Preparation and photocatalytic study of CoFe 2 O 4 /TiO 2 /Au nanocomposites and their applications in organic pollutant degradation and modeling by an artificial neural network (ANN)

Due to the increase of environmental pollution by various industries in recent decades, preparing drinking water has become one of the most vital issues for many countries. The organic pollutant, such as different azo dyes, is one of the most important issues. Using photocatalyst materials is considered to be an optimal solution to prevent environmental pollution. In this work, novel ternary catalysts of CoFe 2 O 4 /TiO 2 /Au were synthesized for the photocatalytic reduction of methyl orange (MO) under UV light illumination. The localized surface plasmon resonance (LSPR) property of Au nanoparticles is widely exploited for their photocatalytic activities. In this research, both CoFe 2 O 4 and TiO 2 nanoparticles (NPs) were prepared by sol–gel method. Hydrothermal treatment was also used to synthesize the nanocomposite. Au nanoparticles were successfully loaded on the CoFe 2 O 4 /TiO 2 surface to get CoFe 2 O 4 /TiO 2 /Au magnetic nanocomposites. To characterize the shape of the structure, morphology, purity, and particle size of the nanocomposite, scanning and transmission electron microscopy (SEM and TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), Zeta potential analysis, dynamic light scattering (DLS), photoluminescence spectroscopy (PL), Brunauer–Emmett–Teller (BET), and Fourier transform infrared (FT-IR) spectroscopy were employed. Alternating gradient field magnetometer (AGFM) studies show the superparamagnetic properties of the CoFe 2 O 4 nanostructures. Finally, we investigated the catalytic performance and recyclability in reducing MO of synthesized nanocomposites by monitoring a UV–visible spectrophotometer. The composite catalysts can then be easily separated from the reaction solution using a magnet bar and ultimately reused. We used artificial neural network (ANN) to remove expensive experimental research and tried solving and predicting the novel phenomena with huge factors. Initially, information about the degradation of MO was gathered by experimental analyses. We then tried shaping and calculating the special algorithm that could find the best relation and high percentage of accuracy between input variables. The genetic algorithm as one of the most popular algorithms in an artificial world was selected to predict and train the model. In conclusion, the experimental results determined that the CoFe 2 O 4 /TiO 2 /Au magnetic nanocomposites were successfully synthesized and it exhibited a useful effect on the removal of azo dyes from the contaminated solutions. According to the prediction of removal efficiency of pollution by artificial neural network, the results show that using this algorithm has a high percentage of accuracy to investigate the experimental results of the current research.