E8: The Fusion Marathon

0:10Progress in fusion over the last three to four decades has actually outstripped Moore's law. The main measure that we have for advances in fusion is something called triple product, which is a measure of plasma density, temperature, and the time that it can be sustained. And since the 1970s, we've seen orders of magnitude improvement in the triple product for fusion. And that leaves us only, in some cases, a single order of magnitude away from the types of output that we need to support commercial fusion. That's really, really exciting. And I think that it's been easy for people to discount because of how long it's taken the idea that we would ever get there.

0:49But if you look all around us, just outside your window, we are surrounded by vertical curves. If the last two years have told us anything about the advances of technology and what we've seen in the world of AI, it's that the pace is not going to let up anytime soon. And so I think fusion is one of those areas that is going to catch people by surprise at how fast the milestones are hit and how quickly we're able to get past not just this one critical milestone of QGrid and the one, but as Ben Conway, who's the CEO of Zap, likes to say, we're going to turn that into Basecamp.

1:36There's this joke that fusion energy is always 30 years away. 50 years ago, it was 30 years away. 20 years ago, it was still 30 years away. Today, though, we might be within a decade. You just heard from Clay Dumas, a general partner at Lower Carbon Capital, one of the world's leading climate funds and the only one to launch a dedicated fusion fund, which they named Q1 after the key milestone in fusion, the point at which a fusion reactor generates more energy than it consumes. They're betting the fund on the belief that that remarkable triple product curve will continue. And that curve is truly a thing of beauty.

2:18Nearly 80 years worth of work on fusion from governments, academics, researchers, businesses, engineers, and entrepreneurs across the globe captured in one chart. Beyond simply trusting the curve, though, there's evidence that we're going vertical in the rush of activity from entrepreneurs in the last few years alone. After spending most of its life in government and academic labs, startups have taken the fusion baton in the last sprint towards commercial fusion. Earlier this year, I wrote a piece in Not Boring with Rahul Rana called The Fusion Race. We compared what's happening in fusion to a relay marathon.

2:52We wrote, imagine a bizarro relay marathon in which one runner carries the baton for the first 26.0 miles, opens up a backpack full of batons, and hands them out liberally to a waiting horde of sprinters to dash all out for the final 0.2 miles. That's the best analogy we can come up with for this moment in the fusion race. Global governments are the marathon runner. From the race to develop thermonuclear weapons after World War II, to the$22 billion, 50-plus-year cooperative ITER reactor currently being built in France, to the National Ignition Facility's ignition achievement in December 2022, governments have led fusion research efforts for the better part of eight decades.

3:39Companies like Helion, Commonwealth Fusion Systems, Tay Technologies, General Fusion, and Zap Energy are the sprinters. Armed with$5 billion in funding, most of which has come in the past couple of years alone, and decades worth of research made feasible by new tools, fusion startups are locked in a mad dash to the finish line in what might be called Fusion Race 2.0. When Rahul and I wrote that piece, we put together a list of about 30 funded fusion startups, which we'll link to in the resources guide. Today, according to Clay, there are somewhere around 80 or 90 companies working on fusion. The race is on.

4:18Over the next two episodes, we're going to get you warmed up and ready for the race so that you can watch it play out with a fan's understanding of what's happening and who the key players are. Today, we'll cover the basics of fusion energy, what it is, how it works, why it matters, how progress is measured, and what the main approaches are. On the next episode, we'll talk to founders building fusion companies, taking all sorts of different approaches that physicists have imagined for a long time, but are only becoming possible now, as well as a couple of the investors backing them. We spent the first half of the season on nuclear fission and the last episode on many of the other energy sources that will play a role in meeting humanity's growing energy demands and building an age of miracles.

4:59All of them will be critical. But there's a growing chorus of people who believe that if we're going to meet the demand for three to five times the amount of energy we use today, and do it with clean energy, we're going to need fusion. And that as we climb the Kardashev Scale, the framework for measuring a civilization's level of technological advancement based on the amount of energy it can use, fusion is going to be our ladder. The Kardashev Scale, proposed by Soviet astronomer Nikolai Kardashev in 1964, lays out the three stages of a civilization. A Type 1 civilization can use and store all the energy available on its planet, including the sources we've talked about and some we haven't, like earthquakes and volcanoes.

