Forty years ago this spring, two researchers at the IBM Zurich Research Laboratory mixed barium, lanthanum, copper, and oxygen and measured something that should not have been possible. J. Georg Bednorz and K. Alex Müller found that their ceramic began losing electrical resistance at 35 kelvin , 50% higher than the previous record, in a material class nobody had considered a candidate for superconductivity. Their discovery earned a Nobel Prize in 1987, the fastest ever awarded.
This month, a team at MIT and the University of Basel published a discovery that, in its own way, should not be possible either. In rhombohedral pentalayer graphene , pure carbon, five atomic layers thick , they found superconductivity that not only survives a strong magnetic field but is strengthened by it. The paper, led by Long Ju of MIT and Dominik M. Zumbühl of Basel and published in Nature on June 29, reports three distinct superconducting states that withstand in-plane magnetic fields up to 8.5 tesla , tens of times beyond the theoretical limit for a conventional superconductor.
The two papers, separated by 40 years, bookend a field united by a single theme: nature keeps finding new ways to make electrons flow without resistance, and keeps refusing to fully explain how.
The anniversary
The 40th anniversary was marked by a retrospective in Nature by Inna Vishik and Warren Pickett of the University of California, Davis. Their timeline traces the milestones: Bednorz and Müller’s 35 K cuprate in 1986; the 93 K YBa₂Cu₃O₇ in 1987, the first superconductor to work above liquid nitrogen’s boiling point; the 135 K HgBa₂Ca₂Cu₃O₈ in 1993, still the ambient-pressure record holder; the iron-based superconductors in 2008; the 250 K hydrides under extreme pressure in 2019.
The deepest problem remains unsolved after four decades. The microscopic mechanism of high-temperature superconductivity in cuprates is still not known. The leading candidates , spin fluctuations, charge density waves, Anderson’s resonating valence bond theory , each explain part of the puzzle but none accounts for everything. The normal state of cuprates is a “strange metal” whose electrical properties follow rules that conventional Fermi liquid theory cannot describe.
The new platform
Rhombohedral graphene offers an entirely different way to approach these questions. Unlike cuprates , chemically complex, multi-element oxides whose intrinsic disorder and alloy scattering obscure fundamental physics , pentalayer graphene is crystalline carbon: atomically perfect, clean-limit, and tunable.
The MIT-Basel team identified three distinct superconducting states, labeled SC2, SC3, and SC4. SC2 is strengthened by an in-plane magnetic field , counterintuitive, because magnetic fields normally break Cooper pairs through the Zeeman effect. SC3 is boosted by a small out-of-plane field. SC4 is induced by the field itself: superconductivity appears only when a magnetic field is applied.
The mechanism lies in time-reversal symmetry breaking. In conventional BCS superconductors, the magnetic field exerts a Zeeman energy that competes with the superconducting gap; when the Zeeman energy exceeds the gap, pairs break. The Pauli paramagnetic limit for these graphene states, given their critical temperatures of roughly 110 to 300 millikelvin, would be about 0.2 to 0.56 tesla. The observed survival at 8.5 tesla , 40 to 85 times the theoretical limit , indicates spin-triplet or orbital-mediated pairing that is fundamentally immune to Zeeman depairing.
The normal state in these graphene devices is a quarter-metal phase in which both spin and valley degrees of freedom are polarized. The magnetic field aligns these in a way that promotes, rather than suppresses, superconducting pairing.
Two frontiers
The cuprate anniversary retrospective ends with an acknowledgment: room-temperature ambient-pressure superconductivity remains elusive. The graphene paper ends with a different horizon: non-Abelian quasiparticles for fault-tolerant topological quantum computing. The two goals are not in competition , they reflect the field’s evolution from a single-minded pursuit of higher critical temperatures to a broader exploration of what unconventional superconductivity enables.
The MIT-Basel team achieved another advance in this paper: by incorporating spin-orbit coupling through proximity with transition metal dichalcogenide layers, they generated multiple new superconducting states without introducing disorder , a critical requirement for topological states to coexist with superconductivity without being destroyed by impurities.
The 40-year quest has proven one thing definitively: superconductivity has not stopped surprising us. The next decade may finally answer the question Bednorz and Müller opened in 1986, or it may open a new one nobody is asking yet.
Sources:
Seo J, Cotten AA, Ye S, et al. Family of magnetic field-boosted superconductors in rhombohedral graphene. Nature. Published online June 29, 2026. doi:10.1038/s41586-026-10815-x
Vishik I, Pickett W. Forty years of high-temperature superconductivity. Nature. 2026;654:873-874. doi:10.1038/d41586-026-01801-4