@TomSpliker's great answer mentions that ESA might be looking at producing Radioisotope (powered) Thermoelectric Generators (RTGs) based on the radioisotope americium-241. 241Am is a "kinder and gentler" isotope than 238Pu to access since it's sourced from reprocessing nuclear waste from power generation and doesn't have to be "made to order" in a special reactor.

But it decays five times more slowly, with a half life of ~432 years versus ~88 years for 238Pu, and the alpha decay is accompanied by some emission of X-rays. The additional radioactivity of the decay products which increases over time for the two are different as well.

Suppose I had a spacecraft or rover that needed 100W of electrical power, and due to uncertainties I needed to design the mission to allow for either 238Pu or 241Am based RTGs, I'd have to compare all aspects, including mass, additional radiation of the decay products, mission life (say EOM is 80% of initial power) and size.

How would the two then compare?


1 Answer 1


Power and Mass

From this paper (emphasis mine):

The specific power of an 241Am-fuelled RTG cannot match that of a 238Pu system (except perhaps at small power output levels); however, the design work undertaken provides confidence in potential capability and performance of 241Am systems for future space missions. Medium-sized RTGs in the 10 W to 50 W range are predicted to achieve an overall specific electrical power output of around ~1.5 W/kg.

For comparison, the GPHS-RTG Plutonium-238-based RTG used on various deep-space mssions (Cassini, Ulysses, New Horizons) is capable of producing some ~5W/kg

These figures include the mass of radiator fins and other structural parts, as well as the mass of the isotope itself.

The primary decay mode of Am-241 is α-decay, which makes it well suited for use in RTGs as due to its low penetration and high energy-per-decay.

Also worth noting - RTGs typically have incredibly low efficiencies converting thermal energy to electrical - around 5%. This means that small changes in efficiency can have a large impact on the mass/power ratio of the device.


Using the equation for half-life:

$$N(t) = N_{0}(\frac{1}{2})^{\frac{t}{t_{\frac{1}{2}}}}$$

We can show that Americium-241 will still produce 80% of initial power after ~140 years - plenty enough for any mission plans we currently have. Compare this to 80% at ~29 years for Pu-238 with is shorter than some ongoing missions.

The primary decay pathway is to Neptunium-237 on the Neptunium Series. Np-237 also has a primary α-decay mode, however with a half-life of ~2.14 million years, its contribution to the energy output is minimal.


With regards to secondary radiation, Americium-241 has a very weak gamma byproduct. In fact, it is considered to be safe enough to be used in many household smoke detectors, so I believe there wouldn't be many concerns from secondary radiation products.

  • $\begingroup$ Are the decay products from 241Am alpha decay also radioisotopes? After 20 or 30 years, would there be any additional radiation due to their activity as well? $\endgroup$
    – uhoh
    May 24, 2018 at 9:30
  • 3
    $\begingroup$ Am-241 decays via the Neptunium series, however the primary product is Np-237 with a half-life of some 2 million years, therefore minimal byproduct activity $\endgroup$
    – Jack
    May 24, 2018 at 9:39
  • $\begingroup$ Great! That's what I meant by "secondary radioactivity of the decay products", why not add that to your answer, and I'll fix the wording in the question. Excellent answer by the way! $\endgroup$
    – uhoh
    May 24, 2018 at 9:41
  • $\begingroup$ "These figures include the mass of radiator fins and other structural parts, as well as the mass of the isotope itself." What about the mass of thermocouples, is it included too? $\endgroup$
    – Uwe
    May 24, 2018 at 10:01
  • 1
    $\begingroup$ @Uwe The linked paper discusses Thermoelectric Generators as part of the design and the Pu figure is from this table. So I believe yes on both counts! $\endgroup$
    – Jack
    May 24, 2018 at 10:10

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.