Bonus: A World with Abundant Fusion Energy with David Kirtley of Helion Energy

0:10David, welcome to Age of Miracles. Thank you for having me. So we're going to start you out with our traditional opening question, which is, what do you think the pie of energy sources looks like in the U.S. in the year 2050? Yep. So we think about this a lot, actually. You can look at how power sources are growing in the U.S. Renewables are increasing, more solar, more wind. I love that. It's really great. And there are places, however, where we're starting to cap some of those out, actually. and putting in less solar. And I'm up in Washington State where we put in a lot of hydro throughout the last century and now don't really put any more.

0:51But fossil fuels have continued to grow. So I believe renewables will continue to add more market share. And I think that's really great. But I'm also concerned and it's why we founded Helion that we need something else. We need something that's going to replace the fossil fuels that we're continuing to grow when really a lot of us believe you should be replacing those. And so by 2050, we better have Fusion on the grid. There'll probably be multiple players in Fusion as well and other renewable and carbon-free power sources that are all over in the United States in different locations. Great. Fantastic.

1:31Well, David, I get to ask the more softball question after Packies, which is, what's your background? Can you tell us a little bit about yourself and how you got into Fusion? Yep. Great. Thank you. So my background, I got into science and engineering wanting to solve really hard problems. And I, you know, as most kids in high school, you know, you're going into undergrad and you're saying, great, I'm going to go solve the biggest problem in the world. And I want to go learn how to do that and what that means. To me, what I saw that is I saw energy. I saw wars fought over oil and things and said, how do we go out and solve that?

2:09And you look in the world and you see energy density and how you make electricity in the universe. Most of the power is made from fusion. And if you look at other other sources of energy, literally joules per cubic cubic meter or joules per kilogram of fuel. Well, right, nuclear power in all of the forms is right there and seems so obvious. So I got into fusion, spent a lot of years learning fusion, and then discovered, like a lot of people in fusion, that it's so far away. It's so many decades away, and the technologies were so massive and expensive that it wasn't going to happen in my lifetime.

2:44And I'm a builder. I wanted to get my hands on those machines. So I pivoted, worked on space propulsion for a while, building plasma thrusters, Hall effect thrusters, some of the things used in Starlink today. to actually go build some spacecraft. And it wasn't until I found some of the technologies that Helion uses that I said, wow, we actually have a chance to do this in my lifetime. I want to go do this and I want to put all in on it. And then I went to actually go build those machines, get the wrenches out and actually help build them. That's amazing. Maybe a segue there. Can you tell us a little bit about Helion Energy?

3:18Yeah. So Helion, we do fusion. We've done fusion a long time. And our goal is to do fusion in a way that's different than other people. Our goal is to take lightweight isotopes of hydrogen and helium, fuse them together under intense pressure and form heavier atoms and release a tremendous amount of energy. But we don't want to release heat energy. We want electricity. And so our goal is to do fusion in a way where we can directly harness the electricity from that fusion reaction as electrons and get it out on the grid as soon as possible. So there's a whole bunch that goes into there. Our systems are pulsed and electromagnetic, but really always the focus is how do we get electricity out of fusion as fast and efficiently, as efficiently as possible.

4:03What's the answer on that one? We had a really interesting episode where we talked to people who were in favor of solar, in favor of other things, but particularly the solar people were like, yeah, the way we do fission in particular is like kind of caveman ooga booga, where like heat something up, make steam, turn a turbine. Obviously, like being able to just kind of generate electricity directly is great. Like, what does that take to do? Yeah, it takes a level of technology before it actually can happen. So this is something I think about a lot in that the first cars were electric cars. In the 1800s, there were these electric cars in New York.

4:40They were a commercial product. But the batteries didn't exist. The motors didn't exist. The transistor didn't yet exist. And so they couldn't actually make that small niche product into a widespread car. Then gasoline engines took over and we had 100 years of gasoline engines. And we're only now at the place where we have the power electronics that are efficient. We have regenerative braking and electric regenerative braking. We have lithium batteries. Finally, now the electric car makes sense. And so if I was doing fusion in the 1950s, I'd be doing thermal fusion, too. I'd be using the energy conversion that we could do then, even though fusion makes charged particles and electrons already, but I'd be using those technologies.

