Decoding lithium's subtle phase stability with a machine learning force field

Abstract

Understanding the phase stability of elemental lithium (Li) is crucial for optimizing its performance in next-generation battery anodes, yet this seemingly simple metal exhibits complex polymorphism that requires proper accounting for quantum and anharmonic effects to capture the subtleties in its flat energy landscape. Here we address this challenge by developing an accurate graph neural network-based machine learning force field and performing efficient self-consistent phonon calculations for bcc-, fcc-, and 9R-Li under near-ambient conditions, incorporating quantum, phonon renormalization and thermal expansion effects. Our results reveal the important role of anharmonicity and predict 9R-Li as the ground state at zero temperature and pressure. The calculated bcc-fcc phase boundary qualitatively agrees with experiments, albeit with a systematic overestimation of the pressure slope and the transition temperature. The predicted small free energy differences between phases, particularly fcc- and 9R-Li, explain the experimental challenges in obtaining phase-pure samples. These findings provide crucial insights into Li's complex polymorphism and establish an effective computational approach for modeling Li based materials for energy and related applications.