Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
This is what happens when one confuses "random" with "profound."

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Supersimple reverse osmosis

Needed: pipe, pump, membrane.
  (+23, -4)(+23, -4)
(+23, -4)
  [vote for,

For some reason I thought the below described method was common knowledge. But it is not easy to google, so maybe some good discussion will come of a posting.

Stretch a semipermeable membrane across the end of a pipe. This membrane would be the type used for reverse osmosis, allowing noncharged water thru but not ions. Reverse osmosis is what happens next. Walk with your pipe to the end of an ocean pier and lower the membrane end into the water. At a certain depth, the pressure difference between the air filled pipe and the outside water will be enough to force fresh water thru the membrane, where it will accumulate in the bottom of the pipe. The membrane might need to be buttressed against the pressure with an underlying mesh or cap.

Attach pipe to pier. You now have a well. Pump out the water. If you need more water, put another pipe down. There is no accumulation of brine, although a large enough operation of this sort could increase the salinity of an area without much water movement. In which case you would need to lower the pipe further down.

Such an operation could be set up on an offshore oil drilling platform. The amount of water provided is limited only by the amount of pipe and membrane available.

addendum 8/11/06: [Freefall] posted an anno to this idea positing that given a deep enough pipe, a reverse osmosis desalinator could be built such that the freshwater rises to the top of the pipe and fountains up from the top. This is even slicker than my prosaic desalinator as above. Much of the discussion following that anno concerns [Freefall]'s freshwater fountain, not my idea as posted above.

bungston, May 07 2004

Geothermal distillation http://www.halfbake...ater_20for_20Africa
Inspired by interesting discussion here. [bungston, Oct 04 2004, last modified Oct 05 2004]

Howard's link http://www.lhup.edu.../museum/osmosis.htm
link service provided free of charge [Freefall, Oct 04 2004, last modified Oct 06 2004]

Desalination company http://www.lenntech...tion-RO-modules.htm
They assert 50-60 bar pressure needed for reverse osmosis pressure of seawater. [bungston, Oct 04 2004, last modified Oct 06 2004]

Pressure conversions http://xtronics.com/reference/convert.htm
To convert bar into inches of water. [bungston, Oct 04 2004, last modified Oct 06 2004]

Ocean data http://mbgnet.mobot...alt/oceans/data.htm
How deep is the marianas trench? [Freefall, Oct 04 2004, last modified Oct 06 2004]

Degassing Lake Nyos http://www.pbs.org/...no/01/indexmid.html
Seltzer from a lake [Freefall, Oct 04 2004, last modified Oct 06 2004]

The Lake Nyos Fountain http://perso.wanado...alb/nyos/webcam.htm
Webcam view of the Nyos fountain so the scientists can monitor it. [Freefall, Oct 04 2004, last modified Oct 06 2004]

Osmotic Energy http://exergy.se/go...j/98/osmotic/#intro
A related subject [ldischler, Oct 04 2004, last modified Oct 06 2004]

Ocean Salinity Profile http://www.windows....salinity_depth.html
Not constant, but close enough... [ldischler, Oct 04 2004, last modified Oct 06 2004]

(?) Tremendous variability in salinity in Gotland Deep http://www.smhi.se/...tkonferens/Ralf.pdf
See pages 12-14 (PDF file) [ldischler, Oct 04 2004, last modified Oct 06 2004]

(?) Offshore Drilling http://www.gsfdrill.com
Deep water drilling... alive and "well" (sorry about that) [zigness, Oct 04 2004, last modified Oct 05 2004]

APEC domestic water filters http://www.freedrinkingwater.com/
77 cents per gal. Why are we sinking an 8km pipe? [FloridaManatee, Oct 04 2004, last modified Oct 06 2004]

A primer on Diffusion. http://en.wikipedia.org/wiki/Diffusion
A law of Physics [jhomrighaus, Aug 09 2006]

(?) Cesium Chloride Seperation http://www.bio.com/...tocol.jhtml?id=p571
See Part C. [jhomrighaus, Aug 12 2006]

Solvation http://en.wikipedia.org/wiki/Solvation
Where does the energy come from? [ldischler, Aug 14 2006]

Salinity versus depth http://www.windows....depth.html&edu=high
Salinity is highest at the surface, due to evaporation. [spidermother, Dec 26 2008]

Salinity vs depth again http://www.es.flind...roOc/lecture05.html
change is miniscule, about 3g per kg water [TheLightsAreOnBut, Jul 27 2009]

Water Density Calculator http://www.csgnetwo...om/h2odenscalc.html
[ldischler, Jul 27 2009]

The Deepest Hole http://www.damninte...om/the-deepest-hole
12.5km, only 2.5km short of the end of the upper crust [TheLightsAreOnBut, Jul 27 2009]

A "perpetual salt fountain" as an energy source http://en.wikipedia...wiki/Salt_fingering
"Once primed by moving cold fresh water upwards, such a "perpetual salt fountain" would be able to draw energy from the local ocean stratification. Such a perpetual salt fountain could be a renewable energy source similar to a solar updraft tower." [ldischler, Oct 21 2009]

Deepwater drilling http://en.wikipedia.../Deepwater_drilling
...just joking [pashute, Jun 24 2013]

The practice of deep earth brine disposal http://www.newyorke...weather-underground
Scientifically nondense, I am afraid. Larded with sociopolitical commentary. [bungston, Apr 21 2015]

Size holes for various filtrations http://www.h2odistr...-particle-sizes.asp
Big enough for water, but too small for salt. And Baby bear said... [popbottle, May 03 2015]

California Startup Hopes to Harvest Desalinated Drinking Water from the Ocean Floor https://news.slashd...rom-the-ocean-floor
OceanWell says its technology can use up to 40% less energy by harvesting the water in pods placed at depths of about 1,400 feet, where naturally immense water pressure can help power the filtration process... [xaviergisz, Sep 26 2023]

Force required push open a 'drawer' at depth under the ocean https://physics.sta...-buoyant-force-work
[xaviergisz, Sep 27 2023]


       Clever. You'd need a powerful downhole pump, though, to pump the water up that huge head. And if you had the pump, you could do it directly, and dispense with the pipe. (And this will be a very long pipe, since you have to go down a couple thousand feet. So, if you're not near a trench, there will be a long horizontal run.)
ldischler, May 07 2004

       The pump would not need to be any more powerful than those used on dryland wells. I expect this device will actually be considerably shallower than dryland wells - you don't have to be that deep to have a really big pressurehead.
bungston, May 07 2004

       Eh? It's about 1/2 psi per foot, wherever you are. And don't forget—the pressure needed is much greater for seawater, as compared to purifying well water. Ironically, the pressure you need to pump it from the bottom of the pipe to the top is exactly the pressure you’d need to run the same set-up at sea level, so you gain nothing using the pipe.
ldischler, May 07 2004

       Why would you need to pump from the bottom of the pipe? Am I missing something, or would the pipe not fill to sea level?
justaguy, May 07 2004

       So the idea is to take advantage of the difference in density between sea water and plain water over a great depth to drive the flow? Could work, but as howard said, you'd need a L-O-N-G pipe. Plus, you occasionally need to replace the filters.   

       Edit: checked howard's calculation, I also get approx. 8000 meters (8268 meters, to be exact.) That's a lot of pipe, but it could work.   

       Assuming that we use a large desalinization plant at the bottom of the pipe, buoyancy of the pipe itself shouldn't be a concern. Assuming a non-reusable filter, one filter assembly could simply be attached to the outside of the pipe and allowed to sink down the line and hook up, allowing the old assembly to drop off the end of the line. If recyclability or disposal issues are a concern, the old filter assembly could simply inflate a buoyancy bag to return to the surface.
Freefall, May 07 2004

       What you're missing is that the pressure is not hydrostatic pressure--it's a pressure differential. If you have an equal head on both sides of the membrane, nothing happens (except regular osmosis). You have to exceed the osmotic head to get reverse osmosis, and for seawater that's a lot of pressure.
ldischler, May 07 2004

       ldischler, there is a pressure difference across the membrane due to the difference in density of seawater and fresh water. I'll show my math.   

       Assuming it takes 20 bar to begin reverse osmosis:   

       d=depth of pipe
P1=pressure at the inside of the pipe at the bottom
P2=pressure at th outside of the pipe at the bottom
p1=density of fresh water = 62.428 lb/ft^3
p2=density of sea water = 1.025 * p1 1 atm = 14.7 psi = 2116.8 lb/ft^2

       P1 = p1 * d
P2 = p2 * d = P1 + 20 atm
p1 * d + 20 atm = p2 * d
20 atm = p2 * d - p1 * d = d * (p2 - p1)
d = 20 atm / (p2 - p1) = 20 atm / (0.025 * p1)

       d = 20 (2116.8 lb/ft^2) / .025 * 62.428 lb/ft^3   

       d = 27126.3 ft
d = 8268.1 m
Freefall, May 07 2004

       Sehr interresant... So what do you make of the 8000m number?
justaguy, May 07 2004

       Yah, no fountains are going to shoot out of the ocean. I did some math. These folks (linked) assert 50-60 bar of pressure is needed for reverse osmosis of seawater. That is, to squeeze the freshwater out of the salty stuff. Converted, 60 bar is 24060 inches of water, or 2005 feet. So that is a pretty long pipe - I guess I am not going to be carrying it down to the pier. But 2000 feet should be achievable within a few miles of the california coast.   

       Assume a pipe 2050 feet long. When the freshwater on the inside rises up to a depth of 50 feet, the pressure difference between inside and outside will no longer be adequate to drive reverse osmosis. You will need to pump that water up from a depth of 2000 feet, which is exactly as [ldischler] said: damn deep.   

       The plus: no brine accumulation! Which I understand is a big problem on land.
bungston, May 07 2004

       So bungston, are you refuting your own idea? 2000 feet isn't going to get you much, but if you build your plant in the mariana trench (36,200 ft) or puerto rico trench (28,374 ft), you should be able to get an unlimited supply of free-flowing fresh water.   

       Sadly, if it does take much more than 20 bar to start the process, there's nowhere on the planet that this would work.
Freefall, May 07 2004

       Freefall, it looks like you have the pipe full to the top with freshwater, and you are running this off the density differences between fresh and salt. Which is very cool, because you would not need a pump and it might overflow as a minifountain. It makes me glad I posted this.   

       But that is not the idea as posted. This is brute force reverse osmosis, run off the pressure difference between a column of seawater and a column (pipe) of air. You need less pipe to do it this way, making it more widely applicable.
bungston, May 07 2004

       Yes, the pressure difference between a column of air and water would let you do it, but then you run into the same problems already stated: you still need to pump the water up out of the hole, meaning it would be cheaper and easier to just have everything on land.   

       Even if my implementation isn't exactly what you had in mind, it's probably the most thought-provoking idea I've seen here in a while. I don't think I've ever seen an idea get posts fast enough to act as a near-real-time discussion. I only wish I had more than one bun to give.   

       Upon more thought, it may not take 27000 feet of pipe to do this. Think about dissolved gases. If the dissolved gases that permeate the membrane begin to come out of solution as they rise up the pipe, they may begin to act as a natural bubble pump that keeps it going.   

       (See Lake Nyos link)   

       A similar example was covered a couple years back where archaeologists uncovered that the area surrounding lake Nyos (in Camaroon) appeared to be hit by a massive explosion every several hundred years. The scientists discovered that the lake was unusually high in carbon-dioxide concentration, and the pressure at the bottom of the lake would cause the water to become supersaturated. At this point, an occurance of seismic activity would cause the CO2 in the lake to begin coming out of solution (like shaking a soda bottle). As it would rise up, it would become depressurized (much like taking the cap off said shaken soda bottle) and even more would come out of solution. Once this started, the entire lake would explode with expanding CO2, decimating the surrounding forest and any local villages.   

       The solution was remarkably simple: the scientists took a long pipe with a buoy at one end and jagged edges at the other (to make turbulence). When lowered into the lake, they pumped some water up through the pipe to start the process. As the water de-gassed inside the pipe, the difference in density was enough to keep the process going, resulting in a fountain shooting 20-30 feet in the air.   

       By continually bringing the supersaturated water up in a controlled manner, they're hoping to prevent another spontaneous degassing explosion.   

       I think that something similar may happen within such a long pipe, but it would take more knowledge than I have about dissolved deep ocean gases to answer that question with any confidence.
Freefall, May 07 2004

       [Howard] and [freefall] Did you take into account the change in salinity with depth (which the web site mentioned but didn't quantify).   

       I don't actually know how much the salinity changes with depth, but I suspect that it would exactly cancel out the boyancy of pure water in salt water.   

       Consider this: If you put two tubes side by side, one very deap, and the other shallow, and it didn't cancel, then the water level in one would be different than the water level in the other. If that happened, you could connect the two and let the water flow, and harness a little energy. In the process however, you would be moving pure water from deep to shallow, increasing the salinity of the deep ocean until the extra pressure needed for reverse osmosis balanced out the height difference. In reality, nature has already reached that balance, so there's no energy there to be harvested.   

       Conclusion: no matter how deep you put the pipe, the fresh water level will always be at the same distance down, even when you take into acount the boyancy of fresh water.
scad mientist, May 07 2004

       Supersimple, eh?
waugsqueke, May 07 2004

       It looks to me that howard and Freefall are right (except possibly for the assumption of 20 bar)—you could desalinate at no energy cost if the ocean were only deep enough. It's rather counterintuitive. And if there were sufficient dissolved gasses, that would make it a lot easier.

