Towards Sustainable WRRFs: Exploring the Potential of Internally Stored Carbon for Efficient BNR

Introduction Water resource recovery facilities (WRRFs) may rely on external carbon sources like methanol for denitrification. However, carbon may be stored within the biomass to be used later in the biological nutrient removal (BNR) process for denitrification either post-anoxically or via simultaneous nitrification-denitrification (SND) and may reduce these external carbon requirements. Evidence of internally stored carbon (ISC) denitrification has been observed at full-scale (Bauhs et al., 2022; Regmi et al., 2023), although to a limited extent. 5-stage BNR facilities that are the focus of this analysis include the Seneca WRRF and Damascus WRRF (Washington Suburban Sanitary Commission [WSSC]), and the Virginia Initiative Plant (VIP) and Nansemond Treatment Plant (NTP) (Hampton Roads Sanitation District [HRSD]). Seneca and Damascus show indications of ISC denitrification: low to no methanol addition and yet 4-7 mg N/L of post-anoxic nitrate reduction. Similarly, profiles at VIP attribute 33% of post-anoxic N removal to ISC. NTP serves as a reference facility with no ISC denitrification. This analysis presents a framework for further implementation of ISC denitrification, drawing on an extensive literature review and observations from full-scale WRRFs. Operational Factors Several factors have been theorized to contribute to ISC denitrification at the WSSC and HRSD facilities. High selector F/M and influent C/N: High food-to-microorganism (F/M) ratios promote substrate storage in selector zones, as is the case for densified systems (Sturm 2020). The F/M at VIP far exceeds the recommended value of 2.4 g BOD/g MLSS/d for anaerobic selectors (Jenkins et al., 2003). All three ISC denitrifying facilities have high influent carbon-to-nitrogen (C/N) ratios above 12. Low IMLR: Lower internal mixed liquor recycle (IMLR) rates promote deeper anaerobic conditions upfront in the process for better carbon storage. IMLR rates at Seneca were reduced from 400% to 200%, and Figure 1 shows the significantly lower batch test post-anoxic ISC denitrification rates for the 400% IMLR rate. Low DO: This results in less oxidation of ISC prior to denitrification. Implementation of ABAC at Seneca (averaging 0.3 mg/L DO) improved SND and reduced post-anoxic methanol addition (Regmi et al, 2023). Batch tests at HRSD showed longer aerobic HRT led to lower ISC denitrification rates (Bauhs et al., 2022). Bio-P performance: Sources have identified PHA (Carvalho et al., 2007; Qin et al., 2005) and/or glycogen (Coats et al., 2011; Vocks et al., 2005; Winkler et al., 2011) as the likely form of ISC for denitrification, both of which are forms of carbon in the metabolism of organisms involved in bio-P. Batch testing of HRSD biomass showed a positive correlation of ISC denitrification rates with P uptake rates (Bauhs et al., 2022). Plug flow reactor configuration: The high substrate gradient may enable access to this carbon source. The post-anoxic zone configuration at VIP is equivalent to over 30 continuously stirred tank reactors (CSTRs) in series, compared to facilities like NTP with 3 CSTRs. Figure 2 shows an aerial of the post-anoxic zones at all four facilities. Benefits There are many benefits to implementation of ISC, as follows:

*Chemical Savings: Elimination of external carbon reduces chemical costs. Furthermore, less external carbon addition lowers sludge production and associated solids handling costs.

*Energy Savings: Lower IMLR reduces associated pumping energy while low DO operation leads to significant aeration energy savings.

*Settleability Improvements: Use of a selector zone to promote carbon storage is similar to biological selection factors for densification, so promoting ISC could improve settleability.

*Simplification of Process: Pre-anoxic zones could ultimately be eliminated (use an anaerobic/oxic/anoxic process as in Coats et al. (2023)). Furthermore, pending SND performance, the post-anoxic zone could be as well. Discussion Several knowledge gaps still remain, which if addressed could significantly increase implementation of ISC denitrification. First, how resilient is ISC for N removal? Responses to variation in WRRF loading or other process upsets is not yet understood. For instance, Figure 3 shows the inconsistency in methanol dose at VIP over many seasons. The viability of operating without a 'backup' system (i.e. external carbon addition) should be explored. Next, while several factors may promote ISC denitrification, which factors are the highest priority and what level of control do they offer? Table 1 summarizes these factors for each of the four full-scale facilities, but notably for Damascus, most are absent. Clearly some aspects of operation/configuration may be more important than others, or haven't yet been identified. Some, like ABAC for low DO, may be relatively straightforward. Others, like boosting C/N, may require more guidance. Related to the somewhat surprising observations of low methanol use at Damascus, there arises another question-does reduction of methanol itself promote ISC denitrification? In other words, there could be some interference of methylotrophs with ISC denitrifying organisms (Bauhs 2021). Nitrate concentrations in these systems are not limiting; nevertheless, cease of external carbon feed could trigger a microbial population shift.