Some folks suggest that the answer to the problems of nuclear power stations is to change to using thorium as fuel instead of uranium. There are really four myths involved. The claims are that:
1. Uranium reactors produce large quantities of long-lived radioactive waste, whereas thorium reactors only produce short-lived radioactive waste;
2. Uranium reactors only use a small percentage of the uranium, whereas thorium reactors would use all the thorium;
3. Thorium reactors can’t produce bomb material;
4. Thorium reactors are simple.
All nuclear reactors, apart from fusion reactors (which have their own serious problems, see Nuclear Fusion?), produce large quantities of radioactive fission products. The quantity produced per unit of energy generated is virtually the same regardless of reactor type.
In addition to fission products, reactors whose primary fuel is uranium also produce radioactive actinides. Reactors whose primary fuel is thorium would in fact produce some actinides too, but in smaller quantities.* (The expression primary fuel needs some explaining – I will be writing a page on nuclear reactor types shortly which will clarify this.)
Some of those actinides have very long half-lives, and in the long run the radioactivity of the waste from uranium reactors will be dominated by the radioactivity of the actinides.
The half-life of the fission products is generally much shorter.
Much shorter – but to call them “short-lived” is grossly misleading. Some of them are indeed short-lived – seconds, hours, or weeks – but many of them will be around for centuries. The “short run” during which they will dominate the radioactivity of the waste is at least several hundred years. For at least the first several hundred years (and very likely a millennium or two, depending on the design of the reactors and any associated reprocessing plants) the waste from a thorium reactor would be just as bad as that from any other fission reactor.
There are several types of nuclear reactor. The simplest types do indeed only use a small percentage of the fuel. All commercial reactors are of these types. They cannot use pure thorium, although they could use a small percentage of thorium. There might be a marginal advantage in this.
Several types of experimental reactors exist, or have existed, that can use all, or almost all, of the fuel, whether it’s uranium or thorium or a mixture, but they’re much less well-established than ordinary reactors, and the safety and reliability of them and the necessary associated fuel reprocessing plants are highly questionable. They’re certainly not going to be ready to produce a significant percentage of our electricity – never mind our overall energy consumption – on any timescale relevant to the global warming crisis.
You couldn’t make a thorium bomb any more than you could make a natural uranium bomb. But thorium reactors don’t burn thorium – they burn uranium 233 (233U) which is produced from thorium by neutron irradiation, just as plutonium is produced from 238U by neutron irradiation. No-one has ever built a pure 233U bomb, but there’s no theoretical reason why one couldn’t be made. 233U is fissile, like 235U or 239Pu, which is why it can, like them, be used as fuel in a nuclear reactor – and for the same reason it could, like them, be made into bombs.†
No power-generating nuclear reactor is “simple.” Even the simplest ones involve fairly advanced engineering if you try to make them reasonably safe. Those simplest reactors can’t use more than a few percent of thorium in their fuel. The engineering of reactors that could use more than a few percent thorium (breeder reactors) is so demanding that they’re still experimental, and still have a poor safety and reliability record. There’s been a lot of work on them – mainly with a view to using them to breed 238U into plutonium fuel, not thorium into 233U. The problems with thorium are rather worse than with 238U, because 233U fission produces fewer neutrons than Pu fission does.‡
(N.B. These advantages would not make thorium power preferable to non-nuclear power – just slightly less bad than uranium power.)
It’s true that there is more thorium in crustal rocks – about three times as much of it, in fact. However, as with uranium, most of it is at very low concentrations in a wide range of rocks. How much is in reasonably concentrated ores is less clear. (There has been some work on extracting uranium from seawater. The concentration is a mere 3.3 parts per billion – but the concentration of thorium in seawater is far lower still: just 0.04 parts per trillion.)
There would be real safety benefits to thorium, for the miners and those living in the vicinity of the mines: it, and its ores, are significantly less radioactive than uranium and its ores. But only significantly less; they are still radioactive.
In the far distant future, when the fission products have decayed away to insignificance, the waste from thorium reactors would be significantly less radioactive than any unreprocessed used fuel from uranium reactors. (Reprocessed used fuel from uranium reactors would be just as harmless by then as that from thorium reactors – and you can’t run a thorium reactor without a very similar process to extract the 233U from the thorium.)
* Thorium enthusiasts like to talk about transuranics, which are a subset of actinides, because most (not all) of the actinides produced by thorium reactors would not be transuranic. This exaggerates the advantage of thorium, which is real enough anyway – for our very distant descendants, if not for us or our less distant descendants.
† The USA has exploded a bomb using a mixture of 233U, 235U and 239Pu – which, incidentally, had a very peculiar and interesting history – see Operation Teapot, MET test:
‡ See Fission Neutron Yields.