Every spacecraft ever sent to operate beyond Jupiter has depended on plutonium-238—a reactor-made isotope that cannot be mined. The United States stopped producing it in 1988, then spent nearly three decades drawing down old reserves and supplementing them with material purchased from Russia. In 2015, American production finally restarted with a batch of just 50 grams.

Every spacecraft that has gone on operating beyond Jupiter has carried the same answer to a basic problem: plutonium-238.

Pioneer 10, Pioneer 11, Voyager 1, Voyager 2, Ulysses, Cassini and New Horizons all depended on the heat released by this isotope. Yet plutonium-238 is not something a mining company can extract from the ground. It has to be manufactured in reactors, chemically separated, turned into a ceramic and sealed into heat-source modules through a chain of specialised facilities.

When the United States stopped making it in 1988, that chain disappeared. NASA and the Department of Energy spent the following years drawing down inherited stock and buying additional material from Russia. Then, in December 2015, Oak Ridge National Laboratory announced that it had produced 50 grams.

Fifty grams sounds almost absurdly small beside a planet-sized mission. It was. But the point was not the quantity. The point was that, after nearly three decades, the United States had remembered how to make it.

Jupiter is where sunlight becomes an engineering constraint

Solar power does not stop at Jupiter. NASA’s Juno spacecraft has operated there with enormous solar arrays, and improving cells have made sunlight useful farther from the Sun than engineers once assumed.

But the arithmetic becomes severe. Jupiter receives roughly one twenty-fifth of the sunlight that reaches Earth. At Saturn the fraction is closer to one hundredth, and at Pluto it is around one thousandth. Panels have to grow while the electricity available to run heaters, computers and instruments shrinks.

This is why the wording matters. Not every mission to Jupiter has used plutonium, but every spacecraft on NASA’s list of machines that continued operating beyond Jupiter has relied on a radioisotope power system.

Sunlight also comes with operational vulnerabilities. As I wrote in my piece on Opportunity’s fourteen-and-a-half-year life on Mars, the rover finally fell silent when a planet-wide dust storm cut the energy reaching its panels. A plutonium-powered spacecraft does not need a clear sky, a favourable season or a particular orientation towards the Sun. Its source is always warm.

It is a battery only in the loosest sense

Plutonium-238 is often described as nuclear battery fuel. That phrase is useful, but it can create the wrong picture. There is no miniature reactor inside Voyager or New Horizons. There is no controlled chain reaction, turbine or moving generator.

The isotope decays naturally, mostly releasing alpha particles. Their energy becomes heat inside a tough ceramic form of plutonium dioxide. A radioisotope thermoelectric generator, or RTG, then uses thermocouples to turn a small fraction of the temperature difference between its hot interior and cold exterior into electricity.

The conversion is not particularly efficient. Only about six per cent of the heat becomes electricity, but an RTG can work continuously for decades with no moving parts. Plutonium-238 has a half-life of 87.7 years, long enough for a mission to cross the outer Solar System and continue talking after its original builders retire.

I explored that endurance more closely in an earlier article about the surprisingly small quantity of plutonium that has kept Voyager alive. But the isotope’s longevity creates a strange political illusion. Because old fuel keeps working, it can hide the loss of the industrial system that made it.

You cannot mine a launch manifest

Trace plutonium exists in nature, but not in quantities that can supply spacecraft. According to NASA’s account of the fuel programme, usable plutonium-238 has to be made by exposing neptunium-237 to neutrons in a reactor.

A neptunium-237 nucleus captures a neutron and becomes neptunium-238. That unstable isotope then undergoes beta decay, turning into plutonium-238. The irradiated targets must be dissolved and processed so the new plutonium can be separated while the unused neptunium is recovered for another cycle.

This is also why calling plutonium-238 “reactor-made” does not mean that an RTG is a reactor. The reactor is back on Earth, doing the manufacturing. The hardware launched into space simply harvests the isotope’s decay heat.

Nor is this the plutonium most closely associated with weapons. Plutonium-239 can sustain a chain reaction and is used in nuclear weapons and some reactor fuels. Plutonium-238’s intense heat and isotopic properties make it unsuitable for that job. For spaceflight, the heat is precisely the valuable part.

The modern process begins with neptunium oxide mixed with aluminium powder and pressed into targets. Those targets are irradiated at Oak Ridge’s High Flux Isotope Reactor or Idaho National Laboratory’s Advanced Test Reactor. Oak Ridge dissolves and separates the material, Los Alamos National Laboratory turns purified oxide into ceramic fuel pellets and seals them in protective cladding, and Idaho stores components and assembles power systems. Restarting production meant reconnecting all of those steps.

