On-line Monitoring for Phosphorus Removal Process and Bacterial Community in Sequencing Batch Reactor*

CUI Youwei (崔有為), WANG Shuying (王淑瑩),** and LI Jing (李晶)

?

On-line Monitoring for Phosphorus Removal Process and Bacterial Community in Sequencing Batch Reactor*

CUI Youwei (崔有為)1, WANG Shuying (王淑瑩)1,**and LI Jing (李晶)2

1Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, China2China Aeronautical Project and Design Institute, Beijing 100011, China

For efficient energy consumption and control of effluent quality, the cycle duration for a sequencing batch reactor (SBR) needs to be adjusted by real-time control according to the characteristics and loading of wastewater. In this study, an on-line information system for phosphorus removal processes was established. Based on the analysis for four systems with different ecological community structures and two operation modes, anaerobic-aerobic process and anaerobic-anaerobic process, the characteristic patterns of oxidation-reduction potential (ORP) and pH were related to phosphorous dynamics in the anaerobic, anoxic and aerobic phases, for determination of the end of phosphorous removal. In the operation mode of anaerobic-aerobic process, the pH profile in the anaerobic phase was used to estimate the relative amount of phosphorous accumulating organisms (PAOs) and glycogen accumulating organisms (GAOs), which is beneficial to early detection of ecology community shifts. The on-line sensor values of pH and ORP may be used as the parameters to adjust the duration for phosphorous removal and community shifts to cope with influent variations and maintain appropriate operation conditions.

on-line monitoring, phosphorus removal, sequencing batch reactor, pH, oxidation-reduction potential

1 INTRODUCTION

Phosphorus pollutants in wastewater are responsible for increasing the eutrophication in surface waters, which results in a great financial loss and potentially ill effects on health. The enhanced biological phosphorus removal (EBPR) process is considered as a cost effective and environmentally friendly technology to remove phosphorous from wastewater. EBPR is based on the enrichment of phosphorus accumulating organisms (PAOs) [1]. It has been demonstrated that the biological population in phosphorous removal comprises of at least two groups: one group utilizing either oxygen or nitrate as an electron acceptor (denitrifying PAOs), and the other utilizing only oxygen (aerobic PAOs) [2-5]. In the anaerobic phase, PAOs take up volatile fatty acids (VFAs) and store them as polyhydroxyalkanoates (PHAs). The energy for this anaerobic process is from the hydrolysis of intracellular polyphosphate and the glycolysis of glycogen followed by the release of ortho-phosphate to the bulk liquid [6]. In the subsequent aerobic/anoxic phase, PAOs use oxygen or nitrate as an electron acceptor and use PHAs to generate energy for growth, glycogen synthesis, and phosphate luxury uptake [7, 8]. Therefore, EBPR can be achieved by altering anaerobic- aerobic (A/O) or anaerobic-anoxic (A/A) conditions. As one of EBPR processes, a sequencing batch reactor (SBR) has some advantages,.., single-tank design and the flexibility for different treatment objectives, by adjusting the duration of anaerobic, anoxic and aerobic phase to achieve the EBPR effect. The SBR is well suitable for relatively small amounts of wastewater to be treated, while it also performs satisfactorily in larger applications [9, 10]. However, the flexibility in operation requires a higher level of process control and automation management.

An important aspect in SBR process management is the duration control of batch mode or step timing. Biological wastewater treatment plants (WWTPs) are normally designed and operated under nominal operating conditions, in which the loading rate is assumed constant [11, 12]. The SBR is also operated on a fixed-time schedule, but this steady-state assumption is inappropriate since the process is subjected to fluctuations in flow rate and loading. The primary drawback is the inapplicability for adjusting the cycle duration in the process according to the characteristics and loading of wastewater, which will result in an inefficient operation,.., inefficient energy consumption or poor control on the effluent quality. Moreover, inadequate timing may result in serious process degradation, especially in phosphorous removal [13]. A possible control strategy is based on the real-time measurement of relevant process variables,.., COD, ammonium-N, nitrate-N and phosphate-P, to identify the end of a biological reaction. However, there is a formidable obstacle in this scheme, represented by the complexity in on-line measurement of the variables, which is neither simple nor economical [10]. For this reason, in recent years, it has been received much attention to use oxidation-reduction potential (ORP), pH and dissolved oxygen (DO) as the parameters for real-time control of the cycles in SBR systems [14-32]. It has been shown that the ORP and pH pattern under a given wastewater treatment conditions can be used as indirect process indicators to identify specific control points, in particular, in biological nitrogen removal process [14-27].

However, very little work has been conducted on the real-time control of SBR systems for phosphorous removal, although their inherent time-oriented nature and the available automatic control equipment make the attempts quite attractive. The reasons may be related to (1) many factors affecting the phosphorous removal process [33, 34], and (2) the microbial competition between PAOs and GAOs in EBPR ecology community. In recent years, poor performance or even complete failure of EBPR processes has been reported. In addition to conventional factors affecting the process, GAOs as a potential competitor to PAOs in EBPR system has attracted considerable interest [35-40]. GAOs can compete effectively with PAOs under anaerobic-aerobic conditions. However, GAOs do not store polyP and use glycogen as their sole energy source for VFA uptake. Consequently, the identification of specific control points becomes complicated because phosphorous removal is highly dependent on the competition behavior in EBPR ecology community. Therefore, on-line monitoring and detection of the state of ecology community is required for successful control of EBPR SBR to make the process more effective. A few studies on control of EBPR SBR reported that the profile of OPR and pH could give signals for controlling phase duration in activated sludge in treating synthetic wastewater under A/O condition [10, 28-32]. However, these studies were limited to the treatment of synthetic wastewater, investigation on the sole community highly enriched by PAOs, and application of A/O cycle to develop the control strategy for phosphorous removal, which is insufficient for practical applications. A complete control system should be developed based on the application of real wastewater, the investigations on various microbial communities and all phosphorous removal processes (A/O and A/A cycle mode). Besides, on-line monitoring of population shift should be added to control the phosphorous removal process. These considerations will make the control system more effective, reliable and safe.

Based on the above concepts and analysis, the objectives of this study are to establish an on-line information system for phosphorous dynamic control using ORP and pH as parameters under A/O and A/A conditions, and to detect the state of EBPR community by the on-line parameters. For this purpose, four bench-scale SBR systems fed with actual sewage are used and evaluated in three cases of EBPR community and two operation modes.

Table 1 Characterization of domestic wastewater

Note: All values are in mg·L－1except pH; No. of samples is 60.

2 MATERIALS AND METHODS

2.1 Reactor and operation

Four lab-scale SBRs with working volume of 10 L were inoculated with sludge from a lab-scale modified University of Cape Town (MUCT) process, which purified the domestic wastewater well (the mean removal efficiency for phosphorous and nitrogen is 93% and 74%, respectively). Three reactors were operated under A/O conditions in parallel and the other under A/A cycle. All SBRs were placed in a room with the temperature controlled at 18 to 22°C. These reactors were equipped with ORP, pH and DO electrodes, which were positioned 6 cm below the liquid surface in the reactor. A data acquisition program was used to continuously store and record the measured data on a personal computer. These reactors were operated with the mixed liquor suspended solids (MLSS) at 2500-3000 mg·L－1. The effective sludge age (SRT ) was controlled on about 10 days. The cycle of each SBR involved in 6 min feeding, 0.5 h decanting and 2.4 h settling, while the reaction time was flexible according to experimental design and objectives. In each cycle, the volumetric exchange ratio in the reactor was 100%. During the anaerobic phase, the content of the reactor was kept in suspension using a low speed mixer fixed in the reactor. During the aerobic phase, aeration was introduced into the reactor though fine bubble diffusers fixed at the bottom of the reactor and connected to a pneumatic compressor. The air flow rate at the inlet was controlled by a mass flowmeter. In the anoxic phase, to obtain the electron acceptor, sodium nitrate was added into the SBR at the beginning of anoxic phase. The different process sequence in the operating cycle was controlled manually. All reactor systems were operated daily. After a start-up period of 30 days, the SBR was continuously operated for more than 90 days.

2.2 Composition of domestic wastewater

The feed wastewater was collected from the residential area of Beijing University of Technology, Beijing, and first treated in the septic tank. The organic content in the wastewater was moderate, but the total nitrogen content was high. During the investigation period for more than three months, 60 influent samples were analyzed. The composition of domestic wastewater used is presented in Table 1.

