Oxygen Reduction Reaction Activity in Non-Precious Single-Atom (M–N/C) Catalysts─Contribution of Metal and Carbon/Nitrogen Framework-Based Sites

Abstract

We examine the performance of a number of single-atom M–N/C electrocatalysts with a common structure in order to deconvolute the activity of the framework N/C support from the metal M–N4 sites in M–N/Cs. The formation of the N/C framework with coordinating nitrogen sites is performed using zinc as a templating agent. After the formation of the electrically conducting carbon–nitrogen metal-coordinating network, we (trans)metalate with different metals producing a range of different catalysts (Fe–N/C, Co–N/C, Ni–N/C, Sn–N/C, Sb–N/C, and Bi–N/C) without the formation of any metal particles. In these materials, the structure of the carbon/nitrogen framework remains unchangedonly the coordinated metal is substituted. We assess the performance of the subsequent catalysts in acid, near-neutral, and alkaline environments toward the oxygen reduction reaction (ORR) and ascribe and quantify the performance to a combination of metal site activity and activity of the carbon/nitrogen framework. The ORR activity of the carbon/nitrogen framework is about 1000-fold higher in alkaline than it is in acid, suggesting a change in mechanism. At 0.80 VRHE, only Fe and Co contribute ORR activity significantly beyond that provided by the carbon/nitrogen framework at all pH values studied. In acid and near-neutral pH values (pH 0.3 and 5.2, respectively), Fe shows a 30-fold improvement and Co shows a 5-fold improvement, whereas in alkaline pH (pH 13), both Fe and Co show a 7-fold improvement beyond the baseline framework activity. The site density of the single metal atom sites is estimated using the nitrite adsorption and stripping method. This method allows us to deconvolute the framework sites and metal-based active sites. The framework site density of catalysts is estimated as 7.8 × 1018 sites g–1. The metal M−N4 site densities in Fe−N/C and Co–N/C are 9.4 × 1018 sites–1 and 4.8 × 1018 sites g–1, respectively.

Chemical Kinetic Method for Active-Site Quantification in Fe-N-C Catalysts and Correlation with Molecular Probe and Spectroscopic Site-Counting Methods

“…This requirement may be somewhat limiting and could also apply to the NO-stripping technique if some MN x sites do not adsorb NO with sufficient strength. This hypothesis is consistent with recent data from Kucernak and co-workers that show a lack of a clear NO stripping peak on materials containing Zn, Ni, Sn, Sb, Bi, and Mn . Active-site saturation requirements do not limit HQ oxidation, but the generality may be limited by different considerations.…”

“…The electrochemical stripping of NO derived from NO 2 – was measured on a Nafion-bound film of 0.4Fe-N-C m‑ht deposited onto a rotating-disc electrode following protocols adapted from Kucernak and co-workers. ,, The material was exposed to a 0.125 M NO 2 – solution, and then the adsorbed NO 2 – was converted to NO by exposure to H + in a buffered electrolyte solution (0.5 M acetate, pH = 5.2, Figure B). The electrochemical ORR kinetic current density (0.8 V RHE ) decreased by a factor of 12 after NO poisoning (Figure S28b, Supporting Information), which is consistent with adsorption of NO to FeN x active centers, rendering them inactive.…”

“…This second poisoning and stripping protocol was adopted to ensure that the stripping curve was not perturbed by exposure to O 2 during the “poisoned” ORR LSV, as recommended in a recent report . The “pre-baseline” (e) and “baseline” (f) steps were repeated after this final stripping, which was recently reported by the same group to be suitable for baseline correction . The FeN x site density ( N FeNx , mol g –1 ) was calculated according to the following equation: N normalF normale N x = Q normals normalt normalr normali normalp n normals normalt normalr normali normalp F where Q strip (C g –1 ) is the charge associated with the reductive stripping of NO, F is Faraday’s constant (C mol –1 ), and n strip is the number of electrons associated with the reduction of a NO–FeN x species.…”

Intrinsically Functionalized Fe/C3N4 Nanotubes Display pH-Dependent Oxygen Reduction Overpotentials

“…We summarize the overpotentials of ORR (η ORR ) on different surfaces with the pH corrections in Figure . A general trend is observed that most of the η ORR ’s are lower in the alkaline media than in the acidic media, which is consistent with the experimental findings irrespective of catalytic materials. ,,,− Some of the irregular trends of zN-6 in alkaline media could be due to the specific active site that can be found only in small CNNTs. Since the diameter of zN-6 is as small as 6 Å, it is difficult to compare with the existing experimental data of planar C 3 N 4 or larger C 3 N 4 tubes.…”

Atomically Dispersed Dual-Metal ORR Catalyst with Hierarchical Porous Structure for Zn–Air Batteries

“…The carbonate salt template produces CO 2 via the reaction of Na 2 CO 3 deposition under a higher temperature. CO 2 could etch the carbon skeleton via the reaction of C + CO 2 → 2CO↑, which leads to a porous structure with a higher surface area. − It creates a slower diffusion-limited reaction phase on support during carbonization to expose more active surface area. The dual-atom site construction and template method are helpful in improving Fe–N–C, but it is still a challenge to employ the two methods simultaneously to design dual-atom catalysts with rich porous structure, higher specific surface area and high efficiency.…”