See also the link on Osmotic energy. The water flow in this pipe could likewise be used to generate energy. Where this energy comes from is a good question. (Negative heat of solution of the salts?) But there's a feeling of perpetual motion about it, because you could just dump the fresh water back in the ocean and keep going forever. God doesn’t like that.
ldischler, May 07 2004

       If physics didn't exist it would be nessecary to invent them.
justaguy, May 07 2004

       Alternative idea: lower a sealed pipe by rope. When it is full of fresh water, pull it back up.
Ling, May 08 2004

       The free energy implication is disconcerting, but I can't immediately see why it wouldn't work. I was wondering if you could model this with less extreme depths by using liquid much saltier than seawater, this making the difference more extreme. But I suspect as the amount of solute rises, the pressure needed to do reverse osmosis also rises because of the osmotic pull of all that solute. This is, I think, how the bubbles help - instead of increasing the density of the salty column, with bubbles you effectively decrease the density of the freshwater column.   

       Yeah, [Freefall] I know I just reiterated what you said. But I was so proud that it finally made sense to me.   

       But back to the model issue. It seems like there must be some way to model this osmotic energy thing on a more reasonable scale. Maybe this could be done using really heavy molecules as your solute. As regards osmotic pull, a molecule is a molecule - as I understand it a 1 molar solution of NaCl and a 1 molar of PbCl would have the same pull. But the lead salt solution would be a lot heavier.
bungston, May 08 2004

       Sleeping on it, I realized there would only be a perpetual motion problem if the ocean were in equilibrium, which is isn’t, because it’s constantly stirred up by currents, and those currents are driven by the sun, and by volcanic, tidal, and seismic energy. So there’s nothing mysterious about it.

If the ocean were in equilibrium, the method wouldn’t work. If you totally isolated a column of seawater miles high, so that it became quiescent, subject only to ordinary diffusion, eventually salt would precipitate out at the bottom, and the water at the surface would become fresh. In that case it would be impossible to get a flow in bungston's pipe, since the osmotic head would always be precisely enough to keep it from working. So it’s only the stirring up of the ocean that makes it possible. Mechanical energy that eventually derives from natural sources. No extra magic needed.
ldischler, May 08 2004

       Would it precipitate though? And what percentage? There's a coefficient of chemical balance in operation here of some sort.
RayfordSteele, May 08 2004


You could even use this notion to back calculate what the equilibrium profile of solute would be.

But getting back to the main idea, another way of making the pipe idea practical would be to build it near a volcanic vent, allowing the first leg of the pipe to pass through superheated water, while insulating the rest of it. At 80C you’ve already doubled the density difference, and potentially you could go much further—to several hundred degrees.
ldischler, May 08 2004

       I'm with [Ling] on this one. Who needs a pump or a perpetual diffusion machine, when we could go deep-sea fresh water fishing? The cylinder could be small enough to be carried on sailboats, or on inflatable emergency rafts.
Laughs Last, May 08 2004

       A small cylinder...and a thousand feet of rope.
ldischler, May 08 2004

       In defense of the idea as it has evolved:
First, the link is wrong.
Second, the pipe is not buoyant.
Third, the whole point of this is to do it without a pump.
And fourth, the economics are insane, sure, but so what? Eventually, the cost of oil will reach $100/barrel, and then all sorts of formerly insane technology will get funded.
ldischler, May 08 2004

       [Toadinnov] Isn't that exactly what they do when they drill for oil offshore? Save the contamination worries, of course.
justaguy, May 08 2004

       [laughs last] I didn't think about the emergency use for fishing for fresh water, but you are right: what a wonderful application.
Ling, May 08 2004

       Just to be clear, there are 2 ideas cohabitating here - my original supersimple version, which uses a pump and 2000 feet of pipe, and the more exotic one, which uses no pump, 27,000 feet of pipe and holds the promise of free energy.   

       As regards the simple version and [Toads] complaints. I agree with expensive and large, but not sophisticated. It is a pump and a pipe! The best way to deal with the troubles of oceanic pipes is to adapt oil well platforms for this use. At the very least, such an outfit could supply the oilmen with water, as opposed to having it shipped in. I agree that transoceanic pipes can be dodgy. But still, such pipes could be built with 1960s technology.   

       Back to the free energy. Salts do not precipitate out of solution. Absent evaporation, a flask of salt solution will stay salty. A solution is different from a suspension. Answering [ld]'s overnight thoughts on the free energy aspect lets consider two tall pipes: one full of saltwater and within it, another full of freshwater. The reverse osmosis membrane is at the bottom. Freshwater spills out the top of the interior pipe, mixing again with the saltwater. God frowns.   

       [ld] proposes that absent weather, the fresh and saltwater will stratify out in the salt tube, leaving the top fresh because of replenishment from the fresh tube, and the bottom of the salt tube presumably extra salty. This is an osmosis topic! The fresh water will be osmostically pulled down into the saltier stuff. Conversely, mass action will make the dissolved salt ions move up into the fresher stuff. No wind, rain, agitated jellyfish or other agents are needed. It still seems like there would be perpetual motion.
bungston, May 08 2004

       True, bung, you’ll never see a glass of salt water stratify, because random thermal motions won’t let it. To see that happen, you’ll have to have something as deep as an ocean. Look at my link showing the Salinity Profile. The deep ocean is saltier, becoming less salty as you approach the surface, and this trend continues until about 500 meters from the surface, where it suddenly gets salty again (because of evaporative concentration). Without evaporation, the curve would continue off to the left as you follow it from the bottom. And this effect would be much greater if there were no currents mixing things up.

As for salts not precipitating out of solution, sure they do, if the concentration is high enough—if the water becomes supersaturated.
ldischler, May 08 2004

       "...lets consider two tall pipes: one full of saltwater and within it, another full of freshwater. The reverse osmosis membrane is at the bottom. Freshwater spills out the top of the interior pipe, mixing again with the saltwater. God frowns."
But you're still presuming the presence of gravity, expressed as pressure. It's the difference in pressure that causes the flow.

       It's the whole Lake Nyos thing, but with H2O instead of CO2. You can accomplish the same effect for the same reason, you just need more relative pressure.
phoenix, May 09 2004

       With the 8000m pipe idea, it is interesting to think about where the energy is coming from.
In the two pipe theory where "god frowns", the salt concentration at the bottom of the outer pipe will increase as fresh water is pumped into the top. The Osmotic pressure will increase and stop the flow, until the fresh water and salty water mix again.
So the driving energy is in the mixing of fresh and salty water, against the natural tendency for more concentration at lower levels.

Or something like that.
Ling, May 09 2004

       It seems like the "perpetual motion" osmotic energy machine is actually driven by the forces which mix up the ocean - currents, winds, tides etc. So it actually could run forever, or at least until those things stop.   

       I am thinking about the Black Sea, which I understand does have a supersalty, anaerobic dead zone in the depths. This happened as a result of a lack of mixing. If the osmotic engine is positioned in a deep ocean trench (is it would have to be), it could happen that the deep pocket containing the tube end might get saltier.
bungston, May 09 2004

       //Oil platforms work in shallow water//   

       [Toadinnov], Ever hear of deep water drilling? see link.   

       <quoted from link> Our fleet includes... Three dynamically positioned, ultra-deepwater drillships, two of which are designed for the ultimate drilling capability of up to 12,000 feet of water <qfl>   

       Also, a pipe is not necessarily bouyant just because it's not full of water... I'm not sure what you mean by that one. The pipes used in oil exploration are a lot of things, but they sure aren't bouyant.   

       I've been out on 10 different oil rigs by the way -- I will agree that the economics are rather severe.
zigness, May 09 2004

       Whatever. He just didn’t seem very thoughtful. More of a knee-jerk debunker. And I’m wondering if that’s his job. Does somebody actually pay him to debunk?
BTW, I started off thinking this idea was bunk too, but then some people here--people a lot smarter than me--convinced me I was wrong.
ldischler, May 10 2004

       At first glance, it seems like it would be possible. However, taking into consideration the increased salinity at depth (which would likely increase osmotic pressure to well above 20 atm) and the possibility of stagnation at the bottom of the trenches (the only places deep enough for this to work) I'm thinking this idea may not be feasible.   

       While there may be locations that would make this more feasible (possibly shallower water near volcanic vents) it seems to be stuck in the realm of technologically possible but logistically impossible.
Freefall, May 10 2004

       Did that [Toadinnov] delete his account? Hopefully not because of stuff in this idea! A shame either way.
bungston, May 10 2004

       What's a shame is that people like him--only interested in casting aspersions and ridicule--are in such positions in government. And it's amazing that he'd seen this idea before--and shot it down--and yet had so little understanding of it.
ldischler, May 10 2004

       Oddly enough, we do.
ldischler, May 11 2004

       Drill holes in the side of the pipe, cover entire pipe with membrane, more fresh water. +
DesertFox, May 11 2004

       Yeah, that's what [Toadinnov] apparently thought.
ldischler, May 11 2004

       It's just a fundamental difference in attitude, I think. I see the proposal and try to think of solutions to the problems of depth and buoyancy. He sees them as reasons why it wouldn't work. Not that this is a bad thing - the world needs people like him or people like me would be selling perpetual motion machines on the internet. Just different.
justaguy, May 11 2004

       I worked on a large reverse osmosis project in 1999 and the engineering necessary is staggering for what seems a simple process in terms of physics. In the end, the engineering consultant scrapped the HP stages, relying on filtration and cationic systems.   

       So I'm afraid I'm with the naysayers.   

       First, if the pipe is full, there will be no pressure differential across the membrane. Therefore no reverse osmosis. If the pipe is to be emptied and the pressure maintained, you're pumping water up an 8km pipe.   

       Secondly, have you any idea what an 8km depth submarine pipe costs to build and maintain? You'd be making the most expensive water on the planet.   

       Finally, there's only a few places in the world you could even test it.   

       I'm having trouble finding financing for even straightforward biomass and wind projects. They don't make financial sense in many parts of the world.   

       As for an 8km depth pipe? Dream on!
FloridaManatee, May 12 2004

       //As for an 8km depth pipe? Dream on! //   

       So would that be a pipe dream? +
sartep, May 12 2004

       Hello [Manatee]. I have 2 things to correct in your complaint. 1. "If the pipe were full, there would be no pressure differential." The pressure differential is from the difference in weight between a column of fresh and a column of salt water. The difference is not huge, which is why such a long pipe is necessary to parlay that small difference into enough pressure to run r.o.   

       2. "If the pipe were empty you would have to pump the water up 8 km.". Actually only 2000 feet - see my math above. The difference in weight between a column of saltwater and a column of air is a lot greater, so the pipe does not need to be as deep. There are lots of places where open ocean is 2000 feet deep.   

       I would be interested to see a summary of the engineering challenges your team encountered in what I presume was a land-based r.o. desalination project.
bungston, May 12 2004

       [DeGroof] - that is a good link. It lists 27 bar as the osmotic pressure for seawater. I used 60 bar to get my 2000 foot pipe. But the excess pressure is only necessary because the osmotic pressure of concentrated seawater is greater, and so the system needs to exceed the osmotic pressure of the most concentrated fluid to keep that fluid from pulling fresh water back across the membrane.   

       In the open water pipe, no concentrated saltwater accumulates. Therefore the pipe only needs to generate 27 bar - 910 feet deep!
bungston, May 12 2004

       Convert it to steam—a truly halfbaked suggestion! (If you did that, you would already have fresh water.)
ldischler, May 12 2004

       There is another aspect of reality that has been overlooked in this discussion. It's the biota of the ocean. Almost anything that lies in the ocean gets encrusted with innumerable living and formerly-living things (think barnacles). How do you keep the fancy semipermeable membrane clean so that it can work? My only answer is to sink the pipe into a "dead" zone of the sea, such as the anoxic area in the Gulf of Mexico. Then at least you'd have fewer critters gluing themselves onto the membrane.
Genki, Jul 15 2004

       Why not set up a "vortex" pump using the temperature differential of the water. A lot cheaper and less clumbsy than 8000 metres of pipe
tasman, Jul 15 2004

       More on vortex pipes, please, good [tasman].
bungston, Jul 15 2004

       I’m bumping this idea up because it’s so brilliant and counter-intuitive. And because it occurred to me just now that the discussion about salinity equilibrium could explain the formation of salt domes. The usual thinking is that a sea got cut off and dried up. But if the oceans were frozen over (the Snowball Earth theory) there would have been no solar input to the oceans for millions of years, and no mixing currents to keep salt from precipitating out. The upper water would have become fresh, while a layer of salt would have covered the bottom everywhere. This would have been reversed when the Earth warmed up again, except where sediments covered it up quickly, such as near the outflow of rivers.
ldischler, Aug 09 2006

       What if you created a cone shaped outer body inside of which you construct a helical, spiral membrane. Attach this assembly to a pipe with a free spinning coupler. This whole rig can then be suspended in an area with a current flow. As the current passes through the membrane the assembly would rotate like a turbine, this rotation would power a positive displacement type pump that would accomplish 2 goals, 1 it would move the fresh water up the pipe and 2) it would reduce pressure inside the membrane to increase production at shallower depths and allowing for maximum production at all times. This would also increase production by ensuring fresh seawater at the interface at all times and at a slightly higher pressure as the water is forced through the cone.   

