Influence of voltage profile and fitting technique on the accuracy of lithium-ion battery degradation identification through the Voltage Profile Model

Abstract

Accurate estimation of degradation in lithium-ion batteries is essential in predicting remaining useful life and understanding how to better operate batteries for extended lifetimes. A common way of estimating degradation modes in lithium-ion batteries is to analyse the cells voltage profile through techniques such as incremental capacity analysis, differential voltage analysis or direct fitting of half-cell potential profiles. To extract degradation modes such as loss of lithium inventory or loss of active material requires accurate knowledge of the individual electrode half-cell potential profiles, which is often not available for commercial cells without performing a cell teardown. This work investigates how the choice of half-cell potential profile influences the accuracy of a parametric voltage profile model to estimate electrode capacity and simulated degradation modes through a combination of half-cell and three electrode testing on a LiNi0.5Mn0.3Co0.2O2/Graphite cell. Results demonstrate that whilst half-cell potential data from the same electrode material batch gives the most accurate voltage profile fit, other data sources for the same electrode chemistry can also accurately estimate individual electrode capacity in a fresh cell. However, when degradation modes are induced into the voltage profile, only the model using half-cell profiles obtained from similar sources to the full-cell configurations are able to accurately distinguish between loss of lithium inventory and loss of active material of the positive electrode. It is also shown that the diagnostic accuracy of the parametric voltage profile model can be improved for all data sources through combined fitting of the voltage, incremental capacity and differential voltage analysis compared to voltage profile fitting alone.