How a Thorium Reactor Works

The Economist:

Thorium itself is not fissile. If bombarded by neutrons, though, it turns into an isotope of uranium, ²³³U, which is. Thorium can thus be burned in a conventional reactor along with enriched uranium or plutonium to provide the necessary neutrons. But a better way is to turn the element into its fluoride, mix that with fluorides of beryllium and lithium to bring its melting-point down from 1,110ºC to a more tractable 360ºC, and melt the mixture. The resulting liquid can be pumped into a specially designed reactor core, where fission raises its temperature to 700ºC or so. It then moves on to a heat exchanger, to transfer its newly acquired heat to a gas (usually carbon dioxide or helium) which is employed to drive turbines that generate electricity. That done, the now-cooled fluoride mixture returns to the core to be recharged with heat.
This is roughly how America’s experimental thorium reactor, at Oak Ridge National Laboratory, worked in the 1960s. Its modern incarnation is known as an LFTR (liquid-fluoride thorium reactor).
One of the cleverest things about LFTRs is that they work at atmospheric pressure. This changes the economics of nuclear power. In a light-water reactor, the type most commonly deployed at the moment, the cooling water is under extremely high pressure. As a consequence, light-water reactors need to be sheathed in steel pressure vessels and housed in fortress-like containment buildings in case their cooling systems fail and radioactive steam is released. An LFTR needs none of these.
Thorium is also easier to prepare than its rivals. Only 0.7% of natural uranium is the fissionable isotope ²³⁵U. The rest is ²³⁸U, which is heavier because it has three more neutrons, and does not undergo fission because of the stability these neutrons bring. This is why uranium has to be enriched by the complicated process of centrifugation. Plutonium is made by bombarding ²³⁸U with neutrons in a manner similar to the conversion of thorium into ²³³U. In its case, however, this requires a separate reactor from the one the plutonium is eventually burned in. By contrast thorium, once extracted from its ore, is reactor-ready.
It does, it is true, need a seed of uranium or plutonium to provide neutrons to start the ball rolling. Once enough of it has been converted into ²³³U, though, the process becomes self-sustaining, with neutrons from the fission of ²³³U transmuting sufficient thorium to replace the ²³³U as it is consumed. The seed material then becomes superfluous and can, because the fuel is liquid, be flushed out of the reactor along with the fission products generated when ²³³U atoms split up.

I'm surprised nobody in the States has done much with this for a while.  In the "all of the above" energy policy, this would seem to be one of the above.  Especially considering that we are building a couple of light water reactors (with federal loan guarantees).