For decades, spent nuclear fuel has been the nuclear industry’s most persistent headache. Roughly 400,000 metric tons of it sit in cooling pools and dry casks around the world, with its most hazardous components remaining radioactive for hundreds of thousands of years. Now, a major advance in nuclear fuel recycling has demonstrated that one chemical process, running on real fuel from a commercial reactor, can strip out nearly 90 percent of that long-lived danger in a single day.
In March 2025, Moltex Energy Canada announced that researchers at Canadian Nuclear Laboratories (CNL) in Chalk River had validated the company’s Waste to Stable Salt (WATSS) process using actual spent fuel bundles from an operating CANDU reactor. The tests, conducted inside shielded hot cells (the only facilities in Canada equipped to handle irradiated fuel), extracted 89.4 percent of the transuranic elements in 24 hours. After 60 hours, that figure rose to 94.3 percent.
This was not a computer simulation or a surrogate experiment using non-radioactive simulants. It was real spent fuel, containing real plutonium, americium, and other long-lived actinides, run through a real chemical process under conditions that mirror commercial operation.
What Is the WATSS Process, and How Does It Work?
WATSS stands for Waste to Stable Salt. It is a pyrochemical process, meaning it operates at high temperatures using molten salts rather than water-based solutions, and it targets a specific fraction of spent nuclear fuel: the transuranic elements.
Spent fuel from a typical reactor is mostly uranium (roughly 96 percent), with smaller amounts of fission products and transuranic elements such as plutonium, americium, and curium. The uranium left behind after one fuel cycle still contains usable energy, but it is the transuranics that drive the multi-hundred-thousand-year hazard profile that makes deep geological disposal such a daunting engineering challenge.
The WATSS process begins by converting the spent oxide fuel into a powder through voloxidation, which heats it in oxygen to break the ceramic pellets apart. The powder is then immersed in a high-temperature molten chloride salt containing a carefully chosen reducing metal. In the CNL tests, that metal was niobium. The reducing metal donates electrons to the transuranic oxides, converting them into chloride or oxychloride species that dissolve readily into the molten salt. Meanwhile, the uranium oxide remains chemically stable and settles out as an insoluble solid.
The result is a clean separation: the transuranics move into the salt phase, and the bulk uranium stays behind. Uranium carry-over into the salt was just 0.05 percent after 24 hours and 0.02 percent after 60 hours. The process pulled out the hazardous material without dragging the bulk uranium along with it.
Critically, WATSS is a group-separation process, not a method for isolating pure plutonium. This distinction matters for non-proliferation. The transuranics come out as a mixed stream, unsuitable for weapons use, which sidesteps one of the main objections that halted civilian reprocessing in the United States decades ago.
The CNL Validation: Real Fuel, Real Results
The experiments at Chalk River represent a significant milestone precisely because of where and how they were performed. Canadian Nuclear Laboratories operates the only hot-cell facilities in Canada licensed to handle full-length irradiated fuel bundles. A chemistry claim about spent fuel is only as credible as the facility allowed to test it, and CNL’s involvement gives the WATSS results a level of validation that simulant-based lab work cannot match.
The numbers are worth dwelling on. In the first 24 hours, the process extracted 89.4 percent of the plutonium, the primary long-lived hazard, into the molten salt. Extending the run to 60 hours pushed extraction to 94.3 percent. These are the first-stage results; Moltex has since reported achieving over 95 percent extraction in follow-up testing presented at the American Nuclear Society’s Advances in Nuclear Fuel Management Conference in July 2025.
What happens to the material left behind is equally significant. According to Moltex, a spent fuel bundle that starts as 100 percent high-level waste emerges from WATSS with roughly 98 percent reclassified as intermediate-level waste, a lower hazard category with far less demanding storage requirements. The high-level fraction shrinks to approximately 1 percent of the original total. The dangerous portion does not disappear, but it is drastically concentrated, making whatever permanent disposal solution is eventually needed far smaller in scale.
Why Doesn’t the United States Recycle Nuclear Fuel?
The WATSS breakthrough lands in the middle of an evolving policy landscape. The United States, home to the world’s largest fleet of commercial reactors and roughly 94,000 metric tons of accumulated spent fuel, does not recycle any of it.
The reasons trace back to the 1970s. In 1974, India tested a nuclear device using plutonium separated from spent fuel produced in a research reactor supplied through the Atoms for Peace program. The test demonstrated that civilian reprocessing technology could be diverted to weapons production. President Ford suspended commercial reprocessing in 1976; President Carter formalized the ban in 1977, citing proliferation concerns. When President Reagan lifted the moratorium in 1981, he offered no federal funding to restart it, and no commercial reprocessing plant has operated in the United States since.
