Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comparative Study of Molecular Dynamics Approaches for Simulating Ionic Conductivity in Solid Lithium Electrolytes

Dounia Shaaban Kabakibo, Félix Therrien, Yoshua Bengio, Michel Côté, Hongyu Guo, Homin Shin, Alex Hernandez-Garcia

Abstract

Accurate prediction of ionic conductivity is critical for the design of high-performance solid-state electrolytes in next-generation batteries. We benchmark molecular dynamics (MD) approaches for computing ionic conductivity in 21 lithium solid electrolytes for which experimental ionic conductivity has been previously reported in the literature. In particular, we compare simulations driven by density functional theory (DFT) and by universal machine-learning interatomic potentials (uMLIPs), namely a MACE foundation model. We find comparable performance between DFT and MACE, despite MACE on one GPU more than 350 times faster than DFT on a 64-CPU node. The framework developed here is designed to enable systematic comparisons with additional uMLIPs and fine-tuned models in future work.

A Comparative Study of Molecular Dynamics Approaches for Simulating Ionic Conductivity in Solid Lithium Electrolytes

Abstract

Accurate prediction of ionic conductivity is critical for the design of high-performance solid-state electrolytes in next-generation batteries. We benchmark molecular dynamics (MD) approaches for computing ionic conductivity in 21 lithium solid electrolytes for which experimental ionic conductivity has been previously reported in the literature. In particular, we compare simulations driven by density functional theory (DFT) and by universal machine-learning interatomic potentials (uMLIPs), namely a MACE foundation model. We find comparable performance between DFT and MACE, despite MACE on one GPU more than 350 times faster than DFT on a 64-CPU node. The framework developed here is designed to enable systematic comparisons with additional uMLIPs and fine-tuned models in future work.

Paper Structure

This paper contains 13 sections, 3 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Contributions
  3. Methods
  4. Ionic Conductivity from Molecular Dynamics Simulations
  5. Materials and parameters choice
  6. Results
  7. Discussion and conclusion
  8. Appendix
  9. Related work
  10. Future work
  11. Materials Selection
  12. AIMD Parameters
  13. Additional analysis

Figures (3)

  • Figure 1: Parity plot comparing ionic conductivity computed using (a) MACE and (b) DFT foundation model with experimental values reported by therrien2025obelix. The size of the circles reflect the number of simulations (at 5 different temperatures) that reached at least $\mathrm{MSD} = 20~\mathrm{\AA}^2$.
  • Figure 2: Parity plot comparing ionic conductivity computed using DFT and MACE foundation model. The size of the half circles reflect the number of simulations (at 5 different temperatures) that reached at least $\mathrm{MSD} = 20~\mathrm{\AA}^2$.
  • Figure 3: Logarithmic error between experimental and computed ionic conductivities for each material. The horizontal dashed line indicates perfect agreement. The x-axis lists materials equally spaced for clarity, with experimental $\log_{10}\sigma_{\rm exp} \,[\mathrm{mS/cm}]$ values shown below the labels. Materials are ordered by increasing experimental conductivity.