Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2026 Apr 1;148(14):15114–15124. doi: 10.1021/jacs.6c00705

Operando Cu Aggregation-Induced Spin State Modulation in Fe–Cu Single Atom Catalyst for Enhanced Tandem Electrochemical Nitrate Reduction Reaction

Seongin Hong †, Jaewoo Jeong ‡, Euichan Yoo †, Dongwoo Shin †, Suhwan Yoo †, Eunchong Lee †, Hyungjun Kim ‡, Hyeyoung Shin §,*, Yun Jeong Hwang †,*
PMCID: PMC13088231  PMID: 41920551

Abstract

The electrocatalytic nitrate reduction reaction (NO3RR) provides a sustainable pathway to convert excess nitrate into ammonia, yet realizing high selectivity requires a fundamental understanding of dynamic structural changes occuring at active sites during reactions. Here, we investigate how in situ Cu clustering dynamically activates dual catalytic sites in Fe–Cu bimetallic single-atom catalysts (FeCu–N–C) during NO3RR, through combined density functional theory calculations and operando spectroscopy. Under reductive potentials, atomically dispersed Cu spontaneously aggregates into nanoclusters that efficiently activate NO3 –. Concurrently, Cu clustering induces pronounced structural strain and electronic distortion in adjacent Fe–N x moieties, triggering a spin-state transition in the Fe active site from low-spin to high-spin configuration. This spin modulation dramatically enhances the activity for subsequent NO2 – conversion to NH3. The synergistic coupling between Cu clusters and spin-modulated Fe establishes a highly effective tandem pathway, yielding superior NO3RR activity and NH3 selectivity, compared to Cu–N–C and Fe–N–C counterparts. These findings provide new insights into the rational design of advanced multicomponent electrocatalysts with dynamically tunable active site properties.

graphic file with name ja6c00705_0007.jpg

graphic file with name ja6c00705_0005.jpg

Introduction

The nitrogen cycle is a crucial biogeochemical pathway that has been disrupted by artificial nitrogen fixation and human industrial activities, leading to the excessive accumulation of nitrate (NO3 –). , Sustainable electrocatalytic reactions are being investigated, particularly the electrochemical NO3 – reduction reaction (NO3RR), which aims to convert surplus NO3 – to valuable chemical, NH3, via deoxygenation and hydrogenation steps. − Achieving high selectivity toward NH3 requires a detailed understanding of the structure–activity relationships of electrocatalysts, especially to suppress the competing hydrogen evolution reaction (HER) and incomplete reduction pathways to NO2 –. Recent studies have shown that active sites undergo dynamic changes during electrochemical reactions, in both atomic structures and electronic states, in response to applied potentials and interactions with intermediates. Therefore, these dynamic transformations complicate the identification of the actual active sites, highlighting the need for operando characterization to elucidate and control catalytic activity under realistic reaction conditions.

Among electrocatalysts, single-atom catalysts (SACs) have attracted significant interest as an ideal platform for various electrocatalytic reactions, including NO3RR, − due to their high atomic utilization, tunable coordination environments, and well-defined active sites, which enable precise structure–activity correlations. By engineering the coordination environments of active single atoms, it is possible to selectively modulate energy barriers for multiple proton–electron transfer steps. Among the SACs explored for NO3RR, Fe and nitrogen-doped carbon (Fe–N–C) catalysts have demonstrated high activity, primarily due to Fe–N4 coordination, which enhances electron transfer and stabilizes key reaction intermediates. However, single-atom sites often face selectivity challenges in multistep reactions that involve more than one intermediate species. The formation of NH3 competes with HER and other nitrogen-containing intermediates, such as NH2OH, depending on the reaction conditions. , These limitations have motivated strategies to introduce additional metal centers to enhance product selectivity across complex reaction pathways. , Meanwhile, Cu-based single-atom catalysts (Cu-SACs) have shown promise as electrocatalysts, , but, interestingly, Cu single atoms in the Cu–N–C frameworks are thermodynamically unstable and tend to aggregate into nanoparticles (NPs) under cathodic potentials. This aggregation induces in situ structural and electronic modifications that can significantly alter catalytic behavior. , Importantly, the evolved Cu NPs are not merely a byproduct of structural reconstruction but can actively participate in NO3RR by influencing adsorption and reaction pathways.

When SAC catalysts are designed to have both Fe and Cu sites (FeCu–N–C), the coexistence of dual active sites can provide tandem catalysis when they interact closely. However, a significant gap remains in understanding whether the aggregation of Cu atoms within such tandem catalysts can induce structural distortion of the carbon matrix and modulate the electronic structure of neighboring Fe active sites. Fe centers are known to exhibit distinct high-spin and low-spin states depending on ligand-field splitting, which can strongly influence adsorption energetics and reaction kinetics. Therefore, exploring this secondary interplay under the electrocatalytic operational conditions may uncover new factors that tune the active sites.

Here, we employ density functional theory (DFT) calculations together with operando spectroscopic techniques to investigate the role of Cu clustering in activating both Cu and Fe sites in FeCu–N–C catalysis under electrochemical NO3RR conditions. DFT calculations and operando X-ray absorption spectroscopy (XAS) reveal that Cu single-atom dynamically aggregates into clusters, which preferentially facilitate NO3 – activation. In contrast, although Fe–N sites remain atomically dispersed, they undergo electronic and structural distortions induced by adjacent Cu clusters, generated under the electrochemical reduction conditions. The electrochemically induced spin-state modulation of the Fe active site, identified by electron paramagnetic resonance (EPR) analysis, is proposed to enhance the subsequent reduction of NO2 – to NH3. Complementary electrochemical analysis and attenuated total reflection surface-enhanced infrared spectroscopy (ATR-SEIRAS) further corroborate the synergistic catalytic activity. These findings provide mechanistic insights into the dynamic evolution of active sites and spin-state modulation driven by in situ structural distortion, offering design principles for advanced multicomponent electrocatalysts.

Results and Discussion

DFT Simulation to Clarify the Role of Cu Clusters During Nitrate Reduction on Fe–N–C

Given the promise of FeCu–N–C as a tandem electrocatalyst, understanding the dynamic behavior of Cu under reductive potentials is critical. Under reductive potentials, isolated Cu atoms tend to aggregate into clusters, which can introduce new active motifs and simultaneously perturb neighboring Fe–N–C sites within the same carbon matrix. Such aggregation-induced structural effects are expected to modify the local electronic environment and reactivity of neighboring Fe sites within the framework. To elucidate how these coupled effects reshape NO3RR activity, we employed DFT calculations to probe the role of Cu clusters both as direct reaction sites and as structural modulators of Fe–N–C. Cu aggregation was modeled using a Cu55 cluster (Cu55@Fe–N–C), a well-established theoretical model corresponding to a particle size of approximately 1 nm (Figure S1), providing a representative description of cluster-induced effects on the Fe–N–C framework. , Direct Fe–Cu bonding was intentionally excluded by preserving localized N–C coordination around each metal center, thereby isolating the intrinsic roles of Fe and Cu. Computational details are provided in the Supporting Information. To assess structural feasibility, we calculated the cluster formation energy per atom (E Formation) for both Cu and Fe clusters using the N = 55 model (Figure S2). Cu clusters exhibit a significantly lower E Formation (0.83 eV atom–1) than Fe clusters (1.32 eV atom–1), indicating that Fe clustering is energetically unfavorable under NO3RR conditions, whereas Cu aggregation is thermodynamically preferred as reported in previous studies. ,

We first assessed how Cu cluster formation alters the overall NO3RR reaction energetics by comparing free energy diagrams under representative structural conditions (Figure a). Incorporation of Cu clusters leads to a systematic reduction in activation barriers and limiting potentials, accompanied by a reorganization of the reaction pathway. The activation barrier for *NO3 dissociation to *NO2 and *O decreases from +0.94 eV on Fe–N–C to +0.66 eV on Cu55@Fe–N–C, indicating significant stabilization of the transition state (Figure S3). In addition, the limiting potential shifts favorably from +0.20 to +0.13 V, accompanied by a change in the potential-determining step (PDS) from *NOH → *N on Fe–N–C to *N→ *NH on Cu55@Fe–N–C. The modified free energy landscape suggests a functional decoupling of reaction steps, in which early stage NO3 activation is preferentially facilitated at Cu-associated sites, while the subsequent hydrogenation sequence (*NOH → *NH3) is energetically governed by Fe sites. This redistribution of reaction energetics demonstrates that the influence of Cu clusters is not confined to the initial NO3 activation step but extends across multiple elementary steps, providing a thermodynamic basis for tandem catalysis.

