Role of surface reactions in the stabilization of n-CdS-based photoelectrochemical cells

The potential use of II-VI semiconductor-aqueous junction cells for the conversion of optical energy to electricity has previously been limited by semiconductor photodecomposition processes combined with low energy conversion efficiencies. Decomposition processes in the prototypical n-CdS photoelectrochemical cell can be efficiently suppressed by addition of an appropriate polychalcogenide redox couple to the electrolyte^1-3. However, conversion efficiencies remain low (∼5% at 488 nm)⁴. Moreover, although increased optical to electrical energy conversion rates can be obtained by using a redox couple such as Fe(CN)^4-/3-₆ (∼8% conversion efficiency at 488 nm), the cell lifetime is greatly diminished^5-7 (t_1/2∼1/2h). We report here that the photodecomposition of n-CdS in a Fe(CN)^4-/3-₆ electrolyte can be dramatically decreased and cell output parameters significantly improved by the presence of an appropriate combination of K⁺ and Cs ⁺ ions. Monochromatic (488 m) conversion efficiencies in excess of 20% have been observed, with fill factors (a measure of the ideality of the cell) in the range of 65%. The enhanced stability and efficiency are associated with in situ chemical derivatization of the n-CdS surface with a layer of K _x Cs_y[Cd^IIFe^II(CN)₆]. (This species is an analogue of Prussian blue having a C-bound Fe ^II/III centre and a nitrogen bound Cd^II centre. See, for example, ref. 8.)