Desalination 287 (2012) 109–115

Contents lists available at ScienceDirect

Desalination

j ourna l homepage: www.e lsev ie r.com/ locate /desa l

Comparative study of SMBR and extended aeration activated sludge processes in thetreatment of high-strength wastewaters

Hamed Mohammadi a, A. Sabzali b, M. Gholami c,⁎, E. Dehghanifard c, R. Mirzaei c

a Environ. Health Dept., School of Public Health, Zanjan University of Medical Sciences, Zanjan 4515786349, Iranb Department of Environmental Health Engineering, School of Public Health, Isfahan University of Medical Sciences, Iranc Department of Environmental Health Engineering, School of Public Health and Institute of Public Health Research, Tehran University of Medical Sciences P.O. Box: 6446-14155, Tehran, Iran

⁎ Corresponding author. Tel.: +98 21 88777674; fax:E-mail address: gholamim@tums.ac.ir (M. Gholami)

0011-9164/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.desal.2011.05.045

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 February 2011Received in revised form 17 May 2011Accepted 18 May 2011Available online 12 June 2011

Keywords:Membrane bioreactorActivated sludgeOrganic matterNitrificationHigh-strength wastewater treatment

This study was conducted to compare the performance of extended aeration activated sludge (EAAS) withsubmerged membrane bioreactor (SMBR) systems in the treatment of high-strength wastewater under thesame condition. The chemical oxygen demand (COD) concentration of the influent wastewater for the EAASand SMBR systems was adjusted between 500–2700 and 500–5000 mg/L, respectively. Results showed thatthe SMBR system produced a much better quality effluent than EAAS system in terms of COD, biochemicaloxygen demand (BOD5), total suspended solids (TSS) and ammonium. By increasing the COD concentration,the concentration of mixed liquor suspended solids (MLSS) and the removal efficiency of organic matter in theSMBRsystemwere increased regularly; however, the removal efficiency of COD in theEAAS systemwas irregular.The average BOD5/COD ratio of effluent in the EAAS and SMBR systems were 0.708±0.18 and 0.537±0.106,respectively. These showthat theorganicmatter in the effluent of theSMBRsystemwas lessdegradable, and thus,more biological treatment was achieved. Nitrification was completely achieved in the SMBR system, while theEAAS system could not complete the process.

+98 21 88779487..

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The activated sludge process (AS) does exist in a large number ofmodifications and variations. In the last decade or so the mostimportant development in practice can be observed in industrialwastewater treatment, nutrient removal (N and P), and bulkingcontrol technologies. Extended aeration is one of the modification ofAS which has been used for most sanitary wastewater treatmentplants. High retention time (HRT), low organic loading rate and lowactive biomass limit application of this process for treating industrialwastewater [1]. This is characterized by low BOD loadings and iscommonly used to treat wastewater from small communities, housingcolonies and schools. The aeration period is 24 h or greater. Theextended aeration can accept periodic loadings without becomingupset. Stability of the process results from large aeration volume andcomplete mixing of the tank contents [2]. Many advantages have beenreported for EAAS including low BOD effluent, low waste activatedsludge and low ammonia effluent [3]. These benefits are obtained atthe expense of the large bioreactors required to achieve the long SRTs.But for many small installations the benefits outweigh the drawbackssuch asmixing limitation in large reactors and high required energy tomeet the oxygen requirement [4].

Currently, membrane bioreactors (MBRs) are widely used inwastewater treatment to achieve high effluent quality, which is oftendifficult to be met effectively by conventional activated sludgeprocess. The advantages of MBR are a higher mixed liquid suspendedsolid (MLSS) concentration, lower excess sludge production, and thetreated water produced can be reused [5]. Because of high solidcontents in aeration basin and sedimentation basin from treatmentunits, a MBR system requires less space than activated sludge.

