Co-Founder and CEO
November 16, 2022
Bret Kugelmass [00:01:36] We are here today with Derek Sutherland, who is the Co-founder and CEO of CTFusion. Derek, welcome to the show and thank you so much for being here.
Derek Sutherland [00:01:44] Yeah, thanks for having me. Glad to be here today.
Bret Kugelmass [00:01:46] Yeah, no, I'm super excited to go through this with you. I mean, I'm always so curious and interested to learn more about, you know, more the deep technology side of things. So, it's great to have such an expert here with me today. But before we get to that deep technology, why don't you just tell me a little bit about yourself as an individual? Where did you grow up?
Derek Sutherland [00:02:05] Sure. Yeah. I grew up in Orlando, Florida, so I guess Florida man technically. And so, yeah, I just grew up there and I went to a lot of different... had these exceptional teachers during my period in middle school and high school. And that really gave me this awesome trajectory towards science and technology and ended up landing in fusion, which was pretty exciting.
Bret Kugelmass [00:02:31] Well, tell me a little bit more about how you landed in fusion.
Derek Sutherland [00:02:34] Yeah. So I give you, I guess, a short origin story. So, when I was in like elementary school and middle school, I was very much one of the cool, popular kids. I was a total science nerd. Right? So, I was... My first real fascination was actually meteorology. So, I was really interested in being the weatherman in a sense. And so I kind of ran with that for a long time. And then in middle school, I had this just exceptional teacher, Mrs. Maderos, since I want to share this link with her afterwards, probably. And she's kind of the Miss Frizzle from the Magic School Bus of my origin story. And she was just awesome. So you have this like environment where you had these, you know, preteens that are like jacked up on five pounds of sour patch kids that we smuggled in for 7-Eleven and then come into the room and say, "What do you guys want to do today? Do you want to blow something up?" And it was just this energy in the room was palpable. So, it was just this awesome experience. And so that just kind of concretized my love of science. And then I continued on in high school, and I knew I wanted to do science, but I wasn't sure exactly which way to go. So I took chemistry, bio physics, and I also picked up rowing as a crew member as part of Winter Park High School's crew team. That was a lot of fun, but funny enough, my coach for the crew team was the physics teacher at Winter Park High School, and so I took his IP... AP and IB physics class and there was this like one week segment, I think, on nuclear energy, which like maybe fourth fifths of it was on fission. But then there was like one day on fusion, and he basically positioned as, "This is a really hard thing to do. It's how the sun works. Replicating it on earth is really hard. But if any of you guys can figure it out before I check out, I'll give you 100 bucks." That was enough. And then I was like, "That's a lot of money then, and as a grad student, even now." So, that really was quite motivating. So, after that, I was fortunate enough to get accepted into MIT and I really was quite interested in physics. I knew I wanted to do that, but then I was just... Further studied fusion since that class in high school, and I wanted to do that. I just knew that was where I belong. And so during my time as an undergrad, I kind of went to spend my summer vacations or summer between classes going to various fusion groups. So, my first job at their spry age of 19 was at Los Alamos National Lab, working in the fusion group there with Dr. Glen Wurden, who I'll also share this with.
Bret Kugelmass [00:05:18] Los Alamos? That was building weapons then.
Derek Sutherland [00:05:19] Yeah, well, it wasn't fusion weapons, so this was actually the plasma physics group there. And so it was this idea of compressing filled reverse configurations but with like a metal liner. So, it's kind of like you throw a current through the liner, it's like a Z-pinch, and then you compress an FRC. So, I don't think anyone's pursuing that specific variant right now. But the closest analog is like what Helion is working on and where they're using magnetic compression of an FRC. So that was my first experience. And then beyond that I just loved it. And so then I... During the rest of my undergrad, I worked at General Fusion in Canada. In my second summer, that I got to work at General Atomics, in my third summer, on big tokamaks. And then in the latter stages of MIT, my Dennis White was my undergraduate thesis advisor, and I was taking graduate level fusion classes at the end of my undergraduate career. And one of those classes was the ARC Reactor Design class that really motivated the formation of Commonwealth Fusion Systems. So, I was a Co-Author on there and just continued concretizing my love of fusion, and that's kind of how I landed at U.W. to pursue my Ph.D. in Fusion and founded a company after finding some very compelling results.
Bret Kugelmass [00:06:38] Okay, so tell me about that. What was compelling about your specific angle of fusion? And how does that differ than, let's say, the ARC Project at MIT and some of the other approaches that are out there?
Derek Sutherland [00:06:50] Yeah. So, the fusion landscape right now is actually pretty diverse. There's a lot of different projects, a lot of different approaches, and I think of them as all being similar in that we're trying to produce more power out than you put in. That's a basic technical viability requirement for any power plant to work. Otherwise, it's just a net entropy generator, not a net energy generator.
Bret Kugelmass [00:07:11] And, sorry, when you run that calculation net out more than net in, does that include the whole system or does that stop the reactor? Because if it takes electricity as your input, then you also have to go even beyond that in terms of heat because you have to account for the efficiency loss.
Derek Sutherland [00:07:30] Yeah. So, there's a lot of different definitions of gain. There, typically when you hear net gain in this context right now, it's talking about the plasma gain. So, this is the total fusion power you're producing in a plasma, whatever the geometry, divided by the input power into the plasma to keep the system going or to do the pulse. And it doesn't...
Bret Kugelmass [00:07:53] And is there some special significance of that like 1.0 ratio or is 0.98 into, you know, is that like good enough? Does that also see something quite significant or is there actually something according to the physics that's special at that demarcation?
Derek Sutherland [00:08:09] I wouldn't say there's a physics like, "Oh, something dramatically changes there." It's just more of a I guess, a philosophical, emotional milestone. But of course, you need to get to higher gain than that to actually build a power plant because no system is 100% efficient. So, when you include all the power conversion and other things, you have to have above net gain. But since we've been so focused on net gain for decades, that's kind of our our demarcation point.
Bret Kugelmass [00:08:39] So it's one... It's the first, one of the first, critical milestones, on the way to then... So, I guess the other milestones - And you help me because I'm less familiar with this base but let's try to lay them out together - So, first you've got a net gain in terms of just raw energy in to energy out, then you've probably got... Break even in terms of electricity and maybe need three times more energy out in order to make the system not consume more electricity than it can produce on its own.
Derek Sutherland [00:09:12] Right.
Bret Kugelmass [00:09:13] And then there's probably some sort of economic milestone as well, because if you build $1,000,000,000 facility, but you're just using exactly the same amount of energy, it's not very economical. So you probably need to produce, I don't know, what is it? Like ten times more out in that stage or what is there another critical milestone that your community has defined?
Derek Sutherland [00:09:36] Yeah. So I think...
Bret Kugelmass [00:09:38] Or is it like a ratio of like CapEx like energy in to energy out over CapEx of your facilities, something like that?
Derek Sutherland [00:09:45] Yeah. Yeah. So, that's a great question. So, I think what you're getting at is that there's different types of milestones. I would say the lion's share of our focus to date has been on technical milestones. So, power balance is one of those. But you're completely correct. The thing that actually matters to bring something to the grid is commercial milestones, namely, what are the commercial attributes that you need to hit in order to make it compelling enough for someone to buy it? So, this is when you have aspects like levelized cost of electricity, what's the LCOE? But it's much more than just that because you could build a $10 billion, you know, fission system and the LCOE looks good, but the problem is the $10 billion. So it's not just that.
Bret Kugelmass [00:10:30] That's what nuclear fission is running into now.
Derek Sutherland [00:10:33] Right. So, like the total capital cost matters a lot. So, the lower that is with a competitive LCOE, the better. But also considering what Fusion is trying to do, it's really... It's an alternative to, you know, fossil fuel generation and others. And so we really need it to be very reliable. So, it's not just that it's cheap and and low cost. It's also that it's dispatchable, and it's firm generation that you flip a switch and it's there. So just like fission, it has that reliability aspect that we want.
Bret Kugelmass [00:11:08] And I assume that'd be the case, right? Like nothing I've heard about fusion makes me think that it would be intermittent. Right? Like when you turn it on, you're just going to run it continuously for as long as...
Derek Sutherland [00:11:16] That's right. It doesn't require any atmospheric conditions to make it work. Put it that way.
Bret Kugelmass [00:11:23] Exactly. Okay, cool. All right. So, come back to the different types of fusion. Break 'em out for me. You said a lot of them are similar in the sense they're trying to achieve this breaking milestone. What are some of the differences?
Derek Sutherland [00:11:34] Yeah. So, you'll often hear triple product. That's a very popular quantity, often called the Lawson Criterion. So, the in Tau T, this is the density of the plasma times the temperature, times the energy confinement time. That actually is just conservation of energy. So, it comes from the same equation as dividing net gain. So, we often basically put progress in terms of the triple product, but the triple product alone can't tell you what gain the system's at. You have to know what the temperature is as well. Because you can imagine you could have this exceptionally great confinement system, but if the plasma's room temperature or like let's say a million degrees or something, that's not going to generate a lot of fusion power. So, it's not sufficient to just talk about triple product, you need to know the temperature too. So, the bit of physics to pull out of this is that that temperature that you need to run at is largely set by the type of fusion you're pursuing. So, whether it's deuterium tritium fusion, which has the lowest temperature requirements because it has very large cross-sections at low temperature or D Helium-3 or Proton-Boron, which all have their own temperature requirements. So after you select...
Bret Kugelmass [00:12:48] When you say type of fusion, just so I can slow it down for the audience a little bit, you're talking about, let's call it, "fuel." You're simply smashing two really small things together. And there's a different array of different small things you can smash together.
Derek Sutherland [00:13:02] Exactly.
Bret Kugelmass [00:13:04] So, deuterium or tritium which are both just, you know, how would you describe them?
Derek Sutherland [00:13:08] Yes. So, the deuterium and tritium are... It's they're both hydrogen. They're all hydrogen. It's just like H2O in your water. The only difference is that the number of neutrons that are present in the nucleus. So, regular hydrogen, what you think of in your water bottle, is just a proton with an electron bound to it, whereas deuterium has an extra neutron attached to that proton, and then tritium has two extra neutrons attached to the proton.
Bret Kugelmass [00:13:33] And then you can use a couple of other ones too... Your same boron. And what else?
Derek Sutherland [00:13:38] Yeah. So, D Helium-3. So, that's fuzing deuterium and helion, which is a Helium-3 nucleus.
Bret Kugelmass [00:13:47] Helium-3... Where does that come from?
Derek Sutherland [00:13:49] So, interesting question. So, there's actually a lot on the moon that comes from deposition from like the solar wind, again, very much...
Bret Kugelmass [00:13:58] I think I remember a movie about this...
Derek Sutherland [00:13:59] Yeah, that's right. So, you can collect it that way. It's kind of a high CapEx way of making tritium or getting Helium-3. But one other way is that you can actually just make tritium and let it decay. And when it decays, it turns into Helium-3. And so that's something you don't have to go to the moon to do.
Bret Kugelmass [00:14:19] Okay. And then tritium still is relatively artificial or I guess you have to make it inside of some sort of reactor.
Derek Sutherland [00:14:26] That's right. And we use Lithium-6 in particular and we react it with the neutrons from fusion. And that actually generates the tritium self-consistently.
Bret Kugelmass [00:14:35] And we say reacted... Lithium is bigger than tritium. So, you're splitting it, is that right?
Derek Sutherland [00:14:41] Yeah, you can think of it kind of like a little fission reaction, but you're fissioning lithium.
Bret Kugelmass [00:14:46] A little baby reaction. A light element.
Derek Sutherland [00:14:49] It actually is exothermic. So, you do get a little boost to your power because of it.
Bret Kugelmass [00:14:55] Cool. Okay. Sorry. Did we categorize the main ones? The main different types?
Derek Sutherland [00:14:59] Yes. So. So the so the takeaway is that after you select your type of fusion, so whether it's the Helium-3, Proton-Boron, that sets the T, so the in Tau...
Bret Kugelmass [00:15:10] When you say, "T," you mean temperature. That's a shorthand for temperature.
Derek Sutherland [00:15:13] Yes, sorry. The temperature. And then the other two quantities of the triple product is density and energy confinement time. And so there's basically three main camps of fusion which are throttling up and down those quantities with the understanding that the product is what actually matters. And so magnetic fusion is one version. So, that's the camp that we belong to at CTFusion. And so those are characterized by low densities and high energy confinement times. So, some familiar suspects, these are where the tokamaks live, where the stellarators live. This is where the spheromak lives, which is what CTFusion is developing it off. And also the FRC, like it's being used at T-80 Technologies for a steady state magnetic fusion reactor.
Bret Kugelmass [00:15:56] And so, remind me again, what are the inputs and what are the outputs of this category?
Derek Sutherland [00:16:05] Yeah. So from an energy standpoint, you have to produce that magnetic bottle that you're making in some fashion. And so the main thing that you have at your disposal are two technologies primarily. You have external magnets, which are most commonly superconductors of some type when you're going towards a commercial system. So, like for Commonwealth, they're very much predicated on REBCO HTS magnet technology scaling for their total.
Bret Kugelmass [00:16:32] What is that, REBCO? I've never heard of that before.
Derek Sutherland [00:16:34] Rare earth barium copper. So, it's long acronym but basically it's the high temperature superconducting technology.
Bret Kugelmass [00:16:41] And high temperature is still relatively low temperature.
Derek Sutherland [00:16:44] Still cold. Still cold. Better than liquid helium, which is four Kelvin. But you're looking at, you know, something on the order of liquid hydrogen temperature so like 20 Kelvin. But the thing... The REBCO itself can actually go superconducting at much, much higher temperatures. We're talking like in the neighborhood of 70 to 90 Kelvin. So, that's liquid nitrogen. But to get to really, really large magnetic fields, it has to be so cold.
Bret Kugelmass [00:17:10] Got it. Okay, cool. Yeah, we're talking about the inputs. So which fuel type is this? So we got...
Derek Sutherland [00:17:17] Yeah. So, most of these approaches are DT, so that you're using DT fusion. But there are some that are pursuing like, for example, T-80 Technologies is pursuing Proton-Boron.
Bret Kugelmass [00:17:28] And then Proton-Boron is what you call an aneutronic reaction, whereas the DT is a neutronic reaction because it produces... It spits out some neutrons at the end of the day.
Derek Sutherland [00:17:36] Yeah. So, aneutronic is kind of an interesting term because every fusion reaction, including Proton-Boron, does produce some neutrons. It's just that the neutronicity, another word, is less than DT, so generally neutronic is... aneutronic is meaning you're making 1% or less of your total fusion power out in the form of neutrons.
Bret Kugelmass [00:17:58] Got it. And then the DT is, if I recall correctly because at one point I was comparing fission versus fusion for every... Yeah. How many neutrons produced per how many electron volts? Is that a good way of thinking about it?
Derek Sutherland [00:18:13] Yeah. So you can think of it in two ways. So like for a fission reaction, you know, the average yield per reactions per splitting is like around 200 NED or something like that, right? But for fusion, like DT fusion, it's 17.6 NED. So, per reaction, fusion actually generates less energy than fission, but per nucleon, meaning how many neutrons and protons are involved at a time, it's actually higher power density or energy density than fission. So just two different ways on this.
Bret Kugelmass [00:18:47] Higher energy you're saying on a neutronic or on a nucleon basis, but not on a... Like a if you measure like the volume of the reactor basis.
Derek Sutherland [00:18:57] Right. And fission also is very high power density just because it's in a solid state.
Bret Kugelmass [00:19:02] Right. Right. Right.
Derek Sutherland [00:19:03] And plasmas are not... They're in a plasma state.
Bret Kugelmass [00:19:07] Like this ephemeral gas... You can wave your hand in it, there'll be nothing there, right?
Derek Sutherland [00:19:11] That's right. Yeah. It's MFE in particular is pretty much a vacuum according to our standards.
Bret Kugelmass [00:19:17] Yeah. Dennis was describing this to me, "It's like 100 million degrees, but you can put your hand in it."
Derek Sutherland [00:19:21] Yeah, exactly. So, it's kind of there's more thermal energy in a bathtub than there is in a fusion reactor at full power which is kind of crazy.
Bret Kugelmass [00:19:30] So crazy. So, you say - and we'll come back to that in a little bitbecause I want to hear how some of these things translate into like power plant economics, too - But let's continue the technology differences. Okay. So you are in the DT magnetic camp, and then we'll just put the other ones aside for now. Help me differentiate the different strategies within that. What's different between a spheromak and a tokamak?
Derek Sutherland [00:19:54] Yeah. Yeah. So, the answer to that is really driven by how you choose to make the magnetic fields. So the first option, like I mentioned, is those external magnets. So whether they're superconducting or copper or something else, or you can choose to actually drive an electric current in the plasma itself. We call that a plasma current. And from Ampere's Law, we know that flowing electric current produces a magnetic field no matter where it's at. And so the differentiation is that we actually choose to use more plasma current in the spheromak and the FRC in particular, versus the tokamak in a stellarator, which uses much, much more external magnets instead. And those have implications for the technical approach in detail, but also the commercial viability aspects later on too.
Bret Kugelmass [00:20:47] Go one more step deeper. Explain those implications to me.
Derek Sutherland [00:20:51] Yes. So, a superconductor by definition means it has no resistance. And so when you're flowing a current in there, so long as you're keeping the magnets cold, there's really no input power in steady state to keep it going. So long as you're actively cooling the magnets so that they don't heat up beyond the superconducting temperature. On the other hand, if you're flowing a electric current in the plasma itself, the plasma actually is a fantastic conductor of electricity. And it's one of those weird materials where it actually becomes a better conductor the hotter it gets. And so, like, if you're looking at 150 million degrees or so, you're actually a much, much better electrical conductor than copper.
Bret Kugelmass [00:21:36] In the plasma itself as a conductor?
Derek Sutherland [00:21:37] In the plasma itself. Yeah.
Bret Kugelmass [00:21:39] How do you get energy into a plasma? Where does it come from? Wires in the wall? How do you get in?
Derek Sutherland [00:21:45] Yeah. So, there's various ways to do that. So, the way that we use it is as inductive current drive. So, you're basically changing the magnetic flux somewhere else and then you're inducing an electric field inside of the plasma, which if you apply an electric field to a conductor, it drives a current, just like in a regular circuit.
Bret Kugelmass [00:22:02] Okay, so you still have magnets surrounding this sphere thing?
Derek Sutherland [00:22:06] Yeah.
Bret Kugelmass [00:22:07] And those magnets have, like, giant cables plugged into the wall or something? I mean, I guess, like, what is the external power input to the system? Just so I can help visualize, like, are these like giant cables that would come from a transformer yard or how much power are pumping into it initially?
Derek Sutherland [00:22:29] Yeah. So, two different parts of it. So, let's remove the external coils for a second. Let's just look at the plasma for now. So, in order to drive a plasma current, we have to drive an electric field in some capacity so that we can drive a car. So what we do is we use this technology called helicity injection, which is a form of plasma current drive. And the idea here is that you're actually injecting this quantity called magnetic helicity, which is basically the twistiness of magnetic fields into the confinement chamber. And what's kind of amazing is that that is a conserved quantity over very short timescales, but over longer time scales like, let's say, the resistive time, how long it takes for the current to decay from resistivity, that's the only thing that can destroy it. So basically what those helicity injectors do is it provides constant helicity injection to balance resistive dissipation in the plasma, which is steady state operation. What does helicity injectors look like actually varies depending on the specific form of it. And so there's three main versions of helicity injection current drive. There's coaxial helicity injection, study inductive helicity injection, which is what we're doing, and then local helicity injection, which is actually being pioneered by Pegasus in Wisconsin. And so the closest analog to something that you might... I think you've seen before, is that take Zap's gun approach for how they form their Z-pinches and add in an external magnetic field around their electrodes. That actually is a coaxial helicity injector. So, the technology is very similar to that. In their case, they're not injecting any helicity because they don't have any external feel, but you can add it on and actually do it. So, we've actually used that form of helicity injection on tokamaks before and previous spheromaks as well.
Bret Kugelmass [00:24:31] And so spheromak is what you're doing, right?
Derek Sutherland [00:24:34] That's right.
Bret Kugelmass [00:24:35] So once again, paint like a picture my mind and for our audience as well, I'm assuming it's a sphere of some sort like some kind of spherical chamber?
Derek Sutherland [00:24:42] Yeah, so the word comes from a spherical topology. So, it's a spherical chamber that confines its plasma. The plasma itself is actually a self-organized plasma that is in a toroidal geometry.
