1 CONTROL OF COLOR, MOLECULAR ORDER AND CHARGE TRANSPORT IN CONJUGATED POLYMERS By UNSAL KOLDEMIR 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 2012
2 2012 Unsal Koldemir
3 To Dr. Fikret Koc
4 ACKNOWLEDGMENTS I would like to first thank my advisor Prof John R. Reynolds, for his guidance and su pport in all aspects of my graduate school work during my 5 years at the University of Florida He has been a role model for me in terms of conducting research, mentoring students, guiding a research team, teaching an undergraduate level course, as w ell as a good family member. He has been very influential in building my academic career not only for research purposes but also in teaching undergraduate level courses H e supported me for a research visit to the Max Planck Institute for Polymer Researc h in Mainz, Germany, which has advanced my skills through research collaboration important for future work with other groups. Through his deep insight s in research, I was exposed to many of the challenges of organic chemistry, polymer chemistry, electroche mistr y, physical characterization, as well as transistor device applications. I would also like to thank the members of my committee, Prof. Kenneth Wagener, Prof Aaron Aponick, Prof. Yun C. Cao and Prof. Xiangeng Xue, for their valuable discussions and con tribution through out my studies at the University of Florida. I would like to thank Prof. Kenneth Wagener for leading the Butler Polymer Laboratories This work would not be possible without the contribution of my research collaborators, Prof. Franky So, P rof. Jiangeng Xue in the Department of Materials Science and Engineering at UF and Klaus Muellen at the Max Planck Institute for Polymer Research. I would like to thank Prof. Franky So for his valuable feedba ck in solar cell devices over the past 4 years. He offered critical insights into the design of polymers for high impact solar cell devices. Additionally, I really enjoyed the SoRey meetings with the So group members held every month. Dr. Wojtek Pisula and Dr.
5 Sreenivasa R. Puniredd have c ontributed sig nificantly to my work through their knowledge of X ray characterizations and organic field effect transistor devices. I want to specifically thank to Dr. Chad Amb and Dr. Pengjie S personality helped to create a wonderful team environmen t for my research f or 3 years. His feedback has helped me overcoming many synthetic chal lenges. Pengjie has been a close friend in and outside of the group and I enjoyed spending time with him especially for fishing trips. During my 5 years in the Reynold s group, I had the opportunity to work with many great researcher s. Dr. Frank Arroyave was a walking library; Dr. Jianguo Mei was always full of great ideas and new synthetic routes; Dr. Romain Stalder was full of kindness; Dr. Mike Craig was full of laugh ter and rarely known organic chemistry purification methods; Dr. Stefan Ellinger was full of motivation for research; Dr. Ken neth Graham was full of details. I also owe thanks to Dr. Aubrey Dyer, Dr. Svetlana Vasilyeva and Dr. David Liu for helping me to p erform electrochemistry experiments. I also enjoyed working with the younger graduate students in our group, James Deininger and Caroline Grand. Special thanks should be given to Cheryl Googins and Sara Klossner who devoted their time in facilitating admi nistrative procedures I really enjoyed the time when Sara took me for a Christmas show when I had to stay in Gainesville over the winter holidays Lastly, I would like to thank my parents who supported me unconditionally through all my endeavors Without their support, I would have never made it to this point.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 1.1 Fundamentals of Conjugated Polymers ................................ ............................ 17 1.1.1 Band Gap Control ................................ ................................ .................... 18 1.1.2 Color Control ................................ ................................ ........................... 22 1.1.3 Molecular Order and Charge Transport Control ................................ ...... 25 1.1.3 .1 Molecular Order and Charge Transport in Small Molecules ........... 26 220.127.116.11 Molecular Order and Charge Transport in Polymers ...................... 30 1.2 Po lymer Synthesis ................................ ................................ ............................ 35 1.2.1 Oxidative Polymerization ................................ ................................ ......... 36 1.2.2 Metal Catalyzed Polymerizations ................................ ............................ 37 1.2.3 End Group and Stoichiometric Control ................................ .................... 44 1.3 Processing Methods ................................ ................................ ......................... 46 1.3.1 Small Scale Processi ng Methods ................................ ............................ 47 1.3.2 Large Scale Processing Methods ................................ ............................ 49 1.4 Selected Device Applications ................................ ................................ ............ 52 1.4.1 Field Effect Transistors ................................ ................................ ............ 52 1.4.2 Photovoltaics ................................ ................................ ........................... 55 1.5 Thesis of This Dissertation ................................ ................................ ................ 61 2 EXPERIMENTAL METHODS AND CHARACTERIZATIONS ................................ 64 2.1 General Synthetic Methods ................................ ................................ ............... 64 2.2 Purification of Polymeric Materials ................................ ................................ .... 64 2.3 Materials Characterization ................................ ................................ ................ 68 2.3.1 Structural Characterization ................................ ................................ ...... 69 2.3.2 Polymer Molecular Weight Characterization ................................ ............ 70 2.3.3 Thermal Characterization ................................ ................................ ........ 70 2.3.4 Electrochemical Characterization ................................ ............................ 70 2.3.5 Optical Spectra Characterization ................................ ............................. 72 2.3.6 Spectroelectrochemistry ................................ ................................ .......... 72 2.3.7 Two Dimensional Wide Angle X Ray Scattering (2D WAXS) .................. 73
7 2.3.8 Grazing Incidence Wide Angle X Ray Scattering (GIWAXS) ................... 75 .2.3.9 Atomic Force Microscopy ................................ ................................ ....... 77 2.4 Photovoltaic Devices ................................ ................................ ......................... 77 2.5 Organic Field Effect Transistors (OFETs) ................................ ......................... 79 3 THE IMPORTANCE OF SYNTHESIS AND PROCESSING METHODS ON THE PERFORMANCE OF GREEN COLORED PHOTOVOLTAIC MATERIAL: PGREEN CASE ................................ ................................ ................................ ...... 80 3.1 Introduction ................................ ................................ ................................ ....... 80 3.2 PGreen by Oxidative Polymerization ................................ ................................ 84 3.3 PGreen by Still e Polycondensation ................................ ................................ ... 89 3.4 Conclusions on the PGreen ................................ ................................ .............. 93 3.5 Utilization of APFO Type Polymers for Obtaining Green Color ......................... 94 3.5.1 Polymer Synthesis and Characterization ................................ ................. 95 3.5.2 Conclusions ................................ ................................ ........................... 101 3.6 Experimental Details ................................ ................................ ....................... 101 4 THE INFLUENCE OF THE BRIDGING ATOM IN THE FUSED THIOPHENE BASED DONOR ACCEPTOR POLYMERS ON CONTROLLING MOLECULAR ORDER AND CHARGE TRANSPORT ................................ ................................ 107 4.1 Introduction ................................ ................................ ................................ ..... 107 4.2 Polymer Synthesis and Characterization ................................ ........................ 114 4.3 Optical a nd Electrochemical Characterization ................................ ................ 119 4.4 2D WAXS and GIWAXS Characterization ................................ ...................... 125 4.5 OFET and Solar Cell Performances ................................ ................................ 129 4.6 Conclusions and Perspectives ................................ ................................ ........ 132 4.6 Experimental Details ................................ ................................ ....................... 133 5 KEY ROLE OF END GROUPS IN CONTROLLING MOLECULAR ORDER AND CHARGE TRANSPORT ................................ ................................ ....................... 140 5.1 Introduction ................................ ................................ ................................ ..... 140 5.2 Polymer Synthesis and C haracterization ................................ ........................ 143 5.3 OFET Performances ................................ ................................ ....................... 155 5.4 X ray and Morphology Analyses ................................ ................................ ..... 157 5.5 Conclusions and Perspectives ................................ ................................ ........ 161 5.6 Experimental Details ................................ ................................ ....................... 162 6 CONCLUSIONS AND PERSPECTIVES ................................ ............................... 164 LIST OF REFERENCES ................................ ................................ ............................. 167 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 180
8 LIST OF TABLES Table page 2 1 GPC molecular weights in THF for NE(DTG BTD) polymer with Soxhlet extraction.. ................................ ................................ ................................ .......... 68 3 1 GPC molecular weights in THF and elemental analysis of t he PGreen series. .. 85 3 2 S olar cell performance parameters for the PGreen series ................................ .. 87 3 3 GPC molecular weights in THF and element al analysis of the Stille PGreen polymers. ................................ ................................ ................................ ............ 90 3 4 Summary of solar cell performance parameters for PGreenStille1 and PGreenStille2 ................................ ................................ ................................ ..... 92 4 1 GPC estimated molecular weights in THF and elemental analyses of the polymers. ................................ ................................ ................................ .......... 118 4 2 Absorption parameters for polymers both in toluene and thin film s .................. 121 4 3 Estimated HOMO LUMO energy levels and band gaps by CV and DPV ........ 124 4 4 2D WAXS (bulk) and GIWAXS (Thin film) analysis for the DTS and DTG series. ................................ ................................ ................................ ............... 128 4 5 H o le mobilities and on/off ratios for DTS and DTG series ............................... 130 5 1 GPC molecular weights in THF and TCB, and elemental analy sis of the polymers. ................................ ................................ ................................ .......... 144 5 2 H ole mobilities for E(DTG BTD) and NE(DTG BTD). ................................ ....... 157
9 LIST OF FIGURES Figure page 1 1 The concept of band gap as in conjugated polymers. ................................ ........ 19 1 2 Reduction of the band gap in a donor acceptor copolymer. ............................... 21 1 3 Representative electron poor and electron rich units employed in donor acceptor type polymers. ................................ ................................ ..................... 22 1 4 Colors obtained from 3,4 dioxythiophene based polymers along wit h their normalized absorption spectra ................................ ................................ ............ 23 1 5 Repeat unit structures for the APFO type polymers displaying green and purple color along with their normalized absorption spectra. .............................. 24 1 6 Examples of chemical modifications to s tudy the molecular order ...................... 27 1 7 Control of molecular order in poly(3 alkylthiophenes). ................................ ........ 33 1 8 A FM phase images for different molecular weight P3HT .. ................................ .. 34 1 9 X ray analyses of PCPDT BTD for different molecular weights. ......................... 35 1 10 Formation of Herrmann complex with Pd2(dba) 3 :P(o tol) 3 catalyst system. ...... 38 1 11 Catalytic cycle scheme as in Stille coupling. ................................ ...................... 39 1 12 Proposed mechanism on the chain growth Suzuki polymerizations. .................. 43 1 13 Direct arylation method for the synthesis of conjuga ted polymers ...................... 44 1 14 Two dimensional grazing incidence X ray diffraction patterns for Poly(3 octylthiophene) ................................ ................................ ................................ ... 48 1 15 Illustratio n of six fabrication steps of deposition of organic solar cells components via roll to roll process. ................................ ................................ .... 51 1 16 Representative hole and electron transporting polymers exceeding 0.1 cm 2 /V.s.mobili ties in OFET devices. ................................ ................................ ... 54 1 17 Schematic illustration of the electronic processes in a bulk heterojunction organic solar cell ................................ ................................ ................................ 57 1 18 Representative J V curve obtained from an organic solar cell and the evaluation of important parameters from the curve. ................................ ........... 5 9
10 1 19 Representative high performing OPV polymers with their sola r cell parameters. ................................ ................................ ................................ ........ 60 2 1 Pd scavenging by using diethylammoniumdiethyldithiocarbamate ..................... 66 2 2 Demonstration of soxhlet extractio n process. ................................ ..................... 67 2 3 Preparation and analysis of a polymer via 2D WAXS.. ................................ ....... 74 2 4 Analysis of GIWAXS patterns ................................ ................................ ............. 76 2 5 Picture of a bottom gate bottom contact OFET covered with a polymer. ............ 79 3 1 Properties of PGreen. ................................ ................................ ......................... 82 3 2 Synthesis of PGreen via oxidative polymerization. ................................ ............. 85 3 3 Employed solar cell architectures a nd performance of PGreen series ............... 86 3 4 Performance of PG reen3 in large area solar cells ................................ .............. 88 3 5 The performance of large scale processed blue green cells .............................. 89 3 6 PGreen synthesis through Stille polycondensation. ................................ ............ 90 3 7 MALDI MS of PGreenStille polymers ................................ ................................ 91 3 8 Performance of PGreenStille polymer in large area solar cells.. ........................ 93 3 9 Chemical composition of APFO copolymers with different donor and acceptor units employed in the repeat unit structure. ................................ ........................ 94 3 10 Chemical composition and synthesis of APFO DOT copolymer by Suzuki polycondensation. ................................ ................................ ............................... 96 3 11 Absorption and emissi on characteristics for APFO DOT ................................ .... 97 3 12 Current density and radiant emittance as a function of appli ed bias for APFO DOT based PLED ................................ ................................ ............................... 98 3 13 Electrochemic al characterization of APFO DOT ................................ ......... 99 3 14 Solar cell characteristics of APFO DOT ................................ ........................... 100 4 1 Various cyclopentadithiophene based copolymers and their chemical composition. ................................ ................................ ................................ ..... 108 4 2 Copolymers based on dithienosilole and their chemical compositions. ............ 110
11 4 3 X ray diffractograms of donor acceptor polymers of cyclopentadihiophene and dithiensilole. ................................ ................................ ............................... 111 4 4 Dithigermole based donor acceptor copolymers reported in 2011. .................. 112 4 5 The literature examples studied the C Ge bond lengths. ................................ .. 113 4 6 Repeat unit structures for the discussed n dode cyl functionalized dithienogermole and dithienosilole copolymers ................................ ................ 114 4 7 Schemes for the synthesis of n dodecyl functionalized dithienosilole and dithienorgermole monomers and the six donor ac ceptor type polymers. .......... 116 4 8 Comparison of 1 H NMR spectra of the distannylated monomers and their precursors. ................................ ................................ ................................ ........ 117 4 9 MALDI MS of donor acceptor polymers. ................................ .......................... 119 4 10 Normalized UV VIS absorption profiles for PGe12 BTD PSi12 BTD PGe12 TPD, PSi12 TPD PGe12 PT and PSi12 PT. ................................ ................... 120 4 11 Electrochemical character ization of DTS and DTG polymers ........................... 123 4 12 Extruded fiber 2D WAXS patterns me asured for DTS and DTG polymers ....... 127 4 13 GIWAXS patterns recorded for thin films of DTS and DTG polymers.. ............. 128 4 14 Bottom gate bottom contact OFET architecture for charge transport measurement s of the DTG and DTS polymers. ................................ ................ 130 5 1 Reaction schemes for the synthesis of E(DTG BTD) and NE(DTG BTD) ........ 144 5 2 NMR charact eriza tion of toluene end groups ................................ ................... 146 5 3 MALDI MS spectra of polymers ................................ ................................ ........ 149 5 4 Possible scenario showing the loss mechanisms of debromi nation and destannylation. ................................ ................................ ................................ 151 5 5 XPS spectra of NE(DTG BTD) with two different X ray sources of Mg and Al and E(DTG BTD) with Mg source.. ................................ ................................ ... 152 5 6 Absorption spectra for dilute solutions of E(DTG BTD) and NE(DTG BTD) in toluene and for thin film.. ................................ ................................ .................. 154 5 7 Differential Pulse Voltammograms of E(DTG BTD) and NE(DTG BTD). .......... 154 5 8 Transistor output curves for E(DTG BTD) and NE(DTG BTD) thin films. ......... 156
12 5 9 GIWAXS characterization of E(DTG BTD) and N E(DTG BTD) ........................ 158 5 10 GIWAXS integrations of the wide angle scattering region along qz at qx,y = 0 for E(DTG BTD) and NE(DTG BTD). ................................ ............................... 159 5 11 Tapping mode AFM height images of drop cast and spin coated E(DTG BTD), and NE(DTG BTD).. ................................ ................................ ............... 160
13 LIST OF ABBREVIATION S AFM Atomic force microscopy APFO Alternating polyfluorene BTD 2,1,3 Benzothiadiazole CV Cycli c voltammetry DOT 3,4 Dioxythiophene DPV Differential pulse voltammetry DTG Dithienogermole DTS Dithienosilole EC Electrochromic ECD Electrochromic device ECP Electrochromic polymer EtHx 2 Ethylhexyl Fc/Fc + Ferrocene/Ferrocenium FET Field effect transistor FF Fill Factor GPC Gel permeation chromatography GIWAXS Grazing Incidence Wide angle X ray Scatterring HOMO Highest occupied molecular orbital ICBA Indene C 60 bisadduct ITO Indium tin oxide IR Infrared Spectrum J sc Shor t current density LED Light emitting diode LUMO Lowest occupied molecular orbital
14 MALDI Matrix assisted laser desorption/ionization M n Number average molecular weight M w Weight average molecular weight NMR Nuclear magnetic resonance OPVs Organic photo voltaics OFET Organic field effect transistor OLED Organic light emitting diode PCE Power conversion efficiency PCBM [6,6] Phenyl C61 butyric acid methyl ester fullerene PDI Polydispersity index ProDOT Propylenedioxythiophene PSC Polymer solar cell p TSA p Toluenesulfonic acid PV Photovoltaic SCE Saturated calomel electrode TGA Thermogravimetric analysis TLC Thin layer chromatography UV Ultraviolet spectrum Vis Visible spectrum V oc Open circuit voltage
15 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 CONTROL OF COLOR, MOLECULAR ORDER AND CHARGE TRANSPORT IN CONJUGATED POLYMERS By Unsal Koldemir D ecember 2 012 Chair: John R. Reynolds Major: Chemistry Conjugated polymers with a high degree of molecular order were developed and integrated into organic electronic applications, where the mechanical robustness, processibility, and high charge transport properti es of these materials were taken advantage of to yield high performance devices. This dissertation examines strategies for improving device performance first through optimizing synthetic protocols for the scalable preparation of photovoltaic materials with aesthetically pleasing color; and also through synthetic manipulation, via heavy atom substitution and end capping, of commonly used polymers to yield improved charge transport in field effect transistor applications. A dioxythiophene donor acceptor based polymer, PGreen, was synthesized through oxidative polymerization (later optimized through Stille polymerization) and implemented in photovoltaic devices. The synthesis was both reproducible and scalable. PGreen was chosen because of its aesthetically ple asing green color, achieved via a trough in its absorption profile in the green region of the spectrum. This work is the first to study the scalability of both the synthesis and processing of a polymer other than poly(3 hexylthiophenes) for large area sola r cell applications. In
16 particular, the performance differences between spin coated, small area solar cells, and large scale slot die coated solar cells were investigated to highlight the importance of processing techniques on device performance. In the s econd portion of this thesis, a germanium atom was substituted in place of a silicon atom as the bridging atom in fused thiophene aromatic units in six polymers obtained by Stille polycondensation by 1% stoichiometric imbalance. Heavy atom substitution is discussed as a means of increasing the molecular organization of conjugated polymers, leading to higher charge transport properties. Ge atom substitution provided a way to tune both the HOMO and LUMO levels and thereby the absorption profiles of the polyme rs in the visible spectrum. Furthermore, longer C Ge bond lengths enhanced Pi stacking with closer intermolecular interactions, as shown by X ray analyses, resulting in higher charge carrier mobilities. As an alternative approach, the effects of end group s on the molecular order of the best performing dithienogermole benzothiadiazole based copolymer were studied and compared with the performance of a non end capped control polymer. End groups were confirmed by 1 H NMR, 2D NMR and MALDI MS analyses. The pres ence of end groups resulted in an increased tendency to stack with closer intermolecular interactions. Independent of the processing conditions, a much higher hole mobility was obtained by end capping.
17 CHAPTER 1 INTRODUCTION 1.1 Fundamentals of Conjugate d Polymers With the new millennium came a revolution in consumer electronics which would forever change the public view of electronic devices. Through miniaturization and sophistication, products such as ipads and E readers have been created, which years a go could only be imagined. This has turned the attention of the electronics industry towards alternative high performance semiconducting materials, and the subsequent development of materials that can be easily deposited in nanometer thicknes ses even onto flexible surfaces This is where conjugated polymers entered the arena as a family of new materials that combine the optoelectronic properties associated with inorganic materials with the plastic features of organic materials. Extensive research in this fi eld has led to the realization of real life applications such as SONY organic light emitting diode displays, 1 E paper, 2 and Konarka solar cell laptop bags. 3 Synthetic chemistry is an invaluable tool for the development of high performance, optoelectronic materials because it encompasses an infinite number of strategies for the design and preparation of conjugated polymers, allowing them to be custom tailo red for specific applications Moreover, by studying the structure property relationships of conjugated polymers, the performance of optoelectronic devices can be further enhanced. In C hapter 1 the fundamental properties unique to conjugated polymers a re discussed along with important criteria for achieving high performance materials based these polymers. To this end, discussion of the band gap formation of a color, and molecular order. Next, polymerization methods are described that facilitate
18 the synthesis of high molecular weight, scalable, low dispersity polymers. Lastly, a detailed analysis of processing methods is introduced along with a discussion of selected opto electronic applications of conjugated polymers. 1.1.1 Band Gap Control The electronic structure of conjugated polymers relies on a backbone of alternating single and double bonded carbon atoms with delocalized orb itals along the polymer chain. From the simplest ethylene molecule to an infinite theoretical chain of polyacetylene, molecular orbitals overlap to form continuous energy levels; these are the so called energy bands as shown in Figure 1 1a. 4 In theory, polyacetylene h as a fully filled valence band (VB) and an empty conduction band (CB), with a zero energy difference between the two bands. Thus it should yield equivalent resonance structures such as in Figure 1 1b. 5 In reality, there exists an energy difference between the highest occupied molecular orbital (HOMO), and lowest unoccupied molecular orbital (LUMO) in the infinite chain This difference is defined as the band gap and is denoted as E g For example, consider the geometrical difference in structure between the aromatic and quinoid forms of polythiophene as depicted in Figure 1 1c. An energy difference arises as a result of bond length alternation or double bond localization, which results in the formation of non degenerate HOMO and LUMO levels. In solid state physics, this phenomenon is known as Peierls distortion. 4 Additionally; planarity, substitution, aromaticity and interchain interactions can play a role in determining the band gap of a materia l, as shown in Figure 1 1d. To broach these factors, E BLA defines the difference of length between the double and single bonds. The higher the bond length alternation, the larger the bandgap. 6 For instance, in the aromatic form of polythiophene, electrons
19 are more localized in the ring system than they are in the quinoid form. Thus, the carbon carbon bond lengths between the thiophene rings exhibit more single bond character. Figure 1 1. The concept of band ga p as in conjugated polymers. a) Formation of bands in a conjugated polymer. b) Reson ance forms of poly(acetylene). c ) Aromatic and quinoid resonance structures for polythiophene. d) The structural factors affecting the energy of the band gap in an alkylate d polythiophene. (Adapted and modified with permission from R oncali ) c ) a) b ) d)
20 The parameter E describes energy differences that arise due to torsion between two aromatic rings (resulting in a twisting of the polymer described by the angle theta) caused by the steric congestion between the atoms or substituents on the rings. 7 The torsional angle can be decreased by forming more planar geometries, thus fa voring a better overlap of the orbitals. Increased deloc a lization of the electrons along the polymer chain reduces the band gap. From a synthetic organic chemistry perspective, placing double or triple bonds in between aromatic rings can be effective in reducing the amount of torsional strain. 8 10 E sub is related to the effect of the substituents on the aromatic rings. These moieties can be of the electron donating kind (e.g. OR), resulting in increased HOMO energy levels, 11 or can be of t he electron withdrawing kind ( e g CN), operating in decreased LUMO energy levels. 12, 13 Apart from these, the substituents are mostly bare alkyl chains that induce solubility to the polymer. 14, 15 Strong interactions occurring between polymer chains enable the delocalization of electrons, thus governing the energy difference due to interchain effects, E Int Lastly, the energy difference between resonance structures, E res, is related to the degree of aromaticity in the quinoid form, and thus describes energy difference between the aromatic and quinoid forms. This strategy has been applied in realization of the very low band gap conjugated polymers based on isothianaphtene and thienothiophene. 16 18 The donor acceptor strategy first introduced by Havinga et al ., also known as the push pull method, is an effective method to control the band gap 19 This approach finds its remarkable power through reducing the bond length alternation in a polymer by bringing together electron rich (donor) units with the electron poor (acceptor) units. The difference in electronegativity between two units causes polarization of the bond, such
21 that a resonance struct ure can be written as (D A D + = A ). This enhances the delocalization of electrons along the polymer chain by reducing the bond length alternation. Figure 1.2 shows this phenomenon schematically. The molecular orbitals of the donor ized so that the y form new molecular orbitals with a reduced band gap. The effectiveness of this method in reducing the band gap has been realized by Reynolds and coworkers, who report a 0.5 eV band gap copolymer of dithienopyrrole and benzobisthiadiazole. 20 Figure 1 2. Reduction of the band gap in a donor acceptor copolymer. There are many possible donor and acceptor moieties that can be selected for creating conjugated polymers for device applications, including integration i nto OFETs, 21, 22 OPVs, 23 26 electrochromics 27, 28 and OLEDs. 29 S ome o f the available electron rich and electron poor units are depicted in Figure 1 3. The large variety of available units allows polymer properties to be tailored through judicious choices of donor and acceptor combinations. For instance, the Mats Anderson gr oup has reported band gaps between 2.1 to 1.3 eV by varying the electron poor unit in the structure of alternating polyfluorene (APFO) type donor acceptor polymers. 30
22 Figure 1 3. Representative electron poor and electron rich units employed in donor acceptor type polymers. 1.1.2 Color Control The color of a conjugated polymer depends on its light absorption behavior in the visible region of the spectrum and the relative intensity of its absorption bands. There are multiple reasons for the occurrence of absorption bands, such as the extent of conjugation in the polymer and the energy levels of the HOMO and LUMO of the polymer. Along with the considerations mentioned for band gap control, relative energies of the HOMO LUMO levels become a challenge to be addressed by synthetic chemistry. This phenomenon has been important in controlling the color of non emissive electrochromic materials, where a neutral color is switched to a transmisssive color upon redox processes. The design of electrochromic polymers with specific colors has been reviewed by our group recently. 27 Particularly, 3,4 dialkylateddioxythiophenes have been employed in the design of polymers to obtain red, orange, magenta, blue,
23 cyan, green, yellow and black colors. Figure 1 4 describes the absorption profiles of the desired colors for e ach specific p olymer. 31 Figure 1 4. Colors obtained from 3,4 dioxythiophene based polymers along with their normalized absorption spectra. (Adapted with permission fr om R eynolds ) It is worth noting here that previously, electrochromic polymers enjoyed low band gap polymers with high HOMO levels to obtain the colors stated above with the exception of yellow. A neutral yellow colored copolymer was obtained by employing di oxythiophene benzene moieties. 32 This polymer had a band gap of 2.38 eV with an absorption maximum at 455 nm and an onset of absorption of 520 nm. The absorption profile in the blue region yields a vibrant yellow color. Removing the benzene unit and changing the propylenedioxy bridge to 2 ethylhexyloxy groups on the thiophene shift s the absorption maxima to 490 nm, achieving an orange colored polymer. The same electron rich moiety was randomly copolymerized with 3,4 dimethoxythiophene to yield
24 a red colored polymer by r educed steric interactions because of the smaller size of the me thoxy groups. Additionally, a random copolymer approach has been successfully applied with 3,4 propylenedioxythiophene and benzothiadizole units to control the whole absorption in the visible spectrum in order to obtain neutral black colored polymers. 33 For organic solar cell polymers, the c olor has not been the central concern. The highly studied P3HT is orange in color due to minimal absorption above 650 nm. Cylopentadithiophene benzothiadiazole (PCPDT BTD) copolymer, also commonly used in solar cell applications, is blue green 34 and dithienosilole benzothiadiazole (PDTS BTD) 35 achieves a red shift of absorption leading to its green color. In this regard, the Andersson group has shown a systematic approach to span the visible spectrum using the alternating polyfluorene (AP FO) type donor acceptor polymers. 36 As seen in Figure 1 5, by changing the nature of the acceptor unit, the absorption can be expanded up to 1000 nm by varying the energy Figure 1 5. Repeat unit structures for the APFO type polymers displaying green and purple color along with their normalized absorption spectra. (Adapted with permission from A ndersson When th e acceptor moiety is benzothiadiazole, the polymer is purple in color due to minimal absorption in the blue and red region. By changing the acceptor to
25 thiadiazoloquinoxaline, the polymer achieves a green color due to absorption in the blue and red region with transmission in the green region, ca 500 550 nm. The absorption bands are due to transitions. Beaujuge et al. have proposed concepts in an attempt to explain the origin of dual band absorption, particularly in the donor acceptor type polymers. 37 In particular, the concentration of donor and acceptor units has been varied in the repeat unit structure of the polymers based on 3,4 propylenedioxy thiophene and benzothiadizole. By increasing the electron rich moietie s in the polymer backbone, the longer wavelength absorption band is reduced and slightly blue shifted. At the same time, the shorter wavelength absorption band is increased in intensity and red shifted, minimizing the trough between this band and the longe r wavelength absorption band. In a simple assumption, the donor acceptor polymers are treated as consisting of two distinct chromophores. One chromophore involves the donor units and increasing the number of electron rich moieties decreases the band gap of these chromophores and leads to the aforementioned red shifted absorption. A second chromophore results from the donor acceptor segment, and increasing the electron rich units in the repeat unit causes the dilution of donor acceptor interactions. Conseque ntly, leted and slightly blue shifted. 1.1.3 Molecular Order and Charge Transport Control Many inorganic semiconductors form crystalline solids through the periodic arrangement of their atoms into a three dimensional ne twork bound together with strong ionic and covalent bonds. Organic semiconductors, on the other hand, rely on attractive non bonded interactions, such as stacking between aromatic units through delocalized
26 to each other, determines the stacking ability or so called molecular order In fact, molecular order/packing determines the transfer integral and reorganization energy of a polymer. 38 The transfer int egral is determined by the difference between the HOMO and LUMO energy levels, and is directly related to overlap between adjacent molecules. 39 The reorganization energy is the energy lost concomitant to the charge transport through a molecule. It is dependent on the extent of conjugation in organic molecules and on the molecular order in their films 40 43 Lower reorganization energy correlates with higher charge carrier transport, a necessary component for organic molecules in optoelectronic applications. Improved molecular ordering decreases the reorganization energy. 44 Therefore, this section describes important synthetic tools to increase molecular order in small molecules and polymers, in conjunction with t heir charge transport properties. While this dissertation does not focus on the synthesis of small molecules, it is important to cover these strategies as an a priori for understanding the molecular order in polymers. Highly ordered films made of organic m olecules can form semicrystalline materials, which, in general, correspond to higher charge carrier mobilities as suggested by Street et al. 45 Conjugated molecules/polymers result in films with slig htly less order, but can still be treated as distinct from random structures. In the subsequent two sections, examples are presented of strategies to induce order in the solid state. Small molecule conjugated systems are analyzed first, as their physical a nalysis relatively simple, followed by the more complex macromolecular systems. 18.104.22.168 Molecular Order and Charge Transport in Small Molecules Alkyl chain substitution is a powerful way to create, not only soluble molecules, but also to control the mesosc opic organization these molecules. 46 This method was
27 used to prepare two oligomers based on anthracene and thiophene in order to study the effect of side n hexyl chains, as illustrated in Figure 1 6a. Figure 1 6. Examples of chemical modifications to study the molecular order. a c) Alkyl chain substitution. d) Alkoxy substitution. e) Halogen atom substitution. f) Heavy atom substit ution Oligomers were prepared by Suzuki coupling between pinacolatoboronic ester substituted 2,6 anthracene and the corresponding 2 bromothiophenes. 47 Both oligomers present ed with a herringbone type packing geometry, similar to that of pentacene, while the dihexyl substituted oligomer demonstrated smaller unit cell dimensions in the solid state. The difference in packing structures of these two oligomers revealed distinct ch arge transport properties when implemented in OFET devices. T he dihexyl
28 substituted oligomer exhibited a hole mobility of 0.5 cm 2 /V.s. with an on/off ratio as high as 10 7 whereas the non substituted oligomer showed a hole mobility value of only 0.063 cm 2 / V.s. with an on/off ratio of 10 5 This order of magnitude difference in the charge transport behavior can be attributed to the hexyl chains promoting long range order in the molecule by favoring self assembly processes during crystallization. This analysis is separately discussed by Garnier et al. through the examination of oligomers composed of sexithiophenes. 46 In their work, three sexithiophene oligomers were prepared as shown in Figure 1 6b. sexithiophene presented excellent solubility of 400 mg/mL in chloroform, whereas non substituted and dihexylsexithiophene showed poor solubility in either dichloromethane or chloroform, ca 1 mg/mL. Thi s observation was attributed to the strong i nteractions in non s ubstituted sexithiophene and Additionally, melting temperatures observed for these compounds were, 80 o C, 280 o C, and 290 o C, respectively, for non substituted sexithiophene and dihexylsexithiophene, confirming the increased conformational freedom of the dihexylsexithiophene system. On the ot her hand, substitution on the position enhanced the cohesive forces between the conjugated molecules. This was mainly caused by the h ydrophobic lipophilic interactions existing between alkyl chains. Furthermore, X ray measurements showed long range order for the substituted sexithiophene compared to the non substituted form, whereas the substituted sexithiophene did not show any cry stallinity. The crystallinity of each molecule correlated with its measured hole mobility. Hole mobilities of 0.05 cm 2 /V.s. and 0.002 cm 2 /V.s. were obtained the substituted sexithiophene and non substituted
29 sexithiop hene respectively, whereas the subst ituted sexithiophene did not show any charge transport behavior. Consequently, alkyl substitution has proven effective in improving the structural organization of molecules on the mesoscopic scale, resulting in highly ordered solid architectures, and there by improving hole mobilities. Additional studies on the impact of alkyl side chains on the crystallinity and charge transport were conducted using thiophene oligomers subs tituted with n hexyl, n decyl, n dodecyl and bulky cyclohexyl side chains at the position as shown in Figure 1 6c. 48 It was observed that t he modified molecules aligned themselves with their long molecular axis perpendicular to the substrate, while unsubstituted thiophene oligomers render randomly oriented thin films. Bulky cyclohexyl functionalized quarterthiophenes in particular showed good solubility in common organic solvents, allowing the fabrication of OFETs through drop casting. Drop cast samples demonstrated hole mobilities of 0.06cm 2 /Vs, a value 3 times higher than the di n hexyl substituted quarterthiophenes. Small organic molecule s were substituted with alkoxy derivatives to increase their solubility as well as to promote ordering in the solid state. A derivative of disc shaped hexabenzocoronene (HBC) was functionalized with four dodecyloxy side chains as illustrated in Figure 1 6 d. 49 Though its aromatic core is severely distorted out of planarity into a corrugated structure, the molecule was found to self assemble into two dimensionalcolumnar structures due to the tetra n dodecyloxy substitutients. Solution processed OFETs fabricated with dodecyloxy functionalized HBC showed hole mobilities of 0.02 cm 2 /V.s. with an on/off ratio of 10 6 Anoth er synthetic strategy to control the solid state ordering in organic molecules involves the incorporation of halogen atoms. Halogen atoms such as chlorine and/or
