Hacker News Reader

Comment by eatonphil | Hacker News Reader

The article doesn't seem to mention why exactly but I'm guessing it's some combination of 1) visible climate change (extreme heat, wildfires, drought, etc.) so people want better energy sources and 2) the mayhem Russia started seeing as it is one of the bigger energy providers.

Or has there been anything else?

View on HN

Actual advances in fusion and related energy generation tech. For example, at least one startup, Helion, is developing a way to generate electricity directly from the fusion reaction and magnetic field, instead of indirectly by heating water into steam that then turns a turbine generator. The increased efficiency from that might enable it produce net electricity. And quite frankly, it's way past time that we advanced beyond converting energy -> heat -> motion -> electricity, and shortened that cycle to energy -> electricity.

Magnetohydrodynamic generators were invented for coal plants 70 years or so ago but turbines are more efficient.

I thought they were ultimately limited by problems of corrosion at the electrodes?

Helion's scheme to turn fusion heat into pushing E&M fields back through their magnet to generate electricity is pretty different from a standard MHD generator.

And it's only possible because the reactor produces a series of small explosions of alpha particles.

The taste of immense profit is in the air. In the minds of potential investors, fusion is no longer a "lol maybe in 50 years" technology. Advances in magnetic modeling and magnet technology have resulted in cheaper, easier-to-develop alternatives to the classic Tokamak. Timelines are now believably "this decade" for demonstration reactors.

Energy production/harvesting is a foundational element of our civilization. Literally all of the comforts of modern society require the expenditure of some form of energy. Any entity that develops an unlimited, clean and cheap source of energy essentially has access to a money printer.

If nuclear fusion were ever developed, I imagine the government would seek to seize its patent protection under the Invention Secrecy Act.

The first part, the taste of immense profit, sure. But the idea that people are thinking of profits on the timescale of "this decade" is almost laughable, given the situation we are already in with climate change.

A much more rational explanation would be "they are trying to grab as much cash as they can right now, for as long as people believe that this is a thing that we should be doing, and may be possible", regardless of whether it's a good idea (it isn't), or whether it would work (doesn't look like it). But sure, maybe they are completely sincere and really believe in it all, who knows?

But the issue isn't just that "all of the comforts of modern society require the expenditure of some form of energy" it's also the case that "exponentially increasing energy expenditure is just accelerating the rate at which we are asset stripping and polluting the earth" and the latter continues to be true even if we stop emitting CO2.

I think it should now be apparent to all that comfort-seeking and suicide are essentially the same process (ask any addict), at least for the definitions of "comfort" that our current political and economic system are organised around. Of course, quite a lot of people still do not experience a lot of "comfort" even with the practically limitless energy that we have from fossil fuels.

Edit: sorry for the ramble, but to summarize, I think you give people too much credit to imagine they can design such a complex money printer with a 10 year plan, and if they could plan to that timescale, they are probably going to see that society so far in to collapse by then, that even with infinite energy, there will not be much value in the money that it prints.

> In the minds of potential investors, fusion is no longer a "lol maybe in 50 years" technology.

No. Potential investors are used to extremely skewed investments results for startups. Most startups fail, but some of those that succeed, succeed spectacularly.

A successful fusion startup is almost guaranteed to have a spectacular success, something that would make Facebook or Uber or Tesla look like small potatoes. Given that, a VC is ok with investing in it even if they think the likelihood of success is only 1%. They would not if they think that likelihood is only 0.01%, because at that level estimates are not reliable (who's to say it's 0.01% and not 0.00001% ?). But if there's a reasonable argument that P(success) ~ 1%, then it's a no-brainer to invest, because even a mistake of 2 orders of magnitude would still result in a positive return.

For the rest of us though, the "fusion is 50 years in the future" can simply be replaced with "there's a 99% chance fusion will not be done in the next 50 years". Or, what the heck, just leave it at "fusion is 50 years in the future" and you'll be much more right than wrong.

The cost effectiveness of fusion is completely up in the air. People speak about fusion as if it somehow is just going to produce infinite power at no cost.

> Any entity that develops an unlimited, clean and cheap source of energy essentially has access to a money printer.

Well, one thing that should be clear to anyone watching the space is that fusion will be neither unlimited, clean nor cheap. At the very least, not with any approach being looked at today.

First of all, being based on state-of-the-art science and tech, it's clear that it cannot be cheap for another 20-30 years even if it was working today.

Secondly, fusion reactors will produce large quantities of radioactive waste, probably larger than fission reactors, because everything that comes in close proximity to the reactor will be bombarded by 30+ times as many neutrons than in a fission reactor (or 5x-20x as many for DD fusion), and much higher energy neutrons at that, making it brittle and radioactive. Most of the reactor components are expected to need changing every 2-4 years, by which time they will be radioactive waste. Not to mention, fusion plants will work with large amounts of tritium, which is notoriously hard to contain, making tritium leaks all but guaranteed. So, definitely not clean.

Thirdly, fusion reactors require large amounts of tritium, which is virtually non-existent on Earth and must be created in fission power plants. The neutrons from the fusion reaction can theoretically also be used to recycle tritium (when hitting the lithium blanket), but you would have to recapture 100% of the tritium produced this way to recreate fuel, which is not possible. So, to run a fusion reactor you will be limited by tritium availability, and that will mean you also need a small fission reactor, and that in turn requires uranium - so, not really that unlimited.

Not to mention, even Deuterium is not actually that easy to obtain, as you need to either get it from fossil fuels (the way it is mostly obtained today) or from water hydrolysis, which is another waste of your produced energy.