Unlike the conventional activated sludge system, the MBRs arecharacterized by a complete retention of the biomass inside of thebioreactor by using membrane filtration. This controls and increases thesludge retention time (SRT) independently from the hydraulic retentiontime(HRT) [5].HighSRTsenableone to increase the sludgeconcentration,to apply organic loading, and to enhance the pollutant degradation. Thisallows the development of slow-growingmicroorganisms able to removepollutants in the wastewater, resulting in improved removal rate [6].

The disadvantages associated with the MBR are mainly cost-related. High capital costs due to expensive membrane units and highenergy costs due to pressure gradient requirements characterize thesystem. Membrane fouling problems can lead to frequent cleaning ofthe membranes, which stops operation and requires clean water andchemicals. Another drawback is that waste activated-sludge mayexhibit poor filterability properties [7]. Additionally, when operated athigh SRTs, non-filterable inorganic compounds accumulating in thebioreactor can reach concentration levels that can be harmful to themicrobial population or membrane structure [8,9].

Table 1Specification of membrane used in the project.

Parameter Specification

Pore size 0.2 μmBase PolypropyleneSurface area of filtration 0.25 m2

Filter length 25.4 cmInner diameter 28 mmOuter diameter 69 mm

110 H. Mohammadi et al. / Desalination 287 (2012) 109–115

Some research has been done on the effect of operating parameterssuch as organic loading rate (OLR), solid retention time and food tomicroorganism ratio (F/M) on the performance of MBR. Trussell et al.[10] reported that themembrane fouling increasedwith increasing F/M,whereas How and Salwomir found that excellent organic removalefficiencieswere achieved in theMBRoperatedatdifferent SRTs but thatnitrification ceased when the SRT was less than 2.5 days [11].

MBRs have been used to treat various types of wastewaters with achemical oxygen demand (COD), ranging from about 100 to morethan 40,000 mg/L and hydraulic retention time (HRT) varying from4 h to several days [9]. Therefore, MBRs present a way to intensivelybiologically treat high COD or BOD5 [12]. There are successful full-scale MBR units installed in paper mills in the Netherlands and Franceshowing high effluent quality in terms of COD, BOD5 and TSS. It shouldbe noted that most of the MBR studies and reports focused on thetreatment of raw wastewater with just mechanical pre-treatment[13].

Besides efficient organicmatter removal, it has been demonstratedthat an MBR can easily obtain efficient nitrogen removal. Theimproved retention of nitrifiers and prolonged SRT, at which MBRsgenerally operate, provide an anoxic zone for denitrification [14–18].

Two types of MBR systems have been developed. One is the side-stream MBR system where the membrane is located outside ofaeration tank and where the membrane filtration occurs externallythrough recirculation subjected to a pressure drop by a fitted pump.The other is the submerged MBR (SMBR) system where themembrane is submerged in the aeration tank [12].

In the SMBR system, the membrane is submerged directly in theaeration tank. By applying low vacuumor by using the static head of themixed liquor, effluent is driven through the membrane leaving thesolids behind. In SMBRs, the scour required to control solid accumula-tion on themembrane surface is provided by coarse bubble aeration [9].

In this study, the performance of extended aeration activatedsludge (EAAS) was compared with a SMBR system for treating asynthetic high-strength wastewater. Experiments were carried out ina continuous stirred tank reactor (CSTR) with a submerged, hollowfiber membrane system. First, the EAAS was set up, and the treatmentprocess optimized. Then, the Membrane module was installed, andthe treatment process was investigated. Lastly, results for COD, BOD5,TSS and nitrogen removal of the two modules were compared.

The objectives that guided this study were: (1) to evaluate SMBRfor treating high strength wastewater; (2) to evaluate COD removalefficiency and nitrification in both systems; (3) to determine the effectof operating conditions (i.e. organic loading rate and flow rate) onCOD removal efficiency under continuous-flow conditions; and (4) tocompare SMBR with EAAS for treating high strength wastewater.

