The Star Builders

by Arthur Turrell

  “You blink and your fuel is no longer there,” Fernanda says. The effect of this is to make the whole machine strain at its moorings, with forces equivalent to hundreds of tons and jumps of a centimeter or more. For this reason, JET rests on giant springs.

  Just as seriously, a sudden energy release can melt the walls—even though they can withstand a few thousand degrees. Unleashed electromagnetic forces are strong enough to twist the reactor structure, misshaping it and ruining its confining properties. Disruptions are bad news.

  Fernanda says that predicting disruptions is a big physics challenge facing tokamaks. One large disruption could terminally damage the reactor. Lorne Horton told me that industrial partners who might actually build a fusion reactor are adamant that the technology must be predictable to have a chance of being commercially viable. A reactor that stops without warning, potentially breaking itself in the process, is just not going to allow star builders to achieve their ambitions.

  It’s no coincidence that Culham’s star-maker-in-chief, Professor Ian Chapman, made his (scientific) name by trying to pin down the causes of instabilities in tokamaks. He’s published more than 110 scientific papers and has been invited to speak to other physicists at conferences all over the world. His speciality is the sawtooth instability; it causes temperatures and densities to slowly climb, only to suddenly crash back down again, firing off hot plasma into the reactor walls as it does so.

  Disruptions can even threaten the quality of the energy confinement of the reactor in future experiments. When particles from the plasma strike the wall and melt it, even on a microscopic scale, heavy wall materials are released into the fusion fuel. This heavy metal is definitely not Fernanda’s favorite, as it acts like a pollutant and has a terrible effect on the ability of the reactor to confine energy.

  JET’s internal walls are mainly made of beryllium and tungsten, atoms that have four and seventy-four electrons respectively, compared to a hydrogen atom’s one electron. Because deuterium and tritium just have one electron each, and because it’s so hot in tokamaks, they get completely dissociated from their nuclei when they turn into plasma. It takes much more energy to remove all of the electrons from beryllium and tungsten, and that is precisely why they cause problems when they pollute the plasma. As beryllium and tungsten become partially stripped of their electrons, they suck energy away from the hydrogen fuel.

  Worse, the electrons that remain attached to the beryllium and tungsten nuclei can temporarily go into an excited, higher energy state. When they relax again, the energy of their excited state has to go somewhere, and it tends to come out in the form of light. This light is an energy escape route straight out of the hot plasma.

  The pattern of colors that the transitions of electrons create as they jump into different states are like fingerprints; they say what particles are present, how many electrons they have, and what the temperature and density around them is. This is why a hot flame burns blue while a colder flame burns red, or even yellow. The pattern of colors tells a story about what’s going on in the hot fuel—which we know is too hot for any thermometer to work—and star builders often use this technique of looking at the light emitted, called spectroscopy, to infer conditions in their star traps. The light radiation from atoms whose electrons are partially stripped away can be useful. But when metals are radiating away energy from the center of a tokamak, it’s really bad news.

  Even tiny amounts of wall material getting into the fuel ruin the reactor’s performance. At less than 0.02 percent tungsten, the radiation emitted in the form of light carries away a whopping 50 percent of the energy from the helium nuclei created in fusion reactions. Elements with more electrons light up like Christmas trees.10

  For this reason, most of the walls of JET are made of beryllium, with only four electrons. Only the points where the very worst, and hottest, fusion fuel lashes, whiplike, into the walls are made out of the more robust tungsten. Tungsten has the highest melting point of any metal, at 3,422 degrees Celsius (a little less than 6,200 degrees Fahrenheit), while beryllium melts at a comparatively chilly 1,287 degrees Celsius (approximately 2,350 degrees Fahreneheit). A Q greater than one tokamak would see the part of the wall specifically designed to take the biggest pounding, the divertor, exposed to a heat load equivalent to that of a spacecraft reentering Earth’s atmosphere. Truly, star building requires mastery of extremes that make bringing a shuttle down from orbit seem a cinch. The design of tokamaks has cleverly incorporated the divertor to be the one place that takes an absolute beating. It also serves, as the name suggests, as a place where impurities that flake off the main beryllium walls are isolated and removed, leaving the fuel in the doughnut’s tube pure.

