Operando particle-scale characterization of silicon anode degradation during cycling by ultrahigh-resolution x-ray microscopy and computed tomography

February 05, 2021

Paul Choi (1), Bharathy S. Parimalam (1), Laisuo Su (1), B. Reeja-Jayan (1), Shawn Litster (1)
ACS Applied Energy Materials, 4, Issue 2, February 2021: 1657–1665. DOI: 10.1021/acsaem.0c02823


lithium-ion battery; silicon anode; time-resolved; high-resolution; operando X-ray microscopy; in situ X-ray computed tomography


A combination of two-dimensional (2D) operando transmission X-ray radiograph sequences and three-dimensional (3D) in situ X-ray computed tomography was used to characterize a composite silicon anode during cycling in an electrode and coin cell format consistent with commercial lithium-ion (Li-ion) batteries. Silicon (Si) particle expansion and phase transformation within the porous electrode were imaged continuously during cycling at various rates at O(100 nm) resolution within a large O(100 μm) region of interest that capture electrode-scale effects. The imaging utilizes the substantial change in the 8 keV X-ray absorption coefficient with lithium (Li) alloying of Si during charging. At low rate cycling, the X-ray signal attenuation over the Si particles decreased with increased lithiation. In contrast, at high rate cycling, we observe increased attenuation at the electrode scale. A useful feature of this operando imaging is the simultaneous imaging of a large number of particles in close proximity. To capture the transformations of such a large number of Si during cycling, we introduce a standard deviation analysis of the operando transmission X-ray radiograph sequences. At key instances in the cycling, the same region of interest from the radiographs was reconstructed into 3D volumes. Si particle fracture, electrode expansion, and particle detachment from the current collector were all observed in the reconstructed volumes. This study demonstrates the unique capability of the combined 2D operando and 3D in situ X-ray imaging techniques in investigating the dynamic behavior of battery materials at the sub-micrometer particle scale in commercially relevant electrode formats.

How Our Software Was Used

Dragonfly was used to visualize segmented volumetric structures of an Si electrode.

Author Affiliation

(1) Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States.