- Email: [email protected]
Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature
Accepted Manuscript Short Communication Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature Yuhai Liang, Dong Li, Huiping Zeng, Cuidan Zhang, Jie Zhang PII: DOI: Reference:
S0960-8524(15)01105-0 http://dx.doi.org/10.1016/j.biortech.2015.08.003 BITE 15352
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
2 July 2015 30 July 2015 3 August 2015
Please cite this article as: Liang, Y., Li, D., Zeng, H., Zhang, C., Zhang, J., Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.08.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature Yuhai Liang a, Dong Li a ∗, Huiping Zeng a, Cuidan Zhang a, Jie Zhang a,b Key Laboratory of Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, Chinaa; State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, Chinab Abstract: The successful suppression of nitrite-oxidizing bacteria in the partial nitrification (PN) stage was the main challenge for the application of autotrophic nitrogen removal process treating mainstream sewage. In this study, two identical PN granular reactors (P1 and P2) were rapid started-up using the simultaneous PN and granulation strategy, for treating the domestic sewage. P1 was seeded with 30% PN granular sludge to induce nucleation, in which the granule size achieved to more than 400 µm in 12 d, with ammonia oxidation rate and nitrite accumulation rate of 80% and 95%, respectively, while P2 realized granulation in 42 d. The presence of organic matters and specific structure of granules were profitable for the stability of PN for treating sewage with low ammonia. High-throughput pyrosequencing results indicated the biodiversity of both reactors decreased after start-up, and Nitrosomonas was the predominant specie of aerobic ammonia-oxidizing bacteria in PN granular sludge. Keywords: partial nitrification; granular sludge; domestic sewage; High-throughput pyrosequencing
∗
Key Laboratory of Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, China, [email protected] (D. Li). Tel.:+8601067392099-4. 1
1 Introduction The autotrophic nitrogen removal process in base of partial nitrification (PN) and anaerobic ammonium oxidation (anammox) has been successfully applied in hundred sewage plants, as one efficient and economic nitrogen removal process (Lackner et al., 2014). However, most of the sewage treated in those plants were sidestream sewage with high temperature (>30°C) and high ammonia concentration (>100mg L-1). With the development of autotrophic nitrogen removal process and energy autarkic wastewater treatment plant, the process has been used to treat mainstream sewage such as domestic sewage with low temperature (<30°C) and low ammonia concentration (<100 mg L-1) (Khiewwijit et al., 2015; Wett et al., 2013). But the nitrate accumulation significantly affected the stable operation in the present autotrophic nitrogen removal plants (Lackner et al., 2014). So the effective suppression on nitrite-oxidizing bacteria (NOB) was the key factor to ensure the stable operation. For the domestic sewage with low temperature and low ammonia, the suppression on NOB was certainly an inevitable and major challenge in PN process. As previously reported, many environmental factors could effectively inhibit NOB, including high free ammonia (FA) (Hulle et al., 2010), high free nitrous acid (FNA) (Gabarro et al., 2012), low dissolved oxygen (DO) (Hulle et al., 2010), high temperature and suitable sludge retention time (SRT) (Hulle et al., 2010). Whereas the ammonia concentration and temperature was relative low in domestic sewage, so the FA, FNA and temperature could not effectively suppress NOB. Although low DO 2
could effectively suppress NOB, it would also limit the reaction rate of PN. And SRT was a secondary factor that needs to combine with other factors to inhibit NOB. In conclusion, new strategy for suppressing NOB was necessary in the domestic sewage. The organic matters exist in domestic sewage could induce heterotrophic bacteria, which would compete DO with NOB (Mosquera-Corral et al., 2005). In addition, the high biomass quantity, profitable settle ability, specific microbial distribution and mass transfer gradient of aerobic granular sludge, make it be an interesting alternative for PN of domestic sewage (Lopez-Palau et al., 2011). PN granular sludge has been successfully cultivated in several studies (Lopez-Palau et al., 2011; Shi et al., 2011; Vázquez-Padín et al., 2010; Wan et al., 2013). They cultivated the PN granular sludge with ammonia concentration of 800, 350, 400 and 125 mg L-1. These researches about PN granular sludge focused on the sewage with high ammonia, and few reports were available about domestic sewage with low ammonia. Moreover, the variations of microbial characteristic should be deeply investigated after granulation. In this study, PN granular sludge was cultivated in two SBR reactors, using the strategy of simultaneous PN and granulation, and one reactor was seeded with a small amount of granules to shorten the start-up period. Then the PN granular sludge was applied for treating domestic sewage at room temperature. High-throughput pyrosequencing was used to detect the variations of microbial characteristic of the sludge after granulation. 2 Materials and Methods 2.1 Experimental setup 3
Two reactors (P1 and P2) with identical setup (diameter: 80 mm, height: 1000 mm, effective volume: 4.0 L) were adopted in this study. During the experiment, the reaction temperature was controlled around 25°C by the water bath, and the aeration rate was set as 2.0 L min-1 with DO as 4.0-8.0 mg L-1. The seeded sludge for P1was nitrification activated floc sludge together with 30% PN granular sludge, while that of P2 was totally the nitrification floc sludge. The mixed liquid suspended solids (MLSS) of the seeded sludge was 6.5 g L-1. P1 and P2 was operated as follows: inflow 1 min; aeration 60-180 min (adjusted with the cycle experiment results); settling time 10 min (P1) and 18 min (P2); outflow 2 min; volumetric exchange ratio: 62.5%. The domestic sewage used in this study was obtained from a living quarter of a university, and the detailed contents as follows: chemical oxygen demand (COD, 250-350 mg L-1), NH4+-N (60.0-90.0 mg L-1), NO2--N (0-1.0 mg L-1), NO3--N (0-3 mg L-1), alkalinity (as CaCO3 : 492-614 mg·L-1) and pH (7.3-7.7). During phase I (day 1-30), (NH4)2SO4 and NaHCO3 were added to the domestic sewage to increase NH4 +-N (110-130 mg L-1) and alkalinity (as CaCO3: 1000-1600 mg·L-1) for the rapid start-up. In phase II (day 31-60), both reactors were used to treat the domestic sewage directly. 2.2 Analytical methods According to Standard Methods (APHA, 1998), the NH4+-N, NO2--N and NO3--N concentrations in influent and effluent were daily analyzed by ultraviolet and visible spectrophotometers (UV755b, Youke, China). COD was detected by 5B-1digestion instrument. The temperature, pH and DO were measured with online 4
pH/DO instruments (pH296/Oxi296, WTW, Germany). The granule size were detected by laser particle analyzer (Malvern Mastersizer 2000, U.K.). Ammonia removal efficiency (ARE) and nitrite accumulation rate (NAR) were calculated as Eq. (1) and (2), respectively. ARE =
NAR =
[ NH 4+ ]Inf . − [ NH 4+ ]Eff . [ NH 4+ ]Inf .
×100%
− NO2 - N ×100% NO2 - N + NO3− - N
(1) (2)
−
2.3 Microbial sampling and phylogenetic assignment In this study, the seeded sludge and the granular sludge on day 60 were obtained from the two reactors. According to the methods (Zhang et al., 2015), the DNA were extracted from each sludge samples. The extracted DNA was used for high-throughput pyrosequencing and phylogenetic assignment (Sangon, China). 3 Results and discussion 3.1 Reactor performance The reactor performance of P1 was depicted in Fig.1 (A). During phase I (day 1-30), the influent ammonia was between 110-130 mg L-1. By adjusting the aeration time, the ARE reached around 80%. Since 30% PN granular sludge was added to P1, and the FA in P1 was around 5-12 mg L-1, NOB was effectively suppressed. As a result, the NAR was always more than 95%. In phase II (day 31-60), no additional ammonia was supplied to the domestic sewage. The influent ammonia decreased to 40-60 mg L-1, and the FA was 1-3 mg L-1. So aeration time was shortened to ensure the ARE higher than 80%. And the NAR was still more than 95%. This results suggested that the high FA in the initial period had effectively suppressed or washed 5
out the NOB, and the PN granular sludge performed a stable ability under the condition of low FA. After an initial decrease, the average particle size of P1 increased to more than 400 µm on day 12, and then achieved to 922 µm on day 60 (Appendix Fig. A). The MLSS was decreased from 6.5 g L-1 to 2.7 g L-1.This results proved that the simultaneous PN and granulation, and together with the inducement of a small addition of PN granular sludge was an effective strategy for the start-up of PN granular reactor. The reactor performance of P2 was depicted in Fig.1 (B). Since the seeded sludge of P2 was totally the nitrification floc sludge, the initial PN ability was worse than P1. The ARE and NAR reached 60% and 95% on day 12, respectively. It indicated that the NOB was effectively inhibited by high FA. The same as P1, the influent was changed to domestic sewage without addition of ammonia from day 31. The ARE was increased to more than 80% by adjusting the aeration time, and the NAR was always higher than 95% in phase II. This results also indicated that the PN granular sludge was suitable for treating domestic sewage. To compare with P1, the MLSS of P2 was decreased significantly from 6.5 g L-1 to 2.4 g L-1. The average granule size gradually increased from 195 µm to 400 µm on day 42, and to 625 µm on day 60 (Appendix Fig. A). These results shown that the strategy of simultaneous PN and granulation could shorten the start-up period, and the addition of granular sludge could further shorten the period. And the PN granular sludge could apply for treating domestic sewage at room temperature. 3.2 Influence factors for rapid start-up and stable operation of PN granular sludge 6
In this study, ammonia was added to the influent to elevate the ammonia concentration to 110-130 mg L-1 during phase I, and the FA was 5-12 mg L-1, which effectively suppress NOB. On the other hand, the short settling time of P1 and P2 (10 and 18 min, respectively) together with the air upflow velocity of 0.425 cm s-1, made a favorable condition for the start-up of granulation (Adav et al., 2007). By controlling the above conditions, the PN and granulation were simultaneously achieved, which effectively shortened the start-up period of PN granular sludge. Moreover, addition of small PN granular sludge could induce granulation, and further shorten the period (Pijuan et al., 2011). The high air upflow velocity not only accelerate the granulation, but also lead to a high DO concentration in the reactor (4.0-8.0 mg L-1), which was adverse to the PN. However, both reactors showed prominent PN ability for treating domestic sewage with low ammonia at room temperature. The possible reasons could be as follows: Firstly, the domestic sewage used in this study contained organic matters, and the influent COD was about 250-350 mg L-1, while the effluent was below 50 mg L-1. The heterotrophic bacteria would consume massive oxygen to oxidize the organic matters, and then micro anoxic environment could be formed. NOB could not outcompete DO with the heterotrophic bacteria and aerobic ammonia-oxidizing bacteria (AOB). So the presence of organic matters was profitable for the suppression of NOB, and then ensured the stable operation of PN. When the influent was changed to totally domestic sewage, the FA decreased to1-3 mg L-1. The FA could not effectively inhibit NOB since it was lower than the suppression concentration (5-10 mg L-1). So the effective 7
inhibition of NOB by organic matters could be proved by the reactor performance which could also sustain a suitable ARE and NAR. Secondly, the granule structure was profitable for PN when compared to the floc sludge. In the granular sludge, heterotrophic bacteria survives in the outer layer to consume oxygen for the COD oxidation, and the oxygen gradient makes the inner layer be oxygen-limited environment, in which the AOB outcompete with NOB and got dominate (Terada et al., 2010). The combined effects of all the above factors made the rapid start-up and stable operation with low ammonia of PN granular reactor. 3.3 Microbial community The seeded sludge and PN granular sludge on day 60 of P1 and P2 were used for high-throughput pyrosequencing, to detect the microbial variation of the PN granular sludge. And P11, P21, P12 and P22 represented the seeded sludge of P1, seeded sludge of P2, PN granular sludge of P1 and PN granular sludge of P2, respectively. Operational taxonomic units (OTUs) for community analysis were defined at 3 % variation in sequences, and OTUs were regarded as relating to genus. For P1, from phase I to II, 1249 OTUs were washed out while 1125 OTUs emerged, and 674 OTUs existed in both samples of P1, which took a share of 22.11% in all the OTUs. For P2, 1241 OTUs were washed out while 935 OTUs emerged, and 540 OTUs existed in both samples of P2, which took a share of 19.88% in all the OTUs. These results suggested that the microbial communities have significantly changed in both reactors after PN granulation. The Shannon index of P11, P12, p21 and P22 were 5.817, 5.385, 5.860 and 5.460, respectively. These results indicated that the biodiversity of both 8
reactors decreased in phase II. The PN and granulation washed out the organisms, which could not adapt to the influent quality and hydraulic conditions. The genus results were shown in Fig.2. The uncultured_Saprospiraceae, Thiothrix and Thauera were detected in the samples, which had the ability to degrade organic matters. The share of uncultured_Saprospiraceae, Thiothrix and Thauera in P1 were increased from 5.59% to 11.84%, from 0.10% to 8.17%, and from 1.10% to 3.18%, respectively. The share of uncultured_Saprospiraceae, Thiothrix and Thauera in P2 were increased from 3.24% to 14.1 %, from 0.31% to 3.85%, and from 1.27% to 4.67%, respectively. The presence of heterotrophic bacteria was profitable for the suppression of NOB, and then ensured the stable operation of PN. In PN system, AOB was the most important functional bacteria, while NOB should be inhibited. And 145 AOB sequences were detected from P11, in which 113 sequences were related to Nitrosomonas, 14 sequences was related to Nitrosococcus, and 18 sequences was related to uncultured_Nitrosomonadaceae. 24 NOB sequences were detected, all of which were related to Candidatus Nitrotoga. For P12, 155 AOB sequences were detected, in which 145, 1 and 9 sequences were related to Nitrosomonas, Nitrosococcus and uncultured_Nitrosomonadaceae, respectively. 6 NOB sequences were detected, in which one was related to Nitrospira and 5 sequences were related to Candidatus Nitrotoga. During the experimental period, the share of AOB and NOB in P1 were decreased from 0.94% to 0.85%, and 0.16% to 0.04%, respectively. This semi-quantitily indicated the washout of NOB and the prominent PN ability of P1. The Nitrosococcus sequences decreased to 1 from 14, indicating the adverse effect on 9
this specie of PN granular system. Nitrosomonas was the predominant AOB in P1. For P2, 116 AOB sequences were detected in P21, in which 82, 19 and 15 sequences were related to Nitrosomonas, Nitrosococcus and uncultured_Nitrosomonadaceae, respectively. 24 NOB sequences were detected, in which 2 sequences were Nitrospira and 22 sequences were Candidatus Nitrotoga. 39 AOB sequences were detected in P22, in which 30 sequences were Nitrosomonas and 9 sequences were uncultured_Nitrosomonadaceae. No NOB sequences were detected. the share of AOB and NOB in P2 were decreased from 1.06% to 0.33%, and 0.22% to 0, respectively. Both AOB and NOB in P2 decreased significantly, indicating the greater impact on microbial community of PN and granulation. Nitrosomonas was also the predominant AOB in the P2. NOB was not detected, suggesting the more effective washout of NOB by seeding floc sludge. The two reactors ran for 30 d in the condition of low ammonia, high DO and room temperature, and the quantity of NOB had not increased. P1 and P2 showed prominent stability of PN, indicating the feasibility of PN granules for treating domestic sewage. And the successful PN would provide a reliable influent to the subsequent anammox process, thus promote application of autotrophic nitrogen removal process in treatment of mainstream sewage. 4 Conclusions The simultaneous PN and granulation could significantly shorten the start-up period, and the addition of granular sludge could further shorten the period. The presence of organic matters was profitable for the PN granulation. The ammonia oxidation rate and PN rate were about 80% and 95%, respectively. The specific 10
