Powering Up in Space: Is Nuclear the Answer?


Speed, power, and response — these factors decide success and failure in space. Players who want to lead in space have to push the envelope, and maybe even take a few longshots. At the Defense Innovation Unit, we believe that compact nuclear power will get us there in space.

On paper, the United States should be light-years ahead of other nations in nuclear space tech. Six decades ago, America launched a nuclear reactor into space (it’s still up there), and the nation has since spent more than $15 billion on a dozen government programs to develop a nuclear space capability, without a single launch. Meanwhile, Russia is building a nuclear space tug, and China has announced a nuclear system 100 times more powerful than current U.S. designs. And while these claims may oversell the technical reality, those in the field have to ask: Is the United States still in the lead?



Programs currently in the works at the Defense Advanced Research Project Agency (DARPA) and NASA promise to launch fission-powered nuclear thermal propulsion before the end of the decade. These worthwhile efforts will lead to spacecraft with two to three times more maneuverability than traditional chemical propellants. Using the nuclear core to heat hydrogen gas, nuclear thermal propulsion allows for responsive in-space maneuvers by maintaining a high thrust-to-weight ratio. In addition to nuclear thermal propulsion, NASA is also researching a fission reactor to power electric-propulsion systems (Nuclear Electric Propulsion), which could generate even greater capability for future missions to Mars and other interplanetary missions.

The drawback to these fission reactors is scale, in both size and weight. When you include the fuel, moderator, shielding, power conversion, and radiators, the smallest fission reactor is still pretty heavy. As the Department of Defense continues to source smaller and disaggregated spacecraft, physics is pushing us to find alternative solutions (that is, not fission) for nuclear propulsion and power. While NASA and DARPA are working on these traditional nuclear fission approaches, the Defense Innovation Unit is supporting non-traditional and non-fission approaches to nuclear.

As a program manager at the Defense Innovation Unit, I’m leading the Department of Defense’s effort to build prototypes of these novel nuclear power and propulsion systems for small spacecraft. This work will have a direct impact on how the United States employs spacepower, ushering in an era in which spacecraft maneuver tactically in cislunar space. If the Department of Defense wants starcruiser-like spacecraft before the end of the decade, America needs a smaller, faster, and safer approach to nuclear. In a volume nearly 2,000 times larger than geostationary orbit, cislunar space requires Department of Defense spacecraft with advanced maneuver and power capability that could help enforce “norms of behavior” and commercial activities in this new domain.

The good news is that commercially developed concepts that may fit the bill already exist — U.S. companies are spearheading the development of next generation radioisotopes and compact fusion reactors that could enable big improvements in maneuverability over current Department of Defense space platforms (e.g. X-37B). Let’s review these nuclear options, the hurdles they face, and the future they may enable.


The approach is straightforward: Radioactive materials undergo nuclear decay, producing heat that can be converted into electricity. This electrical power can run spacecraft sensors, communications, and electric propulsion systems (e.g., ion drives). Radioisotope power systems have been around since the early days of the space age, and plutonium-238, with its consistent heat output and low gamma/neutron emission, is still the preferred source. Despite the expense and scarcity, plutonium-238 radioisotope sources continue to power experiments and payloads on the moon and Mars.

With a half-life of 88 years, plutonium-238 can produce sustained power for decades — proven through its use on the Voyager interstellar probes, which are still communicating with Earth nearly half a century after their launch. However, the leading radioisotope power system is a microwave-sized device providing roughly 100 watts of electrical power at somewhat low efficiency (around 5 percent). At around 2 watts per kilogram, these units are too heavy and produce too little power to be useful for propulsion on future Department of Defense satellites where much shorter timelines are at play.

If plutonium is expensive, scarce, and lacks necessary power density, could shorter half-life radioisotopes be a better option? Could higher-performance radioisotope sources feasibly power both sensor payloads and electric propulsion systems?

Cobalt, europium, and strontium could be those sources. Policy updates from the White House (e.g., Space Policy Directive-6 and National Security Presidential Memorandum-20) and pending regulatory guidance from the Federal Aviation Administration have opened a pathway for commercial entities to obtain launch and operational licenses for these radiological materials. From a launch-safety standpoint, a 100-watt plutonium-238 radioisotope source is in the same regulatory category as a 27,000-watt europium source or a 17,000-watt cobalt source. These shorter half-life (5 to 15 years) radioisotopes could achieve energy density 30 times higher than plutonium — up to several hundred watts per kilogram.

One path towards high power (more than 1,000 watts) radioisotope power sources is being developed at USNC-Tech, a Seattle-based company, with funding from NASA, where the technology will be used to rendezvous with the first known interstellar object, ‘Oumuamua, currently speeding away from Earth at roughly 30 kilometers per second. Such a staggering power system would not only outperform plutonium-238, but also offers power density at least 10 times higher than a similar-sized fission reactor power system, and could be ready years before the first fission systems. Companies developing these new radioisotope power systems have their work cut out — they will have to work out new irradiation schemes, novel encapsulation techniques, shielding and remote handling, and power conversion challenges, but the payoff could be huge.

Fusion: No Longer 30 Years Away?

Building a compact fusion reactor in your garage is possible. The problem is getting more energy out of it than you use to run it. This ratio of energy out to energy in is called the Q-factor. To date, a fusion reactor with a Q-factor greater than one has not been built, although there are dozens of fusion startups, a fledgling industry association, and persisting hope that fusion is within grasp. The closest anyone has come is a Q-factor of 0.33 for 5 seconds, achieved at the Joint European Tokamak, per a report published this year.

If nuclear fusion is right around the corner, how might fusion reactors be used in space? Let’s take a look at our options.

