* 2005

Overview of developments and projects

Future research directions

Innovation by co-operation

Dutch MBR grows to maturity

April

Magazine for water supply and water managementMBRSPECIAL IIIMBRSPECIAL III

The STOWA is an organisation for the initation, coordination

and embedment of applied research for the benefit of all water

authorities being responsible for water management in

The Netherlands.

The products of the STOWA are scientific publications

in the fields of:

� urban water management

� wastewater treatment

� sludge disposal

� surface water

� groundwater

� aquatic ecology

� flood protection

The finances needed for the research programme and

the STOWA secretariat are provided bij the water authorities

in the foundation.

STOWA, Foundation for Applied Water Research

Arthur van Schendelstraat 816

P.O.Box 8090, 3503 RB Utrecht-NL

Phone +31 30-2321199

Fax +31 30-2321766

E-mail stowa@stowa.nl

Innovation by co-operationWater: the most essential requirement for all living organisms. Along with the growth of

the world’s population and prosperity the demand for clean, fresh water increases. Water

is scarce in various places across the globe. However, it can also lead to opportunities, such

as for companies specialised in the purification and supply of water.

The Netherlands and water are often referred to as synonymous. Even today we are still

known for our Delta Works and dikes. But in addition to reclaiming land from the sea,

water management is equally important in a densely populated country like the

Netherlands. This is an area of expertise in which the Netherlands historically has a great

deal of experience and do excel at international level.

Water management is a field of business stimulated by the Dutch government. I recently

applied to study the importance and potential of water treatment technology for our

economy. I also asked to analyse patent applications and to examine the research projects

supported by the government. You can find the results on internet. I concluded that the

Netherlands has considerable knowledge, skills and experience for water purification.

The Dutch authorities have high demands on the quality of both drinking water and of

wastewater to be discharged. Water quality requirements will only become stricter in the

future because of new types of pollution and greater focus on health and environment.

The implementation of the European Water Framework Directive will lead to urgent calls

for innovative water purification techniques.

Innovation is necessary to comply with the strict legislation on water quality. And close

co-operation is needed to realise such ambitious innovations. The membrane bioreactor is

a good example of Dutch innovation by co-operation. This concept is created by

combining biotechnology, membrane technology and process control. However, the

dedicated teamwork of potential consumers, suppliers of knowledge (such as universities,

institutes and consultants), manufacturers of components, contractors and authorities

also plays a key role. This collaboration resulted in further development of the MBR

technology which improves the quality of purified water, on a smaller footprint. You can

find more details on this innovation in this magazine.

The studies and above-mentioned example prove that the right conditions and

appropriate parties are available in the Netherlands to realise innovations for water

purification. An integrated approach and co-operation are needed to take full advantage

of these conditions, to produce new processes, services and products that can be applied

and marketed in many parts of the world.

I challenge all parties involved to work together on

innovations that contribute to competitiveness and

sustainability. The results of that co-operation will be

beneficial for people, planet and profit.

H2O, magazine for water supply and water

management, appears two-weekly.

Official publication of

Foundation to publish the magazine H2O

and her participants:

- Vereniging van Waterbedrijven in

Nederland

- Koninklijke Vereniging voor

Waterleidingbelangen in Nederland

- Nederlandse Vereniging voor Waterbeheer

- Kiwa N.V.

Publisher

Rinus Visser

Editorial staff

André van Bentem (DHV Water BV)

Peter Bielars (editor-in-chief)

Marjon Hoogesteger

Helle van der Roest (DHV Water BV)

Michiel van Zaane

Secretariat

Carolina Lemmens

Editorial adress and publishing

business

P.O. Box 122, 3100 AC Schiedam,

Netherlands

Phone +31 10 427 41 65

Fax +31 10 473 26 40

E-mail h2o@nijgh.nl

Office adress: ’s-Gravelandseweg 565,

3119 XT Schiedam, Netherlands

Editorial comittee

Harry Tolkamp (chairman/NVA)

Marjan Huisman (KVWN)

Frits Vos (VEWIN)

Gerda Sulmann (Kiwa)

Advertising exploitation

Roelien Voshol phone +31 10 427 41 54

Brigitte Laban phone +31 10 427 41 52

Media-order

Carola Sjoukes phone +31 10 427 41 41

Suzanne Klüver phone +31 10 427 41 40

Fax +31 10 473 20 00

Print and lay-out

Den Haag offset b.v., Rijswijk

Copyright

©2005 Nijgh Periodieken B.V.

The copyright on the content of this

magazine is explicit reserved. Taking over

articles only after a written permission of

the publisher.

Laurens Jan Brinkhorst

Minister for Economic Affairs

27 Industry is ready forMBR

Henk Brons

31 MBR applicationsoutside theNetherlands: driversand case studies

Berend Reitsma, Luc Kox and Cora Uijterlinde

36 Dutch MBRdevelopment:reminiscing the past five years

Darren Lawrence, Chris Ruiken,Dennis Piron, Ferdinand Kiestraand Ruud Schemen

14 Membrane bioreactorfor domesticwastewater: currentexpectations from theDutch WaterAuthorities

Jan Willem Mulder, Kees de Korte,Hans Ellenbroek, Philip Schyns andDennis Piron

19 MBR technology: futureresearch directions

Harry Brouwer, Hardy Temmink,Maxime Remy and Stefan Geilvoet

24 The role of consultancyfirms in the applicationof new wastewaterpurification technology

Bert Geraats, Peter de Jong, Jans Kruit and Helle van der Roest

3 Innovation by co-operation

Laurens Jan Brinkhorst

INTRODUCTION6 Co-operation and

innovation for asustainable and safeliving environment

Helle van der Roest, Henk van Brink and Monique de Vries

PLATFORM9 The MBR in broad

Dutch perspectiveCora Uijterlinde, Maarten Hofstra,Erik Kraaij and Jacques Leenen

Contents H2O MBR special III April 2005

Page 14

Page 31

Page 58

Page 62

Page 66

74 MBR research at WWTPLeeuwarden for the posttreatment of effluent

Sybren Gerbens, Hans de Vries,Sameh Sayed and Bonnie Bult

EDUCATION78 The growing

importance of educationfor MBR staff, operatorsand management

Edwin de Buijzer

Photo on the frontpage:Future quality effluent(photo: DHV Water)

58 Wastewater treatmentplant Hilversum:ambition and challenge

Kees de Korte

62 The Ootmarsum hybridMBR project

Dick de Vente, Bert Geraats, Hans Boenders and Harry Futselaar

66 Hybrid MBR - theperfect upgrade forHeenvliet

Jan Willem Mulder, Freek Krameren Eric van Sonsbeek

70 Comparison of the MBRwith continuous sandfiltration at theMaasbommel WWTP

Ferdinand Kiestra, Dennis Piron,Jacques Segers and Jans Kruit

41 Dutch MBR grows tomaturity

André van Bentem, Coert Petri, JanWillem Mulder, Herman Evenblij, Dick de Vente and Bert Geraats

48 POSTER VARSSEVELD

PROJECTS50 MBR Varsseveld, first

Dutch full scaleexperience

Coert Petri, Philip Schyns, André van Bentem and Frans Durieux

6

As an introduction to the officialopening of MBR Varsseveld, the third H2OMBR special has been created. After the firstand second MBR specials in 2001 and 2003this edition will effectively close the firstMBR development phase, which dealt withthe possibilities of the MBR technology forthe specific Dutch wastewater situation. Therealisation of the MBR Varsseveld representsthe beginning of the second phase of thenational MBR development, which willdemonstrate all the facets of scale up.

MBR Varsseveld can be seen as a productof a combined effort from the Dutchwastewater sector. Since the year 2000,fundamental scientific organisations,suppliers, consultants and water boards haveall been involved with the development ofthe technology, and a positive spin off is thatthis has spread beyond the Dutch borders. Inrecent years the Dutch contribution to theMBR development has received worldwide

recognition, and through this third editionof the H2O MBR special, the initiators, waterboard Rijn en IJssel (WRIJ), the Foundationfor Applied Water Research (STOWA) andDHV Water BV (DHV) hope to give anincreased impulse to the technology. We areproud to present you this H2O MBR Specialand wish you pleasant reading.

MBR technologyThe MBR technology is based on the

combination of the activated sludge processand membrane filtration in one treatmentstep, where the separation of the activatedsludge and effluent is achieved with thehelp of membranes. The MBR technologymaintains the good performance andflexibility of the conventional activatedsludge process, but also has two majoradvantages:• The required space is small as secondary

clarification is not necessary and the

sludge concentration in the aerationtank is two to three times that ofconventional systems;

• The effluent quality is significantlybetter as all the suspended and colloidalmaterial is removed. Furthermore extraremoval of heavy metals, microcontaminants, bacteria, viruses andcolour is achieved and sludgedisturbances no longer cause pooreffluent quality.

Especially in Holland where almost allwastewater treatment plants are of theactivated sludge type, where space is limitedand the quality of surface waters must bestrongly improved, the MBR technology hasgreat potential. Until now, the Dutch havefocused on possible improvements in theeffluent quality and the space saving wasconsidered less important. However, theMBR technology offers potentially compactsolutions where space saving can offeradvantages. The latter, particularly insituations where the treatment works islocated in or nearby large cities whereinnovative solutions with MBR can befeasible.

National developmentThe national MBR development in the

Netherlands began in 2000, and five yearslater can be considered to have pushed thetechnology to new levels. The nowworldwide famous pilot research at thetreatment works Beverwijk was the startingpoint of the first phase of the Dutch MBRdevelopment. During an extremely shortperiod of seven months the MBR technologyhad to be proven viable for the specificDutch municipal wastewater characteristicsand give reliable data for scale up. Waterboard Hollands Noorderkwartier and DHVin co-operation with four membranesuppliers and a number of foreign partiesinitiated this challenge, and within the first

H2O # 2005

M B R s p e c i a l I I I

Dutch MBR development enters next phase

Co-operation andinnovation for asustainable and safeliving environmentWithin the Dutch wastewater sector the national MBR development programme has grown into aclassic example of co-operation and innovation, where fundamental research organisations, suppliers,consultants and water boards have been involved. With the official opening of the demonstrationinstallation at Varsseveld on May 3, the second phase of the Dutch development programme shallcommence. Here, amongst others, an intensive research programme will be carried out to address scaleup issues. It has been seen that co-operation in the Dutch wastewater sector is of great importance andshould be copied for other developments within the water sector.

Monique de Vries.Henk van BrinkHelle van der Roest

7

year of the study the group was expanded toinclude other Water Authorities and fellunder the coordination of the STOWA forthe national MBR development programme.

From an economic perspective, thepotential impact for the environment,innovation power and co-operationthroughout the water sector, thesignificance of the MBR technology wasaddressed at administrative and politicallevel. At the end of 2000 it brought the watersector and policy bodies together. It wasunderstood that only through‘togetherness’, a technology could bedeveloped whereby the technical andfinancial risks were not for the individualbut rather for the group. Shortly thereafteran innovation fund was created to initiatethe correct start for the technology and notlimit it’s further development. Based onthese initiatives the Dutch water boards andthe Ministry of Traffic and Public Workswere able to agree on long term financialcommitments.

We are now a few years down the lineand the national MBR development is at fullsteam. After the successful completion ofthe research phase the commissioning of thedemonstration plant Varsseveld signals thestart of phase II. In the next 1.5 years WRIJ,STOWA, DHV, TNO, Delft University ofTechnology and Wetsus shall cooperate inintensive research in order to address theconsequences of scale up. The waterauthorities Hollandse Delta and Regge enDinkel will also realise MBR installations in2005 to address the applicability of varioushybrid configurations. All three full-scaleprojects are being supported out of theinnovation fund.

Already, preparation is being made for

the realisation of the full-scale system ofMBR Hilversum and will signal the start ofphase III of the National development. This phase has the goal of generating amature product for the MBR market with anemphasis on economics. Figure 1 is aschematic of the national MBR development.

All activities in the MBR developmentprogramme are coordinated via the STOWA,and through a steering committee, asupervisory commission and platformmeetings, all the participants are informedregarding knowledge dissemination andproject progress. The educational branchWateropleidingen has been carrying outcourses over the last few years to ensure thefuture of the technology.

The future for MBR?The future of the MBR technology

depends on many factors. The fact that costplays an important role is obvious, eventhough the membrane cost has significantlyreduced in recent years. The latter isreinforced by further developments incountries such as China, Korea and Taiwan,and together with the rapid technical andtechnological development the costdifferential between MBR and conventionaltechnologies is narrowing, and in somecases the MBR is a viable economicalternative. Future European guidelines, thedrive for innovation in combination witheconomic perspectives, possibilities to freeup expensive inner city ground, can all leadto arguments for further MBR development,despite the fact that on the short term thetechnology is more expensive than availabletraditional technologies.

Further technical and technologicaldevelopments will concentrate on solvingeveryday problems and expanding the MBR

advantages, and through a combined focuson the development issue the progress willbe accelerated. For this it will be clear thatcourage at managerial level is indispensable,the example of the Dutch MBR programmehas shown the positive consequences. It is ofgreat importance that the generated co-operation in the Dutch wastewater branchcan be learnt from, and applied to other(technological) developments, so that thewater authorities can develop a beautifulperspective for the future.

Book markerThis third edition of the H2O MBR

special is a celebration of the officialopening of the MBR Varsseveld and also togive an idea into the current activitiessurrounding the Dutch MBR developmentprogramme. This edition is in English andalso available as a complete pdf-format onthe world wide web(www.mbrvarsseveld.nl), which wasespecially set up by WRIJ, STOWA and DHVfor the Varsseveld project. The website willbe available until July 2006.

All participants in the developmentprogramme will be addressed in the thisedition of the H2O MBR special. The firstfive articles will describe the vision of thegovernment, water authorities,fundamental research institutions,consultants and industry. Thereafter, ashort overview of the development of othercountries outside of the Netherlands isenlightened, and an insight given to twoarticles covering the first two phases of thenational development programme. Lastly, insix articles the running MBR projects in theNetherlands will be openly described, wherethe last deals with the educational aspect.Once again the initiators, water board Rijnen IJssel, STOWA and DHV are proud topresent you this H2O MBR special and wishyou pleasant reading. ¶

Helle van der Roestprincipal consultant DHV Water BVP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 24 07e-mail: helle.vanderroest@dhv.nl

Henk van Brinkgeneral manager Water Board Rijn en IJsselP.O. Box 148, 7000 AC Doetinchemphone: +31 314 36 93 69e-mail: h.vanbrink@wrij.nl

Monique de Vrieschairman of the board STOWAP.O. Box 8090, 3503 RB Utrechtphone: +31 30 232 11 99e-mail: stowa@stowa.nl

H2O # 2005

Figure 1: Development of MBR technology in the Netherlands (2000-2010).

Kiwa Water Research

Kiwa Industrie en Water

Groningenhaven 7P.O. Box 10723430 BB NieuwegeinThe Netherlands

Tel. + 31 30 606 95 11Fax + 31 30 606 11 65E-mail alg@kiwa.nlInternet www.kiwa.nl

Knowledge only becomes valuable if you

put it into practice

That is why Kiwa Water Research also supports you with researchand counselling regarding water systems and water chain

Kiwa Water Research

Kiwa Water Research is the leading know-

ledge institute in the field of water and rela-

ted environmental and nature aspects.

Worldwide it provides made-to-measure

knowledge to water works, government

agencies and other players involved in the

water industry.

For the water works in the Netherlands and

their (international) partners Kiwa Water

Research carries out the joint research pro-

gram. Also we counsel water works,

Ministerial Departments, local authorities

and District Water Boards in the field of

groundwater, surface water, nature, water

production, (waste) water purification and

water infrastructure.

Kiwa Industry and Water

Industrial partners may obtain access in an

efficient way to water knowledge geared to

their practice via the specialized branch

which is Kiwa Industry and Water. Jointly

with enterprises, industry organisations and

water works Kiwa Industry and Water deve-

lops and implements innovative water tech-

nology. It supports individual parties in the

solving of day-to-day water problems,

implementation projects and it helps out as

trouble shooter. Apart from this Kiwa

Industry and Water offers extensive networ-

king facilities and a platform to exchange

knowledge and experience.

Technological innovations often arisefrom a desire to optimise and/or increaseefficiency. In the industrial sector in particular,innovations are often driven by technologypush. Besides these incentives, innovations indomestic wastewater treatment, which in theNetherlands is the exclusive domain of thewater boards, are also determined by policydevelopment (national and international).

Policy developmentsAt the national level the recent Fourth

Policy Document on Water Managementincludes an explanation of the MaximumTolerable Risk (MTR) standards for surfacewater. Purification technology research aimsprimarily at removing phosphate andnitrogen. Until now relatively little attentionhas been given to the other substancesmentioned in the MTR standards. We mustrealise though that surface water standards arenot effluent standards. Nevertheless, surfacewater standards are used in practice asreference values against which theperformance of (new) purification technologiesis measured.

Phosphate and nitrogen were highlightedwith the introduction of the discharge policy

for urban wastewater and the Europeandirective on urban wastewater. The limits ofcurrent purification systems are now beingtested by shifting the focus to lower effluentconcentrations (instead of 10 mg/l nitrogen the

target is 2.2 mg/l and the phosphate target forresearch projects is 0.15 instead of the usual 1or 2 mg/l). Besides nutrients, more and moreattention is being given to priority substances(heavy metals, organic micro-pollution andhormone-disruptive substances, amongstothers). These substances will also get theattention they deserve during implementationof the EU Water Framework Directive and theywill throw new light on purificationtechnology and techniques. The role ofpurification in this framework still has to beweighed against other measures (tackling thesource of the problem, for example).

It may be concluded that the issue of thereuse of effluent is getting sufficient attentionwithin the water boards recently. There arevarious initiatives/studies which are examiningthe (partial) closing of the aquatic cycle.

The comparison of purificationmanagement operations is an importantsource of inspiration for the water boards’innovation-consciousness. This has partlycontributed to participation in research intoinnovative technologies gaining greatersupport.

External succes factorsBesides the general policy developments

mentioned and the need for commercialoptimalisation and a desire for purificationtechniques which lead to better effluentquality in general, there are also otherimportant factors for successfully gettinginnovative technology operational. Firstly, it isvitally important that sufficient confidence isgenerated in the considered technology.Experience with pilot projects helps toengender this confidence. Design

H2O # 2005

MBR special III

9

The MBR in broad Dutchperspective

Cora Uijterlinde, STOWA

Maarten Hofstra, Rijkswaterstaat / RIZA

Erik Kraaij, Unie van Waterschappen

Jacques Leenen, STOWA

The path that leads from theoretical innovation to practical implementation is often a troubled one.This also applies to MBR technology. However, the Dutch wastewater sector has succeeded in furtherdeveloping this (in the Dutch wastewater field) new technology into a system that can be applied tothe Dutch context in a relatively short timeframe. The article below gives a brief overview of waterboard developments which have contributed to MBR technology. It also reviews MBR developmentsin the Netherlands, starting with the MBR aspirations of (the former) Water Board of UitwaterendeSluizen in Hollands Noorderkwartier, and leading right up to the present. Furthermore, an overviewis given of current and recently completed STOWA research projects, and the purpose anddevelopment of the water boards’ innovation fund as an important success factor is also brieflyexplained. Finally, the article also discusses a second important success factor for the furtherdevelopment of new technology, i.e. broad co-operation. This means sharing expertise amongst thewater boards, but also includes collaboration with commercial parties, consultancy companies andresearch institutions, both in the Netherlands and abroad.

Dutch delegation visits MBR plants in the United Kingdom (October 2003).

fundamentals can be substantiated throughpilot studies and specific questions are moreeasily answered with test installations than inpractice (for example, initiating an MBR can beeffectively imitated with a test installation).

Positive business experience leads to thenecessary confidence in the technology. Visitsby water quality managers to MBR plants inthe UK, for example, have definitelycontributed to confidence in the technology1.This despite the fact that the application of thetechnology is different in the UK than isenvisaged for the Netherlands.

A positive incentive for embracing the newtechnology also includes the compactness of anMBR. This plays an unmistakably importantrole in MBR development. MBR will scorehighly against conventional sewage treatmentplants where purification space is at apremium. Possibilities for multifunctionalground use are also created.

Otherwise cost remains the mostappealing factor, though this is usually lessfavourable during development whencompared to conventional technology. Toensure broader application of this innovativetechnology in the future, these technologieswill also have to compete economically withconventional technology. As the marketexpands, free-market processes will arise.

The above-mentioned developments andsuccess factors can reinforce each other.Confidence in the technology can lead to moreapplications to which the market can react,both economically but certainly alsotechnically. It may be concluded that this wasindeed the case for MBR technologydevelopment. Running through the variousdevelopment phases, much ‘profit’ has beenachieved both economically and in terms ofproduct improvement. Furthermore, a three-phase structured approach has played animportant part. The knowledge acquired invarious studies is further expanded upon withsimultaneous scale enlargement to theeventual large-scale application. The greatadvantage of this structured and phasedapproach is that damage risk stays limited andincreasing understanding can easily beintegrated in subsequent development phases.

MBR development in theNetherlands in terms of scale

The development of MBR technology forthe Dutch context has therefore been tested invarious research set-ups. This raised importantresearch questions such as those regardingeffluent quality and operational managementaspects (performance under different

conditions, cleaning procedures in relation tothe use of chemicals, energy consumption,etc.). The first Dutch MBR for domesticwastewater on a working model scale is nowoperational in Varsseveld. This installationfulfils a demonstrative function for Dutchwater quality managers. It is the first projectrealised using the innovation fund set up forthis very purpose by the water boards.

A couple of hybrid MBR plants inHeenvliet and Ootmarsum are also underconstruction now. These projects are alsosupported by the previously mentionedinnovation fund. They are both relativelysmall plants. Both projects also have ademonstrative function for the hybridapplication of MBR systems.

The first large scale working model MBRwill probably be the Hilversum STP. Seriousplans for the construction of an MBR plant arebeing prepared. The current STP is to bemoved, and there is only a limited surface areaavailable at the new location. Integration inthe surroundings and effluent quality play arole in the decision to switch to MBR. Researchhas already been carried out for some time atthis location using a pilot plant.

A review of MBR development inthe Netherlands

A large study was carried out in 2000/2001at the Beverwijk STP into the application ofMBR for domestic wastewater in the Dutchcontext. For this purpose, MBR pilot plantsfrom four different suppliers were testedunder various conditions2). Within anextremely short timescale of seven months, ithad to be shown whether MBR technology inthe Dutch context was both applicable andexpandable. These challenges were at firsttackled by the then Water Board ofUitwaterende Sluizen in HollandsNoorderkwartier and DHV in collaborationwith four membrane suppliers (Zenon,Kubota, Mitsubishi and Norit). In that sameyear, STOWA took over coordination ofnational MBR development the group of watermanagers was expanded with the inclusion ofthe then Water Treatment Board of HollandseEilanden en Waarden (now the HollandseDelta Water Board), the Veluwe Water Board,the Rijn en IJssel Water Board, the Regge enDinkel Water Board and the Water BoardAmstel, Gooi en Vecht/DWR. This wasextensively covered in the last two H2O-MBR-specials3),4).

The promising research results of thesepilot studies at the Beverwijk STP led to thedecision to build a working model of an MBRdemonstration plant at Rijn en IJssel WaterBoard’s Varsseveld STP. In order to make thisdemonstration plant a successful workingmodel, and to keep damage risks to a

minimum, research projects were initiated atVarsseveld and elsewhere at various pilotplants to further expand (still somewhatlacking) expertise.

Simultaneous research projectsIn order to get a broader understanding of

the possibilities and limits of MBRs, as well asto ensure the success of the Varssevelddemonstration plant, various parallel studieswere conducted at different locations besidesBeverwijk STP, as has already been mentioned.Therefore, research was started in 2002 by theSTOWA in co-operation with the RivierenlandWater Board at the Maasbommel STP to studythe applicability of a membrane bioreactorcompared to a conventional active sludgesystem with linked sand filtration5). Duringthe two-year-long study, both systems provedthat it is possible to remove most phosphateand nitrogen. Besides removing nitrogen andphosphate, the study increased insight intothe removal of various other components. Thestudy has offered a great deal of usefulinformation about the possibilities andlimitations of both methods. A pilot study hasalso been started in Hilversum to prepare forthe MBR plant to be constructed there.

In the province of Friesland, STOWA iscarrying out research in co-operation withFryslân Water Board into the working of alinked MBR. In this pilot study at theLeeuwarden STP, an MBR is linked to aconventional active sludge plant. Central tothis is the removal of special substances (non-biodegradable organic micro-pollution). In thissense, it is an innovative application of MBR.

Finally, it should be mentioned that in2002 STOWA carried out market research intoMBR technology for use with domesticwastewater6). Through interviews and surveys23 of the 26 regional water managers present inthe Netherlands were approached. For thelonger term (2020) 69 projects with anaverage/high probability were assessed by thewater managers. This study revealed that theNetherlands is primarily a market for small-scale custom builds and does not supportlarge-scale construction.

Hybrid applicationsBesides the Varsseveld demonstration

plant two demonstrations have now beenstarted with MBR hybrid applications. Thisconcerns a joint project of the Regge en DinkelWater Board and the Hollandse Delta WaterBoard at Ootmarsum STP and Heenvliet STPrespectively. Hybrid systems combine theadvantages of MBR (high effluent quality,space savings) with the advantages ofconventional active sludge plants which can

H2O # 200510

MBR special III

process large volumes of wet weatherdischarges. During dry periods, all wastewater(or at least most of it) is processed by an MBR.During wet periods, rainwater is alsodischarged via the ‘conventional path’. In thisway, it is expected that, compared to an MBRplant through which wastewater iscontinuously flowing, energy would be usedmore efficiently with slightly lower totalremoval efficiencies. Especially at thoselocations in the Netherlands wherewastewater and rainwater are often collectedtogether, much will be expected of this hybridapplication. The application will be installedin particular in plants needing attention dueto capacity problems. However, it will alsooffer a solution which could be used toimprove existing conventional purificationwhere effluent is discharged into fragilesurface waters. The Heenvliet STP andOotmarsum STP joint project runs to the endof 2006. After Varsseveld STP, the MBR hybridapplication demonstration project is thesecond one to be supported by the innovationfund set up by the water boards.

Innovation fundSince they are still in the development

stage, innovative techniques like MBR cannotyet compete with conventional technologyand, by extension, do not have sufficientmarket profile. This means that scaling up of

the technology to working models is notwithout risk and there is consequently acertain damage risk. Finally, the water boards’infrastructure must be used to scale up thetechnology. To spread the damage risk andextra costs, the water boards set up the so-called innovation fund in 2001. The fund worksas follows: the plant to be developed isbudgeted and this is then debited from theestimated costs of a comparable fictionalconventional plant. After payment of asubstantial extra contribution from the ‘hostwater board’, the difference is paid by the fund.In the unlikely event that the project is a totalfailure, recovery costs are also borne jointly bythe fund. The fund has been assigned to theSTOWA and is financed on the basis of thenumber of pollution units in a water board’smanaged area. The annual contribution hasbeen based on the costs of the Varsseveldproject. The commitment of the water boardsapplied for four years, and was at firstexclusively designated to upscaling the MBR. Itwas agreed with the water boards that, afterfour years, the fund’s function would beevaluated and, on the basis of this, decisionsfor the future would be made. When theinnovation fund was set up, the Ministry ofTransport and Public Works contributed about1.4 million Euros. The water quality managerscollected more than 4.4 million Euros in theperiod 2002-2005.

Within the Varsseveld project subsidieswere received as part of LIFE (EU subsidy) andEINP (Ministry of Economic Affairs subsidy forenergy investment deductions for non-profitorganisations). After deducting the Varsseveldcontribution, the balance has recently beenassigned to the hybrid MBR projects: Heenvlietand Ootmarsum.

Future of innovation fundAs has previously already been mentioned,

the water boards committed themselves toconducting both an evaluation of the fund’sfunction and further decision-makingregarding the fund’s continuation. On 15 April2004 there was a meeting of the participants ofthe STOWA, that is all water boards, whereamongst other things the innovation fund’saim, scope and finance were considered.During this meeting it was also suggested tocontinue contributing to the innovation fundafter 2005. A desire was also stated to widen thefund’s objective: not only for MBR applicationsor projects related to wastewater systems. Allthe various tasks of the water board (waterchain, water systems and water barriers)should be considered. Projects which are notspecifically technical (‘alpha’ like applications),should also be considered for a donation fromthe fund. In the summer of 2004 STOWA’smanagement decided in the light of thissuggestion that from the 2005 fiscal year theinnovation fund would be integrated with theSTOWA’s research program so that a matureR&D policy can be developed for the waterboards. Innovation forms a separate themethroughout all the tasks in the multi-yearplanning.

Co-operationThe course followed by the research into

development of the MBR in the Dutch contextis an outstanding example of successful co-operation. This co-operation has taken variousforms during the different phases of the study.Besides the STOWA this included involvementby various water boards, nearly all the largeDutch advisory agencies, all suppliers of MBRplants and many technical universities(national TU Delft, Wageningen UR and TUTwente and international TU Aachen) andtechnological top-institutes like TNO andWetsus.

All those (who have been) involved withthe study are convinced that this broad co-operation partly ensured the results achieved.