microbial community and structure of PN granular sludge made it feasible to treat domestic sewage with low ammonia concentration at room temperature. High-throughput pyrosequencing results suggested the biodiversity decreasing of both reactors after the start-up of PN granulation. Nitrosomonas was the dominant AOB in the PN granular system. Acknowledgements This work was supported by water project of National Science and Technology Major Project (Grant No.2012ZX07202-005). References 1. Adav, S.S., Lee, D.J., Lai, J.Y. 2007. Effects of aeration intensity on formation of phenol-fed aerobic granules and extracellular polymeric substances. Appl Microbiol Biotechnol, 77(1), 175-182. 2. APHA. 1998. Standard methods for the examination of water and wastewater, 20th ed., American Public Health Association. Washington, DC, USA. 3. Gabarro, J., Ganigue, R., Gich, F., Ruscalleda, M., Balaguer, M.D., Colprim, J. 2012. Effect of temperature on AOB activity of a partial nitritation SBR treating landfill leachate with extremely high nitrogen concentration. Bioresour Technol, 126, 283-289. 4. Hulle, S.W.H.V., Vandeweyer, H.J.P., Meesschaert, B.D., Vanrolleghem, P.A., Pascal Dejansb, A.D. 2010. Engineering aspects and practical application of autotrophic nitrogen removal from nitrogen rich streams. Chemical Engineering Journal, 162(1), 1-20. 11
5. Khiewwijit, R., Temmink, H., Rijnaarts, H., Keesman, K.J. 2015. Energy and nutrient recovery for municipal wastewater treatment How to design a feasible plant layout. Environmental Modelling & Software, 68, 156-165. 6. Lackner, S., Gilbert, E.M., Vlaeminck, S.E., Joss, A., Horn, H., van Loosdrecht, M.C. 2014. Full-scale partial nitritation/anammox experiences--an application survey. Water Res, 55, 292-303. 7. Lopez-Palau, S., Pericas, A., Dosta, J., Mata-Alvarez, J. 2011. Partial nitrification of sludge reject water by means of aerobic granulation. Water Sci Technol, 64(9), 1906-1912. 8. Mosquera-Corral, A., Gonza´lez, F., Campos, J.L., Me´ndez, R. 2005. Partial nitrification in a SHARON reactor in the presence of salts and organic carbon compounds. Process Biochemistry, 40(9), 3109–3118. 9. Pijuan, M., Werner, U., Yuan, Z. 2011. Reducing the startup time of aerobic granular sludge reactors through seeding floccular sludge with crushed aerobic granules. Water Res, 45(16), 5075-5083. 10. Shi, Y.J., Wang, X.H., Yu, H.B., Xie, H.J., Teng, S.X., Sun, X.F., Tian, B.H., Wang, S.G. 2011. Aerobic granulation for nitrogen removal via nitrite in a sequencing batch reactor and the emission of nitrous oxide. Bioresour Technol, 102(3), 2536-2541. 11. Terada, A., Lackner, S., Kristensen, K., Smets, B.F. 2010. Inoculum effects on community composition and nitritation performance of autotrophic nitrifying biofilm reactors with counter-diffusion geometry. Environ Microbiol, 12(10), 12
2858-2872. 12. Vázquez-Padín, J.R., Figueroa, M., Campos, J.L., Mosquera-Corral, A., Méndez, R. 2010. Nitrifying granular systems_ A suitable technology to obtain stable partial nitrification at room temperature. Separation and Purification Technology, 74(2), 178-186. 13. Wan, C., Sun, S., Lee, D.J., Liu, X., Wang, L., Yang, X., Pan, X. 2013. Partial nitrification using aerobic granules in continuous-flow reactor: rapid startup. Bioresour Technol, 142, 517-522. 14. Wett, B., Omari, A., Podmirseg, S.M., Han, M., Akintayo, O., Gomez Brandon, M., Murthy, S., Bott, C., Hell, M., Takacs, I., Nyhuis, G., O'Shaughnessy, M. 2013. Going for mainstream deammonification from bench to full scale for maximized resource efficiency. Water Sci Technol, 68(2), 283-289. 15. Zhang, X., Zhang, H., Ye, C., Wei, M., Du, J. 2015. Effect of COD/N ratio on nitrogen removal and microbial communities of CANON process in membrane bioreactors. Bioresour Technol, 189, 302-308.
13
effluent
ARE
NAR
160
140
140
-1 nitrite concentration（ mg·L ）
120 100 80 60 40 20 0 0
10
20
40
50
120 100 80 60 40 20 0
60
0
10
20
Time（ d）
8
50
60 1200
5 4 3 2 1
80 800 60 600 40 400 20
0
granule size（ µm）
1000
6
200
0 10
20
30
40
50
60
0
10
20
Time（ d） effluent
ARE 160
140
140
-1 nitrite concentration（ mg·L ）
160
120 100 80 60 40 20 0 0
10
20
30
40
50
NAR
40
50
60
50
60
granule size
120 100 80 60 40 20 0
60
0
10
20
Time（ d）
50
30 Time（ d）
influent
-1 ammonia concentration（ mg·L ）
40
100
7
0
30
40
Time（ d）
1200
100
45
1000
40 35
ARE， NAR（ %）
-1 nitrate concentration（ mg· L ）
30 Time（ d）
ARE， NAR（ %）
-1 nitrate concentration（ mg· L ）
30
granule size
30 25 20 15 10
80 800 60 600 40 400 20
200
5 0
0 0
10
20
30
40
50
60
0
Time（ d）
10
20
30
40
50
60
Time（ d）
Fig.1 Nitrogen removal performance and granule size of P1 (A) and P2 (B)
14
granule size（ µm）
-1 ammonia concentration（ mg·L ）
influent 160
Fig.2 The taxonomic identities of sequences retrieved from P1 and P2 at the Genus level
15
HIGHLIGHTS Partial nitrification granular sludge was rapid started-up. Addition of granular sludge could significantly shorten the start-up period. Partial nitrification granular sludge was feasible for treating domestic sewage. Microbial characteristics were analyzed by high-throughput pyrosequencing.
