Chemosphere 174 (2017) 399e407

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Simultaneous domestic wastewater and nitrate sewage treatment by DEnitrifying AMmonium OXidation (DEAMOX) in sequencing batch reactor Rui Du a, Shenbin Cao b, Baikun Li a, Shuying Wang a, Yongzhen Peng a, * a

Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering, Engineering Research Center of Beijing, Beijing University of Technology, Beijing 100124, China State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, China

b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


DEAMOX was developed to treat domestic wastewater and NO 3 -N sewage simultaneously.
TN removal efficiency achieved 95.8% þ with influent NO 3 -N/NH4 -N ratio of 1.09.
Performance recovered rapidly after deterioration for mass transfer limitation.
The dominant Thauera genera possibly played key role in high NO 2N accumulation.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 May 2016 Received in revised form 7 December 2016 Accepted 3 February 2017 Available online 4 February 2017

A novel DEAMOX system was developed for nitrogen removal from domestic wastewater and nitrate   (NO 3 -N) sewage in sequencing batch reactor (SBR). High nitrite (NO2 -N) was produced from NO3 -N reduction in partial-denitrification process, which served as electron acceptor for anammox and was removed with ammonia (NHþ 4 -N) in domestic wastewater simultaneously. A 500-days operation demonstrated that the efficient and stable nitrogen removal performance could be achieved by DEAMOX. 1 and The total nitrogen (TN) removal efficiency was as high as 95.8% with influent NHþ 4 -N of 63.58 mg L þ 1 NO 3 -N of 69.24 mg L . The maximum NH4 -N removal efficiency reached up to 94.7%, corresponding to the NO 3 -N removal efficiency of 97.8%. The biomass of partial-denitrification and anammox bacteria was observed to be wall-growth. The deteriorated nitrogen removal performance occurred due to excess denitrifying microbial growth in the outer layer of sludge consortium, which prevented the substrate transfer for anammox inside. However, an excellent nitrogen removal could be guaranteed by scrapping the superficial denitrifying biomass at regular intervals. Furthermore, the high-throughput sequencing analysis revealed that the Thauera genera (26.33%) was possibly responsible for the high NO 2 -N accumulation in partial-denitrification and Candidatus Brocadia (1.7%) was the major anammox species. © 2017 Elsevier Ltd. All rights reserved.

Handling Editor: Chang-Ping Yu Keywords: Partial denitrification Anammox Domestic wastewater Nitrite accumulation High-throughput sequencing

1. Introduction

* Corresponding author. E-mail address: [email protected] (Y. Peng). http://dx.doi.org/10.1016/j.chemosphere.2017.02.013 0045-6535/© 2017 Elsevier Ltd. All rights reserved.

Anammox process capable of converting NHþ 4 -N to nitrogen gas (N2) using NO 2 -N as the electron acceptor has attracted much

400

R. Du et al. / Chemosphere 174 (2017) 399e407

attention (Mulder et al., 1995; Jetten et al., 1998) for its low demand of oxygen, no organic carbon sources required (Van Dongen et al., 2001), high nitrogen removal and low sludge yield (Jetten et al., 1998). Autotrophic anammox process has been applied for ammonia-rich wastewater (Peng et al., 2008; Joss et al., 2009), but limited studies focused on its widely application in treating lowstrength sewage, such as the domestic wastewater. The main challenge of anammox for domestic wastewater treatment is the stable source of NO 2 -N generation (Lackner et al., 2014). Traditional NO 2 -N production was via nitritation process, in   which NHþ 4 -N is oxidized to NO2 -N without going further to NO3 N. Successful nitritation-anammox process is favorable for ammonium-rich wastewater due to the free ammonia (FA) inhibition on nitrite oxidation bacteria (NOB). However, nitritation is difficult to maintain stably for the low-strength wastewater (Ma et al., 2009; Peng et al., 2012), consequently the effluent contains  undesirable concentration of NO 3 -N due to NO2 -N oxidation by NOB (Fux et al., 2004; Erguder et al., 2008). On the other hand, NO 2 -N can be produced from partialdenitrification process. According to our previous study, NO 3 -N could be reduced to NO 2 -N with the transformation ratio about 80% in the denitrification (Cao et al., 2013). Moreover, excess NO 3 -N in effluent of anammox reactor could be converted to NO 2 -N efficiently and removed with NHþ 4 -N by backflow to anammox reactor for advanced nitrogen removal performance (Du et al., 2015a). This demonstrated the feasibility of providing NO 2 -N for anammox by NO 3 -N reduction, which hold great potential for treating lowstrength wastewater by integrating partial-denitrification to anammox process. In the DEAMOX system, NO 2 -N is generated from heterotrophic NO 3 -N reduction by inoculated partial-denitrification sludge, then þ the NO 2 -N and NH4 -N was removed by anammox bacteria in single reactor. In this case, NO 2 -N could be accumulated without complex control (Cao et al., 2013) compared with nitritation process. Therefore, it provided an efficient option for simultaneous nitrogen removal from domestic wastewater and NO 3 -N sewage, such as some industrial wastewater for fertilizer, metal finishing and explosive, which usually contained high concentration of NO 3 -N. Previous study reported that the average nitrogen removal was 81% in an anammox-denitrifying system when fed with synthetic  wastewater containing NHþ 4 -N, NO3 -N, and acetic acid (Gao et al., 2012). Kalyuzhnyi and Gladchenko (2009) found that it was less efficient when using organic matter as electron donor for DEAMOX with NHþ 4 -N removal around 40% compared with that using sulphide. However, the efficiency and stability of NO 2 -N production during partial-denitrification as well as the microbial community of the system have not been well understood, especially for the treatment of real domestic wastewater. In this study, the DEAMOX based on partial-denitrification for simultaneous NO 3 -N sewage and domestic wastewater treatment was developed for the first time. Denitrifying sludge with high NO 2 -N accumulation was inoculated to supply substrate for anammox (Fig. 1). The nitrogen removal performance and the characteristics of active sludge were investigated over a 500-days operation. The microbial community of the system was also estimated by high-throughput sequencing. 2. Materials and methods 2.1. The DEAMOX SBR The DEAMOX SBR (working volume: 2.0 L, diameter: 120 mm, and height: 250 mm) was operated with 8 h cycles per day, consisting of four stages: 5 min for feeding with 1.2 L wastewater, 6 h for anaerobic reaction with a magnetic mixing (250 rpm), 60 min

