Pharmaceuticals, such as dipyrone (DIP), paracetamol (PCT), and propranolol (PPN), are widely used analgesics and beta-blockers with the greatest presence in wastewaters and, consequently, in natural waters. The present work evaluated solar light-driven photocatalyst from petrochemical industrial waste (PW) as a strategy for the degradation of three pharmaceuticals in different water matrices (distilled water-DW, simulated wastewater-SWW, and real hospital wastewater-RHWW). All experiments were carried out in a solar photo-reactor with a capacity of 1 L and the experimental condition employed was a catalyst concentration of 350 mg L-1 at pH 5.0; these conditions were selected considering the Doehlert design validation spreadsheet and the desirability function. All materials prepared were conveniently characterized by zeta potential, small-angle X-ray scattering (SAXS), diffuse reflectance ultraviolet-visible (DRUV), and infrared spectroscopy. According to the results of the characterization, significant differences have been observed between the PW and the photocatalyst such as vibrational modes, optical absorption gap, and acid-basic characteristics on the surface, which suggests the potential use of the photocatalyst in the degradation of contaminants of emerging concern. Based on pharmaceutical degradation, DIP showed the highest photosensitivity (87.5%), and therefore the highest photocatalytic degradation followed by PPN; both compounds achieved final concentrations below the limit of quantification of the chromatographic method in DW. However, PCT was the most recalcitrant pharmaceutical in all matrices. Radicals from chromophoric natural organic matter (NOM) could improve PCT degradation in the SWW matrix (56%). Nevertheless, the results in RHWW showed a matrix effect with decreased the oxidation percentages (DIP-99%; PPN-71%; PCT-17%); hence, the addition of an oxidant such as H2O2 was studied as a pharmaceutical oxidation boost in RHWW. PPN was the molecule most sensitive to this strategy of oxidation (98%). Furthermore, 20 transformation products (TPs) generated throughout the treatment were identified by LC-QTOF MS using a customized TPs database. According to quantitative structure activity relationship (Q)SAR analysis, more than 75% of the TPs identified were not biodegradable. About 35% of them have oral toxicity characteristics indicated by Cramer's rules, and the DIP TPs represent high toxicity for different trophic levels.