A predictive machine learning force field framework for liquid electrolyte development

2024-04-10

AI Summary

To address key limitations of machine learning force fields (MLFFs) in molecular dynamics (MD) simulations of liquid electrolytes—namely, insufficient accuracy, poor generalizability, and large deviations from experimental densities—this work proposes BAMBOO, a novel MLFF framework. First, it introduces a physics-informed, graph-equivariant Transformer architecture that rigorously enforces rotational and translational covariance. Second, it pioneers an integrated knowledge distillation strategy to substantially suppress trajectory fluctuations during MD sampling. Third, it incorporates a density alignment algorithm enabling end-to-end calibration of simulated densities to experimental values. By synergistically integrating graph neural networks, equivariant deep learning, and quantum-mechanics-driven training, BAMBOO achieves an average density error of only 0.01 g/cm³ across 15+ lithium-ion battery electrolyte components, while accurately predicting viscosity and ionic conductivity. Its overall performance sets a new state-of-the-art.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Despite the widespread applications of machine learning force fields (MLFF) in solids and small molecules, there is a notable gap in applying MLFF to simulate liquid electrolyte, a critical component of the current commercial lithium-ion battery. In this work, we introduce BAMBOO ( extbf{B}yteDance extbf{A}I extbf{M}olecular Simulation extbf{Boo}ster), a predictive framework for molecular dynamics (MD) simulations, with a demonstration of its capability in the context of liquid electrolyte for lithium batteries. We design a physics-inspired graph equivariant transformer architecture as the backbone of BAMBOO to learn from quantum mechanical simulations. Additionally, we introduce an ensemble knowledge distillation approach and apply it to MLFFs to reduce the fluctuation of observations from MD simulations. Finally, we propose a density alignment algorithm to align BAMBOO with experimental measurements. BAMBOO demonstrates state-of-the-art accuracy in predicting key electrolyte properties such as density, viscosity, and ionic conductivity across various solvents and salt combinations. The current model, trained on more than 15 chemical species, achieves the average density error of 0.01 g/cm$^3$ on various compositions compared with experiment.

Problem

• Develop MLFF for liquid electrolyte simulation in batteries
• Improve accuracy in predicting electrolyte properties like density
• Align simulation results with experimental measurements effectively

Innovation

• Physics-inspired graph equivariant transformer architecture
• Ensemble knowledge distillation for MLFFs
• Density alignment algorithm for experimental accuracy