> 1.5 gallons of fresh drinking water per hour for every square meter
7 litres per m^2 per hour.
7 litres per m^2 per hour.
2.46kWh/m^3 (energy consumed per unit of water produced) is claimed for reverse osmisis [1]. This equates to 8.86MJ/m^3.
Output for this still is 7L/(m^2.h)
Assume a solar flux of 1kW/m^2.
Energy consumed per unit of water is 1kW/m^2 / [7L/(m^2.h)] = 3.6MJ/0.007m^3 = 514MJ/m^3.
Assuming the above is correct (check anyone?), the still uses 58 times more energy than reverse osmosis. The solar energy may be "free" but with a 20% PV cell efficiency a reverse osmosis system would produce about 11 times more water per unit area of solar collector?
I think this is much more important to a great many people than the theoretical efficiency of reverse osmosis. Percentages greater than 100% always do well in media reports about these topics, but they're not necessarily the point of the exercise. If the goal was to produce a system that's as efficient as possible, the researchers wouldn't have used household-style supplies but more expensive, advanced materials.
Reverse osmosis is great for a central area such as a large city in a place with reliable distribution, but in many places around the world, an independent, affordable system that can turn seawater into drinking water for a family or two has much more value.
The actual membranes are $9 for enough for 100 gallons per day (retail prices, [1]).
High pressure pumps and hoses scale linearly. Solar panels scale linearly. Filters scale linearly.
In fact, using this solar fountain [2] as the basis for the design, and switching the pump impeller for a high pressure version, and the nozzle for an RO membrane and hose, you immediately have a drinking water machine for a few people for $20. The fountain already has a pre-filter built in.
[1]: https://www.aliexpress.com/item/32669709750.html [2]: https://www.aliexpress.com/item/1005002883892948.html
Expect to pay around $300 minimum for a small RO solar plant, and a couple bucks a month for upkeep. And logistics. Granted that's USA prices (upstate NY, not SV) but that prices out a lot of developing regions.
That combined with reverse flow flushing will probably last many years. And it can all be controlled by a 10 cent microcontroller, one pump, one valve, and a pressure sensor.
I don't doubt that high pressure pumps are bad... but that's just a design issue - there is nothing theoretically expensive about them.
Free chlorine (technically assorted chlorine oxides like hypochlorite) attacks RO membranes, so now you have to deactivate your reactive species first. Usually UV lamps, you use sun, but now you need UV-clear tubing, not easy.
Reverse flow flushing can be done with the components you mention, that's another $20/50 plus sourcing logistics.
Yes, the theoretical expense in quality pumps is quality. There are tighter tolerances, beefier components, better polymers, and more QA. It all adds up.
It's still all fairly cheap by Western standards, but it's a tall ask for a lot of places that barely have potable water.
There was a big push towards developing chlorine-resistant chemistries a few years ago, but as far as I can tell, that has fallen into a “researchers don’t know what they’re doing, the plants are already designed around this problem” narrative. Of course, those plants are huge investments, and maybe it’s correct that one wouldn’t be able to take advantage of chlorine-resistant membranes until a new plant was built.
Cellulose triacetate RO membranes have chlorine tolerance, but have inferior chemical stability, productivity and selectivity to polyamide membranes, so it is sometimes used where chlorine tolerance is an issue. CTA membranes are also available in hollow fiber format, while polyamide membranes are essentially exclusively found in spiral wound configuration (some operators want fibers for higher fouling/solids feed). CTA is limited to a much smaller pH window (like 4-6 versus 2-12), and are not suitable for more aggressive cleaning methods, so I’m not sure if it overall provides a benefit versus polyamide RO with more aggressive cleaning cycles.
The osmotic pressure you need to overcome to perform desal is directly proportional to the concentration. So 500-1000ppm water that you don't love is a lot less challenging than seawater which is about 3.5% or 35000ppm.
You can do "I want extra pure drinking water from 'normal' water RO" for a few hundred, sure because the pressures are in the range of 20-60 psi. Reasonably efficient seawater starts somewhere around 400-600psi and I've heard of plenty of systems that work at more like 800-1000psi. Different pumps, membranes, membrane housings, piping, all of it. It's a couple thousand dollars plus a fair amount of energy unless you buy even more expensive energy recycling pumps that use the pressure in the brine output to help on the input.
But a few hundred bucks will still get you from either "potable but not great" to fairly pure, or brackish to potable. If you have effectively infinite seawater and power, you can still pump at lower pressure and get pure permeate, it's just less efficient per unit pump and filter lifetime. If you had a super reliable pressure system (low friction ceramic pumps for example, not cheap but last forever) and cheap first-pass filters, you can run it for quite some time. But you're spot on in that you get to a point of engineering-give-a-mouse-a-cookie and it just makes sense to optimize the whole stack.
Also, I wonder which system would be easier to slap together in a garage out of leftover junk. Not everyone can rely on access to commercial-grade, turn-key solutions to a problem. A design that could be reasonably DYIed could be better in certain contexts, even if less efficient.
They also don't seem to keep good control of salinity within the stages - I suspect after a few hours operation the efficiency will drop as the salinity gets higher and higher within the paper towels.
Regardless - this isn't a scale able solution, but doesn't need to be - should work fantastic for a small family with negligible environmental impact.
> it's nice
come on dude, have some self-awareness
If you look at any map of the modern world, there is almost nowhere that we haven't transformed into farmland, cities, etc, and it's pretty good (nice) that we have left the uninhabitable islands to nature. We've consumed most of the land, which is both pretty amazing and also quite bad for biodiversity, as everyone knows.
I'm not debating the "cancer" metaphor which is definitely harsh - it has been catastrophic for other life, but great for us. Maybe that wording can be less harsh, but IMO that's not the point.
If I calculate one unit with 300 dollars this would mean 5.5 cents per person per day for one year for fresh drinking water. And after that only a little bit for maintenance probably.
From the article "The team estimates that a system with a roughly 1-square-meter solar collecting area could meet the daily drinking water needs of one person.", which sounds very different.
On this Q&A page [1] Zhang is quoted "Our current strategy, for example, is to use an assembly of 100 of these devices to achieve an area of 1 m2, which will increase the total production by 100 times to create 10-20 liters of clean water per day."
I don't want to sound too negative, but something doesn't add up here. Still, enough water for one person per square meter of desalination plant is an impressive result I think. The oceans are big.
[1] https://www.techbriefs.com/component/content/article/tb/feat...
(My faith in my own maths is weak ..)