Thorium – the kinder, safer nuclear power?

My problems with nuclear power are more or less the standard worries about meltdowns and waste. In my opinion, those problems will get worse as the planet warms. I also think that we can’t afford to discard nuclear power out of hand. There are some versions of it that seem better than what we have now, including Liquid Fluorine Thorium Reactors(commonly referred to as LFTR):

If you found this post useful or enjoyable, please share it, and please consider becoming a patron over at my Patreon page. Your donations make this blog possible, and even as little as one dollar per month adds up to make a difference. If you feel you can afford more than that, you can get access to all sorts of other content and perks! Your patronage allows me to put more of my time and energy into making this blog a useful resource. Thanks for reading!



  1. Dunc says

    LFTRs are an interesting idea, but my understanding is that there’s a number of potentially tricky engineering details still to be worked out, and that nobody’s even built a prototype yet. It’s also my understanding that taking a technology like this from from prototype to full-scale build-out usually takes decades… And we don’t have decades.

  2. Dunc says

    Fusion’s still around the corner. :/

    Yeah, it’s been 20 years away for my entire life. (And I’m quite a lot more that 20.)

    Funnily enough, the last time I saw somebody put a number on how long it would take to get LFTRs up and running, that number was… 20 years.

  3. Callinectes says

    I thought the reason Thorium reactors were never developed was because you can’t get a weapons program out of it.

  4. says

    I thought the reason Thorium reactors were never developed was because you can’t get a weapons program out of it.

    That may be pretty much correct, unfortunately.

    I always suspected that’s why pebble-bed reactors didn’t get popular – because a nuclear power could actually sell them and they’d no longer have “non-proliferation” as a club to keep other countries under fossil fuels. Maybe now that the US has lost control of the solar industry, that will change.

  5. says

    Funnily enough, the last time I saw somebody put a number on how long it would take to get LFTRs up and running, that number was… 20 years.

    Yeah. To be fair, the expenditures on fusion research (compared to oil exploration…) are very small. The planet should be collectively funding fusion as if it’s our last chance for survival…

  6. Art says

    Two thoughts:
    The old joke is that fusion is twenty years in the future, and always will be.

    Thorium was, and is promising. It has been held back by institutional inertia. Once we found something that worked the machine picked up speed. Light water worked for the military. Yes, you could make bombs, but the main thing was it worked and, for small reactors operated with strict discipline, it has a very good safety record.

    Incidentally, the military nuclear program is still, and always has been, the main feeders for nuclear operators and engineers. This locked in the thinking. GE was heavily invested and jealously undercut any developments that might undercut its investments. GE and related nuclear industries also supplied a lot of the materials and funding for the few civilian nuclear programs.

    Thorium, presented as ‘safe’ is not entirely safe. It is safer than uranium but if you breath the dust or ingest it can be cancer causing, and potentially deadly. IMHO even thoriated TIG electrodes, about 2% thorium, are handled far too casually by most welders. Particularly when grinding.

    Nothing is completely safe, driving is still far more hazardous, and sensible precautions can go a long way, but there are still issues with even this safer form of nuclear power. I think they skated past that point a little too fast.

  7. Raucous Indignation says

    I agree Art. Safer than uranium is an incredibly low bar. An any couldn’t walk under that sucker.

  8. wereatheist says

    I’m just having fun imagining this discussion with German participants 🙂

  9. wereatheist says

    As for, “you can’t get a weapons program out of it.” (Callinectes), of course you can. But the bombs might well be some bigger than with Pu.

  10. Allison says

    I used to work in nuclear fusion research (mainly magnetic confinement.) The “20 years” thing was bogus and we knew it. The real answer as to how long before we can get usefull amounts of electricity (or steam) out of it has always been “damned if I know,” but you can’t say that to the funding agencies.

    The thing is, getting controlled nuclear fusion is an engineering nightmare. Each step of the way has involved lots of unforseen problems, and each has taken considerable effort to manage or work around, and each time, the whole system becomes more complicated. Fission is easy: you just bring the fissionables together and off they go. Fusion is very, very hard. The only relatively straightforward approach is gravitational confinement — what the sun uses. A little hard to recreate on Earth.

  11. EnlightenmentLiberal says

    Let me add my bits in:

    I believe that LFTR as a breeder reactor (turning thorium into uranium 233 into energy) is not yet ready. Kirk Sorensen claims to have worked out the details, but won’t release it publicly to keep a competitive advantage – but I’ll believe it when I see it, e.g. when the details are made public.

