The photocatalytic degradation of antibiotics requires a good separation efficiency of photogenerated electron-hole pairs and a wide visible light absorption range. Current studies have discussed the successful preparation of ferroferric oxide/graphite carbon nitride/reduced graphene oxide (Fe3O4/g-C3N4/rGO). The phase structure and morphology of the Fe3O4/g-C3N4/rGO composites were characterized by XRD, HR-TEM, SEM, and EDS. The obtained composites were used to degrade tetracycline hydrochloride (TCH) to evaluate its photocatalytic activity. The effects of four variables on the degradation of TCH were analyzed by the response surface method and artificial intelligence (gradient regression tree, random forest, artificial neural network, etc.). The results showed that the graphite carbon nitride in the catalyst maintained its original structure and that the photocatalytic activity was significantly improved. The degradation rate of TCH was 86.7% under the optimal conditions (the Fe3O4/g-C3N4/rGO dosage was 0.1 g, pH = 7.0, the initial concentration of TCH was 20 mg/L, and the visible light irradiation time was 60 min). At the same time, the degradation rate of TCH changed little after the material was used five times, which indicates that the stability and recyclability of the Fe3O4/g-C3N4/rGO photocatalyst were excellent. Finally, a possible photocatalytic mechanism of the Fe3O4/g-C3N4/rGO photocatalyst is proposed in this paper.
Keywords: Artificial intelligence; Graphitic carbon nitride; Metal doping; Tetracycline hydrochloride degradation; Visible light photocatalysis.
© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.