       The motion of the assembly should also drasticly reduce the buildup of encrustations on the membrane. A manual backflush cycle could be setup as well that would reverse the flow of the pump to over pressure the membrane and help keep it clean between cleaning cycles. This might allow for high production and significantly lower depths with no outside input of energy. Placed in a location with a constant flow, these modules could produce constantly.   

       Serviced by a floating barge that is very low to the waters surface would allow for thier placement in locations near land with no deleterious visual impact. Exisiting pipeline technology as is used for oil wells would allow the water to be pumped back to land.   

       Any thoughts?
jhomrighaus, Aug 09 2006

       Fine, if you must. But it ruins the simplicity of the idea.
ldischler, Aug 09 2006

       As presented the idea is grossly inefficient and doesnt really address the cited objections. I was offering a variation that is really quite simple and addresses many of the issues cited.   

       The concept is simple, but execution is anything but.   

       [ldischler]Why did you bump the idea if you are going to poo-poo any discussion of the topic.
jhomrighaus, Aug 09 2006

       //Why did you bump the idea if you are going to poo-poo any discussion of the topic.//
If it seemed I was poo-pooing, it was because your idea is only peripherally related to the original idea. There's so much that could be discussed that it would make more sense to post it separately. (Anyway, this is the halfbakery, not a corporate brainstorming session, so poo-pooing is allowed.)
ldischler, Aug 09 2006

       //promise of free energy//   

       Its not free energy it would just be free to us. You still need to expend energy dissolving the salt into the water to create the salt water in the first place. There is also a small cost in the passage of the membrane as well. But if, Theoretically you could find a deep enough place in the ocean you could get a steady flow of water to form at a pressure slightly lower than the overall differential less the Osmotic presure of the seawater. But as stated you would need 10s of thousands of feet to accomplish this.   

       As to the points on Stratification of Salt water, this is not a process that occurs spontaneosly. If you begin with a fixed concentration volume of Saltwater and place it in an enclosed tube, the concentration throught the tube will be the same for all time. This is basic diffusion. The solution is at equalibrium becase it is a solution, not a suspension(which is a Physical Process) Stratification in the ocean occurs due to temperature differentals and Large masses of water. It can be simulated in a lab but as time goes on the solution will equilabrate and a uniform density will occur.
jhomrighaus, Aug 09 2006

       //it was because your idea is only peripherally related to the original idea//   

       As is 90% of the discussion on this page if that is your assesment.
jhomrighaus, Aug 09 2006

       //As to the points on Stratification of Salt water, this is not a process that occurs spontaneosly//Actually, it does, at least on large scales. It's ocean currents that keep things stirred up. Even so, there is quite a bit of stratification, esp. in the more quiescent waters of the Arctic ocean. Too much stratification, and this idea won't work. (And that this idea works implies that there must be stratification, absent mixing currents.)
ldischler, Aug 09 2006

       [ldischer] Please link some data that shows that a homogenious solution of saltwater placed in a column will spontaneously stratisfy. This certianly occurs in the ocean but it has a lot to do with moving water masses, temperature differentials and currents. In a homogeneous, static solution this is not the case. I could be wrong, but 10 years in chemistry and a degree later, i,m not sure how it would happen in the situation I described.
jhomrighaus, Aug 09 2006

       Look at page twelve of the Gotland Deep link, which shows the bottom salinity to be twice the surface salinity. This is more than the open ocean, presumably because there is less mixing in this area. That kind of variation is almost universal across the oceans, although the degree is generally less severe. See the "Standard Salinity Profile" in the link above it. Except for the surface water (made salty by evaporation--more evaporation than precipitation), the ocean typically gets saltier with depth, even with currents that tend to mix it up. Take out the evaporation effect (as in the cold Baltic waters), and the red curve would continue bending to the left, all the way to the surface.
ldischler, Aug 09 2006

       [jhomrighous], I like your spinner. It is an idea in and of itself but I appreciate your putting it under this heading so discussion of membrane desalination can occur in one place.   

       No-one doubts that one can use pressure to carry out reverse osmosis desalination. The question is whether you can rig it to operate without obvious energy inputs (the energy input here is probably the mixing of the ocean by currents etc, as discussed above). Your hybrid solution harnesses energy inputs from currents to sidestep problems with depth and mixing. As a different issue, it may be a good idea to have membranes in motion to avoid fouling with ocean life.
bungston, Aug 09 2006

       [bungston] Thanks that means a lot coming from you.   

       [ldischler] I have already stated that Ocean stratification is a proven process. The question is not does it happen it is a question of in a closed system(addressing some earlier discussions) will it stratify and i still say no to that.
jhomrighaus, Aug 09 2006

       Okay, have you ever seen that demonstrated? Can you provide a link where someone has isolated a solution for a time and measured the concentration top and bottom with precision instruments, and shown that there was no gradient? Because if there isn't, then we've got a perpetual motion machine here.
ldischler, Aug 09 2006

       Has anyone accounted for the Osmotic potential of the water back the other way? This would directly counteract the incoming flow of water once equilabration of the water column pressures has matched up(Fresh water taller than salt) Water will flow one way or the other until everything balances out. The more I think about it the more I think that this would equalize and no water would flow unless you removed water from one side or the other(requiring energy input)   

       If you started with a to matched columns one salty one not.(the fresh one taller by the prescribed amount to generate equal pressure at the membrane) with a membrane in between and a valve, what would happen when you opened the valve?   

       The answer is nothing, the colomns are balanced and there is no flow.   

       If you shorten the fresh column a little you would get some flow towards the fresh side but eventually it would equilibrate at the new level as the salinity of the salt side was reduced.   

       If you removed some fresh water then you would get flow towards the fresh side which would stop when the salinity stablized.   

       Remove salt water and flow would reverse to the salt side of the system until diluted sufficiently to equilabrate again.   

       I would expect some harmonic cycling in a non idealized system but in all cases it would eventualy cease to flow.   

       In a real world trial in the ocean I suspect it would look an awful lot like a perpetual motion machine, but in reality, the energy budget would be negative.
jhomrighaus, Aug 09 2006

       //The answer is nothing, the colomns are balanced and there is no flow.// You're neglecting the difference in density. Look at Freefall's calculations above.
ldischler, Aug 09 2006

       No Im not. In my example I cited the Fresh tube being higher than the slat tube. This difference was for the exact differential in height required to equalize the pressure at the interface of the tubes. From there follow the scenarios and you will see what will happen if tried.   

       If the columns are of equal height at outset you will see flow for a period of time sufficient to dilute the salt water to a concentration at which the flow of water at the membrane can no longer be sustained, at which time the flow will cease and the system will equilabrate.(it may oscilate in a non ideal scenario but it will equilabrate in the end) Keep in mind that such a scenario could lead to water flowing from the fresh to the salt side which could permit salt into the freshwater side of the system. Either way in the end it will equilabrate.   

       Freefalls calculation explains why it will look like a perpetual motion machine in the real world, but left to run for a sufficiently long period of time(eons and eons i suspect) the Ocean would eventually equilabrate with the water in the tube and the flow would stop.(RO pressure would increase as salinity increased, or vice versa the osmotic presure from the Fresh water would increase until eventual the system equalizes)   

       See link for discussion of diffusion which governs the Osmotic process. This is an energy cost free activity.   

       Perhaps another way to look at this is to say that, left to its own in a fully sealed system(neglecting heat losses) you would have a perpetual fountain, but the moment you try to extract ANY energy from the fountain you go negative an d the process stops(ie a turbine would increase backpresure and would force the flow back in the other direction and so no energy can be extracted.)   

       Also of note is when I say equilabrate that could mean either pressure equilabration, concetration equilibration or a combination of both depending on the scenario.   

       This is tough to visualize.
jhomrighaus, Aug 09 2006

       But it isn't two columns. Just one for the fresh side. The other side is the ocean. The top level of each is constant. Water spills from the fresh side continuously, with no energy input. That's what's so neat about it: fresh water by reverse osmosis, for free. (This is not the original idea--it's the better idea Freefall came up with. Bung ought to note that at the bottom of the idea.)
ldischler, Aug 09 2006

       Not free but at a cost that is not visible to your eye because it is spreadout over the entire ocean.   

       I agree that it is a very cool concept and could actualy work(assuming the engineering obstacles could be overcome) But it is certinaly not a perpetual motion machine. Which was what I was hoping to show with the closed loop example(perhaps I have failed miserably to do so however)   

       Bungstons point is still valid, this is brute force RO and as such i suspect that even a relatively small system of several hundreds of feet could be used to generate fresh water.
jhomrighaus, Aug 09 2006

       It's perpetual, but it's not perpetual motion, because the energy source is the non-equilibrium state of the ocean, which gets that way from mixing currents. And the energy for that ultimately comes from the sun.
ldischler, Aug 09 2006

       It is brilliant, and counterintuative. I'm still not sure I understand it though.   

       [ldischler] and [jhomrighaus], could you tell me if I'm getting close:   

       If I imagine a simplified model -an isolated system with a column of sea water and a column of fresh water linked by an osmotic membrane with a fixed pressure difference across it - at the top of the column, you get a head of fresh water, which could drive a turbine.   

       You could allow the fresh water (after extracting energy from it) to pour back into the sea water column. It looks like a perpetual motion machine now - it appears that we can extract work from the system, with no energy input.   

       But, after running for a while, the water at the bottom of the salt water column has become more saline - eventually, too salty for RO to function at that pressure. For the system to continue running, the fresh water at the top of the sea-water column has to be mixed with the salt water, either by stirring (requiring kinetic energy) or by diffusion (requiring thermal energy)   

       From the isolated system model, it looks like the energy in the ocean-bound system would come from the mixing of the ocean by currents.   

       So the energy output is actually coming from the gravitational potential energy of Na+ and Cl- ions (and others) that make the upper ocean waters more dense than they would be if allowed to settle without solar input.   

       Is that about right? I really have trouble getting my head around this.   

       [+] for the idea, and if I could, i'd [+] for the mental exercise too.
Frankx, Aug 10 2006

       That's it, except the column of sea water is just the sea itself (which I'm sure you meant). It'd be interesting to calculate the potential energy for the oceans--huge, I'm sure.

A similar effect has been suggested for the internal heat source of Jupiter--stratification of the originally well-mixed He and H2.
ldischler, Aug 10 2006

       To my knowledge there is no Gravitational Potential Energy associated with Na or Cl ions in solution. Thier mototion and diffusion within the water solution is an energy nuetral process. As such the static Model(and im am guessing here) may flow if left in an undisturbed state. The instant that any amount of energy is extracted will shift the balance and the entire process would stop and equilabrate.   

       I think the Salinity stratification in the ocean is significant at an operational level for this system, but when looking at an ideal closed loop system it does not enter into the picture.(if you took a colomn of ocean and isolated it, it would eventualy equilabrate and become uniform over its entire volume, as is dictated by the laws of thermodynamics.)   

       I am going to try and do some calculations on my lunch break and see if i cant quantify this question somewhat.
jhomrighaus, Aug 10 2006

       [ldischler] I think the thing that is confusing is that stratification comes about through the interaction of a complex system of energy flow causing variable potentials to move about the system. I think the other key is that these are short term(relatively) phenomnons and given time would give way to equilabration.(like food coloring in water, eventualy it is evenly distributed) This is the second law of thermodynamics at work and is the eventual state of the universe(a still quiecent pool of matter evenly distributed and equal in makup at all points devoid of all motion)
jhomrighaus, Aug 10 2006

       You have it backwards. It's the relative lack of stratification that's due to mixing currents. In a classroom you can say that diffusion is the only effect, and you won't be far off. But on large scales, you can't ignore gravity. Look at the earth itself, which was once molten. Did it stay all nice and mixed up?
ldischler, Aug 10 2006

       Sorry [ldischler], I should have been clearer: I meant to describe an imaginary system of two columns of water isolated (from the ocean/surroundings), to understand where energy was crossing the boundary of the system.   

       From that simplification, it was clear that energy must enter the salt-water column from the surroundings (from outside the system), causing the salinity at the top of the column to be higher than it would be otherwise (by stirring and/or diffusion).   

       So I saw that a column of sea-water of uniform salinity has a higher energy than a column with a salinity gradient from top to bottom, and that effectively, this is gravitational potential energy (of the ions contributing that density).   

       So the energy available from a “real” system – and there would be a real, useable energy output – comes from the elevated density of sea water in the upper layers of the ocean, elevated relative to the density gradient that would develop if there were no solar input to the ocean.   

       [jhomrighaus] – it’s true that there are a number of factors influencing salinity distribution - solar evaporation will tend to produce a higher salinity near the surface; Mixing by ocean currents will tend to produce homogeneity, as will diffusion. All of these factors oppose the tendency towards the gravitational distribution of low density to high with increasing depth – it’s gravitational “rest state”.   

       Each mechanism depends on energy input to the ocean from the sun. The energy is expressed as the displacement from it’s rest state.
Frankx, Aug 11 2006

       Hmm. Just done some rough calculations on the depth required. Taking a mean density of sea water as 1025 kg/m^3 and fresh water as 1000 kg/m^3, to develop a 60 bar pressure difference across the membrane would need a depth of around 24,000m. Challenger Deep is around 10,911m.   

