Linköping says AI-assisted materials research cuts hazardous chemical and water use

Linköping University says an AI-assisted search of old materials records helped researchers create 11 new materials and develop a synthesis method using fewer hazardous chemicals and less water. Its October 29 account highlights a practical manufacturing benefit, though the authors’ technical description involves high temperatures—not a process that stays at room temperature.

The findings appear in Science under the title “Data-driven discovery and rapid, direct synthesis of MXenes.” The research was already public in a preprint submitted August 17, 2026; the university’s October coverage is not the study’s first disclosure.

Old records supply new laboratory leads

The Swedish team worked with colleagues in Estonia and Finland, using AI to search materials databases. Their target was MXenes: materials made from metals combined with carbon or nitrogen. Only a few atomic layers thick, these sheets have a large surface area relative to their weight.

MXenes were recognized as a materials family in 2011. The search looked backward for compositions that had already been made but whose significance researchers in the field had missed. Professor Johanna Rosén describes the work as “The Treasure Chest Project,” drawing on overlooked knowledge in literature and databases.

A rapid reaction, not room-temperature production

The authors describe a self-propagating high-temperature synthesis completed within minutes without sustained external heating. The university also reports significantly lower water and hazardous chemical use than previous methods, without quantifying those reductions.

Beyond rediscovery, the team made 11 new rare-earth-based MXenes inspired by the search. Experiments and theory revealed semiconducting behavior and varied magnetic states in this family. The university identifies magnetism as a route toward additional applications, including data storage.

MXene sheets can be stacked, allowing different chemical reactions between layers depending on their composition. Possible uses include water purification, energy storage, telecommunications and biomedicine. These remain potential applications; Rosén expects scaling up manufacturing to be the step that drives substantially wider use.