Co-operation and information exchangeare central to STOWA projects. Around theMBR theme, various supervisory committeesand a steering committee were created. Twice ayear, a symposium is organised for all thoseinvolved in STOWA MBR-studies. Informationexchange is central to this. When bringing a

H2O # 2005 11

MBR special III

The Maasbommel research report is officially given out at the third Dutch MBR conference (Echteld, November 2004).

new development onto the market, it isimportant to learn from each other’sexperiences in order to prevent unnecessarilynegative signals thwarting the developments.

Water boards work mutually with MBRprojects; an example of this is the previouslymentioned co-operation between theHollandse Delta Water Board and the Regge enDinkel Water Board in the field of hybrid MBRplants. Lastly, the co-operation with theStichting Wateropleidingen is also an exampleof this. Stichting Wateropleidingen has alreadybeen holding an MBR course for a couple ofyears, ensuring the required education forfuture users.

International joint venturesCo-operation also occurs at an

international level, besides the above namedspecific co-operation in the study into MBR inthe Dutch context. The STOWA participates,together with KIWA in the Global WaterResearch Coalition (GWRC), a collaborationbetween twelve global informationinstitutions involved in research in the field ofthe water chain (UK, USA, South Africa,Australia, France, Germany, Switzerland andthe Netherlands, amongst others).

MBR has been placed on the collectiveresearch agenda. Currently, a ‘state of thescience’ report is being prepared, in which areassessed, besides current expertise, gaps andneeds in expertise. This report is evaluated inan international workshop, after which theGWRC will review which of the joint projectsshould be tackled. The said workshop is linkedto an international symposium about MBRthat will be held on site on the occasion of theopening of the Varsseveld STP. In this waystrengths are combined, preventing the wheelfrom being invented twice. More and more co-operation is occurring in Europe. Variousresearch projects discover how to collaborate inan European context. The Varsseveld MBRdemonstration project (Life subsidy) is anexample of this.

ConclusionsStagnation means decline. In order to

progress, it is important to reposition one’shorizons by focusing on the long(er) term.Current MBR applications are maybe (still)expensive. But in the long term, MBR can beeconomical. Flexibility, effluent quality, spacesaving and potential for multiple ground useand development of membrane prices all play apart in this. It is expected that a temporaryhigh investment in MBR is justified.

New technology must be handled withcare. Research is necessary to ascertain whatthe possibilities are, and especially what is notpossible with the technology. It is importantthat the appropriate expectations are assumed

for a new technology. Disappointingexperiences do not help technologicaldevelopment. Research and practicalexperience contribute to realistic expectations.

Study results must be seen in their correctperspective as local conditions can have astrong influence on whether a technology isattainable.

A precondition for new technology is thatthe water sector should generate greatercollaboration and expertise. Confidence in anew technology arises as joint experience isacquired. The impulse to win over confidencein the MBR comes from pilot studies anddemonstration plants.

Furthermore, co-operation in innovationcan form an important stimulus to strivejointly for greater progress. The sum 1 + 1 = 3(more than the sum of its parts) applies here.

The merits of innovative applicationsmust be assessed in a wise manner. Here liesthe challenge for the water world: give MBRthe chance to prove itself for applications inthe Netherlands. ¶

Jacques Leenen managing director STOWAP.O. Box 8090, 3503 RB Utrechtphone: +31 30 232 11 99e-mail: leenen@stowa.nl

Maarten. Hofstramanaging director of the department of waterquality and information Rijkswaterstaat, atpresent RIZAP.O. Box 17, 2800 AA Lelystadphone: +31 320 29 84 69e-mail: m.hofstra@riza.mws.minvenw.nl

Erik Kraaijhead of department of watermanagement Unievan WaterschappenP.O. Box 93218, 2509 AE Den Haagphone: +31 70 351 97 51e-mail: ekraaij@uvw.nl

Cora Uijterlinde research manager STOWAP.O. Box 8090, 3503 RB Utrechtphone: +31 30 232 11 99e-mail: uijterlinde@stowa.nl

literature

1) STOWA 2004-W-02. Van Stonehenge tot MBR: waarin een

groot land ‘klein’ kan zijn. Verslag van STOWA-

studiereis.

2) STOWA 2002-11A. MBR for municipal wastewater

treatment; Pilotplant research Beverwijk WWTP.

3) H2O-MBR-special (October 2001).

4) H2O-MBR-special II (April 2003).

5) STOWA 2004-28. Vergelijkend onderzoek MBR en

zandfiltratie RWZI Maasbommel.

6) STOWA 2002-13. MBR-marktstudie.

H2O # 200512

MBR special III

SamenvattingDe weg die bewandeld moet worden om innovaties in de praktijk tot uitvoering te laten komenis doorgaans een moeilijke. Zo ook voor de MBR-technologie. De Nederlandse afvalwatersectoris er echter in geslaagd om in een relatief kort tijdsbestek deze, voor de Nederlandseafvalwaterwereld, nieuwe technologie verder te ontwikkelen tot een systeem dat onderNederlandse omstandigheden kan worden toegepast. In onderstaand artikel wordt hiervan eenbeeld geschetst door in vogelvlucht aan te geven welke ontwikkelingen in de omgeving vanwaterschappen hebben geleid tot een bijdrage aan de ontwikkeling van de MBR-technologie.Er wordt een terugblik gegeven op de MBR-ontwikkeling in Nederland, beginnend bij deMBR-aspiraties van (destijds) het Hoogheemraadschap van Uitwaterende Sluizen in HollandsNoorderkwartier tot de dag van vandaag. Daarbij wordt een overzicht gegeven van de lopendeen de recent afgeronde STOWA-onderzoeksprojecten en wordt kort de opzet en ontwikkelingvan het innovatiefonds van de waterschappen als belangrijke succesfactor toelicht. Tenslottewordt stilgestaan bij een tweede belangrijke succesfactor voor de verdere ontwikkeling van eennieuwe technologie, zijnde de brede samenwerking, binnen de waterschappen onderling omkennis te bundelen, maar ook samenwerking met marktpartijen, adviesbureaus enonderzoeksinstellingen, zowel in Nederland als daarbuiten.

WWW.AQUATECHTRADE.COM

THE WORLD’S LEADING TRADE EVENTS FOR PROCESS, DRINKING AND WASTE WATER

Organisation:Amsterdam RAIP.O. Box 77777, 1070 MS Amsterdam, the NetherlandsTelephone +31 (0)20 549 12 12Fax +31 (0)20 549 18 89E-mail aquatech@rai.nl

WQA AQUATECH USA 200629 - 31 MARCH

AQUATECH ASIA 20056 - 8 OCTOBER

AQUATECH AMSTERDAM 200626 - 29 SEPTEMBER

H2O # 200514

MBR special III

Pilot research has been carried out in theNetherlands since the beginning of 2000 on theuse of membrane bioreactor (MBR) technologyfor domestic wastewater treatment4).

Based on experiences abroad, full-scaleapplications were expected to be possible in theshort term. Several wastewater treatmentplants were scheduled for an MBR upgrade fordifferent reasons3),6),8). With respect to theWWTPs at Beverwijk (452,000 p.e.), Hilversum(91,000 p.e., 1,500 m3/h) and Dordrecht (265,000p.e.), lack of space to accommodate anextension played an important role. For theHilversum and Varsseveld (23,150 p.e.) WWTPsand the smaller Maasbommel WWTP (7,400p.e.), another reason to consider MBRtechnology as an option was the requiredeffluent quality.

One major expectation with respect to theMBR was a superior effluent quality. The aimwas to achieve maximum tolerable risk (MTR)quality without major problems, and theexpectation was that many micro pollutantswould be removed more efficiently whencompared to conventional techniques.Examples of micro pollutants are heavy metals,pesticides and endocrine-disruptingcompounds. These expectations were not basedon research data, however, and the ongoingresearch programme was expected to confirmthem. Neither process engineers nor decision-makers had any serious doubts about thepotential of the MBR.

Problems in developmentAlthough the aim was to develop a large-

scale practical application, some problems hadto be solved five years ago. The MBR was muchmore expensive than conventional techniques,especially when treating large hydraulic peakflows. Combined sewerage systems dominatein the Netherlands, resulting in large-volumeflows during storm weather. Anotherdisadvantage was the higher energyrequirement caused by intensive membraneaeration and by lower aeration efficiency in theactivated sludge tanks.

Some uncertainties remained, for examplevarious operational aspects and the lifetime ofmembranes. There were several membranesuppliers, but it was uncertain which supplierand which system were favourable.

Membrane bioreactor fordomestic wastewater: currentexpectations from the DutchWater Authorities

Jan Willem Mulder, Water board Hollandse Delta

Kees de Korte, Dienst Waterbeheer en Riolering

Hans Ellenbroek, Water board Regge en Dinkel

Philip Schyns, Water board Rijn en IJssel

Dennis Piron, Water board Rivierenland

In 2000, a large-scale pilot study was started into the use of membrane bioreactors (MBR) for thetreatment of municipal wastewater in the Netherlands. Under Dutch conditions, with wastewatertreatment plants having to handle large volumes of rainwater, a very compact plant should be able tobring about a considerable improvement in effluent quality. Lower membrane costs were also predicted.Moreover, significant cuts in energy consumption appeared feasible. It was found that the flux can beincreased, so that less membrane surface area is needed. The membrane cleaning procedure can also beimproved. Furthermore, it was found that a significant improvement in the quality of the effluent canbe achieved, although some expectations, especially with regard to micro pollutants, could not befulfilled. It proved possible to reduce energy consumption, but not to the extent required, and this,together with the higher costs of an MBR, is still a major bottleneck with regard to future (large-scale)applications. In certain situations, however, an MBR, possibly in hybrid form, may be the best solution.

Artist’s impression of the Hilversum WWTP’s office building; the plant itself will be constructed in the hill (contaminated soil).

H2O # 2005 15

MBR special III

Foreign experiencesAs many MBR facilities had been built

abroad prior to 2000, the suggestion was to copysuch concepts and use them in the Netherlands.It became clear that further research wasrequired for several reasons before MBRtechnology could be applied in the Netherlands.

The first reason was the scale ofapplication. Many of the previous plants werebuilt in Japan and have a very small capacity.Factors such as costs and energy requirementare less decisive at smaller scales. Copying suchconcepts for large-scale applications wouldresult in extremely expensive MBR facilities.

The second reason was the requiredeffluent quality. Several MBR plants have beenbuilt in the UK, for example, but none of themhas reached MTR quality. In some cases, theplants are not even required to remove nitrate.As a result, such MBR plants are of a muchsimpler construction than those built to meetMTR quality.

Several MBR plants in Germany were builtto produce an improved effluent quality. Thesame problems arose at those plants as duringthe pilot research programme in theNetherlands, with the conclusion being thatsome of them could have been built moreefficiently with the knowledge we have nowacquired.

Results of five years of researchResearch involving pilot plants has been

carried out at Beverwijk, Hilversum andMaasbommel for the past five years5),7). A largenumber of suppliers have demonstrated theirMBR systems and it was possible to achievemany optimisations. The research has broughtMBR technology for MTR quality to the pointwhere large-scale application is now possible.Note that five years ago, it was alreadyexpected that the technology would advancealmost to this point.

Membrane performance improvedimpressively following the research, resultingin higher permissible fluxes and, as a result, inonly a limited membrane surface beingrequired. This has had a favourable impact oninvestment costs, operational costs and theenergy requirement.

The energy requirement itself has also beenoptimised. Discontinuous aeration in themembrane tanks limits the energy requirement.Improving the biology may have a favourableeffect on the alpha factor, and therefore on theaeration efficiency in the activated sludge tanks.Sludge concentrations of 20 g mlss/l turned outto be unfavourable, and design concentrationsare currently limited to approximately 10 gmlss/l. Even with this restriction, the MBR canstill be considered very compact.

Improved pre-treatment is essential forsafe operation. Screening at less than 1 mm

will considerably reduce the risk of membranefailure. In addition, knowledge of chemicalcleaning contributes to the safer operation ofMBR plants.

The effluent quality was less favourablethan expected, however. It may still be possibleto achieve MTR quality for nitrogen (2.2 mg/l)and phosphorus (0.15 mg/l), although severalpilot plants were only able to reach thesevalues after addition of an external carbonsource and an iron salt.

With respect to micro pollutants, theresults were disappointing. At Maasbommel,the effluent of the pilot MBR was compared tothe effluent of the conventional WWTP1),7), andno significant difference was found in theremoval of micro pollutants. Most of thesecomponents may well be dissolved or adsorbedto natural organic matter and thus able tobypass the membranes. Although the MBRand the conventional effluent did not differsignificantly with respect to the measuredconcentration of endocrine disruptingcomponents, the endocrine potential was 70%lower.

The MBR was an effective disinfectionoption. Both bacteria and viruses were foundto have been reduced to very low effluentconcentrations.

Present statusAlthough much progress has been made,

MBR plants are still more expensive thanconventional activated sludge systems builtaccording to the latest designs. More effort willbe required to achieve a further cost reduction.

Costs can be reduced by improving membraneperformance, and the unit costs of themembrane surface may also decrease in thefuture due to the larger-scale application ofMBR. More full-scale plants will have to bebuilt to achieve both factors.

The energy requirement for MBR stillexceeds the requirement for conventionalactivated sludge systems. The requirement canbe further optimised to some extent bylimiting the membrane surface, but also byoptimising the performance of the biology,improving the alpha factor and consequentlythe aeration efficiency. Full-scale applicationscan contribute to both developments. Fromthe point of view of sustainability, it should benoted that a further reduction of the energyrequirement is considered essential.

Several pilot and full-scale experiencesdemonstrate that the operation of an MBR ismuch more critical than the operation of aconventional plant. Well-trained processoperators are required, as well as asophisticated process control and automationsystem. Basically, this is an issue that can besolved, but it will require more attention. Itwill be easier to handle this aspect when morefull-scale MBR plants are in operation.

The effluent quality falls short of theexpectations of five years ago. There is hardlyany doubt as to the potential of the MBR withrespect to nitrogen and phosphorus removal,but it is unlikely to remove micro pollutantseffectively enough. On the other hand, MBReffluent is free of suspended solids and issuitable as a starting point for more advanced

MBR pilot plant at the Maasbommel WWTP.

H2O # 200516

MBR special III

techniques when further treatment isindicated. Clarifier overflow is less suitable atthis point. The MBR was also shown to be anefficient technology for disinfection.

What is the future?Because MBR still has two major

disadvantages (costs and energy requirement),the question is whether the MBR technology hasa future in the Netherlands. In spite of thesedisadvantages, there are still several good reasonsto embark on the full-scale application of MBR.The main reasons will be discussed below.

Although effluent quality does not live upto the original expectations, especially inrespect of micro pollutants, it is still betterthan the quality achieved by conventionaltreatment. Suspended solids are absent in theeffluent; as a result, nitrogen, phosphorus andheavy metals, part of the suspended solids, arereduced to some extent. In theory, theconcentration of micro pollutants can also bereduced, as these components will be partlyadsorbed to suspended solids. Further researchon this possibility is necessary.

The MBR blocks all bacteria and someviruses. Disinfection is not very common inthe Netherlands, but is favourable from ahygienic point of view.

A second reason to select MBR technologyis the compact set-up of MBR plants. The spaceavailable to upgrade WWTPs is sometimeslimited. This problem is expected to grow inthe future, as the population figures rise andurban areas expand quickly.

A further advantage besides the space-saving aspect is that MBR plants can becovered more easily than conventional plants,thereby limiting the environmental impact interms of noise and odour. Occasionally, it maybe easier to introduce a short-term extensionin an MBR plant than a conventional one.

A third reason in favour of MBR is thepossibility of reusing WWTP effluent. MBReffluent in itself may not yet be suitable fordirect reuse, but in combination with othertechniques MBR can play an important role inthe production of water for several differentpurposes, for example agricultural use andindustrial water of different qualities. Its directreuse in drinking water production is not verylikely in the Netherlands, but it may be apossibility in more drought-prone regions ofthe world. An example of the direct reuse ofwastewater to produce drinking water can befound in Namibia and Singapore.

Further developmentIn view of the potential of the MBR, more

research will be necessary to solve the cost andenergy requirement problems. These twoaspects have been optimised in the pilotresearch carried out during the past five years.Further optimisation can only be achieved bybuilding full-scale applications and byoptimising these MBR plants on a practicallevel. For this reason, the Dutch water boardsco-operated in setting up a full-scale MBRfacility at the Varsseveld WWTP. This willresult in an improved design for the nextgeneration. In addition, the cost price for

membrane modules will fall when membranesare produced on a larger scale.

Further optimisation will not cease withthe development of MBR systems. Wastewaterproperties can also have an important effect onthe cost effectiveness of the MBR, as well as theenergy requirement.

An important factor influencingwastewater properties is the sewerage system.In the Netherlands, combined sewersdominate, resulting in large RWF to DWFratios. Disconnecting rainwater drainage fromwastewater sewers will result in much smallerhydraulic capacities, which is favourable forthe MBR. There has been a trend in theNetherlands to disconnect rainwater drainagefrom the sewers, but separate collection hashad only a minor impact as yet. It will in anyevent take several decades before there is anysubstantial effect on the RWF/DWF ratio.

A second factor involved in wastewatercollection is the inflow of infiltration waterinto sewers, for example groundwater andsurface water. A study by STOWA pointed outthat the dry weather flow increases by 60% onaverage owing to other water sources2),9). Evenwhen rainwater is disconnected from sewers,the hydraulic capacity can be further reducedif the sewers are in good condition, preventingthe inflow of groundwater and surface water.This would naturally be favourable for MBRapplications.

Another development is the hybrid MBR.This concept is suitable when a conventionalWWTP is upgraded with an MBR. Two hybridplants are currently under construction atWWTPs in Heenvliet and Ootmarsum. In thisconcept, the MBR is not supposed to receivethe entire hydraulic load, nor, consequently,the entire organic load. Wastewater isdistributed between the MBR and theconventional plant. In dry weather, the MBRreceives relatively more wastewater andmembrane capacity is therefore usedefficiently. In storms, the MBR has limitedhydraulic capacity and receives relatively lesswastewater. The overall effluent quality resultsfrom mixing the MBR effluent and effluentfrom the conventional plant, so removalefficiency will be a compromise between costsand result. By using a wastewater storage tankin dry weather conditions, or even in storms,the compromise can be optimised further.

The end result must be borne in mind inany further development. The present designswill become obsolete and suboptimal after afew years, but they can play a decisive role inthe MBR development. It would be well to bearin mind the status of MBR technology that wewill have achieved in the future, for exampleafter ten years of practical use.

MBR Varsseveld with old aeration tanks and clarifier in the background (photo: Aerofoto Brouwer - Brummen).

ConclusionsThe following conclusions can be drawn,

based on five years of research in theNetherlands:• Owing to more stringent effluent

requirements, more advanced designs andoperational aspects are needed for MBRplants in the Netherlands compared toplants in most other countries;

• Pilot research appeared to be essential inorder to make MBR technology suitable forthe Dutch situation. Experiences abroadare inadequate by themselves;

• After five years of pilot research, the time isright to construct the first full-scale MBRplants. The first plant recently came onstream at the Varsseveld WWTP. The timeschedule seems to make sense;

• The present state of MBR technology meansthat it is still not suitable for widespread,large-scale application. Further

optimisation must be achieved, especiallywith respect to costs and energyrequirement. This only becomes possible bybuilding full-scale plants and learning fromthem. The experience gained operating theVarsseveld WWTP and, next year, theHeenvliet and Ootmarsum WWTPs, and theresearch data produced at these WWTPs willcontribute to such optimisation;

• The future of MBR technology has to beborne in mind, both with respect towastewater collection and the status ofMBR technology. Existing and upcomingMBR plants will not be fully representativeof the future. Close co-operation withinthe Dutch water sector and financialincentives within a span of about eightyears after the start of the Dutch MBRresearch should produce enough expertiseand experience to achieve a competitiveand reliable MBR system. ¶

Jan Willem Muldersenior policy advisor Water board HollandseDeltaP.O. Box 469, 3300 AL Dordrechtphone: +31 78 633 16 68e-mail: j.mulder@wshd.nl

Kees de Korteproject manager Dienst Waterbeheer enRioleringP.O. Box 94370, 1090 GJ Amsterdamphone: +31 20 460 27 85e-mail: kees.de.korte@dwr.nl

Hans Ellenbroekproject manager Water board Regge en DinkelP.O. Box 5006, 7600 GA Almelophone: +31 546 83 25 25e-mail: H.J.Ellenbroek@wrd.nl

Philip Schynsproject manager Water board Rijn en IJsselP.O. Box 148, 7000 AC Doetinchemphone: +31 314 36 96 13e-mail: p.schyns@wrij.nl

Dennis Pironprocess engineer Water board Rivierenland P.O. Box 599, 4000 AN Tiel phone: +31 344 64 95 04 e-mail: D.Piron@wsrl.nl

literature

1) Kiestra F., D. Piron, J. Segers and J. Kruit. Comparison

membrane bio-reactor and sand filtration (in Dutch). H2O

37 (2004), no. 23, pp. 29-32.

2) Kerk A. van de, B. Palsma and A. Beenen. Infiltration

water: the ‘DWAAS’ method uses available data only (in

Dutch). H2O 37 (2004), no. 2, pp. 7-9.

3) Korte K. de, J. Mulder, A. Schellen, R. Schemen and P.

Schijns. MBR will be the wastewater treatment plant of

the future, vision from the water authorities. H2O MBR

special, October 2001, pp. 22-25.

4) Lawrence D, A. van Bentem, P. van der Herberg and P.

Roeleveld. MBR pilot research at the Beverwijk WWTP, a

step forward to full-scale implementation. H2O MBR

special, October 2001, pp. 11-15.

5) MBR for municipal wastewater treatment. Pilot plant

research Beverwijk. H2O report 2002-11A.

6) MBR marketing research (in Dutch). Dutch Foundation

for Applied Research Water Management (STOWA).

STOWA report 2002-13.

7) Research for the comparison of MBR and sand filtration at

the Maasbommel WWTP (in Dutch), STOWA report 2004-

28.

8) Roest H. van der, J. Leenen, M. Hofstra, J. Boeve and J. van

der Vlist. The Dutch contribution to the MBR development

in perspective. H2O MBR special, October 2001, pp. 7-10.

9) STOWA report ‘Infiltration water into sewers’ (in

progress).

H2O # 2005 17

MBR special III

SamenvattingIn 2000 begon een grootschalig pilotonderzoek naar de toepassing van de membraanbioreactorvoor de zuivering van communaal afvalwater in Nederland. De verwachtingen warenhooggespannen. In een zeer compacte installatie zou onder Nederlandse condities, met onderandere veel regenwater op de rwzi, een belangrijke verbetering van de effluentkwaliteitkunnen worden bereikt. Het pilotonderzoek heeft de nodige resultaten opgeleverd. Zo kon deflux worden opgevoerd, waardoor minder membraanoppervlak nodig is. Ook kon dereinigingsprocedure voor de membranen worden geoptimaliseerd. Daarnaast bleek eenbelangrijke verbetering van de effluentkwaliteit mogelijk, hoewel sommige verwachtingen,met name voor wat betreft de microverontreinigingen, niet waar konden worden gemaakt. Hetenergiegebruik kon weliswaar worden verlaagd, maar nog in onvoldoende mate, en vormtsamen met de hogere kosten van een MBR nog steeds een hindernis voor toekomstige(grootschalige) toepassingen. Daar waar sprake is van bijzondere situaties zal een MBR echter,al dan niet in hybridevorm, uitkomst kunnen bieden.

The DWR deignteam visits the Varsseveld MBR.

combining scientific excellence with commercial relevance

info@wetsus.nl

www.wetsus.nl

Agora 1

P.O. Box 1113

8900 CC Leeuwarden

The Netherlands

t +31 (0)58 - 284 62 00

f +31 (0)58 - 284 62 02

Wetsus, centre for sustainable water technology is a research institute in which renowned universities and industrial partners have joined forces. To face the steadily increasing global water problems, Wetsus focuses upon the development of treatment technologies for sustainable water. The main added value lies in the multidisciplinary use of biotechnology and separation techno-logy. Wetsus has the ambition to become the European Centre of Excellence for sustainable water treatment.

Despite tremendous research effortstowards development of new MBR technology,a worldwide technological breakthrough,especially towards domestic wastewatertreatment is lacking. The most importantreason is the high costs of MBR technology incomparison with conventional wastewatertreatment concepts. Cost reduction will be themain driver for research and development onMBR technology in the coming years. Shortterm research should focus on identification,characterisation and behaviour of foulants. Themechanisms of fouling should be studied. Thiswill yield knowledge for required membraneoperation, membrane properties to be chosenand new module design & operations thatfinally will result in reduction of costs andenergy. New methods to improve the aerationefficiency at high sludge concentrations shouldbe developed and attention is required for theeffluent quality. Long term research shouldfocus on much more sustainable MBRconcepts. New MBR concepts to treat differentwastewaters (e.g. grey and black) or to treatwastewater anaerobically should be explored.MBR technology offers opportunities toproduce energy out of waste (membrane fuelcells) and to minimize emission of greenhousegasses. Cutting down the operational costs ofMBR technology will be the key driver for

research. This article outlines some researchareas and specific topics that potentially willcontribute to lower costs. Special attention to

these topics should be given the coming years.Long term research should focus on sustainableMBR concepts. A few innovative developmentswill be presented.

Research drivers and topicsFor a real worldwide breakthrough of MBR

technology in domestic water treatment asignificant cost reduction is necessary. Thepresent generation membranes show very lowpermeabilities due to a high degree of fouling.Besides, operation of membranes is energydemanding and, although prices aredecreasing, membranes are still relativelyexpensive. In addition, one of the advantages ofMBR, i.e. a high sludge concentration iscounteracted by inefficient aeration at highsludge concentrations. To reduce theoperational costs of MBR technology researchshould be directed into three areas: fouling,filtration and aeration. Within these areasseveral research topics can be defined.

Fouling When research regarding membrane

fouling in MBRs is critically reviewed theconclusion must be that the main questionsstill are not answered. Even today it is unclearwhich group of compounds dominates thefouling process, whether these compoundshave their origin in the wastewater or areproduced by the biology, and what the mostimportant fouling mechanisms are (figure 1).

Origin fouling Without supporting scientific evidence the

H2O # 2005

MBR special III

19

MBR technology: future research directions

Harry Brouwer, TNO Science & Industry

Hardy Temmink, Wetsus

Maxime Remy, Wetsus

Stefan Geilvoet, TU Delft

The last decade R&D on MBR technology has shown a rapid development. In spite of its advantagescompared to conventional treatment systems, MBR does not yet provide a competitive alternative. Thesmaller required footprint and the excellent water quality are partially counteracted by high costs formembrane filtration and inefficient oxygen transfer. Reduction of costs will be the main driver forR&D on MBR technology the next years. R&D should be directed into three areas: fouling, filtrationand aeration. Long-term research should focus on the improvement of the sustainability of MBRtechnology by exploring new concepts of wastewater treatment. Examples worthwhile to be exploredare decentralised treatment of wastewater offering possibilities to treat separate streams of grey andblack water to enable water reuse; anaerobic water treatment to minimize emission of greenhousegasses and to decrease energy requirements and sludge production and to use MBR in combinationwith electricity production in a fuel cell to produce electricity and treat wastewater simultaneously.

Figure 1: Membrane fouling in MBR.

literature mentions dominant foulants varyingin size or molecular weight (flocs, colloidalmatter en solubles) and in chemical nature(extracellular polymers such as proteins,polysaccharides and humic compounds andinorganic precipitants). The origin of thefoulants still is subject to debate although fordomestic wastewater it seems that biologicalmaterial such as colloidal matter produced byshear and (soluble) metabolites or lysisproducts are more important than wastewaterconstituents. In a way this is an advantage asthis implies a generic character of foulingwhich also may allow for a generic solution.

Fouling mechanismsThe mechanism of membrane fouling in

MBRs is extremely complex. Formation of acake or gellayer, pore blocking, adsorption onthe membrane surface or in the membranepores could all be important. Equallyimportant could be the interaction betweenthese mechanisms. For instance, a (dynamic)cakelayer provides an additional resistanceagainst filtration, but at the same time mayprotect the membrane against (irreversible)fouling. Figure 2 and table 1 provide anexample of such behaviour.

Two sludge samples were taken from themembrane tank in the same MBR, operated atdifferent conditions. The sludge samples andthe supernatant of these samples were filteredin a specifically designed set-up and theresistance against filtration was monitored as afunction of the permeate volume during cross-flow filtration (Figure 2). In addition, free andsludge bound proteins and polysaccharideswere determined (Table 1). The supernatant ofsample 2 exhibited a higher resistance than thesupernatant of sample 1. Possibly, this can beattributed to higher concentrations of freeproteins or polysaccharides present in sample2. However, for the (total) sludge samples theopposite applies with a much higher resistancefor sample 1 than for sample 2. This suggeststhat with sludge sample 2 a cake layer wasformed which was much more protectiveagainst fouling than with sludge sample 1 andat the same time this cake layer was morepermeable than the cake layer with sample 1.

Foulant process interaction The example above illustrates that

identification and characterisation of the mainfoulants allow a better focus towards theproblem of fouling. First of all this concernsresearch towards design and operation of thebiological reactor with the objective to reducethe concentrations of these foulants. Today thebiological reactors of MBRs are not designedand operated differently from conventionalsystems applying secondary settlers. This seems

odd as the microbial population in MBRs can beexpected to be different from conventionalsystems and much higher sludgeconcentrations are maintained. Shear, waterhardness, sludge and hydraulic retention time,the substrate gradient, the COD/P/N ratio ofthe wastewater and the redox regime all areexpected to be important factors determiningthe concentration of foulants. Knowledge aboutthe behaviour of these foulants in response tosuch factors is extremely important to allowmodification of the design and operation of thebiological reactors in MBR systems with theobjective to minimize fouling.

In situ foulant monitoring A second approach is to study membrane

fouling more in detail, preferably using in situand direct monitoring techniques. It isappreciated that this is extremely difficult butmay become feasible once more information isavailable about the dominant foulants. Thisapproach should lead to better membraneoperation, not only during stable conditions,but particularly in response to varyingconditions such as rain weather incidentswhich are known causes of severe foulingproblems.

Also, this will result in a morefundamental basis for the selection of theappropriate membrane properties such as poresize, membrane material, etc., and for thedevelopment and design of new membranemodules which nowadays more or less is basedon a trial-and-error approach.

FiltrationFoulants will undoubtedly have their

impact on the permeability of the membranes.Other parameters that affect the permeabilityare the trans-membrane-pressure (TMP), shearrate, temperature and sludge composition.Especially membrane properties andmembrane material are crucial elements thatinfluence the membrane permeability.

Enhanced hydrodynamics Current developments in membrane

operation are aimed at reducing membranefouling by selecting the proper hydrodynamicconditions. Generally this is done by creatingturbulent flow conditions near the membranesurface. The shear flow rate along theboundary layer of the membrane stronglyaffects the membrane flux. A standard methodto promote shear nowadays is to dose airbubbles. Although air bubbles enhance themembrane flux, they also have their impact onthe particle size distribution and flocstructure. Moreover, enforced aerationdemands energy. Research should be directedto optimisation of the current coarse aerationmethods for submerged membrane modules.Secondly, alternative filtration concepts topromote shear locally near the membranesurface, e.g. by mechanical means, should bedeveloped.

Improved membrane propertiesThe past few decades the main areas of

attention for membrane development were

H2O # 200520

MBR special III

sample 1 sample 2bound free bound freemg.gVSS-1 mg.l-1 mg.gVSS-1 mg.l-1

polysaccharides 64 118 49 170proteins 53 17 40 62

Table 1: Composition sludge samples 1 and 2.

Figure 2: Filtration resistance as function of filtered permeate volume.

membrane morphology, hydrophobicity orhydrophilicity and charge. In time foulingleads to changes in porosity and pore sizedistribution. The size of pores decreases as aresult of pore narrowing due to internaladsorption. The number of small pores alsodecreases due to clogging. Therefore flux andselectivity are changing during the filtrationprocess and during the lifetime of membranes.Very little is known about the interactionbetween foulants and membrane properties.Understanding of the relation betweenfoulants and membrane properties will be achallenging research item the coming years.

AerationConventional activated sludge processes

have an insufficient aeration performance. Lessthen 10-15% of the oxygen supplied istransferred to the water phase using finebubble diffusers1). Hence, the electricity neededfor aeration provides more than 50% of the totalenergy costs of municipal water treatment. InMBR systems the oxygen transfer is even lessefficient due to the high concentrations ofsolids and therefore the energy requirementseven will be higher. This stresses the need for abetter understanding of the potential causesand measures to be taken to improve the pooraeration efficiency in MBR systems.

Understanding nature and impact on α-factor MBR has many advantages over

conventional treatment. One of them is theability to apply high concentrations ofactivated sludge, over 10 g/l and higher.However, high levels of biomass will decreasethe efficiency of oxygen transfer. The researchin Beverwijk has shown there is a correlationbetween high viscosity and low aeration

efficiency. The contribution of EPS towardslower oxygen transfer efficiencies was lessobvious. More knowledge is required to allowtranslation of these observations to a betteroperation of the biology in order to improvethe aeration efficiency.

New efficient oxygen transfer devicesThe most efficient aeration devices

currently being employed in activated sludgeprocesses are fine pore aeration systems, eitherusing ceramic or membrane rubber materials.Membrane diffuser technology offers thehighest oxygen transfer efficiencies comparedto other type of aeration systems like coarsebubble diffusers, surface aerators, brushaerators, jet aeration and venturi aerationsystems.

However, the oxygen transfer efficiency athigh solids concentration not only depends onthe type of aeration but also on sludgecharacteristics. Especially mechanical stressenforced by shear could promote the oxygentransfer efficiency. On the other handmechanical stress will change the sludgestructure and will require energy. Moreresearch is needed, preferably on full scale, toexplore the possibilities and limitations of newand existing aeration systems.

Enhanced oxygen transfer new configured flatsheet-frame modules

The depths of conventional activatedsludge processes are usually in range of 3 to 4meter. In case of diffused air aeration,consequently, there is a limited time foroxygen transfer from gas to liquid phase.Transfer of oxygen can be improved by newdesigns of flat sheet and frame modules. Theidea is to increase the retention time of the air

bubbles. Potential energy savings for aerationcould be in the range 10-30%.

Enhanced oxygen transfer by new reactordesigns

An example of a new reactor design thatcould promote the oxygen transfer efficiencyspectacularly is the deep-shaft MBR, whichcould be a unique modification of theconventional MBR with even smallerfootprints and oxygen transfer rates that aresignificantly higher. The main objective of thissystem is to increase the efficiency of oxygentransfer. The deep-shaft configurationincreases the partial pressure of oxygen,thereby causing a high saturationconcentration in the reactor. The deep-shafttechnology has been successfully applied forhigh strength polluted wastewaters. Morethen 80 full-scale references are known.Possibilities for domestic water treatmentcould be explored.

New innovative future MBRtreatment concepts

The present generation of MBR systems isfar from sustainable. The treatment process isenergy demanding, produces an immenseamount of waste sludge, nutrients aredestroyed, greenhouse gasses are produced andpotential energy sources are wasted. Moresustainable solutions are needed in order torecycle the water in dry areas, in order torecover nutrients like phosphates andnitrogen, and in order to produce or minimizethe energy use to prevent production ofgreenhouse gasses. For future sustainablewater treatment new concept approaches arenecessary.

Decentralised sanitationNewly developed decentralised sanitation

concepts2),3) offers possibilities for MBRtechnology. Within the decentralisedsanitation concept wastewater is treated closeto a household with a strong focus onsustainable use of resources. Generally threewastewater streams with different origin arecollected and treated: grey water (shower,kitchen), yellow water (urine) and brown water(toilet water without urine). These wastewaterstreams differ strongly in amount andcomposition. Especially towards water reuseopportunities new MBR applications should bedeveloped.

Anaerobic MBRGenerally anaerobic treatment of

wastewater has two main benefits: low energyrequirements (no aeration, formation ofbiogas) and low sludge production. Bothfeatures are very attractive for thedevelopment of anaerobic MBR concepts for

H2O # 2005 21

MBR special III

Direct production of electricity out of wastewater using microbial fuel cells (TNO).

domestic water treatment. The role of foulingwill be crucial, even more then with aerobicMBR systems. Anaerobic conditions promotethe formation of struvite, an insoluble saltbuilt up from magnesium, ammonium andphosphate. Especially near the membranesurface where concentration polarisation(accumulation of compounds) occurs, thedanger of struvite being precipitated on themembrane is high4)). Another point of interestis the way the membrane configuration ischosen. Due to the absence of air, submergedmodules, attractive from an energy point ofview, are less appropriate. Most anaerobic MBRconcepts use hollow fiber cross flowconfigurations that are energy demanding.New alternative filtration / operation modulesshould be developed: e.g. recirculation ofbiogas to enhance shear locally on themembrane surface5).

Microbial Fuel Cell The present generation water treatment

systems are wasting potential energy sources.Theoretically the combination of MBR withthe microbial fuel cell principle could be used

to produce electricity directly fromwastewater6), while at the same timeaccomplish water treatment7). This newapproach could lead to significant reduction ofoperation costs, but also creates opportunitiesto produce products with market value. Proofof principle is demonstrated. The next step tobe taken is to study the technological andeconomic feasibility of the process8). Wetsus aswell as TNO are working on the production ofenergy from wastewater in bio-fuel cells,which is one of their main research themes.

ConclusionsIn spite of its advantages compared to

conventional treatment systems, MBR doesnot yet provide a competitive alternative. Thesmaller required footprint and the excellentwater quality are partially counteracted byhigh costs for membrane filtration andinefficient oxygen transfer. Therefore thecoming years MBR research should be directedtowards minimization of costs for filtrationand aeration. Further it would be worthwhileto improve the sustainability of MBRtechnology by exploring new concepts of

wastewater treatment. Examples aredecentralised treatment of wastewater offeringpossibilities to treat separate streams of greyand black water; anaerobic water treatmentwith low energy requirements and low sludgeproduction and direct energy production bycombining MBR with microbial fuel celltechnology. ¶

Harry Brouwerproject manager / scientist TNO Science &IndustryP.O. Box 342, 7300 AH Apeldoornphone: +31 55 549 39 91e-mail: Harry.Brouwer@tno.nl

Hardy Temminkscientific project manager WetsusP.O. Box 1113, 8900 CC Leeuwardenphone: +31 58 284 62 00 / +31 317 48 48 03e-mail: hardy.temmink@wetsus.nl /hardy.temmink@wur.nl

Maxime RemyPhD student / researcher Wetsusphone: +31 58 284 62 08e-mail : Maxime.remy@wetsus.nl

Stefan GeilvoetPhD student / researcher Delft University ofTechnologyStevinweg 1, 2600 GA Delftphone: +31 15 278 40 26e-mail: S.P.Geilvoet@CiTG.TUDelft.nl

Literature

1) Koot A. (1980). Behandeling van afvalwater.

2) Otterpohl R., M. Grottker and J. Lange (1997). Sustainable

water and waste management in urban sreas. Water

Science & Technology, vol. 35, no. 9., pp. 121-133.

3) Lens P., G. Zeeman and G. Lettinga (2001). Decentralised

sanitation and reuse - Concepts, systems and

implementation. ISBN 1900222477.

4) Choo K-H., I-J. Kang, S-H. Yoon, H. Park, J-H. Kim, S.

Adiya and C-H. Lee. (2000). Approaches to membrane

fouling control in anaerobic membrane bioreactors. Water

Research, vol. 41, pp. 363-371.

5) TNO (2002). AMBR: Experimental study on biofouling

behavior. TNO report R 2002/391.

6) TNO (2004). Biobrandstofcel. TNO report R 2004/340.

7) Logan B., C. Murano, K. Scott, N. Gray and I. Head

(2005). Electricity generation from cysteine in a microbial

fuel cell. Water Research vol. 39, pp. 942-952.

8) TNO (2005). Geïntegreerde biologische waterzuivering en

elektriciteitsproductie (In preperation).

H2O # 200522

MBR special III

SamenvattingDe laatste decennia is wereldwijd veel onderzoek verricht naar de MBR-technologie.Desondanks ontbreekt vooralsnog een technologische doorbraak voor zuivering vancommunaal afvalwater. De kosten van de MBR-technologie blijven hoger dan voor deconventionele zuiveringstechnieken. Kostenvermindering zal de komende jaren dan ook debelangrijkste stimulans zijn voor verder onderzoek en ontwikkeling van MBR-technologievoor communale waterzuivering. De aandacht zal zich moeten richten op drie gebieden:vervuiling, filtratie en beluchting. Naast onderzoek gericht op kostenreductie is ontwikkelingvan innovatieve, meer duurzame MBR-concepten gewenst. Voorbeelden hiervan zijndecentrale zuivering om te komen tot gedeeltelijk hergebruik van water en wellichtnutriënten, anaërobe zuivering van afvalwater en het toepassen van MBR-technologie incombinatie met de productie van elektriciteit in een brandstofcel. De potentiële voordelenlegitimeren een verdere verkenning van deze innovatieve MBR-concepten.

3D-analysis of flat sheet and frame membrane with Confocal laser scanning microscopy. Microbes (DNA) are visible as green.

The responsibility for the purification ofhousehold wastewater in the Netherlands hasbeen incorporated in approximately 25 waterboards. The traditional procedure in theNetherlands when adapting a regional waterpurification installation can be generallydescribed as follows.

First, a technological concept is chosenbased on the end user’s requirements. This isthen fleshed out in a basic design, a detaildesign and a specification. Finally, aftertendering, realisation and start-up arecommenced. Consultancy firms are involvedby end users in all these steps as designers andmanagers of the construction phase.

In essence, the activities of the consultancyfirms in the typical project route describedabove are multidisciplinary by nature.Integration between the disciplines concernedand co-operation in the design andconstruction process are essential for a goodprice/quality ratio. Consultancy firms are theliaison between knowledge institutions,suppliers and end users. In this role the

consultancy firms constantly look out forglobal developments in both technical andtechnological fields. The market supply of

suppliers is closely monitored anddevelopments among suppliers and knowledgeinstitutions are also identified in good time.Initiating and redirecting developments intorelevant practical conditions is performed bymany consultancy firms. Because they areinvolved in the construction, start-up andoptimisation of regional water purificationinstallations, consultancy firms possessknowledge about the practical situation; theyare excellent discussion partners for the endusers. As intermediaries, consultancy firms canensure the correct communication level in aproject thanks to their knowledge about thecustomer’s circumstances. This is because theconsultancy firms are more willing and able tocreate an open atmosphere for specificknowledge transfer, whereas confidentialityand secrecy of information play a much greaterrole among suppliers.

MBR and other new developmentsIn 2000, a start has been made in the

Netherlands with the pilot research on theregional water purification installation inBeverwijk, with a plan for the developmentand large-scale introduction of MBRtechnology for the treatment of householdwastewater. The pilot research has lead to newinsights into the design and the operationalmanagement of MBR. These insights havebeen implemented in practise at the recentlyopened MBR Varsseveld. Furthermore, thepreparations for the realisation of anotherthree large-scale MBRs for householdwastewater are in full progress. This concernsthe projects in Heenvliet, Ootmarsum andHilversum. The plan mentioned above ischaracteristic for the ‘Dutch school’ and for therole of consultancy firms in applying newwastewater purification technology.

H2O # 200524

The role of consultancy firms in the application of newwastewater purificationtechnology

Bert Geraats, Grontmij

Peter de Jong, Witteveen+Bos

Jans Kruit, Royal Haskoning

Helle van der Roest, DHV Water BV

The Netherlands has a strong foundation in the field of wastewater treatment. The country has along history of developing and introducing new biological wastewater purification technologies on apractical scale. Some of the most talked about examples of these technologies, which are well knownthroughout the whole world, are anaerobic wastewater purification in accordance with the UASBprinciple, the Carrousel system, BCFS, and the Sharon and Anammox nitrogen removal technologies.Important success factors in these are the traditionally strong microbiological basis in theNetherlands and a unique work method that aims to integrate disciplines. The primary matter ofimportance is that only a practice-oriented application of scientific developments to the practicalscale will lead to a successful application for the end user. In the Netherlands, consultancy andengineering firms fulfil a pivotal role in this application.

MBR special III

Carrousel, Geestmerambacht.

A typical feature of an MBR is theinteraction and harmony between biology andmembranes. Particularly the focus on thisinteraction and the multidisciplinary approachhas formed the foundations of crucialimprovements in the design of MBR. Theoutstanding Dutch microbiological and processtechnological knowledge and the actions takenin service of total solutions geared to practicalsituations have proven themselves again.

You could compare an MBR with anorchestra in some ways. In order to realise asymphony, co-operation and harmony betweenthe soloists under the leadership of a conductoris vital. And so, as you can guess, consultancyfirms are the leaders of orchestras. The leader ofthe orchestra knows which instruments to useand which musicians will be selected to playthese. An MBR is a fine example of anintegrated wastewater purification concept.Biological, technological, civil and mechanicalaspects, membrane technique andmanagement form the backbone for asmoothly operating MBR. It is clear that theconsultancy firms possess integratedknowledge of the purification process.

New developments and project risksThe development and large-scale

introduction of new technology ischaracterised by various learning curves. Thisis inevitably accompanied by risks. At first,during laboratory and pilot-scale research, the(harmful) risks are still limited. But in the firstlarge-scale applications, both the financial andthe technical (harmful) risks are on a verydifferent level. Minimizing these risks is vital -and this is where professional and integratedadvice comes into play.

End user’s must be made aware of possiblerisks and must accept part of these. Settingrealistic requirements is essential. Animportant question is how long it is actuallypossible during the phases of a project to

include various systems and to introduce newdevelopments: liberty versus clarity. Aconsultant can present new technologies andsuppliers from rough to fine, determine andevaluate distinguishing characteristics.During the supervision of the system choiceprocess it will be determined which choiceswill be fixed and which will be left to themarket and left open during the tenderingphase. In the design phase a decision will haveto be made about whether a chosen conceptwill be maintained or that flexibility will bebuilt in to include various systems and newdevelopments. Certainly in the case of newtechnologies, a realistic responsibility for theresult must be placed with thesupplier/builder of vital components. Thismust be taken into account when a choice ismade about the form of tendering and thecontract. Is a functional specification suitablefor a development project, or is a detailedspecification better to minimize the risks? It iscertainly important that the aspectsmentioned above link up with the culture andwork method of the end user.

ConclusionThe consultancy firms see a role for

themselves as leaders of the ‘wastewater

Purification Orchestra’. It is clear that thisorchestra can only be successful given a goodquality of the leader of the orchestra as well asthe members of the orchestra. Communicationand co-operation between the members of theorchestra, the public and the principal areessential in this.

The Dutch ‘wastewater PurificationOrchestra’ has already proven on numerousoccasions with different symphonies (MBR,UASB, Carrousel, BCFS, Sharon) that it deservesto play on the world’s stages. In order to keepon doing this in the future, continuous andstructural work must be performed on thequality and commercial strength of the watersector. ¶

Bert Geraatssenior consultant Grontmij Infrastructure andEnvironmentP.O. Box 203, 3730 AE De Biltphone: +31 30 220 79 11 e-mail: Bert.Geraats@grontmij.nl

Peter de Jongsenior consultant Witteveen+BosP.O. Box 233, 7400 AE Deventerphone: +31 570 69 79 11e-mail: p.djong@witbo.nl

Jans Kruitsenior consultant Royal HaskoningP.O. Box 151, 6500 AD Nijmegenphone: +31 24 328 42 84e-mail: j.kruit@royalhaskoning.com

Helle van der Roestprincipal consultant DHV Water BVP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 24 07e-mail: Helle.vanderroest@dhv.nl

H2O # 2005 25

MBR special III

Sharon, Zwolle.

SamenvattingNederland kent een lange en succesvolle traditie op afvalwaterbehandeling. De advies- eningenieursbureaus vormen een belangrijke functie in de vertaling van nieuwe ontwikkelingennaar de praktijkschaal. Door de betrokkenheid bij de bouw, opstart en optimalisatie van rwzi’shebben de Nederlandse adviesbureaus kennis van de praktijk en zijn een volwaardigegesprekspartner voor de waterbeheerders. De recente MBR-ontwikkelingen in Nederland zijnkenmerkend voor de Nederlandse werkwijze en de rol van adviesbureaus in de toepassing vannieuwe afvalwaterzuiveringstechnologie. Een multidisciplinaire aanpak en integrale kennisvan het zuiveringsproces zijn hierin cruciaal. Adviesbureaus bieden integrale advisering metkennis en kunde waardoor eventuele risico’s bij nieuwe ontwikkelingen kunnen wordengeminimaliseerd. Om het succes van de Nederlandse afvalwatertechnologie ook naar detoekomst toe te verzekeren dient de aandacht voor investeringen in kwaliteit en commerciëleslagkracht van de Nederlandse watersector niet te verslappen.

HUBER Vacuum Rotation MembraneBioreactor VRM®

HUBER Complete Plant BioMem®

for up to 500 PE

HUBER MembraneClearBox® MCB SmallBiological Wastewater Treatment Plantfor up to 50 PE

➤ Future oriented solution forincreasing requirements

➤ Free of solids, bacteria andgerms

➤ Allows for reuse of used water

➤ Complies with the latest legalstandarts

➤ Suitable for decentralised municipal and industrial applications

➤ Minimum space demand

Hans Huber AG · Maschinen- und AnlagenbauPostfach 63 · 92332 Berching · Telefon: +49 - 8462 - 201 - 0Telefax: +49 - 8462 - 201 - 810 · E-Mail: info@huber.de

www.huber.de

Membrane Bioreactor forany type of wastewater

The QualityCompany –Worldwide

At Water Quality Europe (the annualDutch conference for industrial water), thecentral issue was how to boost Dutchinnovation in the water sector. Due to the newEuropean Water Framework Directivedemands on wastewater discharge and acontinuing motivation to improve costperformance, there is a permanent drive forinnovation in industry. MBR was one of themost promising technologies discussed at theconference.

Important criteria for investment in watertechnology in the process industry are thecontribution to the continuity of theoperations (the so called ‘license to operate’),cost minimisation and customer satisfaction,local stakeholders included. Under whatcircumstances can MBR meet these criteria?Companies have different reasons to consideran investment in MBR. As in the case oftreatment of wastewater from households, astable quality of the effluent is an importantargument and the compact design of MBR isattractive. However, the most promising seemsto be the application of MBR for water reuse.Last year, most of the respondents to a VEMWquestionnaire saw the future for MBR inindustry in in-process applications. Companiesstill need more information about operationalcosts of MBR and about expecteddevelopments in energy consumption andsludge reduction. Although compact andflexible systems have the future, not allquestions have been answered. Industryhowever, is confident that suppliers will comewith sufficient solutions in coming years.

Johan Raap of CSM comments on thewebsite of the Dutch Ministry of EconomicAffairs on innovation in the water sector that

companies experience “a gap between what theindustry develops and what is permitted bythe authorities.” There is tension between theimplementation of solutions and dischargepermits. “There is a lack of reason behindmany water rules: these rules might evenprove to be contra-productive to their initialaim.” In his view a critical flaw in the currentwater policy is a lack of an overall vision on theso-called water chain. This water chain extendsfrom intake of water to discharge aftertreatment. Industrial water usage, with itspractice of reuse and (process) water treatment,forms an essential element of the water chain.In legislation and discharge permits, thecontribution of industry to the water systemshould be recognised in full.

In a water chain orientated approach, theuse of more flexible and cost oriented

instruments is essential. In the existinglegislative framework real barriers to such anapproach are the rigid and non-transparentfinancing schemes of municipal sewers incombination with a narrow view in thedischarge permit, focussing on the receivingsewer service. A cost oriented fee shouldaccommodate an environmental and economicbeneficial option, where pre-treatment ofindustrial wastewater occurs. In these casescurrent restrictions to further treatment of‘thin’ or diluted industrial wastewater shouldbe mitigated in order to encourage treatmentfacilities. A cost oriented fee contributes to thepredictability of tariffs and fees and thereforeimproves the investment climate.

CargillIn Bergen op Zoom in the southern part of

the Netherlands, Cerestar, a Cargill company,operates a factory where both maize and wheatare used as raw material to supply customerswith food and non-food products. Their MBRcame online in 2001 with a capacity of 35 m3/hand is used to upgrade process water forcooling purposes, and lead to a water saving of270,000 m3/year, or 20 % of the total waterusage. The total projects costs were about onemillion Euro.

Subsidies turned out to be a real incentiveto the project, and the Dutch authoritiesfinanced about 30 % of the investment costs.On the other hand, a problem is therequirement in the discharge permit for whatis called ‘thin’ or diluted water, where thetreated wastewater should not be dilutedfurther than 350 l/pollution equivalent (p.e.)and limits the back up-provision for the MBR-facility. At the moment, the company has tobypass the MBR if there is no or reduced

H2O # 2005

MBR special III

27

Industry is ready for MBR Henk Brons, VEMW

In the Netherlands about twenty highly diverse industrial MBRs have been realised in the past tenyears. This respectable number could have been even higher if the water policy of Dutch authoritieshad given more support to the industrial initiatives. Why do companies consider investing in MBRand does governmental policy hinder them? VEMW, the Dutch Association of non-domestic waterusers, pleads for clear guidelines on fees and market conditions to enable investment in MBR.Industry is ready for innovation; an active approach from the water authorities is called for.

Cerestar.

demand for cooling water in the productionprocess. Although the MBR has both a higherefficiency and a better environmentalperformance than the receiving communalwastewater treatment plant Cerestar is forcedto bypass contaminated wastewater.

Due to the presence of salts, the treatedwater cannot be directly discharged intoreceiving waters. Cerestar would like toconsider replacing the existing aerobictreatment plant with a second MBR-facility.

This investment should contribute tolower costs; due to higher efficiency and savingof the fee per discharged population equivalentpayable to the water board, a lower sludgeproduction and a better performance on COD-reduction, where the energy balance has to beconsidered. However, this option will beblocked by the discharge requirement for thinwastewater and therefore Cerestar have notprepared an investment decision until now.

Recently Cargill constructed a MBR at theoil seed crushing plant in Saint Nazaire(France). Cargill started the MBR last January.It is a submerged MBR with tubular PVDFcoated membranes with a capacity of 850 kg/dCOD. This design was selected due to planteffluent characteristics, small footprint (spacelimitation) and minimal sludge production.The choice for a submerged MBR reducedenergy consumption compared with thealternative external cross-flow membranes.The tubular type membranes allow reversedflow at regular time intervals while coarsebubble aeration creates enough turbulence toavoid biomass built-up at the surface of themembranes. The effluent quality should allowreusage as a make-up water-source to providecooling water in the near future.

In Amsterdam Cargill is investigating theoption of using MBR technology for aerobicside stream treatment of plant effluentcontaining sulphates in the range of 20,000-40,000 ppm. The objective is to discharge thesulphates directly into the dock andconsequently to eliminate the constraints todischarge via the municipal sewer. Sulphatesmay contribute to corrosion as sulphates arebiologically reduced to sulphides underanaerobic conditions. Also H2S-formationraises concerns regarding occupational healthand safety conditions at wastewater treatmentplant Westpoort. In this application MBR lookspromising as the biomass is retained in thereactor as in the classical design, sludge-settling characteristics are poor due todispersed growth. A pilot plant trial has beensuccessfully completed and now that the saltscontaining water can be discharged into thereceiving brackish water, the approval by theauthorities is to be expected.

Diosynth Oss Diosynth is an Akzo Nobel company and

produces active pharmaceutical ingredients.The manufacturing facility in Oss produceswastewater (20,000 p.e.) that contains priority

substances. Anticipating on expectedlimitations on effluent components under theWater Framework Directive, Diosynth aimedto improve knowledge on stabilising effluentquality. Diosynth therefore installed an MBR-

H2O # 200528

MBR special III

company location sector year capacity(m3/h)

Essent Landgraaf landfill 1995 13Essent Montfort landfill 1996 10Afvalzorg Middenmeer landfill 1996 5Recept Rotterdam composting 1996 2Nijhoff Wassink Rijssen tank cleaning 1997 1Smink Hoogland landfill 1996 20Den Ouden OTT Alblasserdam tank cleaning 1999 1Royal Dutch Navy Den Helder bilgewater 1999 5Dekker Transport Oudekerk a/d IJssel tank cleaning 1999 1,5Dekker Transport Dendermonde tank cleaning 2000 2Vos Logistics Zuidbroek tank cleaning 2000 2,5Vos Logistics Rotterdam tank cleaning 2000 12Gentenaar Moerdijk tank cleaning 2000 4Biocel Lelystad compost 2001 2Pieter Bon Zaandam tank cleaning 2002 8Van Gansewinkel Weert liquid waste treatment 2002 16TCL Maastricht tank cleaning 2003 3Ecopark De Wierde Oudehaske landfill 2003 30Essent Haps landfill 2004 5Hoyer Rotterdam tank cleaning 2004 8ATM Botlek tank cleaning 2004 7,5Vos Logistics Hoogerheide tank cleaning 2004 1,53e Merwedehaven Dordrecht landfill 2004 36Stubbe Gouda tankcleaning 2005 2 Sources: DHV, Triqua, Logisticon, Zenon, Solis, RWB Afvalwater, TNO-MEP

Tabel 1: MBR in waste treatment.

company location sector year capacity(m3/h)

Noviant Nijmegen coatings 1996 20SCA Suameer paper 1998 0,7Driessen Dongen leather 1998 10Rendac Bergum rendering 1999 40VWS Broek op Langedijk flowerbulb cleaning 1999 3VHP Ugchelen paper 2000 12Platvis Volendam food 2000 2Cerestar Bergen op Zoom wheat refinery 2001 35Rentex Floron Bolsward textile cleaning 2002 35Astra Faam Harlingen food 2002 2,5Rendac Son rendering 2003 100Akzo Nobel Oss pharmaceutical 2003 1,2Du Pont / Invista Dordrecht chemical 2003 60Fuji Tilburg photo/film 2004 35Bavaria Eemsmond malt 2005 55Noviant Nijmegen coatings 2005 3Cargill Amsterdam soja bean crushing 2005 3Sources: DHV, Triqua, Logisticon, Zenon, Solis, RWB Afvalwater, TNO-MEP, VEMW

Tabel 2: MBR in industry.

pilot with a capacity (membrane flow) of 1,200litres per hour.

