View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/C4RA05762G (Communication) RSC Adv., 2014, 4, 52892-52897

Simultaneous nutrient removal and reduction in sludge from sewage waste using an alternating anaerobic–anoxic–microaerobic–aerobic system combining ozone/ultrasound technology†

Shanshan Yang, Wanqian Guo*, Yidi Chen, Xianjiao Zhou, Heshan Zheng, Xiaochi Feng, Renli Yin and Nanqi Ren
State Key Laboratory of Urban Water Resource and Environment, School of Municipal and Environmental Engineering, Harbin Institute of Technology, P.O. Box 2614, 202 Haihe Road, Harbin 150090, China. E-mail: guowanqian@hit.edu.cn; Fax: +86 451 86283008; Tel: +86 451 86283008

Received 15th June 2014 , Accepted 21st August 2014

First published on 29th August 2014

Abstract

A newly developed ozone/ultrasound technology combined with an alternating anaerobic–anoxic–microaerobic–aerobic (AAMA + O3/US) system achieved a 59.54% reduction in sludge production compared with a control system. Pyrosequencing showed that higher relative abundances of the microbial consortia responsible for nutrient removal were observed in the AAMA + O3/US system.

Introduction

Eutrophication, an excessive growth of plant material in water bodies usually caused by the presence of nutrients – particularly nitrogen and phosphorus – is a well-recognized environmental problem worldwide.1 The rapid increase in sludge production caused by eutrophication is a major challenge in the biological treatment of waste water.2 With improvements in living standards and the acceleration of urbanization, stringent standards for the discharge of sewage have been developed, especially for nitrogen and phosphorus in the effluents from biological sewage treatment systems.3 The development of efficient technologies for the simultaneous removal of enhanced nutrient levels and the in situ reduction of excess sludge has attracted increasing attention4,5 in view of the environmental burden and expense of dealing with these problems.

Guo et al.5 divided in situ methods of activated sludge reduction into four groups: chemical and/or physical cell lysis and cryptic growth methods in biosystems; uncoupling of metabolic processes; worm predation; and improved or novel processes. Of these four groups, the reduction of sludge by cell lysis and cryptic growth has generated much public concern and interest.6–8 We have previously shown9 that ozone/ultrasound (O3/US) cell lysis and cryptic growth technology can produce the enhanced disintegration of sludge. In addition, an enhanced efficiency in the removal of nitrogen and phosphorus may be achieved by using an appropriate distribution of aeration concentrations and by supplying suitable external sources of carbon in biological sewage treatment systems.10 Thus the combination of O3/US cell lysis and cryptic growth technology with alternating aeration in a bioreactor was thought to be feasible. This combined system could be a useful biosystem to simultaneously achieve enhanced sludge reduction and nutrient removal in future practical applications.

We report here the application of combined O3/US cell lysis and cryptic growth technology to a biological sewage treatment system under alternating anaerobic–anoxic–microaerobic–aerobic (AAMA) conditions. Verification experiments were conducted to compare the performance of this simultaneous enhanced sludge reduction and nutrient removal system (AAMA + O3/US system; AAMA2#) with a control system (an alternating AAMA system without recycling of sludge lyses material; AAMA1#). To further specify how the alternating dissolved oxygen (DO) conditions and the recycling O3/US sludge lysis influenced the microbial composition and structure of the microbial communities, high-throughput 454 pyrosequencing was used to provide an insight into the evolution of the microbial communities. The objectives of this study were: (1) to examine the impact of O3/US sludge lyses return on sludge reduction and nutrient removal; and (2) to explore the relationship between the performance of the waste water treatment and the functional composition and structure of the microorganism communities in these two AAMA systems.