2.3 Analytical methods

2.4 Cultivation of bacterial community

Domestic wastewater usually contains a carbohydrate and an organic acid mixture, which are used as carbon and energy sources in biological treatment. Previous studies have the consensus that GAOs often compete with PAOs for carbon sources in A/O systems [41, 42], supported by various accepted biochemical models [43, 44]. Operated properly, the EBPR process can reach relatively high phosphorous removal efficiency. However, under certain circumstances the EBPR system may experience upsets and deterioration. The breakdown of EBPR was found to be due to the outgrowth of GAOs overwhelming PAOs in lab-scale reactors [38, 45-48] and in wastewater treatment plants [47, 49-51]. Mino. [52] proposed a metabolic model for the breakdown of EBPR by GAOs, and the process deterioration was attributable to the presence of GAOs [36, 38, 47-52]. How to establish PAOs and GAOs enriched communities was addressed [35-42]. It was found that carbon sources favored PAOs over GAOs based on the difference in their abilities to utilize different carbon source. Some carbon sources, such as acetate, offered a competitive advantage to PAOs with respect to GAOs [35, 36, 42], while glucose could not be used as the sole carbon for phosphorous release [37-39]. In additions, Liu. [37] indicated that the P/C feeding ratio strongly affected the dominance of PAOs and GAOs in the community. Higher ratio seemed to favor the growth of PAOs, while lower ratio resulted in a shift of the dominant population to GAOs. All above studies showed that the predominance of PAOs was attributable to good performance, while the predominance of GAOs was related to deteriorated performance by 16S rRNA/DNA-based molecular methods. Therefore, adjusting the competitive ability of PAOs and GAOs by selecting carbon sources fed to the EBPR system is a promising means.

In this study, these methods to control the proliferation of GAOs or PAOs were adopted. During the cultivation period, three A/O SBRs, SBR1, SBR2 and SBR3, were fed with different wastewater by adding glucose or acetate into domestic wastewater, which also resulted in the change of the P/C ratio in influent to form different selections to EBPR community (Table 2), as the previous studies [36, 37]. After a 30-days cultivation period, the same domestic wastewater was fed into the three SBRs in the experimental period. The performance in SBR2 kept deteriorating to about 50%. Over 90% phosphorous removal was obtained in SBR3 and 85% for SBR2. According to the reports [35-42], this was due to the dynamic shift of EBPR population structure. From the conditions established for controlling growth of GAOs or PAOs [36, 37], it can be concluded that PAOs were dominant in SBR3, GAOs were dominant species in SBR2, and PAOs and GAOs coexisted in SBR1. In SBR4, A/A cycle was applied to cultivate denitrifying PAOs (DPAOs) with the actual domestic wastewater (Table 2). At initiation of the anoxic phase, some sodium nitrate was added into SBR4 so that the electron acceptor (nitrate) was available for DPAOs.

Table 2 Controlled conditions

3 RESULTS AND DISCUSSION

3.1 SBR systems in A/O mode

PAOs store phosphorus under sequential anaerobic- aerobic conditions. VFA are taken up anaerobically and stored as PHA through the release of phosphorus (P) and degradation of glycogen. More P is taken up when an electron acceptor is supplied in aerobic phase. Biochemistry of PAO strongly influences the pH and ORP value in SBR, so that it is important to find the relation between on-line parameters and the extent of biological metabolism for realizing on-line control.

Figure 1 Relation of ORP with phosphorous concentration in different SBRs

3.1.1

Table 3 Linear regression equation in anaerobic phase

Figure 2 The first derivative of ORP with time curves in the anaerobic phase

3.1.2

For different microbial community, pH profiles in the anaerobic stage are different (Fig. 3). This observation is not in agreement with previous studies, in which pH decreased as the phosphorous release increased [24-29]. This may be explained by the highly-enriched PAOs sludge used in their studies. In this study, the activated sludge system was cultivated to form three community structures composed of PAOs and GAOs. Some bioprocesses that can affect pH in anaerobic and aerobic phase are exhibited in Table 4. In anaerobic phase, VFA uptake and CO2stripping increase pH, while phosphorous release and CO2production decrease pH. In aerobic phase, phosphorous uptake is important to variations of pH. GAOs show similar features as PAOs, but they produce different PHA and lack the ability of EBPR or poly-P accumulation. Since both PAOs and non-PAOs utilized fatty acid in the anaerobic stage, the amount of acid, population size of PAOs, and population size of non-PAOs determined the pH variation because all the conditions were the same in the three SBRs except the ratio of PAOs and GAOs.

