From electrons to phase diagrams with classical and machine learning potentials: automated workflows for materials science with pyiron

TL;DR

This paper introduces a comprehensive, automated workflow within the pyiron environment for developing, fitting, and validating machine learning and empirical potentials in materials science, demonstrated on diverse potential types and a phase diagram for Al-Li.

Contribution

It presents an integrated framework for automated MLP development and validation in pyiron, covering database creation, fitting, and testing across different potential models.

Findings

Framework successfully automates potential development process.

Demonstrated on empirical, neural network, and basis set potentials.

Applied to compute a phase diagram for Al-Li alloy.

Abstract

We present a comprehensive and user-friendly framework built upon the pyiron integrated development environment (IDE), enabling researchers to perform the entire Machine Learning Potential (MLP) development cycle consisting of (i) creating systematic DFT databases, (ii) fitting the Density Functional Theory (DFT) data to empirical potentials or MLPs, and (iii) validating the potentials in a largely automatic approach. The power and performance of this framework are demonstrated for three conceptually very different classes of interatomic potentials: an empirical potential (embedded atom method - EAM), neural networks (high-dimensional neural network potentials - HDNNP) and expansions in basis sets (atomic cluster expansion - ACE). As an advanced example for validation and application, we show the computation of a binary composition-temperature phase diagram for Al-Li, a technologically important lightweight alloy system with applications in the aerospace industry.