I expect that the bulk of the improvement is in avoiding a bunch of the complexity that’s needed to do effectively the same thing at sea level.
Normally, in a desalination plant, you have feed water entering a membrane at a pressure P_feed. Brine comes out at P_brine and permeate (the desalinated output) at P_permeate. P_feed is very high, P_brine is nearly as high as P_feed, and P_permeate is much lower. The flow rate of the feed and brine is considerably higher than the permeate, and one tries to adjust the parameters to get the permeate flow rate as high as possible for a given feed rate because obtaining feed water is expensive and disposing of brine is expensive.
One can engage in trickery. There’s a very clever device called a pressure exchanger that uses the large P_brine to help pressurize some of the incoming feed water. One might imagine a simpler hack of making P_permeate negative so that the feed and brine could be at low pressure, but that’s not going to work (water will not remain liquid at excessively low pressure, and negative pressure has all kinds of problems).
Now move the whole device deep under water. The feed water is (a) all around you and (b) already at plenty of pressure to let P_feed be the ambient pressure. You need to pump feed water in or brine water out to get P_feed - P_brine to be correct, but no pesky pressure exchanger is needed. You need to pump the permeate out — one might think of pumping it “up”, but really the only hard work is producing the pressure difference P_feed - P_permeate or so — water is buoyant to an extent that almost exactly negates its weight. (You’re moving the permeate up, but the pressure difference between the plant and the air helps you out. This is just like how swimming from the bottom of a pool to the surface while carrying your entire body weight is easy, while you almost certainly could not swim well enough to lift your body weight entirely above the water.)
For bonus points, it seems likely that one could dispose of the brine water immediately outside the plant on whichever side is downstream relative to the ocean currents.
I dug around and found:
https://www.flocean.green/subsea-desalination
(Scroll down to 'Subsea SWRO') for their explanation
which is interesting; so yes they take advantage of the pressure - but the other thing they say is because they're taking it from the deep low oxygen area they don't have to fight with sealife etc.
(It seemed easier to look for that rather than fight the paywalling)
Desalinated water is also less dense than normal seawater, so the water column inside the output pipe would create a pressure imbalance with the water column outside the pipe, assisting in the outflow? I'm having trouble figuring out how to resolve this seeming perpetual motion machine
I think the claim about higher efficiency is due to the fact that the sea temp is stable and they don’t have to deal with algae blooms at the bottom of the ocean.
I don’t see how taking advantage of the pressure at lower depths makes much sense. The water would still need to be pumped to the surface, which I think would take as much energy as just pressurizing it.
It's not the pressure difference that other comments write, that does not make sense.
I would assume it's the result to waste water ratio. Afaik, reverse osmosis produces 3 to 4 litres of waste water per liter of fresh water. Since you do not have to pressure the waste water, only depressure the fresh water, you save energy.
It's that you have the pressure difference for almost free-- you get it without investing anything more than the work required to filter the water, whereas you otherwise have to invest enough to put it under pressure.
Suppose that you've got a pipe to the deep sea and a filtration system at the bottom, then a pump on the surface, so that the pipe is mostly filled with air.
Then you have a sufficient pressure difference for the membrane at the bottom and what goes through the membrane only has to go through the filter system.
Meanwhile if you want to achieve this on the surface, then it has to go through the filter, then through a high-pressure pump. The pressurized water will contain salt and some will go through the membrane, so it will be enriched in salt. So now you have a choice: keep letting it try to get through the membrane, or feed it back through the pressure recovery system and use that to repressurize new water.
Since the pressure exchanger is something like 90% efficient, you don't just feed everything back through the pressure exchanger immediately.
Meanwhile, when the membrane is at the bottom of the sea, you can feed in as much new water as you like.
I had this idea many years ago, but didn't think it was worth pursuing, so it's nice to that it's being tried.
Isn't one of the issues here the pressure gradient across a very long segment of pipe? How easy would this be to build and how hard would it be to maintain?
Theoretically, as water is pumped from the surface of the desalinated pipe, the resulting pressure imbalance drives water through the lower desalination filter at high pressure, continuously restoring the water level at the top.
You have had to spend energy to get the floating container to the bottom.
If you filled it with something heavier than water, or left it open to the elements to sink, you still would have to spend a bunch of energy to pump it clean at the bottom.
Then you could also do it at the surface. But they do it a depth because they want a pressure difference on the two sides of the osmosis membrane. You somehow need to generate that pressure difference and the energy you need for that is minimum equal to the amount you need to move the freshwater.
Oh, and you will have to do it continuously, not with a 'container'. Existing desalination plants produce hundreds of thousands of cubic meters of fresh water per day.
This seems like it would work nicely if you removed the concept of pipes and pumps, and replaced them with containers and gravity.
I imagine a barrel of air at the surface with an osmosis filter at the opening and a big ass rock tied to it. Kick it off your barge, let it drop to the bottom and fill with filtered water. Then cut the string and let it float up for collection.
What is the energy expenditure of getting this rock there? The size of the rock is directly proportional to the amount of freshwater this container can hold right?
How much energy does the barge, or whatever pulls it, spend getting itself and the rock and the container into place and back out?
