Q IWA Publishing 2010 Water Science & TechnologyWST | 61.10 | 2010Water Science & TechnologyWST | 61.10 | 2010 Ch. Brepols et al. | Design and financial feasibility of full-scale membrane bioreactorsConsiderations on the design and financial feasibility of full-scale membrane bioreactors for municipal applicationsCh. Brepols, H. Scha fer and N. EngelhardtABSTRACTBased on the practical experience in design and operation of three full-scale membrane bioreactors (MBR) for municipal wastewater treatment that were commissioned since 1999, an overview on the different design concepts that were applied to the three MBR plants is given. The investment costs and the energy consumption of the MBRs and conventional activated sludge (CAS) plants (with and without tertiary treatment) in the Erft river region are compared. It is found that the specific investment costs of the MBR plants are lower than those of comparable CAS with tertiary treatment. A comparison of the specific energy demand of MBRsand conventional WWTPs is given. The structure of the MBRs actual operational costs is analysed. It can be seen that energy consumption is only responsible for one quarter to one third of all operational expenses. Based on a rough design and empirical cost data, a cost comparison of a full-scale MBR and a CAS is carried out. In this example the CAS employs a sand filtration and a disinfection in order to achieve comparable effluent quality. The influence of membrane lifetimeon life cycle cost is assessed.Key words | energy consumption, investment cost, MBR, membrane lifetime, operation cost, tertiary treatment Ch. Brepols (corresponding author)H. Scha ferN. EngelhardtErftverband,Am Erftverband 6,50126 Bergheim/Erft, GermanyE-mail: info@erftverband.deINTRODUCTIONMarket reports indicate that the membrane bioreactor (MBR) market is currently experiencing accelerated growth, which is expected to be sustained over the next decade (Judd2006). For Europe alone Lesjean et al. (2009) report an expected growth of 6 large installations (. 5,000 m3/d) per year. Although MBR technology is often viewed as being state-of-the-art, it is also sometimes perceived as high-risk and prohibitively costly compared with more established conventional technologies such as activated sludge plants (Judd 2006). It is often difficult to compare cost and operational data from conventional activated sludge (CAS) plants and MBRs because of the regionally diverse design and construction practices and different boundary conditions.The Erftverband, a German regional water managementassociation, owns and operates 44 municipal wastewater treatment plants (WWTPs) within its region, of which three are MBRs. A total of 1.06 million population equivalents (PE) are served. Biological nutrient removal is common at these 44 WWTPs. One third of the CAS additionally employ tertiary treatment to meet stringent discharge criteria. The increasing investment cost for high quality wastewater treatment spurred the companys interest in MBR technol- ogy and finally led to the commissioning of the three MBR plants with a maximum hydraulic capacity ranging from3,240 to 45,000 m3/d.Some design recommendations based on the Erftver- bands actual planning and operational experience are given. The investment costs and energy consumption of existing conventional and membrane wastewater treatmentplants are presented as cost functions. Life-cycle costs for adoi: 10.2166/wst.2010.179MBR and a CAS plant are assessed based on a cost comparison method.PLANT DESIGNSince 1999 the Erftverband has commissioned three full-scale membrane bioreactor plants: Ro dingen (1999), Nordkanal (2004) and Glessen (2008). Their main technical data are displayed in Table 1.Because the Ro dingen plant was designed and built as a large-scale pilot to introduce MBR technology for municipal installations, it features a highly variable process layout with two identical bioreactors both equipped for denitrification and nitrification and two lines of submersed hollow fibre ultra-filtration membranes installed in