Hacker News Reader

Comment by opwieurposiu | Hacker News Reader

You can improve the current density a lot if you can make a single crystal, or at least make your crystal grains larger.

Impure superconductor samples often come out as a spongy mixture of superconducting and non-superconducting bits, the critical current is limited because less then 25% of the cross section is actually carrying current. When I was DIYing YBCO this is what happened most of the time. Every now and then you would get a good one.

See this patent for growing single crystals of YBCO. https://patents.google.com/patent/US6046139A/en

edit: Actually if you look at the sample picture in the paper on page 7, it looks like spongy crap. Nobel prize winning spongy crap, but still. I would expect the numbers to improve as better crystal growing methods are found.

View on HN

Yeah, it’s been a pretty solid trend line, new superconductor, first batch is usually kinda shitty, but proves the basics, refining the mix nails down its exact performance characteristics.

It’s spongy grainy crap indeed… but as long as their analysis holds up to scrutiny and replication… and whoa boy do I bet there are people already trying to replicate this result as I’m typing my reply and reading the rest of the comments…. As long as the results hold up and this isn’t an abnormally low performing superconductor… i have no doubt this is going to win a Nobel prize. This has been the prize for a long time in this whole discipline, and they may have finally nailed it.

We as a species may be on the bring of a revolutionary step forward in what we can achieve in engineering and science. Better instruments and more powerful or sophisticated motors and power systems. It’s heady stuff to think it may happen in my lifetime.

As someone who doesn't follow superconductors, what sorts of things about life/engineering/society would change, how dramatic would it be, and more importantly: which stocks would you pick :)

The direct impact could be relatively limited, at least for a while. Having a room-temperature superconductor is really awesome, but existing high-temperature superconductors are fragile and expensive. You can make motors, electromagnets, and power grids with much better efficiency—but that’s not so useful if the parts break when you look at them wrong.

You’ll still get better magnets and sensors, probably. Maybe even get new types of circuitry.

Just for comparison—we use silicon for integrated circuits. Not because it has the best performance, but because it’s convenient, it’s readily available, silicon dioxide is a good insulator, etc.

Sensors could be huge I think, even with a material that's a total dud in terms of the superconductor power revolution. I'd be surprised if there wasn't a range of novel sensor approaches that haven't really been explored due to the practicality threshold of low temperature superconductivity.

https://en.wikipedia.org/wiki/SQUID

Isn't the big win of superconductors that you can build batteries with them? Like, you just pump them full of power that goes round and round forever with no or trivial losses. I always heard that this was why they were interesting.

It is an option, but there are two downsides: - such a current generates a huge electromagnetic field. So it won't work for a car battery, but may work for grid storage. - price - there is a limit to how much current you can store, and so far this was the limiting factor - i.e. we don't really care about room temperature superconductivity in this case, but we care about the price of materials to build such batteries

I'm pretty sure you can pick coil geometries that cancel external magnetic fields. There may be some stray fields, but they can be quite modest with tight manufacturing tolerances.

It's an interesting idea worth exploring. The two places where I think feasibility may face challenge is in the energy density gated by critical current density and magnetic field and in raw discharge rate (giant inductors are not known for being able to change their current quickly).

Knowing peak capacity and aging is also tricky since you can't measure critical limits without hitting a quench (a very, very bad scenario). You'll need to maintain healthy margins so you don't have things blowing up on sunny days or after so many charge/discharge cycles.

The energy density is absurdly low.

https://en.wikipedia.org/wiki/Superconducting_magnetic_energ...

Compare with this table:

https://en.wikipedia.org/wiki/Energy_density#In_chemical_rea...

Until someone sneezes and all the power is released at once.

That's only if you run close to the limit. You'd only ever do that in controlled labs.

Back in the 90s or maybe early 2000s, everyone was convinced that silicon was almost dead for high-performance chips like CPUs, and that we'd all be switching to GaAs (gallium arsenide) very soon. Turns out that GaAs wasn't that practical and silicon's limitations could be overcome, so we still use silicon today.

Would this revolutionize MRIs?

One thing I can imagine is desktop MRI machines, or, at least, much cheaper and less finicky big MRI machines that don’t take days to chill to operating temperature.

Maglevs are also a popular guess - safer, faster, and more power efficient ground mass transport would be a huge thing.

Maybe even magnetic rail space launches.

And, of course, military applications (the last few examples I mentioned involve acceleration of big-ish masses to surreally high velocities, which is popular approach to weaponry).

Those are two I see mentioned the most often (MRI and maglevs). To be honest though, those are great things to improve (especially the MRI), however, the amount of hype in this thread and on the internet tells me there must be more than just improved MRI and better trains....

