational mode of the monomer unit (symmetric, in plane C =C stretch is responsible for the appearance of this absorption peak splitting. A profound understanding of these materials optical and electrochemical properties allows the construction of electrochromic devices with enhanced contrast, long term switching stability and fast optical response. Variable reflectance electrochromic devices based on PXDOTs have been optimized to exhibit contrast ratios of 60-80% from the visible to mid-IR regions. They exhibit remarkable lifetimes, with no significant loss in electrochemical contrast after 10,000 redox cycles. The use of carefully designed optically complementary polymers PProDOT-Me 2 and PProDOP-NPrS is a promising route for achieving a high level of control over the color, brightness, and redox stability of a transmission/absorption window. These devices operate at low applied voltages ( 1.0 V), exhibiting an optical contrast of up to 70% at max as well as an overall luminance change of 53%. They retain 86% of their initial color (96% retention after break-in) after 20,000 redox cycles. xv

PAGE 16

CHAPTER 1 INTRODUCTION 1.1 Brief History of Conducting Polymers Twenty years after the discovery by Heeger, MacDiarmid and Shirakawa that an organic conjugated polymer can reach metallic-like electrical conductivity upon doping,1,2 the field of conjugated polymers has enjoyed a tremendous development culminating in 2000 when the Nobel Prize for Chemistry was awarded to the above researchers for launching a new era in polymeric materials. Conjugated polymer synthesis reaches as far back as 1862, when H. Letheby first synthesized polyaniline (PAni, 3).3 Known as aniline black, this material was formed by oxidation of aniline under mild conditions and was used in the printing industry.4 The first polymerization of acetylene to form polyacetylene (PAc, 1) was reported in 1958 by Natta and coworkers.5 Because PAc was obtained as an insoluble and infusible powder, the material received little attention at that time. The idea that conjugated polymers could be good electrical conductors has roots back to the 1960s when MacDiarmid and others discovered that poly(sulfurnitride) (SN) x a polymeric inorganic explosive,6 has a high conductivity.7 The interesting electrical properties of (SN) x represented a step towards conducting polymers as they are known today. The modern era of conducting polymers began at the end of 1970s when films of PAc were found to exhibit a 12 order of magnitude increase in electrical conductivity 1

PAGE 17

2 when exposed to iodine vapors.1,2 The procedure for synthesizing PAc was based upon a route discovered in 1974 by Shirakawa through an accidental addition of 1,000 times more catalyst during the polymerization of acetylene. Although initially it was thought that PAc would replace dense metals in air and space applications, its instability to environmental conditions constituted a major obstacle for any practical use. However, PAc, being the simplest model from this class, remains the archetype of conducting polymers and is still subject to much theoretical and experimental work. n n NHn NnH Sn SOOnPAc, 1PPV, 2PAni, 3PPy, 4PT, 5PEDOT, 6 n n NHn NnH Sn SOOnPAc, 1PPV, 2PAni, 3PPy, 4PT, 5PEDOT, 6 Figure 1.1 Common conjugated polymers The opportunity to synthesize new conducting polymers with improved properties began to attract the attention of synthetic chemists in the early 1980s. During this time, the discovery that polypyrrole (PPy, 4) can be obtained as highly conducting and homogeneous, free-standing films via oxidative electropolymerization8 focused the research efforts towards the development of conjugated poly(heterocycles). Electrochemical polymerization was rapidly extended to other aromatic compounds such as aniline, thiophene, furan, indole, carbazole, indole, azulene, pyrene, and fluorene.9,10 Although exhibiting lower conductivities than doped PAc, the most common conjugated poly(heterocycles) shown in Figure 1.1 (compounds 4-6) posses superior environmental

PAGE 18

3 stabilities in their p-doped form, along with other interesting properties such as electrochomism, volume change during doping, electroand photoluminescence, etc. 11 Among these so-called first generation conducting polymers, polythiophene (PT, 5) has rapidly become the subject of considerable interest, mainly due to its structural versatility and enhanced stability in both neutral and p-doped states. From a theoretical point of view, PT has been often considered as a model for the study of charge transport in non-degenerate ground state polymers. However, PTs are unstable at the potentials required for their formation. This effect is called PT paradox and means that the polymer degradation competes with its deposition, leading to polymers with a high content of overoxidized, non-electroactive material. A successful strategy to control the properties of PTs involves the modification of the monomer structure to obtain easily polymerizable and eventually solution processable species. This led to the synthesis of a vast family of PT derivatives with varied interesting properties.12,13 As a member of conducting polymers second generation, the 3,4-ethylenedioxy derivative of PT (PEDOT) stands out as an excellent candidate for use in a variety of applications such as electrochromic devices, LEDs, capacitors and sensors. Initially developed to give a processable polymer (as an aqueous dispersion) with a high degree of order due to the lack of and couplings, PEDOT showed other very interesting properties worth pursuing by both industry and academia. In addition to its high conductivity (300-400 S/cm), PEDOT exhibits high electrochromic contrast with the major advantage of being almost transparent in thin, doped films.14,15 Other advantages include low monomer oxidation potential, high stability in the doped form and an ease of derivatization at the ethylenedioxy ring, thus allowing for state-of-the-art tuning of the

PAGE 19

4 materials electronic and optical properties. The industrial use of the polymer as an antistatic layer in photographic films makes it the most widely used conducting polymer to date.16 In this context, the past few years have witnessed the emergence of an impressive variety of EDOT derivatives.16 Functionalization of the monomer with long alkyl or alkoxy pendant groups affords materials that can be processed by common methods including spraying, printing, spin coating or solution casting.17,18 Moreover, the introduction of water-soluble sodium salt units as pendant groups allows water-processability along with self-doping. This type of polymer has been deposited on several substrates via the electrostatic adsorption technique.19-21 To expand the range of substituted 3,4-alkylenedioxy thiophenes, our group synthesized monomers having several ring sizes, such as 7-membered (ProDOT)22 and 8-membered (BuDOT) rings.17 ProDOT has received special attention, as the monomer can be symmetrically derivatized at the central carbon of the propylene bridge, resulting in a regiosymmetric polymer. The bandgap of EDOT-containing polymers can be modified by varying the degree of -overlap along the polymer backbone via steric interactions and by controlling the electronic character of the -system with electron donating or accepting units. The latter is accomplished by using substituents and co-repeat units that adjust the energy of the highest occupied molecular orbital (HOMO) or valence band and the lowest unoccupied molecular orbital (LUMO) or conduction band, thus obtaining polymers with a broad range of colors. In this manner, materials with higher gaps than the PEDOT parent have been prepared, some of which are used as anodically coloring polymers in electrochromic

PAGE 20

5 devices.23-25 Further work using a donor-acceptor methodology has led to low band gap polymers able to pand ndope along with exhibiting multi-color electrochromism.26 1.2 Metals, Semiconductors and Insulators Energy bands. Electrons in an isolated atom are described by their atomic orbital wave functions. Their energies are determined solely by the interaction with the nucleus and other electrons of the atom. When atoms are brought together to form crystals, energy bands evolve, as illustrated in Figure 1.2 for sodium. At infinite separation (Figure 1.2 A), all the atomic orbitals have discrete energies. As the atoms are brought together, their orbitals begin to overlap, and energy bands result (Figure 1.2 B). 04s3p3s2p R=10 3.67 ABC 04s3p3s2p R=10 3.67 ABC Figure 1.2 Energy bands developing in metallic sodium: A) large atomic separation, B) atoms separated by 10 C) actual separation in metallic sodium (from Slater27) When the interatomic distance becomes comparable to or less than the spatial extension of the electronic wavefunction associated with a particular atom, the valence electrons do not belong to any single atom, but to the entire material (Figure 1.2 C). However, the Pauli exclusion principle states that the electrons of a certain atomic orbital

PAGE 21

6 cannot have the same energy. Consequently, the electrons form a band of energy levels with very small energy differences between individual levels.27 Metals, semiconductors and insulators. According to the band theory, the electrical properties of a material are the direct consequence of the energy difference CBCBCBCB VB VB VB VBMetalASemimetalBSemiconductorCInsulatorDSmallEgLarge Eg CBCBCBCB VB VB VB VB VB VB VB VBMetalASemimetalBSemiconductorCInsulatorDSmallEgLarge Eg Figure 1.3 Description of materials according to their band structure: A) Metal, B) Semimetal, C) Semiconductor, D) Insulator between HOMO and LUMO, as well as of the level of band filling and band overlapping. The basic difference between metals and other solids (semiconductors and insulators) is that HOMO is only partially occupied (Figure 1.3 A), whereas this band is filled in semiconductors and insulators (Figure 1.3 C and D). As a consequence of this energy band configuration, there is a sharp distinction between the materials electrical conductivity, in particular as the temperature approaches 0 K.28 Figure 1.4 shows a comparison of electrical conductivities of metals, semiconductors and insulators on a logarithmic scale (A) along with temperature dependence of the conductivity for materials from each conductivity class (B). Electrical conductivity is described by the equation:

PAGE 22

7 = ne where n is the density of charge carriers contributing to conductivity, e is the elementary charge, and is charge carrier mobility (velocity per unit electric field). In a metal, both n and have large values in the range of 10 22 cm -3 and 10 3 cm 2 /Vs,28 respectively, while the elementary charge is 1.60210 -19 C. They combine to give large values of conductivity of 10 5 -10 6 S/cm, as shown in Figure 1.4 A and B (Cu). In addition, the number of charge carriers contributing to conductivity is constant at all temperatures, due to the lack of a band gap between HOMO and LUMO. The progress of electrons in a metal is interrupted after traveling a distance corresponding to mean free path or a time equivalent to the scattering relaxation time. The dominant scattering mechanism is the interaction of itinerant electrons with lattice atoms as they vibrate with thermal energy. Therefore, the mobility of charge carriers, and consequently the conductivity of a metal, increases as the temperature approaches absolute zero and the electron-lattice scattering is diminished. Semiconductors have filled valence bands and empty conduction bands. The gap energy is small as compared to the larger gap in insulators. For intrinsic semiconductors, electrons in the valence band can be thermally activated into the conduction band, and the density of such electrons follows an Arrhenius law. During the activation process, a corresponding hole is created in the valence band and both free electrons and free holes contribute equally to the conductivity. These semiconductors have a positive temperature

PAGE 23

8 AB AB Figure 1.4 Conducting properties of metals, semiconductors and insulators. A) Conductivity scale comparing metals, semiconductor and insulators (adapted from Skotheim11) B) Temperature dependence of the conductivity for four materials. The inset represents the mobility, charge carrier, and semiconductor conductivity variations as a function of temperature (adapted from Epstein29)

PAGE 24

9 dependence of the conductivity (see PPy behavior in Figure 1.4 B). Even though the carrier mobility is affected by the same scattering process as in a metal, the number of charge carries at the certain temperature dominates the overall temperature dependence of the conductivity. In addition, the room temperature conductivity of semiconductors is lower than the conductivity of metals due to a reduced charge carrier density. In extrinsic semiconductors, the number of charge carriers can be increased by introduction of impurities (dopants). In these materials, the carrier density is affected more by dopant nature and concentration than by the temperature. Classical extrinsic semiconductors are silicon-based doped with small amounts of gallium (p-type semiconductor) or arsenic (n-type semiconductor). In insulators, the electrons are strongly localized between the atoms, forming chemical bonds. There is no significant overlap between -orbitals of adjacent atoms and the valence band is completely filled. Furthermore, a large gap separates it from the conduction band, which is not accessible by thermal excitation. For example, polymers having only bonds are good insulators. There is a limited number of materials that have filled valence bands that overlap with conduction bands to generate two incompletely filled bands, as show in Figure 1.3 B and 1.4 B (SN x ). These materials are called semi-metals and they can reach metal-like conductivities. One example from this class is graphite that shows a pronounced anisotropy of conductivity, with a much greater value in the aromatic plane due to a good overlap than the conductivity measured along the stack axis.30 Conducting polymers. Trans-PAc offers an excellent model to illustrate the formation of energy bands.31 The simple chemical structure -(CH 2 ) x implies that each

PAGE 25

10 systems of (CH)x Long chain PAc Eg~ 1.4 eV x =24816Uniform C-C bond order Alternating C-C bond order ABCDE Figure 1.5 Evolution of the band gap in PAc. A) Schematic representation of the -molecular orbitals energy levels with increasing chain length in PAc, (adapted from Chien30) B) uniform C-C bond order in PAc, C) dimerized structure resulting from Peierls distortion, D) cis-PAc, E) degenerate ground state phases in trans-PAc

PAGE 26

11 carbon contributes a single p z electron to the band (HOMO) (Figure 1.5 B). As a result, this band would be half filled, leading to a one-dimensional metal-like conduction along a neutral PAc chain. Figure 1.5.A shows the increase in the orbital overlap as the length of the PAc chain increases. However, experimental studies show that neutral PAc is a semiconductor with a band gap of 1.4-1.5 eV. In the case of trans-PAc, the doubly degenerate ground state is unstable, and undergoes Peierls distortion32 (equivalent to Jahn-Teller effect), resulting in an alternation of bond lengths along the backbone. Further, this process leads to a splitting of the band into an empty conduction band and a fully occupied valence band, thus rendering PAc semiconducting. In the case of cis-PAc (Figure 1.5 D), the ground state structure is non-degenerate and the valence band is filled. Evolution of the band structure from monomer to polymer for conjugated heterocycles is similar to the one sketched in Figure 1.5 A.33 The magnitude of the room temperature conductivity for different types of conducting polymers is shown in Figure 1.6 and compared to the conductivities of copper, platinum, and the range of conductivities of amorphous metals. Data for PAc doped with MoCl 5 FeCl 3 and ClO 4 are represented under the label PAc(Cl) and with iodine, labeled PAc(I).34 As seen in Figure 1.6, the conductivity of PAc exceeds that of the amorphous metals and is nearly identical to platinum conductivity. This is remarkable considering that the density of states at the Fermi level in highly conducting PAc is much smaller than that of conventional metals, and that the polymer is only partially crystalline.35 Samples on the insulator side of the metal to insulator transition (M-I transition) (see Chapter 4 for details) are represented by open symbols. These samples have zero conductivity as the temperature approaches 0 K. The samples in which the

PAGE 27

12 10-210-1100101102103104105106 RT(S/cm)PAc (Cl)PAc (I)PAniPAni blendsPPyPT, PMeTPPVPEDOT Cu Pt Amorphous metals 10-210-1100101102103104105106 RT(S/cm)PAc (Cl)PAc (I)PAniPAni blendsPPyPT, PMeTPPVPEDOT Cu Pt Amorphous metals Figure 1.6 Magnitude of room temperature conductivity for different types of conducting polymers (adapted from Kaiser 34). PEDOT conductivity values are taken from references36-41: Open circles represent samples on the insulator side of M-I transition, while closed circles represent samples on the metallic side of M-I transition. conductivity retains a non-zero conductivity value at 0 K are represented by solid symbols. They have delocalized electronic states at the Fermi level to allow conduction without thermal activation, and therefore are on the metallic side of M-I transition. The general trend is that low room temperature conductivity samples show an insulator-type temperature dependence of the conductivity, whereas most samples having conductivities higher than 100 S/cm are on the metallic regime. However, there are polymers, especially PPV (Figure 1.6), that possess high room temperature conductivity without the metallic signature at low temperatures, probably due to a high content of disorder existent in the material. An increase of the effective conjugation length along the backbone and consequently a reduction in the materials band gap would allow for achieving polymers

PAGE 28

13 with enhanced electrical conductivities. The band gap of polyaromatic compounds is determined by five contributions: bond length alternation, planarity of the monomer unit, aromatic resonance energy of the monomer cycle, donor-acceptor effects of an eventual substitution, and the extent of interchain coupling.13 Therefore, the main synthetic strategies adopted for controlling the band gap are based on the modification of one or more of the above parameters. 1.3 Electrochemical Polymerization Conducting polymers can be prepared by using chemical and electrochemical methods. From the applied point of view, electrochemical polymerization of easily attained, simple aromatic benzenoid, or heterocyclic compounds is of utmost interest, especially if the polymeric product is used in microtechnology, as a thin layer sensor, or as a polymer film electrode. Electrochemical formation of conducting polymers such as PT, PPy and PEDOT proceeds via oxidation of the neutral monomer at the anode surface.42-44 This oxidation step requires 2 electrons per molecule, the excess of charge passed during synthesis being necessary for oxidation of the resulting polymer film. Figure 1.7.A represents the mechanism proposed for the polymerization of heterocycles, where X can be S, O or N-R. The first step consists in oxidation of the monomer that yields the radical cation. Monomer diffusion towards the anode is the rate-determining step, being much slower than the electron-transfer reaction. Consequently, a high concentration of radical cations is constantly maintained near the anode. The further fate of these highly reactive species depends on the experimental conditions including electrolyte composition, temperature, applied potential, nature and morphology of

PAGE 29

14 XX+X++' coupling' coupling' couplingXXX+ XXXX XXXXX XX+-e-X+2XXXXHH+++2H+ XX-e-XX+H XX+HX++XXHH++X XXX+2H+BA XX+X++' coupling' coupling' couplingXXX+ XXXX XXXXX XX+-e-X+2XXXXHH+++2H+ XX-e-XX+H XX+HX++XXHH++X XXX+2H+BA Figure 1.7 Electropolymerization mechanism. A) Polymerization of heterocycles (X = S, O, NH), B) Competitive reaction pathways in unsubstituted poly(heterocycles)

PAGE 30

15 deposition support, etc. In favorable cases, the next step is a dimerization reaction (coupling of two radical cations) that produces a dihydro dication dimer. Further, this species loses two electrons and rearomatizes to form the neutral dimer. Due to extended conjugation over two rings, the dimer has a lower oxidation potential than the monomer itself, and therefore it oxidizes easily to form the radical cation. Stepwise chain growth proceeds via association of radical ions or, less likely, through coupling of a radical cation with a neutral monomer. As chain length increases, resulting oligomers become insoluble in the electrolytic medium and precipitate onto the anode. First generation heterocycles such as thiophene and pyrrole have two possible reaction pathways, as shown in Figure 1.7 B. Polymerization proceeding exclusively through couplings affords polymers with a linear backbone and enhanced electrical properties. The occurrence of a linkage in a given chain modifies its electronic distribution and could promote the formation of branching in energetically favorable sites. Furthermore, the presence of these linkage defects generates twists in adjacent chains, and thus modifies their electronic distribution promoting the propagation of more defects in the resulting material. The relative reactivity of the and positions is about 95/5 for thiophene and decreases as the polymerization proceeds, leading to an increase of the number of undesired couplings and consequently to a decrease in the polymer effective mean conjugation length.44 This is consistent with the considerable increase in the content of disorder as well as the decrease in conductivity as the polymerization proceeds. Therefore, limitation of the polymer growth to very thin films allows for the synthesis of a highly compact material with fewer defects and enhanced electrical properties.45-48

PAGE 31

16 Undesired and couplings can be eliminated through monomer substitution at positions. Substitution of thiophene by electron donating groups leads to a decrease in monomer oxidation potential, allowing for electrosynthesis under milder conditions and thus decreasing the possibility of obtaining overoxidized material. For example, in EDOT, both positions are substituted and the polymerization proceeds exclusively through the desired -couplings. Judicious selection of the substituent affords materials that are soluble, easily polymerizable and possess an enhanced degree of order. 1.4 Doping Induced Property Changes Conducting polymers, once studied solely for their high conductivities, are now extensively used in more dynamic applications where rapid switching from doped to neutral forms is desirable. Charge injection (doping) in these materials leads to a wide variety of interesting and important phenomena, which now defines the conducting polymer field. This reversible intercalation of ions in the polymer matrix triggers significant changes in the materials optical, ionic, electrical and morphological properties.49 These properties can be tuned by varying dopant size and nature from small molecules to high molecular weight polymers as well as by using different preparation techniques.11 Table 1.1 summarizes several properties of conducting polymers that change according to their charge state. As sketched in Figure 1.8, doping can be accomplished in several ways depending on the polymer nature and its intended application. The initial discovery of the ability to dope conjugated polymers involved chemical doping by charge transfer redox chemistry.1,2 Oxidation (p-doping) was

PAGE 32

17 Table 1.1 Qualitative properties of conducting polymers in two extreme redox states. Property Neutral P-doped Stoichiometry Without anions (or with cations With anions (or with cations) Content of solvent Small Higher Volume Small Higher Color: cathodically coloring anodically coloring Transparent or bright Dark Dark Transparent IR optical properties Highly transmissive Highly absorptive Electronic conductivity Semiconducting Metallic Ionic conductivity Smaller High Diffusion of molecules Dependent on structure Surface tension Hydrophobic Hydrophilic accomplished by exposing the polymer to iodine vapors, whereas reduction (n-doping) involved treatment with sodium naphthalenide (Figure 1.9 A). In this case, complete doping results in high quality materials with metallic-like conductivities. Another unique chemical doping procedure is PAni protonation by acid-base chemistry. This leads to an internal redox reaction converting the semiconducting form of PAni (emeraldine base) to a metal (emeraldine salt).50 Although chemical doping is an efficient process, controlling the level of dopant ions is rather difficult. Attempts to reach intermediate doping levels resulted in inhomogeneous doping. As an alternative, electrochemical doping allows for fine tuning of the doping level by simply adjusting the potential between the working and counter electrodes (Figure 1.9 B).52 The working electrode supplies the redox charge to the

PAGE 33

18 Doping of conjugated polymersChemical Electrochemical Photochemical Interfacial High electrical conductivitySolubilityTransparent electrodes, antistaticsConducting fibers, EMI shieldingControl of doping levelElectrochemical batteriesElectrochromism and smart windowsLight-emitting electrochemical cellsHigh performance optical materialsTunable NLO propertiesPhotovoltaic devicesOrganic electrical circuitsTunneling injection in LEDs Doping of conjugated polymersChemical Electrochemical Photochemical Interfacial High electrical conductivitySolubilityTransparent electrodes, antistaticsConducting fibers, EMI shieldingControl of doping levelElectrochemical batteriesElectrochromism and smart windowsLight-emitting electrochemical cellsHigh performance optical materialsTunable NLO propertiesPhotovoltaic devicesOrganic electrical circuitsTunneling injection in LEDs Figure 1.8 Doping mechanisms in conjugated polymers and their applications (adapted from Heeger51) conducting polymer, while ions diffuse in or out of the electroactive film to compensate the electronic charge. Thus any doping level can be achieved by setting the electrochemical cell to a desired potential and waiting for the system to attain an equilibrium state. This type of doping is permanent, meaning that the charge carriers remain in the film unless a neutralization potential is purposely applied. Some high performance optical materials require a different type of doping based on localized oxidation via photo-absorption (Figure 1.9 C). Around these oxidized areas, the material becomes reduced and charge separation occurs (electron-hole separation).

PAGE 34

19 A. Chemical doping by charge transfer a) p-type doping (CP)n + 3/2 ny(I2) [(CP)+y(I3-)y]n b) n-type doping (CP)n + [Na+(C10H8)-]y [(CP)-y(Na+)y]n + C10H8 B. Electrochemical doping a) p-type doping (CP)n + [Li+(BF4-)]soln [(CP)+y(BF4-)y]n + Lielectrode b) n-type doping (CP)n + Lielectrode [(CP)+y(Li+)y]n + [Li+(BF4-)]soln C. Photo-doping (CP)n + h (CP)*[(CP)+y + (CP)-y]n D. Charge injection at a metal-polymer interface a) Hole injection (CP)n ye[(CP)+y]n b) Electron injection (CP)n + ye[(CP)-y]n Figure 1.9 Doping methods in conjugated polymers demonstrated for chemical, electrochemical, photo-, and interfacial doping

PAGE 35

20 Upon photoexcitation from the ground state to the lowest energy excited state, recombination back to the ground state can be radiative (luminescence) (PPV, PPP) or non-radiative (PT, PAc).53 Photoconductivity lasts only until the excitations are either trapped or decay back to the ground state and the material becomes neutral. Another way to dope a conducting polymer is the electrical injection of electrons and holes into HOMO and LUMO respectively. However, the polymer is not doped in the sense of chemical or electrochemical doping because there are no counter ions present in the film. By charge-injection at the metal-semiconductor interface, the polymer (semiconductor) can be used as an active layer in field effect transistors and thin film diodes.54 Dual-carrier injection in a polymer film sandwiched between two metal electrodes provides the basis for light emitting diodes. In conclusion, as summarized in Figure 1.8 and Table 1.1, charge injection in conducting polymers opens up a remarkably wide range of possible applications for this relatively young class of materials. 1.5 Applications As emphasized throughout this chapter, conducting polymers exhibit novel properties not typically available in other materials. These unique properties enable a large number of applications, some of them summarized in Table 1.2. There are applications involving the neutral or doped state of conducting polymers such as LEDs, antistatic materials, corrosion protection, and dynamic applications using the reversible change in properties upon doping, including smart windows and actuators. Some of these applications are briefly described below.

PAGE 36

21 Table 1.2 Applications of conducting polymers General function Special function Application System/references e -conducting film Antistatic coating PEDOT 15,22 e -conducting film in holes Printed circuit boards PEDOT/epoxy55 e -conducting film Capacitor Metal/dielectric/PEDOT56 Thin film technologies band structure, optical transitions LEDs Metal/PPV/ITO 57 Matrix for functional molecules Sensors PAni/glucose-oxidase58 Material for fillings, porous membranes, composites Membrane Separation PPy59 Wettability Offset-printing PT 60 Electrochromism Smart windows, camouflage, thermal control PEDOT and derivatives61, PT60 Intercalation Batteries PT/Li62 Redox process Change of volume Actuators PPy63 Others Inhibition, protection Corrosion protection PAni/Fe64 Doped thin film deposition and microstructuring of conducting polymers. PAni, PPy and PT derivatives are used as antistatic protection and electromagnetic interference shielding materials.15,22 They are incorporated as fillers in common polymeric materials such as poly(vinylchloride) and poly(vinylacetate). PEDOT is used as an antistatic layer for photographic films (AGFA). Another commercial use of PEDOT is in printed circuit boards. PAni has been proved useful as a discharge solution in electron beam lithography (IBM). It protects the insulating electron beam resist from charging and distorting the image. Materials for energy technologies. The possibility of a reversible switching of conducting polymers between two redox states seemed promising for rechargeable

PAGE 37

22 batteries (Figure 1.10 A). Owing to their low density, it was thought that batteries with power densities much higher than those of the ordinary lead/acid battery could readily be obtained. Since the charge on the polymer backbone is delocalized over 3 to 4 monomer units, the charge capacity per unit weight for conducting polymers in not much better that that of metals. The first prototype of commercial batteries with conducting polymers was based on Li/PAni (BASF/Varta). The possible use of conducting polymers as electrode material in supercapacitors is given by their high ionic conductivity allowing high discharge rates. As supercapacitors require high capacitance along with quick charge/discharge of the electrode material, conducting polymers are promising as compared to classical used carbon materials.65 Very recently, photovoltaics have emerged as an important application of this class of materials (Figure 1.10 B). Photovoltaics find potential applications as solar cells and photosensors, and consist of thin films of organic materials sandwiched between two metal electrodes. A built-in electrical field formed in the semiconductor in contact with the electrolyte (photochemical cell) or in the organic layer at the interface with the metal electrode (photovoltaic device) is responsible for the photogeneration of charge carriers. Conducting polymers have attracted attention because of their light weight, potentially low cost and facile fabrication of large area, thin film devices. For example, devices of Al/C 60 modified PTs/ITO show a conversion efficiency of 15 % with zero bias and 60 % with a bias of 2 V.66,67 Conducting polymers bearing both donor and acceptor units are expected to afford even higher quantum yields for photogeneration of charge carriers.68

PAGE 38

23 Discharged ChargedPolymer/MetalPolymer/MetalP0/M0P+/M-ABElectron injecting electrode (Ca, Al)Hole injecting electrode (ITO)Conducting polymer layer C e-h+ h Transparent electrodeElectroactive polymer 1Gel electrolyteElectroactive polymer 2Transparent electrode Glass substrateAcceptor layerDonor layerITO or PEDOT/AuGlass substrateAl or CaD Discharged ChargedPolymer/MetalPolymer/MetalP0/M0P+/M-A Discharged ChargedPolymer/MetalPolymer/MetalP0/M0P+/M-ABElectron injecting electrode (Ca, Al)Hole injecting electrode (ITO)Conducting polymer layer C e-h+ h C e-h+ h Transparent electrodeElectroactive polymer 1Gel electrolyteElectroactive polymer 2Transparent electrode Transparent electrodeElectroactive polymer 1Gel electrolyteElectroactive polymer 2Transparent electrode Glass substrateAcceptor layerDonor layerITO or PEDOT/AuGlass substrateAl or Ca Glass substrateAcceptor layerDonor layerITO or PEDOT/AuGlass substrateAl or CaD Figure 1.10 General design for A) Polymer based battery; B) Photovoltaic device; C) OLED, D) Dual polymer electrochromic device

PAGE 39

24 Electroluminescent devices. A significant event occurred when Friend and co-workers published electroluminescence studies on the neutral form of PPV.54 This work has opened up a new avenue of research and a potential market for the material. Organic electroluminescent devices (OLEDs) are a possible alternative to liquid crystals displays and cathode ray tubes, especially for the development of large displays, and now evolve beyond the lab research into the market. The basic set-up of a polymeric LED is presented in Figure 1.10 C and consists of ITO/light-emitting polymer/metal layers. A thin ITO electrode deposited on an optically transparent electrode (glass or plastic) serves as the anode, whereas metals such as Al, Ca or Mg are being used as cathode materials. After the application of an electric field, an electron is injected into the polymer film from the cathode and a hole is injected from the anode. The electrons and holes migrate towards the center of the film, where they recombine producing light. One of the most important benefit of polymeric OLEDs is the chemical tuning of the HOMO-LUMO gap through judicious synthesis. Therefore, emission of red, green, and blue light has been reported,66 along with a newer type of emission in the NIR region of the spectrum.69 This emission is obtained by the incorporation of lanthanide metals either by chemical bonding or blending into the conjugated polymer structure and is based on energy transfer from the absorbing conducting polymer (usually PPV derivatives) to the emitting lanthanide atom. In this way, the electroluminescence is enhanced by the absence of overlap between absorption and emission spectra. Typical materials are PPV and its derivatives and substituted PTs. OLEDs efficiency is improving steadly along with novel developments such as flexible LEDs, polarized light emitting LEDs and IR emitting LEDs. One major challenge is balancing the electron mobility to that of hole mobility.

PAGE 40

25 This is accomplished by adding electron transport layers and hole transport layers to reduce the barrier height and to encourage holes and electrons to combine near the center of the film. For example, a thin layer of PEDOT is now commonly used as hole transport layer in OLEDs.70 Electrochromic devices. There are two types of successful electrochromic devices: variable transmittance windows and variable reflectance mirrors, described in great detail in Chapters 6 and 5, respectively (Figure 1.10 D). The principle of such devices is based on the redox driven change in transmittance/reflection of a thin conducting polymer layer (or two complementary polymers) over various regions of the electromagnetic spectrum. Organic electrochromic devices are characterized by low-voltage operation, good optical contrast and wide-view angle, making them potentially appealing for controlling the sun radiation in buildings and cars, automotive rear-view mirrors and display devices.71 Moreover, military interest in camouflage materials and thermal control for air and space applications promotes a great deal of research in dual polymers electrochromic window and mirror area.11 Improvement in the cyclability of the coloration/bleaching process and the design and synthesis of new redox systems are current issues being addressed by researchers. Sensors and artificial muscles. Since conducting polymers change properties by incorporation of ions and solvent, it is possible to develop ion-specific sensors based on these materials. Properties changing upon the incorporation of ions or molecules into the polymer matrix include conductivity, electrochemical potential, optical absorbance and fluorescence. Sensing capabilities have been demonstrated for gases like SO 2 and NO 2 (PAni), alkali and alkaline-earth metal cations (through crown ether or polyalkyl ether

PAGE 41

26 functionalization), as well as for more complicated molecules of biological interest (glucose, urea, hemoglobin, etc). A thorough review comprising a wide range of conjugated polymer-based chemical sensors was written by Swager et. al in 2000.72 Since conducting polymers show a swelling with increasing oxidation level, they are able to convert electrical energy into mechanical work. The incorporation of counter ions into the polymer leads to a structural change of the polymer backbone and to an increase in volume up to 30%.73 These electromechanical properties are used in actuators, commonly called polymer based artificial muscles. Starting with first polymer actuator reported by MacDiarmid et al.,74 oxidation induced strain in PAni, PPy, and recently in PEDOT has been investigated.63 There are several other interesting applications of conducting polymers such as controlled drug release, corrosion protection, membrane and ion exchanger, lasers, etc. that award the conducting polymer field the deserved importance.11 1.6 Structure of this Thesis The main characteristic (and beauty) of this work is the combination of fundamental studies on the electrical and optical properties of conducting polymers from the PXDOT and PXDOP class with more application-driven studies of electrochomic devices with an operation window covering the visible and IR regions of the spectrum. This work was extremely rewarding as it allowed probing of theoretical deductions through practical application in devices. Multiple and content-varied projects as well as several useful research collaborations made this work broad in the sense that it covers a wide range of concepts defining the conducting polymer field. Experimental procedures

PAGE 42

27 employed to acquire data presented throughout this dissertation are summarized in Chapter 2. Environmentally stable organic materials with metal-like conductivities are still the focus of many researchers. Understanding the mechanism behind achieving a high conductivity allows for an accurate design of such materials. Chapter 3 details the mechanisms for electronic transport in conducting polymers along with a discussion about the metal to insulator transition in these materials. Optical reflectance, microwave and dc conductivity as well as EPR and X-ray measurements are employed in order to gain insight into the electronic transport. Several PXDOT and PXDOP free-standing films bearing various doping ions were analyzed. These materials electronic properties can be tuned from metallic to insulator by varying the dopant nature and monomer unit planarity. As many possible applications revolve around optical properties of conducting polymer thin films, the work presented in Chapter 4 focuses on doping-induced electronic structure and infrared active vibrational modes as a function of redox state in PXDOT films deposited on ITO substrates. This chapter includes a detailed discussion on the frequently debated origin of the multiple peaks or fine structure often seen in the absorption spectra of neutral polymers. Polymers from the PXDOT family are the basis set for this study. Chapter 5 deals with the optimization of electrochromic devices that operate in the reflective mode and are able to modulate the reflectivity of a metal surface in the visible, NIR and mid-IR regions of the spectrum. Optical contrast, switching time and lifetime of

PAGE 43

28 devices containing PEDOT or PProDOT-Me 2 as the active layer are also reported in this chapter. The use of carefully designed complementary polymers is a promising route for achieving a high degree of control over the color, brightness, switching speed and redox stability of an electrochromic window. As such, the last chapter (Chapter 6) addresses the optical and redox complementarity of polymer pairs required to attain a high contrast and a long lifetime transmissive ECDs. In addition, a new way of achieving a broadband response in an ECD by using bilayers of polymers having different absorption maxima and comparable oxidation potentials is presented.

PAGE 44

CHAPTER 2 EXPERIMENTAL TECHNIQUES The intent of this chapter is to provide necessary background on the most common experimental techniques employed in the characterization of conducting polymers. These methods will be frequently referred to throughout the subsequent chapters. 2.1 Conductivity Measurements Among available conductivity techniques, four probe methods have several advantages for measuring electrical properties of conducting polymers. First, four probe techniques eliminate errors caused by contact resistance, since the two contacts measuring the voltage drop are different from the contacts applying the current across the sample. Second, this technique allows for conductivity measurements over a broad range of applied currents, usually varying between 1 A and 1mA for conducting polymers studied in this work. These current values produce potential differences ranging from 10 V to 10 V, depending on the resistance and thickness of the sample. Polymers for conductivity measurements are usually prepared as free-standing films, although thin films deposited on non-conducting substrates can be easily analyzed. In the latter case, a soluble polymer is solution cast or deposited by spin coating on a glass microscope slide. However, many conducting polymers are insoluble and infusible, and in addition, some do not allow the synthesis of high quality, thick free-standing films (5 microns or more) that can be easily removed from the supporting electrode. Generally, 29

PAGE 45

30 this happens for less conducting samples where the formation of oligomers in solution predominates once the electrode is covered with a thin polymer film. An alternative solution for these cases is the electropolymerization at low charge density to afford thin films, followed by polymer removal from the polymerization support by adhesion with a pressure sensitive adhesive tape. The materials presented in Chapter 3 were prepared electrochemically on glassy carbon or freshly polished titanium electrodes. For a rigorous surface cleaning, glassy carbon electrodes were immersed in acetone and sonicated for 30 minutes, then washed with deionized water and dried under nitrogen. A Pt flag or a polished stainless steel sheet was used as the counter electrode. Free-standing films with thicknesses ranging from 2 to 100 m were obtained by slow galvanostatic deposition at an applied current of 0.02 to 0.08 mA/cm 2 The temperature was maintained at C by immersing the electrochemical cell in a NaCl/ice bath, and at C by using a dry ice/ACN bath. Due to the long polymerization time required to prepare thick films (24-48 hours), the assembly electrochemical cell/cooling bath was kept in a cooler to maintain a constant temperature over night. Generally, thick free-standing film synthesis requires a higher monomer concentration than common electropolymerization of thin films on ITO (10mM). Consequently, the films were prepared from 60 mM monomer in 0.1 M electrolyte (TBAPF 6 LiCF 3 SO 3 or Li(CF 3 SO 2 ) 2 N) dissolved in a low vapor pressure solvent, usually propylene carbonate (PC), that acts as a plasticizer for the resulting film. When the monomer had a reduced solubility in PC, the concentration was limited to 20 mM monomer/PC. After reaching the desired film thickness, the polymer-coated electrodes were removed from the reaction medium and the wet films were detached

PAGE 46

31 from the electrode using a razor blade. Subsequently, the films were washed in PC to remove unreacted monomer and electrolyte salt, placed between two glass slides to flatten the films, and dried under vacuum for 24 hours. Most likely, a small amount of PC remains in the film after drying. Although ACN seems more suitable as a washing solvent due to its high vapor pressure, polymers washed with ACN were brittle and less shiny, probably owning to detrimental morphological changes (cracks, holes) created by fast drying. The morphology and conducting properties of the resulting polymer depend on many experimental variables such as solvent nature, concentration of reagents, temperature, cell geometry, nature and surface morphology of electrodes, and applied electrical conditions. Due to the interdependence of many of these experimental variables, electrosynthesis optimization constitutes a complex study. The influence each of these factors exerts on the electrical properties of the resulting material is described in detail in Section 3.5. There are three types of four probe conductivity techniques that can be employed in the study of conducting polymers: Van der Pauw,75 four-wire76 and four-point probe (Signatone),77 and their use depends on the instrumentation available as well as sample quality and geometry. The Van der Pauw method is a four-point probe technique used for irregularly shaped thin films and was first developed at Philips Laboratories in the Netherlands.75 It consists of four conducting wires (Cu or Ni) attached with silver paste in four points A, B, C, and D, located at the circumference of the sample (see Figure 2.1 A). This method

PAGE 47

32 abcdtATop viewSide view i V 1234 + V t dB t l 1234+ V wC abcdtATop viewSide view i i V V 1234 + V t dB t l 1234+ + V wC Figure 2.1 Four-probe conductivity methods: A) Van der Pauw, B) four-point probe, C) four-wire

PAGE 48

33 affords good measurements if the contacts are sufficiently small, the sample is homogeneous in thickness and does not have defects such as holes. If the contacts are placed such as they form a square on the sample surface, then the conductivity value is given by: = ln2 (Rt) -1 (2.1) where R is the resistance of the sample, and t is the thickness. R is measured by connecting contacts A and B to a constant current source, and C and D to a voltmeter. R=V/I (2.2) where V is the voltage drop and I is the current applied. Another four-point probe method has been developed at Bell Laboratories in 1958.77 This method employs devices with predefined electrode geometry, such as the Signatone S-301-4 available in our laboratories. Figure 2.1 B shows the simplest form of a four-point probe measurement setup. A row of pointed electrodes touches the surface of a polymer film taped or spin cast on an insulating substrate. A known current I is injected at the electrode 1 and is collected at the electrode 4, while the potential difference V between contacts 2 and 3 is measured. For this arrangement, the volume resistivity () of the sample is given by an equation derived by Valdes:78 = 2d V/I (2.3) The result is independent of the electrode contact area as long as the distance between the points (d) is much greater than the film thickness (t). Another equation that describes the resistivity of a sample of thickness t by taking into consideration the ratio of sample thickness to the distance between contact points is given by: = (V/I) (t /ln2) F (2.4)

PAGE 49

34 where F (t/d) is a correction factor that approaches unity as t approaches zero, and its values have been calculated by Uhlir et. al. 79 This four-point probe method has several advantages that make it extensively used in solid-state electronics laboratories. First, if the voltmeter used to measure V has a high impedance, then the conductivity measurement is independent of the contact resistance. As polymeric materials have a high degree of disorder, it is expected to report a conductivity value obtained by averaging multiple measurements in different areas of the film. The four-point probe method allows for the contact points to be easily repositioned in various areas of the film, thus allowing for several conductivity measurements on the same sample. When four point probe devices are not available or when the experimental setup needs to be placed in a cryostat for low temperature conductivity measurements, a four-probe technique (Electrodag method) can be used. In this method, leads can be directly attached to the film using conducting silver paste, colloidal graphite (water or alcohol dispersion) or vacuum evaporated metal. The sample should be cut in a rectangular form, and for best results its length should be much larger than its width. Four thin wires such as Cu or Au (for low temperature conductivity measurements) are attached to the film such that the distance between contacts 2 and 3 (see Figure 2.1 C) is much larger that the distance between contacts 1 and 4. Volume conductivity is calculated from the following equation: = l/Rtw (2.5)

PAGE 50

35 where l is the distance between inner leads 2 and 3, t is the film thickness and w is the width of the sample. l, t and w can be measured with a micrometer unless w is too small and requires the use of a profilometer. Both the current source (Keithley 220) and the multimeter (Keithley 195 A) used for low temperature conductivity data acquisition were computer controlled. The sample with attached Au wires was immersed in Janis Dewar flask containing liquid He and the temperature was controlled from 4.2 K to 300 K by using a LakeShore 82C Temperature Controller and LakeShore DT500 Temperature Sensor. Microwave dielectric constant and conductivity were measured using the same temperature control as above. This method is called cavity perturbation method or wireless conductivity. In this work, a homemade cavity was used in combination with a Hewlett-Packard 8350B microwave source providing a frequency of 6.5 GHz. Often, information about conductivity at different doping levels of thin polymer films is necessary, especially for the determination of conductivity onset and comparison of conductivity in nand pdoped states. This cannot be achieved easily by ex-situ methods described above and an in-situ method is required. Along with the disadvantage that this non-routine method requires a fairly complicated setup such as two potentiostats and expensive, problematical to obtain interdigitated microelectrodes, it affords a conductivity value called pseudo-conductivity that cannot be converted to volume conductivity due to the lack of information about the thickness of the film deposited on interdigitated electrodes. Conducting polymer in-situ conductivity data reported in literature are usually compared to the standard P3MT conductivity (60 S/cm), by

PAGE 51

36 assuming that the all the variables affecting film conductivity are similar (dopant, thickness, electrode geometry, etc.).9,80 2.2 Electrochemical Methods The arsenal of electrochemical methods that can be applied to the study of conducting polymer films deposited on a conducting surface is fairly broad and it has been thoroughly reviewed by Doblhofer et al.81 Among these methods, cyclic voltammetry (CV) has becoming increasingly popular as a mean to study redox states, due to its simplicity and versatility. The electrode potential at which a polymer undergoes reduction or oxidation can be rapidly located by CV. Furthermore, CV reveals information regarding the stability of the product during multiple redox cycles. Since the rate of potential scan is variable, both fast and slow reactions can be followed. A very important aspect of this method is its ability to generate a new redox species during the first potential scan and then probe the fate of species on the second and subsequent scans. Therefore CV allows the growth of a polymer film along with its further characterization during a single experiment (see Figure 2.2). Fundamentals of CV. This method consists of cycling the potential of an electrode, which is immersed in an unstirred solution, and measuring the resulting current at the working electrode. Therefore, the obtained voltammogram is a display of current (vertical axis) vs. potential (horizontal axis). The reducing or oxidizing strength of the WE is precisely controlled by the applied potential (Figure 2.2). Typically, the polymerization of electron rich monomers starts at low potentials (a) where no redox

