On February 24, 2026, engineers at NASA’s Jet Propulsion Laboratory fed liquid lithium into a prototype plasma thruster, cranked the power to 120 kilowatts, and watched tungsten electrodes glow past 5,000 degrees Fahrenheit. They did it five times. Nothing melted. Nothing failed. And that, for a technology that has been stuck in concept papers since the Eisenhower administration, counts as a breakthrough.
The test took place inside JPL’s Condensable Metal Propellant Vacuum Facility, known as CoMeT, a 26-foot water-cooled vacuum chamber designed specifically for liquid-metal thruster experiments. According to JPL’s account of the session, the five ignitions confirmed that the lithium-fed magnetoplasmadynamic (MPD) design can survive the extreme thermal stress that a deep-space mission would impose. JPL has described CoMeT as the only large-volume vacuum chamber in the United States currently configured to test megawatt-class, liquid-metal-fueled thrusters, though that claim is difficult to independently verify. The chamber’s walls condense and recover lithium vapor after each firing, allowing engineers to run repeated high-power tests without contaminating the facility.
The goal behind the hardware is ambitious: pair clusters of these thrusters with a nuclear reactor powerful enough to drive them for years, and you could slash the travel time to Mars well below what chemical rockets can manage. Where a conventional mission might keep astronauts in transit for seven to nine months each way, nuclear electric propulsion, or NEP, could shorten that window significantly by sustaining continuous, efficient thrust over the entire journey.
Why lithium, and why now
Most electric thrusters flying today, including the Hall-effect engines on NASA’s Psyche asteroid mission, run on xenon gas. Xenon works well at low power levels, but it is heavy, expensive, and poorly suited to the megawatt-class systems a crewed Mars vehicle would need. Lithium offers a different set of properties: it is the lightest metal on the periodic table, it ionizes easily, and it can be stored as a compact liquid before being vaporized and accelerated by powerful electromagnetic fields. Those characteristics make it a strong candidate for MPD thrusters, which channel very high current densities to produce both high exhaust velocity and meaningful thrust.
NASA’s overview of the project describes the February test as a stepping stone toward individual thrusters operating at 500 kilowatts to 1 megawatt. A reference architecture published in a 2021 study on the NASA Technical Reports Server outlined opposition-class Mars mission profiles that would require 2 to 4 megawatts of electrical power directed into propulsion for more than 23,000 operating hours. Those figures represent a design target from that study, not a formal 2026 program requirement. Even at the upper end of the per-thruster target, that means multiple units firing in concert on a spacecraft bus that can distribute power and shed waste heat at levels no existing flight system has achieved.
A long history of false starts
Nuclear electric propulsion is not a new idea. Engineers first proposed it in the late 1950s, and variations have surfaced in programs such as SNAP, the Space Exploration Initiative of the early 1990s, and Project Prometheus in the early 2000s. Each effort was eventually shelved or sharply reduced, undone by budget cuts, shifting political priorities, or technical risks that proved harder to solve than anticipated.
NASA’s own planning documents acknowledge that key NEP technologies remain at Technology Readiness Level 4 or below, meaning they have been validated in laboratory conditions but not yet in environments that simulate actual spaceflight. The agency’s Space Nuclear Propulsion roadmap aims to push megawatt-class hardware to TRL 5, where components are tested in simulated operational settings, but no firm public timeline exists for that milestone. The February thruster test moves one piece of the puzzle forward, but the broader system remains in its early stages.
The engineering gaps that remain
Surviving five firings at 120 kilowatts is not the same as running reliably for thousands of hours at a megawatt. NASA has not yet released detailed efficiency metrics from the February session. Specific impulse, thrust output in newtons, and lithium consumption rates are all absent from the public record. Without those numbers, independent analysts cannot compare the lithium-fed design against competing electric propulsion concepts on a performance-per-kilogram basis.
The thruster is also only one component in a much larger engineering challenge. No flight-qualified nuclear reactor currently exists that can deliver 2 to 4 megawatts of electrical power in space. Recent NASA and Department of Energy investments have focused on fission surface power for the Moon, targeting tens of kilowatts to support lunar habitats, not the megawatt output a Mars transit vehicle would demand. Scaling from kilowatts to megawatts involves higher core power densities, more robust radiation shielding, more complex heat transport loops, and new launch safety protocols.
Thermal management may be the most stubborn problem of all. A megawatt-class reactor generates enormous waste heat that must be radiated away in the vacuum of space. Designing radiators with enough surface area, low enough mass, and high enough reliability to function for years is an unsolved challenge. A thruster that performs well inside a water-cooled chamber on Earth does not automatically translate to one that stays stable at the end of a long boom, connected to flexible power lines and radiator panels, millions of miles from any repair crew.
Crew safety adds another layer. A nuclear reactor on a spacecraft means radiation shielding between the power source and the habitat module, and mission planners must account for scenarios like partial thruster failure or cooling loop anomalies while astronauts are aboard. None of these integration-level questions were addressed by the February test.
How it fits alongside competing approaches
NASA is not betting on a single path to Mars. The agency’s DRACO program, a collaboration with the Defense Advanced Research Projects Agency, is developing a nuclear thermal rocket that heats hydrogen propellant directly with a fission reactor. Nuclear thermal propulsion offers higher thrust than electric systems but lower fuel efficiency, making it better suited for shorter, more powerful burns rather than the long, steady acceleration that NEP provides. The two technologies could eventually complement each other in different mission phases.
Meanwhile, SpaceX continues to develop Starship as a fully reusable chemical rocket capable of Mars transit through orbital refueling. That approach sidesteps the nuclear question entirely but accepts longer trip times and the physiological toll they impose on crews, including muscle atrophy, bone density loss, and elevated radiation exposure from cosmic rays during months in deep space. A faster NEP transit would reduce that exposure window, which is one of the strongest arguments in the technology’s favor.
What the February 24 firing actually proved
The February 24 firing is best understood as a proof of principle. It confirmed that a lithium-fed MPD thruster can be pushed into the high-power regime without immediate catastrophic failure, and it validated JPL’s CoMeT facility as a viable testbed for future runs at even higher power levels. Those are real, measurable results from a primary source: the engineering team that built the hardware and collected the data.
But the test does not, by itself, guarantee faster crewed missions to Mars or a specific launch date for an NEP-powered spacecraft. The gap between a laboratory demonstration and a flight-ready system capable of running for years remains vast. Closing it will require parallel advances in reactor design, thermal management, systems integration, and long-duration reliability testing, each carrying its own schedule and budget risks.
What changed in February is that one of the hardest questions in nuclear electric propulsion, whether a lithium-fed thruster can physically handle the thermal and electrical punishment that a Mars mission would demand, now has a preliminary answer. The next questions are harder, and more expensive. But for the first time in decades, the hardware is catching up to the theory.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
