Researchers began to investigate the impact of the C/N ratio on N₂O production during BNR as early as around 2000. At first, the influent of urban wastewater was simulated at a laboratory scale to measure N₂O. During the denitrification process of the mixed flow reactor, when the COD/NO₃ ⁻-N ratio was 1.5 or 2.5, 3%–12% of the influent NO₃ ⁻-N was converted into N₂O. When the COD/NO₃ ⁻-N ratio was 3.5 or 4.5, however, the N₂O conversion rate dropped below 4% (Table 1). Additionally, an insufficient sludge retention time (SRT) was also likely to enhance N₂O emissions (Hanaki et al., 1992). Studies of cyclic operation demonstrate that for intermittent aeration wastewater treatment systems, the maximum N₂O emission rate occurs during the initial aerobic stage, not the anaerobic stage (Park et al., 2000).

For the intermittent aeration BNR process of high concentration wastewater, in a bioreactor with an influent COD/N ratio less than 3.5, 20%–30% of the influent nitrogen is discharged in the form of N₂O; However, with the increase of the C/N ratio to 5.0–5.5, the N₂O conversion rate dropped below 1% (Itokawa et al., 2001). The nitrification/denitrification process of aquaculture wastewater also has a similar trend. When the C/N ratio of wastewater increased from 2.6 to 4.5, the conversion rate of N₂O to influent nitrogen decreased from 17.7% to 1.71% (Kishida et al., 2004).

With the advancements of BNR process, the hydraulic retention time (HRT) for nitrification and denitrification has progressively decreased from a maximum of 10 days to less than 1 day. In the complete nitrification process carried by the A/O SBR reactor, when the influent C/N increased from 7.5 to 14.5, the removal rates of TN and NH₄ ⁺slightly increased, but the conversion rate of N₂O (N₂O-N/TN _removed) decreased from 6.0% to 1.3% (Hu et al., 2013b). Researchers have also explored a broader range of C/N ratios. When the C/N ratios of synthetic wastewater were 6.5, 3.3, and 1.3, respectively, N₂O accounts for 0.32%, 88.57%, and 16.97% of the nitrogen loss (TN _loss) respectively (Yang et al., 2021), indicating that N₂O emissions do not increase monotonicity as the C/N ratio of influent water decreases, and further in-depth mechanism analysis may be needed to explain the tendency of wastewater with low C/N ratios to have higher N₂O or overall GHG emissions. For the nitrification and denitrification separation system, the ratio of N₂O production rate to NO₂ ⁻ accumulation rate is defined as the N₂O conversion rate (rN₂O-N/rNO₂ ⁻-N). When the C/N ratio was 4 and 1, the N₂O conversion rate was 11.98% and 22.63%, respectively (Wu et al., 2014). The N₂O emission factor during the nitrification stage ranged from 0.24% to 0.78%, while that of the denitrification stage decreased as the C/N ratio increased, spanning from 12.0% to 26.6%. According to the research findings of other independent nitrification and denitrification processes (Hanaki et al., 1992; Alinsafi et al., 2008; Lee et al., 2019), it can also be observed that the N₂O emissions from the denitrification process are generally greater than those from the nitrification process. Different studies have different nitrogen indicators compared to the generated N₂O when defining the N₂O conversion rate for different processes. Generally, influent nitrogen content (TN _load) and nitrogen removal (TN _removed or TN _loss) are taken as reference. Considering that most studies employ different statistical units for N₂O emissions and emission rates, the N₂O conversion rate can serve as a relatively unified quantitative standard for N₂O emissions in BNR processes. However, when comparing emissions across different studies, it is necessary to take into account the variations in the conversion rate based on nitrogen indicators for scientifically accurate comparisons.

In addition, monitoring the dynamic changes in N₂O emission concentration at different stages of the nitrogen removal process and paying attention to emission peaks can provide insights into the emission situation in wastewater treatment with different C/N ratios. In the process of nitrite denitrification, when COD/N was 6, the cumulative amount and duration of N₂O were significantly lower than that when COD/N is 1 or 4. The peak N₂O concentrations at COD/N ratios of 6 and 1 were 1.976 and 9.028 mg/L, respectively (Wang et al., 2019). These values can serve as a basis for characterizing emission characteristics.