       True, density will change with temperature and pressure, but I’ve assumed that the relative effect between fresh and sea water of these is negligible.   

       Even if osmotic membranes were greatly improved, it doesn’t look promising. Shame.
Frankx, Aug 11 2006

       Thanks, Frankx. That’s a lucid review. And yeah, the gods seem to have conspired against this. The absolute minimum pressure differential is >24 bar, but probably closer to 40. That makes the situation better, but still, the ocean is not that well mixed, and there are salinity and temperature gradients working against it.
ldischler, Aug 11 2006

       Croissant for making me think hard.   

       Is the best refutation that this perpetual freshwater well would not flow in a perfectly still ocean, because in the absence of stirring, the concentration of salt would go up with depth, so the osmotic pressure would go up at exactly the same rate as the water pressure? ?   

       (I'm Ignoring the practical fact the fact that the ocean is not deep enough.)
Heathera, Aug 11 2006

       If Salinity varies by depth due to gravity(as has been positid and I have refuted) Could someone please supply the calculation that indicates the salinity level at a given depth?   

       If this is the way salt water behaves in an isolated column of know length then this vavlue should be able to be calculated.   

       I have thrown the gaunlet and await someone to pick it up!
jhomrighaus, Aug 11 2006

       Just saying it ain’t so is not a refutation. But I’ll argue it like this. Lets say you’re right, without mixing currents, in a quiescent ocean, a salt gradient won’t exist because of diffusion. Okay, that’s great, because we can set up this long pipe and start making fresh water (given a deep enough ocean, of course) and let's say we dump it back in the ocean where it becomes salty again because of diffusion. We can do this forever! But that can’t happen, can it, because it’s perpetual motion. So one of two things are wrong: either this pipe cannot make fresh water after all, or there must be a salt gradient in your quiescent sea that prevents us from making fresh water this way. And this salt gradient should be just sufficient to keep this idea from working.
ldischler, Aug 11 2006

       The concept of a salt gradient goes against the concept of osmosis, which it would seem most posters here believe in. Just as we are using reverse osmosis to remove fresh from salt water, in a column of salt water, the osmotic pull of lower salty water should pull freshwater down from the top, with a consequent even distrubtion of salt. My google for ocean salinity shows the main differences in the ocean to be at the top, probably because of evaporative losses as was mentioned above.   

       Imagine a two column problem: one full of seawater and one full of salt. They are long and connected at the bottom with a membrane. A fountain results and the fresh overflows into the salt column at the top. The top of the salt column would become less salty and the bottom more so, and the fountain would decrease. However, the second law (entropy) states that even without external currents, ions in solution tend to move towards an area of less concentration, and so _even absent external currents and stirring_ the salt column will tend to equilibrate. As it does, and the bottom of the salt column gets less salty, the fresh fountain will periodically dribble over into the salt. Forever?   

       It looks like perpetual motion. It seems to me it is actually running off of entropy. Entropic engines are really heat engines. It seems to me that over time, these two columns will freeze and that will be what stops the flow.
bungston, Aug 11 2006

       It freezes! What ever happened to the heat? Did we destroy it? What if we fed in a little heat so that the water never freezes? Could we continue destroying heat forever? No, so the problem is still there--either this doesn't work, in spite of the math that says it does, or there's a gradient.
ldischler, Aug 11 2006

       [bungston] That is exactly what i think would happen, Diffusion and by extension Osmosis are "energy free" processes and so do not consume energy on thier own, however molecular motion is a function of thermal energy input and so with no input of thermal energy I believe that a freeze is what would ultimatly result. IF you could preserve the heat in a perfectly insulated environment then the fountain could theoreticaly "flow" forever, this is decieving though because it appears like perpetual motion but in reality is balanced motion. As nothing is ideal loses due to friction(water flowing in columns, molucules bumping etc.)you would slowly lose energy through heat loss and the system would stop and eventualy freeze up along witheverything else in the universe.   

       As to feeding it heat, I think the answer is Yes, we make up its losses and it will perpetually dribble water.   

       This would be a incredibly efficient molecular machine that like many other perpetual motion experiments would require a minimal input of energy to sustain.(think like really long pendulums)   

       Thanks for make my point for me [Bungston] I failed rather dismally.
jhomrighaus, Aug 11 2006

       /What ever happened to the heat? /   

       We used it to do work.
bungston, Aug 11 2006

       //This would be a incredibly efficient molecular machine that like many other perpetual motion experiments would require a minimal input of energy to sustain.(think like really long pendulums) // No no no! You cannot destroy energy! That's another law of thermodynamics.

//We used it to do work.//What work? If we stop everything, the ocean is exactly the way it was when we started, except we've mysteriously destroyed energy from the outside.
ldischler, Aug 11 2006


       You are not destroying heat the heat is radiating away(kind of like diffusion) to the surroundings, this is the fundmental process of Thermodynamics, everything will eventually equilabrate and there will be no differental anywhere. Due to the overall temperature of the universe and its infinite size and volume of matter and energy, the ultimate fate of the universe is an even distribution of all energy which will never reach absolute Zero but shall continue to grow colder and colder on into infinity.(ie it will never ever freeze all the way) For our purposes there will be no discerable energy to detect. The universe will be dead.
jhomrighaus, Aug 11 2006

       //The universe will be dead.//
And so is this discussion.
ldischler, Aug 11 2006

       I have been wondering how to model this on a more reasonable scale. One way would be to substitute a heavy solute for the NaCl in ocean water.   

       A problem is that the salt does not contribute that much to the weight of the water. Even using PbCl2 at the same molar concentration (0.56 M) produces a solution with a density of 1.156 - 112% the denisty of seawater, which allows a tube only 12% shorter.   

       I wonder if really large macromolecules could allow the density of the solution to greatly increase, while keeping the osmotic properties the same? For example hemoglobin has a molecular weight of 64000 - a 0.56 M solution would provide 35840 grams/liter of solute for a density of - can this be right - 36.84 kg/m2? This solute would allow for a tube 35x shorter - 775 feet, which is less than the Eiffel tower. Hemoglobin is big, but maybe some of those huge macromolecules could allow an even shorter column. Dextrans are cheap and have a molecular weight of around 200,000.
bungston, Aug 11 2006

       The discussion has been going on for over two years, and it seems to be alive and well again.   

       The question of the moment seems to be "is there a salt gradient?". Yes, there is natural diffusion in the ocean. But there is also gravity. Let's look at the individual NaCl ion in homogeneous solution. The osmotic pressure is equal in all directions, but there is still gravity pulling the ions downward. If there were a tall column of undisturbed salt water, there would be a gradient determined by the point at which unbalanced upward osmotic pressure due to the gradient exactly balances the downward pressure due to gravity. On a large scale, this is observable (look at the Gotland deep link).   

       There is also molecular motion due to non-zero temperature. At small (laboratory) scale, this motion totally overwhelms any gradient that may form, eliminating any measurable level of non-homogenaeity.   

       There is a saline gradient in the ocean. it has been measured. It's fairly linear, with a very shallow slope, once you go below the halocline.   

       If the salinity at the bottom of the two useable trenches is low enough that the pressure difference between sea water and fresh water is sufficient to drive osmosis, then this device will work.   

       If there are ocean-floor currents to continuously move the concentrated saline away from the bottom of the pipe, then this device will work continuously.
Freefall, Aug 11 2006

       [Freefall], since you have been good enough to weigh in again and it was your math I looked at, tell me how you think this could be modeled on a more reasonable scale?
bungston, Aug 11 2006

       [Freefall] If this is a function of gravity as you say then there should be a calculation to determine the change in solution strentgh based on depth and mass of the ionized molecule. Based on this information one should be able to calculate that the salinity of the water at XXX feet will be and that number should be a constant.   

       There is to my knowledge no such calculation and no such constants(doesnt mean they arent out there, but I have found no evidence of them anywhere) The observed gradients are not I believe a funtion of gravity but rather are a funtion of changes in the ocean salinity levels and incomplete diffusion process.   

       The question here is not is the a gradient in the ocean, but, if you put a homogeneous mixture of water and salt at a constant temperature into a column 27000 feet deep and 2 feet wide, what will happen. One side here says that the chloride and sodium ions will sink to the bottom and a gradient will form. There is no scientific justification for this position other than the statement that the ocean has a gradient. The other thought is that no gradient will form and a uniform solution will remain, unless some outside action causes energy transfer(ie heat or reverse Osmosis or something) There is scientific backing for this under the headings of the second law of thermodynamics and the principle of diffusion, which both predict a stablized concentration accross the entire solution(the definition of homogeneous mixture, or solution)   

       WHere is the scientific defense of the former position? where is the calculation that determines the nature of this gradient? Provide that, and I will consider myself educated and remove all my posts on the subject, but I have been looking for days now for any evidence to refute my position and I cannot find it. This is an intriguing question and I would love to know the true answer to the above question.   

       [Bungston]s idea is really neat, the overall concept would probably work, I even recommended some additional ideas to support it.
jhomrighaus, Aug 12 2006

       [Bungston] This could be modeled in a pressure chamber I suspect. Seal the two tubes and pressurize then adjust the pressure on the Freshwater side of the system to simulate a lower overall head(reduced water column Height) and see if the pressure begins to shift. If this would work then the pressure should change until the differential matches the calculated RO pressure and the difference in the water column pressure at which point it will stablize.   

       As far as real world the thing would probably be unable to flow free at the surface but if you pumped from a lower depth it would work no matter what the Salinty level(which doesnt vary that much)
jhomrighaus, Aug 12 2006

       There is a method for separating cellular fragments of roughly molecular sizes, where a concentration gradient of caesium chloride is established by spinning the solution in a very fast centrifuge. The cellular fragments float or sink to where they are neutrally bouyant. This clearly shows that acceleration (gravitational or otherwise) can create a concentration gradient.   

       This points to a method of producing a model - in a high speed centrifuge with ceasium chloride.   

       I guess the effect would be of short duration, unless energy were supplied to the external solution to oppose the formation of a concentration gradient, in which case you might get your minifountain.   

       Thanks [bungston] for this thought-provoking idea. I also like that no one here seems to have swallowed the incorrect explanation of osmosis as told in schools (in Australia at least). The correct version, as implied here, is that osmosis is a tendency for water to flow towards a region of greater solute concentration and that the osmotic pressure is equal-and-opposite to the plain-old pressure which exactly balances this tendency.
spidermother, Aug 12 2006

       [spidermother] Is this "solution" of Cellular fragments actually a solution where the particles disolve chmically into the water or is it a very very fine suspension of colloidal suspension which could be seperated in a centrifuge. If as you say these are "particles" of Cesium Chloride that due not dissociate into component Ions then this may not be a direct comparison. I will need to look into that example a little. Interesting thought though.   

       Keep in mind also that spinning in a centrifuge changes the rules a little bit because the gravitational acceleration experinced by the material in the Tube is Raddically different than normal and so things will not behave as they would in the real world where the acceleration due to gravity is constant at 9.8 ms2.
jhomrighaus, Aug 12 2006

       [jh] - I think the cesium example is pretty good proof. Cesium chloride is a soluble salt, and the acceleration provided by the centrifuge is no different than that of gravity except in degree. If a salt can stratify in a centrifuge than it certainly can under any acceleration, including gravity. [Freefall] laid this out above: gravity tends to counter osmosis. But like any force, on certain scales or situations it would be hard to measure because other forces overwhelm the effect.   

       Do not feel any obligation to delete your posts as you mentioned you might. They are fine.
bungston, Aug 12 2006

       If this is the case then solutions would not remain in solution. If you put food coloring into a cup of water it would settle to the bottom and you would get a layer of food coloring at the bottom. This is an all or nothing deal, it doesnt happen selectively, Gravity does not change with depth. You would not end with a gradient if this were the case, you would end with a full separation. If Gravity can not be overcome by some of the salt molecules it must be able to be overcome by all of them given time, but that is not what happens.   

       Gravity is a weak force and generaly requires Large bodies to have any meaningful impact on thier motion.   

       *Edit* Found some info on the Cesium Chloride Procedure and from what I can see the gradient is formed by centrifuging at 275000 gravityies.(holy crap batman thats a whole lot) And even at 275000 gravitys the most that happens is a slight gradient formation, the gradient only being able to partially impact the effects of Diffusion. Also keep in mind that Cesium is more than 5 times heavier than Sodium. It also appears as if the Change in angular momentum on the particles is what leads to the gradient and not the gravitation alone.
jhomrighaus, Aug 12 2006

       I don’t know about the diffusion/gravity debate, although it might be worth thinking about the atmosphere as an example of a gravity-driven concentration gradient.   

       My point earlier was that, in the “two columns of water isolated from outside influences” model, if the fresh water is allowed to flow into the top of the salt water column (having extracted work from it), the model will grind to a halt unless there is a flow of energy into the salt-water column from outside the system.   

       The energy crossing the boundary of the system could be: 1) thermal energy, giving energy to the ions in solution, and driving the diffusion of the ions, or 2) kinetic energy, in the form of stirring, to distribute the ions through the column.   