PAGE 52

37 -0.6-0.4-0.20.00.20.40.6-10010203040 I(mA/cm2)Potential(V vs. Ag/Ag+) abcd SOOSOOOOnOOOOe(Epa, ipa)(Epc, ipc)xx-0.6-0.4-0.20.00.20.40.6-10010203040 I(mA/cm2)Potential(V vs. Ag/Ag+) abcd SOOSOOOOnOOOOe(Epa, ipa)(Epc, ipc)xx Figure 2.2 Cyclic voltammogram of BEDOT-B(OG) 2 in 0.1 M TBAP/ACN at a scan rate of 20 mV/s. a represents the beginning of the experiment, b is monomer oxidation potential onset, c is monomer oxidation peak potential, d is the polymer reduction peak potential, and e is the polymer oxidation peak potential reactions occurs, followed by scanning in the anodic direction. At potential (b), the electrode has sufficient oxidizing character to oxidize the monomer to its radical cation. The anodic current increases rapidly (b-c) until the concentration of the monomer at the electrode surface approaches zero, causing the current to peak (c), and then decay as the solution surrounding the electrode is depleted of monomer. Monomer oxidation is immediately followed by chemical coupling that affords oligomers in the vicinity of the electrode. Once these oligomers reach a certain length, they precipitate onto the electrode surface where the chains can continue to grow in length. The electroactivity of the

PAGE 53

38 polymer deposited onto the WE can be monitored by the appearance of a peak corresponding to the reduction of the oxidized polymer while scanning in the cathodic direction (d). A second positive scan reveals another oxidation peak at a lower potential than the monomer oxidation peak (e). This is due to the neutral polymer now becoming oxidized. Another noticeable fact is the increase in monomer oxidation peak current in the second and subsequent scans. As the peak current is directly proportional with the electrode area (Randles-Sevcik equation),82 this increase in the peak current could be attributed to an increase of the WE area due to a more porous morphology of the electrodeposited polymer. Important parameters of a polymer cyclic voltammogram are the scan rate, switching potentials, as well as the magnitudes of the anodic peak current (i pa ), cathodic peak current (i pc ), anodic peak potential (E pa ) and cathodic peak potential (E pc ). Continuous deposition of the polymer onto the WE can be monitored by the increase in the polymers anodic and cathodic peak currents, while the polymer redox properties are characterized by the magnitudes of its peak potentials. Electrochemical cells and electrodes. Modern potentiostats, such as Perkin Elmer PAR 273 A available in our laboratories, utilize a three-electrode configuration. The potentiostat applies the desired potential between a working electrode (WE) and a reference electrode (RE). WE is the electrode at which the electrolysis of interest takes place. An auxiliary (counter) electrode (CE) provides the current required to sustain redox processes developing at the working electrode. This arrangement prevents large currents from passing through the reference electrode, which could change its potential. Typical electrochemical cell consists of a glass container with a cap having holes for

PAGE 54

39 introducing electrodes and argon (see Figure 2.3). Oxygen removal from the reaction medium is carried out by bubbling argon prior to electropolymerization, whereas maintaining the cell oxygen-free during an experiment is accomplished by passing argon over the solution. to ArCEWERE to ArCEWERE Figure 2.3 Electrochemical cell setup There is a large variety of REs that are commercially available. For use with aqueous electrolytes a saturated calomel electrode (SCE) or an Ag/AgCl/satd KCl electrode can be used. In non-aqueous solvents, a silver wire in contact with 0.1 M AgNO 3 dissolved in a particular solvent (eg., ACN) is used and denoted as Ag/Ag + reference. Ag wire can be immersed directly in the reaction medium, but its potential

PAGE 55

40 changes with the time (pseudo reference) and consequently frequent calibrations with a solution of ferrocene/ferrocinium (Fc/Fc + ) are required. Figure 2.4 illustrates the relationship of these REs to the standard hydrogen electrode (SHE), which is arbitrarily assigned a value of 0.000 V on the redox potential scale. The auxiliary or counter electrode (CE) is usually a Pt foil or gauze with a relatively larger area as compared to the WE. As with WE materials, the common practice is to use electrochemically inert materials such as the noble metals Pt and Au, or glassy carbon. These electrodes consist of a conductive rod of 0.02 cm 2 diameter encapsulated in a Teflon casing (Bioanalytical Systems, Inc.). While Pt button working electrodes can be used for most CV experiments, Au has a lower oxidation potential and it should not be used for experiments requiring potentials higher than 1.2 V vs. Ag/Ag + SHE (0.000 V) Ag/Ag+(0.499 V)Ag/AgCl(0.197 V) SCE(0.241 V)0Ag wire pseudo reference0.10.20.30.40.50.6 Fc/Fc+(0.569 V) V vs. SHE SHE (0.000 V) Ag/Ag+(0.499 V)Ag/AgCl(0.197 V) SCE(0.241 V)0Ag wire pseudo reference0.10.20.30.40.50.6 Fc/Fc+(0.569 V) V vs. SHE Figure 2.4 Electrode potential relationship between common reference electrodes

PAGE 56

41 2.3 Spectroelectrochemistry and Other Optical Measurements Spectroelectrochemistry plays a key role in probing the electronic structure of conducting polymers as well as in examining the optical changes that occur upon doping. It provides information about the materials band gap and intraband states created upon doping as well as gives some insight into a polymer color through the location of the absorption maxima and the ratio of peak intensities if the material shows fine structure on the main peak (see Section 4.3). Measurements were carried out with a UV-Vis-NIR Varian Cary 5 spectrophotometer using a specially designed three-electrode cell to allow potential application while monitoring the absorption/transmission spectra. Typical polymer samples are thin films (100-500 nm) deposited potentiostatically on transparent ITO/glass electrodes purchased from Delta Technologies, Ltd. These electrodes have surface resistivities in the range of 10 to 20 / and are cut to fit a spectroelectrochemical cuvette (0.7 cm wide). Usually a Ag wire pseudo reference is used as a reference electrode and a Pt wire is used as a counter. For potential control, all three electrodes were connected to an EG&G PAR 273A potentiostat. A typical spectroelectrochemical experiment is carried out with the spectrophotometer operating in double beam mode. An accurate baseline is obtained using two cuvettes containing bare ITO/glass electrodes immersed in a liquid electrolyte placed in both reference and sample compartments. After collecting the baseline, the polymer coated ITO is placed in the sample compartment. Polymer analysis is obtained by sequentially stepping the applied potential until the polymer reaches its fully oxidized and neutral forms. These extreme redox states are attained when the polymer absorption

PAGE 57

42 envelope stops changing with increasing/decreasing applied potential. Usual potential steps are between 50 and 100 mV. In order to ensure that a polymer film is fully neutralized, chemical treatment with a solution of 85 % hydrazine in water follows the electrochemical reduction. Dual polymer transmissive devices presented in Chapter 6 are analyzed in a similar manner. However, devices used to collect the baseline are assemblies consisting of two bare ITOs sandwiched together with gel electrolyte. In order to apply voltage across the device, the counter and the reference leads are connected to one another. The reference and the sample holders used for single polymer spectroelectrochemistry do not have the suitable geometry to hold devices, thus they have to be replaced with other supports available in the Cary 5 kit. These holders are basically rectangular non-transmissive sheets having a variable size hole to allow the transmitted light to reach the detector. Devices are scotch-taped on these supports with the area of interest covering the holes. A kinetics experiment allows for measuring polymer and device switching times between the two extreme redox states. This measurement follows the spectroelectrochemistry experiment described above because it requires knowledge of polymer(device) potentials applied for a full switch as well as the wavelength of maximum contrast. Once these values are known, the spectrophotometer can be switched to transmission mode to monitor changes in a samples opacity at the wavelength where the optical contrast is the highest. Concurrently, a square potential waveform is applied at desired time intervals (usually between 0.5-3 s). This allows monitoring % T as a function of time at the specified wavelength, usually located in the visible region. The

PAGE 58

43 rate at which the material is able to access both fully doped and neutral states is determined by the time necessary to attain ca. 95% of its full contrast. Long wavelength optical measurements described in Section 4.4 were performed in the Physics Department at UF, using three different spectrophotometers: a Bruker 113v Fourier Transform Infrared (FTIR) spectrometer for the region 400-5,000 cm -1 a Zeiss MPM 800 microscope photometer for 4,500-45,000 cm -1 region, and a modified Perkin-Elmer 16U for 3,700-45,000 cm -1 region. All measurements were performed at room temperature and the reflectance data have been corrected using an Al mirror as the reference. 2.4 Colorimetry In-situ colorimetric analysis is rapidly becoming a popular technique in the study of electrochromic polymers.83 This method allows for accurately reporting a quantitative measure of the color and graphically representing the track of doping-induced color changes of an electrochromic material or device. There are three attributes that are used to describe the color: hue, saturation and brightness. Hue represents the wavelength of maximum contrast (dominant wavelength) and is commonly referred to as color. Saturation takes into consideration the purity (intensity) of a certain color, whereas the third attribute, brightness, deals with the luminance of the material, which is the transmittance of light through a sample as seen by the human eye. A commonly used scale that numerically defines colors has been established in 1931 by The Commission Internationale de lEclairge (CIE system). This method takes into consideration the response of a standard observer to various color stimuli, the nature

PAGE 59

44 of the light source, and the light reflected by the object under study. These functions are used to calculate tristimulus values (XYZ) that define the CIE color spaces. To simplify the representation of a color, the tristimulus value Y represents the brightness of a material, whereas X and Z are used to define its chromaticity, and they can be easily represented on a two-dimensional graph called spectral locus (x-y chromaticity diagram).84 This chromaticity diagram is a horseshoe shaped area where all visible light wavelengths are represented (Figure 2.5). The location of a point on the chromaticity diagram affords the hue and saturation of a color. From this graphical representation it is possible to establish the dominant wavelength, or hue, by drawing a straight line from the point of interest to the white point (W). Fully saturated and consequently pure colors lie along the periphery of the horseshoe diagram. 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx W 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx W Figure 2.5 x-y chromaticity diagram

PAGE 60

45 Luminance analysis of a material or device is especially valuable in the study of electrochromic polymers because it relates the transmittance over the entire visible region as perceived by the human eye to the applied potential. Such measurements are very useful for the characterization of transmissive/absorptive electrochromic windows because a low luminance value corresponds to an opaque device while a high luminance corresponds to a transparent window. Moreover, changes in a device luminance over time can be used to quantify the devices stability to multiple redox switching. Such experiments are presented in Chapter 6 and were performed by continuously stepping the voltage such as the device would attain its extreme bleached and colored states with a certain delay at each voltage allowing a full color change and hold period. During this time, the decrease in the luminance of the device in the bleached state and the increase in the brightness of the devices dark state were monitored over a several day period. The experimental setup for colorimetric analysis is similar to the one used for spectroelectrochemistry. The applied potential is controlled either by a Pine Bipotentiostat model AFCBP1 or an EG&G 273 potentiostat. The polymer coated ITO/cuvette assembly is placed in a black painted light booth (chromaticity box) that has a light source D50 (5000 K) located in the back. Y, x and y data at different polymer doping levels are recorded with a Minolta CS-100 Chroma Meter colorimeter. In order to obtain accurate values, a background measurement on a bare ITO/cuvette assembly was taken either at the beginning or at the end of the polymer colorimetric analysis. 2.5 Surface Analysis Both atomic force microscopy (AFM) and profilometry techniques give valuable information about polymer thickness and surface morphology. These contact techniques

PAGE 61

46 have similar principles of operation involving a stylus or cantilever that moves across a sample surface. This gives rise to a response signal that is further transmitted to a detector and converted to a digital signal. Although profilometry technique is fairly easy to operate, AFM allows for a much better resolution of the sample surface. Profilometry. This technique is extensively used by our group for measuring polymer film thicknesses and employs a Dektak 3030 Surface Profile Measuring System purchased from Vecco Instruments, Inc. Measurements are made electromechanically by moving the sample beneath a diamond-tipped stylus. A high precision stage moves the sample beneath the stylus according to a user-programmed scan length, speed and stylus force. As the stage moves, the stylus rides over the sample surface causing the stylus to rise vertically and produce electrical signals. These response signals are transmitted to a Linear Variable Differential Transformer mechanically coupled to the stylus that transforms electrical signals in digitized signals and further forwards them to a computer for data processing and storage. In addition to routine step height measurements, this technique allows for the measurement of average surface roughness, average height, and maximum height. A routine experiment performed in our group is relating the polymer film thickness to the amount of charge passed during its electrodeposition. This experiment is carried out for almost every newly synthesized polymer, and if carefully performed, it represents a good calibration curve for future use. For this study, polymers are electrodeposited potentiostatically on ITO/glass electrodes at several charge densities usually ranging from 5 mC/cm 2 to 200 mC/cm 2 Upon deposition, the polymer films were carefully washed with ACN and dried under vacuum for 24 hours to ensure complete solvent evaporation.

PAGE 62

47 Film thicknesses were measured after scratching the film with a razor blade in several small areas to expose the ITO substrate. The depth of the scratch is measured from the cross-section view of the topographic image by placing one cursor at the bottom of the scratch (ITO surface) and the other one at the polymer surface. As the polymer films are often inhomogeneous, multiple measurements in different areas of the sample to give average thickness values were used. Some polymer surfaces are very rough making it difficult to measure their thickness through a simple step height measurement. In these cases, average height method gives very reproducible results and consists of leveling the ITO surface (the scratch bottom) at zero nm and placing the cursors as far apart on the polymer surface as possible. AFM. This contact technique is based on a flexible cantilever with a very low spring constant that induces forces smaller than the interatomic forces, thus the topography of the sample can be monitored without displacing the atoms. Several techniques for sensing the deflection of the cantilever have been investigated and developed. Most AFM used today rely on optical techniques to sense the deflection of the cantilever as a response of surface height variation. For example, the Nanoscope III from Digital Instruments used in this work relies on a beam from a laser diode focused onto the back of the cantilever. The beam reflects off the back of the cantilever onto a segmented photodiode. The amplified differential signal between the upper and lower photodiodes provides a sensitive measure of the cantilever deflection. 2.6 Device Construction Transmissive/absorptive window. The construction of a transmissive type ECD consists of two thin polymer films deposited on transparent indium tin oxide coated glass

PAGE 63

48 (ITO), and separated by a viscous gel electrolyte. Polymer films used to assemble a device were obtained by constant potential oxidative polymerization at potentials slightly higher than the monomer oxidation onset to ensure a slow rate and uniform film deposition. Film thicknesses ranging from 100-300 nm were controlled by monitoring the total charge density passed during electrosynthesis. Polymer-coated electrodes were removed from the polymerization medium and placed in monomer-free electrolyte. The films were electrochemically conditioned by sweeping the potential between .7 and +0.7 V vs. Ag/Ag + for about 15 minutes to guarantee that a maximum doping level, and therefore a high contrast, would be attained. Cathodically coloring films were fully oxidized and anodically coloring polymers were fully neutralized before rinsing with ACN, to ensure charge balance prior to device assembly. The films were then coated with gel electrolyte casting solution until the entire polymer surface was uniformly covered. The cathodically and anodically coloring electrodes selected were then applied to one another and allowed to dry for 24 hours. The gel electrolyte formed a seal around the edges, the devices becoming self-encapsulated. The device construction is carried out with one polymer oxidatively doped while the other is neutral, and both films are simultaneously in either their transmissive or absorptive states. As such, the device is observed as bleached or colored. Complementary polymers thicknesses influence to a large extent the operation of the device as the ability to match the number of redox sites in each film allows for attaining its extreme color states. Therefore, each polymer used in these devices should have a reliable thickness vs. charge density calibration curve described in Section 2.5.

PAGE 64

49 Variable reflectance mirrors. As a device operating in the reflective mode we have used an outward facing active electrode device sandwich structure consisting of five layers. In this construction, two gold-coated Mylar sheets are used as both counter and working electrodes. The top electrode is cut with a series of parallel slits of varied separation (0.5 to 2 mm) across the active surface making it porous to ion transport during switching. Both the active top and the counter polymer films were electrochemically deposited on the gold-coated Mylar electrodes from solutions of 10 mM monomer in 0.1M LiClO 4 in PC at constant potential. Although ACN can be used, the Mylar support swells and curls during an extended polymerization time. Polymerization on gold-coated Mylar electrodes proceeds much faster than the deposition on ITO-coated glass due to an enhanced conductivity of the gold layer. In addition, it leads to a very uniform film owing to the excellent homogeneity of the electrode surface. Following the polymer films redox conditioning, the counter polymer was fully neutralized while the active polymer was fully oxidized to ensure a charge balance prior to device construction. A separator paper, soaked in electrolyte, was used to isolate the back of the working electrode from the counter polymer layer. The top layer is in contact with a window, which is transmissive to the wavelengths of interest, allowing accurate measurements of the active layer reflectivity. We typically used ZnSe for NIR to mid-IR, glass in the NIR and visible, and polyethylene for visible through mid-IR. As only the outward facing electroactive polymer is responsible for the surface reflectivity modulation, the counter electrode polymer optical properties do not affect the device operation. Therefore, the counter polymer layer thickness was typically twice as much as the active polymer layer. The cell was assembled using a high viscosity polymeric

PAGE 65

50 electrolyte composed of LiClO 4 or Li(CF 3 SO 2 ) 2 N dissolved in an acetonitrile (ACN) /propylene carbonate (PC) swollen poly(methyl methacrylate) (PMMA) matrix described below. Device construction begins by placing the neutral polymer-coated back electrode on a poly(ethylene) support, followed by the application of a thin and uniform layer of gel electrolyte on the polymer surface. The separator paper is then gently pressed on the counter polymer/gel surface with care taken not to scratch the polymer layer (neutral polymers are very soft!) and then another layer of gel is applied on top of the paper. The active polymer-coated electrode is placed on top of the separator paper, followed by the application of a third layer of gel and finally, the top window is placed on the active polymer surface. For several purposes the device can be encapsulated in a polyethylene bag. Cu wires are attached to the gold electrodes using a conductive silver paste, then the whole device assembly is placed in a especially-designed polyethylene bag that enables the device to be evacuated prior sealing it. The bag is punctured to take out the Cu leads for electrical contact purposes and then the area around the leads is sealed using an epoxy-based glue. Initially, three of the sides are heat-sealed while the fourth side is used to evacuate the device and then sealed. First, this type of encapsulation provides an oxygen free environment, thus increasing the lifetime of the device. Second, it permits the use of water-free electrolytes, and more importantly of liquid electrolytes. The ability to use liquid electrolytes is a major advantage over the former device assembly, as the polymeric component (such as PMMA) included to toughen the electrolyte can be now eliminated.

PAGE 66

51 Gel electrolyte preparation. A gel electrolyte based on PMMA and Li[N(CF 3 SO 3 ) 2 ] was plasticized by PC to form a highly transparent and conductive gel. To allow an easy mixing of the gel components, we included ACN as a high vapor pressure solvent. The composition of the casting solution by weight ratio of ACN:PC:PMMA:Li[N(CF 3 SO 3 ) 2 ] was 70:20:7:3. Gel preparation begins by dissolving the electrolyte salt in ACN, followed by the addition of PMMA. This polymeric component is not easily dissolved, so vigorous stirring and mild heating (60 C) for a period of about two hours is required. When all of the PMMA has dissolved, PC was introduced to the reaction medium. The gel was stirred and heated on a hot plate for about two more hours until it reached a honey-like consistency and started to stick to the container walls. Preparing the gel in a dry box environment affords a water free electrolyte. Besides acting as an ion transport material, this type of gel electrolyte provides a relative encapsulation of an ECD. At the edges of the device, the ACN in the electrolyte evaporates, leaving behind the PMMA and electrolyte salt in PC. As the PMMA becomes insoluble, it seals the outer edges of the device and provides self-encapsulation. The use of this electrolyte minimizes further solvent evaporation, prevents leaking, and allows for long-term testing of the ECD. 2.7 Purification of Laboratory Chemicals and Materials Since electrochemistry is a very sensitive technique, it is crucial that solvents and electrolytes employed are pure. Reagent grade ACN and PC in Sure Seal bottles were purchased from Aldrich. ACN was distilled over CaH 2 and PC was percolated through

PAGE 67

52 activated type 3A molecular sieves, followed by vacuum distillation (10 mm) and storage in an Ar atmosphere. Tetrabutylammonium perchlorate (TBAP) was either purchased from Aldrich and used as received, or prepared by mixing a 1:1 mole ratio of tetrabutylammonium bromide dissolved in water with perchloric acid. The precipitate was filtered, recrystallized from a 1:1 molar ratio ethanol and water and dried in the vacuum oven for 24 hours at 60C. Tetrabutylammonium hexafluorophosphate (TBAPF 6 -Fluka) and lithium trifluoromethansulfonate (LiCF 3 SO 3 -Aldrich) were recrystallized from ethanol and dried in the vacuum oven for 24 hours at 60C. Lithium bistrifluoromethanesulfonimide (Li(CF 3 SO 2 ) 2 N3M) and lithium perchlorate (LiClO 4 -Aldrich) were used as received.

PAGE 68

CHAPTER 3 ELECTRONIC TRANSPORT AND OPTICAL PROPERTIES OF PXDOT AND PXDOP FREE-STANDING FILMS 3.1 Introduction Traditionally, polymers were thought of as insulators and any electrical conduction in polymers was generally regarded as an undesirable phenomenon. In the last few decades, an opposite trend has been started by the discovery of ionic conductivity in polymers and its wide application to polymer electrolytes for power sources and sensors.49,85,86 In addition to this and somewhat surprisingly, a new class of polymers possessing high electronic conductivity in the doped form has been discovered in 1970s.1,2 These novel materials with interesting and unanticipated properties have attracted the whole scientific community from synthetic chemists to theoretical physicists.31,87,88 After 20 years of research, the fundamental nature of charge propagation is still under debate. Electronic transport can be assumed to occur via an electron exchange reaction (electron hopping) between neighboring redox sites in ion-conducting polymers and by the motion of electrons through conjugated systems in the case of electronically conducting polymers. Owing to the diversity and complexity of these materials, such as chain and segmental motions, changes in morphology and slow relaxation, much research is still needed to achieve a detailed understanding of all processes related to the dynamic and static properties of several interacting molecules confined in a polymer matrix. When debating the conduction mechanism in doped 53

PAGE 69

54 conducting polymers, researchers encountered several difficulties such as unknown molecular weight and morphology, nonuniform doping, instability of the polymer in the doped form, and especially a high disorder of the material due to chain breaks and defects. Theories of transport in amorphous conducting polymers have been advanced by Davis and Mott (1970),89 Cohen (1969),90 Anderson (1958),91 Nagels (1970),92 Prigodin and Epstein (2001),93 Kaiser (2001),34 and some of them are presented in the following section. This chapter details the mechanisms for electronic transport in conducting polymers along with a discussion about the metal to insulator transition in semi-crystalline polymers supported by optical, microwave, dc and EPR results. 3.2 Mechanisms for ElectronicTransport Modified band levels. Conducting polymers fall into the broader category of amorphous semiconductors and their density of states differ markedly from their crystalline counterparts. The main feature of the energy distribution of the electronic states density in amorphous solids is determined by their configurational disorder. This causes fluctuations in the potential, leading to the formation of localized states. These states are localized in the sense that an electron placed in a region will not diffuse at 0 K to other regions, and various models presented here differ in the nature of localized states. Figure 3.1 represents a schematic drawing of density of states showing two new features, as compared to crystalline semiconductors. The first is the tail of localized states, which are narrow and extend into the forbidden gap. Second there is a band of compensated levels near the middle of the gap, originating from defects such as vacancies or twists in the polymer backbone. A refinement of the above model is the splitting off

PAGE 70

55 from the tail states of various localized states, lying at certain energies within the gap (Figure 3.1.C). This mechanism fits better for real disordered semiconductors and explains luminescence and photoconductivity in conducting polymers.30 A)B)C) A)B)C) Figure 3.1 Schematic drawing of density of states: A) band with tail or localized states and a Fermi level near the middle of the gap; B) the center band split into acceptor and donor bands; C) real disordered semiconductor with defect states (from Chien30)

PAGE 71

56 Conduction by thermal activation to extended states. This model implies that the charge carriers are bound and there is a minimum temperature at which the materials become conducting. Conductivity in this regime would be more appropriately considered as Brownian motion. The lowest value of the electrical conductivity before the start of an activation process is called by Mott94 the minimum metallic conductivity and is given by: min =const e 2 / a (3.1) where a represents the interatomic distance. The constant has a value of about 0.026 and min is in the range of 200-300 S/cm. Conduction by activated hopping in band tails. If the charge carriers are localized, conduction can occur by hopping between states in the band tails. An electron moves from one localized state to another with the energy provided by exchanging with a phonon, and the charge carrier mobility follows Arrhenius law.95 Variable-range hopping conduction. This model fits most of the conducting polymers studied worldwide, especially the ones belonging to the insulator side of metal-insulator transition (M-I transition) (described in detail in Section 3.3). Consider a semiconductor with strong interaction between the charge carrier and the atom. Each wave function is confined to a small region of the space, falling off exponentially with the distance. These materials are called Fermi glasses. When Fermi energy lies in the range of energies for localized states, variable-range hopping is possible. For the case of strong localization, hopping will be only between nearest neighbors. Conductivity of materials in this category is given by: dc = 0 (T)exp[-const/T 1/(1+d) ] (3.2)

PAGE 72

57 where d is the dimensionality of hopping.94 For most polymeric materials the electronic conduction is 3-D, therefore a linear log vs. T -1/4 dependence has been widely used as evidence for transport via variable range hopping.96 Fluctuation-induced tunneling conduction (Heterogeneous Model). This model arises from the study of composite materials based on conducting particles embedded in an insulating matrix, which seems to be particularly applicable to conducting polymers.97 First, most polymers seem to have a fibrilar morphology98 (see Figure 3.2) that is not continuous. Second, there is vast evidence that doping can be highly inhomogeneous, consequently creating tiny metallic domains separated by insulation regions. In these materials, the carriers are itinerant and free to move over distances very large compared to their atomic dimensions. For such polymers, the electrical conduction may be dominated by carrier transport between conducting domains rather than by hopping between localized states. The presumption is that the carriers tend to tunnel between conducting regions at points of closest approach and the conducting domains vary in shape. Well-defined crystalline regions surrounded by amorphous material are evident in Figure 3.2 (taken from Percec et al.98). Similar boundaries in conducting polymers dominate the resistance of a sample since the charge carriers have to pass through the barriers surrounding the crystallites. The significance of heterogeneity for electronic transport in PAc was recognized in the early work of Park et al.,99 who proposed a barrier resistance in series with highly conducting crystalline regions. Studies by Travers et al.100 on the evolution of transport properties function of aging provide evidence that heterogeneous disorder plays a key role in conducting polymers, with the barrier regions

PAGE 73

58 Figure 3.2 Scanning force microscopy image of polymer chains on a surface showing separate regions of well-aligned chains (from Percec98) around conducting grains being broadened by aging. A related inhomogeneous localization model for PAni proposed by Prigodin and Epstein,35,93 involves less distinct boundaries between crystalline and disordered regions. This model takes into consideration a detailed description of chain arrangement in conducting polymers. Structural studies showed that there are crystalline regions where the polymer chains are regularly and densely packed. Outside these regions, the chain arrangement is less ordered, and here, the chains form amorphous media. The crystalline metallic islands are coupled into the network with twisted and entangled polymer chains (see Figure 3.3). Electrons move primarily along the polymer chain and sometimes hop between neighboring chains. In the metallic islands, there is a good chain overlap and the charge carriers are delocalized over the entire grain volume. Single polymer chains that pass through disordered, amorphous media provide connections between metallic grains. Therefore the electric connection is not only between the neighboring grains, but between distant ones as well. Depending on the route of synthesis, the crystalline region varies

PAGE 74

59 from a few percent to 50%. The standard percolation analysis101 predicts that 30 % crystallinity provides metallic behavior, but several polymer samples with 50 % crystallinity are still in the insulator regime. The Epstein model allows the interpretation of frequency dependence of electromagnetic response in conducting polymers, including the microwave region of spectrum.96,102 Kaiser et al. has applied a refinement to this model to introduce charge carriers being thermally activated over thin barriers (dominating at high temperatures) or tunneling through the barriers (dominating at low temperatures).34 Figure 3.3 Schematic drawing on the conducting polymer structure. The lines represent the polymer chains and the dashed squares mark the regions where polymer chains demonstrate crystalline order. (from Prigodin102)

PAGE 75

60 3.3 Elementary Excitations in Conducting Polymers Solitons. The localized electronic state associated with the soliton is a nonbonding state at an energy lying at the middle of the gap, between the bonding and antibonding levels of the perfect chain. The soliton is a defect both topological and mobile because of the translational symmetry of the chain.53 Soliton model was first proposed for degenerated conducting polymers (PAc in particular) and it was noted for its extremely one dimensional character, each soliton being confined to one polymer chain (see Figure 3.4). Thus there was no conduction via interchain hopping. Furthermore, solitons are very susceptible to disorder, and any defect such as impurities, twists, chain ends or crosslinks will localize them.103,104 .D.D-D+D-+Positive solitons in PAcNeutral solitons in PAcA) B) VBVBCBCB .D.D-D+D-+Positive solitons in PAcNeutral solitons in PAc.D.D-D+D-+Positive solitons in PAcNeutral solitons in PAc .D-D+D-+Positive solitons in PAcNeutral solitons in PAcA) B) VBVBCBCB Figure 3.4 Band diagrams for neutral and positive solitons. A) Schematic representation of neutral (left) and positive (right) solitons in degenerate PAc, where D represents a dopant ion; B) band diagrams for neutral (left) and positive (right) solitons with associated electronic transitions.

PAGE 76

61 Polarons and bipolarons. The polaron can be viewed as a bound state of a charged and a neutral soliton whose mid-gap energy states hybridize to form bonding and anti-bonding levels. The neutral soliton contributes no charge and a single spin and the charged soliton possesses charge and no spin. The resulting positive or negative polaron is a radical cation or anion, respectively, a particle consisting of a single electronic charge leading to local geometrical relaxation of the bond lengths, as the polymer chain passes from the neutral benzoid form to a partially-quinoid structure (Figure 3.5). Optical and EPR studies indicate that polarons are the main charge carriers when a material is lightly doped.51,105 CBVB Eg(*) CBVB Eg 12 P1P2 CBVB 12BP1 SOOSOOSOO SOO+SOOSOO +SOO+SOOSOONeutralPolaronBipolaronA)B) CBVB Eg(*) CBVB Eg 12 P1P2 CBVB 12BP1 SOOSOOSOO SOO+SOOSOO +SOO+SOOSOONeutralPolaronBipolaron CBVB Eg(*) CBVB Eg(*) CBVB Eg 12 P1P2 CBVB Eg 12 P1P2 CBVB 12BP1 CBVB 12BP1 SOOSOOSOO SOO+SOOSOO +SOO+SOOSOONeutralPolaronBipolaronA)B) Figure 3.5 Polaron and bipolarons in non-degenerate ground state polymers: A) band diagrams for neutral (left), positive polaron (center) and positive bipolaron (right); B) neutral (left), lightly doped (center) and heavily doped (right) PEDOT structures

PAGE 77

62 Similarly, a bipolaron is a bound state of two charged solitons of the same charge with corresponding mid-gap levels, as illustrated in Figure 3.5. Since each charged soliton carries a single electronic charge and no spin, the positive (negative) bipolaron is a spinless dication (dianion). This represents a doubly charged state of two polarons brought together by the overlap of a common lattice distortion (enhanced geometrical relaxation of the bond lengths due to the formation of fully quinoid structure at high doping levels).105 3.4 Metal-Insulator Transition in Conducting Polymers As the extent of disorder increases in a metallic system, the mean free path of an electron becomes equal with the interatomic spacing and consequently some of the carriers become localized, the material becoming insulator.28 The metal-insulator transition (M-I transition) has been proposed by Ioffe and Regel106 and is defined as: E F l 1 (3.3) where E F is the electron wavenumber and l is the mean free path. Based on the above equation, Mott107,108 proposed that a metal-insulator transition occurs when the disorder is sufficiently large that E F l<1 and the material becomes a Fermi glass insulator (term described in Section 3.2). In recognition of Andersons early work on disorder induced electron localization, Mott called the M-I transition the Anderson transition. The M-I transition in conducting polymers is especially interesting; critical behavior has been observed over a relatively wide temperature range in a number of systems including PAc, PAni, PPy, PPV,109 and more recently PEDOT and its derivatives,36-41,110-112 which are the focus of this work. In

PAGE 78

63 each case, the metallic, critical and insulating regimes have been identified. The critical regime is tunable in conducting polymers by a multitude of methods including varying the synthetic path, the dopant, via monomer derivatization or by applying external pressure and/or magnetic field. However, the truly metallic regime with E F l has not yet been achieved. Superconductivity. Experimental studies have demonstrated that for conducting polymers, the electrical conductivity and mechanical properties improve together, as the degree of chain extension and chain alignment are increased. Materials become truly metallic if there is a high probability that an electron will traverse to a neighboring chain before traveling between defects on a single chain. Schn et al. recently claimed the first organic polymer superconductor.113 This gate-induced superconductivity in PT films had granular character, consistent with the heterogeneous model described above. The films of poly(3-hexyl thiophene) used possessed low disorder and high charge-carrier mobility as a result of self-assembly of polymer side chains. Application of a voltage varied the carrier concentration over a wide range, the conductivity varying from semiconducting to metallic. Superconductivity was observed below 2.35 K for samples with room temperature conductivities of about 80 S/cm. The researchers suggested that the superconductivity originated in metallic islands with dimensions of about 10 nm corresponding to well-ordered nano-crystalline regions in the polymer, emphasizing again the key role of heterogeneity in charge transport in conducting polymers.

PAGE 79

64 3.5 Electrochemical Synthesis of Free-Standing Films Electrochemistry has played a significant role in the preparation and characterization of these novel materials. Electrochemical techniques are especially suitable for controlled synthesis and for tuning of well-defined oxidation states.12 The electropolymerization of thiophene derivatives involves many experimental variables such as solvent, concentration of reagents, temperature, cell geometry, nature of electrodes and applied electrical conditions. As a consequence of the diversity of these parameters, electrosynthesis conditions determine to a large extent the morphology and conducting properties of the resulting polymer. However, due to the interdependence of many of the experimental variables, electrosynthesis optimization constitutes a complex study. The results for PXDOTs and PXDOPs obtained in various conditions are presented in Table 3.1. The solvent used must simultaneously present a high dielectric constant to ensure the ionic conductivity of the electrolytic medium, and a good electrochemical resistance against decomposition at the potentials required to oxidize the monomer. Until now, the most conductive PTs and PT derivatives obtained have been prepared in rigorously anhydrous aprotic solvents of high dielectric constant and low nucleophilicity such as acetonitrile, benzonitrile, nitrobenzene and propylene carbonate.9 When used for PEDOT synthesis, ACN led generally to powdery deposits or to brittle films with conductivities typically comprised between 5-20 S/cm. PC allowed for the synthesis of compact and shiny free-standing PEDOT films with conductivities exceeding 100 S/cm. Doped polymers are electrosynthesized in the presence of small anions derived from strong acids such as ClO 4 PF 6 BF 4 CF 3 SO 3 associated with lithium or tetraalkyl

PAGE 80

65 ammonium cations. The nature of these dopants strongly affects the morphology and the electrical properties of the resulting polymer. As seen in Table 3.1, PEDOT films prepared in the presence of PF 6 show higher conductivities than the ones prepared with other dopants. Probably, PF 6 facilitates a more compact packing of the polymer chains, thus increasing the interchain transport contribution to the overall conductivity.112 Table 3.1 Electrochemical synthesis of free-standing PXDOT films. Polymer Dopant Electrode Material TemperatureCurrent Density (mA/cm2) Conductivity(S/cm) PEDOT PF6Glassy carbon -45C 0.08 110 PEDOT PF6Glassy carbon -45C 0.06 110 PEDOT PF6Glassy carbon -45C 0.04 330 PEDOT PF6Glassy carbon -45C 0.02 74 PEDOT PF6Glassy carbon 20C 0.04 10 PEDOT PF6Glassy carbon -5C 0.04 90 PProDOT PF6Glassy carbon -45C 0.04 66 PEDOT CF3SO3Glassy carbon -45C 0.04 95 PProDOT CF3SO3Glassy carbon -45C 0.04 38 PEDOP CF3SO3Ti -5C 0.04 40 PEDOT (CF3SO2)2NGlassy carbon -45C 0.04 8 The electropolymerization temperature has a marked effect on the optical and electrical properties of the resulting material, the films produced at C having

PAGE 81

66 noticeably higher conductivities than the films produced at higher temperatures. This is due to the fact that the activation energy of undesirable reactions such as termination, kinks in the backbone, etc., is higher than the activation energy for the polymerization reaction, thus films produced at low temperatures have a longer mean conjugation length and consequently higher conductivities. The anode material is of crucial consideration since the physical and chemical properties of the polymerization surface determine the nature and strength of the bond between the polymer and the electrode, which can affect both the deposition process and the properties of the resulting material. Moreover, the formation of a conjugation defect in the early stages of the synthesis (at the anode surface) has much more dramatic consequences for the overall stereoregularity and stacking order of the polymer chains than the occurrence of the same defect at a later stage of the process.45 The highest conductivity and best surface quality films were obtained using glassy carbon electrodes that were extremely shiny, flat (at the atomic level) and of high purity. Other substrates such as stainless steel and ITO/glass have been tried, but their surface quality did not allow the synthesis of highly conducting films. Polymers from the pyrrole family did not adhere well enough to glassy carbon, therefore freshly polished titanium electrodes were used as deposition substrates. Although it is likely that the potential applied during the electrosynthesis is the critical electrical parameter, the most homogeneous and conducting films are generally obtained in galvanostatic conditions.46 The current density used is a crucial factor in obtaining high quality materials. If the current density is too high, the polymer deposition proceeds too fast for the polymer chains to align, resulting in a material with a high

PAGE 82

67 content of amorphous phase. When the current density is lower than its optimal value, formation of semi-soluble oligomers in the close vicinity of anode is favored. Due to their limited solubility and low reactivity, these oligomers insert into the polymer film without chemical coupling, thus reducing the cohesion and conductivity of the materials. For example, PEDOT films with conductivities as high as 310 S/cm have been obtained at a current density of 0.04 mA/cm 2 whereas films obtained at higher (0.06 mA/cm 2 ) or lower (0.01 mA/cm 2 ) current densities exhibited much poorer electrical properties. 3.6 Temperature Dependence of the Conductivity To determine the nature of the metallic state in conducting polymers, it is necessary to use a wide variety of techniques. The easiest method to start identifying the various regimes is to measure the dc conductivity over a broad range of temperatures. The transition from metallic to insulating behavior is generally accompanied by a marked change in the temperature dependence of the dc conductivity. In a semiconductor, or insulator, the electrical conductivity increases with the temperature, due to an increased number of the charge carriers thermally excited across the band gap that are able to contribute to the conductivity. The conductivity of a metal decreases with increasing temperature, since the mobility of charge carriers decreases due to an enhancement in electron-lattice scattering. Since the room temperature conductivity of heavily doped conducting polymers are frequently comparable to Motts minimum metallic conductivity, disorder is the dominant factor in determining the temperature dependence of various transport regimes. Table 3.2 summarizes the room temperature conductivities [(300K)] and the conductivity ratios [(10K)/(300K)] of several PXDOT samples.

PAGE 83

68 The conductivity ratio is a useful empirical parameter to quantify the extent of disorder and for sorting out various regimes. In general, as disorder increases, the materials become more insulating and the conductivity decreases more rapidly upon lowering the temperature. Table 3.2 Transport properties of PXDOT and PXDOP free-standing films Doped Polymer dc (S/cm) (300K) (10K)/(300K) mw (T) (300K) Pauli contribution to T) max for (), cm -1 W-plot slope regime PEDOTPF 6 330 0.42 negative major 1900 positive metallic PProDOT-PF 6 66 0.033 positive noticeable negative insulating PEDOT-CF 3 SO 3 95 0.17 negative major 2100 small positive critical/ metallic PProDOT-CF 3 SO 3 38 0.074 positive noticeable 2300 negative insulating PEDOP-CF 3 SO 3 40 0.12 negative small negative critical PEDOT-(CF 3 SO 2 ) 2 N 8.5 0.002 positive negative insulating From the data presented in Table 3.2, the most metallic sample is PEDOT doped with PF 6 anions, sample that shows also the highest room temperature conductivity. PEDOT-CF 3 SO 3 exhibits a slightly lower conductivity ratio, leading to the conclusion that doping with PF 6 allows for more ordered film morphology and a closer packing of the polymer chains. The electronic properties of the polymers can also be fine tuned by derivatization. Comparing the polymers of the thiophene family, PEDOT shows a more metallic behavior than PProDOT, due to a planar structure of the monomer unit. The propylene-dioxy ring in PProDOT is twisted and its morphology is more open, hindering to some extent the interchain hopping contribution to the conductivity. EDOP planar

PAGE 84

69 structure would recommend its use in the synthesis of polymers with a high degree of order. Unfortunately, we do not have temperature dependence of the conductivity measurements for PEDOP-PF 6 However, PEDOP-CF 3 SO 3 has a similar conductivity ratio as PEDOT containing the same dopant, with slightly lower room temperature conductivity. This is probably due to a less environmentally stable oxidized form of the dioxypyrrole polymer. To explicitly describe the characteristic behavior of temperature dependence of the conductivity, we define the reduced activation energy as the logarithmic derivative of the material resistivity (): W = d ln / d ln T (3.4) The temperature dependence of W for the metallic materials is positive. In the critical regime, W is temperature-independent or shows very weak temperature dependence, whereas in the insulating regime W has a negative temperature coefficient. W vs. temperature coefficients for PXDOTs and PXDOPs are summarized in Table 3.2. Figure 3.6 represents the temperature dependence of the conductivity in PXDOT films. As noted before, the conductivity of PEDOT-PF 6 does not decrease too much at low temperatures, maintaining a finite value (about half of its room temperature conductivity value) as the temperature approaches 0 K. This suggests that the sample is on the metallic side of the M-I transition. Changing the dopant to CF 3 SO 3 or (CF 3 SO 2 ) 2 N leads to more disordered materials with a stronger temperature dependence of the conductivity. PEDOT-(CF 3 SO 2 ) 2 N exhibits typical semiconductor behavior, with no conductivity as the temperature approaches 0 K, suggesting that this material is on the insulator side of the M-I transition.