Recently, N₂O emissions during BNR of actual wastewater with different C/N ratios have also been monitored. Continuous monitoring of N₂O emissions from SBR reactors in sewage treatment plants for 1 year revealed a significant positive correlation between N₂O emissions and influent COD/N (R ² = 0.346, p = 0.044 < 0.05), except for August and September. A lower influent COD/N (less than 6) corresponded to a higher N₂O emission (Sun et al., 2013). Similarly, in an A²/O bioreactor system, when the influent C/N decreased from 10.3/10.7 to 3.5/3.8, N₂O-N conversion rate increased from 0.043%–0.061% to 6.15%–9.18% (Ren et al., 2015). The research on A²/O system in 2017 also found that N₂O emission and generation and the total conversion rate of N₂O-N decreased significantly with the increase of influent C/N. And N₂O was mainly produced through the denitrification process in anaerobic and anoxic ponds (Yan et al., 2017) (Figure 1A).

In order to improve nitrogen removal efficiency and reduce resource input, partial nitrification/denitrification (PN-D) process (or shortcut biological nitrogen removal-denitrification, SBNR-D) came into being. The BNR process involves partially oxidizing ammonia to nitrite and directly reducing nitrite to nitrogen (N₂) without requiring complete oxidation through NO₃ ⁻-N (Schmidt et al., 2003). PN reduces the oxygen required by 25%, and the carbon source required for denitrification by 40% (Tseng et al., 1998; Pollice et al., 2002), making it an economical alternative to the complete nitrification/denitrification process. Due to the low dissolved oxygen (DO) concentration and high NO₂ ⁻ concentration increasing N₂O emissions in the PN process, which may attribute to heterotrophic denitrification. N₂O emission characteristics may differ from those of the complete nitrification process (Gabarro et al., 2014).

In 2006, researchers measured N₂O emissions from laboratory scale SBNR processes. For wastewater with a C/N ratio of 1, the process was able to remove 81.2% of TN. And its N₂O is mainly produced in the anoxic denitrification section, but the emissions were over 90% lower than that of the complete nitrification process under the same conditions (Hwang et al., 2006). For the UASB reactor with a COD/NH₄ ⁺ ratio of 1.6 treating the effluent of concentrated black water, PN exhibits a similar effect, with the N₂O conversion rate accounting for 1.9% of the TN _load (de Graaff et al., 2010). In the continuous aeration PN process, for the wastewater with a C/N ratio of 6, the N₂O emissions were found to range from 18.67 to 330.09 μg/h, and the conversion rate (N₂O-N/NH₄ ⁺-N) was relatively low (0.42%) (Liu et al., 2021a), which also reflected the advantages of PN-D in reducing N₂O emissions during wastewater treatment.

In recent years, in addition to complete nitrification/denitrification and PN-D to improve efficiency, several new BNR processes have been developed, such as Completely Autotrophic Nitrogen removal Over Nitrite (CANON) process, Coupled Aerobic-anoxic Nitrous Decomposition Operation (CANDO), Anammox related process, n-DAMO (Nitrate-dependent anaerobic methane oxidation) process, etc. These new processes minimize the demand for organic carbon and the oxygen required for nitrification (Horstmeyer et al., 2017), making them more suitable for treating wastewater with low C/N ratios.

In the BNR process dominated by anammox, the enrichment of anammox bacteria by anaerobic carrier biofilm enhanced the denitrification efficiency under C/N = 2.7–5 (Li et al., 2019). However, A²/O process combined with denitrification and anammox increased the denitrification efficiency of actual wastewater with low COD/N by about 16.9%. With enhanced denitrification, N₂O emissions remained relatively low. In the A²/O process combined with anoxic carrier biofilm (AM-A²/O), the actual N₂O conversion rate for wastewater with C/N ratios ranging from 1.2 to 7.9 was only 0.02%–0.08% (Li et al., 2019). The maximum N₂O emission during sewage treatment with low COD/N (2.7 ± 0.4) was only 0.15 mg/L (Li J W et al., 2020). Recently, researchers have combined nitrification, PN and anammox into a comprehensive process, called INPDA (integrated nitrification, partial denitrification and anammox). The TN removal rate of this process could reach 94.1%, and the N₂O conversion rate (N₂O-N/TN _load) for treating wastewater with a C/N of 2.5 is calculated to be 2.2% (Zhou et al., 2020). Compared with previous processes in A²/O or A/O sequencing batch biofilm reactors (SBBR), the above combined processes have lower N₂O emissions when treating wastewater with low C/N.

