Application of experimental design methodology to optimize antibiotics removal by walnut shell based activated carbon.

Three-level Box-Behnken experimental design with three factors (pH, temperature and antibiotic initial concentration) combined with response surface methodology (RSM) was applied to study the removal of Metronidazole and Sulfamethoxazole by walnut shell based activated carbon. This methodology enabled to identify the effects of the different factors studied and their interactions in the response of each antibiotic. The relationship between the independent variable (sorption capacity) and the dependent variables (pH, temperature and antibiotic concentration) was adequately modelled by second-order polynomial equation. The pH factor exerted a significant but distinct influence on the removal efficiency of both antibiotics. The removal of Metronidazole is favoured by increasing pH values, with the maximum value obtained for pH 8 - upper limit of the study domain; while Sulfamethoxazole displays a maximum value around 5.5, with a decrease in the extent of adsorption as the pH increases. The best conditions, predicted by the model, for the removal of the antibiotic Sulfamethoxazole (106.9 mg/g) are obtained at a temperature of 30 °C, initial concentration of 40 mg/L and a pH value of 5.5. For the antibiotic Metronidazole, the highest removal value (127 mg/g) is expected to occur at the maximum levels attributed to each of the factors (pH = 8, Cin = 40 mg/L, T = 30 °C). The results of isotherm experiments (at 20 °C and pH 6) displayed a good agreement with the models predictions. The maximum sorption capacity, estimated by the Langmuir model, was 107.4 mg/g for Metronidazole and 93.5 mg/g for Sulfamethoxazole.