Uranium fuel pellets are about the least glamorous object in the energy business. Ceramic cylinders roughly the size of a pencil eraser, stacked inside metal tubes, made in more or less the same way for sixty years. Nobody puts them on a rendering.<\/p>\n
An Indianapolis company called First American Nuclear just asked federal regulators for permission to build an entire plant to make them. On August 17 it said the Nuclear Regulatory Commission had docketed its Regulatory Engagement Plan for a Category II high-assay low-enriched uranium deconversion and fuel fabrication facility, which is a long way of saying a building that takes uranium in as a gas and hands back ceramic.<\/p>\n
The customer is the company’s own machine: EAGL-1, a 240-megawatt small modular reactor cooled by liquid lead and bismuth. FANCO, as everyone shortens it to, is pitching the plant as a fix for America’s fuel supply problem.<\/p>\n
Which it might be. The reactor it feeds, though, runs on a coolant whose second ingredient has not been refined in the United States since 1997.<\/p>\n
Advanced reactor announcements almost always stop at the machine. Fuel gets a sentence, usually the one that says HALEU, and then everyone moves on to the megawatts.<\/p>\n
HALEU is uranium enriched above 5% and below 20% U-235, and the awkward truth is that it arrives from the enricher as uranium hexafluoride, a gas. A reactor cannot burn gas. Somebody has to convert it into solid uranium dioxide, press it, sinter it, grind it, and load it into fuel elements. That somebody, for FANCO’s design, does not currently exist.<\/p>\n
So the company is proposing to become it. The letter FANCO sent the NRC on August 7<\/a>, signed by executive vice president for regulatory affairs and licensing Michelle Catts, lays out what the facility would contain: a main process building handling receipt, conversion, fabrication, inspection and shipping, a secure storage vault, a waste management area, an analytical lab, and an administrative building with a security monitoring center.<\/p>\n The filing also spells out what the building will not have. No reactor. No critical assembly. Every operation held subcritical at all times, permanently, by design.<\/p>\n The uranium going in stays below 20% U-235, which puts the plant in the Category II bracket under 10 CFR 70.4, the class the NRC calls special nuclear material of moderate strategic significance. That designation drags in a specific pile of obligations: material control and accounting under Part 74, including a loss detection system and inventory difference limits, plus protected area and material access area security under 10 CFR 73.67.<\/p>\n The public version of the filing includes a section titled Principal Process Steps. It lists nine of them, numbered one through nine. All nine are redacted.<\/p>\n FANCO asked the NRC to withhold that enclosure in its entirety under 10 CFR 2.390, the rule covering proprietary information. The target dates in the submittal schedule are blacked out too. What survives is the list of topics the company plans to argue about with regulators before it files anything formal, and that list is its own little map of where the hard parts are.<\/p>\n Deconversion of the UF6 process gets its own white paper. So do criticality safety, material accounting, physical security, emergency management, and the transportation of HALEU material. The company also wants an audit-friendly setup, offering the NRC an access-controlled electronic reading room with read-only access to its calculation packages during pre-application.<\/p>\n The cadence it proposed is unusually specific for a document this early: biweekly project management calls, single-topic technical meetings mostly open to the public, and a promise to answer any regulator’s request for more information within thirty calendar days.<\/p>\n Lead-bismuth eutectic is 44.5% lead and 55.5% bismuth by atomic ratio, and the bismuth is doing real work: it drops the melting point to about 124\u00b0C, against 327\u00b0C for pure lead, which is the whole reason the alloy is attractive for a compact reactor.<\/p>\n Bismuth is also one of the rarest elements in the crust, sitting 65th in abundance, and it is almost never mined for its own sake. It comes out as a byproduct of refining lead, tin, tungsten and fluorspar ores, which means nobody builds a bismuth mine. You get bismuth when somebody else expands a lead smelter.