Fresh Water Generation/Desalinization Machine? Any Ideas?

[bobroberts170]

Here in the US midwest we've been having a drought lately, and it got me thinking of ways that we might be able to generate water more efficiently. Right now we're pumping an enormous amount of water out of the Ogallala Aquifer, beyond the replacement rate even in a rainy year I think.

My understanding (correct me if I'm mistaken) of current desalinization plants is that they use a molecular sieve or reverse osmosis to separate seawater from fresh water, or a partial vacuum to flash boil the water. I don't know how much energy that sort of thing requires, but I know it is undoubtedlly more efficient than electrolyzing the water and combusting the H₂ and O₂ gases and condensing the water vapor immediately thereafter to obtain fresh water (otherwise they'd be doing that instead).

But what if the energy from that combustion was used to aid the electrolysis? Obviously, you couldn't use it to power the entirety of the electrolysis, but it would definitely reduce the input energy needed while operating continuously. You could use the heat to power a small turbine and generate supplemental energy for the electrolysis.

I'm a mechanical engineer by training, and so my chemistry is a little (okay, a lot) rusty, and I've forgotten how the math on that works. Here's my attempt so far:

From Wikipedia, assume 50% efficient energy to H₂ gas conversion (chemistry folks: is this a good assumption?):

1 m³ of H₂ gas generation requires 7 kWh of energy

Combustion of those gases would release 286 kJ/mol of H₂. Assume 30% of that energy can be channeled back to electrolysis (rosy assumption):

(1 m³ H₂ @ STP)/(0.0224 m³/mol @ STP) = 44.643 mol H₂

Energy obtained from combusting above quantity of H₂ gas:

(286 kJ/mol)(44.643 mol) = 3.55 kWh

(3.55 kWh)(30%) = 1.06 kWh

The total energy required to obtain 44.643 mol H₂O (Hydrogen gas and water are in equal mole amounts on both sides of the chemical equation for combustion of hydrogen and oxygen gas):

7 kWh - 1.06 kWh = 6.94 kWh

44.643 mol H₂O in liters:

(18.0153 g/mol H₂O)(44.643 mol H₂O)(1 L/kg) = 0.804 L

Crap. I don't like where this is going:

(0.804 L)/(6.94 kWh) = 0.1159 L/kWh

Well dang. Does all that math look right? My thinking was that the vast majority of the energy spent during electrolysis could be made up by burning the resulting gases and recapturing that energy. So I guess a better question would be how accurate these efficiency assumptions are. If my math above is solid (correct me if it's not), efficiency losses would have to be pretty close to zero for this to even maybe work.

Any thoughts? I guess I answered my own question. Any creative fresh water generation ideas?

 

[256bits]

Chemical reaction
2 H2O(l) → 2 H2(g) + O2(g)
has 2 directions:
The forward direction is the decompostion of water into Hydrogen and oxygen,
The reverse direction is the combustion of hydrogen and oxygen into water.
Both directions involve the same amount of energy, except that one requires energy input , the other an energy output, ie exothermic, endothermic.
It does not matter in what process the chemical reactions are carried out - suffice to say the energy does not change if the beginning and end products, reactants are water, hydrogen, oxygen.

So what you want to do is electrolize water into hydrogen and oxygen and the combust these gases, and then use some of the energy from the combustion to aid in the powering of the electolyss and whatever is left over for something else.

Your energy balance should be.
Energy in = energy out
electrolysis energy input + losses = combustion energy output + losses

You should see right there that there are 2 losses of energy that are unrecoverable.
One is in the electrolys of water in that you need to supply more energy for the reaction to proceed from water to H and O gases. It is an unrecoverable low quality form of heat.
The other loss is in the combustion process in which case not all of the energy of combustion can be usefully recovered. A typical power generating station is about say 40% efficiient.

The 2 losses add up together and that is the extra energy that must be added in some other form, from another generating plant for the energy balance euation to become equal. That extra energy will be lost to the environment.

Generation plant energy + combustion energy = electrolysis energy + energy lost to the environment

( Edit : that last equation looks kind of wonky as stated, and perhaps someone can clean it up a bit, if it is not understandable )

 

[zoobyshoe]

There's a whole separate problem if you are thinking of electrolyzing sea water, which is that you'll be generating hydrogen and chlorine, not hydrogen and oxygen.

About four percent of hydrogen gas produced worldwide is created by electrolysis. The majority of this hydrogen produced through electrolysis is a side product in the production of chlorine. This is a prime example of a competing side reaction.

2 NaCl + 2 H2O → Cl2 + H2 + 2 NaOH
The electrolysis of brine (saltwater), a water sodium chloride mixture, is only half the electrolysis of water since the chloride ions are oxidized to chlorine rather than water being oxidized to oxygen. The hydrogen produced from this process is either burned (converting it back to water), used for the production of specialty chemicals, or various other small scale applications.

http://en.wikipedia.org/wiki/Electrolysis_of_water

I assume this is why reverse osmosis or simple distillation have always been the only real options in de-salting sea water. (There's also about six other salts in sea water to complicate the electrolysis products.)

The Ogallala Aquifer is much closer to the great lakes than the ocean. I would think in terms of wind turbines that use the plentiful shore winds to generate heat to distill the polluted great lakes water and then pipe it to the aquifer. No entity will pay for that, though, unless you figure out how they can make money off it.

 

[Astronuc]

A lot of energy is lost due to recombination in electolysis, and the presence of halides is other complicating factor.

For many applications, particularly those in desert areas with a coastline, or with an abundance of fossil fuel, it is preferable to use distillation.

One does need a water source however.


I've wondered why areas prone to flooding don't simply pump the flood waters where there is little or no water. For example, where there was floods on the Red River years ago, or the Missouri and Mississippi Rivers last year, why not pipe the water to Oglala region? Now in the same area, the Mississippi is reaching record lows because of a lack of precipitation in the midwest and Missouri River drainage system.

 

[bobroberts170]

@256bit I knew that there would be those losses, but I wasn't sure how large they'd be. That was why I tried to do the math on it; to see if this was a doable idea, and it appears as though it's not. I don't know if I made it clear in my original post, but the water vapor from combustion is the desired output. It could be condensed and then used as fresh water source.

@zoobyshoe I knew that seawater was composed of water and salt almost entirely, but I wasn't aware that you would get that much chlorine gas by electrolyzing it. You learn something new every day!

 

[256bits]

The Ogallala Aquifer is much closer to the great lakes than the ocean. I would think in terms of wind turbines that use the plentiful shore winds to generate heat to distill the polluted great lakes water and then pipe it to the aquifer. No entity will pay for that, though, unless you figure out how they can make money off it.

The surrounding states and provinces of the Great Lakes use that water for their own purposes. Draining the Great lakes to provide fresh water elsewhere would have some political ramifications with those areas and also downstream along the St. Lawrence River and the entity St. Lawrence Seaway used for navigation.

 

[zoobyshoe]

The surrounding states and provinces of the Great Lakes use that water for their own purposes. Draining the Great lakes to provide fresh water elsewhere would have some political ramifications with those areas and also downstream along the St. Lawrence River and the entity St. Lawrence Seaway used for navigation.

If we have to invade and conquer Canada, so be it. Just kidding. This plan will never happen simply because no one ever seems to gear up to correct ecological disasters. If the aquifer were to get pumped dry, Nebraskans would either have to move away or learn to be desert dwellers.