The fusing of small atomic nuclei into larger ones, with the associated release of particles carrying a lot of motion-energy, is the mechanism that powers the Sun’s furnace, and that of other stars. This was first suspected in the 1920’s, and confirmed in the 1930s.
Nuclear fission (the breaking of larger atomic nuclei into smaller pieces) was discovered in the 1930s, and used to generate energy in 1942. Work on fission in settings both uncontrolled (i.e. bombs) and controlled (ie. power plants) proceeded rapidly; bombs unfortunately were quickly designed and built during World War II, while useful power plants were already operating by 1951. Meanwhile work on fusion also proceeded rapidly; in the uncontrolled setting, the first bomb using fusion (triggered by a fission bomb!) was already made in 1951, and in a flash of a decade, huge numbers of hydrogen bombs filled the arsenals of superpowers large and small. But controlled fusion for power plants… Ah.
Had it been as easy to control fusion as it was to control fission, we’d have fusion plants everywhere; fossil fuels would be consigned only to certain forms of transportation, and the climate crisis would be far less serious than it is right now. But unfortunately, it has been 70 years of mostly bad news — tragic news, really, for the planet.
But finally we have a little glimmer of hope. On December 5th, somebody finally managed, without using a bomb, to get more fusion-generated energy out of an object than the energy they had to put into it.
[UPDATE: Not really. Though this was a success and a milestone, it wasn’t nearly as good as advertised. Yes, more energy came out of the fusing material than was put into the fusing material. But it took far more energy to make the necessary laser light in the first place — 300 megajoules of energy off the electricity grid, compared to a gain from the fusing material of about 1 megajoule. So overall it was still a big net loss, even though locally, at the fusing material, it was a net gain. See this link, in particular the third figure, which shows that the largest energy cost was electricity from the grid to run the lasers. In short, well, it’s still a good day for fusion, but we are even further from power plants than we were led to believe today.]
Poster Child for Particle Physics
In the Sun and similar stars, fusion proceeds through several processes in which protons (the nuclei of the simplest form of hydrogen) are converted to neutrons and combine with other protons to form mainly helium nuclei (two protons and two neutrons). Other important nuclei are deuterium D (a version of hydrogen with a proton and neutron stuck together), tritium T (another version with a proton plus two neutrons — which is unstable, typically lasting about 12 years), and Helium-3 (two protons plus one neutron.)
Fusion is a fascinating process, because all four of the famous forces of nature are needed. [The fifth, the Higgs force, plays no role, though as is so often the case, the Higgs field is secretly crucial.] In a sense, it’s a poster child for our understanding of how the cosmos works. Consider sunshine:
- We need gravity to hold the Sun together, and to crush its center to the point that its temperature reaches well over ten million degrees.
- We need electromagnetism to produce the light that carries energy to the Sun’s surface and sunshine to Earth.
- We need the strong nuclear force to make protons and neutrons, and to combine them into other simple nuclei such as deuterium, tritium and helium.
- We need the weak nuclear force to convert the abundant protons into neutrons (along with a positron [i.e. an anti-electron] and a neutrino.)
How can we be sure this really happens inside the Sun? There are quite a few ways, but perhaps the most direct is that we observe the neutrinos, which (unlike everything else that’s made in the process) escape from the Sun’s core in vast numbers. Though very difficult to detect on Earth, they are occasionally observed. By now, studies of these neutrinos, as here by the Borexino experiment, are definitive. Everything checks out.
In the recent experiment on Earth, gravity’s role is a little more indirect — obviously we wouldn’t have a planet on which to live and laboratories in which to do experiments without it. But it’s electromagnetism which does the holding and crushing of the material. The role of the strong and weak nuclear forces is similar, though instead of starting with mostly protons, the method that made fusion this week uses the weak nuclear force long before the experiment to make the neutrons needed in deuterium and tritium. The actual moment of fusion involves the strong nuclear force, in which
- D + T –> He + n
i.e. one deuterium nucleus plus one tritium nucleus (a total of two protons and three neutrons) are recombined to make one helium nucleus and one neutron, which come out with more motion-energy than the initial D and T nuclei start with.
The Promise of Endless Cheap Safe[r] Power?
