Interplay Between Chemistry and Morphology of the Catalytic Layer in Platinum Group Metal-Free Cathodes

Abstract

Among the different Fuel Cell technologies, one is based on the implementation of Polymer Electrolyte Membrane fuel cells (PEMFC). Significant research effort has been directed towards replacement of the platinum by materials mainly consisting of metal–nitrogen–carbon (MNC) network. The knowledge derived from tens of years of optimization of Pt-based PEMFC can not be directly translated to MNC catalytic systems due to different approaches in making catalyst layers. The structure of the active site/sites of the PGM-free ORR electrocatalysts remains contentious even after 50 years of research. The structure of the active site within the Membrane Electrode Assembly (MEA) has not been studied at all. It is necessary to understand the effect of the interaction of ionomer and catalyst on the structure of the active site, morphology of catalyst layer and durability. There is a missing link between durability and activity parameters as well as chemical and morphological changes that occur inside the catalyst layer during the oxygen reduction reaction in a fuel cell. In this study, we are investigating a series of MNC electrocatalysts synthesized by the same sacrificial support route[1] and their performance in MEA. Chemistry of the electrocatalysts and catalyst layers is studied by XPS. The types of nitrogen and iron-nitrogen functionalities that are present in these materials are in-plane defects such as graphitic N and N-coordinated to three or four nitrogens and a multitude of possible edge sites such as pyridinic, pyrrolic, quaternary and Fe-N2/Fe-N sites. [2, 3] In the catalyst layers, the makeup of the active site may be affected by the interactions with negatively charged sulfonate group of nafion. Due to this interaction between groups on the catalyst surface and sulfonate groups of nafion, the rearrangement in high binding energy range has been observed. The higher relative amount of peaks between 401-403 eV in the catalyst layers is due to the shift in the position of other peaks due to interaction with an ionomer. DFT calculations were used to evaluate the strength of the interaction between different types of nitrogen containing defects and sulfonate groups and to calculate binding energy shifts of N 1s spectra upon ionomer binding. [4] Figure 1 shows DFT calculated adsorption energies of sulfonate groups on nitrogen defects. High adsorption energies particularly for protonated nitrogens which may be either pyrrolic or protonated pyridine nitrogen is observed. Moreover, the higher amount of protonated nitrogens in catalyst layer results in worse MEA performance. The pore structure is critical to the transport of oxygen to active sites and removal of water. The change in pore structure induced by the chemical changes introduced during fuel cell operation has to be understood in order to design PGM-free electrocatalyst with highest possible lifetime. The morphology of catalyst layers will be analyzed by focused ion beam/scanning electron microscopy (FIB-SEM) sectioning. This will allow to obtain a 3D visual representation of morphology of catalysts layers and to estimate in detail the evolution of structural parameters such as: specific surface area, total porosity, connectivity of pores and others as a result of degradation studies. 1. Serov, A., et al., Nano-structured non-platinum catalysts for automotive fuel cell application. Nano Energy, 2015. 16: p. 293-300. 2. Jia, Q., et al., Spectroscopic Insights into the Nature of Active Sites in Iron-Nitrogen-Carbon Electrocatalysts for Oxygen Reduction in Acid and the Redox Mechanisms. Nano Energy, 2016. 3. Artyushkova, K., et al., Chemistry of Multitudinous Active Sites for Oxygen Reduction Reaction in Transition Metal-Nitrogen-Carbon Electrocatalysts. Journal of Physical Chemistry C, 2015. 119(46): p. 25917-25928. 4. Kabir, S., et al., Binding energy shifts for nitrogen-containing graphene-based electrocatalysts – experiments and DFT calculations. Surface and Interface Analysis, 2016. Figure 1