[HN Gopher] Exotic new superconductors delight and confound
___________________________________________________________________
Exotic new superconductors delight and confound
Author : ernesto95
Score : 216 points
Date : 2024-12-09 15:47 UTC (5 days ago)
(HTM) web link (www.quantamagazine.org)
(TXT) w3m dump (www.quantamagazine.org)
| jmward01 wrote:
| The great thing about watching advances in superconductors is
| that any day we could discover the first true practical room temp
| superconductor and that one day changes the world immensely. I
| personally think we are likely to find one in the next 5-10
| years, but that estimate is based on nothing but hope and
| optimism on my part.
| NullHypothesist wrote:
| Better scientific understanding of the underlying process (as
| highlighted by this article), better pattern detection (AI),
| and better simulation capabilities (quantum computing) all
| point to accelerating progress on this front.
|
| What makes me particularly optimistic is the wide range of
| scenarios in which superconductivity is observed (also
| highlighted in the article); different mechanisms leading to a
| similar result suggests much better opportunity for the
| existence of a room-temp SC than if it were a highly similar
| pattern.
|
| Certainly such a discovery has some serendipity and luck baked
| in, but given these advances across the board, 5-10 years seems
| like a reasonable bet (then another decade or two to widespread
| adoption). Let's hope we don't blow everything up before then.
| fuzzfactor wrote:
| From the final paragraph:
|
| _experimentalists are still the ones leading the way.
| "Everyone's rushing as fast as they can," Yankowitz said._
|
| In my experience, the final line is what contributes to my
| optimism most:
|
| _"I can't believe that we're six years in and you can't take
| a break."_
| dtquad wrote:
| For the most valuable applications it is also "good enough" to
| find a superconductor that can be cooled with cheap liquid
| nitrogen and retain the magnetic field tolerance, current-
| carrying capacity, and thermal stability of a superconductor
| cooled with expensive liquid helium.
|
| Some so-called "high temperature" superconductors begin
| superconducting at liquid-nitrogen temperature or higher.
| However in real life applications like MRI and particle
| accelerators it turned out that they still need to be cooled
| with much colder liquid helium to get the desired magnetic
| field tolerance, current-carrying capacity etc. Finding a high-
| quality liquid-nitrogen-grade superconductor with these desired
| properties would be a revolution in itself.
| BariumBlue wrote:
| I know we've got cuprates, superconductors formed with copper
| oxide, useful up to 133Kelvin, higher than Nitrogen cooling's
| capability of 77K.
|
| I've read of them being used in wind turbines and particle
| accelerators, as well as concepts for fusion reactors.
|
| Your comment makes it sound like they have insufficient field
| tolerance / current characteristics though. I don't think
| I've heard about Cuprates at all recently.
| throwup238 wrote:
| Cuprates are also brittle ceramics so they're difficult to
| shape and larger pieces and assemblies tend to run into
| issues with grain boundaries that interfere with
| superconductivity, so there's a lot of practical issues.
| The classic superconductors are very low temperature but
| are much easier to cast.
| FredFS456 wrote:
| There's a few fusion startup companies using
| superconducting ReBCO tape to make their magnets. From
| the results that they're getting, I think we've largely
| cracked the engineering problems of making "wire" from
| ReBCO and it's largely just a scaling game now. I do want
| to point out that they're still cooling with liquid
| hydrogen at ~20K for high current capacity though.
| cyberax wrote:
| I invested in a couple of fusion startups (because why
| the hell not?), and the word on the street is that
| flexible tapes/wires are still not a solved question.
| Nobody can make them with consistent quality in large
| enough batches.
| NL807 wrote:
| >and retain the magnetic field tolerance, current-carrying
| capacity
|
| Even if it doesn't meet those requirements, room temperature
| superconductors will have immense value in low-power
| applications, micro electronics, sensing, etc.
| dtgriscom wrote:
| Data point: liquid nitrogen is cheaper than milk.
| jmward01 wrote:
| How interesting..... Of course liquid nitrogen + milk (with
| one or two other things) is worth more than the sum of its
| parts.
| nharada wrote:
| Yeah, haven't you seen how much they charge for Dippin
| Dots??
| saagarjha wrote:
| I mean it's literally selling air :P
| nick3443 wrote:
| How can we take advantage of this to reduce my grocery
| bill? :)
| nine_k wrote:
| Mechanical properties are very important, too. Able to bend,
| able to be cut, able to withstand mild shocks and vibration
| without fracturing. Something you can make actual _wires_
| from, at least at some stage. This, superconducting at liquid
| nitrogen temperature, would produce a revolution.
| Transmission lines alone would be huge.
| anyfoo wrote:
| > Transmission lines alone would be huge.
|
| Or small! :D
| nine_k wrote:
| But either way freezin' cool!
| popol12 wrote:
| How would we cool transmission lines to liquid nitrogen
| temperature?
| yencabulator wrote:
| https://www.youtube.com/watch?v=3gFUwGedayw
| jmyeet wrote:
| > Transmission lines alone would be huge.
|
| Transmission lines are a great example of competing
| demands. Copper is a better conductor so why do we use
| aluminium? Because of weight. And weight is a huge factor
| in supporting large cables over long distances.
|
| Metals are also ductile, which is important for a cable to
| hang in gravity under its own weight, be moved by the wind
| and so on. Assumedly exotic crystals wouldn't have this
| property. Even if they could, what would the weight be?
| Would the cross-section need to be much larger? Particular
| to this family of superconductors, tungsten isn't exactly
| the easiest thing to do deal with, particularly on a
| massive scale.
|
| There's an interesting Reddit thread about this topic [1].
| One issue it raises is we'd need to essentiaally rebuild
| our entire infrastructure and transformers are a big part
| of that.
|
| Personally, I think energy is going to get an awful lot
| more _local_. Solar is our future (IMHO). The ability to
| store excess power generated during the day and then use it
| when it 's dark or cloudy will obviate the need to
| expensive long-line transmission infrasturcture from
| distant power plants.
|
| Lastly, the GP is correct: liquid nitrogen is _incredibly
| cheap_. It 's basically the cost of drinking water. Getting
| something we could use at liquid nitrogen cooling
| temperatures would be incredibly impactful.
|
| [1]: https://www.reddit.com/r/AskPhysics/comments/12etlkr/w
| ouldnt...
| nick3443 wrote:
| Cheaper batteries meeting the storage requirements for
| renewables would be a game changer!
| BurningFrog wrote:
| Maybe superconducting transmission lines need to be
| underground
| kulahan wrote:
| >to get the desired magnetic field tolerance
|
| Interesting, then, that the article actually mentions a
| superconductor that got _stronger_ in the presence of a
| magnet! Wild things we 're learning here!
| Panzer04 wrote:
| For the most valuable applications, sure - but a lot of the
| biggest historical advances came not from the first
| application of that new tech, but from that new tech becoming
| cheap and ubiquitous.
|
| Maybe (cheap) room-temp superconductors won't actually be
| that interesting (maybe it just increases energy transmission
| efficiency 5-10%), but perhaps it's availability catalises a
| whole range of new applications that were never considered
| before.
| raelming wrote:
| If room temperature superconductors existed that were
| practical for transmission scale wires, it would be
| possible to make a global superconducting electrical grid.
| The middle East could fill their desert with solar panels
| and export that electricity to Europe. The sunny side of
| the planet could export solar power to the nighttime side
| of the planet. You could make electrical motors out of it a
| fraction of the size of current motors, like 1000 of
| horsepower out of a soda can, which would revolutionize the
| design of pretty much everything in the world with moving
| parts. Mechanical transmissions would go the way of the
| vacuum tube. Any vehicle or ship that's not completely
| electrified would be hybrid electric at least. You could
| also miniaturize transformers which has a lot more
| implications for making everything that uses a lot of power
| to be much more powerful and cheaper.
| AlexErrant wrote:
| I'm not a believer in that timeline. There's a large distance
| between superconductivity in the lab and commercial application
| (since you specified "changes the world immensely").
|
| E.g. MRIs still use NiTi (critical temperature of ~10 kelvins),
| discovered in 1962, for a number of reasons (this is in spite
| of MgB2 having a critical T of ~39k, ReBCO with a critical T of
| ~90k, and BSCCO with a critical T of ~108k):
|
| > In this paper, we analyze conductor requirements for
| commercial MRI magnets beyond traditional NbTi conductors,
| while avoiding links to a particular magnet configuration or
| design decisions. Potential conductor candidates include MgB2,
| ReBCO and BSCCO options. The analysis shows that no MRI-ready
| non-NbTi conductor is commercially available at the moment. For
| some conductors, MRI specifications will be difficult to
| achieve in principle. For others, cost is a key barrier. In
| some cases, the prospects for developing an MRI-ready conductor
| are more favorable, but significant developments are still
| needed. The key needs include the development of... [omitted]
|
| https://pmc.ncbi.nlm.nih.gov/articles/PMC5472374/
|
| Unfortunately, it probably won't be as simple as "step 1
| discover material, step 2 manufacture, step 3 profit".
| jeffparsons wrote:
| One exciting thing that could still happen in the shorter
| term (if/when a promising novel "high temperature
| superconductor" is confirmed) is an explosion of investment
| into research in the area. So even if it takes decades for
| Material X to end up in MRI machines, there would still be a
| steady stream of juicy progress to read about while we wait!
| FredFS456 wrote:
| There's some fusion startups successfully making >10T magnets
| using ReBCO tape now, so hopefully things will scale up/cost
| down enough to be used in MRIs.
| IAmNotACellist wrote:
| Too soon. I still tear up over LK-99
| tedsanders wrote:
| I'd bet that the first room temperature superconductor does not
| "change the world immensely."
|
| Resistive losses are just one of very many attributes of a
| conductor. Others important attributes include:
|
| - current capacity (will you need humongously thick wires to
| match charge carried by aluminum or copper?)
|
| - ductility (can it be formed into wires cheaply?)
|
| - cost (does CapEx outweigh electricity savings? is it
| expensive enough that people will cut and steal it?)
|
| - weight (can it hang from power poles? can it be transported
| on the backs of trucks?)
|
| - temperature sensitivity (does it crack at low temperature?
| melt at high temperature? change electrical properties
| depending on the weather? stop conducting on hot days?)
|
| - chemical stability (will it oxidize over a 50-year
| lifecycle?)
|
| - toxicity (will kids be poisoned if they touch it / eat it?)
|
| - machinability (can it be formed into tiny wires? can it be
| patterned onto chips?)
|
| - electromigration resistance (will the material break down
| over time from carrying charge?)
|
| - tensile strength (can it be hung from power poles at their
| current spacing? would we need to rip out all power poles
| across the planet? would we need more expensive underground
| lines?)
|
| - abundance in the Earth's crust (will the price skyrocket if
| we suddenly need to produce an annual megaton to replace the
| world's powerlines?)
|
| - geographic concentration (are the primary deposits
| concentrated in a single country, introducing potential supply
| chain and geosecurity risks?)
|
| - etc.
|
| It's very likely that the first material which does better on
| resistivity is going to do worse on these other dimensions.
| Resistivity is rarely the number one criterion in selecting
| conductors, from power lines to computer chips.
|
| One of the reason incumbent technologies are difficult to
| replace is that they win on criteria that are less salient to
| potential innovators. Aluminum is a common metal for power
| lines not because it has the lowest resistivity, but because
| it's by far the best we've got when evaluating this whole
| portfolio of needs.
| SequoiaHope wrote:
| Well lots of these limit the applications but something could
| have these limitations and still be a commercial success in
| some valuable niche fields.
| skeaker wrote:
| I think even if it's only usable in niche scenarios it would
| still be a commercial success. Superconductors are useful
| even when they're hard to use, as seen with existing cold
| temperature superconductors which a room temperature one
| could certainly replace and become a commercial success in
| doing so. With that said, you're likely correct that the
| first attempt won't catch on for mass adoption.
| zardo wrote:
| > as seen with existing cold temperature superconductors
| which a room temperature one could certainly replace and
| become a commercial success in doing so.
|
| High temperature superconductors are only seeing commercial
| use in the last few years (after their discovery in the
| 80s) due to issues like poor ductility.
| ethbr1 wrote:
| Electromagnets.
| EasyMark wrote:
| Why is room temperature important, what if "the world's first
| superconductor at 0C" was the headline?
| jmward01 wrote:
| I agree, 'room temp' isn't exactly precise. Even if it could
| work at 'room temp' if it didn't work just above that it
| would be too finicky for a lot of applications. Somewhere
| around there is likely the 'magic' point for a huge number of
| applications that will change the world. The perfect would
| be, of course, one with a very broad range of environments it
| could perform in. I think we are on track to figure out how
| to make something truly sci-fi in its properties: Strong,
| wide range of temp/conditions, flexible, etc etc. Once we get
| the theory perfect I bet the world will really open up.
| scheme271 wrote:
| I don't think room temperature is important per se. Rather
| it's shorthand for a superconductor that functions without
| special cooling. Even if the temperature range is somewhat
| limited, it's still possible to be pretty useful (e.g. deep
| ocean temps are fairly cold and stable so a superconducting
| cable to power repeaters on fiber optic cables might be
| useful).
| tizzy wrote:
| Is their existence just speculation or is there mathematical /
| empirical observations that suggest they exist we just don't
| know about them? Is it guaranteed, or highly likely, that one
| exists we just haven't found it?
| Ygg2 wrote:
| I don't think we'll really see a superconductor that ticks all
| the boxes.
|
| Reality is too noisy for such effects to take hold. If there
| was I think evolution would have already used it by now.
| philipkglass wrote:
| Do you mean biological evolution? There are a vast number of
| useful materials that terrestrial organisms cannot evolve
| because they're incompatible with the chemical/physical
| conditions inside living things. Organisms never evolved the
| use of aluminum structures, for example, even though aluminum
| has many useful properties and is more terrestrially abundant
| than carbon.
| t0lo wrote:
| I don't think we need more change right now. We're struggling
| to make a healthy society with meaning with the leaps in
| technology of the past 15 years in ML and computing and
| manufacturing power/ actual power
| kurthr wrote:
| One thing that goes unmentioned about room temperature
| superconductors is that they store energy as well. U =
| (B2/(2u0))V. 2u0 is about 2.5E-6N/A2
|
| So a 1m3 7T magnetic field would be about 20MJ or 7KWh. That
| doesn't sound like much, but collapse times could be microsec to
| generate GW of EM.
