Power · Fluoride-salt SMR
Rung three · no date
Fluoride-salt SMR
Told as physics rather than as machinery: what the coolant does, what the fuel does, and why the order of the ladder puts this last.
A fluoride salt-cooled high-temperature reactor carries its heat in a molten salt at close to atmospheric pressure, and burns fuel that arrives as ceramic-coated particles rather than metal-clad rods. Those two choices are the whole design, and everything else on this page follows from them. Xalant is not building one. This is the rung the site reserves land for and does not wait on.
- Bridge
- Geothermal
- SMR
It comes last because it needs a licence, a vendor agreement and a cost model proven against the bridge — not because the physics is unsettled. Every other rung is designed so the site never has to wait for this one. Why this order
Power · SMR → The machine
Four parts
Four parts, all of it held near atmospheric pressure.
Take the pressure out of a reactor and a great deal of the engineering around it changes shape. The vessel stops being a thick pressure boundary, the worst failure stops being a sudden loss of pressure, and the cooling that matters most stops needing a pump.
01Reactor vessel
A thin-walled can rather than a thick pressure vessel, because the coolant inside it is a liquid at close to atmospheric pressure. Nothing inside is trying to escape at high pressure, so nothing outside has to hold it in.
02Fluoride salt loop
Molten fluoride salt carries heat out of the core, through a heat exchanger and back. It is a liquid across a very wide temperature band, which is why it does not have to be held down by pressure the way water does.
03Pebble fuel core
Fuel arrives as graphite spheres, each packed with hundreds of thousands of ceramic-coated particles. Pebbles circulate through the core and are inspected one at a time rather than shut down for refuelling.
04Passive-cooling stack
Decay heat leaves by natural circulation — hot fluid rises, cool fluid falls — with no pump and no operator action in the loop. This is the part of the design that does not depend on anything still working.
Power · SMR → Why salt
The coolant
Why a molten salt instead of water.
Water is an excellent coolant with one expensive property: to stay liquid at useful reactor temperatures it has to be held at very high pressure.
Everything follows from that. The vessel becomes a thick pressure vessel. The piping, the pumps and the containment are all sized for the pressure rather than for the heat. And the accident everyone designs against is the sudden loss of that pressure, because the coolant flashes to steam and stops cooling exactly when it is needed most.
A fluoride salt is already a liquid across a very wide temperature band without being squeezed. It runs at close to atmospheric pressure, so there is no stored pressure trying to escape and no vessel built to contain one. It also carries a lot of heat per unit volume, which is why the machine can be small.
The honest costs of that choice belong on the same page as the benefits. The salt has to stay hot or it freezes, which is a real operational constraint rather than a footnote. Its beryllium content is genuinely toxic and is handled accordingly — a materials and handling problem, and one that any serious account of this design names out loud. And the high outlet temperature that makes the thermodynamics attractive is exactly what makes the materials difficult.
Power · SMR → The fuel
One particle, five layers
Each particle is its own containment.
TRISO stands for tri-structural isotropic. A speck of uranium is wrapped in a carbon sponge, sealed with two shells of dense carbon and armoured with a layer of silicon carbide — and it is that ceramic shell, not the building, that is the first pressure boundary.
01Fuel kernel
The uranium itself, at the centre — a sphere smaller than a grain of sand.
02Porous carbon buffer
A sponge of carbon around the kernel, there to give fission gases somewhere to go instead of building up pressure.
03Inner pyrolytic carbon
A dense carbon shell that seals the buffer and protects the layer above it while the particle is being made.
04Silicon carbide
The load-bearing layer: a ceramic shell that is the actual pressure boundary. It is the reason the particle is described as its own containment.
05Outer pyrolytic carbon
A final carbon shell that takes the handling, bonds the particle into the graphite matrix of the pebble, and shields the ceramic beneath it.
Hundreds of thousands of these particles are bound into a graphite sphere the size of a large marble. Pebbles circulate slowly through the core and can be inspected one at a time, which is a very different operational rhythm from shutting a reactor down to change a fuel assembly.
The safety argument the design rests on is about temperature, and it is worth quoting rather than paraphrasing — trimming a bounded engineering statement into an absolute one is how honest claims become dishonest ones:
Kairos Power states that TRISO fuel “cannot melt inside the reactor”.
That is Kairos Power’s wording about Kairos Power’s fuel, reproduced whole because the qualifier is the claim. Xalant makes no safety claim about any reactor, and no reactor exists on any Xalant site.
Power · SMR → The pathway
The licence is the long pole
What the regulator actually requires.
In the United States a power reactor is licensed by the Nuclear Regulatory Commission under one of two frameworks. Both are measured in years, and both start long before anything is filed.
The NRC pathwayMeasured in years
- Pre-application engagementA vendor talks to the regulator long before filing anything: topical reports on the parts of the design that are new, and agreement on what will have to be demonstrated.
- The applicationPart 50 splits it in two — a construction permit first, an operating licence later. Part 52 combines them into a single combined licence. The route changes the shape of the review, not its depth.
- Safety and environmental reviewA technical safety evaluation, an environmental review, and hearings on the public record. This is the stage measured in years, and it is measured in years deliberately.
- PermissionPermission to build, and separately, permission to operate. Nothing is built on the strength of an expectation.
Where Xalant standsNowhere, and that is the point
Xalant holds no construction permit, has filed nothing with the Nuclear Regulatory Commission, and has no application in preparation. There is no reactor vendor agreement. The third rung is a prospect for the decade after this one, and the entire ladder is designed so that no site ever depends on it arriving.
What does exist is a public record belonging to somebody else. Kairos Power has been issued construction permits by the Nuclear Regulatory Commission for its Hermes demonstration reactors at Oak Ridge, Tennessee. That record is Kairos’s, it is published, and it is linked here rather than summarised so you can read the primary source. kairospower.com
Kairos Power appears on this page as a technology reference for a reactor class. No commercial agreement between Xalant and Kairos Power exists, and none is implied by anything on this page.
Power · Fluoride-salt SMR → FAQ
Asked before anyone asks
Questions worth answering honestly.
- Is Xalant building a reactor?
- No. There is no reactor, no vendor agreement, no licence application and no site. The third rung is a next-decade prospect, and every reference campus is designed to run on bridge generation and geothermal without it. If that changes, it will change because a regulator and a vendor have both said yes in public, and you will read it here with the paperwork attached.
- Why fluoride salt rather than water?
- Because of what pressure does to a design. Water has to be held at very high pressure to stay liquid at useful temperatures, so the vessel around it is a thick pressure vessel and the worst accident is a loss of that pressure. A fluoride salt is already liquid across a very wide temperature band at close to atmospheric pressure, so there is no stored pressure trying to escape and the vessel does not have to be built to contain one. The fuel is the second half of the case: ceramic-coated particles, each one its own containment.
- What would have to happen before a reactor could sit on a Xalant site?
- A licence from the Nuclear Regulatory Commission under Part 50 or Part 52, which is a multi-year process with a safety review, an environmental review and hearings on the record. An agreement with a reactor vendor, which does not exist. And a site whose land was set aside for the pad from the first drawing, because retrofitting a reactor pad onto a finished campus is not a thing anyone does. Any company naming a date before the first two have happened is telling you about its ambitions, not its schedule.
Power · Next step
Tell us the megawatts. We will draw the rungs.
Model the stack against your own assumptions, then send it as a capacity request with the inputs already filled in.