Magnetic graphene oxide supported plasmonic nanoparticles as visible-light driven photocatalysts: Experimental study and artificial intelligent modelling for tetracycline degradation

Abstract

Designing a high-efficiency solar-activated photocatalyst is a distinctive approach to employing green, abundant solar energy to eliminate anthropogenic environmental contaminants, including pharmaceutical effluents. Herein, the novel plasmo-magnetic quaternary nanocomposite of Ag/CuO/α-Fe2O3/rGO, as an original photocatalyst, indicated enhanced photocatalytic efficiency in the visible range with improved synergistic effects of the individual constituent components. It was fabricated by photo-deposition of plasmonic Ag nanoparticles on the hydrothermally grown CuO/α-Fe2O3 p-n heterostructures on the monolayered graphene oxide (GO) surface. The GO and Ag precursor amounts were changed to optimize magnetic and photocatalytic properties. It was fabricated by photo-deposition of plasmonic Ag nanoparticles on the hydrothermally grown CuO/α-Fe2O3 p-n heterostructures on the monolayered graphene oxide (GO) surface. The GO and Ag precursor amounts were changed to optimize magnetic and photocatalytic properties. The fabricated samples were characterized by various techniques such as PXRD, FESEM/EDX/elemental mapping, Raman, DRS, N2 physisorption, PL, TEM, HR-TEM, XPS, VSM, and EIS. The obtained photocatalyst showed a significantly improved percentage of photocatalytic degradation (~73%) of the organic probe pollutant (tetracycline (TC), 30 mg/L) after 360 min of exposure to visible light. Furthermore, the coexistence of rGO single layer and silver nanoparticles led to extending the absorption edge to the visible region and the decrease of charge carrier recombination. The quenching experiments characterized OH•, h+, and O2•− as the active species in the photocatalytic degradation process. The developed ANN and ANFIS techniques successfully anticipated the degradation percentage of TC and illustrated close agreement with the experimental data (R2 =0.96). Moreover, sensitivity analysis determined the relative importance of input variables in the architecture of both artificial intelligence tools. These characteristics confirm the high feasibility of the synthesized photocatalyst for future environmental remediations such as hydrogen production, air purification, antiviral/antibacterial applications, and water/wastewater treatment.