  Purity is so important that the cleanliness of the inside of the tokamak chamber would have even Marie Kondo impressed. There must be no air, no water, no grease, no oil—nothing that could contaminate the fusion fuel and unwittingly cause energy to leak away as light radiation. The vacuuming of a tokamak is so effective in removing particles that it’s equivalent to removing all but a single star from the Milky Way. The scrub job is made harder because atoms like oxygen don’t just float around in molecules, they can get trapped in the walls too. The only way to deep clean the walls is to bake the chamber at a few hundred degrees for long periods of time. Because of this, Lorne Horton tells me that to do anything in the chamber with humans, which inevitably involves a cleanup, takes a minimum of three months, without counting the time taken to actually do the work.

  The need for such care has also seen the Culham Centre for Fusion Energy become a center of excellence in robotics. Remotely controlled robot arms can get in and out without disturbing the chamber’s vacuum. While Lorne and I were wandering around Culham’s main building, I saw trainee engineers trying out the remote-handling arms for the first time via a controller that looked like a periscope with joysticks. Later, in a mock-up of the reaction chamber housed in a corner of the diagnostic hall, I saw the engineers’ every move translated into the surprisingly graceful movements of a metal arm that weighed tons. The sense of touch created by the robotic arm is so good, Lorne told me, that experienced operators can feel if a bolt is not aligned with its thread.

  The robotics technology is being spun out separately to all kinds of industries where remote handling might be useful, including nuclear fission. In one warehouse on site, I saw a three-storey wire mesh cage with robotic arms busily completing tasks inside; another had drones angrily buzzing back and forth. The scientists here really are quietly creating the future.

  The World’s Most Successful Fusion Experiment

  The world is paying attention. JET is, after all, a collaboration between many of the nations of Europe under the name of Euratom. Although the idea was born early in the 1970s, it took years of wrangling before a location was agreed upon. Governments jostled to host the facility that would be a shining example of technological progress and would undoubtedly bring with it good jobs and lucrative contracts. Six different sites were fought over. “Mad” Rebut, Paul Henri-Rebut, the driving force behind JET’s design and eventually its director, was so incensed by the delay that he suggested putting JET on the Queen Elizabeth 2 ocean liner so that it could dock in a different European port each month. Finally, in 1984, JET was completed and opened by Queen Elizabeth II (note, not on Queen Elizabeth II) and then French President François Mitterrand.

  Ian Chapman tells me that JET cost about £2 billion (a little more than $2.5 billion) in today’s money and took four years to build. You might wonder what Culham’s star builders have been doing with it since it opened. Surely, a machine that is more than thirty-five years old can hardly be at the cutting edge of technology, much less usher in an energy revolution.

  JET set the world record for plasma Q with deuterium-tritium fuel in 1997, at 0.67—that is, for a brief moment, 67 percent of the energy put in by Lorne Horton’s blowtorches was released by fusion reactions within the reactor chamber. It produced sixteen megawatts o
f fusion power; very roughly, that would be enough to power thirty thousand homes if it was available continuously.

  This seems so tantalizingly close to a plasma Q of 100 percent that you’re probably wondering why, in the years since, they haven’t managed to get any further than they did in that single landmark experiment. The truth is that this was on the edge of what JET could do in the 1990s and 2000s, and more important, the energy output was only maintained for less than one second because of instabilities that ruined the confinement of the plasma. The 100 percent barrier is very important psychologically, including for those who are funding fusion, but reaching it doesn’t automatically solve fusion, especially with such short running times. There’s also the subtlety that only one-fifth of fusion energy comes out as a charged particle. The neutrons that carry four-fifths of the energy aren’t charged and so fly right out of the magnetic trap. The reactor chamber will lose energy without a Q of at least five (though these neutrons are ultimately what are used to generate electricity, so their energy is critical in other ways).11

  Perhaps JET’s greatest success wasn’t getting to a plasma Q of 0.67 for a brief moment but getting to a Q of 0.18 for five seconds. For commercial fusion energy, tokamak power needs to be sustained for hours, if not indefinitely. JET has been upgraded, repaired, and modified several times over the years, each time missing action for months and months. Having been refitted, it is about to gear up for a new campaign of experiments with deuterium and tritium. It remains the only tokamak in the world able to run deuterium-tritium experiments. Star builders like Lorne Horton, Fernanda Rimini, and Ian Chapman have a good chance of smashing JET’s old record for fusion energy, and a successor to JET that aims to exceed a Q of 1 is already under construction. But it will only happen if the star builders can control the writhing plasma in their fusion furnace.12

  You may wonder why star builders are so optimistic that they will get net energy gain. It’s not just because of JET’s exciting experimental successes; the first fusion devices were in the 1940s, and the tokamak didn’t come along until the 1960s—both long before JET was in operation.

  Star builders have believed in fusion for many years not only because of successes in experiments, but because of a great success in theoretical physics. Just as theory has illuminated nuclear physics, and clarified to star builders what reactions are possible in what conditions, so theory has given guidance on what conditions star machines need to create in order to achieve net energy gain.

  Way back in 1957, a plasma physicist named John Lawson proved a theory that shows that there is no physics-based barrier to igniting fusion reactions in a star machine. “Igniting” here means getting to conditions where fusion sustains itself through its own energy release, like a star.13

  His argument was beautiful in its simplicity: he said that, at a minimum, more energy must be coming into the plasma than leaving it. He then added up all of the ways that energy could get into the fusion plasma, including heating from external sources, like Lorne Horton’s blowtorches, plus the energy of fusion-produced helium nuclei (he assumed that the fusion-produced neutrons would all escape because they can’t be confined by magnetic fields). Next, he added up all of the ways energy could be lost from the plasma, including through bremsstrahlung radiation (X-rays generated by the movement of charged particles in the plasma) and the loss of any plasma that managed to escape confinement. Then Lawson looked at the energy in and energy out every second, on both sides: gains and losses. He realized that all of the gains and losses were somehow dependent on just three properties of the plasma: temperature, density, and the time period over which particles in the plasma remain confined.

  Lawson knew that deuterium and tritium fusion was the easiest possible reaction for star builders to aim for, so he plugged in the numbers for that. Then came the moment of truth. Lawson might have found that there was no combination of temperature, density, and confinement time that could make fusion work. That would have suggested that star builders should just give up trying; their experiments might have improved, but they’d never reach ignition or net energy gain. But Lawson didn’t find that; what he found was that there were combinations of temperature, density, and confinement time that would make fusion work. Lawson’s equation says that star power on Earth will work. A tokamak that could reach temperatures of more than 100 million degrees Celsius (180 million degrees Fahrenheit), densities of more than 10,000 billion particles in each cubic centimeter, and energy in the plasma that is confined for more than 100 seconds would hit his requirements and ignite its fusion plasma.

  What this theory means is that it’s possible to re-create star power on Earth: hard, yes—the conditions required are extreme, more extreme than anything else on the planet—but possible. The implications of this are as enormous today as they were then. It’s this insight that has kept generations of star builders optimistic about delivering on the promise of nuclear fusion: they know it’s scientifically possible; they just need to build a machine good enough to make it happen. And JET has come very, very close.

  Lawson’s equation also shows why stars are fusion reactors that just keep on going; gravity can achieve high temperature, high density, and sufficient plasma energy confinement in perpetuity. More important for fusion on Earth, using Lawson’s equation, star builders know exactly what they have to aim for in their star machines. Tokamaks use high temperatures, with plasma confinement times measured in seconds, perhaps eventually in hours, thanks to using magnetic fields. But Lawson’s equation also tells us that magnetic fields are not the only way to trap a star…

  I. Do not try this!

  CHAPTER 6 HOW TO BUILD A STAR WITH INERTIA

  “… like trying to confine jelly with rubber bands.”

  —Edward Teller, describing trapping plasma with magnetic fields1

  You don’t have to be in California for long to see why the two scientists most responsible for establishing Lawrence Livermore National Laboratory—Ernest O. Lawrence and Edward Teller (the controversial physicist best known as the father of the hydrogen bomb)—were happy to set up shop here. On a February morning that really ought to be cold, there’s so much sunshine that I remark to the person taking me between buildings on Livermore’s sprawling site that I feel like I should be outside enjoying it instead of poking around nuclear facilities. She tells me that she felt like that too, for the first year after she moved from South Dakota. But then she realized that it was sunny and fine almost every day here on the east side of the San Francisco Bay, and that she no longer needed to rush outside to catch it. It would be there waiting for her whenever.

  And, as I arrived at the laboratory, I couldn’t help but notice a mile-long array of solar panels soaking up that bright Californian sunshine. It’s here that scientists are using laser light to create their own sunshine, via the National Ignition Facility (NIF), the world’s most successful inertial confinement fusion experiment. In the hills above Lawrence Livermore, I see more evidence of their pursuit of clean energy, in the form of a sparse forest of wind turbines ponderously revolving under a laid-back breeze. If the juxtaposition of so bucolic a scene with a government laboratory bristling with security, nuclear secrets, and scientific experiments should have seemed awkward, it didn’t. On site, there are reed-ringed ponds and bicycle lanes. As with most of America, cars and roads are ubiquitous too—I’m on foot, but you’re almost obliged to drive around, if not due to the sheer size of the facility then owing to cultural norms. On a previous visit I was asked to drive to a building where I was scheduled to have a meeting. Following my minder, I did as I was told, and we drove down a couple of small roads to get to another parking lot. When I emerged from the car I could see, less than one hundred meters away (a little more than three hundred feet), the spot that I’d been parked in moments before.

  I’m here on another visit to Livermore to find out how scientists at the NIF are using inertia to build a star. Some of the scientists working here argue that although their mac
hine has not yet matched the record energy gain of JET, their inertial confinement fusion machine is the closest in the world to achieving net energy gain. And they are catching up fast.

  Inertial confinement fusion is very different from magnetic confinement fusion; for a start, magnetic confinement tends to run continuously, like those takeout pizza ovens with conveyer belts, while inertial fusion is a batch process, like a bread oven. But the biggest difference is that inertial confinement fusion doesn’t use magnetic fields to confine the plasma. Instead, it uses inertia. You’d be forgiven for not having a clue what that means. CEO of inertial confinement firm First Light Fusion Dr. Nick Hawker later put it to me this way: “There’s nothing holding the plasma together—no big magnets, or external forces.”

  But surely something needs to hold the plasma together while fusion is happening.

  Lawson’s equation for net energy gain tells the star builders that nuclear fusion schemes need a combination of high temperatures, high densities, and good energy confinement to reach net energy gain. But it also tells them that they can mix between these three, like DJs mixing tracks but keeping the overall sound volume the same. Magnetic confinement star builders use plasma far less dense than in the Sun, even less dense than the air we breathe, but they do it by creating the hottest temperatures in the solar system, hotter than the Sun, and by using long energy confinement times.

  Inertial confinement fusion uses a different mix of temperature, density, and confinement. Comparing NIF, the world’s leading inertial confinement fusion machine, with JET, the world’s leading magnetic confinement machine, shows the compromises. NIF’s plasma reaches temperatures that are a shade less hot than on the JET tokamak. But the densities and pressures are far higher. Assuming you’re reading this on Earth, you’re experiencing a pressure of just one atmosphere, but the center of the Sun has a pressure beyond 100 billion atmospheres—and that’s what NIF has too, its director, Dr. Mark Herrmann, told me. “So the centers of our implosions are like star matter.”