It is my understanding that solid-core nuclear thermal rocket reactors are extremely aggressive designs that operate at the limits of materials and have an incredibly high specific thermal power.

Do space nuclear reactors meant for generating large amounts of electrical power share the same properties?

  • 1
    $\begingroup$ Do you realize you are comparing completely different technologies? Also, what do you mean by aggressive? $\endgroup$
    – GdD
    Commented Nov 13, 2020 at 11:17
  • 1
    $\begingroup$ In NTP cores temperature is close to 2000C in order to reach Isp 900+s. Most aggressive compound is hydrogen which is known to ebrittle most of metals. In NTP rockets neutron moderators and absorbents are ceramics which are brittle an prone to crack when vibrations are high. Usual type of nuclear devices meant for electrical power generation in space are RTGs. They have a peace of radioactive material, ususally Pu238 dioxide or Sr90, placed in container. Outer wall of container is surrounded by thermocouples to which heat sinks are attached. Its pretty safe technology. $\endgroup$
    – WOW 6EQUJ5
    Commented Nov 13, 2020 at 11:44
  • $\begingroup$ This question is about fission reactors, not RTGs. Yes, I recognize I'm comparing different technologies, that's the point. $\endgroup$
    – ikrase
    Commented Nov 14, 2020 at 0:53
  • $\begingroup$ To heat a propellant the heat must move from the structure to the gas, so the structure is going to be hotter than the gas, in contrast to chemical rockets which actively cool the structure. So to get a useful exhaust temperature the structure has to be really friggin' hot. Otherwise there's no advantage over chemical rockets. A power reactor doesn't throw its working fluid away, it cycles it, so there's no problem running them at lower temperatures. $\endgroup$
    – Greg
    Commented Nov 14, 2020 at 23:11
  • $\begingroup$ @ikrase - so far nuclear reactor used in space employ thermionic conversion from heat to electricity. In principle they are similar RTG, with difference that thermocouples are not directly attached to heat source. Heat is collected with liquid metals (Sodium, NaK, Li-6) and circulated to thermocouples. Exception is experimental NR Kilopower which uses free piston Stirling engine for heat to electricity conversion. $\endgroup$
    – WOW 6EQUJ5
    Commented Nov 16, 2020 at 15:23

2 Answers 2


A NTR (nuclear Thermal Rocket) runs at power levels from about 1.2 to many hundreds of Gigawatt.

With a reactor core mass less than 1/1000th as much as a "normal" contained ground-based nuclear power reactor of similar power.

A NTR operates at a core temperature of about 3000K
A power reactor operates at a core temperature of about 300c (570K-isk )

Yes, NTR is "Agressive" seen from this viewpoint.

Compared to space nuclear fission power systems: They simply DO NOT scale up very well without the shielding and cooling abilities of ground systems. Unless you incorporate the shielding and containment, and replace your external gravity-assisted water cooling system with a self-contained closed loop radiative cooling system. In which case the space reactor will be MORE massive than the ground power station.

At present tech, the absolute pinnacle of electric power nuclear reactor suitable for use in space is the KiloPower units. Which, as the name suggests, provide power in the single-digit kilowatt range. And it is not quite ready yet, TRL 5-6 or so. (full scale, lab tests, of prototypes, in non-field-conditions)

At 1500kg each for the 1Kw unit, the mass per power is somewhat...disappointing. A cluster matching the power of the smallest NERVA would mass 1700 metric tons.

  • $\begingroup$ I'm not interested in comparing to non-mobile power reactors. $\endgroup$
    – ikrase
    Commented Nov 14, 2020 at 0:54
  • $\begingroup$ 1.21 gigawatts? Great scots, Marty! $\endgroup$
    – DrSheldon
    Commented Nov 14, 2020 at 3:14
  • $\begingroup$ @DrSheldon 1137 MW in the case of the NERVA prototype. maybe that's why they abandoned it, it didn't quite reach wan point twenny wan jiggerwhats? $\endgroup$
    – user38044
    Commented Nov 14, 2020 at 6:30
  • $\begingroup$ @ikrase added info about KiloPower systems, the best we have for space-rated reactors. $\endgroup$
    – user38044
    Commented Nov 14, 2020 at 6:46

I will argue that these reactor types are not directly comparable.

While a high "alpha" (power/mass) is desirable for both NTRs and reactors for electric power, the limiting factor that causes the most headache for NTR materials is reactor temperature, with Isp proportional to the square root of temperature (not quite proportional, due to hydrogen disassociation).

The Isp in turns ends up having an exponential impact on vehicle mass due to the tyranny of the rocket equation.

In a reactor generating electricity on the other hand, temperature has no impact on Isp when used for electric propulsion.

Temperature does have an impact on thermodynamic efficiency though, as a heat engine with a higher temperature gradient is more efficient (with diminishing returns). This is furthermore made complicated by heat engines in space being run "hot" on the cold end to cut down radiator mass.

So while system mass is a shared goal, it's calculated differently. For a NTR the mass is mostly engine + propellant, while for an electric system it's reactor + heat engine + radiators.
And while peak temperature is a shared goald, it's governed by different physical laws, gas velocity and heat engine efficiency.
Furthermore, a NTR provides high impulse over minutes, while electric systems provides a small impulse over months. Not even the mission profiles are comparable.

It's apples to oranges, so "aggressive" does not carry much meaning.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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