Materials and methods

For the two AAMA systems (Fig. 1), air was supplied at the bottom of each reactor using a mass flow controller to produce a good distribution of aeration in each reaction phase. The concentration of controlled DO in the microaerobic phase was about 0.5 mg L−1 and the concentration of controlled DO in the aerobic phase was 3–4 mg L−1. Nitrate from the aerobic phase was cycled back into the anoxic phase. The effective volumes of the alternating anaerobic, anoxic, microaerobic and aerobic reactors were 0.8, 0.8, 0.8 and 1.6 L, respectively (Fig. 1). A mixed liquor suspended solids (MLSS) concentration of 3500 mg L−1 was maintained in both the AAMA1# and AAMA2# systems. The mixed liquor volatile suspended solid (MLVSS) concentration in the two reactors was 2675 mg L−1. Excess sludge withdrawal began when the MLSS concentration was >3500 mg L−1. The systems were operated on an 8 h cycle, with three cycles per day at room temperature. A continuous in-flow was maintained at a flow-rate of 12 L per day. The characteristics of the influent were as follows: glucose concentration, 580 mg L−1; yeast extract, 68 mg L−1; urea, 8 mg L−1; NaHCO3, 80 mg L−1; MgSO4, 66 mg L−1; CaCl2, 6 mg L−1; KH2PO4, 27.8 mg L−1; (NH4)2SO4, 112 mg L−1; FeSO4, 0.3 mg L−1; MnSO4, 6 mg L−1; and microelement solution, 1.0 ml L−1. The pH of the influent was maintained at 7.5 ± 0.2. The pH was not adjusted during the experimental period.
image file: c4ra05762g-f1.tif
Fig. 1 Schematic diagram of the AAMA + O3/US system.

For the AAMA + O3/US system (AAMA2#), 50% of the discharged excess sludge pretreated by the combined O3/US sludge technology was returned to the system. These pretreated sludge lyses were used as an external supply of carbon and were recycled back into the anoxic and microaerobic zones by controlling the sludge return ratio at 1[thin space (1/6-em)]:[thin space (1/6-em)]1 (e.g. 25% pretreated O3/US sludge lyses in the anoxic and microaerobic phases, respectively) (Fig. 2). The discharged excess sludge removed from the AAMA + O3/US system was pretreated using a combined O3/US apparatus with the optimized parameters obtained in our previous study9 (0.154 g O3 per g dry solids and 1.445 W mL−1 ultrasound energy density for 1 h). Ozone for the combined O3/US apparatus was generated from pure oxygen using an ozone generator (DHX-SS-1G, Jiujiu Ozone, Harbin, China). A low-range gaseous flow meter was used to adjust the flow-rate of the ozone. An ultrasound generator (FS-300, 20 kHz, Shengxi Ultrasonic Instrument Co., Shanghai, China) with an operating frequency of 20 kHz and a sonoprobe generator of diameter 8 mm was used.


image file: c4ra05762g-f2.tif
Fig. 2 Schematic diagram of one cycle of an 8 h, four-stage operating cycle in the combined the AAMA + O3/US system.

To determine the sludge retention time (SRT) and sludge yield of the AAMA + O3/US system, the anaerobic, anoxic, microaerobic and aerobic reactors were considered together as the sewage treatment system (Fig. 1). The total amount of solids remaining inside the sewage treatment system can be estimated by the sum of the solids in the anaerobic, anoxic, microaerobic and aerobic reactors, and in the returned O3/US sludge lyses yields. The sludge leaving the AAMA + O3/US system was the daily yield of discharged sludge minus the daily yield of discharged sludge multiplied by the controlled O3/US sludge lyses return ratio. Thus the SRT of the whole system could be defined as the total mass of sludge in the sewage treatment system divided by the sludge leaving the system (eqn (1)):

 
image file: c4ra05762g-t1.tif(1)
where XT is the sludge concentration of the total sludge solids in the sewage treatment system (g MLSS per L), XR is the sludge concentration of the returned O3/US sludge lyses (g MLSS per L), XD is the total mass of sludge solids discharged from the whole system per day (g MLSS per L per day), VT is the effective volume of the sewage treatment tank (L), VR is the relative sludge occupied volume of the returned O3/US sludge lyses (L), VD is the daily discharged sludge volume (L) and R represents the O3/US sludge lyses return ratio (%).

Analytical methods

The chemical oxygen demand (COD), MLSS, MLVSS, total nitrogen (TN), ammonia nitrogen (NH4+-N), total phosphorus (TP) and sludge volume index (SVI) were determined using standard methods.11 The DO concentration was measured using a DO probe (pH/Oxi 340i main engine pH meter, WTW Co., Germany). During the experimental operational period, the COD and MLSS were measured daily and TP, TN and NH4+-N were determined three or four times per week. All the sludge samples were centrifuged at 10[thin space (1/6-em)]000g for 5 min and then filtered through 0.45 μm cellulose acetate filters. Sample analyses were monitored by three replicate measurements at room temperature.

