The atomic age⁠ at last? Manipulating bits to manipulate atoms

February 8, 2020 1:11 AM

After decades of decline, the U.S. national fusion lab seeks a rebirth - "A visionary new leader aims to expand and diversify the Princeton Plasma Physics Laboratory—and get back to building fusion reactors."

also btw...

• Fusion Startups Step In to Realize Decades-Old Clean Power Dream - "Governments have spent billions of dollars studying the emissions-free energy source. Now, private ventures are building smaller, faster, cheaper reactors." (WSJ; tweet)

Bill Gates and Jeff Bezos have bankrolled fusion startups through funds aimed at transforming energy. Few investors are traditional venture capitalists. Since payback could take more than a decade, investments today are “almost philanthropy,” says Mr. Cowley at Princeton. “Investors want to be part of something that will change the world.”

• Calling All Billionaires: Fund Fusion Now - "The reason to be optimistic is that there has been important progress. Much of it has come from innovation in other fields that provide enabling technology. For example, something called REBCO tape, which was originally developed for MRI machines, allows for stronger magnetic fields which in turn makes much smaller fusion spaces possible. Similarly, inertial confinement approaches to fusion have benefitted greatly from the massive advances in laser technology. The Physics Nobel Prize in 2018 was in fact awarded for just this progress in laser intensity."

posted by kliuless (25 comments total)

I read an article a year or two ago about all the little Fusion start-ups and the incremental work they are doing. It compared them to the humongous international reactor that is being built in France that has taken decades to organize, plan, and build.
posted by Fukiyama at 7:46 AM on February 8, 2020

I remember writing a paper on fusion in high school on an Apple IIe, sigh.
posted by Heywood Mogroot III at 8:09 AM on February 8, 2020 [5 favorites]

Anyone else here got totally sucked into the Fleischmann and Pons cold fusion fiasco? For a week, before it was debunked, I walked around thinking we were hurtling toward a brave new future.
posted by umbú at 8:23 AM on February 8, 2020 [4 favorites]

My understanding is that most of the government fusion work is really focused on thermonuclear weapons research, with the fusion energy as a convenient PR. I’m excited that there is at least some work that is focused on energy production specifically. We need to be spending billions on every potential option for base load decarbonization, even if every option doesn’t pan out.
posted by rockindata at 8:32 AM on February 8, 2020 [3 favorites]

I'm not sure I was "totally" into it, but I certainly took the Pons and Fleischmann claims seriously. Had a professor who was rapidly trying to replicate the experiment. Disappointed it turned out to be false.

On the off chance someone reading this hasn't heard the old chestnut, "Fusion is the energy of the future, and always will be." I remember walking through a model of a Tokakmak reactor in the '70s at a museum that was touting it back then.

The call for billionaires to fund fusion research includes the classic "we can do this in the next ten years." I have for years viewed this timeline as a tell that the research is so far from being practical proponents entire plan is basically to hope for a series of incredible breakthroughs in the next five years and then figure that you could probably commercialize something in five more, right? (For comparison, it took 11 years to replace part of the Bay Bridge, and bridge technology is kind of well known.)

Same with the "since payback could take more than a decade" line on private investment. If you are handicapping pitches like this, always remember on this is fusion doesn't just need to get to "energy production", it needs to do better than renewables, which have become incredibly cheap and aren't going to be standing still during this period either.

I'm still in favor of research! It could pay off, and for self-interested reasons money going to scientists usually make me happy. But press coverage of billionaires that presumes they have unique insights on How to Get Things Done vs. the stodgy government scientists should be treated with the normal level of skepticism.
posted by mark k at 8:51 AM on February 8, 2020 [10 favorites]

That really only applies to inertial confinement fusion - shooting a whole bunch of lasers at a small pellet of deuterium/tritium to make a burst of fusion. Most fusion power research focuses on plasmas that are too low density to be relevant to nuclear weapons research.

The main problem is that the US government doesn't really provide fusion research any more funding than any other avenue of basic physics research. Which means we don't really have a fusion research program, we have a basic plasmas/astrophysics research program that sometimes chat about fusion and write about it on grant applications.

Most of the actual fusion projects, be it a big government project like ITER/the civilian side of the NIF or a smaller privately funded operations like General Fusion or Tri-alpha are basically Hail Mary passes that hope that something works out better than expected with no major complications.

A more thorough, careful approach would involve a lot of investment in doing more of the basic footwork: developing really good detectors and diagnostics (there's a lot going on in plasmas and electromagnetic fields aren't something you can measure directly from a distance), developing really good simulations, and the really expensive part, building a large number of experimental devices at different configurations and scales to verify and improve the simulations with real data, and see how different effects scale.

The reason fusion is always 20-30 years away because the people in the 70's were envisioning a research program of that sort of huge scale and we have never actually invested in anything of the sort. Yes, our understanding of plasma physics is better, but we also know more of the complications so it's still a massive scale research project to see if it's something we can do.

Of course the answer to "how do we do fusion?" might still then be "Well, you put about a solar mass's worth of hydrogen in a gravity well..."
posted by Zalzidrax at 9:12 AM on February 8, 2020 [10 favorites]

The excellent Omega Tau podcast has done a few episodes on plasma physics and fusion, including these:

249 – Plasma Physics
304 – The Past, Present and Future of Fusion Energy
312 – The Wendelstein 7-X Fusion Experiment.

Omega Tau episodes are generally quite long (sometimes several hours) but nevertheless compelling. Topics including physics, computing, lots of aviation, and space technology. About half the episodes are in German.
posted by neuron at 10:01 AM on February 8, 2020 [2 favorites]

The other problem with fusion, as compared to renewables, is that it adds heat to the system. Solar and wind repurpose energy that's already part of the system, but neclear, while "clean" in that it isn't producing greenhouse gasses, is creating energy out of materials that would otherwise be stable, so even if it's carbon neutral, it's not heat neutral.
posted by rikschell at 10:04 AM on February 8, 2020 [1 favorite]

Everything adds heat to the system. Thermodynamics, not just a good idea.

I remember a physicist working out that we’d ruin the ecosystem even with free, carbon- neutral energy if the economy kept growing, but it took several hundred years.
posted by clew at 10:28 AM on February 8, 2020 [1 favorite]

Fusion missed its chance. We needed the technology to have matured a few decades ago for it to matter. PV and wind are already the cheapest energy in many places around the world, and given expected learning curves, will be the cheapest almost everywhere. Given the table table climate change has given us for decarbonization, deploying existing tech at scale needs to be the focus, with research for new tech taking a back seat.
posted by gwint at 10:42 AM on February 8, 2020 [8 favorites]

> I remember a physicist working out that we’d ruin the ecosystem even with free, carbon- neutral energy if the economy kept growing, but it took several hundred years.

The 'Galactic-Scale Energy' post on Tom Murphy's Do The Math blog?
posted by genpfault at 11:03 AM on February 8, 2020 [1 favorite]

I think a different one, though that one is great too. What I remember was framed as a response to a specific economist’s claim that we had to have ? Assume? steady rate of growth in “the economy”.
posted by clew at 11:39 AM on February 8, 2020

I went to a talk from Professor Joseph V. Minervini from MIT back in November. His key point was that the minimum size of a fusion reactor depends upon the fourth power of the strength of the magnets.

So ITER was designed with the best superconducting magnets foreseen in 1995 - 11 Teslas. Superconducting magnets have become far stronger than anyone expected and are now pushing 30 Tesla.

The scaling is to the power of four, so going from 10 to 30 Tesla means your machine gets 80 times cheaper. Instead of ITER at US$40 billion, you could build a fusion machine for less than half a billion.

A $40 billion project takes an international partnership and twenty-five years. Spending $0.5 billion can be done by one company in ten. So it's possible that new magnets means new fusion can leap-frog over ITER... but PV and wind are now so cheap that no-one cares.
posted by happyinmotion at 11:51 AM on February 8, 2020 [10 favorites]

We might still really want fusion as part of the energy infrastructure, though, yesno?

Well into the first link, running through the depth of planning-research-appliedphysics-engineering-construction skill required all at once, there’s a comment that getting the tokamak running would plausibly improve chip-making. Huh.
posted by clew at 12:04 PM on February 8, 2020

Not seen link yet. Is the improvement in chip making through x-rays generated by the beams for advanced lithography? If so there's a newish European Initiative to generate them from Free Electron Lasers generated by the newer "Wakefield" mechanism, much smaller/cheaper. They talk about being able to put one in a Hospital (for imaging/other).
posted by aleph at 12:25 PM on February 8, 2020

Never mind. Finally got to link and it looks like they're talking about the "Cold Plasma" work and Semiconductor processing.
posted by aleph at 12:29 PM on February 8, 2020