Economics also played a role. Throughout the 1980s and 1990s, freshly mined uranium remained cheap and abundant, undercutting the financial case for recovering fuel from spent material. The collapse of the Yucca Mountain permanent repository project in 2010 left the United States without a disposal pathway that might have made recycling economics more attractive.
That landscape is shifting. In late 2024, the Department of Energy announced $10 million in funding for spent-fuel recycling research and development. At least two bipartisan bills have been introduced in Congress: one mandating a Department of Energy study on recycling technologies, and another aimed at streamlining the licensing process for commercial recycling facilities. Companies including Oklo and Curio are advancing their own reprocessing technologies, with Oklo planning a $1.68 billion advanced fuel centre in Oak Ridge, Tennessee, and Curio targeting a pilot-scale demonstration of its NuCycle process by late 2027.
Which Countries Are Already Recycling Nuclear Fuel?
While the United States has remained on the sidelines, several countries have been reprocessing spent fuel for decades, though at varying scales and with different technological approaches.
France operates the largest civil reprocessing program in the world. The La Hague facility in Normandy, run by Orano, has a capacity of 1,700 metric tons per year and supplies recycled mixed-oxide (MOX) fuel to Electricite de France’s reactor fleet.
Roughly 10 percent of France’s nuclear electricity comes from recycled fuel, and the country stores separated uranium as a strategic reserve for up to 250 years.
Russia reprocesses fuel at the Mayak facility in Ozersk (400 metric tons per year capacity) and is building a larger plant at the Siberian Chemical Combine in Seversk under its Breakthrough project, which aims to demonstrate a fully closed fuel cycle by the early 2030s. Russian scientists recently announced a plasma-separation technology that could further reduce waste volumes by separating actinides using electromagnetic fields rather than chemical solvents.
The United Kingdom operated reprocessing at Sellafield for decades, though its Magnox reprocessing plant closed in 2021 and the larger THORP plant had already ceased operations in 2018. Japan’s Rokkasho facility, with a capacity of 800 metric tons per year, has faced repeated delays but remains the centrepiece of Japan’s long-term fuel-cycle strategy. India operates smaller reprocessing plants at Tarapur, Kalpakkam, and Trombay, totaling roughly 260 metric tons per year of capacity, as part of its three-stage nuclear program that envisions thorium-based reactors in the long term.
What distinguishes Moltex’s WATSS from these established programs and from other advanced reactor ventures pursuing waste-to-energy cycles is its simplicity and its specific focus. Existing commercial reprocessing, dominated by the PUREX process, uses organic solvents and aqueous chemistry to separate plutonium and uranium. WATSS uses a single molten-salt step driven purely by chemical redox potential, with no electrochemical separation required. It is designed from the ground up to pair with a fast-spectrum reactor that can burn the recovered transuranics directly, closing the fuel cycle without uranium enrichment.
From Waste to Fuel: What Comes Next
The WATSS process is only the first stage of a larger vision. Moltex is developing a companion reactor, the Stable Salt Reactor – Wasteburner (SSR-W), designed to consume the concentrated transuranic stream that WATSS produces. The SSR-W is a 300-megawatt electric molten salt reactor that uses static fuel tubes, a design that avoids the engineering challenges of circulating-fuel molten salt reactors while retaining their passive safety characteristics.
Together, WATSS and the SSR-W form a closed fuel cycle. Spent CANDU fuel enters WATSS; transuranics are extracted and converted into fuel salt; the SSR-W burns that salt, fissioning the long-lived actinides into shorter-lived fission products; and the spent fuel salt from the SSR-W can be recycled back through WATSS to remove fission products and return the remaining actinides to the reactor for another pass. Moltex’s modelling indicates that this cycle can reach an equilibrium where actinide burning continues indefinitely, with fission products and salt activation products as the only waste streams requiring disposal.
The scale of the opportunity is substantial. Canada has accumulated approximately 3.4 million used CANDU fuel bundles, roughly 65,840 metric tons of heavy metal, as of mid-2025, with about 90,000 additional bundles generated each year. If the existing reactor fleet operates to the end of its planned life, the total inventory will approach 5.9 million bundles. Moltex projects that WATSS processing could reduce the footprint of a future deep geological repository by as much as 80 percent by removing the long-lived transuranics that drive repository size requirements.
