Researchers demonstrate the ability to fuse atoms inside room-temperature metals | Hacker News Reader

Researchers demonstrate the ability to fuse atoms inside room-temperature metals

spectrum.ieee.org

71 pointsMindGods39 comments

Getting things to fuse is "easy" in the sense that you can do it on the lab bench with a farnsworth hirsch fusor, but that has no hope of ever reaching break-even. This article doesn't talk about break-even potential at all and yet they jump to potential applications.

"In the latter case, that extra neutron can start the process over again, allowing two more deuterons to fuse."

So sounds like they are hoping for free reactions.

That's future research and doesn't necessarily mean it makes up for losses. They're counting chickens before they hatched.

That still involves heating though, right?

Is breaking even in such an experiment in violation of any thermodynamics laws? Or is it that we just haven't figured it out?

It doesn’t violate thermodynamics; it’s the conversion of mass to other forms of energy. Nuclear fusion is the source of energy from the sun, but the sun has the advantage of its enormous gravity keeping things confined. The attempt to extract usable energy from controlled fusion on Earth is the attempt to, in the absence of this helpful gravity, find some other way to get enough nucleons close enough together. Nature makes this practically impossible, which is why we don't have fusion power after 70 years of trying. Fortunately, we don’t need it, as we can now use the sun’s fusion here on Earth.

In theory anything lighter can fuse toward Iron. Anything heavier can fission the same direction. Splitting or fusing Iron requires input of energy even in theoretical scenarios at 100% efficiency.

In practice a great many nuclear reactions are not chain reactions or don't yield enough energy to be useful. Really large atoms kinda want to fission due to the speed of light: forces can't propagate from one side to the other fast enough to completely balance out. So splitting Uranium is a lot easier to make useful because we have a head start as it were... if you leave it alone in a box some of it will go ahead and split in pieces for you for free.

Fusion is just more difficult. If you leave hydrogen atoms confined in a box approximately none of them will spontaneously fuse to form helium.

>To overcome that barrier requires a sequence of particle collisions. First, an electron accelerator speeds up and slams electrons into a nearby target made of tungsten. The collision between beam and target creates high-energy photons, just like in a conventional X-ray machine. The photons are focused and directed into the deuteron-loaded erbium or titanium sample.

that way of X-ray generation is of very low efficiency. They should have put that deuterium loaded erbium, titanium (or Pt or Pd like in the famous cold fusion experiment) into the Sandia Z-machine. The typical target for the Z is either LiD or frozen D, and i wonder why they have never tried more heavy metals, especially Pt or Pd, loaded with D given how heavy nuclei is supposed to help in the fusion based on the cold fusion effects and which this NASA research seems to hint at too:

>But the lattice helps again. “The electrons in the metal lattice form a screen around the stationary deuteron,” says Benyo. The electrons’ negative charge shields the energetic deuteron from the repulsive effects of the target deuteron’s positive charge until the nuclei are very close, maximizing the amount of energy that can be used to fuse.

Honestly, my best bet is that Musk, who needs at least fission or even better fusion for Mars (space is the only business case for any plausible peaceful fusion), would soon start a venture for it. The inertial confinement, either Z-machine style or laser (modern lasers are much more efficient than NIF) is clearly the way to go, especially for space and when you need real result instead of large government sponsored research.

They don't use heavy metals because they strongly absorb X-rays, lower density materials are better at transmitting x-rays which is why they often use very low Z elements. Sometimes they dope with heavy metals but nothing has yet allowed to reach break even.

definitely, that is in general a factor. Though erbium and platinum doesn't differ that much than it come to X-ray attenuation.

Additionally - the X-ray absorption by the lattice is important in the low energy level setting like that NASA experiment (basically equilibrium conditions) as any absorbed energy is basically lost and would just radiate away, whereis in high energy Z shot (non-equilibrium inertial confinement situation) the absorbed energy would still result in increased temperature and pressure of the target.