30 bromine were attached on tetracene molecules as shown in Figure 1 6e. The halogen substituted tetracene derivatives were all soluble in common organic solvents, facilitating the growth of single crystals. Single crystal X Ray diffraction patterns of dihalogen derivatives showed a face to face slipped stacking motif, and mono halogen substituted derivatives displayed a herringbone type packing motif. The intermolecular distance between neighboring molecules was determined to be 3.4 9 for the dichlorotetracene, resulting in hole mobilities of 1.6 cm 2 /V.s. 50 This can be attributed to the ability of halogen groups to promote c o facial stacking through the formation of sheets between the layers, thus increasing the hole mobility of dichlorotetracene by promoting interactions. 51 Yet another tool for controlling molecular order in organic materials relies on heavy atom substitution, which has been found to enhance intermolecular orbital overlap in aromatic units by reducing intermolecular spacing. Phenyl end capped benzodithiophene derivative s, 2,6 diphenylbenzodichalgenophenes, have been prepared by Takimiya et al. as a prototypical class to study the effect of heavy atom substitution, as shown Figure 1 6f. 52 All three molecules showed moderate OFET charge transport properties wh en deposited under vacuum. An improved hole mobility of 0.17 cm 2 /V.s. was obtained for the heavier selenium derivative, in stark contrast with the hole mobility of 0.018 cm 2 /V.s. obtained for the sulfur derivative. 22.214.171.124 Molecular Order and Charge Transp ort in Polymers As il lustrated with the examples in s ection 126.96.36.199, the simple strategies of synthetic chemistry can be powerful tools to improve not only the solubility, but also to promote molecular order in thin films, of small molecules and oligomers. Polymers,
31 because they are larger in size than small molecules, are limited in terms of their packing arrangements, due to increased requirements for conformational freedom. As a result they tend to form small crystals surrounded by amorphous material. Ad ditionally, polymers exhibit a polydisperse nature, which limits their long range ordering. The connectivity between the ordered domains and the amorphous domains is also an important factor in determining their charge transport properties, increasing the complexity of the analysis of their microstructures. Poly(3 alkylthiophenes) have been benchmark polymers to study the molecular ordering in conjugated polymers due to their simple synthesis and scalable procedures as suggested by Rieke 53 and McCull ough. 54 57 Polythiophenes were synthesized by electrochemical methods and found to be insoluble, impeding the possibility of device applications. Alkyl substituents on the 3 position were applied to increase the pol assembly. 58 As seen in Figure 1 7a, poly(3 alkylthiophenes) with increasing lengths of alkyl chains(four to twelve carbon s) were synthesized. A non monotonic dependence of charge carrier transport properties was measured by OFETs prepared by spin casting. The average hole mobilities reported were: 1.2x10 3 cm 2 /V.s. for n butyl, 0.01 cm 2 /V.s. for the n hexyl, 2x10 4 cm 2 /V.s. for the n octyl, 6.6x10 5 cm 2 /V.s. for the n decyl and 2.4x10 5 cm 2 /V.s. for the n dodecyl functionalized poly(3 alkylthiophenes). Additionally, in order to evaluate the microstructural characteristics of these polymers, two dimensional grazing incidence X ray diffraction measurements were performed by drop casting n hexyl, n octyl and n dodecyl poly(3 alkylthiophenes). 59 As seen in Figure 1 7c, the n hexyl functionalized poly(3 alkylthiophene) derivative displayed long range order, as evidenced by the spots
32 on the q z axis, as well as an edge on orientation on the substrate, as derived from the arc in the q xy axis. The n octyl and n dodecyl functionalized derivatives displayed less order, as evidenced by diffuse and weaker arcs in the q z and q xy axes. The stacking distances were calculated to be 3.74 for poly(3 hexylthiophene) and 3.90 for the poly(3 dodecylthiophene) derivative. In this context, the regioregularity or irregularity caused by the relative arrangements of alkyl chains, further determ ines the molecular order of poly(3 alkylthiophenes). Alkyl chains can be arranged head to tail (H T) and head to head (H H), as shown in Figure 1 7b. The H H arrangement brings steric repulsion between the alkyl chains, which causes twisting of the thiophe ne rings. When this happens, it results in reduced intermolecular interactions of the orbitals, therefore resulting in random orientations of the polymer chains. For instance, regioregular poly(3 hexylthiophene) (P3HT) adopts a lamellar packing structure with preferential edge on orientation on the substrate 60 resulting in charge mobilities exceeding 0.1 cm 2 /V.s. 60 63 By contrast, increasing the amount of H H arrangements in the regioregular P3HT induced a face on orientation on the substrate resulting in charge carrier mobilities of only 10 4 cm 2 /V.s. 59 Furthermore, molecular order can be controlled through tuning the molecular weight of the polymer. Studies on P3HT have shown that the number average molecular weight is a key parameter for controlling mesoscopic and microscopic order. 64 Figure 1 8 shows the AFM phase images for a 3.2 kDa (Figure 1 8a) and 32 kDa (Figure 1 8 b) number average molecular weight regioregular P3HT. Low molecular weight regioregular P3HT presented with a fibrillar morphology, with rod like crystallites, well suited for crystalline stacking through self assembly
33 Fi gure 1 7. Control of molecular order in poly(3 alkylthiophenes). a) Poly(3 alkylthiophenes) functionalized with different length alkyl chains. b) Regioregular and irregular poly(3 alkylthiophenes) obtained from H T and H H couplings. c) Drop cast two dimen sional grazing incidence X ray diffraction patterns for drop cast poly(3 hexylthiophenes), poly(3 octylthiophenes) and poly(3 dodecylthiophenes). (Adapted with permission from L ocklin ) Figure 1 8 shows the AFM phase images for a 3.2 kDa (Figure 1 8a) and 3 2 kDa (Figure 1 8b) number average molecular weight regioregular P3HT. Low molecular weight regioregular P3HT presented with a fibrillar morphology, with rod like crystallites, well suited for crystalline stacking through self assembly. Increasing the mole cular weight to 32 kDa resulted in isotropic nodules, resulting in less crystallinity, due to their relatively immobile long polymer chains, which restricted their rearrangem ent into highly packed fibers. One would expect that highly crystalline, low molec ular weight polymers
34 would have a higher hole mobility compared to less crystalline, high molecular weight polymers, however the charge transport is three orders of magnitude higher in the high molecular weight P3HT. This can be attributed to the presence of grain boundaries formed between highly crystalline polymer fibers in the low molecular weight polymer. Grain boundaries act as trapping sites for holes, leading to reduced charge transport. In the high molecular weight polymer, grain boundaries are smoo thed out by connections between the ordered domains that provide pathways for charge transport. The low molecular weight P3HT lacks this kind of network architecture. Figure 1 8. AFM phase images for different molecular weight P3HT a) Low molecular wei ght (3.2 kDa) b) High molecular weight (32 kDa) ( Adapted with permission from F rechet ) X ray analyses have been performed on PCPDT BTD copolymers substituted with n hexadecyl alkyl chains, as shown in Figure 1 9. 65 A charge carrier mobility of 0.17 cm 2 /V.s. was achieved with a 13 kDa molecular weight PC PDT BTD copolymer, while a PCPDT BTD copolymer with 65 kDa molecular weight presented with a hole mobility of 0.67 cm 2 /V.s. Thus the higher molecular weight PCPDT BTD copolymer demonstrated improved charge transport by almost 4 fold. However, macroscopic o rdering, as observed through X ray diffraction data, for PCPDT BTD contradicts those results found
35 for P3HT. The high molecular weight po lymer exhibited distinct spectr al peaks in its X ray diffraction pattern, as shown in Figure 1 9a, while the low molecu lar weight polymer yielded a completely amorphous diffraction pattern. This data was further supported by the pronounced reflections in 2D WAXS scans performed using extruded fibers of the high molecular weight polymer (Figure 1 8b). These reflections are absent in the low molecular weight polymer (Figure 1 9c). These findings clearly contradict the aforementioned results regarding the correlation between molecular weight and crystallinity found for different molecular weight P3HT samples and warrant furthe r study. Figure 1 9. X ray analyses of PCPDT BTD for different molecular weights. a) X ray diffraction pattern for high molecular weight (56 kDa) PCPDT BTD (red line) and low molecular weight (13 kDa) PCPDT BTD (black line).Two dimensional wide angle X r ay scattering patterns for b) high molecular weight (65 kDa) PCPDT BTD c) low molecular weight (13 kDa) PCPDT BTD. (Adapted with permission from M uellen ) 1.2 Polymer Synthesis This section is dedicated to the most common synthetic methods for the preparati on of conjugated polymers. Important details on these polymerization methods are discussed in conjunction with the control of molecular weight, polydispersity and chemical purity of these materials.
36 1.2.1 Oxidative Polymerization The first method develope d to generate conjugated polymers was through oxidative polymerization, which can be accomplished through either chemical or electrochemical processes. Oxidative polymerizations are advantageous in that they do not require expensive reagents and involve sh orter reaction times to produce high molecular weight polymers. For chemical oxidative polymerizations oxidants such as Fe(III)Cl 3 Fe(III)(OTs) 3 SbCl 5 and NOPF 6 can be used in organic solvents. 66 70 By far, anhydr ous Fe(III)Cl 3 is the most commonly used oxidant. However, it produces large quantities of HCl during the rearomatization via dissociation of two protons. It can be speculated that the generated HCl can overoxidize the growing polymer chains and cause mole cular defects. 71, 72 In addition, McCarley et al. observed that multiple chlorine end groups are added during the polymerization, which could limit the polymerization and also effect the purity of the polymer. 72 In fact, the growing polymer is in oxidized form and needs to be reduced with hydrazine in order to obtain a neutral polymer. Also, the oxidized polymer is complexed with FeCl 4 and has a lower solubility than its neutral form. 73 Moreover, large amounts of Fe residues can be buried in the polymer network which could be detrimental to the performance of the polymer in device applications. 74, 75 Despi te the challenging pros and cons of oxidative polymerizations, when the conditions are well adjusted, it is possible to obtain high molecular weight polymers changing the solvent in oxidati ve polymerizations from chloroform to orthodichlorobenzene (ODCB). 76 Th ey claim that ODCB solubilizes the oxidized polymer better than chloroform solvents. A thidiazoloquinoxaline bithiophene copolymer was prepared by Fe(III)Cl 3 oxidati on to
37 yield an Mn of 240 kDa in ODCB. The same polymer yielded a Mn of 14 kDa when chloroform was used as the solvent. Furthermore, the rate of addition to the solvent is important, as adding the oxidant all at once causes very low molecular weights. 70 However, applications of these methods are limited, as oxidative polymerization is only effective for monomers with low oxidation potentials. 1.2.2 Metal Catalyzed Polymerizations Palladium reagents and catalysts provide numerous possibilities for carbon carbon bond formation in conjugated polymers through Stille, Suzuki Miyaura, Heck, Sonogashira, N egishi, Kumada and Hiyama coupling reactions. 77 Pd catalysts are tolerant to many functional groups, such as carbonyl and hydroxyl groups. Therefore no protection s tep is needed before the coupling reaction. In addition, they are insensitive to moisture, basic conditions or oxygen. 78, 79 The most common sources of Pd(0) catalysts are Pd(OAc) 2 PdCl 2 (PPh 3 ) 2 Pd(PPh 3 ) 4 ,and Pd 2 (d ba) 3. The first two sources, Pd(OAc) 2 and PdCl 2 (PPh 3 ) 2 require the addition of electron rich phosphine ligands to form Pd(0). Common phosphine ligands include PPh 3 P(o tol) 3 P(t bu) 3 PCy 3. 80 These ligands are electron rich phosphines that enable the oxidative addition step in the polymerization process by donating electrons to an oxidized Pd center. The bulky groups can also accelerate the reductive elimination step through the reduction of steric congestion i n the complex. Moreover, they can be used to increase the catalytic life/activity and turnover numbers of the Pd catalysts. For example, P(o tol) 3 can be used in coupling reactions through the formation of the Herrmann complex with a Pd 2 (dba) 3 catalyst, a s shown in Figure 1 10. 81 The complex has an increased catalytic life because it is stable in air and moisture, however,
38 precursors to the complex must be kept in an inert environment to prevent the oxidation of the phosphine ligand. Figure 1 10. Formation of Herrmann complex with Pd2(dba) 3 :P(o tol) 3 catalyst system. The basic steps for the Pd catalyzed reactions include oxidative addition (OA), transmetallation (TM) and reductive elimination (RE) in order to form a catalytic cycle. A Stille coupling procedure utilizing these steps is outlined in Figure 1 11. Oxidative addition involves the addition of an X Y molecule to Pd(0) species. The single bond between the X Y is broken, and two new single bonds are formed by using Pd nonbondin g electrons. Thus, the oxidation state of Pd(0) increases by two leading to Pd(II). The process can be summarized as; Pd(0) + X Pd(II) Y. During the OA step, two of the ligands dissociate from the catalyst system to free vacant coordination sites. As t he Pd metal is oxidized during the OA step, increased electron density on the Pd atom accelerates this reaction. Sigma donor ligands, such as the phosphine derivatives mentioned previously, are particularly attractive for this purpose. On the acceptor ligands do the opposite: they inhibit the OA step. It is worth noting that, the nature of the halogen or pseudohalogen atom bonded to the sp 2 carbons facilitates the OA step in this order; C I > C Br >>> C Cl>>C F. 82 In the transmetallation step, Ar Pd X species react with organometallic
39 The driving force here is the electronegativitit is comprised mostly of main group elements, which are more electropositive than the palladium metal. This step is called transmetallation and can be summarized as: Ar Pd X + Y Ar Pd R + Y X Figure 1 11. Catalytic cycle scheme as in Stille coupling. The next step, reductive elimination, is complementary to the oxidative addition step. A unimolecular dissociation occurs, forming the original Pd(0) catalyst. The ligands that dissociate must be in cis configuration. The loss of two ligands results in the coordination and formal oxidation state of Pd(0) from Pd(II). Bulky phosphine ligands assist this step by bringing the two organic groups on palladium closer together. The rec overed Pd(0) catalyst is now active and undergoes the OA step again, fulfilling the catalytic cycle. The Stille reaction was first reported by Kosugi Migita and Stille separately. 83, 84 It involves coupling reaction s between organostannanes with arylbromides with the
40 advantage of organostannanesare compatible with other functional groups. The preparation of organostannanes is straightforward however; using stoichiometric amounts of organostannanes is a major drawback There are several methods for the preparation of aryl organostannanes. Aryl halides can be reacted with butyllithium to form aryllithium, which can be further quenched with R 3 SnCl. Another method was discovered by Eaborn et al. by reacting arylhalides wi th hexabutyldistannane (Bu 6 Sn 2 ) and a Pd catalyst system. 85 In this case, hexamethyldistannane (Me 6 Sn 2 ) is preferred over hexabutyldistannane as it can facilitate the transmetallation step. 86 Murata et al. have suggested a Pd catalyzed reaction of aryliodides with tributyltin hydride in ambient conditions, using potassium acetate as a base to generate organostannanes. 87 Suzuki coupling involves the reaction of organoborons with arylhalides. 88 90 Unlike Stille reactions, which utilize hazardous organostannanes, 91 Suzuki reactions use organoborons, which bear low toxicity. In addition, boronic acids/esters are stable with respect to changes in heat and moisture and exposure to air. Thus, a large number of organoboranes are commercially av ailable. Suzuki coupling conditions are advantageous as they are compatible with many functional groups and run in mild reaction conditions. On the other hand, the aryl group bonded to the boron atom is less nucleophilic, making the transmetallation step m ore difficult. To circumvent this problem, bases must be added to generate ate complexes in which boron is quaternized in order to facilitate the transmetallation step. The arylboranes can be easily prepared by reacting aryllithium or arylmagnesium specie s with the corresponding trimethylborate, followed by hydrolysis. Alternatively, arylboranes can be prepared through the Pd catalyzed coupling between arylhalide with bis(pinacolato)diboron. 92 Simple arenes can
41 be reacted with p inacol ester derivatives by iridium or rhodium catalysts to generate the arylborane, without requiring bromination or lithiation steps, in high yields via direct borylation. 93, 94 To summarize, the preparation of th e precursors for Stille and Suzuki couplings, the selection of phosphine ligand and the mechanism of these reactions have been discussed in order to highlight important parameters governing polymerization reactions. When the aforementioned precursors are d ifunctionalized with the appropriate halide and the corresponding organostannane or organoboron derivative, it is possible to create A B type alternating polymers. In this dissertation, as detailed above, the Pd 2 (dba) 3 :P(o tol) 3 catalyst system has been ut ilized to produce high molecular weight polymers. This system has been chosen because of its high catalytic life and high turnover number. The ratio of Pd 2 (dba) 3 catalyst was chosen as 2 4% mol with respect to one mole of monomer. In addition, the amount P (o tol) 3 was chosen by doub ling the amount of Pd catalyst. These amounts were sufficient to generate high molecular weight polymers in just hours, but the reactions were stirred over 3 days to ensure polymerization yields near 100%. This was effective in a chieving polydispersities of 2 3 before any purification step. Toluene was selected as the solvent because it allowed for the samples to be heated at temperatures of 90 o C, while maintaining the polymer solubility in the reaction mixture. For Suzuki polyco ndensations, phase transfer catalysts such as Aliquat 336 have been used to facilitate the mixing of organic and negatively charged species. Schluter and coworkers have shown a polyfluorene homopolymer synthesized with Aliquat 336 that exhibited a number a verage molecular
42 weight (M n ) of 48 kDa. 95 Without this phase transfer catalyst the same polymer resulted in only a 9 kDa M n Stille and Suzuki polycondensations producing high m olecula r weight conjugated polymers ha ve been described in the literature. A patent has claimed the synthesis of a spiropolyfluorene polymer with a M n of 267 kDa using toluene, water, K 3 PO 4 and 0.025 % of the Pd 2 (dba) 3 :P(o tol) 3 catalyst system using Suzuki polycondensation conditions. 96 The same polymer has also been shown to have a M n of 410 kDa when prepared from ultrapure monomers. 95 On the other hand, Stille polycondensatio n has produced polymers with M n over 100 kDa. In 2010, the Andersson group reported a copolymer with a M n of 100 kDa synthesized using pyrazino[2,3 g ]quinoxaline and trithiophene with the Pd 2 (dba) 3 :P(o tol) 3 catalyst system. 97 A 207 kDa number average molecular weight copolymer based on phthalimide has been reported by Guo et al. 98 Recently, Bao and coworkers reported an isoindigo bithiophene copolymer with a M n of 138 kDa using the same catalyst system in a microwave assisted Stille polycondensation. 99 In addit ion to Stille and Suzuki polycondensations, which proceed via step growth, a new method involving chain growth mechanism was suggested by Yokoyama et al. 100 This polymerization proces s is illustrated in Figure 1 12. The polymerization starts with the formation of a Pd complex with benzene, which acts as an initiator. The transmetallation occurs with the AB type monomer. Then, the metal migrates over the part of the fluorene and adds itself oxidatively in between C and Br atoms. The species obtained is similar in structure to the initiator. Through the sequential addition of the monomer,a benzene end capped polyfluorene was obtained. The mechanism of
43 polymerization has been confirmed a s a chain growth process, resulting in a M n of 18 kDa with low polydispersities (1.3 1.4). Figure 1 12. Proposed mechanism on the chain growth Suzuki polymerizations. A new method of carbon carbon formation between aromat ic units, called Direct Arylation, provides another alternative to Stille and Suzuki couplings. 101 103 This method eliminates the necessity of synthesizing the organostannanes or organoborons needed for Stille and S uzuki polycondensations, thus lowering the number of steps in the polymerization process. Further, by eliminating these precursors, this process avoids the use of the toxic materials used in conventional metal catalyzed polycondensations. Lemaire and co wo rkers have succeeded in synthesizing P3AT derivatives using this method, via direct arylation using Pd(OAc) 2 K 2 CO 3 base and stoichiometric amounts of tetrabutylammoniumbromide as illustrated in Figure 1 13a. 104 The resulting polymers displayed regioregular structures with low molecular weights with M n of approximately 6 kDa, and with PDI close to 2. Wang et al. were successful in obtaining a higher molecular weight P3HT with a M n o catalyst. 105 The regioregularity was calculated as 98 % with a polymer yield of 99%. The Leclerc group has recently shown a comparison of this polymerization with Stille
44 polycondensation as depicted In Figure 1 13b. 103 The Stille product showed a M n of 9 kDa with a 71% yield, while direct arylation afforded the same polymer with a M n of 56 kDa and a yield of 96%. Figure 1 13. Direct ar ylation method for the synthesis of conjugated polymers. a) Direct arylation synthesis of poly(3 alkylthiophenes) b) Comparison of direct arylation and Stille polycondensation for generating the same thienopyrroledione based polymer. 1.2.3 End Group and St oichiometric Control In Section 188.8.131.52 the importance of molecular weight on crystallinity and performance has been discussed for conjugated polymers. In order to properly control the molecular weight in step growth type polycondensations, extended reacti on times, strict stoichiometric balance between the bifunctional monomers and carefully designed catalyst systems are required. In this regard, reaction time and stoichiometric balance
45 are emphasized in this section, particularly as they pertain to Stille and Suzuki polycondensations. The Carothers equation predicts the number average degree of polymerization as; X n = (1+r)/(1+r 2rp) for the polymerization of bifunctional monomers, where r denotes the stoichiometric imbalance and p represents the extent of reaction. 106 For polymerizations with carefully weighed, ultrapure bifunctional monomers in stoichiometric amounts, r = 1 and the equation becomes X n = (1)/(1 p).Thus, the p groups as determined by p This explains why Stille and Suzuki polycondensations are run for at least 72 hours to ensure the completion of the reaction. For polymerizations wit h a high degree of nonstoichiometry, the polymer will have a low molecular weight. For instance, if bifunctional monomers are not carefully weighed, a 1 mol% excess of one monomer would lead to X n = 201 when p is 100%. In reality, p never reaches unity, he nce the same case with p = 98% results in X n =40. Another scenario arises when one of the bifunctional monomers has monofunctional impurities. This is true for many Stille reaction precursors in which the purification of ditin monomers is difficult, and mo notin monomers may still be present. This not only results in a stoichiometric imbalance, but also the average functionally will be lower than 2. This is the worst case scenario and will dramatically lower the molecular weight. In prac tice, Suzuki reaction monomers boranates/boronic acids are crystalline solids that can be purified with several recrystallizations or using column chromatography. However, ditin monomers decompose during column chromatography, and trying to purify them by recrystallization m ay lead to decomposition by light and heat. Amb et al. have
46 suggested the utilization of reverse phase HPLC columns for large scale purifications of ditin monomers of dithienogermole. 107 Stoichiometric imbalance can be intentionally applied to end cap step growth polymers. The polymerization proce eds until one monomer is completely exhausted. When this happens, all the chain ends will have the same functional group that is present in the excess monomer. Another strategy is to intentionally add a small amount of a monofunctional monomer. The monofu nctional monomer will limit the polymerization, as the growing chains will have no functional groups to resume the polymerization. Yet another strategy relies on, post end capping reactions. In this strategy, a Stille or Suzuki polycondensation is run with stoichiometric balance. After ensuring sufficient reaction time, monofunctional monomers are added sequentially. This method has been applied by Leclerc for carbazole based polymers 108 and by Schluter for the preparation of polyarylenes. 95 The end capping group was a monofunctional benzene moiety. As benzene hydrogen chemical shifts w ould overlap with the aromatic backbone units, their characterization is difficult by 1 H NMR. Recently Bazan and coworkers claimed the utilization of XPS measurements to analyze the absence of Sn and Br atoms after end capping with monofunctional thiophene via Stille polycondensation, claiming that the absence of these atoms implies successful endcapping of the polymers. 109 Soon after this work was published, Marks group attempted to obtain XPS spectra on these polymers, however their results were inconclus ive with regards to the absence of Sn and Br on the chain ends. 110 1. 3 Processing Methods As conjugated polymers are integrated into high performan ce devices, new processing techniques are needed so that improved device architectures can be
47 realized with conjugated polymers. In particular, solution processible polymers are of interest for industrial applications. The methods used to deposit these po lymers dictate the crystal growth mechanism, and as a result determine device performance of the polymers. Therefore, in addition to investigating polymer synthesis methods, it is also important to consider polymer processing techniques. This section exami nes the small scale and large scale processing methods and discusses important parameters. 1.3.1 Small Scale Processing Methods Popular processing methods for depositing conjugated polymers onto relatively small substrates include spin coating, drop castin g, dip coating and spray casting. In spin coating, a thin film is formed while solution is added over a spinning substrate. 111 Dur ing this process, the film dries rapidly and evenly over the surface. As the polymer does not have sufficient time for self assembly, long range directional alignment is uncommon with this method. The morphology of the film and also charge transport proper ties are dependent on the spinning rate, polymer solution concentration and the choice of solvent. 59, 112 Sirringha us and coworkers showed that P3HT exhibits higher crystallinity, and an order of magnitude higher hole mobility, when processed from a high boiling point solvent such as 1,2,4 trichlorobenzene compared to a low boiling point solvent such as chloroform. 63 During the process, most of the polymer solution does not remain on the substrate; therefore a more economic method such as drop casting is desirable when material is limited. In drop casting, the evaporation of the solvent occurs over a longe polymer is provided through this method. 113 The effect of processing conditions, such as thermal annealing, solvent selection and deposition method, have been studied by t wo dimensional grazing incidence x ray
48 diffraction as seen in Figure 1 14. 59 It is obvious that regioregularpol y(3 octylthiophene) displays different crystallinity and molecular orientation on the substrate depending on the processing condition used for its deposition. Drop casting yields poly(3 octyl)thiophene molecules that have long range order, as proved by the (100) peaks in the q z axis. These molecules prefer an edge on structure on the substrate as evidenced by the (010) peak in the q xy axis for the chloroform toluene, THF and dichloromethane solvents. On the other hand, spin casting does not yield the same long range order for chloroform solutions, and results in a face on orientation for both chloroform and toluene solutions. When a volatile solvent is selected, such as dichloromethane, long range order is still absent in spin cast samples, but molecules ad opt an edge on structure, similar to the drop casting results. Upon thermal annealing, long range order can be recovered for all of the solvents and edge on orientation is preferred in all c ases except for THF solutions. Figure 1 14. Two dimensional graz ing incidence X ray diffraction patterns for Poly(3 octylthiophene) processed from drop casting and spin coating from different solvents. (Adapted with permission from L ocklin )
49 Another method for depositing polymer films is through dip coating, in which a substrate is immersed in a polymer solution then drawn out at a specific rate such that the solvent evaporation happens at the rate of crystallization. This method provides homogenous thin films in the dip coating direction. If thicker films are desired, multiple layers of polymer can be coated on the substrate by repeating the process. Muellen and coworkers have shown that by depositing a P CPDT BTD copolymer via dip coating results in improved long range ordering compared to films obtained through spin co ating. 65 Another deposition technique, spray cast ing, involves forming polymer solution droplets and subsequent deposition on a surface, resulting in homogenous microstructures. 114 This method offers the ability to select flexible, soft or flat substrates. The pressure, the solvent concentration and the angle of spraying are critical in determining the thickness of the material. Although it yields rough surfaces, spray casting is the most widely applied method for depositing conjugated polymers for electrochromic devices. 31, 115 One of the few reports of applying this method for p olymer solar cells has shown power conversion efficiencies of 2.8% for a P3HT:PCBM blend from a dichlorobenzene solvent 116 and 3.2% when chlorobenzene was used. 117 Separately, the Yang group has demonstrated the preparation of a multi junction solar cells with active layers composed of P3HT: PCBM and dithienosilole benzothiadizole copolymer:PCBM by spray coating w hich yield PCEs of 2.3 %. 118 1.3. 2 Large Scale Processing Methods The aforementioned method s except spray coating are mostly applied in the lab scale, on small substrates. These techniques are not easily scalable to the larger substrates needed for real life applications. Techniques such as ink jet printing, screen
50 printing, slot die coating a nd combined roll to roll printing have been developed which can deposit polymers over large areas in a timely and costly manner. Interested readers may look at the reviews by Krebs and coworkers on these techniques. 1 19, 120 These methods are preferable to test optoelectronic device performances in solar cells that have sizes closer to commercial products. Screen printing is a conventional, versatile technique used in industry for creating shapes and texts notable fo r its zero waste solution processing. 121 It has been applied to deposit solar cells, field effect transistors and light emitting diodes. 1 22 124 The process involves the deposition of a film by painting a polymer solution onto a screen to create a pattern. It requires a highly viscous and low volatile polymer solution. Additionally, as the polymer solution is exposed to air, it is important that the polymer not dry out during the process. In 2009, Krebs et al. have prepared solar cells with this method in ambient conditions, and introduced the modules to the public. 125 Ink jet printing is akin to the process used in table top laser ink jet printers. It uses ceramic piezoelectric printheads, which allow different ink formulations with a variety of so lvents to be used. 126 128 During the deposition process, the printheads have no contact with the substrate, and therefore a many kinds of substrates are compatible with this method. While it can fabricate ultrasmoot h pixels with high resolutions (300 dpi), the printing speed is slow. Nevertheless, Hoth et al have demonstrated solar cells based on P3HT:PCBM with an output of 3% PCE with this technique. 129 In the slot die coating technique a polymer solution is filled in a coating head and pushed through a slot by pressure with no loss. 130 It allows deposition of low viscosity polymer solutions into stripe patterns. Th e polymer film thickness is highly dependent on
51 the flow rate of the solution and the substrate velocity. Blankenburg et al. have demonstrated P3HT:PCBM solar cells prepared by this method with 1.7% PCEs under AM 1.5 conditions. 131 They were able to optimize production of 10 m 2 /h polymer solar cells. The Krebs group in RISO has dominated the field of large scale processing polymer solar cells by combining all the techniques above in a roll to roll process. Figure 1 15 outlines each step taken to deposit the electrodes, hole and electron transport layers, as well as the active P3HT:PCBM In the first step, a striped pattern is applied onto ITO substrates by screen printing a UV curable etch resist. Next, the ITO was etched, washed, and allowed to dry. Then, ZnO nanoparticles in chlorobenzene were slot die coated onto the striped ITO pattern. Figure 1 15. Illustration of six fabrication steps of deposition of organic solar cells components via roll to roll process. (Adapted with permission from K rebs ) This was followed by deposition of the active layer, consisting of P3HT:PCBM in dichlorobenzene, using slot die coating. After this layer was allowed to dry, PEDOT:PSS in isoproponal was slot die coated on top. Lastly, silver paste was screen pr inted, and
52 the samples were dried in an oven. In this regard, they have shown in numerous publications how to optimize each step and the required device geometry. 132 135 By following such steps the Krebs group was a ble to generate PCEs of 2.1% when the deposition of each layer is performed in air. 119 1.4 Selected Device Applications This section is dedicated to the description of selected examples of state of the art advances in organic field effect transistors and organic photovoltaics. Fundamentals of device design and parameters are also discussed. 1.4. 1 Field Effect Transistors The basic building block of a field effect transistor is the Metal Oxide Semiconductor (MOS) capacitor, comprised of an insulator sandwiched between two conducting metals. A metal gate electrode, used to modulate the current, is covered with a dielectric layer and a semiconducting polymer film is deposited along with source and drain electrodes in intimate contact with the polymer. This particular kind of architecture is known as a bottom gate top contact FET. Most FET devices ope rate in what is known as enhancement mode, where the device is in an off state when zero voltage is applied to the gate electrode. Increasing the potential applied to the gate creates an accumulation of charges at the semiconducting polymer/insulator inter face. When a sufficient amount of charge builds up at the interface, an inversion layer is formed, creating a conducting channel between the source and drain electrodes. The potential existing between the source and drain electrodes, V SD obeys the basic O Law relation to the current between these electrodes, V SD SD. The performance of FETs is determined by measurement of the charge carrier mobility, In the saturation regime, the mobility is given by the equation: I SD = W/2L sat C i (V G V T ) 2 where W denotes
53 the transi stor channel width, L denotes the transistor channel length (the distance between source and drain electrodes), sat denotes the charge carrier mobility, C i denotes the capacitance of the insulator, V G denotes the voltage applied at the gate electrode and V T denotes the threshold voltage, the voltage that is needed to turn the transistor on. State of the art large area electronic displays utilize field effect transistors that use amorphous silicon as the semiconducting material. Amorphous silicon has charge carrier mobility greater than 0.1 cm 2 /V.s., however high temperatures (greater than 350 o C) and high vacuum equipment are needed for processing. This not only increases the manufacturing cost, it also prevents the application of these materials onto flexi ble plastic substrates. In this instance, organic semiconductors are advantageous as they can be processed from solutions in ambient conditions, on even plastic substrates. However, they need to have charge carrier mobilities exceeding 0.1 cm 2 /V.s. in orde r to compete with current display technology. Carefully designed conjugated polymers have shown great progress in terms of charge carrier mobilities by using simple solution processing techniques. Figure 1 16 shows the polymers with the highest performance s. Regioregular P3HT has been the most widely studied material, with hole mobilities as high as 0.1 cm 2 /V.s reported in top contact devices. 136 Following a more careful design, in order to decrease the density of alkyl chains, bithiophene moieties were incorporated between 3 hexylthiophene units. Ong et al. have shown hole mobilities of 0.1 cm 2 /V.s with PQT 12. 137 Further, derivatives of poly(2,5 bis(3 alkylthiophen 2 yl)thieno[3,2 b ]thiophenes)s (PBT T) have been introduced by McCulloch et al. 138 FETs based on these polymers have shown hole mobilities up to 0.6 cm 2 /V.s. after anne aling, in top
54 gate OFET device architectures. Yet more promising results have been obtained with the introduction of donor acceptor type polymers for high performance OFETs. A phthalimide based copolymer, PhBT12 designed by Guo et al., resulted in mobiliti es up to 0.28 cm 2 /V.s, 98 while a diketopyrrolopyrrole quarterthiophene based copolymer, PDQT, achieved 0.89 cm 2 /V.s., attributed to lamellar packing on the substrate with an edge on orientation. 139 In this particular example, the strong diketopyrrolopyrrole acceptor interacted with the neighboring donor acceptor units, resulting in improved molecular o rder through extensive intermolecular interactions. Figure 1 16. Representative hole and electron transporting polymers exceeding 0.1 cm 2 /V.s.mobilities in OFET devices. Much improvement has been observed by Muellen and co workers by casting the PCPDT BTD polymer via dip coating. 113 This method has increased the hole mobility of
55 0.17 cm 2 /V.s.obtained by spin casting, to values as high as 1.7 cm 2 /V.s. Finally, the current state of the art highest charge carrier mo bility that has been reported thus far was obtained by Mei et al. in 2011. 99 By employing an isoindigo bithiophene copolymer, PiI T2, with siloxane pendant groups (these groups increase the attractive interactions by fo rcing polymer chains closer), they were able to obtain a mobility of 2.5 cm 2 /V.s. The aforementioned examples present holes as the dominant charge carriers. On the other side, electron transporting materials are of interest for incorporation into p n junct ion diodes, acceptor molecules in the organic solar cells and in complimentary organic circuits. Compared to p type polymers, the development of n type materials has lagged behind, with just a few examples exhibiting e exceeding 0.1 cm 2 /V.s. This is due t o the environmental instability of most n type polymers due to electron interactions with oxygen and water in air. Nevertheless, a ladder type electron deficient polymer, BBL, has been synthesized and devices prepared by spin coating have shown electron mo bilites of up to 0.1 cm 2 /V.s. 140, 141 A very high molecular weight (250 kDa) naphthalenedicarboximide bithiophene copolymer has also been tested in air and ambient conditions and found to yield stable electron mobil ities of 0.85 cm 2 /V.s. 142 1.4. 2 Photovoltaics Bulk heteroj unction (BHJ) organic photovoltaics (OPVs) can be classified into two subgroups depending on whether they consist of small molecules or polymers. Small molecule OPVs bear the advantage of straightforward synthesis, high purity that can be obtained by recry stallization, and multiple choices of device fabrication processes. Devices can be made by thermal evaporation of small molecules under high vacuum or
56 by processing from solution. Recent progress in this research field has resulted in power conversion effi ciencies (PCEs) close to 7%. 143, 144 Despite the exciting result s with small molecules, the film quality is still not well controlled in these devices, limiting further improvements. On the other side, polymer based OPVs provide many advantages. Firstly, polymer film deposition via solution processing is much more easily controlled than for small molecules because of the wide variety of solvents that can be used for processing. Secondly, polymers have higher absorpti on coefficients than small molecules, leading to less material consumption when making the ink formulations. Thirdly, they do not require the expensive high vacuum equipment necessary for thermal vapor deposition of small molecules. Figure 1 17 shows the basic mechanisms occurring at the microscopic level to generate photocurrent in photovoltaic devices. Upon exposure to light, excitons are formed within the polymer. In order to create more excitons, a polymer must absorb as much light as possible in the v isible spectrum. The excitons then travel to the donor/acceptor interface, where they are separated into electrons at the acceptor, and holes at the donor, if they do not decay via radiative or non radiative processes (described by the recombination rate) before reaching the interface. Therefore, in conventional solar cells, the thickness of donor and acceptor layers has to be very thin to ensure the maximum charge collection. When the charges reach the electrodes, electrons are collected at the cathode and holes are collected at the anode. The cathode is usually a metal with a low work function such as aluminum and the anode is usually a transparent metal oxide such as indium tin oxide (ITO).
