Asymmetrically coordinated single-atom catalysts: from synthetic strategy to structure-activity relationship

Single-metal asymmetric coordination structures

Single-metal asymmetric coordination structures are constructed on the basis of the conventional M–N₄ coordination motif by introducing additional heteroatoms either within the M–N₄ equatorial plane or along the axial direction of the active center, and by tuning the coordination number of the active site. Such modifications can modulate the charge distribution at the catalytic site, the d-band center, and the adsorption state, thereby enhancing catalytic activity. Employing this strategy to improve the overall performance of SACs has emerged as a highly effective and broadly applicable approach.

Based on this, Wan et al. encapsulated triphenylphosphine (PPh₃) within a Zn/Co bimetallic metal–organic framework (MOF) cage and pyrolyzed it under argon at 950 °C to obtain a Co-SA/P catalyst bearing Co₁-P₁N₃ active sites [Figure 1A]^[54]. During the pyrolysis process, the metal Zn in the precursor evaporates and forms isolated Co atomic sites on the carbon support. ZIF-67 was used as the precursor to encapsulate PPh₃, and the Co₁-P₁N₃ site-containing Co particles/Co₁-P₁N₃ composite catalyst was obtained by pyrolysis at 900 °C under a nitrogen atmosphere. The approach of generating single atoms through vacancies induced by Zn evaporation at elevated temperatures is widely applicable. Li et al. synthesized an Fe-N/O-C catalyst composed of Fe-N₃O and adjacent defects^[55]. Formamide, as both a reactant and solvent, serves as a nitrogen and carbon source. Fe salts, Zn salts, and ethylene glycol were mixed and subjected to hydrothermal reaction at 180 °C to obtain a single-atom precursor (Fe/Zn-N/O-C) with metal and N, O coordination structures, which was then pyrolyzed to evaporate Zn atoms and obtain the Fe-N/O-C catalyst. The authors pointed out that formamide molecules could undergo condensation via the Schiff base reaction and be used to chelate metal cations, thereby anchoring metal ions to form single atoms [Figure 1B]. Moreover, employing C₃N₄ as a support for constructing these catalysts represents an effective strategy. Li et al. employed chemical vapor deposition to fabricate an Fe–N/S–C SAC on S-doped C₃N₄^[56]. A melamine/cysteine precursor was carbonized at 600 °C under N₂ to produce N,S-doped C₃N₄ nanosheets rich in carbon defects and C-S-C motifs. Subsequent pyrolysis with ferrocene (20:1 mass ratio) at 900 °C under N₂ anchored Fe atoms at these defect sites, yielding stabilized Fe–N/S active sites and suppressing iron carbide formation [Figure 1C].

Figure 1. (A) Synthesis and morphological characterizations of Co-SA/P-in situ^[54]. Copyright 2020, American Chemical Society; (B) Schematic diagram of synthesized Fe-N/O-C^[55]. Copyright 2024, Wiley-VCH; (C) Typical CVD procedures to synthesize the Fe(Fc)-N/S-C catalyst^[56]. Copyright 2021, American Chemical Society; (D) Schematic illustration of the synthesis of the Fe₁-N₄SO₂/NC catalyst^[57]. Copyright 2025, American Chemical Society; (E) Schematic illustration and characterization of the Fe-N/Cl-C catalyst synthesized from HDPE^[58]. Copyright 2025, Wiley-VCH; (F) Schematic diagram of the synthesis route for SACs-Mn-1000@g-C₃N₄^[60]. Copyright 2021, Cell Press. SA: Single-atom; CVD: chemical vapor deposition; HDPE: high-density polyethylene; SACs: single-atom catalysts.

Furthermore, introducing axially coordinated non-metal atoms into the traditional M–N₄ structures, as well as adjusting the central metal’s coordination number to form M–N_4-x structures, are both key approaches to constructing single-atom asymmetric coordination architectures. A brief overview of their synthesis methods is provided below.

