The Mechanism of Aggravating N 2 O Emission in a Partial Nitritation Reactor with Increasing bCOD/NH 4 + –N Authors: Shenghao Ji, Shen Cui, Hongjun Zhao, Yujie Chen, Tianyi Zhang, Yu Qin, Jianyong Liu, Yu-You Li Published: 2026 ABSTRACT Partial nitritation (PN) is an important source of nitrous oxide (N2O) emissions, and a low biodegradable chemical oxygen demand to ammonium (bCOD/NH4 +–N) ratio tends to aggravate N2O emissions. However, the specific release characteristics and potential mechanisms with increasing bCOD/NH4 +–N within a low ratio range have not been elucidated. In this study, a continuous-flow PN reactor was operated using food waste digestate characterized by fluctuating low bCOD/NH4 +–N ratios. When the ratio rose from 0.12 to 1.36, the gaseous N2O emission factor increased from 3.2% to 25.8%. Three mechanisms were inferred to be associated with this increase: (1) To maintain the nitrite-to-ammonium ratio at higher bCOD/NH4 +–N, the ammonia oxidation efficiency (AOE) was elevated from 57.7% to 75.1%, which increased the electron flux used for nitrite reduction; (2) to sustain the nitrite-to-ammonium ratio, the aeration coefficient was increased, which elevated the N2O mass transfer coefficient (KLa) from 32 to 705 d–1 and consequently enhanced N2O emissions; (3) the increasing bCOD/NH4 +–N induced a nonuniform upregulation of functional genes, the (nirK+nirS)/nosZ rose from 2.62 to 3.14, enhancing the N2O-producing capacity. Overall, AOE-driven electron flux enrichment, aeration coefficient-driven KLa enhancement, and uneven functional gene promotion may have jointly contributed to the increase in N2O emissions with increasing bCOD/NH4 +–N. Introduction Greenhouse gas (GHG) emissions are driving a range of serious environmental challenges. Wastewater treatment plants (WWTPs) are a non-negligible contributor, accounting for roughly 1.6% of global GHG emissions. , Among these gases, nitrous oxide (N2O) generated during biological nitrogen removal (BNR) is of particular concern because its global warming potential is about 273 times that of CO2, and it may account for more than 50% of the carbon footprint of WWTPs. , Anaerobic ammonium oxidation (anammox), as an advanced alternative to the conventional activated sludge process, offers several advantages: low energy demand, no requirement for external organic carbon, low excess sludge production, and reduced GHG emissions, and is therefore regarded as a promising route toward low-carbon wastewater treatment. , However, the partial nitritation (PN) unit that supplies nitrite for anammox unavoidably produces N2O. Thus, understanding the mechanisms of N2O formation and release during PN has become a key research focus. In the PN unit, N2O may be generated through two pathways: (i) the denitrification pathway (including nitrifier denitrification and heterotrophic denitrification), in which nitrite (NO2 –) is used as an alternative electron acceptor and is stepwise reduced to N2O; and (ii) the incomplete oxidation of hydroxylamine (NH2OH). , Moreover, N2O production in PN is also sensitive to several environmental and operational factors, including temperature, pH, free ammonia (FA), free nitrous acid (FNA), and biodegradable chemical oxygen demand to ammonium (bCOD/NH4 +–N) ratio, among others. , Previous studies have reported that, within a suitable range, increasing temperature enhances N2O emissions. For example, when the temperature was raised from 10 to 20 °C, the gaseous N2O emission factor increased from 1.5 ± 0.3% to 3.7 ± 0.5%. Likewise, within a pH range of 6.0–8.0, higher pH promoted N2O production, reaching 0.53 ± 0.04 mg N2O–N/h/g VSS at pH 8.0. By contrast, FA and FNA generally suppress N2O release. In a lab-scale PN reactor, elevating FA to 29.3 mg N/L reduced the N2O emission factor from 0.90% to 0.11%, and exposing PN sludge to 1.5 mg HNO2–N/L FNA for 1 h almost completely inhibited N2O emissions. , Among these influencing factors, some (e.g., temperature) can be directly adjusted through external devices, while others (e.g., pH, FA, FNA) are largely constrained by chemical equilibria in the system. The bCOD/NH4 +–N ratio, however, is fundamentally different: it is difficult to regulate within the PN unit. , Therefore, understanding the effect and mechanisms of bCOD/NH4 +–N on N2O emissions during PN is particularly important. To date, several studies have reported on the influence of large-scale variations in bCOD/NH4 +–N on N2O emissions from PN-related systems. For example, when the bCOD/NH4 +–N ratio increased from 1.3 to 6.5, the N2O released decreased from 25% to 