Holding a miniature sun in a magnetic bottle isn’t easy. But that’s how you make a fusion reactor. In this episode I chat with plasma physicist John McCone about the challenges of building a functional fusion reactor. We talk about plasma blowtorches, neutron bombardment, lumpy magnetic fields, and ways to make a nuclear bomb.
Dustin Driver: Imagine trying to keep a miniature sun burning in a bottle made of nothing but magnetic fields. That’s essentially what a fusion reactor does and why it’s taken almost a hundred years for humanity to figure out how to do it.
John McCone: Temperatures at the center of the fusion reactor are 10 times the temperature at the core of the sun. You know, the radius of the sun is about a million miles, whereas the radius of a fusion reactor is a couple of meters.
Dustin Driver: That’s plasma physicist John McCone. John is also a listener and he graciously offered to chat with me about the challenges of building a fusion reactor. He has a research PhD from [Column 00:00:38] Science Center and spent more than a decade working in fusion, studying how plasmas behave in reactors. John is still an expert in plasma physics and nuclear fusion, and that’s what he’s talking about today.
Dustin Driver: Before we jump into the technical details, let’s review what a fusion reactor is and how it works. A fusion reactor recreates the fusion that happens in a star and harnesses that energy to make electricity. There are a few ways to go about building a fusion reactor, but today we’re talking about magnetic confinement reactors. In a magnetic confinement fusion reactor, hydrogen gas is super heated by electricity within a magnetic field until it turns into a plasma.
Dustin Driver: Plasma is the fourth state of matter. When atoms get hot enough, they break up into charged ions and free electrons that conduct electricity. In a fusion reactor, magnetic fields are used to contain the heat of the hydrogen plasma until it fuses into helium. This fusion releases a ton of energy, which is then used to turn water into steam, which spins a generator to make electricity.
Dustin Driver: It’s simple in theory, but extremely difficult to pull off. The idea for a fusion reactor has been around for a long time. In fact, the Soviets designed the Tokamak reactor in the 50’s and that’s the design that most of the latest reactors that are based on. Imagine a giant metal doughnut surrounded by tons of electromagnets. These magnets create a field that control and insulate a stream of plasma that flows around inside this donut. These fields contain the heat of the hydrogen plasma until it fuses into helium. ITER, the world’s largest fusion reactor being built in France, is a Tokamak.
Dustin Driver: So what are some of the main challenges in building a fusion reactor and why has it taken so long for us to get to this point?
John McCone: Most of the big challenge in nuclear fusion is to keep the plasma hot. Fusion reactions only occur at a certain temperature. So in order to keep a fusion reaction going, you have to maintain the temperature of the plasma above the temperature required for fusion reactions to occur at a reasonable rate. And that requires holding in the heat of the fusion reaction for long enough.
Dustin Driver: And remember super hot. We’re dealing with temperatures that are 10 times the temperature of the sun.
John McCone: So you’ve got this fusion reactor that’s 10 times hotter than the sun, a couple of meters away from a vacuum vessel that is room temperature. And this is of course, it’s very difficult to maintain those temperature gradients because, of course, heat likes to flow from regions of hot, regions of high temperature to regions of low temperature. So when you want to maintain really, really high temperatures, you need to insulate those, the hot spots very, very well.
John McCone: Now fusion researchers use a magnetic field to insulate the plasma. But another problem with temperature gradients, is that temperature, that heat gradients can be used to do work. Essentially, this is how electricity generation works. That to, you have your furnace and then the heat flows from the furnace and then it spins the turbine and that generates electricity.
John McCone: But heat can do work and that means that the plasma can do work on the magnetic field. And quite often we find that plasma has found all sorts of ways of rearranging a magnetic field to lose heat faster to the outside. So plasmas find all sorts of ways to cool down. And sometimes it’s in the form of micro turbulence and they just get a little bit cooler, and other times it’s in the form of a disruption where the insulation is, where the magnetic, the magnetics insulation is essentially torn apart and the plasma loses heat very, very rapidly, as well as current to the wall and that’s called a disruption.
Dustin Driver: And when that happens, the fusion reaction essentially stops.
