Copenhagen Atomics' EIC Accelerator Plan to Burn Waste With Thorium

The pitch is almost too good: a reactor that fits inside a 40-foot shipping container, burns the waste other reactors leave behind, breeds its own fuel from thorium, and could be mass produced at one unit a day per assembly line for €20 a megawatt hour. Copenhagen Atomics, based in Kastrup outside Copenhagen, calls its reactors Waste Burners: molten salt, moderated by unpressurised heavy water, fuelled with a fluoride salt of lithium, thorium and low-enriched uranium, with 100MW of thermal output.

The EIC Accelerator selected the project, Th-MSR, at its March 2025 cut-off for blended finance, and CORDIS records a grant of €2,499,999 from January 2026 to December 2027. The waste claim is the project's strongest argument: spent fuel from these reactors would need above-ground storage for 300 years, against 100,000 years of deep geological storage for conventional waste.

The experiment that would settle it

In July 2024 Copenhagen Atomics signed a four-year experimental collaboration with Switzerland's Paul Scherrer Institute for what would be the first critical experiment on a thorium molten salt reactor in Europe, using the company's Onion Core design at PSI's facilities, scheduled for 2026 to 2027, as World Nuclear News reported. The stated aims go beyond the company: build partner experience in design, construction, licensing, operation and decommissioning, and release open-source data for validating reactor modelling tools.

That experiment is the difference between an engineering programme and a claim. As of October 2026 no start or result has been reported, and the company's own first commercial reactor date, 2028, was set before the experiment was scheduled. CORDIS shows no periodic report for the EIC project, which only began in January 2026.

What has actually been built

The company's verifiable progress is at component level, and it is not trivial. It produces and tests full-scale prototypes at its Copenhagen site along with dozens of smaller loop tests and salt production, and in February 2026 its molten salt pump completed two years of continuous operation at high temperature. Pumps and valves in hot fluoride salt are exactly where molten salt programmes have historically failed, so a pump that runs for two years is a real result, and more informative than any cost projection.

What does not exist is a licence. Molten salt reactors have no precedent in European regulation, the UK design assessment is an entry application, and the Turkish and Indonesian engagements are agreements and studies rather than orders.

1. Components in hot saltTestedMolten salt pump ran two years continuously to February 2026.
2. Critical experimentScheduledWith PSI for 2026 to 2027; no start reported.
3. Regulatory pathEarlyA UK generic design assessment entry application.
4. €20/MWh at mass productionProjectionDepends on assembly lines that do not exist.

A grant that is small next to the task

The €2.5 million grant is a fraction of what any reactor programme costs, and secondary accounts of the award describe access to up to €15 million of EIC equity alongside it, which has not been announced as invested. Separate coverage reports a €20 million round, undated in the sources found, and the company is privately held with no published accounts, so its total funding is not verifiable here. Its commercial model through UK Atomics, owning and operating reactors and selling energy as a service, is a way to avoid asking utilities to buy an unlicensed machine.

Burning the waste Europe already has

Europe is storing tens of thousands of tonnes of spent fuel with no operating deep repository outside Finland, and the political cost of that is as high as the financial one. A reactor that consumes it and shortens the storage problem from geological to civilisational timescales is therefore attractive to governments in a way that a conventional small modular reactor is not. That is the argument behind this grant, and it is also why the Swiss experiment is being set up to publish open data: the stated aim is open-source data for validating reactor modelling tools, and molten salt reactors have no precedent in European regulation.