Response surface optimization of substrates for thermophilic anaerobic codigestion of sewage sludge and food waste

This study investigated the effects of food waste constituents on thermophilic (55 °C) anaerobic codigestion of sewage sludge and food waste by using statistical techniques based on biochemical methane potential tests. Various combinations of grain, vegetable, and meat as cosubstrate were tested, and then the data of methane potential (MP), methane production rate (MPR), and first-order kinetic constant of hydrolysis (k_H) were collected for further analyses. Response surface methodology by the Box-Behnken design can verify the effects and their interactions of three variables on responses efficiently. MP was mainly affected by grain, whereas MPR and k_H were affected by both vegetable and meat. Estimated polynomial regression models can properly explain the variability of experimental data with a high-adjusted R² of 0.727, 0.836, and 0.915, respectively. By applying a series of optimization techniques, it was possible to find the proper criteria of cosubstrate. The optimal cosubstrate region was suggested based on overlay contours of overall mean responses. With the desirability contour plots, it was found that optimal conditions of cosubstrate for the maximum MPR (56.6 mL of CH₄/g of chemical oxygen demand [COD]/day) were 0.71 g of COD/L of grain, 0.18 g of COD/L of vegetable, and 0.38 g of COD/L of meat by the simultaneous consideration of MP, MPR, and k_H. Within the range of each factor examined, the corresponding optimal ratio of sewage sludge to cosubstrate was 71:29 as the COD basis. Elaborate discussions could yield practical operational strategies for the enhanced thermophilic anaerobic codigestion of sewage sludge and food waste.