The San Francisco startup Marathon Fusion has demonstrated a working technology for producing gold from mercury using a thermonuclear reaction. At the same time, the developers themselves emphasize: gold is far from the most profitable product of this scheme. The same facility is capable of producing a substance for cancer drugs that is billions of times more expensive.

Marathon Fusion engineers claim they have surpassed CERN in the speed of converting mercury into gold by 10 times, and have also broken a 40-year-old Soviet record for isotope enrichment. The company is now building its own thermonuclear reactor in San Francisco.

How the scheme works

During a thermonuclear reaction, two types of hydrogen — deuterium and tritium — fuse. Each such fusion releases an extremely fast neutron. Upon hitting a mercury atom, it knocks out another neutron from it. The atom becomes lighter and within a few days turns into gold — specifically, into its only stable isotope, which does not decay further.

Conventional nuclear power plants produce virtually no neutrons this fast, whereas a thermonuclear reactor emits them with every reaction. If radium is substituted for mercury, the result is actinium-225 — an extremely rare isotope on which the pharmaceutical company Novartis is building a new drug for prostate cancer. Actinium is so scarce that it is sold in tiny portions at $1,200–2,500 apiece. Converted to grams, this amounts to tens of billions of dollars — though a gram of such a substance cannot be held in one's hands due to radioactivity.

Why this is needed

Calculations show just how differently the same neutron is valued depending on its application:

  • for electricity — almost nothing;
  • for gold — roughly the same as for electricity;
  • for molybdenum-99 for medical scanners — a thousand times more;
  • for actinium-225 — approximately 3 billion times more than for electricity.

The key problem of the entire thermonuclear industry is that electricity does not yet pay for itself: to sell it at a profit, a reactor must output 10–22 times more energy than it spends on heating the plasma. The best result in history — about one-third of the input — was demonstrated by the JET facility in England.

With isotopes, the picture changes. To produce gold, a reactor needs only a 3x return, and for molybdenum-99 — to return just one-thousandth of the invested energy. That is, a facility that is deeply unprofitable in energy terms can still earn money — the main thing is to direct the neutrons at the right metal.

According to Marathon Fusion's estimates, a small 3 MW machine could supply the entire world with molybdenum-99. Today this isotope is produced in only a few research reactors, and current experimental facilities are already reaching the required level.

That is why the company now positions itself as a producer of isotopes for medicine. Its pipeline includes about 390 clinical trials of drugs with radioactive isotopes. Novartis's Pluvicto costs $42,500 per dose and brought in $1.9 billion last year. Gold remains in the plans: this market is estimated at $700 billion per year and could absorb about 2 TW of thermonuclear capacity — a scale that is significant for the entire energy sector. The logic of the strategy is as follows: earn now from medical isotopes using small facilities, and use the money from gold to build giant reactors — long before electricity begins to generate profit.

Cryptalist analysis: The Marathon Fusion case shows that the commercialization of thermonuclear fusion is shifting from energy toward high-margin isotope markets, where the economics of a facility cease to depend on the energy balance. For the crypto sector, this is a signal of the emergence of a new class of infrastructure assets potentially competing for capital and attention with the mining industry. Investors should track not the promises of "gold from mercury" but the actual contracts for the supply of medical isotopes — these will be the indicator of the technology's viability.