Membrane distillation (MD) is an emerging membrane technology for water purification and desalination. Unlike most other membrane processes, MD does not require a mechanical pressure pump but relies on low grade heat to power the process.
Ideas resembling membrane distillation were proposed by scientists already in the 1940s.
Interesting concepts were developed by Weyl and Pactide Corp (both US) in 1967, by Henderyckx (Belgium) in 1968 and by Sasakura Kikai (Japan) in 1969. The present theory of membrane distillation was developed in Sweden, Japan, Germany and the United States in the beginning of the 1980s. By the end of the 1980s research had started in many additional countries, including Australia, China, Italy, the Netherlands, OSS and Spain. The first scientific article using the word membrane distillation was The SU membrane Distillation system, Bjäreklint, Rodesjö, Torstensson published in Sweden in 1983. A bit later, the same year a Japanese company Nitto described the same technology in a report entitled Thermopervaporation membrane separators. In 1984 two articles called Membrane distillation were published, one by Daniel Curtin in the US and the other by Schneider and Van Gassel in Germany.
Articles describing the technology using names such as Thermal Pervaporation, Thermopervaporation and Transmembrane Distillation appeared in different countries until an international conference in the Netherlands agreed on the originally Swedish name Membrane Distillation. (Terminology for Membrane Distillation, Smoulders and Franken, 1989).
The best summary of research up to date can be found in A framework for better understanding membrane distillation separation process, M.S.El-Bourawia, Z.Dinga, R.Maa, M.Khayet, Journal of Membrane Science, August 1, 2006
Membrane distillation uses hydrophobic membranes as a barrier for contaminated water from which mass transport of vapour is driven by differences in vapour pressure. Although the theory of MD is quite complicated and to some extent as yet unexplored, the practical use of MD is simple: Hot feed water flows alongside a microporous, hydrophobic membrane. The surface tension of the feed prevents it, while in a liquid state, from entering the non-wettable (hydrophobic) membrane. Instead, part of the water evaporates and, as vapor, passes through the pores of the membrane, then condenses on the other side of the membrane. Non-volatile contents cannot enter the membrane. They flow past the membrane. Temperature on the hot side must be below 100 degrees Celsius.
The driving force in the process is created by a difference in vapour pressure on either side of the membrane which, in turn, is created by the temperature difference between the hot feed and a condensing surface.
In contrast to other distillation processes, which are driven by high temperature or vacuum, no entrainment takes place and the quality of permeate remains unchanged - even up to very high feed concentrations. In contrast to other filtration processes, no water enters the membrane, which reduces the need for back-washing and also the need for pre-treatment of the feed
Since the MD-process itself takes place at temperatures below 100°C and at ambient pressure, requirements to withstand high temperatures and/or pressures are eliminated. Capital costs of equipment can therefore be reduced. For the same reason, operation and maintenance of the equipment is less exacting.
Water is purified in the following way:
- Heat is transported from the bulk fluid to the water surface
- Water evaporates from the surface
- Water vapour diffuses through the membrane
- Water vapour condenses on the other side of the membrane
Schematic picture of MD
In practical terms, water is first passed through a pre-treatment filter in order to remove solid particulate matter. Volatiles are then removed in a degasser and, finally, non-volatiles in the MD – module. Since all matter at a given temperature is either volatile or solid, i.e. volatile or non-volatile, has vapor pressure or has not, installed equipment will be able to remove all conceivable contaminants from sea-water, brackish water, municipal sewage, industrial waste water and from any ground- and surface water.
In distillation technologies, the vapour may carry small droplets of water which contain contaminants - particles, ions, bacteria, virus, pyrogens and the like. This is called “entrainment”. This entrainment can be reduced by various devices, but it is theoretically impossible to reach total separation.
In filter technologies, the relatively high pressure may cause small amounts of impurities – particles, ions or pyrogens, even bacteria and virus, to pass the filter.
In MD, there are no theoretical restraints to total separation. Nothing is entrained with vapour and nothing is pressured through the membrane.
When MD is combined with efficient degassing, all types of contaminants will be removed. Consequently, conductivity levels of less than 0.3 µS/cm in the permeate can be reached in desalination of sea-water without pre-treatment or polishing. Taking conductivity of immediately solved CO2 into account, this means that, with proper pre-treatment, water of (UPW) Ultra Pure Water quality can be obtained in only two process steps – in industry called “18 MegOhm quality” – which corresponds to a resistivity of 0.056 µS/cm.
