What about thorium?

Nearly two decades ago I took a short class on nuclear criticality safety – basically a class on how to work safely with things like enriched uranium. When we were reviewing the nuclear properties of U-235 (the isotope of uranium that fissions) I happened to notice that U-233 seemed to have a lot of advantages over U-235, and that U-233 could be fairly easily produced from natural thorium, virtually all of which is Th-232. Why this is significant is that, in a nuclear reactor, Th-232 will capture a neutron and transmute into U-233. From my geochemistry classes I knew that there is four times as much thorium on Earth as there is uranium, and that less than 1% of uranium consists of U-235 to begin with – this is why we need to go to all the bother of uranium enrichment. Given that it seemed possible to use the abundant thorium to breed fissionable U-233, I asked our instructor, why weren’t we using thorium for fuel? His answer was that the radioactive waste from U-233 fission was much more dangerous and harder to work with than that from “conventional” U-235 fission, and I was content to leave it at that.

Fast forward to two weeks ago when I saw a book called “Superfuel” in which the author, journalist Richard Martin asks fundamentally the same question I raised, but actually goes into the details. According to Martin, the US dabbled with thorium power in the 1950s before giving up on it in the 1970s. Not only that, but there is a thorium movement of sorts in the US, but even more so in India and China. The International Atomic Energy Agency (IAEA) has written about thorium power, and there are a number of papers about it in the scientific and engineering literature.

I’ve been going through a lot of this literature for the last few weeks, trying to condense it down into a single ScienceWonk posting. What I’ve found is that the pros and cons of thorium power are not as simple as one might be led to believe by those arrayed in favor or against thorium power. On balance it seems that there is a lot in its favor, but the case is not as simple as the pro-thorium advocates might have us believe. In any event, I’ve come to realize that the issues surrounding thorium power are far too much to summarize in a single posting – Martin struggled to address them all in his book – and doing justice to the topic really calls for several postings. So that’s what I’m going to do – in this posting I will briefly discuss the fundamental science and technology without a great deal of editorial commentary and in future postings I’ll try to address some of the advantages and drawbacks of thorium power compared to nuclear and other sources of energy. Since thorium power touches on issues of radioactive waste management, greenhouse gas emissions, proliferation, and more, a series of postings seems justified. So – on with the basic science, and we’ll take on the other issues in coming weeks!


The fundamental idea here is fairly straight-forward – if we put thorium into a neutron-rich environment then some of those neutrons will be captured by thorium atoms, transforming the natural Th-232 into Th-233. With a half-life of a bit over 22 minutes, Th-233 quickly decays to form protactinium (Pa-233, to be precise), and the Pa-233 decays with a half-life of just a tad under four weeks to form U-233, which fissions quite nicely. Or, put another way, load a reactor with Th-232 and within several months you’ll have a reactor full of U-233, which can continue to power the fission reaction. A thorium reactor is a breeder reactor, producing as much or more fuel as it consumes.

One of the nicest things about thorium fuel is that every atom of Th-232 can be turned into U-233 fuel. Not only that, but there is four times as much thorium on Earth as there is uranium. So think about these two facts – and remember that, of uranium, less than 1% is the fissionable U-235. This means that one gram of thorium has more than 100 times as many potentially fissionable atoms as a gram of natural uranium – and does not require a costly enrichment process – and Earth contains 400 times as much potential fuel from the thorium cycle as from uranium.

The way a thorium reactor would work in practice is that the whole process would have to be initiated by a sort of “starter core” fueled with uranium and the uranium core would be surrounded by a thorium blanket. Neutrons from uranium fission would be captured by the thorium blanket, breeding the U-233 that would fuel the reactor from there. If the thorium is in solid pellets it’ll be necessary to do some chemical processing to separate the uranium, but chemical processing is a lot easier than uranium enrichment. But an intriguing idea is to use liquid fuel – dissolve the thorium and the uranium into a liquid salt – and the liquid can simply be run through chemical processing on a recurring basis to not only separate out the uranium, but also to remove the fission products. Liquid fuel has both pluses and minuses – I’ll go over them in a later posting on this topic.

With this as a starter, here’s a little bit of a roadmap to future postings on this topic:

  1. Thorium reactors can be designed to be inherently safer than the current crop of water-cooled nuclear reactors
  2. The waste from thorium-cycle reactors is shorter-lived than that from U-235-fueled reactors, reducing the problem of radioactive waste disposal
  3. Thorium-cycle reactors have some advantages when it comes to proliferation resistance
  4. And liquid fuel reactors have some advantages over reactors with solid fuel rods.

There’s more, but we’ll leave it at this for the moment. Stay tuned for more postings!

Tags: ,

7 Responses to “What about thorium?”

  1. Mark L. August 10, 2012 at 12:17 PM #

    Thank you for doing this. I just finished reading Superfuel myself a few weeks ago. The LFTR concept seems very intriguing, but as a layman it’s hard to separate fact from hype.

