Abstract
Photocatalytic treatment of Reactive Black 5 dye wastewater was carried out using suspension of commercially available TiO2 catalyst under ultraviolet irradiation in a shallow pond reactor. An artificial neural network (ANN) model was developed to predict the behavior of the process. Six operational parameters (TiO2 dose, initial dye concentration, pH of the dye solution, area to volume ratio, UV light intensity, and time) were employed as input and decolorization and degradation efficiencies were employed as output of the network. The outcomes have been validated experimentally indicating that the ANN provided reasonable predictive performance. The parameteric optimization was done, using multiresponse optimization with desirability function approach, to simultaneously maximize the decolorization and degradation efficiency. Optimization by Box–Behnken design effectively copes with interaction between optimizing variables and its prediction agreed well with the results of ANN model and experimental run. The decolorization and degradation follows the pseudo-first-order kinetics.