Reduzierung der CO2-Emissionen bei der kommunalen Abwasserreinigung - Vergleich verschiedener Verfahren zur Deammonifikation

Abstract

Conventional wastewater treatment plants (WWTPs) are energy intensive and are responsible for the emissions of more than 27 Mtonnes of CO2 year-1 in Europe alone. About 25 % of the total energy consumption in wastewater treatment is spent on removal of inorganic nitrogen compounds. Partial nitritation and anammox (PN/A) for nitrogen removal in mainstream municipal wastewater (MWW) instead of the conventional nitrification/denitrification processes has the potential for significant energy savings and lowering of the CO2-footprint of WWTPs. While PN/A is a proven technology for MWW side-stream treatment, the technology has not been effectively applied to the much larger market of MWW mainstream treatment due to the high COD/N ratio of MWW, lower temperatures, and low and fluctuating ammonium concentrations. The objective of this 3-year research project was the investigation of two different process strategies to overcome the limitations of the PN/A process for application in mainstream MWW treatment. On the Israeli side, the chosen process strategy employed ion exchange to selectively remove and concentrate ammonium from mainstream wastewater. After passing mainstream wastewater through an ion exchange column containing zeolite to remove ammonium (absorption cycle), the ion exchange column was regenerated using a brine regenerant solution. The ammonium rich brine was “bioregenerated” in a hypersaline (4 % salinity; 40g/kg) PN/A reactor where ammonium was converted to nitrogen gas by the activity of halophilic ammonia oxidizing bacteria (AOB) and anammox bacteria. This allowed for the brine regenerant solution to be continuously recycled for ion exchange regeneration. The PN/A reactor first treated recirculating brine from the ion exchange column for 48 cycles of ammonium absorption and bioregeneration with minimal brine blowdown. The concentration of the various cations in the regenerant solution were stable except for calcium that reached very high values upwards of 3,000 mg/L as Ca2+ and finally caused PN/A reactor failure due to mineral precipitation. The build-up of high concentrations of calcium in the brine regenerant was addressed in two ways: 1) 20 % regenerant replacement per cycle, and 2) precipitation of CaCO3 via the addition of sodium carbonate. Both methods were applied to 30 absorption and bioregeneration cycles each and shown to be effective in keeping calcium concentrations from accumulating in the regenerant allowing for stable PN/A reactor operation. N2O emissions from the ion exchange – bioregeneration process were investigated in PN/A reactors under hypersaline conditions with and without organic addition. Results show that N2O emissions in fixed bed and suspended hypersaline PN/A reactors without organic addition ranged from 0.08 % to 1.0 % and 0.09 % to 1.5 % of ammonium removed, respectively. Bulk ammonium and nitrite concentration were shown to be the main controlling factors for N2O emissions. The addition of acetate to both fixed bed and suspended hypersaline PN/A reactors reduced N2O emissions by 52.5 % and 72.4 %, respectively while deammonification was not significantly affected. A preliminary techno-economic analysis of the ion exchange – bioregeneration process for deammonification of mainstream wastewater showed that is competitive with other biological nutrient removal BNR systems available on the market. On the German side, PN/A was also successfully implemented and operated stably in a membrane aerated biofilm reactor (MABR) configuration. The experimental work focuses mainly on two aspects, 1) optimizing the aeration regime and 2) investigating the impact of the C/N ratio on PN/A performance. For both aspects, the associated N2O emissions were also closely monitored. Additionally, a mathematical model was used to gain further insights into the dynamics of the MABR PN/A under varying oxygen partial pressures and C/N ratios. Reactor 2 operation over more than 1000 days revealed that a key factor for successful nitrogen removal was the oxygen partial pressure, with a pO2 of 5-10 % showing the best results. Controlling the pO2 effectively suppressed of NOB activity and enhanced anammox activity, resulting in high and stable total TN removal of up to 80 %. Limiting oxygen supply through Intermittent aeration was not able to deliver these results at a pO2 of 21 %. N₂O emissions were also effectively minimized at 5 % pO₂ by aligning oxygen availability with microbial demands and avoiding pathways that favor N₂O production. Experimental as well as modelling data underlined, that maintaining optimal C/N ratios was also key for process stability. Very low C/N ratios (0-0.4) supported efficient PN/A with minimal N₂O emissions, while C/N ratios of 0.5-1.7 required fine-tuned aeration to mitigate incomplete denitrification and associated emissions. Additionally, the model also predicted that a biofilm thickness of approximately 1000-1500 µm balances microbial activity zones most effectively, allowing stable nitrogen removal while minimizing N₂O accumulation within the biofilm. Both PN/A technologies were also evaluated using a full scale WWTP scenario for economic and CO2-footpring comparison. The application of the PN/A process led to significant energy and GHG emission savings. The MABR showed a slightly better results due to the more efficient oxygen transfer efficiency. The IX-PN/A system was competitive for upgrading smaller scale WWTPs. Overall, both technologies underline the high potential of PN/A for nitrogen removal. Datei-Upload durch TIB