Thorium Reactors

  Thorium Reactors: Fact and Fiction

These next-generation reactors have attracted a nearly cultish following. Here are the real facts.      

by Brian Dunning


Today we're going to take a second look at a technology that, in the past few years, has become something of a cult icon. Thorium reactors, some say, can provide limitless energy; with a fuel that's cheap, safe, and abundant; using reactors that can't melt down, can't be used for nuclear weapons production, and produce almost no appreciable waste. Such a system would seem to be a virtual miracle. So today we're going to examine these claims, and see if we can find some facts in (what appears to be) an inordinate amount of hype. Cultish support for anything should always raise a red flag or two.

First, I have two disclaimers for you. One is that the technical discussions in this episode are oversimplified. They have to be. It takes decades to design a nuclear power plant, and we can't make a comprehensive comparison in only a few minutes. So don't email telling me that I grossly oversimplified something. I'm already telling you now that I will.

Second, there are far more different types of reactors than we could possibly address. There are a lot of basic reactor designs. There are different fissile materials that can be used as fuel. There are different fuel cycles. There are fast reactors versus thermal reactors, and burners versus breeders. Some designs use fuel in solid rods, and some have the fuel dissolved in the cooling liquid itself. Some are high pressure, some are low pressure. Most of these variations can be put into any combination, resulting in more designs than you can shake a stick at. So when you say "nuclear reactor", that's an almost uselessly vague term. And when we talk about thorium reactors, we're only narrowing it down somewhat. There are still competing types with different sets of benefits and drawbacks.

About the only thing they have in common is the basic idea, which is exactly the same as geothermal energy. Heat from radioactive decay — either from the Earth's mantle or in a reactor core — is used to boil water and turn a steam-powered generator. We've just synthesized the Earth's natural process to a point where we can optimize and control it.

Uranium-238 is the one that does most of the Earth's underground heating, because there's a lot of it. Hold a chunk in your hand and it's very safe, because it's quite stable and barely decays at all. But a bit less than 1% of it is fissile Uranium-235, which is what we need for fuel. Reactor fuel has two basic parts: a "fissile" material that emits neutrons, and a "fertile" material that absorbs the neutrons and continues the cycle. For uranium fuel, we take that natural uranium, separate out a lot of the U238 to create "enriched uranium" which has about 4% of fissile U235, and the rest is fertile U238. This gets combined with other material to make ceramic pellets which are stacked into metal rods, and these are the familiar fuel rods that most of today's reactors use. Installed in a reactor core, those rods are very hot because of the reactions happening within them, and they boil the water and make electricity. After about six years, a typical 2-meter fuel rod is no longer hot enough to run the reactor, and it becomes nuclear waste. It's still very hot, and remains so for decades.

Inside that rod, a lot of that U235 has decayed by spitting out heat and neutrons. Some of the U238 has captured those neutrons and become plutonium, which also decays. The uranium decays into thorium, which decays into protactinium, which decays into actinium; we get more thorium, francium, radium, radon, polonium; the chain goes on and on. That spent fuel rod ends up with just about anything you can imagine in there. It's a lot of waste, though some of it can be recycled via any of numerous costly and inefficient processes, after enough time has gone by. That whole process is what we call the uranium-plutonium fuel cycle.

So now let's contrast that with the thorium-uranium fuel cycle, that has so many people excited. Thorium-232 is what's found in nature. However, thorium is not fissile. Thorium is the fertile element of the fuel; it still needs a fissile element to get it started. And here is one place the thorium-uranium fuel cycle differs from the uranium-plutonium fuel cycle. Most thorium-fueled designs are breeder reactors, meaning they produce more fissile material than they consume. So once the thorium fuel cycle is started by adding fissile U235, theoretically, no more fissile material will ever need to be added. We can continue adding only fertile thorium. This arrangement generally works best in a reactor type called an LFTR (pronounced "lifter"), a liquid fluoride thorium reactor. The thorium, and all the other elements that are part of the fuel cycle, are dissolved in molten fluoride salts.

For today's purpose we're basically comparing LFTRs to LWRs, light water reactors, the general category of the majority of reactors worldwide, in which uranium fuel rods are immersed in regular water. Thus, this is not a head-to-head comparison of uranium fuel to thorium fuel, because such a comparison doesn't really make sense; there are too many other important design elements.

