Hacker News Clone new | comments | show | ask | jobs | submit | github repologin
Fusion power is getting closer (www.economist.com)
32 points by doener 2 hours ago | hide | past | web | 83 comments | favorite





I'm happy to see lots of investment in fusion. As a fission guy, I know that plenty of challenges remain even after you get your first net energy out, such as getting it out economically. In fission, there were more than 20 years between these milestones (first chain reaction in 1942 from CP-1, first competitive power in 1965 from Oyster Creek).

As such, some skepticism in people's timelines is always warranted. I'm reminded of the voodoo fusion article. https://vixra.org/pdf/1812.0382v1.pdf


Fusion is cool and all but it will not save us from anything. We already have enough tech to make emission free and cheap electricity.

"All" we need is political action to build more clean power plants (nuclear, solar …) and continuation of research to electrify more sectors.

Fusion will not solve any of those important issues. It will only be the cherry on the cake if it become deployed globally anytime in the future. But we are far from it and we shouldn’t be that excited or hope for it to make a better world.

I very much see fusion as going to the moon : it’s not going to solve any short time issue but it can be a milestone in human advancement if we manage to survive until it becomes our primary source of energy.


> "All" we need is political action to build more clean power plants

We’re building solar about as quickly as we can, and without cannibalising EV battery supply chains are also maxed out on batteries. Fission power remains uncompetitive with natural gas; you’d have to convince Western consumers to bear higher prices for reduced emissions which nobody is willing to do.


How do I square this with claims that the US heavily subsidizes fossil fuels? Are those claims false or am I misunderstanding how the market works?

> How do I square this with claims that the US heavily subsidizes fossil fuels?

By noting it’s irrelevant. Solar production isn’t capital constrained. Certainly not in the West.


That really depends on what you count as a subsidy. The federal government isn't really making direct payments to fossil fuel companies. Some of them are allowed to lease federal lands at what might be considered below-market prices. The US military helps to provide free security for some oil exporting countries. Fossil fuel producers and customers are allowed to emit pollutants without paying for cleanup or remediation costs. Are those subsidies? Depends on your perspective.

Exactly this. If we drop the cost per kWh by half, the market will respond. The market doesn’t ultimately care about the planet which is why we need govt but bringing the price per kWh with clean fusion kills two birds with one stone.

That’s why I talk about political action. Public is exactly here to act when the market doesn’t care about something important.

> why I talk about political action. Public is exactly here to act when the market doesn’t care about something important

Which is why I mentioned consumers. Voters have made this incredibly clear every time polled: we are not wiling to pay more to reduce emissions.


Can you mention a recent time they were polled and make this incredibly clear?

> We already have enough tech to make emission free and cheap electricity.

As much as I am a proponent of green energy, this claim is outlandish. If we had started building more nuclear reactors 10 years ago, sure, but by themselves wind and solar aren’t quite sufficient enough for places like the US to fully replace fossil fuels. There are limitations for both, namely the need for space and acceptable weather conditions. Solar panel energy yield still has potential to unlock, as well. Last update I’d seen about panels said they still only hover around 25% energy yield.


Fission just isn't a competitive source of energy economically. Nobody is willing to pay 10x as much for energy just to get nuclear. Seems clear to me that the actual solution we need are cheap batteries so we can scale renewables

Fusion is irrelevant as a practical energy source here on earth. The simple reason is that we already have all the "fusion energy" we'll ever need from that big fusion reactor in the sky... modern photovoltaics convert this energy to electricity at close to the same efficiency that you could convert the heat from a controlled fusion reactor to electricity. OK, yeah, photovoltaics still don't work at night, but there's a battery revolution underway that is solving that problem, and the two, photovoltaics + batteries, are already vastly cheaper than fusion can ever hope to be.

Fusion will be important for the future of humanity, especially in deep space, so we should keep researching and developing it. But it's not relevant to the climate crisis... the already existing technologies have that covered, we only need to deploy them faster.


While I admire your optimism, this is not realistic. The earth is big and not all parts get the same amount or intensity of sunlight. Nordic countries cannot rely on Solar with batteries, for example.

> it's not relevant to the climate crisis... the already existing technologies have that covered, we only need to deploy them faster

Going all in on solar and batteries either requires nuclear energy, thereby raising power costs, or natural gas, which means we can’t make our 2050 goals. We simply can’t manufacture the panels and batteries fast enough, the latter especially if we also want to decarbonise transportation.

Fusion won’t come fast enough for the climate crisis. But it may be relevant for complete decarbonisation, e.g. replacing the natural gas plants we’re spinning up as quickly as we’re deploying solar.


No, b/c we really need it as a weapon. Fusion lasers will finally provide the death ray of sci-fi. If you think "practical energy source" is what fusion research is really about then you've been bamboozled.

> Fusion lasers will finally provide the death ray of sci-fi

…why? If you want massive lasers you use a nuke to power it [1][2].

> If you think "practical energy source" is what fusion research is really about

Outside the NIF, which is openly about stockpile stewardship, yeah. That’s what it’s about. Tokomaks aren’t militarily useful.

[1] https://en.m.wikipedia.org/wiki/Nuclear_pumped_laser

[2] https://en.m.wikipedia.org/wiki/Project_Excalibur




It is getting late to the party.

Clean enough energy alternatives are being deployed everywhere.

And we might be already running out of time, the climate system could had surpassed a point of no return by now, we keep adding fossil carbon to the system and if, or by when, this became practical enough to start deploying power plants based on it will be too late for a meaningful change.


"no really" is about right. even if you have managed to produce sustainable fusion with a good bit of output energy (a very big if), you still have the problem of producing useable electricity from it, which as I understand it is one of the things that ITER was supposed to prototype.

Steam.

The actual problem is reaction vessel wall ablation.


This is one of the reasons General Fusion, mentioned in the article, uses a liquid metal reaction vessel, spinning to create a cavity. (I think a mixture of molten lead and lithium?)

Yes, incredible idea. I really hope it works out.

Steam is uneconomical versus solar power. If all the fusion is doing is heating water it will always lose on cost even if the fusion was free.

c.f. coal. The bulk of the cost of a coal power plant is the steam turbine etc. And coal is not economical.


My city has a steam/hot water distribution network. No need to convert to electricity for quite a lot of residential energy usage.

Well, we could always just set up solar panels and use it to capture output from the massive fusion reaction 8.5 light minutes away from us. Does that count as a successful utilization of fusion?

> Does that count as a successful utilization of fusion?

In the way a fire is a steam engine.


It’s been getting closer for what, two decades now?

> It’s been getting closer for what, two decades now?

This is low effort. Are you arguing we haven’t made material progress in the last twenty years?

Material science advances in magnets alone should raise your curiosity to not have to wait for a packaged consumer product to notice advancement.


Tbf, the title is kind of low effort too. Of course fusion is getting closer. We all know that at least some people are working on it, the default assumption is that they make some progress. This has been the state of fusion for the last couple decades.

Arguably there has been a time when fusion got further away (compared to predictions) because of reduced interest and funding. But I'd agree that for the last two decades you could print the headline "fusion is getting closer" at any time, and it would always be true.

Maybe that's just me ranting about how titles have shifted from trying to inform you and distill the essence of a news article to just trying to get you to read the article. It's really a comment on journalism rather than fusion.


> We all know that at least some people are working on it, the default assumption is that they make some progress

This is a terrible assumption when it comes to basic research.

> that for the last two decades you could print the headline "fusion is getting closer" at any time, and it would always be true

What are you basing this on?There have been a few discrete steps forward. In between, ideas and experiments.

> really a comment on journalism rather than fusion

Can you quote the line from the article that renders the title misleading?


> This is a terrible assumption when it comes to basic research.

