San Francisco startup Marathon Fusion says it has used neutrons to turn mercury into stable gold, a feat long associated with alchemy. In an Oct. 21 post on X, the company said it was "turning mercury into real, stable gold," at a rate 10 times higher than a record set at CERN, according to Business Standard.

This is a company claim. Marathon has not disclosed how much gold it made, what it cost, its yield or purity, or a method outside researchers can replicate, according to GadgetReview, and no independent verification was public at the time of reporting. The company did post gamma-ray spectra showing signatures of mercury-197 and gold-197.

The reason the story matters goes beyond novelty. Fusion power remains expensive and unproven commercially, and Marathon argues gold made as a byproduct could help finance fusion plants. If that worked, it could reshape energy economics. If it does not, it is a reminder to treat bold startup claims with care.

The Physics Behind Mercury Turning to Gold

The idea is real nuclear physics, not magic — mercury and gold sit next to each other on the periodic table, so changing one into the other means altering the nucleus.

Marathon's approach starts with mercury-198, a naturally occurring mercury isotope that makes up about 10% of natural mercury. Fusion of deuterium and tritium releases fast neutrons carrying about 14 million electron volts of energy. When one of those neutrons strikes mercury-198, it can knock another neutron loose, a process physicists call an (n,2n) reaction, leaving mercury-197.

Mercury-197 is unstable. It decays into stable gold-197 with a half-life of about 64 hours, according to IFLScience. In other words, within a few days, much of the mercury-197 becomes ordinary gold.

There is a bonus for the reactor. The same reaction releases an extra neutron, which helps fusion plants breed tritium fuel, according to the company's research.

Big Promises and Missing Numbers

Marathon's case rests on projections. In a 2025 preprint, company researchers simulated a fusion reactor blanket filled with mercury and estimated gold output of about 2 tonnes per gigawatt of thermal power per year, according to an American Physical Society abstract. They claimed the scheme would not reduce electricity output and could double a fusion plant's revenue.

Company statements cited by Business Standard use a larger figure, about 5 tonnes per gigawatt of generating capacity per year, and suggest 100 plants could produce roughly 14% of last year's mined gold. The two numbers use different measures and have not been independently reconciled.

The preprint has not been peer-reviewed. And commercial fusion plants that sell electricity do not yet exist, so the entire business case depends on technology still being developed.

The Oct. 21 result also leaves basic questions open. It is unclear what the "10 times CERN" comparison measures. CERN's Large Hadron Collider produced only about 29 picograms of gold nuclei between 2015 and 2018, which vanished almost instantly. Marathon has not said how much gold it collected.

Independent Experts Urge Caution

The company itself is young. Marathon began in 2023 working on tritium fuel systems for commercial fusion, and its website lists the U.S. Department of Energy and Breakthrough Energy among its partners. It also describes work on using fusion neutrons to turn radium-226 into actinium-225, an isotope used in cancer drugs. GadgetReview reported that the company's current reactor runs at about 1 kilowatt of fusion power, tiny compared with a power plant.

Transmutation itself is not new. Scientists turned mercury into gold as early as 1941, but that gold was unstable, IFLScience noted. The challenge has always been making stable gold in useful amounts at a reasonable cost.

Outside physicists have been intrigued but careful. Ahmed Diallo, a plasma physicist at the U.S. Department of Energy's Princeton Plasma Physics Laboratory who was not involved in the work, told the Financial Times in 2025, "On paper it looks great," as reported by Gizmodo.

There are real hurdles. The process can also create radioactive gold isotopes, so the gold might need about 14 to 18 years of storage before it is safe to sell, according to company estimates reported by Gizmodo and IFLScience. Enriching mercury-198 adds cost, and mercury itself is toxic, requiring careful handling.

Markets add another uncertainty. A large new supply could pressure prices, much as lab-grown diamonds saw steep wholesale price declines, a comparison Business Standard cited from Euronews.

Possible Effects on Energy and Mining

If fusion gold ever proved viable, it could change how future power plants are financed, giving investors a second revenue stream alongside electricity. It could also, in theory, reduce demand for some newly mined gold, an industry with heavy environmental footprints from land disturbance and chemical use.

Those effects are distant. No household will see lower energy bills or cheaper jewelry from this announcement. Readers interested in fusion investments should be wary of hype and look for peer-reviewed data and independent replication before concluding.

Next, the key milestones are a full disclosure of yield and methods, independent verification of the gold, and progress on the company's planned fusion reactor in San Francisco. Nature World News will track whether the claim holds up.

The bottom line: turning mercury into gold with fusion neutrons is physically plausible, and Marathon says it has done it. But without disclosed amounts, costs, or outside checks, it remains a promising company claim — not yet proof that gold can pay for fusion.

What Readers Want to Know

What did Marathon Fusion claim? That neutrons turned mercury into stable gold at a rate 10 times CERN's record. It has not disclosed how much gold it produced.

How does mercury become gold? A fast neutron knocks a neutron out of mercury-198, making mercury-197, which decays into gold-197 in about 64 hours.

Has the claim been verified? No independent verification was public at the time of reporting, and the company's 2025 preprint was not peer-reviewed.

Could this pay for fusion? The company claims gold could double plant revenue, but commercial fusion plants do not yet exist.

Is the gold radioactive? Some radioactive gold isotopes could form, and company estimates suggest 14 to 18 years of storage.

Will gold prices drop? Not soon. Any large-scale production depends on future commercial fusion plants.