Application of pretrained universal machine-learning interatomic potential for physicochemical simulation of liquid electrolytes in Li-ion batteries

2025/01/01 by Suyeon Ju, Jinmu You, Gijin Kim +3 · 1 voice · 6 citations
Engineering · Materials Science · #Advanced Battery Technologies Research #Machine Learning in Materials Science

paper · pdf · doi:10.1039/d5dd00025d

openalex publication_date 2025/01/01 · openalex created_date 2025/05/07 · openalex updated_date 2026/07/29

Abstract

SevenNet-0, pretrained on inorganic data, performs surprisingly well on liquid electrolytes—an out-of-distribution system—showing interpolation ability and improving further with fine-tuning.

Citations

• Effect of the damping function in dispersion corrected density functional theory
• Molecular design for electrolyte solvents enabling energy-dense and long-cycling lithium metal batteries
• Gaussian Process Regression for Materials and Molecules
• LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Rechargeable Batteries for Grid Scale Energy Storage
• A universal graph deep learning interatomic potential for the periodic table
• Scaling deep learning for materials discovery
• Structure and transport properties of LiTFSI-based deep eutectic electrolytes from machine-learned interatomic potential simulations
• From ultrasoft pseudopotentials to the projector augmented-wave method
• PACKMOL: A package for building initial configurations for molecular dynamics simulations
• A consistent and accurateab initioparametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu

Cited by

• Bridging Quantum Mechanics to Organic Liquid Properties via a Universal Force Field
• Beyond Scaling: Chemical Intuition as Emergent Ability of Universal Machine Learning Interatomic Potentials
• An efficient forgetting-aware fine-tuning framework for pretrained universal machine-learning interatomic potentials
• Neural Network-Driven Molecular Insights into Alkaline Wet Etching of GaN: Toward Atomistic Precision in Nanostructure Fabrication
• Coarse-grained graph architectures for all-atom force predictions
• Prediction and Experimental Verification of Electrolyte Solvation Structure from an OMol25-Trained Interatomic Potential

Discussions

• Ju, You, et al. evaluate pretrained ML potential SevenNet-0 on liquid electrolytes. Trained on inorganic data, SevenNet-0 performs unexpectedly well on these out-of-distribution systems, showing inter [bsky, 1 points, 0 comments]

Related

• Application of pretrained universal machine-learning interatomic potential for physicochemical simulation of liquid electrolytes in Li-ion battery
• Data-driven composition-only machine learning for high-performance solid-state electrolytes
• A Hybrid Physics-Driven Neural Network Force Field for Liquid Electrolytes
• Universal Machine Learning Interatomic Potentials are Ready for Solid Ion Conductors
• Uni-ELF: A Multi-Level Representation Learning Framework for Electrolyte Formulation Design