Marathon Fusion says it has demonstrated a route to create stable gold from mercury using fusion neutrons, but the claim does not yet point to an imminent flood of laboratory-made metal or a near-term threat to gold’s global price. The San Francisco startup’s Oct. 22 announcement describes a nuclear transmutation process that could eventually turn high-energy neutron output into a commercial co-product, alongside medical isotopes and tritium fuel for fusion systems.
The process begins with mercury-198 exposed to 14.1 MeV neutrons from deuterium-tritium fusion. According to Marathon Fusion, the neutron impact removes two neutrons from the mercury nucleus through an “(n,2n)” reaction, creating radioactive mercury-197. Over several days, mercury-197 decays through electron capture into gold-197, the sole stable isotope of gold.
Marathon Fusion said its conversion rate was 10 times higher than a CERN record set in 2025. The company has not presented its announcement as evidence of large-scale gold production, and its first commercial system is targeted for delivery to initial partners in 2027.
Fusion neutrons create a potential new use for reactor output
The project rests on a characteristic of deuterium-tritium fusion: each reaction produces a neutron carrying 14.1 MeV of energy. Marathon Fusion said mercury conversion requires neutron energies above roughly 6 MeV, placing the reaction outside the range of typical fission reactors.
In the company’s proposed setup, neutrons that pass through a mercury target would continue into a lithium-6 blanket. Lithium-6 can absorb neutrons and generate tritium, a fuel required for deuterium-tritium fusion, as well as helium-4. The arrangement seeks to give a fusion facility more than one economic output from the same neutron stream.
Marathon Fusion was founded in San Francisco in 2023 by Kyle Schiller and Adam Rutkowski. The company has received approximately $4 million in U.S. government funding and raised $5.9 million in seed financing, according to the material provided by the firm.
Its initial commercial machine is based on beam-driven fusion rather than the tokamak-style magnetic-confinement reactors being developed for grid-scale electricity. Marathon Fusion’s published system description uses molten lithium deposited on a solid copper plate. The target absorbs tritium, while an accelerated beam of deuterium ions strikes it to produce fusion reactions and high-energy neutrons.
The company cited the rotating tritium target neutron source, or RTNS-II, operated at Lawrence Livermore National Laboratory in the 1980s, as a predecessor architecture. RTNS-II logged more than 25,000 hours at high fusion power density, according to Marathon Fusion. The startup says it is combining that approach with molten-lithium systems, radio-frequency ion generation and plasma centrifuges for isotope separation.
Marathon Fusion also said it had exceeded a 40-year-old Soviet record in plasma-centrifuge isotope enrichment. Its stated technical target is to sustain roughly one trillion high-energy neutrons per second on a target one centimeter in diameter.
Medical isotopes may offer the stronger commercial case
Gold has attracted the most attention because it is widely recognized as a scarce physical store of value, but Marathon Fusion’s nearer commercial rationale centers on medical isotopes. Some radioactive isotopes used in imaging and cancer treatment have short half-lives, require specialized production infrastructure and face constrained supply.
The company identified actinium-225 as one example. Actinium-225 is being developed for targeted alpha therapies, which deliver radiation directly to cancer cells through molecules designed to seek particular biological targets. Marathon Fusion said present global production serves only a few thousand patients annually and described the material as having a gram-equivalent value in the hundreds of billions of dollars.
It has also reported experiments using fusion neutrons on zinc targets to produce radioactive copper isotopes used in cancer diagnostics and treatment. In separate demonstration material, the company outlined a pathway from europium-151 to terbium-149, another isotope used in cancer-related research and therapeutic development.
Aakash Gupta’s analysis of data in Marathon Fusion’s paper framed the economics in terms of the implied value generated by each fusion neutron. Gupta estimated that a neutron used for grid electricity would carry roughly $3 × 10⁻²⁰ of value, compared with about $2.6 × 10⁻²⁰ when used to produce gold. Those estimates place gold close to electricity on a per-neutron basis rather than far above it.
Molybdenum-99, a medical isotope used to produce technetium-99m for nuclear imaging, produced a far higher estimate in Gupta’s framework: about $2.6 × 10⁻¹⁷ per neutron. He described molybdenum-99 as supporting 80% of nuclear medicine scans. Using Marathon Fusion’s actinium-225 pricing assumptions, Gupta estimated an implied value near $10⁻¹⁰ per neutron, or roughly 3 billion times the electricity case.
Those calculations suggest that specialized isotope production could support fusion systems at much lower energy-gain thresholds than electricity generation. Gupta estimated that grid power would require a plasma energy gain of roughly 10x to 22x, while gold as a co-product could reduce that range to about 3x. A machine focused on molybdenum-99 could theoretically operate around 0.001x under the model’s assumptions. For comparison, Gupta cited the UK JET tokamak’s best result at about 0.33x.
Gold economics remain theoretical and storage-dependent
The gold route faces constraints that distinguish it from an immediate substitute for mined supply. Mercury-197 must decay for several days before becoming gold-197, and the production system would need to operate reliably at sustained high neutron flux. The metal would also need to be separated, stored, certified and brought into existing bullion markets.
Michael Green’s estimate, quoted in the supplied material, placed an initial fusion-gold production cost at about $250 per ounce. After approximately 13 years of storage and a 10% cost of capital, Green estimated total cost at roughly $850 per ounce. Those figures remain model-based projections, dependent on reactor performance, target handling, isotope separation and financing conditions that Marathon Fusion has yet to demonstrate commercially.
The supplied material cited a gold spot price above $4,081 per ounce on Oct. 22. That gap between the quoted market price and projected production cost is striking, but it does not establish that fusion gold can be delivered at scale. Earlier transmutation experiments yielded extremely small quantities, including the 29-picogram result cited for 2025, while Marathon Fusion is still pursuing its first commercial deployment.
For cryptocurrency markets, the announcement is likely to revive a familiar debate over physical scarcity and digitally enforced supply. Bitcoin’s issuance schedule is governed by protocol rules and capped at 21 million coins, while gold’s supply depends on geology, mining, recycling and, potentially, future industrial transmutation. Yet a fusion-driven gold supply shock would require operational machines, economic production, regulatory handling of radioactive precursor materials and volumes large enough to affect a market measured in thousands of tonnes of annual demand.
Marathon Fusion’s work therefore offers a more immediate test of whether fusion neutron sources can earn revenue before supplying grid electricity. If medical-isotope output can fund facilities with lower energy gain than power generation requires, gold may become a visible demonstration of fusion transmutation rather than the startup’s primary commercial product.
Curious how fusion‑made gold compares with digital scarcity? Explore our guide on gold vs Bitcoin investment for long‑term strategy insights.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
