David explains that nuclear fusion is the process that powers the universe, particularly stars. It involves taking the most common, lightweight elements, like hydrogen, and fusing their atomic nuclei together to create heavier elements. During this process, the resulting heavier nucleus is slightly lighter than the sum of its original parts. This lost mass, or "mass defect," is converted into a tremendous amount of energy according to Einstein's equation, E=mc².

Nuclear fission, on the other hand, is the exact opposite. It involves taking the heaviest, most unstable elements in the universe, such as uranium and plutonium. These elements are so large they are barely held together. When a neutron is added to one of these large nuclei, it breaks apart into many smaller pieces. The total mass of these pieces is also slightly less than the original nucleus, again releasing a massive amount of energy as per E=mc².

There's a well-known curve in atomic physics that illustrates this principle. Iron is the critical element that acts as a dividing line. Elements lighter than iron release energy through fusion, while elements heavier than iron release energy through fission. David notes how this process unfolds in stars, which start by fusing hydrogen into helium, and can later fuse helium into carbon, and so on, creating progressively heavier elements. This stellar process inspires the search for more advanced fusion fuels.

David further clarifies the E=mc² equation, describing it as a fundamental relationship that shows mass and energy are interchangeable. This principle applies not only to nuclear reactions but also to chemical bonds, like when hydrogen and oxygen combine to form water. However, the change in mass and subsequent energy release in chemical reactions is so minuscule that it's extremely difficult to measure.

Fission and fusion energy use different primordial fuels. The elements for fission, like uranium and plutonium, were created in supernovas and the Big Bang. We mine these materials from the ground. In fact, most plutonium is made from uranium.

Fusion, on the other hand, uses hydrogen isotopes, which are also primordial and the most common atoms in the universe, making up stars and suns. While fission fuel is found in the ground, fusion fuel is essentially everywhere. The primary fuel used is a type of hydrogen called deuterium, a heavier isotope with one proton and one neutron.

Deuterium is found in all water on Earth, including the water we drink and the water in our bodies. It exists in what is called "heavy water" (D2O), mixed with regular water (H2O). It is estimated that Earth's seawater contains enough deuterium to power all of humanity for 100 million to a billion years at current electricity consumption rates. This makes it a nearly endless resource. With such an abundant power source, humanity could access far more energy, enabling incredible technological advancements and even expansion into the cosmos to find more resources on other planets.

In nuclear fusion, lightweight isotopes like hydrogen and deuterium are combined to release energy. This process involves overcoming the natural electromagnetic repulsion between positively charged atomic nuclei. To achieve this, the fuel is heated to extremely high temperatures, around 100 million degrees. This temperature is a measure of kinetic energy; it makes the particles move so fast that they can get close enough for a different fundamental force, the strong nuclear force, to take over.

Once within a very short distance, the strong force attracts the particles, causing them to fuse together. The resulting heavier atomic nucleus has slightly less mass than the original particles combined. This missing mass is converted directly into a large amount of energy, following the principle of E=mc².

This process is fundamentally different from nuclear fission, which is comparatively easy. Fission happens at room temperature with unstable elements like uranium, where adding a single neutron can cause an atom to split and start a chain reaction. Fusion, in contrast, is difficult to initiate and sustain. The sun achieves fusion through gravitational confinement, where its immense mass crushes fuel together. On Earth, different methods like electromagnetic confinement are needed.

Because of these differences, the term "reactor" is not accurate for fusion. David Kirtley explains that the Nuclear Regulatory Commission's definition of a reactor involves two key elements: nuclear fission and a self-sustaining chain reaction. Fusion is neither.

A reactor is self-sustaining. You take your hands off of it and it keeps going. In fusion that doesn't happen. We actually use the word generator because we don't talk about, for instance, a natural gas reactor. If you stop putting in fuel it turns off. And the same thing happens in fusion.

A fusion device is more accurately described as a generator. It requires a constant input of fuel to run, and if that fuel is cut off, the process stops immediately. The ultimate goal of this process is not just to create heat, but to generate electricity.

Nuclear fusion generates energy by bringing lightweight isotopes together, releasing energy directly in the form of charged particles, which is essentially electricity. This process has an inherent safety benefit: the reaction is very difficult to sustain and naturally turns itself off. Nuclear fission, on the other hand, works through a self-sustaining chain reaction. Unstable atoms like uranium split, releasing heat and neutrons. These neutrons then hit other atoms, causing them to split, continuing the cycle.

In a fission reactor, the engineering challenge is to maintain a perfect balance. There must be enough neutrons to sustain the reaction but not so many that it accelerates uncontrollably. Water is typically used as a coolant to absorb the heat and extra neutrons, which then creates steam to power a turbine. While fusion is fundamentally safer because there is no chain reaction, modern fission reactors are also engineered to be very safe. They have passive safety features; for instance, if they get too hot, they are designed to expand and cool down on their own.

David believes the engineering problems for fission have largely been solved. The real challenges and risks associated with nuclear power today are not with the power plants themselves, but with the human elements surrounding them.