1.

1

Open in a new tab

DFT results for the nitrate reduction reaction on Fe-SAC catalysts with Cu clusters of different sizes. (a) Free energy diagrams for nitrate reduction on Cu55@Fe–N–C (red) and Fe–N–C (navy), referenced to the *NO3 state set to 0 eV. The kinetic barrier for the *NO3 → *NO2 step is shown in the inset, including the corresponding transition state. The potential-determining step (PDS) is highlighted by a yellow line, and the associated intermediate is indicated in the reaction diagram. Cu, Fe, N, and C atoms are shown in orange, blue, gray, and brown, respectively, while adsorbed N and H are shown in green and pink. (b) Calculated kinetic barriers for the *NO3 → *NO2 step as a function of curvature. (c) Energy differences between the high-spin and low-spin states of Fe (E HS – E LS) in Fe–N–C at different curvatures. (d) Schematic illustration of Fe d-orbital splitting for planar D 4h and distorted D 4h geometries.

To understand the structural origin of the reaction landscape reorganization described above, we examined the optimized intermediate structures along the NO3RR pathway (Figures S4 and S5). Notably, these structures revealed a discernible distortion of the Fe–N–C framework in the presence of Cu clusters, suggesting that Cu aggregation induces a non-negligible deformation of the local Fe coordination environment. Given that the extent of this distortion is expected to depend sensitively on the size of the adjacent Cu cluster, FeCu–N–C models were constructed by introducing Cu clusters of varying sizes (N = 1, 13, 38, and 55), denoted as Cu N @Fe–N–C. As cluster size increased, the Fe–N–C layer exhibited progressively larger out-of-plane distortion, particularly along the surface normal z-axis. To quantitatively capture this deformation in a unified manner, we defined a curvature parameter (K) as the inverse square of the radius of an idealized sphere encompassing the Fe center and its Fe–N4 coordination plane. The curvature increased systematically with cluster size (K = 0, 0.05, 0.13, and 0.18 nm–2), indicating that larger Cu clusters induce more pronounced structural deformation in the surrounding N–C framework.

Building on the curvature analysis, we next examined how Cu-induced structural deformation simultaneously modulates reaction kinetics and NO3 adsorption behavior. Increasing curvature correlates with a decrease in kinetic barriers for *NO3 dissociation to *NO2 and *O along the NO3RR pathway (Figure b), indicating that Cu-cluster-induced deformation directly facilitates key elementary steps. In parallel, curvature also governs the initial NO3 adsorption preference between Fe and Cu sites (Figure S6). As curvature increased, NO3 adsorption on Fe weakened substantially, from −2.35 eV at K = 0 nm–2 to −0.53 eV at K = 0.18 nm–2. In contrast, NO3 adsorption on Cu sites remained nearly constant at approximately −1.43 eV over the same range. At higher curvature (K ≥ 0.13 nm–2), Cu becomes the thermodynamically preferred adsorption site, indicating a strain-driven shift in site preference. As NO3 adsorption is progressively disfavored on Fe sites with increasing curvature, Fe centers are relieved from participating in the initial activation step and instead become energetically suited for downstream hydrogenation reactions. The free-energy profiles further show that the limiting potential for NO3RR overall pathway decreases systematically with increasing cluster size, from +0.61 eV for Cu1@Fe–N–C to +0.13 eV for Cu55@Fe–N–C (Figure S7).

To further elucidate whether curvature-induced structural deformation gives rise to corresponding electronic structure changes at Fe sites, we performed projected density of states (pDOS) analysis. In the pristine Fe–N–C configuration, the Fe–N4 moiety adopts D 4h symmetry, which stabilizes the low-spin configuration through well-defined d-orbital splitting. Upon Cu cluster formation, the Fe–N4 unit undergoes a pronounced out-of-plane (z-axis) structural distortion, breaking the local D 4h symmetry and modifying the ligand-field environment along the axial direction. This symmetry breaking alters the relative splitting of Fe d-orbitals near the Fermi level (Figure S8). Spatially resolved pDOS comparisons between Fe atoms located near and far from the Cu clusters confirm that these electronic perturbations originate from Cu-cluster-induced structural distortion (Figure S9).

Consistent with the curvature-driven electronic reorganization revealed by PDOS analysis, we next evaluated the spin-state energetics of Fe sites as a function of Cu cluster size. Increasing curvature induces a spin-state transition of Fe from a +2 low-spin (LS) to a +2 high-spin (HS) configuration. The energy difference between HS and LS states (E HS – E LS) decreases from +0.28 to −0.32 eV with increasing cluster size (Figure c), confirming a spontaneous spin crossover in the presence of large Cu clusters. This transition is absent in models containing atomically dispersed or small Cu clusters, indicating that spin-state modulation is uniquely associated with significant cluster-induced strain. In this context, the loss of D 4h symmetry leads to a relative stabilization of the Fe d z 2 orbital, providing an electronic basis for the low-spin to high-spin transition schematically illustrated in Figure d. These results indicate that local physical strain imposed by neighboring Cu clusters is sufficient to modulate the electronic structure of Fe sites, even in the absence of direct Fe–Cu chemical bonding.

Collectively, these results demonstrate that Cu cluster formation in FeCu–N–C reshapes the NO3RR reaction landscape not only through the introduction of additional Cu active sites but also through curvature-driven structural deformation that modulates the electronic state of neighboring Fe centers. Cu clusters preferentially promote the initial NO3 activation step, while the associated structural distortion alters the ligand-field environment of Fe, leading to d-orbital reorganization and enhanced accessibility of a high-spin configuration. This spin-state evolution weakens NO3 adsorption on Fe sites and reassigns their role toward downstream hydrogenation reactions, providing a consistent explanation for the observed changes in adsorption energetics and kinetic barriers. Together, these findings establish a cooperative tandem mechanism in which cluster aggregation, symmetry breaking, and strain-mediated electronic modulation collectively underpin the enhanced NO3RR performance.

Synthesis and Structural Characterization of Fe–N–C, Cu–N–C, and FeCu–N–C Catalyst

Based on these computational findings, we compared the NO3RR activity of the FeCu–N–C, Fe–N–C, and Cu–N–C catalysts (Figure ). All three samples were successfully synthesized by modifying a well-known zinc imidazole framework (ZIF)-derived method, and the amount of metal atoms was precisely controlled by adjusting the Cu and Fe precursor solutions. Complementary X-ray diffraction (XRD) analysis revealed no discernible diffraction peaks corresponding to crystalline metallic Fe, Cu, further confirming the absence of nanoparticles or preformed clusters in the as-synthesized catalysts (Figure S10). The atomically dispersed metal sites were confirmed from high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images (Figure a). , TEM images of each catalyst displayed no visible nanoparticles (Figure S11), and EDS mapping results verified the uniform distribution of Fe and Cu atoms bound to the carbon support (Figure S12).

2.

2

Open in a new tab

Material characterization and electrocatalytic NO3RR performance of FeCu–N–C electrocatalysts. (a) HAADF-STEM image of FeCu–N–C showing the homogeneous distribution of single atoms. (b) Ex situ Fe K-edge EXAFS spectra and (c) Cu K-edge EXAFS spectra of each catalyst. (d) Linear sweep voltammetry of Fe–N–C, Cu–N–C, and FeCu–N–C in 1 M KOH electrolyte (dashed line) and 0.5 M KNO3 + 1 M KOH electrolyte (solid line) at a scan rate of 20 mV s–1. (e) NH3 Faradaic efficiency, (f) NH3 yield rate, and (g) NO2 – yield rate comparison. (h) Isotope labeling test for FeCu–N–C catalyst using 15NO3 – at −0.34 VRHE by 1H NMR. (i) Stability test for 20 h of electrolysis. The electrolyte was refreshed every 2 h.

