I was looking at this answer that talks about how a good alternative to Radioisotope Thermoelectric Generators would be fission reactors. It makes a number of good points, but in the comment thread there was a discussion of the particular issue that in both cases if there is a worst-case-scenario accident and the launcher burns up in the atmosphere, radioactive material is widely dispersed. From Final Environmental Impact Statement for the Mars Science Laboratory Mission:

The predicted mean radiological dose to the maximally exposed individual from an unlikely launch area accident would be about 0.1 rem. No short-term radiological effects would be expected from such an exposure. Each exposure would, however, yield an increase in the statistical likelihood of a latent cancer fatality over the long term... Less likely launch area accidents include explosion on the pad, situations where the spacecraft is detached from the launch vehicle, and accidents where the vehicle safety systems are assumed to fail. The probabilities of these types of accidents range from approximately 1 in 8,000 to 1 in 800,000, and could result in higher mean releases of plutonium dioxide (up to 2 percent of the MMRTG inventory) with the corresponding potential for higher consequences. The maximally exposed individual could receive a dose ranging from a fraction of one rem up to about 30 rem following the more severe types of very unlikely accidents, such as ground impact of the entire launch vehicle.It should be noted that there are large variations and uncertainties in the prediction of close-in dose modeling for such complicated accident situations. Assuming no mitigation actions, such as sheltering and exclusion of people from contaminated land areas, radiation doses to the potentially exposed members of the population from a very unlikely launch accident could result in up to 60 mean additional cancer fatalities over the long term. Results of the risk assessment also indicate that for the very unlikely accident that involves ground impact of the entire launch vehicle, roughly 90 square kilometers (about 35 square miles) of land area could be contaminated above the 0.2 μCi/m 2 screening level.

I asked about this as it wasn't clear to me how the two systems compare, RTGs releasing plutonium and micro fission reactors releasing uranium. The response was that the fresh U-235 fuel in a launched fission reactor is 'a trivial biological risk compared to either RTGs or reactors that have been operating and producing fission products'.

How exactly does it compare quantitatively? I accept the statement but it would be nice to know the details, as plutonium production for RTGs has started up again, while fission reactor development languishes.

  • $\begingroup$ By way of comparison, after the failure of a Transit satellite to achieve orbit on 21 April 1964, about 1 kg Pu-238 of its SNAP-9A RTG were distributed mainly in the southern hemisphere. During the re-entry of the Kosmos 954 satellite on 24 January 1978, about 20 kg of highly enriched uranium from its nuclear reactor and the fission products from its operation were released. $\endgroup$
    – user10840
    Jun 7, 2016 at 21:22
  • 1
    $\begingroup$ @Loong The key term there is fission products. Fresh fuel is vastly less dangerous than RTG material, and that that is vastly less dangerous than a fission reactor can be after operating for some time. Fission produces a nasty soup of all kinds of radioactive materials, but fresh fuel hasn't built up those materials. $\endgroup$
    – AlanSE
    Jun 8, 2016 at 2:20

1 Answer 1


The probability and consequences of a release of Pu-238 from an RTG in a launch accident are very low, due to the protections in place for such an incident. It's not like they never thought of that. The radioactive material is not "widely dispersed".

As for the numbers, the rate of decay is inversely proportional to the half-life. The half-life of U-235 is 700 million years, and that of the remainder of the fuel, U-238, is 4.6 billion years. The half-life of Pu-238, RTG fuel, is 87 years. So Pu-238 is about eight million times as radioactive as unused fission fuel.

You don't really need to worry about unused fission fuel. The usual ceramic form of the fuel could be used for a dinner plate. A really heavy dinner plate. Even the metal toxicity is not a problem in that case, since the fuel is in its oxide form. Though be careful when stacking a lot of those plates together!

(By the way, I have some uranium kitchen ware -- not fission fuel and not consisting only of uranium oxide, but rather glass tinted green with uranium.)

The "unused" is important. Once you start up a fission reactor, the fission products are highly radioactive. So don't ever start up a space fission reactor until it's well out in space and not coming back.

  • $\begingroup$ I added a quote from the study you linked to for context. $\endgroup$
    – kim holder
    Jun 7, 2016 at 17:51
  • 3
    $\begingroup$ "Though be careful when stacking a lot of those plates together!" made me laugh out loud. +1! $\endgroup$
    – ceejayoz
    Jun 8, 2016 at 0:23

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.