Even trying out things that don't work is progress, since it narrows the space of possible solutions. The only ways to really set back progress would be by getting others to believe something that's untrue, for example by faking research.

You could manage to not advance the field by not publishing a finding and leaving the field. You can slow progress for example by popularizing a path that doesn't lead anywhere, discouraging people from going down the right path, or just reducing popularity or funding of the field. But slowing progress doesn't erase progress. If you do genuine work and at least talk about it with colleagues on a conference that's advancing the field.

> What are you basing this on?

The classic graph everyone uses would be 1, which shows consistent progress over ~40 years. It's a bit out of date now, but I assume the trend has held since then.

> Can you quote the line from the article that renders the title misleading?

I didn't call it misleading, I called it lazy and uninformative. A (mild) form of clickbait, because it merely hints at information without providing any. My English teacher would have deducted points for that title.

1: https://www.researchgate.net/figure/Progress-to-Date-in-Achi...


> Even trying out things that don't work is progress, since it narrows the space of possible solutions

Maybe, maybe not. You may be searching an infinite domain, or the wrong domain entirely.

> If you do genuine work and at least talk about it with colleagues on a conference that's advancing the field

Sure. But it’s not advancing a technology.

> classic graph everyone uses would be 1, which shows consistent progress over ~40 years

Fair enough. I’d argue that’s constant progress in one of the problems of fusion power, namely, ignition. The article’s point is the other problems are either being solved or showing similar progress. That wasn’t true until recently.

> it merely hints at information without providing any

It's debunking the assumption at the top of this thread. Especially with the "no really," it's arguing that the ground truth has changed.


It’s been 30 years away for half a century. Then it was 20 years away for the last few decades. Now it is 10 years away. We’re making progress!

since the late 1950s

Very true. The 1958 Atoms for Peace conference prominently featured big fusion exhibits expecting breakthroughs soon.

Time for Musk to announce fusion powered self driving cars for next sear

If you mean fusion powered from the sun by the means of solar panels installed on roofs that charge your car, he's already achieved that.

Far cheaper and more practical.


Honestly surprised he's not bought/started a company to work on it so he can proudly proclame they've cracked it and it's coming next year on repeat.

> It’s been getting closer for what, two decades now?

No reason for pointless cynicism, this is also against the guidelines. Fusion power is like childbirth - yelling at 9 months that it has been 9 months, and is therefore never arriving, would be delusional.


Weird analogy but to work with it: if someone has been shouting that they're about to give birth any moment now since the 1950s, forgive me for being a bit skeptical when they say "but no really I mean it this time".

Comparing the very optimistic deadline given by Commonwealth Fusion, concerning a prototype that has not even began construction versus the latest postponing of the first firing of ITER is kind of meaningless.

As underlined in the article, ITER is already 9 years behind schedule. Not because they're incompetent or underfunded, but because fusion is hard. I don't expect a private company with much less funding to do any better.

That said, I'd like to be wrong and see one of these companies crack commercial fusion in their basement.


The Economist always phrases articles to imply that private investment is a magical elixir that will make problems go away.

The fundamental thing behind SPARC is (iiuc) to do this far faster by being drastically smaller. Iter was designed a long time ago based on materials available at the time. SPARC can be much smaller with much stronger magnets.

The argument from SPARC wasn't that iter was incompetent but that being so vast and expensive requires multiple nation states cooperating which is inherently slow. Slow things being more risks as thing a change too (9 years is probably 3 governments here). Making it smaller makes it cheaper, a d cheaper means a smaller collaboration (or single company).


Fusion power has been getting closer — no really — for a long time.

Now, don’t get me wrong, I’ll be ecstatic when / if it arrives, especially if it does so with sufficient lead time to be relevant within my lifetime, but no number of “we’ve halved the distance to the wall again” articles will get me excited (or us there).


A comparison to Zeno's paradox isn't quite apt, because fusion power can (in this metaphor) go very far beyond the wall.

I think it's better to think of it this way: are you a lay person not in the field? Then most of this work doesn't matter to you. But you should keep funding it.


