Almost two years ago I commented in an HN thread about this,
"DEKA is a strange duck. They have some genuinely brilliant people there, working on some genuinely revolutionary inventions -- but they just can't seem to actually bring them to market. Or, they're not interested in bringing them to market. ... I've yet to hear of the Stirling engine / water purifier being deployed anywhere; ... I really don't "get" them. The best I can figure is that everyone (influential) there is happy just to be working on these puzzles, and they don't really care if anyone uses them or not."
Actually, I'd say the truly brilliant people are in the marketing department, building huge hype for something which... presumably doesn't work as well as advertised or else we'd have heard about it again in the intervening three years. I don't know precisely what is wrong with it, but it's safe to assume that there's something.
They're taking saltwater from brackish aquifers inland, not pumping it from the sea. In terms of pumping costs, it's no different than their current sources of water that are pumping out of the Edwards aquifer or other fresh groundwater sources.
If you assume more than enough energy, likely from advanced solar (a big assumption, but we haven't been to the future yet), then pumping water can serve as energy storage in addition to more direct use of water.
We do the same thing with oil, I don't see why water would be any harder. Though granted, oil has higher cost and different outside factors than water, so I really can't make any 'is it economical' claims.
Out in the archipelagos in Finland (and I guess also Sweden) converting salt water into drinking water is becoming more common as it's quite easy when the salt amount in the Baltic Sea is quite low compared to the oceans.
Most people believe that bottled water is just tap water marked up a few thousand percent, it would be interesting if these desalinator plants sold bottled water. They could run at a profit apparently, and they could make an interesting argument:
"Conserve our precious natural water sources by drinking -Wapure-, purified water."
This would then make it a good thing to drink bottled water since you would be keeping cows and what not alive.
This is one of those things where I wish the report would include their methodology. This report was brought up during the "Great Bottled Water Debate" at Google and a couple of people wondered how they came up with their numbers.
As it turns out, if you sit at a dump site (at least in Northern California, and Google is fortunate to be built right next to one) and watch the trash being unloaded you will note that there is little recyclable material left in the trash when it gets there, further the waste management company does some separation as well because they get to keep any money they get from the recyclables that get that far (unlike the ones in the cans at the curb which they split with the city).
What was clear though was that between a fairly large homeless population which relies in part on recycling fees and waste management contracts that are structured to provide a disproportionate benefit to the waste management company, they both get paid for disposing 'X' tons of waste which is measured on the way in, and they get paid for 'Y' tons of recyclables they recover post ingress weigh-in so its a double win for them.
The result is that in California there are several mechanisms in place which select for recycling. This results in very little recyclable material actually getting into landfills (which is good) but it also makes statements like the one you link, essentially false.
Question: why do desalination plants use a filtering system, instead of evaporation? (like that desert survival trick of spreading plastic over a hole)
(guessing) Is it because (1) evaporation is actually very energy inefficient compared to filtration, especially at scale; or (2) evaporation doesn't separate out impurities that also evaporate.
that desert survival trick of spreading plastic over a hole
Solar stills are cost effective at very small scales, even for devices that are built in a more permanent fashion than plastic sheeting over a hole. There was a program where the US government gave rural families solar stills in the North American southwest a number of years back.
> Reverse osmosis plant membrane systems typically use less energy than thermal distillation, which has led to a reduction in overall desalination costs over the past decade.
But the wikipedia article says that 85% of desalination worldwide still uses distillation.
Seems like a good area to use solar energy though. Water is easy to store and if you can make enough on sunny days to carry you through nights and overcast times, efficiency doesn't really matter as much since your energy source is "free."
I'm sure it's possible, and that someone has sat down and done the mathematics, and figured out that it's just not worthwhile.
The most efficient way to do it is to use solar thermal, not solar electric, power -- build a huge array of mirrors to focus on heaters which boil your water and distill it out. But it's not really "free" -- you need a lot of land, a lot of mirrors, and a staff of people who keep the mirrors clean and replace the broken ones.
Oh, what the hell, I'll do the maths myself. To heat up and boil one litre of water from 20 degrees C takes 2.6 million joules. Sunlight, on a sunny day, is one kilowatt per square metre; let's generously assume we can get 50% efficiency over the eight hours of the day when the sun is high in the sky [don't forget, this is solar thermal power, it's more efficient than photovoltaics]. So each square metre of collecting area in our plant can boil 5.5 litres of water per day. The desal plant in this article produces 27.5 million gallons per day so to replace it with solar we'd need nineteen million square metres, or nineteen square km, of mirrors. That's a lot of mirrors, and a lot of squeegee men.
