Simultaneous Heterotrophic Nitrification and Aerobic Denitrification of High C/N Wastewater in a Sequencing Batch Reactor

Abstract

Heterotrophic nitrification and aerobic denitrification (HN–AD) is an emerging biological process capable of achieving efficient nitrogen removal in a single reactor. This study investigates the HN–AD performance of a sequencing batch reactor (SBR) operated with a simple anaerobic–aerobic cycle for treating high C/N wastewater. Over a 220-day operation, the system achieved average removal efficiencies of 98.6% for COD, 93.3% for NH₄⁺-N, and 87.1% for total nitrogen. Effluent concentrations of NO₂⁻-N and NO₃⁻-N remained negligible at the end of each aerobic phase. Concentration profiles of NH₄⁺-N, NO₂⁻-N, and NO₃⁻-N throughout the operation cycles confirmed the occurrence of simultaneous nitrification and aerobic denitrification. The consistently high COD removal and robust nitrogen reduction highlight the stability of the HN–AD microbial consortia enriched from activated sludge. Phosphorus removal (average removal efficiency 66.3%) may be enhanced by increasing the activity of phosphate-accumulating organisms (PAOs) through process optimization. This study demonstrated effective HN–AD using activated sludge in SBRs. Future work will focus on evaluating the system with real wastewater and continuous-flow setups to further refine operational parameters for sustained HN–AD performance.

1. Introduction

Simultaneous nitrification and denitrification (SND) represents an advanced biological approach for nitrogen removal from wastewater. In conventional pathways, nitrification begins with the oxidation of ammonium (NH₄⁺-N) to nitrite (NO₂⁻-N) by ammonia-oxidizing bacteria, followed by oxidation to nitrate (NO₃⁻-N) by nitrite-oxidizing bacteria under aerobic conditions. Denitrification then occurs as heterotrophic bacteria reduce NO₂⁻ and NO₃⁻ to nitrogen gas under anoxic conditions, using organic compounds as electron donors. Most SND studies focus on conventional autotrophic nitrification and denitrification, typically achieving SND by controlling dissolved oxygen (DO) levels or creating oxygen gradients in biofilms and flocs to establish microaerobic and anoxic environments.

Comparatively fewer studies have explored heterotrophic nitrification–aerobic denitrification (HN–AD), a distinct process in which heterotrophic bacteria assimilate nitrogen and oxidize NH₄⁺, followed by the aerobic reduction of hydroxylamine, nitrite, and/or nitrate to nitrogen gas [1,2,3]. Research indicates that HN–AD can achieve stable and efficient nitrogen removal under aerobic conditions in a single reactor via both cell assimilation and nitrification/denitrification [1]. Numerous HN–AD-capable bacteria have been identified from sources such as activated sludge, landfill leachate, surface water, and marine sediments [1]. Many HN–AD strains demonstrate high efficiency in nitrogen removal. For instance, Bacillus sp. L2, isolated by Li et al. [4], achieved 98.37% nitrogen removal from acetate wastewater at a pH of 9 and C/N = 9. Deng et al. [5] reported that Pseudomonas sp. DM02 removed 10 mg/L ammonium-N in 12 h, with 70.8% converted to nitrogen gas and 28.1% to intracellular nitrogen. Xia et al. [6] showed that Acinetobacter sp.ND7 removed 99.8% NH₄⁺-N, 96.2% NO₂⁻-N, and 97.18% NO₃⁻-N at a C/N ratio of 8. Additionally, Dadrasnia et al. [7] demonstrated that Bacillus salmalaya achieved 78% ammonia-N removal, 88% BOD removal, and 91.4% COD removal from landfill leachate.

Several factors influence HN–AD performance, including carbon source type, carbon-to-nitrogen ratio (C/N), dissolved oxygen levels, pH, and temperature [1,2]. Common carbon sources tested include acetate, succinate, citrate, and glucose, with acetate being the most effective for many strains [1]. C/N ratios from 5 to 25 are typically used, with optimal values ranging between 8 and 16 depending on the microbial strain [6,8]. Most studies apply a pH range of 7–8 and maintain temperatures around 30 °C.

