Advanced battery cathode microstructure analysis through operando synchrotron X-ray tomography and super-resolution deep learning

Lithium ion batteries are pivotal for clean energy storage and mitigating climate change. In this study, we employ operando synchrotron X-ray computed tomography to investigate the dynamic evolution of battery cathode microstructure. We focus on tracking changes in porosity and pore size distribution at the microscale and cathode thickness at the macroscale during the lithiation and delithiation processes within a commercially configured battery. Image quality was enhanced using both conventional image processing methods and a Super-Resolution Convolutional Neural Network (SRCNN) model. Our findings revealed a slight increase in the cathode solid volume fraction and specific surface area as the battery transitioned from its pristine state to fully lithiated, followed by a reduction during delithiation. This behavior was attributed to the expansion of the cathode material and phase transitions during lithiation, which split larger pores into smaller ones, as evidenced by the increase in surface area. Cathode thickness also exhibited expansion during lithiation and contraction during delithiation. These results offer valuable insights into the structural changes that contribute to battery aging, helping researchers better understand how these different parameters change over time. This understanding is crucial for designing more durable and sustainable batteries in the future, both in terms of specific design and material selection, to enhance resistance during charge and discharge cycles to improve performance and longevity. • Operando synchrotron XCT reveals cathode microstructure evolution during cycling. • SRCNN enhances XCT image quality for improved pore size distribution analysis. • Cathode expands during lithiation, dividing large pores into smaller ones. • Framework developed to study commercial batteries with synchrotron imaging.