The problem comes from humans and the problem comes from other things around nuclear power. You have to enrich that uranium to put it in a plant and the plant's safe, but you had to enrich that uranium and that is some of the problem. Or a plant is designed to run for a certain number of decades safely, but do we run it longer than that? And so those are where I think the real challenges happen is more with the humans around these systems than the engineering of the power plants themselves.

The major nuclear accidents at Chernobyl and Fukushima can be attributed to human failure rather than engineering failures. Three Mile Island is in a different category. In fact, there are plans to restart it because of the need for clean baseload power.

In the case of Fukushima, multiple nuclear fission reactors on the same site continued to run successfully through the tsunami. They were only shut down later for political reasons. The problems occurred at the oldest reactor, which had been on site for a very long time. This suggests the issue was not with the engineering of the plants themselves, but with the humans operating the systems. The plants are safe when they are operated as designed.

Fusion power plants cannot be used to make nuclear weapons. The processes in fusion are fundamentally different from those that occur in nuclear bombs. This is a key distinction from traditional nuclear fission reactors, which use uranium and plutonium—fuels that can be used to create weapons.

Even so-called fusion bombs, or hydrogen bombs (H-bombs), are still fundamentally fission bombs. They work by using a primary fission reaction to create radiation, which then induces a fusion reaction with a small amount of fusion fuel. This fusion reaction, in turn, boosts the uranium reaction again.

Most of the energy, in fact 90% of the energy in an H bomb is all still from the uranium reactions themselves.

The topic of creating and distributing nuclear weapons is known as proliferation. Experts in this field, who work to prevent the spread of nuclear weapons, are strong advocates for fusion energy. They worry that a global increase in fission power would require more uranium enrichment facilities, which could potentially be used to create weapons-grade material. They see fusion as the clean energy solution that avoids this risk.

What they told us is please, please go develop fusion power plants absolutely, as fast as possible. The world needs this.

From a geopolitical perspective, fusion power holds another significant advantage. The fuel for fusion, deuterium, is found in seawater all over the Earth. This means every country has access to it.

You can't have a monopoly on the fuel. And no one can control the fuel, and no one can turn off the fuel, no one can cut a pipeline like that. Just cannot happen with fusion.

By deploying fusion plants globally, it's possible to decouple energy from the control of a few countries, fundamentally altering the geopolitical landscape tied to resources like oil, gas, and uranium.

Fusion power is fundamentally safe because the physics of the reaction prevents runaways. To analyze potential risks, an extreme scenario was considered: what if a meteor strikes and vaporizes a fusion power plant? The analysis concluded that even in this catastrophic event, there would be no need to evacuate the nearby populace.

The key to this safety lies in the fuel system. A fusion generator, like Helion's, is continuously fed fuel and only contains about one second's worth at any given moment. If the fuel supply is cut off, the fusion reaction simply stops. This is a stark contrast to other power plants. A coal plant may have a large pile of coal that could catch fire, and a nuclear fission plant contains several years of fuel in its core, representing a massive amount of potential energy.

In fusion, you have literally one second of fuel at any time in the system. And having a tank of deuterium... can't do fusion by itself. It needs that complex system.

The waste from fusion primarily involves neutrons. The reaction creates ionizing radiation, X-rays, and neutrons. While charged particles and X-rays are contained, the neutrons require shielding, much like in a hospital's particle accelerator. This has led to a landmark regulatory decision in the United States. The Nuclear Regulatory Commission (NRC) will regulate fusion under a statute known as Part 30, which governs medical facilities and particle accelerators, rather than Part 50, which is for traditional nuclear fission reactors that handle uranium and plutonium.

David shares a story about applying for a permit in Washington State, where the Department of Health's standard form for a particle accelerator asked, "Where do the patients go?" This highlighted the novelty of regulating a fusion system for power generation. Helion received its first license for a fusion system as a particle accelerator in 2020.

Fundamentally, all nuclear fusion approaches aim to achieve the same physical process. They take lightweight isotopes and heat them to over 100 million degrees so they move at high velocity. Then, they bring a sufficient density of these isotopes together in a volume and hold them long enough for the particles to collide, fuse, and release energy.

There are several ways to accomplish this, which have been explored since the 1950s. Public and federal funding in the United States has focused on two main programs: inertial fusion and magnetic fusion. In inertial fusion, particles are pushed together through physical means. The most common method is laser inertial fusion, demonstrated by the National Ignition Facility. This approach uses very high-power lasers, pulsed together, to trigger fusion for an extremely short period, just billionths of a second.

The other approach is magnetic fusion, which includes devices like tokamaks and stellarators. Stellarators have an elegant mathematical solution that can be solved analytically, but they are very difficult to build. It has taken decades to develop the technology, with the Wendelstein 7-X being the premier stellarator in the world today.

Lex finds the engineering behind all fusion methods impressive, describing them as "badass." He notes the challenge of containing extremely high temperatures and densities, and finds it fascinating that humans can build such machines to generate energy that could power humanity.

In a magnetic fusion system, the goal is not to push particles together quickly, but to hold onto them for as long as possible using magnetic fields. An analogy is the Earth's magnetosphere, which is a magnetic field generated by the planet's core that protects us from cosmic and solar particles. The same principle is applied in fusion reactors.

