Partial nitrification-anammox (PNA) treating sewage with intermittent aeration mode: Effect of influent C/N ratios

Highlights

• Nitrogen removal performance of PNA was enhanced at C/N ratios from 1.1 to 2.5.

• Anammox bacterial activity exponentially increased with C/N ratios from1.1 to 1.5.

• Anammox bacteria were affected by limited nitrite production at a C/N ratio of 2.0.

• Both particle size and settleability of sludge improved with elevated C/N ratios.

Abstract

The partial nitrification-anammox (PNA) process under intermittent aeration is feasible to treat low-strength ammonium wastewater, but the effect of chemical oxygen demand (COD) on mainstream anammox processes is not fully understood. In this study, PNA process was established for sewage treatment (average ammonium = 59.3 mg-N/L) in a sequencing batch reactor (SBR) with intermittent aeration mode. Influent organic carbon to nitrogen (C/N) ratios were gradually increased (1.1, 1.5, 2.0 and 2.5). With increasing C/N ratios, the total nitrogen (TN) removal efficiency increased steadily from 30.8% to 77.3%. When the C/N ratio increased from 1.1 to 2.0, the improvement of nitrogen removal was attributed to the increase of anammox bacteria, which exponentially increased at the C/N ratios from 1.1 to 1.5. Additionally, nitrate reduction using internal carbon was likely to combine with anammox reaction under intermittent aeration operation, which might further improve the nitrogen removal. Lastly, settleability and particle size of activated sludge were also improved with the elevated C/N ratios, which favored the enrichment of anammox bacteria and facilitated the combination of partial denitrification and anammox. When the C/N ratio was 2.5, TN removal efficiency still increased while the abundance and activity of anammox bacteria remained stable, indicating that influent organic carbon promoted denitrifying bacteria and inhibited anammox bacterial growth rate. Overall, this study demonstrates that influent COD has comprehensive effects on PNA systems and that the influent C/N ratios are suggested to adjust within optimal range to achieve long-term stability.

Introduction

The partial nitrification-anammox (PNA) process involves ammonium oxidizing bacteria (AOB) converting approximately half of the ammonium in wastewater to nitrite, followed by the remaining ammonium and nitrite being converted to N₂ by anammox bacteria. PNA is considered a cost saving alternative to conventional biological nitrogen removal via heterotrophic denitrification [1], [2]. The PNA process consumes 60% less oxygen and 100% less organic carbon source, produces about 90% less sludge compared with nitrification/denitrification [2]. To date, PNA has mainly been used to treat high-strength wastewater with low concentrations of biodegradable organic substances with organic carbon to nitrogen (C/N) ratios lower than 0.5 [3], [4], [5]. Recently, the mainstream PNA has been studied in both lab-scale and pilot-scale [6], [7], [8], [9], and the feasibility of PNA for low ammonium sewage treatment has been verified in full-scale plants [10], [11], [12]. However, there are still several challenges for its application in sewage treatment, including anammox bacteria enrichment, seasonal temperature variation of municipal wastewater and strict discharge standards of effluent nitrogen concentration [13], [14], [15]. Among these challenges, nitrite oxidizing bacteria (NOB) suppression is a major challenge in mainstream PNA process due to the lack of the inhibition factors such as high free ammonium and free nitrite [13], [14], [16]. For NOB suppression in mainstream treatment, intermittent aeration has been considered as an effective strategy because during the transition from anoxic to aerobic environment, NOB activity is difficult to recover in a short time [14], [15], [17], [18].

The influent carbon source is proved to be another critical factor for PNA performance [13], [19]. The effect of chemical oxygen demand (COD) on the PNA process has two opposing aspects. On the one hand, the influent COD affects the nitrogen removal performance by shifting the microbial population. High COD concentration leads to growth of heterotrophic bacteria, resulting in suppression of AOB and anammox bacteria. Anammox activity becomes a limiting factor for PNA and the nitrogen removal efficiency decreases from 79% to 52% when the influent C/N ratio increases from 0.5 to 0.75 [20]. Both the AOB and anammox bacterial activity decrease at a C/N ratio of 2.7–2.9 [21]. Specially, it was reported that a high COD concentration would suppress anammox activity and a COD concentration up to 292 mg/L could completely inhibit anammox bacteria [22]. While in another aspect, acetate can be utilized by anammox bacteria to remove nitrate [19], [23].

On the other hand, PNA performance improves under appropriate influent COD concentrations since the enriched heterotrophic bacteria can denitrify the nitrate produced by anammox bacteria, thereby reducing the effluent nitrogen concentration [24]. For example, a high nitrogen removal efficiency of 84–95% was achieved at an influent C/N ratio of 1.2 with glucose as carbon source [25].

Furthermore, the changes in COD can affect the biomass concentration and sludge characteristics, which is important to the enrichment of anammox bacteria. Anammox bacteria are slow growing microorganisms with a doubling time of 11 days [26]. Hence, efficient biomass retention is important for the anammox process. The biomass with dense structure and good settleability, such as granular sludge with larger particle size, could provide a long biomass retention time and a favorable inside anoxic environment, promoting the growth of anammox bacteria [23], [27]. In addition, anammox granular sludge exhibits significant advantage over floc sludge in the aspect of tolerance to fluctuation in external conditions [24]. In PNA process, the anammox bacterial activity decreased when the average particle size decreased [19]. COD was found to affect extracellular polymeric substances (EPS) [28], which promote floc formation by facilitating the aggregation of cells and sludge particles [28], [29].

The effect of influent C/N ratios on the PNA process treating high-strength wastewater has been investigated [20], [30]. There are several studies about the effect of influent C/N ratios on PNA process treating low-strength synthetic wastewater [23], [31]. Few studies focus on the treatment of low-strength sewage [21], in which the organic components are more complex than in synthetic wastewater. Furthermore, the changes in sludge morphology, which greatly influences the enrichment and retention of slow growing anammox bacteria, are not well understood.

The objective of this study was to elucidate the effect of C/N ratios on the nitrogen removal performance and sludge morphology during the PNA process when treating low-strength sewage. A sequencing batch reactor (SBR) under intermittent aeration was found to facilitate the stability of the mainstream PNA process [15], [17]. Specifically, intermittent aeration with a long anoxic phase could successfully inhibit NOB and enrich anammox bacteria. In this study, an SBR was used for PNA in intermittent aeration mode. The sewage was pretreated in an aerobic reactor to adjust the C/N ratio, which was increased stepwise from 1.1 to 2.5. At the highest C/N ratio (2.5), the sewage was fed into the PNA reactor without pretreatment. The absolute abundances and relative abundances of functional bacteria were measured using the quantitative real-time PCR (qPCR) and Illumina high-throughput sequencing analysis, respectively. Changes in sludge morphology were investigated and correlated with the anammox bacterial population and PNA performance.