Nuclear propulsion drives a spacecraft with heat from a nuclear reactor instead of the chemistry of burning fuel. It comes in two forms. Nuclear thermal propulsion uses the reactor to superheat a light propellant and shoot it out a nozzle, giving strong thrust at about twice the efficiency of a chemical engine. Nuclear electric propulsion uses the reactor to generate electricity that runs efficient electric thrusters. Both aim at the same prize: moving heavy payloads through deep space on far less propellant than a chemical rocket needs.
How Nuclear Thermal Propulsion Works
Nuclear thermal propulsion (NTP) works by pumping a propellant through a hot reactor and letting it expand out a nozzle. The reactor core reaches thousands of degrees. A propellant, usually liquid hydrogen, flows through channels in that core, absorbs the heat, and turns into a fast-moving gas that escapes through the nozzle to make thrust. No burning happens. The reactor supplies the energy that combustion would supply in a chemical engine.
The reason this matters is efficiency. An engine’s efficiency is measured by specific impulse (Isp), the velocity it extracts from each kilogram of propellant. A good chemical engine reaches around 450 seconds. A nuclear thermal engine can roughly double that, because hydrogen is very light and the reactor heats it to a temperature that combustion alone would struggle to reach. Doubling the efficiency means a ship can carry far less propellant for the same trip, or push a much heavier payload with the same tanks.
Nuclear thermal keeps the high thrust of a chemical rocket while adding that efficiency. It can fire a strong burn to push a heavy ship onto a fast path toward another planet, then shut down. That combination of real thrust and high efficiency is what makes it attractive for crewed missions, where the ship is heavy and the trip should be short.
Nuclear Electric Propulsion, the Slower Cousin
Nuclear electric propulsion (NEP) trades thrust for even greater efficiency. Instead of heating propellant directly, the reactor generates electricity, and that electricity runs ion or Hall thrusters. Those electric thrusters reach a specific impulse many times higher than any nuclear thermal design, so they use astonishingly little propellant.
The catch is the same one every electric engine faces. Electric thrusters produce very little force, so a nuclear electric ship accelerates gently over months. It cannot deliver the hard, fast push a crewed departure wants. That makes nuclear electric propulsion a better fit for cargo tugs and robotic missions that can afford a slow spiral outward, while nuclear thermal suits the crewed leg where time in transit is a health cost. Some long-range mission concepts pair the two, using thermal for the fast crewed push and electric for pre-positioned cargo.
| Type | Thrust | Efficiency | Best job |
|---|---|---|---|
| Nuclear thermal (NTP) | High | About twice chemical | Fast crewed departure |
| Nuclear electric (NEP) | Very low | Many times chemical | Slow cargo and robotic runs |
| Chemical (for comparison) | Highest | Baseline | Launch from the ground |
Why Nuclear Propulsion Suits a Crewed Mars Trip
A trip to Mars is long, and the length itself is a danger to a crew. Every extra month in deep space exposes astronauts to more cosmic radiation and more time in weightlessness, which weakens bone and muscle. Cutting the transit time directly cuts those risks, and a more efficient engine is what buys a shorter transit.
A chemical rocket has to carry enormous propellant to move a crewed ship and its supplies, and most of that mass is fuel to move fuel. A nuclear thermal engine, roughly twice as efficient, breaks that spiral. It can push the same heavy ship onto a faster trajectory while burning much less propellant, which shortens the journey and shrinks the vehicle at the same time. For a mission carrying people, life support, and a return stage, that efficiency turns an awkward, oversized rocket into a plausible one.
The gain is real but bounded. Nuclear propulsion does not make Mars a weekend trip. It trims a journey of many months by a meaningful fraction and reduces the propellant bill, which is enough to change the mission’s shape without rewriting physics.
The DRACO Program and Where Things Stand
The leading flight effort is a demonstration engine called DRACO, and its purpose is to prove the concept in orbit. DRACO, short for Demonstration Rocket for Agile Cislunar Operations, is a joint program run by DARPA and NASA to build and fly a nuclear thermal rocket engine as a test. Lockheed Martin is the prime contractor developing the experimental vehicle, working with partners on the reactor. As of 2026 the program is in development toward an in-space flight demonstration. Schedules for such first-of-a-kind projects have moved before, so treat any target date as provisional and check the program’s own updates.
Nuclear thermal engines are not a blank-sheet idea. The United States ran a program called NERVA in the 1960s and early 1970s that built and ground-tested nuclear thermal engines, firing them on test stands and proving the basic design worked before the effort was cancelled without ever flying one. That earlier work is why engineers describe nuclear thermal as a known concept awaiting a modern flight demonstration rather than an untried gamble.
One thing has changed since NERVA: the fuel. Modern designs like DRACO run on high-assay low-enriched uranium, known as HALEU, which is enriched well below weapons grade. That switch eases the security and proliferation concerns that surround reactor fuel, and it lets a wider set of suppliers and facilities handle the material. The tradeoff is that the lower enrichment makes the reactor harder to keep small and hot, so engineers accept some performance cost in exchange for a fuel that is easier to license and launch.
Safety shapes how the reactor is handled. The plan is to launch the reactor cold and start it only after the vehicle reaches a sufficiently high orbit. An unstarted reactor holds very little radioactivity, so a launch accident would not spread meaningful contamination, and switching it on only in space keeps the fission hazard far from the ground. This launch-cold, start-high approach is central to why regulators treat a demonstration flight as manageable.
Nuclear propulsion sits alongside the two other engine families worth understanding: the chemical engines in our guide to how a rocket engine works, which still do the launching, and the ion propulsion that nuclear electric designs power on a larger scale. To follow whether nuclear propulsion moves from demonstration toward crewed use, watch the DRACO program’s flight test and the reactor-fuel work behind it.