The project has attracted significant backing. The Government of Canada, the Province of New Brunswick, and the provincial utility NB Power have all provided support. Engineering firms IDOM and CANDU Energy (an AtkinsRealis company) are contributing design and licensing expertise. Indigenous communities in New Brunswick, including the North Shore Mi’kmaq Tribal Council, have invested directly in Moltex’s development. The company entered pre-licensing consultation with the Canadian Nuclear Safety Commission in April 2025 and is progressing toward a first-of-a-kind deployment at the Point Lepreau site in New Brunswick.
Realism is warranted. No electricity has yet been generated from material processed by WATSS. The SSR-W reactor design must still navigate a full regulatory review, a process measured in years. The chemistry worked in a hot cell, but that is not the same as operating a commercial-scale recycling plant paired to a working reactor. The target date for first power is sometime in the 2030s, and advanced nuclear timelines have a track record of slipping.
What exists today is a validated chemical step, demonstrated on the real material it was designed to process, at the scale of individual fuel bundles, in the only Canadian facility qualified to do the work. For a technology that aims to transform one of energy’s most persistent liabilities into a manageable asset, that is a credible place to start.
Frequently Asked Questions
Where does spent nuclear fuel go if it is not recycled?
In countries without recycling programs, spent fuel is stored at reactor sites in two stages. Freshly discharged fuel spends several years cooling in deep water pools. Once its temperature and radioactivity have dropped sufficiently, it is transferred to dry concrete-and-steel casks that sit on concrete pads at the plant. In the United States, spent fuel is stored at more than 70 sites across 35 states. Canada’s 3.4 million used fuel bundles are held in wet and dry storage at reactor sites in Ontario, Quebec, and New Brunswick. Neither country has an operating permanent geological repository, though both are in various stages of siting and licensing one.
Finland is the global leader in this regard: its Onkalo repository, the world’s first deep geological disposal facility for spent fuel, is expected to begin accepting fuel in the mid-2020s. Its Onkalo repository, the world’s first deep geological disposal facility for spent fuel, is expected to begin accepting fuel in the mid-2020s.
How many years of nuclear fuel do we have left?
At current consumption rates, known conventional uranium reserves are sufficient to fuel the existing global reactor fleet for roughly 90 to 130 years, depending on assumptions about exploration and extraction technology. However, that figure changes dramatically if recycling enters the picture. Spent fuel still contains more than 90 percent of its original energy potential after a single pass through a light-water reactor. Fast reactors, CONTEXT: like the SSR-W, can extract roughly 60 times more energy from the same uranium by fissioning transuranics that conventional reactors leave behind. With full fuel-cycle closure, uranium resources could stretch to thousands of years of supply, and the existing stockpile of spent fuel itself becomes an enormous energy asset. Oklo’s CEO has estimated that the energy content in America’s spent fuel alone could power the entire country for 150 years.
Is recycling nuclear waste expensive?
Yes, reprocessing is more expensive than once-through fuel cycles when uranium prices are low. The World Nuclear Association estimates that reprocessing adds roughly 30 to 35 percent to the cost of nuclear fuel, though the figure depends heavily on plant scale, uranium market prices, and how disposal costs are accounted for. However, the economic calculus shifts when permanent disposal costs are factored in. If a country must build a deep geological repository regardless, and the cost of that repository scales with the volume and hazard classification of the material placed in it, a process like WATSS that can reduce high-level waste volume by roughly 80 percent and downgrade much of it to intermediate-level becomes a cost-avoidance strategy, not just a fuel-production expense. In Canada’s case, the NWMO estimates the total cost of a deep geological repository in the tens of billions of dollars, making any technology that shrinks the repository’s required footprint economically relevant.
What is the difference between reprocessing and recycling?
The terms are often used interchangeably, but a useful distinction has emerged in policy discussions. Reprocessing typically refers to the chemical separation of plutonium and uranium from spent fuel, as practiced at La Hague and historically at Sellafield; the output is a separated plutonium stream that requires careful safeguards. Recycling implies a broader fuel-cycle approach where separated materials are fabricated into new fuel and consumed in reactors, ideally without producing a pure plutonium stream at any point. WATSS fits the recycling definition: it produces a mixed transuranic salt suitable directly for reactor fuel, and the group-separation chemistry means plutonium is never isolated as a separate product. This distinction has been central to the policy debate in the United States, where proliferation concerns have historically blocked reprocessing but where recycling, defined as redeployment rather than separation, is gaining political traction.