These powerful magnetic fields are generated by running massive electrical currents through loops of wire. The amount of electricity is immense. Compared to a home's 200 or 400-amp breaker box, fusion systems run on hundreds of mega-amps, or 100 million amps of current. This is based on Maxwell's equations from the 1800s, which state that an electrical current in a wire generates a magnetic field.

When a charged particle, like an electron or an ion, is placed inside this field, a special property emerges. The particle becomes "magnetized," meaning it gets trapped on a magnetic field line and oscillates around it. A real-world example of this phenomenon is the aurora, or Northern Lights, which are charged particles trapped in the Earth's magnetic field. Fusion reactors replicate this process on a smaller scale to contain the plasma.

That electron or that ion, that charged particle, is what's called magnetized. And what magnetized means is that it's trapped on that field line. In fact, even really more interesting is that it oscillates around that field line.

The fundamental physics underpinning fusion is actually quite old and well-understood. The electromagnetic principles date back to the 1800s and the atomic physics to the early 1900s. The primary difficulty is not in discovering new physics but in solving the immense engineering challenge of putting all these known principles together into a functional power plant. While a graduate student can do the math on paper, building the complex magnetic coils in practice is extremely challenging.

Tokamaks and stellarators are both magnetic systems designed to hold fusion fuel long enough for fusion to occur. In these systems, charged particles are trapped on a magnetic field, oscillating in what is called a gyro orbit. These orbits are surprisingly large, measured in inches. The goal is to trap and heat these particles so they collide and fuse faster than they are lost. However, containing these particles, which move at millions of miles per hour, is extremely difficult.

Helion uses a different method called magneto-inertial fusion, which combines elements of both magnetic and inertial approaches. This technique has roots in successful physics experiments from the 1950s. Early pioneers knew they had to heat gases to 100 million degrees and confine them. One of the first methods was the 'theta pinch,' which used a linear series of electric coils, a solenoid, to create a magnetic field. The problem was that the fusion particles would simply escape out the open ends.

This problem led to a split in fusion research. One branch decided to solve the open-end problem by bending the linear solenoid into a donut shape. This ensures that as particles try to escape, they just go around in a circle.

One branch of fusion said, okay, well, to solve that, why don't we take this solenoid and bend it around? Let's just make it a big donut. So as they're escaping, they go around and around in a circle. Great, that's a great approach. And so one branch of fusion went down that direction, and that evolved into the stellarator and the tokamak.

Another branch stuck with the linear design but tried to close the ends. First, they made the magnetic field much stronger at the ends, creating a 'mirror' to bounce particles back and forth. This didn't work well because the hottest, most energetic particles were the ones that managed to escape. The next idea was to not just hold the plasma but to rapidly squeeze it. By quickly increasing the magnetic field, they could crush the plasma, increasing its density and triggering fusion.

This theta pinch approach was highly successful in 1958, reaching temperatures of 50 million degrees and outperforming other methods. However, the researchers hit a technological limit. They were switching millions of amps of electrical current in microseconds without the aid of transistors, CPUs, or modern electrical switches. Unable to build the systems any further, these pioneers shifted their focus to other areas, like laser-based fusion.

An accidental discovery in plasma physics revealed a new way to confine plasma. Initially, when researchers squeezed plasmas in devices called theta pinches, the plasma would just squirt out the ends like toothpaste from a tube. However, decades later, they found that operating these devices in a specific way caused something new to happen: the plasma pushed back and stayed confined, even though the ends of the container were open.

This phenomenon is now known as a field reverse configuration (FRC). The key to creating an FRC is rapidly changing the direction of the magnetic field that contains the plasma. When the field is reversed, the charged particles of the plasma, which are extremely hot, cannot move fast enough to follow the change. Because the particles can't move but the field has been inverted, the plasma internally reconnects and self-organizes into a closed field, effectively trapping itself.

So now what you're left with is an outside magnetic field, an electrical coil. And inside the plasma, where now it was before it was moving along, it's now moving internally, rapidly reversing the magnetic field. Plasma self organizes into a closed field.

This process has to happen incredibly quickly. To trap a million-degree gas, the electrical current and magnetic field must be reversed in about a millionth of a second. This was not possible when theta pinches were invented in the 1950s. Today, it can be achieved thanks to modern semiconductor switching technology, similar to the transistors in a computer CPU that can switch in a billionth of a second.

Understanding self-organizing plasma can be simplified with an analogy to a transformer. According to Lenz's law, when you have an electrical current flowing in an outer coil, an equal and opposite current is induced in a nearby metal conductor. This is a fundamental principle used in everyday transformers.

Now, imagine that the inner conductor is not metal, but a high-temperature gas, or plasma. In this scenario, an electrical current flows through the plasma itself. While all plasma conditions in fusion research involve some current, the Field-Reversed Configuration (FRC) uses massive amounts of it. This is the crucial difference. In an FRC, the plasma itself contains so much electrical current that it functions like an electromagnet, generating its own magnetic field and trapping itself within it.