  2. Jaideep A. Prabhu August 10, 2012 at 12:21 PM #

    How do you compare India’s AHWR (they claim they will start construction in 2016) with the LFTR?

  3. Mark Schanfein August 13, 2012 at 6:09 PM #

    We must also ask about whether there are any advantages from a proliferation standpoint. In this case, U233 is just as good a weapons material as Pu based on the IAEA significant quantity of 8kg for each.

    It also still takes U235 or Pu239 to get the Th fuel cycle into a steady state U233 economy; since no Th isotopes are fissile.

  4. Dr Y August 14, 2012 at 10:25 AM #

    Wow – thanks for all the thoughtful comments! I hope that the other postings in this series are as thought-provoking. Now let’s see if I can add anything to the discussion….

    With regards to Mark’s comment about separating fact from hype, this is a tough one. The author of Superfuel made no secret about his enthusiasm for thorium power, so the question is whether he reached his conclusions thoughtfully, or if he “drank the koolaid” of the thorium movement. Personally, I think that the technology is intriguing and it seems to offer some legitimate advantages over our current reactor design (more on these in coming weeks). But I also can’t help but wonder what problems will arise in a commercial-sized plant (or in a fleet of commercial plants) that we just can’t forsee with the trial-sized reactors that have been operated thus far. But it would be nice to build at least one sizeable thorium reactor to see what crops up, to get us past speculation and into fact.

    As Jaideep points out, India is very likely to be the first nation to test out the commercial-sized thorium reactors. But I have to say that their approach seems more circuitous and more complex than is called for. To my mind, a direct approach seems much better, but perhaps the 3-stage approach will give the nation the ability to start with existing mature technology before moving into less familiar ground. In any case, I know that the first thorium experimental reactors were running over a half-century ago. India is certainly far beyond that level of technology today and I see no reason why they cannot successfully build and operate reactors today. And, all things considered, I think that the lower-pressure thorium reactors will be much easier to operate and inherently safer to use than the water-cooled reactors currently under development.

    Finally, Mark has an excellent point about proliferation! It is true that U-233 is every bit as fissile as U-235 or Pu-239. In fact, I believe that the US exploded at least one U-233 bomb. So proliferation is a concern, just as it is with “conventional” U-235 reactors. On the other hand, there are some reactions with Th-232 that produce highly radioactive U-232, which complicates both radiation safety and weapons design (since U-232 doesn’t fission very well). In addition, it has been claimed that the U-233 can be “poisoned” to make it less useful for weapons design by diluting it with far less fissile U-238 (although you then have the opportunity to breed Pu-239 from the U-238). Finally, a Th-232 atom needs to capture a total of 7 neutrons to become Pu-239, compared to only 1 neutron capture from U-238 – this puts it much farther away from plutonium than is uranium fuel, making plutonium production much more difficult. I have to look more into this – that’ll be the subject of a future posting on this topic – but at the very least it seems reasonable that the thorium fuel cycle is no worse than uranium from a proliferation standpoint, and likely offers some advantages. Oh – and you’re also correct that the whole thing would have to be kicked off with a uranium reactor to start the whole thing going – but after you start to breed U-233 from Th-232 the reactor will run on that and no more U-235 or Pu-239 would be required.

    Anyhow – I know that I haven’t done justice to any of your comments in this short reply. But I will try to address each of these topics in full-length postings in the next few weeks. And thanks for your interest!

  5. Tim W. August 23, 2012 at 12:13 PM #

    From what I have understood of presentations on LFTR, the U233 produced is always contaminated with small amounts of U232 which is a hard gamma emitter. It’s difficult to handle, makes weapons unstable, and acts like a beacon telling everyone where it is. Pu239 is far easier to work with and hide.

    As far as bootstrapping is concerned, the USA has a stockpile of U233 it can use that’s slated to be down-blended and made useless… but nuclear waste stockpiles could also be used as a less efficient neutron source if we do scrap our U233 supply. Particle accelerators are another neutron source in consideration for creating U233 from thorium.

  6. Jim March 28, 2013 at 12:25 AM #

    In the long run it’s not a question of either or, for both U and Th can be and will have to be used. Both have good points and bad, but the energy contents are just to great to ignore. That is unless the perpetual 50 year forcast for the arrival of commercial fusion power reduces to now.


  1. What about Thorium? » Teaching Nonproliferation - August 14, 2012

    [...] http://www.fas.org/blogs/sciencewonk/2012/08/what-about-thorium/ Share Energy Issues, Exclude from Homepage, Nuclear Power Translate this Page function googleTranslateElementInit() { new google.translate.TranslateElement({ pageLanguage: 'en' }, 'google_translate_element'); } Index hiroshima enrichment Europe Japan nuclear world energy iran fukushima nuclear weapons korea MOX geothermal EGS solar power UK terror nuclear power wind government subsidies fission reprocessing three mile island soviet union energy education nuclear nonproliferation treaty nagasaki Sahara united states renewable energy fusion Topics [...]

Leave a Reply