    However, the idea of a liquid fueled, molten salt, non-breeder reactor is quite close to ready. Companies like ThorCon and Moltex and others are quite close to commercial prototypes. From my reading, the main barriers seem to be government regulation and funding. I used to be a regular non-expert participant of some forums with some of the experts of ThorCon, and I trust them when they say that they’re almost ready for full scale commercial prototyping, and most / all of the engineering details are already worked out. Their motto is “no new technology” because they rightly recognize that they need to act now – now in order to combat climate change, and now in order to win emerging markets instead of coal – both as a business opportunity, and to also combat climate change.

    What about limited fuel supplies? Let me paraphrase the sales pitch of ThorCon: Even if the naysayers are right (and they’re probably not), that’s still 20 years of fuel supply, which is still 20 years more than what we have now to figure out what to do with global warming. I’ll take those 20 extra years please. That’s 20 more years of battery R&D, and 20 more years of R&D into a molten salt breeder reactor.

    Nothing nuclear is entirely safe, and nothing nuclear is entirely proliferation resistant.

    At the start of the modern renaissance of thorium power and molten salt reactors sparked by Kirk Sorensen, many claims were made that molten salt reactors could not be use to make nuclear weapons. I think many advocates in the movement have walked back on those claims, but I think many still occupy a reasonable middle ground which is something like “you might be able to make weapons material with one of our for-civilian-use molten salt reactors, but it may be easier to just a make a new reactor from scratch or to use centrifuges if your goal is weapons material”, and I think that’s a defensible position.

    Molten salt reactors are probably much safer and much cheaper than conventional light water reactors, which are already quite safe. Conventional light water reactors are safe, but that often comes with increased costs. Also, the nature of conventional light water makes it more expensive compared to a molten salt reactor – the super high pressures of a conventional light water reactor incur large costs, and a molten salt reactor operates at relatively low pressures, which means it avoids the costs and some of the accident scenarios that are present with super high pressures. Also, the lack of water, and the lack of high pressure water, remove the primary way that radioactive material gets into the atmosphere; there’s no obvious driver for how an accident would cause an airborne release of radioactive material in a molten salt reactor.

    For example, it’s said that even if you had a bomb explode in a molten salt reactor, it appears as though that the really bad stuff, the strontium and cesium, would stay in solution in the salt, and the salt would just cool down wherever it happened to land, keeping the strontium and cesium at the site. Yes, the cleanup might be a huge costly mess, but the damage would be localized to the plant. For my relatively untrained and uneducated person, this is the biggest selling point of a molten salt reactor. In one of the presentations of ThorCon, the presenter makes an engineering joke like “we actually have some design problems when it comes to recycling because of this – the strontium or cesium in the salt won’t even boil until 2500 C or something ridiculous”, e.g. a huge benefit for safety. (I forget the exact number, but it was something ridiculous like 2500 C).

    Are molten salt reactors magic? No. I think in the near term, if we’re lucky, a company like ThorCon will be able to produce them at costs comparable to coal, with a better safety record than light water reactors (whose safety record is already quite good). I don’t think that’s magic. I think it’s quite plausible. I hope that ThorCon and others get their chance to demonstrate their stuff. I’m like 80% confident that it will work as advertised, which is definitely worth the government investment of the 1 billion dollars or so to build and test that commercial prototype.

    It’s not like fusion at all.

  12. EnlightenmentLiberal says

    To respond to some points in the comments:

    To Dunc:
    From my knowledge, A breeder molten salt reactor has some outstanding technical issues, specifically the core design. Either you go high or low power density. For a high power density, the neutron flux will destroy any barrier between the fuel salt and the breeding salt very quickly, and for a low power density, well that increases fuel inventory and costs. (In fact, the fuel inventory may be so great as to preclude starting enough reactors worldwide fast enough to matter for global warming.) I think this is the primary issue which has not yet been satisfactory solved. Last I checked, Kirk Sorensen claims to have solved the problem, but he hasn’t released it to the public, and so I’m dubious.

    As I wrote in my comment above, several companies appear to be ready for commercial prototyping right now for non-breeder versions. We don’t need a breeder right now. A non-breeder would be great.

    To Callinectes
    The reason why we’re not using molten salt reactors appears to be a little more complicated. This is my understanding of the history. The original nuclear research was to make a bomb.

    They saw two obvious ways to make a bomb: uranium enrichment and a reactor to make plutonium. They experimented with both, and they produced both kinds of bombs, and tested both on Japanese cities.

    Then, Admiral Rickover from the navy took over, and he championed light water reactors for nuclear submarines. Nuclear submarines predated civilian nuclear power plants, and nuclear submarines have a different set of requirements than civilian nuclear power plants. This decision may have been a fine technical decision. However, later, when it came time to decide how to build nuclear power plants for civilian electricity, the natural and easy and cheap choice was to use the same technology. The navy had already figured out a lot of the practical problems, and they also had a bunch of operators coming out of the navy who knew the technology and who could run the civilian reactors because it was the same technology.

