Density Functional Theory Study of Oxygen Reduction Reaction Mechanism on Fe-N-C Non-Precious Metal Catalyst

Abstract

Proton exchange membrane fuel cells (PEMFCs) have attracted enormous attention as a clean and sustainable source of power generation. However, the overall performance of a PEMFC is controlled by the sluggish oxygen reduction reaction (ORR) at the cathode which requires an efficient catalyst for effective reduction of O 2 to H 2 O. Currently, Pt-based catalysts are the best performing ORR catalysts. 1 The price of rare and expensive Pt-based catalysts contributes significantly towards the total cost of a PEMFC and hinders its scaled up application. Thus, it is desirable to explore cost-effective non-Pt based ORR catalysts as alternatives to Pt-based catalysts. Of particular interests are transition metals (Me=Fe, Co, Ni) and nitrogen (N) containing materials (Me-N-C catalysts) with embedded MeN 4 clusters in carbon support. 2-5 Previous experimental studies have illustrated that heat treated Fe-N-C catalysts promote a 4e - reduction of O 2 to H 2 O with their catalytic activity for ORR comparable to that of the benchmark Pt catalyst. 2,3 However the chemical nature of ORR active sites and the mechanism of ORR in Fe-N-C catalysts are still poorly understood. Here, we report density functional theory (DFT) study of the mechanism of ORR on Fe-N-C catalysts. Firstly, we studied the thermodynamic stability of various nitrogen N x (x=0, 1, 2, 3, and 4) and combined transition metal and nitrogen FeN x (x=0, 1, 2, 3, and 4) clusters in a monolayer graphene. We found that the formation energies are very high for all N x clusters. In contrast, the formation energies of FeN x clusters are significantly smaller than those of the corresponding N x clusters. Our DFT results show that the formation energy of FeN x clusters decreases with increasing N coordination to Fe and FeN 4 cluster has the lowest formation energy. Thus, we predict that FeN 4 clusters are thermodynamically stable in a monolayer graphene. The possible mechanism of ORR on the FeN 4 clusters in graphene is investigated through DFT transition state calculations. We calculated the activation energies (E a ) for all the possible elementary chemical reactions in ORR on the FeN 4 clusters embedded in graphene. O 2 dissociation on the FeN 4 clusters (Fig. 1a) has high activation energy of 1.20 eV. In contrast, O 2 hydrogenation chemical reaction (Fig. 1b) to form OOH is nearly barrierless (E a =0.01 eV). Following the formation of OOH, the O-O bond scission in OOH can occur either by its direct dissociation to form O and OH (Fig. 1c) or by its hydrogenation chemical reaction to form 2OH (Fig. 1d). The activation energy for the former reaction is 0.64 eV and the latter reaction is 0.84 eV. The activation energies for subsequent reduction of O to OH and OH to H 2 O are 0.24 and 0.44 eV respectively. Based on our calculated activation energies of the elementary steps in ORR on the FeN 4 clusters, we predict that O 2 hydrogenation to form OOH is kinetically favorable over direct O 2 dissociation. At U=0V, OOH formation would be followed by its dissociation to O and OH. Thus our results show that the ORR on the FeN 4 clusters would favorably proceed via an OOH dissociation ORR mechanism. Since the activation energies for O-O bond scission in OOH by its direct dissociation to form O and OH or by its hydrogenation to form 2OH are not very different, a competing (H+OOH) dissociation ORR mechanism cannot be ruled at non-zero electrode potential. In either mechanisms, our calculated activation energy of the rate determining step is comparable to that (0.79 eV) 6 on the Pt(111) surface. Hence we predict a comparable catalytic activity of FeN 4 clusters embedded in graphene for ORR to that of Pt-based catalysts in agreement with the experimental findings. Our theoretical calculations provide an explanation for the experimentally observed 4e - ORR on Fe-N-C catalysts. Acknowledgements This work was funded by Chemical Sciences Research Programs, Office of Basic Energy Sciences, U.S. Department of Energy (Grant no. DE-FG02-09ER16093). We also acknowledge the research grant from the EERE program of the U.S. Department of Energy (Grant no. DE-AC02-06CH11357). References (1) Stamenkovic, V. R.; Mun, B. S.; Arenz, M.; Mayrhofer, K. J. J.; Lucas, C. A.; Wang, G. F.; Ross, P. N.; Markovic, N. M. Nat. Mater. 2007 , 6 , 241-247. (2) Wu, G.; More, K. L.; Johnston, C. M.; Zelenay, P. Science 2011, 332 , 443-447. (3) Lefevre, M.; Proietti, E.; Jaouen, F.; Dodelet, J. P. Science 2009, 324 , 71-74. (4) Kattel, S.; Atanassov, P.; Kiefer, B. Phys. Chem. Chem. Phys. 2013, 15 , 148-153. (5) Kattel, S.; Wang. G. J. Mater. Chem. A 2013 , 1 , 10790–10797. (6) Duan, Z. Y.; Wang, G. F. Phys. Chem. Chem. Phys. 2011 , 13 , 20178-20187.