Evaluation of solar treatment in the absence and presence of TiO2 has been made to assess its effectiveness in reducing bacterial load with respect to drinking water standards. Field experiments under direct solar radiation were carried out using a compound parabolic collector (CPC) placed at the Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. Water contaminated with E coli K12 was exposed to sunlight in different seasons. The obtained results indicate that the presence of TiO2 accelerates the detrimental action of light. Total photocatalytic disinfection was obtained in both periods of year and no bacterial recovery was observed during 24h after stopping sunlight exposure. In the absence of TiO2, total disinfection was not always reached; and bacterial recovery was observed, especially when inactivation was not complete. Bacterial decay was mainly dependent on light intensity. It was also demonstrated that solar UV dose is not a pertinent parameter to standardize solar disinfection. The influence of the following topics on solar water disinfection is also studied in this paper: (a) UV and total solar spectra characteristics (b) volume of phototreated water (c) post-irradiation events. (C) 2004 Elsevier Ltd. All rights reserved.