Typical third party test results for Scarab’s Membrane Distillation equipment
Type of contamination
14 000 (after 7 days)
National Bacteriologic Laboratory, Stockholm
< 0.01 mg/l
Water Protection Ass of South West Finland
31 000 ppm
< 1 ppm
VBB Viak Stockholm
1 000 µg/l
< 1 µg/l
University of Turku, Finland
< 4 Bq/l
Swedish Radiation Protection Institute
< 0.1 Bq
Radiation Physics Department, Univ of Lund
< 0,003 mg/l
< 0,003 mg/l
- 100 % (theoretical) rejection of ions, macromolecules, colloids, cells, and other non-volatiles
- Lower operating temperatures than conventional distillation
- No vacuum is used
- Lower operating pressure than conventional pressure-driven membrane separation processes
- Low sensitivity to variations in process variables (e.g. pH and salts)
- Good to excellent mechanical properties and chemical resistance
- Ambient pressure and low temperature means no metals – and no metals means no corrosion – and lower capital cost than comparable state-of-the-art technologies
- Reduced use of chemicals, filters and other consumables means low running cost
- Self-regulating operation increases safety
- Waste heat sources with temperatures below 100°C can be readily employed to decrease energy cost
- High energy intensity
- Sensitive to surfactants
State-of-the-art water purification / desalination
There are many technologies on the market for purification of water. Each technology has its own advantages. Here are the most important examples:
- Flocculation, sedimentation: Common in municipal plants. Removes particles
- Disinfection, by chlorine for instance: Kills micro organisms
- UV: Kills micro organisms
- Ozone: Kills bacteria and breaks down large organics
- Activated Carbon: Removes particles, bacteria and volatiles
- Aeration, degassing: Removes volatiles
- Ion Exchange: Removes ions
- Electrical methods such as ED, EDR etc.: Remove ions
- Filter methods such as UF, MF etc: Remove particles
- Distillation such as MSF, ME,VC, MVC etc.: Removes particles and ions
- Membrane distillation: Removes particles and ions
- Reverse Osmosis (RO): Removes particles and ions
- Combinations: Because of the limitations in each one of these methods combinations are often used to achieve good results. To make Ultra Pure Water for the semiconductor industry, for instance, several dozens of different methods are used in series
Most technologies listed above are a good choice for a specific purpose. Only the distillation technologies (10) and RO (12) are possible to use for desalination of sea-water. ED (8) can be used for brackish water.
Reverse Osmosis (RO) is presently the closest benchmark for MD. RO is a proven and mature technology which makes very good quality water at a cost of around 0.15 US$ per cubic meter for purification of sweet or brackish water and between 0.5 - 2 US$ for salt water.
Compared to RO, MD has the following features.
- Throughput per membrane area – similar.
- Cost of membrane – similar or higher.
- Life length of membrane and module – similar or higher.
- No high pressure pumps.
- Lower material costs.
- No chemicals to condition the water.
- Not sensitive to changes in feed water quality.
- Up to 100% higher permeate/feed ratio.
Many of these practical advantages can be turned into economic advantages
Typical cost distribution
Emerging solar desalination technologies
The efficiency of RO has been improved considerably during recent years. Energy use has been reduced by recovery and new membranes. Also the performance ratio for electricity production from PV has been improved considerably and will be improved further. However, certain losses in transformation from solar radiation to electricity will also exist in the future and set an upper limit for the use of RO.
MD has been developed from start for use of low temperature solar energy. It uses low grade heat and inexpensive material
Compared to other dedicated solar desalination technologies that also may use low grade heat and inexpensive materials such as solar stills, humidification/dehumidification and dew collection/condensation, the advantage of MD is that evaporation and condensation surfaces are tightly packed and therefore result in compact equipment leading to a good relation between output and capital cost.
Of MD, there are also several different varieties. Scarab has developed Air Gap Membrane Distillation (AGMD). Compared to Direct Contact Membrane Distillation, energy losses are much smaller in AGMD. Compared to Vacuum Membrane Distillation and Sweeping Gas Membrane Distillation, the function of AGMD is less complicated and does not require electrical auxiliary equipment, i.e. vacuum pumps or fans respectively
Other players in membrane distillation
Literature reviews show that a substantial amount of laboratory testing of MD has been carried out over the last decades at research institutes in more than a dozen countries.
Three projects have, this far, gone beyond laboratory testing: Scarab Development AB in Sweden - www.scarab.se , TNO in Netherlands www.tno.nland Fraunhofer in Germany www.ise.fhg.de .
Since the most important cost in membrane distillation is energy, it is optimal to use waste heat during energy generation. That offers to reduce the cost of pure water by an appreciable amount. The concept of co-generation is already widely accepted although what has held it back to some extent is that it is not always possible to find ready buyers of the heat generated.
Membrane Distillation can use waste heat and return it to the system with only a minor drop in temperature. The cost of the system is then allocated between electricity and water. Both become more viable. This is even truer where the heat can be used by some other end-user.
In the long term, solar energy will provide sufficient energy for all human activities on the earth. The daily influx of energy from the sun is more than a thousand times larger than our present use of all types of energy.
While existing energy infrastructure, especially power plants, has been built on the false assumption of growing and non-ending supply of fossil fuels, new more sustainable energy technologies are today in rapid development. The lead in development of specific technologies is often taken by small or medium size specialist companies, but the large international infrastructure companies (“integrators”) are looking for ways to integrate these new technologies both in new and in existing infrastructure in an efficient way.
One of the new concepts that have evolved recently is the co-location of power plants and water works. The most powerful model is one that extends the co-location concept to the co-generation concept and then further into the polygeneration concept.
A power plant turns fuel into electricity and heat. The figures are from an efficient power plant. Often the percentage of electricity produced from the fuel is much less.
Co-location means that power utilities and water utilities are placed next to each other and can use some of the same facilities, but also get close to each others produce, since water plants need electricity and power plants need water.
Co-generation usually means that the same plant produces both electricity and heat. In traditional power plants half the fuel usually goes to waste. Not only is the fuel not put to good use, the heat produced has to be removed by heat-exchangers into rivers or lakes or into the air with costly cooling towers. Using that energy for practical and valuable purposes obviously makes sense.
A proven co-generation concept is the district heating system, very common in Sweden, where waste heat from power plants is used for hot water or heating in buildings. Another is the CHP (Combined Heat and Power) concept that has been introduced especially in the US to provide electricity to individual houses, shops etc. The Swedish district heating systems are large – in the megawatts while some of the CHPs are very small – tens of kW. So, there is experience of a very wide range of capacities.
The efficiency of co-generation is obvious. Yet many power plants waste as much heat energy as they deliver in electricity. The reason is that there is no conventional use for the heat at the location or in the vicinity. Therefore the new concept of polygeneration has been developed.
Polygeneration means using several different fuels in several different processes to produce a range of different results/products. The purpose is both to save on resources and increase profitability.
During the last decades, researchers around the world have fine tuned the polygeneration concept. New fuels have been added, efficiency has been increased as well as internal energy recovery and new uses of energy. Today a standard polygeneration system may use coal, oil, natural gas, biomass, municipal waste or solar collectors alone or in combination to provide electricity, process steam, district heating, hot water and various types of cooling.
Scarab has added yet another dimension by developing and introducing a low grade heat powered water purification/desalination technology. By using waste heat, energy use for water treatment is reduced at least four times compared to comparable existing methods.
Polygeneration in one form or another will be the most economical choice for future implementation of energy and water infrastructure developments from the smallest scale to the largest.
Typical applications of polygeneration are:
- Any conventional power plant can be made more efficient and sustainable in polygeneration mode.
- Solar powered CHP for single houses or small communities producing electricity, hot water, space heating, pure water, space cooling and refrigeration.
- Small PV for electricity production co-producing hot water and pure water.
- Large concentrated solar power steam engine plants, possibly topped up with biomass burning to increase initial temperatures and to permit 24-hr production. This plant could for instance concentrate on delivering electricity and desalinating water if located in a sparsely populated area and be complemented with for instance hot water and space cooling if located in a populous area.
- Large PV field for production of electricity and pure water
- Gas-engines using biogas from agricultural or municipal waste to produce electricity and pure water.
- There are a large number of other variations on the polygeneration theme. In general, saving energy in existing plants, making energy production more efficient in new and existing plants and using the same fuel for multi-purposes whenever possible is more profitable and also more sustainable than just adding new production facilities whether of conventional or new (“alternative” energy) types.