PAGE 85

70 0501001502002503000.010.11 (T)/(300K)T(K) PEDOT-PF6 PEDOT-CF3SO3 PEDOT-(CF3SO2)2N PProDOT-PF6 Figure 3.6 Temperature dependence of the conductivity in PXDOT free-standing films In order to understand the nature of the transport mechanism in these samples, we fit the conductivity data to Motts Variable Range Hopping (VRH) model114 described by the Equation 3.2, for a three dimensional conduction (d=3). We noticed that 3-D VRH fits very well for the samples on the insulator side of the ITM (see Figure 3.7). Therefore, the charge carriers in PEDOT-(CF 3 SO 2 ) 2 N and PProDOT-PF 6 are mainly localized, and they can only hop from a site to another in the 3-D space. Attempts to fit 1-D and 2-D VHR with the conductivity data for PEDOT doped with CF 3 SO 3 and PF 6 failed as well, leading to the conclusion that the conductivity of these samples is mainly due to delocalized charged carriers. Probably, the heterogeneous model proposed by Epstein (Section 3.2) will fit better in this case.93

PAGE 86

71 0.20.30.40.50.60.7101001000 PEDOT-PF6 PEDOT-CF3SO3 PEDOT-(CF3SO2)2N PProDOT-PF6(T)*T1/2(S/cm*K1/2)T-1/4(K-1/4) Figure 3.7 The conductivity dependence of T -1/4 in PXDOT films 3.7 Microwave Experiments The microwave conductivity ( mw ) and dielectric constants ( mw ) were measured using a cavity perturbation technique (at 6.5 GHz) in Dr. Epsteins laboratory at Ohio State University. The mw is a key probe of the charge carrier delocalization. For delocalized electrons at frequencies lower than their plasma frequency, the dielectric constant is negative, at all temperatures, due to the inertia of free carriers in an alternating current field. The value of the negative mw is limited by the relaxation rates and the frequency. For a localized carrier, the charges stay in phase with the applied field and mw is positive at low frequencies.96

PAGE 87

72 Figure 3.8 shows the temperature dependence of the mw for PXDOT samples. In concordance with the dc results, PEDOT-CF 3 SO 3 and PEDOT-PF 6 show metallic behavior, with dielectric constants that remain negative down to 10 K. The weaker temperature dependence of mw for PEDOT-PF 6 ( mw (300 K)/ mw (20 K) ~ 2.7) than for PEDOT-CF 3 SO 3 ( mw (300 K)/ mw (20 K) ~ 4.5) is consistent with a weaker temperature dependence of dc For PEDOT-(CF 3 SO 2 ) 2 N and PProDOT-PF 6 mw stay positive over the entire temperature range, suggesting once again that these samples are on the insulator side of M-I transition. Conductivities obtained by the cavity perturbation method have similar temperature dependence as direct current conductivities, and both data sets are compared in Figure 3.9. However, for the samples exhibiting electronic conduction by hopping, mw deviates slightly from dc at higher temperatures. 050100150200250300-20000-16000-12000-8000-400004000800012000 PEDOT-PF6 PEDOT-CF3SO3 PEDOT-(CF3SO2)2N PProDOT-PF6mw(T)T(K) Figure 3.8 Temperature dependence of the dielectric constant in PXDOT films.

PAGE 88

73 0501001502002503000.50.60.70.80.91.01.1 PEDOT-PF6(T)/(300K)T(K) dc mw0501001502002503000.00.20.40.60.81.0 PEDOT-CF3SO3(T)/(300K)T(K) dc mw0.20.30.40.50.60.7110100 PEDOT-(CF3SO2)2N(T)*T1/2(S/cm*K1/2)T-1/4(K-1/4) dc mw0.20.30.40.50.60.7110100 dc mwPProDOT-PF6(T)*T1/2(S/cm*K1/2)T-1/4(K-1/4)A)B)C)D)0501001502002503000.50.60.70.80.91.01.1 PEDOT-PF6(T)/(300K)T(K) dc mw0501001502002503000.00.20.40.60.81.0 PEDOT-CF3SO3(T)/(300K)T(K) dc mw0.20.30.40.50.60.7110100 PEDOT-(CF3SO2)2N(T)*T1/2(S/cm*K1/2)T-1/4(K-1/4) dc mw0.20.30.40.50.60.7110100 dc mwPProDOT-PF6(T)*T1/2(S/cm*K1/2)T-1/4(K-1/4)A)B)C)D) Figure 3.9 Temperature dependence of the mw and dc conductivity for A) PEDOT-PF 6 ; B) PEDOT-CF 3 SO 3 ; C) PEDOT-(CF 3 SO 2 ) 2 N; D) PProDOT-PF 6 3.8 Magnetic Susceptibility The magnetic properties of localized unpaired electrons differ significantly from their itinerant-electron counterparts. Total electron spin susceptibility () for doped polymers is composed of Curie-type susceptibility ( C ), associated with localized, non-interacting electron spins, and Pauli-type susceptibility ( P ).110,111 The latter represents the susceptibility of delocalized electrons and is essentially independent of the temperature. = P + C ( C = const/T) (3.5)

PAGE 89

74 T = P T + const (3.6) Figure 3.10 represents the temperature dependence of EPR susceptibility for three PXDOT samples. Data for PProDOT-PF 6 fit well a straight line, indicating the presence of localized unpaired electrons as main contributors to the electronic transport. On the other hand, PEDOT data shows a marked deviation from the linear behavior as the temperature increases. This is due to a high contribution of P to the overall susceptibility, thus the electronic transport in these materials occurs mainly via delocalized charge carriers. The number of free electrons increases with increasing deviation from linearity; therefore PEDOT-PF 6 shows the most metallic character. Hence, the susceptibility and the conductivity data are in qualitative agreement, placing PProDOT-PF 6 on the insulating side of the M-I transition, PEDOT-CF 3 SO 3 in the critical regime and PEDOT-PF 6 in the metallic regime. 01020304050607080050100150200250300T (K) T ( 10-3 em u .K/mole ) PEDOT-PF6 PEDOT-CF3SO3PProDOT-PF6 01020304050607080050100150200250300T (K) T ( 10-3 em u .K/mole ) PEDOT-PF6 PEDOT-CF3SO3PProDOT-PF6 Figure 3.10 EPR Susceptibility of Doped PXDOT Films

PAGE 90

75 3.9 X-ray Measurements The X-ray diffractograms presented here were measured by A. Saprigin at Ohio State University. Peaks at ~6.5, ~12.5, ~19, ~25, ~37 are clearly seen in the X-ray diffractograms of all four PXDOT samples presented in Figure 3.11. Based on these measurements, we propose a pseudo-orthorombic structural model with lattice parameters a = 14 b = 14 c = 14 The first four peaks are very close in position to the diffraction peaks of spin cast thin films of tosylate doped PEDOT reported in the literature.115 The presence of relatively sharp peaks in the X-ray diffraction pattern of PXDOTs indicates the presence of crystalline grains embedded in amorphous media. PEDOT PF 6 shows the sharpest peaks, thus having the highest content of crystallinity of all analyzed samples, with the crystalline domain size of about 20 This fact is in agreement with the conclusion that this material is on the metallic side of the M-I transition due to a low amount of disorder. For the other samples, the crystalline domain sizes are in the range of 15-20 3.10 Optical Conductivity Kramers-Kronig analysis of the reflectance data provides the optical conductivity function, (). We can obtain information about the conduction electrons in PXDOTs by comparing this function with the Drude expression for free electrons and the localization models. For PXDOT samples, () demonstrates a intraband peak below 1 eV and a weaker feature at ~ 1.5 eV, attributed to an interband transition (Figure 3.12).112 The () for PXDOTs differs from the Drude model, where Drude () increases continuously with decreasing frequency. As dc increases, the frequency at which the maximum in

PAGE 91

76 () occurs ( max ) shifts monotonically to lower energies (Table 3.2). Since max is lowest for PEDOT-PF 6 this indicates that the scattering time (mean free path) for this sample is the longest, and therefore the material is less disordered. This is in agreement with the transport and structural experiments presented in this chapter. PEDOT-PF6 10203040 PProDOT-PF6 10203040 PEDOT-CF3S03A)B)C)D)2 2 IntensityIntensityIntensityIntensity PEDOT-PF6 10203040 PProDOT-PF6 10203040 PEDOT-CF3S03A)B)C)D)2 2 IntensityIntensityIntensityIntensity PProDOT-CF3SO3 PProDOT-CF3SO3 Figure 3.11 X-ray diffraction for: a) PEDOT-PF 6 ; b) PProDOT-PF 6 ; c) PEDOT-CF 3 SO 3 ; d) PProDOT-CF 3 SO 3 In conclusion, the studied materials have properties varying from metallic to insulating depending on the nature of the dopant or the monomer unit planarity. Selected transport data are summarized in Table 3.2. The most metallic sample, PEDOT-PF 6 has a high room temperature dc weak temperature dependence of dc along with a positive slope of W(T)= dln( dc (T))/dln(T) (reduced activation energy) at low T. Moreover, its 6.5 GHz dielectric response mw stays negative as the temperature is varied. This behavior is similar to previously reported data for metallic polyaniline and polypyrrole. EPR

PAGE 92

77 measured magnetic susceptibility (T) has a major contribution from Pauli component P associated with delocalized spins. 02004006008001000100100010000(), S/cmWavenumbers, cm-1PEDOT-PF6PEDOT-CF3SO3PProDOT-CF3SO3 PProDOT-CF3SO3PEDOT-CF3SO3A)B) 02004006008001000100100010000(), S/cmWavenumbers, cm-1PEDOT-PF6PEDOT-CF3SO3PProDOT-CF3SO3 02004006008001000100100010000(), S/cmWavenumbers, cm-1PEDOT-PF6PEDOT-CF3SO3PProDOT-CF3SO3 PProDOT-CF3SO3PEDOT-CF3SO3 PProDOT-CF3SO3PEDOT-CF3SO3A)B) Figure 3-12 Optical properties of PXDOT films. A) Optical conductivity of PXDOT films as calculated from Kramers-Kronig analysis; B) Photos with highly reflective free-standing films

PAGE 93

78 Stronger T dependence of dc for critical regime compositions (PEDOP-CF 3 SO 3 PEDOT-CF 3 SO 3 ) results in a small positive or small negative W-plot slope at low T. For PEDOT-CF 3 SO 3 mw is negative down to 4 K, demonstrating that this sample is still on the metallic side of the M-I transition. Pauli contribution to the overall magnetic susceptibility is less than for PEDOT-PF 6. PEDOT-(CF 3 SO 2 ) 2 N, PProDOT-CF 3 SO 3 and PProDOT-PF 6 show lower dc negative slope of W-plot, positive microwave dielectric response typical for other dielectric conducting polymers. It is noted that the frequency for maximum of optical conductivity increases for more insulating samples. The optical, magnetic and transport studies demonstrate that this family of inhomogeneously disordered (also confirmed by X-ray diffraction studies) materials spans from dielectric to metal. 3.11 Room Temperature Conductivities of Alkyl Substituted PEDOTs The introduction of substituent groups onto conjugated polymers has been utilized to a great extent in controlling their resulting physical and electronic properties.116 Processability (by solution or melt processing) and electronic tunability (in terms of conducting and optical properties) have been obtained by incorporating different side groups.16 Several groups noted an enhancement in electronic properties of thiophene polymers upon substitution with varied length alkyl chains. Groenendaal et al. synthesized a series of alkyl-PEDOTs with chains varying from CH 3 to -C 14 H 29 .117,118 In-situ conductivity of these materials showed an interesting trend. After a high conductivity value for unsubstituted PEDOT, the conductivity decreased with increasing

PAGE 94

79 substituent length, and then it started increasing again to values even higher (850 S/cm for PEDOT-C 14 ) than for the parent PEDOT (650 S/cm). This trend could be explained by taking into consideration three factors: symmetry of the monomer unit, steric hindrance between repeating units and order induced by mesogenic side chains. PEDOT is highly conductive due to its planar structure along with a negligible steric hindrance between adjacent units. Introduction of a methyl substituent causes asymmetry, especially when the polymerization proceeds via head to head (HH) or tail to tail (TT) coupling. This translates into an increased steric hindrance and a less close 3-D packing, and leads to more disorder and less conjugation. Increasing of the alkyl substituent to a hexyl chain produces more asymmetry, resulting in further drop of conductivity. By extending the chain length to a decyl substituent, the conductivity begins to increase, even though the asymmetry and steric hindrance are similar to those of hexyl substituted PEDOT. The improvement seems to arise from a better film morphology due to self-assembled decyl chains. Our work focused on obtaining free-standing films of these alkyl-substituted PEDOTs to compare the conductivity to the values obtained using the in-situ method. Figure 3.13 shows the conductivities of racemic (R) and chiral (C) alkyl-substituted PEDOT free-standing films with thicknesses up to 35 m. Even though the variation in conductivity with the chain length follows the same trend as described above, none of the substituted materials attained conductivities as high as PEDOT. This could be due to the low solubility of these monomers in the solvents suitable for the electropolymerization. Consequently, we could not use an optimal monomer concentration (of about 0.06 M) or lower the temperature below C. The highest conductivity obtained for an alkyl

PAGE 95

80 substituted PEDOT derivative is 70 S/cm (for PEDOT-C 12 and PEDOT-C 14 ), which is about half of the value obtained for the parent PEDOT. Films prepared in TBAP/ACN showed even lower room temperature conductivities. The slight increase in the conductivity of the chiral compounds is due to a higher degree of side chain ordering, as compared to their racemic counterparts. 0246810121420406080100120 racemic chiral(S/cm)Substituent Length Figure 3.13 Conductivity of alkyl substituted PEDOTs as function of the alkyl chain length. The free-standing films were prepared in TBAPF 6 /PC.

PAGE 96

CHAPTER 4 POLYMER THIN FILM OPTICS 4.1 Introduction Conducting polymers, once studied solely for their ability to replace heavier metal conductors, are now widely used in more dynamic applications where rapid switching from doped (nor p-) to neutral forms is suitable. Applications such as transmissive windows,23-25 displays,51,119 and camouflage materials120 require polymer deposition as thin films on transparent electrodes. A conducting layer of indium-tin oxide (ITO) deposited on boro-silicate glass or flexible poly(ethylene terephthalate) (Mylar) support is widely used as an electrode in the fabrication of conducting polymer devices. Understanding the polymerization mechanism at the polymer-ITO interface is crucial for the quality of resulting films as well as for device lifetime. Section 4.2 deals with preliminary atomic force microscopy (AFM) studies related to polymer nucleation onto an ITO surface. One of the most important characteristics exhibited by the polymers from the PXDOT family is their enhanced electrochromism upon application of a small voltage.17,19 The mechanism behind the electrochromic behavior is related to the doping of materials. In the neutral state, the color is determined by the band gap of the polymer, the absorption peak wavelength, and the relative intensities of the fine structure. Upon doping (removal or addition of electrons), new electronic states are formed in the 81

PAGE 97

82 band gap, thus giving rise to optical absorptions at energies lower than the original band gap. Therefore, in addition to electrochromism in the visible region of the spectrum, these polymers show large absorption changes in the infrared region, making them useful in modulating the reflectance/transmittance off/through a surface over a broad range of the electromagnetic spectrum.121 Owing to this unique behavior, the optical properties of conducting polymers have become the focus of many research groups over the last two decades.122 These studies have focused on PT and its alkyl-substituted derivatives,123,124 and very recently on PEDOT,125-132 and involve IR and Raman spectroscopy coupled with quantum calculations and X-ray diffraction measurements. Due to the strong correlation between electronic structure, morphology and chemical (bond ordering pattern or lattice) structure, the nature of electronic transitions in these materials is still under debate. At the beginning of this chapter, we discuss the doping-induced electronic structure and the infrared active vibrational modes as a function of redox state in PXDOT thin films. Throughout the following sections, we focus our study on the origin of multiple peaks or fine structure on the main absorption peak in these polymers. 4.2 Surface Analysis Understanding the chemical and physical processes taking place at the interface between the electroactive polymer and ITO plays an important role for the quality of the resulting film as well as for the operational capabilities (switching time and lifetime) of electrochromic devices. Tin doped indium oxide (ITO) is routinely used as an electrode in various optoelectronic devices such as electrochromic windows, displays, and OLEDs

PAGE 98

83 due to its unique combination of properties including optical transparency from UV to 2500 nm in the NIR, low electrical resistance (R< 10 ) and excellent surface adhesion.133 ITO electrodes have been reported to interact chemically with electroactive polymers, which possibly contributes to the degradation and eventual failure of polymer devices.134 Even in the absence of oxygen and moisture, MEH-PPV, commonly used in OLEDs, undergoes degradative oxidation losing conjugation through the formation of carbonyl bonds. This oxidation is promoted by the ITO electrode that serves as a source of oxygen. Moreover, after a period of time, devices showed high concentrations of atomic indium originating from the ITO layer.135,136 The general consensus concerning the polymer growth on ITO (and other surfaces) is that during initial stages of the polymerization, growing oligomers deposit on ITO creating nucleation centers randomly distributed on the surface.137 This process is favored by the presence of hydroxyl groups that confer a slight negatively charged ITO surface. Polymer chains are growing from these primary nucleation centers (see Figure 4.1 A). Formation of ordered brush-like film morphology is anticipated at this stage, giving rise to a thin smooth film seen in Figure 4.1 B. Because the ITO surface has been already coated with a polymer film, secondary nucleation occurs on top of the already existing polymer. Precipitated polymer from the proximity of the electrode can also be incorporated into the growing film, leading to a globular surface morphology and increased surface roughness of resulting electroactive layers (Figure 4.2).138 Experimental observations showed that repeated polymer electrochemical depositions followed by gentle wiping off of the polymer layer led to more uniform films

PAGE 99

84 exhibiting enhanced optical properties. In this study, PEDOT films were grown up to 20 mC/cm 2 charge density and then wiped off with a tissue soaked in acetone until the ITO ADCB ADCB Figure 4.1 The model of film formation: A) Primary nucleation centers, B) subsequent polymerization and chain growth, C) Secondary nucleation occurs on the already existing film, D) Incorporation of precipitated polymer at advanced stages of polymerization (from Sapurina137) A B Figure 4.2 Surface morphology of PANI films produced at (A) early stages and (B) advanced stages of polymerization (from Sapurina137)

PAGE 100

85 surface appeared clean to the naked eye. Figure 4.3 A shows an AFM image of a clean ITO revealing its heterogeneous, granular morphology resulting in a very rough surface, also noted by other researchers. For example, Kluger et. al134 has reported ITO surface roughnesses ranging from 25 to 100 nm, depending on the supplier. The image in Figure 4.3 B represents the same ITO surface after a PEDOT electrodeposition/ buffing cycle. The appearance of 1.5 m diameter globular domains across the ITO surface supports the idea that the early stage polymerization proceeds via primary nucleation centers. These centers are very well adhered to the ITO surface (resisting even cleaning with acetone), thus creating anchors for the growing polymer chains. Repeating the above electrodeposition/buffing step several times likely increases the density of primary nucleation centers on the electrode surface, and consequently the uniformity of the resulting films. A B A B Figure 4.3 Surface morphology of a A) clean ITO/glass and B) ITO after a electrodeposition/buffing cycle showing the primary nucleation centers

PAGE 101

86 4.3 Fundamental Spectroelectrochemistry Spectroelectrochemistry plays a key role in probing the electronic structure of conducting polymers as well as in examining the optical changes that occur upon doping.11 The characteristic changes in the set of absorption spectra presented in Figure 4.4 A appear to be general features of conducting polymers. This figure shows the spectroelectrochemistry of an alkyl-substituted PEDOT (PEDOT-C 10 H 21 ), monitored while the polymer was sequentially stepped between its fully oxidized and reduced forms. Neutral PEDOT-C 10 H 21 exhibits an electronic band gap, defined as the onset for the absorbance, of approximately 1.77 eV (700 nm) and shows two well-defined peaks at 592 nm and 650 nm in the visible region, causing the film to appear deep blue. A detailed discussion on the fine structure often seen in neutral conducting polymers is presented in Section 4.5. As the potential is sequentially increased, the intensity of the transition diminishes, along with the concomitant growth of a low-energy absorption located at 1000 nm (1.24 eV) in the NIR region of the spectrum. This is due to the formation of a polaron, a radical cation mobile along the chain (see Section 3.3). In a non-degenerate ground state polymer, single electron oxidation at dilute doping levels results in a partially quinoid-like structure of the polymer chain. This geometric distortion creates new energy states, called polaronic levels and placed symetrically with respect to the band gap center. Basically, four new electronic transitions (a-d) could occur, two of which are dipole allowed, as demonstrated by molecular orbital calculations (Figure 4.4 C).139,140 Transitions c and d, that have the same energy, are weak because associated

PAGE 102

87 electronic levels possess the same symmetry properties. In Figure 4.4 A, transition b occurs at 1000 nm and the NIR tail of transition a starts at about 1200 nm. CBVB Eg(*) CBVB Eg 12 ab CBVB 12 cdefneutralbipolaronpolaron*4006008001000120014000.20.40.60.81.0 Absorbance (a.u.)Wavelength (nm) be a -0.6 V+1 V SOOC10H21nABCD CBVB Eg(*) CBVB Eg 12 ab CBVB 12 cdefneutralbipolaronpolaron CBVB Eg(*) CBVB Eg 12 ab CBVB 12 cdef CBVB Eg(*) CBVB Eg 12 ab CBVB 12 cdefneutralbipolaronpolaron*4006008001000120014000.20.40.60.81.0 Absorbance (a.u.)Wavelength (nm) be a -0.6 V+1 V SOOC10H21n*4006008001000120014000.20.40.60.81.0 Absorbance (a.u.)Wavelength (nm) be a -0.6 V+1 V SOOC10H21n4006008001000120014000.20.40.60.81.0 Absorbance (a.u.)Wavelength (nm) be a -0.6 V+1 V SOOC10H21nABCD Figure 4.4 Doping induced electronic transitions in conducting polymers. A) Energy levels and electronic transitions in a neutral conjugated polymer and the corresponding polaronic and bipolaronic redox states. Strongly dipole-allowed transitions are indicated with a solid arrow.B) Spectroelectrochemistry of PEDOT-C 10 H 21 at applied potentials of .6 to 1 V.

PAGE 103

88 At high applied potentials, the polaron and transition signatures monotonically disappear from the spectrum and a continuous absorption band through the NIR is observed. This new band (e) tails into the visible region of the spectrum starting at about 500 nm, and accounts for the transmissive sky blue color of the highly doped film. During oxidation to high doping levels, two things can happen. Either the second electron is removed from a different polymer chain forming a new polaron or the unpaired electron on an already existing polaron is removed. The latter process produces a spinless dication called a bipolaron. Bipolarons can also be formed when two polarons combine their unpaired electrons into a bond on a doubly charged polymer chain. In bipolaronic state, the intragap levels move closer to the center of the gap. This additional shift could be explained by the fact that the highly doped polymer chain attains a fully quinoid structure. Because the bipolaron levels are unoccupied, only transitions e and f from the top of the valence band can occur. Theoretical calculations predict that transition e is much stronger than the transition f, in agreement with experimental data.140 The theory of electronic transitions for polarons and bipolarons was originated by Su-Shrieffer-Heeger (SSH).53 They predicted that bipolarons are responsible for charge transport at high doping levels. This model was supported by a continuum electron-phonon coupled model proposed by Fesser, Bishop and Campbell (FBC),140 and by magnetic resonance experiments showing a loss in EPR signal as the doping level increases.30

PAGE 104

89 4.4 Reflectance and Transmittance Spectra of PEDOT, PProDOT and PProDOT-Me 2 The spectroelectrochemical experiment described above does not allow the examination of doping-induced optical changes occurring in the mid-IR region of the spectrum. There are important experimental signatures of charge carrier formation in these systems, such as infrared active vibrational modes (IRAV) in the mid-IR due to structural distortions as well as electronic transitions associated with midgap states (polaronic or bipolaronic), lying in the IR region of the spectrum. 11 In order to gain more insight into the electronic structure of PXDOT polymers, thin films of ca. 200 nm in thickness were prepared on ITO/glass using a potentiostatic electropolymerization method described in detail in the Experimental Section. Neutral films were obtained by electrochemical reduction at V, followed by chemical reduction with hydrazine monohydrate (85 % by weight in water), to ensure that all of the dopant ions (ClO 4 in this case) were removed from the film. Neutral films oxidize in air due to their highly electron rich character. Figure 4.5 shows the results of stability studies performed on a 500 nm thick neutral PProDOT film electropolymerized at a charge density of 115 mC/cm 2 This film became lightly p-doped in 24 h, as revealed by the decrease of transition monitored at 580 nm, as well as by an increase in the polaron peak at 900 nm. The doping level reached a saturation level after about 48 hours. Interestingly, after two days of air exposure, both the polaron and peaks decreased concomitant with the increase in the bipolaron NIR tail. The samples analyzed in this ex-situ spectroscopic study were much thinner (about 200 nm) being grown at a charge density of about 40 mC/cm 2 so it was expected that the oxidation, and consequently the change in optical properties, would proceed even faster.

PAGE 105

90 In order to obtain fully reduced samples, both chemical and electrochemical reductions were carried out in an Ar-filled glove bag from Aldrich. Lightly doped samples were obtained from neutral polymers handled in air and analyzed after 24 hours from their synthesis. 0501001502002500.00.51.01.52.02.53.0 Film Thickness: 500 nm (115 mC/cm2) Absorbance at 580 nm Absorbance at 900 nmAbsorbanceTime (hours) Figure 4.5 Neutral PProDOT stability studies performed on a 500 nm thick film by monitoring the decrease in absorption at 580 nm and the variation of the polaronic band at 900 nm. Reflectance and transmittance of PEDOT, PProDOT and PProDOT-Me 2 were measured over a broad range of the electromagnetic spectrum, from the mid-IR (400 cm 1 ) to the UV (45, 000 cm ). All samples consisted of three layers: glass substrate (ca. 0.67 mm), ITO (ca. 250 nm), and polymer (ca. 200 nm), as shown schematically in Figure 4.6. There are several contributions to the reflectance and transmission spectra that are taken into account. First, the light incident on the sample reflects from all three layers, thus absorptions within each layer and reflections at each interface contribute to the total

PAGE 106

91 observed optical spectra. In addition, contributions come from multiple internal reflections within each layer, not shown in Figure 4.6. Figure 4.6 Schematic with polymer/ITO/glass layers Doping induced electronic structure. Optical constants, such as optical conductivity and absorption coefficients can be calculated by applying a multilayer Drude-Lorentz model described in a recently submitted paper.105 For example, absorption coefficients for PEDOT, PProDOT and PProDOT-Me 2 in neutral, lightly doped and heavily doped states are presented in Figure 4.7. All three polymers show similar behavior, with broad electronic absorption bands and narrow vibrational absorption bands. Table 4.1 lists the energies of the *, polaron, and bipolaron transitions, measured at the frequency maximum for each transition. As stated in the previous section, the transition is the main electronic transition occurring in neutral polymers. When the polymers are converted to their doped state, the transition shifts to higher energies due to newly created states from the top of the valence band. This blue shift has a value of about 450 cm -1 for PEDOT, 250 cm -1 for PProDOT and 350 cm -1 for PProDOT-Me 2 Despite the care taken in reducing

PAGE 107

92 and handling the samples, all three neutral polymer films show a very low intensity peak between 2000 and 6000 cm -1 probably due to residual charges trapped in the films. The transition fine structure seen in the spectra of neutral PProDOT and PProDOT-Me 2 will be discussed separately in Section 4.5. ABC ABC Figure 4.7 Absorption coefficients of: A) PEDOT, B) PProDOT, C) PProDOT-Me 2 at three doping levels (neutral, slightly doped and fully doped)

PAGE 108

93 Table 4.1 The electronic structure of PEDOT, PProDOT, and PProDOT-Me 2 in their neutral, lightly doped and fully doped states. E g a, b, and e are defined in Figure 4.4. Polymer Redox State E g (cm -1 /eV) Transition a Transition b Transition e PEDOT Neutral Lightly doped Doped 17,250/2.14 17,700/2.19 5,000/0.62 11,350/1.41 5,700/0.71 PProDOT Neutral Lightly doped Doped 17,750/2.20 18,000/2.23 5,000/0.62 10,500/1.30 5,700/0.71 PProDOT-Me 2 Neutral Lightly doped Doped 17,550/2.18 17,900/2.22 4,000/0.50 11,700/1.45 5,000/0.62 At intermediate doping levels corresponding to polaronic state, three peaks can be easily distinguished, especially for the PEDOT sample. The highest energy peak represents the absorption due to the transition, which is markedly diminished as compared to the neutral polymer spectrum. The middle peak (11,350 cm -1 for PEDOT) is transition b between the two mid-gap polaron levels created upon doping, and the lowest energy one represents transition a from the valence band to the lowest energy polaron level. High doping levels, and consequently bipolaron formation, account for the broad peaks extending from the visible through mid-IR regions of the spectrum. Doping-induced infrared active vibrational modes (IRAV). The resonant frequencies of the phonons (lattice vibrations) occur in the IR spectral region, and the modes that interact directly with light are called IR active. Detailed selection rules for deciding which phonon modes are IR active can be determined by using group theory. Optically active phonons are able to absorb light at their resonant frequency. Photons couple to phonons through the electric field of the light wave. Therefore, this can happen

PAGE 109

94 only if the atoms are charged or the bonds are polar, creating an asymmetric electron cloud between the atoms. This leads to a dipole that can interact with the electric field. Provided the bond has some polar character, its phonons are IR active. If the bonds between atoms are non-polar, there will be no coupling to the light. Upon doping, conducting polymers undergo strong modifications of the local structure around the doping site; electronic charges are displaced, originally non-polar bonds become strongly polar because of charge displacements and dipole changes for a few particular modes become very large, thus giving rise to very strong IR bands.141 These new IRAV are a common feature of doped conjugated polymers and are almost absent in the spectra of neutral materials. For example, doping-induced IRAV modes in PAc are 1,000 times more intense than ordinary infrared active phonons.142 IRAV modes seem to be an intrinsic property of the doped polymers, as they are not affected by the identity of dopant ion.143 The contributions to the absorption coefficient of ordinary and doping-induced phonons are shown in Figure 4.8. Consistent with the previously observed fact that the neutral films contain a small amount of residual dopant, the neutral polymer shows signatures characteristic of the doped form. These doping induced IRAVs are marked with circles in the fully oxidized PEDOT spectrum. In general, these peaks are highly dependent on the doping level, increasing monotonically with increasing oxidation level. The same doping induced IRAVs appear in the neutral PEDOT spectrum and they are marked with asterisks in Figure 4.8 A. The peaks that do not belong to the doped form of PEDOT and do not seem to grow with increasing doping level are identified as being signatures of the pristine polymer.125,126,130,132 This allows us to recognize the peaks

PAGE 110

95 ABC ABC Figure 4.8 Infrared active vibrational modes in: A) PEDOT, B) PProDOT, C) PProDOT-Me 2 A) Doping induced IRAVs are marked with circles in the fully oxidized PEDOT spectrum and asterisks in the neutral PEDOT spectrum. corresponding to the neutral state of the other two polymers as follows: PProDOT (1047, 1367, 1410, 1435, 1470, 1706 cm -1 ) and PProDOT-Me 2 (1027, 1362, 1395, 1435, 1470, 1720 cm -1 ). Peaks at 1047 and 1027 cm -1 are assigned to C-O-C deformation modes, the peaks at 1362 and 1367 cm -1 are due to C -C stretch and C -C inter-ring stretch modes, and the features at 1410 and 1435 cm -1 are assigned to symmetric C =C stretches. At 1470 cm -1 both asymmetric and symmetric C =C stretches appear. The peak at 1706 cm -1 in the PProDOT spectrum, and correspondingly at 1720 cm -1 for PProDOT-Me 2

PAGE 111

96 does not appear in PEDOT spectrum, leading to the assumption that it originates from one of propylene bridge vibrational modes. This approximate peak assignment is in agreement with IR experiments and calculations on PEDOT samples provided by Garreau et.al.125 The above IRAV comparison allows the conclusion that the bonds polarity changes drastically upon charge injection, IR inactive phonons becoming IR active as the polymer is doped. The intensity of the vibrations gives insight into the magnitude of dipole momentum.144 Therefore, the change in symmetry between neutral and doped states could be qualitatively estimated from the difference in their IRAV intensities.128,129 Compared to PEDOT, the propylene analogs have more intense peaks in the doped state and larger differences between the neutral and doped form spectra. This implies that the geometry of these systems undergo deeper changes upon charge injection, as compared to PEDOT. One possible explanation could be that these materials are able to accommodate a higher dopant concentration. Whereas PEDOT monomer unit is planar and the polymer exhibits a high degree of crystallinity, PProDOT and its dimethyl-substituted analog have a more open morphology due to the propylenedioxy ring unit that is bent out of plane and thus increases the interchain distance. This speculation is in agreement with switching speed and transmittance studies performed in our group.17 For example, PProDOT-Me 2 exhibits a 20 % increase in contrast (at max ) as well as a 30 % decrease in redox switching rate, as compared to PEDOT. EQCM measurements could be useful in quantifying the amount of dopant intake upon charge injection for the systems described above.

PAGE 112

97 4.5 The Origin of the Fine Structure in Neutral Polymers The origin of the multiple peaks or fine structure within the main absorption peak in neutral conducting polymers continues to be debated. These peaks seem to be a general feature of conducting polymers in the solid state, although multiple peaks with very low intensities have been reported for the spectra of several polymer solutions.145 Moreover, several groups reported a similar behavior for colloid aggregates with molecular packing similar to that of semicrystalline solids.146 The source of these multiple peaks has been assigned to absorptions by discrete polymer segments with different conjugation lengths,147 to vibronic coupling of excited electrons with a vibrational mode within the polymer backbone,148 and more recently, to Davydov splitting of the exciton levels due to inequivalent sites in the lattice.149,150 Although this issue has been addressed extensively for PT and its alkyl derivatives,148-150 absorption splitting in neutral PXDOT type polymers has not yet been studied. This section deals with a discussion on several possible theories leading to the elucidation of this fine structure, followed by an extensive spectroelectrochemical study on various PXDOTs. Discrete polymer segment absorptions. This theory stems from UV-Vis studies of regioregular alkyl-substituted PTs (PAT) showing three discrete peaks, as seen in Figure 4.9.124 These peaks, due to the transition, are substantial in intensity ranging from 60 to 100% of the max intensity depending on the film thickness. McCullough et. al noted that this splitting corresponds to distinct, reproducible solid state structures with different, yet discrete effective conjugation lengths.151 The long-wavelength absorptions have been interpreted as evidence of structures possessing long

PAGE 113

98 range order,152 and the relative intensity ratio of the peaks is an indication of the extent of order present in the solid films.153 This assumption is supported by UV-Vis studies on films of various thicknesses showing an increased peak resolution with decreasing film thickness.124 Roncali et. al reported similar thickness dependent absorption spectra for electrochemically polymerized poly(3-methyl thiophene).45,47,48 It is well established that thin films have a higher degree of order and consequently higher conductivities than thick films of the same material. Figure 4.9 Solid-State UV-Vis spectra of regioregular poly(3-dodecylthiophene) of various thicknesses (from McCullough124) However, spectra of monodisperse, short-chain oligomers often show a well-defined fine structure similar to the ones shown in Figure 4.9.154 These short-chain oligomers serve as excellent models for investigating, both theoretically as well as experimentally, the properties of their respective conjugated polymers. They are

PAGE 114

99 crystalline, the molecules are planar, and all have the same conjugation lengths. Consequently, the theory that the fine structure originates from chain segments with different conjugation lengths does not seem to apply for all cases studied. Vibronic coupling. Although there is not a general consensus, most of the researchers assign the fine structure to vibronic coupling of excited electrons with phonons.146,155-157 These electron-phonon contributions appear to be a general feature of -conjugated materials. A typical manifestation of lattice relaxations taking place in the excited state is the appearance of vibronic progressions (equally spaced absorption peaks) in the experimental optical spectra.123 As sketched in Figure 4.10.a, vibronic features could not be observed if the equilibrium positions in the ground and excited states were identical and the orthonormality of the electronic transitions would allow transitions between vibrational levels of the two states with the same quantum number. This is based on the classical Frank-Condon principle stating that, during an electronic transition, the nuclei do not have time to move from their original position. In contrast, a displacement of the equilibrium geometries (noted Q) leads to the appearance of several vibronic features, usually between the zeroth vibrational level of the ground state and various vibrational levels of the excited state. It is worth noting that the larger the displacement Q, the more peaks can be seen in the spectra (Figure 4.10 b and c). The lowest energy transition (0-0 transition) is assigned to the relaxed geometry of the excited state (zeroth vibrational level). Conducting polymers are rigid and fully delocalized systems, thus the excited state relaxations (and consequently the vibronic effects) are expected to decrease linearly with increasing degree of polymerization and are thought to

PAGE 115

100 EEEQQQABC0-00-0v=0v=0v=0v=0Q0-0 EEEQQQABC0-00-0v=0v=0v=0v=0Q0-0 Figure 4.10 Schematic representation of the vibronic transition when the ground and excited states are: A) identical, B) weakly displaced, C) strongly displaced (from Bredas123) become insignificant for long chain polymers. This is supported by experiments and calculations on short chain PPV oligomers presented in Figure 4.11.123 The fine structure decreases in resolution, accompanied by a red shift to lower energies, as more monomer units are added to the chain and the effective conjugation length increases. The existence of vibronic coupling in long-chain conjugated polymers indicates that their effective conjugation length is much shorter than the actual chain length due to various

PAGE 116

101 defects that localize the charge carriers. Indeed, single molecule spectroscopy on PPV 158 indicates that its effective conjugation length reaches 10-15 monomer units despite its high molecular weight of about 100,000 g/mol (M w ) that would allow for a much greater electron delocalization. Absorbance (a.u.)Absorbance (a.u.)Energy (eV)Increasing effective conjugation length AB Absorbance (a.u.)Absorbance (a.u.)Energy (eV)Increasing effective conjugation length AB Figure 4.11 Absorption spectra of PPV oligomers with different molecular weights: A) Experimental and B) INDO/SCI simulated. (from Bredas 123 )

PAGE 117

102 However, spectra of alkyl-substituted PT and poly(furans) (PF) are not easy to interpret exclusively with the classical vibronic mechanism.159,160 First, these materials fine structure does not match a vibronic progression of equally spaced peaks commonly seen in systems with strong coupling to only a few vibrations. The transition dipole lies in the plane of the ring, so the transition couples strongly with only one or two totally symmetric, in plane vibrations. Usually, the splitting in PTs has been assigned to C=C symmetric ring stretch, lying in the 1450-1515 cm -1 region.149,150 On the other hand, Curtis et al. reported peak separations of about 1700 cm -1 for alkyl-substituted PF as well as for poly(nonylbisoxazole), too large to be attributed to a totally symmetric ring mode.159,161 The same author noted that the more ordered a material is (i.e. increased head-to-tail content in regioregular polymers), the more resolved is its fine structure. This would contradict the idea that the longer the conjugation length, the less strong the coupling between the vibration mode and the excited electron. Davydov (exciton) splitting. Many conjugated oligomers are crystalline, with two adjacent oligomer chains in one unit cell. These molecules are often translationally inequivalent and the coupling of their dipole moments gives rise to Davydov splitting; in the crystal, the excited states of the single molecule are split into pairs of levels. The optical transition between the ground state and the lowest energy excited state is forbidden.162 The magnitude of the Davydov splitting decreases as the distance between the molecular centers increases and is also correlated to the number of molecules in the unit cell. Therefore, the presence of a high degree of disorder in the system should strongly attenuate this effect. Recently, both the Leclerc and Curtis research groups have reported that the observed fine structure originates from Davydov-type intermolecular

PAGE 118

103 interactions.149,159,161 Both groups analyzed completely crystalline oligomers or polymers with an enhanced degree of order, as confirmed by X-ray diffraction experiments. Moreover, Curtis et. al ascribed the multiple peaks noticed on regioregular, alkyl-substituted PF absorption spectra to a combination of Davydov splitting and exciton band structure exhibiting vibronic coupling.159,160 4.6 Spectroelectrochemical Data of PXDOT Series A major goal in the electrochromic and electroactive polymer field is to tune the materials optical properties via main chain and pendant group structural modification. Polyheterocycles, in particular PEDOT, are of interest to several electrochromic applications due to their outstanding environmental and electrochemical stability, as required by operation conditions.16 As PEDOT and PProDOT derivatives are considered the forefront of electrochromics, we chose them as a basis set for our spectroelectrochemical investigations. Absorption spectra for several compounds in their neutral and fully p-doped sates are presented in Figure 4.12, and their band gaps, visible light absorption peak locations, as well as peak-to-peak separations are summarized in Table 4.2. The absorption spectra in the polymers doped and neutral states were recorded while scanning throughout the visible region, from 350 nm to 850 nm. For an accurate comparison, all of the films were electrodeposited on ITO transparent electrodes at charge densities of about 25 mC/cm 2 This allowed the synthesis of polymer films with thicknesses ranging from 150 to 300 nm. The deposition conditions, including solvent, electrolyte, and monomer concentration (10 mM monomer/0.1 M TBAP/ACN), were identical for all of the

PAGE 119

104 4005006007008000.00.20.40.60.81.01.2 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.8 AbsorbanceWavelength (nm)4006008000.00.51.01.52.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.01.21.41.61.8 AbsorbanceWavelength (nm) SOOn SOOCH3n* SOOCH3n SOOC6H13n* SOOC6H13n SOOC10H21n* SOOC10H21nABCDEFGH4005006007008000.00.20.40.60.8 AbsorbanceWavelength (nm) SOOSOOnp1p2p34005006007008000.00.20.40.60.81.01.2 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.8 AbsorbanceWavelength (nm)4006008000.00.51.01.52.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.0 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.01.21.41.61.8 AbsorbanceWavelength (nm) SOOn SOOCH3n* SOOCH3n SOOC6H13n* SOOC6H13n SOOC10H21n* SOOC10H21nABCDEFGH4005006007008000.00.20.40.60.8 AbsorbanceWavelength (nm) SOOSOOnp1p2p3

PAGE 120

105 4005006007008000.00.51.01.52.02.5 AbsorbanceWavelength (nm)I4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) SnOO SnOOH3CCH3 SOOSnOO SOOC14H29n SOOC12H25n*JKL4006008000.00.20.40.60.8 AbsorbanceWavelength (nm) SnOO4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) SOOOC7F15OnM4005006007008000.00.10.20.30.40.50.60.70.80.9 AbsorbanceWavelength4005006007008000.000.050.100.150.200.250.300.350.40 AbsorbanceWavelength (nm) SOOnOPNp2p1p3p4p5p64005006007008000.00.20.40.6 AbsorbanceWavelength (nm)4005006007008000.00.51.01.52.02.5 AbsorbanceWavelength (nm)I4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm)4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) SnOO SnOOH3CCH3 SOOSnOO SOOC14H29n SOOC12H25n*JKL4006008000.00.20.40.60.8 AbsorbanceWavelength (nm) SnOO4005006007008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) SOOOC7F15OnM4005006007008000.00.10.20.30.40.50.60.70.80.9 AbsorbanceWavelength4005006007008000.000.050.100.150.200.250.300.350.40 AbsorbanceWavelength (nm) SOOnOPNp2p1p3p4p5p64005006007008000.00.20.40.6 AbsorbanceWavelength (nm) Figure 4.12 Visible absorption spectra for neutral (solid line) and oxidized (dotted line) forms of A) PEDOT, B) PBiEDOT, C) PEDOT-CH3 chiral, D) PEDOT-CH3 racemic, E) PEDOT-C6H13 chiral, F) PEDOT-C6H13 racemic, G) PEDOT-C10H21 chiral, H) PEDOT-C10H21 racemic, I) PEDOT-C12H25 chiral, J) PEDOT-C14H29 racemic, K) PProDOT, L) PBiProDOT, M) PProDOT-Me2,N) PProDOT-Bu2, O) PEDOT-Ph, P) PEDOT-F

PAGE 121

106 Table 4.2 Spectroelectrochemical results for PXDOT series Polymer max (nm) E g (nm) eV Peaks and aspect of the neutral polymer absorption Peak to peak separation Comments PEDOT 610 750 nm 1.65 eV Broad single peak Highly planar structure PEDOT-F 608 750 nm 1.65 eV Broad single peak PEDOT-Ph 610 1.65 eV Broad single peak PEDOT-CH 3 (C and R) 600 750 nm 1.65 eV Peaks start to separate forming tiny shoulders PEDOT-C 6 H 13 (C and R) 656 710 nm 1.75 eV p 1 (shoulder): 552 nm (18116 cm -1 ) p 2 : 597 nm (16750 cm -1 ) p3: 656 nm (15244 cm-1) p 1 -p 2 :1366 cm -1 p2-p3:1506 cm-1 p 2

p 3 PProDOT-Me 2 580 645 nm 1.92 eV p 1 (shoulder): 537 nm (18623 cm -1 ) p 2 : 580 nm (17241 cm -1 ) p3: 625 nm (16000 cm-1) p 1 -p 2 :1382 cm -1 p2-p3:1357 cm-1 p 2 >p 3 PProDOT-Bu 2 712 nm 1.74 eV p 1 (shoulder): 531 nm (18832 cm -1 ) p 2 : 575 nm (17391 cm -1 ) p3: 631 nm (15848 cm-1) p 1 -p 2 :1441 cm -1 p2-p3:1543 cm-1 p 2 and p 3 have similar intensities PBiProDOT 514 652 nm 1.53 eV p 1 (shoulder): 421 nm (23753 cm -1 ) p 2 (shoulder): 448 nm (22321 cm -1 ) p 3 : 478 nm (20920 cm -1 ) p 4 : 514 nm (19455 cm -1 ) p 5 : 557 nm (17953 cm -1 ) p6(shoulder): 607 nm (16474 cm-1) p 1 -p 2 :1432 cm -1 p 2 -p 3 :1401 cm -1 p 3 -p 4 :1465 cm -1 p 4 -p 5 :1502 cm -1 p 5 -p 6 :1479 cm -1 P 5 has the highest intensity

