| Abstract Scope |
Large, curated materials datasets and first-principles workflows have transformed the prediction of composition–structure–property relationships, but accelerating materials discovery also requires predictive synthesis. Building on the Materials Project data infrastructure, we present a framework that couples thermodynamics with kinetics to predict phase selectivity during solid-state reactions. A cellular automaton approach (ReactCA) integrates first-principles thermodynamics with transport-informed reaction rules to simulate the time evolution of intermediates and products as functions of precursor composition, atmosphere, and heating profile. Onsager-based transport analyses quantify correlated ionic motion in the amorphous reactive interphase, providing kinetic descriptors for diffusion-limited phase selection. We demonstrate accurate prediction of phase evolution in ternary oxide systems, including composition-dependent polymorph selection driven by correlated ion transport. This integrated strategy—combining thermodynamics, kinetics, and autonomous experimentation—illustrates how closing the modeling–experiment loop can make predictive synthesis an integral part of data-driven materials discovery. |