They’re not. These reactors are going to be based on the liquid sodium moderated breeder design that the DoE built in the early ‘90s. While the liquid sodium moderator used in this design is tricky to handle, the reactor has several advantages: it has a high negative coefficient of reactivity, meaning that if the core gets too hot, criticality is lost and the reactor shuts down. Also, the sodium-moderated design allows the reactor to use fuel elements that are simply cast instead of precision-machined, thus drastically lowering fuel costs. Another advantage is the fuel itself, a plutonium/uranium mix: as the reactor ages, neutron flux from the plutonium gradually converts the uranium into more fuel, allowing the reactor to recover ±99% of the energy in each fuel element this means the core can be sealed, as one fuelling will last the lifetime of the reactor. This configuration furthermore transmutes the usual long-half-life “poison” byproducts of fission into short half-life isotopes, allowing the spent fuel elements to be safely stored on site. Finally, the alloy fuel elements cannot be reprocessed into weapons-grade nuclear material without the use of huge, heavy, easy-to-find-and-bomb centrifuges, thus greatly reducing the likelihood of proliferation.
I’ll be honest: liquid sodium scares the crap out of me. That stuff oxidizes like crazy, and if it touches water, well... don’t ask. However, the basic design of these reactors is sound and the prototype tested well, so I’m confident the new units will work.
Didn’t the Soviets play with Liquid-Sodium reactors on a sub or 2? I don’t think that they had a lot of success with it.