The breakthrough this week? Finally, after decades of promises and disappointments, workers at a US lab, Lawrence Livermore Laboratory in California, working at the National Ignition Facility, have gotten significantly more energy out of fusion than they put in. How this works is described by the lab here. The steps are: make a pellet stocked with D and T; fire up a set of lasers and amplify them to enormous power; aim them into a chamber containing the pellet, heating the chamber to millions of degrees and causing it to emit X-rays (high-energy photons); the blast of X-rays blows off the outer layer of the pellet, which [action-reaction!] causes the inner core of the pellet to greatly compress; in the high temperature and density of the pellet’s core, fusion spontaneously begins and heats the rest of the pellet, causing even more fusion.
Not as easy as it sounds. For a long time they’ve been getting a dud, or just a little fusion. But finally, the energy from fusion has exceeded the energy of the initial lasers by a substantial amount — 50%.
This one momentary success is far from a power plant. But you can’t make a power plant without first making power. So December 5th, eighty years and three days after fission’s first good day, was a good day for fusion on Earth, maybe the first one ever.
If this strategy for making fusion will ever lead to a power plant, this process will have to repeated over and over very rapidly, with the high-energy particles that are created along the way being directed somewhere where they can heat water and turn a steam turbine, from which electric current can be created as it is in many power plants. Leaving aside the major technical challenges, one should understand that this does not come without radioactive pollution; the walls of the container vessel in which the nuclear reactions take place, and other materials inside, will become radioactive over time, and will have to be disposed of with care, as with any radioactive waste. But it’s still vastly safer than a fission power plant, such as are widespread today. Why?
First, the waste from a fission plant is suitable for making nuclear weapons; it has to be not only buried safely but also guarded. Waste from a fusion plant, though still radioactive, is not useful for that purpose.
Second, if a fission plant malfunctions, its nuclear chain-reaction can start running away, getting hotter and hotter until the fuel melts, breaks through the vessel that contains it, and contaminates ground, air and water. By contrast, if a fusion plant malfunctions, its nuclear reactions just… stop.
And third, mining for uranium is bad for the environment (and uranium itself can be turned into a fuel for nuclear weapons.) Mining for hydrogen involves taking some water and passing electric current through it. Admittedly it’s a bit more complicated than that to get the deuterium and especially the tritium you need — the tritium be obtained from lithium, which does require mining — but still, less digging giant holes into mountains and contaminating groundwater with heavy metals.
Meanwhile, both forms of nuclear power have the advantage that they don’t dump loads of carbon into the atmosphere, and avoid the kind of oil spills we saw this week in Kansas.
So even though we are a long way from having nuclear fusion as a power source, and even though there will be some nuclear waste to deal with, there are good reasons to note this day. Someday we might look back on it as the beginning of a transformed economy, a cleaner atmosphere, and a saved planet.
33 Responses
Fascinating, although for me it’s more to do with the social and political science behind the publicizing of the result rather than the physical science. It wouldn’t surprise me if the US military loves this stuff because of the development of cutting-edge, high powered laser technology required; and the scientists and engineers won’t bite the hand that feeds them because this is the stuff they dreamed of doing as kids.
The military certainly loves PR, hype and wonder weapons as a means of securing trillions of taxpayer dollars. Judging by its record of failure, it seems to prefer them to winning wars on the ground. As far as I can tell, fusion is as far away now as it was when I was born, and so is real AI, interstellar travel, and all the other nonsense I was propagandized to expect. It seems to me that this whole civilization desperately needs a reality check.
I think there’s much truth in what you say. But remember the successes, too: GPS, cell phones, speech recognition, cloud computing, fast search, solar panels, international communication, and so on. For all the nonsense, there is also sense; it may be the minority, but unlike the nonsense, it actually changes the world.
My impression is that physics is becoming like almost every other field in the USA now: driven by PR and hype. Seems like about 99% of these hyped scientific and technological breakthroughs amount to nothing, yet they keep coming. There must be a deep cultural and systemic problem with this society, which prefers hype and simulation to cold reality. We see this not just in entertainment, advertising and politics, but across the board, in military, finance, tech and now science. It’s really weird to watch, like something out of a Philip K. Dick story or Baudrillard’s “Simulacra and Simulation”, where a whole society is trying to escape into a world of its own propaganda. Do other people see it this way or am I exaggerating?