Honestly, my best bet is that Musk, who needs at least fission or even better fusion for Mars (space is the only business case for any plausible peaceful fusion)

An interesting assertion. What are some of the reasons why you're certain that fusion will not be a useful source of energy in peacetime/civilian applications? Is it likely to be "too cheap to meter," as we were promised fission power would be, or do you think it'll always be too expensive to be practical?

Cheapest electricity is natural gas one. Imagine a natural gas plant with free natural gas. That would decrease the electricity price by about 1/3 (fuel price is up to 2/3 of natural gas based generation, and the generation is between half and 2/3 of electricity price). The fusion power plant ideally is equivalent of that free natural gas power plant. Though it can't be further from it on practice. Any feasible in the observable future fusion involves neutrons. That in addition to affecting reactor design - cost and complexity of building, operating and maintenance - brings in the whole regulatory regime just a notch lighter than fission. Taken together we'd be lucky to match the current natural gas - I personally think no chance in hell. Carbon emission of natural gas isn't an issue because nobody [who matter] cares about it today, and then, when the issue is finally forced for resolution, adding carbon sequestration would still be a comparatively minor expense. And that not even talking about already free fusion used in solar and wind based generation - hard to beat economics of that fusion reactor.

Whether fusion is expensive depends on the reactor design. There are a lot of possibilities.

On one end of the scale is ITER, the giant tokamak they're building in France for tens of billions of dollars.

At the other end is the petawatt laser boron fusion idea being pursued by HB11 Energy and various other researchers around the world. The main expense would be the laser at tens of millions of dollars.

For power on the surface? Solar will do fine and is dead simple to setup. I have no specific knowledge, but I suspect there is some way to relatively easily manufacture solar cells on mars with minimal equipment, and that would be a much cheaper prospect that shipping a big nuclear plant to mars. Perovskites last I read are simple to make, and the atmosphere of mars may be far less corrosive with the lack of free oxygen.

For propulsion? Nuclear rockets can't be used until you get out of the allen belts, otherwise fission product fall back to earth, it's really what submarined Orion as a concept. Pulse nuclear is probably the primary means for interplanetary transport if a sufficient manufacturing can be bootstrapped.

In space stations / asteroid mining? Again solar is king without pesky atmospheres getting in the way. But once you have pulse nuclear ships flying around, it should be trivial to also have a thorium reactor for power.

propulsion. Additional 3-4km/s of delta-V would allow to not be limited to those once-in-2-years launch windows and significantly decrease travel time. That would be a game changer on top of the game changer that the Mars bound Starship is already going to be. Nuclear launch from Earth isn't an option, yet Starship can easily launch a reactor into LEO, fission or fusion (fusion i think will even be more compact), and it can be used from there.

Reading this kind of news makes me very excited but again afraid at the same time, thinking is it again one more case of Fleischmann–Pons defamation in progress. As it sounds very much as saturation of palladium bar with isotopes of hydrogen.

From the article: “What we did was not cold fusion,” says Lawrence Forsley, a senior lead experimental physicist for the project. Cold fusion, the idea that fusion can occur at relatively low energies in room-temperature materials, is viewed with skepticism by the vast majority of physicists. Forsley stresses this is hot fusion, but “We’ve come up with a new way of driving it.”

How they define "cold fusion" temperature, if we look previous lab fusion experiments they all had very narrow point of fusion, welding arc has temperature of 10000 F withing few millimetre, that is what most of those previous so called cold fusion experiments did. But in reality fusion should occur at 100 million K. Cold fusion is refereed as room temperature, is there some trick in defining room temperature?

Just 20 more years and we'll have cheap, safe, limitless energy!

Just 20 more years? Count me in, that is much better than what they promised 20 years ago, when they said that fusion energy would be viable around 2020, or what they told in the eighties when they said it would be 2000...

Wait a second! What is going on here?

To be fair, the timeline was predicated on spending levels that were never met. In fact, funding levels never made it above the "fusion never" threshold: https://i.imgur.com/3vYLQmm.png

Interestingly, the projected budget was $100B-$300B in 2020 dollars. Certainly puts $4T stimulus plan into perspective!

Thorium reactors would have taken much less than that level of funding with (IMO) less engineering hurdles, and they made a functioning prototype that almost fit in a closet at General Atomics.

This is all boondoggle money now with solar/wind/battery though. Maybe once we actually move a majority of the grid we can move back to major nuclear or fusion research.

I have seen this graph many times before on HN. I'm not sure how accurate it is. Maybe the investments in fusion have recently ramped up (the graph ends in 2012), but the end point of the "actual" fusion investment shows something a little less than half a billion dollars. Is this supposed to be worldwide investment, only US, only US Department of Energy, or what?

Anyway, just in 2019 the US DoE got a budget allocation of more than half a billion dollars ($564 MM to be precise, see [1] page 162). ITER's financial statements for 2019 ([2], page 43) shows member contributions of € 400 MM, which is again about half a billion dollars.

If we could manage to add up all the various research budgets for fusion, we would probably be quite a bit above the "never fusion" line, we could maybe reach the "moderate level" in the graph (the orange line).

But the question is: is this needed? MIT alone has a fusion project (SPARC [3]) that appears to be ahead of ITER. A spin-off of that project is the Commonwealth Fusion Systems [4], which managed to raise about $200 MM of funding entirely from private organizations.

Is it possible that we'll see a repeat of the Human Genome Project scenario, where the US Government invested $10 BN and more than a decade of reasearch, only to see a private company (Celera [5]) come in an steal the thunder at the finish line, with only a 20th of their budget?

My point is that as society progresses, there is a time when a certain thing becomes achievable on a medium budget, which only 50 years before would be impossible on an infinite budget. Just think about sending a rocket to the Moon in 1920 or creating an mRNA-based vaccine for the coronavirus in 1970. It is very, very likely that if the US government had allocated $50-100 BN for fusion research in 1970, we would not be any closer to fusion today. However, today, 50 years later, we are at a point where fusion appears achievable only based on private investments.

[1] https://www.energy.gov/sites/prod/files/2019/05/f62/fy-2020-...

[2] http://e.issuu.com/embed.html?d=2019_iter_annual_report&u=it...

[3] https://www.psfc.mit.edu/sparc

[4] https://en.wikipedia.org/wiki/Commonwealth_Fusion_Systems

[5] https://en.wikipedia.org/wiki/Celera_Corporation

Fusion seems like a rare opportunity to do remarkable impact for human society while also making a remarkable profit. Frankly, if I was a billionaire, I would treat fusion investments as half-charity. That's why I don't understand how so many billionaires have pledged to give a large % of their wealth away but the most promising privately funded fusion energy project only has $200 million in funding. That kind of discrepancy increases my respect for tech evangelists, Ray Kurzweil, etc...because marketing the future might bring it closer. We need more of that.

Bill Gates is doing just that, but with fission rather than fusion. If the aim is to fight climate change, then fission is a proven technology, while fusion is a very risky bet.

Great. We should build standardized SMRs everywhere ASAP. Nonetheless, we don't have to pick between SMRs and fusion reactors. We can do both. The riskiness of fusion can be counteracted by taking an investing-as-charity approach to the whole field. If a billionaire is fine moving humanity closer to a low scarcity, carbon-free future without necessarily getting a commensurate payout, they should invest in fusion. Additionally, if they continuously invest in fusion and spread their investments across multiple fusion "bets," they can maximize their chance of making an outsized return.

As an aside, I think it kinda sucks that we're in the age of information, but I can't google "fusion energy research funding history" and get anything other than op-eds and old news about funding announcements. We should be able to see this information cleanly presented and filter by US, Worldwide, etc. Then we wouldn't have to guess or make assumptions. Is there a source of information I am lacking or do I have to do my own research project on fusion funding just so I can participate meaningfully in this conversation?

I’m working on this exact issue. You can view all fusion energy companies here and filter by location. https://www.fusionenergybase.com/organizations/ You can view each of their funding histories on the company detail page, just click on the company name.

DO NOT BE MISTAKEN This is the same Pons & Fleischmann Cold Fusion effect. I will not argue semantics but Pons & Fleischmann, and Andrea Rossi are also due credit for researching this phenomenon. it makes sense if you think about it... imagine a 3D metal lattice...only say 8 atoms... you tightly pack as much Hydrogen as possible in the middle of the lattice.... everything is under immense heat and pressure.... spinning like crazy.... you hit it with a jolt. or a hammer.... and everything sorta smashes into each other and explodes. or maybe this is the 'inverse beta decay' / controlled electron capture method.

It is certainly reminiscent of P&F, in loading a metal matrix with deuterium. And the effect this paper documents could ... conceivably ... possibly ... maybe explain the erratic and fugitive positive results that P&F-style experimenters sometimes produced. But I don't think your certainty is well-founded yet.

Spot on. Funny enough speaking about semantic, remind me on Star Trek dilithium crystals...

Is there any reason this couldn't be used for general purpose energy? They work for NASA so perhaps they are just mentioning propulsion to keep their funding.

As noted by another commenter, this is a significant result but overall still very early. It's still not even clear if it can be used for energy production at all and there is a lot of work to be done before that will even be worth asking seriously.

But a simple answer is: if it can be made to power a spacecraft it would probably be suitable elsewhere.

>Is there any reason this couldn't be used for general purpose energy?

It's possible to get out energy if you put enough energy in, but currently you can't even retrieve all the energy you started with, let alone generate energy efficiently enough to drop the cost below solar et al.

The idea is that if you fantasize that this might, one day, have gain > 1 it would be much smaller and lighter than other forms of fusion that you might also fantasize would, one day, have gains > 1. But, in this fantasy world, terrestrial fusion reactors would have much higher fantasy gains, so you are accepting a lower fantasy gain for lower weight.

The reality is, fusion energy research has been a major activity since the 1950s, but no device (aside from bombs) has achieved practical break-even, despite occasional breathless but deceptive press releases from national labs.

http://progressive.org/op-eds/let-cut-our-losses-on-fusion-e...

I'm guessing because it'd be expensive and have a low power output. That'd make it uncompetitive for a power plant, but fine for a deep space mission; you'd need very little fuel, so as long as the device itself isn't too massive, it'd be great.

Not sure about the low power output. With the caveat that I'm a nuclear engineer but not a fusion expert, from what I understand the deuterium density is actually quite high compared to plasma-based systems like tokamaks. NASA's press release says the deuterium density in their lattice confinement process is a billion times more dense than the plasma in a magnetic confinement system (https://www1.grc.nasa.gov/space/science/lattice-confinement-...), though I note it's not clear exactly what they mean. So that is promising.

If they can figure out how to sustain a chain reaction or otherwise extract a net energy gain it might still be useful as battery even if they end up having to put more energy to fabricate the fuel in than you get out. Something like how current RTGs are used, except I expect it would have much higher energy density.

This is all speculation, however, and there is a lot of work left.

Interesting point. I just guessed at a plausible reason to focus on deep space, I don't know whether that's actually the case.

It seems like it couldn't have too much power density though, or the solid lattice would melt.

The lattice melting is probably not a major issue, you just have to cool it to remove the heat as it's generated. At a basic level you face the same problem in a fission reactor and the solution is straightforward -- extract the heat into some sort of working fluid (or gas) to keep the fuel from melting. Fission reactor fuel is actually not even terribly dissimilar to the stuff described in the article, most often it's fabricated as a sintered powder.

Again, this is super early so this is all wild speculation, but generally engineering challenges are easier to solve than fundamental science problems. Of which there remain a few. That said, most of the problems with MCF are engineering issues and we still don't have that working yet either so...

Reply on news.ycombinator.com