16
S0960-8524(15)01105-0 http://dx.doi.org/10.1016/j.biortech.2015.08.003 BITE 15352
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
2 July 2015 30 July 2015 3 August 2015
Please cite this article as: Liang, Y., Li, D., Zeng, H., Zhang, C., Zhang, J., Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.08.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid start-up and microbial characteristics of partial nitrification granular sludge treating domestic sewage at room temperature Yuhai Liang a, Dong Li a ∗, Huiping Zeng a, Cuidan Zhang a, Jie Zhang a,b Key Laboratory of Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, Chinaa; State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, Chinab Abstract: The successful suppression of nitrite-oxidizing bacteria in the partial nitrification (PN) stage was the main challenge for the application of autotrophic nitrogen removal process treating mainstream sewage. In this study, two identical PN granular reactors (P1 and P2) were rapid started-up using the simultaneous PN and granulation strategy, for treating the domestic sewage. P1 was seeded with 30% PN granular sludge to induce nucleation, in which the granule size achieved to more than 400 µm in 12 d, with ammonia oxidation rate and nitrite accumulation rate of 80% and 95%, respectively, while P2 realized granulation in 42 d. The presence of organic matters and specific structure of granules were profitable for the stability of PN for treating sewage with low ammonia. High-throughput pyrosequencing results indicated the biodiversity of both reactors decreased after start-up, and Nitrosomonas was the predominant specie of aerobic ammonia-oxidizing bacteria in PN granular sludge. Keywords: partial nitrification; granular sludge; domestic sewage; High-throughput pyrosequencing
∗
Key Laboratory of Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, China, [email protected] (D. Li). Tel.:+8601067392099-4. 1
1 Introduction The autotrophic nitrogen removal process in base of partial nitrification (PN) and anaerobic ammonium oxidation (anammox) has been successfully applied in hundred sewage plants, as one efficient and economic nitrogen removal process (Lackner et al., 2014). However, most of the sewage treated in those plants were sidestream sewage with high temperature (>30°C) and high ammonia concentration (>100mg L-1). With the development of autotrophic nitrogen removal process and energy autarkic wastewater treatment plant, the process has been used to treat mainstream sewage such as domestic sewage with low temperature (<30°C) and low ammonia concentration (<100 mg L-1) (Khiewwijit et al., 2015; Wett et al., 2013). But the nitrate accumulation significantly affected the stable operation in the present autotrophic nitrogen removal plants (Lackner et al., 2014). So the effective suppression on nitrite-oxidizing bacteria (NOB) was the key factor to ensure the stable operation. For the domestic sewage with low temperature and low ammonia, the suppression on NOB was certainly an inevitable and major challenge in PN process. As previously reported, many environmental factors could effectively inhibit NOB, including high free ammonia (FA) (Hulle et al., 2010), high free nitrous acid (FNA) (Gabarro et al., 2012), low dissolved oxygen (DO) (Hulle et al., 2010), high temperature and suitable sludge retention time (SRT) (Hulle et al., 2010). Whereas the ammonia concentration and temperature was relative low in domestic sewage, so the FA, FNA and temperature could not effectively suppress NOB. Although low DO 2
could effectively suppress NOB, it would also limit the reaction rate of PN. And SRT was a secondary factor that needs to combine with other factors to inhibit NOB. In conclusion, new strategy for suppressing NOB was necessary in the domestic sewage. The organic matters exist in domestic sewage could induce heterotrophic bacteria, which would compete DO with NOB (Mosquera-Corral et al., 2005). In addition, the high biomass quantity, profitable settle ability, specific microbial distribution and mass transfer gradient of aerobic granular sludge, make it be an interesting alternative for PN of domestic sewage (Lopez-Palau et al., 2011). PN granular sludge has been successfully cultivated in several studies (Lopez-Palau et al., 2011; Shi et al., 2011; Vázquez-Padín et al., 2010; Wan et al., 2013). They cultivated the PN granular sludge with ammonia concentration of 800, 350, 400 and 125 mg L-1. These researches about PN granular sludge focused on the sewage with high ammonia, and few reports were available about domestic sewage with low ammonia. Moreover, the variations of microbial characteristic should be deeply investigated after granulation. In this study, PN granular sludge was cultivated in two SBR reactors, using the strategy of simultaneous PN and granulation, and one reactor was seeded with a small amount of granules to shorten the start-up period. Then the PN granular sludge was applied for treating domestic sewage at room temperature. High-throughput pyrosequencing was used to detect the variations of microbial characteristic of the sludge after granulation. 2 Materials and Methods 2.1 Experimental setup 3
Two reactors (P1 and P2) with identical setup (diameter: 80 mm, height: 1000 mm, effective volume: 4.0 L) were adopted in this study. During the experiment, the reaction temperature was controlled around 25°C by the water bath, and the aeration rate was set as 2.0 L min-1 with DO as 4.0-8.0 mg L-1. The seeded sludge for P1was nitrification activated floc sludge together with 30% PN granular sludge, while that of P2 was totally the nitrification floc sludge. The mixed liquid suspended solids (MLSS) of the seeded sludge was 6.5 g L-1. P1 and P2 was operated as follows: inflow 1 min; aeration 60-180 min (adjusted with the cycle experiment results); settling time 10 min (P1) and 18 min (P2); outflow 2 min; volumetric exchange ratio: 62.5%. The domestic sewage used in this study was obtained from a living quarter of a university, and the detailed contents as follows: chemical oxygen demand (COD, 250-350 mg L-1), NH4+-N (60.0-90.0 mg L-1), NO2--N (0-1.0 mg L-1), NO3--N (0-3 mg L-1), alkalinity (as CaCO3 : 492-614 mg·L-1) and pH (7.3-7.7). During phase I (day 1-30), (NH4)2SO4 and NaHCO3 were added to the domestic sewage to increase NH4 +-N (110-130 mg L-1) and alkalinity (as CaCO3: 1000-1600 mg·L-1) for the rapid start-up. In phase II (day 31-60), both reactors were used to treat the domestic sewage directly. 2.2 Analytical methods According to Standard Methods (APHA, 1998), the NH4+-N, NO2--N and NO3--N concentrations in influent and effluent were daily analyzed by ultraviolet and visible spectrophotometers (UV755b, Youke, China). COD was detected by 5B-1digestion instrument. The temperature, pH and DO were measured with online 4
pH/DO instruments (pH296/Oxi296, WTW, Germany). The granule size were detected by laser particle analyzer (Malvern Mastersizer 2000, U.K.). Ammonia removal efficiency (ARE) and nitrite accumulation rate (NAR) were calculated as Eq. (1) and (2), respectively. ARE =
NAR =
[ NH 4+ ]Inf . − [ NH 4+ ]Eff . [ NH 4+ ]Inf .
