What did the cuprate superconductivity study find?
Physicists at Utrecht University and Ludwig Maximilian University of Munich have measured when electrons pair up inside a cuprate superconductor. The study was published in Science on 5 November 2026. It reports that electron pairs persist up to the pseudogap energy. That energy reaches about 70 milli-electron volts at some points on the sample.
The material was Bi2Sr2CaCu2O8+delta, often shortened to Bi-2212 or BSCCO. It is a copper-oxide ceramic that superconducts at temperatures far above the reach of conventional metals. The researchers used a scanning tunneling microscope, cooled to 4.2 kelvin, to probe tiny patches of its surface.
Their signal was a step in the effective charge of the tunneling current. Below the pseudogap energy, the current carried noise as if charge arrived in pairs. Above that energy, the noise fell back toward the single-electron level. The step followed the local pseudogap energy at every position the team measured.
The team repeated the procedure at multiple positions on each crystal and on crystals with different doping. It then compared the energy of the noise step with the pseudogap energy measured at the same spot. No single spectrum settles the question, so the correlation across many locations carries the argument.
Two conditions must hold for a material to superconduct. Electrons must pair, and the pairs must move in step. Utrecht compares this to couples on a dance floor. It is not enough for each couple to dance together; the pairs must also keep time with one another. The new experiment addresses the first condition directly and the second only by inference.
The result held across samples with different doping levels, including an underdoped crystal with a critical temperature of 60 kelvin. The first authors are Jiasen Niu and Maialen Ortego Larrazabal. The author list includes Leiden University, Delft University of Technology and a Max Planck institute in Dresden. Brookhaven National Laboratory is also represented.
The paper does not demonstrate a room-temperature superconductor. It narrows the explanation for why cuprates stop superconducting when they do. The authors argue that electron pairing is not the limiting step. Phase coherence, the ability of pairs to move together as one fluid, is the harder condition to meet.
That distinction matters for anyone trying to design better superconductors. If pairing already survives at high energy scales, then research should focus on making pairs march in step. The experiment itself cannot show how that would be done.
How does shot noise reveal electron pairs?
Shot noise is the crackle in an electrical current caused by charge arriving in discrete lumps. A perfectly smooth stream would be silent. A stream of individual electrons produces small fluctuations. Pairs of electrons carry twice the charge, so they change the size of those fluctuations.
The team used that difference as a detector. A scanning tunneling microscope passes a current between an atomically sharp tip and the sample. By changing the voltage, the researchers chose the energy of the electrons involved, much like turning a radio dial.
At each energy, they measured the noise and converted it into an effective charge. A value near one electron meant ordinary tunneling. A value rising toward two electrons meant pairs were crossing the junction. In an ordinary s-wave superconductor, that doubling is clean and easy to read.
Cuprates are harder. They are d-wave superconductors, with nodes where the gap vanishes. Quasiparticles at those nodes add single-electron noise and dilute the pair signal. Simulations in the paper, run at low junction transparency, show an effective charge well below the ideal value of two.
That is why the measurement took years to build. The group developed a homemade cryogenic amplifier and a stable microscope. Milan Allan of LMU Munich told Ars Technica that the advance rested on persistence. "There was not a single trick," he said. "It was electronics craftsmanship."
The scanning approach also solves an interpretation problem. A mesoscopic junction averages over a whole device, so stray noise or barrier effects can mimic pairing. A sharp tip measures one small region at a time. That lets the team compare the noise step with the local pseudogap at the same point. The paper says this design gives a simpler interpretation than the earlier junction experiments.
The magnetic field test provided a check. Noise from drifting atoms or amplifier drift should ignore a magnetic field. Shot noise from pairs should fall as the field breaks them. The team measured at 0, 1 and 1.4 tesla, and up to 6 tesla on one sample. The effective charge moved toward the single-electron level.
That behaviour is what the pairing explanation predicts. It is not a direct photograph of a Cooper pair, but it is consistent evidence from an independent knob. The paper notes that mesoscopic junctions are prone to artifacts, and its local design gives a simpler interpretation.
Why is the pseudogap central to the high-temperature superconductivity puzzle?
Superconductivity was discovered in 1911. For most of the century, it appeared only near absolute zero, at about -260°C, and required liquid helium. In the mid-1980s, copper-oxide ceramics broke that pattern. They superconduct at roughly -173°C, a temperature liquid nitrogen can reach cheaply.
The theory that explains conventional superconductors failed to account for them. Nearly four decades later, physicists still argue about the mechanism. Ingmar Swart of Utrecht University put the problem plainly in the university's release. "We don't understand superconductivity at these higher temperatures at all," he said.
The pseudogap sits at the centre of that argument. Above the superconducting temperature, measurements find fewer electrons than expected at certain energies. One camp says the electrons are frozen into local patterns, such as charge density waves. The other says they have already paired and are waiting for conditions that let them flow.
The new data support the second camp. The pairing step appears at the pseudogap energy, not at the smaller kink feature seen in the same spectra. The correlation held from point to point, with a slope near one when pairing energy was plotted against pseudogap energy. That rules out a pseudogap caused purely by local order.
Doping changes the picture in a revealing way. The pseudogap temperature falls as hole doping increases, while the superconducting temperature follows a dome. The measured pairing energy grows as doping decreases, even after the critical temperature starts to fall. That mismatch is the core argument for a coherence limit. If pairing strength set the critical temperature, the two curves would move together.
It does not rule out a mixed explanation. The authors write that pairing and local order may be intertwined, forming one gap. A pair density wave, a state that combines pairing with spatial modulation, is one candidate they mention. The result excludes a simple two-gap picture, not every role for order.
Earlier evidence pointed the same way, but from a different platform. A 2019 Nature paper by Panpan Zhou and colleagues reported pairing signals above the critical temperature. Those measurements used mesoscopic junctions made from a different cuprate. Those junctions are harder to interpret, and the debate did not settle.
What does the result mean for room-temperature superconductivity?
Utrecht University framed the result as hope for room-temperature superconductivity. Swart said the experiment shows one of the two key conditions can be met at room temperature. "The 'glue' that holds the pairs together is strong enough to keep them together even at room temperature," he said in the release.
That claim rests on an energy comparison, not a demonstration. The measured pairing energy reaches about 70 milli-electron volts, well above the thermal energy at room temperature. If pairs can exist at that scale, then the barrier to a higher critical temperature is not the strength of the glue.
The energy scale is the surprising part. In conventional superconductors, the pairing gap and the critical temperature are linked by BCS theory. In these cuprates, pairing appears at an energy significantly larger than the gap associated with the superconducting state. The authors argue that this breaks the usual link and makes condensation, rather than glue, the variable to engineer.
The team is careful about the next step. Allan told Ars Technica that a room-temperature superconductor would not follow automatically, because other effects are coupled to the pairs. He also said he puts more hope in direct trial and error than in a single theoretical shortcut.
Phase coherence is the missing piece. In a simple picture, superconducting islands can host pairs but stay resistive if the links between them cannot sustain a coherent current. The paper reports no clear puddles in its maps, so the exact mechanism remains open.
The practical stakes are large but distant. Utrecht lists levitating trains, efficient energy storage and quantum computers as long-term possibilities. Superconducting magnets already serve medical imaging and particle physics, and lossless cables could change power grids. None of those applications follows from this experiment alone.
For now, the result is a constraint on theory. It tells materials researchers where to look: not at making electrons pair, but at making pairs cooperate across a macroscopic sample.
What are the limitations of the shot-noise experiment?
The clearest limitation is temperature. The team could not measure above the superconducting transition. Too many quasiparticles appear there, and the amplifier loses sensitivity as the sample warms. So the pairing step was measured below the critical temperature, while the pseudogap is usually defined above it.
That gap in coverage matters. The authors infer that pairing extends to the pseudogap from data taken inside the superconducting state. A direct measurement in the normal state would be stronger, and they say other groups may manage it.
Definitions also vary between instruments. Photoemission sees gaps that differ by momentum direction. Tunneling sees a local gap that may not be identical. The paper acknowledges that naming features consistently across techniques is difficult, because the crossovers are broad.
The modelling is qualitative. Small changes in the number of nodal quasiparticles can shift the effective charge. The simulations therefore predict a shape rather than a precise number. Junction transparency, tip stability and background noise all need careful calibration.
The study covers one material family and several doping levels. That is a strength for internal consistency, but it is not a survey of all cuprates. Independent replication, ideally with a different apparatus, is the next test.
A replication would need to show two things. The noise step should appear at the pseudogap energy in a different lab, ideally with a different tip and amplifier. And the field dependence should repeat, because it separates pairing noise from instrumental drift. Neither test requires the same crystals or the same model.
The paper appeared in Science on 5 November 2026, after a preprint in September 2024. Peer review adds weight, but it does not replace reproduction.
What happens next in cuprate superconductivity research?
Swart plans to push the technique further. He wants higher-resolution maps of the surface and, eventually, an instrument that detects phase coherence directly. That would give separate markers for pairing and for the onset of coherent flow, which the current experiment cannot separate.
Theory has work to do as well. The paper calls for calculations of how coherence would appear in tunneling spectra. Without predictions, the community cannot easily test competing pictures of the pseudogap.
Allan sees materials chemistry as a parallel route. He told Ars Technica that improving cuprates may depend more on direct synthesis and testing. In that view, AI helps with the search, but physics alone is not enough. That is a pragmatic view of how superconducting materials have improved before.
Materials search is where AI enters the story, but only at the edges. Allan's point is that synthesis and testing remain the filter, with computation helping to choose candidates. A measurement like this one gives that search a concrete target. Theory and synthesis teams can compare pairing energies across compounds.
Other groups can test the central claim. If pairing survives above the critical temperature in a directly measurable way, the coherence bottleneck becomes the target. If the step disappears in other cuprates, the picture becomes more complicated.
Who should watch this? Teams building superconducting qubits care about coherence and materials quality. Energy researchers care about cables and storage. Materials labs care about a design rule. All of them need the same thing: a mechanism they can trust.
The study does not end the cuprate debate. It removes one explanation from the shortlist and points the next round of experiments at coherence. After nearly 40 years, that counts as progress.