Theoretical and Computational Chemistry

Machine Learning 3D-Resolved Prediction of Electrolyte Infiltration in Battery Porous Electrodes

Alejandro A. Franco Université de Picardie Jules Verne - LRCS (UMR CNRS 7314) - Amiens, France

Abstract

Electrolyte infilitration is one of the critical steps of the manufacturing process of lithium ion batteries. Along with being the most time-consuming step in manufacturing, electrolyte wetting directly impacts the battery cell energy density, power density and cycle life. We present here an innovative machine learning model to fast and accurately predict electrolyte infiltration in three dimensions in lithium ion battery electrodes. Our machine learning model is able to speed up the infiltration predictions by several orders of magnitude compared to a physics-based model based on Lattice Boltzmann Method, paving the way towards massive computational screening of electrode mesostructures/electrolyte pairs to unravel their impact on the cell wetting and optimize electrolyte infiltration conditions.

Content

Thumbnail image of Battery electrolyte Infiltration_Machine Learning_Shodiev Duquesnoy Arcelus Chouchane Li Franco 2021.pdf
download asset Battery electrolyte Infiltration_Machine Learning_Shodiev Duquesnoy Arcelus Chouchane Li Franco 2021.pdf 1 MB [opens in a new tab]