Advanced machine learning modelling and experimental analysis of ortho-nitrophenol degradation during UV254/chloramine treatment in the presence of Fe(III)

UV₂₅₄/chloramine (UV₂₅₄/NH₂Cl) efficiently degrades recalcitrant contaminants in the wastewater, while Fe(III) notably impacts the performance of the advanced oxidation treatment. This study investigated the combined and separate effects of UV₂₅₄, chloramine, and Fe(III) on ortho-nitrophenol degradation under varying influencing factors experimentally and computationally. The degradation of ortho-nitrophenol was enhanced from 33 % to 85 % during the UV₂₅₄/chloramine treatment via OH• (54.24 %) and RCSs (31.06 %) with the addition of Fe(III). Beyond the optimum level, increasing the oxidant concentration inhibited ortho-nitrophenol degradation. Among the four models, the Gradient Boosting and XGBoost models showed the best prediction performance. Possible degradation pathways of ortho-nitrophenol were proposed based on the identification of intermediate products (IPs) and density functional theory (DFT) calculations. The ECOSAR model predicted the ecotoxicity of ortho-nitrophenol and possible intermediate products. Finally, the formation of HNMs and the associated toxicity were explored. The UV₂₅₄/chloramine/Fe(III) treatment demonstrated significantly lower electrical energy per order (EE/O, as low as 2.02 kWh/m³/order), highlighting its superior energy efficiency compared to other advanced oxidation processes for nitrophenol degradation. Overall, the integration of Fe(III) into the UV₂₅₄/chloramine system significantly enhanced ortho-nitrophenol degradation efficiency (from 33 % to 85 %) by promoting •OH and RCS generation, while the use of advanced machine learning and DFT modelling provided a reliable predictive framework for treatment optimization and ecotoxicological risk assessment.