After 35% of the Uranium Has Been Burned: Power Fraction, Plutonium Share, and Uranium Cost

The Direct Answer

“35% of the uranium has been burned” is the setup line of a reactor-physics burnup problem, not an industry burnup metric — real commercial fuel is discharged at roughly 4%-7% of its heavy-metal atoms fissioned, so 35% refers to the fraction of the initial fissile uranium-235 consumed, not 35% of all uranium in the fuel. As that fissile inventory depletes, plutonium-239 bred from uranium-238 takes over a growing share of the fissions, and by the World Nuclear Association’s own account, plutonium is already responsible for “about one-third of the total heat output” of a light-water reactor averaged across a full cycle — near 60% in a heavy-water CANDU-type reactor. There is no single universal percentage for “the power fraction at 35% burned”: it depends on the reactor’s conversion ratio, which a real problem always supplies. What follows is the method to compute it, plus the real-world numbers, refueling data, and uranium-cost figures that a search-results summary can’t hand you.

What “35% of the Uranium Has Been Burned” Actually Means

Two different things get called “burnup,” and mixing them up is exactly what makes this phrase confusing. The industry’s operating metric is GWd/tU — gigawatt-days of heat produced per metric tonne of uranium loaded. The World Nuclear Association’s fuel cycle overview puts current common practice at 40 GWd/t, with 55 GWd/t already demonstrated and 70 GWd/t “approaching feasibility” within present enrichment limits. On an atom-percent basis, that entire range only fissions a small slice of the uranium loaded: the same source notes that around one-third of a core’s mass — roughly 24.3 tonnes of natural-uranium equivalent out of a 73-tonne, 1,000 MWe core — is replaced each year, or 36.5 tonnes every 18 months, because most of the heavy metal is still there when the fuel comes out.

“35% of the uranium has been burned” is a different number from a different framework. It is the classic phrasing used in university reactor-physics courses, where a core is loaded with enriched uranium and the question tracks what fraction of the original U-235 inventory — not the total uranium mass — has fissioned or transmuted by a given point in the cycle. Read that way, 35% is a plausible, late-cycle depletion fraction for the fissile atoms specifically, while the industry’s GWd/tU figures describe total energy extracted per tonne of heavy metal loaded (fissile plus fertile). Confusing the two is why a plain-language read of “35% burned” can sound impossibly high next to real 40-70 GWd/t fuel: it isn’t the same 35%.

The Physics: Power Fraction, Plutonium, and Burnup

Every version of “what fraction of the power comes from plutonium after X% has burned” rests on the same three moving parts:

• Burnup fraction (B): the share of the initial U-235 inventory that has fissioned or been transmuted — 0.35 for “35% burned.”
• Conversion ratio (C): new plutonium-239 atoms bred from uranium-238 for every U-235 atom consumed. This is set by the reactor’s neutron spectrum and the enrichment given in the problem — a light-water reactor and a heavy-water reactor do not share a conversion ratio, which is exactly why the two reactor types end up with very different plutonium power shares (below).
• Power fraction from plutonium (f_Pu): the plutonium-239 fission rate divided by the total fission rate. As B rises, U-235 atoms available to fission fall while Pu-239 atoms built up by C keep accumulating, so f_Pu climbs through the cycle — slowest at beginning-of-cycle, highest near discharge.

Because C is specific to the reactor and enrichment a given problem specifies, “35% burned” alone can’t be turned into one universal power-fraction number — you multiply B and C together with the relative fission cross-sections your problem provides. What a real, operating fleet shows once fuel reaches a comparable point in its life is the useful anchor: averaged over a full cycle, plutonium supplies “about one-third of the total heat output” in a light-water reactor, and “about 60% of the heat” in a CANDU-type pressurized heavy-water reactor, because CANDU’s natural-uranium fuel and heavy-water moderator give it a much higher conversion ratio. Both numbers describe the whole cycle, not one instant — the plutonium share is lower than that at beginning-of-cycle and higher than that by the time fuel approaches discharge, which is the direction any “after 35%” problem is testing.

