Green-synthesized Ag/Ag3O4/CuO from Inula viscosa coupled with peroxydisulfate for Safranin O degradation: modeling and optimization using DT_LSBOOST and dragonfly algorithm
Reguia Boudraa, Atmane Djermoune, Djahida Touati-Talantikite, Hadjar Blkacemi, Kamilia Madi, Rachida Bouallouche, Farid Ait Merzeg, Mohammed Kebir, Farid Fadhillah, Fekri Abdulraqeb Ahmed Ali, Hichem Tahraoui, Abdeltif Amrane, Aymen Amine Assadi
Abstract
The growing environmental pollution caused by organic contaminants and the limitations of conventional treatment methods have made the development of sustainable degradation strategies increasingly urgent. This study proposes a green synthesis as an eco-friendly alternative to the traditional solvent-based preparation of photocatalysts. Ag/Ag 3 O 4 /CuO nanocomposites were successfully synthesized using Inula viscosa leaf extract, a natural reducing and stabilizing agent. Physicochemical analyses confirmed the formation of crystalline heterostructures (50.0–57.6 nm) with band gap energies ranging from 1.76 to 2.81 eV. The AAC-4 sample, containing 30% silver, achieved 99.3% degradation of Safranin O dye within 100 min under visible light and in the presence of 10 mM potassium persulfate (PDS). The synthesized catalyst maintained good efficiency over five reuse cycles. Machine learning (DT_LSBOOST) accurately predicted the degradation results ( R = 0.9981, RMSE = 0.1116), and the dragonfly algorithm identified optimal conditions with only a 0.34% deviation from experimental data. These results highlight the synergistic effect of green nanomaterials and artificial intelligence for cost-effective and eco-friendly wastewater treatment.