PAGE 122

107 polymers studied. The potential was maintained at .7 V while recording the neutral polymer spectrum, and at +1 V for the p-doped case. We observed that while neutral PEDOT, PBiEDOT, PEDOT-F, PEDOT-Ph and PEDOT-CH 3 feature a single broad absorption peak, all of the other polymers show a fine structure on the main absorption peak. As a general trend, the absorption spectrum is split into three peaks: two very well defined low energy peaks (noted p 2 and p 3 in this study), and a distinguishable shoulder (p 1 ) located at shorter wavelengths, as marked on Figure 4.12 E. The exception is PBiProDOT (where BiProDOT is the starting material) that shows a total of six peaks, of which three are well-defined and three are shoulders. The energy differences between adjacent peaks range from 1290 cm -1 to 1507 cm -1 All polymers switch between a dark blue or purple neutral form to a highly transmissive blue oxidized state, exhibiting high contrasts up to 78 % (for PProDOT-Me 2 ) at their max We observed interesting trends while comparing the results obtained for PEDOT and its alkyl-substituted derivatives. First, the absorption spectra of chiral and racemic alkyl-PEDOTs are identical to the smallest detail. Theoretically, systems having alkyl substituents linked to the polymer backbone through a chiral center are more ordered. In this case, the order induced by the chiral carbon does not influence the polymer absorption spectrum, mainly due to the fact that these materials are regiorandom, as they were obtained by electropolymerization. Second, increasing the substituent length from methyl to hexyl induces the appearance of the fine structure on the absorption spectrum. Unsubstituted PEDOT shows a broad featureless peak centered at 610 nm. PEDOT-CH 3 behaves similarly, with the observation that tiny shoulders start to be distinguishable. In contrast, PEDOT-C 6 H 13 spectrelectrochemistry features very sharp peaks at 597 and 656

PAGE 123

108 nm, and a shoulder at 552 nm. Moreover, its band gap is slightly higher (1.75 eV) than that of PEDOT and PEDOT-CH 3 (1.65 eV). The band gap energy increases along this series reaching a maximum for PEDOT-C 10 H 21 and then decreases as the alkyl chain length is further increased to twelve and fourteen carbons, as reflected in Figure 4.13. Besides exhibiting the highest band gap of 1.77 eV, the absorption spectra of PEDOT-C 10 H 21 features two peaks of similar intensities at 592 and 650 nm, whereas in the other alkyl-substituted PEDOT, p 3 is much more intense than p 2 024681012141.601.651.701.751.80 Band Gap (eV)Alkyl Substituent Length Figure 4.13 Band gap dependence of the side chain length in alkyl-substituted PEDOTs. The curve represents the best polynomial fit for the data. The fine structure is seen in all of the neutral polymers derived from PProDOT. As compared to PEDOT derivatives, PProDOT and PProDOT-Me 2 have higher band gaps of 1.84 eV and 1.92 eV respectively, probably due to a slightly less planar polymer backbone. Interestingly, the same relationship is seen in the dioxypyrrole derivatives

PAGE 124

109 family, where PEDOP exhibits a band gap of 2 eV whereas PProDOP has a higher band gap of 2.2 eV. Contrasting with the results described above for PEDOT derivatives, PProDOT and PProDOT-Me 2 feature a more intense middle peak p 2 than p 3 However, changing the substituents from dimethyl to dibutyl induces an unexpected transformation in the envelope of the neutral polymer absorption spectrum. For PProDOT-Bu 2 p 2 has a slightly lower intensity than p 3 and the shoulder p 1 becomes more pronounced. Moreover, p 2 and p 3 are less defined and appear broader than for PProDOT and PProDOT-Me 2 Theoretically, there are four electrochemical routes leading to the synthesis of the same compound (for example PProDOT), as depicted in Figure 4.14.163 The leaving group in electrochemical desilylation is the trimethylsilyl (TMS) cation, similar to the loss of hydrogen during the electrochemical polymerization of conventional monomers. PProDOT growth during accumulative synthesis is presented in Figure 4.15 A-C. While polymer growths from PProDOT and PProDOT-TMS 2 show similar behavior regarding both electrochemical (Figure 4.15 B and C) and optical aspects (Figure 4.15 D and E), the polymers derived from PBiProDOT or PBiProDOT-TMS 2 exhibit several new features. First, the monomers oxidation potential is lower (0.5 V vs. Ag/Ag + ), as expected for a molecule possessing a higher extent of electron delocalization and enhanced capabilities of balancing the positive charge created during oxidation. Second, the resulting polymer spectroelectrochemistry is somewhat surprising, showing six distinct peaks spaced 1400-1500 cm -1 apart, and a much lower band gap of 1.53 eV. In addition, its bipolaron NIR tail lies in the visible region (starting at 450 nm), thus reducing the transparency of doped films. Similarly, PBiEDOT has a deeper blue color in the oxidized form as compared to

PAGE 125

110 PEDOT, although the spectroelectrochemistry of PBiEDOT does not feature a fine structure. SOOSOOSOOSi(CH3)3(CH3)3SiSOOSOOSi(CH3)3(CH3)3SiSOOSOOnHHHH Figure 4.14 Four electrochemical polymerization routes to PProDOT Interpretation of results. In this study we aim to get insight into the origin of the fine structure seen on the absorption spectra of neutral polymers from PXDOT family of electrochromes. Although electropolymerization is a suitable synthetic method for a large number of applications, it gives rise to polymers that are mainly amorphous with a low content of crystallinity.12 Of the entire PXDOT family, the parent PEDOT shows by far the highest conductivity due to its planar structure along with a negligible steric hindrance between adjacent units.112 This suggests that PEDOT has the longest effective

PAGE 126

111 -1.0-0.50.00.51.01.5 E (V) vs. Ag/Ag+AB SOOSOO SOOSi(CH3)3(CH3)3Si SOOC4006008001000120014000.00.20.40.60.8 AbsorbanceWavelength (nm) SOOSi(CH3)3(CH3)3Si4006008001000120014000.00.20.40.60.81.0 AbsorbanceWavelength (nm) SOOSi(CH3)3(CH3)3SiSOODE -1.0-0.50.00.51.01.5 E (V) vs. Ag/Ag+AB SOOSOO SOOSi(CH3)3(CH3)3Si SOOC4006008001000120014000.00.20.40.60.8 AbsorbanceWavelength (nm) SOOSi(CH3)3(CH3)3Si4006008001000120014000.00.20.40.60.81.0 AbsorbanceWavelength (nm) SOOSi(CH3)3(CH3)3SiSOODE Figure 4.15 Potential sweep growth of PProDOT from A) BiProDOT, B) PProDOT, and C) PProDOT-TMS2 at a scan rate of 20 mV/s in 0.1 M TBAP/ACN. Spectroelecrochemistry of D) PProDOT-TMS2 and E) PBiProDOT-TMS2 in 0.1 M TBAP/ACN at applied potentials from .7 to 1 V vs. Ag/Ag +

PAGE 127

112 conjugation length due to absence of defects that would localize the carriers. The planarity of the monomer unit allows for a better packing of the monomer chains, thus enhancing -stacking and consequently the materials conductivity via interchain hopping. In addition, X-ray diffraction studies performed by our group (Section 3.9) and others115 confirm that PEDOT has the highest amount of crystallinity among its derivatives. Moreover, substituted PEDOTs are polydisperse and regiorandom, confirmed by the fact that introduction of a chiral center at the carbon responsible for the alkyl chain linkage does not affect the absorption characteristics. Figure 4.16 is a schematic of a regiorandom chain with H-H and T-H couplings resulting in a twisted backbone, as compared to a regioregular chain with planar backbone. The introduction of alkyl substituents of lengths up to ten carbons in regiorandom polymers causes asymmetry and steric hindrance, resulting in less order and less conjugation,117 as confirmed by the blue shift in absorption spectra and the increase in polymers band gap. Even though the steric hindrance is the same as above, extension of the alkyl chain to twelve and fourteen carbons appears to have a positive effect, inducing some degree of order in the system by a side-chain self-assembly mechanism. This fact is confirmed by conductivity studies on doped alkyl-substituted PEDOT films performed in our group on free-standing films (Section 3.12), and by Groenedaal et. al on thin films deposited on interdigitated electrodes, using poly(3-methyl thiophene) as conductivity standard.117 In this context, it is unlikely to consider the observed splitting of the absorption spectrum as originating from polymer segments with discrete absorptions. This is in agreement with the fact that the more ordered neutral PEDOT films exhibit a broad featureless absorption. Moreover, this is strongly supported by Mass Spectrometry

PAGE 128

113 experiments performed by Welsh et. al on soluble PProDOT derivatives.145 These experiments reveal the presence of a broad distribution of chain lengths in polymers exhibiting multiple peaks in their absorption spectra. SOOSOOSOOSOOSOOSOOSOOSOOSOOSOOSOOSOOHead to TailHead to HeadTail to TailAB SOOSOOSOOSOOSOOSOOSOOSOOSOOSOOSOOSOOHead to TailHead to HeadTail to TailAB Figure 4.16 Regioregularity of alkyl-substituted PEDOTs. A) Regioregular chain showing exclusively head-to-tail interactions, resulting in a planar backbone, B) Regiorandom chain, showing head-to-tail and tail-to-tail interactions resulting in twisting of the backbone

PAGE 129

114 In a similar manner, Davydov splitting mechanism has been proposed for materials with a high content of crystalline regions. However, the magnitude of this phenomenon decreases drastically when the crystallite size is reduced or when introducing disorder in a system. Again, if this theory would apply to our basis set, then unsubstituted PEDOT would have to exhibit multiple absorption peaks. When considering the origin of the fine structure discussed in this chapter, vibronic coupling of excited electrons with a vibrational mode of the monomer unit seems intuitively appealing. To further investigate this possibility, we calculated the peak-to peak separations for our basis set and the results are summarized in Table 4.2. The polymers showing three peaks appear to follow a general trend. The separation between p 2 and p 3 is in the range of 1500 cm -1 whereas the separation between the shoulder p 1 and p 2 is in the range of 1300 cm -1 These peak-to-peak separations correspond to in-plane vibrations of the monomer unit rings (the totally symmetric, in-plane vibration of C =C bond or weakly coupled stretching vibration of C -C bond or C -C inter-ring stretch). The three vibrations described above possess the strongest force constants of all the possible vibrations in PEDOT-type polymers, as reported by Garreau et. al.125,126 Moreover, the absorption spectrum of PBiProDOT is charateristic of a vibronic progression of almost equally spaced peaks, commonly seen in systems with a strong displacement of the equilibrium geometry in the excited state. Inter-peak distances in this material range from 1401 to 1502 cm -1 suggesting that the C =C symmetric stretch vibration is responsible for coupling. PEDOT, that presumably has a longer effective conjugation length as evidenced by X-ray115 and conductivity measurements,112 does not show vibronic coupling. This is in excellent agreement with the theory that coupling

PAGE 130

115 between delocalized electronic excitations and the localized vibrational modes decreases as the electrons are free to move over more rings. Alkyl-substituted PEDOTs have a shorter effective conjugation length due to some content of H-T coupling that disrupts and shortens the polymer conjugation length.117,118 Consequently, these materials exhibit a strong vibronic coupling, with the exception of PEDOT-CH 3 where the small methyl substituent does not pronouncedly affect the conjugation length. Garreau et al. reached the same conclusion while comparing PEDOT and PEDOT-C 14 Raman and IR spectra.128,129 Similarly, PProDOT and its derivatives bear out of plane propylenedioxy rings, thus helping localize the charge carriers and enhancing the vibronic coupling. The intensities of the peaks are weighted by the overlap of the vibrational wavefunctions, the lowest energy transition being the transition to the relaxed state (v=0) of the excited state.123 Alkyl-substituted PEDOTs show a very intense low energy peak (p 3 ), meaning that the excited state is only slightly displaced from the equilibrium position. On the other hand, the middle peak (p 2 ) is more intense in PProDOT and PProDOT-Me 2 suggesting that the transition to first excited vibrational level (v=1) dominates the spectrum and the differences between geometries of ground and excited states are larger. Somewhat surprisingly, PEDOT-Ph and PEDOT-F do not possess vibronic coupling, presumably because the above substituents do not disrupt PEDOT conjugation length. XDOT monomers solutions show the fine structure on the UV absorption peak, suggesting once again that an in-plane vibration of the monomer unit is responsible for the splitting. In addition, EDOT, EDOT-Ph and EDOT-F absorption spectra reveal the vibronic coupling, thus supporting the theory that the increased effective conjugation

PAGE 131

116 length is responsible for the disappearance of this fine structure in the respective neutral polymers spectra. In conclusion, we showed that optical spectra of conjugated polymers are a very rich source of data. Vibronic coupling is an appealing theory for the origin of splitting in PXDOTs, as the polymers exhibit multi-peak absorption spectra attributed to in-plane, symmetric ring stretch modes. Performing the above experiments at low temperatures would give insight into the degree of inhomogeneous broadening of these peaks. Effective conjugation length studies based on Raman spectroscopy will nicely complement our deductions.

PAGE 132

CHAPTER 5 VARIABLE REFLECTANCE MIRRORS 5.1 Introduction Electrochromic (EC) materials exhibit reversible and highly stable changes of their optical properties upon the application of a voltage.164 Lately, the definition of electrochromism has been extended from that of a color change in the visible spectrum to multi-spectral energy modulation that might cover ultraviolet, near infrared, mid infrared, and microwave regions, with color corresponding to the response of detectors at these wavelengths.165 There is a need for flexible flat panel displays for use across a broad portion of the electromagnetic spectrum. The displays must be thin and able to cover objects of various shapes and sizes.166 In addition, they must exhibit large changes in reflectivity when switched between doped and undoped states, long lifetimes and rapid redox switching.167 Interest in these types of advanced materials and devices stems from applications such as thermal control of low earth orbit satellites and camouflage countermeasures against night vision. 168 Moreover, the NIR region is of particular technological interest, being used for optical data transmission through silica fibers, which exhibit absorption minima at 1300 and 1550 nm.169 There are several categories of EC materials that are capable of modulating both visible and infrared light. Among them, highly disordered transition metal oxides (especially tungsten oxide) have been studied extensively, due to their broad polaron 117

PAGE 133

118 absorption in the 0.4-2 m range.100,170 Difficulties encountered in processing, as well as relatively slow response times (many seconds to minutes), have opened up opportunities for the development of different types of EC materials. 169,171 Of these, redox electroactive and conducting polymers represent a particularly attractive class of EC materials, as they are easily electrochemically deposited or spun coat as thin films.168 Several hybrid organic-inorganic infrared switching devices have been reported, with the conducting polymer as the active layer and a transition metal oxide as a counter electrode material.172 Although several U.S. patents have been advanced,173,174 there are only few publications concerning all polymer variable reflectance mirrors.121,168 Of the many redox electroactive polymers reported, some of the most promising for use in ECDs are based on poly(3,4-ethylenedioxythiophene) (PEDOT) and its derivatives, in particular PProDOT-Me 2 In this chapter we address the optimization of ECDs that operate in the reflective mode and are able to modulate the reflectivity in the visible, NIR and mid-IR regions of the spectrum. 5.2 Devices as Platforms to Study Electronic Spectra As a device platform which conveniently allows EC property characterization in a reflective mode, we have used an outward facing active electrode device sandwich structure originally described in the patent literature, and shown schematically in Figure 5.1. This ECD structure has several benefits. First, the properties of the EC material of interest can be probed through a window chosen to be highly transmissive over the wavelength range of interest. Second, all of the materials can be flexible, allowing

PAGE 134

119 significant mechanical deformation without hampering device operation. Finally, by using a high viscosity electrolyte, the device can be made self-sealing. Gold / Mylar Counter Electrode Counter Electrode Polymer Porous Electrode Separator SlittedGold / Mylar Working Electrode Surface ActiveRedoxPolymer RadiationTransmissiveWindow Polyethylene Substratehh Gold / Mylar Counter Electrode Counter Electrode Polymer Porous Electrode Separator SlittedGold / Mylar Working Electrode Surface ActiveRedoxPolymer RadiationTransmissiveWindow Polyethylene Substratehh Figure 5.1 Side-view schematic diagram of a dual polymer electrochromic device for surface reflectivity control in both visible and IR regions of the electromagnetic spectrum. In the construction used in this work, gold-coated Mylar sheets are used as both counter and working electrodes. The top electrode was cut with a series of parallel slits, separated by about 2 mm, across the active surface (see Figure 5.2) making it porous to ion transport during switching. The cell was assembled using a high viscosity polymeric electrolyte composed of LiClO 4 dissolved in an acetonitrile (ACN) /propylene carbonate (PC) swollen poly(methyl methacrylate) (PMMA) matrix. At the edges of the device, the ACN in the electrolyte evaporates, leaving behind the PMMA and LiClO 4 in PC. As the PMMA becomes insoluble, it seals the outer edges of the device and provides self-encapsulation. The use of this electrolyte minimizes further solvent evaporation, prevents leaking, and allows for long-term testing of the ECD. Both the active top layer and the

PAGE 135

120 counter polymer film were electrochemically deposited on the gold-coated Mylar electrodes from solutions of 10 mM monomer in 0.1M LiClO 4 in ACN at constant potential. A separator paper, soaked in electrolyte, was used to isolate the back of the working electrode from the counter polymer layer. The top layer is in contact with a window, which is transmissive to the wavelengths of interest, allowing accurate measurements of the active layer reflectivity. We typically use ZnSe for NIR to mid-IR, glass in the NIR and visible, and polyethylene for visible through mid-IR, with somewhat lower performance from the latter on account of IR absorption bands. With this design, only the outward facing electroactive polymer is responsible for the surface reflectivity modulation, whereas the counter electrode polymer is used for charge and coloration balance. Figure 5.2 Top-view photograph of a PProDOT-Me2/PBEDOT-NMeCz dual polymer electrochromic device in its two extreme states. Left: surface PProDOT-Me2 in neutral state at -1.1 V, Right: surface PProDOT-Me2 in oxidized state at +1.1 V. Figure 5.2 shows a photograph of a device based on PProDOT-Me 2 as the active top layer, and poly[3,6-bis(2-(3,4-ethylenedioxy)thienyl)-N-methylcarbazole] (PBEDOT-NmeCz) as the back layer. As the top film on the device is switched from its neutral, colored state, to its p-doped, bleached state, a gradual and controllable transition from a dark opaque violet to a reflective yellow is observed, as the transparent polymer allows

PAGE 136

121 the gold to be viewed. This color change is due to the doping process that modifies the electronic band structure of the polymer, producing new electronic states in the gap and bleaching the -* transition; consequently the electronic absorptions are shifted toward lower energies outside of the visible region. 5.3 In-situ Reflectivity Measurements In order to probe the electronic properties of PEDOT compared to PProDOT-Me 2 we constructed devices having these two polymers as the active layers. Figure 5.3 shows the reflectance of the sandwich structures over 0.3 m (4-0.25 eV) for a PProDOT-Me 2 based device and up to 15 m for a device having PEDOT as the active layer at several cell voltages. Visible and near-IR results for the PProDOT-Me 2 cell were measured on a 150 nm thick polymer film under a glass window; the mid infrared data on a 200 nm thick polymer film with a ZnSe window. The in-situ reflectance data for the PEDOT-based cell were recorded for a 350 nm thick active layer covered with a polyethylene window. The parasitic reflectivity of these windows has been subtracted in these spectra. However, for practical applications, the top windows can be coated with an anti-reflective material. Although the device based on PProDOT-Me 2 exhibits higher optical contrast in the visible, as well as NIR regions than PEDOT-based device, both variable reflectance mirrors show similar features. Several layers in our device influence these spectra: the electrochromic polymer, the electrolyte gel, and the underlying gold electrode. Two strong absorption bands from O-H (2.8 m) and C-H stretching (3.3.4 m) are seen in all of the spectra; these arise from water in the electrolyte, electroactive polymer and the

PAGE 137

122 gel electrolyte, and are relatively constant as the polymer is oxidized and reduced. The fully reduced polymer (-1.5 V; solid line) is strongly absorbing in the visible region (0.40.65 m) and, thus the reflectance of the device is low in this region. At wavelengths longer than 0.9 m (1.4 eV), this polymer layer becomes quite transparent, so the gold layer underneath the polymer dominates the reflectance. At still longer wavelengths vibrational features in the polymer become evident, and the device reflectance is diminished. When fully oxidized, i.e., doped, (+1.0 V, dotted line), the visible absorption is bleached; at the same voltage a strong infrared absorption appears, hiding the underlying gold electrode. The infrared absorption (and the contrast of our devices) is strongest at ca.1.8 m for PProDOT-Me 2 and 1.3 m for PEDOT. At 1.8 m, we detect a reflectance contrast, R, of greater than 90% for PProDOT-Me 2 This contrast ratio is highly enhanced when compared to results for PEDOT (R=50-55%) and polyaniline derivatives.173,174 At longer wavelengths (4-5 m), the devices reflectance increases somewhat with the polymer in the doped state, and decreases with the polymer in the undoped state, reducing the contrast to R = 60 % for PProDOT-Me 2 and 25 % for PEDOT. At intermediate oxidation states (0 V, -0.5 V, and -1 V) the devices have a reflectance that generally is intermediate between that of the fully oxidized and fully reduced states, as seen in Figure 5.3 A. The exception to this behavior is the doping induced band around 1.2 m (1.2 eV) which is strongest at these intermediate doping levels. The physical interpretation of these spectra is that the undoped insulating polymer has its interband transition in the visible, and is transparent (except for vibrational absorptions) in the IR region. Light doping produces two sub-gap absorption

PAGE 138

123 B0.02.55.07.510.012.515.00.00.10.20.30.40.50.60.7 polyethylene windowReflectanceWavelength (m) doped +1V neutral -1.5 Va)b) PEDOT/PBEDOT-NMeCz PProDOT-Me2/PBEDOT-NMeCzReflectanceWavelength (m)1.02.03.04.05.00.00.20.40.60.81.01.21.02.00.50.3Photon Energy (eV)Aa)b)c)d)e)f)B0.02.55.07.510.012.515.00.00.10.20.30.40.50.60.7 polyethylene windowReflectanceWavelength (m) doped +1V neutral -1.5 Va)b) PEDOT/PBEDOT-NMeCz PProDOT-Me2/PBEDOT-NMeCzReflectanceWavelength (m)1.02.03.04.05.00.00.20.40.60.81.01.21.02.00.50.3Photon Energy (eV)Aa)b)c)d)e)f) PProDOT-Me2/PBEDOT-NMeCzReflectanceWavelength (m)1.02.03.04.05.00.00.20.40.60.81.01.21.02.00.50.3Photon Energy (eV)Aa)b)c)d)e)f) Figure 5.3 Reflectance spectra: A) Reflectance of the PProDOT-Me2/PBEDOT-NMeCz electrochromic device spanning the visible to the mid-IR regions of the electromagnetic spectrum at various applied potentials. a) uncoated gold surface; b) .5 V (PProDOT-Me2 insulating, undoped); c) .0 V; d) .5 V; e) 0.0 V; f) +1.0 V (PProDOT-Me2 conducting, p-doped). B) Reflectance of the PEDOT/PBEDOT-NMeCz dual polymer electrochromic device spanning the visible to the far-IR regions of the electromagnetic spectrum at various applied potentials. a) .5 V (PEDOT insulating, undoped); b) +1.0 V (PEDOT conducting, p-doped)

PAGE 139

124 bands, due to the presence of polaron states, and partial bleaching of the interband transition. At full doping, bipolarons are formed, and the polaron interband transition is absent, as described in detail in Chapter 4. 5.4 Active Polymer Layer Thickness Optimization In order to obtain a high contrast over a broad infrared frequency range, we studied the influence that the electroactive polymer film thickness exerts on the reflectance contrast for devices having PEDOT as the active polymer film. In these experiments, PBEDOT-NMeCz was used as the counter redox polymer and polyethylene as transmissive top window. We prepared several devices containing PEDOT films of various thicknesses ranging from 30 to 750 nm. Figure 5.4 shows the contrast of devices over a broad frequency range for three active layer thicknesses. In-situ change in reflectance for a 62 nm thick film is small at low frequencies because both neutral and doped PEDOT are highly transmissive in this range. However, at higher frequencies (midand near-infrared) the reflectance of the neutral and doped states differs. This is due to a less absorptive neutral polymer (* transition tail), and a highly absorptive doped material (bipolaronic band). In the visible, at wavelengths shorter than the polymers isosbestic point, the relative absorption is reversed, the active layer exhibiting a lower reflectance in neutral state than in the doped state. The device having a 250 nm thick active layer shows similar absorption trends, with a modest contrast in the far infrared and a relatively large contrast in the mid-infrared and visible regions. On the other hand, the contrast for the device based on a 500 nm active polymer film is large in the far-infrared range. However, at high frequencies

PAGE 140

125 (midand near-infrared) the contrast between the neutral and doped states is slightly lower than that of the device containing a 250 nm thick PEDOT layer. Note that there is zero contrast in some areas of the mid-infrared region (500 cm -1 ), where strong absorption bands from gel electrolyte and polyethylene window occur. These absorptions mask the reflectance of the device, thus rendering it impossible to distinguish between reflectance data of neutral and doped states. In addition, there is a zero-crossing at the near infrared/visible edge, which is related to the isosbestic point of the PEDOT, suggesting that at this frequency, the absorption is independent of doping level. Figure 5.4 Device contrast as a function of frequency for 62, 250 and 500 nm thick PEDOT active layer

PAGE 141

126 Figure 5.5 shows the thickness dependence of the device contrast at various frequencies in the far-, mid-, and near-infrared range, allowing contrast optimization for the frequency of interest. Interestingly, in the far-infrared (239 cm -1 ) the contrast is greatest for the thickest film (750 nm). At higher frequencies, enhanced optical contrasts are attained by using film thicknesses ranging from 100 to 350 nm. The highest contrast obtained for a PEDOT-based device is 90 % for a 350 nm thick active layer at 7,692 cm -1 (1.3 m) in the NIR region of the spectrum. Figure 5.5 Active layer thickness dependence of the electrochromic contrast in PEDOT based devices, monitored at several NIR, mid-IR and far-IR frequencies

PAGE 142

127 5.5 Switching Time The response time of a device is defined as the time needed to switch between the two extreme redox states of the active polymer layer. In order to study the switching time, the device was connected to a power supply that applies a square-wave potential, allowing for potential period and amplitude control. The measurement was performed with a Bruker 113v FTIR spectrometer. We used a PProDOT-Me 2 electrochromic cell having 200 nm thick PProDOT-Me 2 films as both active and counter polymer. The time interval between in-situ reflectance measurements was 1.5 seconds. Figure 5.6 shows a typical cycle: the square wave potential represented in the upper panel and graphs of reflectance vs. time represented in the lower panel for two fixed frequencies (2,650 and 3,850 cm -1 ), where the gel electrolyte absorptions are very small. The cell voltage was switched between +1.12 V and .52 V, and held at these voltages for about 20 seconds to ensure a full redox switch. The reflectance change during doping and dedoping processes initiated at the slits and spread over the whole active polymer surface. As expected, the switching time in this type of devices is strongly dependent of the distance between slits, and the position of monitored area relative to these cuts. A switch from the neutral to fully p-doped state requires about 3 seconds, whereas switching from the doped to the neutral state requires about 7.5 seconds. This difference in switching times suggests that the redox exchange ions leaving the film during the p-doping process and entering the polymer while dedoping are Li + ions from the gel electrolyte. It should be noted that this switching time is due to the diffusion of the ions through the gel electrolyte and along the working electrode slits. Improving the design of the ECD by using a highly porous working electrode (gold coated microporous

PAGE 143

128 membrane) and increasing the electrolyte conductance will decrease the optical response time and may enhance the chromatic contrast of the device. AB AB Figure 5.6 Switching time of PProDOT-Me2 based device recorded at two IR frequencies.

PAGE 144

129 5.6 Long-Term Switching Stability A key characteristic of an ECD of this type for displays or for thermal control applications is the lifetime of the device, i.e., the number of switching cycles a cell can undergo before a certain amount of electrochromic contrast is lost. The redox stability of the cell was determinated by continuously switching the device between its fully absorptive and transmissive states using a switching period of 25 s. After several double potential steps, the reflectivity was measured at a fixed frequency (7,692 cm -1 1.3 m), where the device contrast is high using a Zeiss MPM 800 microscope photometer. There are several factors affecting the long-term switching capabilities of these variable reflectance mirrors including polymer film thickness, polymer adhesion to the conducting substrate, electrolyte composition, and operational voltage. For example, a PProDOT-Me 2 based cell using the LiClO 4 electrolyte performs hundreds of switches without any degradation of the working electrode. After 1500 switches, the reflectivity of the oxidized form of the polymer has the same initial value, but the neutral form exhibits a decrease in reflectivity. As the device was assembled in air, the oxygen-sensitive neutral form is likely irreversibly oxidized during switching. Preparing the device in an oxygen and water free environment should drastically increase its cyclability, while also opening up a further window in the 2.7-3.1 m region. By changing to a lithium trifluoromethylsulfonylimide (3M salt) based electrolyte we have constructed devices which could be switched 10,000 times over a period of 6 days without any significant degradation of the active polymer layer (see Figure 5.7) During the above lifetime experiment, the device was switched between and +1 V. By increasing the applied voltage to 1.5 V, the device lifetime drastically decreases, the contrast diminishing by

PAGE 145

130 about 85% after only 400 cycles. This overoxidation is generally triggered by nucleophiles such as H 2 O or OH attacking highly oxidized thiophene rings, and leading to a breakdown of the polymer backbone -conjugation. Figure 5.7 Reflectance as a function of number of deep double potential switches for PProDOT-Me2 based device 5.7 O-H and C-H Removal As described in this chapter, devices with contrast ratios of 55% at 0.6 m in the visible, greater than 80% between 1.3 to 2.2 m in the NIR, and greater than 50% between 3.5 to 5.0 m have been assemblied. Their electrochemical stability to thousands of double deep switches was remarkable, with no significant loss in electrochemical contrast after 10,000 cycles.

PAGE 146

131 However, two strong absorption bands from water (2.8 m) and C-H stretching (3.3.4 m) are seen in all of the spectra (Figure 5.3). They arise from both the polymer and the gel electrolyte, and are relatively constant as the polymer is oxidized and reduced. These absorptions hinder the high reflectance of the device when the PProDOT-Me 2 layer is in the oxidized form, thus drastically reducing the electrochromic contrast in the mid-IR region. Our recent work focuses on the removal of these absorptions due to O-H and C-H vibrational stretches. This would allow for achieving an extremely high contrast throughout the entire 2.5 to 3.5 m area of the spectrum. As the polymer layer is fairly thin (150-200 nm), presumably these absorption peaks in the device are mainly due to the gel electrolyte, therefore our immediate efforts focused toward O-H and C-H stretch removal from the electrolyte. In order to remove water from the device assembly, the gel electrolyte was prepared in dry conditions. Li(CF 3 SO 2 ) 2 N was dried in an vacuum oven for 24 hours at 60C, whereas ACN and PC were freshly distilled over CaH 2 following a procedure described in the experimental section. The gel electrolyte preparation was carried out in a dry box to ensure that the resulting material will not be exposed to ambient humidity. As seen in Figure 5.8, these steps were sufficient for eliminating the absorptions due to O-H stretching mode from the gel electrolyte. Using anhydrous gel electrolyte, assembling and further encapsulating a device in a water-free environment affords electrochromic cells that show reflectance contrasts of 70 % over the entire 2000 to 6000 cm -1 region, as seen in Figure 5.9. The amount of gel electrolyte deposited on the active layer surface has an important effect on the intensity of absorptions seen in the 2.53.5 m region of the spectrum (C-H stretch).

PAGE 147

132 20003000400050000.00.20.40.60.8 TransmittanceWavenumbers, cm-1 Gel prepared in airGel prepared in dry conditions 20003000400050000.00.20.40.60.8 TransmittanceWavenumbers, cm-1 Gel prepared in airGel prepared in dry conditions Figure 5.8 Transmittance spectra of two gel electrolyte batches prepared in air (dashed line) and in dry conditions (solid line) Therefore, minimizing the upper gel layer thickness significantly reduces the C-H stretch absorption and increases the device contrast in the mid-IR. Tuning the gel electrolyte thickness was achieved by placing a thin spacer (25 m in thickness) between the top window and the active polymer layer. The device was then assembled without covering the top polymer surface with gel, but rather pressing the active electrode to the top window to ensure that a controlled amount of gel will pass through the slits and spread onto the active polymer surface. By using this new device assembly method, the C-H stretch appearing at 3000 cm -1 was diminished by about 80 % as compared to the variable reflectance devices reported earlier in this chapter.

PAGE 148

133 200030004000500060000.00.20.40.60.81.0 ReflectanceWavenumbers, cm-10 V+1 V-0.5 V-1 V200030004000500060000.00.20.40.60.81.0 ReflectanceWavenumbers, cm-10 V+1 V-0.5 V-1 V Figure 5.9 Reflectance spectra of a PProDOT-Me 2 based device at four applied voltages Future plans involve further repression of the C-H stretch absorption by using different types of electrolytes. For example, one can use a liquid electrolyte based on LiCF 3 SO 3 LiBF 4 or Li(CF 3 SO 2 ) 2 N, and fluorinated acetone or (CF 3 SO 2 ) 2 O as solvents lacking C-H bonds (C-F vibrational modes lie in the 7.5-9 m region). Another class of electrolytes suitable for devices operating in IR due to lack of absorptions in the region of interest comprises solid inorganic electrolytes such as LiF-AlF 3 films. Further experiments concerning the use of fluorinated polymers are now being discussed.

PAGE 149

CHAPTER 6 ABSORPTION-TRANSMISSIVE WINDOWS 6.1 Introduction There are numerous processes possible for the modulation of visible light. Among them are electrochromic techniques that use the reversible change of color and/or optical density obtained by an electrochemical redox process in which the oxidized and the reduced forms have different colors, indices of refraction, or optical densities. These techniques are readily employed in a multitude of applications such as display panels,175 camouflage materials,176 variable reflectance mirrors,121 and variable transmittance windows.23,165,177-179 For example, Donnelly and Gentex electrochromic mirror systems have been successfully commercialized in the automotive industry and the first electrochromic glazing window for architectural purposes is now going to be introduced onto the market by Pilkington/FLABEG, Germany. Electrochromic devices (ECDs) based on inorganic semiconductors have a long history, and their performance has improved steadily since their creation.166 When viewed in this context, the recent rapid progress made with organic conducting and electroactive polymers in a variety of fields suggests they may find numerous practical applications in the near term.180 These materials have made valuable contributions to the emerging fields of electrochromic devices, as well as organic light emitting diodes,181,182 and photovoltaics.68 In terms of electrochromics, the remarkable 134

PAGE 150

135 advances in their performance can be viewed from several fronts. First, the range of colors now available effectively spans the entire visible spectrum83 and also extends through the near and mid-infrared regions.121 This is due to the ability to synthesize a wide variety of polymers with varied degrees of electron-rich character and conjugation. For example, a fine adjustment of the band gap, and consequently of the color, is possible through modification of the structure of the polymer via monomer functionalization,16 copolymerization,183,184 and the use of blends,185 laminates,120 and composites.70,186 Second, it has been the marked increase in device lifetimes. The key to this is control of the degradation processes within the polymeric materials (by lowering the occurrence of structural defects during polymerization) and the redox system.24,187-189 Third, the polymer based ECDs have achieved relatively fast switching times of a few hundreds of milliseconds for large changes in optical density. This fast switching is attributed to a highly open morphology of electroactive films, which allows for fast dopant ion transport.25 Other beneficial properties of polymers are outstanding coloration efficiencies190 along with their general processability. In this chapter, we show that the necessary control over the color, brightness and environmental stability of an electrochromic window can be achieved through the careful design of anodically and cathodically coloring polymers. 6.2 Absorption/Transmission Device Construction The construction of a transmissive type ECD is depicted in Figure 6.1. It consists of two thin polymer films deposited on transparent indium tin oxide coated glass (ITO), and separated by a viscous gel electrolyte based on LiN(CF 3 SO 2 ) 2 dissolved in an

PAGE 151

136 Gel Electrolyte Transparent Electrodes AnodicallyColoring PolymerCathodicallyColoring Polymer PProDOT-Me2 neutral PProDOT-Me2 doped PProDOP-NPrSdoped PProDOP-NPrS neutral Gel Electrolyte Transparent Electrodes AnodicallyColoring PolymerCathodicallyColoring Polymer Gel Electrolyte Transparent Electrodes AnodicallyColoring PolymerCathodicallyColoring Polymer PProDOT-Me2 neutral PProDOT-Me2 doped PProDOP-NPrSdoped PProDOP-NPrS neutral Figure 6.1 Variable transmittance window schematic. The color representations shown on top and bottom of the scheme are L*a*b* determined color coordinates. acetonitrile/propylene carbonate swollen poly(methyl methacrylate) matrix. The device construction is carried out with one polymer oxidatively doped while the other is neutral, and both films are simultaneously in either their transmissive or absorptive states. As such, the device is observed as bleached or colored. Application of a voltage neutralizes the doped polymer with concurrent oxidation of the complementary polymer, inducing color formation, or bleaching. The ability to match the number of redox sites in each film enhances the contrast of a device, as the extremes of absorption and transmission can be

PAGE 152

137 attained. We define a cathodically coloring polymer as one that passes from an almost transparent state to a highly colored state upon charge neutralization (reduction) of the p-doped form, while an anodically coloring polymer is highly transmissive neutral and absorbs visible light in the oxidatively doped state. 6.3 Cathodically Coloring Polymers An ideal cathodically coloring material has a band gap around 1.8-2.2 eV (550-650 nm) in order for the transition to occur in the visible region where the human eye is highly sensitive. In addition, it should undergo large optical changes when switched between neutral and doped states. Some of the most promising conducting polymers for use in ECDs are based on PEDOT and its derivatives, as they exhibit high electrochromic contrasts, low oxidation potentials and high conductivity, as well as good electrochemical and thermal stability. Some electrochromic polymers developed in our laboratories are represented in Figure 6.2. These materials exhibit colors spanning the entire visible region. Of the PEDOT derivative palette available, PProDOT-Me 2 stands out as the best cathodically coloring polymer for use in electrochromic devices. The decision to use this material as the cathodically coloring layer in transmissive window type devices came from the fact that PProDOT-Me 2 shows the highest contrast of all polymers we have studied in the visible region: about 78% at max (580nm) and a luminance change of 60% measured by colorimetry. This high contrast at 580 nm corresponds to a wavelength where the human eye is highly sensitive, the polymer switching from a highly transmissive light blue in the doped state to a dark blue-purple in the neutral state. The color representations shown in Figure 6.1 are L*a*b* determined color coordinates. They

PAGE 153

138 are calculated conform 1931 Commission Internationale de lEclairage (CIE) by taking into consideration the hue, saturation and brightness of the samples. We have already successfully incorporated PProDOT-Me 2 into a variable reflectance ECD showing a remarkably high contrast throughout the visible, NIR and mid-IR regions of the electromagnetic spectrum. Figure 6.2 Cathodically and anodically coloring polymers (from Thompson83)

PAGE 154

139 6.4 Anodically Coloring Polymers Ideally, an anodically coloring polymer is chosen to have a high band gap (E g ) > 3.0 eV ( to transition onset < 410nm) with all of the absorption lying in the ultraviolet region of the spectrum. In addition to the required complementary optical properties, proper ECD operation demands a high degree of electrochemical reversibility and compatibility. Most of the high gap conjugated polymers synthesized to date (e.g. poly(phenylene vinylene) (PPV), poly(p-phenylene) (PPP), etc.) have high oxidation potentials. In order to prepare an ECD that can undergo a distinct switch from a highly transmissive state to a deeply absorptive state, we have pursued easily electrochemically polymerized monomers that can yield polymers having these desired complementary EC properties to the cathodically coloring PProDOT-Me 2 Our group has designed and synthesized several anodically coloring polymers based on carbazole, biphenyl and other aromatic units linked via electropolymerizable EDOT moieties.189,191 While ECDs incorporating these types of polymers showed outstanding coloration efficiencies, lifetimes and switching times,24,25 the band gaps of the anodically coloring polymers were not sufficiently high for the absorption to be excluded from the visible region, thus providing a coloration to the transmissive state of the devices. There are several ways of controlling the band gap in polymers, including monomer unit planarity, donor-acceptor effects, resonance, interchain effects or bond length alternation. Our approach was to start with an electron rich monomer with low oxidation potential and derivatize it at the redox center. Recently, we reported the synthesis,192 electropolymerization and redox switching properties of a new series of

PAGE 155

140 conducting polymers based on 3,4-alkylenedioxypyrroles.193-195 As pyrroles have somewhat elevated LUMO levels, the band gaps for poly(3,4-ethylenedioxypyrrole) (PEDOP) (2.0 eV) and the propylene bridged analog (PProDOP, 2.2 eV) are higher than their thiophene counterparts (PEDOT and PProDOT) which have band gaps of 1.6-1.7 eV. In order to elevate the band gap of the dioxypyrrole polymers even further, we have prepared a series of N-substituted ProDOPs.192,196 N-substitution is known to induce a torsional angle between heterocyclic rings along the backbone, resulting in a decrease of the -conjugation and therefore an increase in the band gap, as illustrated by Hyperchem calculations on trimers of PProDOP and PProDOP-NPrS (see Figure 6.3). In addition to the advantage of this high band gap, these polymers preserve the electron rich character of the PXDOP family, allowing electrosynthesis and switching under mild conditions. Polymerization is forced through the 2and 5positions as desired giving rise to a material with few structural defects, and consequently better electrochemical cyclability when compared to the parent polypyrrole. Finally, N-alkyl substitution inductively increases the electron density in the monomer, thus exhibiting a lower oxidation potential than the underivatized ProDOP. Table 6.1 summarizes band gaps and monomer oxidation potentials for a series of N-substituted ProDOPs as compared to PProDOT. Increasing the substituent length from hydrogen to methyl induces an increase in band gap from 2.2 to 3.0 eV respectively, with a concurrent decrease in monomer oxidation potential from 0.8 to 0.5 V vs. Ag/Ag + By judicious selection of the N-alkyl substituent, we have prepared an N-propane sulfonated PProDOP (PProDOP-NPrS) where the onset of the transition is located at the boundary of the visible and ultraviolet regions of

PAGE 156

141 the spectrum (E g = 3.2 eV). As such, the electronic absorption is transferred to the UV and the polymer is colorless in the neutral state. PProDOP-NPrS switches from a gray-green state to an almost clear neutral state, as seen in Figure 6.1 bottom. Moreover, the presence of a sulfonate group at the end of the propyl chain offers the possibility of self-doping along with water solubility of the resulting polymer. As a result, PProDOP-NPrS exhibits a fast and regular growth even in the absence of the supporting electrolyte. A)PProDOPB)PProDOP-NPrS A)PProDOPB)PProDOP-NPrS Figure 6.3 Trimer of A) PProDOP and B) PProDOP-NPrS Hyperchem illustrations. Top figures represent the front view of the oligomer chain and bottom figures represent the side view along the chain.

