Electrochemical impedance spectroscopy (EIS) is a reliable and non-invasive measurement technique for the characterization of electrochemical systems, ranging from material and electrode level up to cells and systems. The impedance spectrum resulting from the measurement contains valuable information about properties and states of the electrochemical system, including the ongoing electrochemical processes [2, 3]. In practice, direct access to this information is difficult, so the spectrum must be processed using physical or equivalent circuit models. However, the discovery and parameterization of such models is challenging because a detailed a-priori understanding of the system is required.
The Distribution of Relaxation Times (DRT) analysis improves the mentioned disadvantageous by transforming frequency and/or time domain data into a more distinguishable form in the τ-domain. This makes it possible to reliably detect, deconvolute, and identify processes. This involves preselecting a kernel function, in this case the elementary transfer function of a resistive-capacitive element, and the assumption of serially connected processes, emphasizing the predominantly model free approach of the DRT method. To date, there are several methods to determine the distribution function [6]. Here, we want to focus on the gDRT method. The method is a discretization and Tikhonov regularization-based approach, where the polarization contributions are calculated for predefined time-constants by solving an ill-conditioned linear system of equations.
Electrochemical processes within the battery’s anode and cathode often overlap in the spectrum, making it challenging to distinguish between them. Assigning polarization contributions to anodic and cathodic reactions successfully typically requires assembling experimental cells with extracted electrodes from a reference cell. When compared to the distribution function of the reference cell, additional peaks, shifts in time constants, and polarization amplification or damping can be observed [5, 7]. These observations complicate the transfer of electrochemical properties across different cell formats.
To better understand the implications, impedance spectra of experimental cells in the EL-Cell case in full-cell configuration, 2032 format coin cells (in full-cell and half-cell configurations with lithium metal as the counter electrode), and 80 mm x 60 mm single-layer pouch cells are analyzed using both DRT methods. These analyses make it possible to investigate the effect of cell geometry and format. To minimize ambiguity in the study, the number of variable parameters is minimized by using uncycled active material for both the anode and cathode, identical separators, and the same electrolyte in every cell. For each cell and configuration, impedance spectra at different SoCs and temperatures were measured, analyzed, and compared in terms of the effects of cell geometry and format.