Accelerated Stress Tests on Fuel Cell Cathode Catalysts: A Material Balance Approach Combining Modeling and Experiment

Wednesday, October 14, 2015: 14:40
211-A (Phoenix Convention Center)
C. A. Rice (Tennessee Technological University), P. Urchaga (Tennessee Technological University), J. Hu (Automotive Fuel Cell Cooperation Corporation), T. Kadyk (Simon Fraser University, Department of Chemistry), and M. Eikerling (Simon Fraser University, Department of Chemistry)
Proton Exchange Fuel Cells (PEFCs) represent an environmentally benign power sources for automotive transportation. The current major hurdles for the mass commercialization of automotive PEFC technology are cost and durability of the cathode catalyst layer. Catalyst degradation occurs due to voltage transients experienced by the cathode catalyst during normal drive cycles and start-up/shut-downs (SUDS). Urban drive cycles involve rapid stop-go transients (idle-to-peak power) that take the cathode catalyst potential from >0.9 V (idle) to around 0.6 V (peak power). Frequent potential changes with high peak potentials (> 1.2 V) drastically accelerate the loss of active catalyst particles and the decay of the of  electrochemically active surface area (ECSA). The main degradation mechanisms affecting the carbon-supported Pt-based catalyst are dissolution/redeposition, coagulation and detachment.

Accelerated stress tests (ASTs), performed to decouple and quantify contributions of different catalyst degradation mechanisms are typically performed ex situ to eliminate degradation effects caused by the other materials within the cell, such as the proton conducting membrane and the gas diffusion media. The AST results presented herein evaluated Pt/C degradation under operationally relevant conditions temperature (22°C and 70°C), upper potential limit (UPL, 0.9 V and 1.2 V) and potential wave form (triangle vs. square waves). A comprehensive material balance analysis was performed to track changes in the state of Pt during the AST. This analysis included the determination of ECSA, particle size distribution and amount of dissolved Pt in the electrolyte solution. These results were analyzed with a dynamic model that couples the three catalyst degradation mechanisms to quantify their relative contributions. The model relates the kinetic rates of the degradation processes to the evolution of the particle size distribution.

The loss of ECSA during 25,000 AST cycles at 70°C by periodically performing cyclic voltammetry (0.02 V « 0.6 V at 10 mV sec-1) and integrating the hydrogen desorption charge (210 µC cm2). Generally, ECSA loss was found to be accelerated by the SW profile and by the high UPL of 1.2V.

The Pt particle size distribution was measured by Transmission Electron Microscopy (TEM). The beginning of life (BOL) particle median size was 2 nm. Figure 1 shows the change in the particle size distribution after 25,000 AST cycles at 70°C. The tests with higher UPL had the most significant increase in mean particle size and width of the distribution.

The combined modeling and material balance preliminary results demonstrate a significant role of the effective surface tension of the catalyst dominating the kinetics of the dissolution/Redeposition mechanism at 1.2 V, suggesting that Pt dissolution is strongly coupled to Pt oxide formation and reduction. The analysis fit the enhanced degradation due to the square wave acceleration of dissolution/Redeposition compared to the triangular wave.

Figure 1. Histograms from TEM particle size analysis for 25k AST cycles at 70°C. Upper potential limit of (A) 0.9V and (B) 1.2V. Lines are simulated particle size distributions from dynamic model analysis.