Under the best experimental conditions in DW and SW (pH 6.0, 150 mg L-1 of TiO2 and 6 mg L-1 of H2O2) and, in STP (pH 7.4 and 450 mg L-1 of TiO2), the influence of the radiation source (black light and sunlight) was evaluated. High degradation efficiencies were obtained for all pharmaceuticals, resulting in values below the limit of detection (LOD) (25, 3.3, and 0.76 μg L−1 for GEM, HCTZ and NAP, respectively) in all matrices evaluated under radiation with black light. However, a higher dose of energy was required with the increase in the complexity of the matrices: STP (527 kJ m-2), SW (371 kJ m-2) and DW (448 kJ m-2). Similar results were obtained with the use of sunlight, when working under the same conditions used for black light. These results show that the components of the matrices affect the degradation efficiency. This is justified by the presence of optically active substances that limit the penetration of light, the presence of inorganic anions and organic matter that compete with the target compounds for hydroxyl radicals.
Although heterogeneous photocatalysis was efficient to degrade target compounds, there was no complete mineralization. In DW, mineralizations of 75% (sunlight) and 85% (black light) were obtained after 448 kJ m-2. Using the same dose of energy, 64% (sunlight) and 72% (black light) of mineralization were obtained for the SW matrix. In STP, there was 75% mineralization for both sources of radiation after 523 kJ m-2. It is noteworthy that although a higher percentage of mineralization was obtained in DW compared to the matrices of SW and STP, the amount of organic carbon removed in mg L-1 was higher for SW and STP, since the initial DOC value in DW is 0.77 mg L-1 while in SW it is 6.7 mg L-1 and, in STP, 25 mg L-1.
Since complete mineralization was not achieved, by-products of the target compounds or from organic matter naturally present in the SW and STP matrices were formed, being necessary to evaluate the evolution of the solution's toxicity before and after treatment. In DW, the initial inhibition was 1% and increased to 99% (black light) and 83% (sunlight) when 448 kJ m-2 was applied. At the same dose of energy, in SW, the initial toxicity of the matrix (in natura and after the addition of pharmaceuticals) was 3.3% and increased to 77 and 99%, respectively, with black and sun lights (with or without the addition of pharmaceuticals). In STP, the initial matrix toxicity (with or without the addition of pharmaceuticals) was 50% and increased to 79% (fresh matrix) and 85% (addition of pharmaceuticals) with the application of 523 kJ m-2, with no significant differences being obtained with the use of sunlight. The increase in toxicity observed for DW comes from the by products generated during the degradation of pharmaceuticals.