http://hdl.handle.net/10012/20645

Lakshminarasimman Meanakshi Sekar, Narasimman (2024) Investigating the impacts of upgrading a full-scale conventional activated sludge process with a hybrid membrane aerated biofilm reactor, UWSpace - Theses, http://hdl.handle.net/10012/20645

Membrane aerated biofilm reactors (MABR) are an emerging wastewater treatment technology that offer several process advantages such as higher aeration efficiency, simultaneous nitrification and denitrification, and reduced nitrous oxide (N₂O) emission. However, the current knowledge on MABR operations is largely based on bench- and pilot-scale systems that differ in mixing conditions, biofilm thickness control strategies, and cassette arrangement from those employed in full-scale. This study bridged this critical knowledge gap by investigating the long-term performance (20 months) of one of the largest MABR installations in North America and reported the findings on a wide range of responses related to effluent quality, electricity consumption, N₂O emissions, and biofilm characteristics. The key performance metrics related to nitrogen removal, aeration tank operations, and electricity consumption were monitored in a full-scale conventional activated sludge (CAS) before and after upgrading with a hybrid MABR. The inclusion of the hybrid MABR process improved the seasonal nitrification observed in the CAS process before the upgrade to year-round ammonia removal. Denitrification in the hybrid MABR doubled the TN removal in the plant from 30-40% before upgrade to 70-80% afterwards. Operation at reduced MABR airflow (4.5 NL m⁻² hr⁻¹) resulted in lower nitrification rates due to insufficient biofilm thickness control that led to diffusional limitations. Temperature was found to impact nitrification in the MABR with a 22% decrease in nitrification rate from 1.8 ± 0.2 g-N m⁻² d⁻¹ during warm weather to 1.4 ± 0.2 g-N m⁻² d⁻¹ during cold weather conditions. Operation at higher MABR airflow (6 NL m⁻² hr⁻¹) increased the NH₄-N removal efficiency in the aeration tank during cold weather conditions suggesting increased nitrifier seeding due to enhanced sloughing of biofilm into the suspended sludge. The hybrid MABR process achieved high denitrification efficiency (80-97%) throughout the study and was not substantially impacted by ammonia loading, process airflow, or wastewater temperature. Aeration related electricity consumption, as described by volumetric energy intensity (kWh m⁻³), decreased by 50% after the upgrade due to the efficiency of oxygen supply by the MABR and the reduction of aeration requirements downstream aeration basin. Overall, the improved N removal under reduced electricity consumption at full scale demonstrated the potential of MABR as a suitable process intensification technology. Pollutant removal in biofilm processes such as membrane aerated biofilm reactors (MABR) is directly influenced by the biofilm thickness and microbial community functions. However, these biofilm characteristics have not been studied in full-scale MABR systems before. This study addressed this knowledge gap by characterizing the spatial and operational dynamics of key biofilm properties such as thickness, and microbial community structure and functionalities in a full-scale MABR facility. The arrangement of the MABR cassettes in an array along the length of the plug flow MABR train resulted in a longitudinal biofilm thickness gradient. The biofilms on the front cassettes were more than twice as thick as those on cassettes at the back. Examination of biofilm thickness as a function of MABR process airflow indicated that a lower airflow (4.5 NL m⁻² hr⁻¹) resulted in a thicker biofilm (> 1000 µm) throughout the tank. Consistent with the trends in thickness, analysis of Bray-Curtis dissimilarity index showed that there were differences in the biofilm microbial community composition along the length of the MABR tank and between operating phases. Thicker biofilms in the front cassettes of the full-scale tank were predicted to have a higher relative abundance of organisms with anaerobic functions such as fermentation and sulfur reduction and lower relative abundance of organisms with aerobic functions such as aerobic heterotrophy and nitrification. Nitrosomonas was identified to be the main ammonia oxidizer and Nitrospira and Nitrotoga were the main nitrite oxidizers in the biofilm samples. The 16s RNA gene profiles were strongly correlated with biofilm thickness (R² = 0.8) and MABR nitrification rate (R² = 0.4). Phases with thinner biofilm showed higher relative abundance of nitrifiers which corresponded to higher nitrification rates. Thus, optimizing the process airflow to provide adequate biofilm thickness control is key to maximizing nitrification rate in full-scale MABR. Biological nitrogen removal often results in emission of nitrous oxide (N₂O) which is a highly potent greenhouse gas. Current published models for N₂O emissions in MABR have several simplifications that are not representative of full-scale systems. This study developed an improved MABR N₂O model that captured commonly overlooked phenomena such as back diffusion of generated N₂O into MABR lumen gas and the recirculation of the N₂O laden lumen gas for tank mixing and biofilm thickness control. The improved model was validated with measured N₂O concentrations in the lumen gas phase and bulk mixed liquor in a full-scale hybrid MABR facility. The validated model was used to obtain insights into N₂O bioconversion pathways. Model predictions revealed that all N₂O generated in the inner layers of the biofilm, which back diffused into the lumen gas, was via ammonium oxidizing organism activity. The N₂O transported to the outer biofilm layers was reduced via the heterotrophic denitrification pathway. The N₂O gas model predicted that up to 70% of the N₂O carried by the recirculated lumen gas was scrubbed into the mixed liquor which was further denitrified. An N₂O emission factor of 0.18 ± 0.01% N₂O-N/N load was estimated for the full-scale MABR process which removed up to 50% of the influent N load, highlighting the potential of this technology to mitigate N₂O emissions when compared to conventional activated sludge. Release of organic micropollutants (OMP) such as pharmaceuticals and personal care product ingredients in the treated wastewater effluents is concerning as these compounds could have harmful effects in the aquatic ecosystem. This study monitored the removal of 16 OMPs over multiple seasons in a full-scale CAS before and after an upgrade with a hybrid MABR process. A comparison of OMP concentrations in the plant effluent showed that 12 out of 16 target compounds were present at lower concentrations after the upgrade. An examination of plantwide removal efficiencies revealed that highly removable compounds (> 75%) such as acetaminophen, ibuprofen, naproxen, triclosan, triclocarban, and norfluoxetine and the recalcitrant carbamazepine were not impacted by the addition of the MABR process. However, six compounds namely gemfibrozil, sulfamethoxazole, trimethoprim, atorvastatin and its ortho- and para- hydroxy metabolites, that were poorly removed (< 25%) by the CAS configuration had moderate removals (25 to 75%) with the hybrid MABR/CAS configuration. After the MABR upgrade it was found that six OMPs showed higher removal under warm weather conditions (19.3 ± 1.6℃) when compared to cold weather conditions (13 ± 1.2℃). Mass balance analyses on the MABR tank revealed a broad range of compound specific responses such as complete biotransformation (acetaminophen), partial removal (naproxen), and compound formation from unmeasured precursors (sulfamethoxazole, carbamazepine). Overall, long-term monitoring of the full-scale facility before and after the upgrade revealed that upgrading of CAS to a hybrid MABR configuration can enhance the removal of some OMPs that are poorly removed by the CAS process alone.

https://uwspace.uwaterloo.ca/bitstreams/14491a28-b1e8-460e-9fd6-4a2de86cdde9/download