For much of the past century, nuclear energy and space exploration were the emblems of the future.
These technologies offered a path to liberation from energy scarcity and gravity, and together they promised to open up the solar system for human exploration.
More than a decade before there were bootprints on the moon, NASA recognized the enormous potential for nuclear energy in spaceflight. Despite some early promising tests, NASA’s dream of a nuclear-powered Mars rocket never came to pass. But now, more than half a century later, a startup in Maryland is bringing the concept to life.
X-energy was founded in 2009 by Kam Ghaffarian, a deep-tech entrepreneur behind companies like Axiom, which is building the world’s first commercial space station, and Intuitive Machines, which could be the first to place a commercial lander on the moon. In a previous life, Ghaffarian was the cofounder of SGT, which became one of NASA’s largest services providers and endowed him with a fortune that he used to advance game-changing technologies on and off the planet.
When Ghaffarian founded X-energy he tasked the company with designing an advanced nuclear reactor that was cheaper and safer than any nuclear reactor in history. To achieve this lofty goal, he assembled a team of South African scientists who had been working on a next-generation nuclear reactor called the Pebble Bed Modular Reactor. This reactor had been under development since 1994 and was supposed to help wean South Africa off its coal addiction with safe and abundant clean energy.
But bringing the reactor to life proved to be a more difficult—and expensive—task than the South African government anticipated. The program was mothballed when the government cut funding in 2010. As the program wound down, Ghaffarian offered some of the South African scientists an opportunity to continue their work on pebble bed reactors at X-energy. The result is the Xe-100, an advanced small modular nuclear reactor that is designed to be meltdown-proof and smaller than any reactor in operation today.
The key to the reactor’s safety are the billiard ball-sized fuel pebbles that are rotated through its core like a gumball machine. Known as TRISO-X fuel, each pebble contains thousands of poppyseed-sized grains of uranium wrapped in layers of high-temperature ceramics. This shell contains the uranium fission reaction and prevents thermal runaway—otherwise known as a meltdown. This built-in safety feature allows X-energy’s nuclear reactor core to safely operate at temperatures above 3,000 degrees Fahrenheit, which is more than 3 times hotter than most conventional nuclear reactors.
There are many advantages to running hot—the steam can be used for process heat to decarbonize industrial processes or produce clean hydrogen, and the reactor is also highly fuel-efficient, which means it can squeeze more energy out of a smaller package. Each Xe-100 core generates about 80 megawatts of electric power, enough to power ~80,000 homes. Whereas a conventional nuclear reactor would take up several city blocks to make room for the bulky safety features, the inherent safety and fuel efficiency of X-energy’s reactor means it can be placed and operated on as few as ten acres of land.
X-energy’s fuel and reactor designs have attracted the attention of the Department of Energy. In 2015, the DoE selected X-energy to stand up a pilot TRISO-X production facility at the Oak Ridge National Laboratory, where the company has successfully demonstrated its ability to produce commercial quantities of its nuclear fuel. And last year, X-energy became one of two companies selected by the DoE to build their advanced nuclear reactor, which will be delivering power to residents of eastern Washington by 2027. This fast of a turnaround is unheard of in the nuclear industry, which typically requires at least a decade to go from design to operation.
X-energy’s timing was fortuitous. Just as the company’s advanced nuclear reactor design was maturing, NASA began to emphasize the role of nuclear thermal propulsion for a crewed mission to Mars. Now that the agency had plans to land humans back on the moon as a stepping stone to the Red Planet, nuclear-powered rockets were back in the game. And after a decade of work on an advanced high-temperature terrestrial reactor, X-energy had all the ingredients.
Space exploration and nuclear energy have a long history together.
Many of our most ambitious robotic missions into deep space—from the twin Voyager spacecraft hurtling through interstellar space to NASA’s new Mars Perseverance rover—have been powered by plutonium. But these robots don’t have a reactor on board. Instead, they use a radioisotope thermoelectric generator or RTG, which converts the heat from the natural decay of radioactive material into a steady drip of electricity for decades. Once the plutonium is inserted into the RTG, there’s no way to turn it off.
RTGs are fundamentally different from nuclear reactors, which require initiating and controlling a fission chain reaction in the core. This involves firing a neutron at a uranium atom and splitting it, thereby releasing more neutrons that split other uranium atoms and so on. The splitting of the atom releases heat, which can either be captured directly or used to heat a gas that spins an electric turbine. That’s the way nuclear reactors work on Earth, but it didn’t take physicists long to realize that the hot gas could also be blown out the back end of a reactor through a nozzle to create a rocket engine.
NASA’s flirtation with nuclear thermal propulsion predates the Apollo program, but the technology has been stymied by significant technological hurdles and fickle space exploration priorities. Since the end of the Apollo era, NASA’s human spaceflight program has been focused on low Earth orbit. Nuclear rockets are designed for crossing large distances quickly and in the cis-lunar environment, there wasn’t much need for all that extra power.
Still, space stations in low Earth orbit and a return to the moon are pathways to more ambitious crewed expeditions into deep space. And there are immense technical challenges to overcome before NASA can send humans to Mars. Those astronauts will face a litany of hardships on their excursion through the void, ranging from months spent in cramped crew quarters to exposure to deadly cosmic radiation. They would have a weeks-long launch window that only came around every two years and little opportunity to abort a mission once it was launched.
Scientists have acknowledged nuclear thermal propulsion as the solution to many of these problems since at least 1949. That year, two physicists published a series of papers in the Journal of the British Interplanetary Society detailing the use of “atomic rockets” for interplanetary travel. The physicists realized the key benefit of a nuclear thermal rocket is fuel efficiency. In fact, today, nuclear thermal rocket engines are calculated to be about twice as efficient as the Saturn V, the rocket that carried humans to the moon with what remains the most efficient rocket propulsion system ever made. That means it could cut the travel time to Mars in half and significantly reduce astronauts’ exposure to deadly cosmic radiation.
The idea of nuclear thermal rockets soon captured the attention of NASA. From 1959 until the end of the Apollo program, a small group of engineers at the agency worked on bringing a nuclear engine into the real world. The research program, called NERVA, resulted in more than 20 ground tests of a nuclear-powered rocket engine over the course of a decade. Even Werner von Braun, the director of Marshall Space Flight Center and architect of the Saturn V, became a strident advocate for a nuclear Mars rocket. The NERVA program revealed the potential for the incredible performance gains in a nuclear thermal propulsion system, but also the immense technical challenges that come with operating one.
There are very few materials that can withstand those temperatures.
The way a nuclear thermal propulsion system works is by pumping liquid hydrogen through a reactor core, bringing the gas up to around 4,400 degrees Fahrenheit, and then expelling the hot hydrogen through a nozzle to create thrust. The problem the NASA engineers faced was that the blistering hot gas eroded the components of the reactor core as it was operating, which quickly led to catastrophic failures. “It would disintegrate during operation,” says Hans Gougar, the manager of product engineering at X-energy.
But it’s not just a matter of finding materials that can take the heat. The nuclear reactor must be small enough to fit inside of a rocket. That’s a tall order given that the size of most nuclear reactors today are measured in acres. Even X-energy’s mobile version of the Xe-100, designed to fit in a shipping container, is too large for an orbital launch system.
To send a crew to Mars or beyond, it will have to be shrunk to the size of a large trash can, sustain temperatures several times higher than a conventional nuclear reactor, and maintain thrust for several hours to boost a spacecraft on its way to our neighboring planet. “To squeeze that much power out of a trash can-sized reactor means power densities and temperatures that would melt a traditional, land-based reactor into a slag heap,” says Gougar.
Fortunately, materials science has come a long way since NASA shuttered the NERVA program in 1973. Although there are still only a handful of materials that can withstand the high temperature and radiation environment in a nuclear thermal propulsion system, Gougar is confident that the materials X-energy is exploring for its nuclear thermal propulsion system will be able to handle the heat. “We think it will hold up,” Gougar says. “It’s not inexhaustible at those temperatures, there will still be erosion of the reactor core, but it will stay intact long enough for you to complete a mission.”
The push for a nuclear thermal rocket is heating up both in NASA and in Congress.
Over the past few years, Congress has earmarked hundreds of millions of dollars for NASA’s nuclear propulsion program and encouraged the agency to seek commercial partners that will lead the development of a nuclear rocket engine. NASA also enlisted a panel of experts from the National Academies of Sciences, Engineering and Medicine to study the feasibility of using a nuclear thermal rocket for a crewed Mars mission in the 2030s. After meeting more than a dozen times in 2020, the panel concluded that an “aggressive program” could develop a nuclear thermal propulsion system in time for an uncrewed Mars mission in 2033 followed by a crewed mission by 2039.
Moving quickly, NASA partnered with the Department of Energy to request proposals for preliminary nuclear thermal propulsion reactor designs. X-energy submitted a design and later this year, NASA and the DoE will select a few of these proposals for further study before narrowingly selecting one or two designs for a flight demonstration by 2026. That test will be the first time that a fission-powered propulsion system has ever been used in space.
X-energy’s nuclear thermal propulsion concept is still evolving, but it is fundamentally based on its commercial Xe-100 reactor. Both the terrestrial and the space version of the reactor will be fueled by TRISO-X particles, but the materials used in the spacefaring reactor will have to be modified to handle the extreme conditions required for a nuclear-powered rocket engine. Like NERVA, the company’s reactor core is graphite-based, but it will include advanced materials that make it more resistant to erosion. And whereas the terrestrial reactor is used to heat liquid helium, X-energy’s design uses liquid hydrogen propellant, which creates additional challenges because it must be stored cryogenically.
Gougar says one of the biggest challenges for X-energy or anyone trying to design a nuclear thermal propulsion system is cutting weight. The company’s commercial Xe-100 reactor puts out 200 MW of thermal power and weighs several hundred tons. But to meet NASA’s requirements, it will have to double its thermal power output while reducing its weight to under 2 tons. It’s a huge engineering feat that Gougar compares to transitioning from vacuum tube computers that were the size of a building to more powerful transistor-based computers that can fit in your pocket.
After years of working on a small advanced terrestrial nuclear reactor, Gougar and his colleagues at X-energy are well on their way to developing a nuclear thermal propulsion system. Today, most of their work is concentrated on using simulations to test their reactor design and doing some basic experiments on their TRISO-X fuel at their pilot facility in Oak Ridge, Tennessee. But if NASA selects X-energy’s design for the next phase of its nuclear thermal propulsion program, Gougar says the program will step into high gear.
“NASA is looking for a flight demonstration test in 3 or 4 years, which is rather different than traditional nuclear energy applications,” says Gougar. “It takes a long, long time to do a detailed design of a nuclear power plant.