Chemists: sink my water car!

September 16, 2008 11:08 AM

Why is a water-powered car said to be impossible? (Via electrolysis.)

First, I'm not asking why they're not being produced by GM today, nor am I asking why the people selling "plans" for them are fraudsters. I understand that current designs use more energy "splitting" water into hydrogen and oxygen than is produced by burning the two.

But in the Water-fuelled car page (and in the page on Oxyhydrogen), Wikipedia states that it's impossible to do: a water-powered car would essentially be a perpetual motion machine. I don't understand why it's impossible, however. (I'm not a chemist, so please don't explain this by listing out the chemical reactions like Wikipedia does!)

The simplest explanation I've seen is essentially that you split H₂O into H and O, and it gets recombined to H₂O for combustion (though not in liquid form), and thus it's impossible that the reaction would give off energy. I suspect that I'm massively misunderstanding, but I don't see what this has to do with anything: the splitting and recombining isn't how the energy is released, just like cutting up firewood isn't how I heat my house: it's burning it that releases the energy. (I suspect that this shows just how bad my understanding of chemistry is.)

Furthermore, I've learned that:

- Catalysts, or even just salt, can make electrolysis considerably more efficient. (From the Wikipedia page, Electrolysis of water, "Pure water has an electrical conductivity about one millionth that of seawater. It is sped up dramatically by adding an electrolyte (such as a salt, an acid or a base)." Would a salt-water powered car be possible?

- Above 2000 degrees Celsius, water breaks into hydrogen and oxygen on its own (see Thermolysis). I recall reading elsewhere that hydrogen, mixed with oxygen, burns around 3,000 degrees Celsius. Once the engine is warmed up, couldn't the excess heat given off by the engine be used to "split" water? (Momentarily ignoring the fact that "splitting" it at 3,000 degrees would cause the hydrogen and oxygen to burn immediately?)

Again, I completely understand that current systems use more energy to "split" the water into its respective elements than they produce by burning them. What I don't understand is why it's said to be impossible to ever produce a net gain of energy by burning the two. (It intuitively makes sense that it's not possible, and if the thermolysis idea were possible, one could also use that to blow up the world's oceans?)

Every explanation I've seen for why a "water car" is imossible attempts to use complex chemistry to "prove" that it's impossible, which loses me. Further, every explanation I've seen seems to address electrolysis of pure water, not saltwater or water with a catalyst added... Nor do they address the thermolysis concept. Can someone explain this such that a dummy like myself can understand?

posted by fogster to science & nature (29 comments total)

You start with water. Burning the hydrogen and oxygen together produces water. So you start and end with the same thing. Where does the energy come from?
posted by GuyZero at 11:17 AM on September 16, 2008

Every explanation I've seen for why a "water car" is imossible attempts to use complex chemistry to "prove" that it's impossible, which loses me.

Also, this does not mean the explanation is wrong.
posted by GuyZero at 11:17 AM on September 16, 2008 [3 favorites has favorites]

All the catalyst will do is make the reaction a little more efficient. As GuyZero says, the fundamental "something for nothing" perpetual motion problem remains: burning hydrogen and oxygen results in water, which is what you started with. This is very different from burning normal fuels.
posted by exogenous at 11:20 AM on September 16, 2008

You can't gain energy by splitting water and then putting it back together. A physical analogy might be "You can use an elevated weight to turn stuff, so why not use a car that's powered by lifting up weights and then dropping them?" You are always going to have to do more work to lift the weight than you'll get back by dropping it.

For this kind of question it's worthwhile looking at how much energy the whole system has, and ignoring the steps in between. Thermodynamics says your system cannot gain energy.

So: at time 0 you have some water. If at time 2 you still have the water and some kinetic energy, that is impossible. It doesn't matter if at time 1 you turned the water into hydrogen and oxygen (by turning electricity into chemical potential energy), or into cream cheese (by .. I have no idea). You still can't gain energy for free.
posted by aubilenon at 11:21 AM on September 16, 2008

the splitting and recombining isn't how the energy is released, just like cutting up firewood isn't how I heat my house: it's burning it that releases the energy.

