oceanic distillation

[Sumaleth]

We have a bit of a water crisis developing here in Australia, and I had this idea:

 

First, we allow the ocean to naturally flow into a small chamber, above which sits an array of mirrors that redirect and focus sunlight onto the smallest possible area* of the water's surface.

 

The sunlight boils the water at the focal point, releasing steam which is allowed to condense into a collection pipe.

 

Lots of details to work out, but I'm wondering whether (with fine-tuning) this has the potential to generate enough drinking water to make setup worthwhile?

 

I don't expect it would be a great lone source of clean water, but maybe as an efficient top-up system. By its nature, it would work on the days when it's not raining.

 

Rowan.

[insane_alien]

so a solar furnace.

 

it would work but you need to remove the large concentration of salt that would build up in the system, this can be done simply by flushing the system with normal sea water (at night probably since it wouldn't be operating anyway).

 

you would also have to thermally isolate the chamber to get any sort of efficiency and remember its going to take a lot to heat it all up after you've flushed the system or even just starting up in the morning.

 

it would also have to be massive to provide enough water for a town.

 

it could work, but it would really only be viable for small towns and only the onesnear the coast at that.

[Sumaleth]

I had in mind a chamber that is open to the ocean, so that water is free to come and go. There'd be a small entrance, to keep turbulance to a minimum, but the coming and going of water would take the salt.

 

I also didn't see the chamber being heated up, which as you say would have build-up time. I'm thinking instead that it could focus the light so finely that you generate spontaneous evaporation at the point of contact with the water.

 

Even with something as small as a magifying glass you can generate a lot of heat at the focal point, so I imagine that an array of mirrors focused tightly onto water should be able to evaporate it, even with the water moving around and even with salt in the water.

 

If the focus is tight and water movement minimal then there shouldn't be too much wastage, and nothing is wasted building up heat in the system.

 

As you say, this couldn't supply a town with water, but I'd be curious to know how what volume of water it could generate from a plant of a given size.

[insane_alien]

why would turbulence be a problem? also the finely focused light would be inefficient as you would dump a lot of the energy into the vast cloud of mist that would suddenly spring up. you'd maybe be better with a more dispersed beam working on a reasonably big area and getting a good boil going on. since the whole place would be hot there would be less condensation in the air and less heat would be lost to water in the air.

 

i'll write up an equation to work out how much water you'd get out of it. give me a while on it.

 

<update>Okay, so i took valuable webcomic reading time and worked it out.

 

power required

 

= Q*(Cp*dT + dHv)

where Q = mass flowrate of water to be vapourised

Cp= heat capacity of water (4186 J/kgK)

dT= temperature change (75 K)

dHv= latent heat of vapourisation (2260000 J/kg)

 

power in

 

= A*I

where A = area of mirror(assuming 100% effectiveness)

I=solar irradiance (W/m^2)

 

so A=Q(Cp*dT +dHv)/I

 

plug the numbers in and out pops A=2524*Q (roughly) call it 2600 for all the losses. (this is called a fudge factor among engineers)

 

so, for 1 kg/s you would need an area of 2600 m^2

 

this is pretty big. and thats at peak irradiance so its going to be greater even than that.

[Sumaleth]

wow, great post.

 

re turbulence, I thought the less movement of the water the more efficient it would be.

 

re steam, I hadn't thought about the generated steam obstructing the generation. Perhaps if the beam entered the chamber at an angle, and the steam was sucked away on the other side. Or, as you say, boiling up the whole chamber, but there you need to regulate water in/out plus there's heat-up time, so you lose efficiency two ways.

 

I guess this is where practical experimentation would be worthwhile, to find the optimum setup. There might even be different optimums for different environments.

 

re 2600 m^2, that's not as efficient as I hoped it might be, but it's still a smaller footprint than a recycling or desalination plant, and would be a lot cheaper to run. 1kg/second sounds pretty good too -- about 3600 litres per hour of sunlight.

 

I'll see if I can find how that compares to desal/recycle/rain.

 

Thanks for the great post.

[Sumaleth]

It doesn't stack up well against desalination. You can get a truck-sized desalination plant that can produce 10000 litres per hour, and the really big plants in the middle east can produce 34 million litres of clean water per hour, 24 hours a day.

 

The cost of maintenance and volume of greenhouse gas emissions are probably more important here in Australia that straight volume, but it's still not looking as attractive as I'd hoped.

 

And I found this, which notes that it's an old idea anyway:

 

http://www.itdg.org/docs/technical_information_service/solar_distillation.pdf

 

End of exploration I think. Thanks for helping me work through that.

[insane_alien]

yeah no problem, reverse osmosis is probably the better way of desalinating stuff, just pump it trough a membrane.

 

the analysis i done was for optimal conditions, they aren't going to happen. i would probably say that an average over the whole day would be 4-5 thousand m^2 per kg s^-1

[OilMan2007]

Ok I am not ready to call this a dead issue I am very interested in using a solar furnace to distill water. My though would be to focus on a heat exchanger which would allow temp concentration to a level of flash steam. Since it would be closed loop no refraction would occur from the steam.

[insane_alien]

the fact still stands that you'll need 2600 m^2 for every kilogram per second. That analysis was done for 100% efficiency. which just isn't possible, the mirrors will absorb some energy, heat exchanger will reflect some. there are a whole lot of problems with this when we have a better way of doing it. namely reverse osmosis.

[imp]

What will the Ph be of the distilled water produced by the device? imp

[insane_alien]

roughly 7. its just pure water.

[Rocket Man]

2600 m^2

50 x 50 metres. not that unreasonable.

the only problem is actaully getting around to building a plant that scale.

the only way i can see it being feasable is if the energy could be recovered in the condenser to run... reverse osmosis in parrallel?

or would you pre heat the feed water?

[ecoli]

roughly 7. its just pure water.

 

don't you get carbon dioxide dissolved in it, which lowers the pH?

[insane_alien]

you could preheat the feed water using the condensing water stream i suppose. it wasn't in the spec given. but ou could use it. and the 50m by 50m is for a litre per second. that isn't much if its a big plant. i'll carry out the calculations for a preheated one later. i'm studying now and i'll be away with the green fairy after my exams.

[insane_alien]

you would get some carbon dioxide but it would be tiny. it would be less than the CO2 in your tap water. in fact, the treatment plant next to it would be putting stuff in rather than taking anything out.