×100%
− NO2 - N ×100% NO2 - N + NO3− - N
(1) (2)
−
2.3 Microbial sampling and phylogenetic assignment In this study, the seeded sludge and the granular sludge on day 60 were obtained from the two reactors. According to the methods (Zhang et al., 2015), the DNA were extracted from each sludge samples. The extracted DNA was used for high-throughput pyrosequencing and phylogenetic assignment (Sangon, China). 3 Results and discussion 3.1 Reactor performance The reactor performance of P1 was depicted in Fig.1 (A). During phase I (day 1-30), the influent ammonia was between 110-130 mg L-1. By adjusting the aeration time, the ARE reached around 80%. Since 30% PN granular sludge was added to P1, and the FA in P1 was around 5-12 mg L-1, NOB was effectively suppressed. As a result, the NAR was always more than 95%. In phase II (day 31-60), no additional ammonia was supplied to the domestic sewage. The influent ammonia decreased to 40-60 mg L-1, and the FA was 1-3 mg L-1. So aeration time was shortened to ensure the ARE higher than 80%. And the NAR was still more than 95%. This results suggested that the high FA in the initial period had effectively suppressed or washed 5
out the NOB, and the PN granular sludge performed a stable ability under the condition of low FA. After an initial decrease, the average particle size of P1 increased to more than 400 µm on day 12, and then achieved to 922 µm on day 60 (Appendix Fig. A). The MLSS was decreased from 6.5 g L-1 to 2.7 g L-1.This results proved that the simultaneous PN and granulation, and together with the inducement of a small addition of PN granular sludge was an effective strategy for the start-up of PN granular reactor. The reactor performance of P2 was depicted in Fig.1 (B). Since the seeded sludge of P2 was totally the nitrification floc sludge, the initial PN ability was worse than P1. The ARE and NAR reached 60% and 95% on day 12, respectively. It indicated that the NOB was effectively inhibited by high FA. The same as P1, the influent was changed to domestic sewage without addition of ammonia from day 31. The ARE was increased to more than 80% by adjusting the aeration time, and the NAR was always higher than 95% in phase II. This results also indicated that the PN granular sludge was suitable for treating domestic sewage. To compare with P1, the MLSS of P2 was decreased significantly from 6.5 g L-1 to 2.4 g L-1. The average granule size gradually increased from 195 µm to 400 µm on day 42, and to 625 µm on day 60 (Appendix Fig. A). These results shown that the strategy of simultaneous PN and granulation could shorten the start-up period, and the addition of granular sludge could further shorten the period. And the PN granular sludge could apply for treating domestic sewage at room temperature. 3.2 Influence factors for rapid start-up and stable operation of PN granular sludge 6
In this study, ammonia was added to the influent to elevate the ammonia concentration to 110-130 mg L-1 during phase I, and the FA was 5-12 mg L-1, which effectively suppress NOB. On the other hand, the short settling time of P1 and P2 (10 and 18 min, respectively) together with the air upflow velocity of 0.425 cm s-1, made a favorable condition for the start-up of granulation (Adav et al., 2007). By controlling the above conditions, the PN and granulation were simultaneously achieved, which effectively shortened the start-up period of PN granular sludge. Moreover, addition of small PN granular sludge could induce granulation, and further shorten the period (Pijuan et al., 2011). The high air upflow velocity not only accelerate the granulation, but also lead to a high DO concentration in the reactor (4.0-8.0 mg L-1), which was adverse to the PN. However, both reactors showed prominent PN ability for treating domestic sewage with low ammonia at room temperature. The possible reasons could be as follows: Firstly, the domestic sewage used in this study contained organic matters, and the influent COD was about 250-350 mg L-1, while the effluent was below 50 mg L-1. The heterotrophic bacteria would consume massive oxygen to oxidize the organic matters, and then micro anoxic environment could be formed. NOB could not outcompete DO with the heterotrophic bacteria and aerobic ammonia-oxidizing bacteria (AOB). So the presence of organic matters was profitable for the suppression of NOB, and then ensured the stable operation of PN. When the influent was changed to totally domestic sewage, the FA decreased to1-3 mg L-1. The FA could not effectively inhibit NOB since it was lower than the suppression concentration (5-10 mg L-1). So the effective 7