For the CANDO process that focuses on eliminating NO₂ ⁻, when the C/N ratio was 3, the N₂O yield (N₂O-N/NO₂ ⁻-N) was 62.5%, and when the COD/N ratio was 5, the N₂O yield decreased to 60.8%, accompanied by an increase in the nitrogen removal rate (Weissbach et al., 2018). In the unipolar CANON system, a similar pattern was observed. When the C/N ratio decreased from 1 to 0, the total N₂O emissions increased from 1.32% to 1.62%. Comparisons reveal that carbon source has no significant impact on N₂O emissions during hydroxylamine (NH₂OH) oxidation and heterotrophic denitrification (HD). The enhancement of nitrifying bacteria denitrification in wastewater with lower C/N ratio is the primary reason for the rise in total N₂O emissions in this process (Fang et al., 2020).

Summarizing the N₂O emissions from BNR in different processes, it can be seen that the processes of low C/N wastewater treatment tend to result in higher N₂O production, whereas new processes like anammox combined with PN or PD and treatment systems with added biofilm exhibit lower N₂O conversion rates under low C/N conditions.

Microbial-mediated denitrification plays a crucial role in the denitrification of wastewater. Nitrate (NO₃ ⁻) and nitrite (NO₂ ⁻) are reduced to harmless nitrogen gas (N₂). This process is usually mediated by heterotrophic bacteria (HB), including four steps mediated by nitrate reductase (NAR), nitrite reductase (NIR), nitric oxide reductase (NOR) and nitrous oxide reductase (N₂OR) respectively (Zumft, 1997). HB utilize carbon sources as electron donors. Specifically, after decomposing the carbon source in the environment, these denitrification bacteria can store the carbon source in the form of polyhydroxyalkanoates (PHAs) and glycogen in the anaerobic stage, as electron donor for endogenous denitrification, used for biomass production and anaerobic reduction of nitrogen oxides, and generate energy (Jetten et al., 1997; Oehmen et al., 2005; Miao et al., 2015).

When the denitrification process is not in optimal balance, the intermediate gaseous products nitric oxide (NO) and nitrous oxide (N₂O) may be discharged into the environment. Among them, N₂O is generated through the sequential action of NO₃ ⁻, NO₂ ⁻, and NO reductase (Richardson et al., 2009) (Figure 2). Taking Alcaligenes faecalis, a denitrification culture, as an example, N₂O emissions are higher in the absence of carbon sources; When the electron donor is increased due to the addition of carbon source, the output of N₂ increases, but the production of N₂O does not increase (Schalk-Otte et al., 2000), which proves that HB may not be able to complete the denitrification process under the condition of insufficient carbon source, and incomplete denitrification makes a large amount of intermediate N₂O produced. However, the inclusion of a carbon source facilitates complete denitrification and the generation of the final product, N₂. It was observed in batch experiments that N₂O accumulated at C/N ratios of 1.28 and 2.57, while complete denitrification occurred at C/N ratios of 5.14 and 12.85 (Lee et al., 2019).

In the process of denitrification, when the electron supply rate of the oxidation process cannot meet the demand for electrons of the four reduction steps, electronic competition will occur between the four reduction steps (Richardson et al., 2009), so the availability of electrons can adjust the activities of various denitrification enzymes. In research, PHAs are commonly used as the primary endogenous electron donor, serving as a bridge between the external carbon source and the electron utilization of denitrification enzymes. The electronic availability in the system can be known by monitoring the PHAs content, and the PHAs will increase with the increase of influent C/N, which will alleviate the competition of each denitrification process for carbon to some extent (Ge et al., 2018).

