One of the key figures for the success of proton exchange membrane fuel cells (PEMFCs) in automotive applications is lifetime. Damage of the cathode carbon support, induced by hydrogen/air fronts moving through the anode during start-up/shut-down (SUSD), is one of the lifetime limiting factors. In this study, we examine the impact of varying the temperature at which SUSD events take place, both experimentally and by a kinetic model. For MEAs with conventional carbon supports, the model prediction of carbon oxidation reaction (COR) currents as a function of temperature matches well with the temperature dependence of experimentally determined SUSD degradation rates (predicting ≈8-fold lower COR currents compared to ≈10-fold lower measured degradation rates at 5 • C compared to 80 • C). This, however, is not the case for MEAs with graphitized carbon supports, where a factor of ≈39 lower COR currents are predicted when decreasing SUSD temperature from 80 to 5 • C, in contrast to the measured decrease by a factor of ≈10. As we will show, this is explained by a change of the governing degradation mechanism from predominantly carbon corrosion induced losses at higher temperature to predominantly voltage cycling induced platinum surface area losses near/below room temperature. several practical aspects of automotive PEMFC operation remain challenging to date.5 One such phenomenon is the degradation caused by start-up and/or shut-down (SUSD) of the PEMFC, where a H 2 /air anode gas front moves through the anode flow-field, which was first discussed in the scientific literature by Reiser et al. 6 in 2005 (note that it was described in the patent literature as early as 2002 7 ). Typically, during an uncontrolled shut-down, air will leak slowly into the H 2 compartment through leaks in the back-pressure valve, through the stack sealing or by crossover through the membrane, which was found to lead to substantial damage of the cathode catalyst carbon support in the MEA (membrane electrode assembly). 6 One of the early methods to mitigate this damage was the use of a controlled shut-down, in which a high-flow air purge of the anode compartment was applied in order to minimize the H 2 /air anode front residence time, which is proportional to the induced damage. 8The reactions occurring during the passing of the H 2 /air anode front through the anode flow-field are listed in Figure 1a (slightly modified from what was shown in Ref. 9 and 10), illustrating the partially H 2 -filled (lower left segment, colored in red) and partially air-filled (upper left segment, colored in blue) anode, opposite of the air-filled cathode flow-field (right blue segments) and separated by the proton conducting membrane (orange). While electrons can be well conducted in-plane, mainly via the diffusion media (DM) and flow fields (FF), the in-plane proton conduction resistance through the membrane is very high, so that significant proton conduction in-plane can be supported only over very short distances across the H 2 /air anode front, namely over a ...

The oxygen reduction reaction (ORR) activity of ZrO 2 based, carbon-supported nanoparticles is not conclusively reported in literature. This study examines the dependence of the ORR activity on the used precursors as well as on the heat-treatment atmosphere and temperature. We further determine the ORR activation energy and the ORR mechanism. Various precursors containing Zr and/or N were employed in the synthesis, and the ORR activity was measured by rotating (ring) disk electrode (R(R)DE) voltammetry in both acidic and alkaline electrolyte as well as by measurements in a single-cell polymer electrolyte membrane fuel cell (PEMFC) configuration. We show that even the most active ZrO 2 based ORR catalysts exhibit an activity gap of ca. two orders of magnitude compared to the DOE target of 300 A/cm 3 for PGM-free ORR catalysts, thus requiring further development. Our RRDE analysis suggests a primarily 2-electron ORR mechanism in the case of the tested catalysts in acid, which in turn provides a consistent temperature dependence between RDE and PEMFC experiments, allowing also for a mechanistic (re-) interpretation of experimental results in the literature.

For automotive applications, one of the main challenges for proton exchange membrane fuel cells (PEMFCs) is to increase the lifetime of membrane electrode assemblies (MEAs), especially during transient operating conditions such as start-up/shut-down (SUSD) cycles. During SUSD, the carbon support in the cathode layer is known to be oxidized as a consequence of hydrogen/air anode gas fronts moving through the anode. In this work, we focus on the effect of relative humidity (RH) during SUSD events. Here we show the significant impact on PEMFC performance by both experiments with 50 cm 2 single-cell PEMFCs and by a simple SUSD model using the RH-dependent kinetics for the carbon oxidation reaction (COR) rate. The kinetic parameters of the COR are determined by on-line mass spectrometry, yielding a COR reaction order with respect to RH of one. Utilizing the thus determined COR kinetics in the SUSD model predicts a ≈ 3-fold lower COR during SUSD events at 80 • C for an MEA with a conventional high surface area carbon support when the RH is decreased from 100% to 25%. This agrees perfectly well with the experimentally determined factor of ≈ 3.

In this experimental study, we determine the size dependent activity as well as accelerated voltage cycling stability of various carbon supported palladium electrocatalysts for the oxygen reduction in acidic medium. Furthermore, ex-situ transmission electron microscopy studies before and after accelerated voltage cycling provide a deeper understanding regarding particle stability during voltage cycling. Regarding oxygen reduction, a particle size effect on the specific activity is observed, with bulky Pd-black (≈4 m 2 g −1 Pd ) exhibiting ≈x6 times higher activity than Pd supported on Vulcan carbon (≈190 m 2 g −1 Pd ). Mass activities, however, exhibit a strong correlation with catalyst surface area at small electrochemically active surface area (ECSA) values, but are observed to be nearly constant between ≈50−200 m 2 g −1 Pd . As stability tests during voltage cycling reveal a benefit for smaller surface areas, i.e. bigger particles, a limited gain in stability can be achieved by increasing catalyst particle size at a negligible cost of electrocatalytic mass-based activity. Moreover, we provide a deeper insight regarding the oxygen reduction reaction mechanism and show significant hints that a sequential two-plus-two electron reduction mechanism via intermediate hydrogen peroxide is likely to occur on carbon supported Pd catalysts.

We report here the synthesis of carbon-supported ZrO 2 nanoparticles from zirconium oxyphthalocyanine (ZrOPc) and acetylacetonate [Zr(acac) 4 ]. Using thermogravimetric analysis (TGA) coupled with mass spectrometry (MS), we could investigate the thermal decomposition behavior of the chosen precursors. According to those results, we chose the heat treatment temperatures (T HT ) using partial oxidizing (PO) and reducing (RED) atmosphere. By X-ray diffraction we detected structure and size of the nanoparticles; the size was further confirmed by transmission electron microscopy. ZrO 2 formation happens at lower temperature with Zr(acac) 4 than with ZrOPc, due to the lower thermal stability and a higher oxygen amount in Zr(acac) 4 . Using ZrOPc at T HT C900°C, PO conditions facilitate the crystallite growth and formation of distinct tetragonal ZrO 2 , while with Zr(acac) 4 a distinct tetragonal ZrO 2 phase is observed already at T HT C750°C in both RED and PO conditions. Tuning of ZrO 2 nanocrystallite size from 5 to 9 nm by varying the precursor loading is also demonstrated. The chemical state of zirconium was analyzed by X-ray photoelectron spectroscopy, which confirms ZrO 2 formation from different synthesis routes.