inhibition of NOB by organic matters could be proved by the reactor performance which could also sustain a suitable ARE and NAR. Secondly, the granule structure was profitable for PN when compared to the floc sludge. In the granular sludge, heterotrophic bacteria survives in the outer layer to consume oxygen for the COD oxidation, and the oxygen gradient makes the inner layer be oxygen-limited environment, in which the AOB outcompete with NOB and got dominate (Terada et al., 2010). The combined effects of all the above factors made the rapid start-up and stable operation with low ammonia of PN granular reactor. 3.3 Microbial community The seeded sludge and PN granular sludge on day 60 of P1 and P2 were used for high-throughput pyrosequencing, to detect the microbial variation of the PN granular sludge. And P11, P21, P12 and P22 represented the seeded sludge of P1, seeded sludge of P2, PN granular sludge of P1 and PN granular sludge of P2, respectively. Operational taxonomic units (OTUs) for community analysis were defined at 3 % variation in sequences, and OTUs were regarded as relating to genus. For P1, from phase I to II, 1249 OTUs were washed out while 1125 OTUs emerged, and 674 OTUs existed in both samples of P1, which took a share of 22.11% in all the OTUs. For P2, 1241 OTUs were washed out while 935 OTUs emerged, and 540 OTUs existed in both samples of P2, which took a share of 19.88% in all the OTUs. These results suggested that the microbial communities have significantly changed in both reactors after PN granulation. The Shannon index of P11, P12, p21 and P22 were 5.817, 5.385, 5.860 and 5.460, respectively. These results indicated that the biodiversity of both 8
reactors decreased in phase II. The PN and granulation washed out the organisms, which could not adapt to the influent quality and hydraulic conditions. The genus results were shown in Fig.2. The uncultured_Saprospiraceae, Thiothrix and Thauera were detected in the samples, which had the ability to degrade organic matters. The share of uncultured_Saprospiraceae, Thiothrix and Thauera in P1 were increased from 5.59% to 11.84%, from 0.10% to 8.17%, and from 1.10% to 3.18%, respectively. The share of uncultured_Saprospiraceae, Thiothrix and Thauera in P2 were increased from 3.24% to 14.1 %, from 0.31% to 3.85%, and from 1.27% to 4.67%, respectively. The presence of heterotrophic bacteria was profitable for the suppression of NOB, and then ensured the stable operation of PN. In PN system, AOB was the most important functional bacteria, while NOB should be inhibited. And 145 AOB sequences were detected from P11, in which 113 sequences were related to Nitrosomonas, 14 sequences was related to Nitrosococcus, and 18 sequences was related to uncultured_Nitrosomonadaceae. 24 NOB sequences were detected, all of which were related to Candidatus Nitrotoga. For P12, 155 AOB sequences were detected, in which 145, 1 and 9 sequences were related to Nitrosomonas, Nitrosococcus and uncultured_Nitrosomonadaceae, respectively. 6 NOB sequences were detected, in which one was related to Nitrospira and 5 sequences were related to Candidatus Nitrotoga. During the experimental period, the share of AOB and NOB in P1 were decreased from 0.94% to 0.85%, and 0.16% to 0.04%, respectively. This semi-quantitily indicated the washout of NOB and the prominent PN ability of P1. The Nitrosococcus sequences decreased to 1 from 14, indicating the adverse effect on 9
this specie of PN granular system. Nitrosomonas was the predominant AOB in P1. For P2, 116 AOB sequences were detected in P21, in which 82, 19 and 15 sequences were related to Nitrosomonas, Nitrosococcus and uncultured_Nitrosomonadaceae, respectively. 24 NOB sequences were detected, in which 2 sequences were Nitrospira and 22 sequences were Candidatus Nitrotoga. 39 AOB sequences were detected in P22, in which 30 sequences were Nitrosomonas and 9 sequences were uncultured_Nitrosomonadaceae. No NOB sequences were detected. the share of AOB and NOB in P2 were decreased from 1.06% to 0.33%, and 0.22% to 0, respectively. Both AOB and NOB in P2 decreased significantly, indicating the greater impact on microbial community of PN and granulation. Nitrosomonas was also the predominant AOB in the P2. NOB was not detected, suggesting the more effective washout of NOB by seeding floc sludge. The two reactors ran for 30 d in the condition of low ammonia, high DO and room temperature, and the quantity of NOB had not increased. P1 and P2 showed prominent stability of PN, indicating the feasibility of PN granules for treating domestic sewage. And the successful PN would provide a reliable influent to the subsequent anammox process, thus promote application of autotrophic nitrogen removal process in treatment of mainstream sewage. 4 Conclusions The simultaneous PN and granulation could significantly shorten the start-up period, and the addition of granular sludge could further shorten the period. The presence of organic matters was profitable for the PN granulation. The ammonia oxidation rate and PN rate were about 80% and 95%, respectively. The specific 10