The pilot aimed to investigate thebiodegradability of substances in thewastewater streams in the production processusing MBR-technology. The researchprogramme was specifically aimed at thebiodegradability of COD, nitrogen and prioritysubstances (volatile organic solvents,chlorinated carbon-hydrogen and toluene).The pilot also gave information aboutoperational costs, the necessary maintenanceand operation tasks and the capacity of theinstallation. During the pilot also technicalquestions had to be answered which led tomodifications of the design. This was the casefor e.g. the composition of the membranes, theaeration and adaptations to improve theoperations (e.g. cleaning of the membranes,tuning of the process parameters and

maintenance schemes). Diosynth paid muchattention to stabilise operations and to thetraining of operators.

The pilot was evaluated at the end of 2003.The MBR led to a reduction of contaminationto 3,500 p.e. The pilot was run without subsidy.The Water Board was very interested in theresults of the pilot, especially in itscontribution to improving the effluentquality. New environmental standards wereexpected according to the Water FrameworkDirective, and the wastewater of Diosynthcomplied entirely with the discharge permitsubmitted by the Water Board, also on theaspects of thin water. A possible replacement ofthe pilot with a definitive plant might haveconsequences for the wastewater profile.

Changes in the water sectorAs the Dutch representative for non-

domestic water users, VEMW has concernsabout exclusive rights of dominant suppliersof water services covering industrial waterusers in the Netherlands. These rights lead toefficiency loss in the water market andtherefore to unnecessary high prices. Pricesthat are not set by the market are as such not aclear indicator for investment for industry.This lack of both transparency and efficiencyimplies a strong need for an independentauthority with sufficient and effectiveinstruments to enhance efficiency and lowerprices in water services. To counter thisinefficiency in the water market, VEMW callsfor an independent supervisor to stimulate theinvestment climate.

As illustrated in the cases, a possibleconflict of interest in the wastewater sectoremerges as the water boards combine botheconomic dominant positions in wastewatertreatment and granting wastewater dischargelicences. In combination with the legalmonopoly on treatment of domestic sewage,

this competence for granting permits is aserious threat to industrial initiatives for thetreatment and reuse of processwater. Due tothe ability of water boards to place restrictionsto industrial initiatives, in order to protecttheir own sewage treatment plants (i.e.: of thewater boards), they can prevent or discourageinitiatives for water reuse which are as suchbeneficial both for economic andenvironmental reasons. This combination offunctions of the water boards leads to theundesirable situation in which the referee isalso a player in the match. Therefore, thecurrent combination in the Netherlands ofresponsibilities in the field of operations,permits and inspections of the treatment ofwastewater, should be separated. Clear optionson permit-conditions are necessary to improvethe investment climate.

Due to geographic circumstances, wateravailability should be an incentive forcompanies to settle in the Netherlands.Specialised and applicable knowledge aboutwater technology is widely available. Changingthe water sector in order to reach transparentand structural low tariffs, and eliminatingconflicts of interest will boost innovation.Industry is ready for MBR -The governmentmust now set the right conditions. ¶

Henk Boenink (AKZO Nobel) , Chris Velzeboer(Cargill), Ernst Covers (Cerestar), Johan Raap(CSM) and Joan Brouwer (Diosynth) made thisarticle possible by adding information.

Henk Bronsdirector water VEMWHouttuinlaan 8, 3447 GM Woerdenphone: +31 348 48 43 54e-mail: hb@vemw.nl

H2O # 2005 29

MBR special III

Diosynth.

SamenvattingVanaf 1995 zijn circa 20 industriële toepassingen van de membraanbioreactor gerealiseerd. Metname voor wat betreft het sluiten van de waterkringloop is de MBR kansrijk in de industrie.Cerestar te Bergen op Zoom realiseert sinds 2001 door gebruik van een MBR hergebruik vanproceswater voor koeling. Diosynth te Oss heeft in een pilotinstallatie een aanzienlijke reductievan prioritaire stoffen bereikt. Toepassing van MBR binnen de industrie wordt echter beperktdoordat vanuit het beleid onvoldoende oog is voor de rol van industriële waterbehandeling inde waterketen. Het waterschap kan in de lozingsvergunning de mogelijkheden van voorzuive-ring door bedrijven beperken, bijvoorbeeld door een eis voor ‘dun water’. Bovendien is definancieringsstructuur in de waterketen star en ondoorzichtig, wat investeringsbeslissingenbemoeilijkt. Doordat economische machtsposities in de watersector bestaande inefficiënties instand houden of zelfs bevorderen, zijn watertarieven te hoog. Onafhankelijk toezicht op dewatersector draagt bij aan het terugbrengen van deze inefficiënties, geeft meer kostentrans-parantie en levert door beter voorspelbare watertarieven een bijdrage aan het investerings-klimaat. Een beter investeringsklimaat draagt bij aan de toepassing van MBR in de industrie.

BERGHOF Filtrations- und Anlagentechnik GmbH & Co. KGHarretstraße 172800 Eningen • Germany

Tel.: +49 7121 894 205Fax: +49 7121 894 200E-Mail: info@http://www.berghof-filtration.de

berghof-filtration.de

Hyperm-AE module

exchangeable 11.5 mm membranes

chemical resistance up to pH 13

high pressure applications (10 bar)

high temperature applications (80 °C)

�

�

�

�

HyperFlux module

potted membrane2

high packing density (>39 m )

reuse of module housing

NEW: 10-inch module

�

�

�

�

Tubular membranes

Organic membranes

PVDF and PES

8 up to 13 mm

robust

�

�

�

�

Our experience and innovative

know-how are based on more than

30 years of experience with tubular

membranes in demanding high-

solids applications. Our tubular

membranes have been used in

literally thousands of applications.

BERGHOF is one of the main

suppliers for the industrial MBR-

market.

BERGHOF supplies his products

and engineering know-how via

original equipment manufacturers

(OEM's) and contractors world-wide.

BERGHOF's cooperation is based

on long-term Partnerships.

BERGHOF products are used for:

biomass separation, oil-water

separation, beverage, automotive,

wastewater treatment and

biotechnology.

Fore more information please visit

us at: www.berghof-filtration.de

BERGHOF MBR-experience

easy cleanability and high reliability

high fluxrates (80 -150 lmh)

largest cross-flow MBR of 210 m³/h

more than 85 references in MBR

NEW: MBR Low Energy Concept

�

�

�

�

�MBR-leachate; BERGHOF module 83 37100 I10Built by WEHRLE Umwelt GmbH, Germany

Innovation is our tradition

Scope of supply

membranes and modules

complete assembled racks

engineering and commissioning support

piloting study

�

�

�

�

First applications of MBR technology wereestablished in the 1960s. The first installationswere mainly found in Northern America and

Japan and were accomplished in two distinctlydifferent settings. In Northern America theinitial application of MBR technology

comprised treatment of wastewateroriginating from the petrochemical industry.In Japan first experience with MBR technologywas gained by applying the concept as awastewater treatment concept for apartmentblocks. One thing the initial applications allhad in common was the limited scale on whichthe technology was applied; the drivers tochoose for MBR technology were differenthowever. This article addresses the differentdrivers for the application and development ofMBR technology over the years in differentcountries. Furthermore two case studies arehighlighted: MBR in the UK and waterreclamation at Ben Gurion InternationalAirport (Tel Aviv).

In line with the differences in the motivesfor choosing MBR technology in the firstapplications as mentioned above, todaycountries have different drivers for applicationand further development of MBR technology.In this first paragraph the global drivers forMBR technology are stipulated from East toWest.

AsiaMBR technology is being considered at

many locations all over Asia, the main driverbeing water reclamation. Examples of settingsvary from small-scale applications in Japan,where MBR product water is reused as toiletflushing water in apartment blocks, mediumsized industrial applications in variouscountries and large-scale municipalwastewater treatment plants in China.

Middle EastClean water shortages are the obvious

main driver for MBR applications in theMiddle East, in treatment of both municipal aswell as industrial (petrochemical) wastewater.

EuropeIn Western Europe water reclamation is

not the main driver. In the UK an importantdriver is compactness (e.g. Swanage) and strictdischarge limits due to bathing waterrequirements. In Germany and the Nether-lands important push factors are strictdischarge requirements due to ecologicallysensitive surface waters and the innovativecharacter of the technological developmentsrelated to MBR. In Southern Europe waterreclamation can be considered as the maindriver.

Northern AmericaIn the U.S. and Canada MBR initiatives are

predominantly driven by strict dischargerequirements due to ecologically sensitivesurface waters. At some locations waterreclamation is another important driver.

H2O # 2005

MBR special III

31

MBR applications outside theNetherlands: drivers and casestudies

Berend Reitsma, Tauw

Luc Kox, DHV Water

Cora Uijterlinde, STOWA

Today countries have different drivers for application and development of Membrane BioReactortechnology. Due to changes in legislation, enforcing strict discharge requirements is the main driverfor Northern America, Germany and the Netherlands. In water scarce areas like Israel, SouthernEurope and even Japan, the application of the MBR is driven mainly by the need for waterreclamation. In the UK (case study 1) the effluent quality driver focuses mainly on a high hygienicquality, giving less attention to the removal of nitrogen and phosphate. This is different from theNetherlands where possibilities towards far reaching nutrient removal are the key driver for thedevelopments regarding MBR. In the UK additional drivers are compact and simple facilities, fittinginto the landscape, with zero emission for odour. For the Ben Gurion International Airport (case study2) compactness and superior product water quality are the drivers. The reclaimed water will be used inboth irrigation purposes and industrial applications on the airport.

MBR Lowestoft.

MBR in the United KingdomThere are currently more than 15

operational urban MBRs in the UK. Most ofthese have a capacity of 500-1,500 m3/d. Thetwo largest MBRs each have a capacity of about12,000 m3/d (Swanage and Glasgow).

At the end of 2003, ten Dutch water boardsand five consultancies visited the MBRs inLowestoft, Porlock and Westbury. This casestudy presents the findings of this visit. Theemphasis is laid on describing the drivers foran MBR on these sites.

The Lowestoft WWTP is a demonstrationproject with three different compactwastewater treatment processes. Seventypercent of the wastewater is industrial. TheMBR has a treatment capacity of 46,800 p.e. anda maximum hydraulic flow of 590 m3/h.

The facility was to be fitted into thelandscape with a zero emission for odour. Thetreated wastewater is discharged into theNorth Sea. The effluent requirements arelenient compared to the Netherlands, as longas the water is hygienically safe (bathing waterstandard). The EC standard for non-vulnerablesurface water must be met, which means thatBOD must be reduced to 25 mg/l, and drymatter to 30 mg/l. There are no effluentrequirements for nitrogen and phosphate.

Porlock is a small seaside resort (3,800inhabitants) in Exmoor National Park on thecoast of Somerset. In connection with itsfunction as seaside resort, Wessex Waterconsidered the discharge of untreatedwastewater unacceptable. Determining factorsin choosing the system for the facility were theeffluent quality for viruses and bacteria andthe minimum space demand. Furthermore,the system was cheaper than a long pipelineinto the sea. In 1998, the Porlock MBR facilitywas put into service and has been working tothe full satisfaction of Wessex Water ever since.

Effluent requirements pertain to BOD (40mg/l) and suspended solids (60 mg/l), not tonitrogen and phosphate. It was decided toinstall an MBR in Porlock mainly to achieve agood hygienic effluent quality. Ammonia isremoved to a reasonable extent. Good resultswere presented for the removal of coliformsand bacteriophages (a measure for virusremoval), up to log 6 and log 4 reductions,respectively.

Porlock WWTP was built in a protectedarea near a tourist beach, which is why thestarting points for its design were completelydifferent from those at Lowestoft. The exteriorof the facility was to merge with itsenvironment and the effluent outlet was notto disturb the tourists. The MBR facility issituated close to the town centre near somefarms. Given its location and the tourist

function of the area, it was decided tominimize the MBR facility’s conspicuousnessin the landscape. This starting point yielded anexceptionable design. The facility was placed ina building designed according to the style ofthe surrounding farms. The coarse screenswere built in a basement and the sludgestorage tanks were made as flat as possible.

The Westbury WWTP consists of atreatment plant of trickling filters and arecently constructed MBR parallel with theexisting plant. The existing facility wasextended on account of the connection andextension of a dairy plant. First, the dairywastewater is pretreated with a flotation unitand pH correction. The pretreated wastewateris discharged to the WWTP for furthertreatment. The driver for the application of anMBR is that the high effluent quality of theMBR causes the combined effluent to meet theeffluent standards.

Because the flat sheet membranes needintense aeration to prevent contaminationnormally one tank with membranes is used.The oxygenation for the oxidation is usuallycompletely provided by the coarse bubbleaeration for the membranes. However, onaccount of the specific influent composition tobe expected at Westbury (dairy industry andpercolate), i.e. concentrated wastewater, alarger activated sludge system (four tanks) wasrealized. Only 1/3 is taken up by themembranes (12 sets of 200 flat sheets per tank).The remainder is intensively aerated with finebubble aeration.

The combined effluent must meet therequirements for BOD (13 mg/l), suspendedsolids (20 mg/l) and NH4-N (5 mg/l). For theMBR the effluent requirement for phosphorus

is 1.5 mg Ptotal/l. This requirement is being metby means of chemical phosphate removal. Theguaranteed effluent quality for the MBR is 5mg/l for BOD, suspended solids and ammonia.

Some MBRs in the UK have beenoperational for over five years. The systemsseem to function without problems and theirreversible contamination of the membranesalso seems to be no problem. In the UK MBRsare mainly used to remove organic compounds(BOD) and solids (suspended matter) and toobtain disinfected effluent. The dischargerequirements for permeate are very lenientcompared to Dutch requirements. Often, thereare no requirements for nitrogen or phosphate.Sometimes a requirement for ammonia appliesand even less often one for phosphate.

The available surface area for WWTPs isoften limited and this means that thetreatment process has to take place in acompact facility. This results in a very simplestructure consisting of one aeration tank inwhich also the membranes are placed. Themembrane aeration in the British facilitiesprovides a significant part of the requiredoxygen input. In addition, fitting the facilitiesinto the landscape and the zero emission forodour are important.

In the Netherlands, however, the interestin the MBR is mainly for its extensive capacityof removing nutrients and suspended matter,sometimes even to the maximum allowablerisk requirements for N and P for the receivingsurface water (2.2 and 0.15 mg/l, respectively).This means that the biological process in theNetherlands will be more important than inthe UK, which will require a structure withmembranes in a separate membrane tank.

H2O # 200532

MBR special III

MBR Porlock.

Especially in case of very stringent effluentrequirements, the nitrogen removal must thenlargely take place outside the membrane tanks.As a result, the structure of the aeration tankwill be more complex, with anoxiccompartments, aeration control, recirculationflows, process control, influencing of thesludge volume index and perhaps evenadditional carbon source dosing. Thedependency on measuring instruments (suchas nitrate, ammonia and oxygenmeasurements) plays a much greater role incontrolling the process. In addition, little

attention was paid in the UK to theoptimisation of the energy consumption.

Water reclamation in IsraelIsrael suffers from a lack of sufficient clean

water resources. In recent years the applicationof MBR technology has received a lot ofattention. The drive for this attention isobvious: water reclamation. In the year 2003DHV was involved with the development ofthe plans for the wastewater treatment plantfor Ben Gurion International Airport in TelAviv (Israel).

In some years time Ben GurionInternational Airport will process 16 millionpassengers per annum. The existing airportand facilities will not meet the futurerequirements, therefore it was decided thatnew infrastructure was to be built. After theperformance of an Environmental ImpactAssessment study the Israel Airport Authority,owner of the airport, was forced to applywastewater treatment technologies that wouldenable water reclamation by the IsraeliGovernment. In a feasibility study MBRtechnology was selected, due to the twostrongest advantages MBR technology offersover conventional technologies: compactnessand superior product water quality.

The wastewater to be treated in the MBRat the airport originates from three individualcontributors: the terminal building, foodprocessing by catering facilities and industrialactivities related to the aircraft industry.

The design capacity of the MBRinstallation is 1,200 m3/h and 59,300population equivalents. The required effluentstandards are 10 mg/l for nitrogen and 0.2 mg/lfor phosphorous. In the plant set-up specialattention was paid to the pretreatment of theraw wastewater. Since the feed water maycontain toxic wastes from heavy industrialproduction processes and product waterrequirements are to be met at all time, anequalisation tank is incorporated.

The product water of the MBR plant,reclaimed water, will be used in bothirrigation purposes and industrial applicationson the airport. No additional treatment

H2O # 2005 33

MBR special III

Ben Gurion Airport MBR building.

MBR Westbury.

process is required for this type of reuse. Onlywhen the product water will be used as boilerfeed water application of Reverse Osmosis willbe required.

A special feature for the design of the MBRplant is the architectural design. To createuniformity this needed to comply with theoverall architectural design for all thebuildings on the airport. So, instead of thecommon looks of a water treatment facility:concrete and steel tanks and buildings, thisplant is housed in a modern building. Thebuilding’s exterior is finished with glass wallsand aluminium roofing in the shape of anairplane wing. The latter is influenced by thetop view of the buildings, as seen bypassengers in the airplanes.

When the new airport will be used, waterreclamation will be put to practice by thewastewater treatment concept of the future:MBR technology. This example shows the

strong advantages MBR offers and theenvironmentally sustainable solution it canbring to areas suffering from clean watershortages.

EpilogueThis article intends to give an

introduction to the drivers of the attentionthat MBR technology receives. This positiveattention is growing continuously andsubsequently pushes further development ofthe technology, which enhances thepossibilities for applications in differentsettings.

As indicated, the first applications of MBRin the UK were mainly driven by the demandfor a reliable and compact treatment conceptthat reaches strict discharge requirementsregarding hygienic parameters. However, highlevels of nutrient removal are not driving theseinitiatives. Possibilities towards far reachingnutrient removal are the key driver for the

developments regarding MBR in theNetherlands. This is mainly due to future onchanges in legislation, in which ecologicallysensitive surface waters will be protected bymeans of enforcing strict dischargerequirements.

Obviously the drivers in a water scarce arealike Israel are of a different kind. The superiorwater quality that is produced by MBRinstallation enables water reclamation to beput to practise. Without doubt this will be themajor pushing factor for MBR technology allover the world. The technology can contributeto relieving the current pressure on theenvironment and contribute to sustainabledevelopment of our living surroundings. ¶

Berend Reitsmaprocess engineer TauwP.O. Box 133, 7400 AC Deventerphone: +31 570 69 98 63e-mail: bar@tauw.nl

Luc Koxbusiness development & sales manager DHVWaterP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 22 42e-mail: Luc.Kox@dhv.nl

Cora Uijterlinderesearch manager STOWAP.O. Box 8090, 3503 RB Utrechtphone: +31 30 232 11 99e-mail: uijterlinde@stowa.nl

H2O # 200534

MBR special III

SamenvattingOp het moment zijn er in diverse landen verschillende motieven voor de toepassing enontwikkeling van membraanbioreactoren. In Noord-Amerika, Duitsland en Nederland is devoornaamste drijfveer de strengere lozingseisen die door veranderende wetgeving van krachtzullen worden. In waterschaarse gebieden, zoals Zuid-Europa, het Midden-Oosten en zelfsJapan, is waterhergebruik de voornaamste drijfveer. In Groot-Brittannië (casestudie 1) wordt degoede effluentkwaliteit met name beschouwd met betrekking tot hygiënische kwaliteit en nietten aanzien van de verwijdering van stikstof en fosfaat. Dit wijkt af van de Nederlandsesituatie, waar juist de mogelijkheden voor vergaande stikstof- en fosfaatverwijdering devoornaamste drijfveer vormen. In Groot-Brittannië bestaat daarnaast nog de mogelijkheid vancompacte en eenvoudige installaties die goed landschappelijk zijn in te passen met eenminimum aan geuremissie, bepalend. Bij het internationale vliegveld Ben Gurion (casestudie2) ligt de motivatie in de compactheid en de goede effluentkwaliteit. Het opgewerkte water zalzowel worden toegepast voor irrigatie als voor industriële toepassingen.

Nordkanal, the world largest MBR (Erftverband).

Do you jump just because others do?

Tauw bv,

P.O. Box 133

NL-7400 AC Deventer

phone: +31-570-699111

fax: +31-570-699666

email: info@tauw.nl

www.tauw.com

Tauw consultants and engineers, intelligent about your options.

In 2000 and 2001 large-scale pilot researchat WWTP Beverwijk was carried out under thesupervision and coordination of DHV,commissioned initially by the Water boardHollands Noorderkwartier, and followedthrough by the STOWA. The goal of the pilotresearch was to confirm the technicalfeasibility, to further develop the technology,to eliminate uncertainties, and finally tocompare the MBR-technology with theconventional activated sludge technology.

In co-operation with four membranesuppliers (Kubota, Mitsubishi, X-flow &Zenon) various MBR pilot systems with acapacity up to 10 m3/h were commissioned. Animportant aspect of the first phase of the pilotresearch was to integrate the knowledge ofmembrane technology and the activatedsludge process. From 2002 until mid 2004 theresearch at WWTP Beverwijk was extendedwith various other membrane suppliers(Memfis, Seghers-Keppel, & Huber-VRM).

In 2002 other initiatives were taken to givea better insight into the first phase of the MBRdevelopment programme. In co-operationbetween Water board Rivierenland and RoyalHaskoning, a Zenon MBR pilot installationwith a maximum hydraulic capacity of 20 m3/hwas commissioned in April 2002 at WWTPMaasbommel. This research was primarily

directed towards the feasibility of MTRquality, and a comparison was made withclassical secondary effluent sand filtration. Theresearch received a participation allowancefrom the STOWA and traversed a two-yearduration. Co-operation with the Beverwijkresearch team occurred regularly with an indepth evaluation of membrane fouling andcleaning.

The end of 2002 saw the start-up of aKubota MBR pilot with a maximum hydrauliccapacity of 5 m3/h on WWTP Hilversum undersupervision of the Water Authority DWR. The

research was also directed primarily to thefeasibility of MTR quality. Much effort hasbeen directed to the design and automation ofthe pilot so that a better insight could be maderegarding the N- and P-removal. This researchalso received a participation allowance fromthe STOWA.

All knowledge and experience from thethree pilot research programmes was broughttogether in the Dutch MBR-committee, whichwas organised via the STOWA. Furthermore, inan effort to disseminate the knowledge, anMBR website has been opened in co-operationwith the STOWA at Waterforum Online.

MBR BeverwijkBeverwijk stands synonymous for MBR

since the Water board Hollands Noorder-kwartier, STOWA and DHV began the large-scale pilot research in 2000. In a short time, thefour pilots of Kubota, Mitsubishi, X-Flow andZenon were commissioned and more than twoyears of broad research carried out. This wasreported in 2002 via STOWA and via the IWA1)

for the worldwide audience. However, theMBR research was not finished and remainedactive till mid 2004. Other MBR pilots, fromMemfis, Seghers-Keppel and Huber-VRM weretested alongside some of the original four MBRsystems2).

The foundation blocks of the MBRknowledge were born out of twelve pilotconfigurations from seven membranesuppliers, over a period from March 2000 toJuly 2004, as summarised in table 1.

H2O # 200536

Dutch MBR development:reminiscing the past five years

Darren Lawrence, DHV Water BV

Chris Ruiken, DWR

Dennis Piron, Water board Rivierenland

Ferdinand Kiestra, Royal Haskoning

Ruud Schemen, Water board Hollands Noorderkwartier

The MBR research that has taken place at WWTP Beverwijk since 2000 has yielded positive insightsinto the applicability of the MBR technology for the specific character of Dutch municipalwastewater. The research has been innovative and has proven inspirational to the development of theMBR technology worldwide. The research has fundamentally revolutionised the technology,particularly in the areas of chemical cleaning and process control and automation. In 2002 otherinitiatives came to light via the WWTPs Hilversum and Maasbommel studies, where the pilotresearch programmes were specifically aimed at MTR effluent quality. Through Beverwijk,Hilversum and Maasbommel, the first phase of the Dutch MBR development programme has beenfulfilled.

MBR special III

system speciality type pore surface design research size area capacity period(µm) (m2) (m3/h)

Zenon ZW500a-c-d hollow fibre 0.035 184-60-95 8 03/2000-10/2003Kubota SD / DD flat sheet 0.4 240 10 05/2000-07/2002Mitsubishi 3-layer hollow fibre 0.4 314 7 05/2000-03/2002X-Flow AirFlush tubular 0.03 220 9 05/2000-04/2002Memfis MTR flat sheet 0.035 112 5 05/2002-06/2003Toray DD hollow fibre 0.08 137 5 02/2003-02/2004Huber E rotated

flat sheet 0.035 360 15 10/2003-07/2004

Table 1: Overview of the Beverwijk pilot research project.

In 2000 and 2001 the research stood in therealm of development and applicability to thespecific Dutch wastewater characteristics, i.e.low process temperatures and variable flowconditions3). The continuous comparisonbetween the four systems made very efficientresearch possible, where the process operationof the MBR systems could be improved in avery short timespan. The research in 2002 and2003 directed more to the optimisation of thechemical cleaning of the membranes, and onseveral pilots various techniques wereintensively tested for better cleaningmethodologies.

Feed conditioningFrom the seven MBR suppliers mentioned

two suggested that extensive feed screeningwas overdone and could be carried out on amore cost effective base with far less stringentfinal filtration of the nominal 0.8 mm punchedholes. Eventually, both pilots were refittedwith slightly more extreme final filtration,this was due to the fact that the membraneswere relatively free of debris, but the auxiliarymembrane equipment was prone tocontamination, e.g., the aeration system,module sides and distribution points.Eventually all MBR systems tested at theBeverwijk were fitted with some form of finalfeed stream filtration between 0.8 mm to 1 mmpunched holes. Wedge wire and square meshwas tested on several occasions but proved lessefficient at hair removal than the selectedpunched hole screens. Due to the extrameasures required to condition the feedstream to an MBR innovative new ideas havecome to light to achieve the required MBR feedstream quality in one simple step.

Biological conditioningIn Holland we are conscious of the fact

that the biology and not the membrane doesthe work in the MBR. The membrane is asimple reliable tool able to achieve a solids freeeffluent and nothing more. The goal hastherefore been set to condition the MBR sludgein such a way that the membrane only seeswater and predominantly inert suspendedsolids. The latter promotes low trans-membrane pressures and high sustainablepermeability.

But how do we achieve the optimalbiology? This has been a study item ever sinceaerobic biological treatment systems were firstenvisaged, the rules of thumb that apply to anoperationally perfect conventional treatmentworks also applies to the MBR, only the speedof events occurs three times faster due to thelower hydraulic retention time. The MBRknowledge base is present, but is oftenoverlooked as the technology has beendominated by the MBR suppliers, who, by trialand error, have generated viable marketableproducts. Many of these products differ fromwhat would be considered as a ‘normal’biological solution - the membranes havedictated the configuration rather than thebiological configuration dictating where andhow the membranes should be utilised.

Most pilots at Beverwijk where designedfor a total nitrogen of < 10 mg N/l with asimple biological process, as the knowledgebase increased the discharge levels were forceddown to < 5 mg N/l and the biologicalconfigurations and automation increased incomplexity. The step to MTR quality required

advanced biological treatment in the form ofRacetrack bioreactors, extended plug-flowdesign, or plug-flow and racetrack incombination; the latter are displayed in thepilots of Varsseveld, Maasbommel andHilversum respectively. Key similarities of allthe pilots are the energy input devices. Whereenergy is directly put into the sludge viamixing, pumping or aeration, the device issuch that the sludge flock structure is leastmechanically affected. Low energy inputrelates to better sludge quality and bettersludge characterisation, and ultimately higheralpha-factors and better membraneperformance. As the configurations becomemore complex, the dissolved oxygen profilebecomes more critical throughout thebiological reactor. This is a major area of R&Dand is detrimentally affected by the air inputvia the membranes.

Membrane cleaningIn the beginning of the research, the

cleaning methodology of that time wasapplied; this allowed the membranes to foul toa certain point before a recovery clean wasnecessary. This technique was deemed a largerisk to full-scale installations, as at the pointwhen the full hydraulic capacity was necessary,for instance during RWF, the fouled membranecapacity would be insufficient to treat therequired throughput. This required a newmembrane cleaning philosophy. The solutionwas simple - don’t let the membranes foul. Formost membrane types, this standpoint yieldeda new cleaning methodology based on‘Maintenance Cleaning in Air’.

This MC in Air is carried out once a week ortwo weeks with considerably less chemicalscompared to the classical recovery cleaningtechniques. Overall the MC in Air procedureshas lead to a more stable process operation.Further optimisation of the procedures atBeverwijk yielded even better results withintermittent MC in Air back flushing withwarm water/permeate, by some 10 to 15°Cabove the normal process temperature.

The year 2002 also saw the beginning ofadvanced automation of several pilots, both forthe membranes and the process control, thelatter was in foresight of larger practicalinstallations. The dynamics of the MBR systemvaries tremendously from that of aconventional installation and little was knownabout the performance of high-techmeasurement devices in the higher sludgeconcentrations. In co-operation withEndress+Hauser, Dr. Lange and Danfossvarious measurement devices were tested forreliability, reproducibility and accuracy andnumerous processes were automated.

H2O # 2005 37

MBR special III

The technical advisory committee of the Beverwijk research project.

MBR HilversumAt the Hilversum STP, an MBR pilot was

envisaged to help the design process for thefull-scale installation Hilversum, and severalgoals were specified. Firstly, to see if the pilotcould generate knowledge directly related toan improved full-scale design, to establish ifthe required effluent discharge criteria of MTRfor nitrogen and phosphorus could be achievedand to give an idea of the chemicals needed,and finally, to gain the practical experience ofrunning an MBR for at that time an unknowntechnique within DWR.

