Dynamic Mechanical Properties of Polymeric Materials Aged in PEM Fuel Cell Conditions (Experimental and Applied Mechanics)

ABSTRACT

Gaskets/seals in PEM fuel cells are exposed to acidic, humid air, mechanical compressive pressure and cyclic temperature environment. Chemical degradation of three elastomeric gasket materials in a simulated and an aggressive accelerated fuel cell solution at 80oC up to 63 weeks was investigated in this work using dynamic mechanical analysis (DMA) which assesses the change of dynamic mechanical properties of the three material samples as they aged. The three materials tested are copolymeric resin (CR), liquid silicone rubber (LSR), and fluorosilicone rubber (FSR).

Keywords: DMA, glass transition temperature, elastomeric gasket material, PEM fuel cell

INTRODUCTION

Gaskets in PEM fuel cells are typically made of elastomeric materials and are exposed to acidic liquid solution, humid air and hydrogen, as well as subjected to mechanical stress. The long-term stability and durability of the gaskets materials are therefore critical to both sealing and the electrochemical performance of the fuel cells. In this study, three polymeric materials that are or have potential to be used as gasket or seals in PEM fuel cells were first aged in a simulated or an accelerated fuel cell environment solutions at 80oC which is close to the PEM fuel cell operating temperature. The dynamic mechanical characteristics of the virgin and aged materials were then studied using a DMA (TA Instruments, RSA-III).

EXPERIMENT DESCRIPTIONS

Materials and Simulated Fuel Cell Environment Solutions

Three elastomeric sealing materials, namely, Copolymeric Resin (CR), Fluorosilicone Rubber (FSR), and Liquid Silicone Rubber (LSR), are, among others, currently been used, or are considered to use, in PEM fuel cells as the gasket/seal.

Two solutions were used to age the three materials. The first is an accelerated durability test (ADT) solution. The final composition is 1M H2SO4, 10ppm HF and reagent grade water having 18Mega Ohm resistances. The pH value is less than one. The second solution is termed a "Regular" solution. Chemical composition of the solution is 12ppm H2SO4 and 1.8ppm HF with reagent grade water having 18Mega Ohm resistances.

Aging and Characterization Methods

Rectangular-shaped specimens were prepared and exposed to either the ADT or the Regular solution. The dimensions of the specimens are 70mm in length, 20mm in width, and 2.0mm in thickness, except Copolymeric Resin (CR) which has a thickness of 1.5mm. The samples were submerged in the ADT and Regular solutions in different bottles placed in an oven at a temperature of 80oC. The aged samples were taken out of the test bottles at selected times for observation and tests. The process completed in 63weeks.

RESULTS AND DISCUSSIONS

Material Properties from DMA

Copolymeric Resin (CR)

The CR samples at different aging weeks were used to reveal the dynamic mechanical properties. Fig. 1(A) presented the storage modulus (E’) and the loss modulus (E”). Fig. 1(B) is the Tan 5 versus temperature for the samples aged in the Regular solution. Fig. 2 is for CR in the ADT solution.

Figs. 1 and 2 show that the glass transition temperature (Tg) is about -48oC and no change after aging. Tg is identified in Fig. 1 as the temperature where sharp decrease in E’ or the peak in Fig 2 occurs.

In the Regular solution (see Fig. 1(A)), the E’ and E” increased slightly as the samples aged; but in the ADT solution (see Fig. 2(A)) they tend to decrease as it ages.

Both E’ and E” are stable and maintain a nearly constant value for temperatures higher than Tg for both virgin and aged samples. Note that in our previous work [1], the surface chemistry of CR changed notably after exposure to the environment (particularly in the ADT solution), see the FTIR results in [1]. CR material in Regular solution is more chemically stable than in ADT solution under current test conditions. The current data shows however that the bulk, dynamic mechanical properties of the material is affected very little by aging in the environments.

(A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged CR samples exposed to the Regular solution

Fig. 1. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged CR samples exposed to the Regular solution

(A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged CR samples exposed to the ADT solution. The optical pictures show the surface conditions of the virgin and 63-week aged specimens.

Fig. 2. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged CR samples exposed to the ADT solution. The optical pictures show the surface conditions of the virgin and 63-week aged specimens.

Fluorosilicone Rubber (FSR)

Test data for FSR samples are shown in Figs. 3 and 4 in the Regular and the ADT solution, respectively. Figs. 3(A) and 4(A) present the storage modulus (E’) and the loss modulus (E”). Figs. 3(B) and 4(B) are the Tan 5 versus temperature.

Figs. 3 and 4 show that the glass transition temperature (Tg) is about -50oC and has no change after aging. Above Tg, both E’ and E" decrease fast between Tg and room temperature and slowly after room temperature. The aging appears to increase the modulus slightly.

Note that in our previous study [1], no FTIR spectrum change was observed for FSR in either Regular or ADT solution. FSR appeared to be the most stable materials among the three tested in FTIR tests.

 (A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged FSR samples exposed to the Regular solution

Fig. 3. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged FSR samples exposed to the Regular solution

 (A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged FSR samples exposed to the ADT solution

Fig. 4. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged FSR samples exposed to the ADT solution

Liquid Silicone Rubber (LSR)

Test data for LSR samples are shown in Figs. 5 and 6 in the Regular and the ADT solution, respectively. Figs. 5(A) and 6(A) present the storage modulus (E’) and the loss modulus (E”). Figs. 5(B) and 6(B) are the Tan 5 versus temperature.

Figs. 5 and 6 show that the glass transition temperature (Tg) is about -42oC and has no change after aging. The Tg is consistent with our previous data reported in [1]. Above Tg, E’ is nearly constant, but E" gradually decreases with temperature. The aging appears to increase the modulus slightly in both solutions.

Note that surface appearance of the LSR in ADT changed from black to white after 63weeks (see the discussion in [1]). It was concluded in [1] that there are significant surface chemical changes for LSR samples exposed to ADT solution at 80oC over time. The chemical degradation is likely due to de-crosslinking and chain scissoring in the rubber backbone in the environment. However, the bulk, dynamic mechanical properties of the LSR material had very little change aged in both solutions according to the current test data.

 (A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged LSR samples exposed to the Regular solution

Fig. 5. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged LSR samples exposed to the Regular solution

(A) Storage modulus E' and loss modulus E'', and (B) tangent delta versus temperature of virgin and aged LSR samples exposed to the ADT solution. The optical pictures show the surface conditions of the virgin and 63-week aged specimens

Fig. 6. (A) Storage modulus E’ and loss modulus E”, and (B) tangent delta versus temperature of virgin and aged LSR samples exposed to the ADT solution. The optical pictures show the surface conditions of the virgin and 63-week aged specimens

CONCLUSIONS

From the DMA data and discussions above, it appears that CR, LSR and EPDM are good candidates for PEM fuel cell applications, in terms of both Tg and the storage modulus E’. Among the three, EPDM seems to be the best, as

(a) In ADT solution, CR showed cracks and both CR and LSR lost weights to about 45% after ageing [1].

(b) CR and LSR had significant amount of silicone leachants in ADT solution [1].

(c) CR and LSR had chemical degradation in ADT solution as shown by our FTIR analysis [1].

Next post:

Previous post: