Physics / mechanism
HALEU (high-assay low-enriched uranium) is the enriched uranium feedstock specified by most advanced and small modular reactor designs. In the sources reviewed it appears not as a reactor physics topic but as a supply-chain variable: HALEU is described as the binding fuel constraint on the advanced-nuclear build-out, meaning that reactor licensing and construction schedules can be met while fuel availability still gates first criticality and refuelling ref.
The consequence is that the fuel part of the advanced-nuclear stack moves on a different clock from the reactor part. Reactor vendors are working through multi-year regulatory reviews (only one SMR design, NuScale’s 77 MWe unit, was fully design-certified as of 2023; X-energy’s Xe-100 was in an approximately 18-month construction-permit review with a possible late-2026 decision) while the enrichment and fuel-fabrication capacity that those reactors require is being funded separately ref. Public capital has been directed at this bottleneck: the US Department of Energy committed $2.7bn over ten years to HALEU supply ref.
Fuel materials is characterised as the fastest-moving component slice of the advanced-nuclear investment landscape, with Standard Nuclear (Oak Ridge) named as an example of a private company operating in that wedge ref. That component wedge, including fuel materials, is a small fraction of the headline SMR market and is not separately sized in the available market data ref.
The sources do not specify HALEU enrichment assay, fabrication routes (metal, oxide, TRISO), per-reactor loading requirements, or existing installed enrichment capacity. Those parameters should be treated as unresolved on this page.
Competitive landscape
HALEU sits upstream of the SMR and advanced-fission designs that create demand for it, so it is complementary rather than competitive with them. Within the wider firm-power set, its exposure is indirect: if HALEU supply slips, the beneficiaries are alternatives that do not need it, including existing-reactor restarts (the Microsoft-Constellation Three Mile Island restart, targeted for 2028 at a roughly $65/MWh PPA benchmark, is the only near-term firm power identified) and next-generation geothermal ref ref. The demand pull is real but almost entirely forward-dated: 2030 is described as the pivotal first-deployment year for SMRs, with the bulk of deployments in 2031-2035 ref.
Evidence base
- HALEU is identified as the binding fuel constraint on SMR and advanced-nuclear deployment ref.
- The US Department of Energy committed $2.7bn over ten years to HALEU (announcement dated January in the source) ref.
- Fuel materials is described as the fastest-moving component part of the advanced-nuclear supply chain, with Standard Nuclear (Oak Ridge) named ref.
- The component wedge including fuel materials is a small fraction of the SMR market and is not separately sized in available reports ref.
- Demand-side pull: datacentre-SMR conditional-offtake pipeline grew from roughly 25 GW at end-2024 to roughly 45 GW in 2026 (Build.inc / World Nuclear Association tracker, industry-tracker grade) ref.
- Reactor-side timing that fuel must meet: X-energy’s first unit at Dow Texas around 2028; Google-Kairos first unit 2030, fleet complete 2035 ref.
Frontier (open questions)
- What is the actual installed and contracted HALEU enrichment capacity in tonnes per year, and does it cover the fuel load of the reactors scheduled for first criticality in 2028-2031?
- How much of the DOE’s $2.7bn has been obligated to specific enrichment or deconversion facilities, and what are their commissioning dates?
- Do the fuel-form requirements diverge enough between designs (Xe-100 TRISO, Kairos, Oklo, NuScale) that fabrication capacity cannot be pooled across the pipeline?
- What price per kilogram of HALEU do announced SMR PPAs (for example the Oklo-Equinix 500 MW and Google-Kairos 500 MW agreements) implicitly assume, and how does that compare with the ~$65/MWh existing-reactor restart benchmark?
Synthesised 2026-08-31 from 4 KB sources by the resynth pipeline; citations are KB source slugs.