What is this container made of that it can be large enough for this to be feasible, it is full of only air, and it won’t just collapse under pressure at depth? How much does it weigh? We might be talking a much bigger rock than you are envisioning.
You’re glossing over all sorts of energy input and engineering issues, at some point it’s easier to just pump the remaining stuff up
Making it more complicated to deploy in order to save energy costs seems like the wrong direction. They should be making it cheaper instead and slapping a bunch of cheap solar down to power it with the money saved.
One thing I'd be curious to know about is the efficiency of pumping from 500 meters deep. That's a real issue.
Pumping up becomes really inefficient. Large buildings, for example, get around this by pumping to intermediate tanks [1].
This isn't really an option underwater so I'm curious how they'd handle it. Depending on how much more expensive that is to build and how much energy it consumes, this may just not be economical.
That would work only if the fresh water is enclosed in some kind of container, like a plastic bag, otherwise it will mix with the salt water before reaching the surface.
Perhaps the difference in weight between 2 columns of water of equal height, but where one of the columns is of fresh water and the other of salt water, which causes a difference in pressure at their bases, can be exploited somehow for pumping the fresh water, i.e. for pushing it inside a pipe towards the surface, but with some kind of piston that separates it from the salt water.
It shouldn't be too difficult to fill up a balloon like container at the bottom of the sea with fresh water. Once the container is filled it will float up to the surface.
The container doesn't need to be super engineered, since it is filled with water so there is no pressure difference between the inside and outside.
If you have balloon-like containers instead of a pipe, then you expend energy to submerge replacement containers down to the depth at which desalination happens.
You're not really displacing all that much mass though surely? The column is surrounded by water on all sides, i.e. you're removing relatively little mass from the top of the tube, and the entire ocean is pressing in on the rest of it.
Not quite. You need a pressure difference at the bottom. The fresh water will come from reverse osmosis at lower pressure than the ocean at that depth. And it will be not just little lower pressure, but pressure created by ~500km of water, because if not this then why to go to so deep?
I expect that the bulk of the improvement is in avoiding a bunch of the complexity that’s needed to do effectively the same thing at sea level.
Normally, in a desalination plant, you have feed water entering a membrane at a pressure P_feed. Brine comes out at P_brine and permeate (the desalinated output) at P_permeate. P_feed is very high, P_brine is nearly as high as P_feed, and P_permeate is much lower. The flow rate of the feed and brine is considerably higher than the permeate, and one tries to adjust the parameters to get the permeate flow rate as high as possible for a given feed rate because obtaining feed water is expensive and disposing of brine is expensive.
One can engage in trickery. There’s a very clever device called a pressure exchanger that uses the large P_brine to help pressurize some of the incoming feed water. One might imagine a simpler hack of making P_permeate negative so that the feed and brine could be at low pressure, but that’s not going to work (water will not remain liquid at excessively low pressure, and negative pressure has all kinds of problems).
Now move the whole device deep under water. The feed water is (a) all around you and (b) already at plenty of pressure to let P_feed be the ambient pressure. You need to pump feed water in or brine water out to get P_feed - P_brine to be correct, but no pesky pressure exchanger is needed. You need to pump the permeate out — one might think of pumping it “up”, but really the only hard work is producing the pressure difference P_feed - P_permeate or so — water is buoyant to an extent that almost exactly negates its weight. (You’re moving the permeate up, but the pressure difference between the plant and the air helps you out. This is just like how swimming from the bottom of a pool to the surface while carrying your entire body weight is easy, while you almost certainly could not swim well enough to lift your body weight entirely above the water.)
For bonus points, it seems likely that one could dispose of the brine water immediately outside the plant on whichever side is downstream relative to the ocean currents.
I dug around and found: https://www.flocean.green/subsea-desalination (Scroll down to 'Subsea SWRO') for their explanation which is interesting; so yes they take advantage of the pressure - but the other thing they say is because they're taking it from the deep low oxygen area they don't have to fight with sealife etc. (It seemed easier to look for that rather than fight the paywalling)
Cause it's not an ecosystem, right? It's just a resource for us to drain.
Desalinated water is also less dense than normal seawater, so the water column inside the output pipe would create a pressure imbalance with the water column outside the pipe, assisting in the outflow? I'm having trouble figuring out how to resolve this seeming perpetual motion machine
I think it would stop in an isolated setup once most of the water is desalinated.
I think the claim about higher efficiency is due to the fact that the sea temp is stable and they don’t have to deal with algae blooms at the bottom of the ocean.
I don’t see how taking advantage of the pressure at lower depths makes much sense. The water would still need to be pumped to the surface, which I think would take as much energy as just pressurizing it.
Did I miss something?
It's not the pressure difference that other comments write, that does not make sense.
I would assume it's the result to waste water ratio. Afaik, reverse osmosis produces 3 to 4 litres of waste water per liter of fresh water. Since you do not have to pressure the waste water, only depressure the fresh water, you save energy.
It's that you have the pressure difference for almost free-- you get it without investing anything more than the work required to filter the water, whereas you otherwise have to invest enough to put it under pressure.