So on to the popular beliefs. Here are some true-false questions for you, LFTRs vs LWRs:
True or False? Thorium reactors are inherently safe.

True, but this is more a feature of the molten salt class of reactors as a whole. It can't "melt down" because the fuel is already molten. The molten salt can't burn or boil away. It has to be kept hot by adding fertile elements, or else the reaction stops. Most interesting are the freeze plugs, drainage holes at the bottom of the reaction vessel, kept closed by active cooling systems that solidify the molten salt. If power is lost or there's any other problem, that active cooling stops, the freeze plugs melt, and the molten salt flows into a series of underground tanks too small for the fuel to remain critical. Basically, it takes constant attention to keep the thing hot; it has nothing that can go out of control as with LWRs.
True or False? Thorium reactors produce hardly any waste at all, and the waste they do produce is much safer.

True. This is mainly because of the inefficiency of uranium fuel rods. They are done being used when they're about 2% spent. A lot of material goes into uranium reactors, and a lot comes out, and it's still highly radioactive.

In a LFTR, what goes in also comes out, but it's completely burned rather than only 2% burned. So in total, far less material goes through the system, and what does come out has had almost all its radioactivity already spent; and the majority of that can be recycled into useful industrial products.
True or False? Thorium is cheaper, easier to get, and far more abundant than uranium, and it will last us tens of thousands of years.

Both true and false, depending on what part of the world you're talking about. In most places, true; but some countries, like China, have more uranium than thorium. Perhaps the greatest potential fuel source is the ocean. Although we don't yet have an economically competitive way to extract these elements from the ocean, it's worth noting that there is about 82,000 times as much uranium in seawater than there is thorium.

But really, this too becomes a question of reactor design, more than of thorium vs uranium. Designs for closed-cycle reactors which burn almost everything exist for both the thorium and uranium fuel cycles, and in either, we already have enough fuel to power us for tens of thousands of years.
True or False? Thorium reactors were never commercially developed because they can't produce bomb material.

This is mostly false, although it's become one of the most common myths about thorium reactors. There are other very good reasons why uranium-fueled reactors were developed commercially instead of thorium-fueled reactors. If something smells like a conspiracy theory, you're always wise to take a second, closer look.

When we make weapons-grade Pu239 for nuclear weapons, we use special production reactors designed to burn natural uranium, and only for about three months, to avoid contaminating it with Pu240. Only a very few reactors were ever built that can both do that and generate electricity. The rest of the reactors out there that generate electricity could have been any design that was wanted. So why weren't thorium reactors designed instead? We did have some test thorium-fueled reactors built and running in the 1960s. The real reason has more to do with the additional complexity, design challenges, and expense of these MSBR (molten salt breeder) reactors.

In 1972, the US Atomic Energy Commission published a report on the state of MSBR reactors. Here's a snippet of what was found:

    A number of factors can be identified which tend to limit further industrial involvement at this time, namely:

        The existing major industrial and utility commitments to the LWR, HTGR, and LMFBR.

        The lack of incentive for industrial investment in supplying fuel cycle services, such as those required for solid fuel reactors.

        The overwhelming manufacturing and operating experience with solid fuel reactors in contrast with the very limited involvement with fluid fueled reactors.

        The less advanced state of MSBR technology and the lack of demonstrated solutions to the major technical problems associated with the MSBR concept.

In short, the technology was just too complicated, and it never became mature enough.

It is, however, mostly true that, if we're going to use a commercial reactor to get plutonium for a bomb, recycling spent fuel from a uranium reactor is easier, and you can get proper weapons-grade plutonium this way. It is possible to get reactor-grade plutonium from a thorium reactor that can be made into a bomb — one was successfully tested in 1962 — but it's a much lower yield bomb and it's much harder to get the plutonium.

The short answer is that reduced weapons proliferation is not the strongest argument for switching from uranium fuel to thorium fuel for power generation. Neither reactor type is what's typically designed and used for bomb production. Those already exist, and will continue to provide all the plutonium that governments are ever likely to need for that purpose.

There's every reason to take fossil fuels completely out of our system; we have such absurdly better options. If you're like me and want to see this approach be a multi-pronged one, one that major energy companies, smaller community providers, and individual homeowners can all embrace, then advocate for nukes. You don't need to specify thorium or liquid fuel or breeders; they're already the wave of the future — a future which, I hope, will be clean, bright, and bountiful.