two separate filtration tanks (Engelhardt & Firk 2000). After years of process optimisations the biological wastewatertreatment today is accomplished by simultaneous aerobic sludge stabilisation and alternating denitrification. The membrane filtration tanks were initially equipped with two identical lines of 6 GE Zenon ZW 500 A cassettes. After a severe clogging and fouling incident, that was mainly the result of inappropriate mechanical pre-treatment of the wastewater, the membrane filters were replaced by ZW500 C modules in 2000/2001. By the end of 2008 the filtration units of the longer serving line were replaced by KMS Puron modules.The Nordkanal plant opened up a new dimension of scale in membrane bioreactors (Brepols et al. 2000). With eight separate filtration lines in four bioreactors it uses84,400 m2 of GE Zenon ZW 500 C membranes inside thebioreactor.At Glessen an existing conventional plant was retro- fitted by using membrane bioreactor technology. The existing bioreactor and the sludge storage tank were retained while the old secondary clarifier was transformedinto a buffer tank. A new pre-treatment facility had to beTable 1 | Technical dataRo dingen, 3,000 PE Nordkanal, 80,000 PE Glessen, 9,000 PEDaily inflowMaximum 3,240 m3/d 45,000 m3/d 6,430 m3/d Average 560 m3/d 12,000 m3/d 2,400 m3/d Influent load COD 423 kg/d COD 9,600 kg/d COD 1,008 kg/d(Design values) TN 40 kg/d TN 897 kg/d TN 99 kg/dPtot 10 kg/d Ptot 123 kg/d Ptot 15 kg/dPre-treatementStep screen 3 mm; Aerated sand/grit chamber, Rotary cutter, Bypass sieve 0.5 mm for sludge screening Step screen 5 mm; Aerated sand/grit chamber,Mesh sieve 1.0 mm Step screen 6 mm; Aerated sand/grit chamber, Mesh sieve0.8 mmBioreactor 400 m3 80 m3 (filtration tanks) 9,300 m3 1,600 m3 320 m3(filtration tanks)Specific volume 160 L/p.e. Specific volume 116 L/p.e. Specific volume 213 L/p.e.Process configuration Simultaneous aerobic sludge stabilisation, alternating or upstream denitrificationMembrane 5,280 m2 hollow fibre modules, separate filtration tanks Simultaneous aerobic sludge stabilisation, upstream denitrification84,480 m2 hollow fibre modules, integrated filtration Simultaneous aerobic sludge stabilisation, alternating denitrification10,200 m2 hollow fibre modules, separate filtration tanksEffluent requirements COD , 30 mg/L COD , 90 mg/L COD , 30 mg/L (Grab sample) NH4-N , 4 mg/L NH4-N , 10 mg/L NH4-N , 1.5 mg/LPtot , 1.5 mg/L Ptot , 1.5 mg/L Ptot , 0.6 mg/LWater Science & TechnologyWST | 61.10 | 2010 Ch. Brepols et al. | Design and financial feasibility of full-scale membrane bioreactorsWater Science & TechnologyWST | 61.10 | 2010 Ch. Brepols et al. | Design and financial feasibility of full-scale membrane bioreactorsbuilt to meet the operating demands of the membrane filtration units. The membrane filtration system itself was built separately. So the old plant could be maintained in full operation during the construction phase. Finally the effluent from the bioreactor tank was shifted from the secondary clarifier to the new membrane filtration units in stages. Each of the four filtration tanks is equipped with two GE Zenon 500 D cassettes. After the full commissioning of the membrane filtration, the biological treatment capacity was increased by doubling the MLSS concentration in the bioreactor from 4 g/L to 8 g/L (Brepols & Schaefer 2009).Generally MBR technology offers several options for retrofitting of existing WWTPs (Brepols et al. 2007). According to Bagg (2009), the most practical and least disruptive way of achieving a retrofit upgrade is by providing external membrane process volume and then integrating operation of this with the existing facility. Similar conclusions were drawn by Hashimoto et al. (2009) who investigated the potential of retrofitting with MBRs in Japan. For the full flow upgrade at Glessen the only viable option was to replace the secondary clarifier by a new membrane filtration system (Figure 1).The underlying design concept of the ErftverbandMBRs changed over the years, reflecting the ongoing development in MBR technology as well as the Erftver- bands own operational experience. The pre-treatment facility at the Ro dingen and Nordkanal MBRs had to be upgraded and the membrane cleaning concept of the plantswas overhauled several times. De Wever et al. (2009) published a decision tree for the configuration of submerged MBR in which the size of the installation, effluent requirements and type of membrane are key factors. Submerged MBR technology offers two basic options for choosing a suitable plant layout: (a) outside configuration with additional filtration tanks and a separate circulation of mixed liquor between the bioreactor and the filtration units as at the Ro dingen and Glessen MBR or (b) inside configuration with the filtration units directly integrated into the bioreactor tank as at the Nordkanal plant. De Wever et al. report that in most cases the outside configuration is preferable because it offers higher oper- ational flexibility, easier membrane maintenance and cleaning and better effluent quality, while the inside solution can be favourable for smaller plants with flat sheet membrane filtration units. By contrast, the large Nordkanal MBR displays the inside configuration. In this case savings in the civil construction were accomplished by reducing the plant footprint and the complexity of the basins although this came at the cost of a reduced flexibility in the membrane cleaning procedures. Operational experi- ence at the Nordkanal MBR suggests, that the inside concept is also feasible for larger installations with hollow fibre membranes (Engelhardt 2005).INVESTMENT COSTTwo thirds of the Erftverbands 44 wastewater treatment plants have to fulfil treatment requirements that are moreFigure 1 | Aerial view of the retrofitted Glessen MBR.stringent than the federal governments standard regu- lations, as they discharge into sensitive rivers. Many of the plants were built or reconstructed during the last 20 years. CAS were often complemented with tertiary treatment in the form of sand-filtration facilities and in some cases post- nitrification reactors.To facilitate the cost comparison between CAS, CASwith tertiary treatment and MBRs the investment cost data of 26 WWTPs from the years 1989 to 2008 were selected. Glessen MBR Nordkanal MBR Rdingen MBR Civil worksMechanical equipment Electrical equipmentMembrane filtersThe WWTPs were either built as greenfield installations or underwent a major reconstruction on site, resulting in a relevant increase of their treatment capacity or a significant improvement of their effluent quality. WWTPs that only underwent minor retrofitting are not included. During the period the construction cost indices for industrial buildings increased by 55.5% (Landesamt 2008). The actual construc- tion prices have been adjusted to correspond with these indices for the reference year of 2000.The plants are all situated in the Erft river region. The WWTPs were built under the awarding authority of the Erftverband and share similar standards in civil construc- tion and mechanical and electrical equipment. The data include plants with simultaneous aerobic sludge stabilis- ation and plants with separate anaerobic sludge stabilis- ation. Plants with separate anaerobic sludge stabilisation are mainly found above a capacity range of 20,000 m3/d. The plants with simultaneous aerobic sludge stabilisation, including the MBRs, are designed for an SRT . 25 d.Figure 2 shows the specific investment cost related to treatment capacity. Two distinct cost functions were retrieved for CAS with and without tertiary treatment, each resembling a hyperbola. The specific costs of CAS with 0% 20% 40% 60% 80% 100%Figure 3 | Investment cost structure of the Erftverband MBRs.tertiary treatment are typically higher than the cost of other CAS. The MBRs are found in the range of the CAS without tertiary treatment. In terms of investment cost the MBRs thus show an advantage compared to CAS with tertiary treatment. It seems that the additional cost for the membrane filtration system is more than compensated by savings