In 1988, an inventory became a countdown

The Savannah River Plant in South Carolina stopped producing plutonium-238 in 1988. At the time, the United States had accumulated a stock for defence and civil space programmes. But a stock of a radioactive heat source is not static.

Every year, about 0.8 per cent of plutonium-238 nuclei decay. That is the process that supplies the heat, but it also means older material gradually loses power. The problem was therefore not just how many kilograms remained. It was how much thermal output those kilograms could still provide.

The United States turned to Russia for part of the gap. A purchase agreement began in the early 1990s, and Russian material supplemented the domestic reserve until the last purchases in 2009. By 2015, Oak Ridge said about 35 kilograms had been set aside for NASA, but only roughly half met the desired power specifications.

Fresh material could extend that reserve because high-output plutonium could be blended with lower-output older stock. Even so, the system had become a countdown. A mission that required several kilograms was drawing from a finite inheritance, while the remaining fuel became a little less energetic each year.

A Government Accountability Office review later described the consequences plainly: NASA and the Department of Energy had to manage both the available inventory and the uncertain schedule of new production. This was not merely a procurement inconvenience. It affected which missions could be designed with confidence.

Fifty grams was proof, not a power supply

Oak Ridge’s December 2015 announcement was careful about what the 50-gram batch represented. It was the first end-to-end demonstration of US production in nearly 30 years. The sample still had to be analysed for its purity and energy output before the process could be scaled.

At about 0.56 watts of heat per gram, 50 grams of plutonium-238 produces roughly 28 watts of thermal power. After thermoelectric conversion, that might amount to less than two watts of electricity. It is enough for a tiny light, not a spacecraft carrying cameras, radios, heaters and scientific instruments.

For scale, Curiosity’s power source contained about 4.8 kilograms of plutonium dioxide. The 2015 batch was around one per cent of that mass, and not all of plutonium dioxide is plutonium itself.

But judging the restart by whether the first batch could power a mission misses its importance. The laboratory had fabricated targets, irradiated them, dissolved them, separated the isotope and recovered unused neptunium. The new material met the beginning of a qualification process. A production line that existed mainly in documents and ageing expertise had become physical again.

A distant mission begins years before anyone builds it

Radioisotope supply changes mission planning long before a spacecraft reaches a launch pad. Engineers need to know how many watts a power system can deliver at launch, how rapidly that output will decline, and whether the fuel will exist on the date the system has to be assembled.

This is why a few kilograms per year matters. In 2023, the Department of Energy completed a shipment of about half a kilogram of newly produced plutonium-238, the largest delivery since the restart. Oak Ridge has been working towards an annual production goal of 1.5 kilograms of plutonium dioxide, although a production goal should not be confused with a permanent guarantee.

New fuel has already been incorporated into the system powering Perseverance on Mars. Future material is expected to support missions including Dragonfly, the rotorcraft planned for Saturn’s moon Titan. These spacecraft do not merely need energy. They need energy through cold nights, dusty skies and environments where sunlight is weak or absent.

The supply question also changes how we read past missions. As I wrote about Cassini’s deliberate plunge into Saturn, the spacecraft was still working when the mission ended in 2017. Its fuel reserve for manoeuvring had become the decisive limit, not an inability to make electricity. Three RTGs had supported nearly two decades of flight, including thirteen years in orbit around Saturn.

That reliability can make the power source seem inevitable. It was not. Cassini launched from a country that had already stopped producing the isotope it carried.

The farthest journeys depend on work no camera sees

The 50 grams made in 2015 never became the centrepiece of a famous photograph. It did not fly past Pluto or return an image of Saturn’s rings. It was weighed, sampled and tested in facilities designed to keep people safely separated from intensely radioactive material.

Yet that modest batch restored something larger than its mass: the possibility of planning missions without assuming that a dwindling Cold War inheritance would somehow last forever.

Deep-space exploration is usually told through spacecraft. We remember the gold dish of Voyager, Cassini between Saturn’s rings and New Horizons sending back the first close view of Pluto. Behind those machines is an industrial chain that begins with powders pressed into targets, continues through reactors and shielded chemical cells, and ends with ceramic fuel sealed to survive both launch accidents and decades in space.

Plutonium-238 does not make distant missions easy. It makes some of them possible. The striking part of the 2015 restart was not that America had made enough fuel for the outer Solar System. It was that, after 27 years, it had made enough to prove the next gram could follow.