I think a big part of the reason there's so much hype around this is that room temperatures superconductors have long been a famous, almost legendary undiscovered material in popsci. Theoretically possible, but with no proof that any such material actually exists. But now not only is this paper claiming that such a material exists, but also that they made some, and it's easy to manufacture and works at temperatures well beyond ambient. Seems almost too good to be true!

So in addition to any immediate practical applications there's also this element of cracking a famous long unsolved problem. It'd be like if we discovered definitive proof that P != NP, or a theoretical basis for FTL communications. Even with no immediate practical applications it'd still be huge news.

> Theoretically possible, but with no proof that any such material actually exists.

I wouldn't say that it was necessarily "theoretically possible," for there has never been, and there still isn't, a grand theory of how any given material's atomic/crystalline structure relates to superconductivity. In other words, with no theory of material superconductivity, it was never quite clear what's possible and what isn't. With this new material, though, we might get a lot closer to a working model, if nothing else.

If we can find one with the right material properties it makes long distance movement of power generation much more plausible and economical.

Which would enable us to put a lot of solar power in the desert and move it around effectively to other places on Earth.

Again, with the right material properties it could also produce efficient storage, and that also has huge implications for electrical generation.

> put a lot of solar power in the desert

If you lose 20% to the grid, then build a 20% bigger solar farm.

Solar farms actually help with reversing desertification by reducing water losses from direct exposure to sunlight.

I got into superconductors shortly after the Fukushima disaster, and it was the poster child for superconductors in transmission lines. One of the perks back then was that you could put these reactors that are sensitive to earthquakes and the like out where they are relatively safe from such things. You can also have fewer, higher output power stations, so fewer municipalities have to operate smaller power plants (which tend to be coal or natural gas).

If its economical it will help with losses in the grid over distance. colliders like the LHC and fermi lab could be a lot cheaper to upgrade with room temp superconductors. More reliable too. Would also be a big help fusion reactors. Anything that uses electric motors will also benefit.

Brute forcing fusion by using insanely strong magnetic fields I suppose. What CFS is trying to do, smaller tokamak but much stronger magnets

Make sure you don't have anything metal lying around your desk.

Maybe if it pulsed the magnetic field very quickly it wouldn't be a big deal.

Why don't you try it first ;)

I have too many hobbies right now.

The changes would be more dramatic than the discovery of fire, but would occur gradually.

The changes introduced by discovery of fire were very gradual. No Internet, no written language at all: any innovation back then could only spread as quickly as humans walked, and would temporarily stop at every significant geographic barrier such as sea.

Compared to the discovery of fire, the changes from room temperature superconductors would be a flash fire.

I totally missed the impact of the internet, you are absolutely correct. Something else that I just noticed buried in the paper is that it can be readily deposited onto a substrate (search for UNIVAC in the paper, page 4). That's an incredibly big deal.

For the idiots in the room (me), could you explain why that’s a big deal? Does that make the material more robust and therefore usable in harsher environments?

We can deposit it on chips, there is lots of copper (relatively speaking) in a chip so this could improve efficiency. Of course the semiconductor (silicon) is intentionally resistive, so waste+heat wouldn't go away entirely.

Thank you!

Note that when we invented the aeroplane, uses weren't immediately obvious.

People would have suggested things like cities in the sky, looking down on things, and traversing marshland easily.

The actual main use for planes has turned out to be fast long distance travel. But we don't actually theoretically need to be up in the air to travel fast or far - in fact, had we never invented the aeroplane, we'd probably have cars or trains by now that moved as fast as present day planes do.

The revolutionary effect of new inventions is often hard to see, particularly for basic science research like superconductors.

You're forgetting the other primary use of airplanes: to drop bombs on things.

https://twitter.com/alexkaplan0/status/1684044616528453633?s...

Can you? The proposed model implies this relies on non-periodic impurities to achieve superconductivity, and the displayed strength is consistent with superconductivity only appearing on some places spread inside the material. This seems qualitatively different from YBCO.

I don't know enough to be sure on the certainty of that model, but it seems well supported. So I'd be surprised if only improving the material quality was enough to make it strong.

You are probably right about the details, but as a general observation: industrial products tend to get better over time as people learn how to make them more properly. The first batches coming out of a lab are usually terrible compared to what's coming out of factories a few decades down the line.

Are we talking the cool crystal selection spirals like RR et al use to make monocrystaline turbine blades? https://www.americanscientist.org/article/each-blade-a-singl...

Reply on news.ycombinator.com