Paper EH-ThM9
Photo-Electrochemical Energy Conversion and Storage

Thursday, December 11, 2014, 10:40 am, Room Lehua

Global (mean) power demands of ca. 1.5x10¹³ W, projected to double by 2050, could be provided from the earth’s ultimate power source, ca. 1.2x10¹⁷ W of incident solar radiation, if adequately efficient, robust and economic transducers are developed. However, the diurnal nature of solar power requires that such transducers are coupled to energy storage, preferably in chemical bonds for high specific energy / energy density, and to fuel cells for subsequent conversion to electrical energy. Such systems could decarbonise power sources, manage intermittency of renewable power sources and smooth the dynamics of electrical power demands. This can be achieved if electrons from photovoltaic panels are used, for instance, to electrolyse water to form (oxygen and) hydrogen, which can be oxidised subsequently in fuel cells. Alternatively, solar energy can be used directly, and potentially more cheaply, for photo-electrochemical reduction (and oxidation) of water in an environmentally benign route to hydrogen (and oxygen).

As in photovoltaic cells, a semiconducting material may be used to absorb solar photons with energies (hv) greater than the semiconductor’s band gap, generating electrons in its conduction band (e^-_CB) and highly oxidising electron ‘holes’ in its valence band (h⁺_VB). The semiconductor needs to be chosen judiciously, so that: (a) electrons at its conduction band edge have sufficient energy to reduce water to hydrogen, and (b) holes at its valence band edge are sufficiently energetic to oxidise water to oxygen. Though the feasibility of such processes is well established, practical reactor systems have yet to be deployed, because the semiconductors also need to be stable, well-matched to the solar spectrum and achieve acceptable photon-to-hydrogen energy conversion efficiencies. Unfortunately, no single material yet meets all these criteria, to enable such artificial photosynthetic reactors to be: efficient, robust and cheap, of which only any two properties are achievable at present.

200 nm thin films of n-type α-Fe₂O₃ photo-anodes were produced by automated spray pyrolysis of iron(III) salts dissolved in ethanol onto F-doped SnO₂ coated glass or perforated titanium substrates heated to 450 °C and with an open area of ca. 17 %. These photo-anodes were deployed in laboratory-scale photo-electrochemical reactors, designed initially for photo-assisted electrolysis only, rather than spontaneous photo-electrolysis; a metal cathode supported hydrogen evolution with an electrical energy input.

Results will be reported for the thermodynamic and kinetic constraints on such processes, together with the effects of experimental variables on H₂ production rates.