In this work, we evaluate the performance of three catalysts by analyzing their impedance at different frequencies using DRT and explaining the catalytic behaviour. The three catalysts are a standard platinum catalyst (Pt on Vulcan X-72, Heraeus), a platinum alloy catalyst (Pt0.7Co0.3, Tanaka), and a novel Fe-C-N catalyst, a highly active and durable catalyst of the platinum-free, transition metal-nitrogen-carbon family, derived from Nicarbazin and possessing an open frame structure which is controlled by a silica sacrificial support method [3]. For each material, we assembled a HT-PEM fuel cell MEA and recorded polarization curves and EIS spectra at 100 mAcm-2. This data is shown in Figure 1 b) and c). The EIS spectra were processed using the DRT method (Figure 1 a) and yielded the frequency resolved impedance, giving insight into the processes causing the performance differences.
Quantitative results were obtained by DRT analysis for the operation of the cells at 100 mAcm-2. We found the platinum/cobalt alloy catalyst to be outperforming platinum with an increased ORR activity, explained by the facilitated adsorption and reduction of oxygen resulting from the lower interatomic distance in the alloyed catalytic system [4]. The Fe-N-C system exhibits a lower performance compared to the noble metal catalysts, showing a significantly increased mass transport region and a higher ORR resistance. Contrary to our expectation, the mass transport resistance is dominating the ORR resistance. This can be explained by the narrow spaces in between the nitrogen-doped graphenic layers in which the reactive sites are present [3]. The reactivity of the non-PGM active sites is not the primarily inhibiting factor of the performance.
DRT proves to be a powerful tool to inspect the impedance of a fuel cell in great detail and to learn about the inner mechanisms, highlighting positive and negative aspects of different catalysts. This tool allows us to systematically investigate the performance of catalysts and to elucidate the influence of single parameters, making the optimization straight forward and focused instead of depending on trial and error experiments.
1. F. Ciucci, C. Chen, Electrochim. Acta, 167, 439 (2015).
2. A. Weiß, S. Schindler, S. Galbiati, M.A. Danzer, R. Zeis, Electrochim. Acta, 230, 391 (2017).
3. A. Serov, K. Artyushkova, P. Atanassov, Adv. Energy Mat., 4, 7 (2014).
4. C.V. Rao, J. Parrondo, S.L. Ghatty, B. Rambabu, J. Pow. Sources, 195, 3425 (2010).