I’m going to lay my cards on the table and say this: I think it’s extremely unlike that a nuclear fusion power plant will send any energy to civilian grids in my lifetime. Having said that:

The fusion hype cycle never ends, it hibernates. In 2006, the international community came together to plan a nuclear fusion test reactor in France, called ITER. 20 years later and it still hasn’t been built (it was supposed to be completed by 2016). But ITER isn’t an actual fusion power plant, it is merely a research testing station designed to help scientists plan what an *actual fusion power plant* might look like, and solve a few of the many unsolved problems that have plagued the industry.

But ITER isn’t the only game in town. In the easy money era of pre-2022, several start-up corporations came online, claiming that *they* would have a working fusion reactor before ITER even completed building. They’d be selling to the grid by 2025 (later 2030, then 2035). Helion, Commonwealth, there were some big names with big backers, claiming unequivocally that they would crack fusion power and start selling it to the grid.

With this, the hype for fusion has come in cycles, waxing and waning as old promises are broken and new promises are made. It’s a joke that gets some people very prickly to say that “fusion power is 20 years away, and has been for 75 years,” because fusion-backers are well aware of the industry’s many many failed promises but think that “this time, things will be different.” And maybe it will be! But I’m almost certain it won’t be.

Just so we’re all on the same page: what is fusion? If you smash two atoms together, you can produce more energy than you put in. Atoms don’t like smashing together though, they naturally repel each other very effectively. The only way to overcome this barrier is to crush the atoms under incredible pressure (in stars), or heat them up to millions or billions of degrees Kelvin (here on Earth). These super-heated atoms create a “plasma,” with electrons stripped from nuclei, and the theory is that the plasma can self-perpetuate: the fusion reactions produce enough heat to keep the plasma superheated, with excess heat being used to boil water and drive a steam turbine (generating electricity).

This post is probably too late in the hype cycle. I think 2022 was the sweet spot when I saw these startup discussed commonly on social media. But I want to make a post and here’s my thesis: none of the fusion startups, nor government projects have any hope of successfully making a fusion power plant that brings power to the grid.

That’s honestly a hard claim to defend because there are actually many different *types* of fusion, and even if you conclusively argue that the ITER-type fusion reactors are basically impossible in our lifetimes, the National Ignition Lab-type fusion supporters will come out of the woodwork to say that *their* ideas are actually better and more feasible, and then you have to start the argument all over. So I’m going to try to take this in stages.

First of all: there are several different atoms you can fuse nuclei together to make energy. You can fuse any atoms lighter than Iron together and get more energy than you put in (supermassive stars do this in their cores). But the realistic scenarios of fusion here on earth focus on a few select atoms. To get the nomenclature down, Hydrogen comes in 3 isotopes: a single proton with no neutrons (normal boring hydrogen), a single proton with a single neutron (deterium aka “heavy hydrogen,” which also makes up “heavy water”), and a single proton with 2 neutrons (aka tritium). So anyway here’s how you’d do fusion here on Earth

• Tritium-Deuterium fusion – nearly every serious fusion proposal wants to do this
• Deuterium-Deuterium fusion – harder to do, but deuterium is extremely abundant compared to tritium
• Something else (Deuterium-Helium3, Hyrogen-Boron, etc) – occasionally proposed when people are tired of naysayers like me pointing out how the previous two are infeasible

I’ll take these in reverse order:

Something Else (Deuterium-Helium3, Hyrogen-Boron, etc)

Using “something else” besides Tritium or Deterium for fusion is a pretty dead-on-arrival proposal in my opinion, mostly due to “Bremsstrahlung radiation” which I will directly translate to “braking radiation” because I’m not good at German.

When atoms are superheated (which is necessary if you want to get them to fuse), their nucleus and electrons separate. The negatively charged electrons whip around freely, wholly unchained to the positively charged nucleus.

But what happens when an electron and a nucleus fly past each other? Their opposite charges will interact and pull towards each other. The electron (which is WAY lighter than the nucleus) will slow down immensely, “braking” as it is pulled by the charge of the nucleus. When a charged particle slows down, it must emit energy, and so the electron emits X-ray radiation as it breaks to a near screeching halt compared to its previously unimaginable speeds.

That’s why it’s called “braking radiation.”

The thing about braking radiation is that it becomes more powerful as the atoms involved get bigger. That’s because bigger atoms have a bigger nucleus which carry a bigger charge. The Hydrogen nucleus has a measly +1 charge, it causes a tiny braking radiation on the zipping electrons. Boron nuclei have a +5 charge. They don’t cause 5 times as much braking radiation, they actually cause 25 times as much because the radiation goes up with the square of the charge.

This braking radiation produces massive amounts of X-rays, but it’s not the danger of the X-rays that is the problem, it’s the fact that they take with them all the heat of the plasma. Remember that this is happening in a fusion reactor, the reactor *must* be superheated to hundreds of millions degrees Kelvin (or Billions in the case of Boron and Helium based fusion). But the braking radiation produces so many X-rays, that they all radiate away and steal the heat from the plasma.

So if you try to do fusion with a Boron or Helium-based plasma, you need the heat from the fusion to keep the plasma hot enough to *continue fusing*. However the braking radiation will unavoidably be *cooling the plasma down faster than the fusion heats it up*. You’ll start up your fusion reactor at several billions of degrees Kelvin, only to see it rapidly cool to below-fusion temperatures, as fusion cannot keep it hot faster than braking radiation cools it down.

Fundamentally, it’s not possible on earth to do this kind of plasma fusion with Boron and Helium. The startups try to handwave this away, proposing for instance that they can heat the nuclei (so they’ll fuse) without heating the electrons (so they won’t produce braking radiation), but that’s pretty much nonsense (the two trade heat back and forth almost instantly).

Anyway that’s my two cents. Fusion that relies on elements heavy than hydrogen is impossible in the mid to near future, as there’s no reasonable way to keep the plasma hot while the braking radiation cools it down.

Deuterium-Deuterium fusion

D-D (D alone usually means “Deuterium” when talking fusion) is sometimes presented as an alternative to Tritium. But it isn’t. D-D fuses nearly 300 times more slowly than T-D (Tritium-Deterium) at the temperatures ITER and other startups are capable of reaching (a few million degrees Kelviun). That doesn’t just mean you need a power plant 300 times bigger, it realistically means D-D fusion *cannot heat the plasma faster than it cools*, meaning just like above you’d turn on your fusion reactor only to see it quickly cool as the fusion can’t sustain the temperature.

To make D-D fusion self-sustaining, you’d need temperatures of a few billion degrees Kelvin, not just a few hundred million. An order of magnitude bigger with an order of magnitude more problems keeping that plasma confined (ie not destroying the power plant) while fusion is kicking off.

No one has realistically proposed confinement methods that use known materials and keep the plasma from escaping or cooling down. There’s usually talk at this point of proposed or speculative materials, things that *might* work, but either haven’t been tested yet, haven’t been made in more than microgram quantities, or just haven’t been invented yet.

Tritium-Deuterium fusion (there’s a lot to cover)

Why then does everyone come back to T-D fusion? Why does every serious proposal use this and only this? Why is the international collaboration ITER, a collaboration between most of the world’s advanced economies, still only have plans to fuse Tritium to Deuterium? Basically it’s process of elimination. I don’t want to explain resonance states and Helium-5, but T-D fusion is the only atomic fusion that is reachable with materials that are currently under study, needing a mere 150 million degrees Kelvin to get going, unlike D-D fusion. And because both T and D have a charge of just +1, the braking radiation is marginal.

So of ALL the elements on the periodic table, and ALL the isotopes of those elements, Tritium-Deuterium is the only pair we can reasonably fuse in a controlled manner with the technology available today and for the next 100 years.

But just because it’s possible to fuse these atoms, doesn’t mean it’s possible to power the world with them.

See, fusing Tritium isn’t like putting gas in your car, Tritium is rarer than Diamonds, rarer than Gold, rarer than Platinum and Palladium and Plutonium. If Tritium is considered separately from Hydrogen, it would be one of the rarest element on Earth. There are about 25 kilograms of Tritium in the entire world. There are over 200,000,000 kg of Gold currently in human hands, and there are billions to trillions more deep within the Earth’s crust and oceans.

But to power a city like Chicago for a year would need about 275 kg of Tritium. The entire world’s supply of Tritium would power America’s 4th largest city for less than 2 months.

How are fusion advocates claiming they’ll power our cities, if the resources necessary to do so are so incredibly rare?

Well it helps to know where Tritium comes from. None of the Tritium on Earth was formed naturally. Tritium has a half-life of about 12 years, so if any of it WAS here when our planet formed 4ish billion years ago, it has long since decayed away. Instead, Earth’s Tritium was formed when Deuterium is hit with high energy radiation. Deuterium is a rare isotope of Hydrogen, so this doesn’t happen much in nature. But it DOES happen a lot in Canada, because of the CANDU fission reactors.

CANDU reactors use uranium to create power, and they moderate their energy production with Deuterated water, aka heavy water, aka water where the normal Hydrogen has been replaced by Deuterium. A CANDU reactor creates the perfect conditions for the creation of Tritium, although they consider this a biproduct that needs to be removed, not a desired effect. But again, because of the 12 year half-life, the Tritium that Canada produces quickly decays away to nothing while in storage, so there’s never been a chance to create a whole lot of it for fusion power.

So back to the question: if there isn’t enough Tritium to run a fusion reactor, how can fusion power our cities? Fusion advocates want to CREATE new Tritium, using their fusion reactors, and feed the created fusion back in to continue the process. Canada’s 25 kg of Tritium would be like a “pilot light” for a fusion reactor. Once the reactor has been turned on with a little Tritium, it can keep producing more and more Tritium as it needs, continuing the fusion process indefinitely. Deuterium by the way is relatively rare, but Earth has SO MUCH WATER that it won’t be *that* hard to get the Deuterium needed from that.

But here’s the thing: I believe it will be IMPOSSIBLE for fusion reactors to create the Tritium they need with the technology available to use within the next 100 years. I will continue this thought in my next post, because this one is getting long, but to summarize my thesis:

• Fusing anything except Tritium and Deuterium for power will be impossible for the foreseeable future.
  • The temperatures needed are too high
  • The braking radiation will steal power from the reactor faster than power is created
• Tritium-Deuterium fusion is the only possibility for fusion power for the foreseeable future
  • But it will be impossible to produce enough Tritium to do this for the foreseeable future

I haven’t proven that last point, but I will devote an entire post to it soon. I want to get this posted because I’ve been writing of it for a week, and it’s already very long. But stay tuned for the next post, because if I can prove that point then it will support the final bullet point of my thesis:

• It will not be possible to power humanity with fusion for the foreseeable future.