3% of the nitrogen removed. These results indicate that higher bCOD/NH4 +–N ratios can lead to lower N2O emissions, mainly because organic carbon enhances the denitrification process. Theoretically, complete reduction of nitrite to N2 requires about 2.7 g bCOD/g NO2 ––N. However, in PN systems, an ammonium oxidation efficiency of roughly 57% must be maintained. Therefore, only when the influent bCOD/NH4 +–N exceeds approximately 1.54 can sufficient carbon be supplied to fully reduce nitrite to N2. Under low bCOD/NH4 +–N conditions, carbon becomes limiting, denitrification may terminate at N2O, and N2O will consequently accumulate and be released. Therefore, the N2O emission characteristics with increasing bCOD/NH4 +–N within a low range may not be consistent with those observed across the large-scale range. In practice, many high-strength ammonium wastewaters exhibit low bCOD/NH4 +–N characteristics, such as food waste digestate and mature landfill leachate. , In particular, food waste digestate often shows seasonal fluctuations, making the bCOD/NH4 +–N ratio even more unstable. Although it is recognized that low bCOD/NH4 +–N favors N2O emissions, the effects of fluctuations within the low ratio zone have not been systematically quantified. Moreover, the potential pathways and mechanisms that bCOD/NH4 +–N stimulates N2O formation remain insufficiently elucidated. In this study, a continuous-flow air-lift PN reactor was used to study the N2O emissions during the treatment of food waste digestate exhibiting fluctuating low bCOD/NH4 +–N ratios. The specific objectives were to (1) analyze the key indicators affecting N2O emissions in the PN process; (2) evaluate the release characteristics of N2O with increasing bCOD/NH4 +–N within a low range; and (3) elucidate the potential pathways and mechanisms of N2O generation and release with increasing bCOD/NH4 +–N. Materials and Methods Reactor and Wastewater Types An air-lift reactor (working volume 10 L) was employed in this study (Figure S1). The reactor temperature was maintained at 30 °C using an electric heating rod. Aeration was supplied by an air pump and introduced at the bottom of the reactor through a diffuser stone, and the airflow rate was measured with a rotameter. During operation, pH and temperature were continuously monitored through a remote online system. The influent was food waste digestate obtained from the anaerobic digestion plant (JNEX) in Japan; its characteristics are described in Table S1. Its chemical oxygen demand (COD) fluctuated over time, providing realistic conditions for evaluation. Seed sludge was taken from a partial nitritation reactor that treated the same type of wastewater. Basic Analytical Methods Liquid samples collected from the PN reactor were passed through 0.45 μm membrane filters and stored at 4 °C until analysis. The concentrations of nitrate (NO3 ––N), ammonium (NH4 +–N), and nitrite (NO2 ––N) were determined using a capillary electrophoresis system (Agilent 7000, USA). Chemical oxygen demand was analyzed according to standard methods. N2O Analyses Gas samples were collected in gas bags using a custom gas-collection device positioned just below the liquid surface and were analyzed by GC–MS (6890N-5975B). Liquid samples, including both the influent and effluent, were likewise collected for N2O determination. The detailed procedures for preservation and analysis of N2O in liquid samples are described in Figure S2 and Text S1. The calculation methods for the gaseous and liquid N2O emission factors are as follows (eqs and ). Gas-emission factor (G-EF, %) represents the proportion of N2O released to the gas phase relative to the influent nitrogen load rate, and liquid-emission factor (L-EF, %) represents the proportion of N2O in the liquid phase relative to the influent nitrogen loading rate. 1 G‐EF=G‐N2ONLR×V×1000 2 L‐EF=L‐N2ONLR×V×1000 where G-N2O (mg N/d) refers to the daily mass of N2O released in the gas phase, and L-N2O (mg N/d) refers to the daily mass of N2O in the liquid phase. The calculation methods are referred to Text S2. NLR (g N/L/d) refers to the nitrogen loading rate of the PN reactor, and V (L) refers to the working volume of reactor. The N2O transfer rate (J, mol/L/d) and N2O mass transfer coefficient (KLa, d–1) are important indicators for evaluating N2O mass transfer in the reactor, and their values can be calculated as described below (eqs and ): 3 J=G‐N2O44×1000×V 4 KLa=J×44×1000CL−C* where CL refers to the actual concentration of N2O in the liquid phase (mg N/L). C* refers to the equilibrium concentration of N2O in the liquid phase (mg