I believe the misunderstanding is that the "burning" is actually identical with the recombining.
posted by grobstein at 11:22 AM on September 16, 2008

It takes more energy to split the water into oxygen and hydrogen than you get in return. That's the basic answer. If you want details, for instance, as to why keeping an engine at 2000 degrees C will take more energy, you're going to get into the math... but most car engines operate at ~80-100 degrees C if that helps.
posted by fusinski at 11:24 AM on September 16, 2008

Oh, and nevermind the fact that typical engine materials would totally break down at temperatures that high. :)
posted by fusinski at 11:26 AM on September 16, 2008

It's not impossible, it's just stupid.

Step one: Split water into H and O.
Step two: use H and O as fuel

The equipment and energy supply that you use to split the water to make your fuel, weighs a lot. Why carry a half-baked miniature version of this equipment, and the water feedstock, and energy source, around with you in the car, (which thus requires more energy to get moving, and has less luggage space) when you can do it properly with a bigger, more efficient equipment and energy supply at a gas station, have that produce the fuel, and only carry around the fuel in the car?

Ok, maybe you object to that gas-station model because you like the idea of not needing to go to a gas station via having all the equipment, water, and energy source, carried around in the car so you can make your own fuel as you go. Once again, this is stupid - an engineer would cut out of the middle man - the equipment and the water and the fuel it produces - and just use the energy source to power the car directly.

The closest thing to feasible would be something like a solar powered car, where your energy source is too small to power the car directly, but accumulates enough over a day for a few miles, and you store this energy as fuel from electrolysis rather than put it in a conventional battery. Again, it makes more sense to offload the solar cells and electrolysis to your garage, so the cells can be optimally positioned to face the sun all day, but if you desire the independence of a fuel source even when out in the middle of nowhere, then a solar car that uses electrolysis instead of NiMH chemistry, might have a place.
posted by -harlequin- at 11:28 AM on September 16, 2008

It takes energy to split the water. The energy released by recombining the hydrogen and oxygen is, according to the laws of physics, always less than the energy required to split the water. So if you have the energy to split the water in the first place, you can just use that to directly power the car instead of messing around with a whole chemistry project on the side.
posted by mullingitover at 11:28 AM on September 16, 2008

the splitting and recombining isn't how the energy is released, just like cutting up firewood isn't how I heat my house: it's burning it that releases the energy. (I suspect that this shows just how bad my understanding of chemistry is.)

Yeah, this is where you go off the rails. The "burning" is the combining. Some chemical reactions give off energy, they're called "exothermic". Others consume energy; they're "endothermic". What you think of as "burning" is just a series of self-sustaining exothermic reactions happening in one place.

So to oversimplify, the water car would be doing an endothermic reaction followed by its opposite exothermic reaction. They cancel each other out. (And require extra input energy to get the reactions started, to boot.)
posted by ook at 11:30 AM on September 16, 2008

Note, however, that you can do these two steps of (1): splitting water into H & O and (2) recombining H with O to make water + energy, as long as you split the steps.

So you have a factory that splits water apart, powered by the grid. It packs you capsules of hydrogen and of oxygen. You buy these, and power your car with it. It's still a net loss of energy, but it shifts the fuel source for your car in a useful way. That's the hope, at least.

This is known as a hydrogen Fuel Cell.
posted by Lemurrhea at 11:30 AM on September 16, 2008 [1 favorite has favorites]

"Also, this does not mean the explanation is wrong."

Agreed, I'm just trying to better understand why.

Perhaps again showing my naivety, I'd figured that burning hydrogen / oxyhydrogen would just result in combustion, and thus the kinetic energy. I didn't think you'd get the water "back" out of it (which would certainly be impossible), but that... something... would be expelled, just like burning gasoline in car.

In other words, I didn't see it as "free" energy, but as extracting the energy from water, just like with gasoline.
posted by fogster at 11:36 AM on September 16, 2008

What I don't understand is why it's said to be impossible to ever produce a net gain of energy by burning the two.