Bret Kugelmass [00:24:57] Within the sphere because tokamaks also have that donut like structure to help create that toroidal...
Derek Sutherland [00:25:05] Right. Yeah. I think of it like in the spheromak, the plasma has a donut hole, the machine does not. And a tokamak, the plasma has a donut hole and the machine has a donut hole.
Bret Kugelmass [00:25:15] Are there advantages to that physical orientation, or is it more about how you create the fields?
Derek Sutherland [00:25:21] No, there's a considerable advantage. So, one thing is that there is nothing physically in the donut hole region in a spheromak. So, there is no toroidal field coils, there's no central solenoid, there's just plasma. And so the benefit of that is that you don't have to protect any of those superconductors from high energy neutrons because they're not there. And so it results in a much more compact version of magnetic fusion than if you were to compare it to a tokamak or a stellarator system.
Bret Kugelmass [00:25:51] Okay. So, you don't have to put that kind of thing inside. Awesome. How do you protect stuff on the outside? Because you still a bunch of neutrons that flow outward from your reaction, is that right?
Derek Sutherland [00:25:58] That's right. Yeah. So, outside of it, it looks more common, I guess, to other MFE approaches where you have some plasma chill interface face that's going to see a plasma heat load and a neutron flux. So, what you make that interface out of is important. And then behind that we're using... We're envisioning using a liquid immersion blanket, which can be either led lithium, which is very popular material that everyone's really looking at in detail.
Bret Kugelmass [00:26:26] And there's also a coolant, right? When you say immersion you mean like a coolant.
Derek Sutherland [00:26:30] Yeah, it's a coolant. It's a moderator for neutrons. And it also has the lithium in it to produce tritium.
Bret Kugelmass [00:26:36] And when you say a moderator, you don't mean it in the sense of like fission reactor or moderators is necessary to keep their criticality. What you mean more is you just want to slow down the neutrons, almost like a shield like manner, so they don't stay high speed, high energy and hit the wall behind it. Right?
Derek Sutherland [00:26:57] Yeah. So I think of moderator I mean, what are you moderating?
Bret Kugelmass [00:27:01] You're actually trying to thermalize the neutrons, so they react to your lithium. Okay, sorry. Sorry. Now I understand. I spoke too quickly I didn't hear what you were saying at the end there, which is that you also have the lithium inside.
Derek Sutherland [00:27:11] Yes, that's right. And the lithium has a big cross-section for thermal neutrons.
Bret Kugelmass [00:27:16] Got it. Okay. So, I got it. Okay. So, you have this immersion blanket. It's a coolant. So, it's like a... Okay, so you're starting with a sphere, then you have like another chamber filled with liquid that has your... The lithium in it, which is going to become your fuel and lithium. Lithium-6? Lithium-7? Which lithium is it?
Derek Sutherland [00:27:35] Lithium-6 is the dominant isotope and it's actually the minority in naturally occurring lithium, I think it's like 7.6% of naturally occurring lithium.
Bret Kugelmass [00:27:43] So, you have to enrich the lithium first. That's not a problem. We do that commercially all the time, right?
Derek Sutherland [00:27:47] Well, actually not quite. It depends. So, there's actually different designs and some you have to enrich the Lithium-6 above natural, but some it looks like you might be able to get away with natural lithium. So, it just depends if you're making enough tritium or not in practice. And that's something that we have to work on.
Bret Kugelmass [00:28:06] Got it. Okay. So, you mix that up with lead. Now you've got your liquid-y blanket. What's beyond that? Where do magnets actually sit in this whole thing?
Derek Sutherland [00:28:15] Yeah. So the nice thing about a spheromak is that the only external magnets you actually need are the equilibrium filled coils. And that's a hard physics requirement you can't get away from for a continuously operating system. You have to have some magnets. So, we strip it down to the bare minimum of external magnets that we need for our potentially operating system. And that's it. We don't add anything else to it. And those magnets are outside of the blanket and they're outside of the shield. So, it's extremely easy to make them last decades because of the geometry. You don't have to have it in the middle of the donut hole where it's more challenging.
Bret Kugelmass [00:28:54] Cool. And but are they exactly or are they just coiled up copper wires or something? Or is there some sort of... Like when you say magnet, what type of magnet?
Derek Sutherland [00:29:03] Yeah. So, they're just big circular polloidal fill coils we call them or equilibrium fill coils. And so for the commercial system, they'll be either LTS, low temperature superconductors, or HTS magnets, or they can be copper. Just we're going to use copper in our next device just because it's cheaper and we don't really care about how much energy we're spending during a short pulse in copper. But for the commercial system, they'll be superconducting.
Bret Kugelmass [00:29:30] And help me understand that distinction a little bit there. When you're wrapping it in just copper coils, like I think of like, you know, I've seen copper coils that can create a magnetic field before. It's just, you know, you do a little loop and you pass a current through it, right? Is the reason that that doesn't work on the commercial system because like they get hot or something or what's the problem?
Derek Sutherland [00:29:51] Well, you think that your goal is to sell a watts to customers and so you want to reduce the amount of watts you're using just to heat up copper instead. So, I think the cost benefit of using superconductors is better and is actually, in a tokamak, you can actually calculate since they use such a giant toroidal magnetic field. If you were to have that in copper instead of superconductors, the cooling requirements would be amazing. But it's also just the raw amount of input power you need to keep it going basically means that you're spending all the energy you're making to keep the system going. So, it's just not... It's a low gain system. It's not as attractive. So it's just a way to get... increased gain.
Bret Kugelmass [00:30:30] So good for the prototypes to get the rest of the system. But then you eventually want to move to these low temperature superconducting magnets. What are low temperature superconducting magnets? Like what do they physically look like? Because I know what a copper coil looks like. What do these things look like?
Derek Sutherland [00:30:46] Yeah. So, yeah, there's some cool pictures online. So, there's a couple, two primary versions of LTS technology is Niobium Titanium, which has been extensively used in fusion devices around the world already. And there's also Niobium-10, which is what's being used for the toroidal fill coils on ITER. So, you can kind of do a cross-section of these coils and you'll find that most of the coil is actually not superconductor. Most of it is just like substrates and copper stabilizers and different structural mechanical things to make the coil stable when you have a lot of current going through it. And there's just little slivers of superconductor laced into them and that's where all the current's being held in a steady state sense.
Bret Kugelmass [00:31:36] And then... So you get something really well... Okay. So, the coolant that sits outside of your fusion chamber needs to get hot enough to create energy... Or you need direct electricity conversion or are you doing a coolant and then you got a normal power plant?
Derek Sutherland [00:31:53] No, for the DT version of our systems, it'll be just the heat engine because 80% of the power comes out of neutrons. So, you're moderating neutrons in the blanket. The blanket heats up. Push it through a heat exchanger to a secondary cycle and you spin a turbine to make electricity.