N/L), which can be calculated based on Text S3. Quantitative PCR Sludge samples collected from the reactor under different bCOD/NH4 +–N conditions were subjected to DNA extraction using the FastDNA Spin Kit for Soil (MP Biomedicals, USA). The concentration and purity of the extracted DNA were assessed with a NanoDrop 2000 spectrophotometer (NanoDrop Inc., USA). Copy numbers of the three target functional genes (nirK, nirS, and nosZ) were quantified by quantitative PCR (qPCR). PCR amplification was carried out on a GeneAmp 9700 thermocycler (Applied Biosystems, USA). The primer sequences are listed in Table S2. Metagenomics Sequencing Total genomic DNA was extracted from the collected sludge samples using the E.Z.N.A. Soil DNA Kit (Omega Biotek, Norcross, GA, USA) in accordance with the manufacturer’s instructions. The integrity and purity of the DNA were evaluated by 1% agarose gel electrophoresis and spectrophotometric analysis. Genomic DNA was subsequently sheared to an average fragment size of approximately 300 bp, and paired-end libraries were prepared using the NEXTFLEX Rapid DNA-Seq Kit (Bio Scientific, USA). The libraries were quantified, pooled to the appropriate concentration, and sequenced on the Illumina NovaSeq 6000 platform (Illumina Inc., San Diego, CA, USA) using NovaSeq Reagent Kits at MajorBio Co., Ltd. (Shanghai, China). Raw reads were quality-filtered with fastp to remove low-quality sequences, and high-quality reads were assembled into contigs using MEGAHIT. Subsequent bioinformatic analyses were performed on the MajorBio Cloud Platform. Calculations The ammonium oxidation efficiency (AOE, %) and ammonium oxidation rate (AOR, g N/L/d) in the PN process are directly correlated with the electron flux and can be used to indirectly evaluate its magnitude. The calculation methods are as follows (eqs and ): 5 AOE=[NH4+−Ninf]−[NH4+−Neff][NH4+−Ninf] where NH4 +–Ninf and NH4 +–Neff refer to the ammonium concentrations (mg/L) in the influent and effluent of the reactor, respectively. 6 AOR=NLR×AOE where AOE refers to ammonium oxidation efficiency mentioned above, and the calculation of NLR was described in Text S4. The aeration coefficient (AC, m3 air/kg N) is a parameter that reflects the contact ratio between the unit volume of air and the unit concentration of ammonium. It is also significantly correlated with N2O emissions in the PN process. The calculation formula is shown as follows (eq ): 7 AC=ARNLR×V where AR refers to aeration rate (L/d), and V refers to working volume of the reactor (L). Biodegradable chemical oxygen demand (bCOD, mg/L) represents the COD that can be biologically utilized and removed during the PN process, and it is calculated as follows (eq ): 8 bCOD=CODinf−CODeff where CODinf and CODeff (mg/L) refer to COD concentrations in the influent and effluent of the PN reactor. Results and Discussion The Performance of the PN Reactor under Different bCOD Conditions The bCOD is a critical parameter influencing the performance of the PN process. Excessive bCOD not only intensifies competition for oxygen between ordinary heterotrophic organisms (OHOs) and ammonia-oxidizing bacteria (AOB) but also promotes the overgrowth of heterotrophs, thereby occupying the ecological niche of AOB. In this study, the operational period was divided into three stages based on fluctuations in bCOD: Stage I, stable operation; Stage II, high-bCOD shock; and Stage III, low-bCOD shock, as shown in Figure . Long-term operation performance of the air-lift partial nitritation under different bCOD conditions. During Stage I, bCOD remained relatively low at 258 ± 149 mg/L, accounting for 21.9% of the influent COD. Under these conditions, an AC of 294 ± 67 m3 air/kg N was sufficient to maintain a stable nitrite-to-ammonium (NO2 ––N/NH4 +–N) ratio of 1.20 ± 0.18. Due to the presence of bCOD, denitrification occurred within the continuous-flow PN system, resulting in a modest total nitrogen removal efficiency (TNRE) of 9.0 ± 6.7%. Moreover, PN is also a major source of N2O generation under the two-stage PNA system. N2O was mainly released in the gas phase, and the G-EF was 2.64 ± 1.07%. In Stage II, seasonal fluctuations in upstream processing led to a substantial increase in digestate bCOD, reaching 1540 ± 689 mg/L (58.6% of influent COD). Given the comparable effluent COD concentrations between Stage II and Stage I (949 ± 381 mg/L and 1140 ± 298 mg/L, respectively), the rise in influent COD was primarily attributed to the increase in bCOD. To maintain PN performance, the AC was increased to 622 ± 161 m3 air/kg N, stabilizing the NO2 ––N/NH4 +–N ratio at 