Despite offering relatively high conversion efficiencies, dye-sensitised solar cells using liquid electrolytes containing either I-/I3 - redox couple or Co2+/Co3+ redox couple suffer from durability problems, such as electrode corrosion and electrolyte leakage. Replacements for liquid electrolytes have been extensively studied, but the efficiencies of the resulting devices remain low. One of the factors that limit the efficiency is the sensitising dye. Large sized hole-transport material results in poor pore-filling and thus leads to a fast back electron recombination that reduces the effective electron diffusion length to few micrometeres. The optimal TiO2 layer thickness (2 μm) for maximal power conversion in solid-state dye-sensitised solar cell is much smaller than the 6-10 μm layer thickness required for quantitative light absorption by many dye molecules. Thus, dyes that can absorb in both visible and near-IR region with high extinction coefficient are needed. In order to achieve this, novel oligomers and ruthenium (II) complexes are designed, synthesized and studied as sensitisers for both liquid state and solid state dye-sensitised solar cells in this thesis. Series of ‘donor-free’ dyes including oligo(3-hexylthiophene) (oligo-3HT) (Chapter 3) and oligo(4,4-dihexyl-4H-cyclopenta[1,2-b:5,4-b’]dithiophene) (oligo-CPDTs) (Chapter 4) functionalized with cyanoacrylic end groups are easily synthesized using cross-coupling. They were fully characterised through electrochemical, spectroscopic and computational techniques, showing versatile colour-tuning, as well as outstanding absorption coefficients up to 75000 M-1cm-1. Liquid and solid-state DSSCs device performances are studied and discussed in terms of the dye structures. These dyes are effective sensitisers for liquid-state and solid-state dye-sensitised solar cells, although they do not contain a typical donor group, thus open a new strategy of designing dyes in the future. New dyes containing different azo ligands as an additional chromophore moiety to enhance light harvesting of Ru complexes (Chapter 5) have been prepared using a protection/deprotection strategy that allows for convenient purification. The absorption spectrum of the dyes showed an enhanced light harvesting compared to the N719 dye that lacks the azo ligand and electrochemical study also showed properties suitable for application as sensitisers in DSSCs. Following hydrolysis, the complexes were investigated in DSSCs, with performance investigated using I-V measurements. Poor performance was observed and we attribute this as mostly likely due to poor charge injection due to short excited-state lifetime. Although the application of these current dyes in DSSCs is not feasible due to their poor performance, this study allowed us to determine the positions of the HOMO and LUMO orbitals and correlate it to the π-acidity of the dyes.

This thesis describes the synthesis of a novel family of porphyrins fused to anthracenes, together with investigations into their optical and electrochemical properties, as well as exploring their potential for application in dye-sensitised solar cells. Chapter 1 gives an overview of the structure-property relationships of large planar pi- systems for organic electronic applications. Porphyrins are introduced as suitable building blocks for such systems, and approaches for extending the pi-conjugation of these macrocycles are presented. A literature review of porphyrins fused to aromatic units is presented in Chapter 2, with a focus on the influence of structure on the optoelectronic properties of such systems. The chapter concludes with a summary of my previous work on the synthesis of anthracene-fused porphyrins, and the aims of this project are stated. Chapter 3 describes the syntheses of fully and partially fused bis-anthracene porphyrin monomers and dimers. By varying peripheral substituents, it was possible to solve problems of aggregation encountered for these systems. Fusion of anthracene units to a porphyrin core was found to result in systems displaying strong absorption in the near-IR, small HOMO-LUMO gaps, and low oxidation potentials. Chapter 4 explores the synthesis, crystal structure and optoelectronic properties of a porphyrin fused to four anthracenes, revealing this system to exhibit the longest wavelength absorption of any porphyrin monomer. The synthesis of a liquid crystalline tetra-anthracene-fused porphyrin was proposed, and attempts to synthesise the necessary anthracene precursors were undertaken. Chapter 5 describes the molecular design and synthetic pathway to a mono-anthracene fused porphyrin, and its unfused analogue, for use in liquid electrolyte dye-sensitised solar cells. By varying the metal oxide layer or lithium ion concentration of the device, it was possible to achieve incident photon to current conversion efficiency (IPCE) responses at wavelengths beyond 1050 nm. Chapter 6 details the experimental synthetic procedures and characterisation data for all the compounds synthesised during this project.

<p>Dye-sensitised solar cells (DSC) have attracted much attention during the last few years due to their high efficiencies and their potentially low production costs. The technology is based on a thin layer of nano sized, high band gap (3.2 eV) TiO2 film. A Ru containing dye (from hereon called the Ru dye) is chemisorbed onto the TiO2 film. This combination acts as the working electrode. The counter electrode consists of a platinum layer which is the catalyst for the regeneration of the Iodine/Iodide electrolyte. The work that is presented here is focused on the results that were obtained during studies of the performance of the DSC under certain reverse bias conditions. When one cell in the series connection in a module is shaded, the current will pass this cell in reverse bias. In such a case the shaded cell will be subjected to a voltage in the reverse direction coming from the other lit cells in the module. This reverse voltage could permanently modify or damage the cell if it is not properly protected. Although the work is focussed on the chemical stability of the dye, various techniques were employed to determine the physical changes in the cell. It was found that a cell that was subjected to a reverse bias of 2 V for 500 min showed a 58% recovery and a cell that was subjected to 4.5 V reverse bias was irreversibly damaged. The UV-vis spectra showed a blue shift (higher energy), the Raman showed no peak at 1713 cm-1 (which indicates the absence of free carboxylate groups) and the FT-IR showed the disappearance of the NC-S absorption band at 2100 cm-1. The combined conclusion is that the &ndash / NCS ligand has been depleted and replaced with I3- ions. When measuring the impedance, the Nyquist plots showed an increase in the charge transfer resistance at the counter electrode when subjected to a reverse bias potential of 2 V. This is confirmed by the Bode plots. This indicates a partial oxidation of the Pt catalyst on the counter electrode. It can therefore be stated with confidence that the changes in the cell after being subjected to a reverse bias potential of 2 V for 500 min are changes on the -NCS bonds on the Ru dye as well as the Pt in the counter electrode.</p>

<p>Dye-sensitised solar cells (DSC) have attracted much attention during the last few years due to their high efficiencies and their potentially low production costs. The technology is based on a thin layer of nano sized, high band gap (3.2 eV) TiO2 film. A Ru containing dye (from hereon called the Ru dye) is chemisorbed onto the TiO2 film. This combination acts as the working electrode. The counter electrode consists of a platinum layer which is the catalyst for the regeneration of the Iodine/Iodide electrolyte. The work that is presented here is focused on the results that were obtained during studies of the performance of the DSC under certain reverse bias conditions. When one cell in the series connection in a module is shaded, the current will pass this cell in reverse bias. In such a case the shaded cell will be subjected to a voltage in the reverse direction coming from the other lit cells in the module. This reverse voltage could permanently modify or damage the cell if it is not properly protected. Although the work is focussed on the chemical stability of the dye, various techniques were employed to determine the physical changes in the cell. It was found that a cell that was subjected to a reverse bias of 2 V for 500 min showed a 58% recovery and a cell that was subjected to 4.5 V reverse bias was irreversibly damaged. The UV-vis spectra showed a blue shift (higher energy), the Raman showed no peak at 1713 cm-1 (which indicates the absence of free carboxylate groups) and the FT-IR showed the disappearance of the NC-S absorption band at 2100 cm-1. The combined conclusion is that the &ndash / NCS ligand has been depleted and replaced with I3- ions. When measuring the impedance, the Nyquist plots showed an increase in the charge transfer resistance at the counter electrode when subjected to a reverse bias potential of 2 V. This is confirmed by the Bode plots. This indicates a partial oxidation of the Pt catalyst on the counter electrode. It can therefore be stated with confidence that the changes in the cell after being subjected to a reverse bias potential of 2 V for 500 min are changes on the -NCS bonds on the Ru dye as well as the Pt in the counter electrode.</p>

Philosophiae Doctor - PhD / Dye-sensitised solar cells (DSC) have attracted much attention during the last few years due to their high efficiencies and their potentially low production costs. The technology is based on a thin layer of nano sized, high band gap (3.2 eV) TiO2 film. A Ru containing dye (from hereon called the Ru dye) is chemisorbed onto the TiO2 film. This combination acts as the working electrode. The counter electrode consists of a platinum layer which is the catalyst for the regeneration of the Iodine/Iodide electrolyte. The work that is presented here is focused on the results that were obtained during studies of the performance of the DSC under certain reverse bias conditions. When one cell in the series connection in a module is shaded, the current will pass this cell in reverse bias. In such a case the shaded cell will be subjected to a voltage in the reverse direction coming from the other lit cells in the module. This reverse voltage could permanently modify or damage the cell if it is not properly protected. Although the work is focussed on the chemical stability of the dye, various techniques were employed to determine the physical changes in the cell. It was found that a cell that was subjected to a reverse bias of 2 V for 500 min showed a 58% recovery and a cell that was subjected to 4.5 V reverse bias was irreversibly damaged. The UV-vis spectra showed a blue shift (higher energy), the Raman showed no peak at 1713 cm-1 (which indicates the absence of free carboxylate groups) and the FT-IR showed the disappearance of the NC-S absorption band at 2100 cm-1. The combined conclusion is that the - NCS ligand has been depleted and replaced with I3- ions. When measuring the impedance, the Nyquist plots showed an increase in the charge transfer resistance at the counter electrode when subjected to a reverse bias potential of 2 V. This is confirmed by the Bode plots. This indicates a partial oxidation of the Pt catalyst on the counter electrode. It can therefore be stated with confidence that the changes in the cell after being subjected to a reverse bias potential of 2 V for 500 min are changes on the -NCS bonds on the Ru dye as well as the Pt in the counter electrode. / South Africa