The electronic oxidation states of Fe and Cu were characterized using X-ray photoelectron spectroscopy (XPS). Fe 2p and Cu 2p XPS spectra revealed that each metal in the FeCu–N–C had binding energies nearly identical to those of Fe–N–C and Cu–N–C, respectively, indicating no direct electronic interaction between Cu and Fe. Each Fe and Cu state is similar except for slight changes in the relative proportions. The Fe–N–C catalyst showed two distinctive 2p3/2 electronic states of Fe3+ at 714.6 eV and Fe2+ at 710.5 eV, consistent with the previous study of Fe–N–C materials (Figure S13a). , Cu 2p XPS spectra of Cu–N–C catalyst showed 934.7 and 932.4 eV associated with Cu2+ and Cu0/+, respectively (Figure S13b). The Cu LMM Auger spectra (Figure S14) show features at 570.2 and 568.8 eV indicating Cu+ and Cu2+ states, respectively, and the Cu+ to Cu2+ ratio was similar to that obtained from the Cu 2p spectra (Table S1). To obtain the ligand coordination environment of the catalysts, N 1s XPS spectra were also collected and deconvoluted into three distinct peaks corresponding to graphitic N (400.9 eV), pyrrolic N (399.3 eV), and pyridinic N (398.5 eV), respectively (Figure S15). Compared to the Fe–N–C catalyst, Cu-incorporated Cu–N–C and FeCu–N–C catalysts exhibited an increased proportion of pyrrolic N relative to pyridinic N. Considering that pyridinic N was associated with M-N4 sites, it appeared that Fe–N–C predominantly features a stable Fe–N4 structure. The Cu–N–C and FeCu–N–C catalysts exhibited increased pyrrolic N characteristics, indicating that the Cu–N x sites have more defective structures and are more susceptible to structural changes during NO3RR.

The ex situ XAS analysis was performed to further analyze the electronic states and local structures of the catalyst. Again, the Fe K-edge X-ray absorption near-edge structure (XANES) spectra indicate that the Fe–N–C and FeCu–N–C have similar features (Figure S16a), with an observed pre-edge at the same region. The Cu K-edge XANES exhibit similar features for both Cu–N–C and FeCu–N–C (Figure S16b). The extended X-ray absorption fine structure (EXAFS) results of the three catalysts exclusively exhibited strong features corresponding to metal-N sites near 1.4 Å, without the appearance of the M–M peak near 2.2 Å (Figure b,c, Tables S2, S3, Figures S17 and S18). This indicates that the synthesized catalysts contain local Fe and Cu structures bound to N. The coordination number (CN) of Fe–N was close to four, and the Cu–N–C and FeCu–N–C exhibited a slightly decreased Cu–N CN of 3.6. Overall, the electronic state analysis suggests that each metal center is primarily bound to the N/C network, and no direct bonding is observed between Fe and Cu.

Electrochemical NO3RR Performance of Fe–N–C, Cu–N–C, and FeCu–N–C Catalysts

Next, the NO3RR performance of each catalyst was compared in 0.5 M KNO3 with 1 M KOH electrolyte, and all potentials are reported versus the reversible hydrogen electrode (RHE) (Figure d–g). To evaluate the catalytic activity per metal atom of the catalyst, the atomic percentages of the three catalysts were adjusted to approximately 0.25, as confirmed by ICP-OES (Table S4). An alkaline electrolyte was employed to suppress competing HER and to achieve greater selectivity for NO3RR. The M–N–C series is known as catalysts requiring high overpotentials for HER in the alkaline condition due to the slow kinetics of forming H–H recombination on their isolated single active sites. ,

First, the activities of NO3RR and HER were estimated by linear sweep voltammetry (LSV) with and without KNO3 in the KOH electrolyte (Figure d). Notably, under electrolytes with NO3 –, the current density increases in Fe–N–C, Cu–N–C, and FeCu–N–C, suggesting a synergistic effect between Fe and Cu in NO3RR. Compared to the Fe–N–C, the Cu–N–C exhibited a distinctly smaller overpotential during the NO3RR at low current density. Meanwhile, the current density was largely suppressed in the absence of KNO3, where HER proceeds. This indicates that Cu–N–C showed poor HER kinetics, confirming the highest Tafel slope and overpotential of HER compared to Fe–N–C (Figures S19 and S20). The FeCu–N–C exhibits moderate HER activity, falling between that of Cu–N–C and Fe–N–C.

Next, for a detailed comparison of the NO3RR catalytic activities, Faradaic efficiency (FE) and yield rate for NH3 and NO2 – products were measured in the range of −0.1 to −0.5 VRHE (Figure e–g). All measurements were repeated three times independently, and the individual values for FE and yield rate at each applied potential are summarized in Tables S5–S7. In these potential regions, dramatic changes were observed in the current densities and product distribution of NO3RR while the HER remained low (Figure S21). At each potential, NH3 and NO2 – products were quantified using a colorimetric method based on UV–vis calibration curves established across a range of concentrations to eliminate matrix effects. The ammonia quantification was additionally validated by 1H nuclear magnetic resonance (1H NMR) spectroscopy (Figures S22–S29). The measurement details are provided in the Supporting Information. Comparing the activities of Fe–N–C and Cu–N–C first, the Fe–N–C showed higher FENH3 values (Figure e) than the Cu–N–C at overpotentials below −0.35 VRHE, despite having a lower NH3 yield rate (Figure f). Meanwhile, the Cu–N–C primarily generates NO2 – near the onset overpotentials, and the NH3 production increases sharply at more negative potentials, reaching a FENH3 ∼ 92% under −0.4 – –0.5 VRHE. This product trend of Cu–N–C is consistent with metallic Cu electrocatalysts, where desorption of NO2 – intermediate is feasible, and its reduction to NH3 requires a more cathodic potential (Figure g). , Therefore, the enhanced NH3 selectivity of the Fe–N–C catalyst at low potentials implies its higher activity for the hydrogenation of NO2 –. However, when the potentials were more negative than −0.35 VRHE, the NH3 FE of the Fe–N–C rather decreased. In contrast, the Cu–N–C catalyst maintained high FE for NH3 production due to its larger overpotential for HER (Figure d,e).

Notably, the FeCu–N–C catalyst maintained NH3 FE near 100% even at more negative potentials and exhibited the highest NH3 yield rate of 14.0 mg h–1 cm–2 at −0.39 VRHE, among the three catalysts, demonstrating effective HER suppression (Figure e,f). In terms of mass activity normalized to the metal loading, the catalyst achieved a remarkable value of 1180 mg h–1 mgFe+Cu . These high FE values and elevated NH3 yield rates of our FeCu–N–C catalyst demonstrate outstanding performance compared to other reported NO3RR catalysts, including M–N–C catalysts (Figure S30 and Table S8). Unlike the Cu–N–C, the FeCu–N–C also showed well-suppressed NO2 – production at low potential regions. These performance results highlight that the FeCu–N–C catalyst exhibited synergistically enhanced activity, combining the high NH3 selectivity of the Fe–N x sites with the suppression of the HER and the high NO3RR current density of the Cu–N x sites even at low overpotentials. Isotope labeling experiments with 15NO3 – confirmed that NH3 was produced exclusively via NO3RR10, as evidenced by the two distinct peaks characteristic of 15NH3 in 1H NMR spectra (Figure h). Furthermore, the FeCu–N–C catalyst maintained stable NH3 FE and yield rate over 20 h of operation at −0.39 VRHE, with the electrolyte refreshed every 2 h, confirming excellent long-term stability (Figure i).