While many studies confirm the ability of HN–AD bacteria to perform SND aerobically, most focus on the HN–AD performance of isolated HN–AD bacteria strains [1,9,10]. The main objective of this study is to evaluate whether HN–AD nitrogen removal can be achieved in a sequencing batch reactor (SBR) seeded with activated sludge and treating high C/N wastewater. The effect of the C/N ratio on HN–AD performance was investigated by monitoring the removal efficiencies of COD, NH₄⁺-N, NO₃⁻-N, NO₂⁻-N, total nitrogen, and phosphate during a 220-day operational period.

2. Materials and Methods

2.1. Wastewater Characteristics

Synthetic wastewater that mimicked medium-strength wastewater discharged from a local brewery was used in this study. The compositions of the synthetic wastewater were adjusted based on varying COD/N/P ratios during Phases I to III over 220 days of operation (

Table 1. Compositions of synthetic wastewater and trace element solution.

Chemicals	Concentration (mg/L)
Organics and nutrient	Phase I	Phase II	Phase III
Sodium acetate	3080.0	3080.0	3080.0
NH4Cl	200.0	400.0	200.0
KH2PO4	33.0	33.0	33.0
K2HPO4	42	42	21.0
CaCl2·H2O	80	80	80
MgSO4·7H2O	421.5	421.5	421.5
NaHCO3	300	300	300
Trace element solution	Concentration (mg/L)
FeCl2·4H2O	2000
MnCl2·4H2O	500
CoCl2·6H2O	2000
NiCl2·6H2O	142
ZnCl2	50
Na2SeO3	123
AlCl3·6H2O	90
CuCl2·2H2O	38
H3BO3	50
HCl	1 mL (36%)
EDTA	1000
(NH4)6Mo7O24·H2O	50

).

In Phase I (Days 1 to 179), the synthetic wastewater contained (per liter): 3080 mg sodium acetate; 200 mg NH₄Cl; 32.5 mg KH₂PO₄; 42 mg K₂HPO₄; 80 mg CaCl₂·H₂O; 421.5 mg MgSO₄·7H₂O; and 0.1 mL of trace nutrient solution. The concentrations of COD, NH₄⁺-N, PO₄³⁻-P, Ca²⁺, and Mg²⁺ were 3100 ± 31.5, 49 ± 3.6, 14.6 ± 0.9, 28.8 ± 1.2, and 84 ± 0.9 mg/L, respectively. The COD/N/P ratio was 220/3.5/1, corresponding to a C/N/P mass ratio of 60.2/3.5/1 (C/N = 17.1). In Phase II (Days 180 to 208), the NH₄⁺-N concentrations were adjusted from 49.0 ± 3.6 to 98.5 ± 7.7 mg/L to achieve a C/N/P ratio of 60.2/7/1 (C/N = 8.6). The COD concentration remained unchanged and the PO₄³⁻-P concentration was 15.5 ± 2.1 mg/L. The remaining components of the synthetic wastewater were the same as those in Phase I. In Phase III (Days 209 to 219), the C/N/P ratio was adjusted to 120.4/7/1 (C/N = 17.1). The NH₄⁺-N and PO₄³⁻-P concentrations were reduced to 48.2 ± 3.9 mg/L and 7.2 ± 0.8 mg/L, respectively. The COD concentration remained unchanged. Throughout all three phases, 300 mg/L of NaHCO₃ was added to the synthetic wastewater to maintain steady alkalinity and pH conditions.

Trace elements are essential for enzymatic reactions in microbial cells and bacterial growth [11]. The composition of the trace element solution used in this study is shown in Table 1 and was prepared as described by Scampini [11]. All chemicals were supplied by Fisher Scientific (Mississauga, ON, Canada).

2.2. Experimental Set-Up and Operation

Figure 1 shows the lab-scale SBR used in this study [12]. The reactor consisted of a cylindrical column with a working volume of 4 L, equipped with an overhead analog mechanical stirrer (Cole-Parmer, Montreal, QC, Canada), a Marina 300 air pump (Hagen Marina, Montreal, QC, Canada), and a Masterflex L/S peristaltic pump (Cole-Parmer, Montreal, QC, Canada). The wastewater pump, air pump, and a Masterflex solenoid-operated two-way pinch valve (Cole-Parmer, Montreal, QC, Canada) were connected to timer plugs (GE, Shelburne, ON, Canada) for cycle time control. A DLXB pH/ORP pump control system (Etatron, Cole-Parmer, Montreal, QC, Canada) was used for real-time pH regulation [12].