PAGE 157

142 Table 6.1 Band gap and monomer oxidation potential values for a series of propylenedioxy-derivatized pyrroles and thiophenes Polymer PProDOT PProDOP PProDOP-NMe PProDOP-NPrS Band gap 1.8 eV 2.2 eV 3.0 eV 3.2 eV E ox Monomer vs. Ag/Ag + 1.1 V 0.8 V 0.7 eV 0.5 eV 6.5 Matching Two Complementary Coloring Polymers Optical property match. Here, we report our research efforts focused on optimizing the contrast in dual polymer ECDs by matching the anodically and cathodically properties in order to obtain a color neutral, highly transmissive window in one state, that shows low absorption throughout the entire visible region. Upon switching by applying a small bias potential (ca. 1.0V), the window is converted to a highly colored dark state. Figure 6.4 demonstrates this concept by superimposing the UV-visible spectra of the individual polymers in their doped and neutral states. Figure 6.4 A represents the absorptive state of the device where PProDOP-NPrS is in the oxidized form, while PProDOT-Me 2 is in the neutral state. The summation of the two absorption spectra represents the most probable behavior of the colored state of a device based on these two polymers. There are several factors that combine to attain a very saturated color and broadband absorbance window. First, neutral PProDOT-Me 2 has its absorption in the middle of the visible region ( max = 580 nm) where the human eye is the most sensitive. In addition, the vibrational splitting of the HOMO-LUMO transition broadens

PAGE 158

143 40060080010001200140016000.00.20.40.60.81.01.21.41.61.82.02.2 AbsorbanceWavelength (nm)0.00.20.40.60.81.01.21.41.61.82.02.22.4 AbsorbanceColored state Bleached state40060080010001200140016000.00.20.40.60.81.01.21.41.61.82.02.2 AbsorbanceWavelength (nm)0.00.20.40.60.81.01.21.41.61.82.02.22.4 AbsorbanceColored state Bleached state Figure 6.4 The superimposition of the UV-Vis-NIR spectra of the individual polymer films deposited on ITO/glass substrates. A) Polymers absorption spectra where PProDOP-NPrS is in the oxidized form and PProDOT-Me2 is in the neutral form. The summation of the two absorption spectra (dashed line) represents the most probable behavior of the colored state of a device based on these two polymers. B) The sum of the neutral PProDOP-NPrS and doped PProDOT-Me2 spectra provide a highly transmissive device (dashed line). For clarification purposes, the visible region of the spectrum (400-800 nm) is marked with dashed vertical lines.

PAGE 159

144 the absorption peak over a larger area of the visible spectrum. The absorption of the oxidatively doped PProDOP-NPrS begins at 500 nm, and its absorbance increases into the NIR region where the contribution from PProDOT-Me 2 is small. The bleached state of the polymer films and device is represented in Figure 6.4 B. The sum of the neutral PProDOP-NPrS and doped PProDOT-Me 2 spectra provide a high contrast with the colored state throughout the entire visible region, generating a highly transmissive device. Thickness match. In addition to the required complementary optical properties, matching the number of redox sites in each polymer film ensures proper device operation, such as long lifetime and increased contrast, as the extreme redox states can be reached. This goal can be attained by matching polymer layer thicknesses. Figure 6.5 A represents the charge density required to grow a film of a certain thickness for the cathodically coloring PEDOT-F {poly[pentadecafluoro-octanoic acid 2,3 dihydro-thieno(3,4-b)(1,4)dioxin-2-ylmethylester]} and for the anodically coloring PBEDOT-NMeCz. For this study, polymer films were grown potentiostatically on ITO/glass electrodes at several charge densities ranging from 10 mC/cm 2 to 120 mC/cm 2 Film thicknesses can be measured with either a profilometer or an AFM after scratching the film in a small area to expose the ITO substrate. The depth of the scratch can be measured from the cross-section view of the topographic image (see Figure 6.6). For example, if films of 200 nm are desired (Figure 6.5 A, dotted line), then one should deposit 9 mC/cm 2 of PBEDOT-NMeCz and 22 mC/cm 2 of PEDOT-F. The difference in charge densities required to grow these anodically and cathodically coloring polymers suggests a substantial difference in the polymers electrodeposition mechanism and consequently in the

PAGE 160

145 resulting film morphology. As viewed with the profilometer, PEDOT-F film surface is fairly smooth, whereas PBEDOT-NMeCz has a very rough surface indicating that in the latter, the dendritic growth starts in early stages of the polymerization. 020406080100120020040060080010001200140016001800 PBEDOT-NMeCz PEDOT-FThickness (nm)Charge density (mC/cm2)10152025304080120160200240 PProDOT-Me2 on PEDOP PProDOT-Me2 on ITO/glassThickness (nm)Charge Density (mC/cm2)AB020406080100120020040060080010001200140016001800 PBEDOT-NMeCz PEDOT-FThickness (nm)Charge density (mC/cm2)10152025304080120160200240 PProDOT-Me2 on PEDOP PProDOT-Me2 on ITO/glassThickness (nm)Charge Density (mC/cm2)AB Figure 6.5 Thickness as a function of charge density passed during electrodeposition for two complementary electrochromic polymers (A) and PProDOT-Me2 film thickness vs. charge density for two deposition substrates (B)

PAGE 161

146 510m 510m Figure 6.6 AFM topographic image representing a scratch through a polymer film 6.6 High Contrast ECDs Figure 6.7 shows the transmittance spectra and photographs of devices using PProDOT-Me 2 as the cathodically coloring polymer, and PBEDOT-NMeCz (A) and PProDOP-NPrS (B) as the anodically coloring layers with the devices in the two extreme (colored and bleached) states. In this study, all the polymer films are approximatively 200 nm, as measured by profilometry. Both devices switch between the two forms when a bias voltage of .0 to 1.2 V is applied. By applying voltages continuously between +1 and V, the devices change from colored to transmissive and the spectra evolve from one extreme to the other. Comparison of the results shows that using PProDOP-NPrS as the high band gap polymer has several advantages over the carbazole counterpart. The main benefit is the opening of the transmissivity window throughout the entire visible

PAGE 162

147 PProDOT-Me2/PProDOP-NPrS PProDOT-Me2/PBEDOT-NMeCz40050060070080001020304050607080 +1.2 V -1.2 V% TransmittanceWavelength (nm)01020304050607080 +1.2V -1.2V% Transmittance BAPProDOT-Me2/PProDOP-NPrS PProDOT-Me2/PBEDOT-NMeCz40050060070080001020304050607080 +1.2 V -1.2 V% TransmittanceWavelength (nm)01020304050607080 +1.2V -1.2V% Transmittance BA Figure 6.7 Transmittance spectra and photographs of devices using PProDOT-Me2 as the cathodically coloring polymer and PBEDOT-NMeCz (A) and PProDOP-NPrS (B) as the anodically coloring layers with the devices in the two extreme (colored and bleached) states. The dashed vertical line represents the onset of the cut-off of the neutral PBEDOT-NMeCz.

PAGE 163

148 spectrum by moving the transition into the ultraviolet region. The cut-off of the PBEDOT-NMeCz is evident at ca. 500 nm. Another advantage of the PProDOP-NPrS based device (B) is the noticeable increase in the optical contrast as evidenced by an increase in the transmittance change of the device (%T) from 56% to 68% as measured at 580 nm. Furthermore, while doped PBEDOT-NMeCz is blue and its charge carrier absorption is at the same energy as the transition of the low gap PProDOT-Me 2 absorption, PProDOP-NPrS is gray-green in the oxidized state, thus providing some extra blocking of the transmission of the opaque device in the 400-500 nm region. Switching time. One of the most important characteristics of ECDs is the response time needed to perform a switch from transmissive to colored and vice versa. In order to analyze the switching characteristics of these windows, the variation in monochromatic light at the wavelength of maximum contrast was monitored during repeated redox stepping experiments. For a comparison, we monitored the change in transmittance of a single PProDOT-Me 2 film of the same thickness (about 200 nm) as the films used in the devices. As seen in Figure 6.8, both devices switch quite rapidly. The PProDOT-Me 2 film alone can be effectively switched in about 0.5 s to attain 95% of its total transmission change (%T=76%). While device A switches in about 500 ms, device B has a somewhat larger dynamic range with a total % transmission change in 600 ms. However, it is remarkable that by adding the PProDOP-NPrS layer, Device B loses only 10% in overall contrast compared to a single PProDOT-Me 2 film. Colorimetric analysis. Recently, colorimetric analysis has been used to investigate the properties of electrochromic and light-emitting polymers. Both luminance and x-y chromaticity diagrams provide valuable information for understanding changes in

PAGE 164

149 0.00.51.01.50102030405060708090100 % TransmittanceTime (s)PProDOT-Me2FilmPProDOT-Me2/PBEDOT-NMeCzPProDOT-Me2/PProDOP-NPrS CBA0.00.51.01.50102030405060708090100 % TransmittanceTime (s)PProDOT-Me2FilmPProDOT-Me2/PBEDOT-NMeCzPProDOT-Me2/PProDOP-NPrS CBA Figure 6.8 Transmittance as a function of switching time of (A) PProDOT-Me2/ PBEDOT-NMeCz device, (B) PProDOT-Me2/PProDOP-NPrS device and (C) PProDOT-Me2 film. devices color and/or brightness. For example, the potential dependence of the relative luminance offers a different perspective on the transmissivity of a material as it relates the human eye perception of transmittance over the entire visible spectrum as a function of doping on a single curve. Figure 6.9 shows the hue and saturation x-y track for Device B as the applied potential is changed from .5 V to +1.5 V, which corresponds to the oxidative doping track of PProDOT-Me 2 A relatively straight line that spans between a dark blue area of the color space to a highly transmissive (near white point) blue-green color is observed. Therefore, the dominant wavelength of the color is about the same throughout the bleaching process for a certain illuminant; the absorption decreases in

PAGE 165

150 intensity as represented by the decrease in the saturation of the color. The potential dependence of the relative luminance study (Figure 6.10) shows that, in the bleached state, the window exhibits a luminance of 65% through the positive voltage values. A voltage of .5 V is needed to induce a decrease in luminance; dimming of the device luminance continues to increase until a potential of V is attained. The overall luminance change, that is in essence the optical density change perceived by the human eye, is 55%. 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx Figure 6.9 Hue and saturation (x-y track) for the device B at various doping levels

PAGE 166

151 -3.0-2.5-2.0-1.5-1.0-0.50.00.51.01.52.010203040506070 Luminance (%Y)Voltage (V)%Y=55%PProDOT-Me2/PProDOP-NPrS device-3.0-2.5-2.0-1.5-1.0-0.50.00.51.01.52.010203040506070 Luminance (%Y)Voltage (V)%Y=55%PProDOT-Me2/PProDOP-NPrS device Figure 6.10 Luminance PProDOT-Me2/ PProDOP-NPrS device as a function of the applied voltage Lifetime. The stability of the bleached and/or colored states toward multiple redox switches often limits the utility of electrochromic materials in ECD applications. Main reasons for device failure are different electrochemical windows and/or environmental requirements of the complementary materials. If the two EC materials possess different electrochemical windows for operation, then the applied voltage needed for attaining 100% of the optical contrast increases. Therefore, longer lifetimes are expected for devices operating at low voltages since high-applied potentials are detrimental to the electrochromic films, the electrolyte and even to the ITO layer. The oxidation process of the anodic material should coincide with the reduction process of the cathodic EC material in order to properly maintain charge balance within the ECD.

PAGE 167

152 We employed colorimetry by monitoring the luminance change to investigate the long-term stability for Device B to multiple deep switches (Figure 6.11). This study was carried out by continuously stepping the voltage of the device between and +1 V with a 30 s delay at each potential allowing a full color change and hold period. During this time, the luminance was monitored over a period of 7 days. We noticed that the device loses 10% of its contrast during the first 500 cycles. After this conditioning period, the degradation proceeds remarkably slowly and the device loses only 4% of its luminance contrast over a period of 20 000 double potential cycles. As the devices are sealed to ambient exposure, significantly longer lifetimes are realized. 05000100001500020000203040506070 Luminance (%)Cycles EC Contrast 40%05000100001500020000203040506070 Luminance (%)Cycles EC Contrast 40% Figure 6.11 Lifetime of the PProDOT-Me2/PProDOP-NPrS device to multiple redox switches. The upper trace (squares) shows the decrease in the luminance of the device in the bleached state with the number of cycles performed, while the lower trace (circles) is the fading of the dark state during continuous switching.

PAGE 168

153 6.7 Broadband Absorption Devices: Laminate Alternative Tuning the function of the device over the entire visible spectrum can be achieved by using bilayers of polymers having different max and comparable oxidation potentials (Figure 6.12). By overlapping the conducting polymer layers, not only is more of the visible spectrum covered, but the absorbance of the device in the colored state is also enhanced. In this study, we use PEDOP ( max at 510 nm) deposited on ITO/glass and PProDOT-Me 2 ( max at 580 nm) electropolymerized onto the already existing PEDOP layer. P3: anodically coloring polymerGel Electrolyte Transparent Electrodes athodicallycoloring polymerP2: cathodicallycoloring polymer max(max(P2) E1/2(1/2(P2)[[P2]=[P3] P3: anodically coloring polymerGel Electrolyte Transparent Electrodes athodicallycoloring polymerP2: cathodicallycoloring polymer max(max(P2) E1/2(1/2(P2)[[P2]=[P3] P1: c P1: c P1)>P1)P1)
PAGE 169

154 In general, PEDOP films are smooth and fairly easy to deposit on ITO coated electrodes without requiring several deposition/buffing cycles. In contrast, high quality PProDOT-Me 2 films are difficult to synthesize directly on ITO supports, the polymer spotting the ITO surface for the first three to four deposition/buffing cycles and only repeated electropolymerizations lead to a continuous, homogeneous PProDOT-Me 2 film. Interestingly, when deposited on PEDOP, PProDOT-Me 2 deposition occurs smoothly allowing the synthesis of high quality films from the first trial. Consequently, besides increasing the contrast and the function of a device over the visible region, PEDOP acts as a better support for the deposition of subsequent polymers. Figure 6.5 B illustrates the difference in PProDOT-Me 2 film thickness as the electropolymerization support changes. Films deposited on the PEDOP coated electrode are thinner than the films deposited directly on ITO, at the same charge density value, suggesting that the PProDOT-Me 2 electropolymerized on PEDOP has a more compact and homogeneous morphology. PEDOP/PProDOT-Me 2 bilayer absorption spectra at various applied potentials are shown in Figure 6.13. The two polymer layers are of comparable thicknesses of about 150 nm. At +1 V vs. Ag wire, both polymers are in their oxidized form and the electrode appears highly transmissive. As the potential is decreased to V, the peak corresponding to the transition of the PProDOT-Me 2 layer appears at 580 nm. A further decrease in the potential to V is required in order to observe the transition corresponding to the PEDOP film. When both polymers reached their neutral state, a broadening as well as a large increase in the intensity of the bilayer absorption peak is observed. The need for such a low reduction potential could be explained by the charge trapping mechanism outlined in Figure 6.14. When the polymers are oxidized, and

PAGE 170

155 4006008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) PEDOP max=510 nmPProDOT-Me2max=580 nm4006008000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm) PEDOP max=510 nmPProDOT-Me2max=580 nm Figure 6.13 PEDOP/PProDOT-Me 2 bilayer electrode absorption spectra at applied potentials ranging from V to +2 V vs. Ag wire therefore conducting, they bear negative ions (ClO 4 ) within the film to counterbalance the positive charges on the backbone. At V, the PProDOT-Me 2 layer, which is in direct contact with the electrolyte solution, undopes first becomes neutral and consequently switches to a non-ionic form, rendering difficult for the PEDOP polymer to lose its dopant anions to the solution. Anions difficulty to pass through an non-ionic medium is overcome at V. Further quartz crystal microbalance studies are necessary to validate this theory.

PAGE 171

156 ITOPEDOP +PProDOT-Me2+ +1V ITOPEDOP +PProDOT-Me20 -1V PProDOT-Me20-2VPEDOP 0ITO ITOPEDOP +PProDOT-Me2+ +1V ITOPEDOP +PProDOT-Me20 -1V PProDOT-Me20-2VPEDOP 0ITO Figure 6.14 Charge trapping mechanism in PEDOP/PProDOT-Me 2 bilayer electrode We constructed a laminate device based on the above described bilayer electrode and PBEDOT-NMeCz as the complementary high band gap polymer. Figure 6.15 represents the transmittance spectrum of such a device, exhibiting a contrast of about 20% when switched from +2V to V. Further construction of devices having newly discovered PProDOP-NPrS as the high band gap polymer is being discussed. Figure 6.16 represents the colorimetric data of a laminate device based on PEDOP/PProDOT-Me 2 as cathodically coloring polymers and PBEDOT-NMeCz as the anodically coloring layer. We noticed that the device has 25-30 % luminance change when switched between the bleached and colored state. The right top corner inset represents the x-y coordinates of the CIE color space, and it shows a change from a dark

PAGE 172

157 40050060070080004812162024 -2 V +2 V +3 V +4 V% TransmittanceWavelength (nm) Figure 6.15 Transmittance spectra of the PEDOP/PProDOT-Me 2 /PBEDOT-NMeCz laminate device at four applied voltages blue to a greenish transparent color as the applied voltage increases. The lifetime study (shown in the lower left inset) was carried out by continuously switching the device between the two extreme states every 30 s and taking a luminance measurement every several redox cycles. The upper trace represents the decrease in the luminance of the device with the number of cycles performed, while the lower trace is the lightening of the dark state during continuous switching. We noticed that the device loses most of its contrast (66%) before reaching 1,000 cycles. This short lifetime compared to other devices presented in Section 6.6 is most likely due to the fact that the charge trapped in the film during the reduction process does not allow for the polymers to reach their

PAGE 173

158 extreme redox states. Moreover, the high operational voltage of 3 V required to switch this device leads to active layer overoxidation and thus to fast degradation of the electrochromic window. -3-2-1012345615202530354045 Luminance (%)E(V)020004000600080001000020242832364044485256 -3 V +3 VLuminance (%)Cycles 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx-3-2-1012345615202530354045 Luminance (%)E(V)020004000600080001000020242832364044485256 -3 V +3 VLuminance (%)Cycles 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx Figure 6.16 Luminance analysis of the PEDOP/PProDOT-Me 2 /PBEDOT-NMeCz laminate device. The lower left inset represents the luminance change as a function of the number of switching cycles. The upper right inset represents the hue and saturation of the laminate device at several doping levels

PAGE 174

159 6.8 Broadband Absorption Devices: PEDOT-F Alternative PEDOT-F properties. The optical properties of PEDOT-F are similar to those of unsubstituted PEDOT with regard to max (608 nm), band gap and broadness of the absorption peak in the neutral state (Figure 6.17). This type of broad optical response is especially useful in variable transmission devices where a high contrast between the bleached and dark states, at all visible region wavelengths, is desired. 4006008001000120014000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm)max=608 nm -0.4V0.6V SOOOC7F15O4006008001000120014000.00.20.40.60.81.01.21.41.6 AbsorbanceWavelength (nm)max=608 nm -0.4V0.6V SOOOC7F15O Figure 6.17 Absorption spectra of PEDOT-F in 0.1 M TBAP/ACN at applied potentials ranging from .4 V to +0.6 V vs. Ag/Ag +

PAGE 175

160 In addition, PEDOT-F possesses a high coloration efficiency (CE), which is a key parameter for comparison between electrochromic materials. This parameter is obtained from the ratio of the change in optical density during a redox step and the charge injected as a function of the electrode area. Conducting polymers from the PXDOT family tend to exhibit high CE values on the order of 100-300 cm 2 /C.197 The CE for PEDOT-F reaches values of 586 cm 2 /C, measured at 95% of the full transmittance change. This value is at least 2 times higher than that of PEDOT, making PEDOT-F highly suitable in electrochromic device type applications. Even though the transmittance change is similar to that of PEDOT, the charge required to p-dope is much smaller. There are several factors that could contribute to this small charge value. One of them could be the very porous morphology of the films, allowing for facile diffusion of counterions during doping and dedoping processes. Another possible factor is a low doping level required to attain a very high optical contrast. The doping level corresponding to 95% transmittance change was calculated as a ratio of the charge passed during p-doping and the charge used during polymerization, by assuming 100 % deposition efficiency. We obtained a dopant concentration of about 0.25, which indicates one cation every four rings. This value is common for polymers from the PT family that bear one dopant per two to five rings.11 Water droplet-polymer contact angle measurements performed on thin (ca. 200 nm), ClO 4 doped films of PEDOT and PEDOT-F revealed striking differences in the hydrophobicity of the two materials. While the PEDOT surface is wettable, meaning that the contact angle is less than 30, the PEDOT-F surface is highly hydrophobic, exhibiting a contact angle of 110.

PAGE 176

161 PEDOT-F/PBEDOT-NMeCz device. PEDOT-Fs broad absorbance character, in combination with the increased hydrophobicity and expected enhanced solubility, makes it a viable candidate for electrochromic device applications. Therefore, we constructed transmissive windows based on PEDOT-F as the cathodically coloring layer and PBEDOT-NMeCz as the anodically coloring layer. Figure 6.18.A represents the transmittance of the device assembled using 150 nm thick polymer films, as determined by profilometry. The device exhibits a contrast of 60 % at 590 nm when a bias voltage of 1.2 V is applied. An important observation is that the device exhibits a higher contrast than each polymer alone, owing to the overlap of the two electroactive layers dark states. One of the most important characteristics of ECDs is the response time needed to perform a switch from transmissive to opaque and vice versa. In order to analyze the switching characteristics of these windows, the change in transmittance at 608 nm was monitored during repeated redox switching experiments. The device attains 95% of the total transmission change in about 300 ms, as shown in Figure 6.18.B. Both luminance and x-y chromaticity diagrams provide valuable information for understanding changes in the devices color and/or brightness. The potential dependence on the relative luminance shown in Figure 6.19 shows that, in the dark state, the window exhibits a relative luminance of 32 %. The application of increasingly anodic potentials induces an increase in the relative luminance up to 92%, resulting in a highly transmissive film with a %Y of 60%. The residual yellow color corresponds to the neutral state of the PBEDOT-NMeCz layer. The combination of the properties of PEDOT-F and PBEDOT-NMeCz yields a device that can reversibly switch between an opaque state and a highly transmissive state. Figure 6.19.B shows the hue and saturation

PAGE 177

162 (x-y track) for the device as the applied potential is changed from .2 V to +1.2 V. A line that spans between a dark blue area of the color space to a highly transmissive yellow color is observed. This, along with great luminance and transmittance changes as well as fast switching times place devices based on PEDOT-F at the forefront of organic electrochromic windows. 4005006007008009001020304050607080 % TWavelength (nm) ColoredBleached 0.00.20.40.60.81.020304050607080 % TTime (minutes)AB4005006007008009001020304050607080 % TWavelength (nm) ColoredBleached 0.00.20.40.60.81.020304050607080 % TTime (minutes)AB Figure 6.18 Spectroelectrochemical analysis of PEDOT-F/PBEDOT-NMeCz device. A) Transmittance spectra of PEDOT-F/PBEDOT-NMeCz device in the bleached and dark states; B) Switching time of PEDOT-F/PBEDOT-NMeCz device monitored at 608 nm.

PAGE 178

163 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx-1.5-1.0-0.50.00.51.01.50.30.40.50.60.70.80.91.0 LuminanceVoltage (V) BA 0.00.10.20.30.40.50.60.70.80.00.10.20.30.40.50.60.70.80.9 yx-1.5-1.0-0.50.00.51.01.50.30.40.50.60.70.80.91.0 LuminanceVoltage (V) BA Figure 6.19 Colorimetric analysis of PEDOT-F/PBEDOT-NMeCz device A) Luminance analysis of PEDOT-F/PBEDOT-NMeCz device and photographs with the device in bleached and dark states; B) Hue and saturation of PEDOT-F/PBEDOT-NMeCz device as a function of applied voltage

PAGE 179

164 6.9 Conclusions The dual polymer ECDs reported herein possess the ability to operate at low applied voltages (.0V) with both films being compatible in the same electrochemical environment. This greatly increases their lifetime to 86 % retention of their initial color (96% retention after break-in) after 20,000 cycles. Most importantly, the newly discovered PProDOP-NPrS has the ability to switch between a colorless neutral state to a doped gray-green state, possessing the rare property of being a truly anodically coloring polymer with easily accessible redox switching potentials. Furthermore, doped PProDOP-NPrS broadens the absorption peak of the dark state ECD in both the 400-500 nm and 700-800 nm regions of the visible spectrum, where the contributions from the transition of the PProDOT-Me 2 film are small. The devices based on PProDOP-NPrS and PProDOT-Me 2 complementary polymers exhibit an optical contrast of up to 70% at max as well as an overall luminance change of 53%. They switch from a transparent state to a very dark, almost opaque state in less than 1 second, potentially making them useful polymer displays, such as advertising. The devices based on PEDOT-F and PBEDOT-NMeCz complementary polymers exhibit a 60% optical contrast at max and 60 % overall luminance change. Attaining the same value for the change in luminance and transmittance is unusual, as in most cases the transmittance at max is much higher than the overall luminance value. This implies that the device exhibits a broadband absorption in the dark state, with emphasis in the area where the eye is most sensitive, thus introducing a new dimension in the electrochromic device construction. Another way of broadening the absorption of the device over the entire visible spectrum was achieved by using bilayers of polymers having different max and

PAGE 180

165 comparable oxidation potentials. By overlapping the conducting polymer layers, not only is more of the visible spectrum covered, but the absorbance of the device in the colored state is also enhanced. This performance was realized by building a laminated window with PEDOP/PProDOT-Me 2 as cathodically coloring layers and PBEDOT-NMeCz as the anodically coloring layer. All the above characteristics show that the use of carefully designed complementary polymers is a promising route for achieving the necessary control over the color, brightness, and redox stability of an electrochromic window.

PAGE 181

LIST OF REFERENCES (1) Chiang, C. K.; Fincher, C. R.; Park, Y. W.; Heeger, A. J.; Shirakawa, H.; Louis, E. J.; Gau, S. C.; Macdiarmid, A. G. Physical Review Letters 1977, 39, 1098-1101. (2) Shirakawa, H.; Louis, E. J.; Macdiarmid, A. G.; Chiang, C. K.; Heeger, A. J. Journal of the Chemical Society-Chemical Communications 1977, 578-580. (3) Letheby, H. Journal of the Chemical Society 1862, 15, 161. (4) Noelting, E. Scientific and Industrial History of Aniline Black; J. Matheson: New York, 1989. (5) Natta, G.; Mazzanti, G.; Corradini, P. Atti Accad. Lincei Cl. Sci. Fis. Mat. Nat. Rend. 1958, 25, 3. (6) Burt, F. B. Journal of the Chemical Society 1910, 68, 105. (7) MacDiarmid, A. G. Journal of the American Chemical Society 1976, 98, 3884. (8) Diaz, A. F.; Crowley, J.; Bargon, J.; Gardini, G. P.; Torrance, J. B. Journal of Electroanalytical Chemistry 1981, 121, 355-361. (9) Tourillon, G.; Garnier, F. Journal of Electroanalytical Chemistry 1982, 135, 173-178. (10) MacDiarmid, A. G.; Chiang, J. C.; Huang, W. S.; Humphrey, B. D.; Somasiri, N. L. D. Molecular Crystals and Liquid Crystals 1985, 125, 309-318. (11) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Metal-Insulator Transition in Conducting Polymers; 3rd ed.; Marcel Dekker: New York, 1998. (12) Roncali, J. Chemical Reviews 1992, 92, 711-738. (13) Roncali, J. Chemical Reviews 1997, 97, 173-205. (14) Dietrich, M.; Heinze, J.; Heywang, G.; Jonas, F. Journal of Electroanalytical Chemistry 1994, 369, 87-92. (15) Heywang, G.; Jonas, F. Advanced Materials 1992, 4, 116-118. (16) Groenendaal, B. L.; Jonas, F.; Freitag, D.; Pielartzik, H.; Reynolds, J. R. Advanced Materials 2000, 12, 481-494. 166

PAGE 182

167 (17) Welsh, D. M.; Kumar, A.; Meijer, E. W.; Reynolds, J. R. Advanced Materials 1999, 11, 1379-1382. (18) Schottland, P.; Stephan, O.; Le Gall, P. Y.; Chevrot, C. Journal De Chimie Physique Et De Physico-Chimie Biologique 1998, 95, 1258-1261. (19) Kumar, A.; Welsh, D. M.; Morvant, M. C.; Piroux, F.; Abboud, K. A.; Reynolds, J. R. Chemistry of Materials 1998, 10, 896-902. (20) Stephan, O.; Schottland, P.; Le Gall, P. Y.; Chevrot, C.; Mariet, C.; Carrier, M. Journal of Electroanalytical Chemistry 1998, 443, 217-226. (21) Cutler, C. A.; Bouguettaya, M.; Reynolds, J. R. Advanced Materials 2002, 14, 684-688. (22) Jonas, F.; Lerch, K. Kunststoffe-Plast Europe 1997, 87, 1401-1402. (23) Schwendeman, I.; Hickman, R.; Zong, K.; Welsh, D. M.; Schottland, P.; Sonmez, G.; Reynolds, J. R. Chemistry of Materials, accepted for publication. (24) Sapp, S. A.; Sotzing, G. A.; Reddinger, J. L.; Reynolds, J. R. Advanced Materials 1996, 8, 808. (25) Sapp, S. A.; Sotzing, G. A.; Reynolds, J. R. Chemistry of Materials 1998, 10, 2101-2108. (26) Irvin, D. J.; DuBois, C. J.; Reynolds, J. R. Chemical Communications 1999, 2121-2122. (27) Slater, J. C. Physical Reviews 1934, 45, 794. (28) Edwards, P. P.; Sienko, M. J. Accounts of Chemical Research 1982, 15, 87-93. (29) Epstein, A. J.; Miller, J. S. Scientific American 1979, 241, 52-61. (30) Chien, J. C. W. Polyacetylene: Chemistry, Physics, and Material Science; Academic Press, Inc.: Orlando, 1984. (31) Heeger, A. J. Angewandte Chemie-International Edition 2001, 40, 2591-2611. (32) Peierls, R. E. Quantum Theory of Solids; Clarendon Press: Oxford, 1955. (33) Salzner, U.; Lagowski, J. B.; Pickup, P. G.; Poirier, R. A. Synthetic Metals 1998, 96, 177-189. (34) Kaiser, A. B. Advanced Materials 2001, 13, 927. (35) Epstein, A. J.; Joo, J.; Kohlman, R. S.; Du, G.; Macdiarmid, A. G.; Oh, E. J.; Min, Y.; Tsukamoto, J.; Kaneko, H.; Pouget, J. P. Synthetic Metals 1994, 65, 149-157.

PAGE 183

168 (36) Giurgiu, I.; Zong, K.; Reynolds, J. R.; Lee, W. P.; Brenneman, K. R.; Saprigin, A. V.; Epstein, A. J.; Hwang, J.; Tanner, D. B. Synthetic Metals 2001, 119, 405-406. (37) Aleshin, A.; Kiebooms, R.; Menon, R.; Heeger, A. J. Synthetic Metals 1997, 90, 61-68. (38) Aleshin, A.; Kiebooms, R.; Menon, R.; Wudl, F.; Heeger, A. J. Physical Review B 1997, 56, 3659-3663. (39) Aleshin, A. N.; Kiebooms, R.; Yu, H.; Levin, M.; Shlimak, I. Synthetic Metals 1998, 94, 157-159. (40) Aleshin, A. N.; Williams, S. R.; Heeger, A. J. Synthetic Metals 1998, 94, 173-177. (41) Aleshin, A. N.; Kiebooms, R.; Heeger, A. J. Synthetic Metals 1999, 101, 369-370. (42) Asavapiriyanont, S.; Chandler, G. K.; Gunawardena, G. A.; Pletcher, D. Journal of Electroanalytical Chemistry 1984, 177, 229-244. (43) Genies, E. M.; Bidan, G.; Diaz, A. F. Journal of Electroanalytical Chemistry 1983, 149, 101-113. (44) Waltman, R. J.; Bargon, J. Tetrahedron 1984, 40, 3963-3970. (45) Yassar, A.; Roncali, J.; Garnier, F. Macromolecules 1989, 22, 804-809. (46) Roncali, J.; Yassar, A.; Garnier, F. Journal of the Chemical Society-Chemical Communications 1988, 581-582. (47) Roncali, J.; Yassar, A.; Garnier, F. Journal De Chimie Physique Et De Physico-Chimie Biologique 1989, 86, 85-92. (48) Roncali, J.; Yassar, A.; Garnier, F. Synthetic Metals 1989, 28, C275-C280. (49) Inzelt, G.; Pineri, M.; Schultze, J. W.; Vorotyntsev, M. A. Electrochimica Acta 2000, 45, 2403-2421. (50) Salaneck, W. R.; Lundstrom, I.; Huang, W. S.; Macdiarmid, A. G. Synthetic Metals 1986, 13, 291-297. (51) Heeger, A. J. Synthetic Metals 2001, 125, 23-42. (52) Nigrey, P. J.; Macdiarmid, A. G.; Heeger, A. J. Journal of the Chemical Society-Chemical Communications 1979, 594-595. (53) Heeger, A. J.; Kivelson, S.; Schrieffer, J. R.; Su, W. P. Reviews of Modern Physics 1988, 60, 781-850.

PAGE 184

169 (54) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539-541. (55) Hupe, J.; Wolf, G.-D.; Jonas, F. Galvanotechnik 1995, 86, 3404. (56) Bayer AG Product information AI 1995, 4060. (57) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angewandte Chemie-International Edition 1998, 37, 402-428. (58) Janda, P.; Weber, J. Journal of Electroanalytical Chemistry 1991, 300, 119-127. (59) Schmidt, V. M.; Tegtmeyer, D.; Heitbaum, J. Journal of Electroanalytical Chemistry 1995, 385, 149-155. (60) Schopf, G.; Kossmehl, G. In Polythiophenes Electrically Conductive Polymers, 1997; Vol. 129. (61) Gustafssoncarlberg, J. C.; Inganas, O.; Andersson, M. R.; Booth, C.; Azens, A.; Granqvist, C. G. Electrochimica Acta 1995, 40, 2233-2235. (62) Naegele, D.; Bittihn, R. Solid State Ionics 1988, 28, 983-989. (63) Otero, T. F.; Rodriguez, J.; Angulo, E.; Santamaria, C. Synthetic Metals 1993, 57, 3713-3717. (64) Deberry, D. W. Journal of the Electrochemical Society 1985, 132, 1022-1026. (65) Arbizzani, C.; Mastragostino, M.; Meneghello, L. Electrochimica Acta 1996, 41, 21-26. (66) Sariciftci, N. S.; Heeger, A. J. International Journal of Modern Physics B 1994, 8, 237-274. (67) Roman, L. S.; Andersson, M. R.; Yohannes, T.; Inganas, O. Advanced Materials 1997, 9, 1164. (68) Granstrom, M.; Petritsch, K.; Arias, A. C.; Lux, A.; Andersson, M. R.; Friend, R. H. Nature 1998, 395, 257-260. (69) Harrison, B. S.; Foley, T. J.; Bouguettaya, M.; Boncella, J. M.; Reynolds, J. R.; Schanze, K. S.; Shim, J.; Holloway, P. H.; Padmanaban, G.; Ramakrishnan, S. Applied Physics Letters 2001, 79, 3770-3772. (70) Inganas, O.; Johansson, T.; Ghosh, S. Electrochimica Acta 2001, 46, 2031-2034. (71) Heuer, H. W.; Wehrmann, R.; Kirchmeyer, S. Advanced Functional Materials 2002, 12, 89-94.

PAGE 185

170 (72) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chemical Reviews 2000, 100, 2537-2574. (73) ElRahman, H. A. A.; Schultze, J. W. Journal of Electroanalytical Chemistry 1996, 416, 67-74. (74) Macdiarmid, A. G.; Kaneto, K.; Saito, H.; Min, Y. Abstracts of Papers of the American Chemical Society 1994, 208, 387-PMSE. (75) van der Pauw, L. J. Philips Technical Review 1958, 20, 220. (76) Blythe, A. R. Polymer Testing 1984, 4, 195-209. (77) Smits, F. M. Bell System Technical Journal 1958, 37, 711-718. (78) Valdes, L. B. Proceedings of the Institute of Radio Engineers 1954, 42, 420-427. (79) Uhlir, A. Bell System Technical Journal 1955, 34, 105-128. (80) Tourillon, G.; Garnier, F. Journal of Physical Chemistry 1983, 87, 2289-2292. (81) Doblhofer, K.; Rajeshwar, K. In Handbook of Conducting Polymers; 3rd ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998. (82) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; Wiley: New York, 1980. (83) Thompson, B. C.; Schottland, P.; Zong, K. W.; Reynolds, J. R. Chemistry of Materials 2000, 12, 1563-1571. (84) Kuehni, R. G. Color: an Introduction to Practice and Principles; Wiley: New York, 1996. (85) Linford, R. G. Electrochemical Science and Technology of Polymers; Elsevier: England, 1990. (86) Bruce, P. G. Solid State Electrochemistry; Cambridge University Press: Cambridge, 1995. (87) Shirakawa, H. Angewandte Chemie-International Edition 2001, 40, 2575-2580. (88) MacDiarmid, A. G. Angewandte Chemie-International Edition 2001, 40, 2581-2590. (89) Davis, E. A.; Mott, N. F. Philosophical Magazine 1970, 22, 903. (90) Cohen, M. H.; Fritzsch.H; Ovshinsk.Sr Physical Review Letters 1969, 22, 1065. (91) Anderson, P. W. Physical Review 1958, 109, 1492-1505.

PAGE 186

171 (92) Nagels, P. In Linear and Nonlinear Electron Transport in Solides; Devresse, J. T., Van Doren, V. E., Eds.; Plenum: New York, 1976. (93) Prigodin, V. N.; Epstein, A. J. Synthetic Metals 2001, 125, 43-53. (94) Mott, N. F. Philosophical Magazine 1970, 22, 7. (95) Austin, I. G.; Mott, N. F. Advances in Physics 1969, 18, 41. (96) Kohlman, R. S.; Epstein, A. J. In Handbook of Conducting Polymers; 3rd ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998. (97) Sheng, P. Physical Review B 1980, 21, 2180-2195. (98) Percec, V.; Ahn, C. H.; Cho, W. D.; Jamieson, A. M.; Kim, J.; Leman, T.; Schmidt, M.; Gerle, M.; Moller, M.; Prokhorova, S. A.; Sheiko, S. S.; Cheng, S. Z. D.; Zhang, A.; Ungar, G.; Yeardley, D. J. P. Journal of the American Chemical Society 1998, 120, 8619-8631. (99) Park, Y. W.; Heeger, A. J.; Druy, M. A.; Macdiarmid, A. G. Journal of Chemical Physics 1980, 73, 946-957. (100) Travers, J. P.; Sixou, B.; Berner, D.; Wolter, A.; Rannou, P.; Beau, B.; Pepin-Donat, B.; Barthet, C.; Guglielmi, M.; Mermilliod, N.; Gilles, B.; Djurado, D.; Attias, A. J.; Vautrin, M. Synthetic Metals 1999, 101, 359-362. (101) Stauffer, A.; Aharoni, A. Introduction to The Percolation Theory; Taylor & Francis: London, 1991. (102) Kohlman, R. S.; Zibold, A.; Tanner, D. B.; Ihas, G. G.; Ishiguro, T.; Min, Y. G.; MacDiarmid, A. G.; Epstein, A. J. Physical Review Letters 1997, 78, 3915-3918. (103) Johansson, A.; Stafstrom, S. Physical Review B 2002, 6504, art. no.-045207. (104) Bredas, J. L.; Beljonne, D.; Cornil, J.; Calbert, J. P.; Shuai, Z.; Silbey, R. Synthetic Metals 2001, 125, 107-116. (105) Hwang, J.; Tanner, D. B.; Schwendeman, I.; Reynolds, J. R. Physical Reviews B, submitted for publication. (106) Ioffe, A. F.; Regel, A. R. Prog. Semicond. 1960, 4, 237. (107) Mott, N. F.; Davis, E. A. Philosophical Magazine 1968, 17, 1269. (108) Mott, N. F. Metal-Insulator Transition; Taylor & Francis: London, 1990.

PAGE 187

172 (109) Reghu, M.; Yoon, C. O.; Moses, D.; Heeger, A. J. In Handbook of Conducting Polymers; 3rd ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998. (110) Kiebooms, R.; Aleshin, A.; Hutchison, K.; Wudl, F. Journal of Physical Chemistry B 1997, 101, 11037-11039. (111) Kiebooms, R.; Aleshin, A.; Hutchison, K.; Wudl, F.; Heeger, A. Synthetic Metals 1999, 101, 436-437. (112) Saprigin, A. V.; Lee, V.-P.; Epstein, A. J.; Hwang, J.; Tanner, D. B.; Schwendeman, I.; Zong, K.; Reynolds, J. R. to be submitted (113) Schon, J. H.; Dodabalapur, A.; Bao, Z.; Kloc, C.; Schenker, O.; Batlogg, B. Nature 2001, 410, 189-192. (114) Mott, N. F.; Davis, E. A. Electronic Processes in Noncrystalline Materials; Oxford University Press: Oxford, 1979. (115) Aasmundtveit, K. E.; Samuelsen, E. J.; Pettersson, L. A. A.; Inganas, O.; Johansson, T.; Feidenhans, R. Synthetic Metals 1999, 101, 561-564. (116) Ruiz, J. P.; Dharia, J. R.; Reynolds, J. R.; Buckley, L. J. Macromolecules 1992, 25, 849-860. (117) Groenendaal, L.; Zotti, G.; Jonas, F. Synthetic Metals 2001, 118, 105-109. (118) Shiraishi, K.; Kanbara, T.; Yamamoto, T.; Groenendaal, L. Polymer 2001, 42, 7229-7232. (119) Pei, J.; Yu, W. L.; Huang, W.; Heeger, A. J. Macromolecules 2000, 33, 2462-2471. (120) Boehme, J. L.; Mudigonda, D. S. K.; Ferraris, J. P. Chemistry of Materials 2001, 13, 4469-4472. (121) Schwendeman, I.; Hwang, J.; Welsh, D. M.; Tanner, D. B.; Reynolds, J. R. Advanced Materials 2001, 13, 634-637. (122) Patil, A. O.; Heeger, A. J.; Wudl, F. Chemical Reviews 1988, 88, 183-200. (123) Bredas, J. L.; Cornil, J.; Beljonne, D. In Handbook of Conducting Polymers; 3rd ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998. (124) McCullough, R. D.; Ewbank, P. C. In Handbook of Conducting Polymers; 3rd ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998.

PAGE 188

173 (125) Garreau, S.; Louarn, G.; Buisson, J. P.; Froyer, G.; Lefrant, S. Macromolecules 1999, 32, 6807-6812. (126) Garreau, S.; Louarn, G.; Lefrant, S.; Buisson, J. P.; Froyer, G. Synthetic Metals 1999, 101, 312-313. (127) Garreau, S.; Duvail, J. L.; Louarn, G. Synthetic Metals 2001, 125, 325-329. (128) Garreau, S.; Froyer, G.; Louarn, G. Synthetic Metals 2001, 119, 323-324. (129) Garreau, S.; Louarn, G.; Froyer, G.; Lapkowski, M.; Chauvet, O. Electrochimica Acta 2001, 46, 1207-1214. (130) Kvarnstrom, C.; Neugebauer, H.; Blomquist, S.; Ahonen, H. J.; Kankare, J.; Ivaska, A.; Sariciftci, N. S. Synthetic Metals 1999, 101, 66-66. (131) Kvarnstrom, C.; Neugebauer, H.; Blomquist, S.; Ahonen, H. J.; Kankare, J.; Ivaska, A. Electrochimica Acta 1999, 44, 2739-2750. (132) Kvarnstrom, C.; Neugebauer, H.; Ivaska, A.; Sariciftci, N. S. Journal of Molecular Structure 2000, 521, 271-277. (133) Liau, Y. H.; Scherer, N. F.; Rhodes, K. Journal of Physical Chemistry B 2001, 105, 3282-3288. (134) Kugler, T.; Johansson, A.; Dalsegg, I.; Gelius, U.; Salaneck, W. R. Synthetic Metals 1997, 91, 143-146. (135) Scott, J. C.; Kaufman, J. H.; Brock, P. J.; DiPietro, R.; Salem, J.; Goitia, J. A. Journal of Applied Physics 1996, 79, 2745-2751. (136) Schlatmann, A. R.; Floet, D. W.; Hilberer, A.; Garten, F.; Smulders, P. J. M.; Klapwijk, T. M.; Hadziioannou, G. Applied Physics Letters 1996, 69, 1764-1766. (137) Sapurina, I.; Riede, A.; Stejskal, J. Synthetic Metals 2001, 123, 503-507. (138) Barisci, J. N.; Stella, R.; Spinks, G. M.; Wallace, G. G. Electrochimica Acta 2000, 46, 519-531. (139) Apperloo, J. J.; van Haare, J.; Janssen, R. A. J. Synthetic Metals 1999, 101, 417-420. (140) Fesser, K.; Bishop, A. R.; Campbell, D. K. Physical Review B 1983, 27, 4804-4825. (141) Zerbi, G.; Castiglioni, C.; Del Zoppo, M. In Electronic Materials: The Oligomer Approach; Mullen, K., Wegner, G., Eds.; Wiley-VCH: Weinheim, 1998.

