NASA’s New Shortcut to Fusion Power (2022)
62 points
2 months ago
| 4 comments
| spectrum.ieee.org
| HN
jiggawatts
2 months ago
[-]
This smells of pseudoscience, somewhere between LK99 and cold fusion.

For example, they reference an experiment so bad that I as a layperson could find a dozen glaring flaws in the associated published paper.

The referenced "result" was about changes in a Johnson-Matthey purifier when deuterium and/or hydrogen was cycled through it: https://www.sciencedirect.com/science/article/abs/pii/S03603...

This experiment is destructive, so the team compared two different palladium-silver hydrogen gas purification tubes with months worth of gas cycled applied to them. They then assumed that the tubes started out "the same", and hence they wrote up any differences as changes that occurred because of the gases they used. This blithely ignores the differences in the tubes due to manufacturing variations!

That's shockingly bad science, and I very much doubt that sincerity and honesty of anyone who chooses that specific "result" as something worthy of referencing in a journal article.

reply
perihelions
2 months ago
[-]
- "somewhere between LK99 and cold fusion"

It's cold fusion. Lattice-confinement fusion was an attempted re-branding of the cold fusion field.

- "There are also upcoming conferences: the American Nuclear Society’s Nuclear and Emerging Technologies for Space conference in Cleveland in May and the International Conference on Cold Fusion 24, focused on solid-state energy, in Mountain View, Calif., in July."

reply
DennisP
2 months ago
[-]
The difference from cold fusion is that NASA is blasting the lattice with gamma rays, which is what gives deuterons the energy to fuse.

Whether there's a chance of getting net power that way, I have no idea, but at least it's not depending on previously unknown nuclear reactions.

reply
perihelions
2 months ago
[-]
But in another place (the paper the grandparent comment linked) they're claiming a classic Pons-Fleischmann experiment, where they're just pumping in cold deuterium gas–no heat, no radiation–and say they observed spontaneous nuclear reactions. Party like it's 1989.

I agree with the grandparent commenter: zero credibility here, this group is doing nothing but trolling people.

reply
Phreaker00
2 months ago
[-]
This is an article from 2022. Previous post with more comments: https://news.ycombinator.com/item?id=30503111
reply
throwaway69123
2 months ago
[-]
“The enriched uranium fuel they use must be kept secure. “ really lacking context on this sentence, enriched nuclear waste is stored because of non proliferation treaties not because of limitations of the science or technology
reply
theonlybutlet
2 months ago
[-]
Non-proliferation and more importantly nefarious use.
reply
hutzlibu
2 months ago
[-]
Like dirty bombs, that anyone can make easily, who can make a ordinary bomb.
reply
simiones
2 months ago
[-]
The fallout from a dirty bomb would be relatively easily cleaned up with showers and a few hoses on streets and buildings. The people close to the blast would also inhale some radioactive dust which will be a much bigger problem, but most of them will probably just die because of the conventional explosion itself. The radioactive dust is definitely far less problematic than shrapnel from a fragmentation bomb.

Overall dirty bombs are really not a problem bigger than any other kind of bomb.

reply
XorNot
2 months ago
[-]
Dirty bombs are an overblown threat, compared to a regular bomb parked near a stadium security line or what you can make with off the shelf chemicals.
reply
jabl
2 months ago
[-]
Enriched uranium would not make a particularly impressive dirty bomb. Spent nuclear fuel, or some specific very radioactive isotopes, maybe.
reply
kjkjadksj
2 months ago
[-]
Even more sinister would be to just hide radioactive material some place. No bomb, no one knows whats wrong because no one casually sweeps a geiger counter, and this probably would probably irradiate people nearby for years before its found.
reply
hutzlibu
2 months ago
[-]
Yeah, but the point of terror is to make terror, that people are afraid. Not to actually damage much. A dirty bomb would be a perfect terror weapon as people are afraid of radiation and mostly would not trust government when they declare it is safe.
reply
jjk166
2 months ago
[-]
The US actually does have a radiation detector network that covers most major cities. If you have enough radioactive material to pose a threat to people's health from brief exposure over a non-trivial area, it would be detected. You'd probably also wind up on a lot of intelligence agencies' radars trying to acquire that much radioactive material and the equipment for handling it.
reply
kjkjadksj
2 months ago
[-]
How many smoke detectors would you have to stockpile before getting a knock on your door?
reply
rsaxvc
2 months ago
[-]
In addition to proliferation and other tasks: do not turn into radioactive meteor upon reentry breakup.
reply
RC_ITR
2 months ago
[-]
It will be interesting to see if this helps solve the tritium/helium 3 problem that all current fusion companies face.

I’ve always found it interesting that they are allowed to pitch VCs claiming that they are doing the same thing the sun does, when the sun in fact does proton/proton, proton/deuterium, and helium 3/helium 3 fusion at a relatively slow rate under extremely high pressure.

reply
simiones
2 months ago
[-]
Given that this is a new form of cold fusion, this is merely bad science and will not "solve" anything. Shameful for NASA of all places to spend money on such a hoax.
reply
DennisP
2 months ago
[-]
That's not an actual problem. To breed tritium, bombard lithium with the neutrons from D-T fusion. To breed helium-3, fuse deuterium.
reply
RC_ITR
2 months ago
[-]
It is a problem unless I can borrow your Tokamak that can do D-D fusion at 300mn degrees C without insane energy losses.

And as for breeding tritium - I think you're skipping a few steps. D-T fusion gives you 1 neutron, which is enough to make 1 tritium isotope from Li6 (7.5% of global Lithium supply). Sure you can use neutron multipliers like Be (otherwise, anything less that 100% neutron capture depletes the supply), but how much energy are you losing then (since keep in mind, 80% of the D-T energy is that neutron to begin with)?

And beyond all of that, it's a problem unless you can show me your breeding blanket design and the system you plan to use to capture the radioactive gaseous tritium in a way that doesn't poison any water supplies (keep in mind, as one of the smallest and most buoyant gases, it's really hard to contain hydrogen without noticeable leakage).

Anyway, once those issues are ironed out, sure it's not a problem, but it's a huge problem today.

reply
DennisP
2 months ago
[-]
You don't need to breed He3 unless you can get net power from D-He3. If you can do that, then you can probably also get net power from D-D, which doesn't require as high a triple product. The fusion company Helion is working on exactly this, and attempting net power in 2024 with an FRC design.

Breeding tritium doesn't make the neutron's kinetic energy disappear, or convert into matter. Neither does multiplying neutrons; you just get two neutrons whose combined kinetic energy sums up to that of the original neutron. In the end, as usual, the energy mostly ends up as heat.

I'll grant you tritium capture. Several of the private companies use a molten breeding blanket (CFS, Zap Energy, and General Fusion, possibly others), so at least the tritium won't be trapped in the blanket.

reply
RC_ITR
2 months ago
[-]
In your first paragraph what are you even saying? If you make one complex fusion reaction work then a different one is cake? I don't think helions “smash the plasma linearly” design even works for D-D. Their quartz containment chamber definitely is a weird set up for D-D. I’m aware of what Helion is trying to do and wish them luck, but they won’t even release data on their supposed He3 reactions, so I’m guessing they’re not light years ahead on D-D and ready to use their very customized tech for a different reaction quite yet.

As for your very salient point on where the energy goes, what are you heating? Your new Tritium. So now you not only have to capture the tritium but you have to extract its heat to generate electricity? Or are you just capturing heat from the helium and whatever is absorbed by the rest of the blanket?

No startups use a breeding blanket in 2023. They just talk about them. All the designs are still very much in prototype phase and nobody has blanket IP because they barely have reactor IP (though lots of cool magnet patents!)

In fact if they did have breeding blankets, they could actually make some money selling their tritium!

reply
DennisP
2 months ago
[-]
It's just temperature, density, and confinement time. For each fuel, there's a minimum threshold temperature and a minimum triple product of those three parameters to get net power. The triple product is lower for D-D than it is for D-He3.

For breeding tritium, everybody's focused on just getting net power right now but several companies are planning liquid breeding blankets. For CFS it's a molten FLiBe salt, a couple others are using molten lead-lithium. Either way, you run coolant pipes through it, and the new tritium will bubble out the top. Honestly it doesn't sound all that hard.

reply
RC_ITR
2 months ago
[-]
>It's just temperature, density, and confinement time.

I know what a triple product is (even a Lawson criterion), but what you're saying is like saying curing cancer is just "figuring out how to turn off uncontrolled cell division." There's a lot of practical engineering that is behind that sneakily simple phrase. If D-D fusion is so easy, why don't the other start-ups pursue that since D is abundant and T is one of the rarest materials ever? In fact, since D-D makes T 50% of the time anyway, why don't they do both?

As for the blankets, as I've said, plenty of people talk about them, but have never even come close to building one.

>Honestly it doesn't sound all that hard.

Yeah, I agree it doesn't sound very hard. By chance do you work at a VC firm?

reply
DennisP
2 months ago
[-]
D-D requires a higher triple product than D-T fusion and produces less energy, so most companies go with D-T, and don't attempt to use D-D to produce the tritium.

But D-D takes a lower triple product than D-He3, so if you can get net power from D-He3, then D-D is a fine way to produce the He3.

Helion is the only company that thinks they can get net power from D-He3 so they're also the only company pursuing D-D for breeding.

reply
RC_ITR
2 months ago
[-]
>D-D requires a higher triple product than D-T fusion and produces less energy, so most companies go with D-T, and don't attempt to use D-D to produce the tritium.

