RO (Reverse Osmosis)

I'll give this a GA. It scores very high on the brevity index.

Seems that a conventional unit would pump brine at, for example, 600 psi x 10 gph. (The 600 psi is required to overcome filter resistance.)

This unit would pump fresh water out at 600 psi x 10 gph. (The 600 psi is required to overcome the head that was required to overcome filter resistance.)

In the conventional unit we need to pump 10 gph of salt water for a yield of, say, 2gph (presuming an efficiency of 20%).

Whereas in the suggested unit we need to pump only 2gph of pure water.

Is not the energy saving ratio 10:2?

What would happen if you built a container, submerge it to a depth where the increased water pressure would operate the system. Instead of using a mechanical pump to supply the pressure, you used a DC current to break down the water, using the gas to pressurize the system.

This idea is "borrowed" from A.C. Clark in "Raise the Titanic"

Depends which takes more energy.

Pumping the fluid through the RO filter, which is the norm, or lifting the filtrate from the depth of the filter.

You'll need at least 600PSI to filter seawater, so, you'd be pumping from about 60-80 feet down. Don't check my math.

The energy consumed is less than with a conventional unit. Two reasons:

1. In a conventional unit, the volume of pressurized salt water that must be pumped is several times larger than the volume of filtrate recovered. Most of the energy in the pressurized salt water that does not go through the membrane can be recovered via mechanical pressure exchanger, but not with 100% efficiency;
2. The fitrate -- fresh water -- is less dense than salt water. Therefore the head required to pump it to the surface is less than the head of salt water at the depth of the membranes.

You can take this idea a step further. If you go deep enough, the difference in pressure between the salt water column and the fresh water column will be sufficient to drive the RO. You'll have a fresh water "spring" pouring from a pipe in the deep ocean, powered by the potential energy of descending salt.

This is actually a nice way to desalinate water at very low cost. But you'll need to develop tele-operated robots to do filter installation and maintenance at the bottom of the ocean.

Send me a check after you've made your first $billion.

It strikes me, that either all those engineers designing desalination plants are utterly stupid, or somebody tries to set up an over-unity (perpeetuum mobile) case here.

Hm, deep thinking, hm, hm, let's see.....

The ionic solution - that seawater is - have already separated Na, Cl, H2O by atomic distances. Now, applying a sufficient number of Maxwell's daemons we can sort the molecules properly: water this way, the rest that way. Please do not pester me here, what feeds and energizes those daemons. I am focused on something bigger.

That is partial pressure due concentration. When you transport a molecule, or anything against a pressure differential, you expend energy accordingly. No matter, what method you are choosing.

You set up a scheme to counteract differential pressure with gravity, you pay the same with pumping up against gravity.

The energy of the (ionic) solution is the lowest energy point. Everything is up from there.

I am recently short on daemons, so that is out too.

------------------------------------------------------------------

TANSTAAFL. There ain't no free lunch therabout.

It's definitely not perpetual motion. I mentioned what powers it -- gravitational potential energy of salt moving to greater depth. If you ran it for a while in a closed (but very deep) tank, it would eventually stall out. You'd end up with the lower portion of the tank filled with a heavy brine, and the upper portion filled with fresh water.

It's also not (presently?) an especially cost-effective way to build an RO desalination plant. The cost of energy required to pump high pressure salt water through the RO membrane is a minor part of the cost of the desalinated water. The major part is in capital cost of the plant and membranes, and in the need for maintenance and periodic replacement of the membranes. Locating the guts of the plant under several thousand feet of water does not make cleaning and replacement any easier.

It will work, though there is no need to do it at depth as access for maintenance becomes incredibly difficult.

There are seawater RO plants all over the world, including a major one at Beckton in east London, another Larnaca in Cyprus (to mention just two land-based ones), and in the hulls of many thousands of ships and oil platforms. One still needs more than 40bar to drive one, and it doesn't matter if that pressure is obtained by lifting the water from depth or whether it is used to squeeze the water out of the seawater at the surface.

The idea is a little impractical, though good luck with the approach and please post some images back on this thread when a commercially-viable prototype is in operation. CR4 readers are always keen to learn.

In a recent YouTube discussion, a post by Matt Ferrell., with 980,096 views as on May 13, 2025, goes as below.

How Deep Sea Water is Now Drinkable. Take your personal data back with Incogni! Use code UNDECIDED at the link below and get 60% off an annual plan: http://incogni.com/undecided Freshwater may soon flow from the deep sea to drought-stricken coastal cities. A handful of companies are betting that crushing deep sea pressures can replace the energy-hungry pumps and toxic chemicals that are an inconvenient truth of desalination. By going the extra quarter-mile into the ocean, they’re pioneering a way to turn saltwater into freshwater — slashing costs, curbing pollution, and reducing environmental harm in the process. How on earth…or, under earth…does this work? And could it solve our global water crisis?

Is the concept I suggested still contemplated?

One problem could be that the water outside the membrane gets more concentrated in salts, slowing the process down, perhaps to a stop. Might not be a problem if there's a strong current, but that would add to more structural issues.

How about 'Back-washing' at periodical intervals?