In the preparation of catalysts with axial coordination, Qu et al. employed potassium thiocyanate (KSCN) as the sulfur-oxygen dopant, impregnated it onto a FeCp₂@ZIF-8 precursor via ultrasonication and stirring in methanol, and then pyrolyzed the dried composite under N₂ at 950 °C for 3 h to yield Fe₁-N₄SO₂/NC featuring atomically dispersed Fe₁-N₄SO₂ coordination sites on a carbon matrix [Figure 1D]^[57]. In addition, Ren et al. impregnated high-density polyethylene (HDPE) with FeCl₃, pyrolyzed under NH₃ at 800 °C for 3 h (5 °C·min^-1), then acid-etched in HCl to remove residual iron, yielding graphene-supported Fe-N/Cl-C with atomically dispersed FeN₄Cl sites [Figure 1E]^[58]. Besides, modulating the coordination number between active centers and nitrogen atoms offers an effective route to asymmetric coordination structures. Yan et al. precipitated a glycine–MgCl₂ complex, mixed it with Co(OAc)₂, pyrolyzed under N₂ at 800 °C for 2 h, then HCl-etched to remove MgO/Co particles, yielding carbon-supported Co-N₅C with atomically dispersed CoN₅ sites^[59]. In 2021, Liu et al. hydrothermally self-assembled graphene oxide nanoribbons with FeCl₃ into a 3D hydrogel, freeze-dried, then acid-leached and calcined to yield eFe-N₃/PCF - a porous carbon framework with edge-hosted Fe-N₃ sites, offering high intrinsic activity, fast mass transport, and efficient atom utilization [Figure 1F]^[60]. Qin et al. mixed Mn-ZIF-8 with g-C₃N₄ in ethanol, then subjected the mixture to pyrolysis, acid leaching, and a brief reheating to yield SACs-Mn-1000@g-C₃N₄, a graphene-like, hierarchically porous framework bearing Mn–N₃ sites anchored at defect sites^[61].

Non-contact multi-metal asymmetric coordination structures

Non-contact coordination structures differ from randomly dispersed single-atom structures. In non-contact coordination structures, although the metal atoms do not directly contact each other, the spacing between them needs to be sufficiently small (< 15 Å). Two metal atoms with sufficiently small spacing can interact to achieve the desired catalytic performance. Therefore, non-contact multi-metal asymmetric coordination structures have attracted extensive attention from researchers.