<\/p>\n America stopped getting any. According to the US Geological Survey<\/a>, the last domestic primary production came as a byproduct of lead refining at a Nebraska refinery that closed in 1997. The last bismuth in the National Defense Stockpile was sold off the same year.<\/p>\n Meanwhile China refined 82% of the world’s supply in 2023, per USGS figures published this June<\/a>, with Laos and South Korea splitting most of the rest. US net import reliance on bismuth runs at 89%.<\/p>\n Then February 4, 2025 happened. Beijing added bismuth to its export control and licensing list alongside indium, molybdenum, tellurium and tungsten. Fastmarkets was assessing bismuth metal in Rotterdam warehouses at $30 to $36 a pound by late March 2025, in a market that had spent most of the previous decade in single digits.<\/p>\n None of this stops FANCO from buying bismuth. Prices have come well off that spike, the metal is traded, and a reactor is a one-time fill rather than a consumable.<\/p>\n But a company whose CEO, Mike Reinboth, said in the announcement that deploying nuclear at scale “requires a secure, capable and American fuel supply” has picked a coolant with a supply profile that is the opposite of that. The plant solves the uranium. It does nothing for the bismuth.<\/p>\n Worth being precise about what actually happened here, because the language around advanced nuclear tends to inflate on contact with air. The NRC accepted a plan describing how FANCO intends to talk to the NRC. That is the entire event.<\/p>\n The company says it will submit the actual Part 70 license application in the fourth quarter of 2027. Review follows. Construction follows that. FANCO also told Nuclear Engineering International<\/a> that it has not yet settled on a city or property lines for the plant, only that it lands somewhere inside a roughly $4 billion project footprint in Indiana, and that the site gets finalized as the application comes together.<\/p>\n For scale on what the finish line looks like, TRISO-X got there in February. The X-energy subsidiary received a 40-year Part 70 license for its Oak Ridge plant on February 13, which the Department of Energy<\/a> described as the first Category II fuel fabrication license the NRC has ever issued and the first new fuel fabrication license of any kind in about 50 years.<\/p>\n That application went in during 2022. We covered the Oak Ridge pebble plant and its 700,000-sphere annual target<\/a> earlier this month.<\/p>\n Radiant is in the queue too, with the NRC accepting a Part 70 application in May for its R-50 microreactor production facility, also in Oak Ridge. Oklo took a different road entirely, building its fuel fabrication plant next to its own reactor in Idaho<\/a> under Department of Energy authorization rather than NRC licensing.<\/p>\n FANCO is choosing the NRC route and starting from further back than any of them.<\/p>\n Lead-bismuth reactors are not theoretical. They are historical, which is a different and less comfortable thing.<\/p>\n The only ones that ever operated were Soviet naval units that powered a handful of Cold War attack submarines, and Nuclear Engineering International noted in its April write-up of the EAGL-1 pre-application<\/a> that the program was dropped over coolant solidification, corrosion and polonium-210.<\/p>\n That last one is inherent chemistry, not a design flaw: bismuth-209 absorbs a neutron, becomes bismuth-210, and beta-decays into polonium-210, an alpha emitter with a 138-day half-life that is volatile enough to complicate every maintenance job for the life of the plant.<\/p>\n The freezing problem is just as blunt. Let the alloy drop below about 124\u00b0C anywhere in the primary circuit and it goes solid, and a solid primary circuit is a dead reactor.<\/p>\n FANCO is not pretending the corrosion question away. It is building a lead-bismuth test loop to generate physical data instead of leaning on models, and in August it picked up two Department of Energy Nuclear Science User Facilities awards to irradiate candidate alloys, one on FeCrAl-ODS steels at Oak Ridge and one on MA956 led by EVP of nuclear fuels Philip Jensen.<\/p>\n Those are exactly the materials that have to survive standing in hot liquid metal for decades, and they are the same corrosion fight Sweden’s Blykalla is running with its own aluminum-forming steels<\/a>.<\/p>\nNine process steps, and every one of them blacked out<\/h2>\n
The uranium is the easier half of the supply problem<\/h2>\n
A docket is not a license, and the gap is measured in years<\/h2>\n
The coolant has a track record, and it is not a good one<\/h2>\n