Overall, fusion seems extremely unlikely to be an economically viable source of energy in the coming decades. Every new reactor will require massive investments, it will waste a good portion of its output to keep itself running (powering the superconducting magnets, pumping coolant, void pumping the reactors, water hydrolisis, lithium processing to extract the tritium etc). It will require highly trained engineers to design, build and maintain. It will have a high risk of catastrophic damage to the reactor itself, easily vaporizing much of the investment in an instant if the magnetic containment of the plasma fails, for example, or melting down spectacularly if the cooling system fails. Virtually all components of the reactor, including the high tech magnets, will require constant replacement as they become brittle from the neutron bombardment - and all the old parts will require expensive radioactive storage for at least a few decades to centuries.

> Not to mention, even Deuterium is not actually that easy to obtain, as you need to either get it from fossil fuels (the way it is mostly obtained today) or from water hydrolysis, which is another waste of your produced energy.

This is really not an honest critique. The cost of obtaining deuterium is quite low compared to all the other costs of building and operating a fusion reactor.

REBCO high-temperature superconductors have potentially changed the game. They can support stronger fields which means smaller devices for same confinement. Commercial production seems to only started in the last decade.

MIT's research, spun out as CFS, may be prompting other startups.

Is there any fundamental limit we’re hitting on magnetic fields or might we see another 10-100x increase in field strength in the future?

Nobody knows, is the short and boring answer. We don't have a sufficiently detailed understanding of what drives critical field strength in these materials (otherwise we'd have room temperature superconductors or have ruled out their existence already).

It would be shocking if HTS manufacturing research did anything other than accelerate. If there are fundamental physical limits to run in to then they are very far away. This is deep in the engineering limited area. How much abuse from strain and radiation can your HTS handle? The better your answer at a cheaper cost, the smaller and more powerful of a machine you enable. Humans know how to make pretty strong steel, so the superstructure won't be a limiting factor for a while if ever. Making HTS tapes and divertors that can handle what we ask of them are the material challenges. There is room to be clever with physics to lessen the divertor problem.

Actually, I think the strength of modern steels is a limiting factor for fusion magnets. Plasma "beta"s are only a few %, so for a few atmospheres of plasma pressure your magnetic cage needs to be like a pressure vessel that contains hundreds of atmospheres. And with large holes poked through it for access. If you look at CFS's magnet test from last summer, there's a huge amount of steel.

60% of the mass of the ARC reactor is the steel support structure to resist JxB forces. A magnetic field 10-100x higher would exert such strong forces (magnetic pressure scales as B^2) that no material could resist them. So: strength of materials.

Some breakthroughs in plasma modelling, too. https://www.sciencealert.com/physics-breakthrough-as-ai-succ...

Interesting. By that link, it seems they was able to stabilizing 'droplets' where two plasmas co-existed simultaneously inside Tokamak.

I wonder if they thought if it would be possible to accelerate such plasma droplets in opposite directions for to take Helion's aproach [1] if needed.

[1] https://news.ycombinator.com/item?id=29119180

The really big deal in the last decade has been better magnets -- many of the "novel" approaches attracting recent funding are probably doomed to fail, but the MIT-associated consortia that simply aspire to building better tokamaks with modern magnets look encouraging.

A much more reasonable first guess would be "the desire to increase the next-quarter earnings of industries which make political contributions." Which is the main driving force of all public investment. If politicians can take advantage of climate chaos and put it down as part of their "climate bullshit, whatever" policies then they can get a larger pay day, so why not?

Here in Korea for example, they are building out the worlds most dangerous fission reactors because domestic heavy industry, concrete companies, and so on are extracting very good earnings from that. Meanwhile, Korea is the leader in offshore wind (but only to export to other countries, and it's all manufactured in Thailand) presumably the lack of interest in the Korean government in stimulating domestic demand, and the cheap labour in Thailand are factors, along with oil prices, in why it is so cost-effective.

A belief in the free market would lead us to expect that eventually the green options would become just as corrupt and be able to hold the world hostage and capture policymakers, but it does not seem to have happened yet, maybe there is just a lead time for industries to develop their capacities for political corruption?

There seems to be a lot of innovative ideas and a lot of them seem to be quite doable, which is interesting for investors. One I really like is FirstLightFusion [1]. They are using a ballistic system to shoot at a small fuel cube, which creates a fusion reaction and then use the generated heat to power a turbine. There is a nice behind the scene video with some interview from FullyCharged [2].

[1] https://firstlightfusion.com/ [2] https://www.youtube.com/watch?v=M1RsHQCMRTw

Inertial confinement fusion is orders of magnitude less likely to ever be economically viable than magnetic confinement. The targets (and projectiles in this case) will require such high precision to be able to achieve fusion in the moment of impact that it is virtually impossible to imagine this could ever be done economically.

Further, you will need huge quantities of such targets and projectiles, as the power plant will have to destroy them at a rate of one per second or so, in constant operation. So even for a single day, you will require 86,400 targets and as many projectiles - each being a marvel of precision engineering. Also, there is a good chance that you wouldnt even be able to recycle the material from spent targets to make new ones, as they will become radioactive from being in contact with the fusing plasma.

Essentially an ICF plant would actually be running on the world's most expensive fuel, and consuming it at an extraordinary rate.

MIT's SPARC figured out how to manufacture high temperature superconductor magnets. That shrinks the size of the magnets and power consumption.

Those things might explain some rising investment in alternative energy in general, but I don't think either them would be specific to fusion.

Better high temp super conductor lowering barrier to entry maybe? (im pretty ignorant of this field)

missing 3) tons of public funding lead to recent advances that the private sector is ready to take over and profit.

capitalism innovation as usual.

Reply on news.ycombinator.com