Tohoku University and Fujitsu Use AI to Discover Promising New Superconducting Material

The AI technology was utilized to automatically clarify causal relationships from measurement data obtained at NanoTerasu Synchrotron Light Source
Image Credit: Scientific Frontline / stock image

Tohoku University and Fujitsu Limited announced their successful application of AI to derive new insights into the superconductivity mechanism of a new superconducting material. Their findings demonstrate an important use case for AI technology in new materials development and suggests that the technology has the potential to accelerate research and development. This could drive innovation in various industries such as environment and energy, drug discovery and healthcare, and electronic devices.

The two parties used Fujitsu's AI platform Fujitsu Kozuchi to develop a new discovery intelligence technique to accurately estimate causal relationships. Fujitsu will begin offering a trial environment for this technology in March 2026. Furthermore, in collaboration with the Advanced Institute for Materials Research (WPI-AIMR), Tohoku University , the two parties applied this technology to data measured by angle-resolved photoemission spectroscopy (ARPES), an experimental method used in materials research to observe the state of electrons in a material, using a specific superconducting material as a sample.

Causal discovery from ARPES measurement data.
Image Credit: ©K. Fujita, K. Nakayama et al.

Tohoku University and Fujitsu established the Fujitsu x Tohoku University Discovery Intelligence Laboratory in October 2022 as part of the Fujitsu Small Research Lab initiative which sees Fujitsu researchers stationed at universities to accelerate joint research, discover new themes, develop human resources, and build long-term relationships. The aim is to contribute to solving societal issues through the development of new technologies and human resource development, by integrating Tohoku University and Fujitsu's technologies, achievements, and knowledge. The two parties are engaged in joint research to develop and socially implement discovery intelligence that uses AI to find solutions to various problems from data, including those in materials science.

NanoTerasu Synchrotron Light Source, which began operation in April 2024, enables the measurement of molecular, atomic, and electronic states with nanometer-level high spatial resolution. The facility works to develop new functional materials to drive innovation and contribute to resolving societal issues, including environmental challenges. However, as measurement performance improves, the amount of data created increases. Efficiently extracting only useful information without relying on human experience or intuition and advancing the automation of scientific research processes are key priorities for the future.

ARPES measurement data is very large. A causal graph of the data has a massive number of nodes making it difficult to find useful information. The technique developed in this collaboration significantly compresses the scale of the causal graph by performing fitting based on a model equation for the measurement data and constructing a causal graph from only the extracted parameters. In addition, the two parties developed a technique to further simplify the graphs and reduce noise impact. This technology reduced the size of the causal graph to less than 1/20 of the conventional size, enabling the efficient discovery of new insights.

Tohoku University and Fujitsu applied this technology to ARPES measurement data of cesium vanadium antimonide (CsV3Sb5), a kagome superconducting material. Cesium vanadium antimonide has potential applications as a high-temperature superconductor, but its superconductivity mechanism is not yet fully understood. They found that the superconductivity mechanism is due to the interaction of vanadium, antimony, and cesium electrons.

Moving forward, both organizations will further leverage this technology along with NanoTerasu's world-class capabilities in spatial resolution to automatically clarify the causal relationships between phenomena at the microscopic level. This will contribute to the development of new functional materials that address global environmental issues - one of Fujitsu's materiality priorities - in areas such as high-temperature superconductivity and next-generation low-power consumption devices.

Published in journal: Scientific Reports

Title: Extracting causality from spectroscopy

Authors: K. Fujita, K. Nakayama, Y. Fujiki, T. Kato, H. Suito, H. Higuchi, and T. Sato

Source/Credit: Tohoku University

Reference Number: ms122325_01

Privacy Policy | Terms of Service | Contact Us