What’s going on with the reports of a room-temperature superconductor?

Rumors are flying of confirmation, but the situation is still frustratingly vague.

See full article...

 

[klnn]

Isn't that like 90% of everything that makes use of flow of electrons?

only if you ignore transportation of other things than electrons, heat generation, light generation... so most things in fact.

 

[O/Siris]

Power transmission and computers seem like the most obvious applications, any others that I am not thinking of?

I think I've seen superconducting mentioned in the context of Mag-lev trains, too.

In fact, I think mag-lev could have all kinds of industrial applications.

Actually, I've got this really weird idea of using a superconductive shell around a package inside a matrix of magnets to ship something in a really low-impact scenario. Not sure that'd be practical, but I keep thinking about it.

 

[wagnerrp]

I think I've seen superconducting mentioned in the context of Mag-lev trains, too.

In fact, I think mag-lev could have all kinds of industrial applications.

You still need a magnetic track. The use of superconductors just means you’re passively stable in a loss-of-power situation.

Actually, I've got this really weird idea of using a superconductive shell around a package inside a matrix of magnets to ship something in a really low-impact scenario. Not sure that'd be practical, but I keep thinking about it.

Why not just use springs?

 

[cbum]

Fusion has been consistently 30 years away for the last 60 years

You are much too pessimistic. It's consistently been away for 20 years.
;-)

 

[Defenestrar]

If.
So far, that hasn't been conclusively demonstrated, let alone reproduced.


Ahem.

View attachment 60396

Hmm... so I'm torn between trying to explain this as a subtly suave xkcd reference or blame it on autocorrect. Couldn't have been a mistake while I was typing on the bus cause that wouldn't make me look good

 

[wagnerrp]

Actually, I've got this really weird idea of using a superconductive shell around a package inside a matrix of magnets to ship something in a really low-impact scenario. Not sure that'd be practical, but I keep thinking about it.

Why not just use springs?

You just quoted me admitting it's not practical, then question the practicality.

And you wonder why I'm so often critical of replies to me?

You're critical of people confirming your uncertain expectations?

 

[slugabed]

Power transmission and computers seem like the most obvious applications, any others that I am not thinking of?

If you can now route power and ground on printed circuit boards without having to use entire layers for each, PCB's will get smaller and (theoretically) cheaper to build. (edited out misinfo)

Looking at the bigger picture, if transformers, motors and generators become significantly more efficient, then we take a big chunk out of our electrical load and impact global warming as a result.

 

[inecatus]

If you can now route power and ground on printed circuit boards without having to use entire layers for each, PCB's will get smaller and (theoretically) cheaper to build. And (correct me if I'm wrong on this) superconductors cannot function as antennas, so signal lines will not radiate or operate as receivers, making circuits quieter and the world safer for AM radio.

Looking at the bigger picture, if transformers, motors and generators become significantly more efficient, then we take a big chunk out of our electrical load and impact global warming as a result.

You are somewhat wrong. Superconductors can still function as electromagnets (and in fact are extremely good at it) which also means they can act as EM emitters. It’s been a while since grad school so I can’t comment on their ability to be used to receive though.

 

[chabig]

You are somewhat wrong. Superconductors can still function as electromagnets (and in fact are extremely good at it) which also means they can act as EM emitters. It’s been a while since grad school so I can’t comment on their ability to be used to receive though.

Any moving charge will generate a field, and a field will move a charge. So a superconductor should function normally as both a transmitting and receiving antenna.

 

[slugabed]

Any moving charge will generate a field, and a field will move a charge. So a superconductor should function normally as both a transmitting and receiving antenna.

Yah, kinda goofed that up. Was recalling my old training on Smith charts which required a real impedance to calculate insertion losses and forgot that it's the impedance differential, not the impedance itself. As it turns out, supercondutor films are already in use as hyper-sensitive receiving antennas in some applications.

https://www.birmingham.ac.uk/Documents/college-eps/eece/research/EDT/antenna.pdf

 

[malor]

Another thought is that antennas made out of superconductors might be substantially better than the existing wire ones. At least hypothetically, wouldn't the signal to noise ratio for any particular antenna size and shape be the best possible, given the source's strength and distance?

