Synthesis of Zinc Oxide Nanoparticles on Montmorillonite for Photocatalytic Degradation of Basic Yellow 28: Effect of Parameters and Neural Network Modeling

ZnO/montmorillonite (ZnO/MMT) nanocomposite was prepared using MMT as a support, zinc chloride as ZnO synthesis precursor and cetyltrimethylammonium bromide as a surfactant. The prepared ZnO/MMT nanocomposite was characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscope, Fourier transform infrared spectroscopy and N-2 adsorption/desorption analysis. Results indicated the appropriate immobilization of the ZnO nanoparticles with 20-40 nm width on the surface of the MMT. The effect of ZnO immobilization on its adsorption and photocatalytic activity was evaluated by the removal of basic yellow 28 (BY28) in aqueous solution under UV light irradiation. The ZnO/MMT nanocomposite was more effective in adsorption, and photocatalytic degradation processes than pure ZnO. The performance of used light source for photocatalytic degradation of BY28 was found to be in the order of UV-C (wavenumber region of 200-280 nm) > UV-B (wavenumber region of 280-315 nm) > UV-A (wavenumber region of 315-400 nm). An artificial neural network (ANN) model was developed to predict the photocatalytic degradation process under UV-C radiation. The ANN model with reasonable predictive performance (R-2 = 0.999) indicated that the influence of initial concentration of BY28 on decolorization process (49.66%) was higher than that of nanocomposite dosage (36.44%) and UV radiation time (13.09%). Results of reusability tests indicated that the ZnO/MMT was stable and appropriate for long time application.