In a tokamak and your stellarator, you make the magnets and you trap your plasma in it. In an FRC, you make the plasma, which makes the magnets and it traps itself.

This phenomenon is not just a laboratory creation; it occurs in nature. Solar flares, for instance, are essentially arcs of plasma with electrical current flowing through them. When a flare rips off the sun, it can travel through the solar system as a self-contained structure called a plasmoid. This plasmoid holds itself together with its own magnetic field, at least for a while.

The primary challenge is control and stability. A solar flare is a transient event that eventually dissipates, which is not suitable for a sustained fusion reaction. The difficult part of the job is learning how to create these self-organized plasmas in a controlled way and keep them stable.

In this fusion system, an electrical current must be reversed in a millionth of a second. The setup involves a series of magnets creating an external magnetic field around a donut-shaped plasma, known as a Field-Reversed Configuration (FRC). This FRC has its own electrical current and magnetic field, creating an internal pressure that pushes back against the external compression.

This internal pressure has two components: particle pressure, which is like hot gas expanding in a balloon, and magnetic pressure from the plasma's own electromagnetism. David compares this to an electric motor. In a motor, electrical current flows into windings to generate an electromagnetic force, which then induces a current in the spinning central part, the armature. Similarly, an electromagnetic force is induced in the plasma, which can be used to compress or expand it. However, this configuration is inherently unstable.

This leads to the concept of plasma beta, the ratio of the plasma's particle pressure to the confining magnetic field's pressure. It essentially measures how well the plasma is trapped. For an FRC, the beta is very high, close to one, meaning the internal particle pressure almost perfectly balances the external magnetic pressure. This balance is incredibly useful, as it creates a direct equation between the magnetic field and the plasma's density and temperature—the key ingredients for fusion.

B squared over 2 mu naught is nkt. So for a known magnetic field, I know what the density and the temperature of the plasma is. And just to circle back to it, we talked about fusion, we talked about it had to be hot enough and it had to be dense enough, and that's N and that's T. So now I have a very clear equation between magnetic field and density and temperature of the fusion fuel. And that's really critical.

The major downside is that high-beta plasmas are typically unstable. While a tokamak is stable because its plasma is firmly held by magnetic coils, the FRC is unconfined. The entire plasma donut can simply flip over, a phenomenon called a tilt instability. David likens it to a motor's armature spinning without a physical axis to hold it in place. Since you cannot put mechanical parts into a 100-million-degree system, there is nothing to physically hold the plasma, making it unstable. Discovering how to stabilize this was a fundamental challenge.

A key factor in understanding Field-Reversed Configurations (FRCs) is a parameter called S over E. The concept can be understood through the analogy of a spinning top. A top is inherently unstable and will fall over, but if you spin it fast enough, the angular momentum and inertia will keep it upright. Similarly, in an FRC, if the particles are driven fast enough with sufficient kinetic energy and inertia, the system will remain stable.

Another analogy helps explain the geometry aspect. A spinning coin will eventually fall over. However, a thicker, longer object like a roll of duct tape spinning on the same axis will stay stable for much longer due to both its inertia and its geometry. In the S over E parameter, 'S star' represents the kinetic energy (the spinning top), while 'E' represents the elongation (the roll of duct tape). Fortunately, the process of creating FRCs in long solenoids naturally makes them very long. By designing these systems to be long and driving the ions at high velocities, they can be stabilized against instabilities.

This understanding has led to significant progress. While the basic theory suggests these systems should only last for a few microseconds, researchers have successfully made them last for thousands of microseconds. This is thousands of times longer than the basic criteria would predict, demonstrating a clear understanding of how to design for stability.

In this fusion system, a parameter known as S star over E is a requirement of the design. This parameter is also a measure of temperature. The challenge is to satisfy this requirement throughout the entire process. Higher temperatures create more stability, similar to how a top is more stable when it spins faster. The main difficulty is heating the plasma hot enough, quickly enough. If it's not heated fast enough, it becomes unstable and tilts over. This is why a significant focus is on electrical engineering, designing electronics fast enough to heat the plasma and maintain stability.

Understanding temperature at 100 million degrees requires a shift in perspective. As a substance is heated, it moves through states of matter: from a solid lattice structure, to a liquid, to a gas where particles move freely. Heating a gas further can lead to a rarefied state, like in space, where particles are so far apart they rarely collide. Hotter still, at around 10,000 degrees, electrons are stripped from their atoms, creating a plasma of charged particles. When this plasma is heated to 100 million degrees, the conventional idea of temperature as random particle motion no longer applies.

What it's really is, is a measurement of its velocity. It's really a measurement of how fast is that particle moving. And that's how I really think about temperature when you get to that hundred million degrees.

At these extremes, particles are moving at immense speeds, on the order of a million miles per hour. The goal of fusion is to have these high-velocity particles collide and fuse. Because they would destroy any material they touch, they are contained within a magnetic field, where they can bounce around without making physical contact with the reactor walls.

Achieving fusion requires controlling systems at incredibly high speeds. Because high temperature equates to high particle velocity, the reactions occur on the scale of microseconds. A particle might move at meters per microsecond, meaning the entire fusion process is a literal flash that begins and ends faster than the human eye can respond. This necessitates a control system that can react to the universe in millionths of a second.

Early fusion pioneers attempted this without modern computers, a monumental challenge. Today, the process relies on gigahertz-scale computing. While older megahertz processors could only respond once per microsecond, modern gigahertz processors can perform a thousand operations in that same timeframe, enabling complex, real-time control.

But now gigahertz means I can do a thousand operations in that one microsecond. So I can do more useful things.

The system is too fast for any human to manage directly. Instead, operators use programmable logic, pre-programming sequences based on numerical simulations. The languages used range from legacy codes like Fortran to modern ones like Python and Java, and even low-level assembly language for the fastest hardware control. To trigger the electrical currents, signals are sent via fiber optics, as traditional wires are too slow. Fiber optics transmit photons at the speed of light, enabling responses in nanoseconds.

A key challenge is coordinating the tens of thousands of parallel electrical switches needed to generate the immense electrical currents. Each switch has fiber optic signals going in to trigger it and coming out to report its status. Real-time monitoring systems use electromagnetic coils to confirm that all switches are conducting properly, ensuring the entire system operates in perfect harmony.

The operation of a fusion system happens on a timescale too fast for human operators. Consequently, the process, known as a "shot," must be pre-programmed. Computers handle the triggering, operation, and real-time performance measurement. To design these systems, a combination of numerical simulation tools are used, many of which are based on decades of research from government and national lab programs.

One such tool is a magneto-hydrodynamic (MHD) code. For engineers familiar with computational fluid dynamics (CFD), this is very similar. It takes the same set of fluid dynamics equations and adds electromagnetic equations on top. However, due to computational limitations, simulations must be run at different levels of detail because computers are still not fast enough to simulate everything at once.

One level uses "fluid codes," which treat particles like ions and electrons as fluids or gases. These codes can simulate the entire system, from the electrical circuitry (starting with a spice model) to the plasma physics. This helps in design work and in predicting how the machine will run. A deeper level of simulation uses "particle-in-cell" codes, which treat individual ions as particles. These require several orders of magnitude more computational resources. Advancements in GPUs, driven by the AI industry, have been directly applicable and have significantly sped up this work.

These more advanced codes have only been practical for the last few years. They allow for the simulation of complex behaviors, like stability criteria, that were previously understood through empirical tests. Now, it's possible to simulate these phenomena, understand why they work, and make predictions.

Different simulations are used to design various parts of the machine. At the MHD level, a circuit model helps the design team make decisions about the circuitry, including which capacitors, switches, and cables to use, down to the specific size of the cable. These tools are used daily in power plant design.

A deeper level of analysis uses more advanced, but slower, computational tools to assess stability. These focus on the magnetic field topology, informing the design of the magnet's shape and the precise timing for triggering a sequence of magnets. Because these simulations can take a day or two to run, the current process is not real-time. Operators run many simulations ahead of time, collect data from the machines, and then a simulation team compares that data with the simulations to inform the next set of tests.

Artificial intelligence and reinforcement learning are being explored to accelerate this process. The goal is to move from a multi-day cycle of simulation, testing, and comparison to a real-time system. An operator could see what an AI model would have predicted and use that insight to understand and control the generators as they are running.

To achieve fusion, you need particles at a high temperature (T), at a high enough density (N), and held together for a long enough time (tau). Magnetic pressure, represented as B squared, is equal to the plasma's pressure, represented as NKT. This means that to increase density and temperature, you must maximize the magnetic field. Most magnetic fusion approaches aim to do this.

Fusion power output scales very strongly with the magnetic field, to the power of approximately 3.77. This emphasizes the critical need for the strongest possible magnetic field. Pulsed magnetic field systems have a significant advantage here. Researchers have demonstrated over 100 Tesla magnetic fields in pulsed systems, whereas steady systems have only reached the high 20s. This massive difference in magnetic field strength can lead to huge increases in fusion power output.

However, pulsed systems imply a shorter confinement time (tau). Different fusion approaches play with this trade-off. Inertial fusion uses a very short tau (nanoseconds) but extremely high pressure. Tokamaks and stellarators aim for a very long tau but have much lower density. The approach discussed, magneto-inertial fusion, occupies a middle ground with extremely high magnetic fields and a confinement time between 100 microseconds and a few milliseconds.

The process involves filling a chamber with a magnetic field, injecting and ionizing fusion fuel into a plasma, and then rapidly increasing the magnetic field to compress and heat the plasma. As fusion occurs, new charged particles are created, increasing the plasma's internal pressure. This high internal pressure pushes back against the external magnetic field, inducing an electrical current. This current can be captured to directly recharge the capacitors that power the system, representing a direct energy conversion method.

There are two primary ways to generate energy from fusion, which can be understood through analogies. The first is the "campfire" method, used in steady systems like a tokamak or stellarator. In this approach, fuel is heated to ignition, creating a steady, hot reaction like a bonfire. The standard fuel, a mix of deuterium and tritium, fuses to create helium and a neutron. The helium particle stays inside, stoking the flames, while the uncharged neutron flies out. This escaping neutron carries energy, which is used to heat water and create steam. That steam then drives a turbine to generate electricity, a process that is about 30-35% efficient, similar to coal or nuclear fission plants.

