PEMFC Model Parameterization By Means of Differential Cell Polarization and Electrochemical Impedance Spectroscopy

Abstract

Understanding the contributions of the loss mechanisms in PEM fuel cells remains a key issue for the optimization of the membrane electrode assembly (MEA) with respect to performance and ageing.A full parameterization of all the physics via adapted experiments is considered as too challenging and/or too expensive, thus the parameters of performance models are often only estimated or fitted to insufficient experimental data from single cell steady-state measurements. In this study, we aim to remedy this issue by combining polarization and impedance measurements over a wide range of operating conditions on a differential cell setup with 12 cm2 and a straight channels flowfield. For this purpose, a fuel cell test station is entirely automated and allows therefore to design a full factorial experiment with relative humidity, temperature, oxygen partial pressure and current density as parameters, leading to more than 1000 operating points. Self-developed Matlab-tools enable the analysis of the impedance spectra by fitting them to equivalent circuit models (ECM) and thus to understand the impact of the operating conditions on the cell parameters, cf. Fig. 1. This step is supported by the evaluation of the distribution of relaxation times (DRT) for well-chosen conditions.Since it is commonly known that the main contribution to the overall performance loss at low and medium current densities originates from the sluggish oxygen reduction reaction (ORR) on the platinum catalyst of the cathode and that there is still no consensus in literature concerning its proper modeling and parameterization for a full cell, this study focuses on this very mechanism. In order to get a proper parameterization, the measurements are corrected by the anode contribution, the impact of hydrogen crossover through the membrane and by the ohmic resistance as well as the protonic resistance. Figure 1