Yes, because again, all the ways we know how to make Tritium are implausible from either an energy balance standpoint or a "where do we get all these energized neutrons" standpoint. That's always been my argument.

>You don't need to breed He3 unless you can get net power from D-He3. If you can do that, then you can probably also get net power from D-D

Going back to this, you're aware D-D results in 50% T and 50% He3, right? Helion relies on D-T fusion working so they can sell their "waste" Tritium (lol).

None of these things you are claiming to be easy factually pencil out and that is why we are so far from commercial fusion.

reply
DennisP
2 months ago
[-]
D-T: Where you get all those energized neutrons is easy: from D-T fusion. You need a neutron multiplier like lead or beryllium, then it looks like you can average about 1.15 tritium atoms per D-T reaction.

Helion: to use their tritium they just have to wait. Tritium decays to He3 with a 12-year half-life. (Of course if they didn't want to use the tritium at all, it would just mean building twice as many D-D reactors and it'd still work. That'd be silly though.)

reply
wheelerof4te
2 months ago
[-]
"when the sun in fact does proton/proton, proton/deuterium, and helium 3/helium 3 fusion at a relatively slow rate under extremely high pressure."

We actually don't know how the Sun does fusion. If we did, we'd have created it here on Earth already.

reply
ben_w
2 months ago
[-]
> We actually don't know how the Sun does fusion. If we did, we'd have created it here on Earth already.

We know we can't do what the sun does here on Earth, and we also know that we don't want to what the sun does.

We can't, because the cube-square law means the sun stays hot enough for long enough for this to happen even though a reactor on the ground would cool down almost immediately.

We don't want to, because the reaction happens over the course of billions of years giving the sun a specific power output similar to rotting cabbages.

reply
andruby
2 months ago
[-]
There’s a difference between knowing the physical and chemical reactions and being able to reproduce it. We have reproduced it, but not yet sustainably and in an economically profitable way.

We can’t reproduce exactly what the sun is doing, because we don’t want the whole planet to run fusion. One of the large problems to solve, as I understand it, is containment. No need for containment on the sun.

reply
simiones
2 months ago
[-]
> One of the large problems to solve, as I understand it, is containment. No need for containment on the sun.

Containment, in the context of fusion, doesn't refer to preventing the reaction from running rampant. It refers to keeping the reactants in a small confined space so the reaction can continue at all.

The sun achieves this containment through the least efficient means: it's gigantic mass means that the reactants are forced together by the force of gravity. Plus, the shear quantity of superhot plasma helps ensure that energetic particles have an excellent chance to hit another core, keeping the reaction going. Note that the solar core, where the vast majority of solar fusion happens, is ~0.25 of the sun's radius, so ~25 Earth radii. Quite hard to get the amount of plasma on Earth.

The entirety of the Earth doesn't have even a fraction of the Sun's mass, so that is impossible to reproduce. So we need to rely on much more violent forces to achieve a similar level of containment as the Sun.

reply
ryanpandya
2 months ago
[-]
I thought we're pretty clear that it's thanks to absolutely colossal mass, which... Can't really happen on earth since it requires the mass of a huge number of earths.
reply
wheelerof4te
2 months ago
[-]
Then we need to find another way to generate fusion power, or admit that it's impossible to do on this planet.

Throwing lumps of cash at fusion power venture capital feels almost like a money-laundering fraud.

reply
ben_w
2 months ago
[-]
> Then we need to find another way to generate fusion power, or admit that it's impossible to do on this planet.

Yes. Here are the currently known alternatives off the top of my head:

* Fusion bomb

* Inertial electrostatic confinement (14 year olds have done this for highschool science fair projects)

* Virtual cathodes (Polywell, technically also an IEC variant)

* Magnetic confinement (Tokamaks: the big expensive doughnuts)

* Field-reversed configuration (technically also magnetic, but these are magnetic "smoke rings", Helion Energy does this)

* Z-pinch (again magnetic, but different)

* Laser confinement (NIF)

* Muon-catalyzed fusion (no chance of this becoming mainstream unless we can convince muons to stop decaying)

IIRC all of these have demonstrated that it's physically possible; some even show promise.

(I've been wanting to write a plasma physics simulation to see if I can improve on the Fusor design since I was at university; now chatGPT writes acceptable code, I keep asking myself why I've not yet made this happen…)

reply
varjag
2 months ago
[-]
The whole fusion energy effort since 1950s is about finding another way.
reply
Maken
2 months ago
[-]
At least it's not crypto and it's not actively hurting anyone.
reply
RC_ITR
2 months ago
[-]
Until the tritium leaks...
reply