A different approach uses a pulsed magnetic system, which is more like a "piston engine." This method creates a high-beta state where an electromagnetic force compresses the fusion fuel. As fusion occurs, the newly created charged particles don't just heat the system; they actively push back against the magnetic field. This expansion applies pressure, inducing an electrical current that can be extracted directly.

In a piston engine, you use the motion of the piston, the pressure on it and the motion of it to do something useful... What we do is use the expansion of the magnetic field to extract that electricity. And we believe you can do it much, much higher efficiencies... not 30 to 35% efficiency like a steam turbine can do, but 80% efficiency, 85% efficiency.

This direct conversion is far more efficient. It also has another major advantage: the ability to recover the initial energy put into the system. The magnetic energy used to compress the fuel can be recaptured with over 95% efficiency. This means you get back nearly all the energy you put in, plus a highly efficient conversion of the new fusion energy. This method, however, makes the traditional deuterium-tritium fuel less than ideal, as its energy-carrying neutrons do not push back on the magnetic field to generate current.

Beyond deuterium-tritium, other more interesting fusion fuels exist. One promising candidate is a combination of deuterium and helium-3. Normal helium, found in balloons, has a nucleus with two protons and two neutrons. Helium-3, also called light helium, is a stable but uncommon isotope. It is so lightweight that it escapes Earth's atmosphere and drifts into space, making it scarce on our planet. Potential sources for helium-3 include manufacturing it, or mining it from space, with the moon and Jupiter being possible locations. Jupiter, in particular, has massive amounts of it.

When deuterium and helium-3 are fused, they produce an alpha particle (a helium nucleus) and a proton. Unlike the neutron produced in other reactions, the proton is a charged particle. This is a significant advantage.

This, that proton is now trapped in the magnetic field, pushes back and you can extract that electricity.

This ability to directly extract electricity makes it the ideal fuel for a high-beta system, like a pulsed magnetic fusion system. However, there are prices to be paid for using helium-3. These costs are multifaceted, involving physics, engineering, technical, and business challenges.

Helium-3 fusion has different requirements than the more common deuterium-tritium fusion. While deuterium-tritium works well at 100 million degrees, helium-3's optimal temperature is much higher, sometimes 200 or 300 million degrees. This presents a challenge. For a given magnetic field, as temperature increases, density decreases. This means you might have fewer particles available for fusion, requiring the system to be bigger to achieve the same reaction rates.

However, this is offset by a major advantage. A helium-3 system can recover energy at a much higher efficiency, around 80%, compared to roughly 30% for deuterium-tritium. When considering the final goal of electricity output, a helium-3 system can be a similar size to a deuterium-tritium one despite the higher operating temperature.

Size is a critical constraint because it directly impacts cost. The ultimate goal is to produce clean, low-cost electricity that people will actually buy. David Kirtley explains that the cost of a power plant is fundamentally tied to its physical size and the materials used to build it.

The goal is to manufacture a product for as low of cost as you can, so you can sell it for as low price as you can. It asymptotes to the material cost because you can never get cheaper than that. So it's literally, in some first principle sense, how much concrete goes into building the power plant, how much steel, how much copper and aluminum. At the end of the day, the cheapest function is the least amount of materials.

This economic reality shapes the entire research and development strategy. Instead of building massive, complex, multi-decade projects, the focus should be on smaller, mass-producible systems that can be built and learned from quickly. David found that this approach of rapid iteration is not only better for business but also accelerates scientific discovery.

What I've found in my career at this is that they're actually the same thing and that the faster you can build a thing, the faster you can learn if that thing works, the faster you can iterate on that and build the next thing. What I found is actually small, iterative, just building as fast as possible gets you there faster. Because you can learn, you can build, you can iterate, you can solve the problems and then you can learn the fundamental physics way sooner than if you would have just started on one mega project and then waited decades to get to the answer.

The constraints of striving for a simple, low-cost, manufacturable product ultimately push the science and innovation forward more effectively.

Helion has built seven fusion systems, iterating at an incredible pace. The first six were a series of prototypes focused on scaling the process of creating and compressing plasma to thermonuclear conditions. These early systems were named after beers, with the most successful being the IPA (Inductive Plasmoid Accelerator). Later systems were named after Starbucks cup sizes: Tall, Grande, Venti, and finally Trenta.

The Trenta system, which came online in 2020, successfully reached 100 million degrees and was the first to achieve deuterium and helium-3 fusion. This rapid progress is driven by a core philosophy: get electricity to the world as quickly as possible. This mission informs every decision, from materials to manufacturing.

To move fast, the team prioritizes commonly available materials like simple aluminum and copper alloys, engineering around rare materials that cause supply chain bottlenecks. They also design for mass production. Instead of building one giant, complex magnet, they create a composite of 100 smaller magnets. Each small magnet can be made on a simple machine, is light enough for a person to handle, and can be mass-produced, making the process faster than building a single large component.