5:41Type 2 civilizations can harness the total energy of its star, not just the solar energy that lands on the planet. You'd need something like a Dyson Sphere for this. And a Type 3 civilization can control energy at the scale of the entire galaxy, including the billions of stars in it. Type 4 and type 5, and I think someone even came up with a type 6 afterward, all get more speculative. They talk about capturing all the energy in the universe and then in alternate universes and alternate dimensions, but we're a ways away from that. We're not even type 1 yet. In his 1973 book, Cosmic Connection, astrophysicist and legend in the field Carl Sagan came up with a formula for measuring where humanity was then, and he came up with a 0.7 on the scale.

6:27Today, using the same formula, we're at about 0.73. One way to climb the Kardashev scale is to capture as much of the energy from the stars as we possibly can. The other way is to make that energy ourselves, the same way that the stars do. Fusion. Take a second and appreciate how lucky we are to be alive right now to get to see that happen. In that context, it's easy to see why people are so excited about fusion. But how does it work? Andrew Cote is a fusion engineer and one of our favorite follows on X. He breaks down complex ideas in deep tech, physics, and energy in a way that's easy to understand.

7:07So we asked him to explain fusion to us. Well, the easiest way to understand what's fusion all about, you look at the sun, right? The sun shines by fusion energy, and that's like a great place to start. And how it does that is the sun is very big. And so it has a huge amount of gravitational pressure at the center of the sun. So it's very hot. it's very dense. And at very high temperatures and densities, you can take individual atoms and fuse their nuclei together. And when they come together, they kind of fall downhill in energy. And as they fall downhill in terms of energy, they gain a lot of kinetic energy, like just like a ball rolling down a hill.

7:43And that kinetic energy comes up as something we observe as heat, right? So many different types of atoms, many different elements can fuse. Stars are mostly powered by hydrogen fusion, right? That's the main sequence or that's the main lifetime of a star. Towards its later life stages, it'll run through its hydrogen fuel and start to burn heavier and heavier elements, but those don't release as much energy. And so that's how you get like the end of a lifetime of a star. So fusion energy for us as humans, most companies are trying to burn hydrogen, isotopes of hydrogen, specifically deuterium and tritium.

8:17You know, there's different isotopes, which is when you have more neutrons or a different number of neutrons. Each type of fusion fuel has some different characteristics, but fusion is a broad sort of physical phenomenon that is the opposite of fission. Fission is when large heavy atoms split apart. Fusion is small light atoms coming together. It all comes back to Einstein's formula, the one we talked about on episode two, E equals MC squared. In nuclear fission, a neutron strikes the nucleus of a heavy atom, typically uranium-235 or plutonium-239, which absorbs the neutron, becomes unstable, and splits into two or more lighter nuclei, releasing energy along with additional neutrons, which can trigger more fission reactions in a chain reaction.

8:58Fusion is almost a mirror image of that. In fusion, two light nuclei, typically the hydrogen isotopes deuterium and tritium, merge to form a single heavier nucleus. The process releases energy because the total mass of the resulting single nucleus is less than the mass of the two original nuclei. The leftover mass becomes energy, E equals mc squared. Simple, right? Sahela Gonzalez, the global director of fusion energy at our sponsor, Clean Air Task Force, and former senior expert in fusion as a fusion physicist at the International Atomic Energy Agency, Toto said the principle behind fusion actually is pretty simple.

9:36So the principle is very simple. You have two light nuclei. Usually isotopes of hydrogen. And as you can imagine, in fusion, you fuse the two nucleus. Fine. The problem is the nucleus are positive. So when you try to join two positive charges, they repair each other. So you need to overcome these forces by putting more energy in that system to make them diffuse, to collide, let's say. So the whole trick is how can I make that thing happen to overcome these Columbus forces. So it was found by Lawson, it was a criteria which say if you want to have this fusion reaction, so to be able to put as close as possible this nuclei, to be able to fuse, to overcome these repulsion forces, you need to go beyond a threshold.

10:30And this is defined by three parameters, the density, the temperature, and the time that you keep things together. So this density, time, and temperature is called the Lawson criteria. If the multiplication of these three factors go higher than a certain number, you are having fusion reactions. So if you are able to imagine a machine that is able to put this hydrogen or hydrogen isotope inside and reach this threshold, you have fusion. As simple as that. Of course, The challenge is to have a machine where you put the nuclei and you heat up them high enough for long enough and you have a density high enough that nothing happens, which is really complicated because these temperatures are hundreds of millions of degrees.

