Density Functional Theory Studies Combined with Experimental Investigations of the Oxygen Reduction Reaction on Fe- or Co-Containing Carbon Fibers

Abstract

The oxygen reduction reaction (ORR) is responsible for a large potential loss for anion-exchange membrane fuel cells (AEMFCs), and researchers continue to develop non-Platinum-group-metal (non-PGM) electrocatalysts that attenuate the related kinetic losses. Moreover, it is desirable to use non-PGM electrocatalysts to mitigate fuel cell costs, and with respect to the ORR, AEMFCs provide an operating environment that allows a shift to non-PGM electrocatalysts.1,2 Many research investigations have been performed on non-PGM electrocatalysts such as M-phthalocyanines (MPc, where M = Fe, Co or Mn for example)3,4, heat-treated metal macrocycles5,6 and nitrogen-doped carbons devoid of metal nanoparticulates7. In some cases, these non-PGM electrocatalysts exhibit ORR activities comparable to Pt/C electrocatalysts. As a contribution to non-PGM electrocatalysts research efforts, we studied the ORR in 0.1N KOH at room temperature on Fe- or Co-carbon fibers derived by pyrolyzing electrospun nanofiber precursors. Electrospinning provides a facile approach for developing high-surface area carbon materials with a wide range of morphologies. However, as is the case with most heat-treated non-PGM electrocatalysts, the active site(s) for heat-treated metal-containing carbon materials are not well defined. Therefore, we combine our physical characterizations and electrochemistry findings with density functional theory (DFT) calculations and attempt to probe potential ORR pathways for Fe- or Co-N4-containing graphitic electrocatalyst clusters.We probe H2O, O2, HOOH, OOH, and OH adsorption on the Fe- and Co-containing graphitic clusters. In addition, we compare each of the adsorbate binding energies as a function of the spin and oxidation states. We also compare various DFT methods for each of the systems including M06-L, M062X and B3LYP. Lastly, we provide rotating ring disk electrode (RRDE) voltammetry, cyclic voltammetry and other physical characterizations such as XRD, XPS, TEM and SEM to better relate the electrocatalyst activity, morphology and surface chemistry to the DFT models developed. Figures 1 a-d show typical Fe- or Co-containing graphitic clusters used for the DFT calculations. Additional graphitic cluster configurations are considered and may be presented. Acknowledgements We gratefully acknowledge the fuel cell team at the U.S. Army Research Laboratory. We acknowledge the U.S. Department of the Army and Army Materiel Command for funding and support. The authors would also like to thank ARL for facilitating this opportunity through the Science, Mathematics and Research for Transformation (SMART) program.