PAGE 189

174 (142) Etemad, S.; Pron, A.; Heeger, A. J.; Macdiarmid, A. G.; Mele, E. J.; Rice, M. J. Physical Review B 1981, 23, 5137-5141. (143) Fincher, C. R.; Ozaki, M.; Tanaka, M.; Peebles, D.; Lauchlan, L.; Heeger, A. J.; Macdiarmid, A. G. Physical Review B 1979, 20, 1589-1602. (144) Fox, M. Optical Properties of Solids; Oxford University Press: New York, 2001. (145) Welsh, D. M., Ph. D. Dissertation, University of Florida, 2001. (146) Cornil, J.; Beljonne, D.; Heller, C. M.; Campbell, I. H.; Laurich, B. K.; Smith, D. L.; Bradley, D. D. C.; Mullen, K.; Bredas, J. L. Chemical Physics Letters 1997, 278, 139-145. (147) Chen, T. A.; Wu, X. M.; Rieke, R. D. Journal of the American Chemical Society 1995, 117, 233-244. (148) Sakurai, K.; Tachibana, H.; Shiga, N.; Terakura, C.; Matsumoto, M.; Tokura, Y. Physical Review B 1997, 56, 9552-9556. (149) DiCesare, N.; Belletete, M.; Marrano, C.; Leclerc, M.; Durocher, G. Journal of Physical Chemistry a 1999, 103, 795-802. (150) DiCesare, N.; Belletete, M.; Leclerc, M.; Durocher, G. Journal of Physical Chemistry A 1999, 103, 803-811. (151) McCullough, R. D.; Tristramnagle, S.; Williams, S. P.; Lowe, R. D.; Jayaraman, M. Journal of the American Chemical Society 1993, 115, 4910-4911. (152) McCullough, R. D.; Lowe, R. D.; Jayaraman, M.; Anderson, D. L. Journal of Organic Chemistry 1993, 58, 904-912. (153) Macdiarmid, A. G.; Epstein, A. J. Synthetic Metals 1994, 65, 103-116. (154) Politis, J. K.; Curtis, M. D.; Gonzalez-Ronda, L.; Martin, D. C. Chemistry of Materials 2000, 12, 2798-2804. (155) Spano, F. C. Journal of Chemical Physics 2002, 116, 5877-5891. (156) Spano, F. C.; Siddiqui, S. Chemical Physics Letters 1999, 314, 481-487. (157) Cornil, J.; Beljonne, D.; Shuai, Z.; Hagler, T. W.; Campbell, I.; Bradley, D. D. C.; Bredas, J. L.; Spangler, C. W.; Mullen, K. Chemical Physics Letters 1995, 247, 425-432. (158) Hu, D. H.; Yu, J.; Barbara, P. F. Journal of the American Chemical Society 1999, 121, 6936-6937.

PAGE 190

175 (159) Politis, J. K.; Nemes, J. C.; Curtis, M. D. Journal of the American Chemical Society 2001, 123, 2537-2547. (160) Curtis, M. D. Macromolecules 2001, 34, 7905-7910. (161) Koren, A. B.; Curtis, M. D.; Kampf, J. W. Chemistry of Materials 2000, 12, 1519. (162) Harrison, M. G.; Friend, R. H. In Electronic Materials: The Oligomer Approach; Mullen, K., Wegner, G., Eds.; Wiley-VCH: Weinheim, 1998. (163) Sotzing, G. A.; Reynolds, J. R.; Steel, P. J. Advanced Materials 1997, 9, 795-&. (164) Monk, P. M. S.; Mortimer, R. J.; Rosseinsky, D. R. Electrochromism: Fundamentals and Applications; VCH: Weinheim, 1995. (165) Rauh, R. D. Electrochimica Acta 1999, 44, 3165-3176. (166) Lampert, C. M. Solar Energy Materials and Solar Cells 1998, 52, 207-221. (167) Hale, J. S.; Woollam, J. A. Thin Solid Films 1999, 339, 174-180. (168) Chandrasekhar, P.; Zay, B. J.; Birur, G. C.; Rawal, S.; Pierson, E. A.; Kauder, L.; Swanson, T. Advanced Functional Materials 2002, 12, 95-103. (169) McDonagh, A. M.; Bayly, S. R.; Riley, D. J.; Ward, M. D.; McCleverty, J. A.; Cowin, M. A.; Morgan, C. N.; Varrazza, R.; Penty, R. V.; White, I. H. Chemistry of Materials 2000, 12, 2523-2524. (170) Granqvist, C. G. Electrochimica Acta 1999, 44, 3005-3015. (171) De Paoli, M. A.; Casalbore-Miceli, G.; Girotto, E. M.; Gazotti, W. A. Electrochimica Acta 1999, 44, 2983-2991. (172) Trimble, C.; DeVries, M.; Hale, J. S.; Thompson, D. W.; Tiwald, T. E.; Woollam, J. A. Thin Solid Films 1999, 356, 26-34. (173) Benett, R. B.; Koronasky, W. E.; Hannan, M. J.; Boxall, L. G. US Patent 5,446,577, 1995. (174) Chandrasekhar, P. US Patent 5,995,273, 1999. (175) Kido, J.; Kimura, M.; Nagai, K. Science 1995, 267, 1332-1334. (176) Brotherston, I. D.; Mudigonda, D. S. K.; Osborn, J. M.; Belk, J.; Chen, J.; Loveday, D. C.; Boehme, J. L.; Ferraris, J. P.; Meeker, D. L. Electrochimica Acta 1999, 44, 2993-3004. (177) Rosseinsky, D. R.; Mortimer, R. J. Advanced Materials 2001, 13, 783.

PAGE 191

176 (178) Arbizzani, C.; Mastragostino, M.; Zanelli, A. Solar Energy Materials and Solar Cells 1995, 39, 213-222. (179) Arbizzani, C.; Mastragostino, M.; Meneghello, L.; Morselli, M.; Zanelli, A. Journal of Applied Electrochemistry 1996, 26, 121-123. (180) Kim, J.; Swager, T. M. Nature 2001, 411, 1030-1034. (181) Sirringhaus, H.; Tessler, N.; Friend, R. H. Science 1998, 280, 1741-1744. (182) Montali, A.; Bastiaansen, G.; Smith, P.; Weder, C. Nature 1998, 392, 261-264. (183) Mudigonda, D. S. K.; Meeker, D. L.; Loveday, D. C.; Osborn, J. M.; Ferraris, J. P. Polymer 1999, 40, 3407-3412. (184) Burn, P. L.; Holmes, A. B.; Kraft, A.; Bradley, D. D. C.; Brown, A. R.; Friend, R. H.; Gymer, R. W. Nature 1992, 356, 47-49. (185) Meeker, D. L.; Mudigonda, D. S. K.; Osborn, J. M.; Loveday, D. C.; Ferraris, J. P. Macromolecules 1998, 31, 2943-2946. (186) Gazotti, W. A.; Casalbore-Miceli, G.; Geri, A.; Berlin, A.; De Paoli, M. A. Advanced Materials 1998, 10, 1522. (187) Sotzing, G. A.; Reynolds, J. R.; Steel, P. J. Abstracts of Papers of the American Chemical Society 1997, 214, 288-POLY. (188) Sotzing, G. A.; Reddinger, J. L.; Katritzky, A. R.; Soloducho, J.; Musgrave, R.; Reynolds, J. R. Chemistry of Materials 1997, 9, 1578-1587. (189) Sotzing, G. A.; Thomas, C. A.; Reynolds, J. R.; Steel, P. J. Macromolecules 1998, 31, 3750-3752. (190) Rauh, R. D.; Wang, F.; Reynolds, J. R.; Meeker, D. L. Electrochimica Acta 2001, 46, 2023-2029. (191) Sotzing, G. A.; Reynolds, J. R. Journal of the Chemical Society-Chemical Communications 1995, 703-704. (192) Zong, K.; Reynolds, J. R. Journal of Organic Chemistry 2001, 66, 6873-6882. (193) Thomas, C. A.; Zong, K.; Schottland, P.; Reynolds, J. R. Advanced Materials 2000, 12, 222. (194) Schottland, P.; Zong, K.; Gaupp, C. L.; Thompson, B. C.; Thomas, C. A.; Giurgiu, I.; Hickman, R.; Abboud, K. A.; Reynolds, J. R. Macromolecules 2000, 33, 7051-7061.

PAGE 192

177 (195) Gaupp, C. L.; Zong, K. W.; Schottland, P.; Thompson, B. C.; Thomas, C. A.; Reynolds, J. R. Macromolecules 2000, 33, 1132-1133. (196) Sonmez, G.; Schwendeman, I.; Schottland, P.; Zong, K.; Reynolds, J. R. Macromolecules, submitted for publication. (197) Gaupp, C. L.; Welsh, D. M.; Rauh, R. D.; Reynolds, J. R. Chemistry of Materials, accepted for publication.

PAGE 193

BIOGRAPHICAL SKETCH I was born in Bucharest, Romania, to a family of chemists, so I was exposed to science since I was a baby. After finishing at a special physics and mathematics high school, I decided to enroll in the Polymer Technology Program at the Polytechnical University of Bucharest, from where I graduated in 1994 among the top of my class. I made the decision to continue my education in the US in 1996, while I was finishing my masters in polymer science at the Polytechnical University of Bucharest. The fact that I had not taken an English class before did not keep me from learning this language by myself and passing the required tests. In June 1997 I came to the University of Florida and I immediately joined Dr. Reynolds group. Since then, I have been involved in optical and transport studies of conducting polymers, and their application to electrochromic devices. I met John in August 1997, during the first week of classes. In the fall of 2000, we got married and our daughter, Laura, was born in 2001. 178


Permanent Link: http://ufdc.ufl.edu/UFE0000606/00001

Material Information

Title: Optical and transport properties of conjugated polymers and their application to devices
Physical Description: Mixed Material
Creator: Schwendeman, Irina ( Author, Primary )
Publication Date: 2002
Copyright Date: 2002

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000606:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000606/00001

Material Information

Title: Optical and transport properties of conjugated polymers and their application to devices
Physical Description: Mixed Material
Creator: Schwendeman, Irina ( Author, Primary )
Publication Date: 2002
Copyright Date: 2002

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000606:00001


This item has the following downloads:


Full Text











OPTICAL AND TRANSPORT PROPERTIES OF CONJUGATED POLYMERS AND
THEIR APPLICATION TO ELECTROCHROMIC DEVICES











By

IRINA SCHWENDEMAN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2002




























Copyright 2002

by

Irina Schwendeman





























To John, Laura and Ileana















ACKNOWLEDGMENTS

First, I would like to thank my advisor, Dr. John Reynolds, for sending me a nice

contact letter in 1996 when I was deciding what university to join for graduate studies.

Since then, his continuous support and advice, which were not limited to science, made

this work possible. I greatly appreciate that Dr. Reynolds was there when I needed

counseling, and seemed always interested and appreciative of my work. Learning new

concepts is not hard and can be done independently, but becoming a good scientist

requires a great deal of fine guidance and effort, and for this I thank Dr. Reynolds the

most. Several professors from Polytechnical University of Bucharest had a major

contribution to my interest in polymer science. Alexandru Stefan, my undergraduate

advisor, Nicolae Cobianu, my master's advisor, and Mihai Dimonie, who was about to

become my Ph. D. advisor, prepared me for the next step in my scientist life.

For being outstanding coworkers and friends, and helping me to adjust to a new

country, I thank Jennifer and David Irvin, Fatma Sotzing and Hiep Ly. I was fortunate to

work with two wonderful undergraduate students, Jessica Hancock and Roberta

Hickman, who made my work day twice as productive and ten times more fun. I do not

attempt to point out which experiments they helped me with because their contributions

are spread over each project presented in this thesis. Room 300 in Leigh Hall is a special

environment, where many people come and go in a short period of time. Over the last two

to three years, besides myself, there has been only one other long term resident of this

lab, Avni Argun. I would like to thank Avni for taking over all the lab duties, for his









immediate help when I needed it and for the comfortable thought that there is somebody

to count on around. I thank Chris Thomas, who took the time to teach me the basis of

electrochemistry and spectroelectrochemistry like nobody else, despite his approaching

oral examination. I also thank Chris Thomas, Phillippe Schottland and P.H. Aubert for

being so knowledgeable and answering any question that I might have. Kyukwan Zong,

Dean Welsh and Carl Gaupp have significantly contributed to this work by providing the

monomers and allowing me to take the fun part of the work, which is the characterization

of new compounds and their application to devices. Thanks go to Mohamed Bouguettaya

and Charlotte Cutler for fun and relaxing conversations over a coffee at Java. I would like

to thank Barry Thompson, Nisha Ananthakrishnan and all of the other Reynolds group

members for being helpful when I needed it and generally fun to be around.

My work is characterized by having multiple collaborations, especially with

physics departments ("You are turning into a physicist!" Frechet, Fall 2000). Dr. Arthur

Epstein from Ohio State University made a great impact on my work. Data presented in

Chapter 3 were collected by his students as follows: Alexey Saprigin did the reflectance

and the X-ray measurements, Won-Pil Lee did the low temperature mw and dc studies,

and Keith Brenneman performed EPR experiments. I would like to thank Dr. Epstein and

his group, and especially Alexey, for being such nice hosts when I visited Columbus, OH.

Dr. Tanner from the Physics Department at UF has been like a second advisor for me,

and was always extremely nice and helpful. The collaboration with his group resulted in

the reflectance/transmittance data from Chapter 4 and the variable reflectance mirror

project presented in Chapter 5. Jungseek Hwang from Dr. Tanner's group (my three year

long collaborator), and recently Maria Nikolou and Matt Cornick, performed the above









experiments. Bert Groenendaal from AGFA provided the alkyl-substituted EDOTs and

EDOT-F.

The greatest acknowledgment goes to my family for their love and understanding

that make my life complete. To John, thanks go for being such a loving husband and

good team member, and for sharing with me every joy and every worry. My daughter,

Laura, actively contributed to this thesis by being such an adorable baby and sleeping

through the night when we needed to work, and loving us unconditionally even when we

were tired and not in the mood to play. I thank my mom and dad, my aunt and

grandmother for giving me the intellectual and emotional guidance that made me who I

am.
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ .. ............ ix

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT ........ .............. ............. ...... ...................... xiv

CHAPTERS

1 IN TR OD U CTION .............................................. .. ....... .... .............. .

1.1 Brief History of Conducting Polymers .......................................... ...............1
1.2 M etals, Sem iconductors and Insulators ....................................... ............... 5
1.3 Electrochem ical Polym erization ...................................................................... 13
1.4 Doping Induced Property Changes........... .......... .......................... 16
1.5 A applications ............. ................................................................. ............. 20
1.6 Structure of this Thesis ................................................ .............................. 26


2 EXPERIM ENTAL TECHNIQUES.......................................... .......................... 29

2.1 Conductivity M easurem ents ............................................................................ 29
2.2 Electrochem ical M ethods............................ ............................ 36
2.3 Spectroelectrochemistry and Other Optical Measurements...............................41
2.4 C olorim etry ....................................................... ............ ......... 43
2.5 Surface A analysis .......................................... ... .... ........ ......... 45
2.6 Device Construction .................. .............. ............ ..... .....................47
2.7 Purification of Laboratory Chemicals and Materials............................................51


3 ELECTRONIC TRANSPORT AND OPTICAL PROPERTIES OF PXDOT AND
PXDOP FREE-STANDING FILM S ........................................ ....................... 53

3.1 Introduction .................. ............ ............................................. 53
3.2 Mechanisms for ElectronicTransport ......................................... ...............54
3.3 Elementary Excitations in Conducting Polymers ...............................................60
3.4 Metal-Insulator Transition in Conducting Polymers ..........................................62
3.5 Electrochemical Synthesis of Free-Standing Films .............................................64









3.6 Temperature Dependence of the Conductivity .....................................................67
3.7 M icrow ave Experim ents ............................................... ............................. 71
3 .8 M agnetic Su sceptibility .............................................................. .....................73
3.9 X -ray M easurem ents ...........................................................................75
3 .10 O optical C ondu activity .................. ......... .............................. ............... ...75
3.11 Room Temperature Conductivities of Alkyl Substituted PEDOTs.....................78


4 POLYM ER THIN FILM OPTICS........................................... ........................... 81

4.1 Introduction ............................................................................................. ........... 81
4 .2 Surface A naly sis ..................... ........................................ .......... ............82
4.3 Fundamental Spectroelectrochemistry .............. .......................................... 86
4.4 Reflectance and Transmittance Spectra......................................................89
4.5 The Origin of the Fine Structure in Neutral Polymers.........................................97
4.6 Spectroelectrochemical Data of PXDOT Series.............................................. 103


5 VARIABLE REFLECTANCE MIRRORS ............ .............................................117

5.1 Introduction ......... .......................... ............. ... ........ ............... 117
5.2 Devices as Platforms to Study Electronic Spectra..............................................118
5.3 In-situ Reflectivity M easurements....................................................................... 121
5.4 Active Polymer Layer Thickness Optimization...................... ...............124
5.5 Sw itching Tim e ...................................................... ...... .... ............... 127
5.6 Long-Term Sw itching Stability ........................................ ....................... 129
5.7 O -H and C -H R em oval ............................................... ............................. 130


6 ABSORPTION-TRANSMISSIVE WINDOWS .................................. ...............134

6.1 Introduction ............................... .. .................. .......... ........... 134
6.2 Absorption/Transmission Device Construction.................... ...............135
6.3 Cathodically Coloring Polym ers................................... .................................... 137
6.4 Anodically Coloring Polymers ................................................. 139
6.5 Matching Two Complementary Coloring Polymers ................. ... .................142
6.6 H igh Contrast ECD s ...................................... .. ................................. 146
6.7 Broadband Absorption Devices: Laminate Alternative.................. ........... 153
6.8 Broadband Absorption Devices: PEDOT-F Alternative............... .......... 159
6 .9 C on clu sion s............................ ..................................................... ............... 164


LIST OF REFEREN CES .................................................................. ............... 166

BIO GR A PH ICA L SK ETCH .................................... ........... ......................................178





viii
















LIST OF TABLES


Table page

1.1 Qualitative properties of conducting polymers in two extreme redox states............... 17

1.2 Applications of conducting polym ers ............................................... ............... 21

3.1 Electrochemical synthesis of free-standing PXDOT films............... ...................65

3.2 Transport properties of PXDOT and PXDOP free-standing films ............................68

4.1 The electronic structure of PEDOT, PProDOT, and PProDOT-Me2 in their neutral,
lightly doped and fully doped states................................ ................................. 93

4.2 Spectroelectrochemical results for PXDOT series ......................................... 106

6.1 Band gap and monomer oxidation potential values for a series of propylenedioxy-
derivatized pyrroles and thiophenes ......................................... ...............142
















LIST OF FIGURES

Figure p

1.1 Comm on conjugated polym ers .................................. ......................................2

1.2 Energy bands developing in m etallic sodium ........................................ .....................5

1.3 Description of materials according to their band structure...............................6

1.4 Conducting properties of metals, semiconductors and insulators ..............................8

1.5 E evolution of the band gap in PA c ......................................................................... 10

1.6 Magnitude of room temperature conductivity for different types of conducting
p o ly m ers ............................................................................12

1.7 Electropolym erization m echanism ...........................................................................14

1.8 Doping mechanisms in conjugated polymers and their applications...........................18

1.9 D hoping m methods in conjugated polym ers ........................................ .....................19

1.10 General designs of conducting polymer devices ................................................. 23

2.1 Four-probe conductivity m ethods.................................................................... ...... 32

2.2 Cyclic voltammogram of BEDOT-B(OG)2 .................. ...........................37

2 .3 E lectrochem ical cell setup ........................................ .............................................39

2.4 Electrode potential relationship between common reference electrodes...................40

2.5 x-y chrom aticity diagram ............................................................................. ............44

3.1 Schematic drawing of density of states.................................................................. 55

3.2 Scanning force microscopy image of polymer chains.............................................58

3.3 Schematic drawing on the conducting polymer structure ........................................59

3.4 Band diagrams for neutral and positive solitons .....................................................60









3.5 Polaron and bipolarons in non-degenerate ground state polymers ...........................61

3.6 Temperature dependence of the conductivity ...........................................................70

3.7 The conductivity dependence of T-1/4 in PXDOT films............................................71

3.8 Temperature dependence of the dielectric constant in PXDOT films.........................72

3.9 Temperature dependence of the mw and dc conductivity ................ ..................73

3.10 EPR Susceptibility of Doped PXDOT Films ......................................................74

3 .1 1 X -ray d iffractio n .............................................................................. ................ .. 7 6

3.12 Optical properties of PXDOT films ............... ................................. ...............77

3.13 Conductivity of alkyl substituted PEDOTs..................................... ............... 80

4.1 T he m odel of film form ation.................................................. ................. ..................... 84

4.2 Surface m orphology of PANI film s ........................................ ....................... 84

4.3 Surface m orphology ......... .................................................................... .. .... ....... 85

4.4 Doping induced electronic transitions in conducting polymers ...................................87

4.5 N eutral PProD O T stability studies........................................ ........................... 90

4.6 Schem atic w ith polym er/ITO/glass layers ..................................................................91

4.7 Absorption coefficients of: A) PEDOT, B) PProDOT, C) PProDOT-Me2 ................92

4.8 Infrared active vibrational m odes ..................................................... .....................95

4.9 Solid-State UV-Vis spectra of regioregular poly(3-dodecylthiophene) ......................98

4.10 Schematic representation of the vibronic transition.....................................100

4.11 Absorption spectra of PPV oligomers with different molecular weights..............101

4.12 V visible absorption spectra.......................... ................................... ............... 105

4.13 Band gap dependence of the side chain length in alkyl-substituted PEDOTs.........108

4.14 Four electrochemical polymerization routes to PProDOT............... ...............110

4.15 Potential sweep growth of PProDOT............................ .... .................111

4.16 Regioregularity of alkyl-substituted PEDOTs............................... ..................1.13









5.1 Side-view schematic diagram of a dual polymer electrochromic device ................19

5.2 Top-view photograph of a PProDOT-Me2/PBEDOT-NMeCz dual polymer
electrochrom ic device ............................................ .. ......... .............. 120

5.3 Reflectance of the PProDOT-Me2/PBEDOT-NMeCz dual polymer electrochromic
d ev ice .............................................................................12 3

5.4 Device contrast as a function of frequency.............. ................................125

5.5 Active layer thickness dependence of the electrochromic contrast in PEDOT based
devices,. ............................................................ ........ ......... 126

5.6 Switching time of PProDOT-Me2 based device recorded at two IR frequencies......128

5.7 Reflectance as a function of number of deep double potential switches for PProDOT-
M e2 based device .................................................... ..... .. ........ .... 130

5.8 Transmittance spectra of two gel electrolyte batches .............................................. 132

5.9 Reflectance spectra of a PProDOT-Me2 based device.....................................133

6.1 Variable transmittance window schematic. .................................... .................136

6.2 Cathodically and anodically coloring polymers ................................ ............... 138

6.3 Trimer of A) PProDOP and B) PProDOP-NPrS Hyperchem illustrations ..............141

6.4 The superimposition of the UV-Vis-NIR spectra of the individual polymer films
deposited on ITO/glass substrates................................................. ............... 143

6.5 Thickness as a function of charge density passed during electrodeposition............145

6.6 AFM topographic image representing a scratch through a polymer film.................. 146

6.7 Transmittance spectra and photographs of devices using PProDOT-Me2 as the
cathodically coloring polym er......................................... ........................... 147

6.8 Transmittance as a function of switching time .................................. ............... 149

6.9 Hue and saturation (x-y track) for the device B................................. ... ................ 150

6.10 Luminance PProDOT-Me2/ PProDOP-NPrS device........................................151

6.11 Lifetime of the PProDOT-Me2/PProDOP-NPrS device........................ ...............152

6.12 L am inate device scheme atic ......... ................. ...................................................... 153

6.13 PEDOP/PProDOT-Me2 bilayer electrode absorption spectra...............................155









6.14 Charge trapping mechanism in PEDOP/PProDOT-Me2 bilayer electrode..........156

6.15 Transmittance spectra of the PEDOP/PProDOT-Me2/PBEDOT-NMeCz laminate
d ev ice ............................................................................. 15 7

6.16 Luminance analysis of the PEDOP/PProDOT-Me2/PBEDOT-NMeCz laminate
d ev ice .......................................................................... 15 8

6.17 Absorption spectra of PEDOT-F in 0.1 M TBAP/ACN..........................................159

6.18 Spectroelectrochemical analysis of PEDOT-F/PBEDOT-NMeCz device. .............162

6.19 Colorimetric analysis of PEDOT-F/PBEDOT-NMeCz device ..............................163















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

OPTICAL AND TRANSPORT PROPERTIES OF CONJUGATED POLYMERS AND
THEIR APPLICATION TO ELECTROCHROMIC DEVICES

By

Irina Schwendeman

December 2002


Chair: Professor John R. Reynolds
Major Department: Chemistry

This work combines fundamental studies on the electrical and optical properties of

conducting polymers from the poly(3,4-alkylenedioxythiophene) (PXDOT) and poly(3,4-

alkylenedioxypyrrole) (PXDOP) class with more application-driven studies of

electrochromic devices. Understanding the mechanism behind achieving a high

conductivity allows for an accurate design of the next generation of organic materials

with metal-like electrical properties. Optical reflectance, microwave and dc conductivity,

and EPR and X-ray measurements are employed to gain insight into the electronic

transport of doped, free-standing polymer films. These materials' electronic properties

can be tuned from metallic to insulator by varying the dopant nature and monomer unit

planarity. For example, PEDOT-PF6 shows a metallic-like transport behavior, PEDOP-

CF3SO3 and PEDOT-CF3SO3 are in the critical regime, whereas PEDOT-(CF3SO2)2N,

PProDOT-CF3SO3 and PProDOT-PF6 are on the insulator side of metal to insulator

transition and show semiconducting properties.









Since many promising applications revolve around dynamic changes in the optical

properties of electroactive thin films, this work focuses on elucidating the doping-induced

electronic structure and infrared active vibrational modes (IRAV) in PXDOT films.

Transmission and reflectance spectra spanning the UV, visible, near- and mid-IR regions

of the spectrum reveal polaron and bipolaron signatures as the doping level increases.

Neutral polymers have unique visible absorption spectrum envelopes, such as a broad

single peak or multiple sharp peaks, as a function of the nature of the monomer unit. A

discussion on the existing theories for the origin of this fine structure supported by data

taken for a basis set of sixteen PXDOT thin films suggests that vibronic coupling of

excited electrons with a vibrational mode of the monomer unit (symmetric, in plane

Ca=Cp stretch) is responsible for the appearance of this 7-7 absorption peak splitting.

A profound understanding of these materials' optical and electrochemical

properties allows the construction of electrochromic devices with enhanced contrast, long

term switching stability and fast optical response. Variable reflectance electrochromic

devices based on PXDOTs have been optimized to exhibit contrast ratios of 60-80% from

the visible to mid-IR regions. They exhibit remarkable lifetimes, with no significant loss

in electrochemical contrast after 10,000 redox cycles. The use of carefully designed

optically complementary polymers PProDOT-Me2 and PProDOP-NPrS is a promising

route for achieving a high level of control over the color, brightness, and redox stability

of a transmission/absorption window. These devices operate at low applied voltages (+

1.0 V), exhibiting an optical contrast of up to 70% at Xmax, as well as an overall

luminance change of 53%. They retain 86% of their initial color (96% retention after

break-in) after 20,000 redox cycles.














CHAPTER 1
INTRODUCTION


1.1 Brief History of Conducting Polymers


Twenty years after the discovery by Heeger, MacDiarmid and Shirakawa that an

organic conjugated polymer can reach metallic-like electrical conductivity upon

doping,1,2 the field of conjugated polymers has enjoyed a tremendous development

culminating in 2000 when the Nobel Prize for Chemistry was awarded to the above

researchers for launching a new era in polymeric materials.

Conjugated polymer synthesis reaches as far back as 1862, when H. Letheby first

synthesized polyaniline (PAni, 3).3 Known as "aniline black", this material was formed

by oxidation of aniline under mild conditions and was used in the printing industry.4 The

first polymerization of acetylene to form polyacetylene (PAc, 1) was reported in 1958 by

Natta and coworkers.5 Because PAc was obtained as an insoluble and infusible powder,

the material received little attention at that time. The idea that conjugated polymers could

be good electrical conductors has roots back to the 1960s when MacDiarmid and others

discovered that poly(sulfurnitride) (SN)x, a polymeric inorganic explosive,6 has a high

conductivity.7 The interesting electrical properties of (SN)x represented a step towards

conducting polymers as they are known today.

The modem era of conducting polymers began at the end of 1970s when films of

PAc were found to exhibit a 12 order of magnitude increase in electrical conductivity









when exposed to iodine vapors. 1,2 The procedure for synthesizing PAc was based upon a

route discovered in 1974 by Shirakawa through an accidental addition of 1,000 times

more catalyst during the polymerization of acetylene. Although initially it was thought

that PAc would replace dense metals in air and space applications, its instability to

environmental conditions constituted a major obstacle for any practical use. However,

PAc, being the simplest model from this class, remains the archetype of conducting

polymers and is still subject to much theoretical and experimental work.


NH
n n
PAc, 1 PPV, 2 PAni, 3



H

~n n 0 0

PPy, 4 PT, 5 PEDOT, 6
Figure 1.1 Common conjugated polymers

The opportunity to synthesize new conducting polymers with improved properties

began to attract the attention of synthetic chemists in the early 1980s. During this time,

the discovery that polypyrrole (PPy, 4) can be obtained as highly conducting and

homogeneous, free-standing films via oxidative electropolymerization8 focused the

research efforts towards the development of conjugated poly(heterocycles).

Electrochemical polymerization was rapidly extended to other aromatic compounds such

as aniline, thiophene, furan, indole, carbazole, indole, azulene, pyrene, and fluorene.9,10

Although exhibiting lower conductivities than doped PAc, the most common conjugated

poly(heterocycles) shown in Figure 1.1 (compounds 4-6) posses superior environmental









stabilities in their p-doped form, along with other interesting properties such as

electrochomism, volume change during doping, electro- and photoluminescence, etc. 11

Among these so-called "first generation" conducting polymers, polythiophene (PT,

5) has rapidly become the subject of considerable interest, mainly due to its structural

versatility and enhanced stability in both neutral and p-doped states. From a theoretical

point of view, PT has been often considered as a model for the study of charge transport

in non-degenerate ground state polymers. However, PTs are unstable at the potentials

required for their formation. This effect is called "PT paradox" and means that the

polymer degradation competes with its deposition, leading to polymers with a high

content of overoxidized, non-electroactive material. A successful strategy to control the

properties of PTs involves the modification of the monomer structure to obtain easily

polymerizable and eventually solution processable species. This led to the synthesis of a

vast family of PT derivatives with varied interesting properties. 12,13

As a member of conducting polymers' second generation, the 3,4-ethylenedioxy

derivative of PT (PEDOT) stands out as an excellent candidate for use in a variety of

applications such as electrochromic devices, LEDs, capacitors and sensors. Initially

developed to give a processable polymer (as an aqueous dispersion) with a high degree of

order due to the lack of ca-3 and P3-3 couplings, PEDOT showed other very interesting

properties worth pursuing by both industry and academia. In addition to its high

conductivity (300-400 S/cm), PEDOT exhibits high electrochromic contrast with the

major advantage of being almost transparent in thin, doped films. 14,15 Other advantages

include low monomer oxidation potential, high stability in the doped form and an ease of

derivatization at the ethylenedioxy ring, thus allowing for state-of-the-art tuning of the









materials' electronic and optical properties. The industrial use of the polymer as an

antistatic layer in photographic films makes it the most widely used conducting polymer

to date.16

In this context, the past few years have witnessed the emergence of an impressive

variety of EDOT derivatives.16 Functionalization of the monomer with long alkyl or

alkoxy pendant groups affords materials that can be processed by common methods

including spraying, printing, spin coating or solution casting.17,18 Moreover, the

introduction of water-soluble sodium salt units as pendant groups allows water-

processability along with self-doping. This type of polymer has been deposited on

several substrates via the electrostatic adsorption technique.19-21 To expand the range of

substituted 3,4-alkylenedioxy thiophenes, our group synthesized monomers having

several ring sizes, such as 7-membered (ProDOT)22 and 8-membered (BuDOT) rings.17

ProDOT has received special attention, as the monomer can be symmetrically derivatized

at the central carbon of the propylene bridge, resulting in a regiosymmetric polymer.

The bandgap of EDOT-containing polymers can be modified by varying the degree

of 7t-overlap along the polymer backbone via steric interactions and by controlling the

electronic character of the 7t-system with electron donating or accepting units. The latter

is accomplished by using substituents and co-repeat units that adjust the energy of the

highest occupied molecular orbital (HOMO) or valence band and the lowest unoccupied

molecular orbital (LUMO) or conduction band, thus obtaining polymers with a broad

range of colors. In this manner, materials with higher gaps than the PEDOT parent have

been prepared, some of which are used as anodically coloring polymers in electrochromic









devices.23-25 Further work using a donor-acceptor methodology has led to low band gap

polymers able to p- and n- dope along with exhibiting multi-color electrochromism.26


1.2 Metals, Semiconductors and Insulators


Energy bands. Electrons in an isolated atom are described by their atomic orbital

wave functions. Their energies are determined solely by the interaction with the nucleus

and other electrons of the atom. When atoms are brought together to form crystals,

energy bands evolve, as illustrated in Figure 1.2 for sodium. At infinite separation (Figure

1.2 A), all the atomic orbitals have discrete energies. As the atoms are brought together,

their orbitals begin to overlap, and energy bands result (Figure 1.2 B).


0
4s
3p I
3s




2p
R= oA 10 A 3.67 A

A B C


Figure 1.2 Energy bands developing in metallic sodium: A) large atomic separation, B)
atoms separated by 10 A, C) actual separation in metallic sodium (from Slater27)

When the interatomic distance becomes comparable to or less than the spatial

extension of the electronic wavefunction associated with a particular atom, the valence

electrons do not belong to any single atom, but to the entire material (Figure 1.2 C).

However, the Pauli exclusion principle states that the electrons of a certain atomic orbital









cannot have the same energy. Consequently, the electrons form a band of energy levels

with very small energy differences between individual levels.27

Metals, semiconductors and insulators. According to the band theory, the

electrical properties of a material are the direct consequence of the energy difference



CB
CB
CB CB
SmaE Large Eg
Small Eg
VB
1B
VB 1B



Metal Semimetal Semiconductor Insulator

A B C D
Figure 1.3 Description of materials according to their band structure: A) Metal, B)
Semimetal, C) Semiconductor, D) Insulator

between HOMO and LUMO, as well as of the level of band filling and band overlapping.

The basic difference between metals and other solids (semiconductors and insulators) is

that HOMO is only partially occupied (Figure 1.3 A), whereas this band is filled in

semiconductors and insulators (Figure 1.3 C and D). As a consequence of this energy

band configuration, there is a sharp distinction between the materials' electrical

conductivity, in particular as the temperature approaches 0 K.28

Figure 1.4 shows a comparison of electrical conductivities of metals,

semiconductors and insulators on a logarithmic scale (A) along with temperature

dependence of the conductivity for materials from each conductivity class (B). Electrical

conductivity is described by the equation:









o = neCt (1.1)

where n is the density of charge carriers contributing to conductivity, e is the elementary

charge, and / is charge carrier mobility (velocity per unit electric field).

In a metal, both n and /u have large values in the range of 1022 cm-3 and 103

cm2/Vs,28 respectively, while the elementary charge is 1.602x10-19 C. They combine to

give large values of conductivity of 105-106 S/cm, as shown in Figure 1.4 A and B (Cu).

In addition, the number of charge carriers contributing to conductivity is constant at all

temperatures, due to the lack of a band gap between HOMO and LUMO. The progress of

electrons in a metal is interrupted after traveling a distance corresponding to mean free

path or a time equivalent to the scattering relaxation time. The dominant scattering

mechanism is the interaction of itinerant electrons with lattice atoms as they vibrate with

thermal energy. Therefore, the mobility of charge carriers, and consequently the

conductivity of a metal, increases as the temperature approaches absolute zero and the

electron-lattice scattering is diminished.

Semiconductors have filled valence bands and empty conduction bands. The gap

energy is small as compared to the larger gap in insulators. For intrinsic semiconductors,

electrons in the valence band can be thermally activated into the conduction band, and the

density of such electrons follows an Arrhenius law. During the activation process, a

corresponding hole is created in the valence band and both free electrons and free holes

contribute equally to the conductivity. These semiconductors have a positive temperature















SILVER
COPPER
IRON
BISMUTH


InSb


SILICON


GLASS-

] DNA

DIAMOND

SULFUR
QUARTZ


106

- 104

- 102

- 1

- 10-2



- 10-6



10-10

- 10-12

- 10 14

- 10-16

- 10-18
1-1 cm-1


POLY-
ACETYLENE
on>2 X 104 om 1
^


(SN),
TTF'TCNQ
NMP-TCNQ
KCP





TRANS {CH),




CIS (CH)


MOST
MOLECULAR
CRYSTALS


POLY-
THIOPHENE
Ox,= 2000 S cm


DOPED












UNDOPED
POLY
PYRROLE
t*_=5000cm-1
H
N^


106
B (SN) (SEMIMETAL)
104 -


S102 (TTF)(TCNQ) (TYPE III LINEAR CHAIN CONDUCTOR)


-; 1
g I I" a -a

eg -; ciua m iu
S10- 2
0 PPy-DBSA (SEMICONDUCTOR)

104-
s'"Co aran on of
Craia*Careas
10-6

Temperature --
108
0 50 100 150 200 250 300 350
Temperature /K

Figure 1.4 Conducting properties of metals, semiconductors and insulators. A)

Conductivity scale comparing metals, semiconductor and insulators (adapted from

Skotheim 11) B) Temperature dependence of the conductivity for four materials. The inset

represents the mobility, charge carrier, and semiconductor conductivity variations as a

function of temperature (adapted from Epstein29)









dependence of the conductivity (see PPy behavior in Figure 1.4 B). Even though the

carrier mobility is affected by the same scattering process as in a metal, the number of

charge carries at the certain temperature dominates the overall temperature dependence of

the conductivity. In addition, the room temperature conductivity of semiconductors is

lower than the conductivity of metals due to a reduced charge carrier density. In extrinsic

semiconductors, the number of charge carriers can be increased by introduction of

impurities (dopants). In these materials, the carrier density is affected more by dopant

nature and concentration than by the temperature. Classical extrinsic semiconductors are

silicon-based doped with small amounts of gallium (p-type semiconductor) or arsenic (n-

type semiconductor).

In insulators, the electrons are strongly localized between the atoms, forming

chemical bonds. There is no significant overlap between nT-orbitals of adjacent atoms and

the valence band is completely filled. Furthermore, a large gap separates it from the

conduction band, which is not accessible by thermal excitation. For example, polymers

having only o bonds are good insulators.

There is a limited number of materials that have filled valence bands that overlap

with conduction bands to generate two incompletely filled bands, as show in Figure 1.3 B

and 1.4 B (SNx). These materials are called semi-metals and they can reach metal-like

conductivities. One example from this class is graphite that shows a pronounced

anisotropy of conductivity, with a much greater value in the aromatic plane due to a good

7T overlap than the conductivity measured along the nt stack axis.30

Conducting polymers. Trans-PAc offers an excellent model to illustrate the

formation of energy bands.31 The simple chemical structure -(CH2)x- implies that each









nt systems of-(CH)x-


x = 2 4 8 16 Uniform
C-C
bond
order


Alternating
C-C bond
order


7TC* -


Rt


*



t


Akt~


It.


i I

I- I



-- Eg 1.4
t E



f1f


Figure 1.5 Evolution of the band gap in PAc. A) Schematic representation of the n-
molecular orbitals energy levels with increasing chain length in PAc, (adapted from
Chien30) B) uniform C-C bond order in PAc, C) dimerized structure resulting from
Peierls distortion, D) cis-PAc, E) degenerate ground state phases in trans-PAc


D


Long chain PAc









carbon contributes a single pz electron to the '7 band (HOMO) (Figure 1.5 B). As a result,

this band would be half filled, leading to a one-dimensional metal-like conduction along a

neutral PAc chain. Figure 1.5.A shows the increase in the 7t orbital overlap as the length

of the PAc chain increases. However, experimental studies show that neutral PAc is a

semiconductor with a band gap of 1.4-1.5 eV. In the case of trans-PAc, the doubly

degenerate ground state is unstable, and undergoes Peierls distortion32 (equivalent to

Jahn-Teller effect), resulting in an alternation of bond lengths along the backbone.

Further, this process leads to a splitting of the '7 band into an empty conduction band and

a fully occupied valence band, thus rendering PAc semiconducting. In the case of cis-

PAc (Figure 1.5 D), the ground state structure is non-degenerate and the valence band is

filled. Evolution of the band structure from monomer to polymer for conjugated

heterocycles is similar to the one sketched in Figure 1.5 A.33

The magnitude of the room temperature conductivity for different types of

conducting polymers is shown in Figure 1.6 and compared to the conductivities of

copper, platinum, and the range of conductivities of amorphous metals. Data for PAc

doped with MoCl5, FeC13, and C104- are represented under the label PAc(C1) and with

iodine, labeled PAc(I).34 As seen in Figure 1.6, the conductivity of PAc exceeds that of

the amorphous metals and is nearly identical to platinum conductivity. This is remarkable

considering that the density of states at the Fermi level in highly conducting PAc is much

smaller than that of conventional metals, and that the polymer is only partially

crystalline.35 Samples on the insulator side of the metal to insulator transition (M-I

transition) (see Chapter 4 for details) are represented by open symbols. These samples

have zero conductivity as the temperature approaches 0 K. The samples in which the










1 0 6 ........................................................................................................ C u

105 Pt
----- ------e--------------- ---------------------------- e0 ---------
104 -Amorphous metals
103 ---- '8----------------
10 -
i O- *8 *
102 o oo
S0 o 8




O O
101 8
100 0 0
8 0 0

10-1 0 0 0 0
O O
10 -2 I I I I I I I I





Figure 1.6 Magnitude of room temperature conductivity for different types of conducting
polymers (adapted from Kaiser 34). PEDOT conductivity values are taken from
references36-41: Open circles represent samples on the insulator side of M-I transition,
while closed circles represent samples on the metallic side of M-I transition.

conductivity retains a non-zero conductivity value at 0 K are represented by solid

symbols. They have delocalized electronic states at the Fermi level to allow conduction

without thermal activation, and therefore are on the metallic side of M-I transition. The

general trend is that low room temperature conductivity samples show an insulator-type

temperature dependence of the conductivity, whereas most samples having conductivities

higher than 100 S/cm are on the metallic regime. However, there are polymers, especially

PPV (Figure 1.6), that possess high room temperature conductivity without the metallic

signature at low temperatures, probably due to a high content of disorder existent in the

material.

