Sajad Khorshidi - Department of Chemical Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran
Akbar Mohammadidoust - Department of Chemical Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran

In this study a network of high frequency ultrasonic’s transducers without additives was introduced for removing the Reactive Red ۱۲۰ dye from aqueous solution. pH, irradiation time, initial concentration and number of piezoelectric were input variables at constant temperature of ۲۵ °C. The results revealed that the ultrasonic waves played an important role in cracking the hydrocarbon bonds due to the cavitation phenomenon and OH° attacks. The effects of the variables and their interactions were investigated by the central composite design (CCD) method as one of the response surface methodologies (RSM). Maximum dye removal’s efficiency (۷۶.۰۵%) was attained at initial concentration of ۵ mg/l, irradiation time of ۵۰ min; pH ۱۰ and ۵ ultrasonic’s transducers. It was in a good agreement with the experimental, ۷۸%. Finally, to more evaluates, the RSM model was compared to the artificial neural network (ANN) model. Performance’s functions reported that the RSM was better than the ANN in predicting the dye removal’s efficiency (R%). 

Reactive Red ۱۲۰, Ultrasonic transducer’ s network, Optimization, Response surface methodology, Artificial neural network