In essence the pilot system is as follows. AHuber pretreatment on raw influent via a 0.5mm fine screen, a plug flow biology and asludge water separation via a separate Kubotamembrane filtration tank. The company Solissupplied the membrane system with 150 flatsheets with a surface area of 0.8 m2 and amaximum RWF of 5 m3/h.

The pilot installation was commissionedin November 2002 and is expected to remain inoperation on location WWTP Hilversum untilJuly 2007. Unfortunately, start-up problemsforced a rebuild in 2003, and by the end of 2004the system had undergone further rebuilds tooptimize to smaller membrane tanks. Thecoming months will see changes in thecleaning of the fine-screens and a methanoldosing system will be installed. Eventually, asmall man-made lake will be installed after thesystem to further investigate the effects ofpermeate on surface waters, in relation to algaegrowth and ground water infiltrationcharacteristics.

Reflecting on the results so far, a numberof items spring out. The pretreatment withfine screens is most problematic. Blockagethrough fat, hair and toilet paper causes the

necessity for intensive and frequent cleaning.The future design takes into account thesefactors and is foreseen with hot water and highpressure cleaning facilities. An interesting factis that the screenings (paper), makes upapproximately 25% of the sludge productionthat can be further dewatered to some 35% DS.

The achievement of MTR quality effluent(TN < 2.2 mg N/l and TP < 0.15 mg P/l) is noteasy, even with sodium acetate dosing thedischarge criteria remained difficult. At thesame time it was seen that due to the release ofbiologically bound P in the post denitrificationit was not possible to achieve the P dischargecriteria, but is believed with the future use ofmethanol as C-source for denitrification the Prelease shall no longer occur. A complicatedfactor for MTR discharge is the presences ofhumic acid in permeate and the relevantbound phosphorus and nitrogen. For thisreason alone is the MTR discharge criterionalmost impossible to reach.

One item that stood out was the effortlessfunctionality of the Kubota membrane system,this has built up trust within the DWRorganisation for membranes and has increasedthe knowledge of the operational aspects ofmembrane technology.

The sludge production of the pilotinstallation, including the ‘paper production’was much lower than in conventional systems.It is believed that due to the relatively largemembrane tank with continuous aeration,mineralisation of the sludge occurred. Thereduction in the membrane tank size from 12m3 to 3 m3 at the end of 2004 confirmed thisobservation. The sludge production withoutchemical addition rose considerably andproved advantageous for the phosphorusremoval and for the formation of humic acid.

Through the decrease in size of themembrane tank the Kubota principle ofcirculation of sludge in the membrane tankhas been abandoned. Sludge and air, requiredfor the dynamic scouring of the membranehowever, remains turbulent as in the oldsituation.

MBR MaasbommelIn 2002 the Water Board Rivierenland,

together with Royal Haskoning and STOWA,started a two-year research project concerningthe applicability of MBR and continuous sandfiltration for the treatment of municipalwastewater4).

On the wastewater treatment plantMaasbommel, a MBR with an organic loadingof 650 p.e. (136 g TOD) and a hydraulic capacityof 20 m3/h was operated parallel to the existing

WWTP. Two full-scale continuous sand filterswere installed downstream of the secondaryclarifiers to polish the effluent of the existingWWTP at a total hydraulic capacity of 120m3/h. The performance of the two alternativeswas compared on the following aspects.

MTR qualityThe results of the research project

concluded that for both configurations it wasdifficult to comply with the yearly averageMTR effluent standards for nitrogen andphosphate5). During conditions of RWF, thecontact time of the MBR was reduced andprocess conditions deteriorated (especially fordenitrification). As for the removal of heavymetals, compared to conventional treatmentboth configurations displayed no advantage.

The same applied to the removal ofherbicides and pesticides. For the purpose ofdesinfection (as determined using the bacteriallevel and E. coli) performance of the MBR wassuperior. Based on the chemical analyses ofseveral endocrine disruptors, the conventionalsystem with sand filters and the MBR yieldedcomparable effluent qualities. Both systemsachieved removal efficiencies of 95% forbisfenol A, estron and ß-oestradiol. Theoestrogene potential of the effluent(determined using a bioassay) was about 60%lower for the MBR as compared to the sandfilter effluent.

PerformanceTechnical process stability was comparable

for the two alternatives. Achieving a stableeffluent quality for nitrogen and phosphatewas easier for the sand filters than for theMBR. Especially under rainwater flowconditions the effluent quality of the sandfilters was stable while that of the MBR is morevulnerable to influent flow variations.Optimisation of the process configuration andprocess control would lead to improvements inthis respect. When membrane cleaning wascompletely automated, required operatorattention was equal for both configurations.Another problematic aspect was the processmeasurement and control in lowconcentration ranges around the MTR effluentstandards. Current instrumentation is tooinaccurate in these ranges to be reliably used asinput for process control. For future designpurposes this will require additional attention.

EpilogueLooking back at the last five years we must

assess our current position and consolidate theknowledge we have acquired. ‘MBR Beverwijk’as a trigger still lingers on in the thoughts ofmany foreign and domestic end users, and theDutch contribution to the successful

H2O # 200538

MBR special III

The new membrane tank at Hilversum (photo: Solis).

development of the MBR-technology has beenwidely recognised and applauded.

The MBR hype associated with Beverwijkhas found a solid form in the realisation of thedemonstration plant MBR Varsseveld; manyaspects of numerous Beverwijk pilots havebeen combined into this demonstrationsystem. The possibility to interchangemembranes has been addressed, removal ofcassettes has been eliminated, cleaningprocedures have been made totally flexible andintegrated, the biological system has been finetuned to the membrane configuration, and thepre-treatment has been exhaustive and final inthe removal of unwanted debris able to hinderthe performance of the membranes.Understandably, all these items are beingaddressed in the research and developmentprogramme.

At the time of writing this article MBRVarsseveld has been in operation for some fourmonths. Already the fruits of the MBRBeverwijk experience are being harvested: theflexibility in the chemical cleaning procedurewas essential as the wastewater feed fromVarsseveld contains substantially more fat

than at Beverwijk, the biological configurationis already yielding MTR values for totalnitrogen, and total phosphorus will follow.The pre-treatment has been intensively testedto yield better quality influent suitable for themembranes, here, off the shelf technologieshave proved inadequate and the devices haverequired serious modification to achieve adebris free feed stream. The membranes havebeen made maintenance friendly with easyinspection and overall accessibility.

From the pilot research project describedin this article, much knowledge has beengained. Knowledge however, only becomes‘real’ once it has been proven under variousconditions, on several systems, and lastly hasbeen scaled up to a viable full-scaleinstallation. The lessons of the first phaseexperiences will have to be tested in full-scale,with Varsseveld the next phase has beenentered. ¶

Darren Lawrenceprocess engineer DHV Water BVP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 22 23e-mail: darren.lawrence@dhv.nl

Chris Ruikenprocess engineer Dienst Waterbeheer enRioleringP.O. Box 94370, 1090 GJ Amsterdamphone: +31 20 460 28 33e-mail: chris.ruiken@dwr.nl

Dennis Pironprocess engineer Water board Rivierenland P.O. Box 599, 4000 AN Tiel phone: +31 344 64 95 04e-mail: D.Piron@wsrl.nl

Ferdinand Kiestraprocess engineer Royal HaskoningP.O. Box 151, 6500 AD Nijmegenphone: +31 24 328 42 84e-mail: f.kiestra@royalhaskoning.com

Ruud Schemenproject manager Water board HollandsNoorderkwartierP.O. Box 850, 1440 AW Purmerendphone: +31 299 39 14 89e-mail: r.schemen@hhnk.nl

Literature

1) Roest van der H., D. Lawrence and A. van Bentem (2002).

Membrane bioreactors for municipal wastewater

treatment. Water and wastewater practitioner series:

STOWA report 2002. ISBN 184339.011.6.

2) Verkuijlen J., M. Kraan, D. Lawrence and A. van Bentem

(2003). MBR research Beverwijk, back to basics. H2O MBR

special II, April 2003, pp. 7-9.

3) Lawrence D., A. van Bentem, P. van der Herberg and P.

Roeleveld (2001). MBR pilot research at the Beverwijk

WWTP, a step forward to full scale implementation. H2O

MBR special, October 2001, pp. 11-15.

4) Piron D. and F. Kiestra (2001). Feasibility of MAR-quality

at the WWTP Maasbommel. H2O MBR special, October

2001, pp. 46-47.

5) Kiestra F., D. Piron, J. Segers and J. Kruit (2004).

Vergelijkend onderzoek MBR en zandfiltratie RWZI

Maasbommel. STOWA report 2004-28.

H2O # 2005 39

MBR special III

SamenvattingHet MBR-onderzoek dat sinds 2000 plaatsvond op de rioolwaterzuivering Beverwijk, heeft totpositieve inzichten geleid over de toepasbaarheid van de MBR-technologie voor hetNederlandse afvalwater. Het onderzoek stimuleerde de ontwikkeling van de MBR-technologiewereldwijd. Het onderzoek leidde bovendien tot fundamentele wijzigingen in deprocesvoering van een mebraanbioreactor, in het bijzonder op het gebied van chemischereiniging en procesbesturing en automatisering. In 2002 zijn pilotonderzoeken gestart op derioolwaterzuiveringen Hilversum en Maasbommel. Hierbij was de aandacht met name gerichtop de haalbaarheid van de MTR-norm voor het effluent. Met het onderzoek op de rwzi’s vanBeverwijk, Hilversum en Maasbommel is hiermee de eerste fase van de ontwikkeling van demembraanbioreactor in Nederland afgerond.

Module types of Kubota, Mitsubishi, X-Flow and Zenon.

Bea

Geplaatste afbeelding

After five years of pilot research at differentlocations in the Netherlands, the second phaseof MBR development has begun. This secondphase projection of three demonstrationinstallations shall shed light on the technicalaspects of the scale up and on the achievableeffluent quality of MBR plants. The firstoperational years of all three MBRs shall bededicated to broad based research programmes,which will address all aspects of the MBRtechnique: pretreatment, biological treatment,and membrane filtration. The projects aresupported by STOWA via the innovation fund.

Through these demonstration installations andthe research programmes, much knowledgeand experience will be gathered which can berelayed to future systems with larger capacities.The first demonstration installation was startedup at the end of 2004 at Varsseveld WWTP. TheVarsseveld MBR is a full-scale application inwhich all wastewater is treated with this newtechnology. Two other demonstration plants, atOotmarsum WWTP and Heenvliet WWTP, willstart up in 2006, and will be hybrid installationswhere the MBR is integrated in a conventionalactivated sludge system.

MBR technology has great potential in theNetherlands, where almost all wastewatertreatment plants are activated sludge based,where space is often limited and where thequality of the local surface waters must beimproved. The latter was confirmed throughan MBR market research programmecommissioned via STOWA1). The mostimportant driver for the applicability of theMBR technology was the effluent quality. Inregard to the effluent guidelines with respectto nitrogen en phosphorus, the treatmentefficiency for these substances is an importantresearch item. The effluent requirements forthe three MBRs are listed in table 1.

The results of a STOWA pilot researchproject at Maasbommel WWTP 2) showed thata stable MTR quality (Ntotal < 2.2 mg/l, Ptotal

< 0.15 mg/l) is difficult to achieve, with bothMBR and effluent filtration. The full-scaleresearch at the demonstration plants will givemore insight in the effluent quality that can beactually achieved.

Membrane typeIn the past decade the number of

membrane suppliers and membrane types hasincreased drastically. As a result of this, thecompetition between the membrane suppliersis tough. Each membrane type has its ownspecific characteristics, which may or may notbe beneficial for the application in municipalwastewater treatment. For the future of MBRthe competition is good as it inspires themembrane suppliers to produce the bestmembranes at the lowest costs. In othercountries it is shown that if one suppliercontrols the market, like Kubota in Englandand Zenon in Germany, prices remain highand the market penetration of MBR remainslimited. In the Netherlands this situation doesnot occur as three water boards have selectedthree fundamentally different membranetypes (hollow fibres, flat sheets and tubes), asshown in table 2. In this way experience will begained with all types of membranes, and allpaths for future development remain open.

Compared to other countries, the applieddesign fluxes are relatively high. In Germanythe advised net flux rate is approximately 25l/(m2.h). The higher flux rates in theNetherlands are based on the positiveexperiences of the pilot research. As the

H2O # 2005

MBR special III

41

Dutch MBR grows to maturityAndré van Bentem, DHV Water BV

Coert Petri, Water board Rijn en IJssel

Jan Willem Mulder, Water board Hollandse Delta

Herman Evenblij, Witteveen+Bos

Dick de Vente, Water board Regge en Dinkel

Bert Geraats, Grontmij

The MBR development in the Netherlands enters the next phase after five years of pilot investigation.MBRs will be constructed at three wastewater treatment plants. The first, in Varsseveld, treats thetotal influent flow of this town, and has been in operation since December 2004. The MBRs inOotmarsum and Heenvliet will be hybrid systems that are integrated into the existing conventionaltreatment plants. At all three plants the effluent quality is a major criterion. Extensive researchprogrammes, under supervision of STOWA, will shed light on the scale up effects, the achievableeffluent quality and the operational aspects of MBR. The three water boards involved have chosen forthree fundamentally different membrane types (hollow fibre, flat sheets and tubular membranes)that maintains open MBR development in all directions. After the first phase of Beverwijk,Hilversum and Maasbommel, during which all aspects of the membrane and their interactions wereinvestigated, the Dutch MBR has now grown to maturity.

Table 1: Effluent requirements and targets.

parameter unit Varsseveld3) Ootmarsum Heenvlietrequired target required target required target

year - 2015 2005 2018 2006 2010season* - S W S W A S W S WBOD5 mg/l 10 5 2 - -NH4-N mg/l - 1 2 0.8 1 - 1 -Ntotal mg/l 10 5 10 4 5 2.2Ptotal mg/l 1 0.15 1 0.15 0.3 0.15SS mg/l 10 5 5 2 - -E-coli -/ml - - - 20 20ecology - - - + - - * S = summer, W = winter, A = all year

RWF/DWF ratio is relatively high, the averageflux rate of the membranes is much lower.

Complete versus hybrid MBRThe MBR at the Varsseveld WWTP is a

complete system, in which the existingconventional WWTP is taken out of operationand all the wastewater is treated in the MBR.In general, the complete MBR configurationwill be applied in the case of very strict effluentcriteria, for instance at the Hilversum WWTP,or at locations where space is limited orvaluable. Based on the current projects and theresults of the STOWA market research, it isexpected that in most cases MBR will beapplied for the extension of existing treatmentplants. In such cases the application of a hybridconfiguration is a serious option. In a hybridconfiguration the MBR is connected parallel toor even in series (DWF) with the existingconventional WWTP. The hydraulic capacity ofthe MBR is limited, during DWF all (Heenvliet)or most of (Ootmarsum) the wastewater istreated in the MBR. During RWF the surplusof rainwater is treated in the conventionalactivated sludge system with secondaryclarification. In this manner the requiredmembrane surface can be reduced compared toa complete MBR installation and themembranes can be continuously used at anoptimum level. The costs of such a hybridinstallation can be lower whereas the effluentwill have the desired quality to comply withthe guidelines, most of the time.

The hybrid MBR system can be applied inseveral configurations but none of them havebeen tested so far. For two locations, Heenvliet(Water Board Hollandse Delta) andOotmarsum (Water Board Regge en Dinkel),the hybrid configuration has been chosen asthe best solution. In 2005 the construction ofboth installations will commence and in thedesign and research phase, a close corporationbetween both water boards has beenestablished which will lead to better designconsiderations and optimal performance. Thehybrid MBR plant at Heenvliet andOotmarsum will, just like the Varsseveld MBR,be a role model for the Dutch MBRdevelopment. The two installations will haveto prove the applicability of the hybrid conceptfor the Dutch circumstances.

Two different ways of operation of theHybrid MBR are possible (figure 1);• Parallel hybrid operation, the influent is

always distributed over both systems. Thedistribution ratio and the sludge loadingof both systems can vary, depending onthe flow conditions. At Ootmarsum abuffer tank is used to split the hydraulicand biological loading during RWF. At

Heenvliet the conditions are equal duringparallel operation, which enables a directcomparison between the performance ofthe conventional system and the MBR;

• In series hybrid configuration, thepreferred option at Heenvliet. Themembrane filtration step can be optimisedwithout the risk of overloading theconventional plant during storm weather.This provides the opportunity to upgradethe plant with a minimum loss of overalleffluent quality.

ResearchFor a successful introduction of MBR in

the Netherlands, and to improve theknowledge and performance of MBR systemsin different configurations and set-ups, thetechnical aspects of scale up are of greatimportance. Initiated by the STOWAinnovation fund, the start-up and operation ofthe three MBRs will be accompanied byextensive research programmes. Also Europe

supports the Dutch MBR research activities.For the Varsseveld project, a subsidy is receivedfrom the LIFE programme. For the Heenvlietproject the European commission is requestedfor a subsidy for the EUROMBRA researchprogramme as a specific targeted researchproject (STREP). The Ootmarsum project issubsidised by the European Interreg IIIb-programme (UWC project).

MBR VarsseveldThe main goals of the research programme

at the Varsseveld MBR are to demonstrate thetechnical feasibility of the MBR scale-up and toevaluate the design. The following study itemswere selected:- pretreatment: effectiveness and

redundancy- general aspects: overall performance and

operational requirements- MTR effluent quality: heavy metals, micro

pollutants, hormones and hygienic quality- oxygen input and hydraulics of both the

H2O # 200542

MBR special III

Table 2: Membrane type and design specifications.

parameter unit Varsseveld4) Ootmarsum Heenvliet

type - complete hybrid hybrid biological capacity MBR p.e. 23,150 7,000 3,300 average daily flow MBR m3/d 5,000 1,400 2,400maximum flow at RWF m3/h 755 150 100average flow at DWF m3/h 250-300 75 50-100 membrane type - hollow fibre tubular flat sheetmembrane supplier - Zenon Norit/X-Flow Toray/SKGproduct specification - ZW500d LPCF unibranepore size µ m 0.035 0.03 0.08 design net flux at RWF l/(m2.h) 37.5 54 24.3maximum net flux at RWF l/(m2.h) 50 54 45average net flux at DWF l/(m2.h) 12-15 26-40 12.2-24.3membrane surface installed m2 20,160 2,784 4,110number of lanes - 4 6 2

Figure 1: Two hybrid configurations during dry weather flow and storm weather flow.

aeration tank and the membrane tanks- sludge quality, its relation with the

influent characteristics and its effect onthe membrane performance

- process control: an evaluation andoptimisation based on dynamicsimulation

- membrane filtration: optimisation of theoperation and cleaning.

For each study item a separate projectteam has been formed, in which, besides WRIJ,STOWA and DHV, some of the top researchinstitutes in the Netherlands participate,which are TNO, Wetsus, Delft University ofTechnology and BRCC.

Pilot plantThe worldwide experience with the MBR

technology has shown that the use of a pilotplant can significantly speed up and simplifythe start-up of a full-scale MBR installation. Itwas therefore decided that a pilotplant shouldbe installed at the Varsseveld WWTP beforeand during the startup period of this plant.The pilot plant is a reflection of the full-scaleinstallation, with a scaling down factor ofapproximately 200. The most importantadvantage of the pilot plant was the fact thatthe operating software was a copy of that fromthe full-scale installation. This meant that allsoftware problems and other control itemswere dealt with in the pilot plant. The pilotplant was commissioned in May 2004, anduntil the startup of the full-scale plant theoperators MBR knowledge and skills wererefreshed and deepened. This improved thestart up and minimized the risk ofmismanagement of the full-scale plant5).Besides that, the use of the pilot plantincreased the efficiency of the research

activities, especially regarding the membraneoperation and cleaning.

Energy savingsMembrane development occurs quickly

and new module configurations are designedwith energy consumption in mind. Moresurface area is built into a smaller footprintand the modules can be sequentially aerated.Larger systems, like Varsseveld, containmultiple trains. In the train that is notrequired for operation the air can be drasticallyreduced and actually stopped for long periodsof time. All these options are being testedintensively in the pilot plant and the full-scaleinstallation at Varsseveld. From the pilot planttests it can be concluded that the aerationcapacity in the membrane tank can, underspecific circumstances, be reduced by 40%. Theresearch on energy savings will be continuedin the full-scale plant and will focus on themembrane aeration in both process andrelaxation mode.

Membrane tank symmetryAn important design aspect of the

Varsseveld MBR is associated with the typicalDutch condition of much rain that istransported in the sewers, here, the maximumhydraulic loading of the MBR is some threetimes larger than the average flow. The designnet flux of the membranes is 37.5 l/(m2.h) atmaximum influent flow. To be able tomaintain a good functionality of themembranes it is important to achievesymmetrical loading of the membranes.Therefore special attention has been giventowards the design of the sludge supply anddischarge to and from the membranes. Theperformance of this ‘symmetric’ membranetank design will be investigated in more detailand will include the following items:- the hydraulic profile in the membrane

tank,- the distribution of the cleaning chemicals

in the membrane tanks, modules andcassettes,

- the filtration balance: the permeabilityvariations in all membrane modules intime,

- the oxygen input in the membrane tankunder different circumstances.

MBR OotmarsumThe Ootmarsum MBR is a hybrid

configuration that can be operated in parallelmode, as illustrated in figure 1. The existingconventional activated sludge system will below-loaded and extended with a sand filtrationstep to improve the effluent quality. Theeffluent of both plants will be discharged viaan ecological filter system. The typical designfeature of the whole system is that the flow is

distributed in different proportions underDWF and RWF conditions and that an influentbuffer tank is used to equalize the biologicalloading of both systems. At RWF the buffertank will be working as sort of primaryclarifier of which the overflow is treated in theconventional system and the underflow in theMBR. Due to this configuration, specialattention will be required for the largehydraulic variations on the conventionalsystem.

The ecological filter (or ecologicalactivation system) uses water and marshplants as an environment in which an aquaticecosystem will develop (with zooplankton,phytoplankton, macro fauna, fish, birds,amphibians and insects). This shouldtransform the ‘sterile’ effluent into ‘living’water with equal characteristics as thereceiving surface water.

The main objective of the researchprogramme is to gain practical experiencewith, and optimise the operation of, the hybridMBR system. The achievement of the MBRsystem will be compared with theconventional system with sand filtration. Afterstartup when the plant is in stable operation,the following research objectives are defined:- comparison of the effluent quality of both

systems,- investigation of the synergy effects

between the sand filtration, the MBR andthe ‘ecological filter’, to achieve an optimaloverall performance,

- the development of a universal operationalconcept for the (hybrid) MBR,

- monitoring the effect of the ecologicalfilter, as switch between the WWTP andthe receiving water.

The Water Board Regge en Dinkel andNorit Process Technology, the membranesupplier for Ootmarsum, intend to carry out apilot research programme, prior to the startupof the full-scale plant. The main goals are tospeed up, simplify and improve the startupprocedure of the plant, especially regarding thecontrol system.

MBR HeenvlietThe Heenvliet MBR is a hybrid

configuration that can be operated both inseries and parallel mode, as illustrated infigure 1. The pretreatment of the MBR is a 3mm micro screen with perforated holes. It ispossible to adjust this 2 mm, if necessary. Thedesign philosophy was to keep it simple butflexible.

The Heenvliet MBR research programmewill focus on the achievable effluent quality forthe MBR as a sole technology as well as in

H2O # 2005 43

MBR special III

The symmetric membrane tank concept requires an inflow

pipe (bottom right) and an outflow gutter (top left) over the

full tank length.

series Hybrid MBR operation. Besides that, theoperational aspects (process stability andmaintenance), the minimisation of operationalcost and the maximum achievable flux will beinvestigated. Because of the unique activatedsludge configuration during the in seriesHybrid MBR operation special attention willbe given to sludge filtration characteristics andsludge morphology.

A three-year research programme isforeseen; - first year, parallel, non-hybrid

configuration: to compare the effluentquality of the conventional plant and theMBR;

- second year, in series configuration: toinvestigate the feasibility of this system;

- third year: parallel, non-hybridconfiguration: the load of the MBR as asole technology will be reduced in order toachieve effluent concentrations as low aspossible, with MTR values as target.

EpilogueThe development of MBR technology in

the Netherlands has reached a crucial stage.After a decade of laboratory and pilot researchthe first full-scale installations are coming intobeing. In the coming years the essential

experience and required knowledge to judgethe MBR on its merits, will be acquired. TheVarsseveld MBR will have to demonstrate thefeasibility of the scaling up of the applicationunder typical Dutch flow conditions. Thehybrid MBR configuration is an obviousextension to the Beverwijk researchprogramme and the Varsseveld demonstrationplant. The Heenvliet and Ootmarsum MBRswill have to show that MBR can be a seriousalternative for the extension of WWTP’s whereeffluent quality has priority and one seeks anoptimum between costs and performance.

The research programmes are essential forthe MBR development and are supported andsupervised by STOWA and numerous WaterBoards. The national co-operation, supervisedby the STOWA, leads to an increased nestlingand acceptation of this new technology.Learning by doing, and learning from eachother is the basis of success.

With the Hilversum MBR, the third phasewill already be entered in the very near future.This installation is characterized by itsuniversal membrane tank design, in which allavailable membrane types can be applied. Thesecond phase of MBR development suppliesthe required experience on all types available;

hollow fibres, flat sheets and tubularmembranes. The Dutch MBR developmentstrategy works out well and promises aprosperous future for the MBR. ¶

André van Bentemprocess engineer DHV Water BVP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 24 12e-mail: andre.vanbentem@dhv.nl

Coert Petriprocess engineer Water board Rijn en IJsselP.O. Box 148, 7000 AC Doetinchemphone: +31 314 36 95 22e-mail: c.petri@wrij.nl

Jan Willem Muldersenior policy advisor Water board HollandseDeltaP.O. Box 469, 3300 AL Dordrechtphone: +31 78 633 16 68e-mail: j.mulder@wshd.nl

Herman Evenblijprocess engineer Witteveen+BosP.O. Box 233, 7400 AE Deventerphone: +31 570 69 79 11e-mail: H.Evenblij@witbo.nl

Dick de Venteprocess engineer Water board Regge en DinkelP.O. Box 5006. 7600 GA Almelophone: +31 546 83 25 25e-mail: d.devente@wrd.nl

Bert Geraatssenior consultant Grontmij Infrastructure andEnvironmentP.O. Box 203, 3730 AE De Biltphone: +31 30 220 79 11 e-mail: Bert.Geraats@grontmij.nl

Literature

1) STOWA (2002). MBR marktonderzoek. Report 2002-13 (in

Dutch).

2) STOWA (2004). Vergelijkend onderzoek MBR en

zandfiltratie rwzi Maasbommel. Report 2004-28 (in

Dutch).

3) Leunk J., C. Roos and P. Schyns (2001). First Dutch full

scale MBR application on WWTP Varsseveld. H2O MBR

special, October 2001, pp. 44-45.

4) Schyns P., C. Petri, A. van Bentem and L. Kox (2003). MBR

Varsseveld, a demonstration of progression. H2O MBR

special II, April 2003, pp. 10-12.

5) Schyns P., D. Lawrence and F. Durieux (2005). De

voorspoedige opstart van de MBR-installatie in

Varsseveld. H2O nr. 5, pp. 27-29 (in Dutch)

H2O # 200544

MBR special III

SamenvattingDe MBR-ontwikkeling in Nederland is na vijf jaar pilotonderzoek toe aan de volgende fase. Opdrie rioolwaterzuiveringsinstallaties worden MBR-installaties gebouwd. De eerste installatie,in Varsseveld, behandelt de gehele influentstroom en is sinds december 2004 in bedrijf. DeMBRs in Ootmarsum en Heenvliet zullen als hybride systemen, geïntegreerd in de bestaandeconventionele installatie, worden uitgevoerd. Bij alledrie de installaties is de effluentkwaliteiteen belangrijk criterium. Uitgebreide onderzoeksprogramma’s, onder supervisie van deSTOWA, zullen de komende jaren meer inzicht moeten geven in de opschalingeffecten, dehaalbare effluentkwaliteit en de operationele aspecten van MBR. De drie betrokkenwaterschappen hebben gekozen voor drie principieel verschillende membraantypen (hollevezels, plaat- en buismembranen) waardoor de MBR-ontwikkeling in Nederland in de vollebreedte wordt voortgezet. Na de eerste stappen door Beverwijk, Hilversum en Maasbommel,waarbij alle aspecten van het membraan en zijn omgeving zijn verkend, wordt de MBR inNederland volwassen.

The Water board Regge & Dinkel visits Beverwijk to compare the hollow fibre, flat sheet and tubular membranes.

Bea

Geplaatste afbeelding

Bea

Geplaatste afbeelding

50

Water board Rijn en IJssel (WRIJ) had toupgrade Varsseveld STP due to morestringent effluent requirements, to expandthe biological capacity, to reduce odour andnoise emissions and for a complete revisionof the technical installation.

Improving the effluent quality wasnecessary as Varsseveld STP discharges to avery small vulnerable waterway called theBoven Slinge, which serves as a regionalecological stepping stone. Before the MBRoption was selected, WRIJ considered toupgrade the STP in a conventional way, withcontinuous floc filtration in sand filters as afinal treatment step. WRIJ recentlyupgraded two comparable STP’s with thispost-treatment step and obtained goodresults.

