Dye-Sensitized Titanium Dioxide Solar Cells ― Studies on Redox Couples, Ruthenium and Thienyl Dyes

本論文主要目的是探討固態氧化還原對、釕錯合物以及噻吩有機分子於染料敏化太陽能電池之應用及其性能表現。 首先，以N3為染料進行二氧化鈦染料敏化太陽能電池組裝與測試，並從中歸納出影響元件光電位、光電流以及轉換效率之因素。其中，光電位受光激發注入之電子與光電極界面上電子-碘離子再結合之暗電流的影響。因此，增加注入的電子數與減少界面之再結合可以提升光電位。光電流是由激發光敏染料產生，且受到入射單光光子-電子轉化效率的影響，所以具有較高之近紅外線區吸收係數之染料與較長壽命之光生電子可以得到較高的電流，而電子於二氧化鈦薄膜中之壽命與傳遞性質可以經由暫態光電位與光電流量測估算求得。另外，由等效電路之分析與還原電流之量測，可以推導出光電流-電位之經驗模型，並可定量描述其光電流-電位曲線之行為。 本研究初期選用鐵氰化銦（InHCF）為氧化還原對，以組裝出具有高光電位之固態染料敏化二氧化鈦太陽能電池。因為InHCF之氧化還原反應需要鉀離子的嵌入/遷出，所以選用含飽和鉀離子之poly(2-acrylamido-2-methylpropane-sulfonic acid)（K-PAMPS）高分子電解質。由於N3染料與InHCF之接觸界面不佳且鉀離子擴散速率較低，造成此固態元件效率偏低。若將染料分佈於高分子電解質中，則可以改善此界面性質，使得元件較偏向再生式電池的表現。為了更提升N3染料與InHCF之界面性質與可逆反應，直接將鋰離子摻雜之InHCF或鐵氰化銅（CuHCF）氧化還原對鍍於染料敏化二氧化鈦電極表面上，然而，對電極與光電極之接觸界面性質不佳，使電子轉移阻力過大，造成元件之光電位損失，影響元件性能表現。處於氧化態之染料與InHCF或CuHCF之還原反應太慢也是影響電流與效率的重要因素。此外，CuHCF之電子轉移阻力較大，也使得氧化染料之還原再生反應較慢，元件效率較低。 本研究嘗試以由中研院化學所林建村教授實驗室所製備之具有低pi*軌域的phenanthrenyl（TAPNB）配位基之釕錯合物，進行染料敏化太陽能電池組裝。由光譜與電化學量測，可知錯合物之激發能階位置可以搭配二氧化鈦導電帶。由電池性能量測結果顯示，雖然其開環電位相近，但短路電流卻比含N3染料之元件低一個數量級以上，所以能量轉換效率比較低（1%以下），一方面是因光生電子壽命較N3低，另一方面可能是因為立體障礙而使得錯合物吸附量較少。另外，當硫氰配位基被吡啶置換後，會使得金屬-配位基電荷轉移(Ru(II) → TAPNB)能量增加，造成吸收光譜藍位移，但吸附量卻增加，因此使得效率提升。當分子之羧酸基置換成乙酯基後，會使得錯合物在二氧化鈦表面的吸附量減少且與半導體之作用力亦相對減弱，因而產生較低之光電流與效率。 由中研院化學所林建村教授實驗室所製備之四種以benzothiadiazole與benzoselenadiazole為發色團，並以2-噻吩-2'-氰丙烯酸為anchor之有機染料進行染料敏化太陽能電池之性能量測。當分子中之苯基被噻吩基置換後，會使分子共軛性質增加且結構共平面使得光譜吸收紅位移。然而，由暫態實驗量測，此共平面結構並不能加速電荷分離或抑制染料與電子之再結合反應。當以苯基連接時，效率可顯著的提升至4%，這是因為染料之彎曲非共平面結構，而使得電子之再結合反應可以有效的被抑制。此以二苯胺為施體與噻吩氰丙烯酸為受體，並以benzothiadiazole與benzoselenadiazole為發色團之有機染料，能提供有效的電荷轉移並加速電荷分離，使光能量轉換效率達4%。The object of this thesis is to investigate the behavior of the solid-state redox couples and the performance of Ru complexes or thienyl molecules used as sensitizers in the dye-sensitized solar cells (DSSCs). At first the N3 sensitizer is used as a model compound in a TiO2 DSSC to find out the factors that affect the photovoltage, photocurrent, and power conversion efficiency. The photovoltage is influenced by the photoinjected electron and the dark current caused from the recombination of electrons with I3- ions through TiO2/electrolyte and FTO/electrolyte interfaces. Thus, the photovoltage can be raised by increasing the injected electron and decreasing the recombination at the interface. The photocurrent induced from radiant power is affected by the value of incident photon-to-current conversion efficiency (IPCE) that is generated from energetic excitation of the sensitizer. The high absorption coefficient of the sensitizer extending to near infrared and long lifetime of photoinjected electron are necessary to achieve high photocurrent. The lifetime and transport of electrons in the TiO2 film can be estimated by transient photovoltage and photocurrent measurement. From the empirical model for a DSSC derived from the equivalent circuit and the reduction current, the behavior of the photocurrent-voltage curve can be predicted quantitatively. To fabricate a high-voltage, solid-state TiO2 DSSC, indium hexacyanoferrate (InHCF), was initially chosen as the redox couple. Since redox reaction of InHCF involves the K+ insertion/extraction, the KCl-saturated poly(2-acrylamido 2-methylpropanesulfonic acid) (K-PAMPS), a K+-conducting solid polymer electrolyte (SPE), was also incorporated to the cell. The imperfect N3 dye/InHCF contact by the SPE incorporation and slow solid-state diffusion of K+ in InHCF should be mainly responsible for the poor efficiency. Using a dye-incorporated SPE was found to improve the contact and could