Investigation of Structural Distortion-Induced Clustering of Cu by Operando XAS and EPR Analysis

To investigate the oxidation states and structural changes of individual Fe and Cu single atoms during NO3RR, operando XAS measurement was applied to Fe–N–C, Cu–N–C, and FeCu–N–C, respectively, under −0.26 VRHE, and −0.34 VRHE, where NH3 generation was confirmed in Figure e–g. The spectra changes were compared to the spectra obtained at open-circuit potential (OCP). First, Fe K-edge XANES spectra showed that the oxidation states of both Fe–N–C and FeCu–N–C catalysts decreased to less than Fe2+, implying involvement of Fe sites in the NO3RR by accepting electrons (Figure S31). The pre-edge feature remained invariant across potentials for both catalysts, indicating that the Fe–N x coordination framework does not undergo collapse during NO3RR. , The white line intensity, corresponding to the 1s → 4p transition, is markedly higher in FeCu–N–C than in Fe–N–C and increases progressively with more cathodic overpotential (Figure S31e). The enhancement reflects two interrelated structural effects: the curvature-driven geometric distortion of the Fe–N x moiety, which alters the local coordination, and the concurrent low-spin to high-spin transition at the Fe center, which further increases the unoccupied 4p partial DOS. The progressive increase in white line intensity with increasingly negative applied potential, corresponding to larger Cu cluster size and higher induced curvature. Meanwhile, the Fe K-edge EXAFS spectra showed that the Fe–N–C catalysts maintained a CN of approximately 4 throughout the cathodic NO3RR condition, and the unchanged coordination number is consistent with a stable pyridinic N structure (Figures a, S32 and Table S9). In contrast, the FeCu–N–C catalyst exhibited a decreasing trend in CN from 4.0 at OCP to 3.8 at −0.26 VRHE and 3.6 at −0.34 VRHE. This was accompanied by a slight increase in Fe–N bond distance from 1.99 to 2.03 Å (Figures b, S33 and Table ). These results support that the Fe–N4 structure in FeCu–N–C undergoes a distinctive structural distortion during NO3RR, leading to a reduced coordination environment, in contrast to the stable Fe–N4 geometry preserved in Fe–N–C. Nevertheless, the Fe–Fe scattering paths indicative of Fe aggregation were absent in both Fe–N–C and FeCu–N–C during the NO3RR, implying that Fe single atoms remain stable throughout the reaction (Figure a,b).

3.

3

Open in a new tab

X-ray absorption spectroscopy characterization and EPR analysis. Fe K-edge operando EXAFS spectra of (a) Fe–N–C catalyst and (b) FeCu–N–C catalyst. (c) Cu K-edge operando XANES spectra of the FeCu–N–C catalyst exhibit an increasing Cu0 state in the catalyst during NO3RR. Operando EXAFS spectra of (d) Cu–N–C catalyst and (e) FeCu–N–C catalyst also show increasing aggregated Cu clusters as the negative potential increased. (f) Perpendicular mode X-band EPR spectra of FeCu–N–C catalyst reveal the Fe high-spin state formation.

1. Structural Parameters of FeCu–N–C Catalyst Extracted from the EXAFS Fitting.

Sample Shell CN R (Å) σ2 (10–3 Å2) R factor (%)
FeCu–N–C Fe–N 4.0 ± 0.9 1.98 ± 0.02 8.7 ± 4.7 1.0
Cu–N 3.6 ± 1.2 1.93 ± 0.02 7.1 ± 3.9 0.5
FeCu–N–C OCP Fe–N 4.0 ± 0.7 1.99 ± 0.02 4.6 ± 3.1 1.5
Cu–N 3.6 ± 1.3 1.94 ± 0.03 6.0 ± 4.6 0.2
FeCu–N–C -0.26 V RHE Fe–N 3.8 ± 1.0 2.02 ± 0.03 8.5 ± 5.6 1.0
Cu–N 3.3 ± 0.7 1.91 ± 0.01 9.5 ± 3.0 0.5
Cu–Cu 3.1 ± 0.4 (8.3 ± 1.2) 2.54 ± 0.01 7.1 ± 1.2 ―
FeCu–N–C -0.34 V RHE Fe–N 3.6 ± 0.9 2.03 ± 0.02 7.6 ± 4.7 1.5
Cu–N 3.2 ± 0.5 1.89 ± 0.01 10.3 ± 2.3 0.3
Cu–Cu 4.1 ± 0.3 (8.6 ± 0.7) 2.53 ± 0.01 7.9 ± 0.7 ―
FeCu–N–C post OCP Cu–N 3.6 ± 1.1 1.96 ± 0.03 5.1 ± 4.2 1.5
Open in a new tab
a

The real coordination number was estimated using the proportion of Cu0 obtained from linear combination fitting.

Next, in operando Cu K-edge XANES, the oxidation state of Cu–N–C and FeCu–N–C catalysts decreased, and pre-edge and oscillation in Cu foil reference were observed, indicating the evolution of Cu0 state (Figures c and S34). Under further investigation, the linear combination fitting revealed that the Cu0 state increases at more negative applied potentials (Tables , S10, Figures S34 and 35). The Cu0 state does not appear in the as-prepared catalyst, and its evolution during NO3RR represents the formation of metallic Cu. These changes correspond to the Cu K-edge operando EXAFS analysis results in Figure d,e, where new Cu–Cu bonds at 2.54 Å appear only under NO3RR of −0.26 and −0.34 VRHE. The concurrent decrease in the CN of Cu–N from 3.6 at OCP to 3.3 at −0.26 VRHE and 3.2 at −0.34 VRHE supports the collapse of Cu atom sites, leading to aggregation and the formation of Cu clusters (Table , Figures S36 and S37). These changes in oxidation state and aggregation are consistent with previous studies on Cu–N–C-type catalysts for electrochemical reduction reactions. Since EXAFS represents the average ensemble of Cu sites in the entire sample, the measured CN values would contain the information on both Cu clusters and Cu single atoms. Therefore, the CN of the Cu–Cu was estimated using the proportion of Cu0 obtained from linear combination fitting (Tables , and S11, details are provided in Supporting Information Note 1). , This supports that the Cu cluster sizes used in our DFT simulations corresponded to size regimes comparable to those experimentally observed from operando XAS during NO3RR. The average CN of 8.6 at −0.34 VRHE is consistent with a Cu55-dominated cluster distribution, whereas the slightly lower value of 8.3 at −0.26 VRHE indicates a relatively greater contribution from smaller Cu38 clusters. As the Cu–Cu CN increases with more cathodic applied potentials, both NH3 FE and yield rate increase, while the NO2 – yield rate is concurrently suppressed in both FeCu–N–C and Cu–N–C catalysts (Figure S38). , These results indicate that the evolution of the Cu clusters modulates reaction kinetics, lowering the energy barrier to the *NO2 reduction step and thereby promoting the overall NO3RR to NH3 production.

2. Fractions of Cu Foil and Cu Single Atom of FeCu–N–C Catalyst Calculated by Linear Combination Fitting During NO3RR.

Sample Fraction of Cu foil(%) Fraction of Cu single atom (%) R factor (%) Reduced χ2 (%)
FeCu–N–C -0.26 V RHE 37.5 ± 1.3 62.5 ± 1.7 0.2 0.04
FeCu–N–C -0.34 V RHE 48.1 ± 1.0 51.9 ± 1.6 0.2 0.03
Open in a new tab

When the state of Cu was monitored under OCP conditions immediately after NO3RR, the Cu–Cu bonds of FeCu–N–C catalyst completely disappeared during the first scanning of operando XAS, and Cu–N CN of FeCu–N–C was restored in the electrolyte at room temperature (Figure e and Table ). It revealed that reversible redispersion of Cu single atoms occurs upon removal of the applied potential within a stable N–C framework network. Consistent results were observed, demonstrating the absence of Cu particles in FeCu–N–C when HAADF-STEM images were taken within 24 h after NO3RR at −0.34 VRHE (Figure S39). These findings reveal that Cu clusters undergo reverse reconstruction, redispersing into isolated single atoms.

Additionally, the computational results suggest that the formation of the Cu clusters coincides with distortions in adjacent Fe–N x structures and induces changes in the Fe spin-state. To probe these spin-state modifications induced by NO3RR, electron paramagnetic resonance (EPR) analysis was performed with the FeCu–N–C catalyst. The FeCu–N–C catalyst was collected from the electrode immediately after NO3RR at −0.34 VRHE, and EPR measurements were performed at 5 K. The details are provided in the Supporting Information. Figure f exhibits a distinctly strong signal at g ≈ 4.27, indicating a high-spin state of Fe, , in contrast to the FeCu–N–C before the reaction (denoted as FeCu–N–C pre). To verify that this signal originates from high-spin Fe, EPR measurements were also performed on the Cu–N–C catalyst before and after NO3RR under identical conditions (Figure S40b). The Cu–N–C spectra showed no discernible signal at g ≈ 4.27 either before or after the reaction, confirming that Cu species in the M–N–C framework do not contribute to this resonance and that the observed signal is attributable to high-spin Fe centers. To further examine whether the Fe spin-state transition is influenced by the presence of Cu in the catalyst, control EPR experiments were also conducted on the Fe–N–C catalyst under conditions identical to those applied for FeCu–N–C (Figure S40a). In contrast to the slight change observed in Fe–N–C, the FeCu–N–C catalyst exhibited a substantially more pronounced increase in the high-spin Fe signal near g ≈ 4.27 after the NO3RR condition. Because EPR signal intensity scales with the concentration of the paramagnetic species, the spectra were normalized to the Fe atomic content of each catalyst determined by ICP analysis, ensuring a quantitatively rigorous comparison. The pronounced contrast near g ≈ 4.27 between the two catalysts validates our DFT prediction that the LS → HS transition is affected by the presence of Cu. The strain imposed by operando-formed Cu clusters breaks the local D 4h symmetry of the adjacent Fe–N4 moiety and stabilizes the Fe d z 2 orbital, driving the spin-state crossover. In addition to g ≈ 4.27, the post reaction Fe–N–C spectrum (Figure S40a) exhibits weak signals near g ≈ 2.00, 2.08, and 2.15, which may be attributed to an increase in carbon-framework defects, Fe high-spin contributions, or structural perturbations induced by NO3RR conditions. − Collectively, the significant increase in the g ≈ 4.27 signal intensity observed in FeCu–N–C supports computational predictions that Cu cluster formation during NO3RR stabilizes the d z 2 orbital and drives the LS → HS transition by distorting Fe–N x sites along the z-axis, thereby accelerating the reaction kinetics by lowering the energy barrier to NH3 production.

Mechanistic Study with NO2 – Reduction Reaction and Operando ATR-SEIRAS

Next, to elucidate the enhanced reaction kinetics of the FeCu–N–C catalyst compared to Cu–N–C and Fe–N–C catalysts from a mechanistic perspective, the electrochemical NO2 – reduction reaction (NO2RR) activities were evaluated. In the NO3RR pathway, both the activation of NO3 – and the subsequent hydrogenation of NO2 – should be active, and NO2 – has been considered as a key intermediate of the rate-determining step, particularly at Cu-based catalysts. Therefore, the NO2RR experiment can be used to understand the relative kinetic changes across reaction steps or the synergistic effect of tandem catalysts with dual active sites in NO3RR. In contrast to NO3RR, Figure a shows that Fe–N–C exhibited the highest NH3 production from NO2RR, without any apparent synergistic effect between Fe and Cu in FeCu–N–C, which displayed decreased NH3 yield rate probably due to the lower Fe content (Figure S41). This evidence that Fe–N x sites possess intrinsically higher NO2RR kinetics than Cu. The NO2 conversion to NH3 involves deoxygenation and hydrogenation steps, requiring multiple hydrogen adsorption (Hads) from water dissociation under alkaline conditions. This behavior is consistent with the higher HER activity of the Fe–N–C catalyst compared to the Cu–N–C catalyst (Figure S19). Then, Fe–N x sites can be more active in water dissociation than Cu, producing more Hads and thereby facilitating the hydrogenation of NO2 to NH3. A direct comparison of NO2RR and NO3RR performance between Fe–N–C and Cu–N–C (Figure b) reveals that, although Cu–N–C exhibits noticeably inferior NO2RR kinetics, it demonstrates higher reactivity in NO3RR. Cu–N–C produces substantial amounts of NO2 – at low potentials, highlighting its greater ability to activate NO3 –. Meanwhile, considering that Fe–N–C in NO3RR showed lower NH3 yield at low potentials despite high NH3 FE, this can be due to sluggish activation of NO3 – in the initial reaction step. These findings underscore a mechanistic distinction: Cu–N–C facilitates the initial activation of NO3 – to NO2 –, whereas Fe–N–C synergistically accelerates the subsequent hydrogenation of NO2 – to NH3 (Figure c), consistent with the DFT simulation results. Therefore, the FeCu–N–C tandem catalyst, which integrates both Cu and Fe active sites, leverages this complementary functionality to achieve enhanced NH3 FE and yield in NO3RR.

4.

4

Open in a new tab

NO3RR mechanism analysis. (a) NH3 yield rate of Fe–N–C, FeCu–N–C and Cu–N–C in NO2RR. (b) The sum of NH3 and NO2 – yield rate of Fe–N–C and Cu–N–C catalysts in NO2RR and NO3RR. (c) The schematic illustration of the tandem reaction of the FeCu–N–C catalyst. Operando ATR-SEIRAS of (d) Cu–N–C, (e) FeCu–N–C, and (f) Fe–N–C catalysts at various applied potentials under NO3RR conditions. Gray vertical lines from left to right indicate δO–H (∼1630 cm–1), νaNO3 – (1390 cm–1), νsNO3 – (1340 cm–1), and νs*NO2 (1300 cm–1), respectively. Colored regions corresponding to the graph colors highlight the areas where peaks of asymmetric stretching in nitro groups, νa*NO2 (1465 cm–1), were detected. The gray spectra, acquired under OCP conditions following the potential step, were employed to assess the persistence of *NO2.

To verify whether this tandem effect is accentuated in FeCu–N–C, we evaluated the NO3RR performance with the catalyst that physically mixes Fe–N–C and Cu–N–C (P.M FeCu–N–C; Figure S42). If the NO2 – intermediate desorbed from the Cu–N–C surface can migrate and adsorb onto the Fe–N–C surface during NO3RR, a tandem effect is also expected in the P.M FeCu–N–C. Indeed, the mixed catalyst exhibited improved NO3RR activity compared to either Cu–N–C or Fe–N–C alone. However, the FeCu–N–C catalyst still outperformed the P.M FeCu–N–C catalyst, which can be attributed to two primary factors. First, computational analysis, XPS, and XAS revealed that Cu aggregation in FeCu–N–C induces distortion in the adjacent Fe–N x structure, modulating the spin-state and bonding structures of Fe, thereby enhancing its intrinsic catalytic activity. Second, the intermediate spillover of NO2 – is more effective in the FeCu–N–C catalyst than in P.M FeCu–N–C, owing to the closer spatial proximity of Fe and Cu atoms achieved through their co-introduction during synthesis. The spillover efficiency from Cu to Fe sites was estimated using a mass-balance approach based on TOF values (Table S12), which showed the higher spillover efficiencies of 95–100% on the integrated FeCu–N–C catalyst than on P.M FeCu–N–C (i.e., 70–80%). The details are provided in the Supporting Information Note 2. The reduced atomic distance between Fe and Cu significantly pronounces both effects. As a result, the FeCu–N–C catalyst, hosting colocalized dual active sites, exhibits superior NO3RR performance relative to either single-atom catalysts, by enabling a more efficient tandem pathway.

To gain mechanistic insight into the dynamic behavior of *NO2 intermediates on Cu and Fe sites in the tandem pathway, operando ATR-SEIRAS was performed (Figure d–f). ATR-SEIRAS measurements were conducted in 0.01 M KNO3 + 1 M KOH electrolyte, as 0.5 M KNO3 condition resulted in dominant NO3 – consumption signals that obscured *NO2-related features (Figure S43). The absorption spectra exhibited characteristic peaks corresponding to various molecular vibrations: Si–O stretching (νSi–O, 1200 cm–1), symmetric and asymmetric stretching in nitro groups (νs*NO2, 1300 cm–1 and νa*NO2, 1465 cm–1), symmetric and asymmetric vibrations in nitrate groups (νsNO3 –, 1340 cm–1 and νaNO3 –, 1390 cm–1), O–H bending and stretching in H2O (δO–H, ∼ 1630 cm–1 and νO–H, ∼ 3450 cm–1), and N–H stretching (νN–H, 3700 cm–1), across the entire spectral range being examined (Figure S44). To confirm the origin of nitro-containing peaks, a series of control experiments was conducted (Figure S45). When the electrolyte was replaced with a 1 M KOH-only solution, all peaks associated with nitrogen-containing oxide species disappeared, confirming that the observed features originate from NO3 –-derived species. A NO2 – reduction experiment further demonstrated that the 1465 cm–1 peak corresponds to a consumption feature of NO2 –, consistent with its assignment to the asymmetric stretching mode of the adsorbed nitro species, νa*NO2. Additionally, KOD/D2O substitution showed that the 1465 cm–1 peak persists, confirming that this feature is attributable to an H-independent nitrogen–oxygen vibration. Furthermore, an isotope-labeled experiment using K15NO3 showed that a νa*15NO2 peak was observed at around 1424 cm–1, corresponding to a redshift of ∼41 cm–1 relative to the νa*14NO2 peak. , These results collectively validate the assignment of the 1465 cm–1 peak to νa*NO2, originating from the NO3RR pathway.

