How the Chemical Composition Alone Can Predict Vibrational Free Energies and Entropies of Solids

TL;DR

This paper presents a machine learning approach that predicts vibrational free energies and entropies of solids using only chemical formula-based descriptors, achieving high accuracy with minimal data.

Contribution

It introduces a simple, formula-based machine learning model that accurately predicts vibrational thermodynamic properties of solids, outperforming more complex descriptors with small training sets.

Findings

Mean absolute error less than 0.04 meV/K/atom for entropy

Chemical formula alone matches complex descriptors in predictive power

Model enables fast screening of phase diagrams at finite temperatures

Abstract

Computing vibrational free energies ($F_{vib}$) and entropies ($S_{vib}$) has posed a long standing challenge to the high-throughput ab initio investigation of finite temperature properties of solids. Here we use machine-learning techniques to efficiently predict $F_{vib}$ and $S_{vib}$ of crystalline compounds in the Inorganic Crystal Structure Database. By employing descriptors based simply on the chemical formula and using a training set of only 300 compounds, mean absolute errors of less than 0.04 meV/K/atom (15 meV/atom) are achieved for $S_{vib}$ ($F_{vib}$), whose values are distributed within a range of 0.9 meV/K/atom (300 meV/atom.) In addition, for training sets containing fewer than 2,000 compounds the chemical formula alone is shown to perform as well as, if not better than, four other more complex descriptors previously used in the literature. The accuracy and simplicity of the approach mean that it can be advantageously used for the fast screening of phase diagrams or chemical reactions at finite temperatures.