The lab-scale SBR system was operated for 220 days. Synthetic wastewater was fed into the reactor from the bottom via the peristaltic pump during the feeding period. The volume exchange ratio was maintained at 20% from Day 1 to Day 131 and increased to 50% from Day 132 to the end of the operation. The mechanical stirrer was set to 150 rpm to ensure complete mixing of the wastewater and sludge. The airflow rate was maintained at 2.36 L/min using a Cole-Parmer flowmeter, resulting in a superficial air velocity of 0.01 m/s. DO and pH were monitored every two days using Oakton meters (Oakton, Cole-Parmer, Montreal, QC, Canada). The solid retention time was controlled at 18 days. The reactor operated at ambient laboratory temperature, ranging from 19 to 23 °C [12].

The SBR was seeded with aerobic sludge obtained from the secondary clarifier at the Guelph Wastewater Treatment Plant (WWTP), with an initial mixed liquor suspended solids (MLSS) concentration of approximately 2000 mg/L. Each day, 200 mL of MLSS was withdrawn from the system.

The reactor was operated under alternating anaerobic and aerobic conditions in 8 h cycles. Each cycle included a 120 min anaerobic phase with 50 min of feeding, a 240 min aerobic phase, a settling period of 10 min, a 10 min withdrawal phase, and an idle period for the remainder of the cycle. Effluent was discharged through a 10 mm OD pipe controlled by a solenoid-operated two-way pinch valve and a time-sequencing timer. The pH was maintained around 7.50 using the DLXB pH/ORP control system by dosing 0.1 M HCl into the reactor.

2.3. Cycle Test

To monitor the concentration profiles of chemical oxygen demand (COD), nitrogen species, and phosphorus during the anaerobic/aerobic cycle of the SBR, cycle tests on Days 138 and 161 were conducted during Phases I. During each 8 h cycle, which consisted of 2 h anaerobic and 6 h aeration periods, 50 mL of mixed liquor was sampled from the effluent outlet every 30 min. Samples were filtered through 0.22 μm nylon syringe filters (Cytiva Whatman™, Marlborough, MA, USA) and analyzed for COD, VFA, NH₄⁺-N, NO₃⁻-N, NO₂⁻-N, total nitrogen (TN), PO₄³⁻-P, and pH [13].

2.4. Analytical Methods

The synthetic wastewater, effluent, and samples from the SBR were filtered through 0.22 μm nylon syringe filters (Cytiva Whatman™, Marlborough, MA, USA) for the analysis of COD, VFA, NH₄⁺-N, NO₃⁻-N, NO₂⁻-N, total nitrogen (TN), PO₄³⁻-P, and total phosphorus (TP), using HACH test kits in accordance with the APHA Standard Methods [13].

2.5. Statistic Analysis

Differences in effluent concentrations of COD, NH₄⁺–N, TN, and PO₄³⁻–P, as well as the corresponding removal efficiencies, among operational phases were assessed using Welch’s one-way analysis of variance (ANOVA), which is robust to unequal variances and sample sizes. When the overall ANOVA indicated significant differences, pairwise comparisons were performed using the Games–Howell post hoc test, which accounts for unequal variances and does not assume equal sample sizes. Statistical significance was evaluated at the 0.05 level. All statistical analyses were conducted using IBM SPSS Statistics (Version 29.0.2.0(20)).

3. Results

3.1. COD and VFA Removal from Phase I to III

This study was conducted in three operational phases, during which the COD removal efficiency consistently remained above 98%. In Phase I (Days 1–179), the influent concentrations of COD, NH₄⁺-N, and PO₄³⁻-P were 3100 ± 31.5, 49.0 ± 3.6, and 14.6 ± 0.9, respectively, corresponding to the C/N/P ratios of 60.2/3.5/1. The exchange ratio was 20% from Day 1 to Day 131 and 50% from Days 132 to 179, which corresponded to an average organic loading rate (OLR) of 1.86 kg/(m³·d) and 4.86 kg/(m³·d), respectively. The operational period was divided into Phase I (PI1: Days 0–29, PI2: Days 30–70, PI3: Days 71–123, and PI4: Days 124–180), Phase II (PII), and Phase III (PIII), based on variations in effluent quality. In Phase I (PI), effluent COD remained low in PI1 (28.9 ± 9.2 mg/L) and PI2 (28.7 ± 4.3 mg/L) but increased in PI3 (52.5 ± 7.0 mg/L) and PI4 (49.7 ± 5.1 mg/L) (

Table 2. COD, nitrogen, and phosphate removal for different time periods.

