I have been following Peter Zeihan for many years now. Even last night I was sitting on my couch, catching up on his latest videos, learning about things like why the drug cartels in Mexico pose different threats to increased US manufacturing in Mexico. It’s easy to think, this guy knows everything about everything! So you can imagine my surprise when I wake up this morning and there’s an email in my inbox telling me that the latest Peter Zeihan video is about thorium, and why it’s not so great…

Now you have to know that Peter Zeihan really prides himself on understanding supply chains, and it’s consistently one of the things that I find most fascinating about his videos. Whether it’s why certain grains will only grow in certain parts of Russia, or whether it’s about why the rail gauge change makes it hard to ship agricultural products from Ukraine into Europe, here’s a guy you can count on to get the details right, and to you show you why the thing that you thought should be straightforward is just fraught with difficulty.

So when he turned his considerable intellect on the question of thorium, I expected that he would give a similar, in-depth examination of the entire supply chain and what the issues might be. I expected him to talk about the challenge of isotopically separating lithium, or perhaps on the US government’s determination to downblend and destroy our unique and precious supply of uranium-233. Perhaps he might talk about how 95% of the world’s beryllium was sourced from a mine in far west-central Utah or he might talk about very few graphite manufacturers anymore produce the correct graphite grades needed in molten-salt reactors. All of these points and issues would be a valid point-of-departure for discussion about thorium reactors and their supply chain challenges. But no. That’s not what we got at all. That isn’t even what he talked about.

…thorium is a potential substitute, it’s a different element, a different chemical process. And according to its promoters, thorium is better because it is more difficult to turn into nuclear weapons on the back-end, so if you can remove a lot of the proliferation concerns of a uranium power cycle, then maybe we can get rid of some of the obstacles to adopting nuclear power on a broader scale. The short version is probably not.

He starts out by framing the whole “value proposition” for thorium as being based around non-proliferation. I would say that this is an aspect of the value proposition for thorium but it certainly isn’t the key point. The key point for thorium is that we can just about fully consume it in a thermal-spectrum reactor, like the lithium-fluoride thorium reactor (LFTR) we’re developing at Flibe Energy. And we can achieve the goal of very nearly no long-term nuclear waste by fully consuming it. By “long-term” nuclear waste here I’m talking about transuranic nuclear waste, the plutonium, americium, curium, etc, that we have to deal with from today’s nuclear reactors. We can very nearly eliminate the production of that stuff. That’s the stuff that drives the design of long-term, geologic repositories like the proposed (and defunct) Yucca Mountain site in Nevada. Here’s a recent video we released on how thorium will be used in the LFTR we’re designing at Flibe Energy.

But Peter wanted to frame the argument in terms of proliferation. That is already a huge potential discussion, in and of itself. We’ve already built reactors that make lots of plutonium as a byproduct. And all of that plutonium can be potentially separated by chemical means. A number of countries like France already do that chemical separation. Is the production of plutonium in today’s reactors the central issue that holds back the expansion of nuclear power? I would think not. The US, Russia, and South Korea are building uranium-fueled nuclear reactors all over the world. They all make plutonium in the normal course of their operation. The fact that they do so does not appear to be holding back the expansion of nuclear energy. Instead the key challenge appears to be the cost of these huge pressurized-water reactors, both in construction, operation, fuel, and waste disposal. Those costs are big and are growing.

Peter talked about using thorium mixed with low-enrichment uranium (LEU), ostensibly in today’s water-cooled, pressurized, solid-fueled uranium-dioxide nuclear reactors. Peter starts out by talking about how we make enriched uranium-dioxide fuel now (LEUO2).

You’re left with the raw uranium in a form that they call yellowcake, which is a powder, and then you complex that in a compound called uranium hexafluoride, which is solid at room temperature, so you then heat it up a lot and throw it into a gas centrifuge where you spin out the different isotopes. When nuclear material degrades…half-life…all that good stuff…it comes up with different atomic weights. And by putting it into a gas chamber and centrifuging it down, you can increase the cut of the part that is actually fissile, that you can use to achieve an atomic reaction.

I might quibble a bit on how he describes the process of the enrichment of uranium hexafluoride, since the half-lives of the two naturally-occurring isotopes of uranium (U-235 and U-238) are so very long that they never change on any human timescale, and they certainly aren’t altered by the enrichment process itself, but hey, close enough. He talks about how we make solid fuel and irradiate it in a reactor, releasing power from fission as well as neutrons that transform some of the uranium-238 in the reactor into plutonium. But then he talks about adding thorium into this arrangement and observes that doing so doesn’t prevent what uranium that remains from continuing to make more plutonium. It makes a little less plutonium, but it’s still making plutonium.

