One-stage partial nitritation and anammox in membrane bioreactor.

Partial nitritation and anammox (PN/A) was applied in a lab-scale membrane bioreactor (MBR) to investigate its technical feasibility for treating ammonium-rich wastewater with low C/N ratio. The bacterial community was analyzed by molecular cloning and 16S rRNA sequence analysis. Partial nitritation (PN) was first realized in MBR by seeding aerobic activated sludge. With dissolved oxygen control, a steady effluent mixture with NO2 (-)-N/NH4 (+)-N ratio of 1.13 ± 0.08 was generated from the PN process. Subsequently, the MBR was seeded with anammox biomass on day 59. After running 300 days, the one-stage PN/A achieved a maximum nitrogen removal rate of 1.45 kg N/m(3)/day at the nitrogen removal efficiency of 89.5 %. Microbial community analysis revealed that Nitrosomonas sp. HKU and Nitrosospira sp. YKU corresponded to nitritation; meanwhile, Candidatus Brocadia TKU sp. accounted for nitrogen removal of the PN/A system. Specifically, Nitrosomonas sp. were enriched in the reactor at the PN/A phase and then conquered Nitrosospira sp. to be the predominant ammonia oxidizers. Nitrite oxidizers and denitrifiers were detected in symbiosis with aforementioned microbes. Denitrification promised potential plus nitrogen depletion. The present one-stage PN/A process allows a significant decrease in operational costs compared with classical nitrification/denitrification.

Kinetic study on nitrogen removal performance in marine anammox bacterial culture.

Kinetics of anaerobic ammonium oxidation (anammox) reaction in marine anammox bacterial culture was first investigated. The nitrogen removal performance of the reactor was compared with prediction of Monod model, modified Stover-Kincannon model, first-order and the Grau second-order substrate removal models. Based on calculations, Monod model, modified Stover-Kincannon model and the Grau second-order model proved to be more appropriate to describe the nitrogen removal kinetics of the reactor than first-order model with high determination coefficients of 0.993, 0.993 and 0.991, respectively. According to the modified Stover-Kincannon model, the maximal substrate removal rate (rm) and saturation rate constant (KB) were suggested as 7.37 and 6.41 g N/L/d, respectively. In addition, in light of the Monod model, the saturation concentration (Ks) and the maximal specific substrate removal rate constant (Rm) were determined to be 0.107 g/L and 0.952 g N/g MLVSS/d, respectively. Moreover, model evaluation was carried out by assessing the linear correlation between measured and predicted values. Both kinetics study and model evaluation showed that Monod model, modified Stover-Kincannon model and the Grau second-order substrate removal models could be used to describe the kinetic behavior or design of the marine anammox reactor.