In 2000, the co-operation within theDutch wastewater sector regarding MBRdevelopment started and it soon was evidentthat the location, size and situation atVarsseveld were perfect for a first full scaledemonstration plant using MBR formunicipal wastewater treatment. The MBRVarsseveld was commissioned at the end of2004 and has been fully operational sinceJanuary 2005.

The MBR Varsseveld has been designedby DHV Water BV in co-operation with WRIJand Zenon. The design loading of theVarsseveld MBR amounts to 23,150 p.e. (1 p.e.= 54 g BOD/day) and 755 m3/h. The designand engineering aspects of the VarsseveldMBR have been described in the second

edition of the H2O MBR special in April2003.

The pages 48-49 of this MBR editionshows an artist view of the MBR Varsseveld,including some details of the micro-screensand the membranes. This artist view gives agood impression of the main installationparts of this MBR.

Pilot plant researchThe worldwide experience with the MBR

technology has shown that the use of a pilotplant can significantly speed up and simplifythe start-up of a full-scale MBR installation.It was therefore decided to install a pilot-plant at the Varsseveld STP before andduring the start-up period of the plant. Thepilot plant is a reflection of the full-scaleinstallation, with a scaling down factor ofapproximately 200. The pilot plant wascommissioned in May 2004. The tests offeredgreat value on the chemical membranecleaning, the potential for reduction inenergy consumption and the optimisation ofthe process operation and start-up.

Effluent concentrations of 2.2 mg Ntotal/land 0.15 mg Ptotal/l have been accomplishedfor several weeks. The so called MTR(maximum tolerable risk) concentration insurface water was reached. This is veryimportant in view of the emerge of morestringent European legislation (EU DirectiveWater) coming up, requiring substantialimprovements of the effluent quality by2015.

A cleaning procedure including NaOHand temperatures of 35-40°C turned out tobe more effective than with hydrogenperoxide. The full-scale MBR has beenrefitted to cope with the dosing of NaOH, inaddition to other cleaning chemicals.

The testing of the full-scale software inthe pilot plant proved to be very useful.Almost all software bugs have been tackledduring the studies with the pilot plant,lowering the risks during operation of thefull scale plant. Another main advantagewas the training of the WRIJ personnel andresearchers during use of the pilot plant, foroptimal preparation for working with thefull scale MBR. The filtration unit of thepilot plant has now been connected to thefull scale biology, offering possibilities toperform tests with the pilot membranesbefore ‘jeopardizing’ the full-scalemembranes.

H2O # 2005

M B R s p e c i a l I I I

Start of second phase development programme

MBR Varsseveld, firstDutch full scaleexperienceThe first Dutch full scale MBR for municipal wastewater is now operational, supported by all Dutchwater boards, STOWA and financial contributions from the European Life programme and DutchMinistries. The first results indicate that an excellent permeate quality is reached but that specialattention is required to maintain the membrane permeability at a constant high level. An intensiveresearch programme, including pilot plant studies, has been carried out from May 2004 and willcontinue until December 2005.

The Boven Slinge.

51

StartupThe startup of the full scale MBR was

relatively prosperous (H2O 38 (2005), no. 5, pp27-29). In about a month time the sludgeconcentration grew from 2.4 to 10 kg mlss/m3

and after this month the full hydraulic loadcould be handled in the MBR. Scum or foamdid not appear during startup.

PretreatmentThe pretreatment in a MBR is

considered to be very important to avoiddamage to the membrances induced by hair,

fat, sand and other physical pollutions. By-passing the pre-treatment, for instanceduring maintenance, is not acceptable. Itwas therefore decided to double theinstalled capacity of the bar screens and themicro-screens.

There are two types of 0.8 mm micro-screens, due to the European tendering andthe demonstration character of the plant.One type is a fixed ‘half drum’ with rotatingsieve cleaning brushes. The other type is arotating drum with sprayers for sievecleaning. It was necessary to adapt the

spraying system for the rotating drummicro-screens, therefore some sprayers areplaced now directly within the gutters. Bothtypes of micro-screens show a completeremoval of hairs, but were not consistentlycapable to handle rapid changes from dryweather flow (DWF) to rain weather flow(RWF). These problems are probably causedby material that settles down in the mainsduring low flow periods. To cope with thisproblem all the micro-screens can work inparallel (at double capacity) during about 10minutes at the start of RWF.

Biological treatmentAlthough the MBR was started up in a

relatively cold period (average watertemperature of 12°C), nitrification wascomplete from the start. The permeateammonia concentration mostly is about 0.1mg NH4-N/l. After one month the sludgeconcentration reached the design value of 10kg MLSS/m3 and the MBR was fully loaded.From that time, also denitrification andbiological phosphorus removal rapidlyimproved. The nitrate concentration in thepermeate is about 1-2 mg NO3-N/l, whichmeans that the overall nitrogen removal isabout 96%. The average influent andpermeate data are presented in Table 1.

The data shows that the phosphorouspermeate concentration still is relativelyhigh, compared to the design value of 0.15mg/l. Considering the high influentconcentration and the fact that until nowno iron salt has been dosed, it can beconcluded that the biological phosphorousremoval capacity is significant. It is expectedthat the required ferric dosing amount, toreach the required permeate quality, will berelatively low.

The nitrogen removal is stable,although the aeration control was unstableduring the first two months due to softwareand start-up problems. As a consequence,the aeration capacity was insufficient, whichhad a negative effect on the sludgecharacteristics. Besides that, there was arapid temperature decrease in February.

Both factors were responsible for theinduction of a scum/foam layer on both thedenitrification tank and aeration tank. Thespecially designed scum/foam collector hadto be switched on to avoid that thescum/foam-layer would become too thick.

Membrane filtrationThe four parallel membrane tanks are

equipped with Zenon ZW500d hollow fibremembranes. Each membrane compartmentcontains four membrane cassettes, with a

H2O # 2005

parameter influent mg/l permeate mg/l efficiency %

COD 950 26 97NH4-N 29 0,5 -NO3-N - 1,8 -Ntotal 67 3,0 96Ptotal 17 3,3 81

Table 1: Average concentrations MBR Varsseveld from 1 February until 31 March 2005.

Membrane compartment feed/recirculation pumps.

Figure 1: Development of MBR permeability over time.

52

membrane surface of 1,260 m2 each. Thetotal membrane surface amounts to 20.160m2. The maximum capacity of eachmembrane tank is 250 m3/h, the biomassrecirculation flow into the membranecompartments is supplied by the feedpumps, which have a maximum capacity of800 m3/h.

Figure 1 displays the permeability trendof one of the membrane compartments. Thepermeability start value and trend are asexpected, until the permeability decreasedto a level of about 300 l/(m2.h.bar), whichstarted mid February this year.

Mid February, after a rainy period offour days, a membrane guarantee test wasexecuted in one of the compartments. Inthis membrane compartment the followingtest procedure was executed:- 64 hours at a 37.5 l/(m2.h) net flux,

followed by- 8 hours at a 50 l/(m2.h) net flux.

The 64-hour test obtained satisfactoryresults, while the 8-hour test resulted inhigh suction pressures, and had to beinterrupted. After this test the membraneswere inspected visually. It proved to be ofgreat advantage that the compartmentscould be drained completely, for themaintenance cleaning in air procedures.Close inspection showed that part of themembranes were sticking together with aslimy/glue-like substance, which wasprobably caused by a combination ofanaerobic sludge conditions and thedischarge of polymeric substances by a localindustry.

Slimy sludge was also discovered on thecompartment walls and the membraneaerator tubes. As the membrane fibres weresticking together, the available membranesurface had decreased dramatically, whichresulted in high suction pressures duringthe peak test. Under normal conditions, theavailable membranes were workingoptimally and could process the requiredflow without problems, at a permeabilityabove 300 l/(m2.h.bar). Only at maximumflow conditions, the required suctionpressure was near the upper limit, whichrequires special attention.

Currently, the oxygen control is beingimproved, and, in co-operation with thelocal industry, the polymer discharge will bestopped for a specified period. Both optionsare further investigated, including thecleaning method to cope with the specificfouling. As figure 3 shows, a mild cleaningon April 10 already recovered the

permeability up to 350 l/(m2.h.bar). After anintensive cleaning of all the membranes andremoval of the fouling, the peak capacitytest will be repeated.

OperationDuring the first months of operation

the operators were very busy, which is alsothe case after commissioning a conventionalSTP. Nevertheless, starting up and operatinga MBR requires special attention, as thereare several additional and new installationparts to cope with. The data collection andinterpretation requires special attentionfrom the operators. Fortunately, theoperators had followed an extensive MBRtraining in 2001 in Beverwijk, and renewedtheir knowledge and experience duringoperation of the pilot plant starting fromMay 2004.

CostsThe investment costs of the MBR

installation were about 11 million euro(table 2), which is about 475 euro per p.e. Noparts of the old STP were re-used. Theinvestment and running costs are higherthan for a conventional STP. However, thedifference with a completely newconventional STP with sand filtration foreffluent polishing, is relatively small.

The project is financially supported bythe STOWA (Innovation Fund), theEuropean Commission (LIFE subsidy), andby the Dutch Ministry of Economic Affairs.Thanks to those contributions, theadditional costs and risks could be covered,and this innovative project could beinitiated.

The major running costs are consideredto be the additional costs for maintenance,energy and chemicals. However, as mainlystandard and quite diluted chemicals (citricacid, NaOH and NaOCl) are used, the costsfor the chemicals are limited. The extra

energy consumption however is quite large.A 100% increase was expected, and the firstmeasurements seem to confirm this (1,600MWh/year instead of 800 MWh/year).Therefore, energy saving is a main issue inthe further research programme on the pilotand full-scale MBR. The membrane aerationrequires approximately 40% of the totalenergy consumption of the whole plant.

Consequently, large energy savings canbe obtained by decreasing the membraneaeration. At low influent flows, two or threemembrane compartments are not inoperation (no permeate extraction). In thisso-called relaxation mode, the aerationcapacity of the membranes can be reducedsignificantly. This could lead to considerableenergy savings in the future.

The costs for maintenance will turn outto be somewhat higher than for aconventional STP as the number ofmechanical and electrical parts in a MBR islarger. A main uncertainty is the lifetime ofthe membranes, which of course are stillrelatively expensive. ¶

Coert Petriprocess engineer Water board Rijn en IJsselP.O. Box 148, 7000 AC Doetinchemphone: +31 314 36 95 22e-mail: c.petri@wrij.nl

Philip Schynsproject manager Water board Rijn en IJsselP.O. Box 148, 7000 AC Doetinchemphone: +31 314 36 96 13e-mail: p.schyns@wrij.nl

André van Bentemprocess engineer DHV Water BVP.O. Box 484, 3800 AL Amersfoortphone: +31 33 468 24 12e-mail: andre.vanbentem@dhv.nl

Frans Durieuxmanager marketing & sales Zenon B.V.P.O. Box 235, 6920 AE Duivenphone: +31 26 312 05 22e-mail: fdurieux@zenon.com

H2O # 2005

civil parts 2,800,000mechanical parts 1,700,000electrotechnical parts 1,100,000membrane parts* 2,400,000subtotal 8,000,000indirect construction costs** 3,000,000total 11,000,000 * including permeate pumps, piping, and

others.** including engineering, personnel, over-

head, taxes and others.

Table 2: Construction costs of the MBR Varsseveld(in euro).

58

The existing wastewater treatmentplant Hilversum is located in the easternpart of the municipality of Hilversum in anarea known as Anna’s Hoeve. This area liesabout 4 - 5 meters above sea level at the footof the eastern slopes of the GooiseHeuvelrug. Just across the municipal borderin the municipality of Laren a nature reserveis situated. In this nature reserve a numberof lakes (Laarder Wasmeren) is present.Originally the area was composed of sandydeposits formed some 300,000 years ago byglaciers.

Nowadays the whole area is stronglyinfluenced by human activities. Being thelowest area and outside the city limits, thisarea has been used to dump solid waste aswell as for discharging wastewater.

The dumping of solid waste startedbefore 1885. Since 1920 - 1930 the solid wastewas distributed from a transfer station by anarrow-gauge railway. Nearby anincinerator was in operation for dead animalbodies. Combustible industrial waste andwaste from public gardens was incineratedin the open air to reduce the amount ofwaste to be dumped. Ash has been used forfilling up low lying recreational areasnearby. The landfill was closed in 1959 butillegal incineration of industrial waste hasbeen practised longer.

Consequently the soil in Anna’s Hoeve isheavily contaminated with heavy metals,PAH etc. The discharge of wastewaterstarted as early as 1875, when an openstorage area was constructed. From the

storage the wastewater was distributed inthe surrounding heath land and infiltratedin the soil. In 1940 the discharge of rawsewage was upgraded by the introduction oftrickling filters. As a consequence theLaarder Wasmeren and the subsoil in theseareas are heavily polluted with e.g. heavymetals.

The present wastewater treatment plantHilversum was constructed in 1975 and theeffluent is discharged in the Gooyergracht,some 5 km to the North. From there theeffluent flows into the Eemmeer and finallyinto the North Sea. (The photo shows thearea in its present state).

For Anna’s Hoeve and LaarderWasmeren a masterplan was developed. Thisplan includes as main items:• Removing of the deposits in the Laarder

Wasmeren• Soil sanitation of large areas of Anna’s

Hoeve• Storage of contaminated soils in a new

24 m high hill in Anna’s Hoeve• Realisation of new ponds and ditches• Building of some 700 houses• Construction of a new wastewater

treatment plant Hilversum.

These different projects are realised bythe municipality of Hilversum or the waterboard Amstel Gooi and Vecht (AGV). Otherimportant stakeholders are the province ofNoord-Holland and the owner of the naturereserve Goois Natuur Reservaat. As a resultof the limited areas available, the differentinterests and the strong interactions it tookmany years to co-operate, but with anintegrated design master plan as a result tobe proud of.

Wastewater and surface watersystems

The present situation is indicated in theupper part of figure 1. About 50% of thesewersystem is combined. Because nosurface water is available, all stormwater isstored and discharged after treatment in thewastewater treatment plant. Thestormwater from the separated sewersystemis discharged in the local ponds and theLaarder Wasmeren. The effluent of thewastewater treatment plant is dischargedwith a pipe in the Gooyergracht.

One of the important changes in thefuture situation is that 2/3 of the effluent ofthe wastewater treatment plant will beinfiltrated in the soil in order to reinforcethe natural groundwater flows from the

H2O # 2005

M B R s p e c i a l I I I

Operational in 2008, costs will be 35,500,000 euro

Wastewater treatmentplant Hilversum:ambition and challenge Infiltration of effluent of the wastewater treatment plant Hilversum requires almost total removal ofnutrients. The level of ambition is high: the wastewater treatment plant must produce effluent ofsuperior quality with MBR technology on a limited area and will have a beautiful architectonicdesign that fits well in the landscape. The biological treatment is a system of 18 tanks in series;anaerobic, anoxic, aerobic and anoxic respectively. The pilotplant results demonstrated the need touse chemicals to produce low phosphorus and nitrogen concentrations. The design of the membranetanks is based on the principle of the ‘universal’ membrane tank.

Anna’s Hoeve, view to the north; the new wastewater treatment plant Hilversum will be located exactly in the centreof the photo, just east of the present plant.

59

Gooise Heuvelrug to the west, where somenatural lakes suffer from a lack of water. Inorder to facilitate this reinforcement, thechoice was made to apply the best technicalmeans for the wastewater treatment; theeffluent standards for nutrients will be aslow as 0.15 mg/l total-P and 2.2 mg/l total-N. The wastewater treatment plant isdesigned as Membrane Bio Reactor (MBR).

In the future situation the system willchange as indicated in the lower part offigure 1.

After the cleaning of the LaarderWasmeren it is no longer allowed todischarge any water in these lakes. Newlocal ponds will be made to replace thestorage capacity of the Laarder Wasmeren.Most effluent (2/3) will be discharged to theinfiltration area directly by pipe orindirectly via the local pond system. Theremaining effluent will be discharged to theGooyergracht like nowadays. The effluentpipe may be used for discharging the excessstormwater from the local ponds as well.The concentrated wastewater from theseparated system will be introduced as aseparate flow in the wastewater treatmentplant at all times. The other inflow is fromthe combined sewer and/or the stormwaterstorage (not indicated in figure 1).

All opportunities to cooperate and to

design sewerage, wastewater treatment andthe surface water as an integrated systemwere fully utilized.

DesignThe level of ambition is high: the

wastewater treatment plant must produceeffluent of superior quality with MBRtechnology on a limited area and have abeautiful architectonic design that fits wellin the landscape.

Many aspects of the design result fromthe knowledge and experience of thepilotplant, in operation since November2001 and still an indispensable source ofknowledge and experience.

The plant will be operational in 2008and the costs will be 35,500,000 euro.

The separation in the sewer systembetween the separated and combined systemis continued as much as possible in the

wastewater treatment plant (Figure 2).Except for the influentbuffer in the

DWF-line both the DWF and SWF lines havethe same scheme for pre-treatment:screening (3-4 mm), sand removal andsieving (0.50 - 0.75 mm). The influent bufferin the DWF line has multiple functions:equalize flow and concentration of DWF,reduce flow through aeration tanks andmembranes with 11%, removal of floatingmaterial and extra anaerobic retention time(assisted by adding some activated sludge).

The DWF line is connected to theanaerobic tank whereas the SWF linebypasses the anaerobic tank and isconnected to the predenitrification (Figure4). This prevents a reduced retention time inthe anaerobic tank by the (more diluted)SWF. Moreover the possible introduction ofoxygen with stormwater in the anaerobictank is eliminated.

The biological treatment is acascade/plugflow system of 18 tanks ofabout equal size and dimensions in series.The oxygen conditions are anaerobic, anoxic,aerobic and anoxic respectively. Theflexibility in the design is demonstrated bythe possibility to recycle a low nitrate flowfrom tank 6 to tank 1 as well as from tank 18to the anaerobic tank 1. Optionalrecirculation from tank 18 to 7 results inchanging the cascade/plugflow into arecirculation character. Additional flexibilityis introduced by the possibility to convertthe tanks 5, 6, 13, 14 and 18 into aeratedtanks. The pilotplant results showed that itwas not possible to produce low P and Nconcentrations without the use ofchemicals. In Figure 3 the use of acetic acidand methanol is indicated, but it is alsopossible to use ferric chloride and anothercarbon source e.g. a waste product.

Tank 19 is the last tank ahead of theseparate membrane tank. The tank isaerated to prevent EPS to enter the

H2O # 2005

Figure 2: Sewersystem and pre-treatment.

Figure 3: Biological treatment and membranes. preT = pre-treatment, ANT = anaerobic tank, preDEN = predenitrification, NIT = nitrification,postDEN = postdenitrification, MT = membrane tank.

Figure 1: Sewer system, wastewater treatment plantand surface water at present and in future.

Design characteristics MBR Hilversum.

capacity 91,000 p.e.hydraulic capacity 1,500 m3/hDWF (dry weather flow) 450 m3/h average,

618 m3/h maxSWF (storm weather flow) 882 m3/h effluent quality P = 0.15 mg/l

N = 2.2 mg/lphosphate removal biologicalF/M ratio 0.044 g BOD/

(kg MLSS.day)MLSS 7.8 g/ldesign temperature 10°Cbiological treatment 1 lane conceptbiological sludge 3,000 kg/d solidsproduction

sludge treatment mech. thickening and transport tocentral facility

redundancy of equipment based on risk analysis odour and noise zero nuisance

60

membrane tank and also contains thesubmersible pumps for feeding themembrane tanks. The recycling from themembrane tanks can be divided in anyproportion between tank 7 (effect:introducing the oxygen in the nitrificationbut decreasing the retention time in thetanks 7 - 18) and tank 19 (no effect).

The design of the membrane tanks isbased on the principle of the ‘universal’membrane tank. This tank canaccommodate membrane systems of allmajor suppliers, submersed and drysystems, eventually with small adjustments(‘one size, fits all’). This prevents thatdependence on one specific supplier arisesduring commissioning or in the future,when the membranes have to be replaced.The cross section of the membrane tank ispresented in Figure 4. The sludge isdistributed in the membrane tank by twoheaders with a number of restrictedopenings pointing 45° down.

Figure 5 presents the lay-out of the 8membrane tanks, the centrally locatedtechnical facilities and the drain system forthe recycle of activated sludge with theoverflow to the tanks 7 and/or 19.

The process control of the wastewatertreatment plant Hilversum will reflect theflexibility of the design. This control systemwill be designed based on a model of theprocesses of the wastewater treatment plant.The results of the pilot plant will be used for

calibration and validation of the processparameters. This enables to design thecontrol system on a trial and error basiswhile testing the control system forresponse to different dynamic loadingconditions. Once the control system of themodel meets the objectives it will be usedfor source code generation of the plantsSCADA system. The method for ‘modelbased design’ has been chosen because:• the design of a good control system

without ‘seeing what it does’ is notconsidered possible for complicatedsystems like this wastewater treatmentplant with its stringent effluentstandards;

• reprogramming the control system ofthe model for SCADA application is notonly expensive, but also a source oferrors;

• testing and adjusting under operationalconditions is not possible because of thehigh effluent quality required.

The control system will optimize botheffluent quality and costs for energy andchemicals during DWF (89% of the time) andmainly effluent quality during SWF (11% ofthe time).

Landscape and architectureAn important aspect of the ambition

level is a beautiful architectonic design ofthe wastewater treatment plant that fitswell in the landscape. Initially the plant was

designed as a compact but more or lesstraditional wastewater treatment plant. Thelarge buildings with a typical height ofeight meter would have been clearly visiblefrom the adjacent nature reserve. Therequired area was about two ha.

In February 2004 a landscape study wascommissioned by the municipality ofHilversum or the water board AGV. Theresult of this study by Grontmij was thatthe main part of the wastewater treatmentplant should be situated under the hill ofpolluted soil (double land utilization). Basedon this idea the final architectural designwas made by Snelder Compagnons. Theresult is that all buildings will be ‘hidden’under the hill, except the office building.This part is designed as a beautiful 30 mhigh landmark. The excellent opportunityto discharge the treated process and vent airat this altitude with an invisible stack insidethe landmark was immediately recognizedand will be a major advantage to fulfil thezero nuisance objective. The requiredfootprint was reduced to about one ha.

All logistics (trucking of solid waste,sludge and chemicals) will take place froman incision in the hill, accessible from twosides. Figure 6 shows the outlines of the hilland the situation of the wastewatertreatment plant and Figure 7 an artistimpression of the office building.

In such a wastewater treatment planthealth and safety are extremely important.From the start of the design, health andsafety, especially during maintenance andrepair, was carefully taken intoconsideration. Special sessions wereorganised to check the results and/orimprove the standards. ¶

Kees de Korteproject manager DWRP.O. Box 94.370, 1090 GJ Amsterdamphone: +31 20 460 27 85e-mail: kees.de.korte@dwr.nl

H2O # 2005

Figure 4: Cross section of the ‘universal’ membrane tank. Figure 5: Layout of membrane tanks.

Figure 6: Artist impression of the office building as a landmark.

Environmental TechnologyTechnology in Society¤ Cleaner production in industries ¤ Urban sanitation ¤ Integrated solid waste management

Bio Energy

¤ Biological alternatives for chemical methods¤ Removal of acidifying compounds (e.g. NOX, SOX)¤ Removal of heavy metals

Gas Cleaning

¤ Treatment of wastewater for reuse of resources¤ Removal of pathogens and microĭpollutants (e.g. hormones)¤ Membranes and bacteria

Water Treatment

¤ Alternatives for fossil fuels¤ Direct conversion of biomass to electricity¤ Production of bioĭethanol

¤ Treatment and reuse of soils and sediments¤ Methods for assessment of risks and biodegradability¤ In situ treatment of toxic source concentrations

Soil Remediation

Research and academic education for BSc, MSc, PhDSubĭdepartment of Environmental TechnologyContact: Liesbeth KesaulyaLiesbeth.Kesaulya@wur.nl / tel. +31 (0)317 483339www.environmentaltechnology.wur.nl

DELTA IMPEX has years of international experience in waste water

treatment, offers tailor made solutions according to the BOO(T)

principle and works with proven techniques.

DELTA IMPEX owns his own laboratory and pilot installations for

testing and necessary research.

DELTA IMPEX fully operates the installations. The customer does not

need to invest and can focus on its core-business.

DELTA IMPEX is also specialized in sludge dewatering,

24 hours per day, 7 days per week.

Crystal clear in water solutions withmembrane bio reactors (MBR)

• purification without investment

• complete control of your wastewater treatment

• long-term security

• closing the water cycle

• energy cost reduction

Spoorstraat 25 / 4431 NK ’s-GravenpolderPostbus 55 / 4430 AB ’s-GravenpolderTelefoon +31 (0)113 313000 / Fax +31 (0)113 313100info@deltaimpex.nl / www.deltaimpex.nl

Part of

62

This MBR has a limited hydrauliccapacity. The necessary synergy can beachieved by coupling both systems. With ahybrid MBR, the costs can be reducedcompared to a complete MBR plant, withoutmaking many concessions in terms ofeffluent quality. In order to arrive at a formof water management conducive to nature,an ecological filter will be provideddownstream of the treatment systemsdescribed above. The construction work onthe Ootmarsum WWTP will start in mid2005 and will be completed by late 2006 orearly 2007.

OutdatedThe Ootmarsum wastewater treatment

plant processes sewage from Lattrop, Tilligteand Ootmarsum and was constructed in1974. The plant consists of an intake unitwith bar screens and a grit collector, acarrousel oxidation ditch and a secondarysettling tank. The Ootmarsum WWTP isoutdated and must be modernised. In viewof the water flow requirement in the area,the WWTP will remain at the currentlocation.

Not only does the plant need to berenovated, but the biological treatmentcapacity must also be expanded from apopulation equivalent (PE) of 11,500 to14,000. This must be combined with moreeffective treatment of the sewage. TheOotmarsum WWTP discharges the treatedwastewater into a water system withconsiderable ecological potential. Thereceiving surface water is in the catchmentarea of a ‘water pearl’ - a water system inwhich, by 2018, under the water

management plan of the water authority,the quality of the surface water must besuch that the associated risk is negligible. Insuch situations, WWTPs must not have anynegative impact on the water system. Inorder to give practical form to this ecologicalpotential, the water authority and themunicipality of Dinkelland have drawn up arestructuring plan. This includes measuresconcerning both the sewer network and theWWTP. An innovative approach to theurban water chain/cycle in a region ofsignificant natural value requires aco-ordinated effort. This is whyparticipation in the European Interreg IIIBprogramme (North Sea area) was sought. Anumber of participants in the Urban WaterCycle project (Hamburg, Bradford, Karlebo,Province of Fryslân, Regge en Dinkel water

authority) are working on solutions forurban water problems and are comparingnotes about a variety of associated subjects.The core message is: ‘improving the urbanenvironment by improving the urban watercycle’.

The restructuring plan drawn up withthe municipality of Dinkelland does notprovide for an expansion of the hydrauliccapacity of the Ootmarsum WWTP. Instead,additional measures will be taken in thesewer network.

As well as the standard effluent limitsthat now apply to surface waters with ahigh nature value, a maximum limit of 5mg/l applies to undissolved components.This is in line with another measurederiving from the water authority’s watermanagement plan, namely the status of theOotmarsum WWTP as a pilot plant for theapplication of a more comprehensivetreatment technique based on filtration.This approach takes account, as far aspossible, of more detailed target values forthe long term. Effluent limits and targetvalues are summarised in the table below.

Choice of configurationIn order to arrive at a responsible choice

of configuration with regard to the measuresat the WWTP, Grontmij engineeringconsultants carried out several studies. TheWWTP was modelled in SIMBA, and themodel was used to study a number ofpossible wastewater treatmentconfigurations. A feasibility study was alsocarried out for the purpose of validating thechoice of the filtration technique. The choicewas between the comprehensive applicationof downstream sand filtration, and aconfiguration in which the conventionalactivated sludge system and final settlingtank are provided with a sand filter and a

H2O # 2005

M B R s p e c i a l I I I

Coupling ‘old’ and new system

The Ootmarsum hybridMBR projectThe Ootmarsum wastewater treatment plant, one of the 22 WWTP’s managed by the waterauthority Regge en Dinkel, is situated in the municipality of Dinkelland. The plant is outdated andmust be modernised. Renovation must be combined with more effective treatment of the sewage. TheOotmarsum WWTP discharges the treated wastewater into a water system with considerableecological potential. A restructuring plan includes measures concerning both the sewer network andthe WWTP. An innovative approach to the urban water chain/cycle in a region of significantnatural value requires a co-ordinated effort. Therefore also participation in the European InterregIIIB programme was sought. The choice fell on the option, referred to as a hybrid system. Aconfiguration in which the conventional active sludge system and final settling tank are providedwith a sand filter. Alongside the conventional treatment system a membrane bioreactor is installed.

Component Limit Target (2005) value

BOD 5 2NH4-N 1/2 0,8

(summer/ (90 perc.; winter) temp >10°C)

Ntotal 10 4Ptotal 1 0,15Undissolved components 5 2

Table 1. Effluent limits and target values (mg/l).

63

membrane bioreactor is installed alongsidethe existing treatment system. The choicefell on the second option, referred to as ahybrid system. The feasibility study showedthat the hybrid system entails higherinvestment costs, but the predicted effluentquality is significantly better. It is possiblethat a considerable step can already be madein the direction of the target values. This ispartly due to the possibility of achieving thenecessary synergy between both systems(conventional and MBR) by coupling themtogether by means of an intermediate buffer.