Specifically, N₂OR competes with NOR and NIR for electrons. An additional electron supply can enhance the electron pool of cytochrome C, thereby reducing competitive pressure and boosting N₂OR activity (Schalk-Otte et al., 2000). Therefore, under different carbon load conditions, it may be the electronic competition intensity between denitrification reductases, rather than the C/N ratio itself, that determines the N₂O accumulation in the process of denitrification (Lu and Chandran, 2010).

Further experiments have demonstrated that electronic competition occurs not only under carbon-limiting conditions but also in situations with excessive carbon sources. This is because the various denitrification processes are interrelated. In the traditional concept, BNR may be carried out by a variety of organisms in sequence, for example, one denitrification bacterium converts nitrate into nitrite, and then another denitrification bacterium converts nitrite into nitrogen (Zumft, 1997). Is there no electronic competition between nitrate reductase and other reductases in subsequent steps of denitrification?

The current view is limited. The fact is that over 60% of the colonies in the denitrification enrichment belong to Sphingobacteriales and Flavobacteriales, and the bacteria of this order have complete nitrogen removal pathways (Caspi et al., 2012). All denitrification enzymes compete for electrons from a common electronic supply system. As the carbon loading rate changes, electrons will be differentially distributed among various reductases, and when the electron flux of reducing nitrite is greater than that of reducing N₂O, N₂O will accumulate (Pan et al., 2013a). In the same year, the research team developed an electron carrier model, indicating that carbon oxidation provides electrons for carriers, and nitrogen oxides receive electrons from these carriers for reduction. The carbon oxidation process and nitrogen reduction process are closely connected. This model enhanced the prediction of N₂O accumulation ability during denitrification by employing different affinity constants and reduction carriers to describe the relative competitive ability of electrons in each denitrification step (Pan et al., 2013b).

Nitrous oxide reductase (N₂OR) is the sole enzyme responsible for catalyzing the decomposition of N₂O into N₂, and the structural gene nosZ encoding N₂OR is co-transcribed with nosR. N₂OR is not only influenced by the electronic competitive activity of other denitrification reductases but is also highly sensitive to DO in the environment. The anoxic and anaerobic enrichment conditions will also lead to N₂O accumulation (Conthe et al., 2018).

In addition to the contribution of HB, N₂O emission from BNR processes are partially attributed to ammonia oxidizing bacteria (AOB) and ammonia oxidizing archaea (AOA) during the nitrification process in the mixed system. Under low C/N conditions, it was found that the contributions of heterotrophic denitrification activity and autotrophic nitrification activity to N₂O production were similar. Therefore, some researchers have created a process model that includes both heterotrophic and autotrophic denitrification paths, resulting in a slightly better prediction effect for N₂O emissions compared to the single-path denitrification model (Domingo-Felez et al., 2017). In the process of AOB metabolism, the oxidation of NH₂OH produces a byproduct N₂O (Chandran et al., 2011); similarly to HB, it uses ammonia or hydrogen as the electron donor to reduce nitrite, resulting in the production of N₂O (Figure 3). This autotrophic ammonia oxidizing bacteria with denitrification ability is primarily classified as Nitrosomonas (Bock et al., 1995), and N₂O is the end product of its nitrogen metabolism (Kim et al., 2010; Law et al., 2012). The model shows that in most cases, the AOB denitrification pathway is dominant, while the NH₂OH oxidation pathway becomes more significant at high DO levels (e.g., 3.5 mg O₂/L) (Peng et al., 2015a).

Unlike heterotrophic organisms, the nitrogen metabolism pathway of autotrophic AOB bacteria is affected by the availability of inorganic carbon (IC). In nitrifying sludge rich in AOB and nitrite oxidizing bacteria (NOB), autotrophic growth can lead to a lack of IC, which in turn limits the oxidation activity of AOB towards NH₃ (Todt and Dorsch, 2016). In the PN/A process, N₂O is also produced through pathways associated with AOB. Decreasing IC/N will inhibit the activity of AOB, thereby enhancing N₂O generation through NH₂OH oxidation (Ma et al., 2015; Li L et al., 2020) (Figure 2). Enhancing the availability of IC will result in a higher reaction rate between AOB and related N₂O (Peng et al., 2015b). However, further research is needed to elucidate the mechanism of N₂O generation under different IC conditions, and more evidence is required to support the IC/N concentration that can achieve N₂O emission reduction in actual wastewater treatment.

