Longevity is a key criterion for the future success of fuel cells in challenging markets like the automotive drive train as it contributes to cost and user experience. The analysis of degradation processes is therefore currently in the focus of research. It is known, that several degradation processes exist, which over time can cause the failure of LT-PEMFC. Examples are changes in the structure of the catalyst particles due to effects like platinum dissolution or Ostwald ripening, the loss of catalyst particles due to corrosion of the carbon support or thinning and pinhole formation of the membrane due to attack by radicals and other detrimental intermediates. A number of test protocols exists which try to selectively test for the influence of one particular degradation process. Thereby the different degradation processes are assigned to vehicle operation conditions as described e.g. by Ohma et al. for the Fuel Cell Commercialization Conference of Japan (FCCJ) test protocols¹. Common assignments are e.g. the tests of platinum particle degradation to driving at variable loads as it is tested with voltage jumps between 0.6 V and 1.0 V or the membrane degradation to standby as it is tested by potential hold at OCV. The carbon corrosion tests, which usually are assigned to system start-stop and are performed by fast potential cycling between 1.0 V and 1.5 V. the assignment as well as the selected potential window is derived from the results of Reiser et al.², who found very high potential for the transient situation that anode and cathode are both filled with air when hydrogen is admitted to the cathode for the system start. This situation also often referred to as air/air start, can indeed occur in the operation of an automotive fuel cell stack. However, as it is due to longer standstill of the vehicle, it will happen at the vehicle ambient temperature which is likely 20 °C or below.

The effect of temperature on degradation processes at fuel cell catalyst and fuel cell membrane electrode assemblies has been studied before also by our group³ but usually with focus to elevated temperatures well above ambient temperature. Publication of results even on catalyst activity for the relevant temperature window between 0 °C and 20 °C are scarce and even less exist which include degradation effect.

Here we will present on test series in which we studied a commercial Pt/C fuel cell catalysts, both at RDE and single cell level, in the temperature window between 0 °C and 20 °C. In Figure 1 results on the RDE measurement at different temperatures are depicted. As can be seen in figure 1 a), the direct effect of the temperature on the performance is simply a reduced activity without changes in the Tafel slopes. If the carbon corrosion test is performed at different temperatures the change of the electrochemical surface area (ECSA) determined by evaluation of the H_UPD part of the CV differs for the different temperatures. At all temperatures the carbon corrosion test causes a slight increase in ECSA following the first sets of stress cycles followed by a severe decay. Now, at lower temperatures this initial increase is stronger and endures longer than at 20 °C. Also the final decay is much more important at 20 °C than at 10 °C or 0 °C. In contrast to this finding, the kinetic current density at 0.9 V after the entire set of 20.000 cycles shows a stronger increase at 20 °C than at the lower temperatures (cf. figure 1 c). As the current density after the stress cycles is normalised to the remaining surface area, it means that the remaining catalyst has a higher activity. In this context, the results of the CO stripping before and after the stress tests shown in figure 1 d) are of interest. Here it is found that the carbon corrosion tests cause the evolution of a pre-peak oxidation area for CO which according to Maillard et al.⁴ is indicating the presence of particle agglomerates. This behaviour is much lower if the stress test is performed at lower temperatures. In the presentation further results, including test on commercial Pt₃Co/C catalyst and single cell tests will be reported.

1. A. Ohma, K. Shinohara, A. Iiyama, T. Yoshida, and A. Daimaru, ECS Trans., 41(1), 775–784 (2011)
2. C. A. Reiser et al., Electrochem. Solid-State Lett., 8, A273–A276 (2005)
3. C. Cremers et al., J. Electrochem. Soc., 165 (2018)
4. F. Maillard et al., Phys. Chem. Chem. Phys., 7, 385–393 (2005)