Figure 3 Relation of pH with phosphorous concentration in different SBRs

Table 4 Effect of bioprocess on pH in EBPR

In the anaerobic phase, there are four EBPR events affecting pH (Table 4). CO2is produced by glycogen degradation and decreases pH, while CO2stripping increases pH. With about the same MLSS in three SBRs, almost the same amount of CO2is produced in the three systems. The amount of CO2stripping is also equal under the same aeration conditions. Therefore, the amount of CO2in the system (CO2production-CO2stripping) is almost the same. It indicates that different patterns of pH profile in the anaerobic phase are resulted from the relative amount of VFA uptake and phosphorous release. Although PHA is taken up by PAOs with the correspondent proton, pH decreases due to the phosphorus release. GAOs also take up PHA with the correspondent proton, but as polyphosphate is not involved in GAOs metabolism, phosphorus is not released and pH increases. In SBR1 [Fig. 3 (a)], the pH curve shows a sharp decrease stage and a subsequent even stage. The sharp decrease in twenty minutes is due to the rapidly released phosphorous by PAOs and the plateau is due to the balance between phosphorous released by PAOs and VFA uptake by PAOs and GAOs. In SBR3 [Fig. 3 (c)], the pH curve has a sharp increase in about 5 min and then a sharp decrease. With highly-enriched PAOs, the decrease by phosphorus release is more than the increase by VFA uptake, so that pH decreases as phosphorous release increases. The sharp increase is due to the denitrification of residual nitrate or nitrite from the previous cycle, which has been proved in some studies [21, 25, 28, 30]. In SBR2 [Fig. 3 (b)], the pH value decreases in a short time and then increases. With highly-enriched GAOs, although PAOs release phosphate, resulting in proton release to the medium, more protons are utilized during VFA uptake by GAOs and PAOs. The short-time decease in pH profile is induced by metabolism of the small amount of PAOs in the system, which is similar to SBR1. Therefore, a net increase in pH is exhibited in SBR2. Since the pH variation in the anaerobic stage depends on the ratio of PAOs and GAOs population, it can be used for community monitoring by on-line diagnosis, which is important to detect the faults so as to make the adjustment in good time.

In the aerobic phase, pH profiles are similar in the three SBRs, with a rapid increase, a subsequent decrease, a rapid decrease and a plateau. The relevant aerobic EBPR processes are phosphorus uptake for poly-P formation, glycogen restoring from PHA resources, PAO growth, lysis and decay. Among them, phosphorus uptake has the highest energy requirement, which is provided by PHA oxidation. According to the biological events occurring in the aerobic phase, the first rapid increase in pH is due to CO2stripping. The first decrease in pH is resulted from the decrease due to nitrification and the increase due to phosphorous uptake. The valley signals the end of ammonium oxidation (data not shown). The second increase in pH is due to phosphorous uptake. When phosphorous uptake terminates a plateau appears since all aerobic biological events have finished. The knee points (point,andin Fig. 3) in pH profiles indicate the end of phosphorous uptake.

3.2 SBR systems in A/A mode

The existence of PAOs that utilize nitrate as an electron acceptor instead of oxygen has been confirmed experimentally and practically [7]. The merits of the anoxic phosphate removal are that both N and P are removed in the same process. Compared to conventional EBPR, simultaneous denitrification and P removal can minimize sludge disposal, and reduce aeration and the demand for carbon sources. The enrichment of DPAOs can be achieved by A/A cycle mode. In the anoxic phase, nitrate was introduced into the SBR to provide electron acceptor. In a WWTP, the nitrate required for phosphorous uptake is determined by the amount of ammonium in influent and the phosphorous concentration at the end of anaerobic phase. According to the amount of electron acceptor and the phosphorus concentration at the end of anaerobic phase, two cases were investigated.

