Development of Accelerated Stress Tests for Polymer Electrolyte Membrane Fuel Cells

The durability of Polymer Electrolyte Membrane Fuel Cells (PEMFCs) is one of the key barriers to their commercialization in automotive applications.¹ The development of accelerated stress tests (ASTs) is a critical step in the evaluation and improvement of the durability of PEMFCs. The availability of validated component ASTs can lead to the accelerated development of materials with improved durability. However the ASTs need to faithfully reproduce the degradation mechanisms of interest and not introduce new degradation modes while attempting to accelerate the degradation rates. Furthermore, the acceleration factors with respect to the intended application need to be determined accurately to provide manufacturers with a valuable tool that enables informed decision making regarding cost, performance and durability. For example the DOE – Fuel Cell Technologies program has set lifetime durability targets of 5500 hours for automotive fuel cells and the U.S. Drive Fuel Cell Technical Team (FCTT) has developed AST protocols for the various components.^2,3 In this paper we will compare the degradation observed during the ASTs to the degradation observed during simulated drive cycle operation and report on the acceleration factors observed for various materials. We will also propose new ASTs that have greater acceleration factors than the reported FCTT ASTs while retaining the degradation mechanisms of interest.

 The FCTT has recommended potential cycling from 0.6 V to 1.0 V at 50 mV/sec as the electrocatalyst AST. This potential cycling is performed at 80 °C in 100% RH with H₂ (anode) and N₂ (cathode) and results in Pt particle growth and loss in electrochemical surface area (ECSA). This loss in ECSA and associated performance loss correlates well to the degradation observed in the field and during simulated drive cycle operation. Figure 1 compares the degradation observed during this AST (Old AST) to that observed during the wet portion of the FCTT: Drive cycle (current cycling from 0.02 to 1.2 A/cm² at 80 °C and saturated conditions). It is seen that this AST has a 5X acceleration factor for 3 different MEAs that were evaluated. In order to further accelerate this AST, we modified the triangle wave to a square wave and lowered the upper potential to 0.95 V. This new AST consisted of a square wave with upper and lower potentials of 0.95 V and 0.6 V with 3 seconds duration, and was based on literature reports.⁴ This New AST (Fig. 1) demonstrated a 100X acceleration factor compared to the drive cycle which is a 20 times improvement over the old AST. Moreover, this new AST retained the Pt growth mechanism while minimizing carbon corrosion.

 The FCTT has 2 recommended ASTs for membranes with one focusing on chemical degradation (OCV hold at 90 °C and 30% RH) and another focusing on mechanical degradation (RH cycling in Air). Our previous results have indicated that both these ASTs are not representative of membrane failure in the field.⁵ In this paper we will present a new AST that combines the mechanical and chemical degradation and is representative of field failure. This AST consists of cycling the RH between saturated and dry conditions in a H₂/Air atmosphere at 90 °C. The duration of the wet and dry cycles are 30s and 45s respectively. The fluoride release rate observed during this AST using DuPont XL^® membranes, illustrated in Figure 2, is comparable to the chemical AST and higher than in the mechanical AST.

References

1. R. Borup, et al., Chemical Reviews, V. 107, No. 10, 3904-3951 (2007).

2. DOE Cell Component AST and polarization curve Protocols for PEM Fuel Cells (Electrocatalysts, Supports, Membranes and MEAs), Revised December 16, 2010.

3. N. L. Garland, T.G. Benjamin, J. P. Kopasz, ECS Trans., V. 11 No. 1, 923 (2007).

4. A. Ohma, K. Whinohara, A. Liyama, T. Yoshida, and A. Daimaru, ECS Trans., V. 41 No. 1 , 775 (2011).

5. R. Mukundan et al., ECS Trans., V. 50 No. 2, 1003 (2013).

Acknowledgements

The authors wish to acknowledge the financial support of the Fuel Cell Technologies Program and the Technology Development Manager: Nancy Garland. The authors also wish to acknowledge Ion Power, Inc. for supplying the MEAs and SGL Carbon for the GDLs used in this study.