Bret Kugelmass [00:32:09] The DT version. Is there a different version?
Derek Sutherland [00:32:10] So, that we're thinking about what comes after DT, assuming it works, and so spheromaks like FRCs are higher beta approaches to magnetic fusion. So, what this means is the plasma pressure that you get divided by the magnetic pressure you're using to confine it... When that ratio goes up, it makes advanced fuels like D Helium-3 or DD or something like that more attractive because it makes power balance easier. Whereas something that's lower beta like a tokamak or stellarator, it's really challenging to envision using anything but DT because the power balance gets harder as you get to higher temperatures with large magnetic fields.
Bret Kugelmass [00:32:52] But then how do you... If you get something really hot outside of your fusion chamber to create your heat engine and something really cold outside of that to create the magnetic fields, how do you create that stark temp... How do you manage that temperature right now?
Derek Sutherland [00:33:07] Yeah. So, I mean, the nice thing about spheromak is that, you know, you're going from the temperature... The majority of the temperature gradient is in the plasma itself. So, you're going from the the core of the plasma, which is around 150 to 200 million degrees. And then you're going to whatever the wall temperature is, let's say 500 C, between friends, and then so you're pulling out a 150 million degree temperature gradient in a meter or so.
Bret Kugelmass [00:33:37] That doesn't have any mass to it, so nobody cares about that temperature.
Derek Sutherland [00:33:40] So the Delta T temperature...
Bret Kugelmass [00:33:43] Huge block of led or flowing led or whatever that what... Let's say 500 degrees or whatever or 400 degrees Celsius, let's say... But then... Go ahead.
Derek Sutherland [00:33:53] Yeah. The temperature gradient that you're actually solving for is like going from 500 C to something that's a couple degrees above absolute zero and that you can do with proper insulation. So just make sure you put your cozies in the right place when you're designing your thermal.
Bret Kugelmass [00:34:08] What's the distance between those? Because I imagine there's some distance constraint. You got to get those magnets close enough to the fusion reaction. And I can imagine it's really hard. Like even like if you had a meter of isolation to separate 300 degrees C and negative whatever a couple of hundred degrees C... It's hard enough even just to keep something negative a couple hundred degrees, say, in a normal room.
Derek Sutherland [00:34:31] Yeah. So it's on the order of a meter, of total distance, give or take a little bit more for the shielding. But it actually... It's surprisingly... You can do it. It's not... Vacuum is a fantastic insulator and so you can, you have...
Bret Kugelmass [00:34:47] But not very good at stopping neutrons.
Derek Sutherland [00:34:50] Correct. But remember you've already gone through the blanket in the shield.
Bret Kugelmass [00:34:54] Can you really capture all the neutrons or do some still leak out beyond your shield?
Derek Sutherland [00:35:00] You can do pretty darn good. I wouldn't say you're going to capture every one because it's an exponential, but you can do really well.
Bret Kugelmass [00:35:06] Okay. All right. So, you've got most of the neutrons. You stop those, then you've got some sort of vacuum as an insulator. Then you've got your... and you need a separate cooling system to keep those magnets, you know, whatever.
Derek Sutherland [00:35:20] That's right. Now you have to... It's a cryostat system. So, it keeps it cold and superconducting and happy.
Bret Kugelmass [00:35:26] Okay, cool. And then you've got some things that create the magnetic field. Okay, awesome. All right. So, everything we've described are like some mechanical challenges. What are some of the fusion challenges that you have to overcome?
Derek Sutherland [00:35:38] Yeah. So one, I think competitive advantage of our approach is that we're using kind of a similar physics foundation as tokamaks, which we already know work quite well from a physics perspective. And so it's this principle of a toroidal magnetic confinement system. So, donut shaped magnetic model to confine a hot plasma. The main difference between us and them is that we're producing much more of the magnetic field with the plasma current instead of the superconductors on the outside. So we basically have to show that we can achieve the same level of absolute performance from a plasma perspective as a tokamak. But with all this considerably simpler, more compact and less superconductor material machine. And so we haven't done that yet, but that's what our next step in our roadmap is to do, is to reach the same level of triple product and temperature performance as tokamaks with the spheromak instead.
Bret Kugelmass [00:36:38] But what's stopped people from scaling up in power so far? What's the real secret challenge to make this work with an increasing amount of energy production?
Derek Sutherland [00:36:49] Yeah, so I think the central challenge of most fusion approaches is what's the confinement like? And so what we've found is that it's spheromaks are actually quite good at confinement when they're operating in their naturally stable state. And that's great. However, one issue is, is that you have to sustain that plasma current in some fashion. And so we use helicity injection current drive to do that. So, the challenge is if you're looking at a true steady state system, you have to ensure that you have good enough confinement to get to net gain in a commercial system while you're sustaining the system in steady state. And so previous spheromaks kind of struggled with that where they... It was kind of like a either or. You could have either good confinement than have it be pulsed or have relatively poor confinement while being driven. And so now we believe our approach could provide a pathway where you can have your cake and eat it too. Where you can sustain it with good confinement, and you can operate steady state.
Bret Kugelmass [00:37:55] What actually is confinement? Like try to tell me more like physical terms. What is creating confinement?
Derek Sutherland [00:38:01] Yeah. So it's like....
Bret Kugelmass [00:38:02] Being a magnet at different frequencies or what's reading the environment?
Derek Sutherland [00:38:08] Yes. So, the principle of confinement comes from the Lorentz force, which is basically F = ma. It's Newton's second law. And it basically says that a charged particle that's moving in a magnetic field feels a force that is orthogonal or at right angles to both its velocity and magnetic field. So, what this looks like is that you have charged particles that basically gyrate or spin around magnetic field lines. And so you say that they're confined across the magnetic field lines, but they're not confined parallel to them. And so the insight way back when, I guess that was in the fifties, is that, well, the end losses are the problem. So, why don't we just wrap the thing into a taurus so we have cross confinement just fine? And then the dog chases its own tail as the particles swirling around the donut and they stay confined in the object long enough that they can fuse. So, it's the magnetic fields that are doing the confinement in all of these approaches.
Bret Kugelmass [00:39:09] I've actually got two questions based on the stuff you said. I guess first on the continuum on this path, like what is the thing that everyone goes to work during the day like doing to make this confinement work? Are they working on the mechanical system itself? Are they working on the wrapping the coils of the magnets differently? Are they working on like the controller that programs the magnets? What is the thing that is going to allow for this confined problem to be solved?
Derek Sutherland [00:39:38] Yeah. So, the answer is yes to all of those.
Bret Kugelmass [00:39:42] Yeah, am I missing anything? I think I listed three. Am I missing four, five, six?
Derek Sutherland [00:39:43] So, it depends on exactly what approach you're looking at. But in aggregate, designing a system to do this, is a challenge. It doesn't matter what approach to fusion you're doing. And so after you design your system physically, like you build the machine and then you have the magnets you decided to have or not. Then it comes down to you have to produce this plasma, and this plasma has to hold on to its heat good enough to make the triple product even reasonable. And so what that looks like is a lot of different aspects. But from a physics perspective, the main thing that we found that sets the confinement time is actually turbulent transport. And a lot of our effort and work really over the past decades has been better understanding how transport scales with size...
