3 Tons of Gold Per Year from Nuclear Fusion?

Summary

Researchers at San Francisco-based Marathon Fusion propose using neutrons from a deuterium–tritium fusion reactor to produce both tritium fuel and gold.

Their design adds three layers around the reactor:

• Inner layer: Mercury and lithium. Fast fusion neutrons can trigger an (n,2n) reaction in mercury: one neutron enters a nucleus and two leave, changing the isotope. The resulting unstable mercury isotope can decay into stable gold.
• Middle layer: Molten lithium, beryllium, and fluoride salt. This layer slows some neutrons, which lithium can capture. Subsequent nuclear reactions produce tritium.
• Outer layer: Special steel shielding.

A computer simulation of a 1.5-gigawatt tokamak predicts nearly 3,000 kilograms of gold per year—roughly comparable in value to the electricity produced at current gold prices. The design aims to produce enough tritium to fuel the reactor and use other neutrons to make gold.

Uncertainties and limitations

• The work has not been peer-reviewed, and the realism of the simulation figures remains uncertain.
• The gold produced would contain some radioactive gold isotope and would take about 14 years to fall below safety thresholds.
• Fission reactors also produce neutrons, but the video says their neutrons are too slow for this proposed process.
• Selling tritium instead is unlikely to be a major alternative because the market is small and could quickly become saturated.
• CERN has also produced gold in lead collisions at the Large Hadron Collider, but only in picogram quantities.

Researchers and sources featured

• Marathon Fusion researchers: Authors of the proposed fusion-reactor gold-production design and simulation. No individuals are named in the subtitles, and the work is described as not peer-reviewed.
• CERN and the Large Hadron Collider: Cited for producing gold in lead collisions.