5:22So it's taken modern high voltage power electronics, fiber optics, gigahertz speed computing before we can really, you do fusion in the way that harnesses the electricity directly. Can you walk us through the design of Helion's fusion generator? Yep. So we've spent a number of years building prototypes to get to where we are today. The earliest prototypes were funded by the Department of Energy. And then we graduated as they got bigger and started to make more energy to privately funded and venture capital. But the design we have today works like this. On either end, we inject our fusion fuel, deuterium and helium-3, a rare isotope of helium-3.

6:06deuterium, which is a heavy hydrogen found in all water. And then we ionize it. We superheat it into a plasma, heating it in our formation sections. We heat it over about 3 million degrees just to get the reaction started. We then, using pulsed magnets, accelerate it into a central compression, think of a piston. And then in that central compression area, we then squeeze it as fast as possible in less than a thousandth of a second. We squeeze it to a thousand atmospheres, where fusion starts to happen. And once fusion starts to happen, it gets hot enough, over 100 million degrees, fusion starts to happen.

6:41And then it pushes back on that magnetic field. And it's now the same, actually same physics that happens in the regenerative braking in an electric car. We can then recover that electricity right back out of the reaction, expanding it, cooling it, exhausting it, waiting for the next cycle, puffing in the fuel and repeating it and running that. The goal is to run that over once a second and that process repeatedly. So what are the trade-offs that you're making? I mean, we talked to a bunch of fusion companies. Everybody's making trade-offs. This seems like an interesting approach, kind of combining the best of MCF and ICF.

7:17But what are the trade-offs that you're making to get that to happen? Yeah, I think you summarized that exactly right. The goal is to take the best of magnetic confinement, which is that keeps that 100 million degree fuel from touching the wall. because you don't want that hot fuel to ever touch the wall. And the best of inertial confinement, which is don't hold on to it forever. Nature is unstable and doesn't like that. Instead, squeeze it and get to fusion as fast as possible. And then adds that third one that most people aren't doing, which is directly extract that electricity. The trade-off of it, the big trade-off is that all this has to happen fast.

7:54So it's all pulsed. That's the inertial part. The beauty is you get to do it fast. The trade-off is you have to, which means now you need triggering systems that respond in nanoseconds, in billionths of a second. Technology didn't exist 10 or 20 years ago. And you need massive pulse power systems, big high voltage electronics. And so that's the big trade-off there. So we have to design and run these big power electronics. In some way, Helion is more an electronics company than it is a fusion company. And that that's where a lot of our technology and our team focus on is those big power electronics.

8:31So that's one of the big tradeoffs we do in those systems. So, yeah, I think that those are probably the two I would focus on. Speaking of different approaches, we also in this podcast have been covering fission reactors. And in fission reactors, you also see a lot of different, you know, reactor design choices between different types of approaches in fission. Is that a decent analogy to make to the different approaches we're seeing in fusion as well and the tradeoff and decision making that goes into that? I think so. In that also in fusion, you have much like in the fission landscape, you have a lot of different scales because we do these pulse systems.

9:13Our focus is building systems that are smaller, 50 megawatts versus the big scientific programs are five gigawatt scale. So many orders of magnitude smaller in scale. And we focus on different approaches of harnessing the electricity, as well as maybe different business cases. Where are you going to site these? What are you going to use them for? And at Halion, we're focused on right now our first customer, Microsoft. Can we build fusion for data centers? And that means fundamentally low electricity. Thanks for listening so far. Hang on. We'll be right back after a quick word from our sponsors.

9:51We are living, living in the future. And speaking of living in the future, helium three, watch for all mankind. And I don't know if you watched, watch the show, but like the, the boon that you get from going to the moon is that you've got all this healing three and there's not a ton of it on earth. How are you sourcing helium three? Is that another like kind of trade-off that you make? How'd you, how'd you land on helium three? Yeah. The I love for, for all mankind. I do say that if you need to start your business by going to the moon, you probably have a rough business model ahead of you and a rough road ahead of you, as maybe was seen in the TV show.

