Past US Space Nuclear Power And Propulsion Programs

nuclear Extended stays on planetary surfaces will likely require nuclear power sources, to augment solar power systems and/or to provide emergency power. This artist’s impression shows a nuclear reactor providing power to a Moon base. (credit: Antares Nuclear) Atoms for space: Past US space nuclear power and propulsion programs by Dwyane A. Day Monday, July 13, 2026 The Nuclear Age and the Space Age were inevitably intertwined from the start. Missiles with nuclear warheads became rockets for launching spacecraft. Early US studies of satellites in the 1950s included a nuclear power source for the satellite. Because nuclear power was relatively compact and the United States was spending massive sums developing it in the 1950s, satellite designers began evaluating it for powering spacecraft. Rocket engineers also began considering nuclear power for propulsion as early as 1947. Nearly $20 billion in constant-year dollars has been spent over nearly seven decades on space nuclear power and propulsion systems. Although radioisotope thermoelectric generators (RTGs) used for outer planet spacecraft such as Voyager, Galileo, and Cassini are the most well-known nuclear systems, many other technologies have been studied, a few occasionally progressed to advanced testing, but rarely launched. According to one recent study, nearly $20 billion in constant-year dollars has been spent over nearly seven decades on space nuclear power and propulsion systems. This is a brief discussion of many of them. nuclear The NERVA A3 (also known as the NRX-A3) prior to a 1965 test. (credit: Dept. of Energy) Rover and NERVA (1955–1972) Chemical rockets are limited in the amount of thrust that they can produce, prompting engineers to consider more advanced forms of propulsion. In the mid-1950s, the United States initiated a program to develop nuclear propulsion for spacecraft with the expectation that it could be used for future human missions, particularly to Mars. The basic technology involved passing hydrogen through a very high temperature nuclear reactor, where it expanded and blasted out of the reactor at high velocity. There have been various proposals for this technology, which can theoretically produce thrust at much higher efficiency than chemical propulsion. By the 1960s, NASA and the Atomic Energy Commission jointly ran two main programs, a reactor technology development program named Rover, and a program to develop a flight capable nuclear rocket engine known as NERVA, for Nuclear Energy for Rocket Vehicle Application. nuclear The NERVA A3 (also known as the NRX-A3). (credit: Dept. of Energy) Under the Rover program, nuclear reactors were built at the Los Alamos Laboratory’s Pajarito Site and tested at very low power, then transported to the Nevada Test Site for higher power tests. Laboratory work also included developing and testing the fuel elements that powered the reactors. nuclear Relative size of the KIWI test engines. (credit: Dept. of Energy) Phase one of Project Rover was called Kiwi and entailed building and testing eight reactors between 1959 and 1964. Phase two, called Phoebus, involved advanced nuclear reactors. NERVA began in 1961 and by the mid-1960s progressed to hardware development tests in the Nevada desert. Rockets were fired in the desert, producing radioactive exhaust. The NERVA engine utilized a hot-bleed cycle in which a small amount of hydrogen gas was diverted from the thrust nozzle to drive the turbine that pumps fuel into the engine. NERVA reached an integrated system component demonstration readiness level. It did not result in a flight-ready or near-flight ready design. nuclear nuclear nuclear NASA envisioned using NERVA rocket engines for a variety of missions, including the Moon and Mars. (credit: NASA) At various points NASA planned on using NERVA as a rocket upper stage, a space ferry for lunar missions, and a propulsion stage for human Mars missions. A human Mars mission proposal made in 1969 would have used multiple NERVA engines that could be recovered in space for reuse. NASA produced extensive artwork depicting these various missions. However, after Apollo, none of these projects was approved and NERVA therefore had no dedicated mission. nuclear nuclear NASA future plans conceived of reusable nuclear upper stages that could be refueled in Earth orbit. At a time when simply making the nuclear rocket engine work was a challenge, this was a highly ambitious vision. (credit: NASA) Congress showed greater support for NERVA than the White House, but the program was eventually canceled by 1972, when its powerful patron chose not to seek reelection and President Richard Nixon wanted to use the money for the Pentagon. Rover and NERVA were collectively the most expensive space nuclear program pursued by the United States. Some of their radioactive test equipment remains quarantined in the desert even today, relics of a time when nuclear rockets seemed to many like the space propulsion technology of the near future. nuclear Orion was a series of studies for propelling a large spacecraft using nuclear explosions. It was considered for both