An international research team has reported the discovery of two new superconductors, YRu3B2 and LuRu3B2, using a method that combines machine learning with quantum calculations to narrow down promising material candidates. The materials were later synthesized and experimentally confirmed at Rice University, and the proof-of-concept study was published in Physical Review Research. The work is part of the SuperC consortium, a collaboration launched in 2023 with the stated goal of finding a room-temperature superconductor by 2033.
Superconductors carry electricity with no resistance, but currently known examples generally function only at extremely low temperatures, making them costly and difficult to use widely. Researchers said the challenge is partly one of scale: possible combinations of chemical elements are vast, while only a small share yield superconducting behavior. In this study, the team used machine learning to pre-screen large numbers of combinations, then applied more intensive quantum analysis to the most promising options. The two new materials were identified as kagome superconductors, with their properties linked to electrons forming flat bands in a kagome lattice.
Professor Päivi Törmä of Aalto University, who leads SuperC, said the approach could significantly speed the search for additional superconductors by reducing the amount of computation needed for each candidate. She argued that the method could eventually allow researchers to assess materials on a far larger scale, potentially into the billions. Törmä also said a practical room-temperature superconductor could transform energy use, including in computers and data centers, by cutting electrical losses and reducing heat generation.
At the same time, the research does not claim that a room-temperature superconductor has been found. The newly identified compounds still represent an early-stage advance in the search process rather than a direct solution to the longstanding problem. Törmä noted that superconductivity remains difficult to understand fully at the quantum-mechanical level, and that even theoretically promising materials may prove hard to synthesize or scale for practical use. The consortium's results nonetheless point to a faster screening strategy in a field where more than 7,000 superconductors have been identified over time, often through chance rather than systematic prediction.