Suppose that you've got a pipe to the deep sea and a filtration system at the bottom, then a pump on the surface, so that the pipe is mostly filled with air.
Then you have a sufficient pressure difference for the membrane at the bottom and what goes through the membrane only has to go through the filter system.
Meanwhile if you want to achieve this on the surface, then it has to go through the filter, then through a high-pressure pump. The pressurized water will contain salt and some will go through the membrane, so it will be enriched in salt. So now you have a choice: keep letting it try to get through the membrane, or feed it back through the pressure recovery system and use that to repressurize new water.
Since the pressure exchanger is something like 90% efficient, you don't just feed everything back through the pressure exchanger immediately.
Meanwhile, when the membrane is at the bottom of the sea, you can feed in as much new water as you like.
I had this idea many years ago, but didn't think it was worth pursuing, so it's nice to that it's being tried.
Isn't one of the issues here the pressure gradient across a very long segment of pipe? How easy would this be to build and how hard would it be to maintain?
Theoretically, as water is pumped from the surface of the desalinated pipe, the resulting pressure imbalance drives water through the lower desalination filter at high pressure, continuously restoring the water level at the top.
You don't need to pump up the water. Fresh water is less dense than salt water so it will float up to the surface on its own.
That would be a perpetuum mobile. You either have a pressure difference at the membrane or between outside and inside the tube.
The process would be like this:
1. Take in salt water
2. Spend some energy to separate salt from water.
3. Put fresh water into a container.
4. The container containing fresh water will raise to the surface, since it is less dense than salt water.
There is no perpetual motion.
You have had to spend energy to get the floating container to the bottom.
If you filled it with something heavier than water, or left it open to the elements to sink, you still would have to spend a bunch of energy to pump it clean at the bottom.
Probably still easier to just pump the water up.
Then you could also do it at the surface. But they do it a depth because they want a pressure difference on the two sides of the osmosis membrane. You somehow need to generate that pressure difference and the energy you need for that is minimum equal to the amount you need to move the freshwater.
Oh, and you will have to do it continuously, not with a 'container'. Existing desalination plants produce hundreds of thousands of cubic meters of fresh water per day.
I am not sure why getting it up is >= the energy to create the pressure force it through the membrane.
This seems like it would work nicely if you removed the concept of pipes and pumps, and replaced them with containers and gravity.
I imagine a barrel of air at the surface with an osmosis filter at the opening and a big ass rock tied to it. Kick it off your barge, let it drop to the bottom and fill with filtered water. Then cut the string and let it float up for collection.
Seems like you could do that pretty cheaply.
How do you get the floating barrel of air down there? If it floats when full of fresh water it definitely floats when full of air right?
I thought the “tied to a giant rock” part sufficiently explained how to get it down there.
What is the energy expenditure of getting this rock there? The size of the rock is directly proportional to the amount of freshwater this container can hold right?
How much energy does the barge, or whatever pulls it, spend getting itself and the rock and the container into place and back out?
What is this container made of that it can be large enough for this to be feasible, it is full of only air, and it won’t just collapse under pressure at depth? How much does it weigh? We might be talking a much bigger rock than you are envisioning.
You’re glossing over all sorts of energy input and engineering issues, at some point it’s easier to just pump the remaining stuff up
Making it more complicated to deploy in order to save energy costs seems like the wrong direction. They should be making it cheaper instead and slapping a bunch of cheap solar down to power it with the money saved.
https://archive.is/20250811174135/https://www.scientificamer...
One thing I'd be curious to know about is the efficiency of pumping from 500 meters deep. That's a real issue.
Pumping up becomes really inefficient. Large buildings, for example, get around this by pumping to intermediate tanks [1].
This isn't really an option underwater so I'm curious how they'd handle it. Depending on how much more expensive that is to build and how much energy it consumes, this may just not be economical.
[1]: https://www.sloan.com/sites/default/files/2016-06/burj-khali...
You don't need to pump up the water. Fresh water is less dense than salt water so it will float up to the surface on its own.
That would work only if the fresh water is enclosed in some kind of container, like a plastic bag, otherwise it will mix with the salt water before reaching the surface.
Perhaps the difference in weight between 2 columns of water of equal height, but where one of the columns is of fresh water and the other of salt water, which causes a difference in pressure at their bases, can be exploited somehow for pumping the fresh water, i.e. for pushing it inside a pipe towards the surface, but with some kind of piston that separates it from the salt water.
It shouldn't be too difficult to fill up a balloon like container at the bottom of the sea with fresh water. Once the container is filled it will float up to the surface.
The container doesn't need to be super engineered, since it is filled with water so there is no pressure difference between the inside and outside.
If you have balloon-like containers instead of a pipe, then you expend energy to submerge replacement containers down to the depth at which desalination happens.
You're not really displacing all that much mass though surely? The column is surrounded by water on all sides, i.e. you're removing relatively little mass from the top of the tube, and the entire ocean is pressing in on the rest of it.
Not quite. You need a pressure difference at the bottom. The fresh water will come from reverse osmosis at lower pressure than the ocean at that depth. And it will be not just little lower pressure, but pressure created by ~500km of water, because if not this then why to go to so deep?