Hybrid dye-sensitised photocathodes for photoelectrochemical hydrogen evolution

Abstract

Nanostructured materials offer new, cost-efficient and sustainable opportunities for converting sunlight into electricity and fuels. The complexity of photoelectrochemical water splitting needs thorough understanding of the system and new ideas to overcome the limiting factors in the process. My work has been focused on investigating new hybrid dye-sensitised photocathodes for photoelectrochemical water splitting and developing a system that enables efficient photoelectrochemical H2 production. Firstly, the effect of the experimental environment to photoelectrochemical performance of the dye-sensitised NiO photocathode was investigated. For that, an optimal electron acceptor for photocurrent generation was chosen and applied in all experiments, while changing the experimental conditions. The effect of electrolyte composition, the pH of the electrolyte and the concentration of the electron acceptor were studied. It was observed that the highest photocurrent was obtained with 4,4’-dithiodipyridine electron acceptor with 5 mM concentration in pH 3 aqueous electrolyte. After determining the optimal conditions for photocurrent generation using the most robust dye available, the system was tested for H2 evolution with electrodeposited Pt catalyst. In addition, the NiO photocathode was tested with another dye with a polymeric structure and with a commercially available benchmark organic dye. Both systems were studied photoelectrochemically under the optimal conditions determined previously and also tested for H2 evolution with electrodeposited catalysts on the surface. The highest photocurrents were observed with the commercially available benchmark organic dye with an electron acceptor and H2 evolution was also shown on the same system with catalyst on the surface. Secondly, a new approach to increase the faradaic efficiency of dye-sensitised photocathodes for H2 evolution from water was investigated using integrated photocatalysts. Superiority to previously reported photocathodes was demonstrated, producing photocurrent densities of 30–35 μA cm-2 at an applied bias of -0.2 V vs. Ag/AgCl over 1 hour of continuous white light irradiation, resulting in the generation of 0.41 mmol h-1 cm-2 of H2 with faradaic efficiencies of up to 90%. Surface analysis of the photocathodes before and after photoelectrocatalysis revealed that the photocatalyst was photochemically stable, highlighting the benefits of the approach towards robust, hybrid solar-to-fuel devices. Furthermore, CuCrO2 as an alternative material to NiO was studied as a photocathode for photoelectrochemical H2 evolution. The aim was to show that the post-calcination of the photocathodes under N2 atmosphere changes the surface chemistry of the CuCrO2 photocathodes which translates into superior stability and efficiency. The origin of these properties was discussed with the help of X-ray photoelectron spectroscopy surface analysis. Stable H2 production on the CuCrO2 photocathode from aqueous buffer solution was shown and the efficiency of the process was further increased with the use of inorganic cobalt co-catalyst in solution. In addition, successful sensitisation of the CuCrO2 photocathode with an organic dye was demonstrated resulting in considerable increase in photocurrent when using an electron acceptor in solution. An additional study was made where Fe was added to the CuCrO2 to form CuCr0.5Fe0.5O2 mixed metal delafossite. The aim of this study was to see how the Fe3+ addition to the delafossite crystal lattice influences the photoelectrochemical performance and stability of the material. Finally, the photoelectrochemical cell was modelled and studied by building a simplified model in COMSOL Multiphysics and changing the exchange current density value on the electrode-electrolyte boundary based on experimental data obtained from photoelectrochemical measurement with different systems presented in this thesis. Results obtained from the model were discussed along with discussion on future optimisation possibilities of the model to increase the output of meaningful data.