The structure and composition of the microbial community were assessed using 454 pyrosequencing of the 16S rRNA gene. After 2 months (60 days) of operation and cultivation, the characteristics and biophase of the sludge in the two AAMA systems were maintained in a steady-state. Samples for pyrosequencing were collected from the anaerobic, anoxic, microaerobic and aerobic phases. Four parts of samples were combined for DNA extraction. The specific steps of the DNA extraction, the high-throughput pyrosequencing of the 16S rRNA gene, the 454 pyrosequencing and processing of the pyrosequencing data, and the analysis of biodiversity and phylogenetic classification are given in Sections S1–S4 of the ESI.†

Results and discussion

Fig. 3a shows the cumulated excess sludge discharged from the AAMA1# and AAMA2# systems. During the two months of stable operation, the amounts of cumulative excess sludge discharged from the AAMA1# and AAMA2# systems were 173.78 and 70.32 g, respectively. Compared with the AAMA1# system, a 59.54% reduction in excess sludge discharged was achieved in the combined AAMA + O3/US system. Based on the operating performance and structural features of the bioreactor in the AAMA + O3/US system, this in situ reduction in excess sludge was achieved by cell lysis and the cryptic growth of the microorganisms obtained by recycling the O3/US sludge lyses. When the activated sludge is lysed by pretreatment with O3/US, the cell contents and nutrients released from the microorganisms are recycled and used in the metabolic cycles of other microorganisms. In this way, the carbonaceous materials released are reused for the vital activities of other microorganisms and, as a result, the overall production of excess sludge is reduced.
image file: c4ra05762g-f3.tif
Fig. 3 (a) Cumulative yields of discharged excess sludge and (b) changes in the SVI in both the control and the AAMA + O3/US systems.

The SRTs for the AAMA1# and AAMA2# systems were calculated to be 4.83 and 24.49 day, respectively (eqn (1)). The SRT in the AAMA2# system was considerably longer than the SRT in the AAMA1# system. In other studies that have attempted to minimize the excess sludge generated during the activated sludge process, it has been reported that the SRT in the biological sewage treatment system is critical.12 Chon et al.13 demonstrated that about half of the overall sludge reduction occurred in the aeration reactor through a long SRT condition in the anaerobic side-stream reactor process. By comparing the SRT in the two systems, our results led us to conclude that a long controlled SRT in the biological sewage treatment system is necessary to achieve the highest reduction in biological solids in the AAMA + O3/US system. This phenomenon was similar to previous studies of a biological sewage treatment system in which a long SRT reduced the production of sludge.12,13 The results obtained in this study successfully demonstrated that this O3/US cell lysis and cryptic growth technology combined with a bioreactor could achieve a good performance for in situ reduction of excess sludge.

Fig. 3b shows the trend of changes in the SVI in the control and combined reactors. The average sludge SVI values in the AAMA1# and AAMA2# were systems were 84 and 79 mL g−1, respectively. Compared with the control system, in situ sludge reduction using cell lysis and cryptic growth with the recycling of O3/US sludge lyses reduces the SVI and therefore improves the sedimentation characteristics of the sludge. This may because the microorganisms acclimatized to a relatively low return of sludge lyses stimulate the activity of the microbes in the activated sludge.

Fig. 4a shows the changes in efficiency of COD removal for both the AAMA1# and AAMA2# systems. The COD removal efficiency in AAMA1# was 91.82%, whereas the COD removal efficiency in AAMA2# was 89.15%. Recycling of the sludge lysis back into the AAMA2# system results in part of the sludge lyses being biologically assimilated to become new activated sludge; the other part of the sludge lyses that is not assimilated remains as inactive sludge in the effluent. Therefore recycling of the O3/US pretreated sludge lyses may have an impact on the quality of the effluent from the AAMA + O3/US system. However, compared with the aeration pattern in the conventional activated sludge system, the aeration pattern in the AAMA2# system was improved on moving to an alternating anaerobic–anoxic–microaerobic–aerobic pattern; this changing pattern of DO concentration alleviates the impact of this strong aeration on the microorganisms. Hence the efficiency of removal of the COD in the effluent was not significantly influenced in the AAMA + O3/US system.


image file: c4ra05762g-f4.tif
Fig. 4 Changes in removal efficiencies of the effluent (a) COD, (b) TN and (c) TP in both the control and the AAMA + O3/US systems.

