With the depletion of the world's oil resources and the threat of global warming, it has become a long term goal to produce liquid fuels from sunlight. Using thermochemical cycles it is possible to transform heat into chemical energy. Hydrogen, for example, may be produced from a two-step cycle. In the thermal reduction (TR) step, a metal oxide, MOx, is reduced to a lower oxidation state, MOx-δ, with the release of oxygen. The thermal reduction step is highly endothermic and requires high temperatures, which may be obtained by concentrated solar power. In the gas splitting (GS) step MOx-δ is reoxidized by water at a lower temperature accompanied by the release of H2.
At the heart of the thermochemical cycle is a metal oxide that changes oxidation states during the cycles. Several binary metal oxides have been tested. CeO2 (ceria) showed the best results for thermochemical cycles. A conversion efficiency for solar to chemical energy of 0.7 % was demonstrated for a ceria-based material , however much larger efficiencies are possible in theory.
Following the work of Meredig and Wolverton , in order to maximize the efficiency of the process, the thermal reduction and gas splitting reactions must be as thermodynamically favorable as possible i.e. have the lowest possible Gibbs free energies. When choosing a temperature for those two reactions, these conditions impose limits on ΔHreduction and ΔSreduction. It was demonstrated that the vacancy formation energy has to be large enough to split water but should also be as small as possible to maximize the number of vacancies. The outcome of the aforementioned work is that there is an optimal oxygen vacancy formation energy window between 2.5 and 5 eV/O atom. This criteria is used to screen large datasets of compounds.
Due to their high tolerance to oxygen vacancy, lanthanum based perovskites have been experimentally used to perform solar thermochemical water splitting . They showed promising results in term of oxygen release during thermal reduction (up to eight times larger than ceria). Therefore, perovskite oxides (ABO3) seems to be promising candidates for thermochemical watersplitting. However, out many possible perovskites, only a small fraction of them have been experimentally tested. The remarkable stability of the perovskite structure suggests that there are potentially better systems to be discovered. When considering metals and semimetals of the periodic table of the elements as A and B atoms, the number of possible compositions for ABO3 structures is too large to be completely explored experimentally. High-throughput density functional theory (HT-DFT) is thus used to search efficiently and exhaustively for suitable candidates in this wide chemical space.
Solar Thermochemical Fuel Production in the Media