Firing Up the Future: Nuclear Thermal Propulsion

After a half-century nap, nuclear propulsion is poised to fuel a paradigm shift in space exploration

Issue: 128. Subscribers: 58,657.

Humanity’s push toward Mars and other deep space destinations is driving a renewed focus on advanced propulsion systems. Current space missions are constrained by the limits of chemical propulsion, which require long transit times and vast quantities of propellant, making ambitious crewed Mars expeditions exceedingly challenging. To deliver heavier payloads faster and farther, space agencies consider nuclear-powered propulsion as a strategic game-changer. This technology promises to significantly cut travel times and enable larger mission architectures. NASA Administrator Bill Nelson noted that with nuclear propulsion, “astronauts could journey to and from deep space faster than ever – a major capability to prepare for crewed missions to Mars”. Beyond crew transport, both NASA and the U.S. Space Force see nuclear propulsion as vital for an Earth–Moon economy and national security, enabling rapid repositioning of assets in cislunar space and “maneuvering without regret” (i.e., without running out of fuel) in Earth orbit. In short, advanced propulsion is becoming a strategic imperative for the next wave of exploration and commercialization of space.

Specific Impulse Challenge

Today’s launch vehicles and deep-space craft almost universally rely on chemical propulsion – burning fuel with an oxidizer to produce thrust. While chemical rockets deliver high thrust, they suffer from low efficiency in terms of propellant usage. This efficiency is measured by specific impulse I_sp, the impulse (thrust * time) produced per unit of propellant. The best chemical engines (like liquid hydrogen/liquid oxygen engines) achieve about 465 seconds of I_sp. This relatively modest efficiency means missions need enormous fuel fractions; for a high-energy Mars transfer using only chemical propulsion, upwards of 90% of the spacecraft’s mass might be propellant (leaving little room for payload).

Even super-heavy chemical rockets like NASA’s SLS or SpaceX’s Starship would require multiple launches and orbital refueling to send a sizeable crewed vehicle to Mars using conventional propulsion. Travel times with chemical propulsion are lengthy, typically nine months to reach Mars under favorable planetary alignments, or, at best, nearly 80 days for projected systems like Starship, under the condition that ~90% of weight is propellant. Such long voyages expose crews to prolonged microgravity and cosmic radiation. Clearly, to go further and faster – and to do so with meaningful payloads – we need propulsion with much higher I_sp than chemical rockets can provide.

Nuclear Propulsion: Thermal vs. Electric Approaches

To harness nuclear energy for space travel, two primary approaches exist: nuclear thermal propulsion (NTP) and nuclear electric propulsion (NEP). Both utilize a fission reactor as the power source, but in fundamentally different ways:

Nuclear Thermal Propulsion (NTP): In an NTP rocket, a compact reactor’s fission heat is applied directly to a propellant. The reactor core heats the hydrogen to extreme temperatures (~2,500–3,000 K, much higher than in the terrestrial reactors), causing it to expand out of a rocket nozzle to produce thrust. The higher the temperature is, the faster the rocket jet will be, causing higher I_sp. No combustion is involved – the propellant is heated by the reactor and expelled by the nozzle. If the exhaust molecular weight is as low as 2 (hydrogen), the specific impulse of NTP can reach ~1000 s or more – roughly twice the efficiency of the best chemical rockets, and potentially “three or more times” in advanced designs. This translates to using far less propellant for the same speed; for example, a nuclear stage might need on the order of 60% of its mass in hydrogen propellant for a Mars transfer, versus 90% for chemical – a dramatic improvement. NTP also provides high thrust, on the order of what chemical upper-stage engines deliver. This high thrust allows quick burns for orbital insertion and departure maneuvers. The trade-off, however, is that NTP systems are complex nuclear reactors operating at very high temperatures, presenting engineering challenges in materials and safety. The core motivation for exploring NTP: it offers a way to enable higher exhaust velocities without sacrificing thrust.

Nuclear Electric Propulsion (NEP): In NEP systems, a fission reactor doesn’t heat a propellant directly but instead produces electricity (via thermoelectric converters, turbines, or other power conversion). That electrical power then drives high-efficiency electric thrusters – for instance, ion engines or Hall-effect thrusters – which expel an ionized gas (like xenon) to produce thrust. Electric thrusters have exceptional I_sp, often 3,000–5,000 seconds and potentially up to ~10,000 seconds in some concepts. This means minuscule propellant consumption: a well-designed NEP transfer stage might use only a fraction of the propellant that a chemical or NTP stage would for the same mission. The drawback is low thrust, measured in Newtons or millinewtons (it is less than the weight of a 100 gramms of chocolate bar in your hand) rather than the meganewton-level thrust of launch rockets. NEP systems accelerate a spacecraft slowly but continuously over long durations. They are ideal for moving cargo or performing ultra-efficient deep-space missions (for example, scientific probes to outer planets) where trip time is less critical. A nuclear-electric ship could spiral out from Earth and eventually achieve very high speeds, all while using propellant sparingly. However, for crewed missions that demand timely transit, pure NEP is less practical due to the slow acceleration – one wouldn’t want astronauts thrusting for months on end just to escape Earth’s gravity. In this post, we will focus on NTP. Read our next issue to learn more about recent developments of NEP systems.

Artist’s concept of a transfer vehicle powered by a solid-core NTP stage (left, with large hydrogen tanks). Credit: NASA.

NTP Demonstration Missions and Funding Milestones

After decades of research, nuclear propulsion is now moving from paper studies to flight hardware. One of the first major initiatives in NTP was the Nuclear Engine for Rocket Vehicle Application (NERVA) program, jointly led by NASA and the Atomic Energy Commission (AEC) starting from 1961, and RD-0410, a Soviet nuclear thermal rocket engine developed by the Chemical Automatics Design Bureau from 1965 through the 1980s. The goal was to create a nuclear-powered upper-stage rocket engine. Throughout the programs, multiple nuclear reactors were designed and tested for integration into a propulsion system. Although NERVA’s engine and RD-0410 were never deployed in space, the projects are widely regarded as a successful demonstration of the feasibility of nuclear thermal propulsion.

Today's flagship effort is the Demonstration Rocket for Agile Cislunar Operations (DRACO) program, a joint US$499 million project by DARPA and NASA to flight-test a nuclear thermal rocket in Earth orbit by 2027 (but that launch date has been put on hold in May 2025 by nuclear reactor test requirements). DARPA awarded the DRACO spacecraft contract to Lockheed Martin, with BWX Technologies (BWXT) designing and fabricating the nuclear reactor that will serve as the engine’s core. The reactor will use HALEU (High-Assay Low-Enriched Uranium) fuel, which offers high performance while keeping enrichment levels around 20% for safety and non-proliferation reasons. Importantly, if not cancelled or delayed, DRACO will be a full in-space demonstration: a United Launch Alliance Vulcan rocket is planned to launch it to orbit in 2027. Once in orbit, the DRACO vehicle will fire up its nuclear thermal engine – the first time the U.S. has activated a nuclear rocket in space.

Artist’s concept of the DRACO nuclear thermal stage in orbit. Credit: Lockheed Martin

DRACO is backed by significant government investment as a pathfinder. Beyond DARPA’s funding, NASA’s Space Technology Mission Directorate is contributing expertise and resources to the nuclear engine development. The Department of Energy (DOE), which oversees nuclear materials and safety, is heavily involved as well – DOE’s Idaho National Lab is managing contracts for reactor design and will supply the enriched uranium fuel.

In 2021, NASA and DOE jointly awarded $5 million design contracts to three teams to start developing NTP reactors. The teams included major and emerging players: BWXT (partnership with Lockheed Martin), General Atomics (partnership with X-energy and Aerojet Rocketdyne), and Ultra Safe Nuclear Technologies (partnered with Ultra Safe Nuclear Corp., Blue Origin, GE Hitachi, and others). This public-private approach is nurturing a base of suppliers and innovative reactor concepts. Those initial studies have since progressed – by 2023, two of the companies (Ultra Safe and General Atomics) had their contracts extended for further development, and as of 2025, hardware testing is underway on prototype fuel elements and components. For instance, NASA reports that reactor fuel samples have been exposed to hot hydrogen flow at Marshall Space Flight Center to simulate NTP engine conditions, helping validate material choices.

