Jet engines can run on almost any fuel, and the operating temperatures of modern jet engines' hottest sections are anywhere between 3000 and 3150 degrees F (1648 and 1732 degrees Celsius). Does that mean that a hydrogen on-demand system could work on modern jets?

Water is pumped and heated first by the exhaust section, then directed towards the hotter sections of the engine (when hot enough to not cause cooling and lower engine efficiency) where it's broken down into hydrogen and oxygen at a heat above 1472 degrees F (800 degrees Celsius), then those gases are pumped into the engine for combustion.

The advantages are that firstly, water is abundant and therefore cheap. Even sea water could be used because at those temperatures it's easy to design a system that would get rid of the impurities that would otherwise corrode critical engine parts.

Secondly, it would save on manufacturing costs given that non-heat critical parts in the exhaust section would not need to be made of sophisticated and expensive materials and alloys given the cooling effect of water.

Thirdly, the costs of the fuel weight would be reduced given that the energy density of hydrogen is twice that of fossil fuels, so less would need to be carried. And most importantly, the environment problem would be solved in aviation given that there would be little or no carbon dioxide emissions.

  • 38
    $\begingroup$ There is a very strong smell of "free energy" (or perpetual motion) scheme from your post. The energy required to break the atomic bonds will always be greater than the energy released on recombination due to losses in the cycle. This is basic thermodynamics. $\endgroup$
    – Transistor
    Sep 17, 2017 at 17:11
  • 13
    $\begingroup$ This question is about chemistry, not aviation. Let the chemists explain at length why you can't get energy from burning water. If this sort of scheme could work, the primary beneficiaries of it would surely be electrical power plants, not aircraft. But it can't work, it's "free energy". Except that water is never fuel, it's ash from burning hydrogen (which by definition happens in the presence of oxygen). VTC. $\endgroup$
    – Ralph J
    Sep 17, 2017 at 17:34
  • 2
    $\begingroup$ In other words: the amount you'd cool (and thus contract) the hot air, while breaking down the water, will be more than you can heat (and thus expand) it by burning the resulting hydrogen. $\endgroup$
    – yshavit
    Sep 17, 2017 at 19:18
  • 2
    $\begingroup$ This should be asked over on physics.stackexchange.com (or maybe chemistry.stackexchange.com). Not that the answer would be that different. $\endgroup$
    – Jan Hudec
    Sep 17, 2017 at 21:34
  • 7
    $\begingroup$ Um...water is not a fuel. It used to be a fuel (hydrogen) that has burned already. $\endgroup$
    – Koyovis
    Sep 17, 2017 at 22:12

5 Answers 5


You're proposing to use the heat of the engine to break down water into hydrogen and oxygen, and then burn the hydrogen in the oxygen, turning it back into water. And, remember, when I said "the heat of the engine", I mean the heat you obtain by... burning the last lot of hydrogen and oxygen you made.

Even if this process was 100% efficient, you'd just be converting back and forth between water and hydrogen/oxygen. No energy would be created, so there'd be no thrust. And it's not 100% efficient: you'd lose energy, rather than gain it.

  • 1
    $\begingroup$ It also requires having something on hand to start the engines with and get them hot enough to crack the water in the first place, thus needing fossil fuels or a hydrogen storage system (with the associated weight penalties). $\endgroup$
    – FreeMan
    Sep 18, 2017 at 15:06
  • $\begingroup$ @FreeMan Yes, though that "just" requires equipment at airports, as long as you don't want to restart the engines in mid-air. $\endgroup$ Sep 18, 2017 at 15:10
  • 2
    $\begingroup$ @FreeMan, that is pointless. If there was energy to be gained, anything else would be minor technical detail. But there isn't, so the minor technical details are not relevant. $\endgroup$
    – Jan Hudec
    Sep 18, 2017 at 20:32
  • $\begingroup$ If you could add the reason it is not and will never be 100% efficient (the Second Principle of Thermodynamics, nothing to do with insufficient technology like some perpetuum mobile inventors like to postulate) this would be an even more more complete and still succint answer. $\endgroup$ Sep 19, 2017 at 7:45

No. Water is not a fuel, it is hydrogen that has burned already. It is a very stable oxidation product, so stable that it is used to extinguish fires. Commercial production of hydrogen is not done by heating water. From the wiki:

There are four main sources for the commercial production of hydrogen: natural gas, oil, coal, and electrolysis; which account for 48%, 30% 18% and 4% of the world’s hydrogen production respectively.[5] Fossil fuels are the dominant source of industrial hydrogen.

There is quite some interest in R&D projects to produce hydrogen from water, because like the OP states water is very abundant. Many production methods are researched: electrolysis, radiation, chemical reaction. Heat is one of the least promising techniques, since temperatures of up to 3000 °C are required. From this wiki:

Thermal water splitting has been investigated for hydrogen production since the 1960s.[16] The high temperatures needed to obtain substantial amounts of hydrogen impose severe requirements on the materials used in any thermal water splitting device. For industrial or commercial application, the material constraints have limited the success of applications for hydrogen production from direct thermal water splitting and with few exceptions most recent developments are in the area of the catalysis and thermochemical cycles.

And that is in a production facility in a research lab. Now we're going to do this in an aircraft, where safety, weight and reliability are of prime importance. Water has been used for propulsion before (by injecting it into the hot exhaust gas of a turbojet) but that went out of the door: it does very little for high-bypass engines, not enough to justify the extra weight.

But most inventions start with an outlandish idea which is dismissed by the grey eminences at first. The OP has two seedlings that may have merit:

  • Use waste heat that is produced anyway to generate useful power or thrust. This is done in power stations already: in remote communities in Australia one can find gas turbines driving alternators, next to a steam turbine that uses the waste heat from the gas turbine. But this is not viable for aircraft, too heavy and no redundancy since the steam turbine cannot work without the gas turbine.
  • Use water that is on board anyway for propulsion. Waste water from the toilets! Inject that into the hot exhaust gases for an environmentally friendly solution.

This answer shows that:

The top gas temperature in a modern [civilian] jet turbine is more like 1500°C, and the turbine blades tolerate temperatures of around 1200°C.

On the other hand, thermal decomposition of water at 2200°C only splits 3% of the water. 50% for 3000°C, which remains the toughest challenge because of the materials required.

Liquid hydrogen however, is more promising.

Temperatures [and apparatuses] aside, to fix the mentioned perpetual motion problem (it's either splitting or thrust, not both), you'll need a jet fuel gas turbine for the splitting, the un-split water to be recycled, and then separate hydrogen-burning jet engines for propulsion.

This is exactly like having an airplane carry crude oil and an oil refinery to produce its own fuel by burning fuel. The only difference is that breaking down hydrocarbons is much easier than breaking down water.

  • 4
    $\begingroup$ @ymb1: The temperature given in the question is correct for the top gas temperature in military engines which run hotter than civilian engines. But the question is targeting a perpetuum mobile anyway, so in the end it does not matter. $\endgroup$ Sep 17, 2017 at 17:56
  • $\begingroup$ Thermal splitting can be done at lower temperature (e.g. below 1,000°C) by using catalysts. $\endgroup$
    – mins
    Sep 18, 2017 at 18:47
  • $\begingroup$ @ymb1, yes. This is much better comparison. $\endgroup$
    – Jan Hudec
    Sep 18, 2017 at 20:43

@DavidRicherby has demonstrated how breaking down water, and then burning hydrogen back to recreate water is a process with losses that cannot generate a positive budget of energy. But let's go further, as if this was not the case, and see which quantity of water is involved.

(All figures are approximated to make them simple to compute mentally.)

Let's evaluate the quantity of water you would need to replace fuel, based on the energy that can be extracted from hydrogen and from fuel. We need to compare the specific energy of both.

