Pharmaceutical removal from different water matrixes by Fenton process at near-neutral pH: Doehlert design and transformation products identification by UHPLC-QTOF MS using a purpose-built database

TL;DR: In this paper, the authors evaluated the Fenton process in the removal of eight pharmaceuticals (gemfibrozil, nimesulide, furosemide, paracetamol, propranolol, dipyrone, fluoxetine, and diazepam) from three different water matrixes (distilled water, simulated wastewater, and hospital wastewater).

Abstract: This work evaluates the Fenton process in the removal of eight pharmaceuticals (gemfibrozil, nimesulide, furosemide, paracetamol, propranolol, dipyrone, fluoxetine, and diazepam), present at an initial concentration of 500 μg L−1 for each compound, from three different water matrixes (distilled water, simulated wastewater, and hospital wastewater). The Fenton process conditions (iron and hydrogen peroxide concentrations, and pH) were optimized for the distilled water matrix by Doehlert design associated with response surface methodology. These corresponded to initial Fe2+ and hydrogen peroxide concentrations of 12.5 and 533 mg L−1, respectively, and pH 5.0. Mineralization rates and pharmaceutical degradation were monitored for all water matrixes and different experimental conditions employed. Unique iron addition, using low (12.5 mg L−1) and increased concentrations (37.5 mg L−1) were evaluated. These preliminary results motivated the study of the Fenton process employing successive iron additions and using an excess of hydrogen peroxide in the reaction medium. Multiple iron addition favored higher oxidation of the initial contaminants. Finally, the most persistent transformation products generated during the Fenton process were identified by liquid chromatography coupled with quadrupole-time of flight mass spectrometry (UHPLC-QTOF MS) using a purpose-built database that allowed monitoring 97 transformation products in a single injection. In this study, 12 transformation products (TPs) were tentatively identified employing this purpose-built database. Most TPs generated were classified as of high toxicity (Cramer rules) and showed ready biodegradability (START biodegradability).