I do see it this way. It’s part of why so often this blog presents anti-hype viewpoints, rather than breathless speculation about what might happen next.
Fortunately, science (unlike most of society) has a self-correcting mechanism; only the science that *actually* works turn into technology that *actually* changes the world, or allows even more advanced science.
You can hype the fancy resuable rockets that you’re going to send into space, based on fancy new hyped engineering that’s based on fancy new hyped science; and that can dupe investors and get you lots of venture capital. But when Elon Musk’s SpaceX actually does it, then you know for sure: the underlying technology, and the science that is its foundation, isn’t hype.
You can hype how great your quantum sensors and new lasers and fancy vibration cancellation devices are going to be, but when you actually use them to measure gravitational waves from outer space at the same moment as other telescopes observe a bright flash, and the waves closely match what Einstein’s relativity predicted for two neutron stars, well, then you know: it’s the real deal.
So no matter how bad our society gets on this score, you can remain confident that science will continue, and real steps forward will be made.
Clear and concise explanation. Thank you!
Familiar territory. NIF isn’t going to bite the hand that feeds, and is allowing for the gross misinterpretation of some unrelated research to be used as a PR boost to the federal funding of some fusion programs that have very little chance of ever producing anything of utility. The claim of a “net energy gain” is with respect to the energy hitting the target, not the total energy consumed. The NIF laser array uses about 300-million joules to produce that 2-million-joule X-ray pulse at the hohlraum which resulted in the 3 MJ fusion reaction. In a theoretically ideal reaction, a pellet might conceivably produce as much as 10 MJ of energy within 10–100 ps. In terms of return on investment, I think there are kids who’ve made more efficient Farnsworth–Hirsch fusors in mom’s basement.
ITER and a deluge of pump-and-dump startups notwithstanding, the more interesting work is coming from elsewhere. In terms of any promise from inertial confinement, J-PARC in Ibaraki, Japan is currently investigating increasing the efficiency muon-catalyzed fusion through the use of lasers in conjunction with extreme pre-compression of the fuel.
Agreed (as stated in the red-colored update.) I’m increasingly skeptical this is a path that leads anywhere useful.
And thanks for pointing me to the new work on muon-catalyzed fusion; it would be great if that idea could be resurrected and if something as seemingly esoteric as muons turned out to save the planet.
I read that the laser power generated is just of order 1% of the electrical input power. And on the output side in the conversion of radiation, heat to turbines and electricity generation might be 10% efficient through the path. So that is three orders of magnitude. And then there is the laser pulse duty cycle. NIF has a very long cool down period before a second pulse is possible. Any comments on the engineering side of things?
I’m still learning too, so no useful comments from me at this point. Clearly these are huge problems.
“the blast of X-rays blows off the outer layer of the pellet, which [action-reaction!] causes the inner core of the pellet to greatly compress; in the high temperature and density of the pellet’s core, fusion spontaneously begins and heats the rest of the pellet, causing even more fusion.”
I’m assuming this is because the emission of the x-rays is symmetrical? In other words, x-rays traveling in the positive x direction are matched by x-rays traveling in the negative x direction, conserving momentum, so the the “recoil” of the pellet is radially inward, on all sides, compressing it?
I’m a big fan of your posts, whenever I see news of a physics breakthrough, I can count on you to “break it down” for us lesser folk.
Let’s see: the pellet is spherical, the outer layer of the pellet comes off spherically, and that leads the pellet to collapse spherically. However, if and how the X-rays end up symmetrical isn’t obvious to me. There are some questions of how long it takes for the object containing the pellet to thermalize, and for the outer layer of the pellet to thermalize. You’d have to ask the Ignition Facility experts exactly how that works; I’m not sure exactly where to find a good explanation of that.
The explanation linked in the text doesn’t really answer the question, https://lasers.llnl.gov/about/how-nif-works , but maybe if you dig on that website you can find more info.
Funny that that sounds like a problem the fission bomb designers had as well, i.e. getting the symmetrical collapse of a small volume. Kip Thorne had a related, funny story in his book about Black Holes; as I recall it went something like: some colleagues were working on the problem of stellar collapse and couldn’t get the calculations to generate symmetrical core collapse. He apparently talked to another colleague working on bomb related projects that told him he knew how to solve the problem but unfortunately couldn’t tell Thorne about it.