3.2.1

Figure 4 Relation of pH and ORP with phosphorous and nitrate concentration in A/A mode SBR with deficient nitrate for phosphorous uptake

The relevant anoxic EBPR processes are phosphorus uptake for polyP formation, glycogen restoring from PHA resources, DPAO growth, lysis and decay. Among them, phosphorus uptake has the highest energy requirement, which is provided by PHA oxidation and thus, results in the highest nitrate consumption (Table 4). In the anoxic phase in SBR4, pH deceases as the nitrate reduction and phosphorous uptake increase. With the denitrification by DPAOs, nitrate is reduced to nitrogen released from the liquid, contributed to an increased of pH. Phosphorous uptake by DPAOs also results in an increase of pH until nitrate is completely consumed (Fig. 4). In this case, nitrate is deficient to phosphorous uptake. When nitrate is completely reduced, an anaerobic context occurs, and phosphorous is released again, making pH decrease. Therefore, there is an apex point (pointin Fig. 4) in pH profile, which coincides with the end of phosphorous uptake.

The ORP value decreases during the denitrification followed by consumption of nitrate during the anoxic phase. With depletion of nitrate, a minimum value of ORP is observed, and designated as the knee point of nitrate (pointin Fig. 4). This point also indicates the end of denitrification and conversion of nitrate, which is usually used as inflection points to control anoxic phase in nitrogen removal SBR [13, 16, 17, 21-24]. In the case of deficient nitrate for phosphorous uptake, the knee point in ORP profile can be detected and used to terminate the anoxic phase.

3.2.2

The typical track analysis in the case of sufficient nitrate for phosphorous uptake is shown in Fig. 5. In the anaerobic phase phosphorous release takes place substantially companied by continuous decrease of ORP and pH, same as the results for the case of deficient nitrate. In subsequent anoxic phase, with the introduction of nitrate phosphorous uptake initiates. With a sufficient nitrate supply, phosphorous uptake ends prior to the depletion of nitrate. The rise in anoxic pH is due to the denitrification and phosphorous uptake by DPAOs. However, there may be other heterotrophic bacteria responsible for the denitrification with CO2production. The continuous accumulation of CO2results in a decrease of pH. Therefore, when phosphorous uptake rate becomes slow in 320 min, the decrease in pH profile is obvious until the end of phosphorous uptake. Utilizing the surplus nitrate heterotrophic denitrifier reduces nitrate, causing the rise in the pH. Therefore, the infection point (pointin Fig. 5) indicates the end of phosphorous uptake. Anoxic ORP profile is similar to that in the case of deficient nitrate. However, the knee point (point b in Fig. 5) is corresponding to the depletion of nitrate rather than the end of phosphorous uptake. Therefore, the pH profile is reliable to provide the information on phosphorous dynamics, while the knee point in anoxic ORP profile is to detect the end of nitrate completion.

Figure 5 Relation of pH and ORP with phosphorous and nitrate concentration in A/A mode SBR with sufficient nitrate for phosphorous uptake

4 CONCLUSIONS

This paper presents the on-line control method for the timing of anaerobic, aerobic and anoxic phase in EBPR SBR. The pH and ORP are used as the indirect process indicators. Three systems with either PAOs or GAOs as dominant species are investigated by feeding glucose as main carbon source and changing P/C ratio in influent, which are beneficial to PAOs or GAOs. Continuous measurements of pH and ORP are related to the dynamic behavior of anaerobic phosphorus release and aerobic/anoxic phosphorus uptake in the EBPR SBR. The time evolution of these parameters clearly indicates the end of anaerobic phosphorus release and aerobic/anoxic phosphorus uptake, providing information on the state of process for determining the operating time required for each phase or a cycle. For different bacteria community, pH profiles in the anaerobic phase are different, so that it can be used to identify the community structure. An on-line information and diagnosis system is developed to detect pH deviations in the anaerobic phase for prediction of possible abnormal states in EBPR community structure.

NOMENCLATURE

A/A anaerobic-anoxic cycle

A/O anaerobic-aerobic cycle

DO dissolved oxygen

深圳港水上“巴士”航線的經(jīng)濟(jì)性由上述3個指標(biāo)構(gòu)成的成本價值量來決定，通過與其他路徑經(jīng)濟(jì)性的比較，分析深圳港水上“巴士”的優(yōu)劣勢.