The variations in removal efficiencies of TN in these two systems were monitored and the changes in the nitrogen components are shown in Fig. 4b. The average removal efficiencies of TN in the AAMA1# and AAMA2# systems were 64.13 and 78.22%, respectively, i.e. the removal efficiency for TN in the AAMA2# system was increased by 18.01% compared with that in the AAMA1# system. In terms of the microbial denitrification mechanism in AAMA2#, the removal of ammonia nitrogen was realized by aerobic nitrifying bacteria in the aerobic phase; the efficiency of nitrogen removal depends mainly on the denitrification process under anoxic conditions. In addition, an enhanced efficiency of nitrogen removal can be caused by the simultaneous nitrification and denitrification (SND) that occurs in a microaerobic environment.14 Hence, in the AAMA + O3/US system with the recycling of O3/US sludge lyses, the recycling of the O3/US sludge lyses as an extra source of carbon was critical to the process of denitrification and the enrichment of the denitrifier group. A series of nitrification, denitrification and SND processes were induced as a result of the structural features of the bioreactor in these two systems and, as a consequence, an enhanced removal efficiency for biological nitrogen was realized in the AAMA + O3/US system.

In addition to the enhanced nitrogen removal efficiency in the combined AAMA + O3/US system, the relatively higher removal efficiency for phosphorus in the AAMA2# system was also attributed to the alternating DO conditions during the sewage treatment period. Fig. 4c shows that the average removal efficiencies for TP in the AAMA1# and AAMA2# systems were 84.14 and 85.86%, respectively; the TP removal efficiency in the AAMA2# was therefore increased by 2% compared with the control system. Previous studies have demonstrated that a different aeration environment is required for organic substrate competition between phosphorus-accumulating organisms (PAOs) and denitrifying bacteria.15 In the AAMA + O3/US system, as a result of the alternating sequential anaerobic–aerobic conditions, phosphorus is taken up by PAOs and by denitrifying phosphate-accumulating organisms (DPAO) when an oxygen electron acceptor is supplied in the aerobic phase. In the microaerobic and aerobic phases DPAO uses either nitrate or O2 as an electron acceptor to enhance the removal efficiency of biological phosphorus. Therefore, in this combined system, enhanced nitrogen and phosphorus removal efficiencies are induced by the combined effects of PAO and DPAO.16

To understand the structures and functions of the microbial communities in these two systems, high-throughput 454 pyrosequencing was applied to analyse the 16S rRNA gene in the bacteria and its distribution. With an average length of 455 bp, 10 267 (AAMA1#-sample) and 10 006 (AAMA2#-sample) high-quality sequence tags were obtained. Compared with conventional methods in molecular biology (e.g. PCR-DGGE), high-throughput techniques are better for profiling complex bacterial communities as a result of their unprecedented depth of sequencing.17

Sludge samples from the AAMA1# and AAMA2# systems were compared. A phylogenetic spectrum was obtained to identify the similarities and differences in the microbial communities (Fig. 5). The relative abundances of the predominant phyla in the total microbial community of the AAMA1# system were: Proteobacteria, 30.93%; Bacteroidetes, 24.39%; and Actinobacteria, 9.74%. In the AAMA2# system, the dominating phyla were Proteobacteria (58.06%), Bacteroidetes (14.93%) and Actinobacteria (7.25%); the relative abundance of the most dominant Proteobacteria in the AAMA2# system was much higher than that in the AAMA1# system. The relative abundance of the Nitrospira phylum in the AAMA2#-system (1.64%) was significantly higher than that in the AAMA1#-system (0.28%), indicating that the improved conditions applied in the AAMA + O3/US system, including the alternating DO conditions, dosing from the external carbon sources and nitrate recycling, favoured the growth of these nitrite-oxidizing bacteria (NOB).


image file: c4ra05762g-f5.tif
Fig. 5 Structures of the microbial communities based on 454 pyrosequencing at the phylum level: (a) AAMA1# system and (b) AAMA2# system.

When comparing the evolution of the microbial community in the AAMA1# and AAMA2# systems at the class level, differences were observed in the composition of Proteobacteria in the two samples. As shown in Fig. 6, the Proteobacteria were classified into four major classes: α-Proteobacteria, β-Proteobacteria, γ-Proteobacteria and δ-Proteobacteria. β-Proteobacteria, which are the most abundant Proteobacteria, were found at much higher numbers in the AAMA2# system (30.75%) than in the AAMA1# system (10.13%). The relative abundance of γ-Proteobacteria in the AAMA2# system (13.16%) was also higher than in the AAMA1# system (12.77%). In previous studies,18 β-Proteobacteria and γ-Proteobacteria were considered to play critical parts in the removal of biological nitrogen and phosphorus. Thus in the AAMA2# system, the higher relative abundances of β-Proteobacteria and γ-Proteobacteria might explain the good performance for the biological removal of nitrogen and phosphorus in the AAMA + O3/US system (Fig. 4).


image file: c4ra05762g-f6.tif
Fig. 6 Relative abundances of the predominant Proteobacteria at the class level in the AAMA1# and AAMA2# systems.