Does Reactor Power Actually Drop at 35% Burnup?

No — not the way a simple “burnup vs. power output” curve implies. A commercial reactor holds its electrical output at its licensed rated power for the great majority of a fuel cycle by withdrawing control rods and, in a pressurized-water reactor, diluting the boric acid dissolved in the coolant as burnup consumes reactivity. Power doesn’t gradually decay with burnup under normal operation; the operators spend the fuel’s excess reactivity to keep output flat. Only near the very end of a cycle, once boron has been diluted close to zero and the control rods are largely withdrawn, does a plant enter a deliberate “coastdown,” backing off power over the final weeks because there is no more reactivity left to spend. The World Nuclear Association puts fleet-wide capacity factor at around 90% for good performance — a figure driven mainly by planned refueling and maintenance outages, not by a mid-cycle burnup-linked power dip.

That distinction matters for anyone using “35% burned” as an operational trigger: in GWd/tU terms, 35% of a fuel batch’s full discharge burnup is a normal mid-cycle waypoint, not a reliability warning sign. The real planning triggers are the refueling interval itself and the boron/control-rod margin the plant’s engineers track — covered below.

Uranium Cost and the Front-End Fuel Cycle

The other half of these searches — “uranium cost,” “plutonium fraction power” alongside cost questions — points at fuel economics. The World Nuclear Association’s economics breakdown gives a full front-end stack for 1 kg of uranium as UO2 fuel, priced as of September 2021: 8.9 kg of natural U3O8 at $94.6/kg for $842, conversion to UF6 for $120, enrichment (7.3 separative work units at $55) for $401, and fuel fabrication for $300 — a total of $1,663 per kilogram of uranium loaded. At an assumed burnup of 45,000 MWd/tonne yielding 360,000 kWh of electricity per kilogram of uranium, that stack works out to 0.46 ¢/kWh for fuel alone.

Front-end fuel is only 15-20% of the levelized cost for a new nuclear build, where capital dominates. For an already-built plant with its capital largely amortized, the same World Nuclear Association source puts fuel’s share of the remaining operating cost at 30-40% — the smaller the cost base, the bigger fuel’s slice of it.

Uranium Price Benchmarks: 2025 vs. Today

The World Nuclear Association’s $842-per-kilogram uranium line item is anchored to September 2021 prices; the market has moved since. The EIA’s Uranium Marketing Annual Report covering 2025, released July 29, 2026, put the weighted-average price of uranium purchased by U.S. civilian reactor operators at $58.46 per pound of U3O8 equivalent across 46.9 million pounds delivered. Long-term contracts, which carried 87% of those deliveries, averaged $55.91 per pound; spot-market deliveries, only 13% of the total, averaged $76.01 per pound — utilities that lock in supply ahead of need paid noticeably less than those buying on the spot market that year.

Spot moves daily and will already be different by the time you read this. As of August 7, 2026, the live U3O8 spot indicator tracked by Uranium Tracker stood at $86.50 per pound — well above the 2025 EIA weighted average, consistent with the broader run-up in term and spot prices reported through 2026. For your own current number, check the EIA report above for the latest full-year annual figures, or a live spot tracker for today’s number; don’t rely on either of these being current a year from now.

Calculator: Your Own Fuel Cost and Plutonium Share

Plug in a uranium price and burnup to see the front-end fuel cost per MWh, and which reactor type’s plutonium share applies, using the same method as the cost stack above.

Refueling Intervals and Outage Planning

U.S. reactors refuel on an 18-to-24-month cycle, and only about one-third of the fuel assemblies in the core come out at each outage — the rest stay in for another cycle or two. Duke Energy's walkthrough of the process puts a refueling outage at roughly a month, replacing about one-third of the used fuel each time; at its Brunswick plant, each fuel assembly holds 92 rods of about 350 pellets, with 560 assemblies making up the full core. The EIA's outage tracking shows the industry getting faster at this: average outage duration fell from 46 days in 2012 to 34 days in 2018, and in 2017 six reactors completed a refueling outage in under 20 days, the shortest at 15 days.

