Researchers are developing systems to extract tritium, a crucial fuel for future nuclear fusion reactors, from existing nuclear waste, potentially solving two critical energy challenges simultaneously
The world’s growing demand for electricity, driven by everything from electric vehicles to the vast data centres powering artificial intelligence, has underscored the urgent need for sustainable and clean energy sources.
Nuclear fusion, a process that promises to unlock immense energy with minimal waste, stands as a theoretical ideal. However, a significant hurdle remains: the scarcity of tritium, a key fuel for fusion reactors.
A physicist at Los Alamos National Laboratory (LANL), Terence Tarnowsky, is proposing an innovative solution: using the very nuclear waste we struggle to contain to generate this valuable and rare resource.
The current nuclear landscape: Fission and its byproducts
Today’s nuclear power plants rely on nuclear fission, the process of splitting atoms like uranium or plutonium to release energy. While highly efficient, this method produces long-lived nuclear waste that requires complex and costly long-term storage. The United States alone has thousands of tons of this waste, raising concerns about potential environmental contamination and health risks. Meanwhile, nuclear fusion, the process that powers stars, offers a cleaner alternative by fusing light atoms, typically deuterium and tritium. While deuterium is abundant, tritium is not. The global inventory of tritium is a mere 55 pounds, with the U.S. having no domestic production capability, making it a highly expensive commodity.
A novel approach: Harvesting tritium from nuclear waste
Tarnowsky’s research addresses this dual problem by proposing a system that recycles nuclear waste to produce tritium. His work, presented at the American Chemical Society (ACS) Fall 2025 meeting, involves a series of computer simulations for a new type of reactor. This theoretical design uses a particle accelerator to initiate atom-splitting reactions in the nuclear waste, a process that ultimately yields tritium.
This accelerator-driven system offers a key safety advantage over traditional fission reactors, as the reactions can be turned on or off at will, unlike the self-sustaining chain reactions in current nuclear plants.
Efficiency and economic potential
The simulations have shown promising results. Tarnowsky’s theoretical system, powered by one gigawatt of energy, could produce approximately 4.4 pounds (2 kilograms) of tritium annually. This output is comparable to the entire yearly production from Canada, the world’s current major commercial source.
The projected efficiency of this design is a significant breakthrough, as it is estimated to produce more than ten times as much tritium as a fusion reactor of the same thermal power. This efficiency could dramatically lower the cost of a future fusion economy.
Next steps and broader implications
While the research is still in the simulation phase, Tarnowsky plans to refine his models to better evaluate the cost, efficiency, and safety of the proposed reactor. A key aspect of his future work involves developing a model that incorporates a molten lithium salt cooling system, which would not only serve as a safety measure but also make the system less suitable for weapons development. By combining previously engineered technologies in this new configuration, Tarnowsky and his team aim to provide decision-makers with a viable, cost-effective, and safe pathway to a fusion-powered future, all while simultaneously addressing the legacy of nuclear waste.
This innovative approach could fundamentally change how we view and manage nuclear byproducts, transforming a liability into a vital resource for clean energy.