David Kirtley explains that a key to their speed is a scrappy, problem-solving mindset, which includes sourcing parts from unconventional places like eBay. For example, a new turbo molecular vacuum pump might have a nine-month lead time from the manufacturer. Helion's approach is different.

You can go and get three of those turbo pumps that are sitting on eBay right now. Bring those in house, test them. Maybe only one of them meets the specifications you need. But guess what? You just got a pump in two weeks instead of nine months. And you got it. It's in the door and it's operational and it's running and you're moving.

This principle of prioritizing speed over perfection applies across the board. The team would rather use a 10-year-old technology that is 95% accurate and can be built in a month than a state-of-the-art diagnostic tool that is 97% accurate but would take three years and millions of dollars to build. The focus is on solving the immediate problem to keep making progress.

At Helion, the culture is centered on pushing the rate of iteration and building things quickly. This sometimes creates differences of opinion with academic colleagues. While a 3% error rate is better than 5%, if 5% is good enough for now, they will proceed. The key is finding the middle ground without ever compromising on quality and safety. The goal is to build a team that wants to construct things. The company is structured unusually for a fusion company. Today, 50% of the staff are technicians, not scientists, supporting a large in-house manufacturing capability to build as fast as possible.

A core strategy for speed is vertical integration, a concept also championed by Elon Musk. Helion brings critical manufacturing processes in-house to control timelines, especially for components that are not commodity products. They have their own manufacturing lines, including a conveyor belt for producing power supplies, to maximize velocity instead of relying on external suppliers.

David points to a photo of their Trenta machine as a perfect example of this philosophy. A green fiberglass structure (G10) on the end of the machine needed to be larger than standard-sized pieces available off the shelf.

The standard piece of G10 was not big enough to fit the end of the machine. And so we could have had one custom manufacturer manufacture a brand new piece of a custom size, build a new mold and a new machine. It would have taken probably on the order of 6 to 12 months, or I could go to a supplier off the shelf, have that delivered in a week, and now machine it with all the bolts in between.

This approach required clever mechanical and structural engineers to figure out how to bolt the pieces together while meeting the system's needs. This exemplifies Helion's ethos: build teams of hands-on people who move quickly and take shortcuts where possible, without sacrificing quality or safety, to bring fusion online as soon as possible.

In 2023, a deal was signed with Microsoft to build a fusion power plant for one of their data centers. This plant will be plugged into the grid, generating electricity from fusion. The timeline is extremely ambitious, with a hard deadline of 2028 for the first electrons to be generated. Microsoft will be buying the power from this plant.

David explains that Microsoft had been observing their progress for years, watching them build new systems, hit milestones, and demonstrate the ability to scale up fusion. Work on the new power plant has been underway for two years, covering aspects like siting, grid interconnection, environmental impact, and regulation. The team is already building the manufacturing capabilities to support it.

David acknowledges the immense difficulty of the task and the skepticism they face. Some still call it a pipe dream. The team's mentality, however, is to push forward regardless because the world needs it. They believe there is no physics reason it cannot be done; the hurdles are in engineering and manufacturing. This approach has served them well in the past when they were told that certain achievements, like compressing an FRC, were impossible.

If we're not discovering new hard problems, we probably didn't push hard enough. We probably didn't push fast enough.

As the company, Helion, has grown from 50 to over 500 people, it's important to keep the team motivated. Many newer employees haven't seen a system built from scratch. For them, seeing the latest machine, Polaris, go from a computer simulation to a physical reality is awe-inspiring. David describes the first moment a system comes online and you see the pink "fusion glow" as a truly powerful experience.

The glow from a fusion plasma can be seen in a couple of ways. The device has small windows all around it where cameras, lasers, and other scientific diagnostics can look in. The plasma is so bright that its light can even shine through the ceramic vacuum vessels themselves.

What is visible to the human eye is not the light from peak thermonuclear fusion. When fusion is happening, it is so hot that the light is in the X-ray spectrum, which is invisible. The beautiful bright purple, fuchsia color comes from the initial stages when the plasma is at a relatively cool one million degrees. At this temperature, it emits photons in a range that humans can see.

Different types of cameras capture different things. Standard SLR cameras only capture an integrated bright flash of light. They cannot see the plasma forming, accelerating, or compressing. However, high-speed cameras can capture these dynamics. By putting special filters on these cameras to measure different wavelengths, scientists can tell which particles are emitting light and when, such as hydrogen or helium. This allows them to take detailed movies of the process, though the simple, integrated flash remains a beautiful sight.

Connecting a nuclear fusion power plant to the grid involves a few steps. The fusion process pushes back on a magnetic field, which recharges capacitors with high-voltage DC power. This steady DC power can then be easily converted to traditional AC power for the grid using large-scale inverters, similar to how a battery works. A unique aspect of a pulsed fusion system is its ability to adjust power output on demand. By changing the pulse repetition rate, the power can be dialed up or down as the grid requires.

