Can We Ever Achieve Fusion Power?

Fusion power has long held the promise of delivering near-endless energy without as many unfortunate side effects as nuclear fission. But despite huge investment and some fascinating science, the old adage about practical power generation being 20 years away seems just as true as ever. But is that really the case? [Brian Potter] has written a review article for Construction Physics, which takes us through the decades of fusion research.

For a start, it’s fascinating to learn about the many historical fusion process, the magnetic pinch, the stelarator, and finally the basis of many modern reactors, the tokamak. He demonstrates that we’ve made an impressive amount of progress, but at the same time warns against misleading comparisons. There’s a graph comparing fusion progress with Moore’s Law that he debunks, but he ends on a positive note. Who knows, we might not need a Mr. Fusion to arrive from the future after all!

Fusion reactors are surprisingly easy to make, assuming you don’t mind putting far more energy in than you’d ever receive in return. We’ve featured more than one Farnsworth fusor over the years.

51 thoughts on “Can We Ever Achieve Fusion Power?

    1. That’s a lot of billions wasted on nothing. The only good thing coming out of it is all the side development – e.g., cryogenics, superconductive magnets, etc. Fusion will never be practical (unless our understanding of physics is very wrong and we’ll find some new physics suddenly).

      1. Did you actually look at the graph? All those pretty colored lines are _projections_ on how we _might_ have spent to achieve functional reactors. The black line is what we’ve _actually_ spent. We are not, and never have, really spending much money on fusion research. Sure, a quarter billion a year would be a damn lot to _me_, but for governments, it’s nothing. Given the rewards, it’s a joke that we’re not even really trying.

          1. You’re getting lost in large numbers. The US economy is in the trillions of dollars. Spending a billion on something is a drop in the ocean.

            It’s like you refusing to spend a dollar that would save you ten thousand.

          2. Dude, once again, it is not any better than spending money on, say, an astrology based market prediction agency. Not much money in the grand scheme of things, but still pointless and annoying. Fusion is a dead end and every physicist know it.

    2. Fusion should be easy. All we have to do is, instead of going to the moon or mars and collecting samples, is go to the sun, cut off a bit and bring it back to Earth… fusion to go.

  1. Interesting read. I like the way Brian Potter’s article has attributions and links to the author of the graph, the author of the data, the data itself and the tool used to create the graph.

    1. Please stop this “how many years nonsense”. Take for example ITER. Plans to buid it started growing somewhere in the ’70-ies or ’80-ies, it’s now around 85% built and it is not even intended to generate electricity. After Iter, the next project is supposed to be DEMO, and that one is supposed to generate electricity, somewhere after 2050. but not yet to be commercially viable. It’s intention is to demonstrate that technology has advanced enough and fusion can be commercialized. Commercially viable powerplants can only be built when they have figured that out. Anyone who said fusion was / would be easy is an idiot. Please don’t parrot idiots.

      Below a link to DEMO. Please excuse me if I have not kept up with details, it’s still so far into the future that further details are not very interesting for people like me.

    2. Antimatter power is only 36 years away !
      (Zephram Cochranes launch in 2063)
      Zero Point Energy is only 100 years away !
      (Unless a ZPM can be found in an Egyptian tomb) 😁

  2. A very good article to read on the history of fusion research and development. While fusion is being looked to as a pollution free energy source there are still some things to consider. If sustainable fusion is achieved can the reactor be economically duplicated ? Building a multi-trillion dollar plant that can power maybe 200,000 homes isn’t cost effective. Can it be scalable to compete with a fission plant in size and power produced ? Hopefully it will be smaller and technically manageable. What about the heat produced ? Heated water from the steam turbine side has to go to cooing towers like with a fission plant and heat is release into Earth’s atmosphere. Wouldn’t that increase global warming even if CO2 levels were to drop ? Even though the news media says it is safe, neutron radiation is still produced which can cause embrittlement of the reactor walls and is dangerous to humans. Concrete and water shielding can reduce the exposure but there is still radiation. The reactor vessel does become radioactive but only for a shorter span of 50-100 years but would still have to be disposed of. News articles on the hopes of the pollution free power of fusion tend to over hype it as the ultimate solution to a desperate world. Maybe it will be replaced with antimatter power in the future.

    1. The effects of directly adding heat to our environment is a subject that not many people talk about right now, since it’s so massively overshadowed by greenhouse effect global warming, but yea absolutely it’s a real factor and something that fusion power could potentially make worse. That’s one of the big advantages to sun-driven renewables, hydro, solar, wind. They’re just repurposing energy that’s already coming from the sun. Not adding any.

