Cu-doped TiO~ could be considered as a better absorber of sunlight for solar energy conversion. The Cu-doped TiO2 has been prepared by anodic oxidation of a Ti-2.5% Cu alloy. The structural properties of these anodic oxides are examined by electron microdiffraction, XPS, and nuclear microanalysis. The optical properties of both the Ti-Cu alloy and the anodic films have been determined as functions of the thickness of the oxides. It is concluded that Cu doping o~ TiO2 creates an extra state in the forbidden gap of TiO2 which gives two extra peaks in the absorption spectrum of the oxide. But for solar energy applications, the preparation of the doped oxide by anodic oxidation is not suitable because one does not control the level of impurity.Titanium dioxide electrodes created a great interest in the field of semiconductor-electrolyte interfaces when it was shown by Fijishima and Honda (1). that water photoelectrolysis was possible without degradation or decomposition of the semiconductor electrode. For solar applications, however, TiO2 is not suitable because the bandgap of this oxide is 3 eV (410 nm) too far away in the u.v. region to significantly absorb the solar energy. A great amount of work has been done to find the right semiconductor electrode with a bandgap around 2 eV (620 nm) and stability comparable to the TiO2 electrode; two review articles (2, 3) summarize these works. Among the numerous materials or combinations of materials which have been tested to do this water photoe]ectrolysis, TiO2 remains high in the list since it is the most stable material for a long period of time. So the problem is to decrease the TiO~ bandgap or increase the optical absorption in the visible region by increasing the number of defects (holes, vacancies, etc.). Both approaches have been tested on specific systems; Ghosh and Maruska (4) have reported water photoelectrolysis using visible light on a Cr-doped TiO2 electrode, Augustinski et al. (5) have also reported results for TiO~ doped with 6% A120~.Keeping in mind the possible applications of this study we have chosen Cu to fill the bandgap of TiO2. The preparation of Cu-doped TiO2 was made by anodic oxidation of a Ti-Cu alloy which is commercially available and has good mechanical properties. Our purpose is also to show the importance of the knowledge of the optical indexes of both the substrate and the semiconductor film to determine the bandgap shift and the absorption variations due to the incorporation of copper.