civil and military missions. (credit: NASA) Orion (1955–1965) Stanislaw Ulam and Cornelius Everett, who had designed the H-bomb, conceived the nuclear pulse drive at Los Alamos in 1955. The Atomic Energy Commission, and later the US Air Force, sponsored studies of this technology under the name Project Orion. Even Orion’s designers acknowledged that it was a difficult project with limited chance of success. Orion would use nuclear explosions to propel a spacecraft. Nuclear bombs would be ejected aft of a spacecraft and explode some distance away. Propellant (water or wax) surrounding the bombs would be transformed into high-energy plasma and bounce off a pusher plate at the rear of the spacecraft and push it forward. These detonations would have to happen rapidly, approximately every second. Shock absorbers would reduce the substantial vibration caused by detonating nuclear weapons behind the craft. Although the plasma from the explosion would have a temperature of 80,000 kelvins, the impulse would be brief and theoretically only a tiny layer of the ablative pusher plate would sublimate after each explosion. The vehicle would be ground-launched from a remote location, propelled into space atop a nuclear pillar. It would not be kind to the surrounding landscape. In theory, the Orion design allowed vast payloads to be hurled to the planets. The Air Force was interested in developing an Orion into a type of space battleship, equipped with hundreds of nuclear weapons that could be fired at Earth or other targets (see “General Power vs. Chicken Little,” The Space Review, May 23, 2005.) A typical design had a payload of hundreds of tons, meaning no complex environmental recycling systems or lightweight structures or equipment would be required. The pusher plate was about one third of the weight of the spacecraft. In 1957, the project was transferred to the Air Force and produced small-scale demonstration tests involving conventional explosives detonated under a model suspended from a crane. The 1963 Nuclear Test Ban Treaty banned atmospheric nuclear tests, essentially killing the program. Funding was eliminated entirely in 1965. Even Orion’s designers acknowledged that it was a difficult project with limited chance of success. nuclear A 1961 illustration of a nuclear-powered spacecraft. (credit: NASA) Systems For Nuclear Auxiliary Power reactor (1959–1971) In 1946, the Air Force’s Project RAND began studying Earth-orbiting satellites and the early studies assumed using nuclear reactors to power the satellites. At the time there was no other way to provide sufficient, long-lasting power to an orbiting satellite. nuclear The earliest studies of satellites included nuclear reactor power sources. This illustration is from a declassified mid-1950s report on using a satellite for reconnaissance. (credit: RAND Corp.) Over the next decade, the Air Force funded low-level study efforts of space nuclear reactors. Early spacecraft could use batteries for a short lifetime, but the eventual development of solar cells made nuclear reactors unnecessary. Nuclear reactors might have some utility for spacecraft, but they were not required for the vast majority of space missions. By the late 1950s, the Atomic Energy Commission began the Systems for Nuclear Auxiliary Power (SNAP) program to develop fission reactors and radioisotope generators for both terrestrial and space use. Radioisotope and reactor programs were given odd- and even-number designations, respectively. nuclear In December 1957, Disney aired an episode of its Disneyland program called “Mars and Beyond,” which featured a fleet of spacecraft carrying humans to Mars. They were powered by nuclear reactors and propelled by electric propulsion systems. The reactor can be seen at the bottom of the boom of the spaceships, which also featured large radiators for cooling. Although not practical, it was definitely cool. (credit: Disney) Starting in the 1950s, nuclear reactors were envisioned both for providing power to spacecraft, and in powering electric thrusters. In 1957, a Disney television episode featured a fleet of spacecraft (often referred to as “umbrella ships”) heading to Mars using nuclear reactors to power electric thrusters. Although fanciful, the concept was based upon engineering principles. By the early 1960s, NASA science programs examined the use of nuclear reactors to power electric propulsion systems to journey to the outer planes. This concept reemerged several times in later decades, but the high cost of developing a space nuclear reactor was always prohibitive. nuclear nuclear nuclear In the early-mid-1960s, nuclear reactors paired with electric propulsion systems were considered for deep space missions. A typical design placed the reactor far from the instruments and the propulsion system. It also required large radiators to dissipate heat. This remains the basic design even today. (credit: NASA) In 1965, the Air Force launched the SNAP-10A reactor into orbit. It operated for 43 days, producing 500 watts of power until the failure of a voltage regulator caused it to shut down. The SNAP-10A reactor was a small zirconium hydride (ZrH) thermal reactor fueled by uranium-235. To date, it is the only American nuclear reactor flown in space. nuclear SNAP-10A was the only nuclear reactor ever flown in space by the United States. It was operated for 43 days in 1964. (credit: NASA) nuclear The SNAP reactor programs were considered for various civil and military applications during the 1960s. A SNAP reactor could have replaced solar panels to provide longer-term power for an Earth-orbiting space station. Steady improvements in solar panel technology undercut most justifications for space reactors. (credit: US Air Force) Additional space nuclear reactor projects were studied throughout the 1960s. These included SNAP-2 and SNAP-8, as well as the SNAP-50/SPUR. They were much higher power compared to SNAP-10A. nuclear SNAP-8 was a space reactor development program in the 1960s. Although now mostly forgotten, it was an extensive test program. (credit: NASA) SNAP-8 was an extensive space reactor development program that included the construction of several test reactors at a NASA facility in Ohio. The primary application for SNAP-8 was for powering a human space station in low Earth orbit. Although substantial money and effort was spent on SNAP-8, it is mostly forgotten in space history books. nuclear nuclear The SNAP-8 nuclear reactor program was actively developed in the 1960s, with test reactors built and tested in Ohio. One possible use was powering an Earth-orbiting space station. Operating a reactor in space required careful consideration of operations so that astronauts were not exposed to the reactor radiation. (credit: NASA) SNAP-50/SPUR was a mid-1960s study of a more powerful reactor with an output of 300 kilowatts to 1 megawatt of electrical power. It was envisioned for launch on a Saturn IB and used to provide power for a variety of planetary missions, including Mercury, Mars, Jupiter and Saturn orbiters, as well as a solar probe. nuclear SNAP-50 was a space reactor program studied in the 1960s and considered for powering planetary missions. (credit: NASA/JPL/Caltech) Space nuclear reactors involve many different technology questions including the nuclear fuel, and the method for turning the heat into electrical energy. In addition, the systems have major safety requirements that affect the design. None of the SNAP reactor projects were pursued to flight hardware stage due high costs and the lack of a clear requirement. They were canceled by the late 1960s. nuclear The Lincoln Experimental Satellites 8 and 9 both used RTGs for power. They operated in very high orbits. It is rumored that one of these satellites was used to test stealthy technology to hide satellites. (credit: US Air Force) Radioisotope thermoelectric generators (1961–Present) Radioisotope thermoelectric generators (RTGs) convert heat generated by the natural decay of radioisotope fuel (typically plutonium-238 in the United States) into electricity through thermoelectric coupling. Over nearly 80 years, the United States has spent many billions of dollars developing RTG technology and producing Pu-238. Maintaining the infrastructure to produce both the fuel and the RTGs has proven expensive. RTGs produce very low power, but they have no moving parts and are very reliable, often lasting for decades. nuclear A fuel pellet for an RTG. It is about the size of a marshmallow. Although RTGs have no moving parts, they represent sophisticated materials and manufacturing technology, and the plutonium fuel is expensive to produce. (credit: Dept. of Energy) The first two RTG demonstration systems were flown in 1961 (called SNAP-3B) and generated three watts of power. Later ones were flown on several earth-orbiting spacecraft, such as the Transit navigational satellite program. nuclear The SNAP-19 RTGs used on two Nimbus Earth observation satellites in the later 1960s were relatively small. (credit: NASA) nuclear Two SNAP-19 radioisotope thermoelectric generators were lost during a launch accident in 1968 and ended up in the Pacific Ocean. They were found and recovered. (credit: NASA) RTGs have also been used to power terrestrial devices, some classified and some unclassified. One of the more infamous examples was a small RTG, a SNAP-19C, used to power a signals intelligence collection device on top of a mountain in the Himalayas for spying on China. Although one appears to have been successfully installed, another of these RTGs was lost on the mountain in 1965, a story that was later reported by a mountain climber who had been involved in the operation. nuclear Plutonium-powered radioisotope thermoelectric generators have been used for terrestrial purposes. Here an RTG is used to power a top secret signals intelligence station on a mountaintop. (credit: New York Times) From the start, SNAP systems were planned for terrestrial, ocean, and space use, and came in a variety of types. A list of the types planned and used by the mid-1960s is below: Designation Use Power (watts) Isotope SNAP-3 Demonstration device 2.5 Polonium-210 SNAP-3A Satellite power 2.7 Plutonium-238 Undesignated Axel Heiberg weather station 5 Strontium-90 SNAP-7A Navigational buoy 10 Strontium-90 SNAP-7B Fixed navigational light 60 Strontium-90 SNAP-7C Weather station 10 Strontium-90 SNAP-7D Floating weather station 60 Strontium-90 SNAP-7E Ocean-bottom beacon 7.5 Strontium-90 SNAP-7F Offshore oil rig 60 Strontium-90 SNAP-9A Satellite power 25 Plutonium-238 SNAP-11 Moon probe 21-25 Curium-242 SNAP-13 Demonstration device 12 Curium-242 SNAP-15A Military use 0.001 Plutonium-238 SNAP-17 Communications satellite 25 Strontium-90 SNAP-19B Nimbus-B weather satellite 30 Plutonium-238 SNAP-21 Deep sea use 10 Strontium-90 SNAP-23 Terrestrial uses 60 Strontium-90 SNAP-27 Lunar landings 60 Plutonium-238 SNAP-29 Various missions 500 Polonium-210 RTGs were used on Apollo 12, 14, 15, 16, and 17 to supply power to the Apollo lunar surface experiment packages (ALSEP). Several of those RTGs had relatively short lifetimes, although others lasted longer. The ALSEP stations on the Moon were shut down prior to their power running below useable levels, although they would not have lasted much longer. (The Apollo 13 RTG ended up at the bottom of the ocean.) nuclear Apollo 12 astronaut Alan Bean extracts the fuel element of a Radioisotope Thermoelectric Generator (RTG) from storage on the lunar module on the Moon. The RTG was used to power the Apollo Surface Experiment Package (ALSEP). (credit: NASA) nuclear The SNAP-27 RTG used to power the Apollo Lunar Surface Experiments Package (ALSEP) for the Apollo 14 mission. NASA has long and successful experience with RTGs, but the technology has been expensive to develop and sustain. (credit: NASA) RTGs were also used to provide primary power to Viking 1 and 2 (SNAP-19, 85 watts total), Pioneer 10 and 11 (SNAP-19, 165 watts total), Voyager 1 and 2 (each RTG provided 157 watts), Galileo (300 watts total, along with 120 lightweight radioisotope heater units), Ulysses (same as Galileo), and Cassini (three RTGs delivering 888 watts, along with 117 lightweight radioisotope heater units). nuclear Containers for storing RTGs for transport at Cape Canaveral in 1976. (credit: NASA) nuclear Radioisotope thermoelectric generators (RTGs) have powered many spacecraft since the early 1960s. Here the RTG for the Cassini spacecraft is prepared for integration with the spacecraft in the 1990s. (credit: NASA) Other than the two Voyager missions that are still operating nearly fifty years after launch, there are a few operational spacecraft using RTGs for power. An RTG powers the New Horizons mission that flew past Pluto in 2015 and is now traveling through the Kuiper Belt. The Curiosity and Perseverance rovers also use Multi-Mission Radioisotope Thermoelectric Generators (MMRTGs) designed for planetary surface use, and the Dragonfly spacecraft that will be launched to Titan will also use an MMRTG. A spare MMRTG remains in storage and has recently been proposed to power a Moon rover. nuclear Voyager 2, still operating deep out in space, is powered by RTGs produced 50 years ago, demonstrating the value of the RTG technology. (credit: NASA/JPL/Caltech) nuclear The Curiosity rover has operated for nearly fourteen years on the surface of Mars powered by a Multi-Mission Radioisotope Thermoelectric Generator, or MMRTG. The MMRTG is the white object at the rear of the spacecraft. (credit: NASA) Despite the expenditure of considerable time, effort, and money to improve the efficiency of the power conversion of RTGs over many decades, very little progress has been made in those subjects since the early RTGs were developed. The technology is mature, and improvements have been incremental. Many other systems, including all Mars rovers, have used lightweight radioisotope heater units to maintain appropriate temperatures for system electronics. China has also used similar heater units supplied by Russia in several of their lunar landers. These systems have been successfully deployed in space: Power Source Spacecraft Mission Type Launch Date Status SNAP-3B7 Transit 4A Navigational 6-29-61 RPS operated for 15 years. SNAP-3B8 Transit 4B Navigational 11-15-61 RPS operated for 9 years. Satellite operated periodically after 1962 high altitude test. Last reported signal in 1971. SNAP-9A Transit 5-BN-1 Navigational 9-28-63 RPS operated as planned. Non-RPS electrical problems on satellite caused satellite to fall after 9 months. SNAP-9A Transit 5-BN-2 Navigational 12-5-63 RPS operated for over 6 years. Satellite lost ability to navigate after 1.5 years. SNAP-9A Transit 5-BN-3 Navigational 4-21-64 Mission was aborted because of launch vehicle failure. RPS burned up on re-entry as designed. SNAP-19B2 Nimbus-B-1 Meteorological 5-18-68 Mission was aborted because of range safety destruct. RPS heat sources recovered and recycled. SNAP-19B3 Nimbus III Meteorological 4-14-69 RPSs operated for over 2.5 years. RHU Apollo 11 Lunar Surface 7-14-69 Radioisotope heater units for seismic experimental package. Station was shut down 8-3-69. SNAP-27 Apollo 12 Lunar Surface 11-14-69 RPS operated for about 8 years until station was shut down. SNAP-27 Apollo 13 Lunar Surface 4-11-70 Mission aborted on the way to the Moon. RPS re-entered earth's atmosphere and landed in South Pacific Ocean. No radiation was released. SNAP-27 Apollo 14 Lunar Surface 1-31-71 RPS operated