Artificial neural network and genetic algorithm for modeling and optimization of photocatalytic removal of aquatic dye by g-c3 n4 /n-tio2 nanoparticles

Abstract

Herein, we report the modification of nitrogen-doped TiO2 (N-TiO2) photocatalytic activity with graphite-like carbon nitride (g-C3 N4). Also, the photo-degradation performance of Reactive Orange 122 (RO122) was investigated under visible light irradiation. The UV-vis spectrophotometer revealed that the absorption edge of nanocomposite was shifted to the lower energy, as compared with only N-TiO2 nanoparticles. The visible light induced the photocatalytic activity of N-TiO2 /g-C3 N4 nanoparticles; they were significantly increased by combining N-TiO2 with g-C3 N4 nanosheets. The effects of some key factors including pH, dye concentration, catalyst dosage, and reaction time on the photocatalysis removal process of RO122 were also evaluated. Artificial neural network model hybridized with the genetic algorithm (ANN-GA) strategy was proposed to model and optimize the photocatalytic degradation of RO122 in a batch system. The results showed that the pho-tocatalytic process could be well predicted by ANN, using 4:20:1 topology. The genetic algorithm was also utilized to optimize the model conditions of the input parameters. The ANN-GA method was capable of effectively modeling and optimizing the photocatalytic efficiency of the prepared N-TiO2 /g-C3 N4 nanocomposite under visible light.