Fig. 1. Schematic of the cooperation between partial-denitrification and anammox in the DEAMOX system and mechanism for nitrogen removal from domestic wastewater and NO 3 -N sewage.

for settling, 5 min for discharging (1.2 L supernate), and 50 min for idle. The system was operated at 28 ± 0.5  C to provide a suitable condition for anammox bacteria and keep efficient microorganism retention. The domestic wastewater and NO 3 -N was pumped into the SBR using a peristaltic pump, as well as sodium acetate solution was pumped separately as external carbon source, as shown in Fig. 1. The SBR was covered with a black PVC material (5 mm thickness) to prevent the penetration of light (van de Graaf et al., 1996). 2.2. Seeding sludge and wastewater The DEAMOX SBR was inoculated with the anammox and partial-denitrification sludge. The seeding anammox sludge was collected from a lab-scale upflow anaerobic sludge blanket (UASB) reactors (working volume of 3.6 L), which had been operated for 7  months with the influent NHþ 4 -N and NO2 -N concentration of 300 mg L1 and 400 mg L1, respectively. The mixed liquor suspended solid (MLSS) and mixed liquor volatile suspended solid (MLVSS) of seeding sludge was about 3.08 g L1 and 1.79 g L1, respectively. The partial-denitrification sludge was collected from an enriched partial-denitrification system as previously reported   (Cao et al., 2013). It reduced NO 3 -N to NO2 -N with the NO3 -N to  NO2 -N transformation ratio (NTR) of 80% when feeding with syn1 thetic wastewater containing NO 3 -N (30 mg L ) and sodium acetate. The MLSS and MLVSS of the seeding partial-denitrification sludge were 3.0 g L1 and 1.8 g L1, respectively. The MLVSS in the DEAMOX SBR was 2.5 g L1 after inoculation. The DEAMOX reactor was fed with domestic wastewater collected from an on-campus sewer line and the wastewater characteristics were listed as follows: COD: 125.2e264.8 mg L1,   1 1 NHþ 4 -N: 47.32e79.09 mg L , NO2 -N: 0e0.02 mg L , NO3 -N: 0.01e0.46 mg L1, and pH: 7.0e7.6. Increasing nitrogen loadings (NL) were examined during the whole operation. At the start-up phase, domestic wastewater was diluted before feeding, then sodium nitrate was added into the þ wastewater to achieve an NO 3 -N/NH4 -N ratio of 1.0e1.3, in order to  simulate the mixture of NO3 -N contained wastewater and domestic wastewater. The increasing NHþ 4 -N loading was achieved by adding ammonia chloride into the mixed wastewater. Since the concentration of biodegradable organic matter in domestic wastewater dropped after dilution, sodium acetate solution was pumped into

R. Du et al. / Chemosphere 174 (2017) 399e407

the reactor at the end of feeding period to maintain the external COD to NO 3 -N ratio (C/N) of 2.0e3.0.

Denitrification contribution ¼

2.3. Batch test The activity of outer layer sludge scraped from the DEAMOX SBR was examined, which was performed in air tight vessel (volume of 1 L) at constant temperature (30  C). Specifically, the sludge was washed and centrifuged twice to remove the attached inorganic nitrogen compounds. The centrifuged sludge was resuspended to obtain 2.0 gVSS L1. The mixture was sparged with N2 (0.6 L min1) for 20 min prior to the experiment. Prepared solution containing þ 1 1 NO 3 -N (10 g L ) and NH4 -N (10 g L ) was added to the vessel to  þ achieve the initial NO3 -N and NH4 -N concentration of 20 mg L1 and 20 mg L1, respectively. The C/N ratio was kept at 2.5 by adding sodium acetate solution (10 g L1). The pH was not controlled during the batch tests and varied at the range of 7.7e8.2 over time. The samples were taken at different time intervals for a period of 150 min, and then MLSS and MLVSS were measured.

2.4. Analytical procedure Influent and effluent samples were collected on a daily basis.   The parameters, including NHþ 4 -N, NO2 -N, and NO3 -N, were measured with a Lachat QuikChem 8500 Flow Injection Analyzer (Lachat Instruments, Milwaukee,USA). The COD was analyzed using a COD quick-analysis apparatus (Lian-hua Tech. Co., Ltd., 5B-1, China). The MLSS and MLVSS of activated sludge were measured according to the Standard Methods (APHA, 1998). The morphology of activated sludge in the DEAMOX SBR was observed using OLYMPUS_BX51 microscope on day 220 and day 255.

2.5. Calculations The contributions of anammox and denitrification pathways on TN removal were estimated based on the difference between influent and effluent NHþ 4 -N concentration, which neglected microbe assimilation and assumed the NHþ 4 -N removal mainly carried out by anammox process (Du et al., 2017).

2.5.1. Nitrogen removal via anammox pathway contributing to TN removal

Anammox contribution ¼

h

Inf:NHþ 4 

N

Eff:NHþ 4

N



þ þ 1:32 Inf:NHþ 4  N  Eff :NH4 i  N  100%=ðInf:TN  Eff :TNÞ

(1) þ þ where the Inf.NHþ 4 -N and Eff.NH4 -N were the NH4 -N concentration in the influent and effluent during long-term operation, respectively. The Inf.TN and Eff.TN were TN concentration in the influent   and effluent, respectively. TN ¼ NHþ 4 -N þ NO2 -N þ NO3 -N.

