Efficient photodegradation of disulfine blue dye and Tetracycline over Robust and Green g-CN/Ag3VO4/PAN nanofibers: Experimental design, RSM, RBF-NN and ANFIS modeling

Abstract

The g-CN/Ag3VO4/PAN nanofibers (NFs) is prominent from various nanophotocatalysts (NPCs) due to their higher surface to volume ratio, mechanical strength and recyclable characteristics. In this study, g-CN/Ag3VO4/PAN NFs were obtained through in situ method by immobilizing Ag3VO4 on g-CN/PAN NFs. Compared to g-CN/PAN and Ag3VO4/PAN, the g-CN/Ag3VO4/PAN NFs revealed excellent photocatalytic performance toward disulfine blue dye (DB). Also, the g-CN/Ag3VO4/PAN NFs showed highly activity over some interference of inorganic cations/anions and excellent cycling stability. Furthermore, the batch experiments designed by central composite design (CCD) were employed for remove of tetracycline (TC). Afterward, response surface methodology (RSM), Radial basis function artificial neural network (RBF ANN) and an adaptive neuro-fuzzy inference system (ANFIS) have been used for modeling of photocatalyst dose, concentration of TC and irradiation time. Due to strong influence of pH on the photocatalytic degradation, the pH was optimized as one at a time variable. For all three models, the values of the statistical parameters were calculated. The obtained results reveal that the ANFIS models is more accurate to predict the degradation of TC. Furthermore, the effective factors were optimized by employing the Desirability function (DF) and genetic algorithm (GA) approach and then used for real water samples. At this optimum condition, the removal percentage was obtained ∼97% by DF and GA approach, respectively. At the end, the LC-Mass method was employed to detect the intermediates of photodegradation of TC molecules.