IntroductionIt is obvious that cell temperature plays an important role on the performance of a polymer electrolyte fuel cell (PEFC) and durability of the membrane electrode assembly (MEA). This issue is more common in Nafion membrane due to its limitation at higher temperature (above 90 oC at a low humidity level) [1]. One of the outcomes of high-temperature operation is membrane degradation that can increase fuel crossover. Great amount of studies has focused on effects of temperature incorporating other parameters such as relative humidity or total pressure gradient between electrodes on either proton exchange membrane (PEM) or MEA [2,3]. The difference between a PEM and a membrane with catalyst layers could be related to the steps during MEA production techniques [4]. However, the distinction of a PEM and an MEA in terms of degradation has not been clear yet. In the current study, we studied hydrogen crossover through a Nafion membrane and a Nafion membrane with catalyst layer at various cell temperatures and varied hydrogen partial pressure gradient.ExperimentalHydrogen crossover was measured using 4 MEAs (containing NR-211, NR-212, N-115 and N-117) at Pt loading of 0.36 mg/cm2 (anode) and 0.38 mg/cm2 (cathode) with the same catalyst layer thickness (10 μm) and the 2 Nafion membranes (NR-211 and NR-212). The MEAs/PEMs were evaluated in a modified Japan Automobile Research Institute (JARI) standard cell with parallel flow channels (active surface area of 4 cm2). Gas crossover was tested at cell temperature 59 oC – 94 oC and atmospheric pressure with the desired fixed inlet relative humidity (70 % – 79 %).Hydrogen and nitrogen were supplied to one side of the cell to control a hydrogen partial pressure and only nitrogen was supplied to the other side. The gas was collected from the cell outlet for gas chromatography analysis (GC-12A, Shimadzu).A pristine PEM was exposed to heat for 6 min at 120 oC and 140 oC for comparison of the MEA that experienced heat treatment during MEA fabrication. An X-ray diffractometer (Ultima IV, Rigaku) with a scanning rate at 2 o min-1, a differential scanning calorimeter (TGA/DSC 3+ STARe System, Mettler Toledo), and a thermogravimetric analyzer (TGA-50, Shimadzu) with a heating rate of 10 oC min-1 were used for characterization of membranes.Results and DiscussionHydrogen gradient between the anode and cathode drives the crossover as follows: N H (M) = k pH (M) Δ p H (1)where k pH (M) represents the permeance of hydrogen andΔ p H represents the difference in the hydrogen partial pressures at both sides. Hydrogen crossover flux N H (M) measured by the gas chromatography proves that the hydrogen permeation through the PEM and MEA is highly temperature dependent (Fig. 1). The activation energy of k pH (M) was 24.0 kJ/mol.Measurements using PEM (25.4 μm and 50.8 μm) and MEA (with PEM thickness 25.4 μm, 50.8 μm, 127 μm and 183 μm) showed, as expected, increasing the PEM thickness in MEA resulted in more resistance (1/k pH (M)) to hydrogen crossover as shown in Fig. 2. The similar behavior was observed in PEM itself. It should be noted that PEM and PEM in MEA showed different crossover resistance per unit thickness. Heat treatment of PEM did not affect the crossover considerably, although the crystal structure change was observed by XRD. The broad peak around 2θ = 17.5 ° assigned to a crystalline region [5] was reduced during heat treatment at 140 oC.Fig. 3 shows the effects of the moisture content of membranes on the hydrogen crossover. The moisture content slightly enhanced crossover through the PEM by itself as reported [6]. Increase in the moisture content reduced hydrogen crossover through the MEA. Confinement of the PEM in the MEA might cause swelling of PEM in the through-plane direction, reducing crossover flux.ConclusionsIncreasing cell temperature activated hydrogen permeation. Unlike a pristine PEM, an MEA containing the same membrane indicated ca. 2 times faster crossover.
Read full abstract