Department Process Engineering

Gravity-Driven Membrane (GDM) technology

Inadequate access to microbiologically safe drinking water continuously threatens the health and well-being of more than a billion people, primarily in developing countries. In many areas worldwide the central water infrastructure is not available at all, or not reliable, leading to unsafe water at the tap. In such cases, decentralized water treatment can be used.

Ultrafiltration is an effective technology to treat water and in principle can be applied on a decentralized scale. Most ultrafiltration membranes have pores which are smaller than the size of bacteria and viruses. Thus, water filtered through these membranes is microbiologically safe.

During dead-end ultrafiltrtion all macro- and microorganisms, particles and colloids accumulate on the membrne surface and a fouling layer is formed. Backflushing or chemical cleaning are usually used during conventional ultrafiltration to remove fouling layer. This prevents the membrane from clogging, which is expected to occur during filtration on a long term. However, backflushing or cleaning results in complex and maintenance-intensive systems, which are difficult to operate on a long term in developing countries.


Dr. Nicolas DerlonTel. +41 58 765 5378Send Mail

Project lead

Regula MeierhoferGroup Leader Safe Water PromotionTel. +41 58 765 5073Send Mail

New releases

Hydraulic resistance of biofilms

Biological control of biofilms using nematodes

GDM for grey-water treatment


Formation of biofilms on membrane surfaces is usually considered to be detrimental as filtration performances are decreased. However, controlling the formation of biofilms remains challenging and requires a significant energy- and chemical- demand. Our research aims at developing a new paradigm for operating membrane systems that consists in taking advantage of the presence of biofilms on membrane surfaces. The formation of biofilms on membrane surfaces indeed helps to stabilise the permeate flux over several month. The feed water composition determines the physical and biochemical structure of the biofilms, and ultimately its hydraulic resistance. The biofilm also helps to increase the permeate quality. The retention of biodegradable compounds or viruses is higher in the case of a "biofilm+membrane" composite system than for systems based on membrane only.

Hydraulic resistance of biofilms

Several factors influence the hydraulic resistance of biofilms formed during GDM filtration and in turn the permeate flux: the feed water composition in terms of (i) substrate concentration and (ii) microbial diversity and (iii) the operating conditions in terms of external forces acting on the biofilm structure (e.g. TransMembrane Pressure). These factors determine the physical and biochemical structure of the biofilms and ultimately its hydraulic resistance, i.e., the quantity of water that is filtered. Stable permeate flux of 5 to 20 L m-2 h-1 are usually observed during GDM ultrafiltration of surface water.

An increasing substrate concentration in the feed water results in a higher biofilm accumulation and in a lower flux. However, our results indicate that flux stabilisation always occurs, whatever the type of feed water: contaminated drinking water, surface water, used wash-water, etc. 

Effect of predation

Predation by metazoa and protozoa improves the performance of membranes used for the filtration of contaminated water. Membrane filtration allows to securely remove most pathogens from contaminated water but efficient application of membranes is hampered by the development of biofilms on the surface of the membrane reducing the water flux. Strategies available to reduce biofilms on membranes are mostly energy and chemical intensive and lead to increasing operating costs. In our work we are developing an innovative approach that relies on biological mechanisms influencing the permeability of the biofilm to maintain water flux rather than chemical cleaning or energy intensive cross-flow.

Membranes are operated at low pressures and in a dead-end mode. Without mechanical stress a porous and heterogeneous biofilm structure develops that is susceptible to predation by higher organisms. In presence of predators (upper panel in Figure), an open and heterogeneous biofilm structure developed with only partially coverage of the membrane surface (85%). In absence of predation (where higher organisms are chemically inhibited, lower panel in Figure) a flat and compact structure covering the whole membrane surface developed. We were able to demonstrate that metazoa (rotatoria, nematoda, and oligochaeta) were the main group of higher organisms responsible for the development of open biofilm structures.

Dr. Nicolas DerlonTel. +41 58 765 5378Send Mail

Application examples

Several treatment systems based on “biofilm+membrane” composite were developed at Eawag: The Safir water filter and the Autarky Toilet. Both the Safir water filter and the Autarky Toilet are decentralized filtration systems (Table 1). The Safir water filter is a point-of-use system applied to the decentralized production of drinking water in developing countries (Peter-Varbanets et al., 2014). The Autarky Toilet is an updated version of the Blue Diversion Toilet and has been developed as part of the “reinvent the toilet challenge” funded by the Bill&Melinda Gates Foundation (Larsen et al., In press). The Autarky Toilet operates “off the grid” (without connections to piped water, sewer system, etc). It separates fresh urine, feces and used wash-water. The used wash-water is treated on-site for further reuse.

Point-of-use production of drinking water in Kenya

The goal of GDMD project started at Eawag in July 2010 is to develop a novel household water treatment system based on the GDMD technology. This novel GDMD system is designed considering economic, technical and social conditions in urban communities of the developing world. Through this project household filter prototypes with a target capacity of 2-5 litres per hour have been designed for field testing. Treated water is stored safely in an integrated reservoir, to minimize the risk of recontamination. In May 2011, 24 households representing different types of potential users were selected for field trials. Both urban and rural areas are represented, and water sources include boreholes, open wells, river and municipal piped water. 6 month of monitoring shows high efficiency in pathogen removal, and intense interest among communities where filters have been distributed. Flow rates are reportedly good after 6 month of operation, and monitoring will continue for one year. The field tests are conducted in partnership with Kenya Water for Health Organization (KWAHO) and in close collaboration with other Kenyan NGOs and private sector. Inspite of a thick biofilm formed after 5 month of operation, GDM filter works well The eventual production cost of a household filter is estimated at ~30€. Assuming a system lifetime of several years, the running costs of the system will be affordable for poor Kenyan households. However, the initial investment costs could represent a significant barrier to adoption. Alternate financing mechanisms, including rental or leasing options, and possibly extension of microcredit loans will be explored.

Regula MeierhoferGroup Leader Safe Water PromotionTel. +41 58 765 5073Send Mail

Community scale production of drinking water in France/South-Africa

A pilot plant based on the principle of gravity-driven membrane disinfection has been designed and built in 2008 in collaboration with Veolia, Kompetenzzentrum Wasser Berlin and Opalium,. The pilot plant was tested in France (Veolia Research Centre, Annet-sur-Marne (France)) and in South Africa (rural settlement Ogunjini, KwaZulu-Natal) in collaboration with Umgeni Water Works during the period of January 2009 – May 2010. This project was supported by the European Commission (project Techeau).

Placed in a 10 feet-long maritime container, the unit is composed of following componen

  • A submerged flat-sheet UF module (membrane area: 40 m2).
  • A storage tank for residual chlorination to avoid recontamination of treated water (not used in this study).
  • Biological sand filtration was introduced into the system in order to decrease surface load of the membrane. In order to investigate the impact of biological sand filtration, it was in operation during tests in France and was by-passed in the later stages.

The pilot plant has a capacity of 4 m3/d .

Pilot plant based on biological sand filtration and GDM-filtration

The pilot plant was operated in a dead-end mode without back-flushing or cleaning. However, impact of manual drainage was studied and manual drainage was applied 1-7 times a week. In France, untreated water from river Marne, France was feeding the system. This water had turbidity of 3-258 NTU and TOC of 0.9-7.7 mg C/L. Flux values of 5-7 L/m2/h (20OC) were measured.
In South Africa, river water feeding the plant had turbidity of 10-600 NTU and TOC of 1.6-2.5 mgC/L. Flux values of 4-6 L/m2/h (at 20OC) were measured when turbidity of river water was lower than 160 NTU. However, rain events resulting in increase of turbidity of river water over 600 NTU over several days led to a reduction of flux and its stabilization at 2-3 L/m2/h (see Figure 12). It was concluded that a pre-treatment is required for the system when raw water turbidity exceeds 100 NTU.

