Artificial neural network modeling of photoeletrocatalytic removal of a azo dye using mwcnts-TiO2 composite on titanium

In this work, performance of titanium electrode coated with multiwall carbon nanotubes – TiO2 composite (MWCNTs-TiO2/Ti) for the treatment of C.I. Acid Red 33 (AR33) in aqueous solutions was investigated. The MWCNTs-TiO2/Ti electrode was prepared by the electrophoretic deposition (EPD) and was characterized by field emission scanning electron microscopy (FE-SEM) and Transmission electron microscopy (TEM) . The photoelectrocatalysis (PEC) performance of the prepared MWCNTs-TiO2 composite electrode was studied in removal of Acid Red 33 (AR33)from water. The main influence factors on the PEC activity such as pH of solution and current density were studied. The result shows that, in optimum conditions, maximum color removal efficiency (98%) was obtained and the removal of chemical oxygen demand (COD) was reduced to 41.66%. Also, a three-layered feed forward back propagation artificial neural network model was developed to predict the PEC of AR33 using MWCNTs-TiO2/Ti electrode. Obtained correlation coefficient (R2= 0.98) and the mean square error (MSE) MSE of 0.186×103 show the good performance of ANN model in prediction of experimental data within adopted ranges.