I think it's a semantic misunderstanding - you do get a gain by burning the two (this is how fuel cell cars work). But the term net gain would imply that you got more energy from burning than what it took to electrolyze them, which would break laws of physics. Therefore, you didn't get a net gain, you just got a gain in one sub-section of the system for a cost in another sub-section.

You can make a car that runs on H2 and O, but if the car creates the H2 and O fuel, then it doesn't really run on H2 and O, it runs on whatever energy source is being used to create the H2 and O fuel, because the H2 and O fuel is just a middle-man process that connects the energy source to the motion of the car.
posted by -harlequin- at 11:41 AM on September 16, 2008

I didn't think you'd get the water "back" out of it (which would certainly be impossible)

Not only is not impossible, it is the inevitable result of the burning reaction!
Hydrogen + Oxygen -> Water + Energy

All burning is a chemical reaction. With hydrocarbons, you get carbon dioxide and water when you burn them, plus energy. Burning hydrogen is cleaner, but as noted above, it is hard to get pure hydrogren (i.e. it takes lots of energy, which has to come from somewhere).
posted by exogenous at 11:41 AM on September 16, 2008

I didn't think you'd get the water "back" out of it (which would certainly be impossible), but that... something... would be expelled, just like burning gasoline in car.

Yup. Burning H2 and O creates water. Water is burned hydrogen.
posted by -harlequin- at 11:42 AM on September 16, 2008

grobstein has it, I think. "Burning" the hydrogen and oxygen puts them back into a low-energy form, namely the original water you started with. There's nothing special about the fact that it's combusted; the same amount of energy is released if you put the hydrogen and oxygen into a fuel cell and react it with oxygen in there without any flames.

To use your firewood analogy: the tree starts with water and carbon dioxide. It uses the energy of the Sun to combine these constituent chemicals into a higher-energy state, namely wood (cellulose and lignin.) You then burn the wood, liberating the Sun's "stored" energy and turning the cellulose and lignin into water and carbon dioxide again. Similarly, you're proposing to start with water; use some source of energy to convert it to a higher-energy state, namely hydrogen and oxygen; and then burn it, liberating the energy you "stored" in the hydrogen and oxygen but no more than that amount.

You're correct in saying that there's nothing important about splitting the wood, but that's a red herring: splitting the wood isn't the chemical change that causes the sun's original energy to be released.
posted by Johnny Assay at 11:44 AM on September 16, 2008

I suspect that this shows just how bad my understanding of chemistry is.
You've nailed it here. Burning hydrogen is, by definition, the act of combining it with oxygen to make water. This is the key point that I think you are missing.
When you say "the splitting and recombining isn't how the energy is released" you are missing the fact that what you call "recombining" IS burning. (Your wood analogy doesn't work, because when you burn wood you combine it with oxygen to make water and carbon dioxide; the other side of that equation is the conversion of carbon dioxide and water into wood that is done by trees using energy from the sun; the cutting up part is irrelevant).

The two chemical reactions are as follows:
water + heat → hydrogen + oxygen
hydrogen + oxygen → water + heat
You to supply the same amount of energy (heat) to split water into hydrogen and oxygen as you would get from burning the amount of hydrogen that you liberate. Even if things were 100% efficient (which they can never be) you could at best break even.

With this point of chemistry clarified, I'm sure you now see that there will never be a net gain from splitting then recombining water.
Your thermolysis engine won't work because it takes as much energy to separate a given quantity hydrogen and oxygen as they release when you burn them. Even if the engine were 100% efficient (which it could never be) then burning the hydrogen would only provide enough energy to split the same amount of water, with none left over to move the vehicle.
posted by nowonmai at 11:45 AM on September 16, 2008

I think this pretty well covered already but I didn't see this yet:

"burning" MEANS (for most materials) "adding oxygen to" (also called oxidizing)

Burning gasoline ALSO does not have a net energy gain. It took more energy to create the gasoline than you get out of it. However, most of this energy came from the sun, a long long long time ago, so to us it just appears underground (in oil form) and we think of it as a net gain - because we ourselves did not put the energy into it to make it.
posted by RustyBrooks at 11:49 AM on September 16, 2008 [1 favorite has favorites]

Uh, to answer the exact question as stated, a water-powered car is not impossible. Here's at least three ways you can do it.

