Superconductivity offers tantalising changes from electricity to transport | Hacker News Reader

Superconductivity offers tantalising changes from electricity to transport

ft.com

67 pointsjonbaer28 comments

This article is kind of whack... Yes modern superconducting materials are getting warmer and warmer. No we're not at room temp now. No there's not a clear path to simulating band gaps for extended structure materials with the quantum computing..

No, but if you could make Variational Quantum Monte Carlo work better by starting with a good variational guess in your quantum simulator....

What would room temperature superconductivity actually buy society? I can think of a few things but I'm not sure about the implications.

* Long distance transmission is currently high voltage AC to work around losses. The whole distribution network would become efficient but would need reinvestment. This would offset the utility of local instant-response power generation.

* Things that need big magnets would get easier: MRI, particle accelerators, etc. Is this a big deal? Maglev trains would get more attractive.

* What else?

Tokamak fusion reactors need big magnets. Superconductor advances already have made big impacts on the feasibility of practical fusion in the near term, and true room temperature operation would only help.

Much more powerful magnets, so electric motors would get even smaller and more powerful.

Currently a big limitation on small electric motors is heat, so a fully superconducting motor could handle huge loads.

The LHC at CERN would be a lot easier to operate without requiring such an enormous amount of Helium as well as the long time it takes to cool down to superconductivity after maintenance.

Nuclear fusion.

The main problem today is superconductors' size. You need space for refrigerator circuits, that are not superconductors.

If superconductors operational temperature are high enough, you could just refrigerate the entire room. Making superconductors more dense and powerful. Making fusion possible in extremely small spaces and making it even cheap.

That is you could do fusion expending millions instead of billions.

Low voltage distribution networks in data centers. Forget 48v - you could centrally distribute 3.3v power for CPUs. 1000s of 80% step down converters become 100% efficient.

Superconducting ring storage of electricity.

Probably some break throughs in electrical storage because SQUID devices can now be commodity parts in hard disks.

> Low voltage distribution networks in data centers.

3.3v (or lower, for modern chips) requires big cables. I doubt that's worth it.

> 1000s of 80% step down converters become 100% efficient.

How does this happen? Are you excluding the efficiency of the central step-down converter?

> Superconducting ring storage of electricity.

Can you actually fit a meaningful amount of energy in there before it saturates?

Big cables not needed as superconductors can carry more current per unit cross sectional area than copper.

Even one central step-down converter could be 98% (or whatever) efficient, plus you could actually generate power at 3.3V if you wanted to, making step-down not required.

As far as storage of energy, yeah, you can store a lot of energy in a superconducting ring. It is a viable energy storage method.

THAT ALL SAID, the physics of superconductivity mean that even if you ARE able to get room temperature superconductivity to work, you can get even BETTER critical current (i.e. current per cross section area that can be handled while maintaining a superconducting state) so you'll still probably want to cool the cables in some applications, and for a certain current carrying capacity, you'll probably want to use a higher voltage than just 3.3V, even if you don't have any voltage drop, because that would increase the power that can be transmitted for the same cable investment.

Not only is it a viable energy storage method, it's already used commercially (predominantly in power conditioning systems).

Hmmm, wonder what the cascade failure of a "full" superconducting ring heated past it's superconducting point would look like?

Maybe impressive fireworks/explosion?... from a safe distance. :D

Good question! The thing you're referring to is "quench." Answer: https://en.wikipedia.org/wiki/Superconducting_magnet#Magnet_...

Thanks, that's interesting info. It's good there seems to be well established ways to detect and cope with this. :)

Room-temperature levitation with permanent magnets.

A diamagnetic material repels external magnetic fields. Unlike other materials, superconductors have very high diamagnetic constants. [1] This is why one can float a magnet over a superconducting plate (or float a superconductor over a magnet).

If we had room temp superconductors, we could tile a floor with them, mount magnets on carts and float the carts across the floor. Or highway. Or whatever. Lots of uses for levitation that consumes no power.

[1] https://en.wikipedia.org/wiki/Diamagnetism

The improvements already in superconductors over the past decades have already made a fusion reactor that is much smaller than ITER a possibility.

A group primarily from MIT proposed such a reactor called ARC:

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

Personal quantum computing, assuming the press releases about the lastest breakthroughs are right about them being made of superconducting elements.

Assuming the price, mass density, critical magnetic field density, and critical current density are all good, you could launch satellites magnetically against Earth’s magnetic field (but only from the poles and with some cute tricks for efficient horizontal motion).

It depends really. High temperature superconductors were incredibly hyped when they first came out, but the problem is they had a limited critical current and critical field strength so applications were(and to some degree still are) limited. These are the maximum current and magnetic field a superconductor can carry before superconductivity breaks down. If room temperature superconductors turn out to have a high processing cost and the critical current/field strength isn't that high, then the applications are going to be limited.

> Long distance transmission is currently high voltage AC to work around losses.

High-voltage DC actually has less losses.

(Why? Because a high-voltage AC signal is actually low-voltage two times per cycle)

Vanilla DFT is still a ground-state theory, last time I checked, which means it can't tell you the things about the Fermi surfaces you would most like to know for superconductors. I can't speak for quantum simulation.

I wonder, if cooper pairs can be made with ions?

As a rule, the heavier your particles, the harder it is to get quantum mechanics behavior out of them. You won't see atoms forming Bose-Einstein condensates at anything you can cool with liquid nitrogen.

http://archive.is/gE8kw

Haven't carbon nanotubes already been shown to exhibit ballistic conduction at room temperature?

Ballistic conduction is not the same as superconductivity.

Superconductivity relies on Cooper pairing, i.e. charge carriers that are (relatively weakly) bonded together such that the even smaller interactions with imperfections (including just thermal dislocations) in the conducting medium are even smaller than the minimum energy needed to break the Cooper bond. Until this minimum energy is achieved, then the Cooper pair is quantum mechanically prohibited from losing energy (to resistance/heat) from these small imperfections, similar to how an electron is quantum mechanically prohibited from swirling into the nucleus (losing energy in the form of electromagnetic radiation due to traveling in a circle, i.e. accelerated, as classically one would expect) from the ground state.

I know the differences between the two, perhaps I should've elaborated more. My point was that other than the Meissner effect, from a practical standpoint ballistic conduction is just as useful in terms of electron mobility.

Ballistic conduction does raise the possibility that the bulk electrical resistivity of carbon nanotubes or graphene (etc) could be lower than (say) copper, but this is not practically as useful as superconductivity. The Josephson Junction (which is at the heart of SQUIDS--used for measuring extremely small magnetic fields--and also superconducting qubits--probably the most promising of the current quantum computing technologies) relies on the quantum mechanical nature of the phenomenon.

Because of Cooper pairing, you can avoid resistance even at the interface between one superconductor and another, thus allowing the complete cessation of all resistance. This allows you to store energy in the form of continual circulating current with no energy input, something that wouldn't be practical with merely much lower electrical resistance enabled by ballistic conduction (which still has significant tube-to-tube resistivity).

So I'd say that probably most practical applications of superconductivity are not transferable to ballistic conduction. Still nice to beat copper in conductivity (which hasn't happened with carbon nanotubes and graphene except for microscopic samples or with corner cases like high temperatures, high frequencies, etc).

Fair enough, I wasn't thinking about the quantum computing side :)

Reply on news.ycombinator.com