An increase of the effective conjugation length along the backbone and

consequently a reduction in the materials' band gap would allow for achieving polymers









with enhanced electrical conductivities. The band gap of polyaromatic compounds is

determined by five contributions: bond length alternation, planarity of the monomer unit,

aromatic resonance energy of the monomer cycle, donor-acceptor effects of an eventual

substitution, and the extent of interchain coupling.13 Therefore, the main synthetic

strategies adopted for controlling the band gap are based on the modification of one or

more of the above parameters.


1.3 Electrochemical Polymerization


Conducting polymers can be prepared by using chemical and electrochemical

methods. From the applied point of view, electrochemical polymerization of easily

attained, simple aromatic benzenoid, or heterocyclic compounds is of utmost interest,

especially if the polymeric product is used in microtechnology, as a thin layer sensor, or

as a polymer film electrode. Electrochemical formation of conducting polymers such as

PT, PPy and PEDOT proceeds via oxidation of the neutral monomer at the anode

surface.42-44 This oxidation step requires 2 electrons per molecule, the excess of charge

passed during synthesis being necessary for oxidation of the resulting polymer film.

Figure 1.7.A represents the mechanism proposed for the polymerization of

heterocycles, where X can be S, O or N-R. The first step consists in oxidation of the

monomer that yields the radical cation. Monomer diffusion towards the anode is the rate-

determining step, being much slower than the electron-transfer reaction. Consequently, a

high concentration of radical cations is constantly maintained near the anode. The further

fate of these highly reactive species depends on the experimental conditions including

electrolyte composition, temperature, applied potential, nature and morphology of














SI +2H+


2 Q-


+2H


B X X


coupling/ \


coupling ( )


cx x xli n

coupling


Figure 1.7 Electropolymerization mechanism. A) Polymerization of heterocycles (X = S, O, NH),
B) Competitive reaction pathways in unsubstituted poly(heterocycles)


A O


s









deposition support, etc. In favorable cases, the next step is a dimerization reaction

(coupling of two radical cations) that produces a dihydro dication dimer. Further, this

species loses two electrons and rearomatizes to form the neutral dimer. Due to extended

conjugation over two rings, the dimer has a lower oxidation potential than the monomer

itself, and therefore it oxidizes easily to form the radical cation. Stepwise chain growth

proceeds via association of radical ions or, less likely, through coupling of a radical

cation with a neutral monomer. As chain length increases, resulting oligomers become

insoluble in the electrolytic medium and precipitate onto the anode.

First generation heterocycles such as thiophene and pyrrole have two possible

reaction pathways, as shown in Figure 1.7 B. Polymerization proceeding exclusively

through ca-a couplings affords polymers with a linear backbone and enhanced electrical

properties. The occurrence of a ca-3 linkage in a given chain modifies its electronic

distribution and could promote the formation of branching in energetically favorable

sites. Furthermore, the presence of these linkage defects generates twists in adjacent

chains, and thus modifies their electronic distribution promoting the propagation of more

defects in the resulting material. The relative reactivity of the a and 3 positions is about

95/5 for thiophene and decreases as the polymerization proceeds, leading to an increase

of the number of undesired couplings and consequently to a decrease in the polymer

effective mean conjugation length.44 This is consistent with the considerable increase in

the content of disorder as well as the decrease in conductivity as the polymerization

proceeds. Therefore, limitation of the polymer growth to very thin films allows for the

synthesis of a highly compact material with fewer defects and enhanced electrical

properties.45-48









Undesired ca-3 and P3-3 couplings can be eliminated through monomer

substitution at 3 positions. Substitution of thiophene by electron donating groups leads to

a decrease in monomer oxidation potential, allowing for electrosynthesis under milder

conditions and thus decreasing the possibility of obtaining overoxidized material. For

example, in EDOT, both 3 positions are substituted and the polymerization proceeds

exclusively through the desired ca-a couplings. Judicious selection of the substituent

affords materials that are soluble, easily polymerizable and possess an enhanced degree

of order.


1.4 Doping Induced Property Changes


Conducting polymers, once studied solely for their high conductivities, are now

extensively used in more dynamic applications where rapid switching from doped to

neutral forms is desirable. Charge injection (doping) in these materials leads to a wide

variety of interesting and important phenomena, which now defines the conducting

polymer field. This reversible intercalation of ions in the polymer matrix triggers

significant changes in the materials' optical, ionic, electrical and morphological

properties.49 These properties can be tuned by varying dopant size and nature from small

molecules to high molecular weight polymers as well as by using different preparation

techniques.11 Table 1.1 summarizes several properties of conducting polymers that

change according to their charge state. As sketched in Figure 1.8, doping can be

accomplished in several ways depending on the polymer nature and its intended

application. The initial discovery of the ability to dope conjugated polymers involved

chemical doping by charge transfer redox chemistry. 1,2 Oxidation (p-doping) was









Table 1.1 Qualitative properties of conducting polymers in two extreme redox states.

Property Neutral P-doped

Stoic y Without anions (or with With anions (or with
Stoichiomecations cations)
Content of solvent Small Higher
Volume Small Higher
Color: cathodically coloring Transparent or bright Dark
anodically coloring Dark Transparent
IR optical properties Highly transmissive Highly absorptive
Electronic conductivity Semiconducting Metallic
Ionic conductivity Smaller High
Diffusion of molecules Dependent on structure

Surface tension Hydrophobic Hydrophilic


accomplished by exposing the polymer to iodine vapors, whereas reduction (n-doping)

involved treatment with sodium naphthalenide (Figure 1.9 A). In this case, complete

doping results in high quality materials with metallic-like conductivities. Another unique

chemical doping procedure is PAni protonation by acid-base chemistry. This leads to an

internal redox reaction converting the semiconducting form of PAni (emeraldine base) to

a metal (emeraldine salt).50

Although chemical doping is an efficient process, controlling the level of dopant

ions is rather difficult. Attempts to reach intermediate doping levels resulted in

inhomogeneous doping. As an alternative, electrochemical doping allows for fine tuning

of the doping level by simply adjusting the potential between the working and counter

electrodes (Figure 1.9 B).52 The working electrode supplies the redox charge to the


























Photochemical Interfacial


High performance optical materials Organic electrical circuits
Tunable NLO properties Tunneling injection in LEDs
Photovoltaic devices


Figure 1.8 Doping mechanisms in conjugated polymers and their applications (adapted
from Heeger51)

conducting polymer, while ions diffuse in or out of the electroactive film to compensate

the electronic charge. Thus any doping level can be achieved by setting the

electrochemical cell to a desired potential and waiting for the system to attain an

equilibrium state. This type of doping is permanent, meaning that the charge carriers

remain in the film unless a neutralization potential is purposely applied.

Some high performance optical materials require a different type of doping based

on localized oxidation via photo-absorption (Figure 1.9 C). Around these oxidized areas,

the material becomes reduced and charge separation occurs (electron-hole separation).


High electrical conductivity
Solubility
Transparent electrodes, antistatics
Conducting fibers, EMI shielding


Control of doping level
Electrochemical batteries
Electrochromism and smart windows
Light-emitting electrochemical cells









A. Chemical doping by charge transfer

a) p-type doping

(CP)n + 3/2 ny(l2) [(CP)+Y(I3-)y]n

b) n-type doping

(CP)n + [Na+(CloHs)-]y [(CP)-Y(Na+)y]n+ C1oH8

B. Electrochemical doping

a) p-type doping

(CP)n + [Li+(BF4-)]soln [(CP)+Y(BF4-)y]n + Lielectrode

b) n-type doping

(CP)n + Lielectrode [(CP)+Y(Li+)y]n + [Li+(BF4-)]soln

C. Photo-doping

(CP)n + hv (CP)*-[(CP)+y + (CP)-]n

D. Charge injection at a metal-polymer interface

a) Hole injection

(CP)n- ye- [(CP)+]n

b) Electron injection

(CP)n + ye- [(CP)-Y]n




Figure 1.9 Doping methods in conjugated polymers demonstrated for chemical,
electrochemical, photo-, and interfacial doping









Upon photoexcitation from the ground state to the lowest energy excited state,

recombination back to the ground state can be radiative (luminescence) (PPV, PPP) or

non-radiative (PT, PAc).53 Photoconductivity lasts only until the excitations are either

trapped or decay back to the ground state and the material becomes neutral.

Another way to dope a conducting polymer is the electrical injection of electrons

and holes into HOMO and LUMO respectively. However, the polymer is not doped in the

sense of chemical or electrochemical doping because there are no counter ions present in

the film. By charge-injection at the metal-semiconductor interface, the polymer

(semiconductor) can be used as an active layer in field effect transistors and thin film

diodes.54 Dual-carrier injection in a polymer film sandwiched between two metal

electrodes provides the basis for light emitting diodes. In conclusion, as summarized in

Figure 1.8 and Table 1.1, charge injection in conducting polymers opens up a remarkably

wide range of possible applications for this relatively young class of materials.


1.5 Applications


As emphasized throughout this chapter, conducting polymers exhibit novel

properties not typically available in other materials. These unique properties enable a

large number of applications, some of them summarized in Table 1.2. There are

applications involving the neutral or doped state of conducting polymers such as LEDs,

antistatic materials, corrosion protection, and dynamic applications using the reversible

change in properties upon doping, including smart windows and actuators. Some of these

applications are briefly described below.









Table 1.2 Applications of conducting polymers
General function Special function Application System/references
e--conducting film Antistatic coating PEDOT 15,22
e--conducting film Printed circuit PEDOT/epoxy55
Thin film in holes boards
technologies e--conducting film Capacitor Metal/dielectric/PEDOT56
band structure, LEDs Metal/PPV/ITO 57
optical transitions

Material for Matrix for Sensors PAni/glucose-oxidase58
functional
fillings, porous unct
molecules
membranes,
composites Membrane Separation PPy59
Wettability Offset-printing PT 60
Electrochromism Smart windows, PEDOT and derivatives61,
camouflage, pT60
Redox process thermal control
Intercalation Batteries PT/Li62
Change of volume Actuators PPy63

Others Inhibition, Corrosion PAni/Fe64
protection protection


Doped thin film deposition and microstructuring of conducting polymers.

PAni, PPy and PT derivatives are used as antistatic protection and electromagnetic

interference shielding materials.15,22 They are incorporated as fillers in common

polymeric materials such as poly(vinylchloride) and poly(vinylacetate). PEDOT is used

as an antistatic layer for photographic films (AGFA). Another commercial use of PEDOT

is in printed circuit boards. PAni has been proved useful as a discharge solution in

electron beam lithography (IBM). It protects the insulating electron beam resist from

charging and distorting the image.

Materials for energy technologies. The possibility of a reversible switching of

conducting polymers between two redox states seemed promising for rechargeable









batteries (Figure 1.10 A). Owing to their low density, it was thought that batteries with

power densities much higher than those of the ordinary lead/acid battery could readily be

obtained. Since the charge on the polymer backbone is delocalized over 3 to 4 monomer

units, the charge capacity per unit weight for conducting polymers in not much better that

that of metals. The first prototype of commercial batteries with conducting polymers was

based on Li/PAni (BASF/Varta). The possible use of conducting polymers as electrode

material in supercapacitors is given by their high ionic conductivity allowing high

discharge rates. As supercapacitors require high capacitance along with quick

charge/discharge of the electrode material, conducting polymers are promising as

compared to classical used carbon materials.65

Very recently, photovoltaics have emerged as an important application of this

class of materials (Figure 1.10 B). Photovoltaics find potential applications as solar cells

and photosensors, and consist of thin films of organic materials sandwiched between two

metal electrodes. A built-in electrical field formed in the semiconductor in contact with

the electrolyte (photochemical cell) or in the organic layer at the interface with the metal

electrode (photovoltaic device) is responsible for the photogeneration of charge carriers.

Conducting polymers have attracted attention because of their light weight, potentially

low cost and facile fabrication of large area, thin film devices. For example, devices of

A1/C60 modified PTs/ITO show a conversion efficiency of 15 % with zero bias and 60 %

with a bias of 2 V.66,67 Conducting polymers bearing both donor and acceptor units are

expected to afford even higher quantum yields for photogeneration of charge carriers.68












Discharged


P+/M-


B



Glass substrate
Acceptor layer
Donor layer


ITO or PE
Glass


Hole injecting electrode
(ITO)

Conducting polymer layer

Electron injecting electrode (Ca,
Al)


Figure 1.10 General design for A) Polymer based battery; B) Photovoltaic device; C)
OLED, D) Dual polymer electrochromic device


Transparent electrode

Electroacti\e polymer 1

Gel electrolyte

Electroacti e polymer 2

Transparent electrode


Charged









Electroluminescent devices. A significant event occurred when Friend and co-

workers published electroluminescence studies on the neutral form of PPV.54 This work

has opened up a new avenue of research and a potential market for the material. Organic

electroluminescent devices (OLEDs) are a possible alternative to liquid crystals displays

and cathode ray tubes, especially for the development of large displays, and now evolve

beyond the lab research into the market. The basic set-up of a polymeric LED is

presented in Figure 1.10 C and consists of ITO/light-emitting polymer/metal layers. A

thin ITO electrode deposited on an optically transparent electrode (glass or plastic) serves

as the anode, whereas metals such as Al, Ca or Mg are being used as cathode materials.

After the application of an electric field, an electron is injected into the polymer film

from the cathode and a hole is injected from the anode. The electrons and holes migrate

towards the center of the film, where they recombine producing light. One of the most

important benefit of polymeric OLEDs is the chemical tuning of the HOMO-LUMO gap

through judicious synthesis. Therefore, emission of red, green, and blue light has been

reported,66 along with a newer type of emission in the NIR region of the spectrum.69

This emission is obtained by the incorporation of lanthanide metals either by chemical

bonding or blending into the conjugated polymer structure and is based on energy

transfer from the absorbing conducting polymer (usually PPV derivatives) to the emitting

lanthanide atom. In this way, the electroluminescence is enhanced by the absence of

overlap between absorption and emission spectra. Typical materials are PPV and its

derivatives and substituted PTs. OLEDs efficiency is improving steadily along with novel

developments such as flexible LEDs, polarized light emitting LEDs and IR emitting

LEDs. One major challenge is balancing the electron mobility to that of hole mobility.









This is accomplished by adding electron transport layers and hole transport layers to

reduce the barrier height and to encourage holes and electrons to combine near the center

of the film. For example, a thin layer of PEDOT is now commonly used as hole transport

layer in OLEDs.70

Electrochromic devices. There are two types of successful electrochromic devices:

variable transmittance windows and variable reflectance mirrors, described in great detail

in Chapters 6 and 5, respectively (Figure 1.10 D). The principle of such devices is based

on the redox driven change in transmittance/reflection of a thin conducting polymer layer

(or two complementary polymers) over various regions of the electromagnetic spectrum.

Organic electrochromic devices are characterized by low-voltage operation, good optical

contrast and wide-view angle, making them potentially appealing for controlling the sun

radiation in buildings and cars, automotive rear-view mirrors and display devices.71

Moreover, military interest in camouflage materials and thermal control for air and space

applications promotes a great deal of research in dual polymers electrochromic window

and mirror area. 11 Improvement in the cyclability of the coloration/bleaching process and

the design and synthesis of new redox systems are current issues being addressed by

researchers.

Sensors and artificial muscles. Since conducting polymers change properties by

incorporation of ions and solvent, it is possible to develop ion-specific sensors based on

these materials. Properties changing upon the incorporation of ions or molecules into the

polymer matrix include conductivity, electrochemical potential, optical absorbance and

fluorescence. Sensing capabilities have been demonstrated for gases like SO2 and NO2

(PAni), alkali and alkaline-earth metal cations (through crown ether or polyalkyl ether









functionalization), as well as for more complicated molecules of biological interest

(glucose, urea, hemoglobin, etc). A thorough review comprising a wide range of

conjugated polymer-based chemical sensors was written by Swager et. al in 2000.72

Since conducting polymers show a swelling with increasing oxidation level, they

are able to convert electrical energy into mechanical work. The incorporation of counter

ions into the polymer leads to a structural change of the polymer backbone and to an

increase in volume up to 30%.73 These electromechanical properties are used in

actuators, commonly called polymer based artificial muscles. Starting with first polymer

actuator reported by MacDiarmid et al.,74 oxidation induced strain in PAni, PPy, and

recently in PEDOT has been investigated.63

There are several other interesting applications of conducting polymers such as

controlled drug release, corrosion protection, membrane and ion exchanger, lasers, etc.

that award the conducting polymer field the deserved importance. 1


1.6 Structure of this Thesis


The main characteristic (and beauty) of this work is the combination of

fundamental studies on the electrical and optical properties of conducting polymers from

the PXDOT and PXDOP class with more application-driven studies of electrochomic

devices with an operation window covering the visible and IR regions of the spectrum.

This work was extremely rewarding as it allowed probing of theoretical deductions

through practical application in devices. Multiple and content-varied projects as well as

several useful research collaborations made this work broad in the sense that it covers a

wide range of concepts defining the conducting polymer field. Experimental procedures









employed to acquire data presented throughout this dissertation are summarized in

Chapter 2.

Environmentally stable organic materials with metal-like conductivities are still the

focus of many researchers. Understanding the mechanism behind achieving a high

conductivity allows for an accurate design of such materials. Chapter 3 details the

mechanisms for electronic transport in conducting polymers along with a discussion

about the metal to insulator transition in these materials. Optical reflectance, microwave

and dc conductivity as well as EPR and X-ray measurements are employed in order to

gain insight into the electronic transport. Several PXDOT and PXDOP free-standing

films bearing various doping ions were analyzed. These materials' electronic properties

can be tuned from metallic to insulator by varying the dopant nature and monomer unit

planarity.

As many possible applications revolve around optical properties of conducting

polymer thin films, the work presented in Chapter 4 focuses on doping-induced electronic

structure and infrared active vibrational modes as a function of redox state in PXDOT

films deposited on ITO substrates. This chapter includes a detailed discussion on the

frequently debated origin of the multiple peaks or fine structure often seen in the

absorption spectra of neutral polymers. Polymers from the PXDOT family are the basis

set for this study.

Chapter 5 deals with the optimization of electrochromic devices that operate in the

reflective mode and are able to modulate the reflectivity of a metal surface in the visible,

NIR and mid-IR regions of the spectrum. Optical contrast, switching time and lifetime of









devices containing PEDOT or PProDOT-Me2 as the active layer are also reported in this

chapter.

The use of carefully designed complementary polymers is a promising route for

achieving a high degree of control over the color, brightness, switching speed and redox

stability of an electrochromic window. As such, the last chapter (Chapter 6) addresses the

optical and redox complementarity of polymer pairs required to attain a high contrast and

a long lifetime transmissive ECDs. In addition, a new way of achieving a broadband

response in an ECD by using bilayers of polymers having different absorption maxima

and comparable oxidation potentials is presented.














CHAPTER 2
EXPERIMENTAL TECHNIQUES


The intent of this chapter is to provide necessary background on the most common

experimental techniques employed in the characterization of conducting polymers. These

methods will be frequently referred to throughout the subsequent chapters.


2.1 Conductivity Measurements


Among available conductivity techniques, four probe methods have several

advantages for measuring electrical properties of conducting polymers. First, four probe

techniques eliminate errors caused by contact resistance, since the two contacts

measuring the voltage drop are different from the contacts applying the current across the

sample. Second, this technique allows for conductivity measurements over a broad range

of applied currents, usually varying between 1 ptA and ImA for conducting polymers

studied in this work. These current values produce potential differences ranging from 10

[tV to 10 V, depending on the resistance and thickness of the sample.

Polymers for conductivity measurements are usually prepared as free-standing

films, although thin films deposited on non-conducting substrates can be easily analyzed.

In the latter case, a soluble polymer is solution cast or deposited by spin coating on a

glass microscope slide. However, many conducting polymers are insoluble and infusible,

and in addition, some do not allow the synthesis of high quality, thick free-standing films

(5 microns or more) that can be easily removed from the supporting electrode. Generally,









this happens for less conducting samples where the formation of oligomers in solution

predominates once the electrode is covered with a thin polymer film. An alternative

solution for these cases is the electropolymerization at low charge density to afford thin

films, followed by polymer removal from the polymerization support by adhesion with a

pressure sensitive adhesive tape.

The materials presented in Chapter 3 were prepared electrochemically on glassy

carbon or freshly polished titanium electrodes. For a rigorous surface cleaning, glassy

carbon electrodes were immersed in acetone and sonicated for 30 minutes, then washed

with deionized water and dried under nitrogen. A Pt flag or a polished stainless steel

sheet was used as the counter electrode. Free-standing films with thicknesses ranging

from 2 to 100 |tm were obtained by slow galvanostatic deposition at an applied current of

0.02 to 0.08 mA/cm2. The temperature was maintained at -50C by immersing the

electrochemical cell in a NaCl/ice bath, and at -450C by using a dry ice/ACN bath. Due

to the long polymerization time required to prepare thick films (24-48 hours), the

assembly electrochemical cell/cooling bath was kept in a cooler to maintain a constant

temperature over night. Generally, thick free-standing film synthesis requires a higher

monomer concentration than common electropolymerization of thin films on ITO

(10mM). Consequently, the films were prepared from 60 mM monomer in 0.1 M

electrolyte (TBAPF6, LiCF3SO3 or Li(CF3SO2)2N) dissolved in a low vapor pressure

solvent, usually propylene carbonate (PC), that acts as a plasticizer for the resulting film.

When the monomer had a reduced solubility in PC, the concentration was limited to 20

mM monomer/PC. After reaching the desired film thickness, the polymer-coated

electrodes were removed from the reaction medium and the wet films were detached









from the electrode using a razor blade. Subsequently, the films were washed in PC to

remove unreacted monomer and electrolyte salt, placed between two glass slides to

flatten the films, and dried under vacuum for 24 hours. Most likely, a small amount of PC

remains in the film after drying. Although ACN seems more suitable as a washing

solvent due to its high vapor pressure, polymers washed with ACN were brittle and less

shiny, probably owning to detrimental morphological changes (cracks, holes) created by

fast drying.

The morphology and conducting properties of the resulting polymer depend on

many experimental variables such as solvent nature, concentration of reagents,

temperature, cell geometry, nature and surface morphology of electrodes, and applied

electrical conditions. Due to the interdependence of many of these experimental

variables, electrosynthesis optimization constitutes a complex study. The influence each

of these factors exerts on the electrical properties of the resulting material is described in

detail in Section 3.5.

There are three types of four probe conductivity techniques that can be employed in

the study of conducting polymers: Van der Pauw,75 four-wire76 and four-point probe

(Signatone),77 and their use depends on the instrumentation available as well as sample

quality and geometry.

The Van der Pauw method is a four-point probe technique used for irregularly

shaped thin films and was first developed at Philips Laboratories in the Netherlands.75 It

consists of four conducting wires (Cu or Ni) attached with silver paste in four points A,

B, C, and D, located at the circumference of the sample (see Figure 2.1 A). This method












Side view


Top view


/AV
Z1 2) 3


/1y
t,/


4


//
w


Figure 2.1 Four-probe conductivity methods: A) Van der Pauw, B) four-point probe, C)
four-wire


' '"


/









affords good measurements if the contacts are sufficiently small, the sample is

homogeneous in thickness and does not have defects such as holes. If the contacts are

placed such as they form a square on the sample surface, then the conductivity value is

given by:

y = ln2 (TRt)-1 (2.1)

where R is the resistance of the sample, and t is the thickness. R is measured by

connecting contacts A and B to a constant current source, and C and D to a voltmeter.

R=AV/I (2.2)

where AVis the voltage drop and I is the current applied.

Another four-point probe method has been developed at Bell Laboratories in

1958.77 This method employs devices with predefined electrode geometry, such as the

Signatone S-301-4 available in our laboratories. Figure 2.1 B shows the simplest form of

a four-point probe measurement setup. A row of pointed electrodes touches the surface of

a polymer film taped or spin cast on an insulating substrate. A known current I is injected

at the electrode 1 and is collected at the electrode 4, while the potential difference AV

between contacts 2 and 3 is measured. For this arrangement, the volume resistivity (p) of

the sample is given by an equation derived by Valdes:78

p = 27nd AV/I (2.3)

The result is independent of the electrode contact area as long as the distance

between the points (d) is much greater than the film thickness (t). Another equation that

describes the resistivity of a sample of thickness t by taking into consideration the ratio of

sample thickness to the distance between contact points is given by:

p = (AV/I) (t 7t /ln2) F (2.4)









where F (t/d) is a correction factor that approaches unity as t approaches zero, and its

values have been calculated by Uhlir et. al. 79

This four-point probe method has several advantages that make it extensively

used in solid-state electronics laboratories. First, if the voltmeter used to measure AV has

a high impedance, then the conductivity measurement is independent of the contact

resistance. As polymeric materials have a high degree of disorder, it is expected to report

a conductivity value obtained by averaging multiple measurements in different areas of

the film. The four-point probe method allows for the contact points to be easily

repositioned in various areas of the film, thus allowing for several conductivity

measurements on the same sample.

When four point probe devices are not available or when the experimental setup

needs to be placed in a cryostat for low temperature conductivity measurements, a four-

probe technique (Electrodag method) can be used. In this method, leads can be directly

attached to the film using conducting silver paste, colloidal graphite (water or alcohol

dispersion) or vacuum evaporated metal. The sample should be cut in a rectangular form,

and for best results its length should be much larger than its width. Four thin wires such

as Cu or Au (for low temperature conductivity measurements) are attached to the film

such that the distance between contacts 2 and 3 (see Figure 2.1 C) is much larger that the

distance between contacts 1 and 4. Volume conductivity is calculated from the following

equation:


= 1/Rtw


(2.5)









where I is the distance between inner leads 2 and 3, t is the film thickness and w is the

width of the sample. 1, t and w can be measured with a micrometer unless w is too small

and requires the use of a profilometer.

Both the current source (Keithley 220) and the multimeter (Keithley 195 A) used

for low temperature conductivity data acquisition were computer controlled. The sample

with attached Au wires was immersed in Janis Dewar flask containing liquid He and the

temperature was controlled from 4.2 K to 300 K by using a LakeShore 82C Temperature

Controller and LakeShore DT500 Temperature Sensor.

Microwave dielectric constant and conductivity were measured using the same

temperature control as above. This method is called cavity perturbation method or

"wireless conductivity". In this work, a homemade cavity was used in combination with a

Hewlett-Packard 8350B microwave source providing a frequency of 6.5 GHz.

Often, information about conductivity at different doping levels of thin polymer

films is necessary, especially for the determination of conductivity onset and comparison

of conductivity in n- and p- doped states. This cannot be achieved easily by ex-situ

methods described above and an in-situ method is required. Along with the disadvantage

that this non-routine method requires a fairly complicated setup such as two potentiostats

and expensive, problematical to obtain interdigitated microelectrodes, it affords a

conductivity value called "pseudo-conductivity" that cannot be converted to volume

conductivity due to the lack of information about the thickness of the film deposited on

interdigitated electrodes. Conducting polymer in-situ conductivity data reported in

literature are usually compared to the standard P3MT conductivity (60 S/cm), by









assuming that the all the variables affecting film conductivity are similar dopantt,

thickness, electrode geometry, etc.).9,80


2.2 Electrochemical Methods


The arsenal of electrochemical methods that can be applied to the study of

conducting polymer films deposited on a conducting surface is fairly broad and it has

been thoroughly reviewed by Doblhofer et al.81 Among these methods, cyclic

voltammetry (CV) has becoming increasingly popular as a mean to study redox states,

due to its simplicity and versatility. The electrode potential at which a polymer undergoes

reduction or oxidation can be rapidly located by CV. Furthermore, CV reveals

information regarding the stability of the product during multiple redox cycles. Since the

rate of potential scan is variable, both fast and slow reactions can be followed. A very

important aspect of this method is its ability to generate a new redox species during the

first potential scan and then probe the fate of species on the second and subsequent scans.

Therefore CV allows the growth of a polymer film along with its further characterization

during a single experiment (see Figure 2.2).

Fundamentals of CV. This method consists of cycling the potential of an

electrode, which is immersed in an unstirred solution, and measuring the resulting current

at the working electrode. Therefore, the obtained voltammogram is a display of current

(vertical axis) vs. potential (horizontal axis). The reducing or oxidizing strength of the

WE is precisely controlled by the applied potential (Figure 2.2). Typically, the

polymerization of electron rich monomers starts at low potentials (a) where no redox









40-


0 0 0
S S n

200



SE(Epa, ipa)
E 10-


0- adb



-10
(Epo ipc)
I I I I I I II
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

Potential(V vs AgAg)


Figure 2.2 Cyclic voltammogram of BEDOT-B(OG)2 in 0.1 M TBAP/ACN at a scan rate
of 20 mV/s. a represents the beginning of the experiment, b is monomer oxidation
potential onset, c is monomer oxidation peak potential, d is the polymer reduction peak
potential, and e is the polymer oxidation peak potential

reactions occurs, followed by scanning in the anodic direction. At potential (b), the

electrode has sufficient oxidizing character to oxidize the monomer to its radical cation.

The anodic current increases rapidly (b-c) until the concentration of the monomer at the

electrode surface approaches zero, causing the current to peak (c), and then decay as the

solution surrounding the electrode is depleted of monomer. Monomer oxidation is

immediately followed by chemical coupling that affords oligomers in the vicinity of the

electrode. Once these oligomers reach a certain length, they precipitate onto the electrode

surface where the chains can continue to grow in length. The electroactivity of the









polymer deposited onto the WE can be monitored by the appearance of a peak

corresponding to the reduction of the oxidized polymer while scanning in the cathodic

direction (d). A second positive scan reveals another oxidation peak at a lower potential

than the monomer oxidation peak (e). This is due to the neutral polymer now becoming

oxidized. Another noticeable fact is the increase in monomer oxidation peak current in

the second and subsequent scans. As the peak current is directly proportional with the

electrode area (Randles-Sevcik equation),82 this increase in the peak current could be

attributed to an increase of the WE area due to a more porous morphology of the

electrodeposited polymer.

Important parameters of a polymer cyclic voltammogram are the scan rate,

switching potentials, as well as the magnitudes of the anodic peak current (ipa), cathodic

peak current (ipc), anodic peak potential (Epa) and cathodic peak potential (Epc).

Continuous deposition of the polymer onto the WE can be monitored by the increase in

the polymer's anodic and cathodic peak currents, while the polymer redox properties are

characterized by the magnitudes of its peak potentials.

Electrochemical cells and electrodes. Modern potentiostats, such as Perkin Elmer

PAR 273 A available in our laboratories, utilize a three-electrode configuration. The

potentiostat applies the desired potential between a working electrode (WE) and a

reference electrode (RE). WE is the electrode at which the electrolysis of interest takes

place. An auxiliary (counter) electrode (CE) provides the current required to sustain

redox processes developing at the working electrode. This arrangement prevents large

currents from passing through the reference electrode, which could change its potential.

Typical electrochemical cell consists of a glass container with a cap having holes for









introducing electrodes and argon (see Figure 2.3). Oxygen removal from the reaction

medium is carried out by bubbling argon prior to electropolymerization, whereas

maintaining the cell oxygen-free during an experiment is accomplished by passing argon

over the solution.




1 1


to Ar


Figure 2.3 Electrochemical cell setup

There is a large variety of REs that are commercially available. For use with

aqueous electrolytes a saturated calomel electrode (SCE) or an Ag/AgCl/sat'd KC1

electrode can be used. In non-aqueous solvents, a silver wire in contact with 0.1 M

AgNO3 dissolved in a particular solvent (eg., ACN) is used and denoted as Ag/Ag

reference. Ag wire can be immersed directly in the reaction medium, but its potential








changes with the time (pseudo reference) and consequently frequent calibrations with a

solution of ferrocene/ferrocinium (Fc/Fc ) are required. Figure 2.4 illustrates the

relationship of these REs to the standard hydrogen electrode (SHE), which is arbitrarily

assigned a value of 0.000 V on the redox potential scale.

The auxiliary or counter electrode (CE) is usually a Pt foil or gauze with a

relatively larger area as compared to the WE. As with WE materials, the common

practice is to use electrochemically inert materials such as the noble metals Pt and Au, or

glassy carbon. These electrodes consist of a conductive rod of 0.02 cm2 diameter

encapsulated in a Teflon casing (Bioanalytical Systems, Inc.). While Pt button working

electrodes can be used for most CV experiments, Au has a lower oxidation potential and

it should not be used for experiments requiring potentials higher than 1.2 V vs. Ag/Ag+


Ag wire pseudo
reference


SHE Ag/AgCl SCE Ag/Ag+ Fc/Fc+

(0.000 V) (0.197V) (0.241 V) (0.499 V) (0.569 V)



V \VV V V V


0 0.1 0.2 0.3 0.4 0.5 0.6



Vvs. SHE
Figure 2.4 Electrode potential relationship between common reference electrodes









2.3 Spectroelectrochemistry and Other Optical Measurements


Spectroelectrochemistry plays a key role in probing the electronic structure of

conducting polymers as well as in examining the optical changes that occur upon doping.

It provides information about the material's band gap and intraband states created upon

doping as well as gives some insight into a polymer color through the location of the

absorption maxima and the ratio of peak intensities if the material shows fine structure on

the main 7t-7t* peak (see Section 4.3).

Measurements were carried out with a UV-Vis-NIR Varian Cary 5

spectrophotometer using a specially designed three-electrode cell to allow potential

application while monitoring the absorption/transmission spectra. Typical polymer

samples are thin films (100-500 nm) deposited potentiostatically on transparent ITO/glass

electrodes purchased from Delta Technologies, Ltd. These electrodes have surface

resistivities in the range of 10 to 20 Q/I and are cut to fit a spectroelectrochemical

cuvette (0.7 cm wide). Usually a Ag wire pseudo reference is used as a reference

electrode and a Pt wire is used as a counter. For potential control, all three electrodes

were connected to an EG&G PAR 273A potentiostat.

A typical spectroelectrochemical experiment is carried out with the

spectrophotometer operating in double beam mode. An accurate baseline is obtained

using two cuvettes containing bare ITO/glass electrodes immersed in a liquid electrolyte

placed in both reference and sample compartments. After collecting the baseline, the

polymer coated ITO is placed in the sample compartment. Polymer analysis is obtained

by sequentially stepping the applied potential until the polymer reaches its fully oxidized

and neutral forms. These extreme redox states are attained when the polymer absorption









envelope stops changing with increasing/decreasing applied potential. Usual potential

steps are between 50 and 100 mV. In order to ensure that a polymer film is fully

neutralized, chemical treatment with a solution of 85 % hydrazine in water follows the

electrochemical reduction.

Dual polymer transmissive devices presented in Chapter 6 are analyzed in a similar

manner. However, devices used to collect the baseline are assemblies consisting of two

bare ITOs sandwiched together with gel electrolyte. In order to apply voltage across the

device, the counter and the reference leads are connected to one another. The reference

and the sample holders used for single polymer spectroelectrochemistry do not have the

suitable geometry to hold devices, thus they have to be replaced with other supports

available in the Cary 5 kit. These holders are basically rectangular non-transmissive

sheets having a variable size hole to allow the transmitted light to reach the detector.

Devices are scotch-taped on these supports with the area of interest covering the holes.

A kinetics experiment allows for measuring polymer and device switching times

between the two extreme redox states. This measurement follows the

spectroelectrochemistry experiment described above because it requires knowledge of

polymer(device) potentials applied for a full switch as well as the wavelength of

maximum contrast. Once these values are known, the spectrophotometer can be switched

to transmission mode to monitor changes in a sample's opacity at the wavelength where

the optical contrast is the highest. Concurrently, a square potential waveform is applied at

desired time intervals (usually between 0.5-3 s). This allows monitoring % T as a

function of time at the specified wavelength, usually located in the visible region. The









rate at which the material is able to access both fully doped and neutral states is

determined by the time necessary to attain ca. 95% of its full contrast.

Long wavelength optical measurements described in Section 4.4 were performed in

the Physics Department at UF, using three different spectrophotometers: a Bruker 113v

Fourier Transform Infrared (FTIR) spectrometer for the region 400-5,000 cm-1, a Zeiss

MPM 800 microscope photometer for 4,500-45,000 cm-1 region, and a modified Perkin-

Elmer 16U for 3,700-45,000 cm-1 region. All measurements were performed at room

temperature and the reflectance data have been corrected using an Al mirror as the

reference.


2.4 Colorimetry


In-situ colorimetric analysis is rapidly becoming a popular technique in the study of

electrochromic polymers.83 This method allows for accurately reporting a quantitative

measure of the color and graphically representing the track of doping-induced color

changes of an electrochromic material or device. There are three attributes that are used

to describe the color: hue, saturation and brightness. Hue represents the wavelength of

maximum contrast (dominant wavelength) and is commonly referred to as color.

Saturation takes into consideration the purity (intensity) of a certain color, whereas the

third attribute, brightness, deals with the luminance of the material, which is the

transmittance of light through a sample as seen by the human eye.

A commonly used scale that numerically defines colors has been established in

1931 by The Commission Internationale de l'Eclairge (CIE system). This method takes

into consideration the response of a standard observer to various color stimuli, the nature










of the light source, and the light reflected by the object under study. These functions are

used to calculate tristimulus values (XYZ) that define the CIE color spaces. To simplify

the representation of a color, the tristimulus value Y represents the brightness of a

material, whereas X and Z are used to define its chromaticity, and they can be easily

represented on a two-dimensional graph called spectral locus (x-y chromaticity


diagram).84 This chromaticity diagram is a horseshoe shaped area where all visible light

wavelengths are represented (Figure 2.5). The location of a point on the chromaticity

diagram affords the hue and saturation of a color. From this graphical representation it is

possible to establish the dominant wavelength, or hue, by drawing a straight line from the

point of interest to the white point (W). Fully saturated and consequently pure colors lie

along the periphery of the horseshoe diagram.



0.9

0.8-

0.7-

0.6-

0.6-

0.4-

0.3-

0.2-

0.1

0.0-
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
X


Figure 2.5 x-y chromaticity diagram









Luminance analysis of a material or device is especially valuable in the study of

electrochromic polymers because it relates the transmittance over the entire visible region

as perceived by the human eye to the applied potential. Such measurements are very

useful for the characterization of transmissive/absorptive electrochromic windows

because a low luminance value corresponds to an opaque device while a high luminance

corresponds to a transparent window. Moreover, changes in a device luminance over time

can be used to quantify the device's stability to multiple redox switching. Such

experiments are presented in Chapter 6 and were performed by continuously stepping the

voltage such as the device would attain its extreme bleached and colored states with a

certain delay at each voltage allowing a full color change and hold period. During this

time, the decrease in the luminance of the device in the bleached state and the increase in

the brightness of the device's dark state were monitored over a several day period.

The experimental setup for colorimetric analysis is similar to the one used for

spectroelectrochemistry. The applied potential is controlled either by a Pine

Bipotentiostat model AFCBP1 or an EG&G 273 potentiostat. The polymer coated

ITO/cuvette assembly is placed in a black painted light booth chromaticityy box) that has

a light source D50 (5000 K) located in the back. Y, x and y data at different polymer

doping levels are recorded with a Minolta CS-100 Chroma Meter colorimeter. In order to

obtain accurate values, a background measurement on a bare ITO/cuvette assembly was

taken either at the beginning or at the end of the polymer colorimetric analysis.


2.5 Surface Analysis


Both atomic force microscopy (AFM) and profilometry techniques give valuable

information about polymer thickness and surface morphology. These contact techniques









have similar principles of operation involving a stylus or cantilever that moves across a

sample surface. This gives rise to a response signal that is further transmitted to a detector

and converted to a digital signal. Although profilometry technique is fairly easy to

operate, AFM allows for a much better resolution of the sample surface.

Profilometry. This technique is extensively used by our group for measuring

polymer film thicknesses and employs a Dektak 3030 Surface Profile Measuring System

purchased from Vecco Instruments, Inc. Measurements are made electromechanically by

moving the sample beneath a diamond-tipped stylus. A high precision stage moves the

sample beneath the stylus according to a user-programmed scan length, speed and stylus

force. As the stage moves, the stylus rides over the sample surface causing the stylus to

rise vertically and produce electrical signals. These response signals are transmitted to a

Linear Variable Differential Transformer mechanically coupled to the stylus that

transforms electrical signals in digitized signals and further forwards them to a computer

for data processing and storage. In addition to routine step height measurements, this

technique allows for the measurement of average surface roughness, average height, and

maximum height.

A routine experiment performed in our group is relating the polymer film thickness

to the amount of charge passed during its electrodeposition. This experiment is carried

out for almost every newly synthesized polymer, and if carefully performed, it represents

a good calibration curve for future use. For this study, polymers are electrodeposited

potentiostatically on ITO/glass electrodes at several charge densities usually ranging from

5 mC/cm2 to 200 mC/cm2. Upon deposition, the polymer films were carefully washed

with ACN and dried under vacuum for 24 hours to ensure complete solvent evaporation.









Film thicknesses were measured after scratching the film with a razor blade in several

small areas to expose the ITO substrate. The depth of the scratch is measured from the

cross-section view of the topographic image by placing one cursor at the bottom of the

scratch (ITO surface) and the other one at the polymer surface. As the polymer films are

often inhomogeneous, multiple measurements in different areas of the sample to give

average thickness values were used. Some polymer surfaces are very rough making it

difficult to measure their thickness through a simple step height measurement. In these

cases, average height method gives very reproducible results and consists of leveling the

ITO surface (the scratch bottom) at zero nm and placing the cursors as far apart on the

polymer surface as possible.

AFM. This contact technique is based on a flexible cantilever with a very low

spring constant that induces forces smaller than the interatomic forces, thus the

topography of the sample can be monitored without displacing the atoms. Several

techniques for sensing the deflection of the cantilever have been investigated and

developed. Most AFM used today rely on optical techniques to sense the deflection of the

cantilever as a response of surface height variation. For example, the Nanoscope III from

Digital Instruments used in this work relies on a beam from a laser diode focused onto the

back of the cantilever. The beam reflects off the back of the cantilever onto a segmented

photodiode. The amplified differential signal between the upper and lower photodiodes

provides a sensitive measure of the cantilever deflection.


2.6 Device Construction


Transmissive/absorptive window. The construction of a transmissive type ECD

consists of two thin polymer films deposited on transparent indium tin oxide coated glass









(ITO), and separated by a viscous gel electrolyte. Polymer films used to assemble a

device were obtained by constant potential oxidative polymerization at potentials slightly

higher than the monomer oxidation onset to ensure a slow rate and uniform film

deposition. Film thicknesses ranging from 100-300 nm were controlled by monitoring the

total charge density passed during electrosynthesis. Polymer-coated electrodes were

removed from the polymerization medium and placed in monomer-free electrolyte. The

films were electrochemically conditioned by sweeping the potential between -0.7 and

+0.7 V vs. Ag/Ag for about 15 minutes to guarantee that a maximum doping level, and

therefore a high contrast, would be attained. Cathodically coloring films were fully

oxidized and anodically coloring polymers were fully neutralized before rinsing with

ACN, to ensure charge balance prior to device assembly. The films were then coated with

gel electrolyte casting solution until the entire polymer surface was uniformly covered.

The cathodically and anodically coloring electrodes selected were then applied to one

another and allowed to dry for 24 hours. The gel electrolyte formed a seal around the

edges, the devices becoming self-encapsulated.

The device construction is carried out with one polymer oxidatively doped while

the other is neutral, and both films are simultaneously in either their transmissive or

absorptive states. As such, the device is observed as bleached or colored.

Complementary polymers' thicknesses influence to a large extent the operation of the

device as the ability to match the number of redox sites in each film allows for attaining

its extreme color states. Therefore, each polymer used in these devices should have a

reliable thickness vs. charge density calibration curve described in Section 2.5.









Variable reflectance mirrors. As a device operating in the reflective mode we

have used an outward facing active electrode device sandwich structure consisting of five

layers. In this construction, two gold-coated Mylar sheets are used as both counter and

working electrodes. The top electrode is cut with a series of parallel slits of varied

separation (0.5 to 2 mm) across the active surface making it porous to ion transport

during switching. Both the active top and the counter polymer films were

electrochemically deposited on the gold-coated Mylar electrodes from solutions of 10

mM monomer in 0.1M LiC104 in PC at constant potential. Although ACN can be used,

the Mylar support swells and curls during an extended polymerization time.