	Phase 1				Phase 2	Phase 3
Days	0–29	30–70	71–123	124–179	180–208	211–220
Removal (%)
COD	99.1 ± 0.3	99.1 ± 0.1	98.3 ± 0.2	98.4 ± 0.2	98.5 ± 0.1	98.5 ± 0.0
NH4+-N	70.3 ± 26.3	93.8 ± 0.0	99.9 ± 0.1	100 ± 0.0	90.1 ± 13.6	100 ± 0.0
TN	64.4 ± 29.1	88.9 ± 0.9	89.1 ± 5.5	94.9 ± 1.6	87.1 ± 14.2	92.9 ± 0.3
PO43−-P	47 ± 26.9	75.7 ± 15.5	65.6 ± 21.8	65.0 ± 11.0	65.0 ± 16.8	99.8 ± 0.1
Effluent concentrations (mg/L)
COD	28.9 ± 9.2	28.7 ± 4.3	52.5 ± 7.0	49.7 ± 5.1	47.7 ± 1.8	45.9 ± 0.3
NH4+-N	14.6 ± 12.9	3.0 ± 0.0	0.1 ± 0.1	0.0 ± 0.0	9.7 ± 13.4	0.0 ± 0.0
TN	17.5 ± 14.3	5.4 ± 0.4	5.7 ± 2.3	2.5 ± 0.8	12.7 ± 14.0	3.4 ± 0.2
PO43−-P	7.7 ± 3.9	3.6 ± 2.3	5.0 ± 3.2	5.1 ± 1.6	5.4 ± 2.6	0.0 ± 0.0
NOx−-N	0.1 ± 0.1	0.1 ± 0.0	0.1 ± 0.0	0.1 ± 0.0	0.1 ± 0.1	0.0 ± 0.0

). Overall, the average effluent COD concentration for Phase I was 41.9 ± 12.6 mg/L, resulting in an average COD removal efficiency of 98.6 ± 12.6%. No evident impact of the exchange ratio on COD removal was observed.

In Phase II (Days 180–208), while the influent COD concentration and OLR remained unchanged from Phase I, the influent NH₄⁺-N concentration increased from 49.0 ± 3.6 mg/L to 98.5 ± 7.7 mg/L. Despite this shift, the effluent COD concentration remained steady at 47.5 ± 1.8 mg/L, with a corresponding removal efficiency of 98.5 ± 0.1%.

In phase III (Days 209–120), the influent COD and NH₄⁺-N concentrations are the same as those in Phase I; however, the PO₄³⁻-P concentration was reduced to 7.2 ± 0.8 mg/L. The system maintained its COD removal performance, with an effluent COD concentration of 45.9 ± 0.3 mg/L, and a consistent removal efficiency of 98.5 ± 0.0%.

The removal trend for VFA closely mirrored that of COD. Effluent VFA concentrations were 20.0 ± 7.5 mg/L in Phase I, 21.4 ± 1.0 mg/L in Phase II, and 20.9 ± 1.4 mg/L in Phase III, with an overall average of 19.6 ± 5.0 mg/L. In all phases, VFA removal efficiency exceeded 98%. Sodium acetate served as the carbon source, making acetic acid the predominant VFA detected in the effluent. The effluent VFA/COD ratios were 48.2%, 45.0%, and 45.2% in Phases I, II, and III, respectively.

Welch’s ANOVA detected an overall difference among groups (p < 0.01). Games–Howell post hoc tests found significant pairwise differences in COD concentrations between PI1 and PI2 with other time periods (p values < 0.005) and between PI3 and phase III. The COD removal over the entire operation period was 98.6 ± 0.4%, demonstrating excellent COD removal efficiencies across all phases at OLRs of 1.86 kg/(m³·d) and 4.86 kg/(m³·d). These results suggest a stable and resilient microbial community with robust biomass activity throughout the entire operational period when the wastewater composition varied in the range tested.