With increasing concerns of global warming and the impending exhaustion of fossil fuels attention is being turned to renewable sources of energy. The sun supplies 3 x 1024 J per year to the earth which is around 104 times more energy than what the human race consumes. The world’s energy needs would be satisfied if a mere 0.1% of the planet’s surface was covered with solar cells(< = 10%)1, causing the conversion of solar energy (sunlight) into electricity to represent a very practical renewable source. Past research into solar energy has produced a photovoltaic device, which when coupled with highly coloured coordination compounds, enables this conversion. This device is known as a dye-sensitised solar cell (DSSC). Further research has been conducted into the properties of the dyes, and has shown that highly coloured coordination compounds are able to convert solar energy into electrical energy with the highest efficiencies. The dominant compounds in this area to date have been Grätzel’s ruthenium complexes and porphyrins. However, there exists a class of smaller compounds called dipyrrins, described most simply as “half a porphyrin”, which possess many of the attractive qualities for DSSC dyes. Although there are no examples of ruthenium-dipyrrin complexes in the literature, the combination of advantageous properties from both components represent very attractive synthetic targets with huge potential as dyes for DSSCs. The objectives of this thesis were firstly to develop a series of dipyrrin complexes which would be suitable as dyes for DSSCs; then to fully characterise the complexes and investigate the spectroscopic properties of each complex; and finally to determine the suitability of the complexes as dyes for DSSCs. These objectives were fully met, resulting in a set of generic target compounds characterised via 1H NMR, 13C NMR, mass spectrometry (ESI-MS), elemental analysis, and x-ray crystallography. From analyses of the UV-visible, fluorescence, emission, and Raman spectra; and electrochemistry results; the complexes were concluded to be suitable as dyes for DSSC’s. An additional bonus is that the syntheses for these complexes are applicable to any dipyrrin, thus aiding future studies into the use of dipyrrins as dyes for DSSC’s. This thesis summarises the findings of the above outlined research project.

This thesis examines the electro-optic, electric and electrochemical properties of liquid crystal (LC) materials in self-assembly systems, that is, liquid crystal-polymer electrolyte composites (LC-PEs), LC binary mixtures, and their potential application in dye-sensitised solar cells (DSSCs). The birefringence of LCs causes light modulation, which can be controlled by an applied voltage and electric field. In particular, the LCs are used as one of the components for the electrolyte redox couple which is responsible for charge transfer mechanism in DSSCs. In this work, LC-PEs were developed by dissolving LCs in polymer electrolytes; using a homologous series of cyanobiphenyls in a range of concentrations, alkyl chain lengths and dielectric permittivities. We found that doping the polymer electrolyte with 15% 4'-cyano-4'-pentylbiphenyl (5CB) improved ionic conductivity by up to 13 % compared to pure polymer electrolyte. Materials with positive dielectric permittivity and shorter alkyl chain length have been identified to be compatible with iodide/triiodide (I^-/I_3^-)-based polymer electrolytes. In DSSCs, 15% 5CB and 15% E7 LC-PEs exhibited the best efficiencies of 3.6 % and 4.0 %, respectively. In addition to LC-PEs, the self-assembly properties of smectic phase LCs were also utilised as templates for controlling the polymer structure in polymer electrolytes. A porous polymer network was prepared using various techniques including self-assembly, by applying an electric field and using a polyimide (PI) alignment layer. We found that the electrochemical and photovoltaic properties of these materials strongly correlated to the morphology/structure with the self-assembled structure, thus showing the best photovoltaic performance (5.9 %) even when compared with a reference solar cell (4.97 %). Finally, this thesis explores the interaction of LCs with graphene (Gr) in DSSC device architectures. Gr-based DSSCs were fabricated using different processing conditions, with the result being that Gr improved the performance of the DSSCs. The highest efficiency obtained was 5.48 % compared to the 4.86 % of a reference DSSC. The incorporation of LC-PEs in Gr-based DSSCs improved the performance of DSSCs was observed in devices with low concentrations of LCs due to the Gr inducing planar alignment of LCs. These results suggest a new strategy to improve DSSC efficiency by incorporating LC materials in the polymer electrolyte component. Even though these LCs are highly insulating, their self-assembly and dielectric polarisability help enhance ionic conductivity and optical scattering when doped into polymer electrolytes. This work can be extended in a fundamental way to elucidate the ionic conduction mechanism of LC-based electrolyte systems. Furthermore, it would be interesting if the benefits of using LC-PEs and smectic-templated polymer electrolytes (Sm-Pes) can be translated further in commercial electrochemical energy conversion systems.