 

[Machupo]

From a research perspective, posting this on the arXiv is a fantastic way to parallelize the search through the potential space of small crystalline deviations having large impact on superconductivity. Twitter has been like a turbocharged natural selection engine for research variability here.

 

[CenterLess]

Ubiquitous maglev, perhaps?

That would clear a technological barrier, but I have doubts it would change anything on the bureaucratic side.

 

[kanka]

I think The Journal of Immaterial Science has a great take on this.
It's about as convincing as the evidence we've seen so far.

I don't get the downvotes as this was hilarious - Journal of Immaterial Science, a beer-reviewed journal

Thanks for sharing!

 

[Oregano]

Fusion has been consistently 30 years away for the last 60 years

Sustainable fusion has been only 8.5 minutes away since the dawn of civilization.

 

[Wheels Of Confusion]

Sustainable fusion has been only 8.5 minutes away since the dawn of civilization.

Actually that's a non-renewable resource.

 

[malor]

Actually that's a non-renewable resource.

Then so is solar power, hydroelectric, and burning wood for heat.

 

[LibraryCommoner]

Correct me if I’m wrong, but I had learned that the great majority of heat generation in an IC is from the transistors, not the traces (on chip wires). More specifically, the heat generation is primarily from transistors when switching between high and low (either direction). CMOS doesn‘t consume much power when in steady state. This is why raising the clock speed increases the power consumption and heating.
There has been a growing issue of getting current to the the transistors as the trace dimensions fall, but that’s still not a major factor in heating.

After some googling I'm inclined to think you're right. As I understand things, when a transistor goes into a low state it effectively becomes a resistor thereby generating heat. I was thinking because the traces take up so much space that they must be contributing to the heat. Anyway ta for the fact check.

 

[Walker On Earth]

Then so is solar power, hydroelectric, and burning wood for heat.

Very true. When the term 'renewable' includes solar power, hydroelectric, and biomass, but doesn't include, say, deuterium or protium, then you've got to suspect an attempt to propaganize at the very least.

 

[Wheels Of Confusion]

Then so is solar power, hydroelectric, and burning wood for heat.

Yes.
Fortunately, at current rates of consumption we should be good for at least another 3 billion years with our proven reserves.

 

[plugh]

Plenty of conductive things in a chip other than the actual gate.

Actually, not that many. While complex, there aren't that many different structures in an IC. Conductive components consist of the gate, metal traces, and the transistor source and drain. Then there's the transistor channel, which conducts when turned on. The source, drain, and channel are semiconductor material (usually silicon) and probably need to remain so unless you're making a fundamentally different device. The only structure worth investigating would be the traces, and they're not the major source of heat and power loss in a digital IC.

This may be different in high power transistors (GaN and SiC). But I suspect the same is true there.

 

[CraigJ ✅]

I suspect that if this is actually true a race of two headed aliens will develop a bacteria to eat it once industry becomes reliant upon it.

 

[Wheels Of Confusion]

I suspect that if this is actually true a race of two headed aliens will develop a bacteria to eat it once industry becomes reliant upon it.

Hopefully by then we'll have stabilized the ring in orbit around the star and the blocking shields.

 

[Frozen Squirrel]

not really seeing levitation here - balancing yes contact point - but levitating not so much

To kinda offer proof - make some iron filings put them on a sheet of paper. Magnet under the paper and observe the orientation of the filings If you get enough iron filings you can make a mini porcupine.

Just skeptical about this is all I'm driving at

And final under educated thought - what would happen if there was a magnet above being held of course but same polarity as the upward facing magnet? Not interesting enough or curse my under educated brain.

 

[malor]

Yes.
Fortunately, at current rates of consumption we should be good for at least another 3 billion years with our proven reserves.