Magnetic Confinement Fusion

The world-record Joint European Tokamak fusion reactor uses magnetic coils to confine hot plasma in a donut-shaped device (tokamak). This approach, called magnetic confinement, has been under development from the very first days of fusion.

Achieving a Q-factor greater than one using magnetic fusion requires massive plasma volumes surrounded by cryogenically cooled superconducting electromagnets that are the size of buildings. The most expensive science experiment in human history, the International Thermonuclear Experimental Reactor (ITER) is expected to achieve a Q-factor of more than 10, but won’t be completed until 2035. Still, it is possible that other magnetic fusion devices (e.g., SPARC), taking advantage of new superconductor materials, could be producing carbon-free terrestrial electrical power in the coming decade. These, however, will not work very well in space — a reasonably sized spacecraft just won’t be able to support the hundreds of tons of magnets needed for magnetic confinement fusion. Bottom line: Magnetic confinement fusion will be great for Earth, but too heavy for space.

Inertial Confinement Fusion

Another approach to fusion relies on squeezing atoms together until they fuse, called inertial confinement. The United States first successfully demonstrated inertial confinement nuclear fusion during the Operation Greenhouse weapon test in 1951 on the Enewetok atoll in the Pacific. But for our purposes, thermonuclear weapons don’t make very good rockets (both NASA and the U.S. Air Force have tried). With the signing of nuclear test-ban treaties and the advent of the laser in the 1960s, scientists began looking into using photons rather than nuclear explosions to squeeze hydrogen atoms together and reach fusion ignition. This technique has been honed at the Department of Energy’s National Ignition Facility, where 192 lasers, together the size of three football fields, are focused onto a fusion target the size of a pencil eraser in a powerful pulse. In these few nanoseconds, the lasers take up 500 times the entire energy production of the United States — proving that squeezing atoms together using light is extremely difficult. While the physics is close (the facility reached a Q-factor of 0.7 recently), engineering a spacecraft to carry the pulsed laser power infrastructure remains infeasible, or leads to designs that are ridiculously large and expensive.

Electrostatic Confinement

Electrostatic confinement is perhaps the longest-running and most underperforming of the fusion concepts, having received little serious attention since being patented by Philo T. Farnsworth in the 1960s. In electrostatic fusion, electrodes cause ions to accelerate toward a central reactor core volume where they collide with other ions and can fuse together. This method offers a fusion device that doesn’t require house-sized magnets, lasers, or capacitor banks. An electrostatic fusion reactor would be ultra-lightweight, however, pure electrostatic fusion devices have never reached a Q-factor of more than 1 because of a fundamental physics limit: collisions between ions cause losses in confinement much faster than collisions that lead to fusion reactions. Bottom line here: light enough to actually launch into space, but needs some serious physics breakthrough to overcome fundamental limits.

Hybrid Confinement

What’s becoming clear is that a combination of plasma-confinement approaches will be required to build compact-enough spacecraft propulsion and power engines. In recent years, billions of dollars in private capital has poured into these hybrid approaches. Magneto-inertial confinement fusion devices (e.g., General Fusion) start with a low-density magnetized plasma before using a “liner” to compress to fusion ignition conditions. Another promising hybrid approach involves using the plasma fuel itself to generate confining magnetic fields (akin to a self-sustaining smoke ring) while slamming these plasmas into each other (as, for example, Helion is attempting to do) to achieve fusion ignition. An important characteristic of these new devices is that they are small. Avalanche Energy is currently working on a hybrid electrostatic/magnetic confinement concept that could lead to a “hand-held” fusion reactor. At these more compact scales, putting a fusion reactor onto a spacecraft is more science than fiction. The bottom line with hybrid approaches: The physics is still less understood, but a hybrid-confinement fusion reactor may actually be light enough to launch into space.

So where are we on the road to putting fusion reactors on Department of Defense spacecraft? Despite all of the challenges of building things for space, there is one advantage that a space fusion reactor has over terrestrial fusion reactors: The bar is high for fusion to provide commercial terrestrial electricity (a fusion power plant may need a Q-factor of over 50 to be profitable). However, for spacecraft propulsion and power, a Q factor of around two could still be useful because there are fewer energy transformation and transportation steps. Such enabling commercial technologies would be extremely valuable for Department of Defense spacecraft power and propulsion in the near term — something worth taking a risk on.

What’s Next?

The Defense Innovation Unit is focusing on two approaches to accelerate toward ground and flight-testing prototypes: compact fusion and next-gen radioisotope concepts that are likely to exceed the performance of fission reactor power systems for small satellites, with the goal of an orbital prototype demonstration in 2027. This approach is not without risk, both technical and programmatic: Fusion that generates more power than it consumes (a Q-factor of more than 1) has to be demonstrated; manufacturing pathways for high-power radioisotopes should be formed, and, most importantly, both industry and the Department of Defense should assure public safety by working hand-in-hand with regulatory and licensing agencies. These are not easy tasks. In fact, many in the fission, fusion, and space industries will see these approaches as true longshots, but America cannot innovate without taking risks on new technologies. This is the way.



Ryan Weed is leading the Nuclear Advanced Propulsion and Power program at the Defense Innovation Unit as a program manager in the space portfolio. Ryan is a Ph.D. physicist and U.S. Air Force experimental test pilot, logging over 2,000 hours in more than 30 different aircraft. As a NASA Innovative Advanced Concepts Fellow, he has studied radioisotope positron propulsion systems. While at Blue Origin, Ryan designed and implemented an Instrumentation Laboratory for cryogenic rocket fuels. As founder of Positron Dynamics, he has designed and built a positron beamline facility, and developed high-specific impulse propulsion concepts.

Image: Wikimedia