for over 6.5 years until station was shut down. SNAP-27 Apollo 15 Lunar Surface 7-26-71 RPS operated for over 6 years until station was shut down. SNAP-19 Pioneer 10 Planetary 3-2-72 Spacecraft successfully operated to Jupiter and is now beyond orbit of Pluto. SNAP-27 Apollo 16 Lunar Surface 4-16-72 RPS operated for about 5.5 years until station was shut down. RTG "Transit" (Triad-01-1x) Navigational 9-2-72 SNAP-27 Apollo 17 Lunar Surface 12-7-72 RPS operated for almost 5 years until station was shut down. SNAP-19 Pioneer 11 Planetary 4-5-73 Spacecraft successfully operated to Jupiter, Saturn, and beyond. SNAP-19 Viking 1 Mars Surface 8-20-75 RPSs operated for over 6 years until lander was shut down. SNAP-19 Viking 2 Mars Surface 9-9-75 RPSs operated for over 4 years until relay link was lost. MHW-RTG LES 8 Communications 3-14-76 MHW-RTG LES 9 Communications 3-14-76 MHW-RTG Voyager 2 Planetary 8-20-77 RPSs still operating. Spacecraft successfully operated to Jupiter, Saturn, Uranus, Neptune, and beyond. MHW-RTG Voyager 1 Planetary 9-5-77 RPSs still operating. Spacecraft successfully operated to Jupiter, Saturn, and beyond. GPHS-RTG and LWRHU Galileo Planetary 10-8-89 RPSs operated through 9-17-03, when the spacecraft completed its mission. GPHS-RTG Ulysses Planetary/Solar 10-6-90 Mission ended. LWRHU Mars Pathfinder Mars Surface 12-4-96 LWRHU provided essential heat to Sojourner electronics for 84 days GPHS-RTG and LWRHU Cassini Planetary 10-15-97 Mission ended. LWRHU Mars Exploration Rovers Mars Surface 6-25-03 and 7-5-03 LWRHU provided essential heat to electronics GPHS-RTG New Horizons Planetary 1-19-06 RPS still operating. MMRTG Curiosity and Perseverance Rovers Mars Surface 11-26-11 and 7-30-20 RPS still operating. MMRTG Dragonfly Planetary 2030 NOTE: An RPS is a generic name for a family of different technologies, including the RHU, RTG, MHW-RTG, etc. SNAP = Systems for Nuclear Auxiliary Power RPS = Radioisotope Power System RTG = Radioisotope Thermoelectric Generator RHU = Radioisotope Heater Units MHW-RTG = Multi-Hundred-Watt RTG GPHS-RTG = General Purpose Heat Source RTG LWRHU = Lightweight Radioisotope Heater Units The MMRTG is not suitable for all space missions. Recently, L3Harris Technologies finalized the design of a next-generation nuclear-based power source for future NASA deep space missions. Known as the Next-Generation Radioisotope Thermoelectric Generator (Next Gen RTG), flight units could power NASA deep space probes starting in the early 2030s. nuclear The Next Generation RTG program has developed flight hardware to support deep space missions. (credit: L3Harris) Now that NASA has announced plans for a Moon base, RTGs may also be pressed into service to provide emergency power in some circumstances, as well as power deployed for long-duration instruments. At least one American company has begun work on an RTG powered by americium, an element that is easier and cheaper to produce than the plutonium-238 used in previous American RTGs. However, it has a much lower energy density, meaning that more of it must be carried to produce the same amount of power as RTGs powered by Pu-238. For deep space missions, this could have a deleterious effect on spacecraft design, because bigger RTGs will impact many other aspects of the spacecraft such as propulsion and stabilization, but it may be acceptable for lunar surface use. One other downside is that NASA equipment and procedures have been molded by decades of Pu-238 handling, and new equipment and procedures would be required for using a different isotope. nuclear The Tau Mission was a 1986 proposal for a mission to deep space. It required a nuclear reactor and electric propulsion systems that would have had to operate reliably for a very long time. (credit: NASA/JPL/Caltech) SP-100 (1983–1995) After the cancellation of SNAP-8 and SNAP-50, no space nuclear reactor programs were pursued by the United States during most of the 1970s. By 1981, a project known as the Space Power Advanced Reactor (SPAR) was underway to develop a space nuclear reactor producing 100 kilowatts of electrical power with a seven-year design life. This work was curtailed, but by 1983, the National Aeronautics and Space Administration, Department of Defense, and Department of Energy began development of a 100-kilowatt-electric-class space nuclear reactor, called the SP-100. nuclear nuclear The SP-100 was a space nuclear reactor program studied in the 1980s. Possible users included NASA and the Department of Defense, but there was no clear requirement for it and the program was canceled. (credit: General Electric) The SP-100 was designed with a two-megawatt-thermal fast reactor unit and thermoelectric system delivering 100 kilowatts-electric for a seven-year design life. This reactor was intended to support a wide variety of military and civilian requirements, none of which emerged. One possible application included powering spacecraft as part of the Strategic Defense Initiative: the program never had a clear need for such a system, although one of the primary proposed uses was to harden an important satellite by replacing large and vulnerable solar panels with a much