Solid vs Liquid
Thorium/U233 cycle is a breeder cycle, and like all breeders there's a requirement for timely chemical processing of the reactor core.
Thorium is more dependent on this than the U238/Pu breeder cycle because thorium involves an intermediate which can be poisoned by neutrons if it remains within the core. Conventional reprocessing of solid fuel can work for U238/Pu, but the Indian design of solid thorium oxide in among the fuel is, apparently, too hard.
So the approach that's being talked about now is an old reactor design where the nuclear materials are disolved as fluoride salts in molten lighter fluorides. The core is just a cauldron of the molten salts with a moderator, and the same material fills the primary coolant loop.
This approach has a number of intrinsic safety features as the core can swell out of the moderator (and thus throttle down) when it gets too hot and can flow into subritical, cooling efficient shapes to shutdown with passive cooling. And because it's not using water it's not pressurised, so it won't explode in the TMI/Fukushima style.
But breeder designs benefit from liquid core for a different reason. Liquid chemistry is easier than the engineering needed to open up fuel rods and dissolve them in nitric acid. The re-processing can be done conveniently at the plant, little and often, without shipping rods and radioactive stuff all over the country. I'm guessing that a liquid core is required to make thorium/U233 practical.
So this one approach offers:
- The huge waste reduction and fuel efficiency of breeders
- Passive walkaway safety of liquid core
- And, as a bonus, it'll run at temperatures that can run efficient air-cooled gas turbines, so they can be built on small sites inland.
You can find any amount of stuff. Google "LFTR" or "MSR Thorium" to find stuff from enthusiasts like Kirk Sorenson (who was in London last week BTW).