Scientific Achievement
- Researchers in the Data Driven Synthesis Science program developed a computational workflow to predict phase selectivity in solid-state reactions, explicitly accounting for how ions move at reaction interfaces
Significance and Impact
- This framework predicts which phases form in powder reactions accounting for both thermodynamic and kinetic effects, enabling rapid simulation of powder synthesis outcomes in-silico
Research Details
- Used machine-learning-driven molecular dynamics to extract transport tensors in liquid-like interphases and a cellular automaton to simulate outcomes and benchmarked predictions against decades of experimental data for Ba-Ti-O system
Publication Details
V. Karan, M.C. Gallant, Y. Fei, G. Ceder, K.A. Persson, Nature Materials (2026).
DOI:10.1038/s41563-026-02596-5
Work was performed at Lawrence Berkeley National Laboratory and in part by NERSC.