2. Materials and methods

2.1. Experimental set-up and operating condition

The SMBR system is composed by two different compartments:the aeration tank with an effective volume of 433 L, and the finalsedimentation tank, with a 325 L volume. The lowest solid loadingrate was selected to design the sedimentation basin. It can provide agood condition for comparing with the SMBR system at the bestoperating condition. The SMBR overall volume is 758 L, realized withone Plexiglas tank, in which the abovementioned compartments havebeen obtained. The membrane module installed vertically in theaeration basin, consisted of polyethersulfone (PES) hollow-fibermicrofilter (MF)membranes with a pore size of 0.2 μm and a filtrationarea of 0.25 m2. The aeration effluent passed through the membranesin the SMBR system step. The characteristics of the membrane used inthe project are presented in Table 1.

The initial activated sludge was brought from the Plascokar Saipawastewater treatment plant (Tehran, Iran), then incubated, acclimat-

ed and diluted with synthetic wastewater in the ratio of 2:1 (v/v). Theproduced solution was a diluted sludge with MLSS concentration of540 mg/L. All experiments were performed at room temperature(25 °C). Synthetic wastewater was prepared with ordinary tap waterand glucose as the main sources of carbon and energy, plus balancedmacro and micro nutrients. The synthetic wastewater consisted ofglucose, as the main source of organic carbon, and nutrients (perliter): 0.5 g NH4Cl, 0.25 g KH2PO4, 0.25 g K2HPO4, 25 mg CoCl2, 0.3 gMgCl2, 11.5 mg Zn Cl2, 10.5 mg CuCl2, 5 mg CaCl2 and 15 mg MnCl2,which were added when required. For instance, the concentration ofNH4Cl was changed when the ammonium concentration wasimportant in the nitrogen removal process. Sludge recycling (QR:Q)was changed based onMLSS concentration (Ave. 15%). After a start upperiod of 15 days, substrate was added and the reactor was operatedcontinuously for 60 days.

EAAS reactor was started at an organic loading rate (OLR) of0.5 gCOD/L.day and a hydraulic retention time (HRT) of 24 h for a periodof 20 days. Following this, the OLR was gradually increased togetherwith reduced HRT. Increasing OLR was done by increasing glucoseconcentration in raw synthetic wastewater. In order to reach steadystate condition after each change in COD concentration, laboratoryanalysis of effluentwas performed for three consecutive days. In fact thevalues given are average values and standard deviations for 5–9measurements over the final 2–4 days of operation at each OLR. Someactivated sludge according to MLSS concentration in aeration basin waswaster by butterfly valve near the reactor bottom. After this period, allcontents of the tank were emptied and then some activated sludge(from before process) was inoculated into reactor. After that, MF wasadded to the reactor and the EEAS reactor was optimized to SMBRreactor. The reactor was operated continuously for 60 days. Bothsystems spent the same operation time which enables us to comparethem together. In the SMBR process, the second basin (secondarysedimentation) of the EAAS process was omitted, and the membranemodule was installed vertically in the aeration basin.

To determine the treatment performance of the high-strengthwastewater in the pilot plant, the COD and BOD5 concentrations of thewastewater were increased by the addition of 1000, 1500, 2000, 2500,3000 and 5000 mg/L of glucose. The average COD concentration ofindustrial suborns in Iran was the basis for selecting the influent CODconcentration ranges.

TheHRTwaskept at 24 h. The influentflowtoeach reactorwas ranged0.15 to 0.2 m3/d which was adjusted by an adjustable dosing pump.

To compare the performance of the EAAS and SMBR, similarconditions were maintained in both treatment processes. Initially, theEAAS system was operated separately to establish how the biofilmscarry out the desired reactions. In the SMBR process, the second basin(secondary sedimentation) of the EAAS process was omitted, and themembrane module was installed vertically in the aeration basin. Thefeed pump was controlled by magnetic floaters that control thewastewater level using two states (off and on). A level sensor wasused to control the wastewater volume in the reactor. The required airwas supplied by a low pressure compressor. To improve the exchangeof oxygen, silicon diffusers with an approximate size of 20 cm wereused. A centrifugal pump was used to return the activated sludge to

1- Wastewater influent 5- Air flow meter 9- A ir diffuser

2- Feeding pump 6- Effluent suction pump 10- Aeration tank

3- Wastewater flow meter 7- Pressure meter gauge 11- Wastewater effluent

4- Air pump (Blower) 8- Microfilter membrane

Fig. 1. Schematic diagram of experimental set up.

