Three years ago, physicists discovered that two stacked sheets of carbon with a tiny, 1.1-degree twist between them could exhibit a dazzling array of behaviors. Most famously, when cooled to low temperatures, the material conducts electricity with zero resistance.
Researchers raced to figure out why twisted bilayer graphene (as it’s called) becomes a superconductor, with a form of superconductivity that seems unusually robust. Many theorists hoped the discovery would rewrite their understanding of superconductivity, and perhaps even allow researchers to engineer materials capable of sustaining the phenomenon at higher temperatures.
But the intense focus on that twist between the graphene sheets may have been a case of misdirection. A team of physicists announced today at an online conference that they’ve observed superconductivity in a triple-decker stack of graphene with no twists at all. The discovery, led by Andrea Young and Haoxin Zhou of the University of California, Santa Barbara, could reset discussions about superconductivity in graphene. It has led some theorists to suspect that graphene’s superconductivity is the vanilla variety after all.
“That’s a very important discovery showing that superconductivity [in graphene] is, in some sense, regular,” said Sankar Das Sarma, a theoretical condensed matter physicist at the University of Maryland who was not involved in the research.
But the evidence for conventional superconductivity is not conclusive. And researchers note that twisted graphene’s superconductivity could still be exotic even if untwisted graphene’s isn’t.
Albert Einstein, Richard Feynman and Werner Heisenberg are just a few of the titans of 20th-century physics who tried and failed to understand why many metals carry current without resistance at low temperatures. In 1957, nearly half a century after this standard kind of superconductivity was discovered, John Bardeen, Leon Cooper and John Robert Schrieffer finally explained the phenomenon, an achievement that earned them the Nobel Prize in Physics.
They determined that sound waves in metals — ripples where atoms bunch together, called phonons — create concentrations of positive charge that attract electrons, which are negatively charged. The phonons stick electrons together into “Cooper pairs.” Coupled off in this way, electrons play by different quantum mechanical rules, fusing into a quantum fluid whose flow is no longer gummed up by the atoms in the lattice. This phonon-mediated theory, known (after its authors’ initials) as BCS, matches almost all superconductivity experiments.
Alternative ways of gluing electrons together work on paper, and experimentalists have seen signs of puzzlingly strong “unconventional” glues in some superconductors, but such claims remain unsettled.
“It’s like if someone tells you in some very distant village on some island there are people with three heads,” Das Sarma said. “You should be very, very skeptical.”
In 2018, some researchers thought they might have stumbled upon just such a mythical island of exotic superconductivity, since twisted bilayer graphene appeared to somehow bind electrons much more tightly together than most superconductors do. Excitement rose earlier this year with the discovery of superconductivity in a similar system: three layers of graphene twisted at their own special angle. Both systems shared a rare, 180-degree rotational symmetry, which theorists argued could support an especially exotic form of superconductivity based on electron vortices known as skyrmions.
But the new incarnation of superconducting graphene appears strikingly plain.
ABC trilayer graphene, as Young and his colleagues call their graphene stack, is one of the cleanest and simplest materials they could make. The second and third layers are shifted rather than twisted, each nudged over by an additional half-honeycomb, so carbon atoms below fall in the center of lattices above.
Stacking graphene sheets is hard, with or without twists. Twisted devices are riddled with wrinkles that disrupt the magic angle in different zones, making each apparatus unique. Even when Young and colleagues manufactured their ABC trilayer devices, most attempts snapped back into an alternative stacking pattern. But — unlike the fussy twisted samples — the ones that stayed put were identical down to the last atom. The atoms “lock into place like Legos,” Young said.
Once the team had their first ABC device, they used an adjustable electric field to shuffle electrons between the pristine layers. As they tuned the electron distribution at cryogenic temperatures, they saw that the system behaved much as twisted graphene does, jumping between various types of magnetic behavior, as indicated by shifts in how the device slowed electric current. They posted their results in an April preprint.
When they examined the transitions in more detail, they identified brief flickers of zero electrical resistance — superconductivity — when the material was about one-tenth of a degree above absolute zero.
Although Young and his colleagues have no way to peep at the Cooper pairs of electrons directly, they found behavior that Bardeen, Cooper and Schrieffer would recognize: Moving electrons between the three layers increased the number of possible configurations the electrons could choose from, a quantity known as the system’s “density of states.” At high densities of states, electrons can more easily fraternize among themselves. BCS theory predicts that this electronic liberty aids the formation of Cooper pairs, and that’s what the researchers found: As the density of states rose, the material displayed two blips of superconductivity.
Since the BCS equation appears to hold, ordinary phonons might be responsible for the superconductivity.
“It’s quacking like a duck and walking like a duck,” Das Sarma said. “Phonons are natural to assume.”
Others are less convinced, noting that the evidence supporting phonons in ABC trilayer graphene remains rough. Superconductivity appears to track with the higher density of states, but that doesn’t mean the BCS equation is obeyed in detail, said Mike Zaletel, a condensed matter physicist at the University of California, Berkeley who consulted with Young during the research and helped develop the skyrmion theory of superconductivity.
In Young’s data, Zaletel sees hints of a mildly exotic sort of superconductivity — something like an island with a six-fingered population, rather than people with three heads. He explained that both flashes of superconductivity appeared immediately before the electrons organized into ferromagnetic states, where their spin directions became aligned. As regions of electrons started lining up, these fluctuating pockets of uniformity could have shepherded electrons into Cooper pairs much as phonons do.
Young’s group is already testing whether ferromagnetism is key to the onset of superconductivity in ABC trilayer graphene, or if it’s irrelevant — which would hint at conventional phonons.
Many physicists feel optimistic that Young’s new platform will help them figure out how electrons superconduct in graphene. The idiosyncrasies of each twisted graphene device made it impossible for even an individual lab to identically replicate its own results. ABC trilayer graphene, with its perfect layout, overcomes that challenge.
“Materials are complicated, and they have a way of lying to us,” said Steven Kivelson, a theoretical physicist at Stanford University. “What’s exciting about this development” is that it promises reproducible materials, “so that everybody can get the same answer.”
Since ABC graphene can become a superconductor and various types of magnet, all without twists or other obvious tricks, it also suggests that a much wider range of fairly ordinary materials might hold overlooked magic. This material versatility “may be hiding in plain sight much more ubiquitously than we thought,” Young said.