Advancing Material Science—AFM-in-SEM for Battery Analysis

Although lithium-ion batteries, as we know them today, are not a new invention, they are still nowhere near the theoretical limit of capacity and energy density. A great amount of research is to be done, delving deeper and deeper into details. Therefore, with the advances in batteries, the measurement and imaging techniques must advance as well.<br/>While the Scanning electron Microscope (SEM) is the go-to instrument for observing battery samples on the microscale, electrical measurements are still mainly done using a large-scale statistical approach. Atomic Force Microscope (AFM) can provide detailed information about the electrical properties, but it is usually hindered by the nature of battery samples. Their fragile surface and challenging geometry (e. g., narrow tape cross-sections) do not lend themselves well to physical imaging with a sharp tip. However, combining AFM and SEM can avoid some of these pitfalls. For example, one can quickly survey a large sample area with SEM and select interesting or hard-to-reach places to be scanned by AFM. Navigating the tip precisely to the area of interest without touching the sample before imaging saves the tip, time and effort.<br/>Analyzing solid-state batteries comes with additional challenges, such as their high sensitivity to humidity. This necessitates using a sample transfer system that protects the sample from the atmosphere or performing most of the analysis in situ.<br/>We present a workflow for analyzing air-susceptible samples in-situ using a combination of AFM-in-SEM. The method is showcased in a study of electron conductivity of a mixed active material cathode, containing Lithium-Nickel-Manganese-Cobalt oxide (NMC) and Lithium-Nickel-Cobalt-Aluminium oxide (NCA).<br/>The sample (a cross-section of a slurry-cast tape) was prepared in a glovebox, then polished using a Broad Ion Beam polisher, and then imaged with AFM inside SEM. All transfers were done in a protective argon atmosphere. Conductivity maps were taken on a single polycrystalline particle of NCM and a pair of neighboring NCA and NMC particles for comparison. The particles were distinguished using Enertg Dispersive Spectroscopy (EDS). Contrary to expectation, the conductivity differed by several orders of magnitude between the particles. The reason was likely not just the material difference but an additional effect, such as separation from the current collector. Similar differences were observed between grains inside the polycrystalline NMC particles. Here, we suspect a combination of grain separation and crystallographic orientation is responsible for the conductivity pattern. The distribution of conductivity can help diagnose possible failure vectors, especially if correlated with EDS or other advanced SEM imaging techniques.