Is it just me, or did this article dance around the question?<p>I am not a physicist but let me give it a stab: except for a few specialized steps like UV or oxidizing heavy metals, most filtration is mechanical. A series of filters with smaller and smaller pores capture more and more of the mess in the water like bacteria and particulates while UV breaks down viruses, the oxidizer precipitates out metals, and so on.<p>None of those methods work with salt. Salts in general disassociate through ion-dipole interactions - the water dipoles essentially rip the ionic compound apart and surround each ion in what is called a hydration shell. They're bigger than bare water molecules but not much bigger - much too small to target with pore size. This shell also puts them in a thermodynamically stable state and it takes energy to "jostle" the water molecules away from the ions either through evaporation, distillation, or through another chemical reaction that precipitates out the ions.<p>As it turns out, doing that takes a <i>lot</i> of energy, so we use reverse osmosis as a cheaper alternative: we exploit the hydration shell of the ions by putting them behind a semi-permeable membrane with <i>very</i> small pores, "nanopores" if you will. The pores are too small for water to cross normally, but under high pressures bare water molecules can be forced through the pores while the ions trapped in their shells remain and concentrate into a brine. It takes less energy but produces a concentrated liquid waste stream that must be disposed of.<p>Someone please correct any mistakes I've made
I was surprised how <i>cheap</i> it is. Desalinated water costs ~50 cents per 1000 liters [1]. That's about the same amount of water as a typical American household uses per day.<p>50 cents per day for a fully desalinated water supply is... incredibly cheap.<p>If you're interested in water policy and water management / engineering, I cannot recommend enough reading the book "Let There Be Water: Israel's Solution for a Water-Starved World".<p>[1] <a href="https://en.wikipedia.org/wiki/Desalination#Costs" rel="nofollow noreferrer">https://en.wikipedia.org/wiki/Desalination#Costs</a>
Slightly off topic, however, the post references the Carlsbad desalination facility. If you find yourself in San Diego and like oysters, I would highly recommend you checkout the Carlsbad Aquafarm. Take the tour and pick up some oysters.<p>What's really interesting and relevant to the topic is that the oyster farm serves as a pre-filter to the desalination plant and there's an symbiotic relationship between the plant and the oyster farm.
Relatively high energy cost since you’re undoing an endothermic reaction, you need to do a lot of it since we use water in large quantities… but most of all, the planet naturally does a lot of desalination for us already through various geological processes, so our “price point” for desalination is $0 per liter (infrastructure to capture rain, dam rivers, or tap groundwater isn’t literally free, but it’s pretty close - especially when it comes to the marginal cost for the next liter). It’s not difficult to desalinate <i>per se</i>, it’s difficult to desalinate extremely cheaply and at huge scale.
Some years ago I toured a maple syrup operation that has the opposite goal: Concentrate the dissolved stuff in the water. Their first stage was reverse osmosis, but only to a point. Second stage is boil, but with aggressive heat recovery from the steam to preheat the incoming liquid. All this to keep the energy cost under control.
MIT solar distiller in 2020 demonstrated a gallon and a half of fresh water in one hour using a square meter of close-quarter membrane distillation process.<p>So, 10,000 sq meter of this baby could pump 150,000 gallons of fresh water over a ten-hour solar shift.<p>Seems like the secret sauce is 1.2cm (or is that 80mm) separation between diffuser plates thus taking advantage of solar heating/condensation/collection in one area.<p>Of course, there remains an collection issue of brine discharge which could be removed gradually instead but in same but 3-peat manner (down to 2-3 permille, or 0.2-0.3% salinity level.)<p>At any rate, this MIT method has leapfrogged the passive solar method ahead of reverse osmosis (RO) method by quite a bit, in terms of energy required to extra fresh water. RO still holds the insurmountable lead in base (non-fluctuating) water output rate.<p><a href="https://news.mit.edu/2020/passive-solar-powered-water-desalination-0207" rel="nofollow noreferrer">https://news.mit.edu/2020/passive-solar-powered-water-desali...</a>
Probably very dumb idea, but would it be feasible to just pump and sprinkle sea-water on hot, dry desert and let it naturally evaporate, and then collect it as fresh rainwater back? i.e. how much water would you need to evaporate to have a noticeable increase in rainfall?
