Is it plausible to:

  1. Have a huge nuclear powered aircraft carrier to be propelled by H2-O2 rocket engine (size adjust) where you generate both reactants from sea water through Electrolysis. What would it take to make it work?

  2. Sucks in air from atmosphere, ionise it and propel with ion engine Is there any limit to how much thrust it can generate, given the system can generate and handle any amount of current or voltage, and suck in as much as air you want to ionise

  • 2
    $\begingroup$ Welcome SS. This seems to be two questions, could you edit it down to one, we only do one per thread here. You can hyperlink a new question to this if you like. Please also take our tour and refer to the help center for guidance. $\endgroup$ May 18, 2022 at 2:15
  • 2
    $\begingroup$ For the first question... what are you trying to accomplish? A rocket engine works by converting chemical energy to heat energy, using that heat energy to accelerate a working fluid, and having the momentum of the working fluid perform work on the vehicle. A nuclear reactor converts nuclear energy to thermal energy, and then a turbine converts thermal energy to electric energy. Hydrolysis converts electrical energy to chemical energy. So you're going nuclear->thermal->electrical->chemical->thermal->kinetic... why not just use the nuclear pile's heat on the working fluid directly? $\endgroup$
    – addaon
    May 18, 2022 at 3:13
  • $\begingroup$ For the second question... given infinite energy (!), you're limited by the amount of working fluid you can access. The analysis here is a bit different in the subsonic and the supersonic cases, but to simplify, you don't "suck" air, you produce an area of relatively lower pressure by removing air and have surrounding air fill in the that volume, at a finite rate. Given actually infinite energy you can build a photonic drive without a working fluid, but if you want a working fluid there are limits. $\endgroup$
    – addaon
    May 18, 2022 at 3:15

4 Answers 4


I think this is best answered with the question : "Why would you want to do that?"

A traditional nuclear reactor typically works by

  1. Getting very hot through nuclear reactions
  2. Using that energy to heat water up and convert it to high pressure steam
  3. Using the steam to drive a turbine...
  4. ...which drives a generator to produce electricity

so your first option is to add the steps

  1. Use the electricity for electrolysis to convert H2Os to H2 and O2
  2. Pump these gases into a reaction chamber and burn them...
  3. ...to create thrust.

Far simpler (and more thermodynamically efficient), if you did desire some form of rocket drive, would be to just use steps 1 and 2 and eject the steam directly through a suitable orifice as your propellant.

But there is a second issue too. Presumably you want airplanes to be able to take-off and land from your aircraft carrier. Typically they take-off to the front (to maximise head-wind and effective iniital velocity into the air), and land from the back (to minimise relative velocity of airplane and landing deck). I douubt that even a typically gung-ho pilot would want to head into the superheated jet blast emitted by your rocket engines when attempting to land.

  • $\begingroup$ Firstly, thanks for reply.as to “Why?” , because why not? It’s just a fun little thought experiment? $\endgroup$ May 18, 2022 at 13:05
  • $\begingroup$ I thought Heating steam enough to generate significant thrust would require a lot of heat and quite a high temperature. And getting that heat directly from reactor would increase risk of meltdown, as opposed to separate specialized rocket thruster. Also, this propulsion would only be used in addition to fan propeller, when needing to position yourself somewhere far away asap $\endgroup$ May 18, 2022 at 13:14
  • 3
    $\begingroup$ @SparkShredder you are probably right that creating enough heat from the reactor directly would be problematical - so why do you think jumping through some additional hoops would make the situation any better? $\endgroup$
    – ths
    May 18, 2022 at 14:30
  • $\begingroup$ You could use multiple separate reactors to pool together enough electrical power to make it all work safely. $\endgroup$ May 18, 2022 at 14:36
  • $\begingroup$ @SparkShredder how would that help at all? You're just rearranging the fissile material and adding a bunch more shielding and more plumbing to get steam to turbines. You still need the same amount of electrical power, which means those multiple reactors have to produce the same amount of thermal power to drive the turbine. If anything, thermal losses will be even worse. $\endgroup$ May 21, 2022 at 12:34

The Good, the Bad, and the Ugly...

Is it plausible? Sure. That's the good part. But there are some problems...

  1. Rockets are a great way to move forward — but they're a nearly useless method of turning when you're turning against something with significant mass... like water. Which means you still have a rudder.

