A comprehensive and quantitative SEM-EDS analytical process applied to lithium-ion battery electrodes.

The combination of scanning electron microscopy (SEM) images and energy-dispersive X-ray spectroscopy (EDS) maps (SEM-EDS analysis) enables the analysis of the relationship between the microstructures and elemental compositions of the surfaces of materials. However, conventional SEM-EDS analyses lack comprehensiveness and quantitativeness, resulting in potential inaccuracies in reflecting the properties of the entire sample and variations in the results depending on the analyst. Therefore, herein, we propose an objective SEM-EDS analytical process that addresses the aforementioned issues. Comprehensiveness was addressed by acquiring large volumes of SEM images through automated capturing, whereas quantitativeness was addressed through microstructural analysis of the SEM images based on image features, model-based dimension reduction and clustering methods, and similarity analysis of the elemental distribution in EDS maps based on statistical distances. The proposed method was used to analyze the degradation of lithium-ion battery electrodes, affording objective results that align with subjective insights into the changes in the morphology and composition of solid electrolyte interphase (SEI) films accompanying degradation.