       Without some energy entering the system, some mechanism to re-distribute the salt ions in the sea-water column, we would be looking at a perpetual motion machine.
Frankx, Aug 14 2006

       Thinking more about diffusion, it seems to me that energy is required by the diffusion process. We could even calculate how much:   

       Imagine a cubic metre of fresh water (density 1000 kg m^-3) with a cubic metre of sea water below it (density 1025 kg m^-3), initially separated by an imaginary boundary, preventing any diffusion. The two volumes are perfectly insulated from the surroundings, and the water in each is perfectly still.   

       Initially, the centre of gravity of the system is 6.17 mm below the boundary between the two volumes: (1025*0.5-1000*0.5)/2025 = 6.17^-3 m   

       Now we remove the boundary (without disturbing the water), and allow diffusion to mix the two volumes.   

       After diffusion is complete, the volumes of water are completely mixed and have a uniform density. The centre of gravity is now exactly in the plane of the boundary.   

       The energy change of the system is +(2025 kg * 9.80665 N/kg *6.17^-3m) = +122.5 J   

       Because no energy has entered the system, the total energy of the system must be the same. Where did the 122.5 J come from? I suspect you would find, if you could measure it accurately enough, that the temperature of the water had dropped:   

       Cp = 417000 J kg^-1 K^-1   

       Delta T = 122.5/(417000*2025) = 0.145x10^-6 K   

       So the thermal energy required to drive diffusion is barely noticeable – an apparently “energy free” process.
Frankx, Aug 14 2006

       interesting approach frankx.   

       In the atmosphere the density gradient in air is not caused by changes in concentration but rather by compression. The reason we cant survive in the upper atmosphere is due to lack of air not lack of oxygen in the air that is available.
jhomrighaus, Aug 14 2006

       [Frankx] Another way to look at it--half of 2.5 kilograms of salt are lifted one meter. So that 1.25 kg m * 9.8 m s-2= 12.25 kg m s-2 = 12.25 N m = 122.5 J

So how much gravitational energy is tied up in making the oceans salty?

Average depth of the oceans = 3730 m
Total volume = 1.347 x 10^18 m3 of water

assuming a salt content of 2.5 kg/m3, that’s 3.37 x 10^18 kg of salt
To raise this to half the average depth
3.37 x 10^18 kg * 1865 m = 6.28 x 10^21 kg m
And that will take 6.15 x 10^23 J

Burning one barrel of crude oil gives 6.1x10^9 J
So the gravitational potential energy of ocean salt is roughly equivalent to one hundred million million barrels of oil.
ldischler, Aug 14 2006

       Ummm not much about the change in pressure differential as the pipe fills with water...   

       Ie the pressure differential is not constant.
madness, Aug 14 2006

       Remember the good ol' days when this was an idea for "supersimple" reverse osmosis?
NotTheSharpestSpoon, Aug 14 2006

       //Where did the 122.5 J come from? I suspect you would find, if you could measure it accurately enough, that the temperature of the water had dropped://

Not directly on point, but for NaCl, the enthalpy of solvation is around 4 kJ/mole for NaCl. As this is positive, the temp goes down slightly when it dissolves. For other solutes, it's the opposite. Eg, for KCl it's -17 kJ/mole.
ldischler, Aug 14 2006

       I can't believe I missed this first time around. It took me a while to get my head around the fact that the pressure differential comes from the fact that you take of the top off the fresh water but once I had, I liked it very much. Very cute idea.   

       For practical reasons, this probably wouldn't be feasible but, as a thought experiment you have my croissant.
st3f, Aug 14 2006

       Absolutely brilliant. As I recall desalinators using reverse osmosis on sailboats require about 850PSI.   

       I hope you all note this is a working perpetuum mobile at least as long as the membrane lasts. Go get your $1000000 from Randi.
Kirkmcloren, Aug 14 2006

       No perpetual motion involved. As the pure water crosses the membrane the salinity on the other side increases. In a small closed system, this would soon slow and eventually stop the flow across the membrane. Fortunately seas are not small closed systems.
st3f, Aug 14 2006

       /desalinators using reverse osmosis on sailboats require about 850PSI./   

       I think the reason that most desalinators must use higer pressures is that they are squeezing water from an increasingly salty brine. With the minimum necessary pressure, one would produce 1 ml of freshwater and then reverse osmosis would stop, as the brine would be too salty. The nice thing about doing desalination in open ocean (whether via slick Freefall method or by original openwater well as described) is that a concentrated and toxic brine is not produced.
bungston, Aug 14 2006

       //I think the reason that most desalinators must use higer pressures is that they are squeezing water from an increasingly salty brine.//

I'd be surprised if anyone didn't constantly cycle the salt water side back into the sea, to avoid that very problem. Anyway, 850 psi is 58 bar, which is right in there with the 50-60 bar that is most common.
ldischler, Aug 15 2006

       //squeezing water from an increasingly salty brine//, //I'd be surprised if anyone didn't constantly cycle the salt water side back into the sea, to avoid that very problem.//   

       The energy expended to pump, against the pressure gradient, the portion of the sea water that is expelled as concentrated brine is wasted unless recovered by coupling the seawater pump to the brine outflow (is this done?) In this way the present Idea could be more efficient or simpler or both than near-surface osmosis.
spidermother, Aug 15 2006

       //the portion of the sea water that is expelled as concentrated brine is wasted //
Good point. I didn't think of that. (Though there are systems to recover that energy. Here's an advertisement from a Norwegian company--"The Aqualyng system uses the brine energy in a special equipment, called the Recuperator, to pressurise seawater to the same pressure as the brine. The process is actually a pump, of the reciprocating principle, hydraulically driven by the brine.")
ldischler, Aug 15 2006

       As I understand osmosis and reverse osmosis, they work on completely different principles. Osmosis is a chemical phenomenon in which different liquids will equalize each other chemically through a membrane. Reverse osmosis is a brute force technique that physically pushes salt water through a membrane that allows only the water to pass through, leaving the salt behind. The proposed invention utilizes the ambient pressure of the deep ocean to create the force necessary to ram the sea water through the membrane. There is no chemistry involved. In order to achieve the necessary pressure, the membrane would have to be about 600 meters deep. The resultant fresh water would have to be pumped out. The fresh water will not rise within the pipe to sea level, since the pressure differential will diminish as the fresh water pipe fills up. Nevertheless, this seems like a brilliant idea that should save energy and protect the environment. (Aside from the brine discharge problem in traditional desalination plants, you also have the problem of sucking up millions of sea creatures in the intake pipe.)
Hurricane1008, Apr 17 2007

       //As I understand osmosis and reverse osmosis// which is not all that far. Reverse osmosis is essentially the reverse of osmosis.   

       There is nothing really magic about osmosis - it is simply diffusion of water. Water is diffusing from an area of high concentration (ie, pure water) into a region of low concentration (ie, a salt or sugar solution; the water in this is, in effect, "diluted" by the solute - there is a lower concentration of free water molecules). The semipermeable membrane is just there to stop the solute going anywhere.   

       Reverse osmosis is the exact opposite - you're pushing water up it's own concentration gradient.
MaxwellBuchanan, Apr 17 2007

       //The resultant fresh water would have to be pumped out. //

Read the debate and you will realize that, were the ocean deep enough, and were the salt concentration uniform enough, the idea embodied in the addendum paragraph would work without any pumping.
ldischler, Apr 17 2007

       //this would work without any pumping.// No, it won't. You need a pressure differential across the membrane, and that has to be paid for one way or another. Suppose you need a pressure head of X to drive the reverse osmosis. There are various ways to pay for it:   

       a) You can do it by pumping saltwater into the device at pressure X (conventional approach).   

       b) If you like, you can just stand the tube up in a big tower above ground, pump the saltwater up into it to a height of X, and then let the freshwater run out the bottom (you're paying to get the water up the tower to provide a pressure head of X)   

       c) You can drop the pipe in the ocean like [bungston] proposes. Fresh water will fill the pipe until it is X metres below the sea's surface - no less. You then have to pump the fresh water up X metres to get it out.   

       Come on, guys, think.
MaxwellBuchanan, Apr 17 2007

       Look at justaguy's calculations, 7th anno down.
ldischler, Apr 17 2007

       OK - I did, and they are interesting. But they are wrong. Here's why, in several steps:   

       (a) Suppose that 'staguys calculations are correct. If so, then you can drop this pipe down deep enough and, thanks to the pressure differential created by the difference in densities between fresh and salt water, you wind up with a pipe full of fresh water, with the top surface being level with the surrounding sea, yes?   

       (b) OK, now here's what you do next. Near the top of the pipe (which is full of fresh water, no?) cut a window. Seal a second semi-permeable membrane over this window. Then fit a cylinder and piston over the window (ie, sticking out from the main pipe at right angles, just below the surface of the sea). The cylinder starts out half-full of salt water.   

       (c) Viola! Regular osmosis will pull water from the top of the main pipe, through the second semi-permeable membrane, into the cylinder. Considerable pressure is generated, and the forceful displacement of the piston can be used to do work.   

       (d) When the side-cylinder is fully displaced, a side-valve is opened to let the brackish water out of the side- cylinder; the cylinder is flushed out with sea water, the piston is returned (with zero effort) to its starting position, and the side-valve is closed. Meanwhile, of course, the main pipe has refilled with fresh water.   

       (e) Etc. The various valve-opening and cylinder-restoring operations can have arbitrarily small energy costs. The piston displacement does work. Ergo, perpetuum mobile.   

       (f) Ergo, the original analysis was flawed.
MaxwellBuchanan, Apr 17 2007

       Yes, you could draw energy from this device. Your claim that it is a perpetual motion machine is flawed.   

       (Previously covered by ldischler's anno, May 08 2004)   

       The energy comes from the difference in gravitational potential energy between the column of fresh water in the pipe and the denser column of saline water outside the pipe.   

       If this machine were constructed in perfectly stagnant water, the saline concentration at the lower membrane would continually increase, until the osmotic pressure required to drive the flow was equal to the pressure difference across the membrane. BUT the ocean is not stagnant. There are deep-ocean currents driven by solar and wind energy. These currents cause the high-concentration saline to be continually replaced by lower-concentration saline and the process thus continues.
Freefall, Apr 17 2007

       [Max], this is one of those rare machines in which energy is drawn from otherwise apparently untappable sources. A similar example is the machine which runs on changes in barometric pressure.   

       As re the Lake Nyos mechanism for driving this, I am not sure dissolved gases would traverse the membrane. Not CO2, anyway, since I think it exists as an ion in solution.   

       That Lake Nyos effect in the ocean should be testable without any desalination. It should be true for any long deepwater pipe. In fact, if this happens, it probably happens at deep water oil rigs.   

       It may be that the best compromise for getting some freshwater is a hybrid of the original scheme and the [Freefall] superlong tube. As the tube gets longer, the freshwater rises closer to the surface. It would need to be impracticably long (absent Lake Nyos mechanism) to get the freshwater to fountain out of the top, but there may be places in the world where a pipe can be made deep enough to allow the freshwater to rise to within a few hundred feet of the surface.
bungston, Apr 18 2007

       [Freefall] and [Bungston] Hmmmm - it's possible that you're right, and that I'm talking bollocks. I am too hungover to be sure at the moment, but please accept a provisional retraction in the meantime.
MaxwellBuchanan, Apr 18 2007

       The best part of this invenion is that its super simple.
pashute, Mar 30 2008

       I've linked to a salinity versus depth profile. Salinity is highest near the surface, due to evaportion, and gradually increases again at greater depths, I assume due to the opposing forces of gravity and diffusion. According to another article in the series, there is very little vertical mixing in the deeper ocean water, which is bad news for our freshwater fountain, but there might be a small energy boost from the evaporative surface concentrating.   

       //There is nothing really magic about osmosis - it is simply diffusion of water. Water is diffusing from an area of high concentration (ie, pure water) into a region of low concentration// Not so - osmosis is different from diffusion. Osmosis is more closely related to flow of water due to pressure. Water can flow by osmosis from a starch solution to a sugar solution even if the starch solution has many times the mass of starch to that of sugar. In this case, the water is flowing 1) from a low water concentration to a high water concentration, 2) from a high to a low mass per volume of solute, and 3) from a low molar solute concentration to a high molar solute concentration. The last of these drives osmosis; the first two don't.   

       Incidentally, while looking for salinity gradient power generation I found an article which claims that since water can't be sucked up more than about 10m, it is osmosis which pushes water to the top of tall trees. This is complete nonsense. Osmosis accounts for less than a metre. The water is, in fact, sucked up, and is at negative pressure. No such thing, you say? In this case, there is. Hydrogen bonding among water molecules and between water molecules and the cellulose of the vessels allows the water to be under tension, and therefore at negative pressure.
spidermother, Dec 27 2008

       I don't see any perpetual motion here.   

       You put the filter on the tube, and lower the tube into the water. Pressure differentials force pure water through the filter, and the difference causes the pure water to go up the tube.   

       Pure water is negligibly lighter than salt water, and therefore, a negligible head of pure water in the tube should extend above sea level.   

       It looks to me more like a giant version of maxwells demon, and it looks to me like it is powered by diffusion. In a closed loop, the water from the top would diffuse back into the sea water, due to the fact that the sun is agitating the molecules by keeping it from being frozen.   

       So, solar power is applied to the ocean, which powers diffusion. The filter removes the salt, and creates a pressure differential between the fresh water in the tube, and the salt water in the ocean, and therefore the freshwater is forced up above sea level.   

