As I mentioned in my first posting on thorium reactors, this is a huge topic that deserves a careful discussion – even four postings isn’t enough to do justice to the topic, but it’s a good start. So this will be my last posting on thorium reactors for now, but I might come back to the topic in the future if it seems appropriate. And for this week’s posting I thought I’d take a look at one of the persistent bugbears of commercial nuclear power, and another area in which aficionados of the thorium cycle claim a great advantage. But the best way to get started is to go over where radioactive waste comes from and we’ll go from there. And, incidentally, I should add here that I’ve generated and processed my share of radioactive waste in the Navy and in civilian life (hospital and university waste primarily, in addition to some remediation projects). Although some radioactive waste can be highly dangerous, most of the waste I’ve generated, processed, and handled has been very low-level and quite safe. Dangerously radioactive materials are the exception rather than the rule.
In a reactor there are three basic sources of radioactive waste – activation products, fission products, and transuranic materials. Activation products come from neutron bombardment of stable atoms – adding a neutron, say, to stable cobalt-59 turns it into radioactive cobalt-60. This can happen with any stable atom in the neutron-rich environment of a reactor core – even the water used to cool the reactor core can capture neutrons to form radioactive tritium, and any air dissolved in the water can become neutron-activated as well. But in a nuclear reactor the majority of activation products come from the metals that make up the reactor itself – cobalt (added to some metals to harden them), iron (in the steel pipes and components), manganese (added to high-strength steels), and more. And when the steel suffers from wear and tear, the corrosion and wear products that flake off from the pipes and other components can collect in low-flow parts of the plant or are filtered out by the water purification system – this is a major source of radioactive waste. Luckily, neutron activation products are usually relatively short-lived and decay away over a period of months or years – at most a few decades. Since the formation of activation products is a function of neutron flux (which all reactors have) and the chemical composition of the reactor components (which all reactors have), we can expect that both conventional and thorium-cycle reactors will produce similar suites of activation products. On the other hand, since the molten salt reactors operate at lower pressures, it might be possible to alter the makeup of the metals that comprise the reactor vessel, pipes, and other components – in a reactor that operates at near-atmospheric pressures. it might be possible to adjust the metallurgy to a mix that produces fewer of the longer-lived nuclides.
Fission products are another form of radioactive waste. Hit an atom of U-235 (or U-233) with a neutron and it’s likely to split into two parts (the fission fragments), both of which will be radioactive atoms. While we can’t predict in advance what fission fragments will result from splitting any individual atoms, there are some patterns that we can see. For example, while fission products can have any atomic mass between about 70 and 166 atomic mass units (1 amu is a little less than the mass of a single proton), most of the fission products cluster around atomic masses in the 130s and the 90s. What’s happening is that the atoms split, but not exactly in half, and some neutrons are given off as well. Anyhow – what this means in practical terms is that every uranium atom that fissions produces two radioactive fission product atoms.
In a conventional uranium reactor the atoms of uranium fuel are sealed within a clad that contains not only the fuel but the fission fragments as well. So under normal circumstances the fission fragments will remain locked away – some will leak out, but not many. On the other hand, if there is a defect in the cladding the fission products can escape into the coolant. Either way the fission products will add to the radioactivity of either the spent fuel (if the clad remains intact) or of the reactor plant itself (if the clad leaks). Luckily, with a very few exceptions, the fission products from both U-235 and U-233 fission are fairly short-lived. In fact, there are only 8 nuclides that are both long-lived and produced in abundance – these include some that are pretty well-known (Cs-137 and Cs-134) as well as some lesser-known nuclides (Tc-99, and Kr-85). These are the nuclides that can make spent reactor fuel dangerously radioactive, and some of these (the cesium nuclides) are among those still hanging around in Japan.
If you have a Chart of the Nuclides handy you can look up the fission product yields for U-233, U-235, and Pu-239 – this will tell you, for example, that fissioning U-235 produces something with a mass of 137 amu about 6.19% of the time while a fission fragment of this mass occurs slightly more frequently (6.76) when U-233 fissions. MOst fission products are short-lived so this 137-amu atom will decay merrily away through short-lived nuclides until it reaches Cs-137, with a half-life of 30 years. So by looking at the long-lived nuclides in any atomic mass along with the frequency with which atoms of this mass are produced we can work out a profile for our radioactive waste.