in civil construction cost.While conventional plants take a greater share in civil works cost the cost of MBR shift towards the equipment. Figure 3 shows the investment cost structure of three MBRs. Pre-treatment, membrane filtration and sophisticated pro- cess control account for this cost shift in large measure. Wedi (2003) has published a cost share of 44% for mechanical equipment against only 39% of construction cost. These values resemble also the cost structure of the Nordkanal and Glessen MBRs displayed in Figure 3, while the cost distribution of the Ro dingen plant is mainly influenced by the high cost of the membrane filtration system at the time of construction. Since then the specific tender price of membrane filtration units approximatelydeclined by two thirds.2,000Specific cost (EUR per PE)1,5001,0005000 CASCAS with tertiary treatmentMBRCost function (CAS)Cost function (CAS with tertiary treatment)0 20,000 40,000 60,000 80,000 100,000Maximum treatment capacity (m3/d) MEMBRANE REINVESTMENTThe membrane filtration units remain an expensive part of the equipment and their life-time is limited. Reinvestment in the membrane filters thus plays a decisive role in the long-term economic success of an MBR. Based on an evaluation of the permeability values in a municipal facility, De Wilde et al. (2007) have projected a theoretical lifetime of 13 years before the filtration capacity decreases to a levelFigure 2 | Investment costs for municipal WWTPs in the Erft region. where membrane replacement becomes inevitable. At theRo dingen MBR eight years of membrane lifetime were achieved under suboptimal conditions. The wastewater pre-treatment at Ro dingen does not meet todays standards for MBR plants, and the membrane filters suffered from more clogging, fouling and wear than in other installations.When the membrane filters were replaced at Ro dingen in 2008, it was found that the limited compatibility between the membrane filtration systems of the different suppliers generated additional cost for changes in the peripheral equipment. This might also happen when using a new generation of membrane filters from the same supplier.OPERATIONAL COSTEnergy consumption is a driving factor for the operational cost of membrane bioreactor plants. Figure 4 shows the actual average energy demand of the Erftverbands WWTPs over a period from 2005 to 2007, including the two MBRs that were in operation at that time. The value for the Glessen MBR, which was started up in April 2008, was included based on data from the first year of operation. The energy consumption values in Figure 4 are calculated as the quotient of the total power uptake for the WWTPs including all wastewater and sludge treatment and the actual amount of wastewater treated. Again two individual functions were derived for CAS with and without tertiary treatment.The data include energetically optimised and non- optimised plants. Some plants actually receive dry weather loads that are significantly lower than their design value, leading to a disproportionately high specific energy consumption. These effects interfere with the usual economies of scale. The trend calculated for the CAS is less pronounced and seems to converge with the trend for CAS with tertiary treatment. Comparable guide values for energetically optimised CAS can be found in a range from 65 kWh a21 PE21 (, 1000 PE) to 27 kWh a21 PE21 (. 100,000 PE) (MUNLV 1999). This correlates to a range of approximately 0.9 to 0.3 kWh/m3 depending on the specific wastewater production per PE and the total annual amount of storm water treated.Figure 4 shows that the energy consumption of the MBRs with low or medium capacity is in the range of the CAS with tertiary treatment. For the large Nordkanal MBR the value in the graph is clearly above average. Because of its size, Nordkanal MBR competes here with CAS that employ anaerobic sludge treatment, have shorter SRTs and generally have a lower energy demand than WWTPs with simultaneous aerobic sludge stabilisation. Additional points to keep in mind are that