       If you're drawing freshwater out of the system, the negligible difference of the head in the pipe and the height of sea level will likely be unworthy of calculating.   

       As a pump for power... the filter constricts the flow of the water through the pipe. You folks can calculate it as a power source, but several important questions must apply: how high over the ocean do freshwater rivers get? I doubt your pipe can get any higher. How much energy will it take to build the pipe? How long will the walls of the pipe last? How much energy will it take to replace the filter when it clogs without getting salt water in the pipe?   

       I predict that when you incorporate those costs into building this as an energy source, the net gains will be negative.   

       Still, it should be able to produce freshwater with some decency. Maybe we have found a new use for all those offshore oil rigs that cost so much to tear down.   

       I don't see why the pipe would have to be vertical though. Just stick it on the ground, make it longer, and have it go far out to sea. We've perfected that sort of technology reasonably well.
ye_river_xiv, Dec 27 2008

       There's another reason why this is a good idea:   

       With conventional R.O. desalination, the pump first pressurizes the salt water, the membrane seperates out (much lower pressure) fresh water, and as a form of waste, high pressure brine.   

       *Some* of the pressure of that brine can be recovered by means of a pressure exchanger, but certainly not all of it.   

       But with this idea, the weight of the ocean supplies all of the necessary pressure for the incoming saltwater, *and fully recovers* the pressure of the brine.   

       The pump only has to deal with the *fresh* water, so 100% of it's work is useful work. Furthermore, since the pump is submerged in freshwater, it doesn't need to be as corrosion resistant as a saltwater pump.   

       As a possible added bonus, if the pump is situated deep enough, then the water on the low-pressure side of the pump will be at a fairly high pressure... as a result, cavitation from high-speed pumping can be avoided.
goldbb, Jul 20 2009

       I know very little about the mechanics of osmotic filtration, so this is as much a question as a suggestion.   

       Q: Would a vacuum on the inside of the pipe create conditions where osmotic filtration would improve the efficiency of the differential for reverse osmosis to occur?   

       If so... Make the fresh-water pipe a sealed system (well the top, anyway) and create a vacuum in the pipe. If this works, it might shorten the length of the pipe necessary and make this idea more viable in shallow depths.
sprogga, Jul 21 2009

       It would require more energy to pump water vapor out of the pipe than to pump liquid water out.
goldbb, Jul 22 2009

       Compared with the weight of water at depth in the outside column, I am not sure how much difference the weight of the same column of air vs the same column of vacuum inside the pipe would make. The difference might be useful at very marginal depth. The vacuum addition would only be useful for the original iteration of this idea where fresh water is pumped out of the pipe, not Freefall's self pumping very deep pipe.
bungston, Jul 23 2009

       How long would it take for the pipe to generate one gallon of fresh water?
Gamma48, Jul 25 2009

       WcW, the idea is fleshed out in the annos, starting with justaguy's calculations on May 07 2004. This does work (or would if the ocean were deep enough), and the energy comes from the non-equilibrium state of the ocean.
ldischler, Jul 25 2009

       his math however is incorrect. It is very simple to calculate that below a certain pressure differential the behavior of an osmotic membrane is to flow fluid from the low pressure side out to the high pressure side and that the "head" pressure produced under such a system will never be higher than the pressure differential produced by the difference in density in said pipe. the math is wrong.   

       Assuming that we necessitate 20 bar pressure to overcome the osmotic gradient we drop our pipe to the depth required to produce that gradient, voila the pipe begins to fill with H2O, and the fluid fills the pipe. Now you have 1 gallon of water at -2000ft. That gallon stays there. Any additional water that happens to wander through the membrane displaces other water that (with the slight rise in pressure) can now wander out.   

       If someone can develop an osmotic membrane that ACTS AS AN ACTIVE GRADIENT WITHOUT USING ENERGY then would have found a font of infinite energy. This sea pipe would be a Maxwell's Demon and sadly COMPLETELY IMPOSSIBLE.
WcW, Jul 25 2009

       In addition, the energy required to pump the water through the membrane in a conventional system would be nearly identical to the energy required to pump the water 2000 feet up from the bottom of the ocean where you can get it to pass unassisted through an osmotic membrane. Add vacuum and the math stays the same. Always the same. No free lunch. ever.
WcW, Jul 25 2009

       //the math is wrong//   

       Where is it wrong?
ldischler, Jul 25 2009

       Lets do a thought experiment. Imagine taking a huge hose of osmotic membrane to the ocean. You walk into the edge of the ocean and carry the hose into the water. Nothing happens because at the surface of the ocean the pressure isn't enough to force a gradient: Any D/S water in the hose at this level is sucked out of the hose, literally. So you dive down with the hose, really deep, to the level where D/S water in the hose is not sucked out, very deep: Here the end of the hose begins to fill with water, D/S water. Yay!   

       Now as you try to bring that water back up it leaks back out of the hose and when you get to the surface the hose is DRY (really just like it was when you first entered the ocean). You could pump that water to the surface, but that would take energy, as much energy as it takes to run an efficient R/O plant. Does this help?
WcW, Jul 25 2009

       Lets do a thought experiment. Imagine taking a huge hose of osmotic membrane to the ocean. You walk into the edge of the ocean and carry the hose into the water. Nothing happens because at the surface of the ocean the pressure isn't enough to force a gradient: Any D/S water in the hose at this level is sucked out of the hose, literally. So you dive down with the hose, really deep, to the level where D/S water in the hose is not sucked out, very deep: Here the end of the hose begins to fill with water, D/S water. Yay!   

       Now as you try to bring that water back up it leaks back out of the hose and when you get to the surface the hose is DRY (really just like it was when you first entered the ocean). You could pump that water to the surface, but that would take energy, as much energy as it takes to run an efficient R/O plant. Does this help?   

WcW, Jul 25 2009

       [marked-for-deletion] bad science, bad math, perpetual motion.
WcW, Jul 25 2009

       //Does this help?//   

       No, not really. You said the math was wrong. I looked at this when justaguy originally posted it. I expected to find a math error, but I didn't. As counterintuitive as it is, this would work, given a deep enough ocean. I did worry that there was no energy source, and it wasn't until the next day that I realized it was the non-equilibrium state of the ocean. That is a huge amount of energy. I calculated the gravitational potential energy of salt in the ocean in an anno on August 14, 2006--it's the equivalent of roughly one hundred million million barrels of oil.   

       So, if you're going to say it's bad math, point out where it's wrong.
ldischler, Jul 25 2009

       the idea still does not work. If the ocean was pure fresh water at -24000 feet, you would still need to pump it out and that would take more energy than is used in a simply R/O setup. As to the math I note that while he uses a fixed value for osmotic pressure he does not add the head pressure of the water filling the pipe which would be in balance with the osmotic differential. In every case the osmotic pressure forces fluid from the side of low salinity to the side of high salinity unless acted against with the addition of energy. Every time; No matter what the concentrations or the the differences in pressure to reverse this process energy must be added. Even for that first gallon of water flowing into the deep pipe you must pay for it by forcing an empty pipe down to that depth.   

       Even when you simply lower a RO bladder down to the depth required, seal it, and pull it back up you must pay the full energy penalty.
WcW, Jul 25 2009

       Look at the math and point out where it is wrong. The water column in the pipe doesn't weigh as much as the water column in the ocean because salt water is heavier than fresh water. Given a long enough pipe, the pressure deferential across the membrane becomes sufficient to drive reverse osmosis. It's simple really, and the math is simple. Point out where it's wrong. Why are you avoiding that?
ldischler, Jul 25 2009

       The difference in salinity is far to small to compensate for the difference in depth and the energy required to raise that lower salt water from the deep and then R/O it.
WcW, Jul 25 2009

       You're still making word arguments, and the arguments are wrong. The math is correct. If you don't understand the math, that's okay, but don't just say it's wrong.
ldischler, Jul 25 2009

       //If the ocean was pure fresh water at -24000 feet, you would still need to pump it out and that would take more energy than is used in a simply R/O setup.//   

       If the ocean had that kind of inversion--if it were fresh at 24,000 feet and salty further up, no membrane would be needed and fresh water would be driven up the pipe to make a spectacular geyser at the surface. Similar to the one on Lake Nyos.
ldischler, Jul 25 2009

       WCW, I love learning and I especially love when someone makes wth the numbers to prove their point. Best of all is when two numbers people duke it out, though that sometimes gets hard for me to follow. Go look at Freefall's math from 5/7/2004. Copy it, paste it, and write under the line that is wrong why it is wrong. Or do your own in this style to facilitate comparison. Intuition is a great place to start, but intuition also says the heavier ball falls faster. Math has trump when it comes to physical phenomena.   

       And don't you dare delete your account if you get grumpy. Go have a snack and come back. Ldischer grumpies with the best if them, and if he can come around you can too.
bungston, Jul 25 2009

       Whoa there chillins the difference we need is 27 bar (given 1.1 mole per l salinity, deep seawater) and then we would need about 33000 feet before the math even begins to work but WAIT, SALINITY INCREASES WITH DEPTH AFTER THE INVERSION ZONE. The two lines will never meet. At 33000 feet the salinity is greater than at the surface. The lines never meet. I'm sorry that I lack equations that actually represent the change in density with depth due to their complexity (temp, saline, pressure) but the salinity is clearly rising and even a small increase in salinity over the surface salinity level leaves us hopelessly under water mathematically speaking. There's Lies, Damn Lies, and then there's Math.
WcW, Jul 27 2009

       This is fantastic, my brain hurts from all the arguments and counter-arguments. Sorry if I duplicate other points, I've read every single anno but can't be bothered to follow all the links.   

       WcW, you start off arguing against the original posting, pointing out that pumping out the fresh water would cost significant energy, as bungston told us back in 2004. I like goldbb's argument that there is near-perfect pressure recovery from the waste brine thus the efficiency would be marginally better than that of land-based RO plants - however this might be negated by frictional losses caused by the flow through the length of pipe over which the fresh water has to flow to reach the surface.   

       FreeFall's development, however, has been the focus of the maths, and depends on the difference in density between saline and pure water, hence the need for an improbably deep pipe to generate the pressure difference between the two water columns, with an ocean to match. In the case of the latter concept, however, you are not at the end arguing bad maths so much as bad assumptions - salinity with depth, etc etc. The principle, given a (possibly) mythical situation where the assumptions hold, would work, and all the head scratching as to why god didn't frown would not be in vain after all. Don't go declaring perpetual motion, try to find out where your energy is coming from. For example, a lamp powered by a solar panel powered by the lamp will work perpetually if you use it in space and happen to point it at the sun.   

       I've found this link [added above] (check out Figure 5.2) that suggests that overall the salinity of the ocean decreases with depth, and argues that this is in balance despite Idischler's suggestion of a fountain effect because the difference in density caused by temperature decrease more than counteracts the difference in density caused by salinity. This effect might suggest that the pipe need not be as long as the maths declares, except the variation in salinity is tiny - 0.3% over 6000 meters (unless the salinity minimum point is beyond 6000 meters and the increase in gradient after this point is astronomical). [Oh, the other possibility is that the pressure differential required for reverse osmosis is highly sensitive to salinity, I haven't checked what it is but I doubt it looking at how land based RO plants operate]. As a sense of scale, the salinity of the dead sea, so wikipedia tells me, is 33.7% compared to 3.5% in the seas (ish).   

       The talk in 2006 of osmosis being an "energy free" process offends me (but not as much as a pint over the head). The pressure required to achieve reverse osmosis (when losses are disregarded) points to a very favourable energy gradient in osmosis. That is to say, a separate water / salt scenario represents a higher potential energy than the equivalent brine solution. I believe that this must be counteracted by gravity which represents a different potential energy, and the importance is understanding the relative scale of the two effects. Let's take the case of the two cubic meters of water at different salinity. After the barrier is removed, the change in salinity will occur in two ways: 1. The combined water/salt ion "molecules" in the lower cubic meter will by diffusion exchange with pure water molecules. 2. The relatively high free H(slightly+) and O(slightly-) ends of the pure water molecules will attract the salt ions from the relatively more saturated H(slightly+) and O(slightly-) ends of the saline water molecules by osmosis. In either case, the mass of the salt ions is raised by a height representing an increase in potential energy. The mixing effects WILL therefore be counteracted by the energy required to lift the salt ions from the lower to the upper domain, and the resulting salinity gradient, however slight, will depend on the relative energies of the processes. I suspect that the energies involved in osmosis of salt in water are an order of magnitude more than those caused by gravity on the salt ions. I believe that essentially this is not any different to the case of a suspension, except that the relative gradients in a suspension are balanced in the other direction, probably due to mass vs surface area and charge (or other attractive) interactions of these much more massive particles - this is nano vs micro. However, this all assumes that the water is an even temperature and therefore density is only affected by salinity. This is clearly not the case in the ocean.   

       This leads me to suggest a highly thermally conductive, externally lagged pipe design that internally conducts surface temperature to the bottom, warming it from 2degrees (I'm sure I read somewhere once that the density of water reaches a minimum at 4degrees, though) to, say, 15degrees (Mathematical challenge: At 15 degrees the remaining temperature difference is 5-10 degrees in good weather, how thick does a really good thermal conductor have to be to achieve this, given a fixed water flow: Which begs Gamma48's question - how much volume of fresh water would this enormous edifice produce?). The change in density could, I suspect, be much more effective than that caused by change in salinity and in any case will work in tandem with it, decreasing the required length (and depth) of the pipe.   

