Congratulations to our postdoctoral researcher, Dr. Zhi Li, on the publication of a new paper as co-first author.
In this work, the authors present a data-driven framework to accelerate the discovery and optimization of high-performance thermoelectric materials. By analyzing a curated dataset of more than 71,000 entries, they identify an empirical design descriptor—the lattice-to-total thermal conductivity ratio (κ/κL≈0.5)—that quantitatively captures the well-known phonon-glass electron-crystal (PGEC) concept.
Building on this insight, the team developed machine learning models that predict both lattice and total thermal conductivity, enabling rapid screening of more than 104,000 inorganic compounds. The framework identified 2,522 ultralow-thermal-conductivity candidates while simultaneously evaluating their proximity to the optimal PGEC regime. In addition, a case study on chemical doping demonstrates how the framework can guide materials optimization toward improved thermoelectric performance.
Congratulations to Dr. Zhi Li and all collaborators on this exciting achievement!