John McCone: The fusion reaction, quite simply, stops. I mean, that’s one of the reasons why fusion reactions are relatively safe. There’s no chain reaction risk in a nuclear fusion reactor because it’s so easy to stop the fusion reaction. But of course, you want the fusion reaction to continue because you want to generate electricity. So, essentially, if you have a disruption, then the electric, the fusion reactor will stop generating electricity and you have to start the plasma again.
Dustin Driver: And if you can manage to keep the heat in, it’s still very difficult to keep out impurities. Plasma has a tendency to, well, torch stuff.
John McCone: The plasma is like a kind of blowtorch. No matter how long you confine the plasma for, eventually it’s going to hit the wall. And when it does, you’ve got ions from the plasma that are millions of degree’s hitting the wall. Now the plasma density is low. It’s 100,000 times less dense than the atmosphere. So you’re not talking about incredible amounts of heat and it doesn’t melt the wall or anything like that, but it does knock atoms out of the wall. So you’ve got this constant sprinkling of ions, that are millions of degrees, hitting the inner wall of the vacuum vessel, and that sputters off atoms from the vacuum vessel.
Dustin Driver: And those super charged ions can really mess things up.
John McCone: For example, just to give you some idea of how serious a problem impurities can be to heat loss in the plasma, if one in 50,000 plasma out of 10, those one in 50,010 ions will radiate 10% of the heat in the plasma away. And you have to work hard to contain the heat in the plasma to get a fusion reaction going. So if heat is leaking out, that’s not good. Really light elements, however, like lithium or carbon, are not so serious. And if one in five plasma ions, and it takes one in five plasma ions to be lithium or one in 15 plasma ions to be carbon, to have the same effect.
Dustin Driver: Those impurities can also really mess up the magnetic fields within fusion reactors.
John McCone: Another really damaging thing that impurities can do, is if the impurity distribution as lumpy, they can actually create a lumpy, resistivity profile in the plasma. And lumpy resistivity causes lumpy currents and lumpy currents cause disruptions that causes instabilities and it can actually break, it can tear up the magnetic insulation and cause the heat to suddenly get lost to the vacuum vessel.
Dustin Driver: So it’s really tough to get a fusion reaction started and to keep it going within a fusion reactor. But once fusion starts, you have another problem to deal with. When you fuse two atoms of hydrogen together to form an atom of helium, a neutron is released. These neutrons can damage the wall of the reactor on an atomic level.
John McCone: It’s like Lego, right? Solid material. You’ve got all these different atoms in a particular arrangement and the arrangement of the atoms creates the material property. So if you think about how ductile or brittle or how conductive a different material is, it’s all because of the arrangement of the atoms inside that material. And if the atoms change their arrangement, then the actual material properties of, will change.
John McCone: So what happens with neutrons is, if you imagine a snooker table and you smash a ball into the, a white ball into a red ball and you break the balls or whatever, that’s kind of what neutrons do to solids. If you can imagine the snooker balls, the triangle of red balls is the atom and then the neutron is the white ball, then the neutron causes these cascades.
John McCone: Now it doesn’t smash up the atom completely. It, only a small amount, in fact, only a small fraction of the atoms in the lattice are displaced, but those displacements can have significant effects over a long enough period of time. So if you bombard, you’re just firing neutron after neutron after neutron into a material, you’re causing cascades of defects and stuff like that. So you’re changing the lattice structure and those changes in the lattice structure will affect the conductivity of the material, the brittleness of the material, things like that.
John McCone: And this is actually, this is the really big problem that, that it’s not that, I mean, it’s not undo, it, I mean, the stressors in a fusion reactor are not unheard of. That we’ve engineered things to be, we’ve engineered things to be, to take those kinds of stressors before. The issue is, that when you engineer materials to maintain live stressors, the properties of those materials are very important.
John McCone: So if you have, so if you carefully develop your reactor to maintain pretty, pretty intense, stressors and then you bombard it with neutrons, you’re going to be changing the properties of those materials. Now of course, again, these things have been simulated, but it’s quite challenging to really work out what the, how the materials, how all these material properties, all these different components will change as neutrons hit them and stuff like that, over the lifetime of the reactor.