1.33 ± 0.42. The elevated bCOD significantly enhanced denitrification performance, improving TNRE to 38.7 ± 12.4%. Notably, the fluctuations in bCOD led to pronounced variations in N2O emissions, G-EF was 20.87 ± 7.46%, indicating a significant enhancement in gaseous N2O release during this stage. In Stage III, upstream process improvements reduced bCOD to 636 ± 295 mg/L (38.9% of influent COD), while refractory organics remained relatively stable at 1001 ± 202 mg/L. The AC was subsequently lowered to 532 ± 133 m3 air/kg N, maintaining a NO2 ––N/NH4 +–N ratio of 1.25 ± 0.23. The decline in bCOD reduced denitrification performance, resulting in a lower TNRE of 17.4 ± 7.4%. bCOD fluctuations again influenced N2O dynamics, with G-EF at 7.57 ± 3.58%. Notably, G-EF was substantially mitigated as bCOD decreased. The Analysis of Key Indicators Affecting N2O Emission in PN Process Several studies have examined the factors influencing N2O emissions during the PN process. Most previous investigations were conducted in sequencing batch reactors (SBRs) and focused on key parameters, including bCOD/NH4 +–N, AR, AOR, FA, FNA, and pH. These parameters can likewise fluctuate in continuous-flow PN reactors treating food waste digestate, driven by variations in influent characteristics and operational conditions. An increase in pH within an appropriate range has been reported to promote N2O emissions. Within the pH range of 6.0–8.0, pH showed a positive relationship with N2O emissions, reaching the highest level at pH 8.0, whereas N2O emissions decreased when pH exceeded 8.0, likely due to the decline in ammonia oxidation activity beyond its optimum at pH 8.0. Additionally, enhanced aeration has likewise been shown to promote N2O release. When AR increased from 10 to 20 mL/min, N2O emissions rose from 2.67 to 3.08 mg under an influent NH4 +–N concentration of 100 mg/L. Several negatively correlated parameters have also been studied. Limited bCOD has been shown to increase N2O emissions. For instance, when the bCOD/NH4 +–N ratios were 1.3, 1.6, and 6.5, the corresponding N2O emission factors were 20%, 14%, and 2%, respectively. Both FA and FNA have demonstrated negative correlations with N2O production. Elevated FA can reduce AOB denitrification capability, lowering N2O generation through the AOB-denitrification pathway. High FNA concentrations may cause microorganisms to deactivate specific enzyme systems to resist toxicity and may directly react with the copper-containing enzyme NirK, leading to inactivation or conformational changes, ultimately reducing N2O formation. In addition, dissolved oxygen (DO) is also a widely discussed influencing factor. Higher DO has been shown to suppress N2O release because DO competes with NO2 – for excess electrons during ammonia oxidation, thereby inhibiting the nitrite to N2O pathway. However, in the continuous-flow partial nitritation reactor, the DO in the reaction zone was strictly controlled below 0.1 mg/L, because excess DO would further oxidize ammonium and shift the target NH4 +–N/NO2 ––N ratio, thereby disrupting the partial nitritation state. Therefore, the effect of DO was not discussed in detail. Despite these established findings, two important parameters have received little attention. One is the AC, a critical operational parameter in continuous-flow PN that represents the aeration volume (m3) required per kilogram of ammonium nitrogen to achieve a target NO2 ––N/NH4 +–N ratio. The other is AOE. Although the theoretical AOE in mass-flow controlled PN processes is approximately 57%, the occurrence of nitrite denitrification induced by bCOD often requires higher AOE values to maintain the required NO2 ––N/NH4 +–N balance. The variations in these two parameters are both associated with fluctuations in bCOD/NH4 +–N, and an increase in bCOD/NH4 +–N simultaneously promotes increases in AC and AOE. However, the effects of changes in these two parameters on N2O release have not yet been effectively explained. Moreover, under continuous-flow PN treating real wastewater, it remains unclear which of the positively correlated parameters exerts the strongest influence on N2O release. To address this gap, a Spearman correlation analysis was conducted using long-term operational data (Figure ). The results showed that the bCOD/NH4 +–N ratio exhibited the strongest positive correlation with G-EF (r = 0.78), which differs markedly from previous studies. Most existing work has investigated N2O emissions under large-scale variations in the bCOD/NH4 +–N ratio, such as 1.3–6.5, whereas in this study, the fluctuations in bCOD/NH4 +–N