This thesis presents a study of hybrid solar cells, specifically looking at various methods which can be employed in order to increase the power conversion efficiency of these devices. The experiments and results contained herein also present a very accurate picture of how rapidly the field of hybrid solar cells has progressed within the past three years. Chapters 1 and 2 present the background and motivation for the investigations undertaken, as well as the relevant theory underpinning solar cell operation. Chapter 2 also gives a brief review of the literature pertinent to the main types of devices investigated in this thesis; dye-sensitised solar cells, semiconductor sensitized solar cells and perovskite solar cells. Descriptions of the synthetic procedures, as well as the details of device fabrication and any measurement techniques used are outlined in Chapter 3. The first set of experimental results is presented in Chapter 4. This chapter outlines the synthesis of mesoporous single crystals (MSCs) of anatase TiO₂ as well as an investigation of its electronic properties. Having shown that this material has superior electronic properties to the conventionally used nanoparticle films, they were then integrated into low temperature processed dye-sensitised solar cells and achieved power conversion efficiencies of &GT; 3&percnt;, exhibiting electron transport rates which were orders of magnitude higher than those obtained for the high temperature processed control films. Chapter 5 further investigates the use of MSCs in photovoltaic devices, this time utilising a more strongly absorbing inorganic sensitiser, Sb₂S₃. Utilising the readily tunable pore size of MSCs, these Sb₂S₃ devices showed an increase in voltage and fill factor which can be attributed to a decrease in recombination within these devices. This chapter also presents the use of Sb₂S₃ in the meso-superstructured configuration. This device architecture showed consistently higher voltages suggesting that in this architecture, charge transport occurs through the absorber and not the mesoporous scaffold. Chapters 6 and 7 focus on the use of hybrid organic-inorganic perovskites in photovoltaic devices. In Chapter 6 the mixed halide, lead-based perovskite, CH₃NH₃PbI_3-xCl_x is employed in a planar heterojunction device architecture. The effects of Lewis base passivation on this material are investigated by determining the photoluminescence (PL) lifetimes and quantum efficiencies of treated and untreated films. It is found that passivating films of this material using Lewis bases causes an increase in the PLQE at low fluences as well as increasing the PL lifetime. By globally fitting these results to a model the trap densities are extracted and it is found that using these surface treatments decreases the trap density of the perovskite films. Finally, these treatments are used in complete solar cells resulting in increased power conversion efficiencies and an improvement in the stabilised power output of the devices. Chapter 7 describes the materials synthesis and characterisation of the tin-based perovskite CH₃NH₃SnI₃ and presents the first operational, lead-free perovskite solar cell. The work presented in this thesis describes significant advances in the field of hybrid solar cells, specifically with regards to improvements made to the nanostructured electrode, and the development and implementation of more highly absorbing sensitizers. The improvements discussed here will prove to be quite important in the drive towards exploiting solar power as a clean, affordable source of energy.