2.5.2. Nitrogen removal via denitrification pathway contributing to TN removal

401

h   Inf:NO 3  N  Eff :NO3  N    1:32 Inf:NHþ 4  N  Eff :NH4    N þ 0:26 Inf:NH 4 N i  Eff:NHþ 4 N  100%=ðInf:TN  Eff:TNÞ (2)

  where the Inf.NO 3 -N and Eff.NO3 -N were the NO3 -N concentration of influent and effluent during long-term operation, respectively.

2.6. Microbial community analysis 2.6.1. DNA extraction and PCR DNA was extracted from dried sludge (weight: 0.10e0.20 g) using the Fast DNA Kit (BIO 101, Vista, CA). DNA concentrations were measured with a NanoDrop ND-1000 (NanoDropTechnologies, Wilmington, DE, USA). PCR was conducted to amplify the 16S rRNA gene. Primers for sequencing were 515F (50 - GTGCCAGCMGCCGCGG-30 ) and 907R (50 CCGTCAATTCMTTTRAGTTT -30 ) for the V4 and V5 region of 16S rRNA gene. The DNA extraction PCR was performed as our previous report (Du et al., 2015b). 2.6.2. High-throughput sequencing and data analysis Sequencing was carried out on an Illumina MiSeq platform (Du et al., 2015b). To minimize the effects of potential early-round PCR errors, amplicon libraries were prepared by combining three independent PCR products. The trimmed sequences were grouped into operational taxonomic units (OTUs) using 97% identity thresholds (i.e., 3% dissimilarity levels) by the Usearch software program. The species richness, and rarefaction curves and Shannon-Wiener were generated from OTU numbers counted for the sample. The generated raw sequences of the sludge sample were assigned by Silva to trim off the adapters and barcodes. The raw reads have been archived at NCBI Sequence Read Archive (SRA) database under accession number SRR2179254. 3. Results and discussion 3.1. Start-up and operation of the DEAMOX system with increasing nitrogen loading The DEAMOX system was operated over a period of 500 days divided into 7 phases. Considering the possible inhibition of complex component in the real domestic wastewater on anammox bacteria, the SBR was started up with low NL. During phase 1 (day 1 1e94), the average influent NHþ and NO 4 -N was 22.91 mg L 3 -N 1 was 26.74 mg L (Fig. 2a and b). It could be seen that the NHþ 4 -N and NO 3 -N removal efficiency achieved 76.2% and 89.9%, respectively. This indicated that the partial-denitrification could provide  NO 2 -N from NO3 -N reduction for anammox, which was a crucial  condition for simultaneous NHþ 4 -N and NO3 -N removal. In phase 2 (day 95e170), the influent NL was elevated to  90.57 mg L1 (Fig. 2c). The NHþ 4 -N and NO3 -N removal efficiency þ  increased with influent NH4 -N and NO3 -N of 42.02 mg L1 and 48.55 mg L1, respectively. Clearly, the activity of anammox bacteria was little affected by the increasing organic matter introduced for partial-denitrification. Furthermore, when the influent nitrogen was elevated to the level of raw domestic wastewater in phase 3 1 (day 171e245) with average NHþ 4 -N of 60.69 mg L , satisfactory

402

R. Du et al. / Chemosphere 174 (2017) 399e407

Fig. 2. Nitrogen removal of DEAMOX at different phases over 500-days operation: þ variation of NHþ 4 -N concentration and NH4 -N removal efficiency (a); the concentration   of NO 3 -N, NO2 -N and NO3 -N removal efficiency (b); TN concentration and TN removal efficiency (c); COD concentration in influent and effluent (d).

 removal of both NHþ 4 -N and NO3 -N was observed and maintained stably with the TN removal efficiency as high as 94.2%. The organic matter in effluent maintained at low level during the whole operation, with the average COD concentration varied in range of 35.51 mg L1 to 46.5 mg L1, suggested that desirable performance of COD removal could be achieved in the DEAMOX process (Fig. 2d). Additionally, the nitrogen removal by anammox and denitrification pathways contributing to TN removal was estimated (Fig. 3). The percentage of anammox contribution occupied 87.0% in TN removal during phase 1, and it increased to 93.9% in phase 3, while the denitrification percentage was at relatively low level of 6.1%. It clearly suggested that the anammox could compete over denitrification and maintain dominating in the DEAMOX system during long-term operation. Moreover, the performance stability was