Interestingly, the fission product curves for both U-233 and U-235 are pretty much the same – both nuclides are so close in mass that they have similar fission product profiles. In addition, the amount fission products produced are solely a function of the amount of energy a reactor produces – splitting a U-233 atom releases about as much energy as does splitting a U-235 atom, so producing a MW-hr of energy requires splitting the same number of atoms in both cases. And since each fission produces two fission fragments, the same power output will produce about the same number (and radioactivity) of fission products from each of these devices. The bottom line is that fission product production should be more or less a wash when we compare thorium-cycle reactors to their conventional cousins. What makes the molten salt reactors interesting (for both activation and fission products) is the ability to continually filter out these radionuclides – even were the amount of radioactivity produced be the same, in a molten salt reactor they can be removed as the reactor operates, making them easier to deal with.
So – both the thorium-cycle and conventional reactors should produce roughly the same activation and fission products. But where the thorium units shine is in the production of transuranic nuclides – those elements that lie beyond uranium on the periodic table. As mentioned in last week’s post, the transuranics are produced when uranium atoms capture multiple neutrons to form plutonium, americium, and even more exotic elements. Lighter atoms are further away from the heavy transuranics, making them more difficult to produce. And since it’s many of the transuranics that hang around for centuries, this also reduces the length of time that the fuel is dangerous. This same characteristic, then, accounts for two favorable characteristics – the relative lack of plutonium produced by these reactors (making them more proliferation-resistant) and the lower burden of radioactive waste they produce.
In Superfuel, the author also discusses the possibility of using thorium-cycle reactors to “process” existing spent fuel and radioactive waste by using the neutron flux to transmute this waste into shorter-lived nuclides. While this is true in principle, this is something that can be done in any nuclear reactor – it’s a function of the waste being exposed to a neutron flux, and every operating reactor has a plethora of neutrons. But I suspect that, in real life, this is no more likely to be accomplished in a thorium-cycle reactor than it is in today’s reactors – it’s something that’s possible, but I’m not sure how likely it is to take place. But this is not a negative; rather, it’s the lack of a positive.
So, let’s sum this up:
- Fission products are largely the same for both reactors
- Activation product production for thorium-cycle reactors might be lower if the less stressful environment (lower pressure) allows the use of alternate materials
- Production of transuranic nuclides is decidedly lower in thorium-cycle reactors.
- But the ability to chemically process molten salt (in a homogeneous liquid-fuel core) makes it possible to scrub radioactivity from the fuel on a continuing basis, reducing the build-up of these nuclides over time.
From this it seems reasonable to conclude that thorium-cycle reactors do offer clear advantages over conventional devices in this area, although these advantages might not be as pronounced as the pro-thorium advocates might suggest.
Finally, let’s try to sum up what we’ve discussed in this series of thorium postings.
Reactor design and construction – we are more familiar with building the “conventional” high-pressure and high-temperature water-cooled reactors, so these have an advantage by virtue of experience and familiarity. However, molten-salt thorium reactors can operate effectively at atmospheric pressure and they don’t require many of the components that make our current fleet of nuclear reactors so complex. Overall, the advantage here goes to thorium-cycle reactors because of their potential simplicity and lower mechanical stress (although to be fair it should be noted that it is possible to build liquid-fuel, molten salt-cooled uranium reactors).
Fuel economy – in a thorium-cycle reactor as much as 100% of the atoms are capable of being transmuted into U-233 and fissioned, compared to less than 1% of the atoms in natural uranium. In addition, there are about four times as many thorium atoms as uranium on Earth. Thus, there are more pounds of thorium than of uranium for us to use and each pound contains more fuel atoms. In this category, thorium has a clear advantage.
Proliferation resistance – thorium-cycle reactors produce far less of the plutonium that can be made into weapons compared to uranium reactors, with their high abundance of U-238. In addition, the presence of high-energy gamma radiation from the U-232 decay series and the ability to “poison” the U-233 by adding U-238 adds further complexity to the process of producing nuclear weapons from this type of reactor. The advantage here is also clearly to thorium.
Radioactive waste – both reactors produce similar fission product profiles. The production of activation products depends on whether or not the lower pressures and potentially simpler reactor design permit the use of alternate materials that are less likely to produce radioactive waste. However, thorium-cycle reactors produce far fewer of the more problematic transuranic nuclides, giving them an advantage in this category as well. In addition, the ability to process liquid fuel on a continuing basis and, thus, to sequester the fission and activation products on a continuing basis adds to this advantage.
My conclusion from all of this is that there is a clear advantage to using thorium-cycle reactors, especially in a world that needs baseline energy production without carbon dioxide emissions. If all that stands in our way is the reluctance of Congress to fund this novel technology (not new tech since it’s been around for more than a half-century) and the reluctance of the existing nuclear utilities to experiment with a new style of reactor then we really need to try to get past the entrenched status quo to try to build some of these things. Maybe we can take notes from the Indians and Chinese as they get their own thorium programs up and running.