Establishing catalytic structure−function relationships introduces the ability to optimize the catalyst structure for enhanced activity, selectivity, and durability against reaction conditions and prolonged catalysis. Here we present experimental and computational data elucidating the mechanism for the O 2 reduction reaction with a conductive nickel-based metal−organic framework (MOF). Elucidation of the O 2 reduction electrokinetics, understanding the role of the extended MOF structure in providing catalytic activity, observation of how the redox activity and pK a of the organic ligand influences catalysis, and identification of the catalyst active site yield a detailed O 2 reduction mechanism where the ligand, rather than the metal, plays a central role. More generally, familiarization with how the structural and electronic properties of the MOF influence reactivity may provide deeper insight into the mechanisms by which less structurally defined nonplatinum group metal electrocatalysts reduce O 2 . KEYWORDS: O 2 reduction, electrocatalysis, metal−organic framework, porous catalysts, 2D materials
■ INTRODUCTIONUnderstanding catalytic kinetics and thermodynamics to construct a reasonable reaction mechanism is central for both elucidating the behavior of a given catalyst and gaining predictive power over structure−function relationships. This predictive power aids in efficiently optimizing catalyst performance by systematically tuning the structural and electronic properties of the catalyst. One class of materials that could benefit from mechanism-guided optimization is nonplatinum group metal (non-PGM) electrocatalysts for the O 2 reduction reaction (ORR) to water (4e − reduction) and/or hydrogen peroxide (2e − reduction). Such catalysts typically include abundant transition metals and/or heteroatoms such as N, O, and S doped into a carbonaceous matrix.1−6 Although quite active and stable during ORR, previously reported non-PGM catalysts often consist of amorphous carbon mechanically blended with transition metal macrocycles or other metal and main group heteroatomic sources. These relatively poorly defined materials do not lend themselves to facile mechanistic studies; the inhomogeneous dispersion and irregular orientation of the dopants throughout the carbon matrix engenders structural ambiguity that makes identification, experimental probing, and computational modeling of active sites difficult.Conversely, highly ordered metal−organic frameworks (MOFs) containing well-defined, spatially isolated active sites present an attractive platform for experimental and computational correlation between the chemical and electronic structure of a given catalyst and the electrocatalytic activity and mechanism, a feat that is traditionally restricted to homogeneous molecular systems. We previously showed that the electrically conductive MOF Ni 3 (HITP) 2 (HITP = 2,3,6,7,10, (Figure 1) functions as