That’s a huge compliment! Thanks!
Yeah, this isn’t really my subfield, and I haven’t had my ear to the general-physics gossip networks in almost a decade. But I can at least give a general overview of what’s going on.
Physics has made two really big promises in the past century-or-so that have turned out to be largely vaporware. The first is quantum computing (which… is a complicated and controversial thing for me to call “vaporware” because there exist actual commercial products, some of which aren’t even fakes). And the second technology, of course, is nuclear fusion as a viable energy source.
As the saying goes, fusion is always 20 years away. (Nuclear fusion was first 20 years away about 70 years ago.)
We know for absolute certain that fusion as an energy source could work in principle. That’s the frustrating thing about it. It definitely can work, because it does in nature! It’s how stars work! Usually when we dream up a technology it’s about exploiting natural laws in new and creative ways, but this isn’t even that, this is literally just trying to reproduce an extant and common natural phenomenon, but without all the tools that nature has to do it with.
The gist of it is, if you get a whole bunch of quarks together, what they really want to be is iron, especially iron-56 (which has the quarks arranged as 26 protons and 30 neutrons). That’s their favorite type of quark party (atomic nucleus). They’ll settle for other things, like nickel-62, but iron-56 is the favorite. And by favorite, I mean, lowest-energy configuration. To go from iron-56 to anything else requires adding energy to the system.
Let’s first talk about a system that does work, technologically. If you’ve got a huge quark party like uranium-235–way, way bigger than iron-56–if you disrupt it a bit it’ll fall apart into smaller ones like krypton and barium. In the process you actually gain energy–it’s easier for everything to be arranged as krypton and barium than as uranium. That’s how nuclear power plants work, they harness the energy gained from that split, and then use it to power our electric grid. But you DO first have to power that disruption, shake your uranium up, it won’t just happen on its own. (Well, it will just happen on its own–that’s what radioactive means–but not often enough to power much of anything.) But the important thing is you didn’t spend as much energy in disrupting your uranium-235 as you got from its breakup. Even when you take into account literally everything that went into it (powering the power plant itself, the computers, etc), you still have a total energy gain. Even when you take into account building the power plant! Over time, you’re going to make all that energy back and more, so it’s a viable technology.
Ok, what if you’ve got one that’s smaller than iron-56? Same deal, in principle. Littler ones will squish together into iron if they can. The relief they feel at being squished together, at being more similar to iron, outweighs how much energy has to go into forcing them together. In principle.
Here’s the problem. These quark parties all end up with more up quarks than down quarks, which means that they’re all positively electrically charged. You get two of these little quark parties up next to each other and they’re going to repel like hell up to a point. At some point they get close enough together that the quarks all see each other and reconfigure, and trying to reconfigure to be more iron-like is a stronger effect than electrical repulsion. But only at extremely short range. Up until you reach that point where you’re close enough, the electric fields are going to be fighting each other hard. So, although in the end you’ve taken your two quark configurations and shoved them together to be more iron-like, and they’re all having more fun in this new, larger party, you had to expend a hell of a lot of energy to get them all in the same room in the first place. Here, I drew a picture in MSPaint:
Your best chance is to start as small as you can, as far from iron as you can, with hydrogen. Hydrogen is small and lonely and easier to convince; hydrogen also only has one proton so it’s going to have less electrical repulsion. So, hydrogen is a double-win in terms of this being less difficult. And, again, we know from nature that this is possible! This is literally how stars work. They squish a bunch of hydrogen together, and some of it sticks as helium, releasing energy in the form of light.
This would be an extremely exciting technology! It’s self-sustaining: enough energy can (in principle, or in stars) be released by one fusion event to power the next fusion event AND have some left over for your electrical grid; all you have to do is get it started. It’s clean: all you need is some hydrogen, and the byproduct is helium, which is also clean. And it for sure works in nature as an almost-inexhaustible energy source. (Our sun’s expected lifetime, powered by hydrogen fusion, is about 10 billion years.) All we have to do is… replicate the conditions at the core of a star…
Here’s what stars have that we don’t: pressure and heat. The pressure comes from gravity–stars are so big, and the gravity they generate means they have immense pressures at their core. And then the heat just follows from that–lacking other forces, temperature in an ideal gas is proportional to pressure, so when you raise pressure you raise temperature. Pressure of course forces the hydrogen closer together so two hydrogens are more likely to meet. Heat means (definitionally) that they zoom around faster, so that if they do meet it’s more likely that they’re going to run into each other hard enough to overcome that electrically repulsive barrier, and get close enough that the quarks are like “oh! more quarks! let’s reconfigure.”
Of the two, pressure is by far the harder parameter to generate in a lab, although at the temperatures we’re talking about–around 10 million degrees–it’s not like heating it is easy. As it turns out–as you may have guessed by the fact that we did not successfully build a fusion power plant in the 1970s (as predicted) nor any time since–the whole thing is really beyond our current technology. That said, there are two main approaches that are given significant funding right now:
- magnetic confinement
- inertial confinement
Now, magnetic confinement was more or less the original proposal, the one they’ve been working on for really 70 years. They use heating elements to get the hydrogen hot, and then they use magnetic fields to guide the moving atoms on trajectories, hopefully without losing heat in the process, so that when they do happen to collide with each other they haven’t lost that energy.
Designing your magnets so that the hydrogen circulates stably, without all being lost by colliding into the walls of your machine or whatever, is pretty hard. Designing the magnetic fields to force a pressure increase, on top of that, is even harder still. It’s done in giant machines like tokamaks and I guess stellarators (pics from wikipedia):
The magnetic confinement fields are created by giant electromagnets. So in addition to needing to heat up your hydrogen source, you also need huge, huge amounts of electrical current to run the magnets.
How well does magnetic confinement do? Uh, I can’t easily find a really good answer to that, because the most advanced project isn’t even measuring its energy output. They are successfully achieving fusion! And have been for decades! That’s certainly something. But are they producing as much energy through fusion as they’re putting in to run their electromagnets and their heat source? Big shrug, but if they are they’ve never proved it.
Which brings us to inertial confinement, the newer, shinier (literally) option. In inertial confinement you’re a lot more directly trying to apply pressure to a packet of hydrogen, and hoping everything goes smoothly from there. Inertial confinement makes sense, because the gravity used by stars is a type of inertial confinement! But how do you get that kind of pressure on Earth? Well, the solution they came up with was shining extremely powerful lasers on a packet of hydrogen to force its outermost layer–where the lasers hit–to explode. If you do it evenly all around, in theory you should get significant implosion from your ring of mini-explosions. This, to be clear, is absolutely not how stars do it–in fact, in stars, the light pressure is a counter-pressure to gravity, pushing outwards from the core where it’s generated. It’s what keeps the outer layers of a star nice and fluffy so it doesn’t collapse into a white dwarf or a neutron star. But this method it does have its advantages: the laser light causing the implosion does double-duty by also acting as the heat source for getting the hydrogen packet up to fusion-level temperatures.
(the standard picture of a packet of hydrogen being hit by lasers and having rings of mini-explosions on the outside, resulting in (hopefully) implosion)
Ok. So does this method work? So here’s the thing. Today they’re publishing in the mainstream media that they got more energy out–due to fusion–than they put in with their lasers.
Than they put in with their lasers. Not than they put into powering the lasers. Let alone powering the facility.
They are absolutely eliding that. They are shifting the goalposts to something ridiculous, something that can’t even be cross-compared with a magnetic confinement experiment. And then claiming they won, that they’re the first to break even or better. They absolutely did not. Lasers aren’t even efficient, they’re cheating by at least whole order of magnitude here, possibly two. Oh, and also, each shot is one-and-done, and then you have to get a whole new hydrogen packet and start over–you can’t do anything sustainable with this. It apparently currently takes them a full day to reset their machines.
There is actually no real comparison available between the two approaches right now, but we can say with confidence that neither has succeeded yet. Sure, inertial confinement reached a new milestone today, and that’s great, any incremental progress is great. But it was also an extremely arbitrary milestone that doesn’t necessarily matter that much as a milestone. It’s just incremental progress.
And here’s the other thing. Physics in general tends not to be drama-heavy. Physicists have a bro-code where they don’t talk shit about each other in public. Of course they do behind closed doors, it’s human nature, but that means you’re only going to hear about your own subfield, and your friends’ coworkers. The exception, the one branch of physics that has public blowups, angry letters in physics newsletters, etc, is fusion science.
So now it’s time to get out your popcorn.