1) You have a tank of highly pressurized water stored in an enormous tower on the back. When you want to go forward, just open the spout and the force from it pouring out should push the car forward a bit like a jet.

2) Using electrolysis - so you've got this huge battery and a little electrolysis device, then a combustion device. As long as you have the battery juiced up, you can use electrolysis to drive your car, but with the following caveat: not only will you have to recharge the battery, your electrolysis/combustion cycle will be less efficient than just using the battery to drive a motor directly.

3) Okay, so, skip the charged battery. Get a chamber full of hydrogen and oxygen, allow them to combust, you will get some energy out of it, you can store some in a capacitor, use it to split the water ... but this cycle, you won't get quite as much hydrogen and oxygen as you had to start with. You'll have some water left over because you have lost bits of energy here and there. This cycle rapidly dwindles down to zero.

What you're getting at, though, is something different. You're looking at a closed path - that is, you end up exactly where you start. It's pretty much the same thing as that endless Escher waterfall. You can't really pull water up using a water wheel driven by a waterfall from the water you pulled up and not lose energy. That's a perpetual motion machine, with a bonus of driving your car around.

You are correct in that above 2000 degrees Celsius, water breaks into hydrogen and oxygen on their own, but you can get they get cooler in the process.
posted by adipocere at 12:07 PM on September 16, 2008

Your car burns hydrogen with oxygen. This releases energy.

To make the hydrogen and oxygen, you electrolyze water. This takes energy. More energy than you got from burning the hydrogen and oxygen.
posted by ROU_Xenophobe at 12:28 PM on September 16, 2008

Actually, it isn't impossible. There's a lot of energy in water. If we can ever figure out how to directly convert mass into energy, you could in principle build a car that could run a million miles on a teaspoon of water.

But we don't know how to do that yet. (And I'm not sure I want us to ever figure it out, because the same technology could build a bomb big enough to destroy this planet and create a new asteroid belt around the sun.)
posted by Class Goat at 12:46 PM on September 16, 2008

Can you really get better gas mileage using your car's engine to make "Brown's gas?"
September 12, 2008

Dear Cecil:

An auto mechanic friend claims to have a gizmo that makes his vehicles run at least partially on water. He swears it's true and has about five test vehicles running with this thing now. It will work best on vehicles with carburetors — fuel-injected vehicles need tweaking of the computer chips. He's got one on an old VW Bug and says he gets about 80 MPG — he's trying to win a prize for getting over 100 MPG. He installed one in a large diesel truck that originally got about 8 MPG; it supposedly now gets 20 to 22 MPG with lots more power. My friend says the gizmo uses electricity from the alternator to split water molecules into something called "Brown's gas" that gets input into the intake manifold. Is this true or another myth?

— Walt Bruun, Glen Ellyn, Illinois

Cecil replies:

I'll tell you one scientific reaction involving Brown's gas you can take to the bank: it makes my blood boil. Where miraculous fuel economy schemes are concerned, tricksters abound, preying on marks who distrust "the authorities" and can't tell good science from the pseudo kind. Some mutter of brave souls silenced because they knew too much — like the late Stan Meyer, inventor of the magical "water fuel cell" (ultimately shown to be bunk), who fans claim was poisoned in 1998 by operatives of the government and/or the oil companies.

The device you're talking about is similar to Meyer's but places the emphasis on hydrogen, thus piggybacking on the "hydrogen economy" meme President Bush brought to public attention in his 2003 State of the Union address. Newspapers and magazines subsequently devoted acres of unskeptical column space to on-board hydrogen-generation and -injection technology. In 2005, for example, Wired wrote that big-rig truckers were getting major improvements in fuel economy and power from hydrogen electrolysis systems.