In the treatment of low C/N wastewater, certain specific conditions can cause an increase in N₂O production of AOB. If the nitrite concentration in the influent is high, it will promote the denitrification of AOB and effectively convert nitrite to N₂O (Colliver and Stephenson, 2000). However, the limited COD availability in low C/N wastewater will also cause nitrite accumulation (Hanaki et al., 1992), further enhancing the N₂O generation pathway of AOB denitrification. Other conditions, such as low pH, can cause nitrification stress in AOB, leading to protonation of NO₂ ⁻ into HNO₂. HNO₂ inhibits the activity of NOB in the mixed system, causing the accumulation of NO₂ ⁻, which in turn promotes N₂O production (Peng et al., 2015b). Additionally, nitrifying sludge is more likely to produce N₂O from AOB under anaerobic conditions. Experiments have shown that the N₂O generation rate reaches its maximum at a DO of 0.85 mg O₂/L, while the N₂O emission factor decreased with an increase in DO from 0.35 mg O₂/L to 3.5 mg O₂/L (Peng et al., 2015a).

In terms of metabolism and gene expression, the AOB denitrification pathway includes NIR that reduces NO₂ ⁻ to NO and NOR that reduces NO to N₂O (Kozlowski et al., 2016b). Studies on bacteria, such as eutrophic Nitrosomonas europaea, have found that AOB requires NOR activity to convert NO into N₂O during the processes of nitrification and denitrification, that is, it is not the nitrite reductase gene NirK that is necessary for N₂O production, but the nitric oxide reductase gene NorB (Kozlowski et al., 2014; Kozlowski et al., 2016a). For oligotrophic AOB and AOA lacking NOR activity, it is more important that abiotic reactions (chemical denitrification) convert NO into N₂O and discharge it in vitro (Kozlowski et al., 2016a). Recent experiments have proven that N₂O in PN with higher nitrogen removal efficiency for low C/N wastewater was also produced by mixing biological and abiotic nitrosation (Terada et al., 2017). During the aerobic ammonia oxidation process, extracellular NH₂OH undergoes a non-biological reaction with substances in the growth medium, which also serves as a pathway for N₂O conversion (Liu et al., 2017). Previous studies may have underestimated N₂O emissions caused by non-biological pathways.

The nitrification of AOB is achieved by the membrane-bound enzyme ammonia monooxygenase (AMO) oxidizing ammonia (NH₃) to produce NH₂OH, which is then mediated by the periplasmic enzyme hydroxylamine dehydrogenase (HAO) to produce nitrite. The acidic conditions mentioned earlier (pH < 5) can induce partial inhibition of HAO, and the released NO can be further reduced to N₂O, thereby enhancing the NH₂OH oxidation pathway of N₂O (Jiang and Bakken, 1999).

There remains a research gap regarding the N₂O emission contribution of AOA in BNR of WWTPs. It is known that both AOA and AOB increase in high ammonium states in soil ecosystems, while AOA dominates in low ammonium states (Hink et al., 2018). The advantages of AOA can be demonstrated under conditions of low ammonium, hypoxia, long SRT, and high temperature, indicating that AOA may promote the generation of N₂O to a certain extent under these conditions (Wu et al., 2020). However, currently only Castellano-Hinojosa et al. (2018) have reported a negative correlation between AOA abundance and N₂O emissions in four aerobic sludge wastewater treatment plants in Spain. Therefore, they believe that AOA is unlikely to make a significant contribution to N₂O generation.

