The effectiveness of TiO2 to degrade organic pollutants, and its viability as water treatment technology has already been demonstrated. Nonetheless, most studies use TiO2 nanoparticles, which hinders its application at industrial scale due to economical and operational issues like recovering the catalyst after the treatment. The growth of immobilized TiO2 nanostructures by anodic oxidation seems a solution to this problem. However, the lack of reproducibility and poor understanding of the process has been the main drawback for its application at industrial scale. In the present work, the photocatalytic degradation of two commercial textile dyes in aqueous solutions (Rhodamine B and Methylene blue) was studied with three different types of immobilized TiO2 nanostructures and two different geometries (i.e. foil and net). These nanostructures were produced by anodic oxidation in the form of either nanotubes or nanoporous ASD coatings with good adherence, high aspect ratio and good spatial uniformity. Commercial purity titanium sheets were anodized in three different electrolytes: aqueous solution (ASD), fluoride containing aqueous solution (A-NT) and fluoride containing organic solution (O-NT). The surface pre-treatment, anodizing temperature and time, were changed, in order to obtain the best nanostructures in each electrolyte. After anodizing in solutions containing fluorides, samples were annealed at different temperatures and times to promote the formation of anatase phase TiO2 and to evaluate the influence of these parameters in the photocatalytic activity of the samples. These nanostructures were characterized by SEM and Raman spectroscopy. The degradation of Rhodamine B (RhB) and Methylene Blue (MB), pure or mixed, was performed by immersing a sample (6 cm2) in 25 ml of 10-5 M RhB solution and/or 1.25X10-4 M MB solution, respectively, irradiating for 6 h with a solar spectrum lamp (UV-A 3 mW/cm2) and monitoring color variations which represent dye degradation. In aqueous and organic solutions containing fluorides, self-aligned vertical nanotubes stem from the substrate; in organic solution the detachment of single nanotubes is less pronounced, creating a sort of porous template rather than a nanotubular array, but with higher thickness and better photoactivity efficiency. Conversely, in absence of fluorides and with the application of a larger cell voltage, an oxide with glassy appearance and larger pores is generated, with a photoactivity comparable to that obtained with the nanotubes growth in aqueous electrolyte. The photocatalytic degradation of dyes with these oxides increases when a chemical etching is applied as surface treatment before the anodization. At lower anodizing temperatures (5-7 ˚C) and times, a decrease in the photoactivity is obtained. The annealing process to crystallize the oxide in the nanotubular structures give the best results in terms of crystallinity and integrity of the nanotubes at 500 ˚C for 2 h. Both types of sample geometries (foil and net) show a good photoactivity, which allows more alternatives in the development and configuration of photocatalytic reactors. When mixes of the two dyes are used, a slight decrease in the kinetic of degradation is observed with respect to their pure forms, which may reduce photocatalytic efficiency in complex industrial wastewaters that generally contain more than one pollutant. In conclusion, the processes here described open the way to the production of photocatalytically efficient nanostructured TiO2 films immobilized onto a metallic substrate at low cost with different geometries, avoiding environmental issues related to the use and recycle of conventional non-immobilized photocatalysts, and allow the development of photocatalytic reactors with different configurations.