"Self" living module

“Self” – an energy and water independent transportable living unit “Self” is a project carried out in collaboration between Eawag, EMPA and Zurich University of Arts, to build and demonstrate an autonomous living unit, which is independent of external supply of energy and water. It is a living module for two persons, equipped with solar panels and an advanced system of energy storage and management. The water concept is based on rain water collection on the roof and grey water recycling. In both treatment of rainwater and treatment of grey water the GDM (Gravitiy Driven Membrane) technology is applied. Rain water disinfection Rainwater is collected in a reservoir on the roof of the living unit and is filtered by gravity through an UF membrane with an area of about 1 m2. The treated rain water is collected in a reservoir and is used for drinking, cooking and in the wash basin of the bathroom. A UV lamp in the reservoir is planned to prevent re-growth of bacteria. The drinking water system is designed to provide around 15 L/day of drinking water for 2 inhabitants, and the storage capacity is sufficient to provide water for up to 20-30 days without rainfall. Grey-water recycling The principle of gravity-driven membrane filtration is also implemented in the membrane bioreactor (MBR) used in “self” for grey water treatment, thus no permeate pump is required which saves energy. Additional energy saving is achieved by an intermittent aeration instead of continuous. The MBR is designed for a surface load of < 2 L/m2/h (“conventional” MBR - 20-25 L/m2/h). Similar to the drinking water system the treated grey water is collected in a storage tank (200 L). Treated grey water is reused for the shower, the dishwasher and for toilet flushing (86 L/d and two Persons). A major challenge to guarantee the required water quality is to limit pathogen regrowth in the storage tank. The recycling concept will be tested and evaluated in the coming year.

Dr. Adriano JossTel. +41 58 765 5408Send Mail

Blue Diversion Toilet (Reinvent The Toilet Challenge)

In 2011 the Bill & Melinda Gates Foundation initiated and funded the competition "Reinvent-the-Toilet-Challenge" (RTTC) in which eight universities with promising entries had to proof their concepts. The target is to develop a mass-produced sanitation system operating grid-free (not connected to electricity grid, piped water, or sewer) with total costs not exceeding 5 US cents per person and day. High user comfort and total resource recovery are also key requirements.

Almost 2.6 billion people worldwide use unsafe toilets or defecate in the open. Poor sanitation causes severe diarrhea, which kills 1.8 million people each year. This problem is especially daunting in dense urban settlements especially affecting the urban poor. Therefore people living in under such conditions are the target group for RTTC.

A key feature of the toilet system developed in collaboration with the Austrian design office EOOS is an on-site water recovery system which is based on gravity-driven ultrafiltration and is integrated in the back wall of the toilet. GDM filtration is applied in the RTTC project to the recovery of wash and flush water. The special feature of the gravity-driven ultrafiltration is its passive biological activity maintaining the membranes permeable, thus making chemical cleaning or maintenance unnecessary as long as moderate aeration is provided.

More information about this project

Publications associated

  • Larsen, T.A., Gebauer, H., Gründl, H., Künzle, R., Lüthi, C., Messmer, U., Morgenroth, E., Niwagaba, C.B., Ranner, B. (2015). Blue Diversion: a new approach to sanitation in informal settlements. J. Water Sanit. Hyg. Develop., 2015, 5(1) 64-71.
  • Ravndal, K., Künzle, R., Derlon, N., Morgenroth, E. (In Press). On-site treatment of used wash-water using biologically activated membrane bioreactors operated at different solids retention times. J. Water Sanit. Hyg. Develop.
  • Künzle, R., Pronk, W., Morgenroth, E., Larsen, T.A. (In Press). An energy-efficient membrane bioreactor for on-site treatment and recovery of wastewater.  J. Water Sanit. Hyg. Develop.



Prof. Dr. Tove LarsenTel. +41 58 765 5039Send Mail

About us

Department of Process Engineering (Eng)

Department of Water and Sanitation in Developing Countries (Sandec)

Dr. Sara MarksWater Supply and Treatment GroupTel. +41 58 765 5631Send Mail
Regula MeierhoferGroup Leader Safe Water PromotionTel. +41 58 765 5073Send Mail

Department of Environmental Microbiology

Dr. Frederik HammesGroup LeaderTel. +41 58 765 5372Send Mail
Dr. Tim JulianGroup Leader of Pathogens and Human HealthTel. +41 58 765 5632Send Mail


  • Thomas Neu, Head of group "Microbiology of interfaces", Department of river ecology, Helmholtz Centre for Environmental Research - UFZ.
  • Judith Blom, Limnology Station of Zürich, University of Zürich.
  • Jakob Pernthaler, Prof., Limnology Station of Zürich, University of Zürich.

Former members

  • Richard Johnston, former Head of the water supply group (SANDEC).
  • Maryna Peter-varbanets, former post-doc in the water supply group (SANDEC)
  • Selina Derksen-Müller, former project officer in the water supply group (SANDEC).
  • Anna Chomiak, former PhD student in the process engineering department
  • Joao Nuno Maximino Mimoso, former PhD student in the process engineering department.