Sunscreen agents are chemical compounds widely used nowadays for skin protection from UV sunlight. Recently, their ubiquitous occurrence in aquatic systems has been evidenced, which poses a high risk for the environment and human health as they are associated with endocrine disrupting activity, reproductive toxicity, and genotoxicity. In this work, the feasibility of an economically and environmentally friendly catalytic system based on the thermally modified natural magnetite and hydrogen peroxide (Fe 3 O 4 -R400/H 2 O 2), has been evaluated for the degradation of two representative sunscreen agents: benzophenone-3 (BP-3) and 4-aminobenzoic acid (PABA) in wastewater. The experiments were conducted under circumneutral pH (pH 0 = 5), with temperature control (25 °C). Both compounds (500 μg L−1) were successfully removed from water by using a relatively low catalyst concentration (0.5 g L−1) and the theoretical stoichiometric H 2 O 2 dose for their complete oxidation (∼2.3 mg L−1). Afterwards, a complete operating condition study was performed with BP-3, given its predominant occurrence in fresh waters, analysing the influence of H 2 O 2 dose (1.2–4.6 mg L−1), catalyst concentration (0.1–0.5 g L−1), and temperature (25–45 °C). From the evolution of the identified by-products, a reaction pathway was proposed according to which oxidation of BP-3 gives rise to several aromatic intermediates, which finally evolve to short-chain organic acids. The generation of such aromatic by-products led to a considerably ecotoxicity increase in the initial stages of the reaction, but non-toxic effluents were ultimately achieved. Notably, the mineralization yield reached was above 60%. As a proof of concept, the feasibility of the system was finally demonstrated in real water matrices (WWTP effluent and surface water). [Display omitted] • The Fe 3 O 4 -R400/H 2 O 2 system was effective for the removal of sunscreen agents. • The pollutants were highly mineralized (∼60%), yielding short-chain acids. • A reaction pathway for BP-3 oxidation was proposed. • Non-toxic effluents were obtained despite a slight initial ecotoxicity increase. • The feasibility of the process was demonstrated in different water matrices. [ABSTRACT FROM AUTHOR]