were confined to a relatively narrow range (0–1.36). This finding indicates that the response patterns of N2O emissions to bCOD/NH4 +–N variations are not the same at different fluctuation scales. AR showed strong positive correlations, with coefficients of 0.57, consistent with previous reports. Interestingly, AOR displayed a weak negative correlation with G-EF, contrary to earlier findings. This discrepancy may be explained by the fact that AOR primarily reflects microbial activity rather than the extent of the reaction process. Under operating conditions with a relatively stable sludge concentration, AOR is directly related to the sludge loading rate and therefore mainly indicates the overall activity level of sludge in the reactor. Since electron flux partitioning depends more strongly on the actual conversion extent of ammonia oxidation, AOE (%) is more appropriate than AOR as a key indicator for interpreting the mechanism of N2O production. Consistent with this interpretation, AOE showed a strong positive correlation with N2O emissions (r = 0.73). pH also showed a strong positive correlation with N2O emissions, with a correlation coefficient of 0.73. The detailed pH variations during operation are shown in Figure S3. This trend contrasts with observations reported in other studies at pH values above 8.0, which may be attributed to the strong influence of pH on FNA and FA concentrations. Specifically, increasing pH decreases FNA concentration while increasing FA concentration, thereby affecting N2O release. FNA showed a strong negative correlation with N2O emissions (r = −0.75), whereas FA exhibited the opposite trend (r = 0.33). The strong negative correlation with FNA may be the reason for the apparent positive correlation between pH and N2O emissions. In addition, the contrasting correlations of FA and FNA were mainly attributable to the combined effects of substrate concentration and pH. Specifically, as the bCOD/NH4 +–N ratio increases, nitrite concentration gradually decreases, while denitrification simultaneously increases the reactor pH. These two effects jointly reduce the FNA concentration. Under these conditions, a higher AOE needs to be adopted to maintain the NO2 ––N/NH4 +–N ratio, which in turn lowers ammonium concentration. However, because of the pH increase induced by denitrification, FA may still increase even when ammonium concentration decreases (the calculation methods for FNA and FA are described in Text S5). Notably, the AC, a parameter seldom examined in previous studies, showed a strong positive correlation with N2O emissions (r = 0.72), even stronger than AR. Overall, these results indicate that the bCOD/NH4 +–N ratio is the most influential parameter governing N2O emissions in continuous-flow PN systems. Spearman correlations between N2O emission factors and key operational parameters of the partial nitritation process. Release Characteristics of N2O under Different bCOD/NH4 +–N Conditions The above discussion has demonstrated that the bCOD/NH4 +–N ratio is the most critical indicator influencing N2O emissions during the PN process. However, a quantitative assessment of N2O emission levels under different bCOD/NH4 +–N conditions is still lacking. Therefore, it is necessary to further investigate the specific N2O emission characteristics across varying bCOD/NH4 +–N gradients. Figure a and b is based on the analysis of the long-term monitoring results, in which the bCOD/NH4 +–N ratios were categorized into five gradients (0.12, 0.30, 0.53, 0.76, 1.36). Gaseous and liquid N2O emissions were analyzed separately. The concentration of gaseous N2O increased markedly with increasing bCOD/NH4 +–N. Specifically, gaseous N2O levels were 82 ppm at the lowest ratio (0.12) and increased to 402 ppm at the highest ratio (1.36) (P < 0.0001). In contrast, liquid N2O concentrations decreased substantially with increasing bCOD/NH4 +–N, from 3.0 mg/L at 0.12 to 0.7 mg/L at 1.36 (P = 0.0060). Release characteristics of N2O with increasing bCOD/NH4 +–N ratio: (a) absolute N2O amount in gas phase; (b) absolute N2O amount in liquid phase; (c) N2O emission factors in gas phase; (d) N2O emission factors in liquid phases. Although the above results reveal the absolute N2O concentrations, the emission factor more accurately reflects the overall N2O emission potential. As shown in Figure c and d, the G-EF increased significantly with an increasing bCOD/NH4 +–N ratio, rising from 3.2% at 0.12 to 25.8% at 1.36 (P < 0.0001). Conversely, the L-EF decreased from 0.21% to 0.05% across the same range (P = 0.0456), consistent with the concentration