3.2. Biological Nitrogen Removal from Phases I to III

Figure 2b,c shows the removal performance of NH₄⁺-N, TN, NO₂⁻-N, and NO₃⁻-N throughout the three operation phases. The effluent concentrations of NH₄⁺-N decreased rapidly—from 39.2 mg/L to below 0.5 mg/L—within the first 25 days. However, blockage of the aeration diffusers between Days 30 and 38 caused a temporary spike in the NH₄⁺-N level. Once the diffusers were replaced, NH₄⁺-N concentrations dropped to below 0.1 mg/L. As shown in Table 2, effluent NH₄⁺–N averaged 14.6 ± 12.9 mg/L (70.3 ± 26.3% removal) during PI1 and 3.0 ± 0.0 mg/L (93.8 ± 0.0% removal) during PI2, then decreased to <0.1 mg/L in PI3 (99.9 ± 0.1% removal) and PI4 (100.0 ± 0.0% removal).

The effluent concentrations of NO₂⁻-N and NO₃⁻-N remained below 0.10 mg/L and 0.15 mg/L, respectively, during phase I (Figure 2c). The highest NO₂⁻-N concentration (0.24 mg/L) was recorded during the diffuser blockage period (Days 30–38). Analysis of TKN and TN indicated that the majority of nitrogen remaining in the effluent during Phase I was in organic form, averaging 3.4 ± 2.0 mg/L. Despite fluctuations in NO₂⁻-N and NO₃⁻-N, the average NO_x⁻-N concentration remained around 0.1 mg/L during PI1 to PI4 (Table 2).

In Phase II (Days 180–208), the influent NH₄⁺-N concentration increased from 49.0 ± 3.6 mg/L (Phase I) to 98.5 ± 7.7 mg/L. This increase in the feed ammonium concentration led to a noticeable decline in nitrogen removal performance. As a result, the effluent NH₄⁺-N and TN concentrations were 9.7 ± 13.4 mg/L (90.1 ± 13.6% removal) and 12.7 ± 14.0 mg/L (87.1 ± 14.2% removal), respectively, in Phase II (Table 2). However, despite the increase in effluent NH₄⁺-N concentration, no significant accumulation of NO₂⁻-N or NO₃⁻-N was detected, suggesting that aerobic denitrification had occurred at a relatively high ammonium concentration.

In Phase III, system performance recovered as the influent NH₄⁺-N concentration decreased to 48.2 ± 3.9 mg/L. Effluent NH₄⁺-N levels remained below 0.1 mg/L, and TN averaged 3.4 ± 0.1 mg/L. NO₂⁻-N and NO₃⁻-N concentrations were minimal, at 0.02 mg/L and 0.01 mg/L, respectively.

Welch’s ANOVA detected an overall difference among groups of the time periods (p < 0.01); however, Games–Howell post hoc tests found only one significant pairwise difference—between PI4 and Phase III (p = 0.047). This pair had a small mean difference and a very small standard error, yielding a large test statistic and a low p-value. In contrast, the pairs that involve the high effluent NH₄⁺-N concentration period (PI1 and Phase II) have both larger mean differences and large standard errors, which reduced the test statistic and produced non-significant p-values.

In summary, 93 ± 15.2% of NH₄⁺-N and 87.1 ± 16.0% of TN were removed over Phase I to III. Given that the effluent was discharged right after the aerobic period, the low NO_x⁻-N levels in the effluent suggest that denitrification occurred efficiently during the aeration phase.