I think it's actually about a billion years until the Sun gets hot enough to cook the Earth. I just read that in "The End of Everything", by Katie Mack. Wikipedia supports her:

At present, it is increasing in brightness by about 1% every 100 million years. It will take at least 1 billion years from now to deplete liquid water from the Earth from such increase.^[136] After that, the Earth will cease to be able to support complex, multicellular life and the last remaining multicellular organisms on the planet will suffer a final, complete mass extinction.^[137]

I had thought we would last longer than that, but apparently not, unless we can move or shade the Earth.

 

[LibraryCommoner]

That's complicated.

On one hand, traditional design would require the discovery/invention of a "supersemiconductor".

In existing superconductor (low temp/high pressure) circuitry, logic can be performed using the same superconductive material by creating Josephson Junctions, which can replicate semiconductor component logic to some degree.

To some degree or another, even if this proves out to be true and large chunks of the scientific and manufacturing communities jump on the bandwagon (which is likely if it is, in fact, a real discovery), for something like a CPU/GPU the design and engineering would need to be overhauled and much of what is considered fundamental to package design today would be useless and we would be starting from scratch - educated scratch, but scratch all the same.

That said, the honest answer in the long term is "yes". We already know practically how both bus and logic could be recreated with superconductive materials. After a few years of development and testing, presuming that the same current applications for Si could be used for this material, one should theoretically get significantly (factor of 5 or more, at least) faster processors with significantly lower power consumption and virtually no heat dissipation. If one was to couple the superconductive material and traditional transistors, you'd see processors of about the same speed with SLIGHTLY reduced heat dissipation requirements and a moderately lower power consumption - so still a benefit, but probably not great enough to compensate for the costs of retooling, redesign, testing, et cetera.

I could see incorporating superconducting wires into traditional ICs as a way to gain experience with and do research on leveraging current semiconductor manufacturing equipment and techniques towards manufacturing superconducting logic.

 

[slugabed]

After some googling I'm inclined to think you're right. As I understand things, when a transistor goes into a low state it effectively becomes a resistor thereby generating heat. I was thinking because the traces take up so much space that they must be contributing to the heat. Anyway ta for the fact check.

Semiconductor technology is useful because we can turn the individual devices on and off. It's kind of like the spigot on the side of your house - if it gushed water all the time it wouldn't be all that useful, and if it never released any water it wouldn't be useful at all. So there's a handle available to turn it on and off.

Superconductors are great at conducting current, but unless we can figure out how to turn the conducting feature on and off, we won't be able to use them as switches (transistors).

 

[KBGB]

I can't speculate if it's real, or if there is anything of any value... South Korea and superconductor research both don't have stellar reputations, be we should be careful not to publish this lab for other's sins. Healthy skepticism, not biased skepticism.

If this doesn't pan out... this moment isn't unlike the cheap thrill of deciding what you'd do with your winning ticket before it becomes a losing ticket.

I do wonder what it must feel like in the lab to be on the verge of something with so much potential to shape the future & then believing you've done it. I bet their spouses knew something was up even if they didn't know what (else) it was.

 

[plugh]

I could be wrong about this, but perhaps it could be useful in inter-chip data transmission. A friend of mine is working on using photons instead of electrons to communicate between chips and the speed benefits are apparently quite significant if realized.

Maybe. I worked on some R&D in that area a long time ago (integrated photonics). A lot of the losses and transmission issues for inter-chip transmission is due to line capacitance, not resistance. I have no idea how a superconductor wire/trace would impact capacitance.

That's one of the big pluses of optical interconnects, no capacitance. The minuses are pretty much everything else: new materials, physical dimensions of optical structures, alignment/packaging...

 

[inecatus]

Semiconductor technology is useful because we can turn the individual devices on and off. It's kind of like the spigot on the side of your house - if it gushed water all the time it wouldn't be all that useful, and if it never released any water it wouldn't be useful at all. So there's a handle available to turn it on and off.

Superconductors are great at conducting current, but unless we can figure out how to turn the conducting feature on and off, we won't be able to use them as switches (transistors).