57 T he bulk heterojunction concept developed by Heeger and coworkers in which donor and acceptor materials are blended together to form a continuous interlayer structure, as shown in Figure 1 17. 145 Electrons and holes generated in solar cell devices only have lifetimes of 10 50 fs and diff usion lengths of around 10 nm. Thus many devices are limited by large recombination rates before charge carriers ca n be extracted. The BHJ architecture is particularly effective for minimizing the recombinati on rates of solar cell devices. In this structure, interpenetrating layers of material provide a spatially distributed interface, thus ensuring that all photogener ated excitons are within the diffusion length of a donor acceptor interface. Thus, the morphology of the donor acceptor blend is critical to the dissociation and transport of the charge carriers. Figure 1 17. Schematic illustration of the electronic proc esses (exciton formation, charge separation and charge collection) in a bulk heterojunction organic solar cell and the bicontinuous interlayer between the donor and the acceptor molecule
58 Many groups have used conjugated polymers as the donor material for generating charge carriers in their devices, and used a variety of acceptor molecules including PCBM derivatives, PCBM, PCBM and the ICBA. PCBM derivatives have stood out as the best choice of n type material as they possess strong electronegativit y and high electron mobilities The driving force for charge separation is the energy difference of donor and acceptor LUMO levels. When the charges are separated they travel to the electrodes through the donor and acceptor materials. For efficient charge collection at the electrodes high charge carrier mobilities of the donor and acceptor materials are required. Otherwise, charges will be trapped at the donor and acceptor interface and will recombine. In summary, when developing polymer solar cells the mo st important parameters to consider are the morphology of the donor/acceptor blend, the charge carrier mobilities in the materials, and the energy differences between the donor and acceptor LUMO levels. When testing the performance of OPV devices, power co nversion efficiencies (PCEs) obtained from the J V curves such as the one shown in Figure 1 18, provide a expressed by the following equation where V OC denotes the energ y difference between the donor HOMO level and acceptor LUMO level, J SC denotes the short circuit current and is directly related to the absorption characteristics of the donor polymer, FF denotes the fill factor defined as the maximum power output divided by the product of V OC and J SC
59 Figure 1 18. Representative J V curve obtained from an organic solar cell and the evaluation of important parameters from the curve. An ideal solar cell is one that has a high V OC J SC and high FF values. It is a synthetic chemistry issue to satisfy all of these parameters. Low band gap polymers yield high J SC values due to high coverage of the visible spectrum, however their energy levels can often lead to low V OC values and sometimes cause reduced charge separation. A comp romise of these parameters has been shown in the literature, where polymers resulted in PCE values of above 5.9 % as depicted in Figure 1 19. Of all the polymers studied, regioregular P3HT has been the benchmark polymer for solar cell applications where it s use in devices led to PCEs of 6.5 % when blended with indene C 60 bis adduct (ICBA). The device structure was composed of ITO/PEDOT:PSS/P3HT: indene C 60 bisadduct (ICBA) (1:1, w/w)/Ca/Al. 146 The HOMO level of regioregular P3HT is approximately 5.2 eV and LUMO levels is 3.3 eV. An increase in solar cell PCE has been attributed to the ICBA accepto r molecule, which has 0.17 eV higher LUMO level compared to PCBM. After optimizing fabrication procedures, solar cell PCE values reached efficiencies of 4 5 % when P3HT was blended with PCBM. 147, 148 While the V OC c an be increased for P3HT by this approach, low J SC values limit the output current.
60 Figure 1 19. R epresentative high performing OPV polymers with their solar cell parameters. It only absorbs visible light with wavelengths less than 650 nm, transmitting the rest of the visible spectrum. Thus, to cover more light in the visible spectrum, low band gap polymers have been created via the push pull synthesis strategy based on fused aromatic ring systems such as dithienosilole, dithienogermole, benzodithiphene, and carbazole, for increased conjugation along the backbone. The Leclerc group has worked on carbazole bithienothiadiazole (PCTBT) copolymers to obtain low HOMO energies in pursuit of high V OC values. 108, 149 OPVs employing PCTBT copolymers as the active material have reached PCEs of 6.1 % by this approach. The Yu group has
61 developed fluorinated thienothiophene acceptors i n order to provide decreased bond length alternation in the polymer backbone by aromatization of the quin o d form in the thienothiophene unit. Despite having higher HOMO energies compared to PCTBT polymers, their polymer enjoyed high J SC values of 15.2 mA due to their longer 150, 151 Benzoditihiophene has been popular as it is a planar molecule, which fuses benzene with two thiophenes, allowing for the incorporation of solubilizing alkyl chains on it. When coupled with the thienopyrrolodione acceptor, the copolymer yielded a performance of 6.8%. 152 Bazan and coworkers have used a dithienosilole based copolymer decorated with n dodecyl groups for enhanced solubility. 144 Solar cells prepared using 1 5.9%. Last year, Amb et al. introduced the germanium atom bridged fused thiophene system, dithienogermole, in the pursuit of increasing at tractive interactions between polymer chains due to the presence of the heavy Ge atom. 107 When this new monomer was coupled to a thienopyrroledione acceptor the afforded polymer displayed PCEs of 7.3% in an inverted solar cell structure. 1.5 Thesis of This Dissertation Synthetic organic chemistr y hosts an infinite number of choices of molecule combinations and methods to generate high molecular weight, high performance materials. As discussed in the previous sections, organic chemistry has its own style to resolve and surpass limitations on the p erformance of conjugated polymers for optoelectronic applications. Challenges that are currently being addressed include scalability, color, purity, the nature of end groups, high charge carrier mobilities and intrinsic molecular order in deposited films. These challenges are thought to be critical
62 for prospective conjugated polymers that might find their place in daily life applications. Hence, this thesis approaches these issues with synthetic organic chemistry tools through structure property relationshi ps Chapter 2 summarizes the basic polymer characterization methods used to analyze the purity of polymers. Polymer purification processes are later examined for the purposes of controlling the purity and molecular weight of the polymers. Then, physical characterization methods such as X ray methods are introduced in order to study the molecular order in the polymers. In C hapter 3 approaches to control green colored, aesthetically pleasing, polymers are discussed in conjunction with the scalabil ity of these polymers. The same polymers are made with two different polymerization methods (oxidative polymerization vs Stille polycondensation), and the difference of performances of solar cells prepared using these polymers is discussed. Even though, t he molecular weights obtained using the two polymerization methods are similar, both polymers require different PCBM blend ratios, thus emphasizing the importance of the synthesis techniques on polymer properties. In addition, these polymers were used in t he production of large scale organic solar cells, previously only studied using P3HT. In this context, the performance difference between the lab scale solar cells versus the large area solar cells is addressed. Lastly, an attempt to obtain green colored p olymers by employing APFO type polymers is outlined along with its performance in electrochromic, light emi tting and photovoltaic devices. Chapter 4 details the synthesis of dithienosilole and dithienorgermole based polymers with benzothiadiazole, thien opyrroledione and phthalimide acceptors. Six
63 polymers were obtained via a 1% stoichiometric imbalance approach where the distannylated monomer was added in slight excess. Then the polymers were end capped by adding 4 iodotoluene to control the end groups. All polymers afforded moderate molecular weights, higher than 20 kDA, using this method. Their absorption and electrochemical behaviors are analyzed to highlight structure property relationships in pursuit of understanding the effect of the heavy atom subs titution (germanium versus silicon atom) in the fused thiophene ring. Further, their OFET performances are evaluated in conjunction with the X ray analyses to determine the effect of the heavy atom as it pertains to polymer applications. Chapter 5 focus es specifically on the characterization and analysis of toluene end groups on a dithienogermole benzothiadiazole (DTG BTD) copolymer. DTG BTD polymers, one end capped with toluene moieties and the other left with bare hydrogen atoms, were characterized by 1 H NMR, 2D NMR and MALDI MS. XPS and ICP analyses are also conducted to examine the purity of the polymers. These end groups have led to higher charge transport and increased molecular order.
64 CHAPTER 2 EXPERIMENTAL METHODS AND CHARACTERIZATION S 2.1 General Synthetic Methods All chemicals were purchased from commercial sources and used without further purification unless otherwise noted. Reactions were run in dry argon atmosphere under strict Schlenk conditions with oven dried glassware equipment. Spe cific details describing the synthesis and purification of the compounds used in this dissertation are given in the experimental sections of Chapter 3 5. As a note, Vogel's Textbook of Practical Organic Chemistry has been extensively used as a reference fo r the synthesis and purification of the compounds discussed throughout this dissertation. Additionally, Preparative Polar Organometallic Chemistry by L. Brandsma and H.D. Verkruijsse has been used as a reference book for handling the strong organometallic bases (butyllithium and Grignard reagents). 2.2 Purification of Polymeric Materials The purity of polymerization products can be critical to their performance in optoelectronic applications. Impurities can arise due to unreacted starting materials, low mol ecular weight oligomers, unwanted end groups, oxidized polymer species, chemically doped polymer species, and metal residues. As the list gets longer, the importance of purification becomes more prominent. For example, residual Pd can limit the intensity o f emission and cause changes in the spectral properties of polymers, which can be detrimental to their use in LED applications. 95 In photovoltaics residual Pd decreases the fil l factor and PCE values, as shown independently by Razumov et al. and Seki et al. 153, 154 On the other hand, the effects of heavy metals on organic field effect transistor perf ormance are still under debate. Parneix et al. prepared P3HT via
65 the GRIM method, and fractionated the polymer after washings with methanol, hexane and chloroform. 155 They observed an increase in the OFET performance in the fractions with larger amounts of metal residues, where the metal acted as a dopant to the polymer. By contrast, McCulloch and coworkers applied various washings (acetone, petrol ether and methanol) and chemical treatments (silica, thiol silica and thiourea silica) on a poly[(2,5 bis(3 decylthiophen 2 yl)thieno[2,3 b]thiophene] polymer obtained by Stille p olycondensation. 156 The amounts of metal residues after each procedure were determined and OFETs were prepared from each sample. They were able to decrease the metal r esidues of Pd from 3000 mg/kg to 600 mg/kg and of Sn from 12000 mg/kg to 300 mg/kg. However, polymer hole mobilities were not affected by these procedures and no improvement in the OFET performance was observed. For the Pd catalyzed polycondensations, remo val of palladium impurities can be efficiently performed by adding a complexation agent known as a Pd scavenger diethylammonium diethyldithiocarbamate introduced by Krebs et al 157 The complexation scheme is out lined in Figure 2 1. The Pd scavenger has a great affinity to bind to palladium(0), and the resulting complex is soluble in most organic solvents including methanol, chloroform and light petroleum. Solubility in methanol is rather important as precipitatio n onto methanol and washing with methanol in Soxhlet extraction are commonly applied processes. However, Schluter and coworkers warn that this kind of palladium scavenger is not effective in removing palladium species that are chemically bonded to polymers 95 For the oxidative polymerizations, hydrazine monohydrate is added to reduce the resulting polymer, which has previously been oxidized during the reaction. Next, the polymer solution is added to a pure methanol bath. Low molecular weight oligomers,
66 starting materials and salts ( e.g tin halides for Stille couplings) can go into the methanol phase while the polymer precipitates out. Precipitated polymer can then be separated b y ultracentrifugation or simple filtration over a nylon membrane. Many impurities can be removed from the polymer this way, and can be tracked easily by the change in color of the methanol phase. However, continuous washings with different solvents are sti ll needed to remove the impurities trapped in the polymer network. Figure 2 1. Pd s cavenging by using diethylammoniumdiethyldithiocarbamate Soxhlet extraction involves continuous washing of the polymer with desired solvents to extract the residual unwant ed oligomeric species, unreacted monomers and salts. Figure 2 2 shows each washing step in a soxhlet extractor. The precipitated polymer is collected in a thimble and washed with hot methanol, acetone and hexane to remove the low molecular weight oligomers salts and catalyst residues. Then the polymer is collected with hot chloroform to obtain the high molecular weight polymer. Finally, the solution is precipitated from methanol to collect the polymer as a solid. Parneix et al have made a detailed study on the efficacy of soxhlet extraction as a method for removing impurities from P3HT using methanol, hexane and chloroform. 155 After each washing, the amount of residual catalyst and the number average molecular weights (M n ) of polymer samples were measured. It was observed that the M n increased
67 and the PDI was reduced as the low molecular weight oligomers were removed. Furthermore, the sequence of solvents used was found to be an important parameter in the purification process. Figure 2 2. Demonstratio n of soxhlet extraction process. a) Soxhlet extraction sequence b) Different colors of washings for the low molecular weight polymers does not dissolve most polymers, but d oes however dissolve polar species. This was followed by water and acetone washings for Suzuki couplings to remove salts and low molecular weight species. After acetone, washing with hexane was performed to conclude the removal of low molecular weight olig omers. Finally, the polymer was a) b) b) a) b)
68 extracted using chloroform to dissolve the moderate molecular weight species. The dissolved polymer in chloroform was then recovered by precipitation from methanol. Any polymer left that was not soluble in chloroform could b e further washed with chlorobenzene. A comparative study has been carried out during the purification of non end capped DTG BTD [NE(DTG BTD)] polymers to study the effect of Soxhlet extractions on the polydispersity and molecular weight. The results are pr esented in Table 2 1. After the polymerization, the crude polymer was precipitated into methanol and collected in a thimble, this fraction is labeled NE(DTG BTD) 1. Then it was washed with methanol. A small amount of this sample was taken and the molecular weight parameters were measured by GPC, this portion is labeled NE(DTG BTD) 2. Next, the polymer was washed with acetone (NE(DTG BTD) 3), hexane (NE(DTG BTD) 4) and chloroform (NE(DTG BTD) 3.11 to 2 .66 through removal of the low molecular weight species. At the same time, Mn values increased from 19.9 kDa to 28.1 kDa. Table 2 1. GPC estimated molecular weights in THF after each solvent washing for NE(DTG BTD) polymer with Soxhlet extraction. a The val ues might not be correct due to large amount of impurities. Polymer M n (kDa) M w (kDa) PDI THF THF NE(DTG BTD) 1 a 26.7 83.1 3.11 NE(DTG BTD) 2 19.9 59.8 3.00 NE(DTG BTD) 3 21.1 57.9 2.73 NE(DTG BTD) 4 20.7 49.4 2.37 NE(DTG BTD) 5 28.1 74.9 2.66 T hus, through soxhlet extractions we were able to obtain higher molecular weight polymers with relatively narrow polydispersities Soxhlet extraction based polymer purification is a prominent method in the literature for lab scale prepared polymers.
69 However it is a time consuming process (3 5 days) which makes it impractical for commercial applications 2. 3 Materials Characterization 2.3.1 Structural Characterization The polymers were characterized using a Mercury 300 or Inova 2 500 MHz NMR spectrometers usi ng deuterated chloroform. Chemical shifts were referenced to 7.27 ppm for residual CDCl 3 peaks. High temperature (373 o K) H NMR (500MHz) spectra were obtained by Dr. Manfred Wagner of the Max Planck Institute for Polymer Research, Mainz, Germany. NMR exper iments were conducted using a Bruker Avance III 500 (11,7 T) with a 5 mm z gradient BBFO 1 H/X probe. The proton (1D and 2D) and carbon spectra (1D and 2D) were measured in C 2 D 2 Cl 4 at 373 o K and the spectra were referenced as follows: for the residual C 2 HDCl 4 (H) d 1 ( 1 H) = 5,95 ppm and C 2 D 2 Cl 4 ( 13 C) = 74.1 ppm. 1 H, 1 H NOESY 2D method was applied via the dipolar coupling. The mixing time was 200ms. The frequency sizes were 6000 Hz in both dimensions (f1 and f2) with a relaxation delay of 1,2s. 2D 1 H, 13 C HSQC e xperiments were carried out by 8192 points in f2 and 512 points in f1. Inductively Coupled Plasma (ICP) analyses were completed at the University of Florida Geology Department by Dr. George Kamenov. Elemental analyses were conducted by the University of F lorida, Department of Chemistry spectroscopic services. MALDI mass spectra were acquired by Dr. Tracy D. McCarley at Louisiana State University with a Bruker Daltonics Ultraflextreme MALDI TOF/TOF mass spectrometer operated in linear mode. The matrix used was trans 2 [3 (4 tert butylphenyl) 2 methyl 2 propenylidene]malononitrile (DCTB).
70 2.3.2 Polymer Molecular Weight Characterization Gel permeation chromatography (GPC) was performed at 40 C using a Waters Associates GPCV2000 liquid chromatography system wi th an internal differential refractive index detector and two Waters Styragel HR i.d., 300 mm length) using HPLC grade THF as the mobile phase at a flow rate of 1.0 mL/min.Calibrations were done against narrow molecular weight polystyrene standards (Polymer Laboratories; Amherst, MA). Alexander Pemba is acknowledged for his efforts in taking the GPC measurements for this research. GPC measurements in trichlorobenzene at 135 o C were performed at the Max Planck Institute for Polym er Research, Mainz, Germany using a refractive index detector with calibrated against polystyrene standards. 2.3.3 Thermal Characterization Polymer thermal stability measurements were taken using thermogravimetric analysis (TGA) on a TA Instruments TGA Q10 00 Series in nitrogen. The TGA samples (2 mg) were typically heated to 25 C to equilibrate to a constant mass, and then heated at a heating rate of 10C/min to a maximum temperature of 600C. 2.3.4 Electrochemical Characterization Electrochemical measureme nts were performed in a three electrode electrochemical cell consisting of a counter electrode, reference electrode and working electrode immersed in a supporting electrolyte solution. A platinum wire or a Pt flag, acted as the counter electrode; Ag/Ag + ac ted as the reference electrode. Two different types of working electrode were used depending on the application. The working electrode was either a platinum button (0.02 cm 2 ) electrode purchased from Bioanalytical Systems when used for CV and DPV measureme nts, or an ITO coated
71 glass slide (7x50x0.7 mm, sheet resistance, Rs 8 Technologies when used for spectroelectrochemistry. Corrware software used to control a potentiostat/galvanostat from EG&G Princeton Applied Research model 273 A. The polymer films were solution drop cast onto the Pt button electrode and air brush sprayed on the ITO coated glass slide from 1 mg/mL polymer solution in toluene. For CV measurements, the working electrode was cycled 5 times at a 50 mV / s scan rate to break in the films before doing the electrochemical characterization. The supporting electrolyte was 0.1 M TBAPF 6 /PC solution and it was used for all CV, DPV and scan rate dependent CV measurements. The electrolyte solution was purged with solvent saturate d argon for 30 min before electrochemical measurements and a blanket of argon was present at all times. For electrochemical characterizations, cyclic voltammetry (CV) and differential pulse voltammetry (DPV) were employed. CV relies on ramping the applied potential at a working electrode in both forward and reverse bias while measuring the current. The important parameters to note in CV are cathodic and anodic peak potentials (E pc E pa ) and the peak currents. (i pc i pa ). The peak separation between the cat hodic and anodic peak potentials (E pc E pa ) p is important in terms of characterization of the electron transfer processes whether they are electrochemically reversible or irreversible. In DPV measurements, a series of discrete potential amplitudes (10mV 100mV) is pulsed stepwise. After each pulse, the potential goes back to a slightly higher voltage which is called step size (1 2mV) resulting in a staircase shape. The current is measured before the application of pulse and at the end of the pulse and the difference in currents is correlated to the applied potential. By doing so, the capacitive or charging current is not
72 measured which is still present in CV measurements. Thus, DPV curves become more symmetric with a higher signal to noise ratio, resulting in better defined onsets of oxidation and reductions. DPV has been used to determine the HOMO and LUMO levels of the samples. The oxidation onset and reduction onset are related to HOMO and LUMO levels respectively. As all measurements are referenced to t he Fc/Fc + redox couple, the energy levels can be calculated by adding 5.1 eV to the oxidation and reduction onsets relative to the vacuum level. Even though some sources in the literature use 4.8 eV for Fc oxidation potential, this discrepancy has been w ell discussed in the dis sertation of Barry C. Thompson 158 and recently by Bazan and coworkers 159 2.3.5 Optical Spectra Characterization UV VIS absorption spectra were obtained from a Varian Cary 500 Scan UV VIS near IR spectrophotometer using 1 cm x 1 cm x 5.5 cm quartz cells manufactured by Starna Cells, Inc. Solution spectra were measured in dilute polymer solutions in toluene or chloroform to give an optical de nsity of 0.8. Thin film spectra were recorded by spraying a hot polymer solution in either chloroform or toluene on ITO glass to such a thickness that would yield an optical density of 0.8. 2.3.6 Spectroelectrochemistry Spectroelectrochemistry has been of ten employed in our research group to calculate the optical band gap, the energies of the polaron and bipolaron states and investigate the electrochromic behavior of polymers under redox processes. For spectroelectrochemistry experiments the air brush spra yed polymer films on ITO were immersed in quartz cuvettes of 1 cm path lengths containing 0.2 M LiBTI/PC electrolyte solution. Copper tape was used to make contact between the ITO electrode and the
73 potentiostat connection cables. The electrolyte solution w as purged with argon for 30 min before electrochemical measurements were taken, and the film was cycled 10 times with a 50 mV/s scan rate using CV measurements. 2.3.7 Two Dimensional Wide Angle X Ray Scattering (2D WAXS) Structural characterizations by 2D WAXS have been performed at the Max Planck Institute for Polymer Research. The experiments have been conducted with the help of Dr. S. Reddy Puniredd. In 2D WAXS experiment, a polymer fiber is placed in the field of an X ray beam to study the level of org anization in the material. The fiber preparation process is show n in Figure 2 3a. The polymer fibers are prepared by filament extrusion process using a home built extruder, in which polymer powder is heated at a point where it can undergo plastic deformati on. Then, the polymer is extruded through the nozzle by a constant motion of the piston inside the cylinder. A number of parameters control this mechanical processing such as the velocity of the piston, diameters of the cylinder and the shape of the die. T he extruder is capable of processing polymer fibers with a 0.7 mm diameter. (Figure 2 3b) The extruded fiber is placed inside the X ray chamber perpendicular to the incident X ray beam along with its extrusion direction. As seen in Figure 2 3c, the positi on of reflections in the X ray pattern give information about the level of organization in the polymer. Two pieces of information can be obtained from the resulting X ray pattern as shown in Figure 2 3c. The outermost reflections indicate stacking and the innermost reflections show the intermolecular distances or lamellar distances. The intensity of the stacking reflections corresponds to the crystallinity of the polymer.
