Rapid prediction of electrolyte-based cell failure in retired Li-ion electric vehicle pouch cells

Abstract Lithium-ion batteries (LIBs) exhibit sudden non-linear degradation (known as the cell knee point) in operating use, but a quick, simple, and industrially applicable method of predicting knee point failure remains elusive. Addressing this gap, we show for the first time that the Warburg coefficient (readily obtainable from EIS measurement) can predict electrolyte-based knee point failure, well before extreme capacity degradation, in retired electric vehicle pouch cells. The technique is found to be effective across a wide range of battery states of charge (SOC), and the information can be extracted from the pouch cells in only 10 seconds. In principle this enables early detection of electrolyte-based knee point failure within commercial electric vehicle cells if applied ex situ through offline diagnosis (energy storage system) within second life battery application systems, thus allowing greater predictability with regard to the remaining lifetime of the battery.