Process configurations adapted to membrane bioreactors for enhanced biological phosphorous and nitrogen removal

Enhanced biological phosphorous (Bio-P) removal process was adapted to membrane bioreactor (MBR). One bench-scale pilot plant (BSP, 200–250 L) and two medium-scale pilot plants (2//MSP, 1000–3000 L each) were operated under several configurations, including pre-denitrification and post-denitrification without addition of carbon source, and two solid retention times (SRT) of 15 and 26 d, in parallel to the full-scale Bio-P removal activated sludge plant of Berlin-Ruhleben (12–18 d SRT). The trials showed that efficient Bio-P removal can be achieved with MBR systems, in both pre- and post-denitrification configurations. Bio-P dynamics could be clearly demonstrated through batch-tests, on-line measurements, profile analyses, P-spiking trials, and mass balances. High P-removal performances were achieved even with high SRT of 26 d (around 9 mgP/L was removed with P/TS~2.6%). Under similar operation conditions of sludge age and mass organic load, the MBR system achieved slightly higher P-removal than the conventional technology. This was due to the rejection of particles and colloids through the microfiltration membrane. When spiking with phosphate, high Bio-P removal of up to 35–40 mg/L could be achieved without addition of external carbon source, and P/TS stabilized around 7.5%.

These trials could demonstrate that enhanced biological phosphorous removal is effective in MBR systems, under sludge ages characteristics of traditional wastewater treatment plants, but also under higher sludge ages usually more suitable to MBR systems. Efficient P-removal was achieved with 15 and 26 d sludge age. Two process configurations were tested with both sludge age conditions: one standard configuration including a pre-denitrification stage, and a configuration characterized by post-denitrification without addition of any carbon source. With both sludge ages, both configurations achieved similar and high P-removal performances. Under identical loading conditions, nitrogen removal was greater in the post-denitrification configuration. Average total nitrogen concentration of 3.6 mg/L was monitored in the effluent. Considering the other operational benefits of this configuration, such as better mass repartition, hence lower sludge concentration in contact with the membrane, and lower recycle rates, the post-denitrification configuration seem to be most indicated when high effluent standard is required. Further development of the project will consist in optimizing both process configurations in respect to solid and hydraulic retention times, recycle and return rates, and number of reactors. Tests will be undertaken with another type of raw water coming from a decentralized area with a short retention and separative sewer, and which is not affected by wet industries. Ultimately, optimum loading and configuration will be determined as a function of raw water characteristics and discharge criteria. This will enable to set up recommendations and specifications for full-scale MBR implementations. The first demonstration site could be located in a remote and yet unsewered area of Berlin, while building or upgrading a decentralized or semi-decentralized wastewater facility. The receiving water bodies of these areas, sensitive to phosphorous loading, could well soon benefit first from these very promising developments.