the Ro dingen MBR is considerably underloaded during dry weather conditions and reflects an older level of knowledge in MBR technology and that the Glessen and Nordkanal MBR have not been optimised energetically at that time.Pyo ry (2009) conducted a field study to investigate the main energy consumers and the potential for energy savings at Nordkanal MBR. It turns out that the air scouring is responsible for 49.2% of the energy consumption. Process aeration in the bioreactor (12.1%), biomass circulation (1.3%), bioreactor mixing (11.5%) and permeate suction (2.7%) take significantly lower shares. The remainder (23.2%) is consumed by pumping stations, pre-treatment, sludge dewatering and miscellaneous process units.A conventional optimisation of the given system at2.502.00SED (kWh/m3)1.501.000.500.00 CASCAS with tertiary treatmentMBRFunction (CAS)Function (CAS with tertiary treatment)0 5,000 10,000 15,000 20,000 25,000Average daily amount of wastewater treated (m3/d) Nordkanal would immediately reduce the specific energy demand by 15% while under ideal conditions an energy consumption of 0.45 kWh/m3 could be achieved. Garces et al. (2007) and Tao et al. (2009) also report that a 35% to 40% reduction in energy consumption for air scouring can be achieved by applying a different aeration regime, while Tao et al. give a value of 0.4 kWh/m3 for an energetically optimised MBR with a capacity of 23,000 m3/d.Cost estimation models for MBRs which have beenpublished in recent years (Yoon et al. 2004; Verrecht et al.Figure 4 | Actual specific energy demand (SED) of WWTPs in the Erft region. 2008) focus mainly on the aeration energy demand and1%1%6%30% Rdingen9% 25%7%21% 3%11%7% 4%16% Nordkanal8% 30%21% EnergySludge and waste disposalPersonnel Maintenance Operating fluids Operating materials Wastewater levyMiscellaneousFigure 5 | Distribution of operational costs of the Ro dingen and Nordkanal MBR. Subscribers to the online version of Water Science and Technology can access the colour version of this figure from http://www.iwaponline.com/wstsludge production, while other cost items such as personnel, wastewater pre-treatment, membrane cleaning and main- tenance are not considered. Figure 5 shows that energy demand accounts only for one quarter to one third of all operational cost at the Ro dingen and Nordkanal MBR. Personnel cost and sludge and waste disposal together take42% of the share at Nordkanal and 28% at Ro dingen. The absolute values of these cost items are comparable to CAS plants, as there are no significant differences in labour utilisation or the actual amount of waste produced. There is also a change in the share of the cost items over time. The Ro dingen plant has been operating five years longer than the Nordkanal MBR. The data suggest that the share of maintenance cost is likely to increase with the age of the installation. features sand-filtration and UV disinfection. A rough design is made for both installations assuming that the SRT is at 25 days and the MLSS concentration in the bioreactor is 4 g/L at the CAS and 12 g/L at the MBR. Reactor volumes and plant footprints are calculated. The area of the paved surface at the CAS is an estimated 2,400 m2 compared toonly 900 m2 at the MBR. Property prices for acquiring abuilding site are not taken into account. The investment and operational costs shown in Table 2 are estimated on the basis of empirical specific cost for the respective plant units that are typical for the Erft region.The net present values (NPV) of the MBR and the CAS are calculated for a 30 year period. The lifetime of the civil works at both installations is fixed at 30 years. For themechanical and electrical equipment, excluding theTable 2 | Investment cost and annual operating costDuring preliminary planning it is common to perform cost comparison calculations in order to determine the most economic solution among various options. As an example, a cost comparison method (LAWA 2005) is employed to estimate the life-cycle costs of an MBR and a CAS and to investigate the influence of membrane lifetime on the economic efficiency of the MBR. It is assumed that the WWTPs have a capacity of 10,000 PE, are greenfield installations, use biological nutrient removal and phosphor- ous precipitation and produce a comparable effluentInvestment cost7,824,5006,102,000Civil works4,960,0972,248,507Mechanical equipment2,152,6142,500,029Electrical equipment711,789795,464Membrane filtrationOperating cost per year, OPEX256,730558,000287,726Maintenance and repair96,41196,420Waste and sludge disposal25,66425,664Personnel70,00070,000Energy45,99082,782Operating fluids11,00012,860tanks and hollow fibre membrane filtration. The CASDisinfection7,665EXAMPLE OF A LIFE-CYCLE COST CALCULATION CAS, cost in EUR MBR, cost in EURquality. The MBR is designed with separate filtration16.0NPV (million EUR)15.615.2 CAS(1) MBR(1) CAS(2) MBR(2) consumption than the like sized CAS. However there is still a significant unrealised potential to increase the energy efficiency of MBR plants of all sizes. This holds the promise14.814.40.0 x(1) = 7.1 yrs x(2) = 8.2 yrs Another key to long term economic efficiency is0 2 4 6 8 10 12 14Life time (yrs)Figure 6 | Net present values with constant energy cost (1) and increasing energy cost (2).membrane filters, the lifetime is 15 years. The cost comparison is performed by using different lifetime values of the membrane filters of 5, 7.5 and 10 years. It is assumed that the specific cost for membrane reinvestment remain constant over the years. Interest rates are fixed at 3%. The NPVs shown in Figure 6 are calculated based on the respective reinvestment cycles.The increase in membrane filter lifetime from 5 to 10 years results in a EUR 1.1 million decrease of the NPV. It can be seen that the break-even point (x) of the MBR and the CAS under the given circumstances comes at a membrane lifetime of 7.1 years. A sensitivity analyses with energy cost increasing by 2% annually shows a shift of this point to 8.2 years. The example illustrates that under the specified conditions CAS and MBR can be equally economic when a reasonable membrane filter lifetime is achieved. This example should not be generalised. The precise economic characteristics of any decision between CAS and MBR need to be assessed individually. On a case to case basis monetary but also non-monetary aspects need to be taken into account.CONCLUSIONSA comparison of the actual investment cost of 26 WWTPs for municipal wastewater treatment has shown that the investment cost of MBR plants can be competitive to CAS plants under comparable boundary conditions.According to the operational results for the plants considered here, the small and medium sized MBRs do not have a disadvantage in actual specific energy consumption compared to like-sized CAS with tertiary treatment. In this comparison the large MBR shows a higher energy membrane lifetime. Practical experience at the Ro dingenMBR has shown that a lifetime of eight years is achieved under unfavourable conditions. An Example of a cost comparison demonstrates, that the life cycle cost of an MBR and a CAS plant with tertiary treatment can be equal at a membrane lifetime of seven to eight years.REFERENCESBagg, W. K. 2009 Infrastructure optimisation via MBR retrofit: a design guide. Water Sci. Technol. 59(2), 323 330.Brepols, Ch. & Schaefer, H. 2009 Kla ranlage Glessen, Planung Bau und Betrieb der dritten Membranbelebungsanlage des Erftverbandes, 42. Essener Tagung fu r Wasser- und Abfallwirtschaft, Aachen 2009, ISBN: 978-3-938996-23-2.Brepols, Ch., Engelhardt, N. & Firk, W. 2000 Kla ranlage Nordkanaldie gro te Membrankla ranlage der Welt? (Nordkanal WWTPthe worlds biggest MBR?), 18. Bochumer Workshop Siedlungswasserwirtschaft, Ruhr-Universitat Bochum.Brepols, Ch., Dorgeloh, E., Frechen, F. -B., Fuchs, W., Haider, St., Joss, A., de Korte, K., Ruiken, Chr., Schier, W., van der Roest, H., Wett, M. & Wozniak, Th. (Working group KA-7.4 ofDWA) 2007 Upgrading and Retrofitting of Wastewater Treatment Plants by Means of Membrane Bioreactor (MBR) Technology. Fourth International Water