Distilling Large-Scale Machine Learning Interatomic Potentials to Predict Long-Timescale Electrolyte Transport Properties
Abstract
Accurately predicting the transport properties of liquid electrolytes is critical for next-generation battery design and requires force fields that are both quantum-mechanically accurate and fast enough for long-timescale simulation. Recent large-scale pretrained machine learning interatomic potentials (MLIPs) approach density-functional-theory accuracy, but remain too slow to access the necessary timescales to converge transport properties. We provide a knowledge distillation procedure to transfer the accuracy of a large-scale "teacher" MLIP into fast, system-specific "students", without any additional ab initio data. Our distilled models run ~40x faster, reproduce teacher-predicted structural observables, and produce stable 20 ns trajectories in days, rather than the months the teacher would require. Across a diverse set of sodium- and lithium-ion electrolytes, student-predicted self-diffusivities, ionic conductivities, and shear viscosities track experiment and outperform classical force-field predictions, establishing distillation as a practical route to accurate, long-timescale simulations necessary for high-throughput electrolyte screening.
Related
- Concentration-Dependent Speciation in ZnCl2 Electrolyte Solutions: Transferability of Machine-Learning Interatomic Potentials and Speciation Sensitivity of EXAFS
- Predicting Thermophysical Properties of Halide Salts Using On-the-Fly Machine-Learned Interatomic Potentials Across Multiple Exchange-Correlation Functionals
- Distilling First-Principles Accuracy into Compact Machine Learning Potentials for Condensed-Phase Chemistry
- Correcting the Liquid Densities of Universal Machine-Learning Interatomic Potentials Using Local-Number-Density Repulsion
- Foundation interatomic potentials learn the practical chemistry they were never taught
BibTeX
@misc{amin2026distilling,
title = {{Distilling Large-Scale Machine Learning Interatomic Potentials to Predict Long-Timescale Electrolyte Transport Properties}},
author = {Ishan Amin and Yue Jian and Nitesh Kumar and Christine Zhang and Jianwei Lai and Casey S. Mezerkor and Aaron M. Melemed and Jonathan R. Thurston and Neil P. Dasgupta and J. David Bazak and Ethan J. Crumlin and Samuel Blau and Aditi S. Krishnapriyan},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.39513},
eprint = {2610.39513},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.39513},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Distilling Large-Scale Machine Learning Interatomic Potentials to Predict Long-Timescale Electrolyte Transport Properties
AU - Ishan Amin
AU - Yue Jian
AU - Nitesh Kumar
AU - Christine Zhang
AU - Jianwei Lai
AU - Casey S. Mezerkor
AU - Aaron M. Melemed
AU - Jonathan R. Thurston
AU - Neil P. Dasgupta
AU - J. David Bazak
AU - Ethan J. Crumlin
AU - Samuel Blau
AU - Aditi S. Krishnapriyan
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.39513
UR - https://arxiv.org/abs/2610.39513
AB - Accurately predicting the transport properties of liquid electrolytes is critical for next-generation battery design and requires force fields that are both quantum-mechanically accurate and fast enough for long-timescale simulation. Recent large-scale pretrained machine learning interatomic potentials (MLIPs) approach density-functional-theory accuracy, but remain too slow to access the necessary timescales to converge transport properties. We provide a knowledge distillation procedure to transfer the accuracy of a large-scale "teacher" MLIP into fast, system-specific "students", without any additional ab initio data. Our distilled models run ~40x faster, reproduce teacher-predicted structural observables, and produce stable 20 ns trajectories in days, rather than the months the teacher would require. Across a diverse set of sodium- and lithium-ion electrolytes, student-predicted self-diffusivities, ionic conductivities, and shear viscosities track experiment and outperform classical force-field predictions, establishing distillation as a practical route to accurate, long-timescale simulations necessary for high-throughput electrolyte screening.
ER -