Optimization of azithromycin degradation: Integrating ANN-PSO modeling, intermediates analysis, identification, and microbiological assessment

This study aims to identify the by-products generated during the degradation process of azithromycin (AZM), a prevalent antibiotic during the COVID-19 pandemic, via a novel chlorine/ferrous-based advanced oxidation process. The optimal conditions for AZM degradation were determined using a neural network coupled with particle swarm optimization modeling. The by-products generated during the process were identified using QuEChERS based on dispersive solid-phase extraction and direct-infusion high-resolution mass spectrometry analysis techniques. The efficacy of the proposed method was validated through microbiological assays conducted pre- and post-degradation of AZM. The optimal values determined were [NaClO] = 200 μM, [AZM] = 35.37 mg L–1, [FeSO4] = 3.5 mg L–1, and a pH 2.51, providing a predicted yield of 99.9%. Five photoproducts, Azit-DP1 through Azit-DP5, were successfully identified with respective m/z values of 357.11533, 159.06552, 146.08127, 100.07701, and 83.04969. A proposed mechanism for the degradation pathway was presented. The biotests results confirmed that as the quantity of AZM in the medium decreases due to degradation, its antimicrobial effectiveness diminishes accordingly, eventually disappearing completely after 4 h of the degradation process. The results of the performed biotests confirmed the effectiveness of the proposed low-cost system in removing azithromycin and its by-products from the aqueous medium.