Using Flow to Control Lithium-Ion Battery Crosstalk for Better Lifetime Predictions

Abstract

High-voltage lithium-ion batteries suffer from accelerated capacity fade due to crosstalk reactions between their electrodes. For example, while the battery is operating at a high voltage, transition metals from cathode active material dissolve into the electrolyte and diffuse to the anode. These metals disrupt the formation, growth, and performance of the solid-electrolyte interface (SEI), causing accelerated capacity fade. Although the effects of crosstalk reactions on battery lifetime and coulombic efficiency are well documented in full-cell studies, the mechanism by which accelerated capacity fade occurs is currently poorly understood.In this study, we show that a combination of collector-generator experiments and single particle battery modeling can be used to better understand the effects and mechanisms of crosstalk reactions on capacity fade. Using a microfluidic flow cell, lined with commercial high-voltage lithium-ion battery electrodes and filled with commercial lithium-ion battery electrolyte, we are able to replicate commercial battery conditions while studying crosstalk reaction fundamentals. Because of forced convection in the cell, crosstalk species generated at the upstream electrode react at the downstream electrode and do not diffuse back to the upstream electrode. This allows us to isolate effects of crosstalk reactions on the downstream electrode and more clearly determine their role in capacity fade.From this study, we present a series of proof-of-concept experiments, in which we show that the charge/discharge behavior of each electrode will clearly depend on flow rate and direction. In addition, we also demonstrate that, using electroanalytical techniques, crosstalk reaction thermodynamic and kinetic parameters can be determined. These parameters are then implemented into a single particle model to elucidate the effects of crosstalk on battery lifetime. This application highlights the novelty of the microfluidic flow cell in deconvoluting crosstalk reactions for generating meaningful inputs in physics-based battery lifetime models.