  2. Rockets generate a ton of heat. It's bad enough that the heat would light your aircraft carrier up via thermal detection like the proverbial Christmas tree, but you're dumping that heat into the ocean, cooking every fish that swims near the ship. Those Greenpeace people will be haunting your ship from the moment it sets sail.

  3. You get more than hydrogen and oxygen from sea water. You get salt... and fish... and dirt... The byproducts of electrolysis would be mountains of horrible waste adding mass to your ship until you dump it.

  4. And we won't mention the fact that aircraft carriers sail with a small armada of ships that serve a variety of functions — mostly defensive — so they'd need rockets, too. It would look cool on a Saturday-morning cartoon, but each and every one of those ships would be a very real missile, as much a danger to each other as to the enemy.

So, rockets, when conveniently fueled and properly balanced against the mass of the object being moved vs. the friction of the medium said object is moving through, would believably increase speed in your case, but they're either increasing speed a lot or simply burdening the ship with additional complexity without a commensurate increase in value. That's a long-winded way of saying, while it's plausible, it's not particularly believable.

And then there's the ugly part...

Aircraft carriers are honking big mother-hubbards and about a third of an aircraft carrier's height is underwater. That part of the ship has a tremendous impact on its ability to move. We learn from Thomas Foster, former Navy F-4/F-14 Radar Intercept Officer (RIO) (1971-1995) the following:

...at speed, the ship sets up a “shock wave” that bounces off the sea floor and masks rudder effectiveness. (Source)

In other words, rockets really aren't a benefit to an aircraft carrier because the faster they go the deeper the ocean needs to be or they simply compromise their own ability to turn, leaving them at the mercy of an unpredictable ocean.


A space shuttle main engine (not the solid booster rocket) output about 375,000 pounds of thrust at sea level. They used liquid oxygen and hydrogen. They burned about 1.6 million pounds of fuel in about 8.5 min.

The displacement of a Nimitz class aircraft carrier is about 97,000 tons or 194000000 pounds.

1 pound of thrust accelerates 1 pound of mass at 32 feet per second (1 g).

So if you had the equivalent of the space shuttle main engine the acceleration would be about 0.002 g.

So I don’t think you can produce enough oxygen and hydrogen to get your aircraft carrier to move very fast.

Plus it would have a tremendous heat and sound signature.

  • $\begingroup$ Plus, I would imagine there would also be structural concerns. $\endgroup$
    – jdunlop
    May 18, 2022 at 20:23
  • $\begingroup$ You are mixing and matching problems badly here. One rocket engine being too weak is not the same as not being able to produce enough fuel. $\endgroup$ May 18, 2022 at 21:58
  • $\begingroup$ If you can’t supply enough fuel for one rocket you can’t provide enough for more rockets or the application. You can’t get enough momentum transfer. $\endgroup$
    – UVphoton
    May 19, 2022 at 5:43

Sorry for the down vote mate, but maybe there is another question you could better ask : Would a High Pressure jet of steam be a good method of propulsion in a nuclear powered aircraft carrier.

Let me first talk about what is not going to work: A jet engine develops a large amount of power that is passed to a compressor that uses most of it to compress the air needed by the engine .. only the remainder is ejected out the back end as thrust. It works on aircraft because the weight is low and the air resistance small, compared to the situation of a ship in water

Putting a jet engine on a ship means it is working against the much larger resistance of the water to move the much larger mass of the carrier. There are valid engineering reasons for using heat from a reactor to generate steam to spin turbines that then drive the propellors of the ship. Propellors are more efficient in water than any other method of propulsion.

All that said .. it is not beyond an author's imagination to think that the heat of a reactor could be used to heat seawater to make superheated dry steam that could then be ejected directly to create thrust that would propel the ship. Think of an oversized water-jet boat, but using steam.

In Real Life jet-boats work because the engine is powerful, the water-jet is high velocity, and the boat is light, Really Light!

In a story disbelief could reasonably be suspended for a hull constructed for minimal weight and water resistance, with a number of (possibly steerable) high volume steam jets. It would be best to vent the jets into the water to avoid creating a zone of Death by Scalding around the ship at low speeds if they were to be above the water line.

Your number 2 item .. this from Wikipedia "An ion thruster ionizes a neutral gas by extracting some electrons out of atoms ..." Where are you going to get the gas to ionise? Generating thrust from an Ion Engine won't work in the atmosphere much less in water.


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