With the increase in negative applied potential, NO3 – consumption was evidenced by the emergence of negative bands, whereas the formation of *NO2 was indicated by the appearance of positive bands. The onset potential of Fe–N–C was observed to be more negative, while Cu–N–C and FeCu–N–C exhibited similarly positive values. The trends in current density observed during ATR-SEIRAS measurements followed the same order as the electrochemical NO3RR activity: FeCu–N–C > Cu–N–C > Fe–N–C (Figure S46). A notable difference was observed in the behavior of the νa*NO2 band in the OCP conditions after NO3RR. In the Cu–N–C catalyst, the νa*NO2 peak intensity decreased (Figure e), suggesting that νa*NO2 formed on Cu sites is relatively unstable and readily desorbs in the absence of applied NO3RR potentials. In contrast, the Fe–N–C catalyst retained its νa*NO2 peak (Figure d), indicating a stronger interaction and more stable adsorption of *NO2 at Fe sites. It supports the earlier electrochemical results, where Cu–N–C facilitated rapid NO3 – activation but showed limited NO2 – retention, while Fe–N–C exhibited stronger NO2 – adsorption and enhanced NO2RR activity. The FeCu–N–C catalyst exhibited a combined behavior (Figure f), showing both a low onset potential comparable to Cu–N–C and a sustained νa*NO2 signal under OCP conditions similar to Fe–N–C. These operando ATR-SEIRAS results evidence the cooperative interaction between Cu and Fe single atoms, accelerating overall a multistep reduction pathway to NH3, accounting for the superior FEs and yield rate of FeCu–N–C compared to its single-metal catalyst.

Conclusion

This study elucidates the dynamic structural and electronic interplay between Cu clustering and Fe spin-state modulation in FeCu–N–C catalysts during NO3RR. Through integrated computational and operando experimental approaches, we demonstrate that electrochemically induced Cu cluster formation serves a dual function: directly catalyzing NO3 – dissociation while simultaneously introducing structural strain that modulates the electronic configuration of adjacent Fe–N4 sites. This strain-induced spin-state transition to high-spin Fe enhances binding affinity for NO2 – intermediates and facilitates subsequent reduction steps toward NH3. These structural transformations occur under electrochemical cathodic conditions, highlighting the importance of understanding catalyst evolution beyond initial states. NO2RR and operando ATR-SEIRAS measurements additionally demonstrated the excellent performance of Cu sites in predominantly activating NO3 – at low potentials, while Fe sites excelled in converting NO2 to NH3. Therefore, FeCu–N–C catalyst achieved remarkable activities, recording ∼100% NH3 Faradaic efficiency and a high NH3 yield rate of 14.0 mg h–1 cm–2 at −0.39 VRHE in NO3RR. This approach of leveraging in situ structural distortions opens promising strategies for developing multicomponent electrocatalysts with enhanced selectivity and efficiency to optimize multistep electrochemical processes via tandem reactions.

Supplementary Material

ja6c00705_si_001.pdf (6.1MB, pdf)

Acknowledgments

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2021-NR060090, RS-2025-00515642, and RS-2024-00435493). The authors acknowledge Sun Hee Kim, Wooyeol Ryu, and Donghwa Lee at the Western Seoul Center, Korea Basic Science Institute (KBSI), for their assistance with the EPR measurements. The authors also acknowledge financial support from the New Faculty Startup Fund and the Creative-Pioneering Researchers Program at Seoul National University.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.6c00705.

  • Computational data; XRD pattern; TEM images; HAADF-STEM images; EDS mapping data; XPS spectra; ICP-OES data; Catalytic activity data; UV–vis spectra; NMR data; ex situ/in situ XAS and fitting data; EPR data; Comparison of NO3RR to NO2RR electrocatalytic activity and selectivity; ATR-SEIRAS analysis (PDF)

#.

Department of Chemistry, College of Sciences, Kyung Hee University, Seoul 02447, Republic of Korea

∥.

S.H., J.J., and E.Y. contributed equally to this work.

The authors declare no competing financial interest.