As reported by Zhang et al., a quasi-double-star Ni₇/Fe₃-N-C catalyst was produced via one-pot solvothermal incorporation of Ni and Fe into Zn-IRMOF-3, followed by Ar-atmosphere pyrolysis at 950 °C for 2 h, yielding a carbon-supported material with atomically dispersed adjacent Ni and Fe SA sites^[62]. Han et al. encapsulated Pt(acac)₂ and Fe(acac)₃ in ZIF-8 and obtained Fe-N₄/Pt-N₄@NC by pyrolyzing the precursor^[63]. Density functional theory (DFT) calculations showed that adjacent Pt-N₄ sites can effectively modulate the 3d electronic orbitals of Fe-N₄ sites, resulting in enhanced catalytic performance [Figure 2A]. Building on this concept of adjacent-site electronic modulation, Zhao et al. developed a low-temperature pyrolysis of diethylenetriaminepentaacetic-acid bimetallic complexes to produce dynamically stable carbon dots with embedded bimetallic atomic sites (DMASs-CDs)^[64]. As shown in Figure 2B, more than twenty DMASs-CDs incorporating Fe, Co, Ni, Mn, Zn, Cu, and Mo pairings were prepared. Notably, NiMn-CDs exhibited superior intrinsic activity in the urea oxidation probe reaction. By adjusting DMAS combinations, this method enables precise, target-oriented design of catalysts for various electrochemical processes. Remarkably, non-noble transition metals such as Co, Cu and Ni also furnish exceptional catalytically active sites. As shown in Figure 2C, the bifunctional oxygen electrocatalyst (Cu–Co/NC), comprising copper–cobalt dual-atom sites with optimized geometric and electronic structures on a high-porosity nitrogen-doped carbon (NC) support, was prepared by Li et al.^[65]. Specifically, after introducing copper acetate into Zn-Co MOF, they coated it with a polymer rich in nitrogen sources and then pyrolyzed it in an inert atmosphere at high temperatures to obtain the target catalyst. Experimental and theoretical results showed that Cu and Co atoms exist as atom pairs on the carbon-nitrogen support. Both Cu and Co formed M-N₄ coordination structures, with Cu in a +1 oxidation state and Co in a +2 oxidation state. After loading Cu and Co, the electronic properties of N and C changed, with N predominantly existing in the form of metal-pyridinic N. Similary, Wu et al. synthesized SOD-[Zn(mim)₂] (MAF-4/ZIF-8, Hmim = 2-methylimidazole) by modulating the Zn/Co/Ni feed ratios during MOF synthesis under vigorous stirring^[66]. Under an N₂ atmosphere, the MOF precursors were combined with NH₄Cl and pyrolyzed at 900 °C to yield carbonized samples. Inductively coupled plasma mass spectrometry (ICP-MS) quantification established the Co and Ni contents, and the products were designated AD-Co₁Ni₀, AD-Co_0.72Ni_0.28, AD-Co_0.62Ni_0.38, AD-Co_0.48Ni_0.52, AD-Co_0.41Ni_0.59, AD-Co₀Ni₁, and NC-Co_0.49Ni_0.51. Powder X-ray diffraction (PXRD) and high-resolution transmission electron microscopy (HRTEM) confirmed the absence of nanocrystals or nanoparticles/clusters. High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) imaging revealed numerous bright spots on the carbon matrix, attributable to atomically dispersed Co and Ni atoms, many separated by distances less than 5.0 Å, suggesting the coexistence of multiple bimetallic site configurations [Figure 2D].

Figure 2. (A) Synthesis process of Fe-N₄/Pt-N₄@NC^[63]. Copyright 2021, Wiley-VCH; (B) Preparation and structure illustration of the “carbon islands” of DMASs-CDs (red: O atom, blue: N atom, gray: C atom, green: first metal atom, orange: second metal atom)^[64]. Copyright 2023, National Academy of Sciences; (C) Synthetic scheme of Cu-Co/NC^[65]. Copyright 2023, Wiley-VCH; (D) Schematic illustration of the synthetic processes and structures of AD-Co_xNi_1-x and O-AD-Co_xNi_1-x^[66]. Copyright 2023, Springer Nature. DMASs-CDs: Dynamically stable carbon dots with embedded bimetallic atomic sites.

Directly bimetallic-bonded asymmetric coordination structures

Direct bimetallic-bonded asymmetric coordination structures refer to catalysts that contain two (or more) different or identical single-atom active sites, where these single-metal atoms are bonded to each other. Unlike metal alloys, these directly bonded metals can only be anchored into the substrate in single-atom form. Such interactions between metals often effectively optimize and regulate the electronic structure, coordination environment, electron spin configuration, and charge state of the active sites.

As shown in Figure 3A, Zhang et al. synthesized the Zn/Co/Ni imidazole zeolite framework (Zn/Co/Ni-ZIFs) precursor using cation exchange and cavity adsorption strategies, and finely tuned the structure of the target catalyst NiN₃©CoN₃-NC by adjusting the Zn content and pyrolysis temperature^[67]. Through aberration-corrected HAADF-STEM, 3D atomic overlap Gaussian fitting mapping, X-ray absorption fine structure (XAFS), and X-ray diffraction (XRD), they systematically confirmed the structural characteristics of NiN₃©CoN₃-NC (Ni-Co covalently coupled atom pair embedded in nitrogen-doped graphitized carbon). Moreover, Wang et al. synthesized C₂N supports featuring uniform N₆ cavities and edge N deficiencies by high-temperature pyrolysis of cyclohexanone and urea in the presence of MgCl₂, providing optimal sites for metal anchoring^[68]. Fe and Sn precursors were subsequently introduced via wet impregnation followed by a second pyrolysis, yielding the FeSn-C₂N dual-atom catalyst with Fe and Sn atoms precisely anchored within the N₆ cavities, forming a robust structure. Energy band alignment between Sn p-orbitals and Fe d-orbitals enhanced p-d coupling and improved the catalyst’s thermodynamic stability, underscoring its catalytic potential. XAFS characterization revealed that Fe in FeSn-C₂N exhibited a lower oxidation state than in Fe-C₂N, due to the polarization-charge effects induced by Sn incorporation. Fe K-edge FT-EXAFS confirmed Fe–Sn bonding at 2.22 Å, verifying the bimetallic coordination [Figure 3B]. Sn K-edge X-ray absorption near edge structure (XANES) indicated a reduced Sn oxidation state relative to Sn-C₂N, reflecting charge redistribution from strong Fe-Sn electronic interactions. In summary, the reduced oxidation states of Fe and Sn in FeSn-C₂N reflect the combined effect of Fe-Sn bimetallic electronic coupling and polarization charge redistribution.

Figure 3. (A) Scheme illustration of the synthetic procedure of NiN₃@CoN₃-NC^[67]. Copyright 2023, Elsevier B.V; (B) The dimer structure of active FeSn dual atom sites derived from the EXAFS result^[68]. Copyright 2024, American Chemical Society; (C) Schematic illustration of the synthetic procedure of the diatomic Ni/Cu-N-C catalyst^[69]. Copyright 2022, American Chemical Society; (D) The geometric structure of Fe-Ni-N-P-C^[70]. Copyright 2021, Cell Press; (E) Schematic illustration of the synthesis of the Cu₂-C₂N^[71]. Copyright 2024, Wiley-VCH. EXAFS: Extended X-ray absorption fine structure.

Furthermore, Zhu et al. reported a Ni–Cu atomic pair catalyst (Ni/Cu-N-C) synthesized via a host–guest strategy [Figure 3C]^[69]. A Cu/1,10-phenanthroline complex (~9.5 Å) was encapsulated within a bimetallic NiZn zeolitic imidazolate framework (Ni–ZIF-8; cavity ~11.6 Å), ensuring isolated Cu sites in the ZIF micropores. Zn dilution and pore confinement favored the formation of Ni–Cu pairs over nanoparticles or single atoms. Characterization showed no metal nanoparticles and revealed a Ni–Cu interatomic distance of ≈2.4 Å, indicating atomically dispersed metals with strong electronic coupling. To achieve highly active SACs, Pan et al., guided by DFT and Monte Carlo (MC) simulation calculations, used a dual-solvent method to load Fe³⁺ and Ni²⁺ onto ZIF-8 as a template, followed by co-pyrolysis with sodium hypophosphite to obtain P-doped Fe-Ni dual-atom pair catalysts (Fe-Ni-N-P-C)^[70]. Both theoretical and experimental results successfully demonstrated the precise synthesis of the P-doped Fe-Ni dual-atom pair [Figure 3D].

As shown in Figure 3E, the facile atomic printing strategy to fabricate a series of M₁M₂-C₂N materials was developed by Sun et al.^[71]. In a typical synthesis, a solid mixture of cyclohexanone and urea was heated at 90 °C for 30 min to yield a crosslinked intermediate, which was then milled with anhydrous MgCl₂ in precise ratios. Calcination under N₂ at 900 °C for 2 h transformed the intermediate into C₂N, featuring a uniformly porous network rich in unsaturated nitrogen “clamps” that efficiently anchor metal dimers at vacancy sites via strong N coordination. During annealing, MgCl₂ and its crosslinked precursors generated HCl gas, facilitating the etching of metal nanoparticles and removal of organic ligands, thereby enabling the formation of two-dimensional M₁M₂-C₂N structures. Furthermore Zhang et al. developed a defect-rich sulfur- and nitrogen-co-doped carbon support (SNC) to anchor Fe–Fe dual-atom sites (A-Fe₂S₁N₅/SNC) via a multilayer defect-trapping and coordination-locking strategy^[72]. Co-pyrolysis of 2-benzothiazolylthiol and melamine under inert atmosphere generated highly defective SNC nanosheets through radical polymerization and thermal decomposition, which then stabilized dual Fe atoms by sulfur coordination and vacancy trapping.