74 Figure 2 3. Preparation and analysis of a poly mer via 2D WAXS. a) Schematic illustration of extrusion process. (Adapted with permission from M uellen, Copyright 2005 American Chemical Society) b)The extruded polymer fiber prepared and comparison with a US quarter. c) Representation of an extruded fib er and the aligned polymer chains along the extrusion direction as well as relating stacking and intermolecular distances in an obtained X ray pattern. The intermolecular distances (or chain to chain distance) are found in the equatorial plane of the s cattering pattern as the chains aligned are aligned perpendicular to the incident beam. The length and bulkiness of the side chains attached to the polymer backbone rules the extent of intermolecular distanced. Also, an amorphous halo can be seen in the X ray pattern which is related to the isotropic side chains. The 2D WAXS measurements discussed in this dissertation were performed by a rotating anode (Rigaku18 kW) X ray beam source with a pinhole collimation and a 2D a) b) b) c)
75 Siemens detector. A double graphite mo nochromator for the Cu nm) is used. 2.3.8 Grazing Incidence Wide Angle X Ray Scattering (GIWAXS) The presented GIWAXS data in the Chapter 4 and 5 have been obtained at the Max Planck Institute for Polymer Research, Mainz, Germany via a home built X ray source (a rotating Cu the substrate surface at or just below the critical angle for total reflection with respect to the incoming X ray beam (~0.2), The GIWAXS data were a cquired using a camera comprising an evacuated sample chamber with an X ray photosensitive image plate. The experiments were performed by myself and data analysis was carried out by Dr. Sreenivasa Reddy Puniredd. In the GIWAXS method, X rays impinge on the surface of the polymer film at a small grazing angle (below the critical angle typically 0.2 o ) and are scattered in plane and out of plane X rays are detected. 161, 162 The polymer thin films are deposited by drop c asting or spin coating and placed horizontally on the stage. The incidence angle ( ) rules the X ray penetration depth, scattering intensity and the nature of scattering. It can be adjusted so that only scattering from the film is obtained without the scat tering from the substrate. X rays can be scattered from the thin film as a result of the periodicity, repeat distance and orientation of the crystallites. Typical GIWAXS scattering patterns for conjugated polymers are shown in Figure 2 4. The essential pea ks obtained are the result of stacking which is around 0.4 nm and intermolecular spacing in the direction of side chains which is around 2.0 nm. The out of plane scattering which is the result of the spacing between periodic planes and the orientation of the vector is normal to th e
76 periodic planes give a pattern such t hat as illustrated in Figure 2 4 a. Figure 2 4. Analysis of GIWAXS patterns. a) Representation of the GIWAXS experiment and cartoons showing the X ray pattern for face on and edge on arra ngement on the substrate. b d) The intensity and the shape of the arc at the q z direction with the cartoons of representative ordering in the polymer chains Their reflections are in the q z direction, where q xy =0 indicate a face on orientation of the poly mer backbones on the substrate. In plane scattered X rays form an X ray pattern such as the one shown in Figure 2 5a, indicating an edge on orientation of the polymer backbones on the substrate. In this particular orientation, the side chains are standing
77 vertical on the substrate creating an insulated region between the polymer and the substrate. Furthermore, the intensity of the reflections are important in terms of determination of the anisotropy in the film. Figure 2 4b d shows the representative scenar ios where the reflection diffuse from a ring to a perfect spot in the q z direction where q xy =0. A ring is indicative of a totally isotropic film where crystallites have different orientations. A diffuse arc represents an isotropic film where crystallites h ave higher angles in between each other. A perfect spot shows almost anisotropic film in which crystallites are perfectly aligned with respect to each other indicating a preferred crystallization axis. In addition, continuous spots on the q z direction show the long range ordering in the film where planes of crystallites repeat each other. As a summary, a GIWAXS analysis can give valuable information including the orientation of the film on the substrate, the relative orientations of crystallites, the extent of long range ordering in the crystallites indicating anisotropy in a film, stacking and intermolecular distances. 2.3.9 Atomic Force Microscopy Atomic force microscopy was used to characterize surface morphologies of the end capped DTG BTD and non end capped DTG BTD polymers in Chapter 5. The experiments were done at the Ma x Planck Institute for Polymer Research. AFM images of the spin coated and drop cast thin films were recorded with a Multimode phase AFM NanoScope IV Scanning Probe Microscope Controller functioning in tapping mode. During the measurement, the cantilever t ip was kept stationary and oscillating at 300 MHz resonance frequency while the sample was moved while scanning. 2.4 Photovoltaic Devices All the solar cell measurements and photovoltaic device measurements are the Department of Material Science of
78 University of Florida. The data have been kindly provided by Dr. Jegadesan Subbiah and Song Chen. Organic solar cells were designed with the bulk heterojunction device architecture. For this purpose, a 30nm layer of po ly(3,4 ethylenedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS; Baytron AI 4083 from HC Starck) was spin coated on indium tin oxide (ITO) coated glass substrates which were ultrasonically cleaned and UV ozone treated, prior to the process. Substrates we re then baked on a hot plate at 180 C for 10 min. Next, onto the PEDOT:PSS, the active layer blend consisting of polymer and PCBM (99% pure, Solenne BV) was spin coated from chlorobenzene solvent with a thickness around 100 120 nm. Then, the sample was he ated on a hot plate at 70 C for 30 min. The cathode (LiF (1 nm) and aluminum (100 nm)) were then thermally evaporated under 1 10 7 mbar vacuum. The whole device has 0.04 cm 2 active area. For the inverted device architecture, ZnO (35 nm) obtained by sol gel process is spin coated onto ITO coated glass and then annealed in air for 30 min at 200 C. The active layer blend is pr ocessed and annealed as the same way mentioned previously. Later, PEDOT:PSS (35 nm) is spin coated onto the active layer and the components are annealed at 130 o C for 5 min. Lastly, silver metal (80 nm) is thermally evaporated onto PEDOT:PSS layer. The who le device has an active area of 0.04 cm 2 The solar cells are tested using a 150 W Newport ozone free xenon arc lamp as the light source in conjunction with a Keithley 4200 semiconductor parameter analyzer system. The measurements were carried out under 10 00 W/m 2 AM 1.5G illumination conditions. Solar cell devices are made in nitrogen in glovebox and characterizations are conducted in ambient atmosphere.
79 2.5 Organic Field Effect Transistors (OFETs) OFET devices were prepared at the Max Planck Institute for Polymer Research, device optimization and data analyses were performed by Dr. Sreenivasa Reddy Puniredd. All OFET devices discussed in this thesis were fabricated in the bottom gate, bottom contact architecture. The substrates used were highly doped Si act ing as the gate electrode with a 200 nm thick SiO 2 dielectric layer on top and was purchased from BASF. The SiO 2 layer was then reacted with hexamethyldisilazane (HMDS) overnight at 100 o C to minimize interfacial trapping sites. Polymer thin films were fab ricated by drop casting 2 mg/mL of a 1,2 dichlorobenzene solution onto FET substrates already heated at 100 C in nitrogen atmosphere. Next, the components were annealed at 200 C for 60 min. Additionally, polymer thin films were fabricated by spin coating from 10 mg/mL chloroform solution at 1200 rpm for 60 s. followed by annealing at 200 C for 60 min. were measured in nitrogen atmosphere and the average hole mobilities measured. All the charge carrier transport measurements are carried out by using Keithley 4200 SCS in a glove box under nitrogen atmosphere as shown in Figure 2 5 Figure 2 5 Picture of a bottom gate bottom contact OFET covered with a polymer.
80 CHAPTER 3 THE IMPORTANCE OF SY NTHESIS AND PROCESSI NG METHODS ON THE PERFORMANCE OF GREEN COLORED PHOTOVOLTAIC MATERIAL: PGREEN CASE 3.1 Introduction Recent developments in photovoltaics have focused on integrating conjugated polymers into these devices as the act ive light harvesting materials. Conjugated polymers are particularly attractive due to their potential for low cost, scalable deposition onto a variety of substrates. Much progress has been made to improve the power conversion efficiencies (PCEs) of polyme r based bulk heterojunction (BHJ) solar cells over the last two decades, with the highest performing devices achieving PCEs as high as 8.5%. 163 However, more advancements are still needed in the field according to one of the leading research teams, the Konarka group, which an nounced that to be 10% and above. 164 It is worth mentioning here that, development of the light harvesting materials in the academia is concomitant with the lab scale process ing techniques such as spin coating and most of the reported PCE values are recorded on very small substrates with active areas of less than 0.5 cm 2 Therefore, in addition to designing new light harvesting materials, advancements must be made in large sca le processing these materials, in order for polymer based solar cells to find a place in the commercial applications. In this regard, P3HT has been the benchmark polymer, yielding both a 6.5% PCE in lab scale devices when processed via spin coating, and 3% PCE when processed on a large scale by ink jet printing. 146 129 However, as P3HT has limited light coverage above 650 nm, alternative polymers with longer wavelength absorptions are
81 still being sought after. As such, one of the most i mportant criteria in obtaining high visible spectrum. 165 In this regard, the donor acceptor strategy has been particularly successful. In addition to considerations necessary for obtaining high PCEs, the color of the solar cells is often an important parameter in terms of increasing demand for commercial products. In this context, green colored polymers are desirable for aesthetically pleas ing polymer solar cells that can mimic the appearance of leaves or grass. However, green colored polymers are synthetically challenging to achieve. Pierre polymers with green hues. 166 Particularly, the PGreen polymer originating from his research, financially supported by Sest a r LLC, has been the starting point for the realization o cell characteristics are shown in Figure 3 1. 167, 168 The chemical structure of this polymer consists of alkylateddioxythiophenes, acting as donor moieties, and benzothiadizole acting as the acceptor unit. The combination of these units results in dual band absorption. Dioxythiophenes were functionalized with 2 ethylhexyl side chains to ensure solubility of the target polymer. Bithiophene spacers were employed in the repeat unit structure to fine tune the band gap and induce planarity by separating the dioxythiophenes, which have bulky side chains. In the solid state, PGreen shows a dual absorption maximum at 448 nm, nm, tailing off to 800 nm to cover the yellow red region of the visible spectrum. The
82 nm, which gives the polymer its perceived green color. When blended with PCBM in bulk heterojunction (BHJ) solar cells in a 1:8 ratio, PGreen achieved a 1.90% PCE in conventional device architectures (ITO/PEDOT/PGreen:PC 60 BM/LiF/Al). It is important to no te here that PGreen showed a PCE of only 0.64 % when bl ended with PCBM in a 1:4 ratio. Figure 3 1. Properties of PGreen. a) Design rationale and chemical composition of the PGreen with absorption profile in the visible spectrum resulting its green co lor. b) Solar cell characteristics of PGreen.(Adapted with permission from S o, C opyright 2009 American Chemical Society) The difference in PCE values within different mixtures was due to the higher content of PCBM, which is responsible for the increas ed short circuit current density (Jsc) from 1.99 to 5.56 mA/cm 2 In this c ontext we discuss some important challenges a) b )
83 that are being addressed by current photovoltaic research, with a focus how PGreen can resolve some of these issues. Scalability, reprodu cibility, polymer chemical quality and processing methods are the key factors that have been taken into consideration. Scalability is important, as large amounts of polymers are needed to process solar cells on the industrial scale. In this regard, though many groups have previously reported polymer yields of 100 500 mg, industrial applications require much larger amounts. Reproducibility depends on the strict control over synthetic protocols in order to generate smaller batch to batch differences between p olymerization products. For more information regarding important protocols for polymerization and the purification methods necessary for decreasing batch to batch variations, the reader can refer back to Chapter 1 and Chapter 2 of this dissertation. Lastly small scale solar cell preparations by spin coating and large scale roll to roll processes for processing polymer photovoltaics result in large discrepancies that need to be investigated. Through C hapter 3, we discuss the synthesis of PGreen on the gra m scale using oxidative polymerizations and compare the performance of solar cells prepared by spin coating and slot die coating. Further, effects of changing the polymerization method to Stille polycondensation in order to control the quality and purity ( reduce branching and controlling end groups) of PGreen and solar cell performances are discussed. Finally, a new APFO type polymer is discussed in an attempt to generate green colored polymers. This polymer, which ultimately allowed for electrochromic, lig ht emission, and photovoltaic activity, was chosen as an example of the careful blending of donor acceptor moieties
84 3.2 PGreen by Oxidative Polymerization PGreen was obtained from the oxidation of the monomer by dropwise addition of five equivalents of Fe Cl 3 over 2 hours, as shown in Figure 3 2. During the polymerization, dry air was passed through the reaction mixture to remove HCl generated by the reaction. The reaction was allowed to run overnight, then the oxidized polymer was reduced with hydrazine an d fractionated through washing with methanol, acetone, hexane and chloroform to remove any low molecular weight species. The chloroform soluble fraction was precipitated into methanol, affording the PGreen polymer. The reproducibility of this procedure was tested by running the polymerization with strict protocols by three different researchers using the same written synthetic protocol to achieve a standard operating principle. Additionally, the scalability of the reaction was investigated by conducting the polymerization on both1 and 2 gram scales. All four polymers presented expected CHN percentages in the 0.4% theoretical limit, indicating high quality polymers with few impurities. The molecular weights were determined usi ng a polystyrene calibrated GPC u sing THF as an eluent and the r esults are shown in Table 3 1. PGreen1 had a M n of 57.3 kDa in a 52% yield, in agreement with the reported values. PGreen2 showed a lower M n of only 37.3 kDa and a yield of 25% due to the fast addition of the oxidant (0.33 eq uivalents in 2 min.) because of a problem with the syringe pump used for adding the oxidant dropwise Additionally, the reaction yielded a large amount of insoluble polymer, as a result of crosslinking of the unsubstituted thiophenes during polymerization. PGreen3 had a still lower M n of 28.2 kDa due to the presence of hydrated FeCl 3 which was evident by its yellow color before the polymerization. Hydrated FeCl 3 has a lower reduction potential, thereby decreasing the extent of polymerization. Additionally, PGreen4 was run through a procedure most
85 similar to PGreen1, but with a 2 gram scale of polymerization, resulting in a M n of 48.4 kDa in a 60% yield. This implies that the polymerization procedure is scalable, while yielding reproducible molecular weight results. Figure 3 2.Synthesis of PGreen via oxidative polymerization. Table 3 1. GPC estimated molecular weights in THF and elemental analysis of the PGreen series prepared by different researchers. Researcher Polymer Scale (g) M n (kDa) Yiel d (%) PDI EA Calculated (%) C H N 66.49 7.65 2.87 Experimental (%) C H N Chad A. PGreen1 1 57.3 52 1.70 66.33 7.70 2.83 Mike C. PGreen2 1 37.3 25 1.91 66.39 7. 96 2.87 Unsal K. PGreen3 1 28.2 58 1.71 66.44 7.85 2.81 Unsal K. PGreen4 2 48.4 60 1.83 66.21 7.71 2.70 Sm all scale solar cells prepared by spin coating, have been tested by Franky creen printing provided by these groups. PGreen1, PGreen2 and PGreen3 were integrated into solar cells with an active area of 0.04 cm 2 The solar cells had the conventional architecture (ITO/PEDOT:PSS/PGreen:PCBM [1:8]/LiF/Al) as shown in Figure 3 3a, with the active layer deposited by spin coating. The illuminated J V characteristics of the bulk heterojunction (BHJ) solar cells based on each PGreen polymer (with a 1:8 ratio to
86 PCBM as the acceptor) are shown in Figure 3 3c and solar cell parame ters are summarized in Table 3 2 Firstly, the J V curves for solar cells prepared from each polymer are similar, thus indicating the reproducibility of the polymerizations. PGreen3 pro duced the highest PCE of 1.90 % despite its comparatively low molecular weight. PGreen1 produced a PCE value of 1.73 %, and PGreen2 produced 1.28 %. The discrepancies between the solar cell performances can be explained by the presence of defects during th e oxidative polymerization. It is known that some defect structures can form due to cross linking and coupling of monomers from positions during the oxidative polymerization. 71 In this regard, PGreen3, which resulted in low molecular weights due to the hydrated Fe(III)Cl 3 like ly experienced less defects in the polymer due to the lower reduction potential of the hydrated Fe(III)Cl 3 169 This also explains the low performance of PGreen2, which most likely had more defects in its structure due to the fast addition of the oxidant. Figure 3 3. Employed solar cell architectures and performance of PGreen series. a) Conventional solar cell architecture with its components. b) Inverted soalr cell ar chitecture. c) Solar cell characteristics of PGreen1(black), PGreen2 (red) and PGreen3 ( green) (Adapted with permission from R eynolds, Copyright 2012 American Chemical Society)
87 Table 3 2. Summary of solar cell performance parameters for PGreen series with d evice structure ITO/PEDOT:PSS/PGreen:PC 60 BM( 1:8 )/LiF/Al Thickness J SC (mA/cm 2 ) V OC (V) FF PCE (%) PGreen1 120 nm 5.16 0.75 0.45 1 .73 PGreen2 120 nm 4.56 0.77 0.37 1.28 PGreen3 120 nm 5.14 0.79 0.47 1.90 Next, PGreen3 was integrated into an inverted solar cell, which is the most suitable structure for roll to toll processing. This architecture yields more efficient charge collec tion, resulting in higher currents because of the favorable morphology of the active layer electrodes. 171 172 Inverted solar cells (ITO/ZnO/PGreen3:PCBM [1:8] PEDOT:PSS/Ag) resulted in a 1.98 % PCE with just a slight incr ease in the efficiency compared to conventional solar cells (1.90 %). The reason the improvement was so minimal can be attributed to the very high content of the PCBM in the active layer, which could prevent favorable vertical phase morphology. As PGreen3 had the highest PCE of for large area solar cell fabrication through roll to roll processing. The active layers (PGreen3 and PCBM) were deposited by slot die coating, and t he metal electrodes were fabricated by screen printing. First, single cells with active areas of 4.2 cm 2 were prepared in order to optimize the ratio of PGreen3 to PCBM for roll to toll processing. The variation of PCE values and solar cell parameters are shown in Figure 3 4. The polymer and PCBM were pumped differentially, and solar cells were tested immediately after the deposition. The highest PCEs were around 0.7 % 0.8 %, with a blend ratio of 1:1. (Figure 3 4a) When the PCBM content was kept between 30 % 80 %, the Voc and FF remained fairly constant, while an increase in the Jsc resulted in an increase in the PCE, as shown in Figure 3 4b. It is worth mentioning here that small scale inverted
88 solar cells with spin 98 % with a blend ratio of 1:8. The large difference between the PCE and the optimal blend ratios used in these two methods (roll to roll versus spin casting) can be attributed to differences in the active layer. Thicker films of 240 nm were used for slot die coating to provide mechanically robust, defect free coatings, as well as enhanced color intensity in the solar cells. However, thicker films resulted in less efficient charge collection, as charges buried inside polymer domains were heavily affected by recombination. Figure 3 4. Performance of PGreen3 in large area solar cells. a) PCE optimization in 4.2 cm 2 solar cells as a function of PCBM content in the active layer blend. b) The variation of short circuit current (ISC), open circuit voltage (VO C) and FF (%) as a function of PCBM content. (Adapted with permission from R eynolds, Copyright 2012 American Chemical Society) After optimization of the polymer to PCBM blend ratio, fifty 25x25 cm 2 modules, employing 12 serially connected cells, were prepared by roll to roll processing as shown in Figure 3 5b. The calculated active area (PGreen3 and PCBM) was 450 cm 2 adding up to 2.25 m 2 of total surface area in 50 modules. On a sunny Florida day, an open circuit voltage of 7.59 V was measured in one module. (Figure 3 5b) The PCE of each module was measured and is shown in Figure 3 5a. With the exception of modules 25 35 and 45 50, mod ules consistently displayed PCEs between 0.25 % and 0.3 %. The large decrease of PCE from 0.7 0.8 % in 4.2 cm 2 to 0.3 % in the large area modules can a ) b )
89 2 ) in the larger device due to the the many series r esistances across polymer/electrode interfaces. Figure 3 5. The performance of large scale processed blue green cells. a) PCEs measured for each module. b) One of the slot die coated, inverted solar cell module in a Florida day sunshine. (Adapt ed with permission from R eynolds, Copyright 2012 American Chemical Society) 3.3 PGreen by Stille Polycondensation Stille polycondensation was used for the synthesis of all PGreen samples in an attempt to minimize structural defects caused by coupling of unsubstituted thiophenes du ring oxidative polymerization. This method affords not only reliable bonding, but also in enhanced control over the nature of the end groups of the polymers. Thus, PGreen was prepared from the Stille polymerization of distanylated bithiophene and dibromothiophene benzothiadiazole thiophene monomers by using a Pd 2 (dba) 3 :P(o tol) 3 catalyst system via the scheme shown in Figure 3 6. Two different paths have been taken to control the end groups: The first includes mixing the difunctionalized monomers in a 1:1 stoichiometric balance and adding the monostannylated benzene and bromobenzene after the polymerization. The resulting product is labeled PGreenStille1. In the second method, 1% excess distannylatedbithiophene monome r was mixed with a dibromo monomer in a stoichiometric imbalance, and the bromobenzene was added to b ) a )
90 end cap the polymer after the polymerization. The end capped polymer is denoted as PGreenStille2. The polymers were obtained after precipitations into metha nol and washings with methanol, acetone, hexane and chloroform in a Soxhlet extractor. Chloroform fractions were precipitated into methanol, affording the polymers as black solids in a 65 % yield. The chemical purity was confirmed by elemental analysis, wh ich showed the percentages of CHN elements in the expected 0.4% theoretical value. The molecular weights of the polymers were determined by GPC in THF, calibrated against polystyrene standards. Results are given in Table 3 3. Figure 3 6. PGreen synthesis through Stille polycondensation. Table 3 3. GPC estimated molecular weights in THF and elemental analysis of the PGreen polymers obtained through Stille polymerization. M n (kDa) M w (kDa) PDI EA (Calcd/Found) THF THF C H N PGreenStille1 30.5 60.9 1.99 66.49/66.52 7.65/7.73 2.87/2.72 PGreenStille2 35.1 64.9 1.85 66.49/66.12 7.65/7.76 2.87/2.75 PGreenStille2 had a slightly higher number average molecular weight (35.1 kDa) than PGreenStille1 (30.5 kDa). This could be the result of making the ditin mono mer more available for the polymerization, as destannylation is commonly observed for Stille polymerizations limit ing the extent of reaction in the PGreenStille1. studied with mat rix assisted laser desorption/ionization mass spectrometry (MALDI MS)
91 MS and the spectra shows ions up to 12000 amu.(Figure 3 7) For each polymer repeat units (n=3 10) have been confirmed wit h their expected masses. The residual ions in the 4500 groups. Spectra are shown in the insets of Figure 3 7. Each polymer showed peaks with 165 amu around n=5, thus identifyin g the end groups as either hydrogen or bithiophene. This is indicative of the loss of tin groups during polymerization due to hydrolytic destannylation. Figure 3 7. MALDI MS of PGreenStille polymers. a) PGreenStille1 and b) PGreenStille2 Small scal e solar cells prepared by spin coating, have been tested by Jagedesian Subbiah in following data have been kindl y provided by them. Solar cells were prepared employing PGreenStille1 and PGreenStille2 in the conventional solar cell architecture (ITO/PEDOT:PSS/PGreenStille:PC 60 BM( 1:4 )/LiF/Al). The solar cell parameters obtained are shown in Table 3 4. Similar solar ce ll a) b )
92 thickness. It is worth noting here that, in the conventional cell architecture, the ac tive layer blend ratio was 1:4. The Stille products consistently required less PCBM to achieve an increased PCE of 2.1%. In previously reported solar cell performance, the oxidative polymerized PGreen had a blend ratio of 1:8 resulting and a PCE of 1.90 %. Additionally, when the active layer blend was 1:4, the PCE was only 0.64 %. A threefold increase was obtained with the new PGreen polymers afforded by Stille polycondensation. This can be attributed to the lack of structural defects (often caused by cross linking in other methods) and better defined end groups in the Stille products. Table 3 4. Summary of solar cell performance parameters for PGreenStille1 and PGreenStille2 with different active layer thicknesses in a device structure ITO/PEDOT:PSS/PGreenSt ille:PC60BM(1:4)/LiF/Al. a Reported solar cell performance for the PGreen obtained by oxidative polymerization employing 1:4 polymer to PCBM rat io in the donor acceptor blend. Thickness J SC (mA/cm 2 ) V OC (V) FF PCE (%) PGreenStille1 120 nm 5.82 0.75 0.46 2.02 PGreenStille1 135 nm 6.48 0.75 0.43 2.10 PGreenStille2 120 nm 6.27 0.73 0.47 2.14 PGreenStille2 125 nm 6.46 0.73 0.46 2.19 PGreen a 120 nm 1.99 0.78 0.41 0.64 to test the solar cell performance of the polymer. The solar cells had 4 cm 2 area of the active layer with the inverted solar cell architecture while the thickness of the active layer was 240 nm. The polymer to PCBM ratio was optimized, and the best performan ce was obtained with a PCBM content of approximately 80 %, resulting in 8. The ratio of the polymer to PCBM in the blend was consistent with the earlier optimized 1:4 blend ratio in the small scale solar cells. It is an intriguing fact is that a threefold decrease in PCE was obtained for PGreenStille
93 when compared to the previously discussed PGreen3 sample, showing PCEs of 0.7 0.8 % in this particular experiment. The reason for this does not likely stem from the purity o f the PGreenStille polymer, as no crosslinking is expected and the chemical elemental analysis demonstrated the high quality of this polymer. Further study is needed to confirm and justify these results. Figure 3 8. Performance of PGreenStille polyme r in large area solar cells. a) PCE optimization and FF (%) results in 4.2 cm 2 solar cells as a function of PCBM content in the active layer blend. b) The variation of short circuit current (I SC ) and open circuit voltage (V OC ) as a function of PCBM content 3.4 Conclusions on the PGreen Conjugated polymers require scalable and reproducible polymerization methods, as well as optimized printing methods to generate aesthetically pleasing solar cells. The PGreen polymer was studied for this purpose after its sy nthesis was scaled to 2 grams, while still maintaining strict synthesis protocols. Slight differences in synthesis have been shown to produce different molecular weights, but reproducible solar cell performances. However, solar cell performances were affec ted by changes in the processing method used. Solar cells prepared by spin coating resulted in higher PCE values than those prepared by slot die coating. Employing a different synthetic method, Stille polymerization, resulted in a different polymer to PCBM ratio in the active blend. b ) a)
94 Large area solar cells resulted in PCEs of 0.3 %. This is the first time a large scale fabrication of slot die coated solar cells has been employed using a polymer other than P3HT. 3.5 Utilization of APFO Type Polymers for Obtai ning Green Color In an effort to diversify our green colored polymer library, APFO type polymers structure is based on alternating fluorene with donor acceptor donor (D A D) u nits of thiophene and various acceptors. 36 30 A j udicious choice of the acceptor units in the repeat unit, green and purple colored polymers were achieved. Inganas and coworkers have studied these polymers, and showed it is possible to span the entire visible spec trum by controlling the dual band absorption through incorporating different acceptor moieties in the polymer backbone. Within the repeat unit structure of APFO type polymers, the fluorene moiety determines the mobility and HOMO levels, and also ensures th e solubility of the final polymer by means of the octyl side chains. Figure 3 9. Chemical composition of APFO copolymers with different donor and acceptor units employed in the repeat unit structure. ( Adapted with permission from R eynolds, Copyright 2011Royal Chemical Society ) The thiophene acceptor thiophene (D A D) unit is responsible for controlling the absorption profile and i ntensity, as well as tuning the HOMO and LUMO levels. In
95 particular, APFO3 polymers exhibited green colors and PCEs of 3.5 % when employed in BHJ solar cells. In our experiments, we changed the thiophene donor units to dialkylated dioxythiophenes in order to increase the solubility of our polymers and enhance molecular order between the polymer chains. Additionally, dialkylated dioxythiophenes raise the HOMO levels to lower the band gap of the polymers. The other components of the repeat unit were not chang ed, as dioctylfluorene moieties can tune charge carrier mobility and deep HOMO levels to provide moderate Voc for solar cell applications In the next sections, the new APF O DOT polymer is discussed. It was prepared by Suzuki coupling to afford a Mn of 34. 5 kDa with a 1.94 PDI. It showed purple to transmissive electrochromism, red electroluminescence and 1.66% PCE when used as the active layer in BHJ solar cells. 3.5.1 Polymer Synthesis and Characterization The APFO DOT polymer was synthesized via Suzuki polycondensation from the corresponding monomers using a Pd 2 (dba) 3 :P(o tol) 3 catalyst system with 3 M K 3 PO 4 as a base, as shown in Figure 3 10. After allowing sufficient time for polymerization, monofunctional benzene molecules were added to cap the polyme r. Precipitations and a soxhlet extraction protocol with methanol, water, acetone, hexane and chloroform were then conducted to remove the low molecular weight oligomers and inorganic salts. The chloroform soluble fraction was precipitated into methanol, a ffording the APFO DOT polymer as a purple solid with a 75 % yield and adequate solubility in common organic solvents at room temperature. The number average molecular weight was determined by GPC in THF as 34.5 kDa with a 1.94 PDI. MALDI MS analysis showed
96 the expected repeat unit mass of 919 g/mol, indicating the presence of dioctylfluorene and dialkoxythiophene BT dialkoxythiophene repeat units Figure 3 10. Chemical composition and synthesis of APFO DOT copolymer by Suzuki polycondensation. (Adapted wi th permission from R eynolds, Copyright 2011Royal Chemical Society) The absorption and emission characteristics for APFO DOT are sh own in Figure 3 11. Absorption studies were performed in dilute solutions and thin films. Thin film absorption measurements (black, triangles) exhibited a ca 10 nm bathochromic shift compared to the solution spectrum, due to stacking. Also, a characteristic dual band absorption profile was observed for the thin film spectrum due to donor acceptor interactions in the main chain, with maxima at 390 nm and 555 nm, tailing off to 750 nm while leaving a minimum at 445 nm. The c olor of the polymer was perceived as purple in solution and in thin films because of the low absorbance in the blue ( ca 425 450 nm) and red ( ca 650 750 nm) regions. Furthermore, films of the polymer on glass substrates showed a broad photoluminescence ba nd in the red to near IR region with a maximum at 730 nm (as shown by the blue crosses in Figure 3 11a). Electroluminescence and OLED measurements have been kindly provided by Ken Graha m. Considering the bright photoluminescence in the red to near IR regio n, we
97 Figure 3 11. Absorption and emission characteristics for APFO DOT. a) Absorption (in solution: circles, in film: triangles) and emission (in solution: diamond, in film: crosses) spectra of APFO DOT. b) Electrolumnescence spectrum of a PLED based on APFO DOT at 12 V applied bias. ( Adapted with permission from R eynolds, Copyright 2011Royal Chemical Society ) The PLEDs were fabricated (ITO/PEDOT:PSS/APFO DOT/Ca/Al) by depositing the polymer via spin coating. Sufficiently bright visible red light was obtained from the PLEDs. The electroluminescenc e spectra of the PLEDs is shown in Figure 3 11b. Devices exhibited strong red to near IR electroluminescence, with an emission maximum at 742 nm, thus resulting in an appearance of a deep red color, with a significant proportion of the emission appearing i n the near IR region. The PLED characteristics are shown in Figure 3 12. The PLEDs demonstrated a maximum radiant emittance of nearly 3mW/cm 2 with external quantum efficiencies (EQEs) between 0.2 and 0.3 % at current densities between 20 1000 mA/cm 2 This result places the APFO DOT polymer as one of the highest performing materials for PLED applications in the 700 800 nm range. APFO DOT showed oxidation and reduction processes observed through electrochemical c haracterizations by CV and DPV. (Figure 3 13ab ) CV measurements a) b )
98 reveal the presence of redox processes between 1.8 V to 1.0 V. A peak to peak separation of 0.16 V is obtained with an E 1/2 of 0.68 V versus Ag/Ag + Figure 3 12. PLED properties of APFO DOT. a) Current density and radiant emittance as a function of applied bias for APFO DOT based PLED. b) The evaluation of EQE as a function of current density. ( Adapted with permission from Reynolds, Copyright 2011Royal Chemical Society ) The small peak to peak separation is further studied by scan rate dependence studies between 5 mV/s and 50 mV/s (Figure 3 13c). A linear correlation between the scan rate and anodic peak current c onfirms that the redox processes are surface bound. Next, DPV was used to estimate the HOMO LUMO levels from the onsets of oxidation and reduction. The oxidation onset was 0.40 V (versus Ag/Ag + ), thereby bringing the HOMO level to 5.5 eV, assuming that th e Fc/Fc + redox couple is 5.1 eV relative to the vacuum level. The reduction onset was 1.47 eV, placing the LUMO level at 3.6 eV relative to the vacuum level. The band gap of the APFO DOT polymer, measured simply by the taking the difference between betw een these HOMO and LUMO energy levels, was found to be 1.87 eV. The electrochromic behaviour and band structure were studied with spectroelectrochemical measurements as shown in Figure 3 13d. During these experiments, the APFO DOT polymer films, spray cast onto ITO coated glass in an argon purged 0.2 M LiBTI/PC supporting electrolyte solution, were oxidized stepwise,
99 and the absorption spectra were recorded. The voltage was increased slowly from 0.5 V to 0.82 V versus Ag/Ag + during oxidation, while holding the potential for 5 min at each voltage increment. Figure 3 13. Electrochemical characterization of APFO DOT. a) Cyclic voltammetry of APFO DOT in 0.1 TBAPF6/PC with a 50 mV/s scan rate. b) Differential Pulse Voltammetry of APFO DOT in 0.1 TBA PF6/PC.c) Scan Rate Dependant Cyclic Voltammetryof APFO DOT in 0.1 TBAPF6/PC. d) The spectroelectrochemistry of APFO DOTfilms coated on ITO glass in 0.2 M LiBTI/PC supporting electrolyte between 0.5 V and 0.82 V versus Ag/Ag+ reference electrode. Inset sh ows photographs of neutral and oxidized states. (Adapted with permission from R eynolds, Copyright 2011Royal Chemical Society) Befo re oxidation, when the polymer was in its neutral form, at 0.5V, two absorption maxima at 562 and 391 nm with a trough at 447 nm resulted in a perceived a) b ) c ) d )
100 purple color (the purple thick line in Figure 3 13d). The absorption onset was used to estimate the op tical band gap as 1.78 eV, in agreement with electrochemical measurements. Upon oxidation, the intensity of the bands in the visible region decreased and two new bands appeared in the near IR range of the spectrum, indicating the formation of polarons and bipolarons. With a further increase of the potential up to 0.82 V, the two bands in the visible region spontaneously bleached, rendering a transmissive film, as is des irable for display applications. APFO DOT was also employed as the active material in BHJ solar cells in the hope of achieving high Voc characteristics are shown in Figure 3 14 with the conventional device architecture using an APFO DOT and PCBM blend. Figure 3 14. Solar cell characteristics of APFO DOT processed from chloroform (black) and chlorobenzene (blue)(Adapted with permission from R eynolds, Copyright 2011Royal Chemical Society) Chlorobenzene and chloroform were used as solvents for spin coating of the blend; higher results were obtained with chlorobenzene. The solar cells showed a Voc of 0.87 V, short circuit current density (J SC ) of 4.38 mA/cm 2 and fill factor of 0.43,
101 resulting in a PCE of 1.66 %. The low J SC is attributed to the large domains of the polymer and PCBM, which decreased the efficiency of charge separation. 3.5.2 Conclusions An attempt to obtain green colored ele ctroactive polymers by employing APFO polymers resulted in purple colored polymers due to weak absorption in the blue and red region of the visible spectrum. However, the APFO DOT polymer based on dialkoxythiophenes showed purple to transmissive electrochr omism upon oxidation and strong red to near IR emission in the range of 700 800 nm, making it a suitable material for dual EC/EL devices. Dual EC/EL devices are attractive for display applications that can operate in various lighting conditions. Lastly, th e polymer showed a high V OC of 0.87 V with a moderate photovoltaic performance, with a PCE of 1.66%. In summary, though APFO polymers did not yield the green color initially targeted by these experiments, they may be suitable for other application such as EC/EL displays or photovoltaic devices. 3.6 Experimental Details 4,7 bis(5 bromo 3,4 bis ((2 ethylhexyl)oxy)thiophen 2 benzo[c][1,2,5]thiadiazole, 167 5,5' bis(trimethylstannyl) 2,2' bithiophene, 174 ,2,2' (9,9 dioctyl 9H fluorene 2,7 diyl)bis(4,4,5,5 tetramethyl 1,3,2 dioxaborolane), 30 have been synthesized according to the reported in the literature. PGreen1 168 4,7 bis(3,4 bis((2 ethylhexyl)oxy) [2,2' bithiophen] 5 yl)benzo[c][1,2,5]thiadiazole (1.00 g,1.02 mm ol) was transferred from a storage vial to a smaller vial by spatula. This was then rinsed into a 250 mL round bottom flask with chloroform (HPLC grade, stabilized with 50 ppm pentene), and then more chloroform
102 was added, so that the total volume of chloro form added was 170 mL. The fl ask was placed in a room temperature water bath, which was maintained at a temperature of reaction was then covered by aluminum foil to limit exposure to ambient light, which was only removed periodically to check the status of the reaction. To this solution was added 4.6 mL of a 1.1 M solution of FeCl 3 in nitromethane via syringe pump. The pum p was set at 2.3 mL/h at a syringe diameter of 13 mm, and the addition took 2 h and 20 min to complete. The reaction was then stirred for 21 h and 40 min longer. After completion of the reaction, the mixture was poured into 500 mL methanol and stirred vigo rously for 10 min. The reaction was then filtered on a course paper filter, and washed with 100 mL methanol. The methanol used for washing was then discarded, and a clean flask was placed under the filter. The paper was then punctured to allow solids to fl ow into the flask below, and the solids were washed down with 350 mL chloroform, yielding a dark suspension with a significant amount of soluble material. Hydrazine monohydrate (10mL) was then added, and the mixture was stirred for 2 h at room temperature. The mixture was then concentrated to approximately 150 mL (rotary evaporation at room temperature), and the mixture was pipetted into 350 mL of methanol. The resulting solid was filtered onto a cellulose thimble, and extracted (via Soxhlet extraction) wit h methanol (16 h), acetone (12 h), dichoromethane (12 h), and chloroform (until the extract was clear, 6 h). The chloroform soluble fraction was then cooled to room temperature, and 5 mL of hydrazine hydrate solution (80% in water) was added and stirred for 2 h. This fraction was then concentrated to 150 mL (by rotary evaporation, at
103 room temperature), pi petted into 350 mL of methanol, and the resulting solid was placed under vacuum ( 0.1 Torr) for 2 days to remove solvents, resulting in 520 mg (52%) of a dark solid. (PG reen1) Elemental Anal.Calcd %: C, 66.49; H, 7.65; N, 2.87. Found: C, 66.33; H, 7.70; N, 2.83. 1H NMR: 8.45 (bs, 2H), 7.34 (bs, 2H), 7.16 (bs, 2H), 1 H NMR spectra were not significantly dif ferent between batches PGreen1 PGreen4. PGreen2 was synthesized using a procedure identical to the one used to preparePGreen1, except that 0.3 mL of the Fe(III)Cl 3 solution was added over 2 min and the remainder added over 1.75 h. The temperature was maint (25%). Elemental Anal. Found: C, 66.39; H, 7.96; N, 2.87. PGreen3 was synthesized using identical procedures as PGreen1, except that the solid Fe(III)Cl 3 used was yellow in color (hydrated), and was dispersed in nitromethane, but not all of the 4.5 g used for the stock solution could be dissolved. Elemental Anal. Found: C, 66.44; H,7.85; N, 2.81. PGreen4 was synthesized using an identical procedure to the one used to synthesize PGreen1, except the amounts of th e reagents and solvents were doubled. Elemental Anal. Found: C, 66.21; H,7.71; N, 2.70. PGreenStille1 A 100 mL flame dried Schlenk flask was charged with 4,7 bis(5 bromo 3,4 bis((2 ethylhexyl)oxy)thiophen 2 yl)benzo[c][1,2,5]thiadiazole (0.422 g, 0.434 mmo l) and 5,5' bis(trimethylstannyl) 2,2' bithiophene (0.213 g, 0.434 mmol), Pd 2 (dba) 3 (0.016 g, 0.017 mmol) and P(o tol) 3 (0.010 g, 0.034mmol). The mixture was cycled with argon and vacuum three times. After addition of degassed toluene (25 mL), the reactio n was heated to 90 o C for 3 days. This was followed by the addition of
104 trimethyl(phenyl)stannane (0.010 g, 0.043 mmol).The reaction was stirred for 24 hours. Then, bromobenzene (0.011g, 0.065 mmol) was added, and the reaction was stirred for another 24 hou rs. The reaction was cooled to room temperature and precipitated into methanol (500 mL). The precipitates were collected into a thimble and the polymer was washed with Soxhlet extraction using methanol, acetone, hexane and chloroform. The chloroform fracti on was concentrated to 50 mL and diethylammonium diethyldithiocarbamate was added. After stirring for 2 hours under argon, the mixture was filtered over a 0.45 m size polypropylene membrane onto methanol (350 mL). Precipitates were collected and dried und er vacuum for 2 days to afford 0.277 g (65 %) of PGreenStille1 Elemental Anal. Found: C, 66.52; H,7.73; N, 2.72.The H NMR data obtained for PGreenStille1are similar to those obtained for PGreen1. For PGreenStille2 the synthesis protocol was similar except the distannylated monomer was added in 1% excess. Yield: 300 mg (65%) Elemental Anal. Found: C, 66.12; H,7.76; N, 2.75. 4,7 Bis (5 bromo 3,4 dipropoxy thiophen 2 yl) benzo[1,2,5]thiadiazole (2) In a 100 mL two neck flask 4,7 Bis (3,4 dipropoxy thiophen 2 yl) benzo[1,2,5]thiadiazole (0.35 g, 0.64 mmol) was dissolved in chloroform (20 mL) and N bromosuccinimide (0.25 g, 1.4 mmol) was added in portions under a flow of argon. The reaction mixture was stirred at room temperature for 24 hours, and aluminum foil was used to shield the reaction from exposure to light. The reaction mixture turned from dark orange to ared bright color. The organic layer was washed with water and dried over magnesium sulfate. After concentration under vacuum, the product was purified via column chromatography (1:1 = Hexanes: Dicholoromethane) to yield 0.397 g (90 %) of a red oil. 8.43 (s, 2H), 4.10 (m, 8H), 1.81 (m, 8H), 1.04 (m, 12H). 13 C NMR (75MHz, CDCl 3 ):
105 152.28, 148.19, 147.46, 127.22, 123.78, 121.14, 100.57, 75.43, 74.98, 23. 39,10.65, 10.55 Elemental Anal.Calcd for C 10 H 16 O 2 S: C, 59.96; H, 8,05; found: C, 59.16; H, 8.53 Poly[9,9 dioctyl 2,7 9H fluorene alt 4,7 Bis (5 bromo 3,4 dipropoxy thiophen 2 yl) benzo[1,2,5]thiadiazole (APFO DOT) A 100 mL schlenk flask was charged with 4 ,7 Bis (5 bromo 3,4 dipropoxy thiophen 2 yl) benzo[1,2,5]thiadiazole (0.225g, 0.326 mmol), 9,9 dioctyl 2,7 (0.210g, 0.326 mmol), tri orthotolyl phosphine (4 mg, 0.0130 mmol), Pd 2 (dba) 3 (6.8 mg, 0.00652 mmol), potassium phosphate (3.18 g, 3M in 5 mL H 2 O) and a drop of aliquat 336. The flask was then evacuated and backfilled with argon several times. Toluene (5 mL) and water (5 mL) were then added under argon atmosphere, followed by stirring at 90 o C for 4 days The reaction mixture turned from red to dark purple over the course of the reaction. 1 bromobenzene (0.034 mL, 0.405 mmol) was added and stirred for 1 hour, followed by the addition of 1 phenylboronic acid (0.049 mg, 0.405 mmol). After stirring for a fur ther 1 hour, the reaction was precipitated into methanol (300 mL). Precipitates were filtered into a cellulose thimble and then washed with methanol, water, acetone, hexane and chloroform via a Soxhlet extractor. The chloroform soluble fraction was precipi tated into methanol and precipitates were collected using a 20 m pore size 47 mm diameter nylon membrane. The precipitates were washed with copious amount of water and then re dissolved in chloroform, then the strongly complexing agent (diethylammonium di ethyl dithiocarbamate) was added in order to remove any residual catalyst. After stirring for 2 hours, the mixture was filtered over a 0.45m size polypropylene membrane into methanol (300 mL). The precipitates were collected and dried under vacuum overnig ht affording 225 mg (75 %) of a purple solid. 1 H NMR (300
106 MHz, CDCl 3 ): : 8.52 (s, 2H), 7.92 (s, 2H), 7.78 (dd, 4H), 4.09 (m, 8H), 2.18 0.65 (m, 54 H) Mn= 34,567 g/mol, Mw= 67,391 g/mol, PDI: 1.94 Elemental Analysis for repeat unit C 55 H 70 N 2 O 4 S 3 : Calcld: C, 71,85; H, 7,67; N, 3,05; found C, 70.95; H, 8,05; N, 2,80
107 CHAPTER 4 THE INFLUENCE OF THE BRIDGING ATOM IN THE FUSED THIOPHENE BASE D DONOR ACCEPTOR POLYM ERS ON CONTROLLING M OLECULAR ORDER AND CHARGE TRANSPORT 4.1 Introduction The donor acceptor strategy, a method which involves alternating electron rich and electron poor aromatic rings in conjugated polymers, has been effective in controlling important properties of these materials such as band gap, stacking, light absorption, and charge transport in electroactive polymers. For examples, refer to the C hapter 1 of this dissertation. These properties are desirable for high performance, low cost, processable, organic electronic application s. In particular, an understanding of structure property relationships in conjugated polymers is of great importance when generating rational design rules for the development of polymers for optoelectronics. As donor moieties, fused thiophene aromatic ring s bridged by carbon (cyclopentadithiophene, CPDT) and silicon (dithienosilole, DTS) have been previously employed in the repeat units of donor acceptor type polymers in OPVs and OFETs. (Figure 4 1 and Figure 4 2) Fused thiophenes offer enhanced conjugation by forcing planarity of two thiophene units, as well as through the attachment of solubilizing chains to the bridging atom. As such, alternating copolymers of 2 ethylhexyl substituted CPDT and benzothiadizole (BDT) have been reported by Brabec et al., whi ch showed particular promise for OPV applications due to their long wavelength absorption profile. 34 These polymers showed PCEs exceeding 5% and later 5.5% by Heeger and coworkers, who employed solvent additives to optimize the polymer:PCBM blend microstructure. 175 Additionally, high charge carrier mobilities of 0.02 cm 2 /V.s have been achieved by Brabec group with these polymers. 176 Additionally, Muellen and coworkers
108 have shown that their charge carrier mobilities could be increased to 0.17 cm 2 /V.s when n hexadecyl side chains were incorporated onto the CPDT moiety to increase the length of the alkyl chains, thereby improving solubility and molecular order. This was as evident by the measured stacking distances of 3.7 between the two polymer chains. 177 Another approach was taken by Janssen group, where the acceptor moieties were changed from BDT to four different acceptors as shown in Figure 4 1. 178 This structure property relationship study allowed them to vary the light absorption from 800 nm (for PCPDT:Q) to 1200 nm (for PCPDT:TP). The band gaps of the polymers were tuned between 1.82 eV and 1.34 eV. From the five polymers, PCPDT: BO showed the highest PCE of 2.5 % when blended with PCBM. Figure 4 1.Various cyclopentadithiophene based copolymers and their chemical composition. Following this work, new polymer designs were developed by changing the bridging atom to silicon, in pla ce of the carbon atom, as outlined in Figure 4 2. Initiated firstly by Marks and coworkers, work to develop copolymers of dithienosilole with thiophene units resulted in hole mobilities up to 0.06 cm 2 /V.s. 179 The motivation behind
109 interactions to yield enhanced transpo rt properties. Inspired by this work, Beaujuge et al. carried out a structure property relationship study using a series of dithienosilole benzothiadiazole donor acceptor polymers as shown in Figure 4 3. 180 Given th e structures, the study examined the extent of stacking in the polymers in pursuit of achieving long wavelength absorption and high charge transport properties. Among the five polymers, dithienosilole substituted with 2 ethylhexyl chains showed the hig hest molecular order, which is reflected in the OFET hole mobility of almost 0.1 cm 2 /V.s and 4.6 % PCE obtained when blended with PCBM and integrated into OPVs. In general, it was found that polymers with high molecular weights and appropriate choice o f alkyl chains (2 ethylhexyl versus n octyl) achieved high performance results. Indeed, Yang Yang, employed 2 ethylhexyl substituted DTS with a BTD acceptor based donor acceptor polymer having an Mn of 18 kDa and yielded 5.1 % PCEs in solar cell devices. 26 This was followed by substitution of the rigid thienopyrroledione acceptor in the place of benzothiadizole, which allowed to polymer to then be functionalized with alkyl chains. Solubility of the polymers was improved with this new acceptor, resulting in molecular weights of up to 28 kDa. 1 81 Leclerc and coworkers achieved a PCE of 7.3 % after optimization of the polymer:PCBM with solar cell parameters: Voc = 0.88 V, Jsc = 12.2 mA/cm 2 and FF = 0.68. 181 Yongfang Li and coworkers studied the effec ts of the position and nature of the alkyl chains on the planar thiazolothiazole (TTz) acceptor. 182 The copolymers of DTS with TTz showed longer wavelength absorptio n when n hexyl side chains were moved from the R1 to the R position due to reduced steric crowding. This allowed a 5.88 %
110 PCE for devices made with this polymer, compared to the 5.58 % of PCE for devices using the polymer with n he xyl chains at the R1 posi tion. Figure 4 2.Copolymers based on dithienosilole and their chemical compositions. Additionally, when 2 ethylhexyl side chains were substituted in place of n hexyl the hole mobility increased to 0.03 cm 2 /V.s. 183 To conclude all the work done on dithienosilole and cyclopentaditihiophene, Scharber and coworkers did a careful structure property relationship study to understand the effect of substituting a silicon atom in place of a carbon atom. 184 As seen in Figure 4 3, they compared the crystallinity of the 2 ethylhexyl functionalized CPDT BTD polymer and DTS BTD. They found that when a silicon atom was substituted for the bridging atom, the resulting polymers
111 displayed higher molecular ordering. CPDT BTD resulted in an amorphous polymer whereas DTS BTD showed a pronounced peak at 5.2 o due to intermolecular interact ions, and another small peak at around 20 o due to the stacking. This higher tendency to achieve molecular order is a result of the longer C Si bonds in the fused ring system. Figure 4 3. X ray diffractograms of donor acceptor polymers of cyclopentadi hiophene and dithiensilole. (Adapted with permission from B rabec ) Continuing down along the Group 14 elements, the Ge atom bridged fused thiophenes, dithienogermole (DTG), has recently attracted a lot of attention in the polymer research community. As sh own in Figure 4 4, a number of donor acceptor type polymers based on DTG with various acceptors have been developed in the past year. Our group has reported a copolymer with a high molecular weight of 47 kDa, consisting of 2 ethylhexyl functionalized DTG w ith a TPD acceptor, that resulted in a 7.3 % PCE
112 when blended with PCBM in an inverted solar cell architecture. 107 The polymer with a Mn of 16 kDa afforded a 4.1 % PCE for Leclerc and coworkers. 185 When copolymerized with a BTD acceptor, PCEs of 1.2 % and 2.9 % we re obtained by Kunugi 185, 186 Additionally, Fujita and coworkers have shown the synthesis of a DTG based copolymer with dithieno bithiaziadizole and thiazolothiazole acceptors res ul ting in PCEs of 1.5 % and 2.4 % respectively. 187 Figure 4 4.Dithigermole based donor acceptor copolymers reported in 2011. One reason why DTG is so attractive is due to the longer C Ge bond (1.9 9 ) when compared to the C Si (1.87 ) and C C bonds (1.53 ). 107, 188 (Figure 4 5a) The motivation for work on these polymers is that by taking advantage of the longer C Ge bond, which can push the solubilizing chains away from the aromatic ring system
113 interchain interactions between the aromatic units may be improved. Heeney and coworkers have studied the peripherally alkylated dithienogermole based polymers illustrated in Figure 4 5b. 189 The y compared these Ge atom bridged polymers with Si bridged polymers and performed theoretical calculations on the length of the C Ge bond to confirm a slightly longer bond length for C Ge than for C Si bonds in this particular kind of polymer. Figure 4 5. The literature examples studied the C Ge bond lengths. a) Comparison of theoretical model structures dithienosilole and dithienogermole aromatic units. (Adapted with permission from R eynolds ) b) Peripherally alkylated dithienogermole based copolymers. In all of these cases, understanding str ucture property relationships is crucial to the design of high performance polymers. In this c ontext we expand upon previous DTG research by changing the bridging atom from carbon in CPDT to a silicon atom in CPDT, and further to a germanium atom in DTG. For this purpose, six polymers have been designed and synthesized as shown in Figure 4 6. DTS and DTG donor units were combined with acceptor units including BTD, TPD and Pthalimide (PT ) to
114 investigate the effect of heavy atom substitution on absorption b ehavior, HOMO LUMO energy levels, molecular packing and charge transport properties. As a general rule, to ensure solubility in the final polymers, DTS and DTG monomers were functionalized with linear n dodecyl alkyl chains, which also direct long range or dering in the solid state. We found that DTG polymers consistently have higher hole mobilities, closer stacking distances, slightly longer absorptions in the visible region and higher HOMOs when compared to DTS based polymers. Figure 4 6. Repeat unit structures for the discussed n dodecyl functionalized dithienogermole and dithienosilole based donor a cceptor copolymers in C hapter 4 4.2 Polymer Synthesis and Characterization The synthesis of n dodecyl functionalized DTG and DTS monomers is outlined in Figure 4 7. Briefly, the synthesis starts with the addition of n dodecyl magnesium bromide onto sil icon or germanium tetrachloride, yielding dichlorodidodecylsilane
115 /germane ( 1ab ) on a 15 gram scale after purification via high vacuum distillation. Particularly, 1.7 eq. of Grignard reagent was added to decrease the possibility of substitution of the thir d and fourth chlorides. Subsequently, compound 1ab was added onto ditlithiated 3 between 3 BuLi 78 o C, affording DTS and DTG monomers. Next, these monomers were lithiat ed and quenched with trimethyltin chloride to obtain distannylated monomers. Purification of these compounds was carried out by passing the material through a plug of silica pretreated with triethylamine to remove the excess tin salts. As the elemental ana lyses results varied slightly from the theoretical values, purification of these distannylated monomers using reverse phase HPLC was attempted. However, due to the dodecyl chains, the molecules were too apolar, and formed aggregates on the non polar column Nevertheless, the structure of these compounds and the absence of monostannylated side products were confirmed by 1 H NMR as shown in Figure 4 8. DTS and DTG monomers have two doublets at 7.20 ppm and 7.05 ppm due to the protons at the 2 and 3 positions r espectively on the thiophene ring. After distannylation, the doublet at 7.20 ppm disappears completely (as shown with the black arrow in Figure 4 8), implying the successful completion of the reaction. After the distannylated DTS and DTG monomers were obt ained, Stille polycondensation reactions were performed with these monomers and the dibromobenzothiadiazole (BTD) ,1,3 dibromo 5 octylthieno [3,4 c ]pyrrole 4,6 dione (TPD) and diboromo phthalimide using a Pd 2 (dba) 3 /P(o tol) 3 catalyst system. One percent ex cess of the distannylated DTS and DTG monomers were mixed with dibromo monomers and the reactions were run at 90 o C for 3 days.
116 Figure 4 7. Schemes for the synthesis of n dodecyl functionalized dithienosilole and dithienorgermo le monomers and the six do nor acceptor type polymers
117 Figure 4 8. Comparison of 1 H NMR spectra of the distannylated monomers and their precursors. a) dithienosilole and distannylated dithienosilole, b) dithienogermole and distannylated dithienogermole. Black arrows show the abs ence of Ha protons. Subsequently, 4 iodotoluene was added in excess to end cap the polymers, and the reaction was allowed to run for another 20 hours. The polymerizations were cooled down to room temperature and precipitated into methanol. Solids were coll ected in a thimble and Soxhlet extractions with methanol, acetone, hexane and chloroform were carried out. A Pd scavenger was added onto the chloroform soluble fraction, and the polymers were collected from precipitation into methanol, affording polymers i n yields of 65 70%. All the polymers dissolved in toluene, THF and chloroform at room temperature. The molecular weights of the polymers were determined by GPC measurements using polystyrene standards in THF. Results are shown in Table 4 1. The polymerizat ion procedure using 1% excess ditin monomer yielded reproducible, high molecular weight polymers with M n greater than 20 kDa (except PSi12 TPD and PSi12 PT, which both had an average number of repeat units higher than 20). PGe12 BTD resulted in the highest number average molecular weight of the polymers studied, with a M n of 33 kDa and an average of 46 repeat units. Additionally, all polymers
118 exhibited the expected amounts of C, H and N in the 0.4% purity limit as shown in Table 4 1. Table 4 1. GPC estimat ed molecular weights in THF and elemental analyses of the polymers. M n (kDa) M w (kDa) Average # of Repeat Units PDI EA (Calcd/Found) THF THF C H N PSi12 BTD 28.5 59.8 43 2.10 68.83/68.97 8.21/9.30 4.22/3.66 PGe12 BTD 33.2 81.0 46 2.43 64.49/64.64 7.6 9/8.38 3.963/3.54 PSi12 TPD 16.6 35.6 21 2.14 69.91/70.22 8.54/9.24 1.77/1.65 PGe12 TPD 25.1 67.0 30 2.66 66.18/66.76 8.09/9.04 1.68/1.59 PSi12 PT 15.9 26.5 20 1.67 73.32/73.64 9.10/9.74 1.78/1.70 PGe12 PT 20.6 43.0 25 2.09 69.19/69.08 8.40/8.56 1.72/1 .61 The repeat unit structures of the polymers have been confirmed by 1 H NMR and matrix assisted laser desorption/ionization mass spectrometry (MALDI MS) measurements. 1 H NMR data showed the expected aromatic and aliphatic proton chemical shifts for the polymer repeat units (See Experimental details). Next, we investigated the presence of toluene end groups. Toluene was selected as the end group due to the characteristic proton chemical shift (~2.2ppm) of the methyl group attached to the benzene ring. Sma ll peaks were observed around 2.3 ppm, overlapping with broader peaks from the methylenic protons in the dodecyl chain. To circumvent this problem, in Chapter 5, PGe12 BTD is discussed specifically for the confirmation of the toluene end group by NMR measu rements at 100 o C in deuterated tetrachloroethane solvent with a 500 MHz instrument. Figure 4 9 shows the selected MALDI mass spectra of PGe12 BTD, PSi12 BTD, PGe12 TPD and PSi12 TPD. Using trans 2 [3 (4 tert butylphenyl) 2 methyl 2 propenylidene]malononit rile (DCTB) as the matrix, ions up to 10,000 u were detected using MALDI MS. Higher mass ions were not observed due to
119 absence of desorption from the matrix. For each polymer, ions that correspond to oligomer series (n=2, 3, 4, 5, 6) were observed. The io ns in each series were separated appropriately to indicate the expected polymer repeat units. Due to the complex nature of step growth polymerization, a number of end groups were obtained. In Chapter 5, end group analysis on PGe12 BTD is discussed in deta il. Figure 4 9. MALDI MS of donor acceptor polymers a)PGe12 BTD, b) PSi12 BTD, c) PG12 TPD and d) PSi12 TPD recorded using DCTB as matrix. 4.3 Optical and Electrochemical Characterization The normalized solution and thin film absorption spect ra obtained for PGe12 BTD, PSi12 BTD, PGe12 TPD, PSi12 TPD, PGe12 PT and PSi12 PT polymers are b ) a) c ) d )
120 shown in Figure 4 10, and their respective absorption parameters are depicted in Table 4 2 Figure 4 10. Normalized UV VIS absorption profiles for PGe12 BTD (black), PSi12 BTD (gray), PGe12 TPD (red), PSi12 TPD (magenta), PGe12 PT (blue) and PSi12 PT (cyan) a) in dilute toluene solution and b) in thin film. a ) b )
121 Table 4 2. Absorption parameters for polymers both in toluene and thin film with their calculated band gap energy values from the onset of absorption in thin film Polymer abs (nm) in Toluene abs (nm) Thin film abs (nm)Onset In Toluene abs (nm)Onset Thin Film Eg (eV) PGe12 BTD 684,420 703(772),427 875 878 1.41 PSi12 BTD 667,414 685,423 854 869 1.42 PGe12 TPD 618,675,(430) 615,682,(430) 741 744 1.66 PSi12 TPD 608,664,(425) 610,676,(425) 732 736 1.68 PGe12 PT 490 497 576 623 1.99 PSi12 PT 487 496 576 623 1.99 In general, the absorption profiles show a strong dependence on the nature of the accepto r unit employed in the repeat unit. A bathochromic shift occurred when phthalimide was replaced by thienopyrroledione and subsequently by benzothiadizole, with the magnitude of this shift increasing as the strength of the acceptor increased. Additionally, with the exception of PGe12 PT and PSi12 PT, DTG containing polymers with thienopyrroledione and benzothiadiazole exhibited slightly red shifted absorption, both in solution and in thin films, compared to DTS containing polymers. In particular, in the solu tion spectra of phthalimide based DTS and DTG polymers, one band at 487 nm and 490 nm was observed, which corresponds to transitions. Additionally, a trivial red shift, ca. 8 nm, was found in the absorption maxima for both polymers accompanied by a broadening of the absorption band in the thin film spectra. Incorporation of a stronger acceptor, thienopyrroledione, resul ted in a bathochromic shift in the spectra compared to the spectra of phthalimide polymers. Two low energy bands appeared around 615 and 682 nm, with a merged high energy band at 430 nm for PGe12 TPD in thin film samples. PGe12 TPD showed slightly (~8 nm ) red shifted onsets of absorption both in toluene and thin film samples, compared to PSi12 TPD. A stronger intramolecular interaction, due to the substitution of benzothiadizole for
122 thienopyrroledione, resulted in a bathochromic shift in the solution and th in film absorption maxima of DTS and DTG polymers. Benzothiadizole based polymers showed a dual absorption profile, in which the two bands were separated ca. 270 nm, leaving the blue green region of the spectrum uncovered. Additionally, PGe12 BTD exhibite d a slightly (~9 nm) red shifted absorption onset compared to PSi12 BTD both in toluene and thin film samples. These results are consistent with the analysis in the TPD based DTS and DTG polymers. Therefore, we conclude that DTG based polymers show increas ed aggregation tendency compared to DTS based polymers due to the longer C Ge bonds, which push the solubilizing chains away from the aromatic units. Thus, the band gaps measured from the onsets of absorption in thin film samples resulted in slightly small er band gap values for DTG polymers with thienopyrroledione andbenzothiadiazole. Moreover, there was a general trend that benzothiadiazole based polymers produce the smallest band gap (1.41 and 1.42 eV) compared to thethienopyrroledione (1.66 and 1.68 eV) and phthalimide based polymers (1.99 eV). The redox processes in these polymers have been studied with cyclic voltammetry (CV) and differential pulse voltammetry (DPV) techniques. Firstly, all polymers showed oxidation and reduction processes in the 0.8 V to 1.9 V voltage range, as determined by CV experiments. The results are summarized in Table 4 3. The effect of substituting a Ge atom in the place of the Si atom, as the bridging atom, has been thoroughly investigated with DPV measurements, as shown in F igure 4 11.DTG containing polymers were found to have lower oxidation potentials than DTS containing polymers; the oxidation potentials of PGe12 BTD and PSi12 BTD differed by 270 mV.
