The primary photoefficiency of ten semiconductors to produce reactive oxidising species under ultraviolet (UV) illumination has been evaluated. TiO2 in the anatase phase was shown to be at least an order of magnitude more efficient than the other semiconductors so was chosen for further experimentation. While the anatase-UV treatment was shown not to be feasible for secondary effluent disinfection, the process did lead to substantial reduction in the toxicity to algae of PCB containing wastewater. Also, the process reduced the total organic chlorine content of a pulp and paper plant's chlorinated bleach-plant effluent and left lignin model compounds more amenable to biological degradation. In addition, the process destroyed cyanide in synthetic solutions and in gold and base metal mine effluents. Based on studies of model compounds, the reaction mechanisms for the process are discussed. Also, recommendations are made as to future research requirements for optimization of the process and for study of its economic viability.