What HALEU is and where it comes from

Higher enrichment lets a reactor run longer between refuellings, hold a smaller core, and use fuel forms that conventional enrichment cannot support. That is why so many advanced designs specify it: TRISO-fuelled high-temperature reactors, sodium-cooled fast reactors like Natrium, and most molten salt concepts all assume HALEU.

The 20% ceiling is not a technical limit but a non-proliferation one. Above 20%, uranium is classified as highly enriched and attracts an entirely different security and safeguards regime. HALEU sits deliberately just underneath that line.

The supply position is the problem. Russia, through Tenex, was the only established commercial HALEU supplier in the world, and the 2024 US import ban closed that route. What exists in its place is small and mostly not yet built:

• Centrus at Piketon runs a demonstration cascade producing about 900 kg a year, and
• received a $900 million DOE task order in 2026 to expand to commercial scale.
• Downblending surplus weapons-grade uranium at Savannah River will yield roughly 3.1
• tonnes total. That is a finite stock, not a supply chain.
• Urenco targets HALEU production at Capenhurst in 2031.
• Orano is developing Project IKE at Oak Ridge with DOE support.
• X-energy's TRISO-X plant at Oak Ridge, which turns HALEU into finished pebbles, is
• targeted for completion in the first half of 2028.

Set those dates against the SMR developers promising units in the late 2020s and the mismatch is plain. This is the single strongest reason to be sceptical of any advanced reactor timeline: the reactor can be licensed and built, and still have nothing to run on.