Bret Kugelmass [00:40:39] What's transport?
Derek Sutherland [00:40:40] Transport is how fast does the heat leak out from the plasma. It's directly the energy confinement time.
Bret Kugelmass [00:40:47] And it leaks out in what form? Radiation or what is the form of heat transfer?
Derek Sutherland [00:40:51] Yeah, you can think of it mainly as conduction, but there's also it's... There's convection going on too. There's turbulence.
Bret Kugelmass [00:40:58] But is it not a vacuum inside the reactor?
Derek Sutherland [00:41:01] It is. It is. Well, remember the...
Derek Sutherland [00:41:05] Where does heat move via conduction or convection through a vacuum?
Derek Sutherland [00:41:07] So it's all the types. So it's it's not a true vacuum. You know, there is particles in there and the particles can interact with each other electrically because they're charged, because they basically fill each other's magnetic electric fields. And so you can think that that's a way that you have friction between particles and they can convect or conduct it's heat through that. You also do have radiation, electromagnetic radiation or light and a common term that does...
Bret Kugelmass [00:41:39] I just mean as a form of heat transfer. I'm wondering like how does it leave the system? Because even if they're interacting with each other and creating that friction and creating, you know, what we think of as heat, that's a good thing. Because you're rising the temperature. It's only a problem if it leaves your vacuum. So, it must leave then in the form of radiation, because there's no other way for heat to move through a vacuum, right?
Derek Sutherland [00:42:01] No, it's forming coming out of conduction. So imagine this...
Bret Kugelmass [00:42:05] Oh! So, you're saying it hits the wall of the, of your chamber. I see. I see. So these very tiny, very light particles that have very few of them, it's enough to be bouncing against your wall and actually creating heat along that wall surface. Yeah.
Derek Sutherland [00:42:19] Then it transports through the magnetic field. And that friction, you think it heats it up. But actually, what's happening is like, let's say that there's one part of the plasma that's hotter than the other. If they tell each other, they're like, "Well, I'm going to transfer some of my power to you because you're colder. So, that means the power, the energy went from here to here. And so that's transport. And that's not light. That's through electric fields and Coulomb collisions.
Bret Kugelmass [00:42:45] Yes. Okay. All right. So, all right. So, you've got... So, what you're saying is like the main challenge that you guys are solving to make the system work is this confinement challenge, which means not losing heat while you're running your steady state system. Eventually, either either losing it to a different part of the plasma or losing it to the wall or something. And the strategy is to do this or what? Just understanding the physics better? Manipulating the magnetic field using fast computers to turn on this coil, turn on that coil, or what's happening to to exert control over this?
Derek Sutherland [00:43:23] Yeah. So from a high level, the best way to have good confinement is to have a plasma that's stable. And that stability means that the magnetic structure that you're building, which should have good confinement, that's why you're building it like you are, stays in that structure and doesn't change shape rapidly or do something that you don't want it to do. So, the magnetic topology, we say, is actually what we're trying to get to and maintain. And so making sure the plasma's stable is a key way of doing that. So if you look at like a tokamak or a stellarator or a spheromak or a sheared-flow stabalized Z-pinch. What are you stabilizing? Instabilities, but it's stable. It's not unstable. So, that same recurring motif is that stability is required for good confinement in most cases.
Bret Kugelmass [00:44:14] Yeah, but I guess what's a strategy to do it? Is it a computer program at the end of the day or is it a mechanical setup that is... to control it.
Derek Sutherland [00:44:25] It's both. So, it depends on the design of the system overall, but it also has to do with the feedback control of the plasma in time.
Bret Kugelmass [00:44:33] So, that's what I was wondering, if there's some way to get control over it.
Derek Sutherland [00:44:35] So, that if let's say you move away from your desired operating point, you have to have something to push it back to where you need it to go.
Bret Kugelmass [00:44:41] And is that feedback control like fundamental physics limitation as well? Because I can imagine computer chips can only think so fast and move so fast relative to a flying particle. Is that part a challenge as well?
Derek Sutherland [00:44:54] So, it is part of it. So we actually are... We use the GPU-based feedback control system on our machine and then we've actually been able to get the latency down to around 15 microseconds, which is pretty fast. Thank goodness.
Bret Kugelmass [00:45:08] If it's 50 microseconds, it's still doing some sort of like macro level control, it can't really understand where individual particles are.
Derek Sutherland [00:45:16] Oh, yeah. It's all macro.
Bret Kugelmass [00:45:18] Topology, as you describe.
Derek Sutherland [00:45:20] For sure. Yeah. Yeah. So, the real limitation is not so much the time of the chips, it's the time to do or the computing power. It's more of like how much math you want to do? How much time do you have it? But even more importantly, how long does it take to make the actuator that's going to change some state actually move? So if it's like if it's a big coil or something like that, then you have an inductance and that inductance slows down how fast you can change something unless you have a massive voltage. So, it's an entire system that you have to optimize.
Bret Kugelmass [00:45:56] And what are the things that knock it out of steady state? When I was talking to Dennis at MIT, he was telling me about like ash, like an ash, like product of your fusion leaves this ash. I guess it's whatever the atoms become after they fuze and are no longer desirable. Is that a problem for spheromaks too? Getting that ash out of the system?
Derek Sutherland [00:46:16] Yeah. So, for any steady state fusion device, you're basically creating the exhaust in the core where you're maximizing fusion so the helium is going to get out. And so the benefit is, is that that's why we don't actually want perfect confinement. We want good enough confinement, but we want enough transport to get the ash out after it's been produced because it doesn't fuse again.
Bret Kugelmass [00:46:41] And how do they get out are there like little holes somewhere that they can be scooped out of? Or is there like a vaccuum that sucks them up?
Derek Sutherland [00:46:46] So, funny enough, that same turbulence that sets the transport, it also convects out the helium. And so then once it exits the confinement system, the magnetic bottle, then it basically gets sucked into a pump. And one of the popular technologies do that as a cryopump. So, it's just a really cold surface that the helium sticks to and then we can remove it from the system.
Bret Kugelmass [00:47:13] Cool. And then one more thing that you mentioned before, I think we were talking about like the Lorentz forces and relationship between magnetism and force. I've often wondered, I think when I did a tour down at Oak Ridge, they were describing this challenge to me as well. You tell me how this challenge is handled. So, you've got all of these magnets and they're creating some sort of physical force as well. Right? Like your I cross J physical force. Where is that force going from these high power banks? Is it just going to the structure of the building?
Derek Sutherland [00:47:46] Yeah. So, you can think that when you energize a magnet, there's a, I guess, a I cross B or J cross B force that tends to want to blow the magnet to a bigger radius. So, there is a large body force that you have to withstand against. It's like you do have a structure to basically ensure that those magnets stay where they should be, so they don't move. So, that just means a lot of steel.