For a familiar reference: 1kWh = 3.6 MJ, this is what would consume a steam iron in an hour.

To get 1 kg of hydrogen from water, you need to break down, in the most optimistic case, about 10 kg of water (atomic mass of $H_2$ = 2 , of $O$ = 16, of $H_2O$ = 18).

  • If hydrogen and oxygen could be separated, by some mean, 1 kg of water would produce 100 g of hydrogen. Hydrogen has a specific energy of 120 MJ/kg, that is 12 MJ/100 g.

  • But the specific energy of fuel is much greater: 42 MJ/kg.

Assuming you have the water breakdown reaction for free, which is indeed not the case, you still need 42/12 = 3.5 kg of water to replace 1 kg of fuel.

enter image description here

Not really a good deal... the aircraft needs tanks about 3 times larger and needs a whole redesign to increase the maximum takeoff mass (which in turn will require a lot more hydrogen).

That's without counting fuel which, in real life, is needed to breakdown water and extract hydrogen. With this fuel you reach insane masses.

However... There are many processes for water splitting, some are really not very efficient... but if it's free...

The energy required to extract hydrogen could be solar energy, which you get for free. There is a process known as thermochemical cracking which use cells with catalyst materials exposed to solar rays.
This process is extremely slow and not very efficient (7%), but it works!
While this is not usable for aircraft propulsion, it's possible to use it for large scale solar energy storage, converting solar energy into hydrogen chemical energy which can be used later.
See Almeria Spanish solar power plant.

  • 2
    $\begingroup$ No, the energy density of water is plain and simple ZERO (as long as we stay in the realm of oxygen chemistry). Because to get a 1 kg of hydrogen from water, you have to provide all of those 120 MJ. $\endgroup$
    – Jan Hudec
    Sep 17, 2017 at 20:21
  • $\begingroup$ “assuming you have the water breakdown reaction for free” it is still whatever runs the reaction for you and not the water that has the energy. $\endgroup$
    – Jan Hudec
    Sep 17, 2017 at 20:32
  • $\begingroup$ There is absolutely no way “to be said” that you get any energy after breaking down water for the simple reason that you are starting with water and ending with water. Each substance has chemical energy and this energy depends only on the substance and not on how it was made. During reaction—any reaction—the difference in chemical energy between reagents and products is released (absorbed if negative). Since you are starting with water and ending with water, nothing can be released. The “free” reaction makes no sense and renders any other conclusion pointless. $\endgroup$
    – Jan Hudec
    Sep 17, 2017 at 21:12
  • 3
    $\begingroup$ @JanHudec: Yes, what you say is correct. Perhaps you could write an answer on the loss occurring when breaking down and recombining water components. But that's not the angle I've chosen, I'm on the mass of water vs. the mass of fuel. $\endgroup$
    – mins
    Sep 17, 2017 at 22:39
  • 3
    $\begingroup$ @Koyovis: There is not a large amount of waste heat. The process is not self-sustaining. It's not even self-starting! If you added a battery, for example, to start the process it would quickly shut down. The waste heat would be supplied by the battery. $\endgroup$
    – Transistor
    Sep 18, 2017 at 14:56

There are two ways of get this working, but both seem only very remotely possible.

  1. Venus planet, or any other with the surrounding temperature above the water boiling temperature. In such environment, liquid water would have energy: while boiling, it can produce compressed steam able to rotate a turbine. A compressor, similar to the compressor of the jet engine, could help getting more hot air into the system to boil the water. Same way liquid nitrogen may work on Earth.

  2. There is plenty of nuclear energy in the water, it contains hydrogen that powers the Sun by fusing into helium. A machine the could use such an energy is currently being built while I think initially it will power more something like a large ship.

You cannot decompose the water into oxygen and hydrogen and then burn these components because exactly the amount of energy required to decompose is released during the later burning process. Due inevitable losses, the machine eventually will stop after exhausting the initial energy pool.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .