2,4-Dichlorophenoxyacetic acid (2,4-D), a chlorinated phenoxy-alkanoic herbicide, is used extensively in agriculture. This work investigates TiO 2 /H 2 O 2 mediated UV photocatalytic degradation of 2,4-D in a laboratory-scale photoreactor. Three levels of Box-Behnken design technique, combined with response surface methodology (RSM), were used to design the experiments. Two kinds of multivariate experimental design (pH, TiO 2 , and 2,4-D concentration) and (pH, TiO 2 , and H 2 O 2

more »

... ) were employed to establish two quadratic models (Model 1 and Model 2), showing the functional relationship between degradation rate of 2,4-D and three independent experimental parameters. Model 1 predicted optimum values for pH, TiO 2 , and 2,4-D concentrations to be 5.7, 1.20 g L , and 32 mg L , respectively. Model 2 predicted optimum values for pH, TiO 2 , and initial H 2 O 2 concentrations to be 4.94, 1.34 g L , and 161 mg L . Degradation rate of 2,4-D approached 78.10% for Model 1 and 83.63% for Model 2. For both models, similar results were obtained through optimizing variables by RSM and using single factorial batch reactor operation. Regression analysis showed good agreement between experimental results and predictive values for Models 1 and 2, with R 2 values of 0.9958 and 0.9976, respectively.