123 This difference reduced to 115 mV in thienopyrroledione based polymers an d 110 mV in phthalimide based polymers. Figure 4 11. Electrochemical characterization of DTS and DTG polymers. a) Differential pulse voltammetry ((step size 2mV, step time38 ms, pulse amplitude 100 mV)for thin films of PGe12 BTD (black), PSi 12 BTD (gray), PGe12 TPD (red), PSi12 TPD (magenta), PGe12 PT (blue) and PSi12 PT (cyan)under a) positive voltage bias b) negative voltage bias c) The range of HOMO LUMO energy levels estimated by CV and DPV measurements for the DTG&DTS series. The optical ly estimated band gaps of each polymer are placed at the center of the DPV estimated band gaps and a second approximated HOMO and LUMO levels are defined assuming the energy levels are equidistant from the center. These are highlighted in gray rectangle ar eas. a) b ) c )
124 Table 4 3.Estimated HOMO LUMO energy levels and band gaps for each copolymer by cyclic voltammetry and differential pulse voltammetry. Polymer E OX (V) E RED (V) EHOMO (eV) ELUMO (eV) Eg (eV) CV DPV CV DPV CV DPV CV DPV CV DPV PGe12 BTD 0.17 0.18 1 .52 1.41 5.27 5.28 3.58 3.69 1.59 1.59 PSi12 BTD 0.56 0.47 1.56 1.48 5.66 5.57 3.54 3.62 2.12 1.95 PGe12 TPD 0.61 0.58 1.60 1.50 5.71 5.68 3.50 3.60 2.21 2.08 PSi12 TPD 0.70 0.70 1.58 1.54 5.80 5.80 3.52 3.56 2.28 2.24 PGe12 PT 0 .76 0.76 1.74 1.74 5.86 5.86 3.36 3.36 2.50 2.50 PSi12 PT 0.85 0.87 1.76 1.65 5.95 5.97 3.34 3.45 2.61 2.52 Thus, it can be commented that Ge atom substitution facilitates oxidation and raises the energy of the HOMO levels between 0.1 0.3 eV Additionally, DPV measurements show that the nature of the acceptor in the repeat unit causes the HOMO levels to become deeper. For example, a stronger acceptor, benzothiadizole was used to form PGe12 BTD with a HOMO level at 5.28 eV, while using a weak acceptor such as pht h alimide resulted in PGe12 TPD with a HOMO level at 5.68 eV and PG e12 PT with a HOMO level at 5. 86 eV. These results are in agreement with the estimation of HOMO energy levels from CV measurements. While this is the case for oxidatio n behaviors, the effect of the nature of the acceptor is even more drastic in reduction potentials. The reduction potentials decrease in this order: benzothiadiazole>thienopyrroledione>phthalimide. Thus, the LUMO levels are lower for benzothiadiazole based polymers than for thienopyrroledione or phthalimide based polymers because benzothiadiazole is a stronger acceptor. Ge atom substitution also resulted in slight changes in the LUMO levels of these polymers. PGe12 BTD yielded a 0.07 eV deeper LUMO energy c ompared to PSi12 BTD, and a 0.04 eV deeper level than thienopyrroledione based DTG and DTS polymers. Ge atom incorporation in the fused aromatic ring raised the HOMO energy and lowered LUMO energy levels of the
125 polymers. For these two reasons, DTG based po lymers resulted in lower band gaps compared to DTS polymers with the same acceptor unit. The band gap reduction is depicted in Figure 4 11. DPV measurements of PSi12 BTD estimate the band gap to be 1.95 eV. A drastic reduction (0.36 eV) in the band gap is observed for PGe12 BTD (E g = 1.59 eV). The band gap reduction is on the order of 0.16 eV for thienopyrroledione, and 0.02 eV for phthalimide based polymers. These observations are in agreement with the band gap results obtained from the onsets of thin film absorption spectra, in which DTG based polymers yielded slightly lower band gaps. It should be noted that in general, electrochemical methods resulted in higher band gaps than optical measurements, a trend commonly observed in donor acceptor type polymers 190 4.4 2D WAXS and GIWAXS Characterization The molecular organization of our polymers has been studied in bulk materials with 2D WAXS and in thin films with GIWAXS methods. X ray patterns are shown in Figure 4 12 a nd Figure 4 13. In addition, the structural information derived from these measurements is shown in Table 4 4. We find that phthalimide based DTS and DTG polymers are amorphous, as no reflections for stacking or lamellar organization are evident in the extruded fiber 2D WAXS patterns. For these polymers, there are no distinct features in the X ray pattern, and only a vague halo is present due to the alkyl chains. (Figure 4 12e and Figure 4 12f) The lack of interactions can be attributed to the distortion from planarity of the thiophene benzene bond angle caused by steric congestion between the hydrogen on the thiophene ring and the hydrogen on the benzene unit. While this is the case for phthali mide based polymers, stacking and lamellar organization of the BTD and TPD based polymers are observed as highlighted
126 in Figure 4 12 The outermost reflections due to stacking, indicated by blue arrows, are more intense for BTD based polymers and br oader and more diffuse for TPD based polymers. Thus, calculated stacking distances are smaller for BTD based polymers (3.5 ) compared to the TPD (3.6 ) based ones. This minute decrease in the stacking distance can be explained by the presence of alkyl chains on the TPD moieties, which reduce the non covalent interactions between aromatic units. In addition to closer stacking, BTD based polymers yield closer interchain distances as well. This is again due to the presence of additional alkyl cha ins, which fill some of the extra space between the lamellar stacks in TPD polymers. With this in mind, it can be concluded that the BTD acceptor moiety with no alkyl chains showed increased the propensity for stacking when compared to the TPD acceptor uni t with alkyl chains. The effect of Ge atom substitution is more obvious in the GIWAXS patterns for thin films, as shown in Figure 4 13. PGe12 BTD exhibits a large number of reflections in the q xy direction (highlighted by the yellow arrows in Figure 4 13a) compared to the PSi12 BTD polymer, which shows only one arc in the q xy direction. This can be attributed to increased ordering in the polymer chain stacks of PGe12 BTD due to the presence of the larger Ge atom in DTG, in which the C Ge bond is longer than the C Si bond in the PSi12 BTD polymer. Additionally, PGe12 TPD shows one arc in the q z direction, corresponding to a face on orientation of the polymer chains on the substrate. However, PSi12 TPD exhibits a strong amorphous halo and a broad arc in the q x y direction, indicating that there is less ordering between its polymer chains. As seen in Table 4 4, the organization parameters obtained from thin film measurements are in agreement with the bulk measurements. The stacking distances remain the same a nd chain to
127 chain distances generally increase slightly (except PGe12 BTD) in thin films when compared to mechanical extrusion process, due to solution processing on the substrate. Figure 4 12. Extruded fiber 2D WAXS patterns measured for DTS and DTG polymers. a) PGe12 BTD, b) PSi12 BTD, c) PGe12 TPD,d) PSi12 TPD, e) PGe12 PT and f) PSi12 PT at 30 o C after annealing the fibers at 200 o C. Yellow arrows show the reflections due to stacking and blue arrows highlight the reflections due to lamellar interactions The PGe12 PT and PSi12 PT patterns are shown in grayscale to increase the contrast. a) b) ) c ) d) e) ) f) )
128 Figure 4 13. GIWAXS patterns recorded for thin films of DTS and DTG polymers. a) PGe12 BTD, b) PGe12 TPD c) PSi12 BTD and d) PSi12 TPD. Yellow arrows highlight the important reflections due to molecular organization. Table 4 4.Structural information obtained from 2D WAXS (bulk) and GIWAXS (Thin film) for the DTS and DTG series. La mellar Spacing ( ) Stacking( ) Lamellar Spacing ( ) Stacking( ) Bulk Bulk Thin film Thin film PSi12 BTD 24.0 3.5 25.6 3.5 PGe12 BTD 23.5 3.5 22.4 3.5 PSi12 TPD 26.4 3.6 29.0 3.6 PGe12 TPD 25.6 3.6 27.5 3.6 PSi12 PT NA NA PGe12 PT NA NA
129 4.5 OFET and Sola r Cell Performances OFETs were prepared in bottom gate bottom contact architecture as shown in Figure 4 14 to investigate the effects of molecular ordering on the charge transport properties of the aforementioned polymers. The polymer solutions (2mg/mL in dichlorobenzene) were drop cast onto hexamethyldisilazane(HMDS) modified Si/SiO 2 substrates followed by annealing at 200 o C for 60 min. All polymers showed holes as the dominant charge carriers in the OFET channel when negative voltage bias applied to the gate electrode, as the polymers showed only a small leakage current under positive bias. The device performances are listed in Table 4 5.The hole mobility values were calculated from the saturation regime of the J V transfer plots for source drain voltages of V SD = 60 V with the equation mentioned in Chapter 1.Though upon first inspection an applied voltage of 60V seems high for practical display applications, the purpose of these experiments is to achieve a fundamental understanding of their charge trans port properties so that advancements can be made in the field for more practical devices. On this note it was observed that BTD based polymers had higher hole mobilities compared to TPD based polymers, correlating with the closer distances obtained th rough X ray methods. Secondly, DTG polymers showed increased hole mobilities compared to DTS ones due to longer carbon germanium bond lengths (compared to Si Ge bond lengths), which improved interchain interactions by pushing away the solubilizing alkyl ch ains. PGe12 BTD displayed a hole mobility of 0.6 cm/Vs, which is almost an order of magnitude higher than the mobility of 0.075 cm/Vs obtained for PSi12 BTD.This result makes PGe12 BTD competitive with the high performing polymers that were discussed in Chapter 1.In TPD based polymers, PGe12 TPD
130 exhibited a hole mobility of 0.0075 cm/V s, whereas the hole mobility of PSi12 TPD was 0.0032 cm/Vs. Surprisingly, PGe12 TPD had a hole mobility almost two orders of magnitude lower than the mobility of PGe12 BT D. This can be attributed to the unfavourable face on arrangement of the PGe12 TPD polymer chains on the substrate. In the bottom gate, bottom contact device geometry the face on orientation of TPD polymers resulted in misalignment of their high mobility axis with the channel path, thus impeding charge carrier transport between source and drain electrodes. As a note, not shown in T able 4 5, phthalimide based polymers showed hole mobilities on the order of 10 6 cm/Vs, due to their amorphous nature. Figure 4 14. Bottom gate bottom contact OFET architecture for charge transport measurements of the DTG and DTS polymers. Table 4 5. Summary of holemobilities and on/off ratios and solar cell performance parameters for DTS and DTG series OFET Solar Cells Hole Mobility (cm 2 /V.s) I on/off J SC (mA/cm 2 ) V OC (V) FF PCE (%) PSi12 BTD 0.075 10 5 4.82 0.62 0.36 1.09 PGe12 BTD 0.6 10 5 10 6 8.18 0.50 0.41 1.68 PSi12 TPD 0.0032 10 4 10 5 1.46 0.77 0.39 0.43 PGe12 TPD 0.0075 10 4 10 5 4.03 0.85 0.55 1.86 group.
131 The polymers were also employed in bulk heterojunction (BHJ) solar cells to evaluate the performance of DTG and DTS based polymers in light harvesting applications. BHJ solar cells were prepared using the inverted device geometry, similar to the ones discussed in Chapter 3. Devices were constructed with the following materials: indium tin oxide (ITO)/z inc oxide /active layer (Polymer PCBM) /molybdenum oxide/silver. In order to improve the morphology of the polymer: methanofullerene [6,6] phenyl C71 butyric acid methyl esters (PCBM) blend, diiodooctane (DIO) was used as a solvent additive during the proce ssing of the active layer. The solar cell performanc es were characterized using an A.M. 1.5G solar radiation source with an incident power of 100 mW/cm 2 .The polymers showed power conversion efficiencies (PCEs)between 0.43 % and 1.86 %. The related solar ce ll parameters are shown in Table 4 5. PGe12 BTD and PGe12 TPD demonstrated higher J SC values compared to their PSi12 analogues, as they have slightly longer wavelength absorption in the visible region and lower band gaps. PGe12 BTD, which had a higher HOM O level as measured by DPV measurements, exhibited lower V OC values compared to PSi12 BTD, in agreement with electrochemical measurements. Moreover, DTG based polymers demonstrated higher fill factor ( FF ) values, indicating relatively better charge generat ion. This observation correlates with the previously discussed OFET hole mobility results of the DTG polymers, as the FF is dependent on the charge carrier mobility In general, DTG and DTS based polymers resulted in poor PCEs (0.48% 1.86%) when employed in the active layers of the solar cells. One of the reasons for this may be the extensive phase separation (even visible to the naked eye) between DTG polymers and PCBM, indicative of an unfavorable blend morphology.
132 Larger domains of polymer were segreg ated from the fullerene phase, preventing excitons from reaching the polymer:fullerene interface for charge separation, thus resulting in a low external quantum efficiency. 4.6 Conclusions and Perspectives An understanding of structure property relationshi ps in conjugated polymers is important for the the design of new high performance polymers for optoelectronic applications. The interplay of the many parameters governing high output photovoltaic activity (such as low E g and deep HOMO levels) and high char ge carrier mobilities (such as increased molecular organization) also require special attention. In C hapter 4 synthetic efforts have resulted in six DTS and DTG based copolymers with BTD, TPD and PT acceptors through 1% stoichiometric imbalance of the precursors in a Stille polycondensation. The polymerizations produced polymers with molecular weights reproducibly higher than 15 kDa, with the highest weight polymers reaching M n up to 33 kDa. The six polymers were grouped into two subcategories, DTS and DTG based polymers, so that a clear trend could be obtained through the variation of the strength of the acceptor units, and for the investigation of the effects of the silicon and germanium atom placed as the bridging atom in the fused aromatic ring syste m. DTG polymers with Ge as the brid g ing atom resulted in a slightly red shifted absorption profile in the visible spectrum compared to DTS polymers. This is an important parameter in the realization of high performance photovoltaic materials, as PCEs are d irectly related to the solar cell J SC values. Additionally, the HOMO and LUMO energies were varied through the use of different acceptor units, so that lower band gaps could be obtained. Also, higher HOMO levels and deeper LUMO levels were observed for the DTG polymers compared to DTS. Thus for applications requiring low band gap materials, DTG can be used in place of
133 DTS Futhermore, the band gap was reduced from 2.52 eV for PSi12 PT to 1.59 eV PGe12 BTD. X ray measurements, which displayed improved molecu lar organization for DTG based polymers, confirm that when a heavy a tom such as Ge is introduced, stronger intermolecular interactions are observed. Improved molecular organization resulted in high charge carrier mobilities, achieving values up to 0.6 cm 2 /V.s. for the DTG based polymers. On the other hand, while the polymers performed well inOFET devices, their performance in solar cells was poor due to an unfavorable phase morphology of the active layer blend with PCBM. Lastly, it is worth noting her e that phthalimide based polymers suffered from possible twisting of the aromatic units due to steric congestion between the thiophene and benzene protons, which is routinely observed for polymers with similar chemical backbones. Thus, completely amorphous polymers were also obtained that displayed very low hole mobilities. 4 .6 Experimental Details dibromobithiophene, 179 4,7 dibromo( benzo[c][1,2,5]thiadiazole), 191 1,3 dibromo(5 octyl 4H thieno[3,4 c]pyrrole 4,6(5H) dione) 192 and4,7 dibromo(2 octylisoindoline 1,3 d ione) 193 have been synthesized accordi ng to the reported literature. Dichlorodi n dodecylsilane (1a) 194 In a 1L three neck flame dr ied flask, SiCl 4 (11 mL, 16.3g, 0.096 mol) was dissolved in 300 mL THF and cooled to 78 o C.Dodecylmagnesium bromide solution (250 mL, 0.68 M in diethyl ether, 0.170 mol) was cannulated into the above solution over 15 min. White solid precipitates were ob served immediately. The cooling bath was removed, and the mixture was stirred overnight at room temperature. Then the solvent was evaporated under reduced pressure and the crude product was suspended in hexanes (600 mL). Solids were filtered via suction fi ltration and the filtrate was evaporated under reduced pressure.
134 Crude yellow oil was distilled to afford the product as pale yellow oil (14 g, 34 %) at 190 o C under 75 mTorr. 1 H NMR (500 MHz, C 6 D 6 ): =1.6 1.4 (m, 4H), 1.4 1.1 (m, 36H), 1.0 0.8 (m, 10H). 13 C NMR (125 MHz, CHCl 3 ) = 33.12, 32.55, 30.34, 30.24, 30.10, 29.99, 29.88, 29.83, 23.51, 23.04,20.96, 14.80 Dichlorodi n dodecylgermane (1b) In a 1L three neck flame dried flask, GeCl 4 (7 mL, 13.09 g, 0.06 mol ) was dissolved in 300 mL THF and cooled to 0 o C.Dodecylmagnesium bromide solution (152 mL, 0.69 M in diethyl ether mL, 0.10 mol) was cannulated into the above solution over 20 min. White solid precipitates were observed immediately. The reaction was stirred for 4 hours at this temperature and the n allowed to stir overnight at room temperature. Later, the solvent was evaporated under reduced pressure and the crude product was suspended in hexanes (300 mL). Solids were filtered via suction filtration and the filtrate was evaporated under reduced pre ssure. Crude yellow oil was distilled to afford the product as pale yellow oil (18.23g, 63 %) at 190 o C under 94 mTorr. 1 H NMR (300 MHz, C 6 D 6 ): =1.6 1.4 (m, 4H), 1.4 1.1 (m, 36H), 1.0 0.8 (m, 10H) 4,4 didodecyl 4H silolo[3,2 b:4,5 b']dithiophene (2a) 195 In a 1 L three neck flame dried flask, n butyllithium (20 mL, 2.5M in hexanes, 0.05 mol) was dissolved in THF (400 mL) and cooled to 78 o C. 3,3' dibromo 2,2' bithiophene (8 g, 0.02 mol) in 50 mL THF was added dropwise into the n butyllithium solution over 20 min, resulting in a clear yellow solution. After the addition was finished, dichlor odi n dodecylsilane (10.76 g, 0.02 mol) in 50 mL THF was added dropwise over 40 min,. A large amount of white precipitates were formed. The cooling bath was removed and a water bath was placed so that the reaction would warm up to room temperature. The re action mixture was
135 further stirred for another hour at room temperature. The clear yellow solution turned to a brown solution over the course of the reaction. The reaction was quenched by adding a conc. NH 4 Cl solution in 300 mL of water. 300 mL of hexanes was added and the organic phase was extracted and then washed with brine. The organic phase was dried over anhydrous MgS0 4 Evaporationof the solvent yielded a brown oil as the crude product. This product was purified via column chromatography using hexanes as a solvent affording 2.87 g (22 %) of yellow oil. 1 H NMR (500 MHz, CDCl 3 ): =7.20 (d, 2H, J= 4.5 Hz), 7.06 (d, 2H, J= 4.5 Hz), 1.44 1.34 (m, 4H), 1.32 1.18 (m, 36H), 0.96 0.86 (m, 10H) 13 C NMR (125 MHz, CDCl 3 ) = 149.55, 141.99,130.00, 125.30, 33.53, 3 2.31, 30.02, 29.90, 29.74, 29.58, 24.57, 23.07, 14.50, 12.24 Elemental Analysis for repeat unit C32H54S2Si: Calcld C, 72.38; H, 10.25; S, 12.08; Si, 5.29; found C, 73.65; H, 12.02 4,4 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene (2b) In a 1 L three neck f lame dried flask, n butyllithium (25 mL, 2.5M in hexanes, 0.06 mol) was dissolved in THF (400 mL) and cooled to 78 o C. 3,3' dibromo 2,2' bithiophene ( 9.44 g, 0.029 mol) in 50 mL THF was added dropwise into the n butyllithium solution over 20 min, forming a clear yellow solution. This was followed by the dropwise addition of dichlorodi n dodecylgermane (14.13g, 0.029 mol) in 50 mL THF over 40 min, forming a large amount of white precipitates. The cooling bath was removed and a water bath was placed so th at the reaction would warm up to room temperature. The reaction mixture was further stirred for another hour at room temperature. The clear yellow solution turned to a brown solution over the course of the reaction. The reaction was quenched by adding a co nc. NH 4 Cl solution in 300 mL of water. 300 mL of hexanes was added
136 and the organic phase was extracted and then washed with brine. The collected organic phases were dried over anhydrous MgS0 4 and evaporation of the solvent yielded a brown oil as the crude product. This product was purified via column chromatography using hexanes as a solvent, affording 7 g (42 %) of yellow oil. 1 H NMR (500 MHz, CDCl 3 ): =7.22 (d, 2H, J= 4.5 Hz), 7.05 (d, 2H, J= 4.5 Hz), 1.50 1.43 (m, 4H), 1.32 1.18 (m, 40H), 0.89 (t, 6H, J= 7 Hz) 13 C NMR (125 MHz, CDCl 3 ) = 147.10, 142.98,130.10, 125.07, 33.11, 32.28, 30.11, 29.99, 29.97, 29.88, 29.71, 29.53, 25.92, 23.05, 14.74, 14.47 Elemental Analysis for repeat unit C32H54GeS2: Calcld: C, 66.78; H, 9.46; Ge, 12.62; S, 11.14; found C, 68.07; H, 10.93 4,4 didodecyl 2,6 bis(trimethylstannyl) 4H silolo[3,2 b:4,5 b']dithiophene (3a) 195 In a 100 mL flame dried Schlenk flask, 4,4 didodecyl 4H silolo[3,2 b:4,5 b']dithiophene (1 g, 1.88 mmol) was dissolved in heptanes (40 mL). I t was cooled to 0 o C, then freshly distilled TMEDA (0.65 g, 0.84 mL, 5.65 mmol) was added, followed by the dropwise addition of n butyllithium (2.26 mL, 2.5 M in hexanes, 5.65 mmol). The resulting brown colored solution was stirred for 2 hours at 0 o C. Sol id SnMe 3 Cl (2g, 0.01 mol, excess) was added quickly under argon flow. The reaction was stirred for another 2 hours at 0 o C, The color changed from yellow to light brown accompanied by the formation of a large amount of precipitates. Then, the solvent was e vaporated under reduced pressure and the reaction mixture was suspended in hexanes. The precipitates were filtered using a plug of silica pretreated with triethylamine. The solvent was evaporated under reduced pressure, and the product was kept under high vacuum for 4 days to afford 1.52 g (95 %) of brown oil. Further purification with reverse phase HPLC
137 was not successful. 1 H NMR (300 MHz, CDCl 3 ): =7.09 (m, 2H), 1.46 1.38 (m, 4H), 1.34 1.16 (m, 36H), 0.92 0.84 (m, 10H), 0.38 (m, 18H) 4,4 didodecyl 2,6 bi s(trimethylstannyl) 4H germolo[3,2 b:4,5 b']dithiophene (3b) In a 100 mL flame dried Schlenk flask, 4,4 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene (1.46 g, 2.55 mmol) was dissolved in heptanes (40 mL). It was cooled to 0 o C, then freshly distilled TMEDA (0.89 g, 1.14 mL, 7.5 mmol) was added, followed by the dropwise addition of n butyllithium (3 mL 2.5 M in hexanes, 7.5 mmol). The resulting brown colored solution was stirred for 2 hours at 0 o C. Solid SnMe 3 Cl (2g, 0.01 mol, excess) was added under argon flow, and the reaction was stirred for another 2 hours at 0 o C. Then, the solvent was evaporated and the reaction mixture was suspended in hexanes. The precipitates were filtered using a plug of silica pretreated with triethylamine. The solvent was evapora ted under reduced pressure, and the product was kept under high vacuum for 4 days to afford 2.2 g of brown oil. (97 %) 1 H NMR (500 MHz, CDCl 3 ): =7.09 (m, 2H), 1.54 1.46 (m, 4H), 1.34 1.16 (m, 40H), 0.89 (t, 6H, J= 7 Hz) 13 C NMR (125 MHz, CDCl 3 ) = 152.96, 144.39,138.07, 137.69, 33.24, 32.28, 30.06, 30.04, 30.02, 29.99, 29.73, 29.57, 26.01, 23.05, 14.88, 14.48, 7.32 Elemental Analysis for repeat unit C38H70GeS2Sn2: Calcld: C, 50.65; H, 7.83; Ge, 8.06; S, 7.12; Sn, 26.35; found C, 52.39; H, 8,87 General Experimental Procedure for Polymerizations:Poly[DONOR alt ACCEPTOR)] A 100 mL flame dried Schlenk flask was charged with carefully weighed DONOR (1.01 eq), ACCEPTOR (1 eq), Pd 2 (dba) 3 (4% eq.) and P(o tol) 3 (8 % eq.). The mixture was cycled with argon and vacuum three times. After addition of 20mL of degassed toluene, the reaction was heated to 90 o C for 3 days. This was followed by
138 the addition o f 4 iodotoluene (excess). The reaction was allowed to run for another 20 hours. The reaction was cooled to room temperature and precipitated into methanol (500 mL). The precipitates were collected into a thimble and the polymer was washed using Soxhlet ex traction with methanol, acetone, hexane and chloroform. The chloroform fraction was concentrated to 50 mL, and diethylammonium diethyldithiocarbamate was added. After stirring for 2 hours under argon, the mixture was filtered over a 0.45m size polypropyle ne membrane into methanol (350 mL). Precipitates were collected and dried under vacuum for 2 days to afford the title polymer. Poly5 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene) alt 1,3(5 octyl 4H thieno[3,4 c]pyrrole 4,6(5H) dione) (PGe12 TPD) Th e polymer was obtained as 0.275 g (65 %) of a black solid. 1 H NMR (500 MHz, CDCl 3 ): 8.60 7.40 (m, 2H), 3.74 (m, 2H), 1.8 0.80 (m, 65H) Mn= 28,500 g/mol, Mw= 59,800 g/mol, PDI: 2.10 Elemental Analysis for repeat unit C 46 H 67 GeNO 2 S 3 : Calcld: C, 66.18; H, 8.09; Ge, 8.70; N, 1.68; O, 3.83; S, 11.52 found: C, 66.76; H, 9.04; N, 1.59 Poly5 5 didodecyl 4H silolo[3,2 b:4,5 b']dithiophene) alt 1,3(5 octyl 4H thieno[3,4 c]pyrrole 4,6(5H) dione) (PSi12 TPD) The polymer was obtained as a black solid of 0.288 g. (67%) 1 H NMR (500 MHz, CDCl 3 ): 8.30 7.50 (m, 2H), 3.80 3.40 (m, 2H), 2.10 0.60 (m, 65H) Mn= 16,700 g/mol, Mw= 35,700 g/mol, PDI: 2.14 Elemental Analysis for repeat unit C46H67NO2S3Si: Calcld: C, 69.91; H, 8.54; N, 1.77; O, 4.05; S, 12.17; Si, 3.55 found: C, 70.22; H, 9.24; N, 1.65 Poly5 didodecyl 4H germolo[3,2 b:4,5 b']dith iophene) alt 4,7 ( benzo[c][1,2,5]thiadiazole) (PGe12 BTD) 0.240 g (68 %) of a black solid polymer was
139 obtained. 1 H NMR (500 MHz, CDCl 3 ): 8.40 7.70 (m, 4H), 2.20 0.60 (m, 50H) Mn= 33,200 g/mol, Mw= 81,100 g/mol, PDI: 2.43 Elemental Analysis for repeat unit C 38 H 54 GeN 2 S 3 : Calcld: C, 64.49; H, 7.69; Ge, 10.26; N, 3.96; S, 13.59 found: C, 64.64; H, 8.38; N, 3.54 Poly5 didodecyl 4H sil olo[3,2 b:4,5 b']dithiophene) alt 4,7 ( benzo[c][1,2,5]thiadiazole) (PSi12 BTD) The polymer was obtained as a black solid of 0.251 g. (70 %) 1 H NMR (500 MHz, CDCl 3 ): 8.60 7.60 (m, 4H), 2.20 0.60 (m, 50H) Mn= 25,200 g/mol, Mw= 67,100 g/mol, PDI: 2.66 Ele mental Analysis for repeat unit C 38 H 54 N 2 S 3 Si: Calcld: C, 68.83; H, 8.21; N, 4.22; S, 14.51; Si, 4.24 found: C, 68.97; H, 9.30; N, 3.66 Poly5 didodecyl 4H silolo[3,2 b:4,5 b']dithiophene) alt 4,7 (2 octylisoindoline 1,3 dione) PSi12 PT The polymer wa s obtained as a dark red solid of 0.175 mg. (69 %) 1 H NMR (500 MHz, CDCl 3 ): 7.92 (s, 2H), 7.95 (s, 2H), 3.73 (s, 2H), 1.73 0.86 (m, 65H) Mn= 15,900 g/mol, Mw= 26,600 g/mol, PDI: 1.67 Elemental Analysis for repeat unit C 48 H 71 SiNO 2 S 2 : Calcld: C, 73.32; H, 9.10; N, 1.78;found: C, 73.64; H, 9.74; N, 1.70 Poly5 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene) alt 4,7 (2 octylisoindoline 1,3 dione)PGe12 PT The polymer was obtained as a dark red solid of 0.357 g (65 %) 1 H NMR (500 MHz, CDCl 3 ): 7.95 (s, 2H), 7.85 (s, 2H), 3.73 (s, 2H), 1.71 0.86 (m, 65H) Mn= 20,600 g/mol, Mw= 43,100 g/mol, PDI: 2.09 Elemental Analysis for repeat unit C 48 H 71 GeNO 2 S 2 : Calcld: C, 69.19; H, 8.40; N, 1.72; found: C, 69.08; H, 8.56; N, 1.61
140 CHAPTER 5 KEY ROLE OF END GROU PS IN CONTROLLING MO LECULAR ORDER AND CHARGE TRANSPORT 5.1 Introduction C hapter 1 to this thesis describes some synthetic methods for controlling the molecular order of conjugated polymers and the resulting effects on their charge transport properties. In this sense, end group functionalization is an efficient method f or modify ing the physical and electronic properties of these polymers. However, this area of research has not received considerable attention due to the difficulty of end capping reactions for conjugated polymers. P3HT has been fortunate in this regard as there have been numerous reports of successful end capping reactions with various end groups, particularly by Grignard metathesis (GRIM) involving a chain growth mechanism. 54 57 Different functional groups such as RCO, 196 197 198 OH, 199 RCOOH, 200 201 have been incorporated as end groups. Additionally, Jeffries El et al. have suggested the addition of Grignard reagents of allyl, aryl, benzyl and vinyl to a growing chain of P3HT. 202 The presence of end groups is typically confirmed by MALDI MS and 1 H NMR techniques. After confirmation, the effects of these end groups on polymer performance in optoelectronic devices can then be evaluated. Chen et al have synthesized hydroxyl end group functionalized P3HT in order to induce H bonding between PCBM and P3HT layers. 199 After annealing, they observe a reduction in the size of PCBM crystals along wi th smoother surfaces of the P3HT:PCBM film when 5% of P3HT OH is added. When integrated into solar cell devices, this effect improved the power conversion efficiency to 4.06% from 3.14%. Kim et al. have designed P3HT end functionalized with hydroxyl, ethyl and perfluoro groups. 203 After confirming the presence of t hese end groups on P3HT with 1 H NMR and MALDI MS, they combined
141 these materials with PCBM and integrated the resulting blend into solar cells to analyze the effect of the end groups on the surface energy of P3HT. It was critical that the resulting end func energy in order to reduce phase separation in the P3HT:PCBM blend. P3HT end functionalized with perfluoro groups yielded almost the same surface energy as PCBM while other end groups afford ed higher surface energies. This effect when translated to CF 3 while others showed lower PCE values. Further, the McCullough group studied P3HT end functionalized with two bromine atoms instead of two bare hydro gens. 204 They observed that the bromine groups acted as charge carrier trapping sites, which decreased the OFET mobility and reduced the solar cell performance through exciton quenching. P3HT was also end functionalized with carboxyl groups to adsorb the polymer on TiO 2 in dye sensitized solar cells. 201 196 Besides P3HT, other polymers such as poly(meta phenylenes) (PM P) obtained by Suzuki polycondensation, have been studied as well. 205 Schluter and coworkers have suggested the slow addition of an AB type monomer onto a monofunctionalized end capping unit, before letting the polymerization proceed via a chain growth mechanism. This method, allowed them t o cap both bromo and boron chain ends with independent end groups as confirmed by MALDI MS measurements. In the context of donor acceptor type polymer synthesis, conv entional methods such as Suzuki and Stille coupling methods based on step growth type po lycondensations have been employed. Apart from concerns about the effects on the molecular weight and repeat unit structure, the nature of end groups has widely been overlooked for these kinds of polymers. Hence their effects on molecular structure and
142 ch arge transport have not been evaluated thus far. For the incorporation of end groups, methods have been suggested in C hapter 1 in context with methods for controlling the molecular weight via the Carothers equation in step growth type polyme rs. Following those, there have been few reports on the synthesis and evaluation of the nature of end groups on the donor acceptor type polymers obtained by these step growth type polymerizations. Leclerc 151 and Andersson 36 have reported the synthesis of c arbazole and fluorene based donor acceptor type polymers end capped with benzene moieties. However, there has been no attempt to confirm the presence of end groups on these polymers. Their end capping procedure involves the sequential addition of monofunct ionalized benzene moieties, bromobenzene and phenylboronic acid, after the polymerization. Recently, Bazan and coworkers have reported on the addition of monofunctionalized thiophene moieties following the Stille polycondensation. 109 They employed XPS to c onfirm the absence of reactive functional groups such as bromine and tin atoms. They claimed that the absence of these groups was indicative of successful end capping reactions. Soon after this report, the Marks group attempted the same end capping procedu re and XPS analysis. 110 They argued that the XPS analysis did not show any significant difference between the spectra obtained for the end capped and non end capped polymers. Consequently, there is still much speculation over the effectiveness of this anal ysis method In this c on text we discuss our efforts regarding end capping reactions for dithienogermole benzothiadiazole based donor acceptor polymers via a 1% stoichiometric imbalance approach. End capped dithienogermole benzothiadiazole, E(DTG BTD), has been functionalized with toluene end groups. Toluene was chosen to
143 ease the identification of end groups through taking advantage of the characteristic chemical shift of the toluene methyl protons in the 1 H NMR spectrum. A non end capped polymer with hydr ogen end groups, NE(DTG BTD), was synthesized as well to act a s a model. Differences between the end capped and non molecular order were properties. Toluene end groups were c onfirmed by 1 H NMR and MALDI MS techniques. Additionally, X ray analyses were conducted which illustrated the closer stacking distances for E(DTG BTD). Bottom gate bottom contact OFETs showed maximum charge carrier mobilities of 0.6 cm 2 /V.s. for E(DTG BTD), which is almost an order of magnitude higher than the value of 0.077 cm 2 /V.s. obtained for NE(DTG BTD). 5.2 Polymer Synthesis and Characterization Synthesis schemes are outlined in Figure 5 1, and the end capped DTG BTD copolymer synthesis is descri bed in Chapter 4. The control polymer, NE(DTG BTD) was synthesized in the same way : through Stille polycondensation of 1% stoichiometric imbalance of distannylated monomer to the dibromo monomer, but omitting the post polymerization end capping reaction. A fter the polymerizations, reactions were cooled to room temperature and precipitated into methanol. Collected solids were washed with methanol, acetone, hexane and chloroform in a Soxhlet extractor. Precipitation again into methanol afforded both polymers in mod erate yields of 68 70 % as black solids. The purity of the polymers was analyzed by elemental analysis. Both polymers yielded purity in agreement with the expected amount of C, H and N in the 0.4% theoretical limit, as shown in Table 5 1.