       This invention is very simple in the sense that "Rocket Science" is extremely simple - a rocket only really requires a fuel, an oxidant, and a structure to direct which way it all goes bang, and I've seen it achieved with cardboard, gunpowder, and air. In the same vein, however, the complexity comes from conducting it on a scale that it becomes useful (achieving orbit / getting fresh water to the surface).
TheLightsAreOnBut, Jul 27 2009

       TheLightsAreOnBut: If you used the energy from a thermal vent to heat the pipe water, you could greatly decrease the water depth needed. In that case, you'd want to operate adiabatically--eg, by insulating the pipe.   

       I've added a link to a water density calculator. Just type in the salinity and temperature.
ldischler, Jul 27 2009

       // And fourth, the economics are insane, sure, but so what? Eventually, the cost of oil will reach $100/barrel, and then all sorts of formerly insane technology will get funded. // posted in 2004.
marklar, Jul 27 2009

       Thanks for that, Idischler. On use of the calculator you linked to, I surmise that salinity has a much stronger effect on density than temperature...   

       I see - the change in salinity in the ocean, which is relatively tiny, makes a difference of 2.5kg/m3 from highest to lowest, whereas the difference in temperature, 2deg to say 25deg, makes a 4.5kg/m3 difference.   

       This is very tiny compared to the difference that changing salinity from 3.5%ish to nothing makes on density. bother.
TheLightsAreOnBut, Jul 27 2009

       yes, marklar, I thought that oddly prophetic. Probably still doesn't make it affordable to dig a 10km hole in the mariana trench, though.
TheLightsAreOnBut, Jul 27 2009

       ...but salinity does increase with depth due to the natural tendency for denser fluids to sink below less dense fluids. When you add in the thermodynamic element you come to realize that there is a near uniform osmotic potential from the bottom of the halocine layer to the deepest location that I could find data for.
WcW, Jul 27 2009

       Right, Idischler, I've just remembered why I avoid this site.   

       Using the calculator, and presuming it's correct at extreme temperatures (it doesn't specify pressures required to keep water liquid above 100degC but it does still give a density), at 95degC and requiring 60bar pressure differential, "only" 9400m are required for this to work - this presumes, however, that the surrounding sea water is still at 2degC.   

       Going to the extreme with the underwater geysers, that reach temperatures of up to 400degC, at that temperature, and still presuming sea water at 2degC, only 1300m depth is required - these vents are found at 3000m depth. However, some serious engineering is also required to withstand that temperature, and in any case land based evaporative desalination is probably easier at a land based volcanic location such as Iceland. Not that they need it, really.   

       Fascinating article on the deepest hole ever bored in the earth, though, at the link given.
TheLightsAreOnBut, Jul 27 2009

       //However, some serious engineering is also required to withstand that temperature,// Not so, at that depth...   

       The biggest problem, so far not discussed, is at those depths the inside of your long tube becomes a liability, unless you can add the same amount of pressure as 3000m of seawater. If you add the same amount of pressure, or something remotely similar in order to prevent collapse into an envelope, you won't get your font of elixir vitae. So we are looking for a tube of extrordinary compression strength. That can house a semi-permeable membrane of similar strength. Although as a thought experiment it will work ([WcW], technically the maths does work), as do space tethers, I fear the devil will be in the details.
4whom, Jul 27 2009

       //unless you can add the same amount of pressure as 3000m of seawater...So we are looking for a tube of extrordinary compression strength //   

       Actually no. The pipe would be lowered already filled with fresh water, so the max deferential pressure it would see would only be a few hundred psi at the bottom.
ldischler, Jul 27 2009

       //so the max deferential pressure it would see would only be a few hundred psi at the bottom// And the osmotic pressure would, therefore, be zero?
4whom, Jul 27 2009

       There has to be some pressure differential, not maintained by the inner medium, to allow for osmotic flow. Most notably this pressure would be taken up by some structure, this structure would support a differential membrane operating under similar forces that that structure would have to support. The difference between what the structure supports and what the membrane (carved into however many small pieces) supports is what you can get out the system.   

       Mathematically this allows for a fountain at the top end. However, we have to presume an open structural integrity on the pipe/s, and an osmotic membrane with high structural integrity, at 3000m, with flow. All of these beasts are unicorns of the same feather. The maths is right, we are just not up to it at the moment.
4whom, Jul 27 2009

       The initial assumption was that osmosis would begin at 20 bar--about 300 psi. (Actually, this is probably closer to 30 bar minimum, or about 450 psi.) At that point the flow would be a trickle, so you need more than that--the more the better. How much you can get is dependent on all these factors being discussed--water salinity, salinity profile, temperature and temperature profile, and operating depth. Anything that lightens the water column in the pipe is good, as it reduces the required operating depth.
ldischler, Jul 27 2009

       4whom, the Freefall self-pumping freshwater filled deep pipe occupies the interior of the pipe with incompressible water. The structural requirements for the deep pipe are the same as for my original shallow air filled pipe where you must pump the water out because the pressure involved to make it happen are the same. The membrane must withstand pressure adequate to do reverse osmosis. The pipe must be at least as strong so it does not cave in.   

       My original version is supersimple, I still think. The free energy Freefall version is a unicorn only because of the phenomenal depth needed to make it happen but the tantalizing this is that this great depth is not an order of magnitude away from earth conditions but close enough that various temperature / dissolved gas tweaks might make it possible.
bungston, Jul 27 2009

       [bung...] I bunned this idea long ago, and as it stands it is a good idea (in fact I would prefer to pump out my fresh water from 10m (all things considered) than have it magically fountain out the top).   

       I even agree with [Freefall]'s addendum. It is not so impossible. The oceans, even at those depths contain energy we can only hope for. In theory. I even agree with [ld...]'s points. In theory.   

       But these theories assume pipes that can contain enormous pressures, I don't care if you send farkin mercury down your tube initially. You will have to deal with the pressures at some point, for the desired result. These pressures have to come from some manufactered state. No such thing as a free lunch...   

       Containment will always be an issue, if we want to derive energy that we can use. From fission to fusion to osmotic to ionic. Containment issues are always the consideration, else the energy "evaporates" into the system.   

       The deeper this goes the more energy evaporates into structural integrity. When we solve that problem, we solve a lot more than fresh water from deep sea water...
4whom, Jul 27 2009

       This is simply a perpetual energy idea that has repeatedly been hashed and re-hashed. If it would work for a 95% R/O then it should work gangbusters with an 85% R/O and even better if the membrane ruptures. If you believe in deep pipe fountains then you believe in this idea otherwise you don't and nobody has been able to build one yet.
WcW, Jul 27 2009

       4whom: The pressures needed to be withstood are the pressures required for RO. In my calculations, I have used 60bar because that's what I've read is required for a decent RO. If I only used 30 bar, all the depth calcs would be halved. As we have working RO plants at the moment, I think we can be fairly confident that the engineering required to make a vessel and RO membrane that can withstand at least 60bar already exists. This is not on a par with the engineering constraints required by fusion.   

       For example, a mild steel pipe 0.5m diameter with only 20mm wall thickness could withstand 60 bar internal pressure easily enough (I don't know the calcs for external pressure - I believe it will need to be thicker as it will be a less stable situation)
TheLightsAreOnBut, Jul 28 2009

       //95% R/O then it should work gangbusters with an 85% R/O and even better if the membrane ruptures//   

       This shows where you fail to understand the process. If the membrane ruptures during operation, the flow rate is initially higher, that's true, but then it slows and eventually stops as saltwater rises in the pipe. The driving force for RO is the density difference of the salty ocean and the fresh water in the pipe. This difference is small, but is amplified by the length of the pipe.
ldischler, Jul 28 2009

       This idea is fine so long as it is simply just sinking a fresh water well into the ocean...   

       There is no way the salt/fresh water density differential is going to continuously pump fresh water out the top of the pipe.   

       I think it is probably best to dig a hole (in the sea bed) near the shore about 700 metres deep into which the pipe is suspended/forced. All that is then required is a bucket and about 700 metres of rope...
madness, Jul 28 2009

       Just show me one NODC data point where the salinity temperature and pressure are great enough that, when you actually apply modern version of the Van't Hoff you get a positive value for p2-p1-rOp   

       do this and I will concede the whole thing. I have been mining data sets for the last two days and none of the really deep sample strings has come up with a positive value. It tends to be really cold down there and saltier. I have access to the NODC data so all I need is the geo-position and depth or the ship and date. Even just the vague location would do. Although don't say "Marianas Trench" because I'm not finding any deep samples that aren't to cold or too salty for the depth of the sample.
WcW, Jul 28 2009

       //NODC data point where the salinity temperature and pressure are great enough that, when you actually apply modern version of the Van't Hoff you get a positive value for p2-p1-rOp//   

       Wow WcW, you lost me there. OK, found out what the NODC is, also who Van 't Hoff was, thank you for enlarging my knowledge a little! The whole point is that there is no-where deep enough, allowing for a temperature of 2degC (I believe it doesn't get any colder than that), for getting a positive p2-p1-rOp, my calculations suggested that you would need to go 10km deeper than the Mariana trench, this was also made clear in 2004 in the annotations. However, just because there is no where on earth deep enough to make it work, doesn't mean that, if there had been, it wouldn't have worked. I find it interesting that you say it is saltier down in the depths, my (very minimal) google search suggested otherwise. Could you give us a figure please?   

       I'm still having trouble around the perpetual motion thing myself, however. In the ocean, I can see that in one way or another, you'd be tapping the sun's energy. However, if you did a "20km tube" experiment... I see it like this:
a 20km tube with an RO membrane at the bottom is placed inside a bigger 20.01km long tube with closed ends, and placed vertically, supported by the space lift when they get around to building it. The outer tube is perfectly lagged, and the temperature is a constant 20degC inside it. In the inner tube is placed pure water, in the outer one CaCl2, with the levels coming close to the top of the inner tube. The water fountain effect means that the pure water will pour out of the top of the inner tube into the saline solution. I'm not generating energy with this system, and with perfect lagging, there is no where for heat losses to go, so what now? Will the process of diffusion of the more concentrated saline solution below to the less so above create a temperature differential that will eventually stop the fountain working? How? Why will the edifice not perpetually cycle the flow?
TheLightsAreOnBut, Jul 28 2009

       What will happen in your scenario, TheLightsAreOnBut, is the upper portion of the larger tube will become fresh, while the lower portion will become so salty that reverse osmosis will stop. Not until you input energy by stirring up the outer column, can you get it to run again.   

       (And this suggests that the process ought to happen on its own--if you allow a very long tube of seawater to sit without stirring, some of the salt should precipitate out. So you don't need to have sea basins repeatedly drying out to make salt deposits. All of that could have occurred during the Snowball Earth episodes a billion years ago when the oceans froze over. With almost no mixing, the oceans would have stratified, and vasts amounts of salt would have crystallized at the bottom.)
ldischler, Jul 28 2009

       [WcW] To soothe any exasperation you may be feeling, I think you are right - the pumpless implementation won't work in general becuse the oceans are largely close to equilibrium w.r.t. salinity (ignoring the top few metres) so there's no energy gradient to exploit.   

       Exceptions may include deep ocean thermal vents (as mentioned earlier) and regions where cold, relatively fresh water from melted ice lies beneath more saline water.
spidermother, Jul 28 2009

       Lights, your experiment is similar to one I proposed but using a solute heavier than NaCl. The heavier the solute the shorter the tube needs to be. Very big and heavy soluble molecules would allow this to be tested with skyscraper magnitude tubes.   

       I suspect that the solute side would stratify to the point where it became too salty (or solutey) for reverse osmosis at the fixed pressure. But stratification would not be complete because there would still be osmotic pull of water into regions of higher solute. So one might get an occasional trickle of fresh water over the top from time to time.   

       Why do water molecules move anywhere? They are hot and move about randomly and tend to move in energentically favorable directions. I think that absent external heat inputs this tube model will cool down until it freezes. /It freezes!/
bungston, Jul 29 2009

       [humanzee] You are right, two separate bores will maintain a 700 metre head of pressure at the "junction" membrane (ignoring atmospheric pressure).   

       Going one step further --- drilling the dry side bore wider and deeper than the wet side will give the fresh water a place to build up a head of pressure. And also permanent access to the junction membrane for maintenance.   

       The dry side bore pressure head will then be enough to bring the fresh water to the surface. Just a bit neater and is there any need to work out exact numbers....
madness, Jul 29 2009

//it freezes!// Why? I lagged my pipe, where did the heat go?
//stratified// I think that the osmotic pressure will be greater than gravity, as I've said before. The only way I think this might be counter-acted would be if the processes of solution (at the top) and dissolution (at the bottom) effectively created a temperature gradient as well as a salinity one.

