A Box-Behnken design (BBD) statistical experimental design was used to investigate the degradation of o-toluidine by the electro-Fenton process. This method can be used to determine the optimal conditions in multivariable systems. Fe(2+) concentration (0.2-1.0 mM), H(2)O(2) concentration (1-5 mM), pH (2-4), and current (1-4A) were selected as independent variables. The removal efficiencies for o-toluidine and chemical oxygen demand (COD) were represented by the response function. Result by 2-level factorial design show that the pH and the Fe(2+) and H(2)O(2) concentrations were the principal parameters. Among the main parameters, the removal efficiencies for o-toluidine and COD were significantly affected by pH and Fe(2+) concentration. From the Box-Behnken design predictions, the optimal conditions in the electro-Fenton process for removing 90.8% of o-toluidine and 40.9% of COD were found to be 1 mM of Fe(2+) and 4.85 mM of H(2)O(2) at pH 2. Under these optimal conditions, the experimental data showed that the removal efficiencies for o-toluidine and COD in the electro-Fenton process and the photoelectro-Fenton process were more than 91% and 43%, respectively, after 60 min of reaction. The removal efficiencies for o-toluidine and COD in the Fenton process are 56% and 27%, respectively. (C) 2011 Elsevier B.V. All rights reserved.