Bridged multi-metal asymmetric coordination structures

In Bridged multi-metal asymmetric coordination structures, there is no direct interaction between the metal atoms, but they are connected by non-metallic atoms (such as O, N, S). Under the influence of non-metallic atoms, the electronic density between bimetallic sites can be redistributed, which alters the charge state of the metals and enhances catalytic performance. Non-metallic bridging atoms with different electronegativities often result in completely different catalytic effects.

Based on this, Zhang et al. prepared planar-Fe-Co and stereo-Fe-Co dual-site catalysts (DSCs) through various electrochemical deposition methods^[73]. Experimental characterizations revealed the different spatial relationships between Fe single atoms and Co atoms in CoOOH substrates, confirming that the Fe and Co atoms in the catalysts exist in an Fe-O-Co form. Furthermore, Zhao et al. developed a dual-atom catalyst (DAC) synthesis approach at the simple-level stage by exploiting precise electrostatic control and engineered adjacent vacancies^[74]. As a proof of concept, they fabricated uniformly dispersed dual-iron sites bridged by two nitrogen atoms (Fe-N₂-Fe). Their method anchors a second Fe ion at pre-existing SAC sites using bridging nitrogen precursors; subsequent pyrolysis etches neighboring carbon atoms to form vacancies that trap the second ion [Figure 4A]. Similarly, Sun et al. noted that incorporating sulfur ligands into SACs typically requires high-temperature pyrolysis to convert solid sulfur into gaseous species for substrate etching, a process that usually yields only monometal–sulfur coordination at bimetallic sites, precluding sulfur-bridged architectures^[75]. To address this, they employed wool-derived keratin as a support to construct a dual-metal catalyst featuring novel Cu–S–Ni bridging sites (Cu–S–Ni/SNC). The abundant disulfide bonds in keratin serve as in situ sulfur donors, while the protein’s partial long-range order provides defined anchoring sites for metal atoms. Additionally, X-ray absorption spectroscopy (XAS) elucidated the coordination geometry of the Cu–S–Ni sulfur-bridged bimetallic sites [Figure 4B].

Figure 4. (A) Atomic structure model of Fe-N₂-Fe DAC^[74]. Copyright 2024, Wiley-VCH; (B) Atomic structure model of Cu-S-Ni/SNC^[75]. Copyright 2024, Wiley-VCH; (C) Schematic diagram of the polymerization-pyrolysis synthesis method of Co-N-Ni^[76]. Copyright 2023, American Chemical Society; (D) Schematic Illustration of the MIR strategy for DACs^[77]. Copyright 2024, American Chemical Society; (E) Schematic illustration of the synthesis of DACs via a macrocyclic precursor-mediated encapsulation-pyrolysis process. The model on the far right represents the local structure of the M¹M²-DAC^[78]. Copyright 2023, American Chemical Society. DAC: Dual-atom catalyst; SNC: sulfur- and nitrogen-co-doped carbon; MIR: metal ion recognition.

