I first thought of this Idea about 1974. I was taking a class in chemical engineering, and the textbook had a chapter on a water-desalting technology that was called "Vacuum Freezing & Vapor Compression", or VFVC for short. The technique involves, first of all, using a vacuum to cause water to boil
and freeze at the same time (which it indeed will do). The reason it happens takes a bit of explaining.
First, water consists of a whole lot of water molecules, and the temperature of the water is an AVERAGE of the temperatures (kinetic energies) of the individual molecules. So inside a cup of water at room temperature some molecules are "hot" and some are "cold", and most are near the average, which is room temperature. (Look up "Gaussian distribution" or "bell curve" some time.)
Second, even at room temperature some of the water can evaporate; this happens simply because a few of those water molecules are literally "boiling hot", so they can exactly literally boil out of the water into the air. Note that they can do this because they have enough (kinetic) energy to do that --and when they do, they take their energy with them, out of the water. This makes the average temperature of the water left behind go down a bit, each time a molecule escapes this way. (Look up "evaporative cooling" some time.) The only reason the cooled cup of water can keep evaporating is because: (A) the air surrounding the cup gets a little hotter as those escaped hot water molecules mix with the air; (B) that warmed air in turn can interact with the cooled cup, which the air surrounds; (C) the cup and the air can thus equalize temperature; and (D) now the cup of water has regained the energy it lost with the escaped water molecules. We can think of this as a kind of "heat recycling", until all the water evaporates. It's a slow process, but it does happen this way.
Third, at high altitude atmospheric pressure goes down, and it is widely known that water boils more easily (and food often requires "pressure cookers" in order to be cooked properly). One way to look at this is to think about the surface of the water and the air above it. Lots and lots of molecules are bumping against each other. The air is a barrier against water molecules trying to evaporate. And if the air pressure is lowered, the number of air molecules above the water goes down, making the barrier thinner and more permeable. It means that water molecules that are not quite so hot as those previously described can escape, in addition to the very-hot molecules. So the evaporation rate goes up and the boiling point of the water goes down.
Fourth, if a cup of water is placed in a vacuum chamber, as the air is pumped out, the situation is a bit different from the cup merely sitting at high altitude. That's because the pressure gets lower and lower, instead of staying as some below-normal value. One obvious difference is that the water molecules that escape by evaporation can get pumped out of the chamber before they have time to give their energy to the air that surrounds the cup, so there is a break in the "heat recycling" described above. Another is that since the pressure gets lower and lower, water molecules that are merely warm can begin to escape, not just hot water molecules. What happens next relates to the simple fact that there are many many molecules that are merely warm, compared to the hottest water molecules. When they can evaporate "en masse" in a low-pressure environment, the effect is quite visibly obvious, and matches any ordinary description of "boiling". The steam coming out of the cup will be merely warm instead of hot, but it will LOOK just like ordinary boiling water.
Fifth, remember that "bell curve"? Now we get to contemplate all those water molecules that are colder than average. What are they going to do while the hotter ones are busy boiling off in a vacuum chamber? They are going to clump together to form ice! At the same time the hotter ones are boiling! I once saw a video of this in high school, and it was really nifty.
You can be sure I remembered that video when I first read about the VFVC system. It used a vacuum to boil and freeze water, as previously stated. One of the neat things about ice is, as it forms, it tends to exclude all other substances. In other words, if you freeze seawater, the ice that forms can be mostly salt-free. In a desalting plant that's obviously a Good Thing. So, in a VFVC system the first part, Vacuum Freezing, the main goal is to obtain the ice that forms, and to ensure plenty more forms as seawater boils in the vacuum chamber. Here we can imagine the ice being scraped off the top of the freezing seawater, into the next processing chamber.
In this chamber the ice needs to be washed. That's because while the inside of the ice is relatively salt-free and purified, the outside is wet with actual seawater. It needs to be removed, and we don't want to melt the ice in doing it. A cold-water rinse is the obvious thing to do. The wastewater can be sent back to the main freezing chamber, while the ice can be carried to the next chamber. Note that this rinsing is imperfect and a small amount of ice is going to melt and end up back in the vacuum chamber; it can't be avoided.
The next chamber is where the Vapor Compression part of the system is done. All the low-temperature steam that boiled off in the vacuum chamber can be compressed, and when the air pressure goes up, those water molecules will find themselves condensing together. The result will be fairly warm water, since every molecule in it came from the high end of the bell curve, of the distribution of temperatures of the molecules of the original water. It also is quite pure water, since it arrived via distillation, even if the distillation temperature was lots lower than normal. This warm purified water is added to the pure ice (and every molecule in that came from the low end of the bell curve), and so the ice melts and the temperature averages out to something like it was, before all these descriptions started, except that now we have desalted water, ready for drinking.
I've simplified the descriptions here just a bit; the textbook descriptions included some special heat-transfer fluid that was also involved. Almost immediately upon reading that part of the textbook, I got the Idea which is the basis for posting here. Because I thought (and still think) the textbook description was, bluntly, "stupid". Here is the EFFICIENT way to do Vacuum Desalting:
In the vacuum chamber, we want to maximize the exposure of the surface area of the seawater to the reduced air pressure. Layers of trays can be placed throughout the height of the chamber, and seawater is piped to one end of each of them. At the other end is an outlet for brine. Also in each tray is placed something that resembles the radiator of an automobile (except it needs to meet certain purity standards). If we pump warm fluid through the clean radiator, heat transfer will prevent the ice from forming in the vacuum, and simply encourage more water to boil.
The vacuum pumps furiously remove all the steam coming from all the trays of seawater in the chamber, maintaining the low pressure required for continuous boiling. On the other side of the pumps, the steam is pressurized back into condensed water. As described above, this will be warm water, guaranteed by the physics of the process (it may even be a bit warmer than you might first expect, since likely some energy was added to it as its molecules were mechanically swatted by the pumps). THIS is the warm fluid we pump through the clean radiators in all those trays! The "heat recycling" previously described for simple evaporation is here abbreviated (air is not involved) and employed on a grand scale. It is possible that some additional heat may need to be applied, just to get it STARTED (need quite a bit of warm water in those clean radiators), but in general the separate outputs of brine from the trays, and cooled water from the clean radiators, will be nearly the same temperature, and should average to equal the temperature of the incoming seawater.
Exiting from the clean radiators, the boiled/condensed/cooled water is ready for drinking. No filters, no high temperatures, just a simple kind of mechanical separation (via vacuum pumps) of seawater into brine and fresh water.
===============
A few years after I thought this up (and told various people in my class), I encountered a description in a magazine of a prototype desalting device that worked exactly as I've described here. It was somewhat discouraging, but I haven't seen any widespread adoption of it, so, since the Idea is at least independently original with me, and Jutta allows not-widely-known Ideas to be posted here....