3.3. Biological Phosphorus Removal

Phosphorus removal efficiency showed the most variability across phases. As shown in Figure 2d, phosphorus removal performance fluctuated throughout the operation period. Over PI1 to PI4, the PO₄³⁻-P removal efficiencies were 47 ± 26.9%, 75.7 ± 15.5%, 65.6 ± 21.8%, and 65 ± 11%, respectively (Table 2). The average effluent PO₄³⁻-P concentration declined from 7.7 ± 3.9 mg/L in PI1 to 3.6 ± 2.3 mg/L in PI2, indicating a steady improvement in phosphorus removal efficiency. However, effluent PO₄³⁻-P concentrations increased to 5.0 ± 3.2 mg/L and 5.1 ± 1.6 mg/L in PI3 and PI4, respectively, due to operational disturbances (e.g., diffuser blockage and sludge mixing). These disruptions likely destabilized PAO activity, leading to temporary phosphorus release and accumulation in the effluent. The average PO₄³⁻-P removal efficiency during Phase I remained at 64.7 ± 20.1%.

In Phase II, the effluent PO₄³⁻-P concentration dropped from 9.6 mg/L on Day 180 to 4.6 mg/L on Day 195, then slightly increased to 5.0 mg/L on Day 204. The increased ammonium load in Phase II did not appear to disturb the phosphorus removal. From Day 196 to 208, phosphorus removal performance remained relatively stable, despite a noticeable decline in nitrification efficiency. The average phosphorus removal efficiency in Phase II was 65.0 ± 16.8%, with a mean PO₄³⁻-P concentration of 5.4 ± 2.6 mg/L.

The most notable phosphorus removal occurred in Phase III, where influent PO₄³⁻-P was halved and effluent concentrations dropped to 0.02 ± 0.01 mg/L, yielding removal efficiencies exceeding 99%. This sharp improvement likely resulted from a combination of reduced phosphorus loading and restored nitrification/denitrification balance, further facilitating PAO or denitrifying phosphorus accumulating organisms’ (DPAO)-mediated uptake. Welch’s ANOVA and Games–Howell post hoc tests found significant pairwise differences between phase III and other operation periods.

In conclusion, throughout all three phases, the system exhibited evidence of simultaneous aerobic denitrification and phosphorus uptake. The occurrence of phosphorus removal might involve both traditional aerobic uptake and DPAO-linked phosphorus removal.

3.4. Cycle Concentration Profiles

To analyze the concentration profiles of COD, NH₄⁺-N, NO₃⁻-N, NO₂⁻-N, and PO₄³⁻-P during operation cycles, cycle tests were conducted on Days 138 and 161. During the cycle tests, mixed liquor samples were collected every 30 min over a complete cycle to monitor the dynamic changes of the concentrations of the COD, NH₄⁺-N, NO₃⁻-N, NO₂⁻-N, and PO₄³⁻-P. The reactor’s feeding phase lasted for 50 min, with synthetic wastewater introduced from the bottom of the SBR. Samples were collected from the discharge port, located at 75% of the reactor height.

Figure 3a compares the COD and DO concentration profiles over the anaerobic and aeration phases during the Day 138 and Day 161 cycle tests. COD dynamics were similar in both cases: concentrations rose from ~60 mg/L, carried over from the previous cycle, to ~1126 mg/L by the end of the 2 h anaerobic period, then declined sharply after aeration began. COD fell below 100 mg/L within 2.5 h on Day 161 and within 4 h on Day 138.

In contrast, DO behavior differed substantially. On Day 138, DO increased from 0 to 7.9 mg/L within 1 h of aeration, whereas on Day 161 it required 4 h to reach 7.2 mg/L. Given that the operating conditions were identical, the slower rise on Day 161 was likely due to reduced oxygen transfer efficiency of the aeration diffuser.

Notably, despite the slower DO increase and lower final DO concentration, COD removal on Day 161 was rapid. This may reflect enhanced aerobic denitrification under the lower DO regime, which could have sustained high COD utilization rates even under oxygen-limited conditions.

Figure 3b illustrates changes in NH₄⁺-N. At the end of the previous cycle, NH₄⁺-N concentrations were 0.02 and 0.03 mg/L, respectively, in the Day 161 and Day 138 test. On the Day 161 test, NH₄⁺-N rose to around 17.1 mg/L following feed addition, and once aeration began, ammonium concentrations rapidly declined—dropping from 17.1 mg/L to 9.4 mg/L within 30 min, and further to 1.0 mg/L at 60 min. Similar NH₄⁺-N rise and decrease trend was observed on Day 138, but it took more than 1.5 h to reduce NH₄⁺-N to below 1.0 mg/L. The maximum SND rates observed in the two cycle tests were 8.62 and 11.39 mg NH₄⁺-N L⁻¹ h⁻¹ for the Day 138 and Day 161 tests, respectively, based on the actual NH₄⁺-N degradation times.