The energy costs are a bit of a red herring depending on local conditions. In California we currently "curtail" i.e. discard a huge amount of renewable energy in the spring season. If we can seasonally apply that energy to desalination, and store the fresh water for later, it is essentially a huge time-shifting battery that stores excess spring energy for the summer.
Solar stills are one of those basic survival tools anyone living in an arid region near a salty body of water should understand how to rig. They’re dirt simple and as easy to build as a prison still with the added bonus of only requiring the sun as an external power source.<p>Similarly everyone should know how to rig a basic water purification system using gravel, sand, and charcoal in series.<p>Even just as applied science experiments to do with kids they’re worthwhile.<p>Edit: Water based solar power is generally an area I think that deserves more research. While photovoltaics have their advantages, water is cheap, clean, and reliable. Heating water with sun during the day and using it for household heating at night is the simple application that I’m most familiar with, but I wouldn’t be shocked if there’s some scale where an economically interesting Carnot cycle becomes possible.
I work in microfiltration (a pre-filter step for RO), and my view is:<p>1/ it’s as much an energy and water storage problem as it is a technical problem.<p>2/ commercially, because of 1/, RO is a municipal sale. It is a civil initiative, rather than a commercial one, which means it gets crowded out by other civil decisions.
The author have mentioned that water with more than 1-2 promille of salt is considered not drinkable. But I have a bottle of carbonated mineral water from the mountains and the label says it contains 6-9 g of salts per 1 liter. So I guess it depends on which salt it is.<p>Also, the author mentions that people in US use 1100 liters per day (which is too much in my opinion), but not all this water needs to be drinkable, one probably can not drink more than 3-4 liters per day, and the rest of the water can be salty.
This one time on an episode of Survivorman, Kalahari iirc, I watched Les Stroud use a hole in the dirt and a clear plastic tarp covering it to act as a solar still. The condensation from his pee in said hole ran off to a collection container as pure H20 once evaporated.<p>I just remember thinking to myself "Bear Grylls drinking his own pee is such a philistine, here is an actual pro using science to remove all the water from that pee instead first." Genuine moment of awe personally.
RO does use a lot of energy to overcome the osmotic pressure and to create flux through the membrane. An interesting concept is using the reverse process, "forward" osmosis, to extract the energy where fresh water mixes with seawater, such as a river mouth. This is called pressure-retarded osmosis (PRO) and was tried at pilot scale by Norwegian power company Statkraft. Ultimately this trial was shelved due to being uncommercial [1], perhaps future membrane development will improve the viability of FO. And yes the membranes are quite different, RO membranes are relatively thick due to the transmembrane pressures required. FO requires a much thinner support for the active layer as there is no external pressure applied to push the water through (it is drawn through by the difference in salt concentration).<p>[1] <a href="https://www.powermag.com/statkraft-shelves-osmotic-power-project/" rel="nofollow noreferrer">https://www.powermag.com/statkraft-shelves-osmotic-power-pro...</a>
One thing that this article missed was that it was San Diego centric. In Israel, desalination is a much bigger part of the ecosystem. Over half of its domestic water comes from desalination. Quite a bit of the problem in California, as in almost every industrial application, is just that we make it hard to do anything with atoms.
Given the countless environmental challenges we are facing (and causing), we should more seriously and openly consider putting a stop to exponential population growth as an (at least short-term) solution.
It’s astonishing how some people preach blind faith in our ability to just find solutions for problems caused and exacerbated by never-ending population growth without identifying it as the root cause. Why is it a given that the earth can just withstand whatever we throw at it?
I think he fully gave the wrong answer. The main problem with desalination is capital cost. The Carlsbad plant cost one billion dollars to make. I bet very little of it is the cost of membranes, or the actual RO systems. It's simply a large plant, and building a large plant is expensive. The same problems that plague large nuclear power plants plague many other large construction projects, including large desal plants.
Pick one<p>1. Steam distillation + Product can be perfect DI type 1 water - Expensive: 300+ kJ/L<p>2. RO membrane + Cheaper - Slow - Wastes more water - Requires regular changing of membranes<p>The end.
