Analysis Method of Oxygen Reduction Reaction Rate in PEFC Using Pt-Sputtered Catalyst

Abstract

IntroductionTo establish the generic model of the cathode catalyst layer (CCL),[1] the intrinsic oxygen reduction reaction (ORR) rate which does not include mass transfer resistance through the CCL should be formulated. Neglecting the backward reaction, the ORR rate per geometric area, r gc, is proportional to oxygen partial pressure, p O [Pa], as follows: r gc = k gc p O [mol/(m2·s)][1] (1) k gc = k gc0exp(–E/RT–E c/b c) = k gc0exp{–(E+4aFE c)/RT} (2)where k gc is the reaction rate constant per geometric area [mol/(Pa·m2·s)], E c is the cathode electromotive force [V], E is the activation energy [J/mol], T is temperature [K], b c is the Tafel slope [V] which is proportional to temperature, and 4a is the transfer coefficient.The objective of this study is to develop a measurement method of temperature, relative humidity (RH) and E c dependencies of the intrinsic ORR rate constant using platinum-sputtered catalyst.ExperimentalA Japan Automobile Research Institute (JARI) standard cell whose active area was 50 mm × 50 mm or modified to 20 mm × 20 mm was used in experiments. The membrane-anode assembly (MAA, Eiwa Corp.) was composed of a proton exchange membrane (PEM, Nafion® NR-212), an 8.3‒9.8 μm thick (0.19–0.24 mg/cm2) anode which was made of Pt/Vulcan® XC-72 catalyst (Pt/C weight ratio: 1) with a Nafion® ionomer (ionomer/carbon weight ratio: 0.74). Pt thin-layer cathode (0.054 mg/cm2) was prepared by sputtering (AOV, SPAD-4240UM, Ar atmosphere, 5 Pa, 100 ℃, for 50 minutes) on the microporous layer (MPL) on a carbon paper (SIGRACET-GDL 29 BC, 235‒238 μm thick) to eliminate mass transfer resistance through the CCL.[2] Before sputtering, MPL substrates were cleaned using argon plasma in order to eliminate the hydrophobized coatings on the MPL surface. The MAA and Pt thin-layer cathode were stacked as shown in Fig. 1. A carbon current collector had a single serpentine flow channel, the width and depth of which were 1.0 mm and the pitch of the ribs was 2.0 mm. H2 and O2 were humidified at 60–80 oC. H2 and O2 flow rates were 600 and 300 cm3/min, respectively. The cell was operated at 65–80 oC. The pressure at the cell outlet was ambient pressure (ca. 1 atm).Results and DiscussionFig. 2 shows the SEM images of the surface of MPL on a carbon paper before and after 50 min sputtering. Pt particles whose diameter is approximately 10 nm are observed. Next, Fig. 3 shows the temporal changes in the substrate weight (MPL) while ion cleaning. The slope from 25 to 135 minutes is less than a half of the slope from 0 to 25 minutes. It indicates that 30 minutes is enough to remove almost all of hydrophobized coatings on the MPL surface. In addition, Fig. 4 shows Pt loading is proportional to the sputtering time.[3] 40-minute ion cleaning and 50-minute sputtering were employed in this study.Fig. 5 shows the Tafel plot at different temperatures. The current density at a fixed E c increased with raising temperature as expected from Eq. (2). From these polarization curves, the k gc parameters, E, 4a and k gc0 were determined by using nonlinear least square method under the assumption that ORR rate is proportional to RH. As a result, E = 25 kJ/mol, 4a = 0.48 and k gc0 = 1.3 mol/(Pa m2 s) are determined. Fig. 6 shows the Arrhenius plot of recalculated results shown by dashed lines with observed values. The ORR rate dependencies on temperature, RH and E c were properly reproduced so the ORR rate parameters were appropriately determined.Conclusions Temperature, RH and E c dependencies of ORR rate constant, k gc, were measured using platinum-sputtered catalyst in order to determine ORR parameters, E, 4a and k gc0 which do not include the effect of mass transfer resistance through the CCL. From experimental results at different temperatures, ORR dependency on temperature was measured and ORR rate parameters were appropriately determined.AcknowledgementThis work was supported by the New Energy and Industrial Technology Development Organization (NEDO), Japan.References Kawase et al., AIChE J., 63(1), 249–256 (2017).Kageyama et al., 232nd ECS Meeting (National Harbor, Oct., 2017), #1387.Murase et al., 31st ISChE (Chiang Mai, Dec. 1, 2018), OC06.Hu et al., ChemElectroChem, 6(10), 2659‒2666 (2019). Figure 1