11:22Densities are very high and time, even if it's the scale of seconds or less than seconds, is really difficult to have it. This is why you have magnetic coils, because you want to confine all your plasma. But the principle is really simple. You only need to reach that conditions. Food stop. The principles may be simple, but creating the conditions to make them happen here on Earth is anything but. Some of the smartest people in the world have been working on this problem since World War II, and we're just now getting close to pulling it off. But a ton of progress has been made. As Clay highlighted at the start of this episode, fusion is progressing faster than Moore's Law.

12:03There are two main metrics that you need to keep in mind when you're thinking about fusion. The triple product, which we'll cover in depth in a bit, and Q1. One is roughly the inputs, that's the triple product, and the other, Q1, is the goal. Q1 has been the goal of the fusion industry since the early days. Simply put, it means that a fusion reactor generates more energy than it consumes. That's critical for a power plant. It needs to add more usable energy to the grid than it takes in. How close we are to getting there is measured by the triple product, the multiplication of the three things that matter most in a fusion reaction.

12:43Density, temperature, and energy confinement time. The Lawson criterion that Sahila mentioned states that the triple product must exceed a certain value to achieve Q greater than one. The plasma needs to be dense enough, hot enough, and confined for a long enough period of time for the fusion reaction to occur. Andrew Coté explains. I put this chart on Twitter recently, this plot that had a scatter plot of these different little shapes, right? And they're all going up and to the right. We love plots that go up and to the right. This plot was showing what you call like triple product, or really it's the product of density, temperature, and confinement time.

13:22And those are the three numbers that really tell you whether a fusion reaction is energy profitable, right? Profitable meaning like, would it release more energy than you put in? And importantly, would it become self-sustaining? So the density and temperature, if they have to hit each other going quite fast to fuse together, the confinement time is more just like, if I put in a bunch of energy, how long does it take to leak out, right? So the longer it can be trapped by the plasma, the more efficient the heating and the easier it is to get to ignition. And ignition is when the reaction becomes self-sustaining, right?

13:54So it's like, I can burn a piece of wet wood, right? And it'll just smoke a lot and slowly get singed over time, but I'm constantly putting energy in. Whereas if I burn dry wood, then it becomes like, well, it turns on fire, it becomes on fire, right? It ignites. And then the reaction is self-sustaining. So this, this concept of triple product or like density, time and pressure, there's been a slow march of steady progress on this metric over decades and across different reactor designs. And the plot I put on Twitter really shows that. It shows like stars and circles and squares, which are different kinds of reactor designs overall, as they slowly get closer and closer and closer to these conditions where you start to get break-even energy release and then actually like a self-sustaining reaction, which is like Q infinity or something like that.

14:37We're going to hammer this point home because everything we're going to talk about from here on out is essentially a function of increasing the triple product by using technology, engineering, and physics to increase density, temperature, and confinement time. Here's what each is and why it matters. First, density. The number of fuel particles in a given volume of plasma. Plasma is a hot ionized gas where electrons are separate from their nuclei. More particles, higher chance of collision, higher chance of fusion. Second is temperature. Fusion requires extremely high temperatures often hotter than the core of the sun itself, north of 100 million degrees Celsius.

15:15At such high temperatures, particles move really fast. And the faster they move, the more likely they are to overcome electrostatic propulsion and collide with enough energy to fuse. And third, energy confinement time. The average time the energy remains in the plasma before it's lost. Fusion reactors need to contain and maintain the plasma at high enough temperatures, long enough for fusion to occur efficiently. More time in the plasma means more time for collisions and fusions. In other words, a higher triple product increases the likelihood of achieving Q1 and eventually of Q infinity, or self-sustaining reactions.

15:52So how do you increase the triple product and eventually achieve Q1 and Q infinity? There are different types of fusion reactor designs that place varying emphasis on each of the components in the triple product. The two main categories are magnetic confinement and inertial confinement. Magnetic confinement fusion, as the name suggests, uses magnets to confine plasma within the reactor. Magnetic confinement reactors typically optimize for high temperatures on the order of 100-150 million degrees Celsius and longer confinement times on the order of seconds. Inertial confinement fusion, on the other hand, works by rapidly compressing and heating a small fusion fuel pellet by bombarding it with high energy beams like lasers or ion beams.

16:38Inertial confinement also relies on high temperatures north of 100 million degrees Celsius, but optimizes for density instead of confinement time. The fusion event in ICF, inertial confinement fusion, is measured in nanoseconds or billions of a second. To date, inertial confinement fusion is the only approach to achieve Q1. In December 2022, the National Ignition Facility at the Lawrence Livermore National Lab, or NIF, shot 192 lasers at a deuterium-tritium fuel pellet and generated a little bit more like a battery's worth of energy from the fusion reaction than the lasers put into it. It was a huge milestone.

17:19It confirmed that Q greater than one is possible in a reactor. But as Andrew told us, publicly funded laser inertial confinement has never been part of an energy development program. NIF has always been used for weapons testing. It admits that that's what its job is, using lasers to tune the yield in the hydrogen secondary stage of nuclear weapons. Given the super short fusion event, capturing the energy in the form of neutrons and radiation, and then converting it into useful power in a controlled and sustained way remains a huge challenge. Thanks for listening so far. Hang on, we'll be right back after a quick word from our sponsors.

17:56That said, there are private companies developing inertial confinement fusion for energy production, along with a number of approaches within magnetic confinement, including tokamaks, which are the big donut-shaped ones, and stellarators, which are the big twisty reactors. There are even startups like the Sam Altman-backed Helion pursuing hybrid magneto-inertial reactors, or magnetic target fusion, that combine elements of each approach. These aim for higher densities than typical magnetic confinement systems, but lower than ICF, while also seeking longer confinement times than ICF, but shorter than typical MCF approaches.

18:31Amazingly, many of these approaches are based on old ideas that are only now becoming feasible. Now that we've given you some of the basics, we can go back to the beginning and ride the triple product curve back up to the present day. The history of fusion is one of scientific discovery, military application, and a mix of global competition and cooperation. Once humans discovered how the sun makes its energy, the bizarro marathon began. Since then, it's been about figuring out how to increase the triple product using the best available technology and engineering at any given time. Where do we start?

19:05To start our journey into the history of fusion energy, we actually need to go back to right before World War II. Until 1939, physicists' best understanding of what happens inside stars was based on Lord Kelvin's, yes, that Lord Kelvin, the temperature Kelvin, 19th century theory of gravitational contraction. Stars are balls of gas, and as gravity pulls the gas in, it heats up, causing the star to emit light and heat. Pretty good. But in 1939, building on the previous year's work by George Gamow and Edward Teller, who theorized partially correctly that stars produce energy through fusion as protons tunneled through the strong nuclear force that holds them together?

19:48Two physicists, Hans Beth and Carl Friedrich von Weissacker, independently arrived at the proper explanation. They proposed that stars produce energy through a series of nuclear fusion reactions known as the CNO cycle for carbon-nitrogen-oxygen by which stars convert hydrogen to helium. Beth and Weissacker's theory was timely, coming as it did on the eve of World War II. On episode two, Rod Adams told us that we actually tried to use fission for energy before using it for the bomb. For fusion, the bomb came first, as Andrew explains. We had fission reactors before we had fission bombs. The first one was Chicago Pile I.

20:28It was like underneath a dorm, what's now a dorm at the University of Chicago run by Fermi and Szilard and stuff. So we had fission power very early on. After we had developed the fission bomb, people started working on what's called the super, which is a hydrogen bomb. So they started to understand those reactions more. What the bomb makers figured out at Los Salamos was a clever and destructive trick. The Teller-Ulam configuration, named after the physicists who devised it, uses the energy from the fission bomb to compress and heat the fusion fuel to such an extent that the atomic nuclei begin to fuse together, releasing a huge amount of energy.

21:01The resulting hydrogen bomb, Ivy Mike, yielded 10.4 megatons of energy, over 450 times the power of the bomb dropped at Nagasaki, and obliterated the island on which the test was performed, leaving an underwater crater 6 ,240 feet wide and 164 feet deep where the island had once been. Hydrogen bombs are not a practical way to create fusion energy. You don't want to turn your power plants into smoking craters, but Ivy Mike was the first man-made fusion reactor to achieve Q1. Fortunately, the bomb has never been deployed in combat, but with fusion proven possible, governments around the world sprung into action to harness it.

21:41In the U.S., President Truman signed the Atomic Energy Act and established the United States Atomic Energy Commission as a successor to the Manhattan Engineer District, better known as the Manhattan Project. Given the relative maturities of the two technologies, the AEC focused most of its effort on nuclear fission, becoming both the champion and regulator of the nascent industry, as we discussed on episode two, with all of the problems and benefits that that came with. But the AEC also laid the groundwork for research into fusion energy. Then, in 1951, Argentinian President Juan Perón dropped a bomb of his own, announcing that a scientist named Ronald Richter, who had moved to the country from Germany after World War II, had achieved controlled nuclear fusion.

22:26The claim turned out to be false, but it did accelerate the race worldwide. If our history of nuclear fission was about understanding what slowed nuclear down in order to take those lessons into the nuclear renaissance, what's most salient in the history of fusion? To me, it's really the Bizarro Relay Marathon that we talked about. The race went from various countries competing with a bunch of designs to a big cooperative international effort focused on just mainly one design. That was cool in a kumbaya sense, but bad in that it slowed down the development of fusion in the way that any large international project and lack of competition slows things down.

23:04But governments and academia did make real progress, and today's entrepreneurs get to piggyback on their work. Yeah, the early days of fusion remind me a lot of the early days of fission with just so much experimentation going on with different designs. Starting in 1951 with fusion, after the Argentinian announcement, three separate projects pursuing three different approaches to plasma confinement sprung up and would form the basis for Project Sherwood, a classified U.S. program in controlled nuclear fusion. First was the Stellarator, designed by Lyman Spitzer at the Princeton Plasma Physics Laboratory.

23:38Stellarators are devices designed to confine hot plasma within magnetic fields in a twisted, torus-shaped configuration to sustain nuclear fusion reactions. Next is the toroidal pinch. Designed by James Tuck at Los Alamos, toroidal pinch devices work by confining a high-temperature plasma into a donut-shaped chamber using magnetic fields. The fields are generated in such a way that they pinch the plasma, increasing its pressure and temperature to the point where nuclear fusion can occur. Tuck named the device the Perhapcitron because perhaps it might be able to achieve a fusion reaction. The Perhapcitron, I think, is my favorite product name of all time, not just infusion.

24:17It sounds like it's a nuclear fusion device straight out of Dr. Seuss. Totally, totally. And finally, there was the Magnetic Mirror. I mean, the names are great here. This one was designed by Richard Post at the Livermore National Laboratory. Magnetic mirror devices contain plasma within a chamber using magnetic fields that are stronger at the ends, reflecting the charged particles back to the center, creating a magnetic mirror. Instead of me trying to explain the physics of these crazy devices, let's turn it back over to Andrew. In the early 50s, you kind of have this flourishing of different approaches or conceptions for a fusion reactor.

24:53And they've kind of evolved in an interesting way as we've kind of learned more about the field. You know, plasma is a interesting substance, right? It's a gas of electrons and ions that are separated from each other. So you can get current that flows in it, but you can also get, it also responds to magnetic and electric fields and currents generate magnetic fields and so forth. There's lots of really interesting dynamics. So one of the first devices called Z-Pinch, and that's, you have like a plasma in a tube, basically you drive a current through it. And that current generates a magnetic field that tries to constrict the plasma tube.

25:24And as it gets constricted, it can hypothetically heat the plasma ions up enough so that they hit each other with enough energy to fuse and then release more energy, right? You know, there's issues with that. It didn't work for a lot of reasons like plasma instabilities and magnetic confinement. And the early history of fusion development was really, how can we beat these kinds of scaling laws for how well we can capture the heat energy versus how strong the magnetic fields are? So there's other devices like magnetic mirrors. Those also came along. And that's where you think of like having a magnetic bottle like this, where like ions, really the fusion fuel, it's like bouncing between them along these magnetic field paths.

26:06Right. So another early design. that had some limitations where it would constantly leak fuel or leak heat energy through a certain kind of like window of particle trajectories called the loss cone. That's like a theoretical result. So it wouldn't work too well, even in the best case. And then there's lots of other machines. Z machine was another kind of design. So there's also the stellarator, which is a kind of magnetic confinement fusion device where people thought, okay, look, this magnetic mirror doesn't really work because every time the particles have to bounce back and forth, you lose some, right?