Bret Kugelmass [00:48:14] A lot of steel. And then I guess what I'm wondering is like, okay, so let's say that you have in your steady state reaction, let's say we scale up to a, I don't know, like a 300 megawatt or how big in full production would these reactors be?
Derek Sutherland [00:48:29] Yeah. So we have a couple of... We have a couple of different models. So, one thing since we're more compact and simple than a tokamak, we believe we can do like more of the distributed modular approach. So, like each unit is maybe 100 megawatts electric and then you like multiple and parallel. Or, we can go big and build a big centralized generation like a gigawatt or anywhere in between. And so that's more flexible than I think most companies are looking at right now.
Bret Kugelmass [00:48:59] So, let's say that it's a hundred... I love that it modularity as well... Let's say that it's 100 megawatts electrical output. What does that mean in terms of... and, okay, so you want less electrical input than that, obviously.
Derek Sutherland [00:49:13] Of course, that's 100 megawatt electric net. So, that's how many net watts to the grid.
Bret Kugelmass [00:49:20] Net watts. Okay. So, this gonna be actually much bigger in terms of actual energy.
Derek Sutherland [00:49:25] Yeah. Yeah, it's like around three, 300 megawatts or so.
Bret Kugelmass [00:49:30] 300 megawatts. And then so what would be the current coming into these magnets for a system like this?
Derek Sutherland [00:49:37] So, I need to look at the design point exactly. But you're talking on the order of like mega amps, so millions of amps occur.
Bret Kugelmass [00:49:45] And then what happens and does that translate directly to the force on the steel structures? I mean, there's like a continuous like 100 megawatts worth of force, or in that order of, being pushed back on the steel of the building at all times.
Derek Sutherland [00:50:04] Yeah. So, there is a constant force that the magnets are feeling. But, it's just like being under like, you know, building a bridge where the force is always on the bylines.
Bret Kugelmass [00:50:16] My question is now, what happens when that energy shuts off and we go in a microsecond from that force to none of that force? Does the building itself feel like a delta of that? Like will it feel a hundred megawatts of contraction within the second from when you have maximum full power to no power?
Derek Sutherland [00:50:34] Oh, I mean, I guess I would be relieved because the building is the atlas and it's getting a break when it turns off. But everything's under compression and then when it turns off, the force just is gone.
Bret Kugelmass [00:50:47] And that's just a straight outward force is not a rotational force at all?
Derek Sutherland [00:50:52] Yeah, so most of it is just an outward force. There is, you have to take into account forks and things like that. But the structural engineering of it is actually not as challenging as it sounds, but you just have to make sure F = ma = zero and you're not close to yielding.
Bret Kugelmass [00:51:08] And then does that magnetic change differ from the spheromaks and the tokamaks? Are they experiencing a constant rotational force on your structure?
Derek Sutherland [00:51:18] So, in a tokamak, the... I would say the maximum body forces that the coils are going to feel is going to be bigger than a spheromak primarily because they're operating in much, much larger magnetic fields. And the place where those maximum fields are happening is actually on the part of the coil that's the smallest inside the donut hole. So, since spheromaks don't have that, the actual peak field on coil like around the equilibrium field coils is like less than ten Tesla. So, the body force from the J cross B is actually less than than a tokamak.
Bret Kugelmass [00:51:54] Yeah. Cool.
Derek Sutherland [00:51:55] And a little bit easier to design.
Bret Kugelmass [00:51:57] Yeah. Yeah. It's just the dynamics. Yeah. I'm not worried about designing a building that can withstand that in steady state. I'm worried about like the dynamic state of it. Like when there's a sudden shift in force, what effect does that have on the structure? And then all of this is kind of like leading to the overall economic question, which I promised you would get back to. How do you model the economics of your whole system? Assuming that your reactor works as prescribed, how do you think about the rest of the facility, the OpEx, the CapEx?
Derek Sutherland [00:52:27] Yeah. So, I think every fusion company has to do that type of calculation. And I think the best way with the information we have right now is the bottoms up approach as best as you can, where you basically just like, okay, let's put together a concept, what are all the components we need and what are the costs that we know right now? And you usually like for bulk system components, you can assume some multiple over like raw costs, so like 2 to 3 X for the manufacture cost. And then there are some systems that don't currently exist at scale, like the tritium processing system for the closed fuel cycle and things like that. And so you just do your best on estimating that knowing what you know now. Fortunately for our first generation system, for the secondary cycle, we're just thinking steam, just let's not take on more risk. Let's just do steam for now. So, we can just look at General Electric turbines to see how much they cost. And so that's relatively easy to price out. It's the fusion part and in particular some of the the blanket technology that really no one's built yet at scale. That's the hardest thing to price out in detail right now.
Bret Kugelmass [00:53:41] And then do you have a sense... Okay, so let's say for a 100 megawatt net plant, do you have a sense of what the CapEx of just your normal steam power plant would be for that?
Derek Sutherland [00:53:52] I need to look up exactly the difference. Like the separation between the steam plant and the fusion part. But I think for most of the design points I've seen, it's generally the fusion part is a little bit pricier depending on exactly what you're looking at, but it's kind of... It's not like all the cost is in the fusion part of it. It's just a minuscule amount in the steam plant. Building big steam plant is not, you know, no money. So, I think it's kind of like half and half, but it really depends on the exact approach to fusion.
Bret Kugelmass [00:54:26] And then what about OpEx? I know how many people it takes to run a modern steam power plant. Not that many. What about the fusion part?
Derek Sutherland [00:54:36] Yeah. So the OpEx, I think there's a couple of different parts to just emphasize. So one, the fuel cost is very low. So, it's almost like a renewable in that sense because even though like you know deuterium and lithium, let's say our zero cost since the energy density is so high for fusion, the amount of stuff you need is actually not that much to run a power plant.
Bret Kugelmass [00:54:58] Yeah, I'm familiar with the same logic in the fission industry, but still even so, yeah. Fuel theoretically doesn't cost anything, but you need to process it and handle it.
Derek Sutherland [00:55:13] Oh, sure.
Bret Kugelmass [00:55:14] Pay off the equipment that's involved in all of that. So what is... So, at the end of the day, it's still ends up being, you know, for a nuclear plant, you know, they still spend a fair amount of their OpEx just on fuel, even though it's got the same logical argument. Uranium costs nothing compared to the heat it produces.
Derek Sutherland [00:55:28] Yeah. And I think fusion will be even cheaper than uranium. Just the fuel because it's not a controlled substance whatsoever. It's in your iPhone and your water bottle.
Bret Kugelmass [00:55:40] But wait a minute. That can't be right, though. Once you've done something with neutrons, once you get it to the state, where it can be like... Tritium is definitely a highly regulated substance. Deuterium I assume is also, and I bet even some of the lithium and other stuff is too.