144 Figure 5 1. Reaction schemes for the synthesis of E(DTG BTD) and NE(DTG BTD) Molecular weights were determined by GPC using two different solvents and temperatures; in THF at 40 o C and trichlorobenzene (TCB) at 135 o C, and the results are summarized in Table 5 1. T he end capped polymer showed a M n of 33 kDa with a PDI of 2.43 in THF and a M n of 12.2 kDa with a PDI of 5.34 in TCB solvents. The non end capped polymer had a M n of 28 kDa with a PDI of 2.66 in THF and a M n of 11 kDa with a PDI of 6.48 in TCB GPC measure ments using TCB as the solvent showed broader molecular weight distributions compared to those using THF as the solvent. The difference of measured molecular weights between the two conditions might have originated from aggregation s formed when TCB was use d as the solvent for the elution process. Nevertheless, GPC yielded similar molecular weights for both polymers. Table 5 1. GPC estimated molecular weights in THF and TCB, and elemental analysis of the polymers. M n (kDa) M w (kDa) PDI M n (kDa) M w (kDa) PD I EA (Calcd/Found) THF THF TCB TCB C H N E(DTG BTD) 33.2 81.0 2.43 12.2 65.1 5.34 64.49/64.64 7.69/8.38 3.96/3.54 NE(DTG BTD) 28.2 75.0 2.66 11.0 71.5 6.48 64.49/64.16 7.69/8.10 3.96/3.77 Furthermore, the repeat units and the nature of the end grou ps have been examined by NMR experiments using C 2 D 2 Cl 4 as the solvent at 373 o K. Both polymers
145 showed the expected aryl and aliphatic proton chemical shifts, confirming the repeat unit structure. Toluene end groups are characterized by a distinctive chemic al shift of the toluene methyl protons near 2.3 ppm in 1 H NMR.The E(DTG BTD) polymer showed a peak at 2.38 ppm, indicating the presence of toluene methyl protons. This peak was absent in t he NE(DTG BTD) polymer spectrum ( as seen in Figure 5 2a) Next, to c onfirm the presence of toluene molecule s as the end groups on the polymer, HSQC and NOESY experiments were carried out. The 1 H, 13 C HSQC 2D NMR spectrum of E(DTG BTD) displayed a correlation pattern of the proton chemical shift at 2.38 ppm with a ca rbon che mical shift at 20.6 ppm (Figure 5 2b) This is the characteristic carbon chemical shift for a methyl carbon attached to a benzene molecule. Additionally, through space correlations between toluene methyl protons and toluene aromatic protons have been analy zed. The 1 H, 1 H NOESY spectrum displayed a correlation pattern between the 2.38 ppm peak and a peak at 7.22 ppm, which would be expected for benzene aromatic protons. (Figure 5 2c) It is worth noting here that these two correlation patterns were absent in the spectrum of the NE(DTG BTD) polymer HSQC and NOESY spectra. Furthermore, in the 1 H NMR spectrum of the E(DTG BTD), the area under the 2.38 ppm peak can be integrated and can be compared to the area under the peak corresponding to methyl protons in the n dodecyl chain, to estimate the number average molecular weight of the polymer. T he integration ratio and therefore the number of repeat units was found to be 40. Multiplying the number of repeat units with the repeat unit molecular weight resulted in a n M n of 28 kDa for the E(DTG BTD) polymer
146 Figure 5 2. NMR characterization of toluene end groups. a) 1 H NMR (500 MHz) of E(DTG BTD) and NE(DTG BTD) at 373 o K in C 2 D 2 Cl 4 between 2.2 2.5 ppm. b) 1 H 13 C 2D HSQC NMR ( 1 H (500 MHz), 13 C (125 MHz) of E(DTG BTD) at 373 o K in C 2 D 2 Cl 4 c) 1 H, 1 H 2D NOESY (500 MHz) of E(DTG BTD) at 373 o K in C 2 D 2 Cl 4 This value is quite close to the number obtained from the GPC analysis in THF as shown in Table 5 1. In this regard, using toluene end groups allows for direct estim ation of polymer molecular weights, whereas GPC results need to be calibrated against polystyrene standards. However, it should be taken into account that it is not possible to evaluate the extent of the polymer end capping through these and therefore MALD I MS data are needed. The MALDI MS results for E(DTG BTD) and NE(DTG BTD) are therefore the focus of the next section. Matrix assisted laser desorption ionization time of flight (MALDI TOF) mass spectroscopy was conducted to examine the
147 repeat units and en d gro ups on the polymers. (Figure 5 3 ) Firstly, MALDI TOF mass spectra displayed peaks corresponding to the expected oligomers in the range of n=4 to n=8, while higher molecular weight species were too heavy to be desorbed from the matrix. The oligomer pea ks were separated by 708 amu, consistent with the expected mass of the DT G BTD repeat unit. In Figure 5 3b and Figure 5 3 d, expanded regions of the mass spectrum of each polymer are displayed between m/z 2700 and m/z 3700 in order to evaluate the residual masses of oligomers and to identify the end groups Polymer structures are represented by illustrations over each peak in the MALDI spectrum. For the polymer E(DTG BTD), the three largest peaks in the region of n=4 have bee n assigned as seen in Figure 5 3 b The peak at m/z 2830 is consistent with the tetramer with two hydrogen end groups. The highest intensity peak at m/z 2922 is due to the tetramer with one toluene and one hydrogen end group. The peak at m/z 2966 is attributed to the tetramer with an addit ional BTD unit end capped with hydrogen. Careful observations reveal the complex nature of the step growth polymerization. Previously, Schluter and Janssen groups performed detailed end groups analysis for Suzuki polycondensation polymers in which they obs erve neither bromo nor boron end groups. 205 206 Accordi ng to their suggestions, these groups might have been lost during the Suzuki polymerization by various reasons including cyclization, li gand scrambling, dehalogenation and hydrolytic deboronation. Loss of functional groups such as halogen and tin atoms is also observed for Stille couplings by dehalogenation and destannylation. 207 In this context, possible com peting side reactions result in t he loss of functional groups as depi cted in Figure 5 4 A scenario is illustrated in which a starting AB type oligomer, such as compound 1 in Figure 5 4 can be assumed. It undergo es
148 successful end capping forming compound 2 and further monomer addition result s in longer pol ymer chains. Compound 1 can lose a bromine atom to yield compound 4 and then can follow three paths: it can form a dead chain by destannylation (compound 5) undergo another end capping reaction forming compound 6 or through continue d addition of the BTD monomer the polymer chain can grow longer T he oligomer residues at 2830, 2922 and 2966 m/z can be explained by s ide reactions leading to toluene and hydrogen end groups Additionally, hydrogen end groups prevent further monomer addition explaining why tol uene end capping is only ever observed at one end of the polymer Further, the pentamer peak has been assigned at m/z 3540 and three other peaks can be observed near it. The peak at m/z 3632 is assigned to a pentamer end capped with a toluene moiety. Th is is expected as the polymerization has been run for 1% in stoichiometric imbalance of the DTG ditin monomer. The peak at m/z 3496 represents a tetramer with an extra DTG unit end capped with toluene. Consistently, these oligomer residues undergo the afo rementioned side reactions causing the loss of functional groups during the reaction. Side reactions are also observed in the mass spectrum of NE(DTG BTD) polymer. As seen in the expanded mass spectrum in F igure 5 3 d, the peaks represent only oligomers end capped with hydrogens. As no end capping reaction was performed for this polymer, it can be said that debromination is active during the polymerization. On the other hand, the oligomers with and additional DTG molecule end capped with hydro gen can indicat e the loss of tin groups during the polymerization by destannylation or after the polymerization during precipitations into methanol
149 Figure 5 3. MALDI MS spectra of polymers. a) E(DTG BTD), b) E xpanded region between 2700 3700 m/z for E(DTG BTD), c ) NE(DTG BTD) and d) expanded region between 2700 3700 m/z for NE(DTG BTD). The assigned oligomer residues are highlighted by representative cartoons in the legend. a) b )
150 Figure 5 3. continued. c ) d )
151 Figu re 5 3 continued. Figure 5 4 Possible scenario showing the loss mechanisms of debromination and destannylation resulting in hydrogen end capped oligomers and successful end cappings with toluene units. In addition, we investigated the presence of atom s in the repeat unit structure as well as the residual bromine and tin atoms contained in the polymers through XPS analyses. Figure 5 5 shows XPS spectra for the spin cast films of E(DTG BTD) and NE(DTG BTD) using a Mg anode as the X ray source. Both polym ers exhibited the orbital energies of atoms such as Ge, N, C, and S present in the composition of the repeat unit. Also, an Al anode was used as an X ray source for XPS analysis of NE(DTG BTD) but no significant change s -except the disappearance of high e nergy
152 peaks due to Ge auger electrons were observed Furthermore, we investigated the presence of Br and Sn atoms in the polymers. T he expected Br 3d and Sn 3d 5/2 orbital energies are highlighted by bl ue and red arrows respectively in Figure 5 5 Figur e 5 5 XPS spectra of NE(DTG BTD) with two different X ray sources of Mg and Al and E(DTG BTD) with Mg source. The expected Br 3d and Sn 3d 5/2 orbital energies are highlighted as blue and red arrows respectively. N o peak was observ ed due to the presence o f Br or Sn atoms. In general we found no substantial difference s between the spectra of the polymers, and therefore conclude that XPS is not a valuable tool to assess the absence of Sn and Br atoms in these polymers. We performed inductively coupled plasm a (ICP) analyses to quantify the amount of Sn residues present in the polymers. Th is measurement was done in the G eology D epartment of UF by Dr. George Kamenov. E(DTG BTD) was found to contain 1979 ppm (0.20 % by weight) Sn 122 isotopes, while NE(DTG BTD) contained only 1583 ppm (0.16% by weight). The Sn content is comparable in both polymers, therefore
153 it can be speculated that residual tin atoms are either buried inside the polymer network or they are still bonded to DTG units. UV VIS spectroscopy and DPV electrochemical characterization methods were also used to examine the structural differences between E(DTG BTD) and NE(DTG BTD ). Figure 5 6 shows the absorption profiles of the polymers in dilute toluene solutions and in thin films. Solution absorption s pectrum for NE(DTG BTD) demonstrates dual band absorption maximums at 420 nm and 684 nm with a trough at 422 nm tailing off to 890 nm. The solution absorption spec trum for E(DTG BTD) presents with a similar dual absorption profile, but with an additional shoulder around 800 nm due to aggregation. In thin film absorption spectra, both polymers preserve their aforementioned dual band absorption profiles. For E(DTG BTD) there is a 15 nm red shift in the low en ergy band maximum as well as a shoulder around 775 nm which is more pronounced in the thin film spectrum than it was previously in the solution spectrum This red s hifted shoulder is attributed to a vibronic aggregation band indicative of more ordered planar chains 208 115 For the electrochemical characterization of the polymers differential pulse voltammetry has been conducted and the voltammograms are shown in Figure 5 7 Both polymers demonstrate reduction around 1.5 V and oxidation around 0.25 V. The reduction potential onsets were 1.41 V for both polymers. As both polymers contain the BTD acceptor unit, their reduction potentials are at the same level. However, the end capped polymer has an onset of oxidation at 0.18 V whereas the non en d capped polymer has a slightly hi gher oxidation onset at 0.22 V.
154 Figure 5 6 Absorption spectra for dilute solutions of E(DTG BTD) and NE(DTG BTD) in toluene and for thin film. Arrows highlight the aggregation bands. Figure 5 7. Differential Pulse Vo ltammograms of E(DTG BTD) (black, solid line) and NE(DTG BTD) (red, solid line) recorded in 0.1 M TBAPF 6 /PC electrolyte solution with a scan rate of 50 mV/s vs. Ag/Ag + reference electrode.
155 5 .3 OFET Performances Charge transport measurements were performe d through the fabrication and testing of OFET devices based on E(DTG BTD) and NE(DTG BTD). OFET devices were built in the bottom contact bottom gate architecture where the polymer was deposited onto an HMDS surface functionalized substrate, Si/SiO 2 The po lymers were deposited using two different methods: drop casting from a 2 mg/mL dichlorobenzene solution, and spin coating from a 10 mg/mL chloroform solution. Then, the samples were annealed at 200 o C for 60 minutes. The annealing temperature was chosen as it provided the highest hole mobility as determined through optimization experiments conducted by Dr. S. Reddy Puniredd. As a note, no thermal decomposition occurs at this temperature whereas 2% weight loss is observed at 350 o C as revealed by TGA measu rements. Two different processing methods were used to confirm that charge transport property differences originated from the variations in the chemical structures of the polymers and were not consequences of the processing method. The results of the charg e transport b ehaviors are shown in Figure 5 8 and summarized in Table 5 2. The OFET transfer plots show that holes are the dominant charge carriers in both polymers for both processing method s Table 5 2 summarizes the mobility values for the charge carrie rs. Firstly, it is observed that drop casted samples generate higher mobilities for both polymers compared to spin cast samples. This effect is due to the fact that while drop casting from a low volatile solvent, the polymer has more time for self assembly Spin casting results in uniform films; however the polymer has less time to self assemble itself due to fast evaporation of the volatile solvent.
156 Figure 5 8 Transistor output curves for drop cast and spin coated a) and c) E(DTG BTD), b) and d) NE(DTG BTD) thin films. Hence, more than an order of magnitude difference is observed between the observed mobilities for E(DTG BTD) (0.6 cm 2 /V.s. for drop casting vs. 0.05 cm 2 /V.s. for spin casting) when using the two methods, and an even more pronounced diff erence is seen for NE(DTG BTD) (0.07 cm 2 /V.s. for drop casting vs 0.002 cm 2 /V.s. for spin casting). Next, charge mobilities were compared for E(DTG BTD) and NE(DTG BTD) For the drop cast samples, the difference is again almost an order of magnitude highe r for E(DTG BTD) (0.6cm 2 /V.s) when compared to the mobilities obtained for NE(DTG BTD) (0.07 cm 2 /V.s.) and a 20 fold increase for E(DTG BTD) (0.05 cm 2 /V.s.) compared to NE(BTG DTG) (0.002 cm 2 /V.s) in the spin cast samples.
157 Table 5 2. Summary of hole mo bilities for E(DTG BTD) and NE(DTG BTD) processed from drop casting and spin casting. Hole Mobility (cm 2 /V.s.)& (on/off ratio) Drop cast Spin cast E(DTG BTD) 0.55 0.05 (9 10 5 ) 0.043 0.013 ( 7 10 4 ) NE(DTG BTD) 0.061 0.016 (2 10 5 ) 0 .002 0.0005 ( 1 10 4 ) 5 .4 X ray and Morphology Analyses In order to examine the difference in charge carrier mobilities, the molecular order and surface morphology were investigated using grazing incidence wide angle X ray scattering (GIWAXS) and atomic force microscopy (AFM) t echniques. As seen in Figure 5 9 the GIWAXS patterns show significant variations in molecular order between E(DTG BTD) and NE(DTG BTD) for both spin cast and drop cast samples. In the drop cast film of E(DTG BTD) a small number of relative isotropic reflections appear in the q xy direction due to the ordered polymer backbones in lamellar structures though the crystalline domains remain randomly oriented with respect to each other and the surface. As no spots or arcs are observed in t he q z direction no long range alignment appears to be present in the polymer. This is due to the effect of the drop casting method, which allows for random nucleation and self assembly processes to occur during crystallization. The schematic organization o f the polymer chains is depicted as a ca rtoon in the inset of Figure 5 9 a. stacking distances have been calculated by meridional integration along q z at q x,y = 0 directions to be 0.35 5 nm for E(DTG BTD) (Figure 5 1 0 a). For the drop cast NE(DTG BTD) thin film, no reflections in the q z and q xy direction were observed indicating low interchain order. This low molecular order can be attributed to poor interactions between conjugated backbones (Figure 5 9 b). It is worth noting here that a typical amorphous halo is present in the meridional integration
158 of the pattern due to the fle xible alkyl side chains. Nevertheless, NE(DTG BTD) polymer chains form lamellar stacks as confirmed by the very weak small angle scattering intensities indicative of highly disordered polymer chains as is illust rated in the inset of Figure 5 9 b. These obse rvations of the drop cast films explain the dramatic decrease in the charge carrier mobility for the NE(DTG BTD) samples. In the spin cast film, X ray patterns obtained from both polymers show an evident reflection on the q z direction, along with a weak ar c in the q xy Figure 5 9 GIWAXS characterization of E(DTG BTD) and NE(DTG BTD) a) and c) patterns of drop cast and spin coated E(DTG BTD), b)and d) patterns of drop cast and spin coated NE(DTG BTD). Insets schematically illustrate the surface organi zation of the polymers.
159 Figure 5 1 0 GIWAXS integrations of the wide angle scattering region along qz at qx,y = 0 for E(DTG BTD) and NE(DTG BTD) a) after drop casting and b) spin stacking peak maximum for E(DTG BTD) (red) and NE(DTG BTD) (blue). The arrow in b) indicates the stacking distance from 0.360 nm for NE(DTG BTD) to 0.355 nm for E(DTG BTD). However, for E(DTG BTD) the reflection in the q z direction is higher in intensity and less diffuse. This reflection results in a stacking distance of 0.355 nm for the E(DTG BTD) polymer chains preferentially arranged face on (out of plane) on the substrate.(Figure 5 9 c). The very weak scattering intensities in the equatorial plane of the pa ttern are indicative of a small portion of polymer chains stacking edge on (in plane) on the substrate. Integration of this reflection results in stacking distance of 0.360 nm for NE(DTG BTD) which means that there are less interactions between
160 polymer chains than in the E(DTG BTD) films. In the spin cast NE(DTG BTD) film, the aforementioned q z reflection is more diffuse and weaker in intensity as shown i n Figure 5 9 d. While the polymer arranges face on on the substrate the molecular order is much lower compared to E(DTG BTD). Special thanks should be given to Dr. Wojtek Pisula and Dr. S. Reddy Puniredd for X ray discussions and GIWAXS data analysis. Figure 5 1 1 Tapping mode AFM height images of drop cast and spin coated a) and c) E(DTG BTD), b) and d ) NE(DTG BTD). Height scale is identical for each processing methods. Additionally, AFM analyses have been carried out to examine the film microstructure and topography and AFM topographs are shown in Figure 5 1 1 At first glance, the drop cast E (DTG BTD) and NE(DTG topographs show almost iden tical layered networks. Upon clos er inspection, the E(DTG BTD) microstructure
161 exhibits a dense, finely structured network whereas this network is much coarser for NE(DTG BTD). This provides a clue to the o rigin of the observed improvements in hole mobility. The E(DTG BTD) polymer, with denser and finer network, can enable for more charge percolation pathways for charge transport. For the spin coated films, E(DTGBTD) exhibits fine, short interconnected fiber s, whereas the NE(DTG BTD) film surface shows a globular topography of small spheres. These results complement the observations obtained from the GIWAXS analyses. E(DTG BTD) showed more ordered polymer chains with closer interchain interactions for both pr ocessing conditions in the GIWAXS analyses. This results in a well interconnected fibrous microstructure via improved self assembly of the polymer chains, while reduced interchain interactions in NE(DTG BTD) lead to isotropic globular entities. 5 .5 Conc lusions and Perspectives Conjugated polymers are ideal for high performance OFET applications because their charge transport properties can be enhanced through both chemical design and, unlike their inorganic counterparts, through solution processing tech niques. C hapter 5 described a method for controlling the molecular order in conjugated polymers, in which toluene end groups were placed after a Stille polycondensation with 1% excess of ditin monomer. Characteristic chemical shifts of meth yl protons an d carbon on the toluene moiety allowed for the simple identification of these end groups with NMR. This is the first report in the literature that confirms th e presence of end groups on donor acceptor type polymers. Further, we showed the confirmation of t oluene end groups with MALDI MS. The charge transport properties of the end capped DTG BTD polymers were found to be superior to the properties of its non end c apped analogue. We attribute the s e
162 result s to the enhanced interchain interactions and higher po lymer ordering in thin films as evidenced by GIWAXS analyses. This is also the first time observation of the effects of end groups on the molecular order of conjugated polymers We speculate that toluene end groups can bind together the aggregates in the thin film, increasing the connectivity between crystallites. This enable s more pathways for charge percolation leading to high charge carrier mobilites. 5 .6 Experimental Details Poly[DONOR alt ACCEPTOR)] A 100 mL flame dried Schlenk flask was charged with carefully weighed DONOR (1.01 eq), ACCEPTOR (1 eq), Pd 2 (dba) 3 (4% eq.) and P(o tol) 3 (8 % eq.). The mixture was cycled with argon and vacuum three times. After addition of degassed toluene (20 mL), the reaction was heated to 90 o C for 3 days. This was fol lowed by the addition of 4 iodotoluene ( in excess) and the reaction was allowed to run for another 20 hours. The reaction was cooled to room temperature and precipitated into methanol (500 mL). The precipitates were collected in a thimble and the polymer w as washed wit h Soxhlet extraction using methanol, acetone, hexane and chloroform. The chloroform fraction was concentrated to 50 mL and diethylammoniumdiethyldithiocarbamate was added. After stirring for 2 hours under argon, the polymer was precipitated ag ain into methanol (350 mL). The mixture was filtered over a 0.45 m size polypropylene membrane and the solid was dried under vacuum for 2 days to afford the title polymer. Poly5 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene)alt 4,7 ( benzo[c][1,2,5]t hiadiazole) E(DTG BTD) The polymer was obtained as 0.240 g (68 %) of a black solid. 1 H NMR (500 MHz, C 2 D 2 Cl 4 ): : 8.40 7.70 (m, 4H), 2.20 0.60 (m, 50H) Mn= 33,214 g/mol, Mw= 81,013 g/mol, PDI: 2.43 Elemental Analysis for repeat
163 unit C 38 H 54 GeN 2 S 3 : Calcld: C, 64.49; H, 7.69; Ge, 10.26; N, 3.96; S, 13.59 found: C, 64.64; H, 8.38; N, 3.54 Td= 500 o C (55 % weight loss) 350 o C (2% weight loss) Poly5 didodecyl 4H germolo[3,2 b:4,5 b']dithiophene) alt 4,7 ( benzo[c][1,2,5]thiadiazole) NE(DTG BTD) The end capping step in the above protocol has been skipped but all other processes remained the same The polymer was ob tained as a black solid of 0.247 g. (70 %) 1 H NMR (500 MHz, C 2 D 2 Cl 4 ): : 8.40 7.70 (m, 4H), 2.20 0.60 (m, 50H) Mn= 28,150 g/mol, Mw= 74,959 g/mol, PDI: 2.66 Elemental Analysis for repeat unit C 38 H 54 GeN 2 S 3 : Calcld: C, 64.49; H, 7.69; Ge, 10.26; N, 3.96; S, 13.59 found: C, 64.16; H, 8.10; N, 3.77
164 CHAPTER 6 CONCLUSIONS AND P ERSPECTIVES This thesis has focused on controlling molecular organization in conjugated polymers. Polymer chains with strong intermolecular interactions through delocalized orbitals are important for efficient charge transport in high performance orga nic field effect transistors. Synthetic organic chemistry tools such as the incorporation of heavy atoms in the repeat unit structure of polymers and addition of proper end capping units to polymer chains have been introduced in pursuit of higher molecul ar order and improved charge transport properties in conjugated polymers Chapter 4 and Chapter 5 describe the structure property relationships in dithienogermole (DTG) based polymers. In Chapter 4, the synthesis of six DTG and dithienosilole (DTS) based d onor acceptor polymers was carried out using Stille polycondensation s with 1% stoichiometric imbalance. Specifically we diverged from the common stoichiometric balance method for controlling the nature of end groups. DTG based polymers have been targeted due to their long C Ge bond lengths which can push solubilizing alkyl chains away from the aromatic units of the polymers We observed this interaction to study its effect on optical, electronic and physical properties of the DTG polymers and drew compar isons to properties obtained for DTS based polymers In general, DTG polymers showed longer wavelength absorptions in the visible spectrum as well as higher HOMO and lower LUMO energy levels resulting in lower band gaps. The longer C Ge bond in DTG was e ffective in improving intermolecular interactio ns by pushing the long dodecyl side chains Closer interactions allo wed efficient charge transport for dithienogermole benzothiadiazole copolymer s, resulting in hole mobilities of 0.6 cm 2 /V.s However DTG based polymers showed po or performances in
165 solar cell devices. This demonstrates that not all highly ordered material s are well suited for photovoltaic applications which rely heavily on the morphology of the polymer and fullerene blend. Chapter 5 demonstrated the successfu l end capping of dithienogermole benzothiadiazole copolymer with toluene end gro ups as confirmed by 1 H NMR and MALDI MS. The toluene end groups were effective in directing improved stacking of the polymer chains as shown by UV thin film measurements. The X ray analyses showed improved molecular organizat ion in an end capped polymer compared to its non end capped analogue, which led to higher charge transport independent of the processing conditions. These observations are the first examples of studies investigating molecular order as it relates to charge transport properties in end c apped donor acceptor type polymers. Further insight in this research can be suggested in this field of research. B enzene molecules functionalized with methyl ethoxy ethoxy (MEE) groups i n the para position can be substituted instead of toluene end groups As these MEE side chains can move freely their molecular ordering is expected to be lower than the toluene functionalized polymers discussed in this thesis. The observation of this effect would support our results by confirming end capping as influentia l in controlling molecular order. Furthermore, chain growth type polymerizations can be employed for facil e incorporation of end groups. The chain growth pathway would also allow the choice of various end group moieties such as naphthalene and anthracene that are capable of stacking. This path will also be a useful tool in confirming our specu lation that end groups can join together the polymer chain ends.
166 Chapter 3 discussed the importance of strict synthetic protocols for generating scalable, high qua lity PGreen polymers. Also, changing the polymer synthesis method from oxidative polymerization to Stille coupling resulted in increased solar cell e fficienc ies. Additionally, Chapter 3 demonstrated the preparation of a large scale aesthetically pleasin g solar cell panels that can achieve a 7 .6 Volt output on a sunny Florida day. D iscrepancies between the small spin coated and large slot die coated solar cell performances have been investigated and left questions of how to decrease the series resistance by employing materials that have high charge transport properties. Lastly, an alternating polyfluorene co polymer has been synthesized that display s electrochromic, electroluminescent and photovoltaic activity. The lack of absorption in the red and blue reg ions resulted in a purple color ed instead of green color ed material In conclusion, org anic chemistry is and will be at the cen ter of preparing future large area polymer solar cells with power conversion efficiencies exceeding 10% and in the fabrication of organic field effect transistors with >10 cm 2 /V.s hole mobilities. In the near future high performing conjugated polymers will surround the most pr actical personal and commercial electronic appliances such as smart windows, sophisticated textiles, clean energy sources and transparent laptops. The field will be driven by advancements in conjugated polymers due to their facile processing with low cost equipment.
167 LIST OF REFERENCES 1. Fu, H. S.; Cheng, Y. M.; Chou, P. T.; Chi, Y., Materials Today 2011, 14, (10), 472 479. 2. Kondo, Y.; Tanabe, H.; Kudo, H.; Nakano, K.; Otake, T., Materials 2011, 4, (12). 3. Schmidtke, J., Optics Express 2010, 18, (19), A477 A486. 4. Hoffmann, R., Angewandte Chemie International Edition in English 198 7, 26, (9). 5. Whangbo, M. H.; Hoffmann, R.; Woodward, R. B., Pro ceedings of the Royal Society of London Series a Mathematical Physical and Engineering Sciences 1979, 366, (1724). 6. Roncali, J., Macromolecular Rapid Communications 2007, 28, (17). 7. Walcz ak, R. M.; Cowart, J. S., Jr.; Reynolds, J. R., Journal of Materials Chemistry 2007, 17, (3). 8. Ko, S.; Mondal, R.; Risko, C.; Lee, J. K.; Hong, S.; McGehee, M. D.; Bredas, J. L.; Bao, Z., Macromolecules 2010, 43, (16). 9. Beaujuge, P. M.; Vasilyeva, S. V .; Ellinger, S.; McCarley, T. D.; Reynolds, J. R., Macromolecules 2009, 42, (11). 10. Lu, K.; Sun, X.; Liu, T.; Di, C.; Xi, H.; Yu, G.; Gao, X.; Du, C., J Polym. Sci.Part A: Polym. Sci. 2009, 47. 11. Groenendaal, B. L.; Jonas, F.; Freitag, D.; Pielartzik, H.; Reynolds, J. R., Advanced Materials 2000, 12, (7), 481 494. 12. Lambert, T. L.; Ferraris, J. P., Journal of the Chemical Society Chemical Communications 1991 (11). 13. Thompson, B. C.; Kim, Y. G.; McCarley, T. D.; Reynolds, J. R., Journal of the Ameri can Chemical Society 2006, 128, (39), 12714 12725. 14. Perepichka, I. F.; Levillain, E.; Roncali, J., Journal of Materials Chemistry 2004, 14, (11). 15. Roncali, J.; Garreau, R.; Yassar, A.; Marque, P.; Garnier, F.; Lemaire, M., Journal of Physical Chemist ry 1987, 91, (27). 16. Wudl, F.; Kobayashi, M.; Heeger, A. J., Journal of Organic Chemistry 1984, 49, (18).