Look up Josephon Junctions. They aren’t transistors per se but you can make computers out of them that will function pretty much the same except for switching at hundreds of gigahertz. The question as to whether a room-temp superconductor would allow you to make a room temperature Josephon junction is not clear due to thermal noise though.

 

[slugabed]

Look up Josephon Junctions. They aren’t transistors per se but you can make computers out of them that will function pretty much the same except for switching at hundreds of gigahertz. The question as to whether a room-temp superconductor would allow you to make a room temperature Josephon junction is not clear due to thermal noise though.

Thank you for the tip - I was unaware of things like rapid single flux quantum. Learned a lot, but I had to climb out when the rabbit hole led to some challenges, such as how to reliably integrate things like single-electron devices and 1.0 attoFarad capacitors into LSICs. I'm out of my element here. Have fun y'all.

 

[Nilt]

"The perfect time to write an article on those results would be when they've been confirmed by multiple labs. But these are not perfect times."

In reasonable times, reputable publications did not permit science writers to publish hearsay, speculation, or other forms of gossip merely to generate ad revenue. But these are not reasonable times. Dwarf rapes nun! Escapes in UFO! Kowtow! Kowtow! To the Lord High Clickbait!

This is a load of horseshit. Ars clearly published this because there was already a bunch of bull being published by others. Articles such as this which clarify things other publications are spewing crap about is one of the major reasons why I pay for a sub here on Ars. Since nobody can be an expert in all things, I am not. I am, however, interested in a huge variety of different things. Many of those things are sufficiently complex that I have barely an infant's understanding of them which makes interpreting the sort of other articles this news caused difficult for me. Knowing Ars would publish something before too long meant I could just go, "Huh" and patiently wait. Now that this article has been written on it, I have a much better understanding of what's going on than I did before.

That's not clickbait. That's actually useful journalism.

 

[Veritas super omens]

Yes, this is a good way to get it peer reviewed quickly. It's also a way to get people rushing to attempt to replicate the results.

Question is, why do they want that to happen so quickly ?

Or is the behind the scenes drama over who gets credit causing the rush as a side effect ?

Rush to the record. Possibly targeting mass appeal. We should watch every move. Perhaps designed to decieve, ie orchestrate illusions. It might just be fantasy for sale...presto, that's entertainment.

 

[Veritas super omens]

Just FYI - lead paint is only really banned for residential use. Leaded gasoline is only really banned for cars and trucks. You can buy leaded gasoline for your airplane at thousands of airports across the USA, then spread atomized lead all over the place when flying over it.

Meanwhile the current understanding is that no level of lead exposure is safe.

 

[Veritas super omens]

You still need a magnetic track. The use of superconductors just means you’re passively stable in a loss-of-power situation.


Why not just use springs?

Too much water to deal with...

 

[Tom the Melaniephile]

Meanwhile the current understanding is that no level of lead exposure is safe.

Absolutely. I'm acutely aware of this ongoing problem.

 

[Voix des Airs]

Maybe. I worked on some R&D in that area a long time ago (integrated photonics). A lot of the losses and transmission issues for inter-chip transmission is due to line capacitance, not resistance. I have no idea how a superconductor wire/trace would impact capacitance.

That's one of the big pluses of optical interconnects, no capacitance. The minuses are pretty much everything else: new materials, physical dimensions of optical structures, alignment/packaging...

Yes. Capacitance is generally the issue.

[In this R is the distributed resistance, C the distributed capacitance, L the distributed inductance and G the distributed conductance. ω is the angular frequency]

If you are analyzing the traces as transmission lines then if R<<ωL (and assuming negligible leakage current, specifically G<<ωC) then the resistance can be ignored and the velocity of light in the line is given by v=1/sqrt(LC). This is usually the case in short lines and reducing the resistance doesn't really get you anything.

If it isn't the case then increasing R decreases the velocity, and worse, the line can become dispersive, i.e. different frequencies travel at different speeds down the line. This can be a real problem on very long high-speed transmission lines.

 

[thrillgore]

The fact we're once again as a society, gonna rely on a poison to move us up hundred of years is not lost on me.