Phase-field models learned from ab initio data simulate materials accurately
Learning ab initio phase-field models
Machine Learning
Summary
Simulating how tiny structures inside materials change over time usually requires either very detailed but small-scale methods or larger-scale but less accurate ones. This paper shows a way to create models that combine both accuracy and scale by learning from detailed atomic simulations. The authors use neural networks to connect the detailed atomic data with bigger-scale models that describe material changes, like melting or mixing. They tested this on iron-boron mixtures and hydrogen-helium mixtures, showing it can explain known behaviors and simulate very large systems beyond usual atomic methods.
What this means in practice
- •For materials scientists: Simulate microstructure changes in alloys and compounds at large scales with quantum-level accuracy for advanced material design.
- •For planetary scientists: Model phase separation processes in planetary interiors, such as hydrogen-helium mixing and droplet formation under extreme conditions.
Authors
Mengyi Chen, Peichen Zhong, Zihan Zhang, Qianxiao Li
Abstract
Simulating microstructure evolution requires quantum-mechanical accuracy and mesoscopic reach in length and time scales, a combination that no current method achieves. Classical phase-field models provide this reach, but their accuracy is limited by phenomenological free energies and mobilities. Here we develop a framework for learning ab initio phase-field models, where the mesoscopic equation is not postulated but derived from a Mori-Zwanzig projection of molecular dynamics onto species-density fields under explicit assumptions. The nonlocal free energy and mobility left unspecified by this equation are parametrized by neural networks and learned from short molecular dynamics trajectories generated with machine-learning interatomic potentials of ab initio accuracy. We demonstrate the framework on an iron-boron melt and on hydrogen-helium mixtures under planetary conditions. For iron-boron, the model shows that the melt at the FeB$_4$ composition is spinodally unstable at ambient pressure but stabilized at 10 GPa, offering a thermodynamic rationale for why FeB$_4$ has been synthesized only under high pressure. For hydrogen-helium, the model predicts the immiscibility boundary and captures droplet nucleation and growth in helium-rain simulations of a column corresponding to 2.2 million atoms, far beyond the scale of atomistic modeling at comparable accuracy. Trained across compositions and conditions, such models could provide a mesoscopic counterpart to ab initio molecular dynamics.