Machine-learning rationalization and prediction of solid-state synthesis conditions

Abstract

There currently exist no quantitative methods to determine the appropriate conditions for solid-state synthesis. This not only hinders the experimental realization of novel materials but also complicates the interpretation and understanding of solidstate reaction mechanisms. Here, we demonstrate a machine-learning approach that predicts synthesis conditions using large solid-state synthesis datasets text-mined from scientific journal articles. Using feature importance ranking analysis, we discovered that optimal heating temperatures have strong correlations with the stability of precursor materials quantified using melting points and formation energies (∆G f , ∆H f ).In contrast, features derived from the thermodynamics of synthesis-related reactions did not directly correlate to the chosen heating temperatures. This correlation between optimal solid-state heating temperature and precursor stability extends Tamman's rule from intermetallics to oxide systems, suggesting the importance of reaction kinetics in determining synthesis conditions. Heating times are shown to be strongly correlated

Modernist materials synthesis: Finding thermodynamic shortcuts with hyperdimensional chemistry

“…strain, dimensionality, external fields, and chemical potential. 45 Given the use of databases and natural language processing techniques, [46][47][48][49] modernist materials chemists have many tools at their disposal.…”

Free energy predictions for crystal stability and synthesisability

“…Such data can then be used to make predictions of optimal crystal synthesis procedures. 93 Even if the synthesis of a specific metastable material is possible, a related question is its lifetime before transformation into a lower energy equilibrium configuration. The answer requires knowledge of the free energy landscape and the corresponding barriers and kinetics of the system.…”