References

  1. van Langevelde P. H., Katsounaros I., Koper M. T.. Electrocatalytic nitrate reduction for sustainable ammonia production. Joule. 2021;5(2):290–294. doi: 10.1016/j.joule.2020.12.025. [DOI] [Google Scholar]
  2. Xu H., Ma Y., Chen J., Zhang W. X., Yang J.. Electrocatalytic reduction of nitrate - a step towards a sustainable nitrogen cycle. Chem. Soc. Rev. 2022;51(7):2710–2758. doi: 10.1039/D1CS00857A. [DOI] [PubMed] [Google Scholar]
  3. He W., Zhang J., Dieckhöfer S., Varhade S., Brix A. C., Lielpetere A., Seisel S., Junqueira J. R., Schuhmann W.. Splicing the active phases of copper/cobalt-based catalysts achieves high-rate tandem electroreduction of nitrate to ammonia. Nat. Commun. 2022;13(1):1129. doi: 10.1038/s41467-022-28728-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Zhang G., Li X., Chen K., Guo Y., Ma D., Chu K.. Tandem electrocatalytic nitrate reduction to ammonia on MBenes. Angew. Chem., Int. Ed. 2023;135(13):e202300054. doi: 10.1002/anie.202300054. [DOI] [PubMed] [Google Scholar]
  5. Murphy E., Liu Y., Matanovic I., Guo S., Tieu P., Huang Y., Ly A., Das S., Zenyuk I., Pan X.. et al. Highly durable and selective Fe-and Mo-based atomically dispersed electrocatalysts for nitrate reduction to ammonia via distinct and synergized NO2-pathways. ACS Catal. 2022;12(11):6651–6662. doi: 10.1021/acscatal.2c01367. [DOI] [Google Scholar]
  6. Jung W., Jeong J., Chae Y., Lee W. H., Ko Y.-J., Chae K. H., Oh H.-s., Lee U., Lee D. K., Min B. K.. et al. Synergistic bimetallic CuPd oxide alloy electrocatalyst for ammonia production from the electrochemical nitrate reaction. J. Mater. Chem. A. 2022;10(44):23760–23769. doi: 10.1039/d2ta06316f. [DOI] [Google Scholar]
  7. Wu X., Zhang H., Zuo S., Dong J., Li Y., Zhang J., Han Y.. Engineering the coordination sphere of isolated active sites to explore the intrinsic activity in single-atom catalysts. Nanomicro Lett. 2021;13(1):136. doi: 10.1007/s40820-021-00668-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Kim J. H., Jang H., Bak G., Choi W., Yun H., Lee E., Kim D., Kim J., Lee S. Y., Hwang Y. J.. The insensitive cation effect on a single atom Ni catalyst allows selective electrochemical conversion of captured CO 2 in universal media. Energy Environ. Sci. 2022;15(10):4301–4312. doi: 10.1039/D2EE01825J. [DOI] [Google Scholar]
  9. Zhang W.-D., Dong H., Zhou L., Xu H., Wang H.-R., Yan X., Jiang Y., Zhang J., Gu Z.-G.. Fe single-atom catalysts with pre-organized coordination structure for efficient electrochemical nitrate reduction to ammonia. Appl. Catal. B: Environ. 2022;317:121750. doi: 10.1016/j.apcatb.2022.121750. [DOI] [Google Scholar]
  10. Wu Z.-Y., Karamad M., Yong X., Huang Q., Cullen D. A., Zhu P., Xia C., Xiao Q., Shakouri M., Chen F.-Y.. et al. Electrochemical ammonia synthesis via nitrate reduction on Fe single atom catalyst. Nat. Commun. 2021;12(1):2870. doi: 10.1038/s41467-021-23115-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kim D. H., Ringe S., Kim H., Kim S., Kim B., Bae G., Oh H.-S., Jaouen F., Kim W., Kim H.. et al. Selective electrochemical reduction of nitric oxide to hydroxylamine by atomically dispersed iron catalyst. Nat. Commun. 2021;12(1):1856. doi: 10.1038/s41467-021-22147-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Tan Y., Fu J., Luo T., Liu K., Liu M.. Theoretical Insights into the Selectivity of Single-Atom Fe-N-C Catalysts for Electrochemical NO x Reduction. J. Am. Chem. Soc. 2025;147(6):4937–4944. doi: 10.1021/jacs.4c14021. [DOI] [PubMed] [Google Scholar]
  13. Zhao J., Liu L., Yang Y., Liu D., Peng X., Liang S., Jiang L.. Insights into electrocatalytic nitrate reduction to ammonia via Cu-based bimetallic catalysts. ACS Sustain. Chem. Eng. 2023;11(6):2468–2475. doi: 10.1021/acssuschemeng.2c06498. [DOI] [Google Scholar]
  14. Zhang S., Wu J., Zheng M., Jin X., Shen Z., Li Z., Wang Y., Wang Q., Wang X., Wei H.. et al. Fe/Cu diatomic catalysts for electrochemical nitrate reduction to ammonia. Nat. Commun. 2023;14(1):3634. doi: 10.1038/s41467-023-39366-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Zhu T., Chen Q., Liao P., Duan W., Liang S., Yan Z., Feng C.. Single-atom Cu catalysts for enhanced electrocatalytic nitrate reduction with significant alleviation of nitrite production. Small. 2020;16(49):2004526. doi: 10.1002/smll.202004526. [DOI] [PubMed] [Google Scholar]
  16. Zhao X., Geng Q., Dong F., Zhao K., Chen S., Yu H., Quan X.. Boosting the selectivity and efficiency of nitrate reduction to ammonia with a single-atom Cu electrocatalyst. Chem. Eng. J. 2023;466:143314. doi: 10.1016/j.cej.2023.143314. [DOI] [Google Scholar]
  17. Yang J., Qi H., Li A., Liu X., Yang X., Zhang S., Zhao Q., Jiang Q., Su Y., Zhang L.. et al. Potential-driven restructuring of Cu single atoms to nanoparticles for boosting the electrochemical reduction of nitrate to ammonia. J. Am. Chem. Soc. 2022;144(27):12062–12071. doi: 10.1021/jacs.2c02262. [DOI] [PubMed] [Google Scholar]
  18. Ding J., Liu L., Zhang J., Liu Y., Xu H., Shen Z., Yang H. B., Feng X., Huang Y., Liu B.. Unraveling dynamic structural evolution of single atom catalyst via in situ surface-enhanced infrared absorption spectroscopy. J. Am. Chem. Soc. 2025;147(11):9601–9609. doi: 10.1021/jacs.4c17565. [DOI] [PubMed] [Google Scholar]
  19. Jahangirzadeh Varjovi M., Tosoni S.. DFT Investigation of X 55 (X= Ni, Pd, and Pt) Clusters on Ultrathin Supported MgO Films: Evidence of Oxygen Spillover and Relevance for Catalytic Model Studies. J. Phys. Chem. C. 2024;128(50):21331–21342. doi: 10.1021/acs.jpcc.4c04213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. He Y., Zhang S., Zheng Z., Li G.. Investigating the influence of substrate orientation and temperature on Cu cluster deposition. J. Appl. Phys. 2024;135(10):105302. doi: 10.1063/5.0193758. [DOI] [Google Scholar]
  21. Ding J., Wei Z., Li F., Zhang J., Zhang Q., Zhou J., Wang W., Liu Y., Zhang Z., Su X.. et al. Atomic high-spin cobalt (II) center for highly selective electrochemical CO reduction to CH3OH. Nat. Commun. 2023;14(1):6550. doi: 10.1038/s41467-023-42307-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Zhao C., Dai X., Yao T., Chen W., Wang X., Wang J., Yang J., Wei S., Wu Y., Li Y.. Ionic exchange of metal-organic frameworks to access single nickel sites for efficient electroreduction of CO2. J. Am. Chem. Soc. 2017;139(24):8078–8081. doi: 10.1021/jacs.7b02736. [DOI] [PubMed] [Google Scholar]
  23. Zeng X., Shui J., Liu X., Liu Q., Li Y., Shang J., Zheng L., Yu R.. Single-Atom to Single-Atom Grafting of Pt1 onto Fe-N4 Center: Pt1@ Fe-N-C Multifunctional Electrocatalyst with Significantly Enhanced Properties. Adv. Energy Mater. 2018;8(1):1701345. doi: 10.1002/aenm.201701345. [DOI] [Google Scholar]
  24. Zhou D., Yang L., Yu L., Kong J., Yao X., Liu W., Xu Z., Lu X.. Fe/N/C hollow nanospheres by Fe (III)-dopamine complexation-assisted one-pot doping as nonprecious-metal electrocatalysts for oxygen reduction. Nanoscale. 2015;7(4):1501–1509. doi: 10.1039/C4NR06366J. [DOI] [PubMed] [Google Scholar]
  25. Martin L., Martinez H., Poinot D., Pecquenard B., Le Cras F.. Comprehensive X-ray photoelectron spectroscopy study of the conversion reaction mechanism of CuO in lithiated thin film electrodes. J. Phys. Chem. C. 2013;117(9):4421–4430. doi: 10.1021/jp3119633. [DOI] [Google Scholar]
  26. Lazar P., Mach R., Otyepka M.. Spectroscopic fingerprints of graphitic, pyrrolic, pyridinic, and chemisorbed nitrogen in N-doped graphene. J. Phys. Chem. C. 2019;123(16):10695–10702. doi: 10.1021/acs.jpcc.9b02163. [DOI] [Google Scholar]
  27. Matanovic I., Artyushkova K., Strand M. B., Dzara M. J., Pylypenko S., Atanassov P.. Core level shifts of hydrogenated pyridinic and pyrrolic nitrogen in the nitrogen-containing graphene-based electrocatalysts: in-plane vs edge defects. J. Phys. Chem. C. 2016;120(51):29225–29232. doi: 10.1021/acs.jpcc.6b09778. [DOI] [Google Scholar]
