5 ms·

> can produce 5.8 liters of fresh water per hour in full sun All this from a 10 x 10cm panel. Sounds too good to be true.

LeifCarrotson 4y ago
> On the roof, the device produced 72 mL of water in 4.5 hours. That sounds more in line with what I'd expect from Massachusetts sunshine.
- jeffbee 4y ago
  Why does the article describe the difference between 5.8l/h and .072l/4.5h as 50% lower? It looks to me like this ratio is more than 300x.
teknopaul 4y ago
I wonder why seaside towns don't all have 1m2 of this per 1000 residents. Seems like a no brainer if it really works but I can't believe it does.
sacred_numbers 4y ago
This is incorrect. The statistic of 5.8 liters per hour is actually liters per hour per square meter. A 10x10cm panel would produce 58ml per hour in full sun. If you instead put a solar panel with 20% efficiency and used the electricity for reverse osmosis you would produce about an order of magnitude more water per square meter. This may have lower capital or operational costs, though, and it is certainly more decentralized, so it could be useful in some situations.
- abrax3141 4y ago
  What’s the capital/op cost of r.o.? Isn’t it just a pump and a membrane?
- abrax3141 4y ago
  Just for clarity, is that a 10x10cm solar panel?
  - sacred_numbers 4y ago
    What I mean is that a solar panel of any given size will produce electricity that can be used to desalinate an order of magnitude more water than a device like this of the same size. The device in this article desalinated 5.8 liters per square meter per hour in full sun, which means this 10x10cm device actually desalinated 58ml of water in an hour. A 10x10cm solar panel would produce enough electricity in an hour in full sun to desalinate approximately 600ml of water using reverse osmosis. A solar panel that was 1 square meter would generate enough electricity to desalinate about 60 liters of water in an hour in full sun.
beambot 4y ago
The linked article's production rate is off by two orders of magnitude. If you read the actual paper [1], the quoted production is 5.8L/hr/m^2 -- i.e. 5.8L/hr would require a 1 m^2 system, not the 10cm x 10cm system under test (0.01m^2). https://sci-hub.se/10.1039/C9EE04122B https://sci-hub.se/10.1039/C9EE04122B

LeifCarrotson 4y ago

> On the roof, the device produced 72 mL of water in 4.5 hours. That sounds more in line with what I'd expect from Massachusetts sunshine.

jeffbee 4y ago
Why does the article describe the difference between 5.8l/h and .072l/4.5h as 50% lower? It looks to me like this ratio is more than 300x.

teknopaul 4y ago

I wonder why seaside towns don't all have 1m2 of this per 1000 residents. Seems like a no brainer if it really works but I can't believe it does.

sacred_numbers 4y ago

This is incorrect. The statistic of 5.8 liters per hour is actually liters per hour per square meter. A 10x10cm panel would produce 58ml per hour in full sun. If you instead put a solar panel with 20% efficiency and used the electricity for reverse osmosis you would produce about an order of magnitude more water per square meter. This may have lower capital or operational costs, though, and it is certainly more decentralized, so it could be useful in some situations.

abrax3141 4y ago
What’s the capital/op cost of r.o.? Isn’t it just a pump and a membrane?
abrax3141 4y ago
Just for clarity, is that a 10x10cm solar panel?
- sacred_numbers 4y ago
  What I mean is that a solar panel of any given size will produce electricity that can be used to desalinate an order of magnitude more water than a device like this of the same size. The device in this article desalinated 5.8 liters per square meter per hour in full sun, which means this 10x10cm device actually desalinated 58ml of water in an hour. A 10x10cm solar panel would produce enough electricity in an hour in full sun to desalinate approximately 600ml of water using reverse osmosis. A solar panel that was 1 square meter would generate enough electricity to desalinate about 60 liters of water in an hour in full sun.

beambot 4y ago

The linked article's production rate is off by two orders of magnitude. If you read the actual paper [1], the quoted production is 5.8L/hr/m^2 -- i.e. 5.8L/hr would require a 1 m^2 system, not the 10cm x 10cm system under test (0.01m^2). https://sci-hub.se/10.1039/C9EE04122B https://sci-hub.se/10.1039/C9EE04122B