Association Conference on Membranes for Water and Wastewater Treatment, 15 17 May 2007, Harrogate,ISBN: 978-1-86194-127-5.De Wever, H., Brepols, C. & Lesjean, B. 2009 Decision tree for full- scale submerged MBR configurations. Final MBR-Network workshop Salient outcomes of the European projects on MBR technology, 31 March 1 April 2009, Berlin, Germany, ISBN:978-3-9811684-5-7.De Wilde, W., Thoeye, C. & De Gueldre, G. 2007 Membrane life expectancy assessment after 3 years of MBR operation at WWTP Schilde. Fourth International Water Association Conference on Membranes for Water and Wastewater Treatment, 15 17 May 2007, Harrogate.Engelhardt, N. 2005 Das Gruppenkla rwerk Nordkanal nach einen Jahr Betriebeine Perspektive fu r groe Membranbelebungsanlagen, Membrantechnik in der Wasseraufbereitung und Abwasserbehandlung, 6. Aachener Tagung Siedlungswasserwirtschaft und Verfahrenstechnik, Aachen, 2005, ISBN: 3-86130-775-8.Engelhardt, N. & Firk, W. 2000 Erste Betriebserfahrungen mit demMembranbelebungsverfahren auf der kommunalenKla ranlage Ro dingen (3.000 EW). (First OperationalExperiences at the Ro dingen Municipal MBR) Begleitbuch zur3. Aachener Tagung Siedlungswasserwirtschaft undVerfahrenstechnik, Aachen.Garces, A., De Wilde, W., Thoeye C. & De Gueldre, G. 2007Operational cost optimisation of MBR Schilde. Fourth International Water Association Conference on Membranes for Water and Wastewater Treatment, 15 17 May 2007, Harrogate, ISBN: 978-1-86194-127-5.Hashimoto, T., Suzumura, K., Itokawa, H. & Murakami, T. 2009Study of MBR system with separate membrane tank forthe reconstruction of large-scale WWTPs. Fifth IWA Specialised Membrane Technology Conference for Water and Wastewater Treatment, 1 3 September 2009, Beijing, PR China.Judd, S. 2006 The MBR Book, Principles and Application of Membrane Bioreactors in Water and Wastewater treatment. Elsevier, Oxford, ISBN-13:978-1-85-617481-7.Landesamt fu r Datenverarbeitung und Statistik Nordrhein Westfalen 2008 Ausgewa hlte Baupreisindizes in NRW (Selected price indices for civil construction in North-Rhine Westphalia), http://www.lds.nrw.de/includes/php/ druckversion.php?site/statistik/datenangebot/daten/m/ r323bauindex.html (accessed 05.08.2008).LAWA, La nderarbeitsgemeinschaft Wasser 2005 Leitlinien zur Durchfu hrung dynamischer Kostenvergleichsrechnungen (KVR-Leitlinien), Ausgearbeitet vom LAWA-Unterausschuss Wirtschaftlichkeitsfragen in der Wasserwirtschaft/ KVR-Leitlinien 2005, Kulturbuchverlag Berlin GmbH, ISBN: 3-88961-240-7.Lesjean, B., Ferre, V., Vonghia, E. & Moeslang, H. 2009 Market and design considerations of the 37 larger MBR plants in Europe. Desalination Water Treat. 6, 227 233.MUNLV 1999 Handbuch Energie in Kla ranlagen. Herausgeber: Ministerium fu r Umwelt, Raumordnung und Landwirtschaft des Landes NRW, Germany.Po yry 2009 Abschlussbericht zum Forschungsvorhaben Ermittlung energetischer Beurteilungskriterien fu r Membrankla ranlagen mit Hohlfasermodulen und deren Verifizierung an einer Groanlage, Auftraggeber: ErftverbandBergheim, Verfasser: Po yry GKW GmbH, Essen, den 27.02.2009. Tao, G., Kekre, K., Htun Oo, M., Viswanath, B., Lew, C. H., Kan,L. M. & Seah, H. 2009 Large scale membrane bioreactor plant design (retrofit) and optimisation. Fifth IWA Specialised Membrane Technology Conference for Water and Wastewater Treatment, 1 3 September 2009, Beijing, PR China.Verrecht, B., Judd, S., Guglielmi, G., Brepols, C. & Mulder, J. W.2008 An aeration energy model for an immersed membrane bioreactor. Water Res. 42, 4761 4770.Wedi, D. 2003 Wirtschaftlichkeit des Membranbelebungsverfahrens(Profitability of the MBR process), ATV-DVWK Membrantage,1./2. Juli 2003 in Bonn.Yoon, S., Kim, H. & Yeom, I. 2004 The optimum operational condition of membrane bioreactor (MBR): cost estimation of aeration and sludge treatment. Water Res. 38, 37 46.Copyright of Water Science & Technology is the property of IWA Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.Copyright of Water Science & Technology is the property of IWA Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.