attain a regenerative cell. The imperfect dye/redox couple contact can be further improved by directly coating the lithium-doped redox couple, such as InHCF or cupric hexacyanoferrate (CuHCF), onto the dye/TiO2 surface. However, the poor contact between the counter electrode and dye/TiO2 electrode cause the charge-transfer resistance at the interface and voltage loss to enlarge. The slow reductive reaction between oxidized dye and Li ion-doped InHCF or CuHCF also plays an important role in determining photocurrent and efficiency. Besides, large charge-transfer resistance of CuHCF results in a low regenerating rate of oxidized N3 sensitizer and a low efficiency. Ruthenium(II) complexes with a new low pi* phenanthrenyl ligand (TAPNB) were obtained from Prof. Jiann T’suen Lin in Institute of Chemistry of Academia Sinica. The spectroscopic and electrochemical measurements showed that the excited states of those complexes matched the conduction band of titanium dioxide. The overall power conversion efficiencies of the solar cells utilized these new complexes as sensitizers for TiO2 films were less (under 1%) than that of N3-sensitized cell. Although the open-circuit voltage was similar to that of N3-sensitized cell, the short-circuit current was about one order lower. Such outcome may be attributed to the short lifetime of photoinjected electrons and less amount of dyes adsorbed due to the steric congestion of the complex. When NCS ligand was replaced by pyridyl ligand, the energy of metal-to-ligand charge transfer (Ru(II) → TAPNB) increased and resulted in blue shift of the absorption band. When carboxylic acid anchor was replaced by acetyl ester, the weaker interaction between the semiconductor and the ligand led to diminishing amount of the complex adsorbed and less photocurrent was detected. Four organic sensitizers based on benzothiadiazole and benzoselenadiazole chromophores with 2-thienyl-2'-cyano acrylic acid anchors were obtained from Prof. Jiann T’suen Lin in Institute of Chemistry of Academia Sinica and incorporated in DSSCs. When the phenylene linker is replaced by a thiophene unit, the improvements in donor property and coplanarity causes a red shift in absorption spectrum. However, the coplanar geometry can neither enhance the charge separation nor decelerate the recombination. The result is proved by the transient measurements. With the phenylene linker, the quantum yield is greatly improved and the cell efficiency approaches 4%. It may be due to the twisted nonplanar structure, which decelerates the recombination of charges. The organic sensitizers that contain diphenylamine donors and cyano acrylic acid acceptors bridged through an aromatic linker and a benzothiadiazole or benzoselenadiazole fragment ensure charge-transfer and facilitate charge separation. The conversion efficiency of a DSSC using the benzothiadiazole dye can reach as high as 4%.Acknowledgement I Chinese abstract III English abstract V Table of contents VII List of tables XII List of figures XIII Nomenclatures XX Chapter 1 Introduction 1 1-1 Solar cells 1 1-1-1 Photovoltaic cells 4 1-1-2 Photoelectrochemical cells 7 1-1-3 Dye-sensitized solar cells 10 1-2 Review on the DSSCs 12 1-2-1 TiO2 based DSSCs 12 1-2-2 The redox couples and mediators 13 1-2-3 The dye sensitizers 15 1-2-4 The characteristic of DSSCs 20 1-3 The motivations of this dissertation 23 1-4 The framework in this research 24 Chapter 2 Theories 27 2-1 Principle of semiconductor 27 2-1-1 Energy band structure 27 2-1-2 Fermi level and