trend. Despite these opposite trends, the G-EF was approximately 516 times higher than the L-EF in the highest bCOD/NH4 +–N ratio. Thus, elucidating the potential mechanisms through which the bCOD/NH4 +–N ratio regulates gaseous N2O release is of particular significance. Pathway Analysis of Increasing bCOD/NH4 +–N on N2O Emissions Increasing the Ammonium Oxidation Efficiency The correlation analysis indicated that AOE is a more influential indicator than AOR in increasing N2O emissions. The effect of AOR primarily stems from the stronger electron flux associated with higher ammonia oxidation activity. , During the ammonia oxidation process, the conversion of ammonia (NH3) to NH2OH requires two electrons, whereas the subsequent oxidation of NH2OH to NO2 – releases four electrons. , Thus, two surplus electrons are generated in the overall reaction. If not effectively transferred or consumed, these electrons can react with NO2 – to form N2O. , However, in the high-loading continuous-flow PN reactor, AOR can remain relatively stable under different bCOD/NH4 +–N and is less affected by fluctuations in NLR (Figure S4). Consequently, when AOR does not vary substantially, AOE becomes a more direct determinant of electron flux, because electron generation is primarily governed by the extent of NH4 + oxidation to NO2 –. Moreover, under continuous-flow PN operation, the DO remains close to zero, and the absence of O2 as a competing electron acceptor allows more surplus electrons to participate in N2O production. , Therefore, an increase in AOE may contribute to enhanced N2O formation. When the bCOD/NH4 +–N ratio increases, denitrification occurs under low DO conditions, leading to a corresponding increase in TNRE. As the bCOD/NH4 +–N ratio increased from 0.12 to 1.36, the TNRE rose from 8.5% to 44.9% (Figure a). The increase in TNRE was mainly attributable to nitrite denitrification, which led to a decrease in nitrite concentration from 770 mg/L to 414 mg/L (Figure b). The decrease in nitrite concentration would disrupt the NO2 ––N/NH4 +–N ratio, compromising the stability of the downstream anammox process. To maintain the target NO2 ––N/NH4 +–N ratio of 1.32, the AC was increased to compensate for nitrite loss through denitrification. Under this adjustment, the NO2 ––N/NH4 +–N ratios remained stable across all bCOD/NH4 +–N conditions, with values of 1.26, 1.22, 1.27, 1.31, and 1.21, respectively (Figure c). This stability was largely attributable to the increase in AOE under elevated AC, ensuring that more ammonium was oxidized to nitrite and compensating for nitrogen loss. As shown in Figure d, AOE increased significantly from 57.7% to 75.1% as the bCOD/NH4 +–N ratio increased from 0.12 to 1.36. The increase in electron flux associated with higher AOE provides a favorable condition for N2O formation. Therefore, AOE stimulation under elevated bCOD/NH4 +–N ratio is inferred to be a possible contributor to the observed increase in N2O emissions. Promotion of AOE with increasing bCOD/NH4 +–N ratio: (a) increase in TN removal efficiency; (b) decrease in nitrite concentration; (c) maintenance of the NO2 ––N/NH4 +–N ratio; (d) enhancement of AOE. Increasing the Liquid–Gas Mass Transfer Coefficient of N2O The correlation analysis confirmed that AC is a more influential indicator than AR in determining N2O emissions. As AC reflects the volume of air supplied per unit of nitrogen and is unaffected by fluctuations in the influent nitrogen concentration, it provides a more reliable metric for investigating N2O release mechanisms. Conventionally, increases in AR are considered to enhance gas–liquid mass transfer and thereby promote the stripping of dissolved N2O from the liquid to the gas phase. However, under fluctuating wastewater characteristics, AC and AR do not always follow consistent variation trends (Figure S5). Therefore, whether an increase in AC alone can enhance gas–liquid N2O transfer requires further evaluation. Increasing AC proved to be an effective strategy for maintaining the target NO2 ––N/NH4 +–N ratio under increasing bCOD/NH4 +–N conditions. As shown in Figure a, AC increased from 289 to 776 m3 air/kg N across the increasing bCOD/NH4 +–N. To determine whether differences in AC under varying bCOD/NH4 +–N conditions could promote N2O mass transfer, the N2O transfer rate was examined. As the AC increased, the N2O transfer rate rose from 0.0022 to 0.0116 mol/L/d, indicating that higher AC significantly increases the amount of N2O transferred from the liquid to the gas phase (Figure b). However, the transfer rate reflects the quantity of N2O transported per