DPAO denitrifying PAO

EBPR enhanced biological phosphorus removal

GAO glycogen accumulating organism

MLSS mixed liquor suspended solids

MUCT modified University of Cape Town

ORP oxidation-reduction potential

PAO phosphorous accumulating organism

P/C ratio of phosphorous to organic concentration (COD)

PHA polyhydroxyalkanoate

SBR sequencing batch reactor

VFA volatile fatty acid

WWTP wastewater treatment plant

1 Barnard, J.L., “Biological nutrient removal without the addition of chemicals”,., 9, 485-490 (1975).

2 Kerrn-Jesperson, J.P., Henze, M., “Biological phosphorus uptake under anoxic and aerobic conditions”,., 27 (4), 617-624 (1993).

3 Kuba, T., van Loosdrecht, M.C.M., Heijnen, J.J., “Effect of cyclic oxygen exposure on the activity of denitrifying phosphorus removing bacteria”,.., 34 (1/2), 33-40 (1996).

4 Meinhold, J., Filipe, C.D.M., Daigger, G.T., Isaacs, S., “Characterization of the denitrifying fraction of phosphate accumulating organisms in biological phosphate removal”,.., 39 (1), 31-42 (1999).

5 Freitas, F., Temudo, M., Reis, M.A.M., “Microbial population response to changes of the operating conditions in a dynamic nutrient-removal sequencing batch reactor”,., 28 (3), 199-209 (2005).

6 Mino,T., van Loosdrecht, M.C.M., Heijnen, J.J., “Microbiology and biochemistry of the enhanced biological phosphate removal process”,., 32 (1), 3193-3207(1998).

7 Kong, Y.H., Nielsen, J.L., Nielsen, P.H., “Microautoradiographic study of-related polyphosphate-accumulating bacteria in full-scale enhanced biological phosphorus removal plants”,..., 70 (9), 5383-5390 (2004).

8 Oehmen, A., Zeng, R.J, Yuan, Z.G., Keller, J., “Anaerobic metabolism of propionate by polyphosphate-accumulating organisms in enhanced biological phosphorus removal systems”,.., 91 (1), 43-53 (2005).

9 Wilderer, P.A., Irvine, R.L., Goronszy, M.C., Sequencing Batch Reactor Technology, IWA Publishing, London, England (2001).

10 Marsili-Libelli, S., “Control of SBR switching by fuzzy pattern recognition”,., 40, 1095-1107 (2006).

11 Muller, A., Marsili-libelli, S., Aivasidis, A., Lloyd, T., Kroner, S., Wandrey, C., “Fuzzy control of disturbances in a wastewater treatment process”,., 31 (12), 3157-3167(1997).

12 Artan, N., Orhon, D., “Mechanisms and design of sequencing batch reactors for nutrient removal”, In: IWA Scientific and Technical Report No.19, IWA Publishers, London (2005).

13 Plisson-Saune, S., “Real-time control of nitrogen removal using three ORP bending points: Signification, control strategy and results”,.., 33 (1), 275-280 (2005).

14 Harremoes, P., Capodaglio, A.G., Ellstrom, B.G., Henze, M., Jensen, K.N., Lynggaard-Jensen, A., Ottterpohl, R., S?eborg, H., “Wastewater treatment plants under transient loading—performance, modelling and control”,.., 27 (12), 71-115 (1993).

15 Buitron, G., Schoeb, M.E ., Moreno-Andradea, I., Moreno, J.A., “Evaluation of two control strategies for a sequencing batch reactor degrading high concentration peaks of 4-chlorophenol”,., 39, 1015-1024 (2005).

16 Chen, K.C., Chen, C.Y., Peng, J.W., Houng, J.Y., “Real-time control of an immobilized-cell reactor for wastewater treatment using ORP”,., 36, 230-238 (2002).

17 Cho, B.C., Liaw, S.L., Chang, C.N., Yu, R.F., Yang, S.J., Chiou, B.R., “Development of a real-time control strategy with artificial neural network for automatic control of a continuous-flow sequencing batch reactor”,.., 44, 95-104 (2001).