People can take take the plutonium from spent nuclear fuel rods and make plutonium-based nuclear weapons. Now if you’re going to do something with thorium, the problem is it doesn’t make it too much better. Plutonium is still the byproduct. And while you don’t generate as much plutonium from the use of thorium in your nuclear reactor as you do from uranium, you still get some. In addition, you need a different kind of reactor.

Peter, I gotta hand it to you, technically nothing you’ve said about that process is incorrect. But it’s not the thorium making that plutonium. It’s the uranium in the fuel that’s making the plutonium. But yeah, if you add thorium into uranium fuel rods, you’re still getting plutonium. Maybe a little less per overall unit of power, but you’re still getting it. But Peter, being right about that specific case is nice and all, but you managed to miss the larger argument in favor of thorium entirely.

Peter’s point is akin to saying something like, I put diesel in this gas car and it didn’t work well at all. Everyone knows that you have to put gas in a gas car and diesel in a diesel car. They just don’t work worth a darn otherwise. And when if comes to thorium, you gotta use it in a molten-salt reactor. You can try to use it in a solid-fueled reactor but the results will be pretty lousy. Is that because thorium is lousy? No, it’s because you used it in the wrong machine. Just like diesel fuel in a gas car.

Nowadays, thankfully, when people are talking about “thorium” reactors they’re not talking about pressurized-water reactors anymore, they’re talking about LFTRs. “Thorium” has become a short-hand for LFTR, and I really like that. So when you want to talk about why people are excited about thorium and what it offers, you really have to talk about thorium in LFTRs. Otherwise you just aren’t even talking about the same thing. You’re often talking right past one another. And LFTRs don’t have enriched uranium in them. They’re started on uranium-233, which comes from thorium in the first place (part of the reason we don’t have a lot right now). LFTRs can essentially avoid plutonium production entirely, because again, they’re keeping uranium-238 out of the system. No U-238, no plutonium. LFTRs breed new fuel from thorium. The video explains how. There you go, power without plutonium.

Now Peter says you have to do a completely new kind of reactor. Is he talking about LFTR? Hard to tell. It doesn’t sound like he is. It really sounds like he’s talking about a solid-fueled reactor. Perhaps Peter would care to parse that information for us. I would be very happy to talk to him about this at some point. I am sure he has mechanisms to reach me.

I really find baffling Peter’s assertion that an average one-gigawatt, uranium-fueled reactor would produce enough plutonium for “six to twelve crude plutonium devices.” Is this true? Most nuclear engineers would strongly disagree with Peter, pointing out that the isotopic assay of plutonium in typical reactors would contain too much plutonium-240 in order to be usable in nuclear weapons. Others might further quibble on this point. Still others might point out that the isotopic quality of the plutonium in spent rods is not uniform, and that near the periphery of the reactor and particularly at the ends of solid fuel rods, the isotopic quality of plutonium is much higher and perhaps suitable for nuclear weapons. It’s all a lot of argument, most of which goes nowhere for a simple reason. Every country in the world that has desired plutonium for nuclear weapons has made a simple reactor to irradiate natural uranium and to chemically extract plutonium. They never do it from the spent fuel of commercial reactors. Countries like the US and France have large fleets of nuclear reactors. The French even chemically process that fuel. But they didn’t get any of their weapons-grade plutonium from those commercial reactors. They got it from dedicated production reactors. The US used to have quite a few of these in Hanford, Washington and at the Savannah River Site in South Carolina. No more. They have been shut down for decades. The US has a large quantity of “surplus” weapons-grade plutonium that it is currently trying to chemically adulterate at enormous expense. It is not tempted by the notion of making plutonium weapons from chemically-processed spent fuel from commercial uranium-fueled reactors.

At any rate, what does any of this have to do with thorium? Well, Peter asserts that thorium is of little to no value because it only mildly reduces the march towards plutonium harvest from spent nuclear fuel from weapons. He thinks you’ll only get “four to ten” plutonium warheads from a thorium-fueled reactor. I have no idea from where he gets these numbers. But remember, there is NO plutonium harvest from commercial spent nuclear fuel for weapons. Indeed, the DOE still possesses UNPROCESSED spent fuel from some of its “production” reactors, notably the N-reactor at Hanford. If the US by any chance actually wanted MORE weapons-grade plutonium, processing the remaining N-reactor fuel would be the first place to look.

But of course we’re not trying to get MORE weapons-grade plutonium, rather, we’re spending ghastly sums trying to destroy what we have! Again, what on earth does this have to do with thorium? It would seem that Peter is making an argument against uranium-fueled reactors rather than against thorium-fueled reactors!

