Assessing Nitrous Oxide Emissions from a Demonstration-Scale Membrane Bioreactor Process

Introduction The Metropolitan Water District of Southern California (Metropolitan) and the Los Angeles County Sanitation Districts (Sanitation Districts) are collaboratively exploring water recycling possibilities through the Pure Water Southern California (PWSC) project. The full-scale endeavor would involve the construction of an Advanced Water Treatment Facility (AWTF) with a capacity of 150 million gallons per day (MGD) at the Sanitation Districts' A.K. Warren Water Resource Facility (Warren Facility) in Carson, CA, previously known as the Joint Water Pollution Control Plant. The AWTF would produce treated water for indirect potable reuse, with potential for direct potable reuse in the future. The Warren Facility, with a capacity of 400 MGD, currently operates as a high purity oxygen activated sludge (HPOAS) facility and discharges its treated effluent (non-nitrified secondary effluent) into the ocean. Metropolitan has established the Grace F. Napolitano PWSC Innovation Center (Napolitano Center) to formulate the design and operating criteria for the 150 MGD AWTF, while ensuring compliance with regulations. Within the Napolitano Center, a demonstration plant with a capacity of 0.5 MGD has been developed. The Napolitano Center's process train comprises a membrane bioreactor (MBR), followed by reverse osmosis membranes, and ultraviolet light with an advanced oxidation process. Nitrogen management is a pivotal aspect of the PWSC project. Consequently, the Napolitano Center's MBR process was designed with flexibility to accommodate different nitrogen removal configurations. The full-scale implementation of biological processes aimed at transforming or removing nitrogen may result in an increase in greenhouse gas (GHG) emissions from the Warren Facility. A specific concern revolves around nitrous oxide (N2O), primarily generated during nitrification and denitrification processes in wastewater treatment plants. N2O is a potent GHG, with the Intergovernmental Panel on Climate Change (IPCC) estimating its 100-year warming potential at 296 times that of CO2 (Ehhalt et al., 2001). Estimating, measuring, and, when necessary, mitigating GHG emissions that contribute to climate change are integral components of the broader sustainability and public health mission in wastewater treatment. The Sanitation Districts have demonstrated a strong commitment to sustainability, achieving carbon-negative status in both 2021 and 2022. Greenhouse Gas Emissions from Wastewater Treatment In the context of GHG emissions, wastewater treatment facilities play a relatively minor role in the overall emissions landscape. To provide perspective, in 2022, these facilities accounted for just 0.8% of the total US GHG emissions, equivalent to 41.8 million metric tons of carbon dioxide equivalents (MMT CO2eq). More specifically, wastewater treatment contributed 2.8% of the total US methane (CH4) emissions and 5.5% of the total N2O emissions (EPA, 2022). Despite their relatively modest overall contribution, there is considerable variability among individual facilities and processes. For instance, a 2010 study conducted by Ahn et al., which investigated N2O emissions from ten wastewater treatment plants, uncovered a wide range of emissions, from 0.0001 to 0.071 lb N2O/lb N-day for a given plant. In a more recent study by the IPCC, encompassing the ten plants studied by Ahn et al., along with an additional 20 facilities, emissions were reported in the range of 0.0003 to 0.071 lb N2O/lb N-day (IPCC, 2020). For the purpose of GHG inventorying, the IPCC and the United States Environmental Protection Agency (U.S. EPA) have established default emission factors for wastewater treatment plants based on available data, with values set at 0.016 and 0.015 lb N2O/lb N-day, respectively (IPCC 2019, U.S. EPA 2021). These default values are derived from the available data and, as such, they can result in substantial overestimation or underestimation of actual GHG emissions at an individual facility. Therefore, the accurate measurement of process emissions becomes essential in ensuring precise GHG inventorying and the evaluation of new treatment processes. Project Objective and Emissions Measurement Description To assess the potential N2O emissions resulting from the implementation of a large MBR process at the Warren Facility, emissions were measured from the Napolitano Center MBR process. In May of 2023, the system operated as a secondary MBR process, with a primary effluent feed flow of approximately 0.5 MGD and an average Total Kjeldahl nitrogen (TKN) concentration of 70 mg N/L (Figure 1). In this configuration, primary effluent was introduced into an anoxic tank to supply carbon for denitrification. Subsequently, this liquid was directed to the aerobic tank, where ammonia was nitrified to nitrate. The nitrified water was then recycled back to the anoxic tank for denitrification. These processes were then followed by membrane filtration conducted in separate membrane tanks. A fraction of the solids retained by the membranes was continuously returned to the front of the aeration tank. For emissions measurement, the anoxic and aerobic tanks were temporarily enclosed using scaffolding and plastic sheeting, which was connected to metal ductwork and an exhaust fan (Figure 2, Figure 3). One of two membrane tanks was also covered with plastic sheeting and connected to metal ductwork and an exhaust fan. Each tank was maintained under negative pressure during the emissions measurement process. Full enclosure of the tanks captured the emissions from the tank surface area and, as such, the resulting emission estimate would not be subject to the spatial variabilities that tend to affect studies that utilize other measurement techniques, such as probes or floating emissions capture systems. In the first two weeks of May 2023, gas-phase N2O emissions were monitored from the fully enclosed aerobic, anoxic, and MBR tanks using a calibrated Innova photoacoustic sensor (PAS) (Innova Air Tech, Denmark) configured to record data at two-minute intervals. Standards were run daily, and the lab analyzed 24 grab samples for comparison with the PAS data. Emissions were assessed over approximately 24-hour periods for each tank, involving four days of measurements for the aerobic tank, two days for the anoxic tank, and one day for the MBR tank. In the case of the aerobic tank, both steady-state and induced process upset emissions were recorded. The determination of emissions used the concentration (in parts per million, ppm) as recorded by the photoacoustic sensor and the measured gas velocity within the ductwork. Emission factors were calculated as the fraction of influent nitrogen lost as N2O and as the mass N2O emitted per MGD. Project Results As anticipated, under steady-state conditions, the aerobic tank exhibited the highest N2O emissions factor of 0.0008 lb N2O/ lb N. The average aeration rate of atmospheric air was 300 standard cubic feet per minute (scfm). In contrast, the anoxic tank and the MBR tank recorded emissions of 0.0004 lb N2O/ lb N, resulting in an overall process emission factor of 0.0016 lb N2O/ lb N (0.16%) or 0.96 lb N2O/MG. It's worth noting that during steady-state operation, the anoxic tank utilized low-intensity aeration at a rate of 100 scfm. Replacing this mixing method with mechanical mixing could reduce emissions from this tank by minimizing stripping. The MBR tank maintained an average air flow rate of 180 scfm. In the broader context, it's important to highlight some limitations in this study. The data collection suffered from limited temporal resolution, as it was conducted over a two-week period during a single season. Nevertheless, despite these limitations, the overall process emission factor of 0.0016 N2O-hr/ lb N-hr was an order of magnitude lower than the emission factors established by the EPA and IPCC, which are 0.015 and 0.016 N2O/ lb N, respectively (IPCC 2019, U.S. EPA 2021). This is particularly noteworthy, given the use of aeration mixing in the anoxic tank, which might have increased emissions from that specific tank. Additionally, it's worth noting that the overall emission value falls within the lower end of the range reported in the IPCC study of thirty facilities, where emissions ranged from 0.0003 to 0.071 lb N2O per lb N processed per day (IPCC, 2020).