mainly attributed to the successful cooperation between the partial-denitrification and anammox process. This could be demonstrated by the nitrogen removal activity of partialdenitrification and anammox in typical cycle (Fig. 4). After the domestic wastewater and NO 3 -N introduced to the SBR, a rapid consumption of NO 3 -N and organic matter (COD) was observed (Fig. 4a), along with the significant accumulation of NO 2 -N, namely the NO 2 -N accumulation period. It subsequently achieved a peak value of 11.93 mg L1 within 15 min in phase 1, and 22.42 mg L1 within 30 min in phase 2 (Fig. 4b). When the initial NO 3 -N increased to 56.98 mg L1 at phase 3, the NO 2 -N kept accumulating and reached up to 41.86 mg L1 at 60 min (Fig. 4c), corresponding to the NTR over 70% during the NO 2 -N accumulation period. Additionally, it should be noted that the NO 2 -N accumulation peak occurred just at the point of NO 3 -N consumed completely. It was notably that the NHþ 4 -N decreased when the partial-denitrification was proceeding, suggested that a certain amount of accumulated þ  NO 2 -N and influent NH4 -N was removed by anammox during NO2  N accumulation period. After the NO3 -N depletion, the accumuþ lated NO 2 -N served as an electron acceptor for NH4 -N removal via  anammox process, namely NO2 -N reduction period. Furthermore, there was 11% of nitrogen produced as NO 3 -N in the anammox process in theory, while the NO 3 -N did not increased during this period. This could be explained by that the NO 3 -N produced by anammox was reduced to NO 2 -N with endogenous denitrification and removed with NHþ 4 -N (Cao et al., 2016b), consequently achieved the advanced nitrogen removal performance. On the other hand, partial-denitrification proceeded rapidly within the initial 30e60 min and the NO 2 -N accumulation achieved the peak value once NO 3 -N was consumed completely  (Fig. 4). In comparison, the NHþ 4 -N and NO2 -N consumption during  the NO2 -N reduction period was relatively slower in the next couple hours, implying that the anammox was the rate-limited step during the whole DEAMOX process. During the stable operation period in phase 3, the TN removal þ efficiency reached 94.2% with an influent NO 3 -N/NH4 -N ratio of 1.18. This was much higher than the previous study. Waki et al.  (2013) reported that the NHþ 4 -N removal was 40% but NO3 -N  removal was higher than 90% at the influent NHþ 4 -N/NO3 -N ratio of 1.0 in a simultaneous NO 3 -N reduction and anammox process, since most of NO 3 -N was reduced to N2 gas by complete denitrification. In this study, the inoculated partial-denitrification sludge  had much higher activity for transforming NO 3 -N to NO2 -N, rather than to N2. The mechanism model of the cooperation between the partial-denitfification and anammox in this DEAMOX system was depicted in Fig. 1. It clear indicated that efficient partialdenitrification process could provide NO 2 -N for anammox reliably, consequently reduced the risk of substrate competition between complete denitrification and anammox, which was regarded as the most essential condition for maintaining efficient and stable nitrogen removal performance in the system. Physiological characteristics of biomass in the reactor at different phases were observed using microscope (Fig. S1). Previous studies had found that hydroxylamine oxidoreductase and hydrazine oxidoreductase are two key enzymes of the anammox metabolic pathway (Kuenen, 2008), and these two enzymes are rich in heme c, which adds a carmine color to the anammox granular sludge (Kuenen, 2008; Schmid et al., 2007). The cultivated sludge became reddish-brown after 220 days of operation, which was a visual indication of the anammox microbes in the system (Kartal et al., 2013). Moreover, these microbes was observed to be aggregated and attached on the reactor wall, which was similar to previous study of anammox reactors (Du et al., 2014; Hendrickx et al., 2012).

R. Du et al. / Chemosphere 174 (2017) 399e407

403

Fig. 3. Estimated percentage of anammox and denitrification contribution to TN removal and the amounts of TN removal in the DEAMOX system.

10

120

180

240

300

0 360

Time (min)

-1

50

(c) 120

Nitrite Accumulation

-1

90 30 60

20

30

10 0 60

120

180

240

300

360

COD (mg L )

40

0

15

0

Time (min)

-1

40

10 20

5 0 60

120

180

240

300

0 360

Time (min)

25

150

Nitrite Reduction

60

20

100

Lag phase

-1

60

80

Nitrite Accumulation

25

0

Nitrogen concentration (mg L )

60

30

(b) COD (mg L )

5

Nitrite Reduction

(d)

20

80

15

60

10

40

5

20

0 0

30

60

90

120

-1

-1

20

0

Nitrogen concentration (mg L )

30

-1

NO3--N

COD

COD (mg L )

40

NO2--N

10

0

Nitrogen concentration (mg L )

NH4+-N

Nitrite Accumulation

15

100

35

(a)

COD (mg L )

50

Nitrite Reduction

-1

Nitrogen concentration (mg L )

20

0 150

Time (min)

Fig. 4. Profiles of nitrogen and COD concentration during typical cycle in the DEAMOX system at phase 1(a), phase 2 (b) and phase 3 (c); nitrogen transformation in batch test with the outer layer sludge of the reactor (d).

3.2. Performance deterioration and rapid recovery During phase 4 (day 246e267), the NHþ 4 -N was observed to increase with NO 2 -N remained in the effluent, resulting in the TN removal efficiency declining to 67.9% on day 261. Moreover, the outer layer of active sludge turned to be black (Fig. S1). It could be speculated that, with the biomass growth as consortium on reactor wall, the substrates transfer of bacteria inside suffered a limitation, leading to the anaerobic zone and cell decay in the inner layer for the lack of substrates. This could explain the black color of sludge and deterioration of nitrogen removal performance. Similar result was also observed in the previous study (Kalyuzhnyi et al., 2007). In order to improve the nitrogen removal performance, sludge at  the outer layer was scraped, as well as the influent NHþ 4 -N and NO3 -

N reduced to 44.80 mg L1 and 50.10 mg L1 on day 262, respectively. It was observed that the nitrogen removal efficiency recovered rapidly with a marked decrease of effluent NHþ 4 -N (Fig. 2a), which was caused by the enhanced transfer of nitrogenous species in the sludge aggregation. In this case, sufficient NO 2 -N produced from partial-denitrification could be used as the electron acceptor by anammox bacteria to remove NHþ 4 -N in domestic wastewater. In addition, the waste sludge scraped from the outer layer of the reactor was then collected in order to assess the activity in batch test (Fig. 4d). The initial C/N ratio was set as 2.5 to supply a suitable  electron donor for NO 3 -N reduction to NO2 -N. After the substrates 1  (NHþ -N of 20 mg L and NO -N of 20 mg L1) were added, NO 4 3 2 -N concentration steadily increased corresponding to both of NO 3 -N and COD decreasing gradually. A turning point was observed at the

