Note: Updated the calculation, found an error. With that, the concept isn't economically viable any more.

I wanted to share a concept for a possible cheap large scale energy storage concept that has been at the back of my head for quite some time. I couldn't find a reason that would make this concept both technically or financially unviable, so I'm eager to hear any criticism.

Physical principle

Ocean water has a mean salt content of 3.5% per mass (3.5 g/L). If you put water with less salinity in a bag, it will float, with more it will sink in the ocean.
The maximum possible salt content is 359 g/L. Thus one can calculate the potential energy of a bag of water with 360 g salt dissolved that is at sea level, and sea ground is 300 m below sea level as:

E= (0.36 kg – 0.035 kg )* 300 m * 10 m/s2 = 975 J.

The pure water doesn‘t have any potential energy, and we also need to subtract the salt content of the sea water. Only the difference of the salinity (effective weight) leads to potential energy.

Summary: With water having a higher salinity than ocean water it is possible to store energy at sea level with respect to the ocean floor.

Actual Implementation

The desalination plant in Carlsbad, California produces 50 million gallons of fresh water a day (Source: Can Desalination Plants Quench California's Thirst For Water In A Clean Way?).
For desalination, they produce high salinity water (20% raised salinity, see article) that is piped back into the ocean. But in this concept, this is an effective weight of 6624 metric tons, or at 300 m height difference, this is 5415.52 kWh of stored energy.

An energy storage system using this high salinity water would need 3 main components.

  • A tank at sea level. Here one could think of building an artificial lake at the coast using a dam, or putting a large plastic bag into the ocean near the coast, and securing it against the waves.

  • A tank at ocean floor level. Again, this would be mainly a large plastic bag, but one needs to make sure that it is possible to empty this bag fully for maximal energy storage. A height difference of a magnitude > 100 m can be often found within 10 km of the coast (this is actually the case at Carlsbad)

  • A pump/turbine + pipelines. In order to use pressurized pipelines, one would need to operate the pump/turbine at the ocean floor tank. Using suction enabled pipelines (for example steel) one could also operate the pump turbine onshore.

Limitations of the system

  • need for big amounts of high salinity water. Operation thus only close to desalination plants, or close to potash mines (they often have large salt heaps, at least in Germany)
  • Environmental impact (1): Placing the sea level tank on the coast line will have drastic impact on maritime life. It should be also possible to install the tank offshore, but with the additional need for large floating structures keeping the tank adrift.
  • Environmental impact (2): The tank on the ocean floor will also impact maritime life. However, in depths of > 100 m, maritime life on the floor is less abundant because of the lack of light.

Economic viability

I‘m not an engineer, and costs clearly depend on the solution, especially where and how to construct the upper tank.
However, I think that the costs should within 1-3 times of that of a usual hydroelectric storage system, with

  • costs for power electronics + O&M: 1002.5$/kW (not per kWh! Additionally capacity doesn't affect the price here)
  • cost for storage unit + load balancing: 12.5$/kWh

See here for details: Analysis of the Cost per Kilowatt Hour to Store Electricity
(IEEE Transactions on energy conversion, vol. 23, no. 2, June 2008, page 529, PDF)

That means the system is easily competitive with current battery technology, with the further upside of having an infinite number of charge/recharge cycles.

Conclusion I think this concept might be viable both commercially and technically. The environmental impact needs to be verified, but might be similar to hydroelectric pump storage, and could be reduced by installing all tanks offshore with the additional need for floating devices.

Furthermore, desalination plants at the coast might use this concept to generate electrical energy without using any tanks, by just installing a pipeline to a deep spot at the ocean floor, and letting the high salinity water sink down. But in that case, it needs to be checked if the diffusion of the salt might be too fast, so that no actual water is moved. This could be solved by small temporary tanks on the ocean floor that can be used as a one way sluice.

  • Welcome to Sustainable Living! It's not energy storage, but perhaps it can be combined with your idea; a few years ago I read an article about the world's first 'blue' energy plant that generates electricity from mixing fresh with salt water: utwente.nl/en/news/!/2014/11/349878/…
    – THelper
    Commented Jul 2, 2017 at 10:11

3 Answers 3


It may be technically feasible. But in reality, for places that do desalination, it's never going to be economical, because there's already an easier way to do things.

Storage on the grid is useful because it enables more supply to be matched to more demand. It does that by shifting energy from the time it is generated to the time it is used. There is another way to match more supply to demand, and that's by shifting demand to times when supply is available.

Desalination provides a great opportunity to time-shift demand. A desalination plant can run at full power whenever there's enough power available, running far higher than the demand for pure water at that time would require. All the surplus pure water can be stored for later. And then, when there's much less powre available, the water supply system can draw on that stored surplus pure water, and the desalination plant can be turned down.

All this works, because storing pure water is much easier, cheaper and more scalable than storing energy; and because in a system with desalination, that process is a high consumer of energy and power.

  • Good point! I didn't think about the use of the desalination plant as a variable consumer. However, it doesn't have the possibilty to store energy, it can be just used for load shedding.
    – t.f.b
    Commented Jul 8, 2017 at 20:24

This all seems highly inefficient. Your one calculation based om m*g*h is meaningless, g is not 9.8 here. You have to displace the sea water which is almost as dense as the saltier water you want to descend. I do not understand how you regard bringing slightly heavier water down as "energy storage". Once the saltier water is down, you want to have caught most of its potential energy or it will have moved into the sea.

If you would want to do anything like this, it would be far easier to just dig up some sand from the beach (heavier than sea water) and drop that into your pipes. Then the sea will wash it back onto the shore for you. But I do not think this would be worth while either.

  • g is always 9.8 on the Earth's surface.
    – kingledion
    Commented Jul 5, 2017 at 18:23

The article you linked says that the salt content of the seawater released by the desalinization plant is increased by 20%, not that the effluent is 20% salt. This means that normal 3.5% salt seawater becomes something like 4.2% salt seawater. Plug that difference into your mgh equation and desalinization plant numbers you get 110 kwH per day of desalinated water, assuming both 100% efficiency and a 300m (!!) height differential.

There is no way that these small amounts of stored electricity will be worth the equipment used to operate the system.

  • It is true that the original salt content of the water is only raised by 20 %. However, a lot more than 50 million gallons of salt-enriched water are produced. I did the maths again and came to an amount of 5415.52 KWh that is too low for any reasonable operation.
    – t.f.b
    Commented Jul 8, 2017 at 20:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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