       So, you're saying as long as we don't extract energy it would carry on, even if it was a tiny trickle? That this would effectively turn the nano motion of molecules (heat) into a macro motion? Presuming that the energy can be extracted from the top after all, then, this proposes a fridge that generates electricity rather than consumes it...
TheLightsAreOnBut, Jul 29 2009

       spidermother, I thought the uniformity of salinity works in favour of the pump-free version, not against it. The reason the idea is half baked would be how ridiculously shallow our oceans are?
TheLightsAreOnBut, Jul 29 2009

       You know, looking at the PT graphs, this does in fact become close to impossible, without sea vents that is. I think that is what [WcW] is saying.
4whom, Jul 29 2009

       Even at the thermal vents the water is very hot, but also super salty and the osmotic potential is more in balance than it would be if the water was simply hot. An osmotic pipe here would need to heat seawater from a lower salt source using the heat of the vent but not its output, and then I wonder if the final output would be better than simply using a heat pump on that same differential.
WcW, Jul 29 2009

       Right. What you do is use the heat of the thermal vent, but you don't take hot water from the vent, which is full of nasty stuff. Nor do you run the RO at high temperature; you do that at the 2C temp of the ocean bottom and then heat the fresh water stream by running the pipe through the vent, thus getting a density reduction several times that of fresh water alone. At 200C you'd need to be about a mile down.
ldischler, Jul 30 2009

       If it was done in the Netherlands then you could siphon the freshwater from the top of the pipe. ;- )   

       An empty can, with holes, in the sea is going to fill to full. Just because the holes only allow water shouldn't matter, should it?   

       It does matter. Water has an affinity for the salt, an energy barrier to overcome.   

       Probably in the future we will be able to design the hole structure to mimic salt in solution and therefore allow the water to flow freer. But not currently, as far as I know.
wjt, Jul 30 2009

       Instead of a pipe lowered to nearly 30,000 feet depth, lower a concrete box, a vault the size of a house, say. The RO membranes affixed to holes in the vault. Connect the concrete vault to the surface of the ocean with a flexible tube. Install the whole apparatus pre-filled with fresh water so that there is ever only positive pressure in the vault/tube. If the math is right the heavier salt water will force RO to occur at the membranes and force the lighter fresh water out the tube under pressure (which will decrease proportional to depth). This apparatus is far more practical than a "big pipe with holes in it" and could be done with existing cable-laying ship technology.
JackyD, Aug 14 2009

       The vault is a big pipe bottom. I do not see how the vault is an improvement over putting the membrane at the bottom of the pipe.
bungston, Aug 14 2009

       [Nobody'sHome] Uniform salinity would work in its favour, but salinity equilibrium (which inplies a gradually increasing salinity with depth such that gravity and diffusion are balanced) would not.
spidermother, Aug 18 2009

       John F. Kennedy, 1962: "If we could ever competitively, at a cheap rate, get fresh water from salt water, that it would be in the long-range interests of humanity which would really dwarf any other scientific accomplishments."
ldischler, Aug 18 2009

       "but first let's go to the moon".   

       Okay so you could actually get a head a metre high if you sunk a 40 metre shaft... that is seriously neat. How long do the membranes last ?
FlyingToaster, Aug 18 2009

       oh please, 40m? Nobody has yet come up with a single location on the ocean where the conditions would work. Not one location.
WcW, Aug 19 2009

       huh ? salt water weighs 1.025 x's as much as water water... so a 40m depth will produce a 1m head just to even out the weight/pressure, ie: if you have a hose straight down then filling it full of freshwater to a level 1m above sea level will produce a zero pressure differential at the bottom.   

       Yeah okay so you need some sort of pressure to osmose... how much ?   

       [edit] "how much" is 350psi or about 24 atmospheres.   

       1 atm is about 1kg/cm2 so 24kg/cm2 difference is required so... well that's only 10km down or so.   

       Marianas Trench is 11km deep.   

       so there.   

       and you still can have a 25m head or so.   

       I don't wanna be the guy who has to change the membrane though.
FlyingToaster, Aug 19 2009

       //well that's only 10km down or so.// It still won't work as the salinity is higher at depth than near the surface. If you pumped surface water down to the bottom of the well (outside the membrane, of course) you could get flow, but with no saving of energy over pumping fresh water out of the well (from a depth of about 300 metres) as per the original idea.
spidermother, Aug 19 2009

       //salinity is higher at that depth// as is the weight of salt water. Wikipedia states that seawater can go up to 1.05 g/cm3 (as opposed to 1.025 at the surface) which is twice the difference of 1.0g/cm3   

       we're rehashing the last 297 annos aren't we.
FlyingToaster, Aug 19 2009

       I agree, we're rehashing. I think the numbers work. The issue now is physical practicality. Bungston, the vault IS like the end of the pipe. It's just way more practical to make the assembly using a vault and hose. A 20,000' long pipe is not a practical pysical solution. Lower a big concrete box from a ship using a cable and connect the box to the surface with a hose. The vault(s) could be lowered into deep water off continental shelves and the flexible hoses could be run back to shore. Vaults will have more area than pipes available for membranes, and the hoses will be able to withstand ocean currents and movement conditions.
JackyD, Oct 21 2009

       There are advantages to the membrane not quite being on the ocean floor. A much earlier post mentioned the difficulties of sea life and decay. Fortunately if you go deep enough, these largely dissapear, but ocean bottom grit is a potential issue, especially if you're placing this unit in a current to keep a constant saline supply or near a volcanic vent for heating. With either the pipe or the vault you probably want your working point some distance off the surface.
MechE, Oct 21 2009

       "World's Biggest Drinking Fountain" [+]
FlyingToaster, Oct 21 2009

       I don't think a hose is a good idea. There is 24bar pressure difference at the bottom, right?
Ling, Oct 22 2009

       //24 bar difference// at 10km down.
//hose// I think we're using "hose" in the sense of "very thick steel/concrete pipe"
FlyingToaster, Oct 22 2009

       The hose is full of incompressible water. It is actually made of lamb intestine, for better sensations.
bungston, Oct 22 2009

       Incompressible water that flows away, and lets the hose collapse.
Ling, Oct 23 2009

       It flows away where? Not out the bottom because there is a membrane there. Out the top as a glorious Bellagio-style fountain!
bungston, Oct 23 2009

       I had an idea and while I was writing it, I realised it wouldn't work. But here it is anyway. Can anyone think how to solve the fatal flaw?   

       "To solve the problem of salinity at the bottom, the pipe could be U shaped. Salt water would flow in one end and fresh water would flow out the other."   

       But where does the super salty water go?   

       Anyway, assuming nowhere is deep enough for this to be energy-free, if it's half deep enough, the fresh water will rise half way up the pipe and require half the energy to pump out.
marklar, Oct 04 2010

       if you put surface salt water in the pipe then the entire pipe then what will come out the other end would be surface salt water. the ocean is generally near a thermodynamic equilibrium.
WcW, Oct 04 2010

       I meant with a membrane at the bottom, the thing I forgot is that the salt needs to get out somewhere.
marklar, Oct 04 2010

       // The salt needs to get out somewhere. //   

       If only it was surrounded by a massive body of water with currents.
Alx_xlA, Oct 04 2010

       I tried to read the whole thing.   

       [bungs] Why did you give up on the original idea. It IS supersimple. Its simply a pipe with air in it, submerged to a depth. And not that deep:   

       The depth you calculated: 910 feet, are 277 meters. As far as I know, pumping up from 277 meters takes less energy than creating 28.4 atm (the pressure at that depth).   

       The pipe is closed with membrane that works with with reverse pressures of 55 atm (accroding to Wikipedia: Reverse osmosis # High pressure pump / desal of seawater. )   

       So you'll need to go 500 meters deep. Still, probably worth it.   

       Here's another way to do it: rather than pumping the freshwater all the way up, pump it to a height where pressure is lower, and there into a submerged plastic bag floated through a pressure chamber out of the pipe. The bag is then floated to the surface.   

       A third way would be to have a gradient, where you keep on desalinating in stages. At 500 meters you use a membrane that lets much of the salt through, as you go up the water goes through thinner membranes, but need less pressure, because the diff is smaller.   

       Then you are pumping only a bit, and need less pressure because the water is losing its salt content.   

       Every once in a while you would need to flush the system though.   

       Pipes for these kinds of depths have been made. See Bosphorus Crossing Project for example.   

       So there you are. No perpetual, just pumping. But still, probably worth it.
pashute, Jun 24 2013

       //As far as I know, pumping up from 277 meters takes less energy than creating 28.4 atm. //   

       The first five words are the key here, I think. To pump 1 ton of water up 277 metres will take about 2,777,000 Joules of energy.   

       To create a pressure of 28.4 atmospheres in 1 ton of water takes about 4,500 Joules.   

       Likewise, to rotate 1 ton of water through 180 degrees on a frictionless bearing given sufficient time will take arbitrarily little energy.   

       My point being that your statement is meaningless, because you are comparing between dissimilar quantities.   

       So, "as far as I know", in this context, means "hey, I have no idea and I'm not prepared to expend the 20- 30 seconds necessary to think about it, but off the top of my head I'd say"
MaxwellBuchanan, Jun 24 2013

       I think you'd have problems, in that it would be difficult to find something that was permeable to hydrogen but not helium.
MaxwellBuchanan, Jun 24 2013

       //No perpetual// We (okay, I) did this back when and based on salt water weighing 1.025kg/l and osmosis at 24atm you could get a 25m head using the Marianas Trench (11km deep).
FlyingToaster, Jun 24 2013

       /Engineers at the Rocky Mountain Arsenal, a chemical-weapons manufacturing center near Commerce City, Colorado, disposed of waste fluids by injecting them down a twelve-thousand-foot well/   

       This is from the linked New Yorker article about induced earthquakes. Fracking companies are disposing of collosal quantities (lakes and lakes) of brine pulled up from the deep by shooting it back down there. 12000 feet is deeper than the Marianas trench. I did not know that people routinely moved such huge quantities of water over vertical distances like that.   

       What if the membrane were at the bottom and capitalized on the pressure of the water column? That is a lot of pressure. Plus I think they are pumping it down. Could one put salt water down and get fresh water back up an adjoining pipe (with a membrane at the bottom)?   

       Really this hinges on the membrane. I bet the reverse desalination membranes are not very durable. Probably they get clogged easy.
bungston, Apr 21 2015

       //not very durable//
[bungston], they are designed to handle high pressure (that's how they work, of course) but in your scenario the pressure would be REALLY high, and would also need a "clean in place" set-up, because you won't want to be hauling the membranes back up 12,000 feet to wash them.
neutrinos_shadow, Apr 22 2015

       I wonder how the membranes are washed in topside reverse osmosis operations. Backflush with clean water? Or do you need to get in there with a bristle brush?   

       /pressure would be REALLY high/ Hopefully not the full pressure of a 15 km column against the membrane, but a height of that magnitude is what was required for the mass difference between columns of brine and fresh water to drive reverse osmosis. I am not sure doing this at the bottom of a basement brine disposal well is better than doing it in the Marianas trench. Although there might be more water customers topside, and less giant squid at the bottom.
bungston, Apr 22 2015

       Mariana trench=36,000' deep.
cudgel, Apr 22 2015

       I think the thing to do here would be to pull a partial vacuum in the pipe. This will help the pressure differential a little, in the order of one of those twenty bar needed. Secondly, even at the relatively frigid temperatres of the deep sea, it's above 0. That means there will be water vapour. Now, what you need to do is make it very dry and very cold at the top and condense out the water vapour. Now you will have a temperature and humidity gradient from the bottom to the top, with a bit of tube engineering, you should be able to get a circulation going and use the thermal energy of the sea as your pumping energy source. Lots of insulation needed, mind.
bs0u0155, Apr 23 2015

       Ok that's a lot to read all at once. I'll be back.   

       There is a lot of real and good info in there but also a lot of bunk.   

       The big one(that I can comment on) at the end is that the pipe would need to handle more then the 60Bar of the RO and that's not true. If this thing works 60Bar+removed head = pipe pressure at bottom. so say 66Bar to be safe.
dev45, Apr 24 2015

       I once read that in the ocean, the water is continually separating out - on a very small scale - into its salty and fresh bits, just due to density. The little bits of fresh rise up and get clobbered by the briney bits coming down from above, mixing back to standard saltiness. Which separates again. (There was a picture, but I really cannot recall the scale of it.)   

       (Or maybe I am confusedly recalling that salt-fingering link all wrong.)
baconbrain, Apr 26 2015

       Like mass spec, gas chromatography, in a laminar flow different molecules bounce differently under different dimensions. A differential path might be designed before the holes to gain another gradient.
wjt, May 03 2015

       Oh well, (no pun intended) there goes my catheter/RO filter/stilts idea for recycling water while being in the desert.   

       I should have known, as there's no mention of the very long stilts that come with the Fremen Stillsuit.
not_morrison_rm, May 03 2015

       This idea has floated to the surface again because a startup called OceanWell is planning to do it for real.   

       I haven't read all the annos so I might be repeating what someone has already said.   

       I'm wondering why you need to pump the fresh water to the surface? Would it be possible to fill a capsule/balloon/bag with the filtered water, then somehow release it into the ocean at depth (600 metres). Because fresh water is less dense than sea water the capsule/balloon/bag would float to the surface.
xaviergisz, Sep 27 2023

       //somehow// is a very hard working word in your sentence there!
pocmloc, Sep 27 2023

       // //somehow// is a very hard working word in your sentence there! //   

       Yep, it's going to be tricky. I'm visualising a drawer type of thing; bag is filled with filtered water, then the drawer is opened into the ocean side. To push the drawer open would take a huge amount of force (see link). The drawer would beautifully/perfectly fit in its housing (like the EDM (electro discharge machining)).
xaviergisz, Sep 27 2023


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle