Revisiting Pseudo-OCV Pulse-Based Incremental Capacity Analysis

Abstract

Li-ion batteries are more prone to safety incidents than other battery technologies. This is due to the inherently higher energy density as well as the flammability of especially anode materials, separator, and electrolyte solvents. During charging, metallic Lithium may form on the anode surface and could in the worst case short the battery from within and cause a fire and or explosion [1]. It is therefore vital to be able to monitor the battery’s state-of-health (SoH) and state-of-safety (SoS) [2], to avoid further use of Li-ion batteries that may have aged in a detrimental fashion and can have an increased safety risk. Especially when a battery has aged considerably (SoH < 80%) the path of ageing can significantly affect the safety properties of the battery and cause dramatic differences in the severity of a possible safety incident.Incremental capacity analysis (ICA, dQ/dV) has emerged as a very powerful diagnostic technique for Li-ion batteries. We recently used ICA to identify different degradation mechanisms in commercial Li-ion cells through classification and tracking of selected dQ/dV features [3]. This allowed us to identify safety critical ageing at an early stage. ICA requires constant charge and discharge at slow currents (C-rate < C/10) [4, 5]. However, a slow controlled constant current charge or discharge is normally not feasible and cannot be easily applied to battery systems without access to high precision battery pack testers.In this work we will revisit applying ICA on the Open-Circuit-Voltage (OCV) curve in the capacity space [6]. The OCV curve can be obtained from any sequence of current or power pulses followed by a rest period to allow the cell to reach a pseudo-OCV after each pulse. By pulsing through the entire state-of-charge window an OCV vs capacity curve can be obtained with sufficient accuracy to perform ICA.A direct comparison of conventional constant current ICA (cc-ICA) and high-resolution-OCV ICA (ocv-ICA) is presented. A strong correlation between ageing patterns is observed providing a first proof-of-concept for the method.