Great story!
Perhaps this story has a relation to Kip Thorne’s famous “Hoop conjecture” ( about gravitational collapse) that seems at first glance almost obvious, yet it has many deep implications and difficulties ( about the notion of the trapping horizon, the definition of quasi-local mass in GR etc.) when considered more closely.
Hi professor, I ask you, how is it possible to measure energy gain? What kind of energy was it, thermal? But if they reached millions of degrees energy in the operation, were they also get millions of degrees in the energy obtaining? I’m confused. Could you please explain it us? thanks
Degrees is not a measure of energy; it’s closer to (though not exactly) a measure of energy-per-unit-volume. A small amount of boiling water has much less energy than a large amount of boiling water, even though they have the same temperature.
I think the energy, mainly from particle motions, would not be thermal, because I don’t think there’s time for the fusion products to fully randomize; but I might be wrong about that.
I don’t know precisely how they measure the energy gain/loss. There are probably quite a few ways to do it, and ot’s probably quite complicated, maybe involving monitoring the lasers on the one hand and monitoring the number of neutrons coming out on the other, and then extrapolating from what’s observed to what the energy gain and loss must be. If I find out I’ll let you know.
How do we plan to harvest this energy? Steam engine?
I imagine that step won’t be so different from what’s in existing fission reactors; see step 2 of https://www.edfenergy.com/about/nuclear/what-is-nuclear-energy .
Hey Matt, couldn’t resist making an off the wall comment; your comment reminds of at least one useful thing that came out of the cold fusion fiasco (for me anyway) ! In reading about the fallacious claims I did learn that outgoing neutron count is one of the ways of estimating the amount of fusion that occurs in a reaction.
Less spectacular but probably far more important is the reduction in cost in energy from the fusion reactor in the sky:
https://www.energy.gov/eere/solar/sunshot-2030
Nothing is perfect… we still have to make huge expanses of silicon wafers and put them all over the place, at some nontrivial environmental cost. I’ve actually never seen a serious study of the pros and cons of fusion from the sky vs fusion from the ground. Obviously there may be some place for both, unless battery technology allows us to rely solely on the Sun someday — but even then, you have to work out the pros and cons of the batteries. There is no free lunch, unfortunately.
“Finally, after decades of promises and disappointments, workers at a US lab, Lawrence Livermore Laboratory in California, working at the National Ignition Facility, have gotten significantly more energy out of fusion than they put in.”
I believe this is false, or at best very misleading. I unfortunately can’t find the source right now, so I’m pulling this from memory, but my understanding is that the lasers put in 2.1MW of energy, resulting in 2.5MW of energy released, but that it actually took ~300MW of input to the lasers in order for the lasers to deliver that 2.1MW to the target. So they are *orders of magnitude* away from what most people would consider “net positive”.
Please feel free to discard this comment, as I really cannot find the source at the moment, but hopefully at least it can inspire some further investigation.
You are indeed correct, I found this a little while ago, after I wrote the article. https://physicstoday.scitation.org/do/10.1063/PT.6.2.20221213a/full/media/sizes/full/fig3.jpg
What was different this time compared to the other times NIF has attempted this?
“Campbell credits NIF’s latest achievement to advances in the last four to five years in the understanding of hohlraums and improved capsule fabrication, with contributions from other labs and the private sector.”
The passengers on the fusion gravy train are understandably concerned lest it run out of gravy, therefore every now and then issue a report of a stunning new success.
Also good for Energy Secretary Granholm, who is a lawyer and always will be.
Matt,
One question: do they measure the output energy of fusion by measuring kinetic energies of outgoing particles like in accelerators i.e. are these fusion pallets surrounded by detectors ?
I am hoping this is not a hype like wormholes, Susy particles etc. We have been burnt before several times! Do they include the electric power used to run the lasers? A power company will have to include the cost of making these 192 lasers also!!
Your comment is spot on. Look what I found… https://physicstoday.scitation.org/do/10.1063/PT.6.2.20221213a/full/media/sizes/full/fig3.jpg
Here’s a working link to the article: https://physicstoday.scitation.org/do/10.1063/PT.6.2.20221213a/
PS: Thank you for the great content Matt! I am a subscriber since 2014 and I still love your content ❤️