Physicists discover a “family” of robust, superconducting graphene structures | Hacker News Reader

Physicists discover a “family” of robust, superconducting graphene structures

news.mit.edu

134 pointsfiloeleven25 comments

Fascinating.

Sadly this still needs to be at super low temperature.

This family of superconductors brings us no closer to to room temperature superconductors which would change the world.

What's really neat about this result is that they're probing a link between the band structure of materials and superconductivity. So while no, this isn't at room temperature, stacked graphenes provide a controllable family of metamaterials that can be probed to learn more precisely the conditions for superconductivity. If hot* superconductors exist, we'll have a better chance at finding them if we understand how superconductivity works.

* we don't need "room temperature" superconductors for room temperature operation. Superconductors have a shared budget[1] of temperature, current and magnetic field -- if we want these so-called superconductors to actually carry current at room temperature, room temperature isn't enough!

[1] http://hyperphysics.phy-astr.gsu.edu/hbase/Solids/scbc2.html...

What are you talking about not needing room temperature superconductors for room temperature operation?

If at 300 K a material has zero resistance then that material will carry current at room temperature.

You need one that can handle higher than room temperature for it to be able to carry meaningful amounts of current at room temperature. See the link provided in parent for why this is.

Like I said, there's a shared budget for temperature, magnetic field and current. If you're sufficiently close to the critical temperature, it only takes a small current to "blow the budget" which turns your material back into an ordinary conductor/resistor. If you remove the current or lower the temperature, there's a nonzero switching time before the material will return to its superconducting phase.

There are some neat applications of this phenomenon in superconducting electronics. Cryotrons use transformers to produce large fields from small currents to switch a higher-current-capacity line from superconducting to normal. N-trons and other "current crowding" devices perform a similar trick using currents in a pinched region attached to the high current line. I can't remember the precise details, but different superconducting films are more or less inclined to "localize" the normal-phase region induced by local current/heat.

Sadly, those neat devices are apparently quite finnicky because switching emits heat into their environment, and also depends on temperature of that same environment. As far as I know, cryotrons are used for high power electromagnets to rapidly and safely dump the energy -- they don't appear to be useful for superconducting computers, for example.

I think it's difficult to get away from the prospect of a practical widespread use of superconductors without active cooling. Even if ohmic losses in the superconductors is not present: environmental temperatures change, especially in uncontrolled (outside, in the sun) environments. Running room temperature water adds cost and complexity, but a lot less than refrigerant (especially cryogenic).

I think mass produceable room temperature superconductors would still reshape the world, even if they still required active cooling.

Thanks for your comment, if they broke the temperature requirement that would be BIG news.

The main point I get is that the mechanism for this kind of superconductivity can be predicted fairly well on computers. That means that if it leads to room temperature superconductivity, we have an actual path to get it.

And, also, there is always those old too noisy experiments that detected superconductivity on certain grains of graphite on temperatures up to 700K... That never gathered enough confidence, no matter how many times they were repeated. But maybe there's something there.

EDIT: Anyway, the other article on the front-page about this experiment explains it much better.

Really, I thought it "was" at room temperature?

Higher temperature superconductors are already changing the world. Doesn’t have to be room temp. Some were used to help an MIT project reach >20 Tesla magnetic field, which wasn’t possible before…

I think a great application for these higher temperature (and higher B-field, current) superconductors is reaching higher magnetic fields for fusion :)

Or just MRI machines that can function without the increasingly expensive liquid helium.

Newest MRI scanners aren't as bad about helium. What I understand is the "problem" mostly was that they used to be designed for an environment where helium was cheap. So helium cost wasn't a major design factor prior to the newest generation of magnets. The old designs are still shipped/refurbished/remanufactured though. I also think they're still manufactured new, too, but I'm not sure. In any case most if not all of the major manufacturers have new models that drastically reduce the amount of helium to a few liters and recapture well enough to the point that both refills and even quench pipes are not required for some models.

Has some high powered algebraist or topologist tried figuring out what is happening here mathematically? what structure is generated by all possible rotations that we are sampling here? multiplied by the many layers, what about nonuniform sheets? i’m imagining channels of various shapes on the inside… such a simple concept, yet such fascinating machinery for the measurements. i feel proud when i get a logic analyzer trace off an fpga pin!

Hum... The article actually says it at some point:

https://en.wikipedia.org/wiki/Moir%C3%A9_pattern

https://everettyou.github.io/2018/05/21/Moire.html

I imagine there's no ELI5 explanation of a flat band from condensed matter physics.

That’s just the original discovery, I’m hoping someone would come in and look for more mathematical structure like the current efforts for topological insulators.

It seems like graphene production at scale has been a few years away for a decade or so now. Is there reasonably a future where all these incredible graphene-related technologies come to fruition within our lifetimes? Or is this just another cold fusion of a technology?

A few things:

1. REBCO was discovered in the 80s but took decades to work out the kinks in mass production of reliable tapes.

2. Cold fusion is psuedoscience. These results are real. Speculating that a technology be vaporware is a very different and weaker claim. How do you know that people will not solve the engineering problems preventing mass production? Because it wasn't immediately obvious how to do it in the past?

Cold fusion got rebranded as Lattice Confinement Fusion.

Who knows, the academic research is often focused on finding out new physics and relies on hero samples. The many failed attempts to solve the engineering challenge of scalability and reliability in graphene are not very visible to the world.

Aluminum was extremely difficult to manufacture before the Hall-Heroult process. Only a true pessimist would assume we would never solve manufacturing problems with graphene.

> The findings could serve as a blueprint for designing practical, room-temperature superconductors.

Cool, up to what temperature?

> 1.7 kelvin

Ok.

It is a disgrace to intellect when scientific knowledge comes with references to magic (even in quotes). Can't we just say "exact", "proper", "special", " specific", "just right"? Why magic?

"Magic" doesn't necessarily mean supernatural. Google quotes the Oxford English dictionary to include these definitions:

- something that has a delightfully unusual quality.

- very effective in producing results, especially desired ones.

I think those definitions make "magic" a better term in this case than any of your alternatives.

If I say during a code review that "magic numbers should be documented", I expect the other person to know the phrase and not think I'm talking about The Book of the Sacred Magic of Abramelin the Mage.

I expect similar things in other domains.

The term "magic number" pops up in many domains. Often it's because the underlying mechanism that yields the empirical result is unknown. If that's the case here then that context should not be lost because of a semantics argument.

Reply on news.ycombinator.com