Tyler M. McGee

Lithium-ion batteries (LIBs) are the leading technology used in consumer electronics, electric vehicles, and grid-level electrochemical energy storage applications. The ever-increasing use of LIBs has highlighted a gap in understanding of their behavior throughout their life cycle. Current monitoring systems rely on electrical and sometimes temperature measurements to assess the internal state which limits information about complex electrochemical processes. In response, ultrasonic testing (UT) has shown promise for non-invasive assessment due to its ease of use and sensitivity to mechanical changes which are correlated with electrochemical changes within the battery. We summarize the research in UT methods applied to LIBs throughout their life cycle. We also discuss physics-based and data-driven modeling approaches used to interpret ultrasonic signals in the context of LIBs, with an emphasis on the existing challenge of establishing rigorous links between electrochemical behavior and elastic and poroelastic wave physics to gain insight regarding physical changes in the LIB that can be directly measured using UT. Finally, we discuss the challenges of implementing UT across the LIB life cycle and identify opportunities for further research. This review aims to provide helpful guidance to researchers and practitioners of UT in the growing field of UT for electrochemical battery systems.

Lithium-ion batteries have become a ubiquitous energy source for applications ranging from cell phones to battery-powered electric vehicles. Despite this success, state-of-charge (SOC) and state-of-health (SOH) estimation remains challenging, particularly in applications requiring high numbers of cells like battery electric vehicles. Often the operating conditions and released capacity of individual cells in a pack are not directly monitored. Inaccurate estimates of cell SOC or SOH can lead to overcharge or overdischarge which could result in thermal runaway. This work explores the relationship between the modal response of a clamped-clamped 10 Ah NMC/graphite pouch cell and its SOC and SOH. We employ scanning laser Doppler vibrometry on a cell at 0\% and 100\% SOC for over 80 charge-discharge cycles. Electrochemically-induced mechanical property change is reflected in the shift in resonance frequency and the magnitude of the transmissibility curve at resonance. We also present an effective medium model (EMM) of the cell based on multiscale homogenization for use in finite-element software to simulate the cell’s modal response as a function of SOC. Results indicate that modal analysis can be used to estimate cell SOC and SOH, while the EMM provides insight about which cell components contribute to the cell’s dynamic mechanical response.

Accurately monitoring the safety of lithium-ion batteries to prevent a thermal runaway (TR) event is of utmost importance especially in high-power applications such as electric vehicles. The fire caused by a lithium-ion battery is particularly difficult to extinguish because of its continuous emittance of highly flammable gas and its extremely high temperature. Current battery management systems (BMS) only monitor cell safety by measuring voltage, current, and temperature on the module level. Because the BMS operates on the module-level as opposed to the cell-level, these metrics can fail to predict a TR event, which may begin with individual cell failures. We report on the viability of early detection of cell damage by simultaneously monitoring voltage, current, temperature, mechanical clamping force, and ultrasonic wave transmission through individual battery cells under various electrical, mechanical, and thermal loading conditions. Time- and frequency-domain features of the acoustic signals transmitted along three propagation paths and excitation frequencies between 0.1 and 1 MHz are monitored and correlated with electrical, temperature, and force sensor data under mechanically clamped overcharge and thermal abuse scenarios. Early indicators of damage in the ultrasonic signals are presented and discussed with respect to propagation paths and known damage mechanisms in lithium-ion batteries.