Dustin Driver: Engineers can simulate neutron damage but they won’t really know how neutrons affect the reactor until the reactor has been running for quite a while. Another challenge for engineers is dust. Remember how the plasma sort of blow torches the inside of the reactor? That creates a lot of dust and that dust could contain tritium, which is a radioactive version of hydrogen. That tritiated dust could pose risks for reactor workers.
John McCone: The atoms that get sputtered off the inside of the vacuum vessel get, that go into the plasma? They turn it dust. I don’t, I’m not sure exactly what the mechanism is, maybe somebody knows. But yeah, dust is definitely something that plasma produces. And yes, if, in a hypothetical fusion reactor, there’s a possibility that there could be a lot of dust contained in the, there could be a lot of tritium contained in the dust. Yes.
Dustin Driver: For the first 10 years of operation, ITER will be using deuterium or just plain hydrogen, for its fusion reactions. After that, they’ll switch to a deuterium tritium reaction, which is much more efficient.
John McCone: So during the 10 years when they’re running deuterium, they’ll get an idea of how much tritium would be contained. They get an idea of how much tritium would be in the dust and I presume, they’re probably going to work really hard to make sure that there’s not too much tritier dust produced. ‘Cause it does potentially pose a hazard, especially for people who want to do maintenance work on the machine or something like that.
Dustin Driver: Controlling the dust in a nuclear fusion reactor will be a challenge, but there are ways to do it. Something called a liquid first wall could reduce the dust to almost nothing. Imagine an ultra thin wall of liquid flowing around the inside of the fusion chamber. This wall would catch any of the dust before it becomes airborne.
John McCone: A liquid first wall, which is just, maybe less than a millimeter thick, would instead of producing dust, it would produce droplets. So when those droplets land on the rest of the material, it would kind of fuse with it, so you wouldn’t get this kind of dust in the reactor. So if you had a liquid first wall, that would significantly reduce even the small amount of radiation that say, a vacuum vessel breach might cause. So even the issue of dust is something that there is potential solutions to, and which can be mitigated.
Dustin Driver: But the amount of radioactive dust in a fusion reactor is nowhere near the amount of dust that we’ve already seen during disasters like Chernobyl.
John McCone: Yeah, I mean the amount of dust that the graphite fire would produce, I mean, that’s what happened during Chernobyl. The graphite moderator caught fire and then, and of course, the graphite moderator was bombarded by neutrons, like huge amounts of neutrons from the uranium, raw, the uranium pellets. And yeah, I mean, then it caught fire and just all the radioactive graphite kind of came up with this smoke. I mean, that’s a lot more dust than even a dusty fusion reactor would contain.
Dustin Driver: Another challenge with fusion reactors is something called Wigner energy. Wigner energy builds up in materials that have been bombarded by neutrons. It’s kind of hard to explain, so I’ll let John take over, once again.
John McCone: If you imagine like a Rubik’s cube, right? And every, and each of the cubes in the Rubik’s cube is like an atom and you take one of the atoms and you kind of stick it in between the other, stick it, squeeze it in between the corners of the other atoms and you’ve got like this hole. So that, so there’s all, so essentially you have these atoms that are kind of, when you have lattice defects, you have the atoms that are squashed too tightly around other atoms. And then you have these kinds of holes in the lattice, where there’s kind of gaps in the atomic structure and that’s in a higher energy state than the unperturbed atom, the lattice without defects.
John McCone: So if those atoms start kind of vibrating, the atom’s that are kind of squidged into the other atoms, kind of find their way back into the holes. And then you heat up a material that has sustained a certain amount of neutron damage, anneals and that reduces some of the lattice defects. So when that happens, it releases Wigner energy. Now, the annealing process is actually something that’s been done in a controlled manner and I think reactors are deliberately heated up to anneal them sometimes.
John McCone: So it can be done in a controlled way, but yeah, so if you have a material that’s being bombarded by neutrons and it’s built up all these lattice defects, then if you heat up that material, it can release more heat. There’s the potential for a kind of a, for a heat, for a release of heat to create a release of more heat. Again, you can kind of, you can regularly anneal the materials in the fusion reactor, to reduce the amount of Wigner energy that gets released, but yeah, the release of Wigner energy is one of the reasons, things that called, caused the Windscale fire.
Dustin Driver: The Windscale fire in 1957 was Great Britain’s worst atomic accident. The plant had a control process for releasing Wigner energy in the graphite rods. During one of these procedures, one of the rods, instead of cooling after the Wigner release, continued to gain heat. It eventually caught fire and the entire reactor burned for more than two days. The amount of radioactive material released was substantial, but nowhere near what was released during the Chernobyl incident.
John McCone: If you’re not careful about it, it has the potential to cause sudden releases of heat that can do damage. And of course, and of course, another thing is, of course, and then if you’ve got superconductors and you’ve got Wigner energy, if you’ve got, if the neutron damage has caused a buildup in Wigner energy and you’ve got superconductors as well, then a sudden release of heat might potentially, it could potentially, kind of saturate the ability of the cooling systems to keep the superconductors cool. So that could, so a release of Wigner energy, in a badly designed fusion reactor, a release of Wigner energy could stimulate a thermal quench in the superconductors.
Dustin Driver: A thermal quench means that as superconductors get too warm to, well, super conduct, the magnetic field would collapse and the fusion reaction would stop. So far we’ve been talking about the challenges of making a nuclear fusion reactor, but once it’s made, could it be used by a rogue state to, I don’t know, create more nuclear weapons?
John McCone: Essentially, there are two ways to make nuclear bombs. One is by enriching the levels of uranium-235. So uranium, natural uranium, 1%, 0.7% of atoms in natural uranium are uranium-235 and the remaining 99.3% of the atoms in uranium are uranium-238. Now, the two ways of making, now in an atom bomb, it’s more like, I guess, 80 or 90%, maybe more are uranium-235, so you have to do a lot of enriching. But in an atom bomb, one way of doing it is to separate out the uranium-235 from the uranium-238 and create enriched uranium, which is maybe, 90% uranium-235, highly enriched uranium.
John McCone: The second way is to bombard uranium-238 with neutrons and turn some of the atoms in the uranium-238 into plutonium-239. Now plutonium-239 can then be chemically separated from the uranium-238. So if you have a neutron source, you can just chemically separate the plutonium-239 from uranium-238. So probably what you want to do, if you wanted to use a fusion reactor to make weapons grade plutonium or whatever, is you would get natural uranium, you’d put it in the blanket of the fusion reactor for a little bit and then take it out and then chemically separate out the plutonium. Then put the uranium back in the blanket, chemically separate or add the plutonium and so on and so forth.
John McCone: Essentially, it’s easier to chemically separate plutonium from uranium than to separate uranium-2325 from uranium-238, because uranium-235 has identical chemical properties to uranium-238, whereas plutonium has different chemical properties. Now terrorists couldn’t do that. So if you have a well behaved state, nuclear fusion is actually more proliferation resistant to terrorists because there is no uranium in a nuclear fusion reactor. But yeah, if a rogue state wanted to use nuclear, could use nuclear fusion reactors to produce plutonium from natural uranium and just with natural uranium and a nuclear fusion reactor, you could make atom bombs without needing any enrichment facility.
Dustin Driver: So there is some danger that a fusion reactor could be used to create weapons grade plutonium, but fission reactors can be used to do the same thing and there are thousands of those on the planet. Also, John made a great point about fusion reactors being really resistant to terrorist attacks. Unlike a fission reactor, if you try to blow it up, it would just stop. A fission reactor, on the other hand, well look what happened with Chernobyl and Fukushima. But what about nuclear waste? Fission reactors create a lot of nuclear waste that is really difficult to store and eventually get rid of. Fusion reactors, on the other hand, don’t. Again, I’ll let John explain.
John McCone: Theoretically speaking, a fusion reactor, a fusion reaction doesn’t have to produce any net waste. So in a fusion reaction you have tritium and deuterium and then they combine to create helium and a neutron. The helium is not radioactive and the neutron then goes out into a lithium blanket and the lithium blanket, and when it hits, and then when it hits the lithium, it turns the lithium into helium and tritium.
John McCone: So then you take the tritium out and you put it back in the plasma, you take the tritium and you put it back in the plasma and then you turn it into deuterium. And so then you reacted with deuterium to fuse helium and neutron. So the net result of the closed loop fusion system, is to turn deuterium and lithium into helium. And deuterium is nonradioactive, lithium is nonradioactive and helium is also nonradioactive.
