Li-ion batteries degrade over time. Traditionally, capacity and resistance of the cell have been used as the state of health (SOH) indicators. However, these parameters cannot provide detail information about the degradation mechanisms. Thus, recently, there have been many efforts on developing diagnostics algorithm that can identify the degradation mechanisms. There exist many degradation mechanisms for Li-ion batteries such as SEI layer growth or lithium plating consuming lithium, structure disordering and metal ion dissolution making active material unavailable for insertion and extraction of lithium. Identifying these mechanisms is challenging due to complex inter-correlations between them. Instead, it is possible to categorize these mechanisms into two degradation modes: loss of lithium inventory (LLI) and loss of active material (LAM) at each electrode. On the electrode level, capacity and utilization window of the individual electrode are proposed as electrode-specific SOH parameters that can be related to the degradation modes. The differential voltage analysis (DVA) is one of the common approaches for estimating these electrode parameters. In this method, the phase transition of the electrode material is used as a fingerprint of each electrode in the cell’s differential voltage (dV/dQ) curve using a low C-rate constant current pseudo-OCV data [1,2]. Another interesting type of battery response is the mechanical response measured as a cell expansion. Batteries expand during charge and contract while discharge in repeatable patterns. Hence, similar to the voltage analysis, the expansion of a cell can be used for identifying the electrode parameters [3].

In this study, electrode-specific parameters (electrode capacity and utilization window) were estimated for a 5 Ah graphite/NMC pouch cell (University of Michigan Battery Lab) during accelerated aging cycle testing at the elevated temperature. A fixture was designed in order to measure the mechanical response of the cell as shown in Fig. 1(e) such that the top and bottom plates were fixed in place while the middle plate was free moving. The expansion was measured using a displacement sensor (Keyence) mounted on the top plate, and a battery cycler (Biologic BCS-815) was used for measuring the voltage. The cell was placed inside a climate chamber at 45˚C. The aging cycle consists of 2C constant current charge and discharge cycles. Diagnostic tests were conducted at intervals corresponding to an expected 5% loss in capacity for a cell at room temperature. The cell was cycled at C/20 to get the pseudo-OCV. Furthermore, in order to measure the potential of graphite and NMC electrodes, coin cells were built. The Li/NMC coin cell was cycled at C/50 between 2.8 and 4.35 V. The Li/graphite coin cell was cycled at C/50 between 0.005 and 1.0 V. The lattice expansion data was taken from literature for graphite and NMC.

The terminal voltage, differential voltage, and expansion data of the cell is shown in Fig.1(a), (b), and (d). A model is developed based on [3] to estimate the electrode-specific SOH parameters as the cell ages and to quantify the degradation modes (i.e., LLI and LAM at each electrode). The cell OCV and expansion models share the same electrode SOH parameters and the estimation is done by finding the best fit for the voltage, differential voltage, and expansion curves. The results show a very good match between the data and model for all three different data curves. The estimated parameters then are provided to the equations describing the degradation modes for quantification (see Fig.1(c)). It is found that LAM_NE and LLI are the main degradation modes for this chemistry of NMC/graphite cell under the accelerated aging cycling at the elevated temperature accounting for the cell capacity fade (SOH in Fig. 1(c)).

Lastly, it was found that cell expansion measurement gives much higher confidence for estimating electrode SOH parameters [3]. Since the differential voltage is a derivative of the measured voltage, which is sensitive to the measurement noise and can lead to incorrect estimation. Furthermore, the addition of expansion makes this electrode SOH parameter estimation feasible when using a partial range of OCV data. This is important for automotive applications where battery packs rarely discharge fully.

[1] Suhak Lee et al., "Comparison of Individual-Electrode State of Health Estimation Methods for Lithium Ion Battery," ASME 2018 Dynamic Systems and Control Conference, 2018.

[2] Hannah M. Dahn et al., "User-friendly differential voltage analysis freeware for the analysis of degradation mechanisms in Li-ion batteries," Journal of The Electrochemical Society 159, 2012.

[3] Peyman Mohtat et al., “Towards better estimability of electrode-specific state of health: Decoding the cell expansion,” Journal of Power Sources 427, 2019.