Until recently, NASA faced a severe plutonium shortage that jeopardized future deep-space missions. In 2013 the U.S. Department of Energy announced, after a 25-year pause, that it would restart production of plutonium 238—the backbone of long-lasting nuclear batteries that have powered numerous missions since 1969. Yet the damage from the hiatus was already done. By 2021 the new effort will yield only enough of the radioactive fuel to make about two and a half nuclear battery modules a year. (The Mars rover Curiosity alone needed eight modules.) This paucity, plus the small existing stockpile, is barely adequate for the next decade of planned planetary missions to destinations such as the icy moons of Jupiter and Saturn. So NASA has been investigating alternatives. It recently become interested in a seafaring candidate: a military technology that has propelled U.S. Navy torpedoes.

The navy first experimented with Stored Chemical Energy Power Systems (SCEPS) in the 1920s, but it was not until the 1980s that engineers at Pennsylvania State University adapted the technology for warheads that could go fast enough and deep enough in their hunt for Soviet submarines. SCEPS harnesses the chemical reaction of two energy-dense reactants that remain stored and separated until needed. In torpedoes, the system commonly holds its energy in reserve as a solid block of lithium and a tank of the inert gas sulfur hexafluoride. When triggered, the combustion reaction of the two materials creates heat that turns the weapon's steam turbine to produce thousands of kilowatts of power.

The NASA version would tweak that chemical recipe. Michael Paul, a space systems engineer at Penn State, has proposed a demonstration mission to Venus, where a SCEPS-enabled robotic lander would draw on the planet's atmospheric carbon dioxide to burn the lithium. The resulting heat could drive an electrical generator to produce about three lightbulbs' worth of power, a sizable budget for space missions. (The Mars rovers Spirit and Opportunity ran on about one lightbulb's worth of solar power.) In July, Paul received half a million dollars in funding through the NASA Innovative Advanced Concepts program to measure the exact efficiency of carbon dioxide and lithium in this configuration. He will also work with planetary scientists to make recommendations to the space agency about other mission applications.

Nuclear power remains irreplaceable for deep-space missions intended to operate for years or decades, as Voyager, Cassini and New Horizons have done to much success, says Ralph McNutt, a physicist at the Johns Hopkins University Applied Physics Laboratory and chair of a NASA report this July on the available U.S. plutonium supply. Yet McNutt describes SCEPS as “exciting stuff.”

If Paul's vision comes to fruition, the energy from SCEPS could enhance missions using nuclear batteries or support those on trajectories too far from the sun to rely on solar power. Such technology might someday propel rovers designed to explore permanently shaded craters on the moon, run a Mars lander's drill in a dimly lit region or generate heat to keep a robot's electronics warm on icy Europa. SCEPS could even provide all the muscle behind shorter missions to not too distant destinations that would last days or weeks rather than years.