NO_x⁻-N responses to aeration initiation also differed between Days 138 and 161 (Figure 3c). On Day 138, concentrations increased during the first hour of aeration before declining sharply. On Day 161, they rose only slightly and remained between 0.16 and 0.09 mg/L throughout the aeration phase. These differences in NO_x⁻-N profiles and NH₄⁺-N degradation rates likely reflect the influence of DO concentration. Maintaining relatively low DO concentration, e.g., below ~4.4 mg/L, has been reported as critical for achieving complete HN–AD nitrogen removal [14].

Phosphorus concentration profiles are shown in Figure 3d. On Day 161, before feeding, the PO₄³⁻-P concentration was 5.3 mg/L, rising to 13.7 mg/L by the end of the anaerobic phase. Since the PO₄³⁻-P input from the feed alone could account for a rise to ~9 mg/L, the additional 4.7 mg/L is attributed to phosphorus release by PAOs and/or DPAOs. Once aeration began, the PO₄³⁻-P concentration steadily declined to 6.9 mg/L, corresponding to a phosphorus removal efficiency of 54.1%. The PO₄³⁻-P profile during the cycle on Day 138 was similar to that on Day 161; however, the PO₄³⁻-P concentration at the end of the anaerobic cycle was only 10.2 mg/L, which is only slightly higher than the concentration increase caused by feeding.

The observed P-release and P-uptake rates were 1.2 and 0.56 mg P/(g MLSS·h, respectively, with a P-uptake/release ratio of 1.42 based on Day 161 test result. In comparison, Kuba et al. [15] reported significantly higher rates, of 52 mg P/(g MLSS·h), for P-release and 24 mg P/(g MLSS·h) for P-uptake in an A²-SBR. The much lower rates observed in this study suggest a limited presence of PAOs in the activated sludge. Consequently, while aerobic phosphorus uptake was evident, the modest PO₄³⁻-P removal achieved in this study may be due to a low PAO abundance and activity.

4. Discussion

This study confirmed that simultaneous nitrogen removal via HN–AD can be achieved in an SBR seeded with activated sludge and treating high-C/N wastewater. In HN–AD, bacteria oxidize NH₄⁺-N to NO₂⁻/NO₃⁻ via hydroxylamine and subsequently reduce these intermediates to N₂ or N₂O under aerobic conditions, using organic carbon as the primary energy source and both oxygen and oxidized nitrogen as electron acceptors. This dual electron acceptor capability prevents NO₂⁻/NO₃⁻ accumulation, maintains redox balance, and supports growth under carbon-rich conditions [1,3]. Unlike partial nitrification–denitrification, which relies on sequential autotrophic and heterotrophic steps [16], anammox, which proceeds under strict anoxia using hydrazine as an intermediate [17], or microbial electrochemical denitrification, which requires an external electrical circuit [18], HN–AD achieves SND within a single heterotrophic population under fully aerobic conditions, offering a distinct physiological and operational advantage for the treatment of high-C/N wastewaters.

In this study, the SBR achieved excellent nitrogen removal via the HN–AD pathway in treating high-C/N wastewater. Across three operational phases, average removal efficiencies were 93 ± 15.2% for NH₄⁺-N and 87.1 ± 16.0% for TN. SND under aerobic conditions was confirmed by cycle tests that showed simultaneous nitrification and denitrification without significant NO₂⁻/NO₃⁻ accumulation upon initiation of aeration.