Li et al. prepared a Co-N-Ni catalyst by co-grinding melamine, L-aspartic acid, Co(acac)₃, and Ni(acac)₂; adding ethanol/HCl and evaporating the solvent; then pyrolyzing under N₂ and cooling^[76]. Characterization confirmed that this polymerization–decomposition route precisely controls the N-bridged Co-Ni spacing [Figure 4C]. As shown in Figure 4D, Wang et al. developed an innovative metal ion recognition (MIR) strategy to address the challenges in the synthesis of dual-atom catalysts^[77]. This strategy involves the sequential adsorption of target cations and anions onto NC substrates to form precursors, enabling the synthesis of various DACs. Taking Fe₁Sn₁-DAC as an example, NC derived from ZIF-8 with rhombic dodecahedral morphology and negative surface charge was first prepared. Then, [Fe(bpy)₃]²⁺ and [SnCl₆]^2- were sequentially adsorbed to form a heterometallic bimetallic ion pair, followed by high-temperature pyrolysis to obtain Fe₁Sn₁-DAC. Multiple characterization techniques confirmed the successful synthesis of this catalyst, and theoretical calculations revealed the coordination structure of Fe and Sn. Additionally, other DACs such as Fe₁Co₁, Fe₁Ni₁, Fe₁Cu₁, Fe₁Mn₁, Co₁Ni₁, Co₁Cu₁, Co₂, and Cu₂ were also successfully synthesized, demonstrating the versatility of the strategy. Similarly, Zhang et al. introduced a macrocycle-mediated encapsulation–pyrolysis strategy for precise DAC synthesis^[78]. A Robson-type macrocycle embeds homo- and heteronuclear bimetallic centers within a planar ligand framework, and its confinement in porous carbon preserves the dual-atom motif during heat treatment. Employing this method, a Fe–Cu dual-atom site bridged by two oxygen atoms was constructed [Figure 4E].

Single-metal asymmetric coordination structures

The coordination environment of the central metal in SACs has a crucial impact on catalytic performance. Studies have shown that incorporating the central atom with one or more coordination atoms of different electronegativity can significantly enhance the catalyst’s catalytic performance.

Recent in-depth studies on SACs have pointed out that the symmetric M-N₄ structure of traditional SACs, due to its symmetric charge state, is not conducive to the adsorption-desorption and catalytic conversion of all intermediates in catalytic reactions. This results in high reaction barriers and low catalytic selectivity for M-N₄ structure catalysts. Therefore, replacing one of the nitrogen atoms in M-N₄ with other atoms of different electronegativity can impart strong polarity and a special charge state to the metal center, thereby improving its catalytic activity.

Proton exchange membrane fuel cells (PEMFCs) and zinc–air batteries efficiently convert chemical energy into electricity, offering significant promise for new energy vehicles and flexible electronics. However, the sluggish kinetics of the ORR limits their electrocatalytic efficiency. SACs have gained widespread attention due to their fully exposed active sites and high atomic utilization. Current research mainly focuses on symmetric M-N₄ sites anchored on carbon-nitrogen materials. The symmetric M-N₄ sites, due to their uniform electronic distribution, have relatively limited adsorption capability for different intermediates during the ORR process. In contrast, single-atom sites with asymmetric coordination structures exhibit asymmetric electronic distributions, and through effective site regulation, can enhance the adsorption and activation of various intermediate species.

