Investigation into High Efficiency Visible Light Photocatalysts for Water Reduction and Oxidation

Abstract

Solar water splitting using an inorganic semiconductor photocatalyst is viewed as one of the most exciting and environmentally friendly ways of producing clean renewable fuels such as hydrogen from abundant resources. Currently, there are many diverse semiconductors that have been developed, the majority for half reactions in the presence of sacrificial reagents. However, for industrial facilitation, there exists an essential, non-debatable trifecta of being robust, cheap and efficient for overall water splitting. To date, no system has combined all three, with most examples missing at least one of the necessary trio. Therefore one of the current challenges of the field is to develop low cost, highly efficient and stable photocatalysts for industrial scale-up use. In order to achieve that aim, researchers must focus on novel semiconductors to improve efficiencies and also understand the fundamental mechanisms. The primary focus of this thesis then, is to investigate some of the newest photocatalysts for water photooxidation, reduction, and overall water splitting. In doing so, the thesis aids to shed light on the mechanisms behind what makes certain photocatalysts either efficient or inefficient. Firstly, test station was set up to analyse gaseous products such as hydrogen and oxygen produced from photocatalytic water splitting, by using a custom made high purity borosilicate reactor in conjunction with a gas chromatography unit. Gaseous products could be measured with very small sampling error (<1%), which improved the throughput of experiments. The photooxidation of water using a novel faceted form of Ag3PO4 was investigated. A novel synthetic method was created that made it possible to control the exposing facets of silver phosphate in the absence of surfactants to yield tetrahedral crystals composed entirely of {111} facets. It was found that due to high surface energy of {111}, and low hole (h+) mass in the 111 direction, Ag3PO4 tetrahedral crystals could outperform all other low index facets for the oxidation of water under visible light. The quantum yield was found to be nearly unity at 400 nm, and over 80% at 500 nm. With the exception of Ag3PO4 tetrahedral crystals, no photocatalyst has exhibited quantum efficiencies reaching 100% under visible irradiation. Therefore, the strategy of morphology control of a photocatalyst, led by DFT calculations of surface energy and charge carrier mobility, in order to boost photooxidation yield has been demonstrated to be very successful, and could be applied to improve other semiconductors in future research. Hydrogen production from water was further studied using the only known robust organic photocatalyst, graphitic carbon nitride (g-C3N4). It was discovered that using a novel preparation method, urea derived g-C3N4 can achieve a quantum yield of 26% at 400 nm for hydrogen production from water; an order of magnitude greater than previously reported in the literature (3.75%). The stark difference in activity is due to the polymerisation status, and consequently the surface protonation status as evidenced by XPS. As the surface protonation decreases, and polymerisation increases, the performance of graphitic carbon nitride for hydrogen production increases. The rate of hydrogen production with respect to BET specific surface area was also found to be non-correlating; a juxtaposition of conventional photocatalysts whose activity is enhanced with larger surface areas - believed to be because of an increase in surface active sites. Finally, overall water splitting was probed using Z-scheme systems comprising of a redox mediator, hydrogen evolution photocatalyst, and oxygen evolution photocatalyst. Ag3PO4 was found not be not suitable for current Z-scheme systems, as it is unstable in the pH ranges required, and also reacts with both of the best known electron mediators used in Z-schemes, as evidenced by XRD, TEM, and EDX studies. However, it has been demonstrated that urea derived g-C3N4 can participate in a Z-scheme system, when combined with either WO3 or BiVO4 – the first example of its kind, resulting in a stable system for an overall water splitting operated under both visible light irradiation and full arc irradiation. Further studies shows water splitting rates are influenced by a combination of pH, concentration of redox mediator, and mass ratio between photocatalysts. The solar-to-hydrogen conversion of the most efficient system was experimentally verified to be ca. 0.1%. It is postulated that the surface properties of urea derived graphitic carbon nitride are related to the adsorption of redox ions, however, further work is required to confirm these assumptions.