404

R. Du et al. / Chemosphere 174 (2017) 399e407

60 min, after which the NO 2 -N accumulation became much slower and reached a peak concentration (17.94 mg L1) at 120 min, resulting in the NTR was 97.2%. However, no further reduction of þ NO 2 -N was observed and little NH4 -N was consumed later, which leading to a lag phase occurred from 60 to 150 min as shown in Fig. 4d. This indicated that the activity of partial-denitrification was dominant in the outer layer sludge. In other words, the denitrifying bacteria was assumed to be located at the outer layer of sludge consortium, while considerable anammox bacteria were presumably retained inside and avoided being eliminated from the SBR. In this case, anammox bacteria could grow synergistically with the heterotrophic partial-denitrification bacteria by receiving the NO 2N transformed from NO 3 -N reduction, as shown in Fig. 1. Thus, the  NHþ 4 -N and NO3 -N in wastewater could be removed simultaneously and efficiently. The sludge collected at the outer layer had more smooth and transparent appearance than that of the inside (Fig. S1). There was a dead zone in the aggregation attached closely to the reactor wall which turned into black color. These dead cells would be decayed into organic matter and NHþ 4 -N, which would be used by heterotrophic denitrification bacteria and increase the NHþ 4 -N in effluent. Notably, the nitrogen removal performance recovered rapidly once the excess sludge at the outer layer of the reactor was scrapped, suggested that the regular sludge washout would be conducted to avoid the substrate transfer limitation for anammox bacteria inside. 3.3. Stability of DEAMOX in long-term operation þ The influent NHþ 4 -N loading was kept at a high level with NH4 -N 1 of 55.12 mg L1 and NO -N of 64.99 mg L during phase 5 (day 3 268e388), corresponding to the average TN removal efficiency reached 96.0%. However, in phase 6 (day 389e407), there was NHþ 41 N (10.96 mg L1) and NO 2 -N (6.21 mg L ) accumulating in the effluent, and the TN removal efficiency decreased to 86.2%. This was assumed to be caused by the excess biomass growth as previously observed (e.g. in phase 4). Similarly, some amount of sludge at the outer layer was scrapped on day 389, while the influent NL main1 tained at high levels (NHþ and NO 4 -N of 58.58 mg L 3 -N of 70.13 mg L1) during this phase. Consequently, the removal efficiency of NHþ 4 -N and TN increased gradually and recovered after two weeks, suggesting that the excess sludge washout was necessary during the long-term operation of DEAMOX SBR. Furthermore, the second sludge scrape in phase 6 was carried out after 140 days away from the first time in phase 4, which was shorter than the time interval from start-up to phase 4 (260 days), this might be related to the higher influent NL in phase 3e6. Therefore, the sludge scrapping should be conducted more frequently when the influent NL was higher. After the nitrogen removal performance recovered rapidly, the TN removal maintained at a high level of 95.8% in average with the effluent TN concentration as low as 5.58 mg L1, when the influent þ NO 3 -N/NH4 -N ratio of 1.09 in phase 7 (day 409e500). This confirmed that the DEAMOX process achieved desirable nitrogen removal from low-strength domestic wastewater and NO 3 -N sewage. It was much higher than the previous study by Kalyuzhnyi and Gladchenko (2009), who reported that the removal efficiency  of NHþ 4 -N, NO3 -N, and TN were 40%, 100%, and 66% respectively during a DEAMOX process using volatile fatty acid as carbon source. This also indicated that the stable and high NO 2 -N accumulation during the partial-denitrification played a key role in the advanced nitrogen removal performance of anammox. On the other hand, previous study reported that the anammox activity decreased remarkably at C/N ratios greater than 2.0 due to the predominance of complete denitrification instead of anammox reaction (Takekawa et al., 2014), and no inhibition on anammox

was observed at the low C/N ratio of 0.4e1.3 (Waki et al., 2013). These were much lower than the C/N of this study, but in which the competition for NO 2 -N between the anammox and denitrification was substantially alleviated. This was mainly attributed to the high NO the inoculated partial2 -N accumulation property of denitrification sludge. Anammox maintained the dominated process with the percentage for TN removal as high as 97.1% at stable operation in phase 7. 3.4. Microbial community analysis in DEAMOX system The active sludge sample was collected in stably operational phase on day 480 for the high-throughput sequencing analysis. The phylogenetic classification of effective bacterial sequences were conducted at phylum and genus taxonomic levels. Active sludge had high microbial diversity (Shannon index >3.5) with 30 phyla and 154 genus in the total bacterial population, which was possibly related to the complex components in real domestic wastewater. Phylums Proteobacteria (45.43%), Chloroflexi (15.3%), Bacteroidetes (9.86%) and Planctomycetes (7.39%) were dominant in the community at the stable operation period (Fig. 5a). The most abundant phylum was Proteobacteria, which had been found as the dominant phylum in various municipal wastewater treatment plants previously (Ma et al., 2015). Notably, the genus Thauera belonging to Proteobacteria phylum was dominant accounting for 26.3% (Fig. 5b). Thauera genus was reported to be capable of denitrification and organic compound biodegradation, which had been identified in wastewater bioreactors (Liu et al., 2006; Langone et al., 2014). The progressive onset of different denitrification genes of Thauera strains were classified recently (Liu et al., 2013). There were Thauera strains reported to possess the capability of NO 2 -N accumulation, and the transcription of nirS  gene (encoding NO 2 -N reductase) was not detected until NO3 -N was consumed (Liu et al., 2013). This study confirmed the dominance of the Thauera genera in the DEAMOX system, which would contribute to high NO accumulation during partial2 -N denitrification (Kartal et al., 2007). The symbiosis of anammox bacteria (i.e., Planctomycetes) with other phyla was found in the DEAMOX system. The phylum Planctomycetes was less abundant than expected (7.39%). The Candidatus Brocadia was the major anammox bacteria species in the system (1.7%) (Fig. 5b), while uncultured Planctomycetaceae was detected with the abundance of 1.09%. Previous study also found that in the presence of acetate, Candidatus Brocadia was more competitive than other anammox bacteria (Kartal et al., 2007; Winkler et al., 2012). Since anammox bacteria degraded acetate to CO2 and then  reused as the electron donor to reduce NO 3 -N and/or NO2 -N instead of incorporating acetate into biomass directly (Kartal et al., 2007). In addition, Candidatus Brocadia was much less abundant than heterotrophic bacteria using acetate as the electron donor in this DEAMOX process, which implied that the acetate consumption by Candidatus Brocadia could be ignored. Previous studies also found that the dominant Candidatus Brocadia did not contribute much to acetate oxidation (Jenni et al., 2014). Furthermore, the anammox abundance was much lower than the denitrifying organisms. This would be due to the lower growth rate of anammox bacteria (yield coefficient of anammox bacteria: 0.066 ± 0.01 g VSS/ þ gNHþ 4 -N) compared to the denitrifiers (0.3 g VSS/g NH4 -N) (Strous et al., 1998; Sabumon, 2007). Similar results was also found in previous study (Cao et al., 2016a), which reported that the bacteria was diverse in the anammox UASB reactor and Candidatus Brocadia was the dominant species with very low proportion of 2.37%,  though high efficiency of NHþ 4 -N and NO2 -N removal activities was obtained. Above all, the co-existence of denitrifiers and anammox bacteria in the DEAMOX reactor clearly illustrated the synergetic