Here's what happens. The gizmo is hooked up to a standard internal combustion engine. Like your pal says, it draws power from the car's electrical system to split water into a mixture of hydrogen and oxygen — so-called Brown's gas — which gets fed into the engine and burned along with the usual gasoline/air mix. Alleged result: big gas savings!

But how? On the most basic level, the technology makes no sense. Let's walk through the process:

1. Your car engine burns gasoline or diesel fuel to power the wheels and your alternator (among other things) at about 20 to 25 percent efficiency.

2. Your alternator generates electricity at about 60 percent efficiency.

3. You take said electricity and use it to turn water into hydrogen and oxygen at about 70 percent efficiency, tops.

4. Then you burn the hydrogen and oxygen, or just the hydrogen, in your engine at about 98 percent efficiency.

In short, you're converting fuel A, gasoline, into fuel B, hydrogen, which then helps power the car. Net efficiency of this complicated process: 10 percent. Efficiency of an ordinary car engine (see step 1 above): 20 to 25 percent. Conclusion: Hydrogen gizmos are a fool's bargain.

Advocates claim using hydrogen as a fuel increases combustion efficiency. Problem is, in modern engines combustion efficiency is already close to the max — 95 to 98 percent under optimal conditions in a gasoline engine, 98 percent or better in a diesel engine. Understand, this refers strictly to how thoroughly the fuel burns in the cylinders. Overall engine efficiency is, as seen, much lower, due to heat loss through the engine block and out the tailpipe. Switching fuels won't change that.

So why do hydrogen injector users report improvements? The same reasons people often swear by iffy technology — lack of appropriate comparisons, sloppy record keeping, wishful thinking, a sample size of one. The fact that fuel economy is partly a function of driving habits no doubt also plays a role. If you simply pay more attention to your speed when driving, you can often increase mileage even without a miracle device.

To be sure, a little water can improve internal combustion engine performance under some circumstances. Water injection helped WWII aircraft engines put out more power by reducing knock. BMW has been trying to increase fuel economy and power by using exhaust heat to power what's in effect a small steam engine attached to an internal combustion engine. A Honda hybrid uses a similar concept to turn a generator to recharge the battery packs while cruising. Although some bugs remain to be worked out, a six-stroke engine using water injection for power and cooling shows promise. Bear in mind, though, that water isn't being used as a fuel in any of these cases. If you really want to improve your fuel efficiency, check your tire pressure. Sexy? No. But it does have the advantage that it works.

— Cecil Adams
posted by wfrgms at 1:01 PM on September 16, 2008 [1 favorite has favorites]

This has been explained pretty well already but if you want to see some of the chemistry take a look at Hess' Law: it's the mathsy statement of the rule that, well, you can't get something from nothing. For a water-car as you understand it, you would start at H2O and end at H2O, so by Hess' Law the enthalpy change (the amount of energy you get out) is zero.
posted by katrielalex at 1:23 PM on September 16, 2008

...And of course, then you add in the Second Law of Thermodynamics, and the amount of energy you get out is actually negative. You can't even manage to break even.
posted by Class Goat at 2:31 PM on September 16, 2008

I guess it would be possible to design a vehicle which was a sort of moving water wheel - a great big tank held high up above the roof - then something like a water wheel or a turbine that would drive the wheels. But you would not want to trust it to go very fast - or very far.

It has been mentioned here a few times but "The Hydrogen Hoax" is a good article to read if you believe that a water (or hydrogen) powered car is going to be coming to a show room near you any time soon.
posted by rongorongo at 4:22 PM on September 16, 2008

Water is made up of 1 hydrogen and 2 oxygen, tightly bound together in a pair of chemical bonds. These bond are pretty stable. To break them requires energy. Catalysts don't make it require less energy to break the bond, it just means it takes less time for the reaction to happen, i.e. they get split quicker. Electrolysis is one way to break this bond using electrical energy. That energy you put in goes into the now free atoms, which basically makes them move about quicker. This an endothermic reaction, i.e. one that needs energy from outside to happen.

