Ionic material for photoelectrochemical solar cells
thesisposted on 31.01.2017, 05:17 authored by Armel, Marie Josephe Vanessa
The pressure on governments to find alternative energy sources which are cheaper, sustainable and also less polluting, has triggered a significant amount of research into solar cells. Conventional solar cells are mainly based on silicon, but the cost of production and materials of these types of devices is high. Dye sensitised solar cells (DSSCs) are an alternative type of solar cell that uses cheaper components. Since their discovery in 1991, many research groups across the world have investigated this type of device in an attempt to understand and optimise their operation. There are several factors that limit the performance of these devices, such as corrosion of the counter electrode, leakage of the commonly used molecular liquid electrolyte and insufficient light absorption by the sensitisers. The aim of this research was to synthesise and characterise novel ionic liquids for use as the electrolyte in dye sensitised solar cells, and also understand the behaviour of commonly used ionic liquids, such as the imidazolium-based family, when in contact with the TiO2 photoanode layer. New phosphonium ionic liquids were synthesised for use as electrolytes in DSSCs. Different anions such as tetrafluoroborate, hexafluorophosphate, dicyanamide, bis(trifluoromethanesulfonyl)amide, thiocyanate and bis(fluorosulfonyl)imide, combined with different asymmetric phosphonium cations, were used to prepare either ionic liquids, or solids that show plastic crystal behaviour. The chemical and physical properties of these phosphonium ionic liquids were measured and they all show relatively good thermal stability, except for the bis(fluorosulfonyl)imide series. They also exhibit good ionic conductivity and are considered to be “good” ionic liquids according to their position on the Walden plot. These new phosphonium ionic liquids were utilised as electrolytes in dye sensitised solar cells using dithienothiophene organic dyes, and the results compared to those obtained using the common ruthenium-based dyes. The effect of film thickness, scattering layer and particle size were investigated with the organic sensitisers. With these new dithienothiophene sensitisers, a thin (2 μm) transparent TiO2 layer with a scattering layer of ~ 6 μm is the optimum for the device to perform well. The addition of chenodeoxycholic acid in the dye bath showed an improvement in device efficiency. Addition of small amounts of solvent such as water, valeronitrile or tetraglyme are also discussed. These novel phosphonium ionic liquids show promising behaviour in these cells. The best device performance was achieved with the least viscous phosphonium ionic liquids containing the methoxy group on one of the alkyl chains. The power conversion efficiencies using these dithienothiophene dyes were all > 5 % under full sun and > 6 % at low sun intensity. In order to understand the behaviour of ionic liquids in dye sensitised solar cells, the effect of this class of material in contact with the semiconductor was investigated by measurement of flatband potentials. Depending on the “acidity” or “basicity” of the neat ionic liquids, the position of the conduction band edge of the semiconductor is shifted. The purity of the ionic liquid is also a very important factor as it can affect the position of the flatband potential of the TiO2-ionic liquid junction. For example, C2mimBF4 gives two different flatband potentials depending on the quality of the ionic liquid from the supplier. The effects of acid treatment and the addition of additives such as lithium iodide and N-methylbenzimidazole to the ionic liquid were also investigated. As expected, addition of small ions such as Li+ or H+ shift the flatband potential towards more positive values, whereas the presence of basic materials such as 4-tert-butylpyridine or N-methylbenzimidazole gives more negative potentials. These additives as well as the ionic liquids play an important role in dye sensitised solar cells, especially on the open circuit voltage. The use of imidazolium, ammonium and phosphonium based ionic liquid electrolytes was also investigated with porphyrin sensitisers and transient spectroscopy measurements were performed to elucidate the influence of the ionic liquid on the device performance. Another interesting class of materials, which are related to the family of ionic liquids studied here, are the organic ionic plastic crystals. These were investigated as potential solid electrolytes in dye sensitised solar cells. It is the first time that these materials have been successfully used in DSSCs. Relatively good performance was obtained with an electrolyte based on the organic ionic plastic crystal, C1mpyrN(CN)2. Performance of > 5 % was obtained with Ruthenium as the sensitiser under full sun intensity. Finally, the use of another type of solid electrolyte based on succinonitrile was investigated with both the porphyrin sensitiser (P159) and with a ruthenium based sensitiser (N719). Over 5 % efficiency was achieved at low sun intensity with the porphyrin dye. This is the first time that such a good performance has been obtained with a solid electrolyte in porphyrin based DSSCs.