A research team including Cornell University in the United States has used ultrasound to probe uranium ditelluride (UTe₂) under pressure and discovered a new phase boundary at which one superconducting component disappears on cooling. At this boundary, superconductivity itself does not vanish. Instead, only one of the two components that had been coexisting is lost. The measurements clarify how multiple superconducting states coexist and transition into one another.
Two distinct superconducting states formed collectively by electron pairs, known as "SC1" and "SC2," were already known in UTe₂. The achievement here is not the first discovery of these two states, but the thermodynamic identification of the region where they coexist and of the boundary where one of them disappears. UTe₂ is a candidate topological superconductor, a class of material in which special excitations are expected, and the result offers clues for understanding its properties. However, the study does not demonstrate topological character itself.
The peer-reviewed paper by Sahas Kamat and colleagues, "Thermodynamic demonstration of tetracriticality and multicomponent superconductivity in UTe2," was published in Nature Physics on October 30, 2026. Its DOI is 10.1038/s41567-026-03504-z. The measurement conditions and theoretical details below are based on the author version released by the research team. The author version is a pre-publication draft and is not identical to the published paper.
Three Successive Transitions Appear on Cooling
To describe a superconducting state, researchers use an "order parameter," which expresses the strength and phase of the collective state of the electron pairs. SC1 and SC2 correspond to different superconducting orders, and "SC1+SC2" denotes a state in which both orders coexist under the same conditions. It is not simply a juxtaposition of superconducting states observed in separate samples.
According to Section III.2 of the author version, the team applied a pressure of 0.213 gigapascals (GPa) to UTe₂ and cooled it with no magnetic field. The longitudinal elastic modulus showed three discontinuous changes. The temperature of each, the change in the modulus, and the corresponding superconducting state are shown below.
| Temperature | Observed change in | Phase assigned thermodynamically |
|---|---|---|
| 1.63 K | Downward jump | SC2 |
| 1.58 K | Downward jump | SC1+SC2 |
| 1.0 K | Upward jump | SC1 only |
This phase assignment is an interpretation based on the changes in elastic modulus and thermodynamic relationships. It is not the result of a direct specific-heat measurement, nor an image of the strength of the two components. The key point is that even though the SC2 component is lost at the 1.0 K boundary, superconductivity from SC1 persists. In other words, lowering the temperature does not eliminate superconductivity altogether; the system moves from a state with two coexisting components to one with a single component.
The team named this new boundary . The existence of SC1 and SC2 was known from earlier work, but these measurements made it possible to place both the region where they coexist and the boundary where SC2 disappears on cooling within the phase diagram.
Why Ultrasound Can Tell Four Phases Apart
The team used "pulse-echo ultrasonics," in which ultrasonic pulses are sent into a sample and the reflected waves are analyzed. The elastic modulus, which expresses a material's stiffness, is tied to the change in free energy that accompanies deformation. When a transition into superconductivity changes the character of the free energy, the elastic modulus shows discontinuous changes or kinks. This relationship makes it possible to examine changes in superconducting state thermodynamically.
The team measured the temperature dependence of the longitudinal moduli and and the shear modulus . Pressure was held fixed in the range of 0 to 1 GPa, with particularly detailed study around 0.20 GPa. Of the three changes observed at 0.213 GPa, the first two were downward jumps in , while the last was an upward jump. Using the thermodynamic relationship linking discontinuities in elastic modulus and specific heat, the team interpreted the last change as a transition in which one component of the superconducting order is lost.
This judgment does not rest on a mere correlation in which the graph shapes look similar. However, the elastic-modulus measurements cannot directly reveal the microscopic mechanism that causes competition between electrons. The author version also reports no hysteresis between cooling and warming measurements, which supports the interpretation of a second-order transition without latent heat.
The resulting phase diagram shows a "tetracritical point" near 0.20 GPa and 1.6 K, where four phases meet: the normal state, SC1, SC2, and SC1+SC2. A region previously interpreted as a triple point turned out to contain a new phase boundary, revealing a structure in which four phases meet. This does not overturn any existing law of thermodynamics; it identifies the arrangement of the phases in a way consistent with thermodynamics.
Measuring a boundary this narrow demanded precise control of pressure. The team applied pressure at room temperature and calibrated it at low temperature by matching against known transition temperatures. According to the author version, the absolute pressure accuracy is 0.05 GPa, whereas the pressure width of the boundary they sought to probe was 0.01 GPa. Capturing this boundary required keeping pressure variations within the sample below 0.01 GPa.
The distinction matters here between the measurement accuracy of absolute pressure and the precision of relative pressure control. The team adjusted the volume of the pressure cell to control pressure finely and reduced pressure variation within the sample by keeping the measurement region small. The ultrasonic transducer used was about 100 micrometers across, and the sample was about 300 micrometers thick. Being able to probe a phase boundary 0.01 GPa wide therefore does not mean absolute pressure could be determined to better than that.
The Extended Data released with the formal paper distinguishes at least two samples, S1 and S2. However, the materials we could check do not state the total number of samples used across the study. It therefore cannot be concluded that all measurements came from the same crystal, nor that there were only two samples in total.
How Are the Two Coexisting Superconducting Components Coupled?
That two superconducting orders coexist is a separate question from how they are coupled. Using the experimentally obtained phase diagram as a constraint, the team built a phenomenological Ginzburg–Landau model describing the interaction between the two orders. In this model, SC2 is suppressed as the SC1 component grows stronger and eventually disappears, a behavior that can explain the folded shape of the phase boundary. However, the order-parameter amplitudes shown in the model are theoretical calculation results; they were neither measured directly in experiment nor derived from an independent microscopic simulation.
In Cornell University's announcement, coauthor B. J. Ramshaw said that what is important about uranium ditelluride is that it can show two different kinds at the same time. When two superconducting components coexist, a phase offset between them could produce states that a single component cannot. However, this study did not determine the sign of the coupling coefficient that fixes that relative phase.
The authors discuss the possibility that, if the two components have similar spin structures, a chiral superconducting state could be favored. This is a conditional prediction based on a theoretical model, not an experimental confirmation of broken time-reversal symmetry or topological properties. The observation that the coexistence region widens when a magnetic field is applied can also be explained by multiple mechanisms, so it alone cannot identify symmetry breaking.
The special excitations expected in topological superconductors are also being studied from the standpoint of quantum computing applications. These results provide a foundation for exploring the conditions under which such states arise, but several steps remain before application to an actual quantum computing device, including directly confirming the properties of the superconducting state.
Ambient Pressure and 12–18 T Fields Are Candidates for Next Experiments
The team applied a magnetic field along the crystal's b-axis and made additional measurements at four pressures near the critical point. They then built a three-dimensional phase diagram from the results and estimated that applying a field of 12 to 18 tesla (T) along the b-axis could produce a coexistence region of SC1 and SC2 even at ambient pressure. This field range is not a condition under which the team directly observed a coexisting state in this study; it is a prediction derived from the phase diagram. The author version also treats it as an estimate, and it offers a concrete experimental condition for seeking coexistence at ambient pressure.
Regarding the new boundary , as of November 2, 2026, within the materials and searches we checked, we found no replication by an independent research group. The author version, the formal paper, and the university announcement are all materials about the same study and do not constitute independent verification. The next stage will be to confirm the boundary with other samples and independent measurement techniques, and to directly probe the relative phase of the two orders and topological excitations. The prediction of 12 to 18 T at ambient pressure offers a concrete lead for advancing such verification.