If you wanted a solar-powered desal plant I'm sure you could do it cheaper using photovoltaics plus reverse osmosis.
Wouldn't it be possible to somehow recover the latent heat of the boiling water when it passes back into the liquid phase, minus the salt? I'm pretty sure I learned the theoretical maximum efficiency for this in my undergrad physics class but I can't quite remember.
Maybe you could use the steam to pre-heat the water that enters the system? If the water is at 60C instead of 20C to begin with, that will reduce the amount of energy needed to boil it. Besides, all that steam needs to be condensed back to water anyway if people are going to drink it. (I think that's what they already do in advanced "multi-stage" distillation plants.)
Theoretically, you could also add a steam turbine to make some extra electricity on the side, but I don't know how efficient such an add-on might be.
But then you need a somewhere sunny, with low land prices on a coast with a large demand for fresh water.
Otherwise you have to pipe the fresh water uphill to the inland cities - much easier to pipe it downhill from the mountains.
Apart from Australia, agricultural bits of california and some parts of the middle east there aren't many places that fit.
My understanding is that it's all about efficiency. The impurities I don't think is too much of a problem (think about distilled water, they make that with evaporation), but filtering is much easier to scale. You can increase pressure, increase the number of filters, etc. to get it all to happen at a faster rate. combined with the fact that you don't lose any water from it and you can probably do very very well there.
Depends what your impurities are. If you have a lot of other liquids mixed in (oil , organic solvents etc) then distillation can be easier - it also lets you separate them for disposal.
Small scale pure 'distilled' water eg. for battery top-up, is typically made by ion-exchange resin - it's more effective than osmosis for removing very small amounts of inorganics and cheaper than actual distilling if you are starting with fresh water.
(Your second number is off -- it's 2,260 kJ/kg or 2,260,000 J/kg)
There's no therodynamic objection that you can't recover (virtually) all the heat you spent creating steam, when the steam is recondensed back to water. Which is what they do. Distillation methods (multi-stage flash (MSF), multiple effect (MED)) "only" use around 200-400 kJ/kg of heat per water output.
Some perspective: at $4/thousand cubic feet natural gas (US of course), this is a heating cost of $0.75-$1.50/cubic meter of desalinated water. High, but not impossible.
Your comment about California having a "long coastline" got me wondering how it compares to other places. It turns out that even if you measure the length of the coastline quite generously (i.e. tidal shoreline) the length of the California coast is 3,427 miles.
That sounds pretty impressive until you compare it with places with more crinkly coastlines - Norway has ~13,000 miles of coastline and even tiny little Scotland has 6,158 miles.
However, I suspect that there won't be much call for de-salination in Norway or here in Scotland (where it has been raining for most of the last week).
I think approximate linear distance is closer to a practical measurement than crinkly measurements. Consider a semi-circular bay. You may have more access (water/land interface) points, but all the pipes have to go out "far enough" to the same ocean. I suppose a really large bay might not have this problem, but then it looks much less crinkly, and more linear, than a small bay.
That's a bit deceptive though because that figure for Maine includes all the islands and what not. Maine isn't very big, there's no way it has that much actual coastline.
The length of the coastline helps with delivery. If you have 1 mile of coastline, you have to transport your water everywhere else that you would like for it to be.
Desalination plants are already operating in many coastal areas of CA. The problem is that, like the article mentions, seawater desalination is very expensive and much less efficient than bringing water down from the sierras, so its generally used as a last resort.
I'm not sure that length of coastline is actually the limiting factor.
It's not like a state with a mere 3000km coastline would have difficulty fitting the pipes in
(Production of desalinated water costs 2.1 times more than
fresh groundwater and 70 percent more than surface water,
according to El Paso Water Utilities.)
Desalination works by reverse osmosis and costs energy. Rather a lot of energy, in fact.
Osmosis-based energy production goes by forward osmosis and produces energy. However if you were paying attention back there when I said "thermodynamics is a bitch" it should come as no surprise that the energy you get out by wasting your fresh water isn't as much as the energy you put in to get it in the first place.
> it should come as no surprise that the energy you get out by wasting your fresh water isn't as much as the energy you put in to get it in the first place
Agreed, but maybe the effect can be used to alleviate the cost of getting the fresh water? For example, it doesn't cost so much to heat up a building once it's been heated up if it has heat exchangers in its HVAC: the output heat is used to heat up incoming hot air. A wasteful process can be made more efficient with stupid tricks.
http://www.wired.com/wiredscience/2008/03/colbert-and-kam/
http://www.thedailybeast.com/newsweek/2008/04/04/big-problem...