Metalessons from the LK-99 Saga

In case you have been avoiding the internet for the past few weeks, the world has been abuzz about a purported room-temperature superconductor called LK-99. As of August 11th, the consensus about LK-99 seems to be trending towards it not being a room-temperature superconductor. (See Alex Kaplan here and this paper from the this paper from the Schoop Lab.)

Normally I try to avoid weighing in on The Current Thing. However, there are several lessons about materials, manufacturing, and progress that the LK-99 saga illustrates, regardless of how it concludes:

Materials underpin civilization. Potential room-temperature superconductors are exciting for a reason: they could revolutionize everything from transportation (extremely efficient electric motors with only a few parts; fusion-driven single-stage-to orbit space planes) to energy (drastically less expensive storage and transmission) to our ability to probe the structure of the universe. These applications, in turn, could have second-order effects that address some of the world’s biggest problems: loss-free transmission would make the deployment of renewables drastically easier. But the thing is that superconductors are not unique in their ability to have such wide-ranging impacts. Unlocking new materials consistently has wide-ranging second-order effects: lasers, carbide, plastics, semiconductors, aluminum, steel, fiber optic cables. Life as we know it would be impossible without figuring out these powerful new arrangements of atoms and how to produce them at scale. It will be easy to forget this fact as the saga fades into the past, regardless of its outcome.  

It takes a long time for materials to get from lab to market. Even in the best-case scenario where LK-99 was a room-temperature superconductor and had properties that make it useful for things that we want to use superconductors for (not guaranteed) and there was a way to manufacture it at scale, it would still take decades to find its way into products. Figuring out how to go from micrograms of an impure, non-uniform lump of rock made inconsistently by hand to tons of product with consistent properties is hard. It can take more research to figure out how to manufacture something at scale than it did to invent it in the first place. Even in successful cases, it usually takes decades for a material to go from invention to widespread use.

A table of time from invention to widespread use from the Feb/March 1995 Issue of MIT Technology Review.

It’s hard to capture the value created by materials research. Even if LK-99 was a legit room-temperature superconductor and Lee + Kim patented it, it’s unlikely they would become rich. There are a few reasons for this:

• Patents last at most 20 years, and as we noted above, it takes materials 20-50 years to reach widespread use, so by the time LK-99-based products were penetrating the market, the patent would have run out.

• Even if LK-99-based products got to market quickly, there are so many people who know how to make it that Lee + Kim would be bogged down in lawsuit hell. Large organizations could delay indefinitely. Complicating the matter, other organizations would likely invent other pieces of the system it would take to make LK-99 useful like processing steps, better formulations, etc. It would take the massive resources of a large corporation or specialized organization to enforce the patent claims, leaving the inventors with a dismal small share. (For a good example of these dynamics, see the story of Gordon Gould in Laser.)

• If the patent owners did manage to stifle everyone else’s use, it would drastically limit the amount of use and innovation around LK-99. Arguably, 3D printer development was drastically slower than it would have been until original patents on the process expired.

• The vast majority of the value of an invention tends to go to the company that commercializes it, and the chances that L+K also have the acumen to build a successful company are low.

Live, fast, open research is fascinating, exciting, and potentially useful. A lot of the work to replicate and understand LK-99 was done quickly and many people live streamed+tweeted their process. People made public mistakes, published failures, and highlighted a lot of uncertainty. This is not how most research is done! 

There are many good reasons that most research isn’t done on Twitter: people can be vicious towards still-developing ideas, most work is less exciting to the outside observer, research shouldn’t become an eyeball-grabbing game like many other domains, and it’s just a lot of work to both do the work and livestream the work. That being said, hopefully that the LK-99 experience can nudge us towards more fast, open research on the margin.   

It’s usually much harder to recreate an experiment. One unusual aspect of LK-99 is how easy it is to synthesize. It required common equipment and materials that you can order online. Many discoveries today require rare, expensive, or entirely custom equipment that means only a few labs in the world can replicate them. The two upshots of this fact are 

1. As nice as it would be, it’s unreasonable to demand that every result be replicated quickly.

2. Driving down the price of scientific equipment isn’t particularly profitable, but is extremely valuable for the world.

Beware hype boom-busts. The tagline of the LK-99 saga could be “We’re so back/It’s so over.” People rapidly flipped between visions of an imminent flying-car future and nothing changing whatsoever. You see this sort of attitude everywhere, especially in Silicon Valley, from whence it leaks into other parts of culture. While we should absolutely be excited about potential, the problem with creating the impression that a discovery means that the future is imminent is that it ignores that resource-intensive slog it will take to create that future. The more excitement about an imminent future, the greater the disappointment when it doesn’t happen, which can lead to people pulling resources or changing the directions of their work. The key to building the future is consistent optimism with realistic, buffered expectations. Most impactful technology isn’t just the result of fevered sprints between milestones, but decades of consistent effort. 

Original image from Tyler Angert

A diversity of institutions creates a healthy ecosystem. All sorts of players played important roles in the LK-99 saga: the idea originated at an international institution, had interest driven by folks at private companies, pseudonymous individuals on the internet, and national labs, with people at name-brand American academic institutions closing the book on it (or at least finishing this chapter.) While “meme science” (as Tim Hwang calls them in his excellent piece on this same topic) might criticize institutional science for killing ideas and the institutional science might criticize the meme science for being irresponsible and playing fast and loose, the fact is that this is what progress looks like. Without the fast-and-loose enthusiasm of the memers, institutional science would probably have just ignored the whole thing. Now we collectively have more knowledge which (might) be leveraged to eventually get powerful new materials.  

Magical materials are possible

I think Casey put it very well:

Original Tweet here

Weird phenomena can lead to amazing technology. You can think of technology as humanity harnessing different physical phenomena — from how electrons move in a semiconductor to how hot gasses expand. Many phenomena are not predicted by theory, and as Thomas Kuhn pointed out, new theory often comes about by observing phenomena that don’t fit existing theory. While it now looks like current theory can explain LK-99 behavior, it highlights the importance of the opportunity to say ‘huh, that’s funny.’ 

These types of observations can come about in two broad ways: 

1. New observational tools can get into the hands of a few people (like the Hubble telescope, or the Large Hadron Collider)

2. Old tools can get into the hands of many people, like telescopes becoming cheap enough that amateur astronomers can proliferate.

You could call these two modes “frontier-pushing” observations and “democratized” observations. Both have been historically important. 

Exploring new physical regimes (temperature, pressure, etc) consistently unlocks “huh, that’s funny” observations. We’ve done a great job of pushing into the physical regimes we can access on earth, but space is a giant, extremely cold vacuum with zero gravity …  

Finally, while we’re probably not going to have antigravity any time soon, this is actually how progress happens! Someone observes some weird phenomenon, other people try to observe it, others try to explain it. It doesn’t work, but interested parties continue to tinker. Maybe the whole line of work dies but gives rise to another that yields results. Someone else spends decades optimizing and scaling it. Someone else spends another decade figuring out how to use it. It’s slow and messy, rarely creating a good story until long after the fact. But with patience and the right institutions, there is a lot more wonder to unlock in the world. 

Thanks to Jessica Alfoldi, Casey Handmer, and Sarah Constantine for reading drafts of this piece.