However, a more innovative and efficient connection exists, particularly for data centers, which are projected to be one of the biggest power consumers in the future. The electricity from Helion's direct recovery method is already DC, which is what computers use. This opens up the possibility of a direct DC-to-DC connection between the fusion plant and a data center. David explains the potential of this approach:

Rather than going AC to the grid and having all these transmission losses, just going direct DC to the data center, can you plug right in? That's some of the things that my team is looking at now is can you do that direct DC conversion at super high efficiencies and run those GPUs directly? That would be really powerful.

The synergy extends beyond just power. The cooling systems for the fusion plant's semiconductors are very similar to those in a data center, suggesting a deeper engineering integration is possible. Looking ahead, AI is expected to drive an immense need for energy. The cost of computation will eventually be dictated by the cost of electricity. Fusion is a perfect match for AI data centers, which require high-density, reliable power on-site. David believes that current forecasts for electricity demand growth are a wild underestimate and that fusion is essential to ensure power availability does not limit AI's potential.

The long-term vision is not just to build one plant, but to create a full-fledged fusion industry. This involves designing the systems for mass production from the start. The goal is to move beyond single prototypes to a "gigafactory of these fusion generators rolling off the line." David tasks his team with planning for this scale.

We want to factor a gigafactory of these fusion generators rolling off the line. One a month, one a week, one a day. How do we actually go build this? How do we go build a gigafactory so we can have 50 megawatt generators coming off the line, being deployed on a truck and then driving off the factory every day.

Data centers are an interesting early application for fusion energy. They require a large amount of power in a very small area, solving the chicken-and-egg problem of how to deploy thousands of fusion generators. The process of manufacturing at scale itself accelerates innovation in science and physics, a trend also seen in the space industry.

This technology could help humanity advance on the Kardashev scale, which measures a civilization's technological advancement based on energy consumption. A Type 1 civilization can use all the energy available on its planet. With research suggesting there are hundreds of millions to a billion years of fusion fuel on Earth, there is significant room for growth. This abundance of power could unlock transformational technologies, such as huge AI data centers that can think at rapid speeds and spur innovation.

While powerful technology is always a double-edged sword, vast amounts of low-cost, dense energy can work in harmony with nature. Fusion is incredibly energy-dense. A 50-megawatt facility could fit on about an acre of land, whereas a solar farm with the same output would require at least 2,000 acres. This small footprint enables applications like vertical farming, where food is grown in 500-foot-tall buildings instead of across vast horizontal farmland. This could return huge swaths of the Earth's surface to nature.

Abundant, dense energy could also revolutionize space travel. Instead of burning chemical fuels, rockets could be powered by beamed energy, such as microwaves, sent from a powerful source on Earth. This would allow spacecraft to use electricity as their fuel. The most exciting prospect is not just the applications we can currently imagine, but the unknown innovations that will become possible with nearly limitless power.

Nuclear fusion is a compelling power source for deep space travel. David explains that while solar panels are effective near Earth, the sun's energy diminishes significantly with distance, following an inverse-square law. For missions further out, you have to bring your fuel with you. Since mass is extremely expensive in space, high energy density fuel is critical, and fusion provides that. However, there's a challenge: many fusion systems rely on a steam cycle, which requires a cool environment to function. This is a problem in the vacuum of space.

This challenge highlights the need for direct energy conversion, making the system highly efficient. David notes that this mindset comes from the aerospace industry, where every watt of power and every joule of heat must be carefully managed. This principle of maximum efficiency was carried into Helion's design philosophy: recover as much energy as possible from the fusion process directly as electricity.

The conversation then shifts to the energy sources of potential alien civilizations. David believes fusion is the most likely candidate, as it is the fundamental energy source of stars and the universe. This leads to a discussion of the Fermi paradox: why haven't we encountered any of the thousands or millions of advanced civilizations that might exist?

Lex voices his fear of a "Great Filter," a theory suggesting that technologically advanced civilizations are highly likely to destroy themselves. While he is scared by this, he also has some optimism, noting that humans have a knack for survival, even when creating dangerous technologies. David introduces an alternative, more optimistic solution to the paradox called the "Matryoshka brain."

The civilizations get so advanced and they focus not on expanding physically and expanding in space and expanding their reach by planting flags in new places, but grow their cognition, grow their ability to think, they grow their brain, they grow their intellect.

This concept suggests that a highly advanced civilization might construct a Dyson sphere around its star, using all of its energy to power computation and expand its collective intelligence. Rather than physically colonizing the galaxy, they would explore and expand within the realm of cognition and consciousness. This would also explain why we don't see them; their star would be hidden. David sees a parallel in our own society's recent focus on AI, suggesting that humanity, with the help of fusion, might be heading down a similar path.

When asked about the most beautiful idea in physics or nuclear engineering, David Kirtley expresses continuous awe that it all just works. He finds it hard to believe that the delicate balance of conditions is a mere accident. This includes the precise temperatures where life can exist and the perfect equilibrium between fundamental forces like the electromagnetic and strong forces.

This sense of wonder extends from the large to the small. Looking at the leaves on a tree, the cells, the atoms, and even the quantum substructure of those atoms, he is amazed that all the pieces come together. Humans have somehow been able to find and operate within this perfect balance where everything just works.