For nitrogen removal systems with integrated functions of autotrophic and heterotrophic microorganisms, denitrification is the main contributor to N₂O emissions. For example, in the SBBR reactor for treating low C/N (=0∼1) wastewater, N₂O emissions from hydroxylamine oxidation, AOB denitrification, and heterotrophic denitrification accounted for 5.4∼7.6%, 45.2∼60.8% and 33.8∼47.2% of the total N₂O emissions, respectively. In reactors with varying C/N ratios, the contribution of denitrification to N₂O production ranged from 90% to 96%. And with the decreases of the C/N ratios, the total amount of N₂O in denitrification process increased (Domingo-Felez et al., 2017; Yan et al., 2019; Fang et al., 2020). However, NH₂OH oxidation and AOB denitrification pathways are primarily found in the PN process (Ni and Yuan, 2015). In the aerobic stage, N₂O generated by NH₂OH oxidation accounts for 65% of the total N₂O, and the N₂O generated by AOB denitrification in the later stage is nearly identical to that generated by NH₂OH oxidation (Rathnayake et al., 2013). In addition, N₂O emissions from non-biological pathways account for 1.1% of the TN _load (Soler-Jofra et al., 2016; Liu et al., 2018).

In the partial nitrification-anammox (PN/A) process, which enhances denitrification performance, anammox process also emits a small amount of N₂O in addition to the contribution of PN. The most probable N₂O emission path involves heterotrophic denitrification in anammox particles (Okabe et al., 2011). The research showed that in the primary PN/A particle reactor, 70% of N₂O emissions occurred in the aerobic surface area dominated by AOB, and 30% occurred in the anoxic area dominated by anammox, and NH₂OH oxidation and AOB denitrification had a similar proportion of contributions to N₂O emissions from AOB related pathways (Ali et al., 2016). In the full-size two-stage PN/A reactor, N₂O emissions in PN stage were high (1.2% of TN _load), and the emission source could be located in HD (Hausherr et al., 2022). Due to the fact that the majority of N₂O (approximately 97.5%) was emitted by PN units, the emissions of N₂O from anammox could be almost negligible (Desloover et al., 2011; Okabe et al., 2011) (Figure 1B). In addition, for the monopolar anaerobic nitrogen removal process dominated by Anammox, the batch experiment of ¹⁵N isotope tracing and specific inhibitors demonstrated that when nitrite was the main nitrogen source, the N₂O emissions of HD and AOB denitrification were 64% and 36%, respectively (Li et al., 2017).

Under aerobic conditions, little research is available on enzyme-based N₂O metabolism. The primary metabolic pathway of N₂O is HD, with the majority of escaping N₂O originating from AOB. This indicates that the rate of N₂O consumption by HB is lower than that of N₂O production by AOB, which is the main reason for N₂O emission from aerobic phase (Yang et al., 2021). In SBR with intermittent aeration, it is also found that AOB denitrification is the main way to produce N₂O, and HD is the sink of N₂O (Liu et al., 2021a).

Due to the unclear correlation between microbial community abundance and N₂O generation contribution, it is challenging to make a unified and clear judgment on the N₂O generation mechanism under different denitrification processes for urban wastewater with low C/N ratios. Future research needs to start from multiple factors such as microbial community structure, N₂O metabolic characteristics, enzyme activity, and more in-depth consideration of various N₂O production pathways and their representative microbial contributions, in order to lay a solid scientific foundation for proposing N₂O emission reduction strategies for low C/N wastewater treatment. In terms of methodological applications, emerging single cell metabolic phenotypes can be considered, whose single cell Raman spectroscopy (SCRS) can identify and select individual cells or functional bacterial populations (Jones et al., 2019; Liu et al., 2022), enabling in situ functional analysis. The key relationship between phenotypic heterogeneity and plasticity of PAO populations in enhanced biological phosphorus removal (EBPR) systems and the stability of the EBPR process has been investigated using SCRS, which cannot be solely determined through phylogenetic analysis (Li et al., 2018). Practical experience has shown that SCRS can be widely applied in the fields of biology and environment, addressing key limitations related to omics-centered environmental ecological research methods, such as a lack of cell-level resolution and limited capacity to infer gene functional relationships, especially for highly diverse and featureless ecosystems (Wang et al., 2020). If relevant cutting-edge technologies are employed to directly correspond different bacterial communities and their N₂O production and reduction functions, it could provide a better understanding of the N₂O metabolism mechanism in BNR systems.