      Ultimately at the end of the day, we need to be way more efficient with our power. The amount of energy we expend on moving thousands of tons of steel from one side of a city to another twice a day, for example, is absolutely ludicrous.

      1. In one of Asimov’s Foundation books (maybe one of the sequels after his death), Trantor is overheated, and the polar regions are the richest, charging $$ for dumping heat there… far fetched but interesting

        1. That was also a topic in Arthur C Clarke’s novels, he proposed giant radiators to,well, radiate the heat away from atmosphere to space. Might not be a problem today as we only produce maybe 0.0001% of energy received daily from the sun, but when we throw some unlimited cheap energy source at the world’s problems, then who knows…

      2. “That’s one of the big advantages to sun-driven renewables, hydro, solar, wind. They’re just repurposing energy that’s already coming from the sun. Not adding any.”

        No, this isn’t true. For instance, for solar, the albedo of a solar panel is below Earth’s average, meaning they absorb more of the Sun’s energy than average. There are studies on the radiative forcing effects of large solar panel fields.

        It’s more complicated for hydro: if you just imagine a dam on a river that creates a reservoir, that reservoir also has a much lower albedo than Earth’s average, so again, you’re absorbing more solar energy than the Earth would normally.

        But that’s a surface area thing, so it’s not entirely fundamental – but wind/hydro aren’t directly a “it comes from the Sun” thing anyway, so it’s harder to talk about.

        But… in the end, it doesn’t really matter. Atmospheric effects on Earth’s energy imbalance are like, orders of magnitude higher.

        1. Well no, it’s not fully negligible… But the effects are orders of magnitude lower than directly adding heat through fusion, which is orders of magnitude lower than atmospheric effects.

          We’d need to build a LOT of solar panels to have a significant effect on global climate, rather than just local conditions. Which again just reinforces the idea of needing to be more energy efficient in the first place.

          I could have been more specific I suppose.

          1. “But the effects are orders of magnitude lower than directly adding heat through fusion”

            Um… no…? It depends somewhat on where you’re installing the panels, because it depends on *what* albedo you’re replacing, but it’s not small. If you build the solar panels in, say, the desert, where the average albedo is 0.4, the typical albedo of a panel is around 0.05. Assuming 20% of that is going to electricity, that still generates 0.15 W heat/W elect. If you *directly* compare power in/out, it’s obviously 0.35W/W.

            Obviously any amount of power generated by fusion is “new”, and any efficiency puts it over 1W/W, but that’s not “orders of magnitude.”

            “We’d need to build a LOT of solar panels to have a significant effect on global climate,”

            This isn’t any different than fusion power, though? Albedo effects dominate by absolute tons.

          2. @Pat

            “It depends somewhat on where you’re installing the panels”
            Exactly. It depends where. So your worst case scenario, the numbers aren’t great.
            But with a diverse portfolio of renewables, the direct impact drops off pretty quickly. The additional direct heating of building a hydro plant isn’t anything close to your numbers. Same with wind. Or tidal, if they ever figure out how to use that effectively.

            Whats the albedo of a solar panel compared to a tarred roof? Or a asphalt parking lot? (elevated, of course.)

          3. To put things in context:

            Earth receives ~2E17 W from the Sun. Earth’s energy imbalance right now is approximately 460E12 W. Human energy production is around 20E12 W right now. In other words, if you assume every bit of energy humans produce is going into heat and it dropped to *zero* it’d reduce the imbalance by under 5%.

            It’s just not significant, and if it ever *did* become significant, we’d be so technologically advanced and flush with power that we could deal with it easily.

          4. “Whats the albedo of a solar panel compared to a tarred roof? Or a asphalt parking lot?”

            The choice isn’t “tarred roof + fusion” or “solar panel.” You’d get a *higher* benefit from “fusion + reduce human albedo effects” since in fact you can go significantly *above* ‘average Earth albedo’.

            Note that I’m not saying “solar panels bad” or “fusion awesome.” I’m just saying that using energy that ultimately comes from the Sun is not in any way a factor in choosing an energy source to combat climate change.

          5. @Pat

            I didn’t just pull this out my butt, it’s a studied topic that just doesn’t get much attention because it’s no where near as immediate a problem as GHG warming, and seeing as we can’t even get people to care about that, whats the point?

            This paper, for instance, even specifically brings up your point about the effects of solar panels on albedo.

            “We have shown that thermal effects from human energy consumption will play an
            increasingly significant role in global temperature forcing in the future. Consequently it is important to discriminate between renewable energy sources that inject heat into Earth’s climate system (geothermal energy), those that rely on Earth’s dissipative systems (wind, wave, tidal energy), and those that may potentially remove heat energy (suitably chosen solar technology, OTEC, and perhaps other future technologies).”