R. Du et al. / Chemosphere 174 (2017) 399e407

405

Fig. 5. Relative abundance of 16S rRNA gene sequencing of the DEAMOX system with taxonomic identities at phylum (a) and genus (b) level.

association for nitrogen removal. Additionally, Chloroflexi phylum constituted a large portion (>15%). Previous studies had found that Chloroflexi-like filamentous bacteria played an important role in the formation of biofilms or granules (Cho et al., 2010). The filamentous bacteria grew in anammox reactors had been reported (Chu et al., 2015). Moreover, low amounts of Nitrospira (0.18%) was also observed, which were often found in anammox reactors (Ma et al., 2015; Park et al., 2010). Sufficient NO 2 -N and limited amounts of dissolved oxygen in the influent wastewater would trigger the growth of Nitrospira. 3.5. Application potential of DEAMOX process in domestic wastewater treatment Conventional nitrogen removal processes require tremendous amount of energy for aeration and organic carbon for denitrification. Anammox process has been found to be a cost-effective nitrogen removal process by saving the operational costs of aeration and provision of organic carbon source (Kartal et al., 2010). However, the bottleneck problem of anammox application is

 maintaining a stable production of NO 2 -N. Generally, NO2 -N was produced from NHþ -N oxidation during the nitritation process. A 4 combination of nitritation and anammox is effective when treating high-ammonia contained wastewater (e.g. landfill leachate and digester liquid) (Joss et al., 2009; Li et al., 2014). Special controls were always carried out in order to achieve nitritation, including high temperature, pH and dissolved oxygen control, and ammonia oxidizing bacteria enrichment (Ge et al., 2015). However, these approaches were not deployable for full-scale domestic wastewater treatment, especially for low-strength wastewater, due to unstable performance and poor adaptation capability of inoculated bacteria (Mosquera-Corral et al., 2005). This study demonstrated the feasibility of NO 2 -N production from NO 3 -N via partial-denitrification process. Successful integration of anammox and partial-denitrification was established for cost-effective nitrogen removal from domestic wastewater and NO 3 -N sewage. Compared with nitritation-anammox process, the integrated DEAMOX process could be achieved with simple control and stable NO 2 -N production when treating low-strength wastewater. Moreover, domestic wastewater containing low