I like the analogy at the end regarding nuclear power vs piping in water from freshwater sources. The upfront costs are high, but over time it would be cheaper than desalination, but due to short term governments and borders, it's hard to justify the upfront costs. So instead we are somewhat stuck with more expensive short term solutions.
Tell that to israelis: <a href="https://www.israel21c.org/how-israel-used-innovation-to-beat-its-water-crisis/" rel="nofollow noreferrer">https://www.israel21c.org/how-israel-used-innovation-to-beat...</a>
I’ve got an idea guys! Let’s build some artificial lakes and we can just let rain and river water pool in them! We can let natural precipitation do the desalination for us!
I was never able to find the publication again but I read an article one time about the work of 2 Australian professors (I forget which kind) who designed a giant circle shaped verical axle wind turbine with 2 tubes that suck sea water up along the sides and sprays it up in the air with the brine dropping near the turbine. An array of those they claimed could create dense clouds that produce a lot of rain at some (according to them) surprising ratio to cost.
Late to this thread but I have always asked why nuclear powerplants can't use seawater for cooling and the condense the steam to create desalinated water?<p>Ocean waves can be used to create electricity and it is also possible to create a dam that uses an artificial river sourced by the ocean and make hydroelectric plants to use that energy. Is it a cost issue in both cases that prevents using electricity generated that way to desalinate?
If you built a desalination system say… 500 feet under the ocean and have the pressure above pushing water through the filters, is it possible to lower the amount of required energy just a little? Then you’re more pumping water out of the system than pumping it through heavy filters
For a blog about "practical engineering", it doesn't address the fact that in practice, saltwater isn't just "water and salt" but also all kinds of particules, pollutants and impurities that will clog any membrane after a while.
Difficult is not an absolute. It needs to be "in comparison to something".<p>Desalination is verrrrry easy if nature does it for us (sun on the salty sea -> clouds -> precipitation over land). Compared to this, doing it "with a machine" is hard.
Stupid question time, why not expose the sea water over a larger area and expose to sunshine? There's plenty of room in the UAE for this, somebody is even building a wall that could hold the water which could be dual purpose and act as cooling via evaporation...
You could use simple pricing to influence behavior. 1100 litres a day for each American is so damn much. When you hike and stay in the mountain huts you are charged 3$ for a 4 minute shower.<p>You could probably fix the drought situations by reducing consumption.
It is not hard, just throw the salt water in a steam turbine room and get the output through aquatuners to cool it down from 95C to 20C<p>Put the steam turbines inside the steam room to recover some of the heat as power and you are good to go
Should heat and cool it with standing soundwaves. Salty water takes the heat, cools and falls (brine-fall) rest is less salty.add membranes at intersection points..
Can I shower in salt water?<p>Then all I need to do is desalinate drinking water.<p>“Distributed” home desalination for drinking water seems like the best approach in my mind, then people can pay as much or as little they need, but I have no real data to back this up.
several better ways, multi stage stills and reverse osmosis are leaders of the pack,<p><a href="https://www.veoliawatertechnologies.com/en/technologies/multiple-effect-distillation-med#:~:text=A%20Multi%20Effect%20Desalination%20MED,to%20produce%20clean%20distillate%20water" rel="nofollow noreferrer">https://www.veoliawatertechnologies.com/en/technologies/mult...</a>.<p><a href="https://en.wikipedia.org/wiki/Reverse_osmosis" rel="nofollow noreferrer">https://en.wikipedia.org/wiki/Reverse_osmosis</a>
"Turning sea water into clean, drinkable water costs $2 to $5 for 1000 gallons.<p>Less than half a penny per gallon is obviously absurdly cheap."<p>- Elon Musk, 2023-05-07 <a href="https://twitter.com/elonmusk/status/1655262008898383872?ref_src=twsrc%5Etfw" rel="nofollow noreferrer">https://twitter.com/elonmusk/status/1655262008898383872?ref_...</a>
While on the issue of water:<p>Some countries have floods in one part, and drought in the other.<p>The floods are so bad that large numbers of people die.<p>Here's a great challenge that should be worked on: how to capture the flood water and use it to mitigate the droughts. Could something like Elon Musk's Boring company concept fix this?