unit time and is influenced by concentration gradients and gas–liquid interfacial area. In contrast, the N2O mass transfer coefficient (KLa) represents the intrinsic mass transfer capacity of the system and is independent of concentration gradients. According to the calculated results, KLa increased markedly from 32 d–1 to 705 d–1 as the AC increased (Figure c). This finding demonstrates that increases in AC significantly enhance the N2O mass transfer coefficient. Therefore, under high AC conditions, the substantial increase in KLa is inferred to be one possible contributor to the increased N2O emissions observed at higher bCOD/NH4 +–N ratios. Promotion of N2O mass transfer with increasing bCOD/NH4 +–N ratio: (a) increase in aeration coefficient; (b) increase in N2O transfer rate; (c) increase in N2O mass transfer coefficient. Increasing Differences of Functional Genes From the perspective of functional gene abundance, nirK was the most dominant gene detected, followed by nosZ (Figure a). This can be attributed to the relatively high influent COD, which stimulated denitrification, and to the high nitrite levels in the PN system, which favored the first reduction step of denitrification (NO2 – → NO), thereby leading to higher nirK abundance. Notably, with increasing bCOD/NH4 +–N, nirK, nirS, and nosZ all increased, especially for nirK+nirS, which rose from 517 to 965 (Figure b). This confirms that higher organic carbon input promotes denitrification-related functional genes, but this alone does not sufficiently explain the enhanced N2O emissions. Nonuniform promotion of functional genes with increasing bCOD/NH4 +–N ratio: (a) TPM changes in nitrogen cycle-related functional genes; (b) TPM changes in nirK + nirS; (c) changes in (nirK + nirS)/nosZ based on TPM; (d–f) changes in gene copy numbers of nirK, nirS, and nosZ; (g) changes in (nirK + nirS)/nosZ based on gene copy numbers. What truly reflects the tendency of N2O to accumulate is the (nirK+nirS)/nosZ ratio. , Theoretically, a higher (nirK+nirS)/nosZ value indicates that the system has a stronger upstream reduction capacity (NO2 – → NO → N2O), while the downstream reduction capacity (N2O → N2) is relatively weaker, making N2O more prone to accumulation. As shown in Figure c, when bCOD/NH4 +–N increased from 0.12 to 1.36, the (nirK+nirS)/nosZ ratio rose from 4.10 to 4.80, indicating that with increasing organic carbon supply under low bCOD/NH4 +–N conditions, the overall denitrification potential was indeed activated, but it shifted more toward the N2O-producing steps, thereby increasing the risk of N2O accumulation. To further verify these observations, quantitative analyses of the functional genes nirK, nirS, and nosZ were conducted under different bCOD/NH4 +–N ratios. The absolute abundances of nirK, nirS, and nosZ all increased with increasing bCOD/NH4 +–N: nirK increased from 1.52 × 108 to 8.90 × 108 copies/g (Figure d), nirS increased from 5.82 × 109 to 7.33 × 109 copies/g (Figure e), and nosZ increased from 2.27 × 109 to 2.62 × 109 copies/g (Figure f). The increase in nirK + nirS was greater than that of nosZ, causing the (nirK + nirS)/nosZ ratio to rise from 2.62 to 3.14, which indicates an enhanced potential for N2O accumulation (Figure g). Multiple-Pathway Mechanism of Aggravating N2O with Increasing bCOD/NH4 +–N The above analysis indicates that the stimulatory effect of bCOD/NH4 +–N increasing on N2O emissions in the PN process may not be attributable to a single factor but is more likely associated with the combined contributions of three inferred pathways. Strengthening of ammonia oxidation efficiency and the nonuniform promotion of functional genes tend to promote N2O formation, whereas KLa enhancement facilitates the transfer of dissolved N2O from the liquid to the gas phase. These three pathways act in the same direction and jointly contribute to the observed increase in N2O emissions. When the influent bCOD/NH4 +–N ratio increases, denitrification inevitably occurs and is further intensified under the low-DO conditions of the continuous-flow PN reactor. To maintain the required NO2 ––N/NH4 +–N ratio for PN, the AOE must be increased by increasing AC to compensate for nitrite losses via denitrification. Consequently, both the increased AOE and the elevated AC contribute to higher N2O emissions. On the one hand, an increase in AOE directly amplifies the potential for N2O formation. Ammonia oxidation proceeds via two sequential steps: NH3 to NH2OH (consuming 2 e–) and NH2OH to NO2 – (producing 4 e–), as described in eqs and . Overall, this results in 2 surplus electrons. Under the