10:30But for Helion, in fact, what we named the company after, the nucleus of a Helium-3 is called a Helion. The Helium-3 fuel is one of the older fuels, actually. And the brilliant scientists that did a lot of the early work in fusion recognize helium-3 would be a really great fuel because you take a deuterium and a helium-3, and when you fuse it, it forms a helium-4, regular old balloon helium and a proton, but all charged particles, all electricity, all trapped in the magnetic field. Two challenges. One, just like you pointed out, there isn't a lot of helium-3 on Earth, so how are you going to even get helium-3 to test it?

11:09Then how are you going to generate it in your system? And the other is that it requires higher temperatures to operate. So both of those are are two negatives that you have to overcome. The first one we solved with, we patented this, but we solved this by essentially the high efficiency, this energy recovery, what it really lets you do is do fusion a lot cheaper. And this is the thing that we got really excited about. You take two deuteriums, not a deuterium and a helium-3, but two deuterium atoms, two deuterons, they're called. You fuse them together and they make helium-3 in gas form. So if you have fusion already, then you can make helium-3 to do helium-3 fusion.

11:51But the key to that, unlocking that, is having really efficient fusion and really efficiently putting electricity in and recovering it out. And if you're not doing that process, it's hard to make helium-3. And then it's hard to burn the helium-3 to make electricity. And so it was a little chicken and egg problem that required modern high-speed transistors and fiber optics to unlock that part. Awesome. You had mentioned Microsoft earlier. Perhaps we can take a pivot there. That was really exciting news for anyone following along with Fusion. You know, plan to deliver in 2028, really fast timelines.

12:28What will it take between now and then to get there? Yeah, our focus at Helion has always been, a lot of us at Helion came out of some of the scientific and academic programs where we were focused on discovering physics and doing new diagnostics and learning what we could about fusion, but not delivering a product. When we spun off Helion, our goal was to make electricity on the grid as soon as possible. Even if sometimes it's not as fun, even if sometimes it's not as elegant, what shortcuts can you make to move faster? And so things we really do, and that's part of been the mantra of Helion, is how do we iterate really quickly?

13:09Build now, we're building our seventh prototype for Helion. How do we actually get electrons demonstrated and on the grid as fast as possible and engineer systems that are easy to make? And so that has been the mantra as we built all of these prototypes over the years. And we've worked with Microsoft actually now since 2015. So seven or eight years we've been working with them. And they've come along on several of our prototypes on the journey of the evolution of us hitting our milestones to actually deliver these things. And so we've been working with them for a long time on get them used to Fusion, as well as working with them on what does a customer need?

13:47What is electricity for a data center? What are their requirements? Should you put it next to the data center? Should you put it on the grid? There's all kinds of really interesting questions that they could answer for us. So we've been working together over the years to sort of right size fusion. And that has helped us land on 50 megawatt scale systems as the baseline scale systems, because that's a good scale for their data centers as well. And so we've been working together on those over the years. It's great to have a partner like them. And it's a lot of the focus of the business. How can we iterate quickly and get it out there?

14:21And I'm happy to talk more about some of the ways we do that. Yeah, I mean, I would love to. It's a little off script, but, you know, there's obviously a lot of skeptics. Like there's everybody else saying we're doing this in the 2030s. And then you're saying we're doing it in 2028. What do you think is their best argument against it happening? And then what do you think they miss? A lot of it comes back to looking at how modern hardware technology companies operate. Actually, it's less on the physics and more on how are you building a company? Think about the SpaceX's and the Tesla's of the world and many others.

14:59But a lot of the modern aerospace is a good example of can you build and test as fast as possible and iterate? And where while in January, we were running our sixth generation system while physically building the seventh generation system and engineering the eighth generation system and doing that all at the same time. That's how you speed up the process. And so our first peer reviewed, published, we did lots of thermonuclear fusion happened in 2011 on a small scale system funded by the Department of Energy. And since then, we've now built four more systems iterating on that, increasing the yield, increasing the neutron output, increasing the fusion reaction rate.

15:39Published about a year ago that we had been the first ones to do deuterium and helium-3 fusion at all. In bulk fusion, we were the first ones to do that, we think, ever. And then also that we were the first private company to hit 100 million degrees, that operating temperature for fusion, that key temperature. And so we've set those milestones and those metrics all along. But a lot of that comes back to the philosophy of how can you build fast? Is that diagnostic going to take you four years to build? Well, it's too long. We're not going to build that diagnostic, even though it's the best one.

16:10Is there a cheaper diagnostic that's faster than I could build in six months? That's the one I'm going to pick. And so we keep that at every stage, even though sometimes it's hard. Sometimes it's a bit chaotic to have all of those parallel things happening all at once. So you talked about landing at 50 megawatts as a size that works well for your customer. Maybe talk a little bit about the trade-offs for size of fusion reactors. Again, going back to the fission analogy here, there's trade-offs in terms of economies of scale, but perhaps economies from production lines and things like that, depending on if you're going large or small.

16:45I'm curious to get your take on that for fusion reactors in general, or maybe specifically for what you guys building. Yep, absolutely. The physics is all very similar, and it all comes to surface to volume. You have a volume that the reactions are happening in. If you increase the radius or increase the size, you get volume cubed. You get a lot more reactions for a given amount of hardware or surface or even capital investment. Fusion works the same way. Combustion works the same way. Fission works the same way. And so you get economies of scale to do that as you increase in power. And then also when you're running thermal systems and doing thermal conversion, where you're using the heat to make electricity, there's some very strong scalings there around much more power.

17:31And so that applies to Helion II, that as we build larger and larger individual systems, and right now we don't know a physics reason why we can't build them much, much larger than we do. The cost per unit might go down in theory. But to your point, Julian, I think this is perfect, is production also matters, is that if you have to go build a boutique reactor or build boutique power plant in a faraway place one time, it's very expensive. But if you can build a gigafactory that's mass producing these and then delivering them, well, you probably, there's an optimal curve there where you have a physics cost and an engineering cost and you have a production cost.

18:15And we think that 50, maybe up to 100 megawatts, but say 50 at the ideal case is probably the right size. A good example for this is actually SpaceX has shown this in a lot of ways, right? That they built small rockets that you can put on a train and deliver and mass produce them in California and can do it really cheaply. But building big giant rockets one at a time out in Texas, that's hard and a lot harder to do. And so our focus is how can we mass produce generators, deliver them, get them on site and start making electricity with them. Later, later, we'll talk about the big ones. We can do that later.

18:51Thanks for listening so far. Hang on. We'll be right back after a quick word from our sponsors. So you also announced the 500 megawatt deal with Nucor. Is that just stacking 10 generators? And if so, what does that look like? Are they on top of each other? Are they in a big circle? What does this all look like? So it's a super good question. What we announced with Nucor is an energy development agreement. And so with Microsoft, it's a PPA. We're going to go build a power plant that generates electricity. With Nucor, that's a bigger step. And so what we're working with them is what do they really need as a customer and how are we going to deliver that?

19:29So baseline is a collection of 50 megawatt units, but we're investigating, can we go bigger right now and still keep those economies of scale? Is that 100? Is that 250? I don't know that we know the answer to that yet. And so a lot of that will depend on what they need as a customer and where they want to put it. So this is the other interesting thing that I think is particularly exciting around Nucor, is that what they want is power, what's called behind the meter, at the actual factory itself. So not on the grid, not interfacing with the grid or requiring maybe some of the regulatory concerns or the interface concerns of the grid, but right there on the facility.

20:05And so now you're actually directly for them powering the electric smelters that are so efficient at making steel that require a tremendous amount of electricity. But it will require electricity in periods where they're like they're doing the heating and then there's other periods where they don't need so much of it. And so that's what we're working on there is how do we dial up and down the power as they need it and do that most effectively. And so exactly where that balance is, I can't tell you exactly where it is yet, but we're working hard to make sure we can figure out where that is and works best for them, too.

20:35You mentioned that. Go ahead. Just one follow up to that one, which is how difficult is it to dial up and down the power infusion? Yeah. So there's like a theory and a reality version of this. There's real engineering and there's the paper engineering. On paper for a pulsed fusion system where we're in between the pulses last about one thousandth of a second, but we run it once a second. So there's literally 999 milliseconds sitting around that we can do. We're waiting. And so we can adjust that repetition rate where we and we've done this ourselves dynamically on small scale systems, showing we can go from a repetition rate of once every few minutes all the way up to, in our small systems, over 100 times a second and vary all the way through there.

21:23However, in reality, when you're building big systems, things like thermal shock and structural loading, some of those things start to come into effect. So right now, we believe we could dial from on the order of a few percent operation up to that 100 percent or maybe a little past operation pretty quickly, as in seconds of operation, but all the way to zero, that's hard. That's really hard. And we don't know how to do that one yet. You're talking a little bit about simulation and modeling here. One topic that Paki and I've been thinking a lot about is the why now for Fusion and how much of that is because of the enablement you get to be able to have tools, software tools, but maybe other things that allow this progress to be made on fusion at the moment.

22:07Can you comment a little bit about what some of those enablers are that have been helping push the whole industry forward right now? Yeah, there's a number of the hardware tools, the hardware advancements that we've talked about with pulse power electronics and high-speed fiber optics that allow us to trigger these systems. But on the software side, there's also been some really interesting things. Computationally, one of the challenges that the earliest researchers that tried these pulsed magnetic approaches had was that you have these pulsed coils that all have to fire in sequence very, very efficiently and know and be able to detect and calibrate to what the other systems are operating.

22:48And what that means is if you have to actually turn on a magnetic field to pulse and compress this plasma in thousandths of a second. It must respond in millionths of a second. And it must be able to detect what's happening in billionths of a second, nanoseconds. When I started my career, that was almost impossible. But now you can literally go buy a 10 gigabit fiber optic network card or a 1 gigahertz FPGA programmable logic off the shelf. And there are little teeny chips that live on the machine right now that detect, see how everybody is operating, and then continue operating. So that's been totally enabling.

23:26And also on the simulation side, we can predict a lot of, not all of, the simulations aren't perfect yet, but we can predict a lot of the operation of these in our design phase. And so we have full simulations of the machine that predict within 10 % how they're going to operate. And then we can use those simulations to go design a machine. We still have to go build the hardware though, and then test it, and then feed that back into the codes. And it's incredible. How much of the software that you guys use today is built in-house versus something that exists off the shelf? Yeah, it's a good question.

23:59So on the plasma physics side, that code has been developed in-house, that advanced plasma physics code. And what's also interesting is that when we started running at Helion, it ran on big clusters, still on premises, but big clusters. And now our scientists and engineers actually run them on their laptops. So it's been pretty amazing to watch that evolution. And so that's the main design plasma physics code. But way more codes we use. So on the structural side, the pulsed mechanical structural codes, also those are off the shelf. But again, those are advanced where even when I started my career, you couldn't simulate the structural response of a pulsed electromagnet.

24:40But now you can't off the shelf and it's pretty straightforward. Those still run on big computers. And same thing on the power electronics. We have tens of thousands of circuits all running. Simulating that is pretty hard too, but really important how they couple and work together. And so those are off-the-shelf codes as well. And that just continues. We're continuing to develop new codes. We announced we're working with Princeton to develop a new simulation code that's even more advanced than what we use now that should be able to be higher accuracy and predict more of what's happening after fusion begins.

25:17and the fusion particles. How are you tracking progress internally at Helion? What are the important things, obviously getting each one of the prototypes to work? Sounds like a big one, but what are the main metrics that you have on a dashboard? I mean, one thing that's nice about being iterative hardware focused, our goal is to build hardware as fast as possible, is that tracking, it's pretty easy. I think we've even had some websites, some videos on our website about, you know, time lapse of construction happening. And so it's literally when a system turns on, when it actually gets in, when the hardware is in the door, when it gets assembled, and when it turns on.