microbial community and structure of PN granular sludge made it feasible to treat domestic sewage with low ammonia concentration at room temperature. High-throughput pyrosequencing results suggested the biodiversity decreasing of both reactors after the start-up of PN granulation. Nitrosomonas was the dominant AOB in the PN granular system. Acknowledgements This work was supported by water project of National Science and Technology Major Project (Grant No.2012ZX07202-005). References 1. Adav, S.S., Lee, D.J., Lai, J.Y. 2007. Effects of aeration intensity on formation of phenol-fed aerobic granules and extracellular polymeric substances. Appl Microbiol Biotechnol, 77(1), 175-182. 2. APHA. 1998. Standard methods for the examination of water and wastewater, 20th ed., American Public Health Association. Washington, DC, USA. 3. Gabarro, J., Ganigue, R., Gich, F., Ruscalleda, M., Balaguer, M.D., Colprim, J. 2012. Effect of temperature on AOB activity of a partial nitritation SBR treating landfill leachate with extremely high nitrogen concentration. Bioresour Technol, 126, 283-289. 4. Hulle, S.W.H.V., Vandeweyer, H.J.P., Meesschaert, B.D., Vanrolleghem, P.A., Pascal Dejansb, A.D. 2010. Engineering aspects and practical application of autotrophic nitrogen removal from nitrogen rich streams. Chemical Engineering Journal, 162(1), 1-20. 11
5. Khiewwijit, R., Temmink, H., Rijnaarts, H., Keesman, K.J. 2015. Energy and nutrient recovery for municipal wastewater treatment How to design a feasible plant layout. Environmental Modelling & Software, 68, 156-165. 6. Lackner, S., Gilbert, E.M., Vlaeminck, S.E., Joss, A., Horn, H., van Loosdrecht, M.C. 2014. Full-scale partial nitritation/anammox experiences--an application survey. Water Res, 55, 292-303. 7. Lopez-Palau, S., Pericas, A., Dosta, J., Mata-Alvarez, J. 2011. Partial nitrification of sludge reject water by means of aerobic granulation. Water Sci Technol, 64(9), 1906-1912. 8. Mosquera-Corral, A., Gonza´lez, F., Campos, J.L., Me´ndez, R. 2005. Partial nitrification in a SHARON reactor in the presence of salts and organic carbon compounds. Process Biochemistry, 40(9), 3109–3118. 9. Pijuan, M., Werner, U., Yuan, Z. 2011. Reducing the startup time of aerobic granular sludge reactors through seeding floccular sludge with crushed aerobic granules. Water Res, 45(16), 5075-5083. 10. Shi, Y.J., Wang, X.H., Yu, H.B., Xie, H.J., Teng, S.X., Sun, X.F., Tian, B.H., Wang, S.G. 2011. Aerobic granulation for nitrogen removal via nitrite in a sequencing batch reactor and the emission of nitrous oxide. Bioresour Technol, 102(3), 2536-2541. 11. Terada, A., Lackner, S., Kristensen, K., Smets, B.F. 2010. Inoculum effects on community composition and nitritation performance of autotrophic nitrifying biofilm reactors with counter-diffusion geometry. Environ Microbiol, 12(10), 12
2858-2872. 12. Vázquez-Padín, J.R., Figueroa, M., Campos, J.L., Mosquera-Corral, A., Méndez, R. 2010. Nitrifying granular systems_ A suitable technology to obtain stable partial nitrification at room temperature. Separation and Purification Technology, 74(2), 178-186. 13. Wan, C., Sun, S., Lee, D.J., Liu, X., Wang, L., Yang, X., Pan, X. 2013. Partial nitrification using aerobic granules in continuous-flow reactor: rapid startup. Bioresour Technol, 142, 517-522. 14. Wett, B., Omari, A., Podmirseg, S.M., Han, M., Akintayo, O., Gomez Brandon, M., Murthy, S., Bott, C., Hell, M., Takacs, I., Nyhuis, G., O'Shaughnessy, M. 2013. Going for mainstream deammonification from bench to full scale for maximized resource efficiency. Water Sci Technol, 68(2), 283-289. 15. Zhang, X., Zhang, H., Ye, C., Wei, M., Du, J. 2015. Effect of COD/N ratio on nitrogen removal and microbial communities of CANON process in membrane bioreactors. Bioresour Technol, 189, 302-308.
13
effluent
ARE
NAR
160
140
140
-1 nitrite concentration（ mg·L ）
120 100 80 60 40 20 0 0
10
20
40
50
120 100 80 60 40 20 0
60
0
10
20
Time（ d）
8
50
60 1200
5 4 3 2 1
80 800 60 600 40 400 20
0
granule size（ µm）
1000
6
200
0 10
20
30
40
50
60
0
10
20
Time（ d） effluent
ARE 160
140
140
-1 nitrite concentration（ mg·L ）
160
120 100 80 60 40 20 0 0
10
20
30
40
50
NAR
40
50
60
50
60
granule size
120 100 80 60 40 20 0
60
0
10
20
Time（ d）
50
30 Time（ d）
influent
-1 ammonia concentration（ mg·L ）
40
100
7
0
30
40
Time（ d）
1200
100
45
1000
40 35
ARE， NAR（ %）
-1 nitrate concentration（ mg· L ）
30 Time（ d）
ARE， NAR（ %）
-1 nitrate concentration（ mg· L ）
30
granule size
30 25 20 15 10
80 800 60 600 40 400 20
200
5 0
0 0
10
20
30
40
50
60
0
Time（ d）
10
20
30
40
50
60
Time（ d）
Fig.1 Nitrogen removal performance and granule size of P1 (A) and P2 (B)
14
granule size（ µm）
-1 ammonia concentration（ mg·L ）
influent 160
Fig.2 The taxonomic identities of sequences retrieved from P1 and P2 at the Genus level
15
HIGHLIGHTS Partial nitrification granular sludge was rapid started-up. Addition of granular sludge could significantly shorten the start-up period. Partial nitrification granular sludge was feasible for treating domestic sewage. Microbial characteristics were analyzed by high-throughput pyrosequencing.
16