111H. Mohammadi et al. / Desalination 287 (2012) 109–115

the aeration basin. Synthetic wastewater was pumped through a feedpump to the aeration basin. During reactor operation as a SMBR, thebio-floc separation was done by a MF membrane.

The diagram of the experimental set up is schematicallyrepresented in Fig. 1.

2.2. Sample analysis

Grab samples were taken from the influent and effluent of EAASand SMBR, transported to the laboratory and analyzed to determinethe approximate optimum concentrations of total COD, BOD5, TSS andNH4. The tentative, optimum concentrations obtained were thenoptimized using a two-factorial central composite design. Analysis ofthe response surface was performed using the computer program,Design Expert, Version 7.

The COD, BOD5, TSS, NH4, NO3, DO, F/M,MLSS, MLVSS and SVI wereanalyzed according to the Standard Methods [19]. Dissolved oxygen,pH and EC were measured using a Hach (HQ 40d) analyzer. Solidretention time and food: microorganism (F:M) ratio were calculatedby Eqs. 1 and 2, respectively [6]:

SRT =V × X

QrXr + QexXrð1Þ

Fig. 2. Variation of MLSS concentrations against op

FM

=Q × BODV × MLVSS

: ð2Þ

3. Results and discussion

As expected, sludge production was strongly increased when theCOD concentration was increased about 2000 mg/L and then remainsconstant with increasing COD concentration. When reactor wasmodified to SMBR the MLSS concentration was increased even in highconcentration of COD. The average and maximum values of the MLSSconcentration in the EAAS and SMBR systems were 2038 and3039 mg/L, and 4000 and 5000 mg/L, respectively (Fig. 2). Themaximum COD removal efficiencies of 98.3% and 92.2% were observedat the same HRT and the influent Ave. COD concentration of 2100 mg/Lfor SMBR and EAAS reactors, respectively. Also, the SMBR systemoperatedwith a higherMLSS concentration than the EAAS system. It hasbeen reported that the MLSS concentration range in MBR systems is2000–4000 mg/L. There was a direct relationship between pressuredrop in SMBR andMLSS concentration in aeration basin, as the pressuredrop was increased until 1.0 bar when the MLSS concentration wasincreased more than 4000 mg/L. This value may be increased to more

eration days in the EAAS and SMBR systems.

Fig. 3. Performance of the EAAS and SMBR systems in COD removal.

Fig. 4. Performance of the EAAS and SMBR systems in BOD5 removal.

112 H. Mohammadi et al. / Desalination 287 (2012) 109–115

than 20,000 mg/L depending on wastewater characteristics andoperation conditions [20].

Fig. 3 shows the performance of the EAAS and SMBR systems inCOD removal. It should be noted that in the EAAS system, the CODconcentrations changed with a constant value of HRT (24 h), after thereactor startup and treatment conditions were optimized. Thevariation of COD concentrations was up to 2700 mg/L. In the SMBRsystem, the concentration of the influent COD was the same as in theEAAS system (N500 mg/L) to better compare them. However, the CODconcentrations were increased up to 5000 mg/L due to the betterremoval efficiency of the SMBR system. It is noteworthy that the SRTvalues of the SMBR system differed, as shown in Table 2.

The comparison of BOD5 removal in both systems is shown inFig. 4.

As shown, the removal efficiency of the BOD5 in the SMBR systemwas more than the EAAS system; however, the removal trend of theBOD5 in the SMBR system was smaller than that for the EAAS.

The removal efficiency of TSS in the EAAS and SMBR systems isdisplayed in Fig. 5.

