Lab-grown gold to crash prices? San Francisco’s Marathon Fusion just turned mercury into yellow metal

Marathon Fusion, a San Francisco-based fusion start-up has said that it turned mercury into “real, stable gold,” beating CERN’s record by a factor of 10. The San Francisco startup announced the result on social media platform X on October 21. However, the company has not disclosed how much gold its experiment produced, leaving questions about the scale and commercial potential of the claimed breakthrough.

The claim has revived an age-old question: could science make gold in a lab and eventually make the precious metal less scarce?

Marathon Fusion’s process


The process, called nuclear transmutation, changes an atom’s nucleus to create a different element. Marathon Fusion proposes using high-energy neutrons from a deuterium-tritium fusion reaction to convert mercury-198 into mercury-197, which then decays into stable gold-197. Radioactive byproducts remain a challenge.

The underlying physics has been demonstrated at CERN. Between 2015 and 2018, the Large Hadron Collider produced an estimated 86 billion gold nuclei, or 29 picograms, by colliding lead nuclei. However, these nuclei existed for only fractions of a second before breaking apart. Unlike CERN’s experiment, Marathon Fusion aims to use fusion-generated neutrons to produce gold on a potentially larger scale.

ALSO READ | India’s hidden gold chest is giving imported bullion a run for its money

Is fusion-made gold a threat to prices?

There is little evidence to assess Marathon Fusion’s demonstration. The company did not disclose how much gold it produced, identify the CERN record it surpassed or share experimental data. The rest of its post focused on medical isotopes rather than gold.

For comparison, CERN’s ALICE experiment estimated that the Large Hadron Collider produced about 29 picograms of gold during its 2015–2018 run. The gold nuclei were created in four experiments but broke apart within fractions of a second.

Marathon Fusion’s 2025 research paper offers a projection for commercial-scale production, estimating that a plant could produce about two tonnes of gold annually per gigawatt of thermal power. Global gold mines, meanwhile, produced 3,672 tonnes in 2025, according to the World Gold Council. An estimated 222,600 tonnes of mined gold already existed above ground as of the second quarter of 2026.

Even the projected two-tonne output assumes a facility far larger than anything Marathon currently operates. Its website lists a reactor with a fusion power output of 1 kilowatt, a millionth of a gigawatt.

The gold would also need to be stored before it could be sold. Marathon’s paper says the process would leave radioactive isotopes in the newly produced gold, requiring an estimated seven to 17 years of storage, depending on the applicable safety standards. Purifying the mercury used as feedstock could reduce this period, the researchers said.

Fusion-made gold therefore remains far from commercial markets and is unlikely to pose an immediate threat to prices. Gold had already been declining for months before Marathon’s announcement.

Multicoin Capital founder and CIO Tushar Jain did compare the experiment to lab grown diamonds and said, “Lab grown diamonds crashed the price of diamonds. What will lab grown gold do to the price of gold?”

One user wrote on X, "It’s still meaningless small amounts. The merging of neutron stars—a kilonova—is the only place where the pressure is high enough and the neutron flow fast enough to build such a heavy element. The collision must produce heat up to an insanely high 10 billion degrees C to allow the “rapid neutron capture” required to produce gold. As a reference, our sun’s core is only 15 million degrees C."

Gold prices:

Gold is trading 5.56% below its January 1 level. The metal had risen around 25% above its New Year level in late January before reversing course.

Market account Bull Theory said gold had shed about $406 billion in value. Silver fell 3% over the same period, taking combined losses in the two metals beyond $500 billion.

However, some analysts believe gold’s downturn may be approaching its final stages. Benjamin Cowen, founder of IntoTheCryptoverse, said he thinks the metal is nearing the end of its bear market, although further weakness could persist for some time.

For investors, gold’s price trajectory remains a more immediate concern than the prospect of lab-made supplies. The coming weeks will show whether the metal can recover its losses in 2026 or whether the sell-off has further to run.