            For more info you can look up Anthropogenic Direct Heat Emissions

          6. Yeah, I know the term. And again: it doesn’t matter. It’s entirely academic until the atmospheric portion is handled, and once that is, everything else is easy.

            Heat management is just a ton easier than carbon. Worrying about anything other than carbon is pointless. It’s like speeding towards a cliff and worrying about dinner.

    2. Maybe it will be replaced with antimatter power in the future.
      Maybe in a couple millennia but not until we have a much better grasp on physics. Yes, the amount of energy produced by a matter-antimatter reaction is amazing but the amount of energy needed to produce antimatter is even more incredible. The best use of antimatter seems to be as a lightweight reaction medium for accelerating a ship to a relativistic velocity. However, it’s so dangerous that stockpiling it your own star system is foolish.

      1. Antimatter is kind of like hydrogen fuel cells—it is a (ludicrously risky) energy transport medium disguised as an energy source. Antimatter would only really be useful in exotic edge cases, such as space flight where it produces outrageously high specific impulse and keeps weight down.

  3. The commercial forces investigating this have had their thought process contaminated by Big Utilities. They keep trying to build a huge fusion reactor monstrosities that will power a large city. This is a bad path to be on. Lots of failures that are expensive, and when it doesn’t work the whole thing has to be thrown away. A better way to go is to make a tiny fusion reactor that can power a single house. Once that is working, only then scale up to a larger reactor and then only deal with the problems of scaling up. But, oh, if the small unit works, then everybody will want to buy a reactor of their own to put in the backyard. Oh, the utility companies have been cut out. Hmmmmm….. I guess they will keep wasting money on huge monstrosities that don’t work. To get this to work, maybe try thinking small, and then scale up. But Big Utilities don’t like this path.

    1. Yeah those silly billies, trying to get it to work at all, at any scale—before they start miniaturizing. Why don’t they just skip that part of the development process and go straight for an absurdly difficult micro fusion reactor off the bat?

      1. It might not necessarily be harder to make a smaller-scale reactor: the problem is that before you can *run* a fusion reactor, there’s development and research that you Really Need a larger-scale reactor for (breeder blankets, plasma heating, etc.).

        A lot of the problem with the “fusion is 20 years away!” thing are people talking about different stages in the development process. There’s been a *ton* done on the plasma front, and that’s what gets all the media attention, but there’s a *ton* more logistics involved.

    2. gyroradius matters. there is a physical limit how small you can miniaturize a fusion chamber. you can crunch it down to a spheromak or maybe even a polywell, but if you cant fit a beachball through it, what you have is a really fancy resistor.

      you cant scale everything down like semiconductors.

      1. The reason for the ultra-large size tokamaks actually comes from the magnetic field: ITER was designed before HTS ribbons became common enough that they could be used for high-field uniform magnets, so it’s still built with niobium-tin (Type 1) superconductors.

        You definitely could scale it down using HTS (and that’s what most of the compact reactor designs are doing) but the entire reason for sticking with the low-field design is that it just that it reduces risk. This might seem insane when talking about a project that’s seemingly years behind schedule, but the point is that from a plasma standpoint there are less unknowns, which lets you be confident you can directly do research on stuff like cooling/breeder blanket/divertor design. Stuff that’s all specific to an actual (significant) fusion burning plasma regime.

  4. I wonder what the efficiency would be if you just dropped H-bombs into a gigantic reinforced underground chamber lined with a water jacket. Would it break even? Maintenance might become an issue

  5. It’s already been what, two years? since the Lawrence Livermore lab achieved better than Q on a fusion reaction. The only remaining obstacles are technical. Asking whether we’ll ever have practical fusion power now is like someone in the 30’s asking whether we could ever get to the moon. Fusion is still “impossible” in 2024 in the same sense that a moon mission was impossible in 1930. The fact is that it’s totally impossible, in the sense of doing it with the technology at hand now, and totally possible in that the only obstacles remaining are technical hurdles. And if I’m wrong, there’s a giant fusion reactor in the sky above our heads whose radiated energy has already satisfied the bulk of our historic energy needs, and if we need to gather that more efficiently in the interim to knocking down the remaining technical hurdles toward our own local reactors, we can do that too.

    1. Yeah, no, the LLNL NIF result isn’t relevant to fusion power generation other than for testing ignition simulations. ICF isn’t a power generation scheme, it’s intended to study fusion in an ignition regime, and if you can’t think of why that’s important outside of power generation just ask yourself what else the Department of Energy works on.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.