UV/peroxydisulfate oxidation of C. I. Basic Blue 3: Modeling of key factors by artificial neural network

The present work deals with the photooxidative decolorization of C.I. Basic Blue 3 (BB3), in the presence of potassium peroxydisulfate (K₂S₂O₈) irradiated by a 30W UV-C lamp in a batch reactor. Results showed that photooxidative decolorization rate was affected by the operational parameters such as the reaction time, UV light intensity and initial concentrations of peroxydisulfate and BB3. Photooxidative decolorization efficiency was enhanced by the addition of proper amount of peroxydisulfate. The increase of UV light intensity increased the photooxidative decolorization rate. This increase is due to the enhanced production of sulfate and hydroxyl radicals. The decrease in photooxidative decolorization rate with increasing initial BB3 concentration has been observed. It could be stated that the complete removal of color, could be achieved in a relatively short time, about 20min. The figure-of-merit electrical energy per order (E_EO) was employed to estimate the electrical energy consumption and related treatment costs. An artificial neural network model was developed to predict the photooxidative decolorization efficiency. The findings indicated that ANN provided reasonable predictive performance (R²=0.9877) while the influence of each parameter on the variable studied was assessed, reaction time being the most significant factor, followed by initial concentration of the dye.