Comparing fine-tuning strategies of MACE machine learning force field for modeling Li-ion diffusion in LiF for batteries
Machine learning force fields (MLFFs) are transforming materials science and engineering by enabling the study of complex phenomena, such as those critical to battery operation. In this work, we explore the predictive capabilities of pre-trained and fine-tuned MACE MLFF and compare different fine-tuning strategies for predicting interstitial lithium diffusivity in LiF, a key component in the solid electrolyte interphase in Li-ion batteries. Our results demonstrate that the MACE-MPA-0 foundational model achieves comparable accuracy to well-trained DeePMD, in predicting key diffusion properties based on large scale molecular dynamics simulation, while requiring minimal or no training data. For instance, the MACE-MPA-0 predicts an activation energy $E_a$ of 0.22eV, the fine-tuned model with only 300 data points predicts $E_a =$ 0.20eV, both of which show good agreement with the DeePMD model reference value of $E_a = $ 0.24eV. In this work, we provide a solid test case where fine-tuning approaches, whether using data generated for DeePMD or data produced by the foundational MACE model itself, yield similar robust performance to the DeePMD potential trained with over 40,000 actively learned data, albeit requiring only a fraction of the training data.