John McCone: So theoretically speaking, if you’ve got everything just perfect, you actually wouldn’t produce any radioactive waste at all from nuclear fusion. Now in practice, there’s other things the neutrons are going to hit. So the neutrons are going to hit stuff other than the lithium and in the process, they’re going to make those things radioactive. Now in, but it’s a relatively, you’ve got less radiation produced then in a fission reactor.
John McCone: And another thing is, that when nuclear, when uranium atoms split, they don’t always split in the same way. So when uranium atoms split, they, all these different uranium atoms are splitting in different ways and they’re essentially producing every element under the sun in radioactive form. So fission fragments, fission products contain pretty much every element in the ray, in the periodic table. That’s very complicated to chemically process. And they’re emitting all sorts of different, gamma radiation, beta radiation, everything.
John McCone: When fusion neutrons hit the few, the material in the fusion reactor, you say you got your component that’s made of one material or maybe two materials and then the neutron hits it and maybe turns one of those materials radioactive. So you’ve got maybe like one particular radioisotope, one particular chemically distinct radioisotope contained within a material that contains one or two other atoms. So it’s much easier to chemically separate the radioisotopes from irradiated fusion components, compared to fission components. And there’s also less of them produced.
Dustin Driver: So fusion reactors are a lot cleaner than fission reactors and they use a much more readily available source of fuel. In fact, we have tons of the two things that you need in a fusion reactor, lithium and sea water.
John McCone: The current lithium reserves are sufficient to supply civilization with fusion energy for a thousand years. And another thing is, of course, if you go down in purity, you go up in quantity. So given the value of lithium infusion, it would probably be possible to mine a whole lot more lithium than are contained in known reserves because it would just be so valuable. There’s enough deuterium and sea water to supply all of our fusion needs, for something like a million years, I think.
John McCone: Essentially if we do get fusion to work, it would be, it would supply all of civilization’s energy needs and energy is what civilization needs. Energy is essentially the thing that powers civilization. It’s absolutely essential that we constantly maintain an energy supply or else everything grinds to a halt. And like I say, it, and it doesn’t require any CO2 to be emitted, so it allows you to produce all the energy you want and it doesn’t cause global warming. So that’s a really big win.
Dustin Driver: And that’s why we’ve been spending so much time and effort trying to get fusion to work. Of course, there are other green energy technologies, some that may actually be even better than fusion.
John McCone: Yeah, I think solar is probably the most promising one, to be honest. I think solar is probably more promising than nuclear fusion, in fact. But we already spend more money on solar than we spend on nuclear fusion. That we already spend more money building solar panels then we spend on researching fusion reactors. So nuclear fusion is one answer and it’s probably worth, it’s probably worth investigating it anyway. Although solar power, I think, is more promising.
John McCone: I mean it’s, solar power is currently being, I mean solar installations are currently increasing by about 30% every year. So although solar power is only producing about 1% of our electricity, in a couple of decades, if this sort of exponential rate of growth can be maintained, that could change a lot. But like I say, the amount of money that’s put into solar power is already much larger than nuclear fusion. So if you directed the money going into fusion research into solar energy, it wouldn’t actually make much a difference because it wouldn’t be, because it wouldn’t really, it’s a small percentage of the amount of money that solar energy is already receiving.
Dustin Driver: The ITER reactor is set to go online in 2025 and should be switching over to the more efficient tritium reaction in 2035. MIT’s SPARC reactor is also slated to go online in 2025. It sounds like a wild science fiction fantasy, but it actually does seem like we’ll have fusion power in our lifetime. Will that give us unlimited amounts of energy and solve all of humanity’s problems? Well, probably not. At least not right away, but it’s a huge step forward.
Dustin Driver: If you want to learn more about fusion reactors, go check out an excellent article on howstuffworks.com by Craig Freudenrich. It breaks down all the different kinds of fusion reactors that are possible and how they work. Tremendous thanks to John McCone for guest starring on this episode. Go check out his website, johnmccone.com. There’s a ton of great stuff there, including articles about everything from universal basic income, to the dangers of sex robots. Go check it out.