Effect of cyanide on the photo-electrochemical response of n-CdSe/Fe(CN)^4-/3-₆ electrolytic cell

PHOTOELECTROCHEMICAL cells based on cadmium chalco-genide electrodes in an aqueous ferrocyanide electrolyte^1-9 provide a model system for studying the fundamental processes associated with photo-induced interfacial charge transfer, and also hold the promise of converting solar energy to electricity and useful chemical products with high efficiency. Licht⁹ has suggested that it is possible to devise n-CdSe/[Fe(CN)^3-/4-₆]_aq cells with a maximum solar-energy conversion efficiency of 16.4%, which derives from an open-circuit photovoltage of ∼1.2 V, a value that approaches the thermodynamic limiting value for a semiconductor with a bandgap of 1.7 eV. Licht suggested that improvements in energy conversion efficiency are possible by modifying the electrolyte so as to prevent chemical modification of the semiconductor surface while optimizing the equilibrium concentrations of critical solution species. Here we report that addition of KCN to a ferro/ferricyanide electrolyte does not prevent surface modification on either the (11&2macr;0) face of n-CdX (where X is S or Se) or the (0001) face of n-CdS. We show instead that the current-voltage response observed by Licht is due to a hidden negative ΔG in the system associated with the thermodynamically 'downhill' oxidation of CN^-, which accounts for as much as 700 mV of the reported 'open circuit' photovoltage.