The additional costs of the hybridsystem must be weighed against theexperience that can be gained with the newMBR technology in the coming years, whichmay result in future cost savings and asmaller risk of failure.

The hybrid system consists of a MBRalongside a conventional system. The MBRhas a limited hydraulic capacity. The idea isthat a relatively large part of the dryweather flow (DWF) will be treated with themembranes. During periods of stormweather flow (SWF) the excess rainwater willbe channelled via the intermediate buffer tothe conventional activated sludge systemand final settling tank. In this way, thesurface area of the membranes can beconsiderably reduced in comparison with acomplete MBR plant, and the membranescan be used to the maximum. With a hybridMBR, the costs can be reduced relative tothose of a complete MBR plant, withoutmaking many concessions in terms ofeffluent quality. There is a number ofpossible hybrid solutions. No experience has

yet been gained with any of these options inthe Dutch situation.

The MBR at the Ootmarsum WWTP willtreat 50% of the total amount of sewage inperiods of DWF, while the hydrauliccapacity is only 23% of the SWF. Themaximum hydraulic capacity of the MBRwill be 150 m3/h, while the total sewageinflow to the WWTP under SWF conditionsis 650 m3/h. The intermediate buffer willserve the function of a primary settlingtank. During prolonged periods of SWF thebuffer will have insufficient capacity andtherefore overflow. The overflow water(max. 175 m3/h) will be treated in theconventional system. In this situation, theconventional system will have to treat amaximum of 500 m3/h.

A notable aspect of this configuration isthe large variation in the hydraulic load ofthe conventional system. The table belowshows the distribution of the wastewaterunder DWF and SWF conditions.

In order to arrive at a form of watermanagement conducive to nature, with theWWTP exerting no disruptive influence onthe receiving water system, an ecologicalfilter will be provided downstream of thetreatment systems described above. Thisdownstream ecological filter, also referred toas an ‘ecological activation system’ or‘ecologising step’, consists of a unit which isecological, integrated into the landscape,and in which the ‘sterile effluent’ istransformed to make it ecologicallycompatible with the surface water intowhich it is discharged. During the time itspends in the system, the treatedwastewater is transformed into morenatural water. This can be achieved bymeans of a system of varying depth, inwhich water plants and marsh plants cangrow as a basis for an aquatic ecosystem thatcan accommodate a variety of vital links(zooplankton, phytoplankton, macro fauna,fish, birds, amphibians and insects).

Choice of membraneFor the purpose of selecting a

membrane supplier, a procedure was

followed whereby, on the basis of a scheduleof requirements, suppliers were invited tosubmit bids for designing, supplying andconstructing a membrane extraction unit.The process of selection took place on thebasis of the operating costs and a number ofquality criteria. This yielded two supplierswith a similar price/quality ratio. The twosuppliers then participated in the followingphase, in which the definitive design wasdrawn up, and ultimately NORITMembrane Technology (NMT) was selected.The final design was then modified toaccommodate the findings of a joint riskanalysis carried out by Norit, the Regge enDinkel water authority and Grontmij.

The NORIT AquaFlex MBR systemconsists of a loop with membranespositioned outside the bioreactor tank(Figure 1), rather than having themembranes in the bioreactor or a separatepart of the bioreactor, as in the submergedconcepts. The membrane modules arearranged vertically and are aeratedcontinuously at the bottom. Thiscontinuous aeration is the main drivingforce for the circulation of the activatedsludge, while the feed pump is only used toovercome the hydraulic losses. Permeation isachieved by a suction pump, as in thesubmerged concept. A combination offorward flushing and periodic back-flushingand/or relaxation intervals is used to controlthe cake layer formation inside themembrane tubes and to extend the intervalsbetween maintenance cleaning. Thecontinuous aeration also takes care of thefouling control inside the individualmembrane tubes. The side-streamplacement of the membranes means thatalmost all the options for individualoptimisation of bioreactor and themembrane system are available. Moreover, amuch lower volume of activated sludge isaerated additionally outside the bioreactorthan in a submerged system, so that thebiological processes are influenced as littleas possible.

DesignThe Ootmarsum WWTP is designed for a

biological capacity of 14,000 PE (at 54 g BOD)

H2O # 2005

conventional buffer MBR total

DWF conditions 75 - 75 150SWF conditions while buffer fills 325 175 150 650long-term SWF conditions 500 150 650

Table 2. Flow through the two systems (m3/h).

Figure 1: Basic principle of the NORIT AquaFlexMBR with 8” X-Flow COMPACTmembrane module and continuousaeration.

64

under winter conditions (7.5°C) and acapacity of 18,500 PE (at 54 g BOD) undersummer conditions (17.5°C). The totalhydraulic capacity is 650 m3/h, and the DWFis calculated to be 150 m3/h.

The wastewater from Ootmarsum flowsthrough gravity sewers and is collected at theinlet chamber of the WWTP. The wastewaterfrom Lattrop and Tilligte passes throughpressure pipelines. It arrives downstream of thescrew pumps and is subjected to preliminarytreatment together with the wastewater fromOotmarsum. The preliminary treatment iscarried out with a bar screen (bar separation 6mm) and a grit collector. The screenings arepassed through a washer and a press. The gritextracted from the grit collector is also washed.The screenings and the grit are then disposedof. The conventional activated sludge system,consisting of the carrousel, will be replaced andexpanded by an upstream selectortank/anaerobic tank and a downstreamdiscontinuous sand filter, which can handle amaximum of 250 m3/h. If the flow exceeds thismaximum, the excess is diverted through abypass. The decision to replace theconventional activated sludge system wastaken to simplify the connection to the MBR,and above all to ensure better continuity oftreatment during the construction phase. Allthe wastewater treated in the MBR will first bepassed through a micro screen (perforation size0.75 mm).

The membranes will consist of 6membrane stacks in parallel, each of themequipped with 14 modules, which can beextended to 18 modules (Figure 4). No pilotstudies have been carried out at theOotmarsum WWTP, so the design of themembrane installation is based onexperience with the sidestream concept atother locations, where the design base forOotmarsum has proved itself over a periodof several years (Harry Futselaar, e.a. Theside-stream MBR-system for municipalwastewater treatment. In ‘Proceedings ofmembranes in drinking and industrialwater production’, 15-17 November 2004,L’Aquila (Italy)).

The selector tank/anaerobic tank of boththe conventional activated sludge systemand the MBR can be operated in a number ofways in order to facilitate the selection ofphosphate accumulating bacteria in practiceand to optimise the generation of readilysettleable activated sludge.

The necessary steps have now beentaken to facilitate the design of theecological filter. The design is, however, notyet complete.

H2O # 2005

Figure 3: Schematic presentation of the MBR configuration.

Figure 2: Schematic presentation of the Ootmarsum hybrid configuration.

Figure 4: Artist’s impression of one of the AquaFlex MBR stacks for the upgrading of WWTP Ootmarsum.

65

CostsWhen the above mentioned feasibility

study was carried out, the hybrid systemwas compared with the conventional systemincluding sand filtration. The completeMBR plant option was taken as a reference.The study looked at the investment costsand the operating costs. The result issummarised in the table below as a factor ofthe conventional configuration. The relativeinvestment costs and annual costs do notinclude the costs of the ecological filter.

Subsidies are provided from two sources,namely the STOWA innovation fund and theEuropean Interreg IIIb programme, underthe project name Urban Water Cycle. Theactivities carried out in the context of theInterreg IIIb programme make an importantcontribution to the goals of the Ootmarsumproject, but, for reasons of transparency, arekept strictly separate from the describedactivities in the context of the innovationfund. Therefore there is no overlap. The

contribution from the innovation fund hasits origins in the fact that the Ootmarsumproject, together with the Heenvliet project,has the status of a demonstration project forthe hybrid MBR, which can yield usefulinformation alongside that from theVarsseveld project. On the basis of the MBRmarket analysis, which was commissioned bySTOWA, it can be assumed that a MBR willusually be considered in the context of theexpansion of an existing WWTP. The numberof new MBR sites in the Netherlands isexpected to be much lower. A hybrid conceptwill be a genuine alternative in an expansionscenario, provided it can adequately bedemonstrated that it functions properly.

Final remarksGrontmij recently completed the

specifications. The construction work on theOotmarsum WWTP will start in mid-2005and will be completed by late 2006 or early2007. Measurements of guaranteed valueswill then be carried out, and an extensiveresearch programme will also be pursued,which will yield more knowledge of MBRplants in general and hybrid MBR systemsin particular. Synergy effects will be aparticular area of interest. The researchprogramme will also have to provideinsights into the achievement of certainecological values, and must lead to a plantmanagement concept which focuses oncontinuity and the management of energyand chemical consumption. ¶

Dick de Venteprocess engineer Water board Regge en DinkelP.O. Box 5006, 7600 GA Almelophone: +31 546 83 25 25e-mail: D.deVente@wrd.nl

Bert Geraatssenior consultant GrontmijP.O. Box 203, 3730 AE De Biltphone: +31 30 220 79 11 e-mail: Bert.Geraats@grontmij.nl

Hans Boendersproject manager Water board Regge en DinkelP.O. Box 5006, 7600 GA Almelophone +31 546 83 25 25 e-mail: J.H.M.Boenders@wrd.nl

Harry Futselaartechnology development manager NORITMembrane Technology BVP.O. Box 731, 7500 AS Enschedephone: +31 53 428 70 10e-mail: h.futselaar@noritpt.nl

H2O # 2005

investment annual costs costs

conventional 1.0 1.0MBR 2.0 2.5hybrid 1.4 1.6

Table 3: Relative investment and annual costs.(Conventional is 1.o)

Artist impression of the Ootmarsum MBR.

66

The MBR technique offers uniqueopportunities for upgrading existingwastewater treatment plants. With theEuropean Water Framework Directivecoming into effect by 2006, water boardshave to consider new technologies to extendthe treatment capacity and efficiency ofexisting wwtp’s. The Heenvliet wwtp is atypical example of such a case and will serveas a demonstration project to gatherexperience with the relatively newapplication of hybrid MBR.

The wastewater treatment plant of thecity Heenvliet (the Netherlands) is currentlytreating 8,950 p.e. of domestic wastewater.The hydraulic capacity will increase to 390m3/hr when the Abbenbroek WWTP(capacity 1,650 pe) will be closed and will beconnected to the Heenvliet WWTP. Theproduced effluent is disinfected withsodiumhypochlorite before being dischargedto a local surface water which is also used asbathing water in the summer time.

The Heenvliet WWTP will be upgradedto a hybrid membrane bioreactor system.The new MBR will treat approximately 25%of the total hydraulic capacity, which isequal to the dry weather flow; during stormweather events the remaining influent willbe treated by the conventional part of the

system. This hybrid system allows foroptimisation of both sub-systems in termsof hydraulics and biological loading rate.

Process schemeHeenvliet WWTP consists of a screen, a

selector, a carrousel type aeration tank, aclarifier and a disinfection tank (see Figure1). Sludge treatment consists of thickeningand storage in a lagoon, followed bytransportation for further treatment. Thedesign sludge loading rate amounts to0.054 g BOD/(g MLSS·day). Two surfaceaerators are installed and aeration iscontrolled by oxygen measurement only.

Objectives of the MBR projectThe current capacity of Heenvliet WWTP

is too small to treat the future increasedinfluent flow, therefore the plant will beupgraded to a hybrid MBR system. Themain reason for applying the MBRtechnique is the expected improved effluentquality in terms of nutrients. Nitrogen andphosphorus for example will have to beremoved to lower concentrations, possiblydown to the Maximum Tolerable Risk(MTR) level, 2.2 mg Ntotal/L and 0.15 mgPtotal/L, when the European WaterFramework Directive becomes effective.

Furthermore, other components may bereduced in a MBR system, especially thosethat can be adsorbed to the biomass such asmicro pollutants. In combination with theabsolute barrier provided by the membranethese substances, as well as bacteria andviruses, will be removed completely, makingdisinfection with NaOCl superfluous.

The Heenvliet case provides a goodopportunity to study the system behaviourunder typical Dutch circumstances and theeffects of up-scaling MBR technology. Inthis way the development of MBRtechnology is supported and its potentialscan be explored to the full. Since theHeenvliet MBR is designed as a hybridsystem it is perfectly suited as a researchcase, since the two sub systems(conventional and MBR) can be tested andevaluated in parallel.

From an operational point of view, thehybrid MBR system is advantageous becausethe hydraulic capacity of the membraneswill be utilised to the maximum during dryweather flow. This results in a moreeconomic use of the installed membranesurface compared to an exclusively MBRsystem, where the membranes will have tobe designed at storm weather flows.

Extension of the plantThe Heenvliet wwtp will be upgraded

with a MBR, which will be operated inparallel to the existing plant. The influentwill be divided over the MBR and the

H2O # 2005

M B R s p e c i a l I I I

Research to establish optimal process parameters

and minimum costs for future MBR installations

Hybrid MBR - theperfect upgrade forHeenvlietThe Heenvliet WWTP consists of a low-loaded activated sludge system, type carrousel, with onesecondary clarifier and disinfection of the effluent by means of sodium hypochlorite. The WWTP willbe expanded to a hybrid MBR system in which the MBR will treat 25% of the hydraulic load andwill operate in parallel to the existing conventional activated sludge system. Flat sheet membranemodules have been chosen as most suitable for the Heenvliet situation. The conventional activatedsludge system will be modified in order to improve the current treatment efficiency. In the initialresearch phase the MBR and the modified conventional lines will be run in parallel and thetreatment efficiency of the MBR will be directly compared to the efficiency of the conventional plant.In the following research phase the MBR and carrousel will be operated in series, where as muchwater as possible will be treated in the membrane units (at DWF conditions) and as little as possiblein the existing secondary clarifier (only under SWF conditions). The full-scale MBR research aims toachieve optimal effluent quality. In addition the research aims to establish optimal processparameters and the minimum required costs for future MBR installations. The research will run fora period of three years.

Figure 1: Flow sheet of the current situation at Heenvliet WWTP.

67

conventional plant and the influent qualitywill be equal for both systems. The MBRconsists of the following parts (figure 2):

ScreeningThe influent will be treated in screens

with flat sheets with 3 mm perforations,because perforations are more effective than

screens with bars. The selection of the size ofthe perforation is related to the selected flatsheet membrane system. Flat sheetmembranes are expected to be less sensitivefor pollution than hollow fibre membranes.Besides flat sheet membranes can be cleanedmore easily than hollow fibres in case ofpollution.

Activated sludge tankThe activated sludge tank is divided into

different compartments for P-release,denitrification, nitrification and P-uptake.The sludge passes a continuously aeratedtank before entering the membrane tanks.This is expected to be favourable for thesludge quality, resulting in a bettermembrane performance. The return sludgefrom the membrane tanks passes an anoxictank to limit the oxygen concentrationbefore entering the denitrification tank.

Phosphorus will be removed biologicallyas much as possible. To support thebiological removal, FeCl3 can be added inorder to reach the required very low effluentconcentrations.

Membrane tankThe two parallel membrane tanks are

equipped with Toray flat sheet membranes,provided by Seghers-Keppel. With a pore sizeof 0.08 µm this can be classified as ultra-filtration. By having two membrane tanksthere will be flexibility in the hydraulicmembrane load, because the flow can bedistributed in different proportions over thetwo tanks.

Modifications of the existingplant

The sludge loading rate of theconventional plant will decrease because ofthe planned extension. This effect will bemore pronounced due to the increasedsludge concentration available, which isneeded to ensure identical F/M ratios inboth the conventional plant and the MBR. Amixer will be installed in the carrousel toincrease flexibility of aerobic and anoxiczones and to avoid sludge settling in theaeration tank. The existing activated sludgetank is not provided with an anaerobic tank.Therefore, biological P-removal is expectedto play a limited role, so FeCl3 is added forphosphorus removal.

The hydraulic load will decreasecompared to the present situation, resultingin an improved separation of solids, even atthe higher sludge concentration in theactivated sludge tank.

In table 1 some specifications are givenfor both systems in the future situation.

Two configurations of the hybridsystem

The new hybrid system can be operatedin two ways (figure 3).

MBR in parallel with the conventional laneDuring dry weather flow, the MBR will

be treating relatively more wastewater than

H2O # 2005

Figure 3:. Two Hybrid configurations during dry weather flow and storm weather flow.

conventional MBR

screens mm 6 (bars) 3 (pores)maximum hydraulic load m3/h 290 100biological capacity p.e. (136 gr BOD/p.e/day) 9,660 3,330F/M ratio g BOD/g MLSS · d 0.045 0.045sludge concentration kg MLSS/m3 4.7 10surface load clarifier m3/m2·h 0.51 -net membrane flux l/m2·h (at 100 m3/h) - 24.3maximum possible flux l/m2·h - 56.3disinfection - NaOCl ultrafiltration

Table 1: Specifications of the plant.

Figure 2: Flow sheet of the MBR.

68

the conventional lane. In this way themembrane capacity is utilised as much aspossible. If the flow increases, e.g. duringstorm weather, the hydraulic loading rate ofthe conventional plant will increase morethan proportional. The sludge loading rateof the conventional plant will also increasebut the sludge loading rate of the MBR willdecrease under these circumstances. Thistype of operation is referred to as paralleloperation.

MBR in series with the conventional laneIn this configuration the activated

sludge tanks of the conventional plant andthe MBR are connected in series. This resultsin one biological system, with a possibilityto separate solids both with the membranesand in the conventional clarifier (figure 4).

In this so called ‘in series hybridconfiguration’ the clarifier also acts as ahydraulic buffer. If the flow is lower thanthe membrane capacity, the water level inthe clarifiers decreases and effluent is onlyproduced by the MBR. When the flowexceeds the membrane capacity the waterlevel in the clarifier will increase again untilthe level of the overflow weirs, and theclarifiers will be used for effluentproduction again. In this way the overall

effluent quality is increased compared toparallel configuration.

Research programmeThe hybrid MBR project at Heenvliet

WWTP will serve as a demonstration case forhybrid MBR systems in the Netherlands. Tofacilitate the further development andapplication of this concept an extendedresearch programme has been designed. Partof this research programme is incorporatedin a Europe-wide scientific research projectin close co-operation with universities fromall over Europe. Water board HollandseDelta, Delft University of Technology andUNESCO-IHW are Dutch representatives inthis project. The project proposal wassubmitted to the EU for a subsidy as aspecific tartgeted research project within thesixth framework research programme andwill start summer 2005.

The Heenvliet MBR researchprogramme will focus on the achievableeffluent quality, and minimisation ofoperational cost. For a MBR system, bothinvestment and operating costs are higherat the moment compared to conventionalactivated sludge treatment. To accuratelystudy and optimise the system with respectto energy requirements and othersustainability related aspects, a full-scaleinstallation is a prerequisite.

The first year parallel operation will beinvestigated. The influent will always bedistributed over both systems with aconstant ratio and the sludge loading ratewill be equal for both systems. This enablesa direct comparison between performance ofboth systems of a very low loadedconventional system and a MBR system.

The last two years will be used to testthe serial hybrid configuration. With thistype of operation the membrane filtrationstep can be optimised without the risk ofoverloading the conventional plant during

storm weather. This provides theopportunity to upgrade the plant with aminimum loss of overall effluent quality.

The research programme will start atthe end of 2005 and continue until 2009.Research topics include:• maximum achievable effluent quality to

comply with the EWFD; in terms of- nutrients: nitrogen and phosphorus;

and- micro pollutants, such as heavy

metals, herbicides, pesticides,medicine residuals;

• effects of in series and parallelconfiguration in terms of membraneoperation, activated sludge settleability,floc structure etc.;

• minimisation of energy input foraeration of the membranes and oxygentransfer;

• influent and recirculation screeningrequirements and efficiency;

• comparison of ultra low loadedconventional activated sludge systemand MBR in terms of effluent quality;

• optimisation of the membraneseparation step. Since the membranecompartment consists of two sectionsthat can be operated separately,alternating operation can be applied toincrease membrane lifetime. Thisfeature also allows for critical fluxdetermination tests with a part of themembrane while ensuring treatmentcapacity.

At this moment the MBR is underconstruction and is expected to becommissioned in October 2005 (an artistimpression is presented in figure 5). ¶

Jan Willem Muldersenior policy advisor Water board HollandseDeltaP.O. Box 469, 3300 AL Dordrechtphone: +31 78 633 16 68e-mail: j.mulder@wshd.nl

Freek Kramersenior consultant Witteveen+BosP.O. Box 233, 7400 AE Deventerphone: +31 570 69 79 11e-mail: J.Kramer@witbo.nl

Eric van Sonsbeekproject manager Seghers KeppelHoofd 1, B-2830 Willebroek (Belgium)phone: +32 475 200 551e-mail: Eric_van_Sonsbeek@Seghersgroup.com

H2O # 2005

Figure 4: Flow sheet of the in-series-hybrid configuration.

Figure 5: Impression of the Heenvliet MBR.

ZWOLLE - KAMPEN - LELYSTAD - ALMERE - NIJVERDAL - EMMELOORD

Installatiebedrijf en Ingenieursbureau Koldijk B.V. Schokkerweg 17 P.O. Box 460 8000 AL Zwolle The Netherlands

T +(0)38 - 421 32 20 F +(0)38 - 421 24 94 E-mail: info@koldijkbv.nl URL: www.koldijkbv.nl

A complicated process, combining new membrane techno-

logy with existing sedimentation and aeration processes,

requires very carefully attuned controls, particularly if the

system as a whole has to provide fully automated remote

control. The necessary combination of advanced IT solutions

and far-reaching process know how enabled Koldijk elektro-

techniek to execute this project at a competitive price.

Purelya matter of revolutionary

process control

70

Around 2010 the Dutch Water boardRivierenland expects stricter demands oneffluent quality of ten wastewater treatmentplants (WWTP’s) within rural areas. Untilconcrete legislation comes in effect, theDutch Maximum Tolarable Risk (MTR) isset as standard for receiving surface water.For nitrogen and phosphate concentrationsof 2.2 mg N/l and 0.15 mg P/l, respectively,have been set. With the current WWTP’s

such levels cannot be reached.Consequently, together with RoyalHaskoning and STOWA, the Water boardRivierenland started a research programmeon the applicability of the membranebioreactor and continuous sand filtrationfor treatment of municipal wastewater. Theresearch was located at Maasbommel WWTPand started in March 2002. The main goalswere to determine the feasibility of MBR

technology or end-of-pipe continuous sandfiltration to reach MTR quality for WWTPeffluent and a comparison of MBR andcontinuous sand filtration technologyperformance.

Figure 1 shows a schematicpresentation of the configuration used atMaasbommel. It included a MBR pilot plant(capacity 16 m3/h) with submerged hollowfibre membranes (440 m2) and two full-scaleupflow continuous sand filters (capacity 110m3/h, surface load 15 m/h).

Effluent qualityThe research showed that for both

technologies it is difficult to maintain MTRquality for nitrogen and phosphatethroughout the year. MBR shows betterphosphate removal (minimum values of0.05 mg P/l) than sand filtration (minimumvalues of 0.12 mg P/l). This was mainly dueto the wash-out of ferric sludge from thesand filters. Better nitrogen removal was

H2O # 2005

M B R s p e c i a l I I I

Difficult to reach maximum tolerable risk quality

for nitrogen and phosphate

Comparison of theMBR with continuoussand filtration at theMaasbommel WWTPRecently, a two-year research period in which the membrane bioreactor and conventional wastewatertreatment with continuous sand filtration as polishing step were compared has been concluded. Theaim for both was to reach Dutch Maximum Tolerable Risk Quality. The research was carried out byWater board Rivierenland, Royal Haskoning and STOWA (Foundation of Applied Water Research)at the Maasbommel wastewater treatment plant. Results showed that it was difficult to attain yearlymean MTR quality for nitrogen and phosphate applying either technology.

Parameter Value Unit

BOD influent 50 - 350 mg/lNkj influent 15 - 110 mg/lPtotal influent 3 - 15 mg/lDWF 50 m3/hRWF 150 m3/h

Table 1: Influent composition.

Sand filtration at the Maasbommel WWTP.

71

achieved with sand filtration, however.Residual nitrate concentrations are easilylowered to average values of 0.5 mg/l whenadditional carbon source (acetol) is dosed. Itproved to be difficult to attain such valueswith the MBR by adjusting recycle flowsand carbon source dosing. Furthermore,MBR was more sensitive to RWF than sandfiltration. Under RWF conditions contacttimes and process conditions dramaticallychange. The process configuration used (ahighly divided cascade system) enhancedthis effect. Application of an M-UCT or BCFSprocess may partially neutralise thisnegative effect. Overall, however, both MBRand sand filtration clearly show betternitrogen and phosphate removal than theconventional Maasbommel WWTP withoutsand filtration (figure 2).

When heavy metals are considered(table 2), only the zinc and copper demandsare exceeded by both systems. Other metalsare eliminated to concentrations well belowthe MTR demands. The difference inremoval efficiency between MBR and sandfiltration is minimal. The added valueconcerning heavy metal removal whencompared with the conventional WWTP islimited as well.

Comparable removal efficiencies ofpesticides and herbicides are obtained withboth MBR and sand filtration. For mostcompounds concentrations were below thedetection limit. Of the compounds in thehigher concentration range, only linuronand diazinon exceeded the MTR qualitydemands. No additional removal ofpesticides and herbicides was achieved withMBR or sand filtration compared toconventional wastewater treatment. Onlyglyphosphate (herbicide, active compoundin Roundup) is about 50% more efficientlyremoved than in the conventionalMaasbommel WWTP.

MBR appeared to be more efficient fordisinfection purposes than sand filtration orconventional treatment. Disinfection wasquantified through viability and E. colicounts. Because of the pore size of 0.04 µmpractically no E. coli can pass the membrane.E. coli counts are lowered down to less than1 per ml. This is appreciably lower thanMTR or swimming water quality demands(20 per ml).

Based on wet chemical analyses, theMBR and conventional WWTP with sandfiltration as polishing step show comparableremoval of estrogenic compounds. Bothsystems showed a removal efficiency of 95%

H2O # 2005

Figure 2: MBR and WWTP effluent concentrations of total nitrogen and total phosphorus. From July 1, 2003WWTP effluent was polished using continuous sand filtration.

Figure 1: Schematical representation of the Maasbommel WWTP installation and basic data of the influent. SC = secondary clarifier, SF = sand filter.

parameter unit effluent effluent effluent MTRsecondary sand filters MBR demand

settler

nutrientstotal nitrogen ppm 6.0 2.5 3.0 2.2total phosphorus ppm 2.5 0.5 0.3 0.15metalscopper ppb 6.8 5.2 6.5 3.8zinc ppb 27 23 28 9.4pesticides/herbicidesglyphosphate ppb 7.5 3.5 4.5 -diuron ppb 0.11 0.18 0.15 0.43linuron ppb 1.35 0.9 0.5 0.25diazinon ppb 0.1 0.08 0.09 0.037E.coli cfu/ml 200 130 < 1 20estrogensbisphenol a ng/l 28 33 20 -estron ng/l 4.75 6.9 3.3 -β-estradiol ng/l 1.05 0.85 1 -EEQ (er-calux) nm 0.014 0.011 0.004 -

Table 2: Average effluent concentrations of secondary settler, sand filters and MBR (during periods withoutdisturbances) in 2003-2004.

72

for bisphenol A, estron and b-estradiol. Theestrogenic potential expressed in b-estradiolequivalents (EEQ) was determined as wellthrough a bio-assay (ER-Calux). Sandfiltration shows a 20% lower potential thanconventional treatment alone. Estrogenicpotential after MBR treatment is 60% lowerthan after conventional treatment withsand filtration. An explanation may be theincreased removal of suspended solids dueto the ultrafiltration membranes applied inthe MBR. It is known that the main removalmechanism for phthalates, poly-bromine-diphenyl-ethers (PBDE’s) and alkyl-phenolsis adsorption to suspended solids (STOWA(2004). Vergelijkend onderzoek MBR enzandfiltratie rwzi Maasbommel. Rapport2004-28 (in Dutch)).

Process sensitivity and stabilityDuring the whole research period, the

MBR was more prone to process disruptionsthan sand filtration. This was expected,since sand filtration is a proven technology.Furthermore, it was installed as a full-scaleplant. After several optimization steps theMBR ran relatively stable. The pre-filtrationstep before the MBR system ran without anydifficulty throughout the testing period.With sand filtration, the on-linemeasurements, sand velocity meters andchemical dosing demanded increasedattention. With increased attention stableoperation of the sand filtration wasachieved.

It appeared to be less cumbersome tomaintain a stable effluent quality with sandfiltration than with the MBR. This wasespecially the case for nitrogen andphosphate. Even at RWF Sand filtrationdelivered stable effluent quality, while theMBR effluent quality started to fluctuate.MBR effluent stability may be improved,however, through process configuration andcontrol optimization.

If MBR cleaning is fully automated, thenit is expected that MBR operation willrequire as much operator attention as aconventional WWTP with sand filtration as

polishing step. It may be roughly stated thatmembrane filtration requires as muchattention as sand filtration.

Process measurements and control weredifficult within the MTR quality range.Current analysis techniques are tooinaccurate within this range for goodprocess control. For future design,measuring and control devices requireincreased attention.

ConclusionsReaching MTR quality is difficult for

both MBR and sand filtration. Nevertheless,it may be stated that there is a slightpreference for sand filtration for WWTPexpansion or green field WWTP’s due tostricter nitrogen and phosphate demands.Sand filtration delivers a more stableeffluent quality for these nutrients. It mustbe said, though, that effluent concentrationfor Ntotal and Ptotal of 3 and 0.5 mg/l may bereached with MBR under proper operation.Table 3 shows a qualitative comparison ofMBR with sand filtration.

When disinfection is the main demandfor WWTP expansion or newly built ones,

e.g. due to discharge into swimming water,MBR is preferred. Also for hormone removal,an important parameter for future effluentcriteria, the estrogenic activity after MBRtreatment is considerably lower than aftersand filtration.

Compared to conventional treatment,water quality is not significantly improvedwith either MBR or sand filtration whenspeaking about heavy metals, herbicides orpesticides. To reach MTR quality and lowerconcentrations in general, additionaltechniques need to be applied.

Additional water treatment techniquesare necessary to eliminate prioritycompounds and to comply with expectedstricter demands due to the WaterFramework Directive. Possible treatmenttechnologies are activated carbon,denitrifying activated carbon, ozonisation,selective resins for metal removal, UVirradiation, nanofiltration or reversedosmosis systems. Specific demands oneffluent quality due to the surface waterquality wanted will eventually determinewhich technology will be implemented. ¶

Ferdinand Kiestraprocess engineer Royal HaskoningP.O. Box 151, 6500 AD Nijmegenphone: +31 24 328 42 84e-mail: f.kiestra@royalhaskoning.com

Dennis Pironprocess engineer Water board Rivierenland P.O. Box 599, 4000 AN Tielphone: +31 344 64 95 04e-mail: D.Piron@wsrl.nl

Jacques Segershead of policy and advice department Water boardRivierenland P.O. Box 599, 4000 AN Tiel phone: +31 344 64 95 02e-mail: J.Segers@wsrl.nl

Jans Kruitsenior consultant Royal HaskoningP.O. Box 151, 6500 AD Nijmegenphone: +31 24 328 42 84e-mail: j.kruit@royalhaskoning.com

H2O # 2005

parameter membrane bioreactor conventional + sand filtration

nitrogen removal + ++phosphate removal ++ +E. coli removal + 0heavy metal removal 0 0pesticide/herbicide removal 0 0hormone removal + 0operational aspects 0 0

Table 3: Comparison of MBR with conventional WWTP with sand filtration as polishing step.

Membrane tank MBR pilot plant.

Bea

Geplaatste afbeelding

74

The WWTP has been renovated entirelyin 2001, but the effluent still contains too

many contaminants to be able to use it asrecreation water. Also there was a concern

for micro-contaminants. In the middle of2003 the three-year study on WWTPLeeuwarden into the post treatment of theeffluent of WWTP has started. Two systemsfor treating the effluent, sand filtration (SF)and membrane bioreactor (MBR), have beenexamined more closely. The research of theMBR is carried out by Wetterskip Fryslânand Vitens at the WWTP of Leeuwarden.Two subjects for research are the specificbiological population, which may develop inthe MBR and the removal of organic micro-contaminants. This article focuses on theMBR research project.

System descriptionNormally MBR is applied as an integral

purification technique, meaning treatmentof WWTP influent. At the WWTP site ofLeeuwarden however the MBR is used foreffluent treatment. The MBR system (seefigure 1) is equipped with X-Flow ultrafiltration side-stream membranes (0.03 µmpore size) and has a capacity of 8 m3/h.

The MBR was designed with respectivelyan anaerobic, anoxic and aerobiccompartment. However, the very low BODcontent in the effluent of the WWTP and thecleaning of the membranes with air, incombination with the recirculation flows,ensured aerobic conditions in allcompartments. Consequently the MBR isnot capable to remove phosphate andnitrogen biologically at this time.

Unique micro-organismsThe research workers expect that

treating the WWTP-effluent in a MBR,enables the development of a uniquebacterial population due to extreme highsludge ages. Microscopic analyses (seephotograph) show a typical picture of thistype of sludge.

The sludge quantity increases veryslowly. In 1.5 year time, sludgeconcentration has increased from 2 to 4.5 gSS/l. Still no sludge has been withdrawnfrom the system and sludge age is at thispoint infinite. From the determination ofthe ash content of the sludge it appears thatthe non-organic part has increased to 50%,while normal activated sludge shows valuesbetween 30 and 40%. It is very likely that thesludge is mineralising itself as aconsequence of the low BOD load andtherefore the sludge shows a relative lowactivity and slow growth.

H2O # 2005

M B R s p e c i a l I I I

Good removal capacities for hormones and medicine

residues

MBR-research atWWTP Leeuwarden forthe post treatment ofeffluentThe Frisian capital Leeuwarden wants to make the city canals more attractive and wants the waterin the centre to be separated from the surrounding (nutrient rich) surface water by means of a specialwater separation construction. These constructions allow the boats and canoes to pass but willminimize the flow of surface water to the centre. To improve the water quality of the canals inLeeuwarden, it is the intention to transport treated effluent of the WWTP Leeuwarden to the canalsof the city centre of Leeuwarden. The intention is to make the canals suitable for recreation and todevelop ecological areas.

Figure 1: Lay out MBR for post treatment effluent of WWTP Leeuwarden.

MBR-pilot at WWTP Leeuwarden.

75

Membrane performanceSince the start-up of the pilot MBR in

September 2003, the permeability has showna continuous decrease from 400 l/m2/h/barat the start to 100 l/m2/h/bar in December2004 (Figure 2). Chemical cleaning withhypochlorite and citric acid of themembranes to improve the membraneperformance, only shows a positive effect fora short period of time. In September 2004 theair supply unit of the membrane tank,which ensures a continuous cleaning of themembranes, was also cleaned. This cleaninghas improved the performance of themembranes for a much longer period andthe permeability was ‘stabilized’ to a valuearound 100 l/m2/h/bar.

Sampling and analysing methodsof micro-cantaminants

In 2004 influent and effluent of the MBRhave been analysed for micro-contaminantsat the laboratory of Wetterskip Fryslân andOmegam for determination of micro-contaminants. Specific analyses of hormoneswere performed by AquaSense and RIZAdetermined the ecotoxicity of the samples.In Table 1 an overview is given of theanalysis techniques and the detectedcompounds of each method.

As shown, several persistent micro-contaminants were detected. The Dutchgovernment has made a list of MaximumTolerable Risk (MTR) values for the mostabundant micro contaminants. This list isused to determine the minimal quality levelof surface water. These MTR values are alsoused as a guideline for the determination ofthe quality level of the WWTP effluent. Ingeneral most of these compounds arealready below the MTR values, except forsimazine, cholesterol, diisobutylphtalateand di-n-octylphtalate.

Although the amount of measurementsis limited, so far removal efficiencies tend tobe low for the measured herbicidespesticides and phtalates. However steroidsand nitrogen compounds appear to beeffectively removed.

Furthermore it has to be noted that a lotof pesticide compounds could not bemeasured since their detection limit isrelative high. Since the detection limits ofsome compounds are higher then the MTRvalues it still may be that individualcompounds are exceeding the MTR value,but have not been detected. For examplep,p-DDT has an MTR value of 0.9 ng/l, butits detection limit is 0.1 µg/l, so about 100times higher.

H2O # 2005

analyses analysing detected compounds laboratorymethod

79 non-volatile GC-MS benzylbutylphtalate, bi-sethylhexylphtalate, Wetterskipcompounds di-butylphtalate, di-ethyl-phtalate, Fryslân

di-isobutylphtalate, di-n-octylphtalate phenols en GC-MS - not detected - Wetterskipalcohols Fryslânorganosulfides GC-MS CS2 and di-methylsulfide Wetterskip

Fryslânsteroids, nitrogen GC-MS indole, nicotine, caffeine, cholesterol, Wetterskip compounds di-hydrocholesterol Fryslânchloorphenol GC-MS - not detected - Omegam50 volatile GC-MS chloroform, tetra-chloorethene Wetterskip compounds* Fryslânfenylureum- HPLC diuron Wetterskipherbicides Fryslân41 pesticides LC-MS di-azinon, carbendazim, furalaxyl, Wetterskip(watersoluble) imidacloprid, simazine, propoxur, Fryslân

metzachloor organophosphor, GC-MS di-azinon Wetterskiporganonitrogen Fryslânpesticides heavy metals ICP e.g. chromium, copper, lead, nickel, zinc, Wetterskip

iron, aluminium, arsenic Fryslânmedicines LC-MS carbamazepine, coffeine, di-clofenac, Omegam

erythromycine, gemfibrozil, metoprolol, naproxen, sotalol, sulfamethoxazol

oestrogenic ER-Calux bisphenol A, estron, EEQ** Aquasense***compounds toxicity research bio-essays (not applicable) RIZA * volatile organohalogens, volatile aromates (eg. BTEX) and chlorinated benzenes** EEQ = 17ß-oestradiol equivalents*** analysed by Biodetection Systems, Amsterdam and Waterlaboratorium, Haarlem

Table 1: Analyse methods and detected compounds of micro-contaminants in influent and effluent ofMBR at WWTP Leeuwarden.

MBR sludge Leeuwarden with high sludge age.

76

The single measurement by Omegam sofar showed similar medicine compounds asobserved in a research by RIZA, such asanalgetics (e.g. naproxen), ß-blockers (e.g.sotalol), cholestorol reducing medicines(gemfibrozil) and anti-epileptica(carbamazepine). Most of these detectedmedicine compounds show a reasonableremoval, except for carbamazepine.

Hormone removalAquasense has reported that the

absolute hormone WWTP effluentconcentrations in the effluent of WWTPLeeuwarden were already very low comparedto other WWTPs. However the posttreatment by the MBR ensured even lowerconcentrations and ensured a high totalhormone removal efficiency of on average90% (ER-Calux method).

An additional hormone degradation

activity test figure 3 was performed at theWageningen University. It was shown thatspecific degradation activity of EE2 (17α-ethinylestradiol) by the MBR sludge wasthree times higher than activated sludgefrom the WWTP of Bennekom, whileadsorption characteristics were identical.

Toxicity reductionMeasurements conducted by RIZA

showed that the toxicity of the surface waterand the effluent of the WWTP inLeeuwarden was relative low. The ECf50factor is the concentration factor whichcauses at 50% of the organisms a toxic effect.The higher the number the lower thetoxicity. As can be seen in table 2 the WWTPeffluent samples should be concentrated atleast 70 times to see 50% effect.

The effluent and the canal water havebeen tested again on two occasions (databetween brackets) showing consistentvalues for the WWTP effluent andfluctuating values for the canals. The MBReffluent was only measured once and itshowed on average a decrease of 1.5 times intoxicity in comparison to the MBR influent.

Discussion and conclusionThe pilot MBR is now running for 1.5

years on effluent of the WWTP. Thepreliminary results show a total hormoneremoval about 90% (ER-Calux method) and agood removal of medicine, steroids andnitrogen compounds. Furthermore it isshown that the high sludge age and lowfeed conditions have clearly selected aspecific biological population with a threetimes higher hormone (EE2) degradationcapacity in comparison to activated sludge.

Removal efficiencies tend to be low forthe measured herbicides, pesticides andphtalates. However most of the detectedcompounds are already below the MTRvalue.

Looking at the measured data, simazine,diisobutylphtalate and di-n-octylphtalateexceed their MTR value in WWTP effluentand cannot be effectively removed under thepresent conditions with the MBR. Thesecompounds are therefore of concern andshould be monitored more frequently inorder to determine if advanced treatment isrequired. It has to be noted that a lot ofpesticide compounds could not be measuredsince their detection limit are relative high(even higher then MTR values). The lowlevel of toxicity measured by RIZA in thesame samples at least confirms that eitherthese compounds were not present or had avery low concentration.

H2O # 2005

daphnia IQ test algae test bacterial test

effluent WWTP (= influent MBR) 216 (219 / 162) 74 (70 / 105) 71 (64 / 104)effluent MBR 337 81 113Canal City centre Leeuwarden 190 (841 / 400) 22 (40 / 209) 138 (170 / 333)

Table 2: ECf50 values of influent and effluent of MBR at WWTP Leeuwarden compared to canals of the citycentre of Leeuwarden (August 9th, 2004).

Figure 3: Degradation activity of EE2 by MBR sludge (WWTP Leeuwarden) in comparison to activated sludge(WWTP Bennekom).

Figure 2: Permeability of membranes in street 1 and 2 of the MBR Leeuwarden.

77

focussed on this specific subject.Last but not least, in general efforts

should be made to realise reliablemeasurement techniques, especially forpesticide compounds, which should have adetection limit below the MTR value of thespecific compound.

ProgressThe current situation will be more

closely examined and focussed on theremoval of hormones and the removal ofmicro-contaminants. A second EE2 activitytest at Wageningen University is yet underinvestigation. Since the amount ofmeasurements is very limited monitoringshould be extended to determine the actualquality of the WWTP eflluent andmonitoring is required to support thecurrent findings.

In the following research phase theresearch will be focused at the combineddisposal of micro-contaminants and

nutrients. For that purpose methanol willbe dosed as a carbon source. Because it is theintention to preserve the unique biomass,which has been created up to now, thebiomass will be put into a separate vessel inwhich it will be cultivated and preserved bya continuous feed of WWTP effluent. Also alab-scale installation will be applied toperform specific micro-contaminantremoval tests using the cultivated MBRsludge. ¶

Sybren Gerbenssenior wastewater engineer Wetterskip FryslânHarlingerstraatweg 113, 8914 AZ Leeuwardenphone : +31 58 292 24 11e-mail: sgerbens@wetterskipfryslan.nl

Hans de Vriesproject leader Aquario WatermanagementSmidsstraat 7, 8601 WB Sneekphone : +31 515 48 28 11e-mail: h.devries@aquario.nl

Sameh Sayedwatertreatment expert Vitens WatertechnologySnekertrekweg 61, 8912 AA Leeuwarden phone : +31 58 294 52 35e-mail: SKI.Sayed@pers.vhall.nl

Bonnie Bulthead of department wastewater managementWetterskip FryslânHarlingerstraatweg 113, 8914 AZ Leeuwardenphone : +31 58 292 24 11e-mail: bbult@wetterskipfryslan.nl

H2O # 2005

Previous research by RIZA showed thatlittle is known about the environmental riskof the detected medicine compounds.Therefore new research should determinethe chronical and specific effects of thesemedicine compounds on the aquaticenvironment. The acute toxicity of thedetected medicine compounds howeverseems to be minimal since bio-essays of thesame samples showed low toxicity.

In terms of ecotoxicity a reduction (1.5times) has been observed by the applicationof a MBR. Then again the toxicity of theWWTP effluent was already very low and incase of Algae and Daphnia toxicity lowerthan the surface water in the canals ofLeeuwarden. Therefore the need for MBRtechnology to further purify the effluent onthis matter will be low.

Nevertheless, since the cultivated MBRsludge shows indeed good removalcapacities for hormones and medicineresidues, research will be continued and

78

The growing impor-

tance of education for

MBR staff, operators

and management

The implementation of MBR technology formunicipal wastewater treatment is growing fastin the Netherlands. The first MBR (pilot)installations were intensively and closelymonitored by scientists, consultants and (senior)process-technologists.

For good operational results of MBR-installations it is essential that operatorshave sufficient knowledge of the processesand skills. The course ‘Membrane Bioreactor’by Wateropleidingen, which examines allaspects of the MBR technology, has proveditself for technicians and operators in thelast few years. The course is an interactiveand intensive training given by highlyqualified, enthusiastic and experiencedteachers. The course members experienced apractical course that helped them withdesigning, building, managing or operatinga membrane bioreactor installation.

The membrane bioreactor (MBR) is seenas the most promising wastewater treatmenttechnology for the future. After successfultreatment of industrial wastewaters andrecent research programmes, the MBRtechnique has been optimised in the directionof low effluent concentrations and high flows.Unfortunately, the MBR technology and itsimplementation were growing faster than therelated knowledge to design, build, supportand operate such a wastewater treatmentsystem. Especially the knowledge and skills ofoperators had been neglected, which can leadto bad references, high costs, and neglectionof MBR technology. Wateropleidingen hastherefore organised a number of MBR

training programmes to supporttechnologists and operators.

Nowadays the membrane bioreactor isstarting to become a more ordinarywastewater treatment plant. Suchwastewater treatment plants are controlledby operators, and just followed from adistance by more specialized (and oftenhigher educated) personnel. The knowledgeand skills of the operators becomes thereforemore important for good and efficientoperation of the MBR. Especially in case ofsome less known problems, like theoptimalization of the membrane cleaningprocess. It is therefore very important toeducate operators early in the process!

DevelopmentIn 2001 Wateropleidingen started, in co-

operation with some experienced MBRspecialists in the Netherlands, to develop acourse to support technicians and operators.The focus of this course is to get insight ofthe MBR fundamentals and to help reducerisks in the realisation of full-scale systems.The first course was held in March 2002 andwas visited by procestechnicians from thewater authorities, some consultants andoperators of industrial plants.

In October 2003 three operators of Waterboard Rijn en IJssel followed the coursebecause they became responsible for theoperation of the new pilot membranebioreactor in Varsseveld. This was thebeginning of a new phase in thedevelopment of the course.

At the end of 2004 Wateropleidingenorganised a special course mainly foroperators of Water board Hollandse Delta.They had to operate the new hybrid MBRwhich will be build at Heenvliet. Theregular membrane bioreactor course ofWateropleidingen takes 2.5 days. The in-company course took three days. In thosethree days extra lessons were given by thesenior procestechnologist of the Water boardabout the case of Heenvliet. A guest lecturerof the membrane supplier also gave a lesson

on how to operate the membranes,including the cleaning process.

Regular course descriptionThe MBR course of Wateropleidingen

handles all aspects of the technology. Thecourse is an interactive and intensivetraining given by highly qualified andexperienced teachers. You will get an insightlook into the working of a MBR and learn onhow to evaluate the performance. You willlearn how to assess the critical factorsinvolved and how to control them.

The following items are covered duringthe course:• principles of the MBR,• membranes and their characteristics,• biological processes in de MBR,• interaction of the biology with the

membranes,• process control,• operational costs and performance,

The course members experienced acourse that helped them with designing,building, managing or operating amembrane bioreactor installation.Especially the practical experiences andenthusiastic contribution of the lecturerswere highly appreciated.

Looking forwardIn the (nearby) future there will be more

wastewater treatment plants working withMBR-technology. The operators of thoseplants have partly other needs than the(process)technologists related to the designand building process of the plant. Thereforethe operational knowledge of MBR plantswill gradually be implemented in thewastewater courses like TAZ, about(waste)water treatment techniques andUTAZ about: comprehensive (waste)watertreatment techniques of Wateropleidingen.The membrane bioreactor course willprobably separate in a part for techniciansand a part especially for operators.

ContactFor those who are interested in

following this membrane bioreactor coursein the Netherlands or in your home countryplease visit our website(www.wateropleidingen.nl) or e-mail to:info@wateropleidingen.nl. ¶

Edwin de Buijzerco-ordinator wastewater courses WateropleidingenP.O. Box 1410, 3430 BK Nieuwegeinphone: +31 30 606 94 00e-mail: edwin.debuijzer@wateropleidingen.nl

H2O # 2005

M B R s p e c i a l I I I

Course members are visiting the Beverwijk research project.

Bundelen van expertise in de watersector

Het Netherlands Water Partnership zet zich in voor verdubbeling van de Nederlandse water-inzet en omzet in het buitenland in de komende vijf jaar en is betrokken bij de volgende ontwikkelingen in de drink- en afvalwatersector:

• Toekomstvisie Water Ontwikkelen deelvisie www.toekomstvisiewater.nl

• EuropaNederlandse inbreng in KP7www.wsstp.org

• Bevorderen innovatiekracht www.waterzuiveringstechnologie.nl

• Business Developmentrichting o.a. Midden-Oosten, China en Oost-Europa, PMC Water in de Keten www.nwp.nl

• Kansen door integratie van verschillende kennisgebieden Innovatieplatform

• Lange termijn scenario Afvalwater Blue Forcewww.traverse.nl

• Waterlinksgezamenlijk bijdragen aan het bereiken van de MDG’swww.irc.nl; www.simavi.org; www.aquaforall.nl

Netherlands Water PartnershipTel: 015-2151728info@nwp.nlwww.nwp.nl

80

NORIT cross-flow MBR:

upgrading of process

water for a malthouse

Membrane bioreactor applications are beingused since the mid 1990s in industrial waste-water treatment. The treated water is free ofparticles and bacteria enabling the directdischarge to surface water or the reuse as processwater. This article describes the benefits of theside stream cross-flow MBR systems for an 80%reuse of wastewater as high-grade process for amalthouse.

Since the 1970’s, NORIT MembraneTechnology (NMT) is active in the field oftotal solutions for industrial wastewaterapplications with the tubular membranemodules produced by her sister companyX-Flow. A large variety of applications havebeen designed, built and commissioned.

The cross-flow mode is used mostgenerally for industrial wastewatertreatment applications. The main

characteristic of cross-flow is that a part ofthe feed is withdrawn as permeate, whilethe other part is forced to flow along themembrane surface (Figure 1a). The pressurepump pressurizes the feed, while thecirculation pump recirculates theconcentrate; part of the concentrate ispurged to the bioreactor. The advantage is abetter control of the cake layer build-upresulting in a long time constant fluxwithout any backwashing or cleaning-in-place. Typically, a system consists of severalmodules in-series (one street) and severalstreets in-parallel. NORIT CrossFlow MBRsystems are available as standardised,modular skids. The modules are placedhorizontally resulting in very reliable andcompact installations (Figure 1b). Generally,the capacity of a cross-flow system isrestricted to 100 m3/h due to energyconsumption limitations.

The heart of the cross-flow membraneinstallations is the 8 inch GPR module withthe COMPACT ultrafiltration membraneswith an inner diameter of 5 or 8 mm (Figure1c). NORIT X-Flow has developed theNORIT CrossFlow MBR application highflux membranes with excellent anti-foulingbehaviour. Together with an optimal

Cleaning-in-Place (CIP) procedure theNORIT CrossFlow MBR process is anefficient solution for industrial wastewatertreatment.

Figure 1d shows an example of a fullscale cross-flow ultrafiltration installationbeing part of a MBR at a tank cleaningcompany. In all the MBR projects NMT cansupply everything from the standardisedskids to turn-key projects, where dedicatedcompanies are subcontracted for thebiological treatment part. Typical referencesare found in paper and pulp industry, food,beverage and dairy industry, chemical(process) industry, tank cleaning waterrecycling and leachate water treatment.

Description of the applicationHolland Malt B.V. is erecting one of

world’s largest and most innovativemalthouses at Eemshaven (in the Northernprovince Groningen, the Netherlands)having a malt production capacity of 130,000metric ton. The feed stock of this plant willbe around 165,000 metric ton of barley,which grows at 30,000 hectare of agriculturalland. Holland Malt is a joint company of theDutch brewery Bavaria (Lieshout), theDutch agricultural co-operation Agrifirm

H2O # 2005

A d v e r t o r i a l

Figure 1: Cross-flow system based on 8 inch modules: (a) basic configuration; (b) standard street; (c) 8 inch GFR module; (d) typical full scale installation.

A

B

D

C

81

(Meppel), and around 600 farmers ofbrewer’s barley combining the quality ofDutch barley with the knowledge andexpertise of leading producers of malt andbeer.

In May 2005, the new malthouse willstart up and the wastewater treatmentsystem will be commissioned. One of themain features of this treatment system isthe reduction of the wastewater stream with80% by reusing the treated water in thewashing process.

The first steps in the malt process arethe intensive washing of the barley followedby a soaking step in large tubs. During thesesteps the water is polluted with dustparticles, sugar and starch. In currentlyoperated malthouses the wastewater isbiologically treated in a large bioreactorconverting the organic components intocarbon dioxide and nitrogen gas. After arough clarification to remove the insolubleparticles the water is discharged on thesurface water. This commonly usedtreatment was not allowed anymore for thisnew malthouse and better treatment of thewastewater was required by the localauthorities.

Description of the processIn order to cope with the strict water

discharge requirements Holland Malt hasdecided to implement an advancedmembrane bioreactor/activated carbon andultraviolet disinfection water recyclingplant.

The malt washing water is dischargedthree times a day and is collected in a largebuffer. From this buffer a constant flow ofwastewater passes a drum filter and flowsinto the biological treatment tank. Thebioreactor is an intensive activated sludge(aerobic) process converting the organiccomponents into carbondioxide and

nitrogen. Next, the activated sludge ispumped to a four stage ultrafiltrationsystem to separate the water from thebiomass. The latter is recycled to thebioreactor, while the particle and bacteriafree treated water is polished. First, theremaining odour is removed by NORITActivated Carbon filtration after which thewater is disinfected by ultravioletdisinfection before it is reused as processwater. The quality of the produced processwater is according to the Dutch drinkingwater requirements. This water recyclingplant enables Holland Malt to reduce itseffluent sewer volume dramatically to 20%enabling a five times reuse of the potablewater. Table 1 summarizes the main processparameters.

Description of the systemNMT delivers the project turn-key. The

construction of the water recycling planthas been finished and will be commissionedas soon as the erection of the malthouse willbe ready. Figure 2 gives an impression of thebioreactor and the ultrafiltration cross-flowinstallation.

Concluding remarksThe installation of the wastewater

treatment and water recycling plant has

resulted in the following benefits:• The intake of fresh potable water is

reduced with 80%;• The effluent discharge costs are reduced

significantly; both the amount isdecreased as well as the quality isimproved allowing direct discharge tothe local surface water;

• Reduced incoming water as well asdischarge costs give double savingsincreasing return on investment;

• The plant is compact with a smallfootprint;

• The cross-flow system is robust with asimple process set-up requiring a lowchemical use for a stable long termoperation;

• The side stream modular built cross-flow membrane system reducesmaintenance and gives clean operatingconditions;

• Organic load reduction is achievedalmost exclusively in the biologicalstage minimising chemical use in theprocess;

• The low-fouling tubular membraneshave a proven record of a high life time.

The membrane bioreactor will begrafted in May 2005 followed by thecommissioning of the water recyclingsystem. Comparable water treatmentsystems are under investigation for severalbreweries in Europe. ¶

Henk Schonewille and Harry Futselaar (NORIT Membrane Technology)Ben van Dieren (Bavaria)

For more information: NORIT +31 53 428 70 10 Bavaria +31 499 42 81 11

H2O # 2005

parameter unit influent effluent

feed flow rate m3/h 65 55+10temperature °C 12-35 12-35pH - 6- 8 6- 8TSS mg/l < 200 < 0.1COD mg/l 1,500 < 40BOD mg/l 1,350 < 5N-kj mg/l 50 < 5

Table 1: Process parameters.

Figure 2: Membrane bioreactor system: (left) bioreactor; (right) 4-stage ultrafiltration system.

ProMinent Verder B.V. • Utrechtseweg 4A • NL-3451 GG Vleuten • T: +31 30 677 92 80 • F: +31 30 677 92 88 • E: info@prominent.nl • www.prominent.nl

ProMinent Dosiertechnik GmbH • Im Schuhmachergewann 5-11 • D-69123 Heidelberg • T: +49 6221 842-0 • F: +49 6221 842-617 • www.prominent.com

Solenoid-driven metering pumps

Controllers & sensors

Chlorine dioxide

Tanks

UV

Polymer feed

Ozon

Motor-driven metering pumps

Experts in Chem-Feed and Water Treatment

Global service locally – since ProMinent

www.prominent.nl

SectorSector2005-2006

DutchWater

Business profiles with

extensive background

information

A survey of Dutch

water industries and

organizations focused

on export including all

NWP members www.nwp.nl

A N

ijgh

Per

iod

ieke

n B

V p

ubl

icat

ion

, al

so p

ubl

ish

er o

f H

2O

In co-operation with:

The “Dutch Water Sector” is a reference book with information about the leadingDutch companies and institutes active in the international water sector.This one hundred full-colour pages magazine consists of two separate parts.The first part contains general information about the Dutch water sector anddetailed profiles of more than 50 companies and institutes. The index in theback gives a clear insight of their activities, as well as all members of theNetherlands Water Partnership, and provides you detailed address information.Dutch Water Sector is published by Nijgh Periodieken B.V., also responsible formagazines like H2O.

At the price of € 18,– your copy of Dutch Water Sector can be ordered at dws@nijgh.nlDo not forget the address of delivery.

The worldwide development of the Membrane Bio Reactor for

wastewater treatment is strongly progressing. For successful

and cost-effective application of this innovative technology,

hands-on experience and upfront expertise with biological and

membrane processes is essential. DHV provides added-value

services in the whole life cycle of Membrane Filtration and Mem-

brane Bio Reactors applications.DHV plays an important role in

the development of the MBR concept for large municipal waste-

water treatment plants. For industries we provide full service

solutions, from design to turn-key realisation & operations. DHV

is an international Consultancy and Engineering Group, active in

Building, Manufacturing and Telecommunications; Transport

and Infrastructure; Water; Aviation; and Spatial Planning and

Environment. Within our informal, open organization of approxi-

mately 3,800 professionals, we work on innovative, socially rele-

vant projects. Meet us at our website www.dhv.com.

Acceptation,Regulation or

Innovation?

Gateway to solutionswww.dhv.com