This study investigated the degradation of eight aliphatic halogenated contaminants (one brominated flame retardant and seven disinfection by-products) in synthetic drinking water by the UVA/TiO2 and UVA/Cu–TiO2 processes. The degradation rate constants of 2,2-bis(bromomethyl)-1,3-propanediol and trichloromethane in the UVA/Cu–TiO2 process were 10.1 and 1.29 times, respectively, higher than those in the UVA/TiO2 process. In contrast, the degradation rate constants of dichloroacetaldehyde, monochloroacetonitrile, monobromoacetonitrile and dibromonitromethane in the UVA/Cu–TiO2 process were 8.15, 2.33, 2.84 and 1.80 times, respectively, lower than those in the UVA/TiO2 process. The degradation rate constants of monobromonitromethane and dichloronitromethane were comparable in the two processes. The relationships between the degradation rate constants and the structural characteristics of the selected contaminants were examined to explain the different degradation efficacies of the contaminants in the two processes. As suggested by a quantitative structure-activity relationship (QSAR) model, the UVA/TiO2 process favored the degradation of contaminants with more polar electron-withdrawing moieties and higher degrees of chlorination. While the UVA/Cu–TiO2 process favored the degradation of hydrophilic unsaturated contaminants with multiple bonds. The concentrations of the reactive species (e.g., HO and e−) generated in the two photocatalytic processes were quantified using competition kinetics. The UVA/Cu–TiO2 process generated >10 times higher concentrations of HO than the UVA/TiO2 process, suggesting that the former process was more suitable for the degradation of contaminants that are reactive towards HO, while e− and e−-derived superoxide radicals were non-negligible contributors to contaminant degradation in the UVA/TiO2 process.