more compact reactor. Because of high costs, schedule delays, and changing national space mission priorities, the SP-100 program was suspended by the early 1990s and later canceled. Timber Wind (1987–1992) The Timber Wind space nuclear thermal propulsion program was initiated by the Strategic Defense Initiative Organization in November 1987. Timber Wind was a highly classified program to develop technology for a very-high-acceleration nuclear-powered rocket to launch missile interceptors into space. Similar to NERVA, a propellant would be pushed through a nuclear reactor and super-heated and expelled from a nozzle, but whereas NERVA was planned for in-space use, Timber Wind would have launched from the ground. In 1991, after its existence was leaked to the media, the project was transferred to the Air Force but terminated a year later. Some information on Timber Wind was declassified at that time. The exhaust would be highly radioactive, but the reasoning was that it would only be used during a nuclear attack where radioactive rocket exhaust would be preferable to mushroom clouds over American cities. Project Timber Wind was based on a particle bed reactor using tiny uranium carbide pellets as fuel. The pellets would be used to heat hydrogen propellant. The exhaust would be highly radioactive, but the reasoning was that it would only be used during a nuclear attack where radioactive rocket exhaust would be preferable to mushroom clouds over American cities. The project had selected preliminary designs but had not tested any prototype components before the program was cancelled. nuclear The Topaz-2 project was an early-mid-1990s program where the United States evaluated an advanced space nuclear reactor developed in the former Soviet Union. Unfueled reactors and test hardware were shipped to the United States and tested. They were eventually returned to Russia. (credit: Phillips Laboratory) Topaz-2 (1980s–1990s) The Soviet Union had a far more active space nuclear power program than the United States, although most of its focus was on in-space nuclear reactors rather than RTGs. The country launched dozens of nuclear reactors into space during the 1970s and 1980s where they were used to power radars to detect American ships at sea. The satellites were known as Radar Ocean Reconnaissance Satellites, or RORSATs. By the late 1980s, the Soviet Union had several separate space nuclear reactor programs under development. These were not all replacements for the RORSAT reactor. One of these projects, mistakenly labeled “Topaz-2” in the United States, was actually known as “Enisey.” This project was purchased by the United States’ Ballistic Missile Defense Organization in the early 1990s and known as Topaz-2. (To add to the confusion, two Soviet nuclear reactors known as Topaz were flown in orbit in the later 1980s, but were an entirely different design.) Six Topaz-2 reactors and supporting equipment were flown from Russia to the United States. Several of the reactors were extensively ground-tested by a joint team of US, British, French and Russian engineers. The reactors’ unique design allowed them to be tested without nuclear fuel, using heating units to simulate the fuel. Topaz utilized a thermionic design for directly converting heat energy into electricity without using a circulating heat transfer fluid or turbine. The ground test program in the United States was considered highly successful, but there was limited technology transfer from Russia to the United States. The United States initially retained several flight capable reactors, and some proposals were advanced to test fly one of the reactors in space, but the program was terminated without any flights. The hardware was later returned to Russia. The legacy of this unusual test program is worthy of future analysis. nuclear The Jupiter Icy Moons Orbiter (JIMO) program was an early 2000s project to send a large spacecraft to orbit Jupiter. It was extremely ambitious and had a price tag over $20 billion. (credit: NASA) Prometheus/Jupiter Icy Moons Orbiter (JIMO) (2002-2005) Several other space nuclear power research programs were conducted during the 1980s and 1990s, although funding was always limited and there is little public information about them. The cost of the overall JIMO program (not including the Prometheus-1 flight test) was $21.5 billion, which would have made it by far the most expensive space science program ever pursued. Starting in late 2002, NASA began pursuing development of a large spacecraft that could orbit Jupiter and explore the moons Ganymede, Callisto, and Europa. The mission was known as the Jupiter Icy Moons Orbiter (JIMO). The technology development program was named Prometheus. Prometheus had the goal of developing a 200-kilowatt-electric reactor capable of operating for the ten-year lifetime of the spacecraft. As the program progressed throughout 2003 and 2004, NASA began considering a test flight known as Prometheus-1. Naval Nuclear Reactors was engaged for development of the reactor, and private contractors and JPL were involved in designing the Brayton power conversion system and the spacecraft. Despite the expenditure of $463 million, JIMO was far from entering full-scale development. The cost of the overall JIMO program (not including the Prometheus-1 flight test) was $21.5 billion, which would have made it by far the most expensive space science program ever pursued. It was canceled in early 2005 (see “Prometheus bound: The legacy of the Jupiter Icy Moons Orbiter,” The Space Review, February 23, 2026.) nuclear In 2018, a space reactor designated KRUSTY was tested on the ground in the United States. Despite success, the program was not pursued to a flight test stage. This has been a common story for many American space nuclear programs. (credit: Wikimedia Commons) Kilopower/KRUSTY (2015–) In the 2010s, NASA and the Department of Energy began development of a nuclear reactor for providing surface power for a lunar outpost. The reactor consisted of a solid core of uranium-235, regulated by a control rod. Passive sodium heat pipes transferred heat from the core and Stirling engines converted that heat into electricity. The project was known as Kilowatt Reactor Using Stirling Technology, or KRUSTY. From November 2017 to March 2018, NASA carried out tests of the reactor, culminating in a 28-hour test of a one-kilowatt-electric Kilopower reactor during which the reactor was turned on, brought up to and operated at full power, and then shut down. At that time, the goal was to conduct a flight test of the reactor. However, no flight test was funded. KRUSTY had progressed further than most other recent projects, but it still faced the problem that there was no clear, pressing demand for it, the kind of requirement that could lead to flight testing and deployment. In June 2022, NASA selected three teams for phase 1 studies of fission surface power systems, small nuclear reactors intended to support later phases of the Artemis lunar exploration campaign. The teams were led by Lockheed Martin, Westinghouse and IX, a joint venture of Intuitive Machines and X-Energy. The teams were to design a 40-kilowatt reactor weighing no more than six metric tons and could operate for 10 years. DRACO and JETSON (2023-2025) In early 2023, NASA and the Defense Advanced Research Projects Agency (DARPA) announced they were partnered on the Demonstration Rocket for Agile Cislunar Operations (DRACO), a nuclear-thermal rocket engine project. In July 2023, Lockheed Martin won a contract from the Defense Advanced Research Projects Agency (DARPA) to develop and demonstrate a nuclear-powered spacecraft. Lockheed was working with BWXT on the program, with BWXT providing the nuclear reactor for DRACO and providing its high-assay low-enriched uranium (HALEU) fuel. After the initial announcements, there were few updates on the program until May 2025, when NASA announced that it was zeroing its budget for DRACO. In July 2025, DARPA announced that it was canceling DRACO because decreasing launch costs no longer made it necessary. DARPA also indicated it had encountered “infrastructure barriers” associated with launching a nuclear reactor. nuclear JETSON is a more recent Lockheed Martin program to develop space reactor technologies. (credit: Lockheed Martin) In November 2023, Lockheed Martin was awarded a contract from the Air Force Research Laboratory (AFRL) for the Joint Emergent Technology Supplying On-Orbit Nuclear (JETSON). The contract was to mature high-power nuclear electric power and propulsion technologies and spacecraft design. That work did not lead to any flight hardware. New space nuclear reactor program (2025–) In July 2025, the acting NASA administrator ordered NASA’s Exploration Systems Development Mission Directorate to release a request for proposals (RFP) for a nuclear reactor system. The program seeks to develop a reactor capable of producing at least 100 kilowatts of electric power and weighing up to 15 tons. The program’s goal is to have the reactor ready for flight by the end of 2029. The reactor would operate in the lunar south polar region for at least 10 years. Other than the prolific and successful RTGs, the history of American space nuclear programs is primarily one of stunted efforts and cancellations, and every program, including the relatively simpler RTGs, have underestimated difficulty, translating into underestimation of development time, cost and effort. By early 2026, NASA indicated plans to develop both a Space Reactor-1 (SR-1) for in-space use, and a Lunar Reactor-1 (LR-1) for lunar use. Detailed plans and timelines have not yet been made public. However, the SR-1 was announced to have a power output of 20 kilowatts electric, a modest, but achievable goal. It would be part of the Skyfall mission to send small helicopters to Mars by 2028, a highly ambitious, and some would say unrealistic, goal. Although a nuclear reactor for surface use on the Moon would appear to be a more practical near-term goal for NASA, such a reactor would require a lander capable of putting it on the surface, an additional challenge compared to an in-space reactor. nuclear The Discovery One from 2001: A Space Odyssey, was nuclear-propelled. Early concepts of the spacecraft had large radiators to dissipate heat, but director Stanley Kubrick deleted them because he thought they looked too much like wings. (credit: MGM) Space nuclear power: chasing elusive atoms Science fiction novels and movies, like 2001: A Space Odyssey, depicted spacecraft powered by nuclear rockets. Nuclear reactors and nuclear rockets have always been viewed as a powerful and advanced method to explore space, but successful development of these systems has not been easy. Other than the prolific and successful RTGs, the history of American space nuclear programs is primarily one of stunted efforts and cancellations, and every program, including the relatively simpler RTGs, have underestimated difficulty, translating into underestimation of development time, cost and effort. There are several reasons for this. As Bhavya Lal and Roger M. Myers pointed out in a 2025 report, the three primary reasons for failure were lack of “mission pull,” technological overreach, and timeliness—most nuclear development projects took too long and did not survive changes in Congresses and administrations. Many of these projects exhibited two or more of these factors. NERVA was a rocket in search of a Mars mission that required it, and took so long to develop that political support evaporated. SP-100 was a space reactor in search of a spacecraft that needed it, and flush national security budgets of the early 1980s were diminished by late in the decade. JIMO/Prometheus was not a mission supported by the science community (it’s “mission pull” was the wishful thinking of a NASA administrator), and required development of a reactor more powerful than those studied before. Although it was canceled by a different NASA administrator a few years after inception, its price tag inevitably would have killed it. Because the technologies can do things beyond the capabilities of conventional systems like solar panels, space nuclear power and propulsion programs will certainly be pursued in the future. Hopefully to greater success. Sources Department of Defense Inspector General Audit Report, The Timber Wind Special Access Program, Report Number 93-033, December 16, 1992. James A. Dewar, To the End of the Solar System: The Story of the Nuclear Rocket, University Press of Kentucky, 2003. George Dyson, Project Orion: The True Story of the Atomic Spaceship, Owl Books, 2003. Steven Aftergood, “Background on Space Nuclear Power,” Science & Global Security, Vol. 1, Nos. 1-2 (1989), pp. 93-108. National Research Council, Assessment of the Topaz International Program, June 27, 1996. National Research Council, Thermionics Quo Vadis? An Assessment of the DTRA's Advanced Thermionics Research and Development Program, 2001. General Accounting Office, OSI-98-3R, TOPAZ II Space Nuclear Power Program: Management, Funding, and Contracting Problems. General Accounting Office, T-NSIAD-93-2, Space Nuclear Propulsion: History, Cost, and Status of Programs. General Accounting Office, T-NSIAD-92-15, The SP-100 Nuclear Reactor Program: Should It Be Continued? Jack F. Mondt, “SP-100 Space Reactor Power System for Lunar, Mars and Robotic Exploration,” 43rd Congress of the International Astronautical Federation, August 28-September 5, 1992, IAF 92-0563. Prometheus Project Final Report, NASA, Jet Propulsion Laboratory, October 1, 2005. Bhavya Lal and Roger M. Myers “Weighing the Future: Strategic Options for US Space Nuclear Leadership,” INL/RPT-25-85616, Idaho National Laboratory, July 2025. “NASA, DARPA Will Test Nuclear Engine for Future Mars Missions,” Space News, January 24, 2023. Jeff Foust, “NASA plans for lunar fission power systems face fiscal challenges,” Space News, July 20, 2023. “Lockheed Martin Selected to Develop Nuclear-Powered Spacecraft,” Space News, July 26, 2023. “BWXT to Provide Nuclear Reactor Engine and Fuel for DARPA Space Project,” Space News, July 28, 2023. Sandra Erwin, “Space nuclear power poised for breakthroughs — if NASA and DoD stay committed,” Space News, April 8, 2025. Jeff Foust, “DARPA says decreasing launch costs, new analysis led it to cancel DRACO nuclear propulsion project,” Space News, July 2, 2025. Jeff Foust, “New study calls for rapid development of space nuclear power systems,” Space News, July 16, 2025. Jeff Foust, “Industry supports NASA plans to accelerate work on lunar nuclear reactors,” Space News, August 8, 2025. Jeff Foust, “NASA advances lunar nuclear plan with commercial focus,” Space News, September 2, 2025. Jeff Foust, “Antares raises $96 million for nuclear reactors on Earth and in space,” Space News, December 3, 2025. Jeff Foust, “NASA and DOE to collaborate on lunar nuclear reactor,” Space News, January 24, 2026. Jeff Foust, “NASA to test nuclear electric propulsion with 2028 mission to Mars,” Space News, March 26, 2026. Jeff Foust, “The budget proposal that overshadowed Artemis 2,” Space News, May 5, 2026 Jeff Foust, “Ignition relaunches Artemis plans,” Space News, May 5, 2026. Jeff Foust, “NASA working to streamline development of nuclear electric propulsion demo mission,” Space News, June 4, 2026. Dwayne Day can be reached at zirconic1@cox.net. Note: we are now moderating comments. There will be a delay in posting comments and no guarantee that all submitted comments will be