The wall, in this case, is power positive sustained (and contained) reaction. Fusion power doesn’t exist and can’t do a damned thing until that wall is reached. Once it’s reached I’ll admit we’ve no practical idea where the next wall is going to be.

And yes, absolutely, fund baby fund, keep making those Zeno steps… just don’t write another article promising me that fusion / Jetson cars / AGI / Unobtanium is nearly here.


But fusion doesn't produce any waste does it?

It doesn't have reaction byproducts, if you don't count the reactor itself, which gets lots of neutrino bombardment and ends up glowing scrap metal.

The actual intended products of the fusion reactions are generally not radioactive and are rather harmless. But most proposed fusion reactions produce neutrons. Neutrons, unfortunately, can hit other nuclei and “activate” them, transmuting them into other isotopes that can be quite nasty. So fusion is far from perfect.

Technically the helium produced is the waste product, but fortunately we have tons of uses for helium.


looks like that article didn't read this article https://news.ycombinator.com/item?id=42221756

2 billion dollars over 5 years? That's peanuts! If anything I would increase spending to a billion dollars a year. Would still dedicate most of the budget to accelerating the energy transition, which is trillion of dollars worth of stuff.

fusion is the ultimate clean energy source. we shouldn't give up on it just because it's a difficult problem

lovely, we only need warp drive now :)

If we're talking Star Trek humanity cosplay, our best bet would be to sort out communicators first. We're almost there. It already works reliably, just needs to be made to respect privacy and be completely free.

It's always been close. But now, finally, I think it's really close. As in, I think I may still be alive when the first fusion power comes into the grid, even if I probably won't be alive to see the full dream of infinite cheap energy come true.

Its interesting to think about how our economies will change if the price of energy drops by one or two orders of magnitude.

Desalination could then be used for agriculture and industry. Coastlines could become more valuable.

We could make huge amounts of aluminum for almost nothing. We'd probably end up with aluminum packaging for everything as it's much easier to recycle than plastic if you don't have to worry about electricity.


> won't be alive to see the full dream of infinite cheap energy come true.

You won't. Sorry for being cynical, but the true cost of energy is the capital invested in the equipment to harvest it. A fusion reactor would be at the minimum, as expensive as a fission one, which means all that capital will have to be amortised over the years. This doesn't even take into account the cost of extracting deuterium from sea water or producing helium 3 and tritium from lithium. Fusion doesn't even solve the problem of nuclear waste, since it will create radioactive isotopes by bombarding components of the reactor with neutrons.

You want cheaper, cleaner energy ? Solar and wind with battery storage is the best option, and it's already here, no need need to wait for fusion magic that may or may not happen in our lifetimes.


> A fusion reactor would be at the minimum, as expensive as a fission one

What are you basing that on? Many of the fusion designs being talked about are quite small.


So does fission reactors. Modern French nuclear subs are powered by 48 MW reactors. They're still expensive to build and maintain.

I don't see why would it be cheap.

Nuclear is mostly capital costs, not the fuel. Fusion doesn't look any cheaper to build. Like any other bit of complicated infrastructure, it'll have a limited lifetime.

So, power cost is ((capital_costs + running_costs) / amount_produced) * profit_margin

While I'm very much in favor of research, I don't understand the expectation of that it'll make power cheap. I'm pretty sure nothing will beat solar and wind because they're cheap, simple and very amenable to mass production in ways that fusion and fission can never be.

At this point IMO neither fusion nor fission have a future as a main source of grid power. They're still worth researching, and will still have specialized uses in places like submarines and aircraft carriers.


To my understanding, the time limit of fission is mostly due to political/security reasons.

Of course keeping it indefinitely on will require huge maintenance costs but if the infrastructure is modular enough to be able to replace every elements, that may be doable.

With fission the structure of the building is part of the infrastructure’s security so it’s harder to change when it’s too damaged. Maybe I’m wrong but it seems like fusion doesn’t require huge cement walls that would take a lot of constraint and which would effectively limit the maximum lifetime of the building.