Derek Sutherland [00:56:00] No, I mean, so just to give you a context, so tritium is being produced inside the facility that is being consumed as part of the reaction that's coming from lithium.
Bret Kugelmass [00:56:08] I get it. But it's still highly controlled, even if it's your own facility...
Derek Sutherland [00:56:12] Oh, sure. My my point is that if you're enriching uranium for use in a nuclear power plant, a civilian is not going to be doing that. However, you can buy deuterium and lithium as a civilian because you have lithium ion...
Bret Kugelmass [00:56:28] But not tritium?
Derek Sutherland [00:56:30] Not tritium. No.
Bret Kugelmass [00:56:30] You can buy deuterium as a civilian?
Derek Sutherland [00:56:32] Yeah, we just order bottles.
Bret Kugelmass [00:56:34] Okay. But not the tritium. So, the tritium will still be like highly regulated... Is that by the NRC? Is the NRC involved in regulation of tritium?
Derek Sutherland [00:56:42] It depends on how much there is. But generally, you know, if you look at like a particle accelerator or something like that, that's more what the regulatory environment looks like for that. That substance in particular.
Bret Kugelmass [00:56:54] I think that the NRC... A problem that I've seen with the fission industry. So, I want to kind of give a warning to the fusion people is that once the NRC kind of gets involved, there are a lot of extra precautions and things to do, even if it's all within your own facility and that ends up driving like your kilogram of tritium that comes back into your system, even if it being produced in your system from being produced to come back into your system. All of the extra layers of requirements tends to drive up the cost per kilogram pretty high. But okay, so fuel isn't that much. What about people that operate your system?
Derek Sutherland [00:57:29] Yeah, people I think you can... I mean, there is no working fusion reactors on the grid right now, but I would assume that we can automate quite a bit of it. So, I don't think that the people are going to be a huge part of it, especially on these more modular approaches that are decentralized. You don't need as big of a team to do that. I think the main OpEx is going to be whatever components you need to replace quasi regularly in the system. So, for us it'll be basically the first wall that is going to be replaced probably on an annual basis or so.
Bret Kugelmass [00:58:07] Do you have to take the whole thing apart and drain out the led to get to that first wall?
Derek Sutherland [00:58:11] Yeah, that's right. And then you just remove it and screw in the new light bulb and fill it back up and start her up.
Bret Kugelmass [00:58:18] And that first wall has experienced neutron bombardment, right?
Derek Sutherland [00:58:22] Correct.
Bret Kugelmass [00:58:23] And it's made of some sort of steel or what is made out of?
Derek Sutherland [00:58:26] Yeah, there's a couple of candidates. One of the most popular is radiation resistant foretic steel.
Bret Kugelmass [00:58:30] What does radiation resistant foretic steel mean?
Derek Sutherland [00:58:35] So, it just means that the specific alloying of the metal is done to specifically be tolerant to high energy neutrons or more tolerant than like a run of the mill stainless steel.
Bret Kugelmass [00:58:47] So, it doesn't experience embrittlement. But it still becomes activated, turns into some sort of cobalt or nickel isotope that's embedded in the steel?
Derek Sutherland [00:58:56] Yeah. It's the alloying is chosen to reduce the amount of activation that occurs per neutron irradiation, but also it's more tolerant. So, the way I think of it is like, let's say your regular tires for the steet road have a particular tread thickness, a radiation resistant foretic steel is like putting on off road tires, it's more robust and you can deal with more erosion than you otherwise would.
Bret Kugelmass [00:59:19] But you still have to then handle... So, it becomes activated inner wall and you still have to handle it, which means that you have to go through all like the worker radiation protection stuff too, right?
Derek Sutherland [00:59:33] Yeah. So it depends on exactly what you're choosing. But yeah, one benefit of fusion is that you don't have that intrinsic to the process, but you can choose what you make your reactor out of. So you specifically choose steels that don't activate as much as others.
Bret Kugelmass [00:59:52] The problem, though, with US regulations though is the activation standards are set like six orders of magnitude lower than they cause any harm to a human. And there's no getting around the NRC on that. There's just zero getting around them. So, those standards are set. So, even if it's like ridiculously low, any amount of activation, they are going to classify as like a worker hazard.
Derek Sutherland [01:00:12] Well, I'll defer to the NRC. But just so you know, later this year we're expecting an options paper from the NRC on providing a series of regulatory options for fusion in the U.S. market. And we've been part of that process... I've actually given two talks to the NRC to date at their public forums that they've been holding. So, we don't know yet how it's going to be regulated, but I believe it should be regulated less onerously than fission because of its much better safety.
Bret Kugelmass [01:00:43] But on a per activation basis, it has to be regulated the exact same. They don't change worker safety standards based on technology. They change it on activation level.
Derek Sutherland [01:00:55] So yeah, it's a little bit more complicated, but basically they're looking at what is the correct part of the regulatory code to use for future. Yeah. It's not clear exactly where it's going to land yet.
Bret Kugelmass [01:01:08] Yeah. Yeah. I guess all of this is just going to say, like, okay, like the end of the day, what is our OpEx? What is our CapEx? So we can back out LCOE from that and also no time off the capital. That's really what I'm trying to get through. So I'm trying to like map out all of these different pieces. So say, yeah, let's give you the benefit of the doubt. You are going to make this technology work. I'll give you that. But what's the step next? How do we design a power, a cost effective power plant to wrap around the working fusion core? And what are the economics of that whole system? That's, I guess, where my mind goes.
Derek Sutherland [01:01:45] Yeah. So I mean, I think on the OpEx side, if you're going with steam, you know what the cost is. We're not going into the steam optimization business. There's been hundreds of years under that already. I think what our focus is is that how low of a CapEx can we get for a particular power output? And that's where we think our approach could be potentially much cheaper than some of our competitors because we are more compact, we're more simple and we use a lot less superconductors, so we can prove at scales we think it will be cheaper and if we're cheaper than the total cost would be lower too even with the same fixed steam cycle, let's say.
Bret Kugelmass [01:02:25] And we think we can get this down to like $50 a megawatt hour electricity-wise?
Derek Sutherland [01:02:30] You know, that's our goal. I think $50 per megawatt hour is a good goal to go for. I think ideally you want it to be much less than that.
Bret Kugelmass [01:02:40] Yeah. Yeah. Cool. Awesome. All right. Well, I have dragged you through the technical gamut on this one. I'm sure we could talk for hours more. Are there any final words you want to leave the audience with?
Derek Sutherland [01:02:50] No. Just appreciate your time. And yeah the fusion space is really developing nicely and there's a lot of different approaches. So, if you want to jump into fusion, now is a great time and there's plenty of companies hiring, so come find us.
Bret Kugelmass [01:03:05] Awesome. Love the work that you're doing. Thank you so much for... I just know how much hard work this all is and you're breaking through scientific barriers here and I just think it's amazing. So thank you again for taking the time to walk us through it all.
Derek Sutherland [01:03:18] Yeah, my pleasure. Thanks.