(Invited) Atomic Scale Models of PGM-Free ORR Electrocatalysts: Activity, Stability, and Experimental Signatures of Active Site Structures within the Electrocatalysis Consortium

Abstract

The electrification of transportation will rely predominantly on electrochemical technologies including fuel cells and batteries. In 2016 the U.S. Department of Energy started the Electrocatalysis Consortium (ElectroCat) in order to increase U.S. competitiveness in fuel cell manufacturing. The main focus of the consortium has been how to replace rare, expensive, and non-domestically mined platinum group metals (PGMs) used as electrocatalysts in fuel cell cathodes to improve efficiency of the oxygen reduction reaction (ORR). At predicted production scale, these PGM components will be one of the dominant cost drivers for fuel cell vehicle membrane electrode assemblies and, unlike other components, will not benefit from economies of scale during mass production. Currently the most promising PGM-free ORR electrocatalysts are based on heat treated M-N-C structures (M = Mn, Fe, Co). Great strides have been made over the previous decade in improving ORR activity, 4 electron pathway selectivity, and durability of these PGM-free materials but further improvements are required to meet the application needs of these materials. Thermal treatments in the range of 800-1100 °C have been shown to produce the most ORR active PGM-free materials but these temperatures lead to highly heterogeneous structures. This heterogeneity makes it difficult to identify the atomic scale structures responsible for facilitating ORR activity (the so-called active site conundrum) and also complicates the study of their associated ORR reaction pathways and degradation mechanisms. One approach within ElectroCat to address these knowledge gaps and to improve synthesis through a targeted, rational approach is the application of computational quantum chemistry. In this approach, an inverse problem to the active site conundrum is addressed. Instead of trying to identify “what is the active site”, the approach is to build a library of plausible atomic scale structures based on theoretical and experimental input and to understand the activity, durability, and experimental signatures of each site, enabling a comparison between the structures within the library. These comparisons can address fundamental questions about active sites and guide in the interpretation of experimental data aiding in the identification and behavior of particular atomic scale structures under specific environmental conditions. The developed approaches rely heavily on high performance computing (HPC) and density functional theory (DFT) to provide input about atomic scale interactions. By studying the binding of ORR intermediates to given library member structures and application of thermochemical models, activity descriptors can be obtained as well as relative energies of proposed reaction pathways. Additionally, kinetics of bond breaking associated with proposed degradation pathways can be identified, giving proposed durability descriptors for given active site structures. Both of these approaches have provided valuable insight into the roles of transition metal species and local carbon environment for material activity and durability. Theoretical approaches can also aid in interpretation of experimental data. Quantification of the number of active sites is more complicated in PGM-free materials vs. their PGM counterparts and experimental approaches are still being developed in order to better decouple active site density and active site turn over frequency. Study of active site interaction with probe molecules, those that can be controllably added and removed from (ideally) only ORR active sites enabling site counting, can also be determined via theoretical approaches, shedding light on binding specificity and thus the veracity of proposed approaches. For Fe-bearing materials, a host of Fe-specific spectroscopies have been developed which enable direct study of this subset of proposed active site structures. These include Mössbauer spectroscopy, nuclear resonance vibrational spectroscopy (NRVS), and x-ray adsorption spectroscopy (XAS). One major drawback to these approaches is a lack of well characterized standard structures that enable assigning given peaks to particular components of the material. By producing libraries of atomic scale structures and their experimental signatures, trends in experiments can be identified. This is valuable to determine the impact of synthesis conditions on produced materials and predominance of structures produced. Also, it has been predicted that certain ligands may evolve spontaneously on these structures under certain conditions. By studying structures with various ligands and probe molecules, trends can be identified which help in determining not just predominant Fe-structures from spectroscopy but also what ligands evolve under what conditions in-situ. Combined, theoretical input for ORR activity, active site durability, and experimental interpretation will be discussed.