Philosophiae Doctor - PhD / The first report of a functioning photo-electrochemical solar cell in 1991 attracted a lot of interest from scientists and industrial groups. From an industrial point of view these so-called dye-sensitised solar cells (DSCs) offered the promise of moderate efficiency devices at ultra-low costs, owing to simple processing methods and the use of inexpensive materials. From an academic viewpoint, DSCs raised important scientific questions around the fundamental processes governing their operation and how these processes influence the photon-to-electron conversion efficiency of the cell. Major successes have since been achieved in understanding these processes, however the conversion efficiency of the best manufactured DSCs remains around 11%, significantly lower than that of their silicon photovoltaic counterparts. In traditional DSCs, charge generation is achieved by ultrafast electron injection from a photo-excited ruthenium-based dye molecule into the conduction band of a film of TiO₂ nanoparticles, subsequent dye regeneration by an I⁻ /I⁻₃ containing redox electrolyte and finally hole transportation to a platinum-coated counter electrode. The low DSC efficiencies are attributed to scattering of electrons at the interface between two TiO₂ nanoparticles leading to recombination with holes present in the redox electrolyte. Recent studies have shown that the application of films of highly ordered TiO₂ nanotubes instead of nanoparticles has the potential to improve the overall conversion efficiency of the cell. This is ascribed to the one-dimensional nature of nanotubes, which provides a linear transportation route for electrons generated during operation of the DSC. As a result the recombination probability of the electrons with nearby holes in the device is decreased. This work investigated the synthesis of Al₂O₃-coated TiO₂ nanotubes via the anodisation technique for application in DSCs. TiO₂ nanotube arrays with an average length of 15 μm, diameter of 50 nm and wall thickness of 15 nm were synthesised via anodisation using an organic neutral electrolyte consisting of 2 M H₂O + 0.15 M NH₄F + ethylene glycol (EG) at an applied voltage of 60 V for 6 hours. In addition, scanning electron microscope (SEM) micrographs showed that anodisation at these conditions yields nanotubes with smooth walls and hexagonally shaped, closed bottoms. X-ray diffraction (XRD) patterns revealed that the as-anodised nanotubes were amorphous and as such were annealed at 450 °C for 2 hours in air at atmospheric pressure, which yielded crystalline anatase TiO₂ nanotubes. Highresolution transmission electron microscope (TEM) images revealed that the nanotube walls comprised of individual nano-sized TiO₂ crystallites. Photoluminescence (PL) spectroscopy showed that the optical properties, especially the bandgap of the TiO₂ nanotubes are dependent on the crystallinity, which in turn was dependent on the structural characteristics, such as the wall thickness, diameter and length. The PL measurements were supplemented by Raman spectra, which revealed an increased in the quantum confinement of the optical phonon modes of the nanotubes synthesised at low anodisation voltages, consequently yielding a larger bandgap The annealed nanotubes were then coated with a thin layer of alumina (Al₂O₃) using a simple sol-gel dip coating method, effectively used to coat films of nanoparticles. Atomic force microscopy (AFM) showed that the average nanotube diameter increased post sol-gel deposition, which suggests that the nanotubes are coated with a layer of Al₂O₃. This was confirmed with HR-TEM, in conjunction with selected area electron diffraction (SAED) and XRD analyses, which showed the coating of the nanotube walls with a thin layer of amorphous Al₂O₃ with a thickness between 4 and 7 nm. Ultraviolet-visible (UVvis) absorbance spectra showed that the dye-adsorption ability of the nanotubes are enhanced by the Al₂O₃ coating and hence is a viable material for solar cell application. Upon application in the DSC, it was found by means of photo-current density – voltage (I – V) measurements that a DSC fabricated with a 15 μm thick layer of bare TiO₂ nanotubes has a photon-to-light conversion efficiency of 4.56%, which increased to 4.88% after coating the nanotubes with a layer of alumina. However, these devices had poorer conversion efficiencies than bare and Al₂O₃-coated TiO₂ nanoparticle based DSCs, which boasted with efficiencies of 6.54 and 7.26%, respectively. The low efficiencies of the TiO₂ nanotube based DSCs are ascribed to the low surface area of the layer of nanotubes, which yielded low photocurrent densities. Electrochemical impedance spectroscopy (EIS) showed that the electron lifetime in the alumina coated nanotubes are almost 20 times greater than in a bare layer of nanoparticles. In addition, it was also found that the charge transfer resistance at the interface of the TiO₂/dye/electrolyte is the lowest for an Al₂O₃-coated TiO₂ layer.

Photovoltaic cells are a promising solution to the current energy crisis. Among the different photovoltaic cell technologies developed, dye-sensitised solar cells (DSSC) are emerging as viable low-cost alternatives to Si PV technology. This thesis presents studies on fabrication and characterisation of TiO2 based dye-sensitised solar cells. Chapter 1 gives an overview of different photovoltaic cell technologies and a review of the state-of-the art DSSC technology. Chapter 2 describes the techniques used for characterisation of DSSCs. Chapter 3 describes the fabrication of TiO2 based dye-sensitised solar cells. Chapter 4 presents the analysis of measurements obtained by the characterisation techniques. Finally chapter 5 summarises the work done and suggests directions for future work.