While the surface temperature (T_surf) of a lithium-ion cell is measurable by an external temperature sensor, the internal temperature (T_int) remains unknown. The difference between the surface temperature and the internal temperature increases with charging/discharging rate and with cell volume. A thermal model describing the correlation between load and cell temperature is necessitated, and requires reliable parameter settings.

The electrothermal impedance spectroscopy (ETIS) is an in-situ method, which determines the thermal impedance and parameterizes a thermal model. ETIS is comparable with electrochemical impedance spectroscopy (EIS), but in ETIS the excitation signal is a sinusoidal heat flow q(t), which leads to a sinusoidal temperature response (T_int(t) and T_surf(t)). From the input signal q(t) and output signal T(t) the thermal impedance can be calculated. Using a simple thermal model (Figure 1 b)) and the measured thermal impedance (Figure 1 a)), the temperature change and the difference between the internal temperature and the surface temperature can be calculated as a function of the load current.

Consequently, the ETIS-Method was applied to a high energy (KOKAM 560 mAh) and to a high power (KOKAM 350 mAh) lithium-ion pouch cell. Both are made of a NCA-LCO blend cathode and a graphite anode, but differ in electrode thickness.

The study shows that the simple model from Figure 1 b) sufficiently describes the thermal behavior of both commercial cell types. The parametrized model describes fairly well the thermal gradient from T_surf to T_int and is capable to quantify the thermal gradients in lithium-ion cells. The thermal resistance R_th of the high energy pouch cell is more than doubled (2.5 K/W) comparing to the high power cell (1 K/W), thus reflecting the influence of electrode thickness to the thermal behavior of a lithium-ion cell.