In considering the removal efficiency, it was apparent that therewasa significant difference between the performance of the EAAS and SMBRin TSS removal. Although the removal trend of both systems wasincreasing, the SMBR system, aswith the CODandBOD5 removal curves,had a slighter trend. Unlike the EAAS system, the SMBR systemwas abletobetter tolerate increasing TSS, andhigher TSSdidn't have any effect onits performance. Moreover, the removal trend of the consideredparameters in this study was linear, and it could be concluded that byincreasing the strength of the wastewater, the treatment capability ofboth systems decreased. However, the SMBR systemwas usable for thetreatment of higher CODs (up to 5000 mg/L).

The nitrogen removal efficiency range in the SMBR system withconcentrations of 38–54 mg/L-N was 88–99%, whereas in the EAASsystem, it was 71–98%. It should be noted that the nitrate removaloccurred due to the existence of an anoxic area in both systems, but theSMBR system causedmore nitrate removal due tomore biomass (Fig. 6).In thebiological nutrient removal processes (BNR), the removal efficiencyof nutrients is related to the type of influent organic matters. Chae and

Table 2SRT values in the SMBR system in different concentrations of COD.

COD concentration (mg/L) SRT value (day)

b1500 17±41500–3000 24±2N5000 32±2

Shin showed that by feeding proper biodegradable organic matter, theremoval efficiencies of total nitrogen (TN) and total phosphorous (TP) inan MBR system with an anoxic area were 74 and 78%, respectively [21].IncreasingOLR could also improve the nitrification process, as Yamamotoand Win also noted [22].

The BOD5/COD ratio in the effluent was plotted with respect to theOLR, as shown in Fig. 7.

Fig. 5. Performance of the EAAS and SMBR systems in TSS removal.

Fig. 6. Removal efficiency of nitrogen and nitrification performance against influent NH4–N concentration in the: (a) EAAS, and (b) SMBR systems.

113H. Mohammadi et al. / Desalination 287 (2012) 109–115

The BOD5/COD ratio in the effluent decreased proportionally to theOLR which means that the hydraulic retention time of the reactor waslower than required to degrade organic materials in aeration basin.The average of the effluent BOD5/COD ratio in the EAAS and SMBRsystems were 0.708±0.18 and 0.537±0.106, respectively. Thisshows that the organic matter in the effluent of the SMBR systemwas less degradable. It may be related to microbial metabolism andfinal production of bacterial auto-degradation. Shin et al. also found

Fig. 7. BOD5/COD ratio in the effluent

that the BOD5/COD ratio in the effluent of swine wastewater of aSMBR system was less biodegradable [12].

Based on the design parameters of the EAAS and SMBR systems,the F/M and OLR are important indicators of the wastewatertreatment. Fig. 8 shows the COD removal efficiency with differentrates of F/M and OLR, respectively. The aeration tank in the SMBRsystem could cultivate and maintain a higher biomass concentrationthan the EAAS system.

of the EAAS and SMBR systems.

Fig. 8. COD removal efficiency of the EAAS and SMBR systems with different values ofthe OLR.

114 H. Mohammadi et al. / Desalination 287 (2012) 109–115

In SMBR, the removal efficiency of the COD, BOD5 and TSS is related tothe COD concentration, OLR and other parameters which can affect thebiodegradation of wastewater. Generally, the removal efficiencies ofthese parameters in municipal wastewater plants have been 95, 98 and99%, respectively [23]. In this study, it was deduced that the removalefficiency inboth systemswasnot significantly changedby increasing theCOD concentration. With an OLR of 0.5 kg COD/m3, the removalefficiencies in the EAAS andSMBRsystemswere91 and96%, respectively,and by increasing the OLR up to 2.5 kg COD/m3, the efficiencies were 92and 97%, respectively.

The performance of the SMBR system was significantly higher thantheEAAS system(Pvalueb0.01). TheeffluentCODwas less than100 mg/L,even with high values of the OLR. However, statistical analyses (t-test)showed that the COD concentration of the synthetic wastewater wassignificantly reduced in both systems (the SMBR and EAAS) with highOLRs (Pvalueb0.01). The results demonstrated that the influent CODconcentration in the EAAS system had a positive relationship with theMLSS and SVI, while a negative relationship was observed with the DO(Pvalueb0.01).

Based on themain objectives and results of this study, it appears thatthe SMBR system could treat high-strength wastewaters. The removalefficiency of the COD in the SMBR system was 97.7% (influentCOD=5000 mg/L; OLR=5 kg COD/m3 day). However, the removalefficiencymay vary by changing the type of wastewater organic matter.Visvanathana et al. investigated the performance of a MBR system forleachate treatment and showed that the removal efficiencies for theCOD and BOD were 62–79% and 60–75%, respectively [24].

Nevertheless, the OLR values in MBR systems for medium waste-water are considered similar to activated sludge systems; the CODremoval efficiency in MBR systems is 90–98%. The value for activatedsludge systems is 75–85% under the same conditions. Better perfor-mance for MBR systems depends on:

1 Complete retention of the TSS (including COD and organic matterswith high molecular weights) in the system.

2 Prevention of biomass washout (a widespread problem in activatedsludge systems).

It is noteworthy that the second condition causes refractoryorganic matters to degrade biologically due to the supply of properconditions for microbial growth.

Generally, the OLR in the treatment of industrial wastewaters byMBR systems has beenmore than that for municipal wastewaters, dueto the concentration of industrial wastewaters. The range of OLR has

been 0.25–16 kg COD/m3 day with the removal efficiency of 90–99.8%[16].

4. Conclusion

The comparison of EAAS and SMBR as the bio-treatment stage forhigh-strength wastewater revealed that the SMBR could produce aneffluent of much better quality in terms of BOD5, COD, TSS and nitrogen.SMBR effluent with a uniform quality could exclude the necessity offiltration needed to meet more stringent wastewater discharge stan-dards.Membrane blockage because of scaling and biofoulingmaybe veryserious problems for the MBR system. This investigation demonstratedthat the SMBR can be used to treat high strength wastewater with goodCOD removal efficiency. On the other hand, high quality effluent withvery less suspended solids can be achieved by optimizing activatedsludge to SMBR.

Acknowledgment

Authors greatly appreciate the Research Deputy of TehranUniversity of Medical Sciences for their financial support of this study.

References

[1] W.W. Eckenfelder, Activated Sludge Process Design and Control, Theory andPractice Taylor & Francis Routledge, Lancaster, 1998.

[2] All about wastewater treatment, First Edition: December, 2005, Geostar PublishingLLC. pp:184.

[3] F.R. Spellman, Volume 3, Spellman's Standard Handbook for WastewaterOperators—Advanced Level, Published by Technomic Publishing, 2000 pp:106cbe.

[4] C.P. Leslie Grady Jr., G.T. Daigger, H.C. Lim, Biological Wastewater Treatment, 2nd,Marcel Dekker, Inc, 1999, pp. 381–382.

[5] F.G. Meng, F.L. Yang, B.Q. Shi, H.M. Zhang, A comprehensive study on membranefouling in submerged membrane bioreactors operated under different aerationintensities, Sep Purif Technol. 59 (2008) 91–100.

[6] Y. Xue, F. Yang, S. Liu, Z. Fu, The influence of controlling factors on the start-up andoperation for partial nitrification in membrane bioreactor, Bioresource Technol.100 (2009) 1055–1060.

[7] C. Nazim, J.P. Franco, M.T. Suidan, V. Urbain, J. Manem, Characterization andcomparison of membrane bioreactor and conventional activated sludge system inthe treatment of wastewater containing high-molecular-weight compounds,Water Environ. Res. 71 (1999) 64–70.

