How does the cost of uranium enrichment to 90% compare to lower enrichment levels?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive Summary
The available documents do not provide a single, published dollar figure for the cost of enriching uranium to 90% U-235, but they converge on a clear technical and economic point: the cost of enrichment rises nonlinearly as assay increases, with the largest incremental cost drivers being separative work requirements and cascade complexity. Recent studies and technical reports emphasize modeling of multi-stream cascades, separative-work charges (SWU), and fuel-cycle impacts rather than public price lists for weapons-grade enrichment, leaving direct commercial comparisons to lower enrichments implied rather than enumerated [1] [2] [3].
1. Why a direct 90% price is elusive — the market and secrecy problem
Public sources reviewed do not state an explicit commercial price for enrichment to 90%, reflecting a combination of commercial confidentiality, national-security sensitivity, and the rarity of legitimate civilian demand for such high enrichments. The technical literature focuses on cascade modeling and separative work rather than spot prices, because enrichment to near-weapons levels is tightly regulated and typically confined to state programs or secure military contracts. The absence of a published unit price in the examined reports underlines that cost estimates must be reconstructed from separative work requirements and cascade economics rather than taken from market quotes [1] [4].
2. Technical driver: separative work and nonlinear cost escalation
Fundamental physics and process engineering make enrichment costs escalate nonlinearly with assay. The amount of separative work units (SWU) required increases sharply as product assay approaches 90% because removing the last fractions of U-238 becomes progressively harder. Technical analyses explicitly state that enrichment cost is not linear and that higher assays require disproportionate SWU and more elaborate cascade configurations, translating directly into higher operating energy and capital allocation per kilogram of product [3] [1]. This is the principal reason 90% enrichment is far costlier per kg than lower levels.
3. Cascade configuration matters — multi‑stream economics and tails management
Economic studies model enrichment cost by examining multi-stream cascades and how feed, product, and tails streams interact. Cost per unit of product depends heavily on cascade design, tails assay choice, and how operators allocate separative work among multiple output streams. Reports that calculate the cost of enriched products in multi-stream cascades show that optimizing for lower-enrichment commercial outputs (e.g., <5% for LWR fuel) is materially different from configuring cascades to produce very high-assay streams; achieving 90% typically requires different cascade staging, greater centrifuge numbers, and more complex material handling, increasing both capital and operational costs [1].
4. Context from fuel-cycle cost studies — SWU dominates when assays rise
Fuel-cycle cost reports analyze how changes in natural uranium price and separative-work charges alter overall fuel costs for reactors, and they uniformly show that as enrichment demand shifts to higher assays, separative-work charges become an increasing share of fuel-cycle cost. While these studies focus on reactor-grade enrichments, their sensitivity analyses demonstrate the principle that enrichment cost escalation has outsized impact on total fuel cost if assays increase substantially. That pattern supports the conclusion that moving toward 90% would sharply raise SWU-driven costs even if exact dollar figures are not published [2].
5. Mining and front‑end costs are related but secondary on marginal enrichment
Comprehensive uranium production cost studies highlight factors like ore grade, reserve tonnage, and energy that shape natural uranium pricing, but these front‑end costs are less influential than SWU when comparing enrichment levels for a given mass of enriched product. UxC’s production-cost work provides the broader context for fuel economics but does not substitute for enrichment-specific cost modeling; front-end costs matter when feedstock requirements change because higher-assay enrichment can alter feed mass and tails choices, but the lion’s share of marginal cost to reach 90% stems from separative work and cascade complexity [5].
6. Proliferation-focused analyses: cost models and policy implications
Recent academic work modeling gas centrifuge characteristics ties technical parameters to proliferation risk and cost, emphasizing that practical and detection considerations strongly constrain production to very high assays. These analyses provide analytical tools to estimate SWU and cascade scale needed for 90% enrichment, reinforcing that while costs can be modeled, they are seldom publicly declared due to safeguards and legal prohibitions. The modeling literature therefore supplies the only route to credible cost comparison, but such reconstructions are sensitive to assumptions about tails, cascade efficiency, and unit SWU cost [4].
7. What the reviewed sources agree on — big picture comparison
Across the reviewed sources, the consistent factual conclusion is that enrichment to 90% is materially more expensive per kilogram than enrichment to reactor-grade levels, owing to sharply increasing SWU requirements, more complex cascades, and higher capital and operational burdens. None of the documents provide a simple price-per-kg comparison, but the technical and fuel-cycle studies collectively justify treating 90% enrichment as an order‑of‑magnitude escalation in marginal enrichment cost compared with the lower enrichments used in civilian reactors [1] [2] [3].
8. Missing data, methodological caveats, and where estimates could come from
The principal omission across the literature is published, audited cost-per-kg values for 90% enrichment; therefore any numeric comparison requires explicit assumptions about SWU price, cascade efficiency, and tails assays. The documents recommend reconstructing costs from SWU requirements and cascade models or using sensitivity analysis from fuel-cycle studies to bound likely ranges. Analysts seeking a dollar comparison should combine cascade-cost outputs with current SWU market rates and clearly state tails and efficiency assumptions, since different choices produce substantially different cost projections [1] [2].