A marked decline in NH₄⁺-N removal occurred in Phase II when the C/N ratio decreased from 17.2 to 8.6 due to an increase in feed NH₄⁺-N concentration from 49.0 ± 3.6 to 98.5 ± 7.7 mg/L. This decline was likely caused by (1) reduced C/N ratio, (2) increased NH₄⁺-N loading, and (3) free ammonia inhibition. Literature suggests that C/N > 7 supports robust HN–AD activity, with complete removal reported at C/N ≥ 10 [10,19,20]. High organic content is essential for HN–AD bacteria to sustain growth, energy production, and aerobic denitrification. Although optimal C/N ratios for many HN–AD species have been reported between 8 and 12 [10], high ammonium loading can negatively affect their performance [21]. In this study, increasing NH₄⁺-N from ~49 to ~98 mg/L doubled the aerobic NH₄⁺-N loading rate from ~4.1 to ~8.1 mg L⁻¹ h⁻¹, potentially exceeding the degradation capacity of the HN–AD community (observed maximum 8.62–11.39 mg L⁻¹ h⁻¹ at C/N 17.2 in cycle tests). Furthermore, although free ammonia concentrations in Phase II (~1.75 mg/L at pH 7.51) were below the strong inhibition threshold (~6.12 mg/L [22], this increase may still have contributed to reduced removal efficiency.

Cycle test results showed that DO concentrations strongly influence HN–AD performance. While high DO favors nitrification, it may suppress aerobic denitrification by favoring oxygen as electron acceptors. Complete denitrification by Pseudomonas stutzeri T13 was reported at DO < 4.37 mg/L [14]. An optimal DO of ~5.1 mg/L was observed for Acinetobacter sp. 11 in pig farm wastewater. Jin et al. [23] found that maintaining DO below 3 mg/L was critical for simultaneous nitrification and denitrification by Pseudomonas sp. ADN-42. These findings highlight that balancing DO during the aerobic phase is essential to maximize HN–AD efficiency.

This study also demonstrated that HN–AD can achieve excellent COD and nitrogen removal at high organic loading rates. The organic loading rates in this study ranged from 1.86 to 4.86 kg/m³·day, substantially higher than those typically applied in conventional nitrification process designs (0.17–0.31 kg/m³·day) [24]. Guo et al. [25] reported that increasing the organic loading rate from 0.5 to 1.5 kg COD/m³·day raised the contribution of heterotrophic nitrification from 7% to 21%. Higher organic loading rates stimulate HN–AD by accelerating oxygen consumption, which in turn lowers DO concentrations in the reactor—helping to maintain DO within the optimal range for simultaneous nitrification and aerobic denitrification. This DO-lowering effect, combined with abundant organic carbon supply, promotes aerobic denitrification as a co-respiratory pathway, enhancing overall nitrogen removal efficiency.

Microbial community analysis from our previous study at the Guelph WWTP identified Pseudomonas and Acinetobacter as dominant potential HN–AD genera [26]. Strains within these genera—such as Pseudomonas putida NP5, P. aeruginosa YL, Acinetobacter junii YB—exhibit strong nitrification (4.69–12.36 mg NH₄⁺-N L⁻¹ h⁻¹) and denitrification (2.38–8.68 mg NO₃⁻-N L⁻¹ h⁻¹; 1.97–9.96 mg NO₂⁻-N L⁻¹ h⁻¹) capabilities [1]. The simultaneous nitrification and denitrification rates observed in this study (8.62 and 11.39 mg NH₄⁺-N L⁻¹ h⁻¹) are consistent with those reported for isolated HN–AD strains. Notably, this work demonstrates that a diverse activated sludge community can achieve near-complete SND under high C/N wastewater conditions, suggesting that mixed microbial consortia may provide more stable and higher nitrogen removal efficiencies than pure cultures.

5. Conclusions

This study shows that an activated sludge SBR can achieve near-complete simultaneous heterotrophic nitrification–aerobic denitrification for high-C/N wastewater, with COD, NH₄⁺-N, and TN removal efficiencies of 98.6%, 93.3%, and 87.1%, respectively, and no NO₂⁻/NO₃⁻ accumulation. Stable operation was sustained at organic loading rates of 1.86–4.86 kg m⁻³·day and a C/N ratio of 17.1, which also helped to maintain DO in the optimal range for aerobic denitrification. Performance dropped when C/N fell below ~8 and NH₄⁺-N loading exceeded ~8 mg L⁻¹ h⁻¹, highlighting the need to maintain C/N > 7–8, DO < 4 mg/L, and ammonium loading within the observed microbial capacity. These thresholds provide clear operational targets for optimizing SBRs treating high-C/N wastewater. The results obtained from this study demonstrate that HN–AD is a promising technology for high C/N wastewater treatment.