the junction 32 2-2 Theoretic power conversion efficiency 35 2-2-1 Solar spectrum 35 2-2-2 Charge transfer and power efficiency 37 2-3 Electron transfer of dye sensitizer 42 2-3-1 Corresponding mechanism in DSSCs 42 2-3-2 Electron injection efficiency 45 2-4 Electrochemical impedance and equivalent circuit analyses 47 2-4-1 Electrochemical impedance spectroscopy 47 2-4-2 Equivalent circuit and model 51 Chapter 3 Experimental 53 3-1 Instruments 53 3-2 Reagents and materials 54 3-3 Experimental procedures 56 3-3-1 Substrates and reagents 56 3-3-2 Preparation of dye-sensitized TiO2 film electrodes 56 3-3-2-1 TiO2 film preparation 56 3-3-2-2 Dye-sensitized TiO2 film preparation 57 3-3-3 Preparation of Pt film electrodes 58 3-3-4 Preparation of mediators 58 3-3-4-1 Liquid electrolyte of lithium iodide and iodine 58 3-3-4-2 Solid-state mediator of metal hexacyanoferrate 58 3-3-5 Preparation of organic dyes 59 3-3-5-1 Ruthenium dyes 59 3-3-5-2 Organic dyes 63 3-3-6 Cell assembling 64 3-4 Transient photovoltage and photocurrent 65 3-5 The properties of the TiO2 film 67 3-5-1 Thermogravimetric analysis 67 3-5-2 X-ray analysis 67 3-5-3 Surface area and pore size distribution analyses 68 3-6 Performance of the DSSCs 68 3-6-1 Experimental setup 68 3-6-2 Photocurrent-voltage (I-V) characteristic 68 3-6-3 Impedance measurement 69 3-6-4 Incident photon-to-current conversion efficiency (IPCE) 69 Chapter 4 Properties of TiO2 film and its application in the N3-based DSSC 70 4-1 Analysis and properties of TiO2 and Pt films 70 4-1-1 Thermogravimetric analysis 70 4-1-2 Surface area and pore size distribution analyses 72 4-1-3 TiO2 crystal and size analyses 75 4-1-4 Properties of sputtering Pt film 77 4-2 DSSCs based on N3 sensitizer 79 4-2-1 Power efficiency 79 4-2-2 IPCE measurement 81 4-2-3 Dye adsorption 83 4-2-4 Transient photovoltage and photocurrent 85 4-3 Equivalent circuit of the DSSCs 88 4-3-1 Equivalent circuit 88 4-3-2 Empirical model for I-V curve 91 Chapter 5 Solid-state mediators with high redox potential incorporated in DSSCs 95 5-1 Indium hexacyanoferrate redox couple 95 5-1-1 Electrochemical properties 95 5-1-2 Cell assembling with polymer electrolyte 97 5-1-2-1 Cell configuration 97 5-1-2-2 Cell performance 98 5-1-3 Cell designation 102 5-2 Cells assembling with lithium-doped InHCF and CuHCF films 103 5-2-1 Electrochemical properties in an acetonitrile solution 104 5-2-2 Cell performance 108 5-2-3 Transient photovoltage and photocurrent 111 Chapter 6 Ruthenium complexes designed for the DSSCs 114 6-1 Spectral and energetic properties of ruthenium complexes 114 6-1-1 Spectral properties 114 6-1-2 Energy level analysis 115 6-2 Performances of ruthenium complexes in DSSCs 119 6-2-1 IPCE measurements 119 6-2-2 Transient photovoltage and photocurrent 123 6-2-3 Power efficiencies 124 Chapter 7 Organic dyes incorporated with a thienyl anchor for the DSSCs 129 7-1 Spectral and energetic properties of organic dyes 129 7-1-1 Spectral properties 129 7-1-2 Energy level analysis 130 7-2 Performances of organic dyes in DSSCs 132 7-2-1 IPCE measurements 132 7-2-2 Transient photovoltage and photocurrent 134 7-2-3 Power efficiencies 136 Chapter 8 Conclusions and suggestions 141 8-1 Conclusions 141 8-1-1 N3 and I-/I3- based TiO2 DSSCs 141 8-1-2 Solid-state redox couples 141 8-1-3 Organic dyes 142 8-2 Suggestions 143 8-2-1 The TiO2 film electrode 143 8-2-2 The mediator 143 8-2-3 The sensitizer 144 Chapter 9 References 145 Appendix A Solar and experimental spectra 159 Appendix B Curriculum vitae 16