BasicsTRISO-X, the fuel arm of X-energy, holds the first Category II fuel fabrication license the NRC has ever issued, a 40-year Part 70 license granted in February 2026 for its Oak Ridge, Tennessee campus. It is also the first new commercial fuel plant license in more than 50 years, according to Nuclear News, and TRISO-X says the review finished about three months early. The plant it covers, TX-1, is still under construction, and the NRC must inspect it before any HALEU arrives.
That gap between a license on paper and a working production line is where the TRISO story sits today. The fuel is not new. The Department of Energy says it was first developed in the 1960s, and it now calls TRISO “the most robust nuclear fuel on earth”. What is new is the push to build a domestic supply chain around it, and the “meltdown-proof” label that comes attached. If you track SMR schedules, ask which of your assumptions quietly depend on fuel arriving on time.
What “meltdown-proof” gets right, and what it skips
A TRISO particle is a kernel of uranium oxycarbide or uranium oxide about 0.5 mm across, wrapped in four coatings. Thousands of these poppy seed-sized particles are pressed into graphite pebbles or compacts. Each particle is its own tiny containment vessel, and the silicon carbide layer does most of the work.
Kernel: uranium oxycarbide (UCO) or uranium dioxide (UO2), enriched anywhere from 5% to 20% U-235 for most advanced designs
Porous carbon buffer: leaves room for gas and swelling
Inner pyrolytic carbon: seals the kernel and supports the silicon carbide
Silicon carbide: holds fission products inside the particle
Outer pyrolytic carbon: protects the silicon carbide and bonds to the matrix
The label holds up well under the temperatures these reactors are designed around. Idaho National Laboratory notes that peak fuel temperatures in a high-temperature gas reactor accident are limited to roughly 1,600°C. In the AGR-2 test, UCO compacts sat at 1,600°C for 300 hours, and ORNL and INL researchers detected no failed TRISO or failed silicon carbide in them. Held there for 300 hours, the particles kept their contents.
The label skips the exceptions. The UO2 compacts in that same test showed cesium release from particles whose silicon carbide failed. A later AGR-5/6/7 compact, irradiated at 1,361°C, appears to have had one silicon carbide failure after 200 hours at 1,600°C. “Meltdown-proof” is a fair summary of particle-level retention at design temperatures. It is not a promise of zero failures, and the test data above shows why the distinction matters.
TX-1 and the Xe-100 behind it
The Xe-100 is a high-temperature gas-cooled reactor rated at 200 MWt, or 80 MWe, per unit, built around TRISO pebbles and designed for 60 years of operation. Its first planned customer is Dow. The Long Mott Generating Station would put four Xe-100 units, 320 MWe in total, at Dow’s Seadrift site in Calhoun County, Texas.
On May 18, 2026, the NRC completed an environmental assessment with a finding of no significant impact, in under a year. Per the agency’s project dashboard, as reported by POWER, the final safety evaluation is due by November 2026. A construction permit decision can follow it. Dow filed the application in March 2025.
TX-1 is meant to feed that project first. On August 24, 2026, TRISO-X said the plant moved into interior build-out and equipment installation, with Clark Construction Group staying on as contractor.
Footprint: 214,000 square feet in Oak Ridge’s Horizon Center Industrial Park
Capacity: about 700,000 pebbles per year, or 5 metric tons of uranium
Reach: enough fuel for up to 11 Xe-100 reactors
TX-2: still in design, with two more lines at 20 MTU per year, for 25 MTU across the campus
Gate before HALEU: the NRC still has to inspect the plant before it can receive any enriched material
Timing is the soft spot. X-energy earlier targeted production by December 2027, while Nuclear Engineering International now projects commercial fabrication at TX-1 in early 2028. Treat both dates as moving, and check them against the next TRISO-X filing.
BWXT, Kairos and the DOE pilot lines
TRISO-X is not the only name making particles. BWXT produces fuel at its Specialty Fuels Fabrication facility in Lynchburg, Virginia. In June 2026 it announced that its TRISO compacts supported the first new reactor criticality under DOE’s reactor pilot program, for Antares Nuclear at Idaho National Laboratory. BWXT also processed the HALEU feedstock for that fuel from NNSA scrap, per its press release on the milestone. On its first-quarter 2026 earnings call, CEO Rex Geveden described BWXT as the only TRISO producer at any scale, making hundreds of kilograms a year and weighing a larger site in Wyoming. That figure comes from a transcript summary, so it is worth confirming against the filing.
Kairos Power takes a different route. Its 140 MWe KP-FHR uses TRISO pebbles in a Flibe molten salt coolant, where the pebbles float, which allows refueling without shutdown. Kairos finalized a DOE contract for HALEU in January 2026 to make pebbles for Hermes 1, its low-power demonstration reactor in Oak Ridge, working with Los Alamos National Laboratory. Nuclear News also reports up to $303 million in ARDP funding for Hermes. A five-year, $27 million ORNL partnership signed in February 2026 covers pebble testing. The sources reviewed do not confirm Hermes 1’s current operating status, so check Kairos directly.
Every project, deal, company, and regulatory event we cover is logged, verified, and searchable in SMRbrief Pro, which is the faster way to line up fuel suppliers against the reactors they feed.
BWXT (Lynchburg, Va.): Antares compacts, Project Pele heritage, HALEU downblending
TRISO-X (Oak Ridge, Tenn.): TX-1 under construction, NRC license in hand
Standard Nuclear: received DOE-allotted HALEU to make a full core of TRISO for Radiant’s 1 MWe Kaleidos microreactor
Kairos Power: pebbles for Hermes 1, with a BWXT collaboration on commercial production for Hermes 2
DOE Fuel Line Pilot Program: conditional selections include Oklo, Terrestrial Energy, TRISO-X and Valar Atomics
The bottleneck is enrichment, then throughput
Almost every TRISO design needs HALEU, which is uranium enriched between 5% and 20% U-235. DOE allocates much of the near-term supply through its HALEU Availability Program, and developers like Kairos and Radiant depend on those allocations to make fuel at all. A fuel plant with a license and no feedstock is a very expensive warehouse.
Scale is the second problem. TX-1’s 700,000 pebbles a year covers up to 11 Xe-100 units. X-energy’s stated pipeline is 11 GW, which it equates to 144 Xe-100 reactors. That math is why TX-2 exists, and why BWXT is looking at greenfield capacity. The reactors get the headlines, but the fuel line sets the pace.
Three things worth watching before year end:
November 2026: the NRC’s final safety evaluation for Long Mott
NRC inspection of TX-1: the step that clears HALEU delivery
BWXT’s capacity decision: a new TRISO plant would change the supply math
Which will arrive first, the reactor permits or the fuel to run them?