August 8, 2023

Superconductors Sure Are Heating Up

Welcome to my relaunched newsletter! I used to send these every month, and then stopped because I was launching Small Wonders, the SFF magazine for flash fiction and poetry that I co-edit. Launching magazines takes a lot of work!

But now we're past the initial huge amount of work and into merely a large amount of work, so I wanted to get back to writing this newsletter. This one's a doozy, because the subject of room temperature superconductors heated up¹. For new folks, I don't normally go on at this length about science stuff. This issue's the answer to the question no one was asking, "What happens when social media gets excited about science that's vaguely adjacent to Stephen's thesis work?"

Hot Girl Room-Temperature Superconductor Summer

It's been quite a month for superconductivity thanks to LK-99. Feuding authors! Multiple academic and hobbyist groups racing to replicate the results! Theorists posting a link to her LK-99 paper with a gif of Obama's mic drop followed by a retraction of the mic-drop vibes! Fanfic-like extrapolations of what's going on.

This is total catnip to me. We've got physics that's in the neighborhood of my thesis work² combined with a story that's playing out on social media in ways that's leaving researchers bemused.

So what's the deal? Why are people excited about a possible room-temperature superconductor? Why this possible room-temperature superconductor?

A lot of it has to do with the right combination of drama, low-barrier-to-entry reproduction, and social media. Before I dive into the story aspects of this, though, let's figure out why people are interested in superconductivity in the first place and where superconductivity comes from.

Superconductors: They're the Future!

When you cool down a metal conductor like copper, its electrical resistance slowly drops. The lower a conductor's resistance, the less energy you lose when you put current through it. For some materials, once you get it colder than a certain critical temperature, their electrical resistance vanishes. They become a superconductor.

Superconductors make magnetic resonance imaging possible. Magnetic resonance imagers use superconductors to produce the magnetic fields they use to peer inside us because they're the best way to get the extremely high, extremely smooth magnetic fields that MRIs need. Regular conductors heat up too much to make a practical MRI.

There's other things you can do with superconductors. Hitting them with a brief magnetic field magnetizes them, and that magnetism goes away very slowly, so you can use them in place of traditional permanent magnets. You build confining magnetic fields in tokamak-style fusion reactors with them. Maglev trains? Sure, superconductors could help there as well.

To date, though, they're mainly used in MRIs because you have to get them so cold before they become a superconductor. MRIs bathe their magnets in liquid helium — expensive! dwindling supplies! — to keep them superconduting. Materials that become superconducting at a higher temperature than, like, 4 degrees above absolute zero, would let us do a lot more with superconductors than we do today.

How Do You Get Superconductivity?

I'm so, so sorry: we're going to have to talk about quantum mechanics. If you're only here to find out how we got from superconductivity to Russian soil scientists on Twitter, feel free to jump forward a section header or two.

Elementary particles have a quantum mechanical property known as spin. They act as if they're charged particles that are spinning. Note that word "act". They're not actually spinning! It's an analogy, the kinds that physicists love even though it means we have to spend a lot of time explaining why the analogy — that we chose! — is misleading. If I were you, I'd think of a particle's spin as a number that tells us something about how the particle behaves. It's like mass. A particle's mass tells us how hard it is to get it moving or slow it down. A particle's spin tells us how it behaves around other particles due to its inherent angular momentum.

Spin is measured in units of ħ, the reduced Plank constant. Spin can't be any random value, though. It has to be a multiple of ½ ħ. You can have a spin of ½, or 1, or 1 ½, but not 3/7. It's quantized. It's as if the USA still had a half-cent coin and your grande pour-over at a coffee shop cost $2.905 or $3.000 but not $2.907.

When it comes to superconductivity, we care about whether a particle has a half-integer spin, like ½ or 1 ½, or a whole number spin, like 1 or 2. Half-integer spin particles are called fermions. Particles with a spin that's a whole number are bosons.

Elementary particles are things like quarks and electrons. What happens when you've got matter, like an atom, that's made up of a bunch of those elementary particles? You add up all of their underlying particles' spins. If you stick two fermions together, you end up with a composite boson, because ½ + ½ = 1, a whole number³. Protons and neutrons, the particles in an atom's nucleus, are fermions. Helium-3 has two protons and a neutron, so it's a fermion, with a spin of ½. But helium-4, the most common kind of helium, has two protons and two neutrons. It's a boson with a spin of 0⁴.

Bosons and fermions act very differently when you get them cold. Bosons like each other. When you chill them down, they all gather in the lowest possible energy state. They're happy to share the same quantum mechanical state. And if you have a bunch of bosons cold enough that they're in the same quantum state, then they start to act like they're one big superparticle. If you do this in a gas of bosons, you'll get a Bose-Einstein condensate. Do this in a gas of bosons in 1995 and you can win the Nobel Prize in physics.

You don't have to use gases to create a Bose-Einstein condensate. If you cool helium-4 enough, it creates a liquid Bose-Einstein condensate. When you have a large mass of particles acting like one big super-particle, odd quantum mechanical effects occur at human scale. Supercold helium-4 is a superfluid. It has no viscosity. If you spin it in a bucket, it will keep spinning forever, or at least until it warms up enough that it's no longer a superfluid.

Unlike bosons, fermions don't like each other. Only one of them can be in a given quantum state. When you cool a bunch of fermions down, they each pick a quantum mechanical state and keep the other fermions out, throwing elbows so they can have their own space. Supercool fermions and all you get are very cold fermions.

But remember me talking about how you can combine fermions to get a boson? What if you found a way to pair up two very cold fermions? Then they'd be a boson and you could create the fermionic version of Bose-Einstein condensates.

It turns out you can do just that. If you cool helium-3 enough, to within a few thousandths of a degree of absolute zero, the atoms form what are known as Cooper pairs⁵ and turn into a superfluid.

The same thing happens with electrons, which are fermions. If you cool a metal enough, the electrons in it create Cooper pairs. The metal becomes superconducting. Just like a superfluid can flow forever because it has no viscosity, a superconductor can transmit electricity without loss because it has no electrical resistance.

Never Mind Metals, Give Me Ceramics

I'm done talking about quantum mechanics. From now on it's all material science!

During the 20th century, researchers kept trying to make higher-temperature superconductors. In the 1930s, the metal niobium held the record at around 10 Kelvin. In the 1950s, a mix of niobium and tin gave a metal that becomes superconducting at 18 K. Niobium-germanium raised that record to 23 K in 1973.

Georg Bednorz and Alex Müller, two researchers at the IBM Zurich Research Laboratory, changed that. Starting in 1983, they struck out in a new direction, testing various ceramic metal oxides. Those weren't obvious superconducting materials. Metal oxides conduct electricity, but not very well. But Bednorz and Müller persevered. In 1986, they showed that adding barium to lanthanum-copper-oxide crystals created a substance that became superconducting at 30 K.

Their discovery set off an explosion of research into ceramic superconductors. It was huge, so much so that they won the Nobel Prize in physics one year later. That's a ridiculously short time to go from discovery to a Nobel Prize. Researchers were convinced that ceramic oxides were the path to room-temperature superconductors. Just look at this graph of record superconducting temperatures over the years. The metal oxides are the light blue triangles. They show up in 1986. Three years later, they'd gone from being superconducting at 30 K to nearly 150 K!

) (source)

It was a gold rush, so much so that universities and labs hurried to patent their new materials even before they'd been proven to work. In 1988 DuPont paid the University of Houston $1.5 million to license their patent rights even before the patent was granted, with $3 million more to follow if they did get the patent. Of course, it immediately turned into a legal battle. I can only imagine what the social media feuds among those labs' researchers would have been like.

Look back at that graph above, though. Progress stalled out after 1989. The materials didn't march ever closer to being a room-temperature superconductor. We don't know why. In fact, we don't have a good theory to explain why metal oxides are superconductors. We suspect that the electrons are forming Cooper pairs, but we don't know for sure. The stacks of crystal lattices with their mix of atoms make for what my advisor would call a "non-trivial" system to analyze.

That's where we've been for over thirty years. The latest trend has been testing materials under high pressure. Care for a compound that superconducts at -23 °C? You'll need a pressure of over 2 million times the atmospheric pressure we live in.

Scan the Wikipedia page for room temperature superconductivity and take note of how often the phrases "not been corroborated," "yet to be verified," and "retracted" appear. Ranga Dias at the University of Rochester published a Nature article in 2020 on a near-room-temperature superconductor that was retracted in 2022 over his objections, only to publish a second paper in Nature in 2023 on a different room-temperature superconductor. Oh, and Dias has a second paper in the process of being retracted. And that didn't blow up on social media, but LK-99 did? What gives?

Why LK-99?

Here is the shortest summary about LK-99 I can give. A team of Korean researchers posted two papers about LK-99, a material that they claim is a superconductor at room temperature and pressure. There's drama from the jump, as members of the team are reportedly feuding, which is why there are two papers and not just one. More critically, LK-99 isn't that hard to make, materials-science wise. I mean, you only need one atmospheric pressure, and anyone can get that for free in the comfort of their own home. People decide to reproduce the results.

I cannot stress how many people try this out. Sure, you've got your usual labs, but you've also got Andrew McCalip, who works at a space manufacturing startup and realized he could make LK-99 himself. This led to a Twitch stream where a lot of people watched a kiln heat some chemicals.

https://twitter.com/andrewmccalip/status/1685153801127546880

Iris, whose Twitter bio states that she's a plant scientist in Russia, goes through the papers, reconfigures the synthesis methods, and eventually shows pictures of a floating sample of LK-99. Early reports claim that she's an anime catgirl, but:

https://twitter.com/iris_IGB/status/1686446426107035650

The Times regrets the error.

Much of the online chatter centers on the Twitter account of Alex Kaplan, a physics major turned frozen coffee startup guy, who posted early about LK-99. Throughout the LK-99 saga he's posted through it, documenting his rollercoaster of "it's over/we're back" emotions.

Why is this the dubious claim of room-temperature superconductivity that took off? You've got

1. a long-hoped-for but never-delivered scientific result that people think will change everything (remember DuPont's $1.5 million payment to license a non-existent patent)

2. that arrived with drama (feuding scientists from the same lab!),

3. that can be tested by non-academics,

4. has a lot of potential Main Characters (Russian scientist and catgirl! space manufacturing startup guy! coffee startup guy!) interacting with each other,

5. and that has the boom-and-bust cycle of optimism and pessimism as new results come out.

It's narrative, baby! And it unfolded over the course of a week at a breakneck speed mediated by social media.

That last bit is what's keeping the LK-99 cycle revving. Friday, the ever-insightful Max Read wrote about LK-99 and made a point that I wish I'd been clever enough to realize: the "it's so over/we're so back" cycle maps onto Blake Snyder's "Save the Cat" beat sheet, the Procrustean bed that Hollywood movies have been made to fit for years. It's a story form that we've encountered over and over and over again.

Currently we're at an "it's so over" trough, as researchers at Peking University made LK-99-like samples and determined that they exhibit ferromagnetism, not superconductivity. Realistically, that's always been the most likely outcome. Extraordinary claims, extraordinary proof, etc. etc.

But doesn't a part of you hope that it's not? That there's more to the story?

That's narrative.

What I'm Vibing With

Every newsletter I like to list things I'm enjoying, thinking about, or want to bring to your attention.

• If you don't yet know the source of the quote "Women are my favorite guy" then you owe it to yourself to watch the deliriously wonderful Planet of the Bass video.
• But be warned! The second clip features a new Ms. Biljana Electronica, and people have strong feelings about that.
• If you like fantasy, check out Simon Jiminez's wonderful novel The Spear Cuts Through Water.
• A Malört and Four Loko cocktail. I'm so sorry.
• I keep turning this essay on Barbie, queerness, and performative gender over in my mind. Super recommended.
• "Why would you say we'd train AI models on your faces and voices like our terms of service said we could?" Zoom wonders.

Finally, poet Richard Siken has opinions about his own fanfic.