  28. Mahmood N., Yao Y., Zhang J. W., Pan L., Zhang X., Zou J. J.. Electrocatalysts for hydrogen evolution in alkaline electrolytes: mechanisms, challenges, and prospective solutions. Adv. Sci. 2018;5(2):1700464. doi: 10.1002/advs.201700464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hong S., Kim J. H., Shin D., Bak G., Jang D., Kim W. B., Hwang Y. J.. Synergistic effect of Pt-loaded CoNC electrocatalysts for hydrogen evolution reaction in alkaline conditions. Appl. Surf. Sci. 2023;610:155523. doi: 10.1016/j.apsusc.2022.155523. [DOI] [Google Scholar]
  30. Wei M., Li S., Wang X., Zuo G., Wang H., Meng X., Wang J.. A Perspective on Cu-Based Electrocatalysts for Nitrate Reduction for Ammonia Synthesis. Adv. Energy Sust. Res. 2024;5(1):2300173. doi: 10.1002/aesr.202300173. [DOI] [Google Scholar]
  31. Khalil M., Marcus M. A., Smeigh A. L., McCusker J. K., Chong H. H., Schoenlein R. W.. Picosecond X-ray absorption spectroscopy of a photoinduced iron (II) spin crossover reaction in solution. J. Phys. Chem. A. 2006;110(1):38–44. doi: 10.1021/jp055002q. [DOI] [PubMed] [Google Scholar]
  32. Zhang P., Chen H.-C., Zhu H., Chen K., Li T., Zhao Y., Li J., Hu R., Huang S., Zhu W.. et al. Inter-site structural heterogeneity induction of single atom Fe catalysts for robust oxygen reduction. Nat. Commun. 2024;15(1):2062. doi: 10.1038/s41467-024-46389-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Han G., Zhang X., Liu W., Zhang Q., Wang Z., Cheng J., Yao T., Gu L., Du C., Gao Y.. et al. Substrate strain tunes operando geometric distortion and oxygen reduction activity of CuN2C2 single-atom sites. Nat. Commun. 2021;12(1):6335. doi: 10.1038/s41467-021-26747-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lee J.-J., Sheu H.-s., Lee C.-R., Chen J.-M., Lee J.-F., Wang C.-C., Huang C.-H., Wang Y.. X-ray absorption spectroscopic studies on light-induced excited spin state trapping of an Fe (II) complex. J. Am. Chem. Soc. 2000;122(24):5742–5747. doi: 10.1021/ja9943290. [DOI] [Google Scholar]
  35. Ma L., Hu W., Mei B., Liu H., Yuan B., Zang J., Chen T., Zou L., Zou Z., Yang B.. et al. Covalent triazine framework confined copper catalysts for selective electrochemical CO2 reduction: operando diagnosis of active sites. ACS Catal. 2020;10(8):4534–4542. doi: 10.1021/acscatal.0c00243. [DOI] [Google Scholar]
  36. Xu H., Rebollar D., He H., Chong L., Liu Y., Liu C., Sun C.-J., Li T., Muntean J. V., Winans R. E.. et al. Highly selective electrocatalytic CO2 reduction to ethanol by metallic clusters dynamically formed from atomically dispersed copper. Nat. Energy. 2020;5(8):623–632. doi: 10.1038/s41560-020-0666-x. [DOI] [Google Scholar]
  37. Calvin, S. Fingerprinting, Linear Combination Analysis. In XAFS for Everyone: A Practical Guide to X-ray Absorption Fine Structure Spectroscopy; CRC Press, 2013. [Google Scholar]
  38. Weng Z., Wu Y., Wang M., Jiang J., Yang K., Huo S., Wang X.-F., Ma Q., Brudvig G. W., Batista V. S.. et al. Active sites of copper-complex catalytic materials for electrochemical carbon dioxide reduction. Nat. Commun. 2018;9(1):415. doi: 10.1038/s41467-018-02819-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Benfield R. E.. Mean coordination numbers and the non-metal-metal transition in clusters. J. Chem. Soc., Faraday Trans. 1992;88(8):1107–1110. doi: 10.1039/FT9928801107. [DOI] [Google Scholar]
  40. Xu Y.-T., Xie M.-Y., Zhong H., Cao Y.. In situ clustering of single-atom copper precatalysts in a metal-organic framework for efficient electrocatalytic nitrate-to-ammonia reduction. ACS Catal. 2022;12(14):8698–8706. doi: 10.1021/acscatal.2c02033. [DOI] [Google Scholar]
  41. Mineva T., Matanovic I., Atanassov P., Sougrati M.-T., Stievano L., Clémancey M., Kochem A., Latour J.-M., Jaouen F.. Understanding active sites in pyrolyzed Fe-N-C catalysts for fuel cell cathodes by bridging density functional theory calculations and 57Fe Mossbauer spectroscopy. ACS Catal. 2019;9(10):9359–9371. doi: 10.1021/acscatal.9b02586. [DOI] [Google Scholar]
  42. Li X., Cao C.-S., Hung S.-F., Lu Y.-R., Cai W., Rykov A. I., Miao S., Xi S., Yang H., Hu Z.. et al. Identification of the electronic and structural dynamics of catalytic centers in single-Fe-atom material. Chem. 2020;6(12):3440–3454. doi: 10.1016/j.chempr.2020.10.027. [DOI] [Google Scholar]
  43. Petasis, D. T. ; Hendrich, M. P. . Quantitative interpretation of multifrequency multimode EPR spectra of metal containing proteins, enzymes, and biomimetic complexes. In Electron Paramagnetic Resonance Investigations of Biological Systems by Using Spin Labels, Spin Probes, and Intrinsic Metal Ions, Part A; Elsevier, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wang N., Meng C., Wang B., Tan X., Wan Y., Yang Y., Kong D., Wang W., Cao F., Fielding A. J.. et al. Spin-regulated Fe-NC catalyst enabled by adjusting coordination nitrogen species for robust oxygen reduction. Natl. Sci. Rev. 2025;12(7):nwaf061. doi: 10.1093/nsr/nwaf061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Saha P., Barman S., Amanullah S., Dey A.. Selective electrocatalytic reduction of NO to NH3 by iron porphyrins at physiologically relevant potentials. ACS Catal. 2023;13(20):13181–13194. doi: 10.1021/acscatal.3c02672. [DOI] [Google Scholar]
  46. Ni L., Gallenkamp C., Paul S., Kübler M., Theis P., Chabbra S., Hofmann K., Bill E., Schnegg A., Albert B.. et al. Active site identification in FeNC catalysts and their assignment to the oxygen reduction reaction pathway by in situ 57Fe Mössbauer spectroscopy. Adv. Energy Sust. Res. 2021;2(2):2000064. doi: 10.1002/aesr.202000064. [DOI] [Google Scholar]
  47. Wang R., Zhang L., Shan J., Yang Y., Lee J. F., Chen T. Y., Mao J., Zhao Y., Yang L., Hu Z.. et al. Tuning Fe spin moment in Fe-N-C catalysts to climb the activity volcano via a local geometric distortion strategy. Adv. Sci. 2022;9(31):2203917. doi: 10.1002/advs.202203917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Gao J., Zhang H., Guo X., Luo J., Zakeeruddin S. M., Ren D., Grätzel M.. Selective C-C coupling in carbon dioxide electroreduction via efficient spillover of intermediates as supported by operando Raman spectroscopy. J. Am. Chem. Soc. 2019;141(47):18704–18714. doi: 10.1021/jacs.9b07415. [DOI] [PubMed] [Google Scholar]
  49. Schumacher H., Künzelmann U., Vasilev B., Eichhorn K.-J., Bartha J. W.. Applications of microstructured silicon wafers as internal reflection elements in attenuated total reflection Fourier transform infrared spectroscopy. Appl. Spectrosc. 2010;64(9):1022–1027. doi: 10.1366/000370210792434404. [DOI] [PubMed] [Google Scholar]
  50. Rima F. R., Nakata K., Shimazu K., Osawa M.. Surface-enhanced infrared absorption spectroscopic studies of adsorbed nitrate, nitric oxide, and related compounds. 3. Formation and reduction of adsorbed nitrite at a platinum electrode. J. Phys. Chem. C. 2010;114(13):6011–6018. doi: 10.1021/jp911027q. [DOI] [Google Scholar]
  51. Pérez-Gallent E., Figueiredo M. C., Katsounaros I., Koper M. T.. Electrocatalytic reduction of Nitrate on Copper single crystals in acidic and alkaline solutions. Electrochim. Acta. 2017;227:77–84. doi: 10.1016/j.electacta.2016.12.147. [DOI] [Google Scholar]
  52. Fan K., Xie W., Li J., Sun Y., Xu P., Tang Y., Li Z., Shao M.. Active hydrogen boosts electrochemical nitrate reduction to ammonia. Nat. Commun. 2022;13(1):7958. doi: 10.1038/s41467-022-35664-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Liu L., Xiao T., Fu H., Chen Z., Qu X., Zheng S.. Construction and identification of highly active single-atom Fe1-NC catalytic site for electrocatalytic nitrate reduction. Appl. Catal. B: Environ. 2023;323:122181. doi: 10.1016/j.apcatb.2022.122181. [DOI] [Google Scholar]
  54. Chen S., Fan X., Yan S., McKee M., Terry A., Gao C., Abdolmaleki M., Heise J., Chen M., Kayser Y.. et al. Partial-coverage assembly of graphdiyne-derived fragment-protected Cu (I) clusters generates an ordered single-metal site catalyst. Natl. Sci. Rev. 2026;13(3):nwaf575. doi: 10.1093/nsr/nwaf575. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja6c00705_si_001.pdf (6.1MB, pdf)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES