Impedance in lithium ion cells grows over the lifetime of the cell and can be a large contributing factor to cell failure. This study attempts to understand impedance growth by looking at the factors that contribute to cell impedance. Separating these factors and understanding the origins of impedance in cells provides insight into how and why cells fail and could lead to improvements in cell cycling and lifetime. The positive and negative electrode solid electrolyte interphase (SEI) layer, contact resistances, degradation of electrical conductivity in electrodes, and loss of ionic conductivity in electrolyte are all contributing factors to overall cell impedance. However, separating these factors can be difficult in impedance spectroscopy measurements.

In this study, Li[Ni_0.5Mn_0.3Co_0.2]O₂/artificial graphite pouch cells were used with 1.2M LiPF₆ in EC:DMC 3:7 (w:w) electrolyte containing 2 wt% vinylene carbonate (VC), 1 wt% lithium difluorophosphate (LiPF₂O₂ – called LFO here), or no additive as a control electrolyte. Cells were formed to 4.2 V or 4.4 V and were either left at top of charge or discharged to 3.8 V before disassembly. See Table 1 for a full list of electrolytes and voltages of the cells used in this study. Following pouch cell formation, symmetric cells were made from two positive electrodes or two negative electrodes harvested from the full lithium ion cell to study each electrode separately. Full coin cells were also made with one positive and one negative electrode. Electrochemical impedance spectroscopy (EIS) was performed at a range of temperatures to facilitate the separation of impedance factors. Measurements were taken at -10°C, 0°C, 10°C, 20°C, 30°C, and 40°C. Spectra were measured at 10°C at the beginning, middle, and end of the experiments to ensure repeatability. Figure 1 is an example of measurements taken from one disassembled pouch cell. Note that only -10°C, 10°C, 30°C, and 40°C are shown in this figure for simplicity. The left column in Figure 1 shows the Nyquist plots for negative symmetric cells at the various temperatures. The middle and right columns show the positive symmetric cells and the full cells respectively. The middle, positive symmetric cell column contains two spectra, one for a cell made with aluminum hardware and the other with steel hardware, labelled (A) and (S) respectively. Data from one symmetric cell is shown per type of symmetric cell, however several cells of each type were made from each pouch cell to ensure repeatability of the data.

The low frequency (right side) semicircular feature in these Nyquist plots may be attributed to charge transfer resistance between the electrolyte and the electrodes. The charge transfer resistance used here lumps together ion desolvation, ion transfer through the SEI and combination with an electron at the inner SEI surface. The high frequency (left side) semicircular feature may be attributed to contact resistance. The steel and aluminum hardware positive symmetric cell data have very similarly sized low frequency features – representing charge transfer resistance, which should not change at all with hardware. However, the high frequency feature varies substantially between the two symmetric cells, indicating that this feature is in fact due to contact resistance. In general, for all spectra, charge transfer resistance increases dramatically in magnitude with decreasing temperature, while the contact resistance remains almost constant. Therefore, Figure 1 strongly suggests that researchers interested in studying charge transfer resistance with minimal confusion should make their measurements at low temperature. The full cells have three features, which represent a combination of charge transfer and contact resistances from the negative and positive electrodes in combination. This data shows that the majority of cell impedance growth as temperature decreases originates from the positive electrode SEI.

Equivalent electric circuit models have been used to model the impedance behavior of the symmetric cells. Circuit models include resistance factors representing charge transfer resistance, contact resistance, and solution resistance, as well as imperfect capacitances (constant phase elements) representing electrochemical double layers. Impedance spectra have been fitted using these simple circuit models. Using this technique, charge transfer resistances, contact resistances, solution resistances, and double layer capacitance values can be obtained for positive and negative electrodes/SEI layers separately as a function of temperature. Activation energies for the charge transfer between SEI and electrolyte have been extracted from charge transfer resistance measurements as they follow an Arrhenius temperature dependence. Capacitance values associated with double layers at the SEI/electrolyte interface were also be obtained through these measurements and these will be discussed.