Pretty much every nuclear reactor running today works the same way at its heart: split uranium atoms, use the heat to boil water, and let the steam push a turbine. The fuel is the entire point of the machine. Take it out and you are left with a very expensive metal box.
So it is a little strange to watch what is going up at a research center in the wooded hills near Bologna, in northern Italy. There, a company called Newcleo is assembling a full-scale reactor that will spin a turbine and generate real electricity with no nuclear fuel inside it at all. The main vessel was lowered into place in late May. Empty, it weighs a little over 20 metric tons. Filled with what is coming next, it will pass 155.
What is coming next is molten lead.
Electric heaters do the work the uranium normally would
The machine is called PRECURSOR, and the name is the whole idea. It is a non-nuclear, pool-type demonstrator built at the ENEA Brasimone Research Centre, the Italian government lab that has worked on lead-cooled systems for years. Rated at 10 thermal megawatts, it is designed to copy the size, plumbing, and thermal behavior of the commercial reactors Newcleo eventually wants to sell, without ever going critical.
Instead of a uranium core making heat through fission, PRECURSOR uses banks of electrical heaters to fake it. They pour heat into the pool of molten lead exactly where the fuel would sit, and the rest of the system has no way of knowing the difference. The lead heats up, moves through the reactor the way it would in a live unit, dumps that heat into a steam generator, and the steam runs a turbine. Newcleo says the vessel is only slightly smaller than the one its commercial 200-megawatt reactor will need, so this is close to full scale, not a tabletop rig.
The turbine is the part worth slowing down on. It was built by Fincantieri, the Italian shipbuilder, and it is what lets PRECURSOR close the entire loop from heat to electricity. CEO and founder Stefano Buono has called it potentially “the only facility in the world to do such a complete demonstration.” Plenty of test rigs heat up lead. Almost none of them bother to wire the result to a generator and actually make power, because the power was never the point of a test stand. Here it is.
Getting molten lead in and out is its own engineering problem, which is why the lead-handling system has three separate pieces: a melting tank that turns solid lead ingots into liquid, a storage vessel that keeps the metal molten and chemically conditioned, and a transfer vessel that shuttles it into the main reactor. None of that exists in a normal water-cooled plant. Lead melts at around 327 degrees Celsius (621 degrees Fahrenheit), so it has to stay hot or it sets like a brick.
Lead is a strange coolant until you see what it does
Cooling a reactor with a heavy metal that has to be kept liquid sounds like a headache, and it is. So why do it? The lead-cooled fast reactor is one of six designs picked out by the Generation IV International Forum as a candidate for the next era of nuclear power, and most of its appeal comes straight from the coolant.
Lead boils at roughly 1,743 degrees Celsius, which means the reactor can run hot and at ordinary atmospheric pressure instead of the high pressure that water-cooled plants need to keep their coolant from flashing to steam. It also means the system carries an enormous amount of thermal margin before anything dramatic happens. The metal is dense enough that it circulates on its own once it is hot, so decay heat can be carried away by natural circulation, with no pumps and no operator stepping in, after the reactor shuts down. Lead even traps some of the nastier fission products, like iodine and cesium, instead of letting them escape, and if the vessel ever leaks, the lead simply solidifies in place rather than spreading.
Then there is the fuel, which is where this connects to a problem the United States has been sitting on for decades. Fast reactors like this one run on mixed uranium-plutonium oxide, and they can be loaded with material recovered from spent fuel and other reactors’ leftovers. Instead of treating that material as waste to bury, the reactor fissions it for energy and burns down some of the long-lived elements in the process. It is the same instinct behind the British work turning reactor graphite into diamond batteries: nuclear waste reframed as a resource nobody has used yet.
The American hook is a pile of Cold War plutonium
None of this would matter much to a US reader if Newcleo were purely a European project. It is not anymore. In October 2025, Newcleo signed an agreement with the California reactor startup Oklo to build advanced fuel-fabrication infrastructure in the United States, with Newcleo planning to put up to $2 billion into the effort through an affiliated investment vehicle.
Then, this spring, the US Department of Energy selected Oklo, with Newcleo as its partner, for advanced negotiations under the new Surplus Plutonium Utilization Program. The program, set up in late 2025, is an attempt to take the country’s stockpile of surplus weapons-grade plutonium and turn it into fuel for advanced reactors rather than guarding it indefinitely. Under the partnership, Oklo would lead the actual use of the plutonium while Newcleo brings its fuel experience and potential capital, all subject to definitive agreements and US security and safeguards rules. Oklo co-founder Jacob DeWitte described the idea as using existing surplus material as a bridge fuel to get more reactors online sooner. Interior Secretary Doug Burgum framed the $2 billion commitment as a win for the administration’s energy-dominance agenda.
Newcleo had already started talking to American regulators. Earlier in 2026 it began pre-application engagement with the Nuclear Regulatory Commission for both a fuel-fabrication facility and a lead-cooled fast reactor design aimed at the US market. And the company is about to get a lot more visible to American investors: in late May it announced plans to go public on Nasdaq through a merger with a blank-check company, NewHold Investment Corp. III, at a valuation of around $2.4 billion, with the combination expected to close in the second half of 2026 under the ticker NWCL. The deal still has to clear the SEC, so it is not done yet.
Nobody has a commercial version of this running yet
The technology is nowhere near a finish line, and the gap between a demonstrator and a power plant is wide. No commercial lead-cooled fast reactor exists anywhere in the world today. The only one being built as a genuine, fueled power unit is Russia’s BREST-OD-300 in Seversk, a 300-megawatt reactor that has been under construction since 2021 as part of a closed-fuel-cycle complex run by Rosatom. Even that one has slipped: its reactor vessel was around 70% assembled at the start of this year, and its startup, once planned for 2026, has been pushed to roughly 2029.
Newcleo’s own roadmap is long. PRECURSOR is meant to be finished in Italy this year. The first actual reactor, the 30-megawatt LFR-AS-30, is planned for a site in the Indre-et-Loire region of western France and is not expected to be operating until the end of 2031. A final investment decision on the first commercial plant is penciled in for around 2029, with commercial reactors following from 2033 and a dedicated plant to make the uranium-plutonium fuel due in France around 2030. These are the kinds of dates that tend to move.
The one thing to keep straight is what PRECURSOR is and is not. It is not a power station, and the “non-nuclear” label is doing real work: there is no fission, no uranium, no plutonium inside it. What it is, is a way to find out whether the hardest parts of the design, the molten metal, the heat exchange, the turbine integration, all behave the way the engineers expect before any nuclear fuel is ever introduced.
That is also why the turbine matters more than it looks. A test rig that just proves lead can be pumped around a loop is useful but unremarkable. One that takes that heat all the way to a spinning generator is testing the whole chain at once, including the unglamorous balance-of-plant pieces that decide whether a reactor design is actually buildable. The heavy, physical end of nuclear, the kind that involves lowering a 500-ton steel cylinder into place with the largest crane on the planet, does not care how good the slides look. PRECURSOR is Newcleo’s attempt to prove its version works in metal before it asks anyone to trust it with fuel.
