A sequential aerobic completely stirred tank reactor (CSTR) following the anaerobic migrating blanket reactor (AMBR) was used to treat a synthetic wastewater containing 2,4-dinitrotoluene (2,4-DNT). A Box-Wilson statistical experiment design was used to determine the effects of 2,4-DNT and the hydraulic retention times (HRTs) on 2,4-DNT and COD removal efficiencies in the AMBR reactor. The 2,4-DNT concentrations in the feed (0-280 mg/L) and the HRT (0.5-10 days) were considered as the independent variables while the 2,4-DNT and chemical oxygen demand (COD) removal efficiencies, total and methane gas productions, methane gas percentage, pH, total volatile fatty acid (TVFA) and total volatile fatty acid/bicarbonate alkalinity (TVFA/Bic.Alk.) ratio were considered as the objective functions in the Box-Wilson statistical experiment design in the AMBR. The predicted data for the parameters given above were determined from the response functions by regression analysis of the experimental data and exhibited excellent agreement with the experimental results. The optimum HRT which gave the maximum COD (97.00%) and 2,4-DNT removal (99.90%) efficiencies was between 5 and 10 days at influent 2,4-DNT concentrations 1-280 mg/L in the AMBR. The aerobic CSTR was used for removals of residual COD remaining from the AMBR, and for metabolites of 2,4-DNT. The maximum COD removal efficiency was 99% at an HRT of 1.89 days at a 2,4-DNT concentration of 239 mg/L in the aerobic CSTR. It was found that 280 mg/L 2.4-DNT transformed to 2,4-diaminotoluene (2,4-DAT) via 2-amino-4-nitrotoluene (2-A-4-NT) and 4-amino-2-nitrotoluene (4-A-2-NT) in the AMBR. The maximum 2,4-DAT removal was 82% at an HRT of 8.61 days in the aerobic CSTR. The maximum total COD and 2,4-DNT removal efficiencies were 99.00% and 99.99%, respectively, at an influent 2,4-DNT concentration of 239 mg/L and at 1.89 days of HRT in the sequential AMBR/CSTR. (C) 2011 Elsevier B.V. All rights reserved.