Now, we take that hydrogen and oxygen, and join them back together. Because they're going from a volatile state to a more stable state, forming that bond releases energy, in the form of heat. This recombination of oxygen and hydrogen together is your 'recombining' which is also burning. This is an exothermic reaction, one which releases energy.

The key thing is - breaking the bond, and remaking the bond requires the exact same amount of energy, under the same temperature and pressure. (and if you're not doing it under the same temperature and pressure, the energy to change the states has to come from someplace else). So. Breaking a bond, then remaking the exact same bond shortly afterwards can never give you more energy back than you started with, it's simply one of the rules of physics that energy cannot be destroyed or created, only moved from one place or one form to another.

Inefficiencies of the process, not least dealing with the changes in temperature mean that some of the energy released by splitting the water is lost to the outside world (heating the engine block, sound waves etc), and of all the electrical energy you put in in the first place, some of that is also wasted in the splitting process, so you get back less energy than you started with.

Now what you can do, is split the two jobs up. One place splits up water into hydrogen and oxygen using cheap electricity from a 'free' source; nuclear power, solar power, wave power, whatever, and throw away the oxygen (well actually they sell it for other purposes, but you get the idea). They then put that portable hydrogen in a tank, put that tank in your car, and then you burn it with atmospheric oxygen, and get back the energy the water splitting plant put in it for you. The key advantage is hydrogen is a lot more portable than a nuclear reactor, so you can drive around with it, even if quite a lot of energy is wasted along the way. As long as your electricity source is plentiful, and you can safely store enough hydrogen, we can still have cars.

This split process is effectively what happens with wood, or petrol. Sunlight was used to power a reaction that made a relatively volatile compound, cellulose. When you burn it or rather combine it with oxygen in an exothermic reaction, you make more stable things (like carbon dioxide) which releases that stored energy. Or, you get a lot of animals that ate that plant material while it was fresh, burnt it with oxygen (which is why we breathe in oxygen and breath out carbon dioxide - we're burning sugar for energy!). They had some left over, and made themselves out of it. Those animals then die, and some of the stored energy in their bodies slowly compresses underground into a very dense sludge of oil. We then turn up millions of years later, dig it up, and finally release that stored energy by burning it with oxygen.

One of the fundamental rules is that you can't get something for nothing. Splitting chemical bonds, then re-forming the exact same chemical bond is energy neutral, you can't ever make an energy 'profit' out of it. What you can make an energy profit on is re-forming chemical bonds that were broken someplace else by somebody else using an entirely different source of cheap energy - usually sunlight.
posted by ArkhanJG at 4:52 PM on September 16, 2008

D'oh, I of course meant 2 hydrogen and 1 oxygen in water! I fail at chemistry.
posted by ArkhanJG at 4:56 PM on September 16, 2008

Ok, I think everyone is missing one other important point. You can make a car run on water, you just have to perform the electrolysis somewhere else. Take seawater, run current, collect gas at cathodes/anodes, bottle O2 and H2, stick em in a car, combine in a controlled fashion, drive pistons! Of course, this would cost energy, but hell, making any kind of fuel costs energy. petroleum has to be heated to crazy temperatures to seperate out the gasoline from the black crap, (it's called cracking) coal has to be dug out of the ground with machines that need gas, even ethanol has to be harvested by tractors, mashed up with machines, and heated to ferment. So making energy portable and converting it to a useful form always costs energy. I think it always sucks carrying around Hydrogen. After all, Hydrogen is nasty stuff in pure form, especially when you are storing it next to oxygen. Combining them? I dont want the frickin hindenburg to happen in my car.
posted by wuzandfuzz at 9:12 PM on September 16, 2008

Just to completely derail this (interesting) discussion and to spread delicious confusion: Adding water injection to car engines actually does improve performance. See Wikipedia on water injection. I was surprised when I heard of this, too. For obvious reasons it has nothing to do with actually "burning" water for energy gains, but more with cooling and compression.
posted by Nightwind at 12:52 PM on September 17, 2008