Researchers Use AI to Accelerate Superconductor Discovery

Researchers at Aalto University used machine learning to accelerate the discovery of two new kagome superconductors, YRu3B2 and LuRu3B2, with critical temperatures of 0.81 K and 0.95 K. The study, published in Physical Review Research, demonstrates a workflow combining ML screening, first-principles calculations, and experimental validation by Rice University collaborators.

An Aalto-led SuperC team reported on June 29, 2026 that machine-learning-guided screening helped identify and experimentally confirm two kagome superconductors, YRu3B2 and LuRu3B2 . The Physical Review Research paper reports superconducting critical temperatures of 0.81 K and 0.95 K , so this is not a room-temperature result; the practitioner value is the workflow. The team used ML to narrow a large chemical search space, then applied first-principles calculations before Rice University collaborators synthesized and tested the candidates. For data-science and materials teams, the useful lesson is that physics-informed features, high-precision candidate ranking, and experiment-ready validation loops can matter more than broad benchmark accuracy when ML systems must produce lab-verifiable discoveries.