Further comparison of the microbial communities down to the genus level was conducted to reveal more information about the mechanism of enhanced nitrogen and phosphorus removal in the AAMA2# system. Fig. 7 shows that the genera Nitrosomonas and Nitrospira, which accounted for 0.77 and 1.28%, respectively, of the total microbial community in the AAMA2# system, were higher than in the AAMA1# system (0.34 and 0.17%, respectively). In previous studies, Nitrosomonas was the main species of ammonium oxidizing bacteria and Nitrosospira was the main NOB species. Nitrification comprises the conversion of ammonium to nitrite by ammonium oxidizing bacteria and the subsequent oxidation of nitrite to nitrate by NOB.19 Higher relative abundances of these species could explain the good level of nitrogen removal in the AAMA + O3/US system. In addition, some heterotrophic denitrifiers, including Dechloromonas, Zoogloea and Flavobacterium were all detected at different relative abundances in both systems. Fig. 7 shows that the relative abundances of the genera Zoogloea, Dechloromonas and Flavobacterium in the AAMA2# system were 13.76, 4.98 and 1.50%, respectively; the relative abundances of these genera in the AAMA1# system were 1.23, 1.33 and 0.76%, respectively. In addition, the genera Sphingomonas, representing the onset of nitrification,20 accounted for 0.67% of the total microbial community in the AAMA2# system, but was almost undetectable in the AAMA1# system. These results show that higher relative abundances of these bacteria coexisting in the AAMA2# system might further explain the enhanced nitrogen removal in the AAMA + O3/US system. The primary microorganisms (Dechloromonas, Propionivibrio) responsible for the removal of phosphorus were also detected (Fig. 7).21 The relative abundance of Dechloromonas and Propionivibrio in the AAMA1# system were 1.33 and 2.19%, respectively. The relative abundances of Dechloromonas (4.98%), Propionivibrio (3.18%) and Acinetobacter (0.89%) in the AAMA2# system were higher than those AAMA1#. The Acinetobacter species, which actively take part in simultaneous denitrification and the uptake of phosphorus,22 were only observed in the AAMA2# system. Based on the 454 pyrosequencing analyses, there was a clear distinction in the components and structures of the microbial communities between the AAMA1# and AAMA2# systems, despite the fact that the microbial consortia inoculated were same. Although the existence of these genera shows a major influence on the removal efficiency for biological nitrogen and phosphorus, their relatively low abundance in the control AAMA1# system might led to limited results. Some genera, e.g. Sphingomonas and Acinetobacter species, which actively take part in nitrification and simultaneous denitrification/phosphorus uptake processes, could only be detected in the AAMA2# system. Hence when comparing the evolution of the microbial communities in the AAMA1# and AAMA2# systems at the genus level, higher relative abundances of the particular enriched microbial consortia responsible for enhanced nitrogen and phosphorus removal processes were observed in the AAMA2# system.


image file: c4ra05762g-f7.tif
Fig. 7 Relative abundances of the phylogenetic groups at the genus level in the AAMA1# and AAMA2# systems.

Conclusions

We developed a combined ozone/ultrasound technique for a biological sewage treatment system under alternating anaerobic–anoxic–microaerobic–aerobic conditions. The results show that a 59.54% reduction in excess sludge discharged could be achieved using this AAMA + O3/US system. Comparing the effects of the performance of sewage treatment in these two systems, the removal efficiencies for TN and TP both showed upward trends in the AAMA + O3/US system. A comparison of the microbial communities using high-throughput 454 pyrosequencing showed that higher relative abundances of the particular microbial consortia responsible for biological nitrogen and phosphorus removal were present in the AAMA + O3/US system. Our results support earlier findings that the introduction of alternating DO environments, extra sources of carbon and intermittent nitrate recycling in the AAMA2# system have positive effects on the optimization of the microbial communities and consequently enhance the performance of the sludge reduction and nutrient removal processes.

Acknowledgements

This research was supported by the National Nature Science Foundation of China (Grant no. 51008105 and 51121062). The authors also gratefully acknowledge financial support from the State Key Laboratory of Urban Water Resource and Environment (Grant no. 2014TS06), the Harbin Institute of Technology Fund for young top-notch talent teachers (AUGA5710052514), the special S&T project on treatment and control of water pollution (2013ZX07201007-001), Academician Workstation Construction in Guangdong Province (2012B090500018) and Shanghai Tongji Gao Tingyao Environmental Science & Development Foundation.