168 17. Hong, S. Y.; Marynick, D. S., Macromolecules 1992, 25, (18). 18. Sotzing, G. A.; Lee, K. H., Macromolecules 2002, 35, (19). 19. Havinga, E. E.; Tenhoeve, W.; Wynberg, H., Polymer Bulletin 1992, 29, (1 2). 20. Steckler, T. T.; Zhang, X.; Hwang, J.; Honeyager, R.; Ohira, S.; Zhang, X. H.; Grant, A.; Ellinger, S.; Odom, S. A.; Sweat, D.; Tanner, D. B.; Rinzler, A. G.; Barlow, S.; Bredas, J. L.; Kippe len, B.; Marder, S. R.; Reynolds, J. R., Journal of the American Chemical Society 2009, 131, (8). 21. Kono, T.; Kumaki, D.; Nishida, J. i.; Sakanoue, T.; Kakita, M.; Tada, H.; Tokito, S.; Yamashita, Y., Chemistry of Materials 2007, 19, (6). 22. Ponce Ortiz R.; Facchetti, A.; Marks, T. J.; Casado, J.; Zgierski, M. Z.; Kozaki, M.; Hernndez, V.; LpezNavarrete, J. T., Adv. Funct. Mater 2009 (19). 23. Dhanabalan, A.; van Dongen, J. L. J.; van Duren, J. K. J.; Janssen, H. M.; van Hal, P. A.; Janssen, R. A. J. Macromolecules 2001, 34, (8). 24. Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.; Neagu Plesu, R.; Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M., Journal of the American Chemical Society 2008, 130, (2). 25. Zhou, E.; Yamakawa, S.; Tajima, K.; Yang, C.; Hashimoto, K., Chemistry of Materials 2009, 21, (17). 26. Chen, H. Y.; Hou, J.; Hayden, A. E.; Yang, H.; Houk, K. N.; Yang, Y., Advanced Materials 2010, 22, (3). 27. Beaujuge, P. M.; Reynolds, J. R., Chemical Reviews 2010, 110, (1). 28. Groe nendaal, L.; Zotti, G.; Aubert, P. H.; Waybright, S. M.; Reynolds, J. R., Advanced Materials 2003, 15, (11). 29. Beaupre, S.; Boudreault, P. L. T.; Leclerc, M., Advanced Materials 2010, 22, (8). 30. Admassie, S.; Inganas, O.; Mammo, W.; Perzon, E.; Anderss on, M. R., Synthetic Metals 2006, 156, (7 8). 31. Dyer, A. L.; Thompson, E. J.; Reynolds, J. R., Acs Applied Materials & Interfaces 2011, 3, (6). 32. Amb, C. M.; Kerszulis, J. A.; Thompson, E. J.; Dyer, A. L.; Reynolds, J. R., Polymer Chemistry 2011, 2, (4 ).
169 33. Shi, P.; Amb, C. M.; Knott, E. P.; Thompson, E. J.; Liu, D. Y.; Mei, J.; Dyer, A. L.; Reynolds, J. R., Advanced Materials 2010, 22, (44). 34. Muehlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C., Advanced Material s 2006, 18, (21). 35. Lu, G.; Usta, H.; Risko, C.; Wang, L.; Facchetti, A.; Ratner, M. A.; Marks, T. J., Journal of the American Chemical Society 2008, 130, (24). 36. Inganas, O.; Zhang, F. L.; Andersson, M. R., Accounts of Chemical Research 2009, 42, (11) 1731 1739. 37. Beaujuge, P. M.; Amb, C. M.; Reynolds, J. R., Accounts of Chemical Research 2010, 43, (11). 38. Wang, C.; Dong, H.; Hu, W.; Liu, Y.; Zhu, D.,. Chemical Reviews 2012, 112, (4). 39. Bredas, J. L.; Calbert, J. P.; da Silva, D. A.; Cornil, J., Organic semiconductors: Proceedings of the National Academy of Sciences of the United States of America 2002, 99, (9). 40. da Silva, D. A.; Kim, E. G.; Bredas, J. L.,. Advanced Materials 2005, 17, (8). 41. Hutchison, G. R.; Ratner, M. A.; Marks, T. J., Jo urnal of the American Chemical Society 2005, 127, (7). 42. Bromley, S. T.; Mas Torrent, M.; Hadley, P.; Rovira, C., Journal of the American Chemical Society 2004, 126, (21). 43. Cornil, J.; Bredas, J. L.; Zaumseil, J.; Sirringhaus, H., Advanced Materials 2 007, 19, (14). 44. Bredas, J. L.; Beljonne, D.; Coropceanu, V.; Cornil, J., Chemical Reviews 2004, 104, (11). 45. Street, R. A.; Northrup, J. E.; Salleo, A., Physical Review B 2005, 71, (16). 46. Garnier, F.; Yassar, A.; Hajlaoui, R.; Horowitz, G.; Deloffr e, F.; Servet, B.; Ries, S.; Alnot, P., Journal of the American Chemical Society 1993, 115, (19). 47. Meng, H.; Sun, F. P.; Goldfinger, M. B.; Jaycox, G. D.; Li, Z. G.; Marshall, W. J.; Blackman, G. S., Journal of the American Chemical Society 2005, 127, ( 8). 48. Locklin, J.; Li, D. W.; Mannsfeld, S. C. B.; Borkent, E. J.; Meng, H.; Advincula, R.; Bao, Z., Chemistry of Materials 2005, 17, (13).
170 49. Xiao, S. X.; Myers, M.; Miao, Q.; Sanaur, S.; Pang, K. L.; Steigerwald, M. L.; Nuckolls, C., Angewandte Chemie International Edition 2005, 44, (45). 50. Moon, H.; Zeis, R.; Borkent, E. J.; Besnard, C.; Lovinger, A. J.; Siegrist, T.; Kloc, C.; Bao, Z. N., Journal of the American Chemical Society 2004, 126, (47). 51. Sarma, J.; Desiraju, G. R., Accounts of Chemical Research 1986, 19, (7). 52. Takimiya, K.; Kunugi, Y.; Konda, Y.; Niihara, N.; Otsubo, T., Journal of the American Chemical Society 2004, 126, (16). 53. Chen, T. A.; Rieke, R. D., Journal of the American Chemical Society 1992, 114, (25). 54. McCullough, R. D.; Lowe, R. D.; Jayaraman, M.; Anderson, D. L., Journal of Organic Chemistry 1993, 58, (4), 904 912. 55. McCullough, R. D.; Tristramnagle, S.; Williams, S. P.; Lowe, R. D.; Jayaraman, M., Journal of the American Chemical Society 1993, 115, (11). 56. Loewe R. S.; Khersonsky, S. M.; McCullough, R. D., Advanced Materials 1999, 11, (3). 57. Loewe, R. S.; Ewbank, P. C.; Liu, J. S.; Zhai, L.; McCullough, R. D., Macromolecules 2001, 34, (13). 58. Babel, A.; Jenekhe, S. A., Synthetic Metals 2005, 148, (2). 59. Ba o, Z.; Locklin, J., Organic Field Effect Transistors CRC Press: Boca Raton, FL, 2007. 60. Wang, G. M.; Swensen, J.; Moses, D.; Heeger, A. J., Journal of Applied Physics 2003, 93, (10). 61. Bao, Z. N.; Lovinger, A. J., Chemistry of Materials 1999, 11, (9). 62. Sirringhaus, H.; Tessler, N.; Friend, R. H., Science 1998, 280, (5370). 63. Chang, J. F.; Sun, B. Q.; Breiby, D. W.; Nielsen, M. M.; Solling, T. I.; Giles, M.; McCulloch, I.; Sirringhaus, H., Chemistry of Materials 2004, 16, (23). 64. Kline, R. J.; Mc Gehee, M. D.; Kadnikova, E. N.; Liu, J. S.; Frechet, J. M. J.,. Advanced Materials 2003, 15, (18). 65. Tsao, H. N.; Cho, D.; Andreasen, J. W.; Rouhanipour, A.; Breiby, D. W.; Pisula, W.; Muellen, K., Advanced Materials 2009, 21, (2).
171 66. Niemi, V. M.; Knuu ttila, P.; Osterholm, J. E.; Korvola, J., Polymer 1992, 33, (7). 67. Kossmehl, G., Makromolekulare Chemie Macromolecular Symposia 1986, 4. 68. Jikei, M.; Katoh, J.; Sato, N.; Yamamoto, K.; Nishide, H.; Tsuchida, E., S. Bulletin of the Chemical Society of J apan 1992, 65, (8). 69. Myers, R. E., Journal of Electronic Materials 1986, 15, (2). 70. Ogawa, K.; Stafford, J. A.; Rothstein, S. D.; Tallman, D. E.; Rasmussen, S. C.,. Synthetic Metals 2005, 152, (1 3). 71. Ma, C.; Xu, Y.; Zhang, C.; Xu, Y.; Xiang, W.; M i, O., Journal of Electroanalytical Chemistry 2009, 634, (1). 72. McCarley, T. D.; Noble, C. O.; DuBois, C. J.; McCarley, R. L., Macromolecules 2001, 34, (23). 73. Andersson, M. R.; Selse, D.; Berggren, M.; Jarvinen, H.; Hjertberg, T.; Inganas, O.; Wenners trom, O.; Osterholm, J. E., Macromolecules 1994, 27, (22). 74. Vangeneugden, D. L.; Vanderzande, D. J. M.; Salbeck, J.; van Hal, P. A.; Janssen, R. A. J.; Hummelen, J. C.; Brabec, C. J.; Shaheen, S. E.; Sariciftci, N. S., Journal of Physical Chemistry B 20 01, 105, (45). 75. Abdou, M. S. A.; Lu, X. T.; Xie, Z. W.; Orfino, F.; Deen, M. J.; Holdcroft, S., Chemistry of Materials 1995, 7, (4). 76. Cai, T.; Zhoub, Y.; Wanga, E.; Hellstrm, S.; Zhang, F.; Xu, X.; Ingans, O.; Andersson, M. R., Sol. En. Mat. Sol. C ells 2010 (94 ). 77. Cheng, Y. J.; Luh, T. Y., Journal of Organometallic Chemistry 2004, 689, (24). 78. Bao, Z. N.; Chan, W. K.; Yu, L. P., Journal of the American Chemical Society 1995, 117, (50). 79. Carsten, B.; He, F.; Son, H. J.; Xu, T.; Yu, L., Chem ical Reviews 2011, 111, (3). 80. Ziegler, C. B.; Heck, R. F., Journal of Organic Chemistry 1978, 43, (15). 81. Herrmann, W. A.; Brossmer, C.; Ofele, K.; Reisinger, C. P.; Priermeier, T.; Beller, M.; Fischer, H., Angewandte Chemie International Edition in E nglish 1995, 34, (17). 82. Grushin, V. V., Chemistry a European Journal 2002, 8, (5). 83. Kosugi, M.; Shimizu, Y.; Migita, T., Chemistry Letters 1977 (12).
172 84. Milstein, D.; Stille, J. K., Journal of the American Chemical Society 1978, 100, (11). 85. Aziz ian, H.; Eaborn, C.; Pidcock, A., Journal of Organometallic Chemistry 1981, 215, (1). 86. Benaglia, M.; Toyota, S.; Woods, C. R.; Siegel, J. S., Tetrahedron Letters 1997, 38, (27). 87. Murata, M.; Oyama, T.; Watanabe, S.; Masuda, Y., Synthetic Communicatio ns 2002, 32, (16). 88. Miyaura, N.; Suzuki, A., Chemical Reviews 1995, 95, (7). 89. Suzuki, A., Journal of Organometallic Chemistry 1999, 576, (1 2). 90. Suzuki, A. B., H. C., Suzuki Coupling, Organic Syntheses via Boranes. In Aldrich Chemical Company: Mil waukee, WI, 2003 Vol. 3. 91. Kimbrough, R. D., Environmental Health Perspectives 1976, 14, (APR). 92. Giroux, A.; Han, Y. X.; Prasit, P., Tetrahedron Letters 1997, 38, (22). 93. Tse, M. K.; Cho, J. Y.; Smith, M. R., Organic Letters 2001, 3, (18). 94. Ishiy ama, T.; Takagi, J.; Ishida, K.; Miyaura, N.; Anastasi, N. R.; Hartwig, J. F., Journal of the American Chemical Society 2002, 124, (3). 95. Sakamoto, J.; Rehahn, M.; Wegner, G.; Schluter, A. D., Macromolecular Rapid Communications 2009, 30, (9 10), 653 687 96. Treacher, K.; Stosel, P.; Spreitzer, H.; Becker, A.; Falcou, H. 2005. 97. Wang, E.; Hou, L.; Wang, Z.; Hellstrom, S.; Mammo, W.; Zhang, F.; Inganas, O.; Andersson, M. R., Organic Letters 2010, 12, (20). 98. Guo, X.; Kim, F. S.; Jenekhe, S. A.; Watson M. D., Journal of the American Chemical Society 2009, 131, (21). 99. Mei, J.; Kim, D. H.; Ayzner, A. L.; Toney, M. F.; Bao, Z., Journal of the American Chemical Society 2011, 133, (50). 100. Yokoyama, A.; Suzuki, H.; Kubota, Y.; Ohuchi, K.; Higashimura, H.; Yokozawa, T., Journal of the American Chemical Society 2007, 129, (23). 101. Facchetti, A.; Vaccaro, L.; Marrocchi, A., Angewandte Chemie International Edition 2012, 51, (15).
173 102. Rudenko, A. E.; Wiley, C. A.; Stone, S. M.; Tannaci, J. F.; Thompson, B C., Journal of Polymer Science Part A: Polymer Chemistry 2012 ; Vol. 50, p 3691. 103. Berrouard, P.; Najari, A.; Pron, A.; Gendron, D.; Morin, P. O.; Pouliot, J. R.; Veilleux, J.; Leclerc, M., Angewandte Chemie International Edition 2012, 51, (9). 104. S evignon, M.; Papillon, J.; Schulz, E.; Lemaire, M., Tetrahedron Letters 1999, 40, (32). 105. Wang, Q.; Takita, R.; Kikuzaki, Y.; Ozawa, F., Journal of the American Chemical Society 2010, 132, (33). 106. Odian, G., Principles of Polymerization John Wiley a nd Sons: 2004. 107. Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.; So, F.; Reynolds, J. R., Journal of the American Chemical Society 2011, 133, (26), 10062 10065. 108. Blouin, N.; Michaud, A.; Leclerc, M., Advanced Materials 2007, 19, (17) 109. Park, J. K.; Jo, J.; Seo, J. H.; Moon, J. S.; Park, Y. D.; Lee, K.; Heeger, A. J.; Bazan, G. C., Advanced Materials 2011, 23, (21). 110. Guo, X.; Ponce Ortiz, R.; Zheng, Y.; Kim, M. G.; Zhang, S.; Hu, Y.; Lu, G.; Facchetti, A.; Marks, T. J., Journal of the American Chemical Society 2011, 133, (34). 111. Schubert, D. W.; Dunkel, T., Materials Research Innovations 2003, 7, (5). 112. DeLongchamp, D. M.; Vogel, B. M.; Jung, Y.; Gurau, M. C.; Richter, C. A.; Kirillov, O. A.; Obrzut, J.; Fischer, D. A.; Sa mbasivan, S.; Richter, L. J.; Lin, E. K., Chemistry of Materials 2005, 17, (23). 113. Tsao, H. N.; Muellen, K., Chemical Society Reviews 2010, 39, (7). 114. Degarmo, E. P.; Black, J. T.; Kohser, R. A., Materials and Processes in Manufacturing 9th edition ed.; Wiley: 2003. 115. Mortimer, R. J.; Graham, K. R.; Grenier, C. R. G.; Reynolds, J. R., Acs Applied Materials & Interfaces 2009, 1, (10). 116. Girotto, C.; Rand, B. P.; Genoe, J.; Heremans, P., Solar Energy Materials and Solar Cells 2009, 93, (4). 117. Steirer, K. X.; Reese, M. O.; Rupert, B. L.; Kopidakis, N.; Olson, D. C.; Collins, R. T.; Ginley, D. S., Solar Energy Materials and Solar Cells 2009, 93, (4).
174 118. Chen, L. M.; Hong, Z. R.; Kwan, W. L.; Lu, C. H.; Lai, Y. F.; Lei, B.; Liu, C. P.; Yang, Y., Acs Nano 2010, 4, (8), 4744 4752. 119. Krebs, F. C.; Gevorgyan, S. A.; Alstrup, J., Journal of Materials Chemistry 2009, 19, (30). 120. Sondergaard, R.; Hosel, M.; Angmo, D.; Larsen Olsen, T. T.; Krebs, F. C., Materials Today 2012, 15, (1 2). 121. Garnier F.; Hajlaoui, R.; Yassar, A.; Srivastava, P., Science 1994, 265, (5179). 122. Shaheen, S. E.; Radspinner, R.; Peyghambarian, N.; Jabbour, G. E., Applied Physics Letters 2001, 79, (18). 123. Holdcroft, S., Advanced Materials 2001, 13, (23). 124. Bao, Z. N .; Rogers, J. A.; Katz, H. E., Journal of Materials Chemistry 1999, 9, (9). 125. Krebs, F. C.; Jorgensen, M.; Norrman, K.; Hagemann, O.; Alstrup, J.; Nielsen, T. D.; Fyenbo, J.; Larsen, K.; Kristensen, J., Solar Energy Materials and Solar Cells 2009, 93, ( 4). 126. Steiger, J.; Heun, S.; Tallant, N., Journal of Imaging Science and Technology 2003, 47, (6). 127. Yoshioka, Y.; Jabbour, G. E., Synthetic Metals 2006, 156, (11 13). 128. de Gans, B. J.; Duineveld, P. C.; Schubert, U. S., Advanced Materials 2004, 1 6, (3). 129. Hoth, C. N.; Choulis, S. A.; Schilinsky, P.; Brabec, C. J., Advanced Materials 2007, 19, (22), 3973 130. Wengeler, L.; Schmidt Hansberg, B.; Peters, K.; Scharfer, P.; Schabel, W., Chemical Engineering and Processing 2011, 50, (5 6). 131. Blan kenburg, L.; Schultheis, K.; Schache, H.; Sensfuss, S.; Schroedner, M., Solar Energy Materials and Solar Cells 2009, 93, (4). 132. Thrane, L.; Jorgensen, T. M.; Jorgensen, M.; Krebs, F. C., Solar Energy Materials and Solar Cells 2012, 97. 133. Manceau, M.; Angmo, D.; Jorgensen, M.; Krebs, F. C., Organic Electronics 2011, 12, (4). 134. Sonder gaard, R.; Manceau, M.; Jorgensen, M.; Krebs, F. C., Advanced Energy Materials 2012, 2, (4).
175 135. Espinosa, N.; Garcia Valverde, R.; Urbina, A.; Krebs, F. C., Solar Ener gy Materials and Solar Cells 2011, 95, (5). 136. Sirringhaus, H., Nature 2009, 457, (7230). 137. Ong, B. S.; Wu, Y. L.; Liu, P.; Gardner, S., Journal of the American Chemical Society 2004, 126, (11). 138. McCulloch, I.; Heeney, M.; Bailey, C.; Genevicius, K.; Macdonald, I.; Shkunov, M.; Sparrowe, D.; Tierney, S.; Wagner, R.; Zhang, W. M.; Chabinyc, M. L.; Kline, R. J.; McGehee, M. D.; Toney, M. F., Nature Materials 2006, 5, (4). 139. Li, Y.; Sonar, P.; Singh, S. P.; Soh, M. S.; van Meurs, M.; Tan, J., Journ al of the American Chemical Society 2011, 133, (7). 140. Babel, A.; Jenekhe, S. A., Advanced Materials 2002, 14, (5). 141. Babel, A.; Jenekhe, S. A., Journal of the American Chemical Society 2003, 125, (45). 142. Yan, H.; Chen, Z.; Zheng, Y.; Newman, C.; Q uinn, J. R.; Dotz, F.; Kastler, M.; Facchetti, A., Nature 2009, 457, (7230). 143. Liu, Y.; Wan, X.; Wang, F.; Zhou, J.; Long, G.; Tian, J.; You, J.; Yang, Y.; Chen, Y., Advanced Energy Materials 2011, 1, (5). 144. Sun, Y.; Welch, G. C.; Leong, W. L.; Takac s, C. J.; Bazan, G. C.; Heeger, A. J., Nature Materials 2012, 11, (1). 145. Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J., Science 1995, 270, (5243). 146. Zhao, G. J.; He, Y. J.; Li, Y. F., Advanced Materials 2010, 22, (39), 4355 147. Li, G. ; Shrotriya, V.; Huang, J. S.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y., Nature Materials 2005, 4, (11). 148. Ma, W. L.; Yang, C. Y.; Gong, X.; Lee, K.; Heeger, A. J., Advanced Functional Materials 2005, 15, (10). 149. Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J., Nature Photonics 2009, 3, (5). 150. Liang, Y. Y.; Wu, Y.; Feng, D. Q.; Tsai, S. T.; Son, H. J.; Li, G.; Yu, L. P., Journal of the American Chemical Society 2009, 131, (1), 56
176 151. Liang, Y.; Feng, D.; Wu, Y.; Tsai, S. T.; Li, G.; Ray, C.; Yu, L., Journal of the American Chemical Society 2009, 131, (22). 152. Zou, Y.; Najari, A.; Berrouard, P.; Beaupre, S.; Aich, B. R.; Tao, Y.; Leclerc, M., A Journal of the American Chemica l Society 2010, 132, (15). 153. Troshin, P. A.; Susarova, D. K.; Moskvin, Y. L.; Kuznetsov, I. E.; Ponomarenko, S. A.; Myshkovskaya, E. N.; Zakharcheva, K. A.; Balakai, A. A.; Babenko, S. D.; Razumov, V. F., Advanced Functional Materials 2010, 20, (24), 43 51 4357. 154. Saeki, A.; Tsuji, M.; Seki, S., Advanced Energy Materials 2011, 1, (4), 661 669. 155. Urien, M.; Wantz, G.; Cloutet, E.; Hirsch, L.; Tardy, P.; Vignau, L.; Cramail, H.; Parniex, J. P., Organic Electronics 2007, 8, (6), 727 734. 156. Bjorklund N.; Lill, J. O.; Rajander, J.; Osterbacka, R.; Tierney, S.; Heeney, M.; McCulloch, I.; Colle, M., Organic Electronics 2009, 10, (2), 215 221. 157. Nielsen, K. T.; Bechgaard, K.; Krebs, F. C., Macromolecules 2005, 38, (3), 658 659. 158. Thompson, C. B. Va riable Bandgap Poly(3,4 alkylenedioxythiophene) Based Polymers for Photovoltaic and Electrochromic Applications. University of Florida Gainesville, FL,, 2005. 159. Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C., Advanced Materials 2011, 23, (20), 2367 2371. 160. Pisula, W.; Tomovic, Z.; Simpson, C.; Kastler, M.; Pakula, T.; Mullen, K., Chemistry of Materials 2005, 17, (17), 4296 4303. 161. Fuoss, P. H.; Brennan, S., Annual Review of Materials Science 1990, 20, 365 390. 162. DeLongchamp, D. M.; Kline, R. J.; Fischer, D. A.; Richter, L. J.; Toney, M. F., Advanced Materials 2011, 23, (3), 319 337. 163. Small, C. E.; Chen, S.; Subbiah, J.; Amb, C. M.; Tsang, S. W.; Lai, T. H.; Reynolds, J. R.; So, F., Nature Photonics 2012, 6, (2), 115 120. 1 64. Scharber, M. C.; Wuhlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.; Brabec, C. L., Advanced Materials 2006, 18, (6), 789 +. 165. Bundgaard, E.; Krebs, F. C., Solar Energy Materials and Solar Cells 2007, 91, (11), 954 985.
177 166. Beaujuge, P Conjugated Organic Polymers. University of Florida, Gainesville, Fl, 2009. 167. Beaujuge, P. M.; Subbiah, J.; Choudhury, K. R.; Ellinger, S.; McCarley, T. D.; So, F.; Reynolds, J. R., Chemistry of Materials 2010, 22, (6), 209 3 2106. 168. Subbiah, J.; Beaujuge, P. M.; Choudhury, K. R.; Ellinger, S.; Reynolds, J. R.; So, F., Acs Applied Materials & Interfaces 2009, 1, (6), 1154 1158. 169. Voorhies, J. D.; Schurdak, E. J., Anal. Chem. 1962 (34), 939. 170. Amb, C. M.; Craig, M. R .; Koldemir, U.; Subbiah, J.; Choudhury, K. R.; Gevorgyan, S. A.; Jorgensen, M.; Krebs, F. C.; So, F.; Reynolds, J. R., Acs Applied Materials & Interfaces 2012, 4, (3), 1847 1853. 171. Campoy Quiles, M.; Ferenczi, T.; Agostinelli, T.; Etchegoin, P. G.; Kim Y.; Anthopoulos, T. D.; Stavrinou, P. N.; Bradley, D. D. C.; Nelson, J., Nature Materials 2008, 7, (2), 158 164. 172. Subbiah, J.; Amb, C. M.; Reynolds, J. R.; So, F., Solar Energy Materials and Solar Cells 2012, 97, 97 101. 173. Koldemir, U.; Graham, K. R.; Salazar, D. H.; McCarley, T. D.; Reynolds, J. R., Journal of Materials Chemistry 2011, 21, (18), 6480 6482. 174. Kilbinger, A. F. M.; Feast, W. J., Journal of Materials Chemistry 2000, 10, (8), 1777 1784. 175. Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G. C., Nature Materials 2007, 6, (7). 176. Morana, M.; Wegscheider, M.; Bonanni, A.; Kopidakis, N.; Shaheen, S.; Scharber, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C., Advanced Functional Materials 2008, 18, (1 2). 177. Zhang, M.; Tsao, H. N.; Pisula, W.; Yang, C.; Mishra, A. K.; Muellen, K., Journal of the American Chemical Society 2007, 129, (12). 178. Bijleveld, J. C.; Shahid, M.; Gilot, J.; Wienk, M. M.; Janssen, R. A. J., Advanced Functional Materials 2009, 19, (20), 3262 3270. 179. Usta, H.; Lu, G.; Facchetti, A.; Marks, T. J., Journal of the American Chemical Society 2006, 128, (28), 9034 9035. 180. Beaujuge, P. M.; Pisula, W.; Tsao, H. N.; Ellinger, S.; Mullen, K.; Reynolds, J. R., Journal of the American Chemical Society 2009, 131, (22), 7514 +.
178 181. Chu, T. Y.; Lu, J. P.; Beaupre, S.; Zhang, Y. G.; Pouliot, J. R.; Wakim, S.; Zhou, J. Y.; Leclerc, M.; Li, Z.; Ding, J. F.; Tao, Y., Journal of the American Chemical Society 2011, 133, (12), 4250 4253. 182. Zh ang, Z. G.; Min, J.; Zhang, S.; Zhang, J.; Zhang, M.; Li, Y., Chemical Communications 2011, 47, (33). 183. Subramaniyan, S.; Xin, H.; Kim, F. S.; Shoaee, S.; Durrant, J. R.; Jenekhe, S. A., Advanced Energy Materials 2011, 1, (5). 184. Scharber, M. C.; Kopp e, M.; Gao, J.; Cordella, F.; Loi, M. A.; Denk, P.; Morana, M.; Egelhaaf, H. J.; Forberich, K.; Dennler, G.; Gaudiana, R.; Waller, D.; Zhu, Z. G.; Shi, X. B.; Brabec, C. J., Advanced Materials 2010, 22, (3), 367 185. Gendron, D.; Morin, P. O.; Berrouard P.; Allard, N.; Aich, B. R.; Garon, C. N.; Tao, Y.; Leclerc, M., Macromolecules 2011, 44, (18). 186. Ohshita, J.; Hwang, Y. M.; Mizumo, T.; Yoshida, H.; Ooyama, Y.; Harima, Y.; Kunugi, Y., Organometallics 2011, 30, (12), 3233 3236. 187. Hwang, Y. M.; Ohs hita, J.; Harima, Y.; Mizumo, T.; Ooyama, Y.; Morihara, Y.; Izawa, T.; Sugioka, T.; Fujita, A., Polymer 2011, 52, (18), 3912 3916. 188. Fei, Z. P.; Kim, Y.; Smith, J.; Domingo, E. B.; Stingelin, N.; McLachlan, M. A.; Song, K.; Anthopoulos, T. D.; Heeney, M ., Macromolecules 2012, 45, (2), 735 742. 189. Kim, J. S.; Fei, Z. P.; James, D. T.; Heeney, M., Journal of Materials Chemistry 2012, 22, (19), 9975 9982. 190. Hou, J. H.; Park, M. H.; Zhang, S. Q.; Yao, Y.; Chen, L. M.; Li, J. H.; Yang, Y., Macromolecules 2008, 41, (16), 6012 6018. 191. Mancilha, F. S.; DaSilveira Neto, B. A.; Lopes, A. S.; Moreira, P. F., Jr.; Quina, F. H.; Goncalves, R. S.; Dupont, J., European Journal of Organic Chemistry 2006 (21). 192. Nielsen, C. B.; Bjornholm, T., Organic Letters 2 004, 6, (19). 193. Chen, J.; Shi, M. M.; Hu, X. L.; Wang, M.; Chen, H. Z., Polymer 2010, 51, (13). 194. Ahmed Aisa, A. M.; EnKe, S.; Richter, H., Journal of Organometallic Chemistry 1999 (575), 126 132. 195. Coffin, R. C.; Peet, J.; Rogers, J.; Bazan, G. C., Nature Chemistry 2009, 1, (8), 657 661. 196. Kruger, R. A.; Gordon, T. J.; Baumgartner, T.; Sutherland, T. C., Acs Applied Materials & Interfaces 2011, 3, (6), 2031 2041.
179 197. Boudouris, B. W.; Molins, F.; Blank, D. A.; Frisbie, C. D.; Hillmyer, M. A., Macromolecules 2009, 42, (12), 4118 4126. 198. Moon, H. C.; Anthonysamy, A.; Lee, Y.; Kim, J. K., Facile Synthesis of Well Macromolecules 2010, 43, (4), 1747 1752. 199. Chen, Y. H.; Huang, P. T.; Lin, K. C.; Huang, Y. J.; Chen, C. T., Organic Electronics 2012, 13, (2), 283 289. 200. Lohwasser, R. H.; Thelakkat, M., Macromolecules 2010, 43, (18), 7611 7616. 201. Lohwasser, R. H.; Bandara, J.; Thelakkat, M., Journal of Materials Chemistry 2009, 19, (24), 4126 4130. 202. Jeffries El, M.; Sauve, G.; McCulloug h, R. D., Macromolecules 2005, 38, (25), 10346 10352. 203. Kim, J. S.; Lee, Y.; Lee, J. H.; Park, J. H.; Kim, J. K.; Cho, K., Advanced Materials 2010, 22, (12), 1355 +. 204. Kim, Y.; Cook, S.; Kirkpatrick, J.; Nelson, J.; Durrant, J. R.; Bradley, D. D. C.; Giles, M.; Heeney, M.; Hamilton, R.; McCulloch, I., Journal of Physical Chemistry C 2007, 111, (23), 8137 8141. 205. Hohl, B.; Bertschi, L.; Zhang, X. Y.; Schluter, A. D.; Sakamoto, J., Macromolecules 2012, 45, (13), 5418 5426. 206. Jayakannan, M.; van Do ngen, J. L. J.; Janssen, R. A. J., Macromolecules 2001, 34, (16), 5386 5393. 207. Groenendaal, L.; Peerlings, H. W. I.; Havinga, E. E.; Vekemans, J.; Meijer, E. W., Synthetic Metals 1995, 69, (1 3). 208. Lee, M. J.; Gupta, D.; Zhao, N.; Heeney, M.; McCullo ch, I.; Sirringhaus, H., Advanced Functional Materials 2011, 21, (5), 932 940.
180 BIOGRAPHICAL SKETCH Unsal Koldemir is a native Turkish born in Denizli, Turkey in 1982. He graduated from Denizli Anatolian High School where he spent seven years studying m ath and science courses in English. He was in the top 1% in the national university exam taken by approximately 1.5 million high school graduates. He chose Bilkent University Chemistry D epartment the highest ranked C hemistry D epartment in Turkey and rec eived full scholarship during studies. He gained his B.S in Chemistry in 2005 and M.S in Chemistry in 2007. His interest in organic chemistry and chemistry started in 2004 when he met Dr. Fikret Koc during a paid summer internship in Prof. Peter Eilbrach group in 2007, he was exposed to research on the synthesis of green colored solar cell polymers as well as high performance field effect transistor materials employing donor a cceptor type polymers. He earned his Ph.D in organic and polymer chemistry from the University of Florida in the autumn of 2012. He will work as a post doctoral researcher for organic photovoltaics.