Model-based analysis of sulfur-based denitrification in a moving bed biofilm reactor.

In this study, a biofilm model was developed for sulfur-based denitrification in a moving bed biofilm reactor (MBBR), including mass transport as well as the conversion kinetics of sulfur-oxidizing bacteria (SOB). The experimental reactor simulated received a synthetic wastewater containing nitrate, sulfide and thiosulfate. The substrate affinity of SOB for intermediary elemental sulfur (S0) was found the most sensitive parameter. After estimating this single parameter, the model could adequately describe the steady state performance of the experimental MBBR. The experimental and simulated mass balances indicated that a fraction of influent sulfur accumulated into intermediate S0. Furthermore, the simulations showed that SOB were active over the entire thickness of a 200 µm biofilm. The simulation results allowed to quantify the extent of diffusion and substrate limitation. Scenario analyses indicated that the specific nitrogen loading rate could be increased from 0.05 to 0.20 kg N.kg-1 VSS.day-1 (corresponding to 0.22-0.86 kg N.m-2.day-1 expressed per biofilm surface area) while maintaining nitrogen removal efficiencies above 70%. An increasing specific nitrogen loading rate in this range resulted in an almost linearly increasing specific nitrogen removal rate, independent from whether it was realized through a decreasing HRT, carrier filling ratio or biofilm thickness.