| xbar wrote:
| This is a useful fact to have handy.
| justlikereddit wrote:
| Commercial MRI quenching doesn't result in explosions and
| there's 7T research MRIs so I assume the issue with how to
| bleed off energy is solved in a safe enough manner already
| gaze wrote:
| These novel new superconductors are generally considered to be
| "bad superconductors." They have really low critical fields.
| Animats wrote:
| Wow. Go over to [1] and read the papers. This is good stuff. When
| someone finds new physics, interesting things result.
|
| Tungsten disulfide/boron nitride superconductors? That's a new
| direction.
|
| This article describes a new research result as a new research
| result, not as "trillion dollar industry by 2027". That helps
| credibility.
|
| [1] https://physics.mit.edu/faculty/long-ju/
| simpaticoder wrote:
| _> This article describes a new research result as a new
| research result, not as "trillion dollar industry by 2027"._
|
| I don't like that language either, but it may be wise to
| understand that different audiences are reading this, and it
| may be effective for the author to reach the others in this
| way. And besides, in general it's easier for a rationalist to
| ignore such language than it is for an industrialist to add it.
| exmadscientist wrote:
| Bollocks. "Hype language" has a very strong correlation with
| people who don't know what they're talking about. (Probably
| because most people who use it do not, in fact, know what
| they're talking about, even if some might.) So other experts
| in the field _will_ look down on you if you speak like a
| British university press release.
| simpaticoder wrote:
| I agree with you. I just think it is wasted effort to
| complain about the quality of press releases. I am
| consigned to their poor quality, and my solution is to read
| primary sources.
| nomel wrote:
| As someone who knows nothing about this, how is something like
| "Tungsten disulfide/boron nitride" selected? Is it based on
| some models? Or, is it more of a random walk?
| mensetmanusman wrote:
| They are choosing from a subset of materials that can form
| stable 2D crystals in order to test effects of relative twist
| angles on their energy bands.
| halflife wrote:
| Are 2d crystals the depth of 1 molecule?
| mensetmanusman wrote:
| One layer of somewhat coplanar atoms.
| gaze wrote:
| They're two materials that 2D materials people commonly
| stock. BN is thought to be a pretty innocuous insulator.
| Might as well be the lettuce of the sandwich. However, it's
| now showing that it has effects on the nearby layers, so
| people are playing with it in heterobilayer devices.
|
| A large part of the field of 2D materials is just trying
| stuff.
| m463 wrote:
| otoh, "a graphene device produced a mythical form of
| superconductivity"
|
| :)
| metalman wrote:
| room temperature superconductors are inevitable, all matter is
| conductive and photo active in some way or another,and conduction
| is one of the non optional components of reality, so the number
| of possible compounds that might superconduct is huge and now
| that there apear to be multiple mechanisms that superconduction
| can function from, means that the search will begin in earnist
| its a multi trillion dollar app
| zardo wrote:
| Is a 400C superconductor also inevitable?
| bawolff wrote:
| Given enough pressure it already exists!
| EasyMark wrote:
| Why is it inevitable? Is there some physics principle that
| proves they should exist even they're beyond current scientific
| knowledge?
| steveoscaro wrote:
| Question - is this a field ripe for breakthroughs using advancing
| AI capabilities? Or not likely because LLMs haven't ingested much
| data that could help reason in this field (or whatever reason)?
| MrLeap wrote:
| Could this "reality search" stuff be parallelized?
|
| Could reaction permutation and property testing occur in a more
| automated fashion than is currently?
| spullara wrote:
| I wish all the smart physicists were working on this instead of
| string theory.
| poincaredisk wrote:
| Why not both? We need people working on moonshots in hope of a
| breakthrough that will leap us forward. Even if, in case of
| string theory, we probably both agree that the chance is slim
| at this point (but I'm not a physicist, so nobody asked my
| opinion and nobody should listen to it).
| BoneZone wrote:
| I worked in Cory Deans lab when he did a brief professorship at
| City College. He is the most sharply intelligent person I have
| ever worked with, a savage experimentalist always devising new
| ways to experiment in nanofabrication and his theoretical
| curiosity is boundless. Additionally, he's a really nice Canadian
| when he goes to the pub!
|
| If you ever wonder why these products using graphene aren't
| commercially viable, it is insanely difficult to work with and
| prepare. Imagine trying to make a sandwich that's 5x5 microns in
| area and about 2-3nm thick. Graphene is essentially atomic tissue
| paper subject to all sorts of contamination and small scale
| effects.
___________________________________________________________________
(page generated 2024-12-14 23:01 UTC)