The importance of nitrate reduction in return sludge for phosphorus release has been emphasized. MUCT process could reduce nitrate concentration in return sludge by recycling sludge to pre-anoxic zone and it could enhance phosphorus release. Nitrate inhibition on phosphorus release in anaerobic zone could be increased in low organic condition. Endogenous nitrate respiration(ENR) was suggested to enhance the phosphorus removal efficiency from low strength wastewater by reducing nitrate in the recycled sludge. The objectives of the study were to elucidate the mechanism and the reaction rate of ENR and to investigate the effect on phosphorus removal. This study consists of four major parts as follows: ⅰ) conventional BNR processes such as $A^2/O$ and MUCT and ENR process were tested and compared for the nutrient removal efficiency, ⅱ) the mechanism and rate of ENR, SPRR(Specific phosphorus release rate), SNR(Specific nitrification rate), and SDNR(Specific denitrification rate) were evaluated, ⅲ) full scale plant was operated to investigate the performance of the suggested system, ⅳ) BNR model was run to ascertain real data from full scale plant operation.
In the comparative study on BNR processes, the released phosphate concentration in anaerobic zone of ENR process were 1.64 and 1.3 times higher than $A^2/O$ and MUCT, while the effluent phosphorus concentration from oxic zone was 10% and 25% lower than the $A^2/O$ process and the MUCT process, respectively.
In lab scale experiments of ENR process, , the soluble phosphate increased from 5.2 to 9.8 mg $PO_4-P/L$ while nitrate nitrogen decreased from 3.2 to 1.3 mg $NO_3-N/L$. When the nitrate in the anaerobic zone increased higher than 3 mg/L, the phosphorus release decreased significantly. Therefore, about 3 mg/L of nitrate was a minimum level allowable for the active phosphorus release in the anaerobic zone. Phosphate concentration in anaerobic zone was three or four times higher than in influent when SCOD concentration reduced 40 % of influent. In the batch experiments of ENR zone with MLVSS of about 5,300 mg/L at 21°C, the endogenous nitrate respiration rate(kENR) was estimated to be 1.75 mg $NO_3-N/g$ MLVSS-h. This kENR indicated that 4 mg/L of nitrate nitrogen in the return sludge of 5,000 mg/L MLVSS could be removed by ENR reaction in 30 minutes of HRT. In suggested system, the SPRRs (Specific Phosphorus Release Rate) of anaerobic zone was 2.6 mg $PO_4-P/g$ MLVSS-h. In oxic zone, estimated SNR was 3.05 mg $NH_4-N/g$ Mv.hr and denitrification rate ranged 2.5 to 3.6 mg $NO_3-N/g$ Mv.hr. In anoxic zone nitrate nitrogen decreased rapidly to detention time of 2 hour and went down gradually after that time.
In full scale plant study, retrofitting full scale plant consisted of ENR zone of 0.5hr, anaerobic zone of 1.2hr, anoxic zone of 1.8hr, and oxic zone of 3.5hr within aerobic detention time of existing sewage treatment plant. In BPR operation mode effluent TP ranged 2 to 3 mg $PO_4-P/L$ and decreased to below 1 mg PO4-P/L. In early stage of BPR mode phosphorus content in wasting sludge was 2 to 3 % and increased up to 5 to 6 % at steady state. In BPR mode with nitrification effluent nitrate nitrogen increased significantly and decreased phosphorus release in anaerobic zone. In BNR mode with nitrified recycle removal efficiencies of TN and TP were 65% and 80%, respectively. The influent TCOD/TP ratio ranged 40 to 60 while effluent TP concentrations was 0.8 to 1.5 mg/L. In TCOD/TKN more than 7 effluent nitrate nitrogen was below 10 mg $NO_3-N/L$. During the operation periods, nitrate nitrogen more than 3 mg $NO_3-N /L$ inhibited phosphorus release in anaerobic zone
In application study to BNR model nitrogen removal efficiencies of high readily biodegradable substrate$(S_s)$, high slowly biodegradable substrate$(X_s)$, and low $S_s$ were 80%, 60%, and 40%, respectively. In case of low Ss removed nitrogen decreased 50%. It could explain that concentration of Ss have an large effect on nitrogen removal. The high Ss needed alkalinity of 35 mg $CaCO_3/L$ while high Xs and low $S_s$ were 40 and 60 mg $CaCO_3/L$, respectively. This phenomenon was due to difference of denitrification rate. Real operation data and output by model run were similar nitrification rate more than 90%. However, denitrification rate showed the great difference. This was due to the difference of stoichiometric and kinetic parameters related to hydrolysis and available organics of heterotrophic bacteria between operated plant and model. In case of primary effluent to anoxic zone Ss and Xs concentration was higher than anaerobic effluent and it could increase denitrification rate. The primary effluent to anoxic zone increased anaerobic zone of readily biodegradable COD of 45 % and particulate inert biodegradable COD, Xs of 12 %. The denitrification rate of nitrate nitrogen was 20 % higher than anaerobic effluent.