Notes and references

  1. Y. Zeng, B. H. Zhao, L. F. Zhu, D. M. Tong and C. W. Hu, RSC Adv., 2013, 3, 10806–10816 RSC.
  2. Y. Tian, J. Zhang, W. Zuo, L. Chen, Y. N. Cui and T. Tan, Environ. Sci. Technol., 2013, 47, 3498–3505 CAS.
  3. H. Chen, X. Zheng, Y. G. Chen and H. Mu, RSC Adv., 2013, 3, 9835–9842 RSC.
  4. W. Q. Guo, S. S. Yang, J. W. Pang, J. Ding, X. J. Zhou, X. C. Feng, H. S. Zheng and N. Q. Ren, RSC Adv., 2013, 3, 21848–21855 RSC.
  5. W. Q. Guo, S. S. Yang, W. S. Xiang, X. J. Wang and N. Q. Ren, Biotechnol. Adv., 2013, 31, 1386–1396 CrossRef CAS PubMed.
  6. L. B. Chu, S. T. Yan, X. H. Xing, X. L. Sun and B. Jurcik, Water Res., 2009, 43, 1811–1822 CrossRef CAS PubMed.
  7. H. J. Ma, S. T. Zhang, X. B. Lu, B. Xi, X. L. Guo, H. Wang and J. X. Duan, Bioresour. Technol., 2012, 116, 441–447 CrossRef CAS PubMed.
  8. W. C. Lan, Y. Y. Li, Q. Bi and Y. Y. Hu, Bioresour. Technol., 2013, 134, 43–50 CrossRef CAS PubMed.
  9. S. S. Yang, W. Q. Guo, Z. H. Meng, X. J. Zhou, X. C. Feng, H. S. Zheng, B. Liu, N. Q. Ren and Y. S. Cui, Bioresour. Technol., 2013, 131, 560–563 CrossRef CAS PubMed.
  10. P. Tanwar, T. Nandy, R. Khan and R. Biswas, Bioresour. Technol., 2007, 98, 2473–2478 CrossRef CAS PubMed.
  11. APHA, The Standard Methods for the Examination of Water and Wastewater, American Public Health Association, Washington, DC, USA, 21st edn, 2005 Search PubMed.
  12. D. H. Chon, M. Rome, Y. M. Kim, K. Y. Park and C. Park, Water Res., 2011, 45, 6021–6029 CrossRef CAS PubMed.
  13. D. H. Chon, M. Rome, H. S. Kim and C. Park, Water Sci. Technol., 2011, 63, 93–99 CrossRef CAS PubMed.
  14. S. M. Hocaoglu, G. Insel, E. U. Cokgor and D. Orhon, Bioresour. Technol., 2011, 102, 4333–4340 CrossRef CAS PubMed.
  15. S. Danesh and J. A. Oleszkiewicz, Water Sci. Technol., 1997, 35, 137–144 CrossRef CAS.
  16. P. Brown, S. K. Ong and Y. W. Lee, Desalination, 2011, 270, 227–232 CrossRef CAS PubMed.
  17. F. Guo and T. Zhang, Water Res., 2012, 46, 2772–2782 CrossRef CAS PubMed.
  18. D. Wu, G. A. Ekama, H. Lu, H. K. Chui, W. T. Liu, D. Brdjanovic, M. C. M. van Loosdrecht and G. H. Chen, Water Res., 2013, 47, 3057–3069 CrossRef CAS PubMed.
  19. M. K. H. Winkler, J. P. Bassin, R. Kleerebezem, D. Y. Sorokin and M. C. M. V. Loosdrecht, Appl. Microbiol. Biotechnol., 2012, 94, 1657–1666 CrossRef CAS PubMed.
  20. K. C. B. Krishna, A. Sathasivan and M. P. Ginige, Water Res., 2013, 47, 4666–4679 CrossRef PubMed.
  21. Y. J. Tu and A. J. Schuler, Environ. Sci. Technol., 2013, 47, 3816–3824 CrossRef CAS PubMed.
  22. P. S. Barker and P. L. Dold, Water Res., 1996, 30, 769–780 CrossRef CAS.

Footnote

† Electronic supplementary information (ESI) available: This file contains the specific operation steps of DNA extraction, high-throughput 16S rRNA gene pyrosequencing, biodiversity analysis and phylogenetic classification. See DOI: 10.1039/c4ra05762g
This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.