• Cycle length: 18-24 months between refuelings, timed for spring or fall when demand is lower (EIA, Duke Energy).
• Fuel replaced per outage: about one-third of the core (World Nuclear Association, Duke Energy).
• Outage duration: a fleet average of 34 days as of 2018, down from 46 days in 2012, with best-in-class outages under 20 days (EIA).
• What actually triggers refueling: the point where a plant's boron and control-rod margin can no longer hold 100% rated power for the next planned run — not a fixed atom-percent burnup number.

That last point is the durable planning rule for any site depending on nuclear-adjacent grid power or considering an on-site reactor: schedule around the plant's stated refueling window and its historical outage length, not around a specific burnup percentage, since the industry's own capability keeps shifting the achievable burnup upward.

Applying This in Agriculture, Forestry, and Mining

For agriculture, forestry, and mining operations that depend on grid power fed by a nuclear baseload plant, or that are evaluating a small modular reactor for a remote site, three of the figures above translate directly into planning inputs. First, the 18-to-24-month refueling cycle and roughly month-long outage set the calendar for any load that can't tolerate an unplanned interruption — ore processing lines, irrigation pumping during a critical growth window, or a timber kiln mid-run should have their maintenance and demand-response contracts built around that known cadence, not around a burnup percentage. Second, the fuel-cost stack above ($1,663 per kilogram of uranium at 2021 prices, moving with the uranium price benchmarks in the table above) is the input that feeds into any levelized-cost comparison a site does between grid nuclear power, diesel or gas backup generation, and renewables paired with storage. Third, because front-end fuel cost is a bigger share of an already-amortized plant's operating cost (30-40%, per World Nuclear Association) than of a new build's levelized cost (15-20%), a site weighing a new small reactor against buying power from an existing paid-off plant is comparing two different cost structures, not the same one at different scales.

Watch on YouTube

Securing uranium supply for any of these plans starts upstream, at exploration. Farmonaut's satellite-based mineral detection service maps uranium and other strategic-mineral prospectivity from orbit before a single drill hole goes in, which is the same logic mining and forestry operators already apply to site selection for processing infrastructure — assess remotely, commit capital only where the signal justifies it.

Watch on YouTube

Watch on YouTube

Satellite Intelligence for Uranium and Mineral Prospects

• Farmonaut's platform screens for uranium and rare-earth mineralization signatures without ground disturbance, ahead of the drilling that actually establishes a resource.
• The satellite-based mineral detection service flags prospectivity heatmaps and structural features relevant to both new exploration and ongoing site monitoring.
• A detailed methodology for the underlying 3D mapping approach is in this satellite-driven 3D mineral prospectivity mapping reference, useful for teams evaluating uranium or rare-earth targets alongside a nuclear-adjacent energy plan.

Watch on YouTube

Watch on YouTube

Watch on YouTube

Watch on YouTube

Watch on YouTube

Frequently Asked Questions

Conclusion

"After 35% of the uranium has been burned" describes fissile depletion inside a fuel cycle, not a reliability cliff — plutonium's power contribution rises through the cycle toward a whole-fleet average of about one-third in light-water reactors and about 60% in CANDU-type reactors, output stays at rated power via control-rod and boron adjustment until a deliberate end-of-cycle coastdown, and the actual planning calendar runs on an 18-to-24-month refueling interval with roughly a month-long outage. Uranium's own cost has moved from a $1,663-per-kilogram front-end fuel stack priced in September 2021 to a 2025 weighted average of $58.46 per pound U3O8, with spot trading near $86.50 per pound as of early August 2026 — figures worth re-checking against the EIA and World Nuclear Association links above rather than treating as fixed.