[8] C. Nazim, D. Dionysiou, M.T. Suidan, P. Gienestet, J.M. Audic, Performancedeterioration and structural changes of ceramic membrane bioreactor due toinorganic abrasion, J Membr Sci. 163 (1999) 19–28.

[9] T.A. Mohammed, A.H. Birima, M.J.M.M. Noor, S.A. Muyibi, A. Idris, Evaluation ofusing membrane bioreactor for treating municipal wastewater at differentoperating conditions, Desalination 221 (2008) 502–510.

[10] R.S. Trussell, R.P. Meerlo, S.W. Hermanowicz, D. Jenkins, The effect of organic loadingon process performance and membrane fouling in a submerged membranebioreactor treating municipal wastewater, Water Res. 40 (2006) 2675–2683.

[11] Y.N. How, W.H. Salwomir, Membrane bioreactor operation at short solidsretention times: performance and biomass characteristics, Water Res. 39 (2005)981–992.

[12] J.H. Shin, S.M. Lee, J.Y. Jung, Y.C. Chung, S.H. Noh, Enhanced COD and nitrogenremovals for the treatment of swine wastewater by combining submergedmembrane bioreactor (MBR) and anaerobic upflowbedfilter (AUBF) reactor, ProcessBiochem. 40 (2005) 3769–3776.

[13] M. Lerner, N. Stahl, N.I. Galil, Comparative study of MBR and activated sludge inthe treatment of paper mill wastewater, Water Sci. Technol. 55 (2007) 23–29.

[14] H. Monclús, J. Sipma, G. Ferrero, I. Rodriguez-Roda, J. Comas, Biological nutrientremoval in anMBR treating municipal wastewater with special focus on biologicalphosphorus removal, Bioresource Technol. 101 (2010) 3984–3991.

[15] J.E. Fleischer, T.A. Broderick, G.T. Daigger, A.D. Fonseca, R.D. Holbrook, S.N. Murthy,Evaluation of membrane bioreactor process capabilities to meet stringer effluentnutrient discharge requirements, Water Environ. Res. 7 (2005) 162–178.

[16] S. Judd, The MBR Book: Principles and Applications of Membrane Bioreactors inWater and Wastewater Treatment, 1st ed. Elsevier, Oxford, 2006.

[17] K. Kubin, M. Kraume, W. Dorau, Investigation of nitrogen removal in a cascadedmembrane bioreactor, Water Sci. Technol. 46 (2002) 241–247.

[18] H. Monclús, J. Sipma, G. Ferrero, J. Comas, I. Rodriguez-Roda, Optimization ofbiological nutrient removal in a pilot plant UCT–MBR treating municipalwastewater during start-up, Desalination 250 (2010) 592–597.

[19] APHA, AWWA, WEF, Standard Methods for the Examination of Waters andWastewaters, 21st ed, American Public Health Association (APHA), Washington, DC,2005.

[20] T. Nodel, A. Mostafaei, Evaluation of membrane bioreactors performance inmunicipal and industrial wastewater treatment, 1st Conference of EnvironmentalEngineering, Tehran, Iran (2006) (In Persian).

115H. Mohammadi et al. / Desalination 287 (2012) 109–115

[21] S.R. Chae, H.S. Shin, Characteristics of simultaneous organic and nutrient removalin a pilot-scale vertical submerged membrane bioreactor (VSMBR) treatingmunicipal wastewater at various temperatures, Process Biochem. 42 (2007)193–198.

[22] K. Yamamoto, K.M. Win, Tannery wastewater treatment using a sequencing batchmembrane reactor, Water Sci. Technol. 23 (1991) 1639–1648.

[23] N.L. Aileen, N.G. Albert, S. Kim, A mini-review of modeling studies on membranebioreactor (MBR) treatment for municipal wastewaters, Desalination 212 (2007)261–281.

[24] C. Visvanathana, M.K. Choudharya, M.T. Montalboa, V. Jegatheesan, Landfillleachate treatment using thermophilic membrane bioreactor, Desalination 204(2007) 8–16.