AI-Accelerated Hunt for Room-Temperature Superconductors Yields Two New Materials

The SuperC consortium used machine learning to discover two new superconducting compounds — YRu3B2 and LuRu3B2 — demonstrating a method that could dramatically speed the search for a room-temperature superconductor, the holy grail of energy physics.

An international research consortium has demonstrated that machine learning can dramatically accelerate the discovery of superconducting materials, already yielding two new compounds in a breakthrough that brings the long-sought goal of room-temperature superconductivity closer to reality.

The SuperC consortium, led by Aalto University Professor Päivi Törmä, combined machine learning with advanced quantum physics to screen a vast space of possible elemental combinations. A specialized algorithm identified the most promising candidates, which were then validated through theoretical calculations and synthesized in the lab at Rice University under Professor Emilia Morosan.

The two newly discovered materials — YRu3B2 and LuRu3B2 — owe their superconducting behavior to electrons forming flat bands within a kagome lattice, a geometric arrangement inspired by traditional Japanese basket weaving. The proof-of-concept study was published in Physical Review Research.

Superconductors can carry electricity with zero resistance but only at extremely low temperatures. Over 7,000 have been identified across decades of research, but the process has been largely serendipitous — until now. The SuperC consortium, established in 2023, has set an ambitious target: find a room-temperature superconductor by 2033.

"Superconductive materials that can operate at room temperature would forever change the way we consume energy," Törmä said. Replacing conventional conductors in computers and data centers alone could slash global energy consumption and drastically reduce the heat footprint of the ICT sector.

The new ML-guided approach shifts superconductivity research from trial-and-error to a systematically accelerated pipeline, opening the door to thousands of potential new materials.

Sources: ScienceDaily, SciTechDaily, The AI Insider