Battery noises decoded to reveal cracks, gas, and safety clues

Researchers at MIT have developed a novel method to decode faint acoustic signals emitted by lithium-ion batteries during charge and discharge cycles, linking specific sound patterns to internal degradation processes such as gas generation and electrode material fracturing. By combining electrochemical testing with sensitive acoustic recordings under real-world conditions, the team was able to noninvasively monitor battery health, identifying distinct acoustic signatures even amid noisy data. This approach was validated through electron microscopy, confirming the correlation between sounds and internal battery damage. Unlike previous methods that relied on simple sound thresholds, the MIT researchers employed advanced wavelet transforms to isolate meaningful signals from background noise, similar to techniques used in structural health monitoring of bridges. This acoustic monitoring provides an additional diagnostic tool beyond traditional voltage and current measurements, offering insights into battery lifespan and safety risks, including early warnings of thermal runaway. The technology has immediate applications in material research, manufacturing quality control, and electric vehicle monitoring, with ongoing collaborations such as one with Tata Motors to implement real-world diagnostic systems. The study