Spatial Quantification of Microstructural Degradation During Fast Charge in 18650 Lithium-Ion Batteries Through Operando X-Ray Micro Tomography and Euclidean Distance Mapping

Abstract

Worldwide market adoption of electric vehicles requires a reduction in current recharge times. Unfortunately, fast charging at rates above 1C aggressively accelerates degradation through structural damage induced by increases in local temperature and inhomogeneous transport of charge within the architectures. At the micron scale, the first indication of damage is irreversible expansion and contraction of the electrode layers indicating reduced coulombic efficiency. Electrode damage often involves void formation between the active material and conductive-binder matrix. Quantification of this evolution must be carried out in real time and, thus, non-destructively. We report the operando X-ray microtomography of cylindrical cells used in EVs under fast charge cycling. Two 18650 batteries were measured during cycling after antecedent fast charging cycles; the 3rd, 81st and 82nd cycles, to track morphological damage at different points of battery life. We employed a deep learning segmentation method using the U-Net convolutional neural network to objectively quantify the electrode degradation. Using a Euclidean Distance Mapping method, reversible and irreversible dilation and voids were spatially resolved in three-dimensions within the electrode layers. Insight into how fast charging induces structural damage will better inform research into fast-charge protocols and new battery chemistries for electrolytes, electrolyte additives, and novel electrode architectures.

Keywords: lithium-ion battery, Operando, Microcomputed tomography, Fast charge, Deep learning, Electrode dilation