Simulation of HT-PEMFC AC Impedance Spectra: Relaxation Impedance and Identification of Oxygen Reduction Reaction Mechanism

High Temperature Polymer Electrolyte Membrane Fuel Cells, (HT-PEMFCs), are one of the most promising alternatives of clean power production by converting chemical energy to electrical energy. Even though the operating principles of an HT-PEMFC are understood, the overall system’s behavior is determined by a number of strongly coupled processes, each proceeding at a different rate.In a HT-PEMFC the elusive multistep Oxygen Reduction Reaction, (ORR), taking place at the cathodic electrode takes a major fraction of the open circuit potential. Thus, in order to optimize fuel cell performance by minimizing ORR power losses, an in depth understanding of the various processes and their interactions is necessary. One of the most powerful electrochemical characterization methods is the Electrochemical Impedance Spectroscopy technique, (EIS). It is a very powerful and sensitive in situ technique that has the ability to separate potential losses corresponding to different processes which proceed at different rates, (have different timescales), [1]. Our research revealed that ORR impedance spectrum consists of two arcs, a high frequency arc attributed to the charge transfer reactions across the interface and a low frequency arc due to the hysteresis phenomena between the oxygenated intermediate species coverage and the current dynamic response, (ORR relaxation impedance), [2]. Thus, by simulating the AC impedance spectra through the development of an appropriate physical model, valuable information on the electrochemical and chemical process at the electrode can be drawn. In order to demystify the multistep ORR kinetics a 1-D dynamic mathematical model has been developed in order to simulate AC impedance spectra at the low current linear activation region while at the same time a Monte Carlo type algorithm was used to estimate the values of the respective kinetic constants by simultaneously fitting polarization-dc bias and the corresponding galvanostatic AC Spectra. The model incorporates a three-step proposed mechanism for the ORR as well as a detailed macroscopic description of mass and charge transfer across the various parts of the Membrane Electrode Assembly, (MEA). The low current activation regime of operation was chosen because there the ORR overpotential dominates the AC spectra, the effect of the finite electrolyte conductivity and mass transport limitations are minimized and the adsorbed intermediate relaxation impedance is prominent while it diminishes at higher current densities as can be seen in Figure 1. As a result, the cathodic overpotential is constant throughout the cathodic electrode which may be well considered to operate under ideal kinetic conditions allowing thus an analytic expression for the HT-PEMFC impedance to be developed and used in the identification of the ORR kinetics. As far as validation of our approach is concerned, synthetic AC Impedance Spectra with known kinetic constants where produced and utilized by the Monte Carlo fitting algorithm at the low current operation regime. Figure 2 shows the histograms of the fitting algorithm for the OH intermediate reduction kinetic constant, by employing different number of bias points (I-V) and their respective AC spectra the algorithm successfully converges to the set value for k3+ , (k3+ =2*10-6  mol/(m2s)). By decoupling the various processes of the simulated AC impedance spectra, information was successfully extracted on the various kinetic and equilibrium constants related with the interaction of the reacting gases in the catalytic layer and the electrochemical interface. [1] O. Antoine, Y. Bultel, and R. Durand, Journal of Electroanalytical Chemistry, vol. 499, no. 1, pp. 85–94, Feb.2001. [2] X. Yuan, H. Wang, J. Colinsun and J. Zhang. 2007. International Journal of Hydrogen Energy, Volume:32, 4365-4380. This work has been conducted and supported within the framework of DeMStack, (DEMSTACK 325368, FCH-JU-2012-1). Figure 1