Researchers have discovered two new superconductors using a machine learning-based screening method, demonstrating a faster way to identify materials that may one day enable room-temperature superconductivity.
An international team led by Aalto University researchers has combined machine learning and quantum physics calculations to identify two previously unknown superconductors, YRu3B2 and LuRu3B2. This approach significantly reduces the time required to search through the vast number of possible material combinations.
Superconductors can carry electricity with zero resistance, but only at extremely low temperatures. They are already used in technologies such as quantum computers, MRI scanners, fusion reactors, and maglev trains. Scientists have long sought materials that remain superconducting at room temperature, a breakthrough that could transform power transmission and computing.
The newly discovered material derives its superconducting properties from electrons arranged in a Kagome lattice, a geometric pattern inspired by traditional Japanese basket weaving. After machine learning identified promising candidates, the researchers verified them through theoretical calculations before synthesizing the materials and confirming them experimentally.
AI refines your search
The researchers say the new workflow addresses one of the biggest challenges in superconductor research: the overwhelming number of possible material combinations.
“Superconducting materials that can operate at room temperature will forever change the way we consume energy,” explained Aalto University professor Paivi Torma. “If such materials can replace ordinary conductors in applications such as computers and data centers, they have the potential to reduce global energy consumption and significantly reduce the thermal footprint of the ICT sector.”
The research is part of the SuperC consortium, an international collaboration launched in 2023 with the goal of discovering room-temperature superconductors by 2033.
After computer screening, Rice University collaborators synthesized the candidate materials to create actual samples. The experimental team then confirmed that both compounds exhibit superconductivity, providing evidence that the machine learning-based discovery process works.
Accelerate the discovery process
For decades, scientists have relied primarily on trial and error to discover superconducting materials.
“Over the decades, researchers have recognized more than 7,000 superconductors, most by chance,” Tolma explained. “The process of identifying potential materials is so computationally intensive that researchers have actually only been able to theoretically predict the viability of about 20 of them.”
The researchers say their approach has the potential to dramatically expand the number of materials that can be evaluated.
“Our method uses machine learning-based pre-screening followed by targeted calculations of promising candidates. This approach will significantly speed up the discovery of future superconductors. Using machine learning, we may be able to push the number of materials that can be processed into the billions,” Tolma said. “This brings us much closer to discovering room-temperature superconductors.”
Rather than replacing traditional physical calculations, machine learning systems act as filters, allowing researchers to focus computational resources on the most promising candidates. The researchers believe this approach could uncover thousands of new superconductor possibilities and accelerate the search for materials suitable for large-scale energy and computing applications.
This research Physical review research.
