NASA Bets $2 Billion on Nuclear Spacecraft SR-1 for 2028 Mars Launch

NASA has committed an estimated $2 billion to Space Reactor-1 Freedom, the first spacecraft powered by a nuclear reactor, with an uncrewed Mars mission scheduled to launch by the end of 2028 and arrive in 2029. The agency announced the plan just before the Artemis II crew began its lunar flyby this past April, and it says the Idaho National Laboratory, which is building the reactor, remains on target for the late-2028 date.

"Procurement of major components is underway, and integration work on the power and propulsion element is advancing steadily," Steve Sinacore, program director for SR-1 at NASA, told MIT Technology Review.

How does the propulsion actually work?

SR-1 will use nuclear electric propulsion. A self-sustaining uranium-fission reaction produces heat and ultimately electricity; that electricity ionizes a gas, which magnetic fields then blast out of the spacecraft to generate thrust. The approach trades raw power for endurance.

"It's very low thrust, but very efficient," said Sebastian Corbisiero, the US Department of Energy's national technical director of space reactor programs. "So you can use it for a long period of time."

Concept art from Sinacore's presentations shows the craft shaped like a colossal fletched arrow. The rear houses the power-and-propulsion system; the tip carries a uranium-filled reactor rated at a capacity of at least 20 kilowatts — roughly one fifty-thousandth of a typical Earth-based nuclear plant's output. Large fins along the body radiate reactor heat into the vacuum of space.

If the mission works as planned, the journey to Mars will take about a year.

What stands in the way?

The United States has started and terminated several nuclear space propulsion programs over the past 75 years. The most recent casualty was DRACO, a NASA–Department of Defense collaboration canceled in 2025. DARPA attributed the cancellation to high experimentation costs, an inability to compete with conventional rocket propulsion on price, and difficulty conducting ground tests safely and effectively.

Engineering researchers see no fundamental barrier this time. "I think it's a very doable technology," said Philip Metzger of the Florida Space Institute. "I'm happy to see them finally doing this." The Soviet Union, for its part, flew dozens of reactors into orbit, mostly to power spy satellites, so reactor operations in space have precedent.

The launch itself remains a key engineering risk. "Going through the launch safely is going to be a challenge," said Simon Middleburgh, co-director of the Nuclear Futures Institute at Bangor University in Wales. "You are being shaken, rattled, and rolled." Engineers will switch the reactor on only about two days after launch, once the craft is far from Earth, because uranium is relatively benign on its own while reactor waste products are not.

Why the rush?

The schedule is tight by common assessment. "It's an aggressive timeline," said Lindsey Holmes, vice president of advanced projects at Analytical Mechanics Associates, an aerospace company in Virginia. She suspects the push reflects, in part, China's and Russia's deep-space nuclear ambitions: the two countries have discussed placing a reactor on the lunar surface in the mid-2030s to power a jointly operated base.

The overlap matters for portfolio planning. SR-1's technology is broadly similar to what NASA would need for its own lunar reactor, because the moon's surface, like space, is a vacuum. If an interplanetary nuclear-powered spacecraft works, a nuclear reactor on the moon should work as well — and the Artemis program, which aims to begin laying foundations for a base at the lunar south pole by the decade's end, would benefit directly from the overlap.

Most of the technology for SR-1 already exists, according to the program's own assessment; the decisive test will be assembling it into one vehicle and seeing how it performs in space. With procurement underway and a progress rating of six on a ten-point scale, the program's fate now rests on integration, ground testing and a launch window that is less than three years away.