low DO conditions of continuous-flow PN, where O2 is absent as a competing electron acceptor, these surplus electrons are more likely to react with accumulated NO2 – to form N2O, as described in eqs and . , Therefore, as AOE increases, the flux of surplus electrons also increases, which may drive N2O production (Figure a). On the other hand, a higher AC was employed to buffer the competition for oxygen under higher bCOD/NH4 +–N conditions, which increases the KLa and thereby accelerates the transfer of biologically produced N2O from the liquid to the gas phase (Figure b). It should be emphasized that the increase in KLa does not generate N2O but merely facilitates the release of N2O already formed. 9 NH3+O2+2H++2e−→NH2OH+H2O 10 NH2OH+H2O→NO2−+5H++4e− 11 NO2−+2H++e−→NO+H2O 12 2NO+2H++2e−→N2O+H2O Inferred pathways potentially contributing to increased N2O release in partial nitritation under increasing bCOD/NH4 +–N: (a) the increase of AOE aggravates electron flux; (b) the increase of KLa aggravates N2O release; (c) nirK + nirS increased disproportionately relative to nosZ. From a biological perspective, N2O formation in PN systems may be regulated by three groups of functional genes: nirK and nirS mediate the reduction of NO2 – to NO, norB and norC mediate the reduction of NO to N2O, and nosZ mediates the reduction of N2O to N2. Although an increased bCOD/NH4 +–N ratio can stimulate the abundance of the three groups of functional genes, the degree of stimulation is not uniform. bCOD/NH4 +–N exerts a stronger promotive effect on the upstream nirK and nirS genes than on the downstream nosZ gene (Figure c). As a result, the system’s potential for N2O production may increase more rapidly than its potential for N2O reduction, thereby favoring N2O accumulation. It is worth noting that these three inferred pathways may also help explain the divergence in N2O emission behaviors across a broader range of bCOD/NH4 +–N conditions (Text S6). Implications for Engineering and Future Research This study demonstrates that within the low bCOD/NH4 +–N range, an increase in bCOD/NH4 +–N markedly enhances N2O emissions. In high-ammonium wastewater treatment, many typical waste streams, such as food waste digestate and mature landfill leachate, are characterized by low bCOD/NH4 +–N ratios, which, moreover, fluctuate with seasonal variations. Under such conditions, partial nitritation is prone to elevated N2O production and release. Therefore, although the two-stage PNA can partially buffer influent bCOD in the upstream PN reactor, implementing effective bCOD/NH4 +–N pretreatment remains crucial for mitigating N2O emissions during the PN stage. For instance, based on the findings of this study, reducing the influent bCOD/NH4 +–N ratio from 1.36 to 0.12 through pretreatment could lower the G-EF from 25.8% to 3.2%. Such a substantial decrease highlights the significant environmental benefit of reducing bCOD/NH4 +–N. Notably, lowering influent bCOD/NH4 +–N is not expected to compromise the overall nitrogen removal efficiency in two-stage PNA, because the PN unit primarily functions to establish an appropriate NO2 ––N/NH4 +–N ratio, while deep nitrogen removal is mainly achieved in the downstream anammox unit. Currently, several technologies can serve as effective pretreatment methods prior to the PNA process, including high-rate activated sludge systems, anaerobic membrane bioreactors, and traditional anaerobic digestion. − These technologies can efficiently remove biodegradable organics, lowering the bCOD/NH4 +–N level. Integrating these pretreatment processes with a PNA configuration can greatly suppress N2O emissions during the PN stage in engineering applications. Although pretreatment that reduces the bCOD/NH4 +–N to 0.12 can substantially mitigate N2O emissions, the G-EF remains at 3.2%. Therefore, further abatement of this residual N2O emission represents a key direction for future research. In the PN process, N2O formation generally requires NO2 – as a critical substrate. Accordingly, lowering the NO2 – concentration in the PN reactor is expected to reduce N2O production. This consideration may also help explain why one-stage PNA often exhibits lower N2O emissions than two-stage PNA, because in one-stage PNA, NO2 – is consumed more completely, thereby limiting substrate availability for N2O formation. Based on this rationale, in two-stage PNA, recycling the nitrite-free anammox effluent back to the PN reactor to dilute nitrite concentration may serve as an effective strategy to further reduce N2O emissions under low bCOD/NH4 +–N conditions. Supplementary Material