Exact knowledge of the relationship between open circuit voltage (OCV) and state of charge (SOC), often referred to as OCV curve, is crucial for most algorithms for the estimation of both the SOC and the state of health (SOH) of lithium-ion batteries. Many researchers showed that the form of the OCV curve changes during battery aging. In order to achieve a precise estimation of the SOC and the SOH for aged batteries, these changes in the form of the OCV curve have to be considered by the algorithm. The OCV curve of a full cell corresponds to the difference between the OCV curves of the two electrodes inside the cell. The most popular theory describing the change in the OCV curve during aging assumes two types of “aging modes” that influence the form of the OCV curve: loss of lithium inventory that can be modeled by a shift of the relative position of the electrode OCV curves and loss of active material that can be modeled by linear scaling of the electrode OCV curves.
A central assumption in this theoretical framework is that the shape of the single electrode OCV curves is invariant upon aging. We investigated this assumption experimentally and we share our results concerning possible changes in the shape of electrode OCV curves during full cell cycling in this presentation.
Our experimental approach is to cycle commercial cylindrical cells containing nickel-rich NMC-811 vs. silicon-graphite until a certain level of degradation is reached. We then open up cells at different aging levels inside an argon filled glove box and harvest samples from both electrodes. Afterwards, we built coin cells with the harvested electrode samples as cathodes and lithium foil as anodes. Subsequently, we measure the quasi-stationary OCV curves of these coin cells and compare them with the OCV curves obtained from fresh electrodes.
We show that there is a change in the shape of the OCV of silicon-graphite, which is probably due to a decrease in the relative capacity contribution of the silicon. Contrariwise, the there is no aging-related trend in the shape of the NMC-811 OCV curve.
These results are relevant for science and application because they suggest that aging-related changes in the shape of the half-cell OCV of silicon-graphite should be considered by methods for non-destructive aging diagnostics and by electrical battery models that are used for state estimation of battery packs, in order to have accurate state estimation also for aged cells.