For the denitrification process in SBR reactors, the N₂O production characteristics of microbial systems established on different carbon sources vary. Lack of oxygen inhibits the production of NAR in the methanol carbon source system, leading to reduced N₂O production in the anoxic and subsequent aerobic stages. However, in a carbon source system where ethanol is used, NAR is more resistant to oxygen limitations. The downstream N₂OR is more sensitive to hypoxia and inhibition than other denitrification enzymes (Korner and Zumft, 1989), which makes N₂O generated and accumulated in the subsequent aerobic stage. The above differences may be related to the distinct composition of microbial communities established by methanol and ethanol. Therefore, in the practical operation of WWTPs, strict control of ethanol addition to the anoxic zone is required to minimize the production and discharge of N₂O in the downstream aerobic zone. For the modified Ludzak Ettinger (MLE) processes dominated by denitrification, changing the carbon source from methanol to acetate can reduce the N₂O conversion rate from 3.0% to 1.0%, and the N₂O reduction rate of acetate biomass is higher than that of methanol biomass (Table 2). The disparity in N₂O reduction rate could be attributed to different bacterial communities enriched with different carbon sources (Song et al., 2015). Then some scholars used methanol, sodium acetate and glucose as external carbon sources to optimize C/N of denitrification biofilter. Through comprehensive comparison, it is also confirmed that sodium acetate is more suitable as an external carbon source (Xu et al., 2018).

Compared with one-time dosing, incremental addition of carbon sources enhanced nitrogen removal efficiency and could serve as a strategy for highly automated wastewater treatment systems to reduce N₂O emissions (Chai et al., 2019). Furthermore, within the cyclic activated sludge system (CASS), consistent feeding can improve denitrification efficiency and decrease N₂O emissions as compared to intermittent feeding. The carbon source in the continuous feeding alleviates the electronic competition between denitrification reductases in the non-aeration stage (Richardson et al., 2009). From a microbial community perspective, the continuous feeding system exhibits a high abundance of N₂O reducing bacteria within the denitrification bacteria (Liang et al., 2015). There are significant differences in N₂O emission reduction strategies among the different processes mentioned above, therefore, when designing emission reduction plans, it is necessary to optimize the design by categorizing the processes.

This indicates that the substitution and addition strategy of additional carbon sources can improve the C/N ratio of wastewater, thereby enhancing denitrification efficiency while reducing N₂O. It is a straightforward and more cost-effective strategy for reducing emissions in WWTPs.

In the combined denitrification process of CANON and denitrification, an appropriate carbon source (C/N = 1) could decrease the total N₂O production by 16.7% compared to C/N = 0. This was due to the inhibition of the AOB denitrification process that consumed NO₂ ⁻-N, which accounted for over 94.5% of N₂O emissions (Yan et al., 2019). Based on the N₂O emissions of wastewater with low C/N from various processes in Chapter 2, there are significant differences in carbon source dosage and C/N ratio for different processes to achieve emission reduction effects (Figure 4A). For the CANON system, primarily composed of autotrophic microorganisms, adding a small amount of carbon source with C/N ratio of 1 could achieve a significant reduction in N₂O emission. If a large amount of carbon source (C/N > 1) is added, the heterotrophic NOB activity cannot be inhibited, disrupting the stability of the system (Zhang et al., 2017; Yan et al., 2019). Specifically, due to the predominance of autotrophic denitrification bacteria in this process, the main carbon source required is IC. Considering the weak inhibitory effect of IC/N on AOB activity and the strong inhibitory effect on anammox activity, some studies have shown that using influent with a C/N range of 1.2 to 1.5–2.0 in CANON can achieve stable denitrification effects (Zhang X J et al., 2016; Yue et al., 2018). This indirectly indicates that the CANON process is suitable for treating low C/N wastewater, but the relationship between N₂O emissions and IC concentration requires further exploration.

