Marathon Fusion says its plasma centrifuge enriched lithium and hydrogen

Marathon Fusion announced on August 27, 2026 that it has enriched hydrogen and lithium isotopes in its plasma centrifuge. The San Francisco isotope company is building separation technology for two jobs commercial fusion needs done at scale: enriching lithium-6, the feedstock for breeding tritium fuel inside a reactor, and selectively pumping unburned tritium out of reactor exhaust so it can be recycled. The goal is a process called differential pumping, which the company says could cut tritium flow rates and the size of tritium processing systems by a factor of ten or more; its initial separation results, per the release, are already large enough to substantially reduce those flow rates.

Heatmap News, which reported the results alongside an interview with CEO Kyle Schiller, adds the verification status: the materials testing lab Covalent has certified the lithium-6 enrichment, the hydrogen separation results have not been independently verified, and the tests used deuterium and protium, non-radioactive stand-ins for tritium, which is expensive and tightly regulated by the Nuclear Regulatory Commission. Per Heatmap's reporting, Marathon now plans to raise a Series A and build its first commercial pilot facility, with a goal of reaching full-scale production by 2029.

The problem is on the Department of Energy's books: the June 2026 Fusion Science and Technology Roadmap calls the fuel cycle and tritium processing "the circulatory system of a fusion plant" and progress there "critical for enabling sustained DT operation." Supply is thin today; Marathon adviser and UC Berkeley professor Per Peterson says in the release that the world makes less than a tonne of enriched lithium a year, with a mercury-based process now used only in Russia and China. Marathon's centrifuge work is funded through ARPA-E's VISION OPEN program, and reactor developers keep pouring concrete, most recently Pacific Fusion's $1 billion Albuquerque campus, which broke ground August 25. Schiller's pitch to Heatmap: whichever reactor design wins, it will need the fuel.