Marathon Fusion claims it made gold from mercury, but the cost kills the idea
Marathon Fusion says it turned mercury into gold, beating a CERN record. The cost of enriched mercury and the radioactivity would make the gold unviable.
English · Original discussion in Spanish · Published
Marathon Fusion claims it transmuted mercury into gold, but the cost and radioactivity bury the idea
A San Francisco startup called Marathon Fusion has announced that last month it converted mercury into stable gold, beating the CERN record by a factor of ten. The news has revived the old ghost of alchemy and, along with it, a question the diamond industry already answered with its wallet: can lab-grown gold sink the price of the metal? The announcement, however, has not been verified by anyone independent, and the material available does not allow anyone to know how much was produced or at what cost.
What exactly Marathon Fusion said
The company, founded in 2023 by Kyle Schiller and Adam Rutkowski —a former SpaceX engineer—, develops fuel systems for future fusion power plants. Its backers include Breakthrough Energy, Bill Gates's network, and Anglo American, the mining group that owns De Beers. Its method fires neutrons from fusion reactions at mercury, which turns into an unstable isotope and eventually decays into gold.
The record they claim to have smashed is the LHC's, which between 2015 and 2018 produced about 29 picograms of gold, that is, 29 trillionths of a gram. Ten times that is still an amount you could not see even under a microscope. For the non-scientists: a picogram is 1x10^-12 grams.
Why artificial gold will not move the price
According to a calculation circulating in the debate, a one-gigawatt fusion plant would produce, under the most optimistic models, between 2 and 5 tons of gold a year. The world's mines extracted 3,672 tons in 2025. A single plant would reduce the annual supply by roughly one tenth of one percent. It would take a hundred plants to add 500 tons, 14% of mining production. And no fusion plant sells electricity yet.
The other wall is the cost. According to that same calculation, the enriched mercury used as raw material runs about $15,000 per milligram, while gold trades near $135 per gram. You load the reactor with something 100,000 times more expensive than what comes out of it. On top of that, the resulting gold may contain radioactive isotopes and would have to be stored for up to 18 years before it could be sold. A plant that started up today would deliver its first sellable ounce, with luck, in the 2040s.
The synthetic diamond precedent
The antiestéticar has historical grounds. Lab-grown diamonds are chemically identical to mined ones and their wholesale prices fell between 90% and 95% from 2015, according to analyst Paul Zimnisky. De Beers launched Lightbox in 2018 at $800 per carat and shut it down in 2025. The company recorded an underlying loss of $511 million that year, and Anglo American is negotiating the sale of its 85% stake for about $1 billion.
Gold is a different league. About 222,600 tons have been mined throughout history, central banks stockpile it as a safe-haven asset, and on Thursday it traded at $4,062 per ounce. Unlike synthetic diamonds, transmuted gold, once its radioactivity has faded, would be impossible to distinguish from mined gold.
That is where everything stalls. The physics works; the arithmetic does not. And until someone proves otherwise, gold will keep coming out of a mine, not a reactor.
A San Francisco startup called Marathon Fusion has announced that last month it converted mercury into stable gold, beating the CERN record by a factor of ten. The news has revived the old ghost of alchemy and, along with it, a question the diamond industry already answered with its wallet: can lab-grown gold sink the price of the metal? The announcement, however, has not been verified by anyone independent, and the material available does not allow anyone to know how much was produced or at what cost.
What exactly Marathon Fusion said
The company, founded in 2023 by Kyle Schiller and Adam Rutkowski —a former SpaceX engineer—, develops fuel systems for future fusion power plants. Its backers include Breakthrough Energy, Bill Gates's network, and Anglo American, the mining group that owns De Beers. Its method fires neutrons from fusion reactions at mercury, which turns into an unstable isotope and eventually decays into gold.
The record they claim to have smashed is the LHC's, which between 2015 and 2018 produced about 29 picograms of gold, that is, 29 trillionths of a gram. Ten times that is still an amount you could not see even under a microscope. For the non-scientists: a picogram is 1x10^-12 grams.
Why artificial gold will not move the price
According to a calculation circulating in the debate, a one-gigawatt fusion plant would produce, under the most optimistic models, between 2 and 5 tons of gold a year. The world's mines extracted 3,672 tons in 2025. A single plant would reduce the annual supply by roughly one tenth of one percent. It would take a hundred plants to add 500 tons, 14% of mining production. And no fusion plant sells electricity yet.
The other wall is the cost. According to that same calculation, the enriched mercury used as raw material runs about $15,000 per milligram, while gold trades near $135 per gram. You load the reactor with something 100,000 times more expensive than what comes out of it. On top of that, the resulting gold may contain radioactive isotopes and would have to be stored for up to 18 years before it could be sold. A plant that started up today would deliver its first sellable ounce, with luck, in the 2040s.
The synthetic diamond precedent
The antiestéticar has historical grounds. Lab-grown diamonds are chemically identical to mined ones and their wholesale prices fell between 90% and 95% from 2015, according to analyst Paul Zimnisky. De Beers launched Lightbox in 2018 at $800 per carat and shut it down in 2025. The company recorded an underlying loss of $511 million that year, and Anglo American is negotiating the sale of its 85% stake for about $1 billion.
Gold is a different league. About 222,600 tons have been mined throughout history, central banks stockpile it as a safe-haven asset, and on Thursday it traded at $4,062 per ounce. Unlike synthetic diamonds, transmuted gold, once its radioactivity has faded, would be impossible to distinguish from mined gold.
That is where everything stalls. The physics works; the arithmetic does not. And until someone proves otherwise, gold will keep coming out of a mine, not a reactor.
Also available in: Deutsch
Summary of a discussion on Burbuja.info - Foro de economía, actualidad y política., translated from Spanish and reviewed before publication. Read the full discussion (25 replies).
More summaries
-
Gold holds $4,000 as silver faces 1kg bar shortage
Gold defends $4,000 despite rate hikes while Deutsche Bank predicts a silver surplus by 2027 amid physical stock shortages.
-
Gold pulls back as floor debated between $3,400 and $3,800
Gold is correcting from its highs, with the floor debated between $3,400 and $3,800 according to projections circulating on the forum. A strong dollar and interest rates explain the slide.
-
Gold pulls back from $4,100 while silver holds above $50
Gold corrects from $4,100 and silver holds above $50. China raises margins to 140% and Texas declares the metal legal tender. Floor or mirage?
-
Private Bullion moves metals community to Discord
Private Bullion shifts its metals community from Telegram to Discord with automated auctions, giveaways, and criticism from a rival channel.
-
Buy and hold stalls: big tech portfolios down 50%
Investors sitting on losses of 50% to 60% in major US tech stocks reopen the debate: buy and hold doesn't fail, it's applied badly when you buy at market highs.
-
30-year bond above 5%, gold at 27% of reserves: markets reposition
The 30-year Treasury bond has spent 27 days above 5%, and the 20-year auction yielded 5.469%, the highest since 1986. Gold now makes up 27% of global reserves.