Nuclear fusion is the energy of the future – and it always will be.

It’s an old physicist’s joke I first heard at varsity in the 1980’s.

Nuclear fusion’s reputation for excited talk and practical failure is that bad and that old, which is a shame because it would solve the world’s challenge of having reliable, CO2 emission-free power.

Given that working nuclear fission reactors preceded a nuclear fission bomb, it might have seemed reasonable to conclude that having first developed a fusion bomb in 1952, a working fusion reactor could not be far off.

USA’s Ivy Mike test: first ever Hydrogen Bomb. 10 Megaton yield, November 1952.

Amazing the fundamental problems that can arise from the simple difference between splitting things apart and squeezing things together. But if the physicists had thought about it from a philosophical angle they might have realised how hard the problem is; after all, broken plates don’t put themselves back together.

The physical challenges were obvious enough that optimism should have been low from the start. After all, that first Hydrogen Bomb needed an Atomic Bomb just as a trigger, channelling its incredibly dense light (as dense as matter) to crush and heat the deuterium (an isotope of Hydrogen) that fuelled the beast. Deuterium was chosen because, with one proton and one neutron in its nucleus, it’s heavier than pure Hydrogen, which has only the proton, and therefore needs less heat and pressure to fuse with another deuterium atom, plus that neutron helps with the fusing. Even then a third Hydrogen isotope, Tritium (1 proton, two neutrons) was needed in small doses to make “Mike” go.

But of course Atomic Bombs explode. Given that the nuclear fusion explosion proceeds in billionths of a second just after the fission reaction, that doesn’t matter but obviously it’s not a solution for a reactor, which we very much do not want to explode. The only known fusion reactors are those of nature – the stars in the firmament whose trillions upon trillions of tonnes of hydrogen mass have crushed and heated it in their cores to the point where fusion has started and become self-sustaining.

But how to replicate that in miniature and on Earth?

The two leading potential solutions to this problem – inertial-confinement and magnetic-confinement – are best shown by two huge experimental efforts currently underway: NIF (National Ignition Facility) and ITER (International Thermonuclear Experimental Reactor) And there are possibly two other ways.

NIF and Lasers

The NIF approach is “inertial confinement”: suspend a small pellet of deuterium and tritium, wrapped in a blanket of plastic and metal, inside a chamber, build the world’s most powerful laser, divide that up into multiple beams that hit the target from all directions in an instant, crushing and heating it, and thus generating nuclear fusion. In a way it’s a miniature version of “Mike”, with the laser substituting for an A-Bomb trigger. Because crushing of the fuel occurs in this design it does not have to heat the fuel up as much as things like ITER. You can read about the history of NIF and its predecessors at the link.

At present the NIF is using as this trigger a 500TW (TeraWatt) laser (1,000 times more power than the USA uses at any instant in time), split into 192 beams.

All that energy hitting a small target in a few billionths of a second forces the trapped deuterium and tritium atoms to be stripped of their electrons – creating a “soup” of atomic nuclei, called plasma (a fourth state of matter). Then the heat and crushing pressure will reach the point where the energy of their collisions overcomes the Coulomb force that cause Protons to repel each other, and allows the strong nuclear force to pull them and the neutrons together and hold them. Two Deuterium atoms produce one Helium atom – plus a lot of excess energy, far more than you get from splitting an atom. There are other reactions involving tritium as well.

But the trick is not to get such fusion reactions occurring, tough as that is. The trick is to get enough fusion reactions happening that they create more energy than what was poured in to start them – a threshold point called the Lawson Criteria.

In the case of NIF there have been countless holdups, diversions into other research and so forth. But every upgrade in the system has seen it get closer to that threshold point. In 2021 NIF produced 70% of the energy of the laser, beating the record set in 1997 by the JET reactor at 67% and achieving a burning thermonuclear plasma. The experiment used 477 MJ of electrical energy to get 1.8 MJ of energy into the target to create 1.3 MJ of fusion energy.

But during the following year they had trouble re-creating the test and people began to wonder if it had been a fluke. And then on December 5, 2022:

 A 3-million-joule burst emerged from a peppercorn-sized capsule of fuel when it was heated with a 2-million-joule laser pulse.