Polymerization on gold-coated Mylar electrodes proceeds much faster than the deposition

on ITO-coated glass due to an enhanced conductivity of the gold layer. In addition, it

leads to a very uniform film owing to the excellent homogeneity of the electrode surface.

Following the polymer films' redox conditioning, the counter polymer was fully

neutralized while the active polymer was fully oxidized to ensure a charge balance prior

to device construction. A separator paper, soaked in electrolyte, was used to isolate the

back of the working electrode from the counter polymer layer. The top layer is in contact

with a window, which is transmissive to the wavelengths of interest, allowing accurate

measurements of the active layer reflectivity. We typically used ZnSe for NIR to mid-IR,

glass in the NIR and visible, and polyethylene for visible through mid-IR. As only the

outward facing electroactive polymer is responsible for the surface reflectivity

modulation, the counter electrode polymer optical properties do not affect the device

operation. Therefore, the counter polymer layer thickness was typically twice as much as

the active polymer layer. The cell was assembled using a high viscosity polymeric









electrolyte composed of LiC104 or Li(CF3SO2)2N dissolved in an acetonitrile (ACN)

/propylene carbonate (PC) swollen poly(methyl methacrylate) (PMMA) matrix described

below. Device construction begins by placing the neutral polymer-coated back electrode

on a poly(ethylene) support, followed by the application of a thin and uniform layer of

gel electrolyte on the polymer surface. The separator paper is then gently pressed on the

counter polymer/gel surface with care taken not to scratch the polymer layer (neutral

polymers are very soft!) and then another layer of gel is applied on top of the paper. The

active polymer-coated electrode is placed on top of the separator paper, followed by the

application of a third layer of gel and finally, the top window is placed on the active

polymer surface.

For several purposes the device can be encapsulated in a polyethylene bag. Cu

wires are attached to the gold electrodes using a conductive silver paste, then the whole

device assembly is placed in a especially-designed polyethylene bag that enables the

device to be evacuated prior sealing it. The bag is punctured to take out the Cu leads for

electrical contact purposes and then the area around the leads is sealed using an epoxy-

based glue. Initially, three of the sides are heat-sealed while the fourth side is used to

evacuate the device and then sealed. First, this type of encapsulation provides an oxygen

free environment, thus increasing the lifetime of the device. Second, it permits the use of

water-free electrolytes, and more importantly of liquid electrolytes. The ability to use

liquid electrolytes is a major advantage over the former device assembly, as the

polymeric component (such as PMMA) included to toughen the electrolyte can be now

eliminated.









Gel electrolyte preparation. A gel electrolyte based on PMMA and

Li[N(CF3SO3)2] was plasticized by PC to form a highly transparent and conductive gel.

To allow an easy mixing of the gel components, we included ACN as a high vapor

pressure solvent. The composition of the casting solution by weight ratio of

ACN:PC:PMMA:Li[N(CF3SO3)2] was 70:20:7:3. Gel preparation begins by dissolving

the electrolyte salt in ACN, followed by the addition of PMMA. This polymeric

component is not easily dissolved, so vigorous stirring and mild heating (600 C) for a

period of about two hours is required. When all of the PMMA has dissolved, PC was

introduced to the reaction medium. The gel was stirred and heated on a hot plate for

about two more hours until it reached a honey-like consistency and started to stick to the

container walls. Preparing the gel in a dry box environment affords a water free

electrolyte.

Besides acting as an ion transport material, this type of gel electrolyte provides a

relative encapsulation of an ECD. At the edges of the device, the ACN in the electrolyte

evaporates, leaving behind the PMMA and electrolyte salt in PC. As the PMMA becomes

insoluble, it seals the outer edges of the device and provides self-encapsulation. The use

of this electrolyte minimizes further solvent evaporation, prevents leaking, and allows for

long-term testing of the ECD.


2.7 Purification of Laboratory Chemicals and Materials


Since electrochemistry is a very sensitive technique, it is crucial that solvents and

electrolytes employed are pure. Reagent grade ACN and PC in Sure Seal bottles were

purchased from Aldrich. ACN was distilled over CaH2 and PC was percolated through









activated type 3A molecular sieves, followed by vacuum distillation (10 mm) and storage

in an Ar atmosphere. Tetrabutylammonium perchlorate (TBAP) was either purchased

from Aldrich and used as received, or prepared by mixing a 1:1 mole ratio of

tetrabutylammonium bromide dissolved in water with perchloric acid. The precipitate

was filtered, recrystallized from a 1:1 molar ratio ethanol and water and dried in the

vacuum oven for 24 hours at 600C. Tetrabutylammonium hexafluorophosphate

(TBAPF6-Fluka) and lithium trifluoromethansulfonate (LiCF3SO3-Aldrich) were

recrystallized from ethanol and dried in the vacuum oven for 24 hours at 600C. Lithium

bistrifluoromethanesulfonimide (Li(CF3S02)2N- 3M) and lithium perchlorate (LiC104-

Aldrich) were used as received.














CHAPTER 3
ELECTRONIC TRANSPORT AND OPTICAL PROPERTIES OF PXDOT AND
PXDOP FREE-STANDING FILMS


3.1 Introduction


Traditionally, polymers were thought of as insulators and any electrical conduction

in polymers was generally regarded as an undesirable phenomenon. In the last few

decades, an opposite trend has been started by the discovery of ionic conductivity in

polymers and its wide application to polymer electrolytes for power sources and

sensors.49,85,86 In addition to this and somewhat surprisingly, a new class of polymers

possessing high electronic conductivity in the doped form has been discovered in

1970s.1,2 These novel materials with interesting and unanticipated properties have

attracted the whole scientific community from synthetic chemists to theoretical

physicists.31,87,88 After 20 years of research, the fundamental nature of charge

propagation is still under debate. Electronic transport can be assumed to occur via an

electron exchange reaction (electron hopping) between neighboring redox sites in ion-

conducting polymers and by the motion of electrons through conjugated systems in the

case of electronically conducting polymers. Owing to the diversity and complexity of

these materials, such as chain and segmental motions, changes in morphology and slow

relaxation, much research is still needed to achieve a detailed understanding of all

processes related to the dynamic and static properties of several interacting molecules

confined in a polymer matrix. When debating the conduction mechanism in doped









conducting polymers, researchers encountered several difficulties such as unknown

molecular weight and morphology, nonuniform doping, instability of the polymer in the

doped form, and especially a high disorder of the material due to chain breaks and

defects. Theories of transport in amorphous conducting polymers have been advanced by

Davis and Mott (1970),89 Cohen (1969),90 Anderson (1958),91 Nagels (1970),92

Prigodin and Epstein (2001),93 Kaiser (2001),34 and some of them are presented in the

following section. This chapter details the mechanisms for electronic transport in

conducting polymers along with a discussion about the metal to insulator transition in

semi-crystalline polymers supported by optical, microwave, dc and EPR results.


3.2 Mechanisms for ElectronicTransport


Modified band levels. Conducting polymers fall into the broader category of

amorphous semiconductors and their density of states differ markedly from their

crystalline counterparts. The main feature of the energy distribution of the electronic

states density in amorphous solids is determined by their configurational disorder. This

causes fluctuations in the potential, leading to the formation of localized states. These

states are localized in the sense that an electron placed in a region will not diffuse at 0 K

to other regions, and various models presented here differ in the nature of localized states.

Figure 3.1 represents a schematic drawing of density of states showing two new

features, as compared to crystalline semiconductors. The first is the tail of localized

states, which are narrow and extend into the forbidden gap. Second there is a band of

compensated levels near the middle of the gap, originating from defects such as vacancies

or twists in the polymer backbone. A refinement of the above model is the splitting off









from the tail states of various localized states, lying at certain energies within the gap

(Figure 3.1.C). This mechanism fits better for real disordered semiconductors and

explains luminescence and photoconductivity in conducting polymers.30


Figure 3.1 Schematic drawing of density of states: A) band with tail or localized states
and a Fermi level near the middle of the gap; B) the center band split into acceptor and
donor bands; C) "real" disordered semiconductor with defect states (from Chien30)


E, E


Ev EC









Conduction by thermal activation to extended states. This model implies that

the charge carriers are bound and there is a minimum temperature at which the materials

become conducting. Conductivity in this regime would be more appropriately considered

as Brownian motion. The lowest value of the electrical conductivity before the start of an

activation process is called by Mott94 the "minimum metallic conductivity" and is given

by:

Cmin=const e2/h a (3.1)

where a represents the interatomic distance. The constant has a value of about

0.026 and omin is in the range of 200-300 S/cm.

Conduction by activated hopping in band tails. If the charge carriers are

localized, conduction can occur by hopping between states in the band tails. An electron

moves from one localized state to another with the energy provided by exchanging with a

phonon, and the charge carrier mobility follows Arrhenius law.95

Variable-range hopping conduction. This model fits most of the conducting

polymers studied worldwide, especially the ones belonging to the insulator side of metal-

insulator transition (M-I transition) (described in detail in Section 3.3). Consider a

semiconductor with strong interaction between the charge carrier and the atom. Each

wave function is confined to a small region of the space, falling off exponentially with

the distance. These materials are called Fermi glasses. When Fermi energy lies in the

range of energies for localized states, variable-range hopping is possible. For the case of

strong localization, hopping will be only between nearest neighbors. Conductivity of

materials in this category is given by:

Gdc= O(T)exp[-const/T1/(+d) ] (3.2)









where d is the dimensionality of hopping.94 For most polymeric materials the electronic

conduction is 3-D, therefore a linear logo vs. T-1/4 dependence has been widely used as

evidence for transport via variable range hopping.96

Fluctuation-induced tunneling conduction (Heterogeneous Model). This model

arises from the study of composite materials based on conducting particles embedded in

an insulating matrix, which seems to be particularly applicable to conducting

polymers.97 First, most polymers seem to have a fibrilar morphology98 (see Figure 3.2)

that is not continuous. Second, there is vast evidence that doping can be highly

inhomogeneous, consequently creating tiny metallic domains separated by insulation

regions. In these materials, the carriers are itinerant and free to move over distances very

large compared to their atomic dimensions. For such polymers, the electrical conduction

may be dominated by carrier transport between conducting domains rather than by

hopping between localized states. The presumption is that the carriers tend to tunnel

between conducting regions at points of closest approach and the conducting domains

vary in shape.

Well-defined crystalline regions surrounded by amorphous material are evident in

Figure 3.2 (taken from Percec et a/.98). Similar boundaries in conducting polymers

dominate the resistance of a sample since the charge carriers have to pass through the

"barriers" surrounding the crystallites. The significance of heterogeneity for electronic

transport in PAc was recognized in the early work of Park et al.,99 who proposed a

barrier resistance in series with highly conducting crystalline regions. Studies by Travers

et al. 100 on the evolution of transport properties function of aging provide evidence that

heterogeneous disorder plays a key role in conducting polymers, with the barrier regions























0 rn 400


Figure 3.2 Scanning force microscopy image of polymer chains on a surface showing
separate regions of well-aligned chains (from Percec98)

around conducting grains being broadened by aging. A related inhomogeneouss

localization" model for PAni proposed by Prigodin and Epstein,35,93 involves less

distinct boundaries between crystalline and disordered regions. This model takes into

consideration a detailed description of chain arrangement in conducting polymers.

Structural studies showed that there are crystalline regions where the polymer chains are

regularly and densely packed. Outside these regions, the chain arrangement is less

ordered, and here, the chains form amorphous media. The crystalline metallic islands are

coupled into the network with twisted and entangled polymer chains (see Figure 3.3).

Electrons move primarily along the polymer chain and sometimes hop between

neighboring chains. In the metallic islands, there is a good chain overlap and the charge

carriers are delocalized over the entire grain volume. Single polymer chains that pass

through disordered, amorphous media provide connections between metallic grains.

Therefore the electric connection is not only between the neighboring grains, but between

distant ones as well. Depending on the route of synthesis, the crystalline region varies









from a few percent to 50%. The standard percolation analysisl1 predicts that 30 %

crystallinity provides metallic behavior, but several polymer samples with 50 %

crystallinity are still in the insulator regime. The Epstein model allows the interpretation

of frequency dependence of electromagnetic response in conducting polymers, including

the microwave region of spectrum.96,102 Kaiser et al. has applied a refinement to this

model to introduce charge carriers being thermally activated over thin barriers

(dominating at high temperatures) or tunneling through the barriers (dominating at low

temperatures).34


Figure 3.3 Schematic drawing on the conducting polymer structure. The lines represent
the polymer chains and the dashed squares mark the regions where polymer chains
demonstrate crystalline order. (from Prigodinl02)








3.3 Elementary Excitations in Conducting Polymers

Solitons. The localized electronic state associated with the soliton is a nonbonding

state at an energy lying at the middle of the 7t-t* gap, between the bonding and

antibonding levels of the perfect chain. The soliton is a defect both topological and

mobile because of the translational symmetry of the chain.53 Soliton model was first

proposed for degenerated conducting polymers (PAc in particular) and it was noted for its

extremely one dimensional character, each soliton being confined to one polymer chain

(see Figure 3.4). Thus there was no conduction via interchain hopping. Furthermore,

solitons are very susceptible to disorder, and any defect such as impurities, twists, chain

ends or crosslinks will localize them. 103,104



A)
A) D/VVDV\
D- + D-

Neutral solitons in PAc Positive solitons in PAc



CB CB

B)


VB VB


Figure 3.4 Band diagrams for neutral and positive solitons. A) Schematic representation
of neutral (left) and positive (right) solitons in degenerate PAc, where D represents a
dopant ion; B) band diagrams for neutral (left) and positive (right) solitons with
associated electronic transitions.









Polarons and bipolarons. The polaron can be viewed as a bound state of a charged

and a neutral soliton whose mid-gap energy states hybridize to form bonding and anti-

bonding levels. The neutral soliton contributes no charge and a single spin and the

charged soliton possesses charge and no spin. The resulting positive or negative polaron

is a radical cation or anion, respectively, a particle consisting of a single electronic charge

leading to local geometrical relaxation of the bond lengths, as the polymer chain passes

from the neutral benzoid form to a partially-quinoid structure (Figure 3.5). Optical and

EPR studies indicate that polarons are the main charge carriers when a material is lightly

doped.51,105


CB
A


A) Eg (7-*)



VB


0 0

B) Z Neu
0 0 0 0

Neutral


CB CB

02


SP BP

VB VB

00 0
0 o

+ SB
0 0 0 0 0 0 0 0


Polaron Bipolaron


Figure 3.5 Polaron and bipolarons in non-degenerate ground state polymers: A) band
diagrams for neutral (left), positive polaron (center) and positive bipolaron (right); B)
neutral (left), lightly doped (center) and heavily doped (right) PEDOT structures









Similarly, a bipolaron is a bound state of two charged solitons of the same charge

with corresponding mid-gap levels, as illustrated in Figure 3.5. Since each charged

soliton carries a single electronic charge and no spin, the positive (negative) bipolaron is

a spinless dication (dianion). This represents a doubly charged state of two polarons

brought together by the overlap of a common lattice distortion (enhanced geometrical

relaxation of the bond lengths due to the formation of fully quinoid structure at high

doping levels). 105


3.4 Metal-Insulator Transition in Conducting Polymers


As the extent of disorder increases in a metallic system, the mean free path of an

electron becomes equal with the interatomic spacing and consequently some of the

carriers become localized, the material becoming insulator.28 The metal-insulator

transition (M-I transition) has been proposed by loffe and Regel106 and is defined as:

EF/ =1 (3.3)

where EF is the electron wavenumber and I is the mean free path.

Based on the above equation, Mottl07,108 proposed that a metal-insulator

transition occurs when the disorder is sufficiently large that EF/
becomes a "Fermi glass" insulator (term described in Section 3.2). In recognition of

Anderson's early work on disorder induced electron localization, Mott called the M-I

transition the "Anderson transition". The M-I transition in conducting polymers is

especially interesting; critical behavior has been observed over a relatively wide

temperature range in a number of systems including PAc, PAni, PPy, PPV,109 and more

recently PEDOT and its derivatives,36-41,110-112 which are the focus of this work. In









each case, the metallic, critical and insulating regimes have been identified. The critical

regime is tunable in conducting polymers by a multitude of methods including varying

the synthetic path, the dopant, via monomer derivatization or by applying external

pressure and/or magnetic field. However, the truly metallic regime with EFI >1 has not

yet been achieved.

Superconductivity. Experimental studies have demonstrated that for conducting

polymers, the electrical conductivity and mechanical properties improve together, as the

degree of chain extension and chain alignment are increased. Materials become truly

metallic if there is a high probability that an electron will traverse to a neighboring chain

before traveling between defects on a single chain.

Schon et al. recently claimed the first organic polymer superconductor.113 This

gate-induced superconductivity in PT films had granular character, consistent with the

heterogeneous model described above. The films of poly(3-hexyl thiophene) used

possessed low disorder and high charge-carrier mobility as a result of self-assembly of

polymer side chains. Application of a voltage varied the carrier concentration over a wide

range, the conductivity varying from semiconducting to metallic. Superconductivity was

observed below 2.35 K for samples with room temperature conductivities of about 80

S/cm. The researchers suggested that the superconductivity originated in metallic islands

with dimensions of about 10 nm corresponding to well-ordered nano-crystalline regions

in the polymer, emphasizing again the key role of heterogeneity in charge transport in

conducting polymers.









3.5 Electrochemical Synthesis of Free-Standing Films


Electrochemistry has played a significant role in the preparation and

characterization of these novel materials. Electrochemical techniques are especially

suitable for controlled synthesis and for tuning of well-defined oxidation states. 12 The

electropolymerization of thiophene derivatives involves many experimental variables

such as solvent, concentration of reagents, temperature, cell geometry, nature of

electrodes and applied electrical conditions. As a consequence of the diversity of these

parameters, electrosynthesis conditions determine to a large extent the morphology and

conducting properties of the resulting polymer. However, due to the interdependence of

many of the experimental variables, electrosynthesis optimization constitutes a complex

study. The results for PXDOTs and PXDOPs obtained in various conditions are presented

in Table 3.1.

The solvent used must simultaneously present a high dielectric constant to ensure

the ionic conductivity of the electrolytic medium, and a good electrochemical resistance

against decomposition at the potentials required to oxidize the monomer. Until now, the

most conductive PTs and PT derivatives obtained have been prepared in rigorously

anhydrous aprotic solvents of high dielectric constant and low nucleophilicity such as

acetonitrile, benzonitrile, nitrobenzene and propylene carbonate.9 When used for

PEDOT synthesis, ACN led generally to powdery deposits or to brittle films with

conductivities typically comprised between 5-20 S/cm. PC allowed for the synthesis of

compact and shiny free-standing PEDOT films with conductivities exceeding 100 S/cm.

Doped polymers are electrosynthesized in the presence of small anions derived

from strong acids such as C104-, PF6-, BF4-, CF3SO3, associated with lithium or tetraalkyl









ammonium cations. The nature of these dopants strongly affects the morphology and the

electrical properties of the resulting polymer. As seen in Table 3.1, PEDOT films

prepared in the presence of PF6 show higher conductivities than the ones prepared with

other dopants. Probably, PF6- facilitates a more compact packing of the polymer chains,

thus increasing the interchain transport contribution to the overall conductivity. 112



Table 3.1 Electrochemical synthesis of free-standing PXDOT films.

Polymer Dopant Electrode Temperature Current Conductivity
Material Density (S/cm)
(mA/cm2)
PEDOT PF6 Glassy -450C 0.08 110
carbon
PEDOT PF6 Glassy -450C 0.06 110
carbon
PEDOT PF6 Glassy -450C 0.04 330
carbon
PEDOT PF6 Glassy -450C 0.02 74
carbon
PEDOT PF6 Glassy 200C 0.04 10
carbon
PEDOT PF6 Glassy -50C 0.04 90
carbon
PProDOT PF6 Glassy -450C 0.04 66
carbon
PEDOT CF3SO3- Glassy -450C 0.04 95
carbon
PProDOT CF3SO3- Glassy -450C 0.04 38
carbon
PEDOP CF3SO3- Ti -50C 0.04 40

PEDOT (CF3SO2)2N- Glassy -450C 0.04 8
carbon


The electropolymerization temperature has a marked effect on the optical and

electrical properties of the resulting material, the films produced at -450C having









noticeably higher conductivities than the films produced at higher temperatures. This is

due to the fact that the activation energy of undesirable reactions such as termination,

kinks in the backbone, etc., is higher than the activation energy for the polymerization

reaction, thus films produced at low temperatures have a longer mean conjugation length

and consequently higher conductivities.

The anode material is of crucial consideration since the physical and chemical

properties of the polymerization surface determine the nature and strength of the bond

between the polymer and the electrode, which can affect both the deposition process and

the properties of the resulting material. Moreover, the formation of a conjugation defect

in the early stages of the synthesis (at the anode surface) has much more dramatic

consequences for the overall stereoregularity and stacking order of the polymer chains

than the occurrence of the same defect at a later stage of the process.45 The highest

conductivity and best surface quality films were obtained using glassy carbon electrodes

that were extremely shiny, flat (at the atomic level) and of high purity. Other substrates

such as stainless steel and ITO/glass have been tried, but their surface quality did not

allow the synthesis of highly conducting films. Polymers from the pyrrole family did not

adhere well enough to glassy carbon, therefore freshly polished titanium electrodes were

used as deposition substrates.

Although it is likely that the potential applied during the electrosynthesis is the

critical electrical parameter, the most homogeneous and conducting films are generally

obtained in galvanostatic conditions.46 The current density used is a crucial factor in

obtaining high quality materials. If the current density is too high, the polymer deposition

proceeds too fast for the polymer chains to align, resulting in a material with a high









content of amorphous phase. When the current density is lower than its optimal value,

formation of semi-soluble oligomers in the close vicinity of anode is favored. Due to their

limited solubility and low reactivity, these oligomers insert into the polymer film without

chemical coupling, thus reducing the cohesion and conductivity of the materials. For

example, PEDOT films with conductivities as high as 310 S/cm have been obtained at a

current density of 0.04 mA/cm2, whereas films obtained at higher (0.06 mA/cm2) or

lower (0.01 mA/cm2) current densities exhibited much poorer electrical properties.


3.6 Temperature Dependence of the Conductivity


To determine the nature of the metallic state in conducting polymers, it is necessary

to use a wide variety of techniques. The easiest method to start identifying the various

regimes is to measure the dc conductivity over a broad range of temperatures. The

transition from metallic to insulating behavior is generally accompanied by a marked

change in the temperature dependence of the dc conductivity. In a semiconductor, or

insulator, the electrical conductivity increases with the temperature, due to an increased

number of the charge carriers thermally excited across the band gap that are able to

contribute to the conductivity. The conductivity of a metal decreases with increasing

temperature, since the mobility of charge carriers decreases due to an enhancement in

electron-lattice scattering. Since the room temperature conductivity of heavily doped

conducting polymers are frequently comparable to Mott's minimum metallic

conductivity, disorder is the dominant factor in determining the temperature dependence

of various transport regimes. Table 3.2 summarizes the room temperature conductivities

[o(300K)] and the conductivity ratios [o(10K)/o(300K)] of several PXDOT samples.









The conductivity ratio is a useful empirical parameter to quantify the extent of disorder

and for sorting out various regimes. In general, as disorder increases, the materials

become more insulating and the conductivity decreases more rapidly upon lowering the

temperature.



Table 3.2 Transport properties of PXDOT and PXDOP free-standing films


Doped Gdc o(10K)/ Smw(T) XPauli (Omax for W-plot regime
Polymer (S/cm) o(300K) (300K) contribution (c0), slope
(300K) to x(T) cm1
PEDOT- 330 0.42 negative major 1900 positive metallic
PF6
PProDOT- 66 0.033 positive noticeable negative insulating
PF6
PEDOT- 95 0.17 negative major 2100 small critical/
CF3SO3 positive metallic
PProDOT- 38 0.074 positive noticeable 2300 negative insulating
CF3SO3
PEDOP- 40 0.12 negative small critical
CF3SO3 negative
PEDOT- 8.5 0.002 positive negative insulating
(CF3SO2)2N


From the data presented in Table 3.2, the most metallic sample is PEDOT doped

with PF6- anions, sample that shows also the highest room temperature conductivity.

PEDOT-CF3SO3 exhibits a slightly lower conductivity ratio, leading to the conclusion

that doping with PF6- allows for more ordered film morphology and a closer packing of

the polymer chains. The electronic properties of the polymers can also be fine tuned by

derivatization. Comparing the polymers of the thiophene family, PEDOT shows a more

metallic behavior than PProDOT, due to a planar structure of the monomer unit. The

propylene-dioxy ring in PProDOT is twisted and its morphology is more open, hindering

to some extent the interchain hopping contribution to the conductivity. EDOP planar









structure would recommend its use in the synthesis of polymers with a high degree of

order. Unfortunately, we do not have temperature dependence of the conductivity

measurements for PEDOP-PF6. However, PEDOP-CF3SO3 has a similar conductivity

ratio as PEDOT containing the same dopant, with slightly lower room temperature

conductivity. This is probably due to a less environmentally stable oxidized form of the

dioxypyrrole polymer.

To explicitly describe the characteristic behavior of temperature dependence of the

conductivity, we define the reduced activation energy as the logarithmic derivative of the

material resistivity (p):

W = d In p/d In T (3.4)

The temperature dependence of W for the metallic materials is positive. In the

critical regime, W is temperature-independent or shows very weak temperature

dependence, whereas in the insulating regime W has a negative temperature coefficient.

Wvs. temperature coefficients for PXDOTs and PXDOPs are summarized in Table 3.2.

Figure 3.6 represents the temperature dependence of the conductivity in PXDOT

films. As noted before, the conductivity of PEDOT-PF6 does not decrease too much at

low temperatures, maintaining a finite value (about half of its room temperature

conductivity value) as the temperature approaches 0 K. This suggests that the sample is

on the metallic side of the M-I transition. Changing the dopant to CF3SO3 or

(CF3SO2)2N- leads to more disordered materials with a stronger temperature dependence

of the conductivity. PEDOT-(CF3SO2)2N exhibits typical semiconductor behavior, with

no conductivity as the temperature approaches 0 K, suggesting that this material is on the

insulator side of the M-I transition.











1-






0.1




I--

S0.01 m PEDOT-PF6
PEDOT-CF3SO3
PEDOT-(CF3SO2)2N
V PProDOT-PF6



0 50 100 150 200 250 300
T(K)


Figure 3.6 Temperature dependence of the conductivity in PXDOT free-standing films

In order to understand the nature of the transport mechanism in these samples, we

fit the conductivity data to Mott's Variable Range Hopping (VRH) modelll4 described

by the Equation 3.2, for a three dimensional conduction (d=3). We noticed that 3-D VRH

fits very well for the samples on the insulator side of the ITM (see Figure 3.7). Therefore,

the charge carriers in PEDOT-(CF3SO2)2N and PProDOT-PF6 are mainly localized, and

they can only hop from a site to another in the 3-D space. Attempts to fit 1-D and 2-D

VHR with the conductivity data for PEDOT doped with CF3SO3- and PF6- failed as well,

leading to the conclusion that the conductivity of these samples is mainly due to

delocalized charged carriers. Probably, the heterogeneous model proposed by Epstein

(Section 3.2) will fit better in this case.93







71




1000- PEDOT-PF6
PEDOT-CF SO
PEDOT-(CF3SO22N
V PProDOT-PF6



100 -

C.
C"

I-
10




S I I II I I
0.2 0.3 0.4 0.5 0.6 0.7
T-1/4(K-1/4)


Figure 3.7 The conductivity dependence ofT-1/4 in PXDOT films



3.7 Microwave Experiments


The microwave conductivity (omw) and dielectric constants (Smw) were measured

using a cavity perturbation technique (at 6.5 GHz) in Dr. Epstein's laboratory at Ohio

State University. The Smw is a key probe of the charge carrier delocalization. For

delocalized electrons at frequencies lower than their plasma frequency, the dielectric

constant is negative, at all temperatures, due to the inertia of free carriers in an alternating

current field. The value of the negative Smw is limited by the relaxation rates and the

frequency. For a localized carrier, the charges stay in phase with the applied field and Smw


is positive at low frequencies.96










Figure 3.8 shows the temperature dependence of the smw for PXDOT samples. In

concordance with the dc results, PEDOT-CF3SO3 and PEDOT-PF6 show metallic

behavior, with dielectric constants that remain negative down to 10 K. The weaker

temperature dependence of Smw for PEDOT-PF6 (Emw (300 K)/smw (20 K) 2.7) than for

PEDOT-CF3SO3 (Emw (300 K)/Smw (20 K) 4.5) is consistent with a weaker temperature

dependence of Gdc. For PEDOT-(CF3SO2)2N and PProDOT-PF6, Smw stay positive over

the entire temperature range, suggesting once again that these samples are on the

insulator side of M-I transition. Conductivities obtained by the cavity perturbation

method have similar temperature dependence as direct current conductivities, and both

data sets are compared in Figure 3.9. However, for the samples exhibiting electronic

conduction by hopping, omw deviates slightly from odc at higher temperatures.



12000

8000

4000

0



-8000

-12000
PEDOT-PF6
-16000- PEDOT-CF3SO3
PEDOT-(CFSO2 82N
-20000- V PProDOT-PF6

0 50 100 150 200 250 300
T(K)


Figure 3.8 Temperature dependence of the dielectric constant in PXDOT films.







73




1- PEDOT-PF, 10 PEDOT-CFSO
10-
08-
A) 0 061


079-

06- 02-

05-
00-
0 00 50 200 250 300 50 100 150 200 250 300
0 50 100 150 200 250 300
T(K) T(K)




PProDOT-PF
100- PEDOT-(CF SO2)2N PPrDOT

100 -



C) r D) d"
SC)1



02 03 04 05 06 07
02 03 04 05 06 07
T 'I(K 1") T"(K'")

Figure 3.9 Temperature dependence of the mw and dc conductivity for A) PEDOT-PF6;
B) PEDOT-CF3SO3; C) PEDOT-(CF3SO2)2N; D) PProDOT-PF6.




3.8 Magnetic Susceptibility



The magnetic properties of localized unpaired electrons differ significantly from


their itinerant-electron counterparts. Total electron spin susceptibility (X) for doped


polymers is composed of Curie-type susceptibility (Xc), associated with localized, non-


interacting electron spins, and Pauli-type susceptibility (Xp).110,111 The latter represents


the susceptibility of delocalized electrons and is essentially independent of the


temperature.


X = XP + Xc (Xc = const/T) (3.5)










XT = XpT + const (3.6)

Figure 3.10 represents the temperature dependence of EPR susceptibility for three

PXDOT samples. Data for PProDOT-PF6 fit well a straight line, indicating the presence

of localized unpaired electrons as main contributors to the electronic transport. On the

other hand, PEDOT data shows a marked deviation from the linear behavior as the

temperature increases. This is due to a high contribution of Xp to the overall

susceptibility, thus the electronic transport in these materials occurs mainly via

delocalized charge carriers. The number of free electrons increases with increasing

deviation from linearity; therefore PEDOT-PF6 shows the most metallic character.

Hence, the susceptibility and the conductivity data are in qualitative agreement, placing

PProDOT-PF6 on the insulating side of the M-I transition, PEDOT-CF3SO3 in the critical

regime and PEDOT-PF6 in the metallic regime.


80

70

S60 PEDOT-PF6 'I

50 PEDOT-CF SO, II
4. 'PProDOT-PF6

230


10
0o j 5 I II I
0
0 9 100 150 200 250 300

Figure 3.10 EPR Susceptibility of Doped PXDOT Films









3.9 X-ray Measurements


The X-ray diffractograms presented here were measured by A. Saprigin at Ohio

State University. Peaks at -6.50, -12.50, -190, -250, -370 are clearly seen in the X-ray

diffractograms of all four PXDOT samples presented in Figure 3.11. Based on these

measurements, we propose a pseudo-orthorombic structural model with lattice parameters

a = 14 A, b = 14 A, c = 14 A. The first four peaks are very close in position to the

diffraction peaks of spin cast thin films of tosylate doped PEDOT reported in the

literature. 115 The presence of relatively sharp peaks in the X-ray diffraction pattern of

PXDOTs indicates the presence of crystalline grains embedded in amorphous media.

PEDOT -PF6 shows the sharpest peaks, thus having the highest content of crystallinity of

all analyzed samples, with the crystalline domain size of about 20 A. This fact is in

agreement with the conclusion that this material is on the metallic side of the M-I

transition due to a low amount of disorder. For the other samples, the crystalline domain

sizes are in the range of 15-20 A.


3.10 Optical Conductivity


Kramers-Kronig analysis of the reflectance data provides the optical conductivity

function, o(c). We can obtain information about the conduction electrons in PXDOTs by

comparing this function with the Drude expression for free electrons and the localization

models. For PXDOT samples, c0() demonstrates a intraband peak below 1 eV and a

weaker feature at 1.5 eV, attributed to an interband transition (Figure 3.12).112 The

c0() for PXDOTs differs from the Drude model, where CDrude(0) increases continuously

with decreasing frequency. As Gdc increases, the frequency at which the maximum in








c(Co) occurs (COmax) shifts monotonically to lower energies (Table 3.2). Since (max is

lowest for PEDOT-PF6, this indicates that the scattering time (mean free path) for this

sample is the longest, and therefore the material is less disordered. This is in agreement

with the transport and structural experiments presented in this chapter.


PProDOT-PF6
6



A) B)

PEDOT-CF SO
3 3ProDOT-CF3SO3




C) D)

10 20 30 40 10 20 30 40

20 20
Figure 3.11 X-ray diffraction for: a) PEDOT-PF6; b) PProDOT-PF6; c) PEDOT-CF3SO3;
d) PProDOT-CF3SO3

In conclusion, the studied materials have properties varying from metallic to

insulating depending on the nature of the dopant or the monomer unit planarity. Selected

transport data are summarized in Table 3.2. The most metallic sample, PEDOT-PF6, has a

high room temperature Gdc, weak temperature dependence of odc along with a positive

slope of W(T)= dln(Gdc(T))/dln(T) (reduced activation energy) at low T. Moreover, its

6.5 GHz dielectric response Smw stays negative as the temperature is varied. This behavior

is similar to previously reported data for metallic polyaniline and polypyrrole. EPR









measured magnetic susceptibility X(T) has a major contribution from Pauli component Xp

associated with delocalized spins.



1000
PEDOT-PF6

800 PEDOT-CF3SO,
800


600


400


200


0


100 1000 10000


Wavenumbers, cm-1


PProDOT-CF3SO3


PEDOT-CF3SO3


Figure 3-12 Optical properties of PXDOT films. A) Optical conductivity of PXDOT
films as calculated from Kramers-Kronig analysis; B) Photos with highly reflective free-
standing films









Stronger T dependence of odc for critical regime compositions (PEDOP-CF3SO3,

PEDOT-CF3SO3) results in a small positive or small negative W-plot slope at low T. For

PEDOT-CF3SO3, Smw is negative down to 4 K, demonstrating that this sample is still on

the metallic side of the M-I transition. Pauli contribution to the overall magnetic

susceptibility is less than for PEDOT-PF6.

PEDOT-(CF3SO2)2N, PProDOT-CF3SO3 and PProDOT-PF6 show lower Gde,

negative slope of W-plot, positive microwave dielectric response typical for other

dielectric conducting polymers. It is noted that the frequency for maximum of optical

conductivity increases for more insulating samples.

The optical, magnetic and transport studies demonstrate that this family of

inhomogeneously disordered (also confirmed by X-ray diffraction studies) materials

spans from dielectric to metal.


3.11 Room Temperature Conductivities of Alkyl Substituted PEDOTs


The introduction of substituent groups onto conjugated polymers has been utilized

to a great extent in controlling their resulting physical and electronic properties.116

Processability (by solution or melt processing) and electronic tunability (in terms of

conducting and optical properties) have been obtained by incorporating different side

groups.16 Several groups noted an enhancement in electronic properties of thiophene

polymers upon substitution with varied length alkyl chains. Groenendaal et al.

synthesized a series of alkyl-PEDOTs with chains varying from -CH3 to -C14H29-117,118

In-situ conductivity of these materials showed an interesting trend. After a high

conductivity value for unsubstituted PEDOT, the conductivity decreased with increasing









substituent length, and then it started increasing again to values even higher (850 S/cm

for PEDOT-C14) than for the parent PEDOT (650 S/cm). This trend could be explained

by taking into consideration three factors: symmetry of the monomer unit, steric

hindrance between repeating units and order induced by mesogenic side chains. PEDOT

is highly conductive due to its planar structure along with a negligible steric hindrance

between adjacent units. Introduction of a methyl substituent causes asymmetry, especially

when the polymerization proceeds via head to head (HH) or tail to tail (TT) coupling.

This translates into an increased steric hindrance and a less close 3-D packing, and leads

to more disorder and less conjugation. Increasing of the alkyl substituent to a hexyl chain

produces more asymmetry, resulting in further drop of conductivity. By extending the

chain length to a decyl substituent, the conductivity begins to increase, even though the

asymmetry and steric hindrance are similar to those of hexyl substituted PEDOT. The

improvement seems to arise from a better film morphology due to self-assembled decyl

chains.

Our work focused on obtaining free-standing films of these alkyl-substituted

PEDOTs to compare the conductivity to the values obtained using the in-situ method.

Figure 3.13 shows the conductivities of racemic (R) and chiral (C) alkyl-substituted

PEDOT free-standing films with thicknesses up to 35 itm. Even though the variation in

conductivity with the chain length follows the same trend as described above, none of the

substituted materials attained conductivities as high as PEDOT. This could be due to the

low solubility of these monomers in the solvents suitable for the electropolymerization.

Consequently, we could not use an optimal monomer concentration (of about 0.06 M) or

lower the temperature below -50C. The highest conductivity obtained for an alkyl-









substituted PEDOT derivative is 70 S/cm (for PEDOT-C12 and PEDOT-C14), which is

about half of the value obtained for the parent PEDOT. Films prepared in TBAP/ACN

showed even lower room temperature conductivities. The slight increase in the

conductivity of the chiral compounds is due to a higher degree of side chain ordering, as

compared to their racemic counterparts.


120


100


0 2 4 6 8 10

Substituent Length


12 14


Figure 3.13 Conductivity of alkyl substituted PEDOTs as function of the alkyl chain
length. The free-standing films were prepared in TBAPF6/PC.














CHAPTER 4
POLYMER THIN FILM OPTICS


4.1 Introduction


Conducting polymers, once studied solely for their ability to replace heavier metal

conductors, are now widely used in more dynamic applications where rapid switching

from doped (n- or p-) to neutral forms is suitable. Applications such as transmissive

windows,23-25 displays,51,119 and camouflage materials120 require polymer deposition

as thin films on transparent electrodes. A conducting layer of indium-tin oxide (ITO)

deposited on boro-silicate glass or flexible poly(ethylene terephthalate) (Mylar) support is

widely used as an electrode in the fabrication of conducting polymer devices.

Understanding the polymerization mechanism at the polymer-ITO interface is crucial for

the quality of resulting films as well as for device lifetime. Section 4.2 deals with

preliminary atomic force microscopy (AFM) studies related to polymer nucleation onto

an ITO surface.

One of the most important characteristics exhibited by the polymers from the

PXDOT family is their enhanced electrochromism upon application of a small

voltage. 17,19 The mechanism behind the electrochromic behavior is related to the doping

of materials. In the neutral state, the color is determined by the band gap of the polymer,

the 7t-7t* absorption peak wavelength, and the relative intensities of the fine structure.

Upon doping (removal or addition of electrons), new electronic states are formed in the









band gap, thus giving rise to optical absorptions at energies lower than the original band

gap. Therefore, in addition to electrochromism in the visible region of the spectrum, these

polymers show large absorption changes in the infrared region, making them useful in

modulating the reflectance/transmittance off/through a surface over a broad range of the

electromagnetic spectrum. 121

Owing to this unique behavior, the optical properties of conducting polymers have

become the focus of many research groups over the last two decades. 122 These studies

have focused on PT and its alkyl-substituted derivatives,123,124 and very recently on

PEDOT,125-132 and involve IR and Raman spectroscopy coupled with quantum

calculations and X-ray diffraction measurements. Due to the strong correlation between

electronic structure, morphology and chemical (bond ordering pattern or lattice) structure,

the nature of electronic transitions in these materials is still under debate. At the

beginning of this chapter, we discuss the doping-induced electronic structure and the

infrared active vibrational modes as a function of redox state in PXDOT thin films.

Throughout the following sections, we focus our study on the origin of multiple peaks or

"fine structure" on the main 7t-7t* absorption peak in these polymers.


4.2 Surface Analysis


Understanding the chemical and physical processes taking place at the interface

between the electroactive polymer and ITO plays an important role for the quality of the

resulting film as well as for the operational capabilities (switching time and lifetime) of

electrochromic devices. Tin doped indium oxide (ITO) is routinely used as an electrode

in various optoelectronic devices such as electrochromic windows, displays, and OLEDs









due to its unique combination of properties including optical transparency from UV to

2500 nm in the NIR, low electrical resistance (R< 10 Q) and excellent surface

adhesion. 133 ITO electrodes have been reported to interact chemically with electroactive

polymers, which possibly contributes to the degradation and eventual failure of polymer

devices. 134 Even in the absence of oxygen and moisture, MEH-PPV, commonly used in

OLEDs, undergoes degradative oxidation losing conjugation through the formation of

carbonyl bonds. This oxidation is promoted by the ITO electrode that serves as a source

of oxygen. Moreover, after a period of time, devices showed high concentrations of

atomic indium originating from the ITO layer. 135,136

The general consensus concerning the polymer growth on ITO (and other surfaces)

is that during initial stages of the polymerization, growing oligomers deposit on ITO

creating nucleation centers randomly distributed on the surface.137 This process is

favored by the presence of hydroxyl groups that confer a slight negatively charged ITO

surface. Polymer chains are growing from these primary nucleation centers (see Figure

4.1 A). Formation of ordered brush-like film morphology is anticipated at this stage,

giving rise to a thin smooth film seen in Figure 4.1 B. Because the ITO surface has been

already coated with a polymer film, secondary nucleation occurs on top of the already

existing polymer. Precipitated polymer from the proximity of the electrode can also be

incorporated into the growing film, leading to a globular surface morphology and

increased surface roughness of resulting electroactive layers (Figure 4.2). 138

Experimental observations showed that repeated polymer electrochemical

depositions followed by gentle wiping off of the polymer layer led to more uniform films









exhibiting enhanced optical properties. In this study, PEDOT films were grown up to 20

mC/cm2 charge density and then wiped off with a tissue soaked in acetone until the ITO


A 2 r




B


Figure 4.1 The model of film formation: A) Primary nucleation centers, B) subsequent
polymerization and chain growth, C) Secondary nucleation occurs on the already existing
film, D) Incorporation of precipitated polymer at advanced stages of polymerization
(from Sapurinal37)


Figure 4.2 Surface morphology ofPANI films produced at (A) early stages and (B)
advanced stages of polymerization (from Sapurinal37)









surface appeared clean to the naked eye. Figure 4.3 A shows an AFM image of a clean

ITO revealing its heterogeneous, granular morphology resulting in a very rough surface,

also noted by other researchers. For example, Kluger et. a1134 has reported ITO surface

roughnesses ranging from 25 to 100 nm, depending on the supplier. The image in Figure

4.3 B represents the same ITO surface after a PEDOT electrodeposition/ buffing cycle.

The appearance of 1.5 |tm diameter globular domains across the ITO surface supports the

idea that the early stage polymerization proceeds via primary nucleation centers. These

centers are very well adhered to the ITO surface (resisting even cleaning with acetone),

thus creating anchors for the growing polymer chains. Repeating the above

electrodeposition/buffing step several times likely increases the density of primary

nucleation centers on the electrode surface, and consequently the uniformity of the

resulting films.









Clean ITO








PEDOT Deposited/ Buffed


Figure 4.3 Surface morphology of a A) clean ITO/glass and B) ITO after a
electrodeposition/buffing cycle showing the primary nucleation centers