Modeling and optimization study on sulfamethoxazole degradation by electrochemically activated persulfate process

In this work, empirical kinetics, response surface method (RSM) and artificial neural network (ANN) were employed to model and optimize electrochemically activated persulfate process. Electrochemically activated persulfate using a sacrificial iron anode was used as a cost effective method to degrade sulfamethoxazole (SMX) in aqueous solution. The individual and the interactive effects of operation parameters such as pH, applied current, persulfate concentration and electrolysis time were investigated. Eight unseen experiments beyond the experimental design were performed to test the accuracy of the generalization ability of the models. Based on the unseen experiments, ANN demonstrated the superiority in predicting the results when compared to empirical kinetics and RSM modeling. Response surface analysis was used to illustrate the interactions between parameters pH/applied current, pH/persulfate concentration and pH/electrolysis time. Sensitivity analysis indicated that the relative importance of the influencing parameters were of the following order: electrolysis time > pH > applied current > persulfate concentration. The highest degradation efficiency was achieved with the optimal conditions of pH of 3.43, the applied current of 18.4 mA, the persulfate concentration of 3.54 mM, and the electrolysis time of 60 min. The electrical energy consumption was also calculated at this optimized condition (0.04 kWh ⋅ m−1order−1). The work provides a novel predictive and optimized model for SMX removal under different conditions by electrochemically activated persulfate.