Researchers at the University of California San Diego have unveiled a novel method to characterize lithium metal battery performance and predict failure, leveraging widely used scanning electron microscopy (SEM) in conjunction with a straightforward algorithm. This significant advance, detailed in the Proceedings of the National Academy of Sciences, promises to accelerate the development of safer, longer-lasting, and more energy-dense batteries crucial for electric vehicles and grid-scale energy storage.

Lithium metal batteries hold the potential to double the energy density of current lithium-ion counterparts, offering extended range for EVs and prolonged device runtimes. However, their widespread adoption has been hindered by challenges in controlling lithium morphology—how lithium deposits on electrodes during cycling. Uneven deposition leads to the formation of needle-like dendrites, which can pierce separators, causing short circuits and catastrophic battery failure.

Historically, researchers have relied on visual assessment of microscope images to determine lithium uniformity, a practice that has led to inconsistent analyses and difficulties in comparing results across different research groups. "What one battery group may define as uniform might be different from another group’s definition," stated Jenny Nicolas, a materials science and engineering Ph.D. candidate at the UC San Diego Jacobs School of Engineering and the study’s first author. "The battery literature also uses so many different qualitative words to describe lithium morphology—words like chunky, mossy, whisker-like and globular, for example. We saw a need to create a common language to define and measure lithium uniformity."

To address this, Nicolas and her colleagues, under the guidance of Professor Ping Liu, developed an open-source algorithm that quantifies lithium uniformity from SEM images. The method converts grayscale SEM images into black and white pixels, where white represents lithium deposits. By dividing images into regions and counting white pixels, the algorithm calculates an "index of dispersion" (ID). Nicolas explained, "The closer it is to zero, the more uniform the lithium deposits. A higher value means less uniformity and more clustering of lithium particles in certain areas."

Validation on 2,048 synthetic SEM images confirmed the algorithm's accuracy. When applied to real electrode images, the team observed that as batteries cycled, the ID consistently increased, indicating less uniform lithium deposits and correlating with increased energy required for deposition—a clear sign of degradation. Crucially, the researchers identified that local peaks and dips in the ID frequently appeared just prior to cell failure, suggesting these fluctuations could serve as an early warning signal for impending short circuits. A significant advantage of this method is its accessibility; battery researchers can readily integrate this algorithm into their existing SEM imaging workflows, maximizing the utility of already collected data. "Our tool can be employed as a low-hanging fruit for researchers to take their analysis to the next level by utilizing image analysis to its fullest potential," Nicolas concluded. This work was supported by the Office of Vehicle Technologies of the U.S. Department of Energy through the Advanced Battery Materials Research Program (Battery500 Consortium).