Peter then says, well perhaps India should do thorium reactors because they have so much thorium. That’s not a terrible argument, but it certainly doesn’t exclude us. The US has enormous amounts of potential thorium. The richest thorium deposit in the entire western hemisphere is the Lemhi Pass in Idaho, where enough energy in the form of thorium exists to power the United States for thousands, even tens of thousands of years, assuming LFTR technology is used to realize that thorium energy. And using LFTR technology with thorium would not create plutonium or enhance our susceptibility to “proliferation.” So if countries that have a lot of thorium should pursue thorium, then the United States clearly should be very high on that list.

As a bit of a parenthetical, how is the United States doing in terms of uranium? Are we doing “just super” on uranium, like the DOE likes to assert? Nope, we hardly mine any, we import nearly all of it, and it is all enriched in Russia, Europe, or in a single plant in New Mexico owned by foreign interests and using foreign technology. We have no waste plan for uranium and most of our possible uranium resources are on Navajo land in the western US. And I don’t think anyone wants to go to the Navajo and say, hey, can we try to mine uranium on your land again?

Is thorium a “silver bullet” for the nuclear industry’s problems? In LFTR, it sure is. Peter lists the proliferation problem and the spent fuel (waste) problem. Both problems are solved by LFTR technology. He worries about plutonium in civilian hands. It already IS in civilian hands. The vast, vast majority of plutonium in the United States is in civilian hands in the form of spent nuclear fuel. Peter thinks there is no other use for plutonium. But there is: it can be consumed for energy. Indeed, it must be consumed for energy if we are ever to rid ourselves of it. If we try to consume plutonium in uranium-fueled reactors, as we once tried during the MOX program in the first part of this century, we only make more and more plutonium in the reactor. Thorium reactors are the ONLY way to consume plutonium permanently without making more. They can actually destroy it permanently and extract valuable energy and neutrons in the process. Thorium is the ONLY way out of our plutonium debacle, and the sooner we get working on that, the better it will be for all of us.

I would agree that our current approaches to spent fuel (waste as he calls it) “address” the issue without actually “solving” it. Thorium-fueled reactors have the potential to solve the problem permanently, but consuming the plutonium and avoiding the production of more. He notes that the Yucca Mountain waste repository is “over-subscribed.” Indeed, if it was going forward, it would be over-subscribed, but it isn’t going forward at all. Harry Reid and Barack Obama killed it back in 2009, and President Trump finally accepted that reality in February 2020 when he submitted a budget request to Congress that made no request for funding of Yucca Mountain. So Yucca Mountain is dead, and as regular readers of this blog would know, I shed no tears over that. Had we built it and put the spent fuel there, our descendants would have dug the spent fuel out for its plutonium and cursed our names for ever having built the monstrosity.

Peter gets it “mostly wrong” by saying our spent nuclear fuel is stored in coolant ponds. Not anymore. Most of it is stored in dry casks on the reactor sites and can be stored that way for a very long time. But he does correctly note that what we have today are close to hundred nuclear waste storage sites rather than just one. And that all by itself isn’t a good idea.

I will agree with Peter that a solution to the issue of nuclear waste is desperately needed, and that barring a new reactor design (like LFTR) then the fraction of our nation’s energy generated by today’s nuclear will fall precipitously. That is correct. There is very little that will stop the aggressive closure of today’s nuclear reactors over the next twenty years. Power generated by uranium-fueled, high-pressure reactors will go down and down. Nothing is going to stop this trend. The only real question is whether or not LFTR technology will be developed and expand at a rate sufficient to, not only replace existing nuclear, but to expand the overall production of nuclear energy. But Peter’s assertion that we’ve only had one new plant come online since 1973 is grotesquely incorrect. Most of our nuclear reactors came online in the 1980s and into the 1990s. We have only had two new nuclear reactors come online that were ordered after 1973, and those are the two new units at Vogtle, units 3 and 4. But Watts Bar 1 came online in the mid-1990s and Watts Bar 2 came online about ten years ago. Browns Ferry unit 1, near to where I live, was also restarted in the last decade.

We spend a lot of time modeling this. And yes, it can be done. It must be done. The path forward is to solve the spent fuel issue simultaneously with the need for startup fuel for the LFTR. We need an intermediate class of reactors that can consume spent nuclear fuel while making new startup fuels for LFTRs. We’re working on those at Flibe Energy. We call them LFLEURs. They have a flexible approach to nuclear fuel. They can use spent fuel or fresh LEU sourced commercially. They will produce uranium-233 from their internal blankets of liquid thorium salt mixtures. We can eliminate our spent nuclear fuel while starting a new generation of safe, efficient, thorium-fueled reactors. I’ve been trying to explain this since I started this blog in 2006, and even more particularly in this 2010 talk given at Google.

Peter concludes that thorium isn’t going to get us any closer to the expansion of nuclear energy that we need. I respectfully and vigorous disagree, and I have the modeling to prove it. Thorium, in the LFTR and LFLEUR designs, is the ONLY way to achieve the energy goals before us.

Give me a call Peter. You know I’m a faithful viewer.