I am thinking mainly in $\frac{kWh}{m^3}$ and $\frac{\$}{m^3}$.

In the past some decades a wide range of surprisingly efficient water desalination plants were built, mainly in desert regions (Middle East). These plants use reverse osmosis via a system of multiple pressered membranes. This solution seems to be very effective in the sense of energy usage.

But it is not enough. Comparing the desalination prices (coming mainly from energy costs) to the alternatives, a further 60-90% reduction is yet needed. Comparing them, what development potential is there in water desalination?

I think water desalination has probably a theoretical energy limit, which could be maybe calculated from entropy and binding energy formulas. How near are we to this theoretical limit?

  • $\begingroup$ According to the IWA trend report, more efficient and fouling resistant membranes are an ongoing research effort - since I don't know how to translate that into kWh/m³ or $/m³, I'll leave this as a comment: iwapublishing.com/sites/default/files/ebooks/… $\endgroup$ – mart Mar 27 '17 at 11:20

Considering that reverse osmosis is not the only way to desalinate water, I think that yes, there is a lot of development potential in desalination, but that potential might not lie in improvements to existing techniques.

To justify this conclusion and illustrate some areas where there could be a lot of development potential, I present to you my idea for a combined wave, wind, and solar desalination and power plant. I haven't done any maths on this to calculate the area of land needed, or costs, or output, so it might not be viable as-is. But I think the concepts described below (and remember this is just one idea) demonstrate that there is development potential in the following areas:

  • Using renewable energy sources on-site to power the plant
  • Using direct-drive energy instead of electrically transmitted energy
  • Directing and amplifying natural processes of desalination

Combined wave, wind, and solar desalination and power plant


  • No external energy input
  • Cleverly harnessed wave, wind, and solar


  • Energy (electricity)
  • Fresh water
  • Cool air


This plant requires a hot location with large area of cheap land by the ocean and a relatively consistent wind.

Stage 1 - Wave Pump

A wave-powered pump raises sea water into a large lake on land. Here is an example of a direct wave-powered pump, other types of wave power harnessing typically convert mechanical motion into electricity. However, that motion can be easily used to directly drive a pump.

Stage 2 - Evaporation Lake

The evaporation lake is a large shallow area, covered in a greenhouse-like way to aid evaporation. The sea-water flows away from the ocean along channels in the lake-bed then back again towards the ocean in the next adjacent channel where it drains back into the sea. This prevents the build-up of deposits as the returning sea-water will take them with it and return to the sea more concentrated. The roof may contain Fresnel lenses or other solar concentrators to help evaporation.

A wind-catching tower blows air across the lake to lower air pressure and aid in evaporation. This tower could be like those used in Masdar City, or a standard wind turbine tower with either electrical or direct transmission to a series of fans. The result is a continuous airflow across the lake which carries the water vapor to the far side where it is channeled up a wide column into the next stage.

Stage 3 - Condensing Tower

The water vapor is channeled up a large column to a condensing chamber high on the tower. Here, a series of fins are cooled by a heat-pump driven directly by a wind-turbine on top of the tower. the water condenses on the fins and drains into a fresh-water tank near the top of the tower.

Stage 4 - Power Generation

The water from the condensing tower is lowered to a height suitable for a standard water-tower through one or more water turbines to generate power.

Stage 5 - Filtering and treatment

The salty sea-air will also condense on the fins, and there may be small airborne particles and particles from wear on stages within this process that get into the water, so it will probably need further filtering and treatment to make it drinkable. Some of the power from the water turbine may be used for this.

There you have it, you have clean water, above ground level so pressure is already available, and hopefully some excess electricity and cool dry air as by-products.

  • 1
    $\begingroup$ The question appears to be asking for an overview of the situation, and is not requesting individual engineering solutions for desalinisation. Whilst this may be helpful to somebody interested in developing desalinisation, it is not answering the question. It would be on topic for a question asking, say, "What solution exists with X features?" $\endgroup$ – doppelgreener Jan 21 '15 at 23:02
  • $\begingroup$ You could very well be right, although the title does say "What development possibilities could yet exist.." and I certainly think that the idea I presented (which as far as I know is not an existing solution) meets this criteria, that of a possibility that doesn't exist yet. $\endgroup$ – jhabbott Jan 21 '15 at 23:24
  • $\begingroup$ Right, though if the question were asking for a list of specific solutions like this one, it would be a list question with no specific correct answer and would need to be closed as too broad. A good answer here would summarise the current possibilities people are aware of. Bear in mind that's the title, not the question body - answering the title instead of the body is almost always going to turn out badly! If the title is not sufficiently matched to the body, then the title probably could do with an edit. $\endgroup$ – doppelgreener Jan 21 '15 at 23:28
  • $\begingroup$ I think you are right, therefore I have edited the preamble to actually answer the question posed and use the proposed plant as an example of the areas of development potential. $\endgroup$ – jhabbott Jan 22 '15 at 0:23

Carnegie, through their CETO device, and others have already looked at using wave power to directly pressurise water for reverse osmosis: a wholly mechanical process rather than converting to electricity and back again (giving potential efficiency savings). Two challenges: first, there aren't many places in the world with a really huge wave resource (UK, Portugal being two that come to mind); and second, it's proven very hard to get wave machines working reliably. That's surmountable, but challenging.

The other significant potential development will seem counter-intuitive, and the key to unlocking it is to consider the wider system, rather than just the desalination process. That development is to move to lower-efficiency desalination processes.

That's because lower-efficiency processes can have much lower capital costs. The advantage of that is that they can then be operated for a smaller proportion of the time, without taking a big hit on the cost per cubic metre of desalinated water.

So why would you want to run the desalination for a smaller proportion of the time? Because places that depend on desalinated water have a lot of sunlight. Which makes PV power cheap. But PV has a generation profile that only partly matches demand. There'll be times of insufficient power, and times of excess power. That excess power is really really cheap. And that's a great time to run desalination.

So a combined energy and water system that has a lot of PV and a lot of low-capex, lower-efficiency desalination can work very well. In effect, the desalinated water acts as a form of virtual storage. All electricity systems need storage somewhere in the system. For some countries, that's in the form of storage hydro. For others, it's in the form of gas-holders, coal bunkers or biomass bunkers. Those stores are pre-generation stores. In other systems, there's storage post-generation, in the form of low-grade thermal storage: when energy is going to be used as low-grade heat, it makes sense to store it in that form, as such storage is very cheap and very scalable. Similarly, storage of desalinated water is very cheap and very scalable. It acts as a time-buffer, a flexible delay mechanism, between the supply of PV electricity, and the demand for the energy service itself - in this case, desalinated water.


For reverse Osmosis

This site gives the minimal required energy for sea water desalination by RO as 2.78 kJ/l (fresh water), this is if you consider only the reversible process. According to wikipedia, the best RO desalination plants operate at 3kWh/m³ which trasnlates to 10.8 kJ/l.

AFAIK, energy losses are pressure loss through the membrane (in addition to the osmotic pressure, a membrane introduces irreversible pressure losses), water pretreatment and energy (in the form of pressure) in the brine. Also a lot of water simply needs to be moved about, there's pre treatment steps etc.

According to this IWA trend report, two areas within the wide field of membranes where more research is done are better membranes in terms of pressure loss and fouling resistance (fouling directly influences pressure losses). Recentish developments in RO desalination like forward osmosis mostly benefit from better fouling resistance compared to RO.

For thermal desalination
(will update when I find more info)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.