Helion says they want to make small, modular plants that are amenable to mass production, opening up the same scale economies that have benefited solar.

That's assuming their stuff even works at all, which I believe there's some reasons to doubt. Perhaps all we'll get working in the end is a honking huge tokamak and then we're exactly in the same situation as fission.

But let's suppose their stuff works as well as they imagine. IMO, not even then. Just look at their youtube videos. All sorts of big, complicated, and expensive looking magic. Huge capacitors, vacuum chambers, extremely high temperatures, etc.

Will that scale better than nuclear? Maybe. Will it scale better than a tech that can be pumped out by the million, shipped to normal people and put on any random roof? Highly doubt it.


Probably we will just burn the world with the free energy.

Even if that was literally done, it would still be a better home than Mars :) https://www.youtube.com/watch?v=5GjOuqAlFEk#t=6m45s

(not trying to make any kind of point, just a fun fact / video)


That video is full of widely discredited claims.

Probably? Oh ye of little faith.

What has changed to make you think it’s finally really close now? I found the short article a little light on details, although as a layman I have noticed people seeming cautiously excited about ITER and speaking about SPARC as if its near-term promises may be credible.

Are there other signals that you’d point to as influencing your shift to optimism?


There is a massive amount of investment into fusion startups right now, so much that it is difficult to keep track of all the really innovative approaches that are being tried, often with billions of $$$ backing them.

Lots of people assume that fusion energy will be cheap but this is not true for the tokamak designs, which have the highest chance of success. Helion's approach should be very cost-competitive but it has a high chance of failure.

> Desalination could then be used for agriculture and industry

This is likely going to happen anyway because of the rapidly falling costs of solar power. Although it certainly doesn't hurt to make energy cheaper yet again.


imagine if fusion research could consistently get 1% of the pentagon's accounting hole.

As nice as that would be, if we think about it logically it's not at all in the US governments interests for Fusion to work. Suddenly that massive pot of money from natural resources (e.g Oil) vanishes.

…vanishes, presumably, because the work that it used to do is now done by this fusion technology instead. Surely the US would like to be the nation to dominate that tech should the moment come, right?

Could sure shake up some of the US’ geopolitical relationships, in which the US spends an awful lot to maintain influence over the way oil gets made and sold. That pot of oil money sure creates its share of headaches, and probably means a lot more to smaller nations than it does to the US.


I guess the question would be that assuming the US was to dominate the technology, is that more profitable than a steady stream of revenue from oil exports?

Building fusion plants would be a massive money maker initially as everyone races to adopt it, but is it a long term money maker when no doubt other nations will be trying to cash in on it themselves.


The US doesn’t have nationalized oil companies. The US government doesn’t have skin in the game here. In fact, fusion would make a lot of military stuff cheaper, so if anything they have incentive to adopt it.

> US government doesn’t have skin in the game here

It absolutely does, just in the opposite direction from OP’s hypothesis. Washington likes cheap oil. It wants high production levels from its producers to keep prices low; producers want high and stable prices.


It'd get nowhere. The problem isn't money. One could say there's even too much money thrown at fusion research that could have better use elsewhere. There are just too many incredibly hard problems to solve.

That would solve about ~40% of global climate emissions, and would single-handedly do serious damage to politician's attempt to use climate change as a cattle prod.

Watch them say, "Yay, we've solved climate change" with gritted teeth as they figure out the next useful crisis.


The Economist as scientific harbinger of the fuuuuture!

Even if good net-energy happens, even if they solve neutron flux irradiating the reactor vessel, even if they can sustain power generation for commercial periods, if they can safely operate the superhot plasma

... it comes down to, ECONOMIST, the cost. As someone that hammers fission power for being currently infeasibly expensive, fusion has a long road to economic viability. I suppose the public will at least not NIMBY it to the same degree as a fission reactor




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: