Ohio State researchers can switch superconductivity in magic angle bilayer graphene on and off, and find the relationship runs opposite to what conventional superconductors predict.
For decades, condensed matter physicists have carried a simple story about superconductors. In ordinary materials like aluminum or lead, electrons form pairs that glide through the crystal without losing energy. The pairs survive because the electrons are, in a sense, weakly repelling one another. Push them to repel more strongly, and you should break the pairs. Push them to repel less, and pairing should get stronger. That is the textbook intuition.
A new result from Ohio State University inverts that intuition in one very specific, very strange material. The team, led by physicist Chun Ning (Jeanie) Lau, reports in Nature Physics that in magic-angle twisted bilayer graphene sitting on a strontium titanate substrate, the relationship runs the other way. Push the relevant interaction parameters in one direction, and the superconductivity actually weakens. The pairing survives the wrong way around.
"It's surprising that in this system, increasing the interaction actually decreases superconductivity," Lau said in the Ohio State News release describing the work. Her graduate student Xueshi Gao is the paper's first author.
The result is a new knob. Twisted bilayer graphene, two atomically thin carbon sheets stacked at a slight angle, became one of the most studied materials in physics after the 2018 discovery that it could superconduct at low temperatures when the sheets were rotated to a precise "magic angle" of about 1.1 degrees, where the two lattices interfere in ways that change the material's electronic behavior. Researchers have spent the years since trying to figure out the pairing mechanism. The Lau group's contribution is a way to probe that mechanism by reaching outside the graphene itself.
The trick is the substrate. Strontium titanate is a synthetic diamond-like crystal that the team has used for years as a platform for graphene devices. By changing how strongly the graphene couples to the strontium titanate environment, the team can tune the strength of certain electron-electron interactions. Pairing in the twisted bilayer is sensitive to that local environment. Change the coupling, and the superconducting state can be turned up, turned down, or turned off.
That makes the substrate a control parameter, a dial physicists can set rather than just observe. The team used the dial to test a question that has hovered over the field since 2018: how important are electron-electron interactions in magic-angle graphene, and do they help or hurt the superconducting state?
The textbook answer in conventional superconductors is that too much repulsion kills pairing. The Lau group's result in twisted bilayer graphene points the other way. As described in the OSU release, the paper reports a "double-edged role" for interactions in this system. The same parameter that, in principle, should weaken pairing instead helps pin down the conditions under which the superconducting state is fragile. The framing is that the result sharpens the map of where this superconductivity can exist, and where it cannot.
That framing matters because the popular narrative around twisted bilayer graphene has been one of escalating promise. The 2018 paper was followed by reports of correlated insulating states, magnetism, and several flavors of superconductivity. Each finding prompted a wave of speculation about whether this would be the platform that finally delivered a room-temperature or commercially useful superconductor. None of those promises have been kept.
Lau is careful to keep the new paper in its lane. The work, she said, is an "initial step" toward understanding the mechanism, not a path to applications. "Mechanism of superconductivity in twisted bilayer graphene remains unresolved," the team notes, in language reproduced by ScienceDaily's coverage of the release. That distance is built into the journal paper's title as well: "Double-edged role of interactions in superconducting twisted bilayer graphene."
The collaborators include Aatmaj Rajesh, Emilio Codecido, Daria Sharifi, Zheneng Zhang, Youwei Liu, and Marc Bockrath at Ohio State; Alejandro Jimeno-Pozo, Pierre Pantaleon, and Paco Guinea at Imdea Nanoscience in Spain; and Kenji Watanabe and Takashi Taniguchi at Japan's National Institute for Materials Science, who grew the hexagonal boron nitride used to encapsulate the graphene. The work was funded by the U.S. Department of Energy and the National Science Foundation.
What the result gives the field is a way to test models. Theoretical proposals for the pairing mechanism in magic-angle graphene disagree on whether interactions help, hurt, or are roughly neutral. A dial that lets researchers push the system between regimes is the kind of data those models need in order to be ruled out. The surprise is not only that the dial works, but that it points the wrong way relative to textbook expectation.
For now, the practical implications are confined to a small community of physicists running low-temperature experiments on atomically thin devices. There is no near-term path to a useful superconductor implied by this work. There is, however, a path to a clearer picture of what kind of superconductor twisted bilayer graphene actually is. That is a smaller promise than the headlines usually allow, and a more honest one.