The NIF produced 3.35 megajoules of energy from a 2.05 megajoule input of laser light for an energy gain of about 1.3Mj. This produced a reasonable amount of excitement, even in the MSM. But keep in mind that fact that it’s only equivalent to roughly the energy used in about 15 minutes of running a hair dryer, admittedly delivered in a millionth of a second. There also another catch arising from that fact:

But this latest fusion burst still didn’t produce enough energy to run the laser power supplies and other systems of the NIF experiment. It took about 300 million joules of energy from the electrical grid to get a hundredth of the energy back in fusion.

“The net energy gain is with respect to the energy in the light that was shined on the target, not with respect to the energy that went into making that light,” says University of Rochester physicist Riccardo Betti

Bugger!

And of course there’s an even bigger problem in the future, which is turning this into a practical reactor. Right now it can take hours or even days and weeks to set up the little target and it took hours to “pump” the lasers up to the power levels needed for the shot. What are they going to have, Deuterium “pellets” chucked into the chamber every few seconds just in time to be hit by a laser pulse? It would make a piston engine look smooth running.

What’s needed is a self-sustaining soup of fusing plasma – and that’s just what the next machine is intended to produce.

ITER and Tokamaks

ITER is currently being built in the South of France. It’s huge, the reactor filling the equal of a small office building, with huge amounts of equipment surrounding it for support. It has taken decades and tens of billions of dollars to develop and is supposed to start up in 2025. It uses what’s called a Tokamak design, first invented in the USSR in the 1950’s. The idea is to trap the Deuterium in a giant donut of powerful electromagnetic fields and then slowly heat it up – and then heating further until the velocity of the atoms crashing into eachother causes them to fuse.

How hot does it have to get for this? Anywhere between 10 and 100 times hotter than the core of a star, which is typically 14 million Kelvin (a 35°C day is 308K). The much higher heat is needed because the other factor in a star – pressure – cannot be replicated in a reactor.

Here is a smaller Tokamak in action in this strangely beautiful video (1m)

But that’s not the real trick. Even the first tokamaks caused nuclear fusion reactions. The trick is the same one that applied to the Super Laser design, that damnable Lawson Criteria; to get enough fusion reactions happening that they create more energy than what was poured in to start them.

So not only do you need the atoms to collide and fuse, you need a lot of them to do that, and that means holding all this together for many seconds (geological ages in the world of atoms) and then minutes, hours, days, all while dealing with heat losses, and so forth. It’s an incredibly complex problem, but at least with the tokamak design it seems that the bigger the machine the closer you can get to that precious threshold level. ITER is the biggest tokamak ever.

Here’s a brief (8m)video about ITER’s attempt to create a tiny star.

Despite decades of failed promises everybody seems confident that ITER will work. Recent advances even with smaller scale tokamaks have shown good results:

Now researchers from the EUROfusion consortium more than doubled previous records at the UK Atomic Energy Authority’s Joint European Torus (JET) site in Oxford using the same fuel mixture to be used by commercial fusion energy power plants, the UK government said on Wednesday. JET produced a total of 59 Megajoules of heat energy from fusion over a five-second period, the duration of the fusion experiment. This is not a large amount of energy, but it validates the design choices for ITER

“If we can maintain fusion for five seconds, we can do it for five minutes and then five hours as we scale up our operations in future machines,” said Tony Donné, EUROfusion Program Manager.

Note that it required roughly three times more energy to heat the fuel than it produced.

Computer models that have been tested many times against these smaller tokamaks to benchmark their predictive powers, show that ITER will produce around 500 megawatts of power continuously for 400 seconds while only consuming 50 MW of energy to heat the fuel. The heat won’t be used to produce electricity but simply dumped – it is merely experimental after all. But we’ll know more by 2027.

There are pessimists, and they’ve been correct so far. But Like all scientific research, nuclear fusion reactors should be pressed on with until we know for certain that it’s a dead end. After all lasers themselves are a fairly new technology that’s evolving fast:

“I think we should look at this with optimism,” says Dmitri Orlov, a research scientist at the University of California, San Diego who studies tokamak design. “Today is like watching a baby learning to walk. Eventually, it will run a marathon.”

Not for decades though, as that article notes in its headline, and let’s be blunt: for all the money involved, US fusion research funding has always been done on a relative shoe string. The Manhattan or Apollo projects this is not.

And the thing is that even if all these technical problems are solved there’s another very practical problem in producing electricity from a fusion reactor.

In the movie Ironman the hero, Tony Stark, builds his “Arc Reactor” which produces electricity directly, powering his armoured suit. Which is all very cool – but Science Fiction of course.

In reality either Super Laser or Tokamak machines would simply act as a fancy boiler, producing heat and then steam that drives turbines and generators – which means there’s an efficiency and cost/benefit problem compared to that being done by nuclear fission, or even coal and gas:

In contrast, burning plasma at millions of degrees while confining it with one of the most complicated and costly machines ever built—that is an expensive way to produce steam. … The easiest kind of fusion to achieve, then, may be permanently uneconomical, never able to compete with other forms of producing steam and powering turbines.
…
A conventional combustion engine is about 40 percent efficient at converting the energy it produces into electricity. For fusion, that might be more like 10 to 20 percent

That’s ugly. But is an “Ironman” reactor possible? Maybe?

The “Arc reactor”

Aneutronic fusion reactors would, in theory, directly convert the energy from the plasma into electricity. These reactors would use different atoms, such as helium-3 or boron. But aneutronic fusion is harder to do because it needs higher temperatures and probably more complicated machines, which is not great news considering the challenges already facing the two leading approaches. However that makes the following news intriguing:

Helion, which is backed by OpenAI founder Sam Altman, committed to start producing electricity through fusion by 2028 and target power generation for Microsoft of at least 50 megawatts after a year or pay financial penalties. … “We wouldn’t enter into this agreement if we were not optimistic that engineering advances are gaining momentum,” said Microsoft President Brad Smith. … Helion is building a prototype that it says will demonstrate the ability to produce electricity through fusion next year. “The goal is not to make the world’s coolest technology demo,” Mr. Altman said in an interview. “The goal is to power the world and to do it extremely cheaply.”

Sounds like he’s based at least and Microsoft are not idiots, although they’ve had their share of failures, including very stupid failures. Here’s a video explanation of the Helion idea..

It may or may not work, but it is intuitively simpler than the heat->steam->turbine->generator approach to producing electricity. You can read the skeptics take here in this MIT article.

As it happens Helion is just one of a number of private companies taking entirely different paths toward achieving the goal of providing useful electrical energy from fusion, and their ideas are very interesting. They are spending vastly less money, claim they are making real progress, and have actual plans that make sense to harvest usable power. They may be going down dead end roads as well, but their approaches are promising in ways that the government-led approaches appear not to be.

LCF: Lattice Confinement Fusion

Finally, one of those government agencies, good old NASA, also has another idea, although it’s only aimed at powering space probes and robotic explorers and may not be suitable for scaling up to power a city, Lattice confinement fusion eliminates massive magnets and powerful lasers. The article goes through the laser and Tokamak scene as background before getting to the meat of LCF:

One promising alternative is lattice confinement fusion (LCF), a type of fusion in which the nuclear fuel is bound in a metal lattice. The confinement encourages positively charged nuclei to fuse because the high electron density of the conductive metal reduces the likelihood that two nuclei will repel each other as they get closer together.
…
In a tokamak or a stellarator, the hot plasma is limited to a density of 1014 deuterons per cubic centimeter. Inertial-confinement fusion devices can momentarily reach densities of 1026 deuterons per cubic centimeter. It turns out that metals like erbium can indefinitely hold deuterons at a density of nearly 1023 per cubic centimeter—far higher than the density that can be attained in a magnetic-confinement device, and only three orders of magnitude below that attained in an inertial-confinement device. Crucially, these metals can hold that many ions at room temperature.

And if you already have that density it makes it a hell of a lot easier to get a fusion reaction by firing an electron-beam accelerator at it, which is a much simpler machine than a super-laser, requiring vastly less power. They already have seen fusion reactions from this process in experiments, but again it has to be a self-sustaining reaction – more energy out than in.

Still, if NASA’s LCF approach can be made to work practically there’d be no need to have giant fusion reactors powering cities. Your house could simply have its own little fusion reactor.