In the CANDO that also eliminates NO₂ ⁻, when the COD/N ratio increases to 5, the N₂O yield decreases and the nitrogen removal rate increases compared to C/N = 3 or 4 (Weissbach et al., 2018). Considering the reduction of nitrogen removal efficiency, enzyme activity, and N₂O emissions, it is more feasible to maintain the influent C/N of A/O SBR at around 6.5 (Yang et al., 2021). Similarly, in another anoxic aerobic BNR experiment, step feeding and additional carbon sources allow AOB (Nitrosomonas) to mitigate the denitrification effect of nitrifying bacteria, thereby reducing the N₂O conversion rate by 66.6% and 12.0%, respectively, indicating that these two methods are effective in reducing N₂O emissions during sewage treatment (Hu et al., 2013b).

In addition, for the system that simultaneously removes nitrogen and phosphorus, setting the C/N ratio to 4 will result in a minimum N₂O conversion rate and an ideal nitrogen removal efficiency of 98.3%, but only 27.44% phosphorus can be removed. The phosphorus removal efficiency will reach the optimal value of 82.79% when C/N = 3. This also inspires future research on actual wastewater denitrification to consider the synergistic effects of multiple pollutants removal and N₂O emission reduction (Ge et al., 2018).

In the BNR process, the optimization of operating parameters is the key to reducing N₂O emissions. In the treatment of low C/N (=6) wastewater, compared to continuous aeration, intermittent aeration had a higher TN removal efficiency (93.5% on average), and the N₂O emission factor decreased from 0.42% of continuous aeration to 0.19%, which promoted PN-D. Among them, the complete ammonia oxidizer (comammox) was significantly enriched during intermittent aeration, and the quantitative results showed that their gene abundance reached 24.7%; The abundance of AOB bacteria significantly decreased (Liu et al., 2021a). Comammox lacks NO reductase, and non-biotransformation emits much less N₂O than AOB, showcasing its potential for reducing N₂O emissions (Liu et al., 2017; Kits et al., 2019). Conversely, during the operation of the PN/A process, a transition from low aeration (or hypoxia) to high aeration will quickly increase the N₂O emission rate, while increased continuous aeration will reduce the emission rate, proving that the continuous aeration strategy is effective in reducing N₂O emissions in PN/A (Figure 4B) (Castro-Barros et al., 2015).

In the A²/O process, for actual domestic wastewater with low C/N (=4.4), when the internal circulation ratio of the process is reduced from 300% to 100%, the production of N₂O increases from 9.81 × 10⁻² mg/L reduced to 3.47 × 10⁻² mg/L (Figure 4C). The primary reduction occurs in the form of N₂O produced by denitrification in anoxic section. This phenomenon is due to the reduction of the volume of internal circulating liquid and the reduction of nitrate substrate and oxygen that can be used for denitrification. As mentioned in Chapter 4.1, N₂OR is more sensitive to oxygen, thus, enhancing the activity of N₂OR entering the anoxic zone, making it more conducive to the denitrification process of reducing N₂O to N₂. Experimental evidence indicated that the copy number of nosZ gene increased as the internal circulation ratio decreased from 300% to 100% (Yan et al., 2016).

In addition, electrical stimulation can optimize microbial population structure and enhance microbial autotrophic denitrification (Figure 4D). At the same time, the activity of NAR and NIR is increased to promote denitrification, so as to improve the removal efficiency of NO₃ ⁻ and reduce the accumulation of N₂O. The removal rate of nitrate and TN can be increased by 20%, and the production of intermediate greenhouse gas N₂O can be reduced by 62.6% when weak electric stimulation (0.2V) is added to the denitrification process of influent C/N = 3 (Dong et al., 2022).

This research suggests that future N₂O emission reduction strategies can be developed based on N₂O metabolism mechanisms, such as identifying operating conditions that promote N₂OR activity or nosZ transcription initiation.

At present, more studies on BNR and N₂O emission reduction only investigate the effects of individual parameters on emissions, making it difficult to compare the specific effects and overall impacts of various operating conditions. In light of this limitation, researchers have employed Plackett Burman (PB) multi-factor experimental design and response surface methodology (RSM) to explore emission reduction strategies for N₂O in nitrifying granular sludge systems. The analysis revealed that at a temperature of 22.3°C, the pH value of 7.1, and the aeration rate of 0.20 m³/h, the N₂O emission during the denitrification process is minimal. The predicted results were confirmed using wastewater with COD/NH₄ ⁺ = 5/3 (Liu et al., 2016).