STELLARIA’S MSR | SMRs & ADVANCED REACTORS
A complete chemical purification step is also required at the end of the operating life. This process will rely on a pyrochemical separation process, which Breyton indicated still needs to be developed. However, given the long lead times between installing the first Stellarium and the need to commission an optimised process, this is not seen as a significant obstacle to development: “We have 40 years ahead of us,” he says. Following purification, the salt is intended to be reused in
the same reactor type. “We multi-recycle the fuel,” Breyton says. “We create the salt, it stays for at least 20 years, and then the salt is cleansed for reuse.”
Development Path for ALVIN, MEGALVIN and First-of-a-Kind Stellaria’s early development has centred on integrated multi-physics modelling and Breyton claims some 2,500 simulations have already been performed using nuclear codes derived from historical French fast-reactor studies. With simulation and modelling underway, a three-stage
technical demonstration roadmap has been devised prior to commercial deployment. The first experimental facility, dubbed ALVIN, is a small-scale system designed to demonstrate first criticality and characterise intrinsic safety behaviour, especially the thermal-expansion shutdown mechanism. “We will insert a lot of reactivity inside the reactor, expect the dilation of the fuel, and show how fast it is to stop the fission,” he says. Data from ALVIN will be used to support regulatory licensing and refine core models. The second stage in the development process is
MEGALVIN, a roughly 10 MWth prototype with sufficient neutron flux to test structural materials, pumps, and heat exchangers under representative irradiation conditions. MEGALVIN is intended to validate integrated systems performance and materials ageing. Brayton’s timeline places ALVIN operational at the end of 2030, MEGALVIN at the end of 2032, and the first commercial 250 MWe reactor at the end of 2035. To this end, Stellaria positions itself as a reactor-
technology developer rather than a plant constructor, but is working closely with multiple industrial partners to develop the physical plants. Overall, the approach reflects a modular supply chain. Engineering, procurement and construction (EPC) would be undertaken by Technip Energies, while Schneider Electric would provide electrical systems and automation. The French CEA is a simulation partner, and Orano is responsible for fuel-cycle tasks. “We will not build the reactor ourselves,” Breyton says.
“We provide the core system.” He expects projects to vary by site, depending on whether customers require electricity, hydrogen, steam, or mixed outputs. In November 2025, Stellaria reached a first pre-order
agreement for 500 MW with Equinix for its AI-Ready data centres using its Stellarium design. Commenting on the agreement Régis Castagne, Managing Director of Equinix France said in a statement: “We chose Stellaria because it is one of the few companies in the world capable of making our high-performance AI data centres energy resilient, while combining high security and flexibility.“
Heat and power Stellaria’s commercial nuclear power unit is specified at 250 MWe per reactor, but the units will be built in pairs to share common infrastructure such as cranes and auxiliary
systems. Operating salt outlet temperature is planned at approximately 605°C. At these temperatures, standard superheated steam cycles can be employed. Breyton estimates an electrical conversion efficiency of “at least 45%, maybe 47%,” significantly above the ~33% typical of large light-water reactors. He attributes this higher efficiency to the better thermodynamic qualities of the fast- spectrum molten-salt coolant. Furthermore, because the steam conditions are
compatible with those used in coal-fired power plants, he emphasises the retrofit option. “The steam turbines are the same as coal power plants,” he says. Existing power blocks can therefore theoretically be converted by replacing the coal-fired boiler with the reactor modules placed in an excavated pit matching the original boiler footprint. Brayton suggests that as many as 250 coal-fired plants in European alone could be candidates for decarbonisation using this approach. A high-temperature heat output also opens the door to industrial applications for the design. Breyton describes potential deployments in hydrogen production or high- temperature process heat, siting reactor modules within industrial clusters to serve multiple users. “Very often those industries are together in the same area,” he observes. Industrial operators, Breyton says, increasingly seek
long-term power price predictability for capital-intensive facilities. “You know exactly the initial cost and can depreciate it easily,” he argues. The objective is a fully defined fuel cost trajectory for 20–40 years. He suggests that such conditions could even reverse industrial decline in regions with constrained grids: “If you have access to predictable-price power, you can secure a business plan and build factories.”
Challenges Ahead While scaling from a 10 MWth prototype to a commercial 600°C, 250 MWe fast MSR in five years clearly represents a significant step and major R&D tasks, Breyton argues that the main challenge is actually financial. “We need to have enough financial support to cross the desert,” he says, referring to the pre-revenue period before commercial reactors are built. He says both private and public funding will be necessary to complete both ALVIN and MEGALVIN. Regulatory licensing for a first-of-a-kind fast MSR
will also require unprecedented depth of materials and safety data, although Breyton notes that licensing submissions are already underway with the French regulator ASN. The Stellaria reactor combines established fast-spectrum
physics with numerous passive safety features with inherent negative feedback loops and passive drain-down. There are also multiple claimed specific advantages of the liquid fuel design including a homogeneous composition, long residence times and fuel breeding. While substantial engineering challenges remain, Breyton argues that given the physics is sound, industrial energy requirements are now sufficiently intense to justify pursuing a fast-spectrum molten-salt architecture. The next decade, beginning with the ALVIN experiment, will determine whether the engineering path he outlines can be realised. Breyton, though, believes that Stellaria reactors will do for 21st- century industry what hydroelectric dams did for 20th- century electrification. ■
www.neimagazine.com | December 2025 | 37
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
