Photocatalytic hydrogen evolution over CuCrO 2 S. Saadi, A. Bouguelia, M. Trari * Laboratoire de Stockage et de Valorisation des Energies Renouvelables, Faculte ´ de Chimie, U.S.T.H.B BP 32 Algiers, Algeria Received 23 September 2004; received in revised form 16 February 2005; accepted 18 February 2005 Available online 5 April 2005 Communicated by: Associate Editor Gion Calzaferri Abstract We have been studying the technical feasibility of a photochemical H 2 evolution based on a dispersion of CuCrO 2 powder in aqueous electrolytes containing various reducing agents (S 2 , SO 2 3 and S 2 O 2 3 ). The title oxide combines a fair resistance to corrosion with an optimal band gap E g of 1.32 eV. The intercalation of a small amount of oxygen should be accompanied by a partial oxidation of Cu + into Cu 2+ implying a p-type semiconductivity. The S 2 oxidation inhibits the photocorrosion and the H 2 evolution increases parallel to polysulfides S 2 n formation. Most of H 2 is pro- duced when p-CuCrO 2 is connected to n-Cu 2 O formed in situ. H 2 liberation proceeds mostly on CuCrO 2 while the oxi- dation of S 2 takes place over Cu 2 O surface and the hetero system Cu 2 O/CuCrO 2 is optimized with respect to some physical parameters. The photoactivity is dependent on preparation conditions and lowering the synthesis temperature through nitrate route leads to an increase in specific surface area S sp . The photoelectrochemical H 2 production is a mul- tistep process where the rate determining step is the arrival of electrons at the interface because of their low mobility. Prolonged irradiation (>80 min) leads to a pronounced decrease of the photoactivity; the tendency toward saturation is due to the undesired back reduction of polysulfides S 2 n in a closed system and to their strong absorption in the visible region (k max = 520 nm). Ó 2005 Elsevier Ltd. All rights reserved. Keywords: H 2 evolution; p-type semiconductivity; Sulfide; CuCrO 2 ; Low mobility 1. Introduction The chemistry based on solar energy conversion has been a subject of intensive research over the past dec- ades on a variety of oxides and sulfides to ascertain their practical applications to water splitting (Li et al., 2001). An unbiased photodecomposition implies a conduction band (CB) potential negative of the H 2 O/H 2 level and a valence band (VB) potential more positive than that of O 2 /H 2 O. These restrictions are encountered with stable oxides (El Moustafid et al., 2002) whose photosensitivity is limited to the UV re- gion of sun spectrum and therefore of little practical use. This comes because the O 2 -2p orbital constitutes mainly VB, too low in energy owing to its large electro- negativity. As a consequence, CB is positioned at a po- tential insufficiently negative to liberate gaseous hydrogen with appreciable rates. Some alternatives have been attempted to overcome this drawback and one of them is to use new materials where both VB and CB bands are of cationic character. In the layered 0038-092X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.solener.2005.02.018 * Corresponding author. Tel.: +213 2124 7950; fax: +213 2124 7311. E-mail address: mtrari@caramail.com (M. Trari). Solar Energy 80 (2006) 272–280 www.elsevier.com/locate/solener