There has been quite a bit of talk in the scientific press recently about thorium and its possible use as a fuel for next-generation reactors. Let's take a look at the various pro's and con's of this option, and why it differs significantly from uranium.
The touted advantages include:
Firstly, the nitty-gritty: Any reactor using thorium will still be a fission reactor. As does uranium, it will produce highly radioactive, highly toxic fission byproducts. These byproducts need to be satisfactorily contained, such that there is absolutely no risk of their leaking to the environment. Therefore it could be said that from an environmental point of view thorium is less ideal than fusion, which produces little or no toxic waste.
Also, thorium cannot initiate a reaction on its own. To start the reactor from cold (in the nuclear sense) a certain amount of uranium or other more active fuel will be needed. Once running the thorium reactor is self-sustaining, however.
That said, thorium is an attractive option because it is an easier route than fusion engineering-wise, and although it is still fission with all which that implies, with suitable reactor design the safety and environmental factors are far more acceptable than for uranium. Ideally we'd like to do fusion, but that may take a while to develop. We can do thorium now, and at not much more cost than uranium. It's one of those cases where aiming for perfection is not necessarily the best choice if we could have a safer reactor now instead of a few decades hence, by setting our sights a little lower.
Thorium can in fact be burned in reactors similar to present-day uranium designs. However, that approach offers little improvement in safety or pollution control. To achieve the full benefits of thorium calls for a purpose-designed reactor. The favoured option is the Liquid Fluoride Thorium Reactor, or LFTR. (sometimes written as LIFTR) The fuel in this reactor consists of a salt mixture heated to between 400C and 800C, in which range it becomes a free-flowing liquid. The thorium component in the salt mix is the fuel, of course. The other components serve as the coolant, doing the job of the water in a uranium reactor.
Thus, instead of fixed fuel rods immersed in a liquid coolant, the entire fuel and coolant mix is pumped round and round the reactor core. This has the immediate advantage that fuel replacement is a very simple process, and one which can be performed anytime, even with the reactor still running. A further advantage is that the molten salt mix has a very high boiling point, typically over double the operating temperature. Thus, unlike water coolant, no pressurisation is needed to keep it from evaporating. There would still be a sealed, strong containment vessel, but in this case its function is purely one of keeping toxic substances safely inside. Unlike the 50-100 atmospheres inside a PWR containment vessel, it would not normally be under any pressure.
The higher temperature of the liquid salt also allows for a more efficient use of the energy than is possible with pressurised water.
Should a molten salt reactor overheat or get out of control, then the operator has a straightforward recourse of opening a valve to drain the liquid fuel out of the core into a holding tank. With no moderator in the holding tank the fission reaction stops immediately, and the now empty reactor poses no threat. The fuel in the holding tank will continue to release residual heat for several days, and thus will require some cooling. Passive cooling will likely be sufficient, though. Dumping the fuel in this way will not damage the reactor core, so the decision to dump can be taken as soon as any anomaly in operation is spotted. Since gravity flow will work just fine, the molten salt being about the same consistency as water, a circulating pump failure -as at Fukushima- will not prevent the fuel dump from proceeding.
It goes without saying that you can't do that with solid uranium oxide fuel, at least not until the solid fuel gets so hot -white hot- that it melts of its own accord. By which time your reactor is seriously damaged and you likely have a catastrophic fuel-cladding fire on your hands. With a fluoride salt, even if it were to become white-hot due to its own radioactivity it will absolutely not catch fire or release explosive hydrogen gas, because it is a chemically very stable substance.
All of the above is known to be feasible technology. One area which has yet to be researched is that of reprocessing or purifying fuel onsite. In principle this would consist of tapping-off a small proportion of the molten salt circulation into a processing plant which would separate the fission products -effectively the 'nuclear ash' of the reaction- from the remaining fuel, and then return the purified fuel to the reactor. The removal of the 'spent' fission products would prevent these from poisoning the fuel, as typically happens in a solid fuel reactor, and would thus allow almost all of a given batch of fuel's energy to be extracted instead of only a few percent.
The environmental advantages of onsite fuel reprocessing are substantial. Firstly, no need to transport hazardous cargoes to remote processing facilities. Secondly, far less and shorter-lived waste. In generating a gigawatt of electricity for one year, a thorium LFTR powerstation would produce 170kg of high-level waste. A uranium reactor, 35,000kg. Worse, the uranium reactor's waste would have to be stored in a high-integrity underground facility for thousands of years owing to the longevity of some radioactive components. The thorium reactor's relatively small amount of waste would be safe to recycle into the environment in a few hundred years.
So, LFTR may not be quite the perfect option that fusion could give us, but it offers a workable and environmentally-friendly solution that we can implement quickly if we set about it. The main issue will be in persuading investors to stake their venture capital on a non-mainstream development. Always a difficult task in view of the perceived risk of financial loss. In this case though, there is the carrot of a worldwide market for a successful design, certainly worth multiple trillions of dollars. Whoever gets there first will be in-for a bonanza of sales like nothing before it. Let's make it the UK which does that. We once designed and built our own nuclear powerstations, and some of those, like the AGRs which still provide most of our nuclear energy, were better, safer, designs than our competitors could produce. We can do it again if we have the resolve to succeed.