|
|
| Fig. 1: NASA's Multi-Mission Radioisotope Thermoelectric Generator (MMRTG), the radioisotope power system used on the Mars 2020 Perseverance rover. The MMRTG produces about 110 W of electric power at the start of mission from about 2000 W of thermal power. [4] (Courtesy of NASA) |
NASA's Artemis program is intended to return astronauts to the Moon and establish a longer-term human presence there, especially near the lunar south pole. NASA has emphasized the south polar region because it offers unusual lighting conditions, possible access to water ice in permanently shadowed regions, and a path toward sustained exploration rather than brief Apollo-style visits. At the same time, the south pole is not continuously illuminated. Polar lighting is irregular and terrain-dependent, while equatorial sites experience the familiar cycle of about 14.5 Earth days of sunlight followed by about 14.5 Earth days of darkness. Any lunar mission that must survive darkness, extreme cold, and long communication or science campaigns therefore has a power problem.
One possible solution is the radioisotope thermoelectric generator, or RTG. RTGs do not rely on sunlight. Instead, they convert heat from radioactive decay into electricity and have powered many U.S. space missions, including Apollo surface experiments, the Voyager probes, Cassini, Curiosity, and Perseverance. RTGs are attractive for lunar surface missions because they provide continuous power and dissipate heat through day-night cycles, dust exposure, and extreme temperature swings.
Historically, the isotope used in U.S. RTGs has been Pu-238. Yet Am-241 is often discussed as a possible alternative, especially in Europe. Am-241 is more available from the civilian nuclear fuel cycle and has a much longer half-life than Pu-238, but it produces much less heat per unit mass. [1-3]
Pu-238 has historically been favored because it is a near-ideal radioisotope heat source. It is primarily an alpha emitter, has a long but not excessively long half-life, and has high specific thermal power. A U.S. Department of Energy assessment lists a half-life of 87.7 years and a specific thermal power of about 0.57 W/g for Pu-238. [1] These are excellent properties for space power. A representative modern RTG is NASA's Multi- Mission Radioisotope Thermoelectric Generator, which produces about 2000 W of thermal power and about 110 W of electric power at beginning of mission. Its oxide fuel inventory is about 4832 g, with a Pu-238 mass fraction of about 0.72, corresponding to roughly 3.48 kg of Pu-238. [4]
Am-241 is considered because it offers some different advantages. DOE product information gives it a half-life of 432.6 years, almost five times longer than Pu-238. [2] Idaho National Laboratory notes that it is produced through the decay of Pu-241 and can be harvested from aged plutonium stocks, including material associated with civilian fuel-cycle streams. [3] In that sense, Am-241 is attractive not because it is better as a heat source, but because it may be easier for some countries to obtain in useful quantities. This is one reason the European space community has invested in americium-based radioisotope power development. [5]
Am-241 also has one subtle operational advantage: because its half-life is so long, its thermal output changes only slowly over mission time. On the other hand, the disadvantage is much larger: Am-241 produces far less heat per kilogram than Pu-238. INL summarizes the difference as, about 5 kg of Am-241 is equivalent to about 1 kg of Pu-238. [3] That ratio already suggests the likely outcome of the calculation.
A fair comparison between Pu-238 and Am-241 begins by holding the thermoelectric conversion efficiency fixed. This isolates the effect of isotope choice rather than differences in converter design. The MMRTG produces about 110 W electric from about 2000 W thermal at beginning of mission, corresponding to a beginning-of-mission conversion efficiency of approximately 5.5%. [4] Using this value as a reference, a 100 W electric RTG requires a beginning-of-life thermal power of
|
(1) |
A real example of this power scale is NASA's Multi-Mission Radioisotope Thermoelectric Generator, shown in Fig. 1. It produces about 110 W electric at beginning of mission from about 2.00 kW thermal. [4] The 100 W case is therefore a reasonable reference point for a small lunar radioisotope power system.
The specific thermal power of Pu-238 is about 0.57 W/g. [1] Supplying 1.82 kW thermal therefore requires a Pu-238 mass of
|
(2) |
This result is consistent with actual flight hardware. The MMRTG contains about 4832 g of oxide fuel with a Pu-238 mass fraction of about 0.72, corresponding to roughly 3.48 kg of Pu-238, and produces about 110 W electric at beginning of mission. [4]
A representative specific thermal power for Am-241 is about 0.1146 W/g. [6] Using the same required thermal output as above, the corresponding Am-241 mass is
|
(3) |
The mass ratio is therefore
|
(4) |
Thus, for the same 100 W electric output, an Am-241-fueled RTG would require about five times as much isotope mass as a Pu-238-fueled RTG, assuming comparable thermoelectric conversion efficiency. This agrees well with the Idaho National Laboratory estimate that roughly 5 kg of Am-241 is thermally equivalent to 1 kg of Pu-238. [3]
Am-241 does have one important advantage: its half-life is much longer. The thermal output of a radioisotope source falls with time according to
|
(5) |
Using half-lives of 87.7 years for Pu-238 and 432.6 years for Am-241, the remaining fraction of thermal power after ten years is
|
(6) |
Over a ten-year mission, Pu-238 therefore loses about 7.6% of its thermal output, whereas Am-241 loses only about 1.6%. [1,2] This slower decline is a great advantage of americium, but over a mission lasting only a decade the effect remains fairly small. Its importance grows mainly for much longer missions, such as outer solar system probes or Voyager-like spacecraft expected to operate for many decades, where even a modest reduction in power loss can become significant. Fig. 2 illustrates this difference by showing the remaining thermal power of Pu-238 and Am-241 as a function of mission duration. Nevertheless, the benefit is minimized compared with the roughly fivefold mass penalty in specific thermal power.
 |
| Fig. 2: Remaining thermal power fraction of Pu-238 and Am-241 as a function of mission duration, calculated from Eq. (5) for thermal output of a radioisotope source using T1/2 = 87.7 years for Pu-238 and T1/2 = 432.6 years for Am-241. [1,2] (Image Source: T. Ohal) |
Pu-238 remains the standard RTG fuel for three main reasons. First, its specific thermal power is much higher, which allows a lighter and more compact system. [1] Second, it has extensive flight heritage in U.S. space missions. Third, the surrounding heat-source technology and safety qualification are mature. By contrast, americium has generally been investigated as an alternative where Pu-238 supply is limited rather than because it offers superior performance. [3,5]
This does not mean that Am-241 is unusable. Its long half-life and potential availability from aged plutonium stocks make it attractive for certain long-duration or supply-constrained missions. [2,3] But the numerical comparison has shown that americium can function as a substitute, yet it does so at a substantial penalty in an increased power system for the mission.
The 100 W reference case is useful because it shows the isotope trade clearly: Am-241 can produce the same electrical output as Pu-238, but only with a much larger fuel mass. For Artemis, however, the more important point is that a 100 W RTG is a small auxiliary power source, not a habitat power system. A unit of this size could be useful for remote instruments, communications relays, or scientific packages operating in shadowed or poorly illuminated regions near the lunar south pole. In such applications, continuous operation may matter more than low mass, which is one reason alternative radioisotope fuels are of interest.
A crewed Artemis habitat is a very different problem. NASA habitat studies for the lunar south pole assume a deployable solar array of 10-15 kW, regenerative fuel cells providing up to 3 kW during eclipse periods lasting up to 100 hours, and a thermal control system designed to reject more than 15 kW of heat. [7] These figures place the habitat firmly in the multi-kilowatt to tens-of-kilowatts regime, far above the 100 W scale considered here. For an order-of-magnitude comparison, Eqs. (7) and (8) scale the 100 W RTG result to a habitat-class power requirement:
|
(7) | |||||
|
(8) |
These masses refer only to the radioisotope fuel itself and do not include converters, shielding, structure, or other hardware. Even as an order-of-magnitude estimate, the result shows that neither Pu-238 nor Am-241 RTGs are practical as the primary power source for a crewed Artemis base. [7] Their likely role in Artemis is therefore limited to low-power instruments or remote surface assets where continuous operation matters more than total output.
For the same 100 W electric RTG output, Am-241 requires about five times as much isotope mass as Pu-238. Using a conversion efficiency consistent with NASA's MMRTG, a 100 W electric system requires about 1.82 kW thermal. This corresponds to about 3.19 kg of Pu-238 or about 15.8 kg of Am-241. The longer half-life of Am-241 means that its power output declines more slowly over time, but this advantage is too small to overcome its much lower specific thermal power.
For Artemis, the result suggests that Am-241 is most plausible in small RTGs intended for remote instruments or survival power in difficult lunar environments. It is not a compelling replacement for Pu-238 when system mass and compactness are important, and neither isotope is suitable for the main electrical supply of a crewed lunar habitat. The most cautious conclusion is therefore that Am-241 is a credible alternative for long-duration radioisotope power, but best understood as a mass-penalized substitute rather than a true replacement for Pu-238 in Artemis-class lunar RTGs.
© Tanav Ohal. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
[1] J. S. Dustin and R. A. Borrelli, "Assessment of Alternative Radionuclides For Use in a Radioisotope Thermoelectric Generator," Nucl. Eng. Des. 385, 111475 (2021).
[2] "Specifications: Americium-241," U.S. National Isotope Development Center, November 2024.
[3] S. G. Johnson, "Considerations for Use of Am-241 for Heat Source Material for Radioisotope Power Systems," Idaho National Laboratory, INL/EXT-16-40336, January 2017.
[4] D. J. Clayton et al., Nuclear Risk Assessment for the Mars 2020 Mission," Sandia National Laboratories, SAND2013-10589, January 2014.
[5] "GSTP Annual Report 2021," European Space Agency, 2021.
[6] M. J. Sarsfield et al., "The Separation of 241Am from Aged Plutonium Dioxide for Use in Radioisotope Power Systems and Radioisotope Heater Units," E3S Web Conf. 16, 05003 (2017).
[7] R. G. Schunk, S. D. Babiak, and B. W. Evans, "Thermal Control System Architecture and Technology Challenges for a Lunar Surface Habitat," IEEE Aerospace Conference, IEEE 9843698, 5 Mar 22.