The use case for fusion in a renewable energy infrastucture is for smoothing out the gaps in the variable power provided by solar/wind, instead of relying on combustion engine or large scale power storage. Fusion power, if you could get it to work, would be pretty much as simple as flip a switch and it's on. Of course a lot depends on how exactly you extract the heat produced by the fusion from the plasma, but it would likely be at least as quick to power on as a diesel generator if not faster. And that's what it would be replacing: battery farms and diesel generators, not the baseline solar and wind power. At least not unless or until we could make it really, really cheap.
posted by Zalzidrax at 1:23 PM on February 8, 2020

Not to derail, but high energy physicists and materials scientists: what ever happened to Plasma Gasification?
Using high energy plasma to not just incinerate, but molecularly disassemble things like hazardous materials, solid waste, etc., capturing the waste heat to power the system and repurposing the inert slag, sounded like a good technology. Where did that go?
posted by bartleby at 1:47 PM on February 8, 2020 [1 favorite]

Hey, we're just 10 years away(!), right?
posted by OHenryPacey at 2:28 PM on February 8, 2020

happyinmotion explained far better than I could why this time could be different. It isn't only that the new materials get you a much higher magnetic flux, they also enable easier construction methods and make net energy possible at scales much smaller than ITER.
posted by wierdo at 4:19 PM on February 8, 2020

There is indeed room for greater strength with these new materials as time passes and both manufacturing and design/assembly improves. It helps that REBCO is used in other industries, so the fusion people aren't the (only) ones pushing the technology forward.
posted by wierdo at 9:10 PM on February 8, 2020

The other problem with fusion, as compared to renewables, is that it adds heat to the system. Solar and wind repurpose energy that's already part of the system, but neclear, while "clean" in that it isn't producing greenhouse gasses, is creating energy out of materials that would otherwise be stable, so even if it's carbon neutral, it's not heat neutral.

Yes, but the magnitudes of the effects aren't remotely comparable. By burning enough carbon to generate 1 joule of heat, you're also releasing enough greenhouse gases into the atmosphere to cause several hundred joules of greenhouse warming.

It's true that switching to an energy source that "only" added that one joule to the system wouldn't be perfect (compared to solar or wind power), but it would still be literally orders of magnitude better than burning fossil fuels.
posted by teraflop at 10:40 PM on February 8, 2020 [3 favorites]

The solution has to be out there, and I worry that it won't be found in time.

I don't see the reason to view fusion as if it were a critical option to deal with the climate crisis. We really have the tech now. The big advantage of fusion IMHO is that conceptually and economically it would have the least disruption to our current system of high-capital-cost, centralized power production.
posted by mark k at 8:37 AM on February 9, 2020

My father's PhD involved plasma gasification back in the 60s, and from the few comments he dropped while I was growing up and afterwards he was getting consulted on that up until he retired a few years ago. Most of the times he mentioned it was in regard to waste remediation: destroying retired chemical and biological warheads, the worst of the Superfund site hazardous wastes, even some chemical manufacturing operations which had waste products too expensive to dispose of otherwise.

The thing is, it is really expensive (you have to heat up everything in a plasma jet until it, too, turns into plasma and all the molecules disassociate, then cool it back down so that it doesn't just form more hazardous material.) Far cheaper in most situations to just shove the contaminated stuff into barrels and toss it into a hole in the ground, or to just add cullet and melt it all down into an environmentally stable glass brick. Or, the chemical engineers favorite solution to waste, just dilute it to 'non-hazardous' levels and then dump it back in the environment (still the most common method of disposal).

And, of course, this sort of remediation only really destroys molecules, so any radioactives, mercury or other heavy metals, and reactive elements are still in the gas, and will cheerfully condense themselves out, forming lovely exotic compounds as they cool off and all those electrons start hooking up again. So every new waste material has to be analyzed and the post-gasificaion scrubbers and other clean-up processing systems customized for each waste stream, and the final output inspected to make sure everything went as planned, so the lab setup->pilot plant->factory progressive development cycle will have to be followed, and that takes time.
posted by Blackanvil at 9:45 AM on February 9, 2020 [4 favorites]

Hey, we're just 10 years away(!), right?
posted by OHenryPacey at 4:28 PM on February 8 [+] [!]

Guys! We just need to start work on a fusion based time machine! With the implicit expectation that the first time travel trip is to come back and tell us how it worked.
posted by ian1977 at 5:58 AM on February 10, 2020 [1 favorite]