406

R. Du et al. / Chemosphere 174 (2017) 399e407

concentration of NHþ 4 -N and COD could be treated by mixed with the effluent of nitrification to achieve suitable influent ratio of NHþ 4N to NO 3 -N. On the other hand, it could simultaneously treat  wastewaters containing high amount of NHþ 4 -N and the NO3 -N-rich sewage from industry with low operational cost, since the aeration stage could be abridged and replaced by partial-denitrification  from NO 3 -N to NO2 -N reduction. On the other hand, wastewater containing high NO 3 -N and COD levels could also be treated with domestic wastewater by DEAMOX to save organic carbon sources for complete denitrification. Above all, the DEAMOX process provides great flexibility for nitrogen removal from diverse types of wastewater, which hold advantages of reducing operational complexity, saving aeration energy and carbon source cost in wastewater treatment. 4. Conclusion This study developed a DEAMOX process based on high NO 2 -N accumulated partial-denitrification and anammox in single SBR for simultaneous NO 3 -N sewage and domestic wastewater treatment.  The NO 2 -N accumulation from NO3 -N reduction maintained stably in 500-days operation with NTR over 70%. The maximum NHþ 4 -N and NO 3 -N removal efficiency of 94.7% and 97.8% was achieved  1 with influent NHþ and 69.24 mg L1, 4 -N and NO3 -N of 63.58 mg L respectively. The effluent TN concentration was below 6 mg L1, corresponding to the TN removal efficiency as high as 95.8%. The performance deteriorated due to excess partial-denitrification bacteria growth in the outer layer of sludge consortium and resulting in the mass transfer limitation for anammox bacteria inside. Regular washout of superficial sludge was necessary to maintain stable performance in the long-term operation of DEAMOX. Candidatus Brocadia was the major anammox bacteria (1.7%), and the dominant genus Thauera (26.33%) was presumably played important role in the high NO 2 -N accumulation during partialdenitrification process. Acknowledgements This study was funded by Natural Science Foundation of China (51478013) and the Funding Projects of Beijing Municipal Commission of Education. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.chemosphere.2017.02.013. References APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. American Public Health Association. Cho, S., Takahashi, Y., Fujii, N., Yamada, Y., Satoh, H., Okabe, S., 2010. Nitrogen removal performance and microbial community analysis of an anaerobic upflow granular bed anammox reactor. Chemosphere 78, 1129e1135. Cao, S., Wang, S., Peng, Y., Wu, C., Du, R., Gong, L., Ma, B., 2013. Achieving partial denitrification with sludge fermentation liquid as carbon source: the effect of seeding sludge. Bioresour. Technol. 149, 570e574. Cao, S., Du, R., Li, B., Ren, N., Peng, Y., 2016a. High-throughput profiling of microbial community structures in an ANAMMOX-UASB reactor treating high-strength wastewater. Appl. Microbiol. Biotechnol. 100, 6457e6467. Cao, S., Peng, Y., Du, R., Wang, S., 2016b. Feasibility of enhancing the DEnitrifying AMmonium Oxidation (DEAMOX) process for nitrogen removal by seeding partial denitrification sludg. Chemosphere 148, 403e407. Chu, Z.R., Wang, K., Li, X.K., Zhu, M.T., Yang, L., Zhang, J., 2015. Microbial characterization of aggregates within a one-stage nitritationeanammox system using high-throughput amplicon sequencing. Chem. Eng. J. 262, 41e48. Du, R., Peng, Y., Cao, S., Wu, C., Weng, D., Wang, S., He, J., 2014. Advanced nitrogen removal with simultaneous Anammox and denitrification in sequencing batch reactor. Bioresour. Technol. 162, 316e322. Du, R., Peng, Y., Cao, S., Wang, S., Wu, C., 2015a. Advanced nitrogen removal from

wastewater by combining anammox with partial denitrification. Bioresour. Technol. 179, 497e504. Du, R., Peng, Y., Cao, S., Li, B., Wang, S., Niu, M., 2015b. Mechanisms and microbial structure of partial denitrification with high nitrite accumulation. Appl. Microbiol. Biotechnol. 100, 2011e2021. Du, R., Cao, S., Li, B., Niu, M., Wang, S., Peng, Y., 2017. Performance and microbial community analysis of a novel DEAMOX based on partial-denitrification and anammox treating ammonia and nitrate wastewaters. Water Res. 108, 46e56. Erguder, T.H., Boon, N., Vlaeminck, S.E., Verstraete, W., 2008. Partial nitrification achieved by pulse sulfide doses in a sequential batch reactor. Environ. Sci. Technol. 42, 8715e8720. Fux, C., Huang, D., Monti, A., Siegrist, H., 2004. Difficulties in maintaining long-term partial nitritation of ammonium-rich sludge digester liquids in amoving-bed biofilmreactor (MBBR). Water Sci. Technol. 49, 53e60. Gao, F., Zhang, H.M., Yang, F.L., Qiang, H., Zhang, G.Y., 2012. The contrast study of anammox-denitrifying system in two non-woven fixed-bed bioreactors (NFBR) treating different low C/N ratio sewage. Bioresour. Technol 114, 54e61. Ge, S., Wang, S., Yang, X., Qiu, S., Li, B., Peng, Y., 2015. Detection of nitrifiers and evaluation of partial nitrification for wastewater treatment: a review. Chemosphere 2015 (140), 85e98. Hendrickx, T.L.G., Wang, Y., Kampman, C., Zeeman, G., Temmink, H., Buisman, C.J.N., 2012. Autotrophic nitrogen removal from low strength waste water at low temperature. Water Res. 46, 2187e2193. Jetten, M.S.M., Strous, M., van de Pas-Schoonen, K.T., Schalk, J., van Dongen, U., van de Graaf, A.A., Logemann, S., Muyzer, G., van Loosdrecht, M.C.M., Kuenen, J.G., 1998. The anaerobic oxidation of ammonium-N. FEMS Microbiol. Rev. 22, 421e437. €nig, R., Rottermann, K., Burger, S., Fabijan, P., Joss, A., Salzgeber, D., Eugster, J., Ko Leumann, S., Mohn, J., Siegrist, A.H., 2009. Full-scale nitrogen removal from digester liquid with partial nitritation and anammox in one SBR. Environ. Sci. Technol. 43, 5301e5306. Jenni, S., Vlaeminck, S.E., Morgenroth, E., Udert, K.M., 2014. Successful application of nitritation/anammox to wastewater with elevated organic carbon to ammonia ratios. Water Res. 49, 316e326. Kartal, B., Rattray, J., van Niftrik, L.A., van de Vossenberg, J., Schmid, M.C., Webb, R.I., Schouten, S., Fuerst, J.A., Damste, J.S.S., Jetten, M.S.M., Strous, M., 2007. Candidatus ‘‘Anammoxoglobus propionicus’’ a new propionate oxidizing speciesof anaerobic ammonium oxidizing bacteria. Syst. Appl. Microbiol. 30, 39e49. Kuenen, J.G., 2008. Anammox bacteria: from discovery to application. Nat. Rev. Microbiol. 6, 320e326. Kalyuzhnyi, S., Gladchenko, M., Mulder, A., Versprille, B., 2007. Comparison of quasisteady-state performance of the DEAMOX process under intermittent and continuous feeding and different nitrogen loading rates. Biotechnol. J. 2, 894e900. Kalyuzhnyi, S., Gladchenko, M., 2009. DEAMOX e new microbiological process of nitrogen removal from strong nitrogenous wastewater. Desalination 248, 783e793. Kartal, B., Kuenen, J.G., van Loosdrecht, M.C.M., 2010. Sewage treatment with anammox. Science 328, 702e703. Kartal, B., de Almeida, N.M., Maalcke, W.J., den Camp, H.J.M.O., Jetten, M.S.M., Keltjens, J.T., 2013. How to make a living from anaerobic ammonium oxidation. FEMS Microbiol. Rev. 37, 428e461. Liu, B., Zhang, F., Feng, X., Liu, Y., Yan, X., Zhang, X., Wang, L., Zhao, L., 2006. Thauera and Azoarcus as functionally important genera in a denitrifying quinolineremoval bioreactor as revealed by microbial community structure comparison. FEMS Microbiol. Ecol. 55, 274e286. Liu, B.B., Mao, Y.J., Bergaust, L., Bakken, L.R., Frostegard, A., 2013. Strains in the genus Thauera exhibit remarkably different denitrification regulatory phenotypes. Environ. Microbiol. 15, 2816e2828. Lackner, S., Gilbert, E.R.M., Vlaeminck, S.E., Joss, A., Horn, H., van Loosdrecht, M.C.M., 2014. Full-scale partial nitritation/anammox experiences - an application survey. Water Res. 55, 292e303. Langone, M., Yan, J., Haaijer, S.C.M., den Camp, H.J.M.O., Jetten, M.S.M., Andreottola, G., 2014. Coexistence of nitrifying, anammox and denitrifying bacteria in a sequencing batch reactor. Front. Microbiol. 5, 1e12. Li, H., Zhou, S., Ma, W., Huang, P., Huang, G., Qin, Y., Xu, B., Ouyang, H., 2014. Longterm performance and microbial ecology of a two-stage PNeANAMMOX process treating mature landfill leachate. Bioresour. Technol. 159, 404e411. Mulder, A., Van de Graaf, A.A., Robertson, L.A., Kuenen, J.G., 1995. Anaerobic ammonium oxidation discovered in a denitrifying fluidized-bed reactor. FEMS Microbiol. Ecol. 16 (3), 177e183. lez, F., Campos, J.L., Me ndez, R., 2005. Partial nitrification Mosquera-Corral, A., Gonza in a SHARON reactor in the presence of salts and organic carbon compounds. Process Biochem. 40 (9), 3109e3118. Ma, Y., Peng, Y., Wang, S., Yuan, Z., Wang, X., 2009. Achieving nitrogen removal via nitrite in a pilot-scale continuous pre-denitrification plant. Water Res. 43, 563e572. Ma, Q., Qu, Y., Shen, W., Zhang, Z., Wang, J., Liu, Z., Li, D., Li, H., Zhou, J., 2015. Bacterial community compositions of coking wastewater treatmentplants in steel industry revealed by Illumina high-throughputsequencing. Bioresour. Technol. 179, 436e443. Peng, Y., Zhang, S., Zeng, W., Zheng, S., Mino, T., Satoh, H., 2008. Organic removal by denitritation and methanogenesis and nitrogen removal by nitritation from landfill leachate. Water Res. 42, 883e892. Park, H., Rosenthal, A., Ramalingam, K., Fillos, J., Chandran, K., 2010. Linking

R. Du et al. / Chemosphere 174 (2017) 399e407 community profiles, gene expression and N-removal in anammox bioreactors treating municipal anaerobic digestion reject Water. Environ. Sci. Technol. 44, 6110e6116. Peng, Y.Z., Guo, J.H., Horn, H., Yang, X., Wang, S.Y., 2012. Achieving nitrite accumulation in a continuous system treating low-strength domestic wastewater: switchover from batch start-up to continuous operation with process control. Appl. Microbiol. Biotechnol. 94, 517e526. Strous, M., Heijnen, J.J., Kuenen, J.G., Jetten, M.S.M., 1998. The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms. Appl. Microbiol. Biotechnol. 50, 589e596. Sabumon, P.C., 2007. Anaerobic ammonia removal in presence of organic matter: a novel route. J. Hazard. Mater 149, 49e59. Schmid, M.C., Risgaard-Petersen, N., van de Vossenberg, J., Kuypers, M.M.M., Lavik, G., Petersen, J., Hulth, S., Thamdrup, B., Canfield, D., Dalsgaard, T., Rysgaard, S., Sejr, M.K., Strous, M., Op den Camp, H.J.M., Jetten, M.S.M., 2007. Anaerobic ammonium-oxidizing bacteria in marine environments: widespread

407

occurrence but low diversity. Environ. Microbiol. 9, 1476e1484. Takekawa, M., Park, G., Soda, S., Ike, M., 2014. Simultaneous anammox and denitrification (SAD) process in sequencing batch reactors. Bioresour. Technol. 174, 159e166. Van Dongen, U., Jetten, M.S.M., Van Loosdrecht, M.C.M., 2001. The SHARONAnammox process for treatment of ammonium rich wastewater. Water Sci. Technol. 44, 153e160. van de Graaf, A.A., de Bruijn, P., Robertson, L.A., Jetten, M.S.M., Kuenen, J.G., 1996. Autotrophic growth of anaerobic ammonium-oxidizing micro-organisms in a fluidized bed reactor. Microbiology 142, 2187e2196. Winkler, M.-K.H., Kleerebezem, R., van Loosdrecht, M.C.M., 2012. Integration of anammox into the aerobic granular sludge process for main stream wastewater treatment at ambient temperatures. Water Res. 136e144. Waki, M., Yasuda, T., Fukumoto, Y., Kuroda, K., Suzuki, K., 2013. Effect of electron donors on anammox coupling with nitrate reductionfor removing nitrogen from nitrate and ammonium. Bioresour. Technol. 130, 592e598.