Expensive and rare tritium means ‘breeding’ techniques are needed
Researchers at Los Alamos National Laboratory in the US state of New Mexico have unveiled simulations showing how radioactive nuclear waste could be repurposed to generate tritium, the rare hydrogen isotope that fuels nuclear fusion.
Terence Tarnowsky, a physicist at Los Almos, will present his findings at the American Chemical Society’s autumn 2025 meeting. His research suggests that decades of stored waste might help unlock a cleaner, virtually limitless energy.
Nuclear fusion that uses the isotopes deuterium and tritium as fuel is seen as a promising pathway for generating energy, mainly because it requires lower energy input compared to other fusion reactions.
However, because tritium is rare and expensive, “breeding” techniques will be needed to provide enough of it for sustainable fusion power generation.
Current global supplies, largely sourced from Canadian “Candu” heavy-water reactors, amount to around 25 kg, worth about $33m (€28m) per kg.
Tritium is produced in Candu reactors as a byproduct of neutron interactions with the heavy water that is used as both moderator and coolant. According to ScienceDirect.com, approximately 130 grams of tritium are generated annually in a typical Candu reactor.
Nuclear fusion does not always require tritium. While the most commonly discussed fusion reaction, deuterium-tritium fusion, does involve tritium, other fusion reactions are possible. Deuterium-deuterium fusion, for example, can occur without tritium, although it requires higher temperatures and is less efficient.
Without a reliable tritium pipeline, commercial fusion plants cannot scale up. A single 1,000 MW fusion reactor would require more than 55 kg of tritium per year, far beyond current supply.
Tarnowsky’s proposal centres on accelerator-driven systems that bombard spent nuclear fuel with particles. This process triggers reactions that generate neutrons, which can then be harvested to produce tritium through a series of nuclear transitions.
Computer simulations suggest that a 1,000 MW system could produce about 2 kg of tritium annually, rivalling the total yearly output of all Canadian reactors.
A key advantage to Tarnowsky’s system would be the efficiency of tritium production. He projects that the design would produce more than 10 times as much tritium as a fusion reactor at the same thermal power.
Tarnowsky now plans to work out the cost for tritium production, and evaluate the efficiency and safety of the hypothetical reactor design.