Overcoming Technical and Economic Challenges

For all its promise, nuclear propulsion faces significant technical and economic hurdles:

Reactor safety and launch approval: regulators and the public need assurance that a launch accident wouldn’t spread radioactive materials. To mitigate this, nuclear propulsion systems are designed to remain completely cold (inactive) until reaching space. In the event of a launch vehicle failure, the reactor stays sub-critical and no fission products are generated. These safety measures, combined with trajectories that place nuclear stages in disposal orbits where they will remain for centuries, aim to make the risk low, but the perception issue remains.

Hydrogen storage: NTP rockets mainly plan to use liquid hydrogen propellant – the lightest, highest-performing propellant possible. Using any heavier gas at the same reactor temperature would mean that the exhaust velocity would be much lower, and hence the specific impulse I_sp and propellant mass efficiency. But liquid hydrogen must be kept at cryogenic temperatures (20 K) and is notoriously difficult to store for long durations. It boils off over time, and the tanks require insulation and possibly active cooling. A crewed Mars mission might need hydrogen to remain in a liquid state for many months in space, so zero-boil-off tank technology is a critical area of research. NASA and industry are investigating advanced insulations, cryocoolers, and alternative storage methods to reduce propellant loss. Another angle is choosing mission profiles that minimize coast time – for example, departing Earth shortly after fueling and using the hydrogen before it boils off. Solving in-space hydrogen storage is key to making NTP practical and is an active area of innovation (with spinoffs possible in terrestrial liquified gas handling tech). There are some proposals to use liquid ammonia, though - a loss in I_sp at a cost of fewer problems with propellant storage.

Materials and fuel durability: running a reactor at extreme temperatures and high power density pushes the limits of nuclear materials. In a solid-core NTP, the reactor outlet temperature can exceed 3000 K (around 2700 °C), far hotter than normal power reactors, where turbines do not require high jet speeds, and hence high temperature. At these temperatures, nuclear fuel can crack, and reactor materials can weaken or corrode. Additionally, reactor endurance is a factor – a crewed mission might need the reactor to start, stop, and restart multiple times and run for hours total. Ensuring that the reactor can go through thermal cycles without damage is another aspect being tackled through ground testing. The bottom line is that materials science is catching up, but it remains a schedule and funding risk – unexpected materials problems could slow development. This is why NASA and DOE have ramped up testing campaigns now, before committing to a crewed mission reactor design.

Despite these challenges, the trendline is positive. Modern advances in simulation, materials science, and nuclear engineering (from molten salt reactors to 3D-printed fuels) are being brought to bear on these problems.

Opportunities in a Nuclear-Powered Space

After a half-century nap, nuclear propulsion is poised to fuel a paradigm shift in space exploration. If the first test reactors light up in orbit in the next few years, we will be witnessing the beginning of a new era – one in which voyages to Mars and beyond become faster, safer, and more achievable than ever before. And for the blooming space economy, it could mean freight trains to the Moon and high-speed routes between Earth orbits – the infrastructure needed for space industrialization.

The revival of nuclear space technology isn’t just a government endeavor – it’s also drawing in startups and venture capital. Investors are recognizing that if nuclear propulsion can deliver on its promises, it will unlock new markets in space. Several startups have already entered the contest. Atomos Space, a Denver-based startup (raised $23.4M, now acquired by Katalyst Space Technologies), is developing orbital transfer vehicles (space tugs) and has publicly discussed plans for an NTP-powered tug to deliver satellites to high orbits. In the near term, Atomos will likely use solar-electric propulsion until nuclear reactors are flight-ready, but the company’s long-term roadmap sees nuclear propulsion as the key to moving large payloads economically (for example, repositioning big geostationary satellites or ferrying modules to a lunar gateway). Another player, Ultra Safe Nuclear Corporation (USNC, raised $17M, filed for bankruptcy in 2024 to continue operations under Standard Nuclear’s ownership) – originally a terrestrial micro-reactor company – spun off a division focused on “astronuclear” systems and, as noted, won NASA contracts to design NTP reactors.

In the coming decade, milestones like a possible DRACO 2027 flight, if not cancelled by the current US administration, will likely trigger increased funding for a first generation of operational nuclear stages – perhaps powering NASA’s first crewed Mars transfer vehicle in the 2030s, or being adopted by Space Force for orbital transfer vehicles. Each success will build confidence and spur competition (both between nations and between companies) to push nuclear propulsion further. Just as the once-impossible dream of reusable rockets is now a commercial reality, the notion of routine nuclear-powered space travel may be only years away from realization.