0
Research Papers

Viscoelastic Stress Analysis of Constrained Proton Exchange Membranes Under Humidity Cycling

[+] Author and Article Information
Yeh-Hung Lai1

Fuel Cell Research Lab, General Motors Corporation, Honeoye Falls, NY 14472-0603yeh-hung.lai@gm.com

Giner Electrochemical Systems,  LLC, Newton, MA 02466

Craig S. Gittleman

Fuel Cell Research Lab, General Motors Corporation, Honeoye Falls, NY 14472-0603

David A. Dillard

Engineering Science and Mechanics Department, Virginia Tech, Blacksburg, VA 24061

1

Corresponding author.

J. Fuel Cell Sci. Technol 6(2), 021002 (Feb 20, 2009) (13 pages) doi:10.1115/1.2971045 History: Received February 09, 2007; Revised June 04, 2008; Published February 20, 2009

Abstract

Many premature failures in proton exchange membrane (PEM) fuel cells are attributed to crossover of the reactant gas from microcracks in the membranes. The formation of these microcracks is believed to result from chemical and/or mechanical degradation of the constrained membrane during fuel cell operation. By characterizing the through-membrane leakage, we report failures resulting from crack formation in several PEMs mounted in $50cm2$ fuel cell fixtures and mechanically stressed as the environment was cycled between wet and dry conditions in the absence of chemical potential. The humidity cycling tests also show that the failure from crossover leaks is delayed if membranes are subjected to smaller humidity swings. To understand the mechanical response of PEMs constrained by bipolar plates and subjected to changing humidity levels, we use Nafion® NR-111 as a model membrane and conduct numerical stress analyses to simulate the humidity cycling test. We also report the measurement of material properties required for the stress analysis—water content, coefficient of hygral expansion, and creep compliance. From the creep test results, we have found that the principle of time-temperature-humidity superposition can be applied to Nafion® NR-111 to construct a creep compliance master curve by shifting individual compliance curves with respect to temperature and water content. The stress prediction obtained using the commercial finite element program ABAQUS ® agrees well with the stress measurement of Nafion® NR-111 from both tensile and relaxation tests for strains up to 8%. The stress analysis used to model the humidity cycling test shows that the membrane can develop significant residual tensile stress after humidity cycling. The result shows that the larger the humidity swing and/or the faster the hydration/dehydration rate, the higher the residual tensile stress. This result is confirmed experimentally as PEM failure is significantly delayed by decreasing the magnitude of the relative humidity cycle. Based on the current study, we also discuss potential improvements for material characterization, material state diagnostics, and a stress model for PEMs.

<>

Figures

Figure 12

The membrane stresses in Cases 11 to 15 of Table 2 for a constrained membrane subjected to humidity cycling

Figure 13

The maximum residual stress as a function of λmin for various λmax and dλ∕dt for all cases in Table 2

Figure 1

Crossover leak rate as a function of number of humidity cycles during humidity cycling of Nafion® NR-111, Nafion® N111-IP, Gore™ Primea PFSA membranes, and a noncommercial hydrocarbon membrane. The N111-IP tests were stopped before any crossover was measured. The horizontal line at 10SCCM indicates the test failure criteria.

Figure 4

Creep compliance of Nafion® NR-111 at various temperatures and RHs

Figure 5

Creep compliance master curve of Nafion® NR-111 with the reference conditions of 60°C and 75% RH

Figure 6

Temperature shift factors of the Nafion® NR-111 from creep compliance tests. Solid line represents WLF equation fit using Eq. 7 with a reference temperature of 60°C.

Figure 9

Water content, hygral strain, and membrane stress history in Case 2 of the parametric stress analysis in Table 2

Figure 7

Humidity shift factors of Nafion® NR-111. Solid line represents the curve fit using Eq. 8 with a reference water content of λ=6.

Figure 8

The comparison of experimental results and model simulation of tensile and relaxation tests. (a) illustrates the comparison of stress/strain curves of both tests; and (b) shows the comparison of stress history from the stress relaxation test.

Figure 10

The membrane stresses in Cases 1 to 5 of Table 2 for a constrained membrane subjected to humidity cycling

Figure 11

The membrane stresses in Cases 6 to 10 of Table 2 for a constrained membrane subjected to humidity cycling

Figure 2

Crossover leak rate as a function of number of humidity cycles of Gore™ Primea® MEAs during inert RH cycling tests. Cycle: 2.5min at 150% RH/3.5min at 0%, 50%, and 80% RH.

Figure 3

Water isotherms for Nafion® NR-111 obtained at 30°C, 60°C, and 80°C. Solid lines represent water contents calculated from Eq. 6.

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections