Optimizing multilayer graphite-silicon anodes: A computational approach to enhancing lithium-Ion battery performance

Abstract

This study evaluated the performance of multilayer anodes for lithium-ion batteries, composed of an outer graphite layer in direct contact with the electrolyte and an inner graphite–silicon composite layer, using finite-element simulations and multivariate statistical analysis. Various silicon contents such as 10, 20 percent and 30 %, layer thickness configurations including 30–30 µm, 20–40 µm and 10–50 µm, and graphite particle sizes of 2.5, 5 and 7.5 µm were systematically examined while maintaining a total anode thickness of 60 µm. In addition, the cathode material NMC 622 and the electrolyte LiPF6 in 3:7 EC:EMC were specified in the simulated cell configuration. The methodology integrated COMSOL Multiphysics® simulations with a simulation design (DOE) constructed in JMP, enabling the identification of key response parameters such as capacity loss percentage, solid-electrolyte interphase (SEI) layer thickness, potential drop across the SEI and electrolyte consumption over 2000 simulated cycles. Simulation results indicated that a 30–30 µm configuration, employing 2.5 µm graphite particles and a silicon content in the range of 20–30 % within the composite layer, substantially reduces potential drop, electrolyte consumption and SEI growth compared to modeled single-layer 100 % graphite or homogeneous silicon–graphite anodes. These findings underscore the viability of dual-layer structures for leveraging silicon’s high theoretical capacity without compromising electrochemical stability, and they highlight the crucial role of simulation-driven optimization in predicting long-term performance in batteries with enhanced energy density and extended cycle life.