Hear me out before you downvote; It's not as simple as it seems.

On Earth, nuclear reactions produce heat energy. This is used to heat water, creating steam, powering turbines connected to generators - and voila, electricity.

If you attempt to use thermal energy to produce steam, it will not "rise" because buoyancy does not apply; furthermore, liquid water will mix with vapor in messy circumstances that make pressure systems more complex to construct. Therefore, conventional reactors cannot operate to produce electricity the conventional way, in space.

Additionally, as jamesqf mentioned in a comment,
Heat engines (steam turbines) require temperature differences, which, unlike on Earth, cannot easily be catered to in space. Enormous radiators disposing of excess heat are necessary to make turbines run effectively.

So how would a nuclear-powered spaceship, without artificial gravity, generate electricity effectively from raw thermal energy and the radiation associated with nuclear reactions?

  • $\begingroup$ Energy is released when atom is split, this energy is absorbed by the water molecules inside most nuclear power plant. Imagine this energy is being absorbed by a certain material which causes it to donate it's valence electrons hence producing a current flow analogy to solar panel. $\endgroup$ – user6760 Nov 23 '16 at 6:59
  • $\begingroup$ @user6760 This is science-based - an unobtanium that turns radiation directly into electricity does not help me. - Unless you have a real world example to turn into an answer. $\endgroup$ – Zxyrra Nov 23 '16 at 7:08
  • $\begingroup$ Yes see nanomaterial but must include NASA in your suffix/prefix. $\endgroup$ – user6760 Nov 23 '16 at 7:12
  • $\begingroup$ newscientist.com/article/… $\endgroup$ – user6760 Nov 23 '16 at 7:21
  • 1
    $\begingroup$ Considering humanity has already used nuclear reactors in space ... en.wikipedia.org/wiki/SNAP-10A and en.wikipedia.org/wiki/BES-5 ...your question is kind of moot. But as others have pointed out: radiators... that is how you get the temperature difference you need to use a heat engine. $\endgroup$ – MichaelK Nov 23 '16 at 11:12

Nuclear reactors can work just fine with other coolants, and really high efficiency nuclear reactors here on Earth are designed to use Helium in the primary coolant loop in order to support running the core at a much higher temperature than water cooled reactors generally run at. Other work arounds have included using Sodium (very dangerous because Sodium will ignite in the presence of water, so a leak could become a flaming radioactive nightmare) and metals like lead (the opposite problem has occurred, apparently some Soviet era submarines using high powered, lead cooled reactors have been decommissioned because the operating temperature was allowed to drop too much, resulting in the coolant solidifying in the primary loop).

The biggest reason to avoid "steam" powered generators isn't technical, but rather the laws of physics. Steam generators are considered to be "Rankin" cycle machines, and like most heat engines have a pretty hard upper limit to the amount of energy that can be extracted, known as Carnot's theorem (thermodynamics)

Carnot's theorem states:

All heat engines between two heat reservoirs are less efficient than a Carnot heat engine operating between the same reservoirs.

Every Carnot heat engine between a pair of heat reservoirs is equally efficient, regardless of the working substance employed or the operation details.

The formula for this maximum efficiency is:

efficiency = 1 − TL/TH

where TC is the absolute temperature of the cold reservoir, TH is the absolute temperature of the hot reservoir, and the efficiency is the ratio of the work done by the engine to the heat drawn out of the hot reservoir.

For steam engines (which is what we are talking about), 33% is a high efficiency without reheat or other additional steps (combined cycle systems get much higher efficiency because they use the energy in the fuel several times, i.e. a gas turbine generator which uses the exhaust to heat the steam).

One thing which might work well in space is to use an MHD generator, where the heat of the reactor ionizes a coolant that is them passed through a magnetic field. MHD is not constrained by the Carnot limit. The ultimate expression of that is a "fission fragment" reactor, where the fissile material is introduced into a magnetic chamber in the form of fine dust. The resulting fissioning of the material is captured in the high energy movement of the fission particles (up to .03 c, but in practice usually .01 c due to internal collisions), making a high energy stream of charged particles to be tapped for electrical energy or to be used as a rocket engine. (See this as well)

MHD and fission fragment reactors also require less "plumbing" and usually smaller radiators for the amount of reactor power, which are all advantageous when designing a spaceship or space colony.

  • $\begingroup$ MHD object of same rule, just hot end temperature may be way much higher then usual, but in practice MHD temperature is not way much higher but 90% is possible(kinda) with convention materials. $\endgroup$ – MolbOrg Nov 24 '16 at 4:22

You're starting from a false assumption. Steam turbines don't use the buoyancy of steam to generate power, they use its pressure. This works just fine in space -- separating the steam from the water might be a bit tricky, but that's just an engineering detail.

Alternatively, if you don't want to deal with moving parts, you can use a thermoelectric generator to turn heat directly into electricity. This isn't as efficient as a turbine, but it's also less likely to break down.

  • 8
    $\begingroup$ Actually, the real problem with steam turbines in space is getting rid of the heat. From basic thermodynamics, you need a temperature difference to run any heat engine. On Earth, your cold side is usually a body of water, or the air (as with your car's radiator). In space, you can only use thermal radiation to get rid of heat, which is why the ISS has those big radiators for just the waste heat from equipment and human metabolism: nasa.gov/mission_pages/station/structure/elements/… $\endgroup$ – jamesqf Nov 23 '16 at 5:14
  • $\begingroup$ Edited question to not base my argument off buoyancy, thanks for pointing that out $\endgroup$ – Zxyrra Nov 23 '16 at 8:05
  • $\begingroup$ water might be a bit tricky - its easy, same way as cyclone dust separators work. But easiest way just prevent it to be there, by heating enough. $\endgroup$ – MolbOrg Nov 24 '16 at 4:06

While nuclear reactors do not require steam to rise, making sure only steam (and no water) gets to the turbine might be a bit difficult in microgravity.

1) Use what we are using now - RTGs. No moving parts = no problem with gravity (or lack of thereof)

2) Generate "artificial gravity" using a centrifuge. (Since centrifuges aren't science fiction, I'm guessing this is not what was meant by "no artificial gravity".) It should be sufficient to place the heat exchanger into the centrifuge - neither the high-pressure primary circuit nor the turbines themselves need gravity, which avoids having to spin the reactor core.


I’ve read that stirling Engine designs are being looked into for this purpose, as being much more efficient and higher power than RTGs. That’s probably what you are looking for. You might find some info in real designs now that you know, but off the of of my head I'm guessing that you can use a design that doesn’t have a phase change like steam (“single-phase working fluid”), and uses active circulation so it doesn’t need convection currents.

  • $\begingroup$ You do trade this higher efficiency against moving parts tho, which is probably why RTGs are the go-to standard $\endgroup$ – Hobbamok Apr 1 '19 at 10:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.