18 Furumai, H., Kazmi, A.A., Fujita, M., Furuya, Y., Sasaki, K., “Modelling long term nutrient removal in a sequencing batch reactor”,., 33, 2708-2714 (1999).

19 Hao, O.J., Huang, J., “Alternating aerobic-anoxic process for nitrogen removal, process evaluation”,.., 68, 83-93 (1996).

20 Hu, Z., Ferraina, R.A., Ericson, J.F., MacKay, A.A., Smets, B.F., “Biomass characteristics in three sequencing batch reactors treating a wastewater containing synthetic organic chemicals”,., 39, 710-720 (2005).

21 Kim, J.H., Chen, M., Kishida, N., Sudo, R., “Integrated realtime control strategy for nitrogen removal in swine wastewater treatment using sequencing batch reactors”,., 38, 3340-3348 (2004).

22 Yoo, C.K., Lee, D.S., Vanrolleghem, P.A., “Application of multiway ICA for on-line process monitoring of a sequencing batch reactor”,., 38, 1715-1732 (2004).

23 Sperandio, M., Queinnec, I., “Online estimation of wastewater nitrifiable nitrogen, nitrification dynamics and denitrification rates, using ORP and DO dynamics”,.., 49, 31-38 (2004).

24 Puig, S., Corominas, L., Vives, M.T., Balaguer, M.D., Colprim, J., “Development and implementation of a real-time control system for nitrogen removal using OUR and ORP as end points”,...., 44 (9), 3367-3373 (2005).

25 Ra, C.S., Lo, K.V., Shin, J.S., Oh, J.S., Hong, B.J., “Biological nutrient removal with an internal organic carbon source in piggery wastewater”,., 34, 965-973 (2000).

26 Guo, J., Yang, Q., Peng, Y., Yeng, A., Wang, S., “Biological nutrient removal with real time control using step-feed SBR technology”,.., 40 (6), 1564-1569 (2007).

27 Holman, J.B., Wareham, D.G., “COD, ammonia and dissolved oxygen time profiles in the simultaneous nitrification/denitrification process”,..., 22, 125-133 (2005).

28 Lee, D.S., Jeon, C.O., Park, J.M., “Biological nitrogen removal with enhanced phosphate uptake in a sequencing batch reactor using single sludge system”,.., 35, 3968-3976 (2001).

29 Luccarini, L., Porra, E., Spagni, A., Ratini, P., Grilli, S., Longhi, S., Bortone, G., “Soft sensors for control of nitrogen and phosphorus removal from wastewaters by neural networks”,.., 45 (4/5), 101-107 (2002).

30 Spagni, A., Buday, J., Ratini, P., Bortone, G., “Experimental considerations on monitoring ORP, pH, conductivity and dissolved oxygen in nitrogen and phosphorus biological removal processes”,.., 43 (11), 197-204 (2001).

31 Hong, S.H., Lee, M.W., Lee, D.S., Park, J.M., “Monitoring of sequencing batch reactor for nitrogen and phosphorus removal using neural networks”,..., 35, 365-370 (2007).

32 Pankaj, T., Tapas, N., Pallavi, U., Pravin, M., “Correlating on-line monitoring parameters, pH, DO and ORP with nutrient removal in an intermittent cyclic process bioreactor system”,, 99, 7630-7635 (2008).

33 Smolders, G.J.F., van Loosdrecht, M.C.M., Heijnen, J.J., “Model of the anaerobic metabolism of the biological phosphorus removal process; Stoichiometry and pH influence”,.., 43, 461-470 (1994).

34 Smolders, G.J.F., van Loosdrecht, M.C.M., Heijnen, J.J., “A metabolic model of the biological phosphorus removal process: Effect of the sludge retention time”,.., 48, 222-233 (1995).

35 Jeon, C.O., Park, J.M., “Enhanced biological phosphorous removal in a sequencing batch reactor supplied with glucose as a sole carbon source”,., 34 (7), 2160-2170 (2000).

36 Cech, J.S., Hartman, P., “Competition between polyphosphate and polysaccharide accumulating bacterial in enhanced biological phosphate removal system”,., 27 (7), 1219-1225 (1993).

37 Liu, W.T., Mino, T., Nakamura, K., Matsuo, T., “Internal energy-based competition between polyphosphate- and glycogen-accumulating bacteria in biological phosphorus removal reactors—Effect of P/C feeding ratio”,., 31 (6), 1430-1438 (1997).

38 Cech, J.S., Hartman, P., “Glucose induced breakdown of enhanced biological phosphate removal”,.., 11, 651-656 (1990).

39 Kargi, F., Uygur, A., Savas, H., Kaya, B., “Phosphate uptake and release rates with different carbon sources in biological nutrient removal using a SBR”,..., 76, 71-75 (2005).

40 Kishida, N., Kim, J.M., Chen, M., Sasaki, H., Sudo, R., “Effectiveness of oxidation–reduction potential and pH as monitoring and control parameters for nitrogen removal in swine wastewater treatment by sequencing batch reactors”,.., 96 (3), 285-290 (2003).

41 Oehmen, A., Yuan, Z.G., Blackall, L.L., Keller, J., “Comparison of acetate and propionate uptake by polyphosphate accumulating organisms and glycogen accumulating organisms”,..,91 (2), 162-168 (2005).

42 Levantesi, C., Serafim, L.S., Crocetti, G.R., Lemos, P.C., Rossetti, S., Blackall, L.L., Reis, M.A.M., Tandoi, V., “Analysis of the microbial community structure and function of a laboratory scale enhanced biological phosphorus removal reactor”,.., 4 (10), 559-569 (2002).

43 Carvalho, G., Lemos, P.C., Oehmen, A., Reis, M.A.M., “Denitrifying phosphorus removal: Linking the process performance with the microbial community structure”,., 41, 4383-4396 (2007).

44 Wentzel, M.C., Lotter, L.H., Loewenthal, R.E., Marias, G.V.R., “Metabolic behavior of Acinetobacter ssp. in enhanced biological phosphorus removal—A biochemical model”,, 12, 209-224 (1986).

45 Zeng, R.J., van Loosdrecht, M.C.M., Yuan, Z.G., Keller, J., “Metabolic model for glycogen-accumulating organisms in anaerobic/aerobic activated sludge systems”,.., 81 (1), 92-105 (2003).

46 Oehmen, A., Yuan, Z., Blackall, L.L., Keller, J., “Comparison of acetate and propionate uptake by polyphosphate accumulating organisms and glycogen accumulating organisms”,.., 91 (2), 162-168 (2005).

47 Crocetti, G.R., Banfield, J.F., Keller, J., Bond, P.L., Blackall, L.L., “Glycogen-accumulating organisms in laboratory-scale and full-scale wastewater treatment processes”,., 148, 3353-3364 (2002).

48 Liu, W.T., Mino, T., Matsuo, T., Nakamura, K., “Biological phosphorus removal processes-effect of pH on anaerobic substrate metabolism”,.., 34 (1/2), 25-32 (1996).

49 Kong, Y.H., Ong, S.L., Ng, W.J., Liu, W.T., “Diversity and distribution of a deeply branched novel proteobacterial group found in anaerobic-aerobic activated sludge processes”,.., 4 (11), 753-757 (2002).

50 Saunders, A.M., Oehmen, A., Blackall, L.L., Yuan, Z., Keller, J., “The effect of GAOs (glycogen accumulating organisms) on anaerobic carbon requirements in full-scale Australian EBPR (enhanced biological phosphorus removal) plants”,.., 47 (11), 37-43 (2003).

51 Wong, M.T., Tan, F.M., Ng, W.J., Liu, W.T., “Identification and occurrence of tetrad-formingin anaerobic-aerobic activated sludge processes”,, 150, 3741-3748 (2004).

52 Mino, T., Satoh, H., Matsuo, T., “Metabolism of different bacterial populations in enhanced biological phosphate removal processes”,.., 29 (7), 67-70 (1994).

2008-09-04,

2009-04-09.

the Project of Scientific Research Base and Scientific Innovation Platform of Beijing Municipal Education Commission (PXM2008_014204_050843), the National Natural Science Foundation of China (50808004), and the Doctoral Startup Research Program of Beijing University of Technology.

** To whom correspondence should be addressed. E-mail: wsy@bjut.edu.cn