    Later, there was still research being done on next-gen reactors. There were two research programs going on, the liquid metal breeder program, and the molten salt program at Oak Ridge National Labs. Nixon wanted to make some budget cuts, and he wanted to benefit his home state of California, and the liquid metal breeder program was happening in California and the molten salt reactor program was not, and so he cut the funding of the molten salt reactor program. At least, that’s my highly limited understanding of it. I think there was also some particular internal politics problems around the head of the molten salt program – reportedly he was too critical in meetings about the safety of light water reactors, and he championed the molten salt reactor as a safer replacement for civilian use, and that earned him some internal political enemies which hurt him in his battle for continued funding.

    To complete the story, Carter ended all reprocessing work as an international diplomatic gambit in order to convince other countries to not do reprocessing, in order to combat nuclear weapons proliferation. That put a stop to any more research on breeder reactors. Later, I forget when, reprocessing research continued.

    Tangent: Then, Clinton came along, and to fulfill campaign promises and kowtow to the Green party, he killed funding on the liquid metal breeder reactor research program called the Integral Fast Reactor, which was truly a shame. There was even a battle in the senate IIRC over this where one of the senators was informed about this, and tried to stop it, and pointed out the lies from Clinton about it, but he lost in the end, and funding was cut. It was truly a shame because they were so close to finishing their work. Right now, I think GE is trying to sell the same sort of idea under the name S-PRISM reactor. The liquid metal breeder reactor isn’t a molten salt reactor, and I know less about them, but the particular reactor design called the Integral Fast Reactor, developed by Argonne National Labs, also seems really cool. It seems to be much safer than conventional light water reactors, and it’s also a breeder which solves any fuel supply problems.

    Finally, as Art mentions, why did no other country do it? Institutional inertia is huge. We’re talking about things that can only practically be R&D’d by governments or mega-corps. Loosely, for several reasons, including: every other country assumed that the United States was the best and knew what they were doing, and so they always approaches people asking for money with the question “if it’s so good, then why isn’t the United States doing it?”. Also, the green movement has created a hostile regulatory environment, which means that funding such a program is a very risk proposition for a private investor – they might lose all of their money if the greens get more power and pass more oppressive regulations, and so it’s quite expected that it hasn’t been bankrolled by any private investor.

    To Marcus
    I think pebble-bed reactors have their own sets of problems. Specifically, what do they do about pebbles that corrode or break, and also they need to move the pebbles, and moving a bunch of pebbles under high radiation is difficult – what if the pebble gets stuck?

    Regarding funding, I have never said stop all funding of solar, wind, etc., even though I think they won’t work. I’ve always been a proponent of a “all of the above, please” approach to R&D. The US has spent like 20+ billion dollars on solar and wind subsidies for deployment – not R&D, deployment. I just want a few of those billions to be diverted to commercial prototyping for reactors like ThorCon, Moltex, and S-PRISM.

    To Art
    I agree that nothing is safe, and nothing nuclear is completely safe, but I think you said it yourself. It’s safer than driving. The point that I try to make is that more people die premature deaths from coal worldwide every day than have ever died from radiation poisoning from civilian nuclear power. Almost all radiation deaths from civilian nuclear power are from Chernobyl, and 3000 deaths is a reasonable high estimate for the total death count. Other estimates say it’s only about 300. The W.H.O. is one source. Worldwide, about 3 million people die premature deaths from particulate pollution from coal, again citing the W.H.O. Three million per year is about 8219 per day, and 342 per hour. Deaths that could be entirely prevented by increased use of nuclear.

    Nuclear power is the safest thing since sliced bread – I think it’s true that more people have died choking on sliced bread than have died from radiation poisoning from civilian nuclear power plants.

    Hell, a single dam accident in China killed about 171,000 and displaced 11 million more, which is way more than all nuclear power plant accidents.

    The only part that concerns me is the potential longlasting loss of land, especially farmland. However, I think that compared to the alternative, global warming and ocean acidification, nuclear is the best answer, especially with next-gen tech like ThorCon, Moltex, and S-PRISM which should substantially increase the already stellar safety record of nuclear.

  13. Dunc says

    The US has spent like 20+ billion dollars on solar and wind subsidies for deployment – not R&D, deployment. I just want a few of those billions to be diverted to commercial prototyping for reactors like ThorCon, Moltex, and S-PRISM.

    I’d rather we diverted the money from somewhere else, personally… If it’s only a billion dollars or so, that’s pocket change for the Pentagon. But I’m all for getting prototypes built if the technology is really there. Heck, Vinod Khosla (supposedly) personally invested more that 10% of that in speculative biofuel ventures which never looked likely to work…

Leave a Reply

Your email address will not be published. Required fields are marked *