Primary Effluent- and Glycerol-Driven PdNA for Large-Scale Potable Reuse: Maximizing Benefits from MBBR to IFAS Transition

Introduction
The Los Angeles County Sanitation Districts (LACSD) and Metropolitan Water District of Southern California (MWD) have partnered to implement a full-scale advanced water treatment facility (AWTF) for reuse. One of the AWTF options considered was a new biological nutrient removal (BNR) facility followed by RO/UV/AOP. The BNR upgrades are intended to achieve a nitrate goal of <19 mg N/L, prior to RO/UV/AOP.

A promising approach identified to intensify BNR performance is Partial Nitrification/Denitrification/Anammox (PdNA). PdNA can reduce supplemental carbon requirements and enhance Total Inorganic Nitrogen (TIN) removal capacity without increasing tank volume. The ongoing multi-year pilot study summarized here explores PdNA application for a large-scale reuse program.

The study objectives are: i) to demonstrate reliable nutrient target achievement by intensifying a tertiary Moving Bed Biofilm Reactor (MBBR) and Integrated Fixed-film Activated Sludge (IFAS) configuration with PdNA, ii) to validate the carbon and aeration energy savings achieved through primary effluent-driven and glycerol-driven PdNA, and iii) to assess the benefits and considerations of incorporating suspended-phase biomass and provide insights for future full-scale PdNA design in membrane bioreactor systems.

Methods
Tertiary MBBR: The tertiary MBBR was a step-feed BNR process with alternating aerobic and anoxic zones (Figure 1). Both primary effluent (PE) and glycerol (MicroC 2000) were used to support denitrification in the anoxic zones, allowing for conventional BNR and PdNA processes.

Tertiary IFAS: The MBBR system was modified to include flocculant biomass retaining clarifiers (Figure 2 and 3). The first aerobic zone was converted to an anoxic zone, with RAS recycling NOx for pre-anoxic partial denitrification. PE was the primary carbon source for Anx1, but glycerol was used when necessary. Operational periods with and without PE are labeled in Figures 4, 5, and 6.

Monitoring and System Control: 24-hour composite samples were used to monitor influent and effluent solids, organics, and nutrients, while periodic grab samples from different system cells were analyzed for nitrogen species and organics. Nitrification and aeration were monitored with NOx and DO probes in each aerobic cell. In the IFAS system, a Solitax probe tracked MLSS concentration. Control loops were used for glycerol feeding based on NOx and DO readings. The NOx/NH3 ratio was controlled by adjusting SE step feed flowrates based on NH3 and NOx probe readings.

Operational Phases:
- Hybrid NdN Mode in MBBR: Both PE and glycerol were supplied to the system for full denitrification.
- Hybrid PdNA Mode in MBBR: Both PE and glycerol were provided to perform PdN and anoxic ammonia removal via anammox.
- PdNA Mode in IFAS: Suspended biomass was retained using a clarifier, and only glycerol (no PE) was supplied to perform PdNA.
- Hybrid PdNA Mode in IFAS: Suspended biomass was retained, and both PE and glycerol were supplied to the system.

Results and Discussion
System Performance: During steady-state operation (August to October 2023), the system consistently met the effluent 1 mg/L NHx limit and the 16 mg/L TIN limit (Figure 4). The transition from MBBR to IFAS mode occurred early November 2023 to April 2024, in which suspended solids began accumulating in the system (Figure 5). Despite operational challenges, the system maintained stable effluent NH3 levels throughout this period. However, during transition, effluent TIN levels temporarily rose above 20 mg/L due to fluctuating MLSS concentrations (Figure 4). Once MLSS stabilized at 2,000 mg/L (Figure 5), effluent TIN gradually decreased to less than 16 mg/L, with effluent NH3 consistently remaining below 1 mg/L (Figure 4).

PdNA Efficiency and Anammox Activity: Anoxic ammonia removal (AAR, g/m2/d), commonly used as a proxy for anammox activity during the PdNA process, is presented in Figure 6. In Anx2, ammonia removal was solely driven by glycerol-fed PdNA, while in Anx1, it was either driven by PE or by glycerol. The AAR driven by PE increased from approximately 0.21 g/m2/d during the steady-state hybrid MBBR operation (August to October 2023) to 0.37 g/m2/d during the steady-state hybrid IFAS operation (June to July 2024). Glycerol-driven AAR rates were comparable between MBBR and IFAS steady-state operations. However, Anx2 required lower bulk NHx and NO2 concentrations to maintain similar removal rates during IFAS mode (Figure 7), due to an increased anammox activity rate, as shown in ex-situ activity tests (Table 1). The bulk NHx concentration of <0.5 mgN/L in Anx2 also eliminated the need for Aer2 in the IFAS system.

Activity Migration into Suspended Phase: IFAS differs from MBBR in the incorporation of suspended biomass, which may facilitate the denitrifier migration from the biofilm phase to the suspended phase, freeing media space for slow-growing anammox bacteria. During the IFAS transition period, elevated effluent TIN levels were observed due to unstable MLSS concentrations (Figure 4). The reduction in ARR alongside MLSS reduction from December 2023 to March 2024 (Figure 6) suggested denitrifiers migrated to the suspended phase. The rise in effluent ammonia concentration (Figure 4) and 7/29/2024 biomass loss (Figure 5) suggests nitrifiers may have also migrated to the suspended phase. This was confirmed by ex-situ batch nitrification act