Recently, Li et al. synthesized an Fe-N/O-C catalyst featuring isolated Fe–N₃O sites adjacent to vacancy defects^[55]. By introducing O into the first coordination shell and coupling with nearby defects, they tuned the Fe d-orbital energy and spin state from low to intermediate spin, allowing the unpaired dz² electrons to interact with O π^* orbitals to mitigate over-binding [Figure 5A]. The catalyst achieves ORR half-wave potentials of 0.927 V in 0.1 M KOH and 0.80 V in 0.1 M HClO₄ [Figure 5B and C], and zinc–air and PEMFC peak power densities of 490 and 1,179 mW·cm^-2, respectively [Figure 5D]. DFT shows a more negative Fe d-band center in Fe–N/O–C vs. Fe–N–C, weakening intermediate adsorption and shifting the rate-determining step from OH^* desorption to OOH^* formation. In addition, A coal-pitch-derived Fe SAs/N and S co-doped carbon (NSC)-vacancy defects (vd) catalyst was synthesized via supramolecular self-assembly pyrolysis, yielding asymmetric Fe–N₃S₁ single-atom sites stabilized by abundant vacancy defects^[79]. This vacancy-induced Fe–N₃S₁ coordination synergistically enhances ORR kinetics, delivering markedly improved turnover frequency (TOF) and mass activity in both alkaline and acidic media, far surpassing 20 wt% Pt/C [Figure 5E and F]. Mechanistic analysis indicates that Fe–N sites predominantly facilitate O₂ adsorption activation under alkaline conditions, whereas Fe–S sites govern ORR in acidic media. DFT calculations confirm that the elevated d-band center of Fe–N₃S₁ sites promotes O₂ adsorption and OH reduction, while vacancy defects balance ^*OOH formation and ^*OH reduction to collectively drive the electrocatalytic ORR. Low-electronegativity boron atoms were introduced by Guan et al. to construct an asymmetric Co–N₃B SAC^[80]. Relative to Co-N₄, Co-N₃B reduces ORR free energy and strengthens ^*O adsorption, thereby suppressing the two-electron pathway and H₂O₂-induced corrosion to enhance stability. In situ attenuated total reflection (ATR)-surface-enhanced infrared absorption spectroscopy (SEIRAS) confirmed robust Co–O intermediate interactions [Figure 5G]. In rechargeable Zn–air batteries, Co-N₃B delivers ≈253 mW·cm^-2 peak power, ≈819 mAh·g^-1 energy density, and > 110 h of cycling. Yin et al. introduced a P, S co-doping strategy to fine-tune Ce single-atom sites and their N coordination via long-range interactions^[81]. A two-step pyrolysis yielded P,S-modified Ce SAs without oxidation or aggregation. XAS confirmed that N atoms stabilize atomic Ce dispersion [Figure 5H]. The hollow porous carbon polyhedral support further enhances active-site exposure. DFT calculations revealed that P and S shift the Ce-4f orbitals toward the Fermi level, boosting electron transfer. The optimized Ce SAs/PSNC exhibits improved ORR activity and reduced overpotentials. Moreover, there has been some research on constructing asymmetrically coordinated SACs with main group atoms as the center atoms for ORR. Shao et al. developed a p-block Sn SAC with axial oxygen ligands to modulate the Sn electronic structure and intermediate adsorption^[82]. The asymmetric Sn SAs achieve an ORR half-wave potential of 0.912 V and a mass activity of 13.1 A·mg_Sn^-1, outperforming commercial Pt/C and most transition-metal SAs. In both liquid- and solid-state zinc–air batteries, it likewise surpasses Pt/C in electrocatalytic performance. In addition, Lin et al. designed an asymmetric Sn-N/O SAC (Sn-N/O-C) that delivers exceptional ORR activity and durability^[83]. In alkaline media, Sn-N/O-C achieves a half-wave potential of 0.910 V, surpassing most state-of-the-art catalysts, and maintains stability in both alkaline and acidic electrolytes. In Zn–air batteries, it affords an energy density of 254 mW·cm^-2, outperforming reported M-N-C cathodes. DFT calculations reveal that asymmetric N,O coordination strengthens O₂ adsorption and charge transfer vs. symmetric SnN₄ sites, thereby accelerating ORR process [Figure 5I]. This work establishes an N/O coordination strategy for crafting robust, high-performance p-block metal single-atom ORR catalysts.

Figure 5. (A) Schematic diagram of d-orbital spin-electron filling states; (B) LSV curves of FeN/OC, FeNC and Pt/C 20% in 0.1 M KOH at the rotation speed of 1,600 rpm; (C) LSV curves of FeN/OC, FeNC and Pt/C 20% in 0.1 M HClO₄ at the rotation speed of 1,600 rpm; (D) Polarization curves and corresponding power density plots of FeN/OC, FeNC and Pt/C-assembled 2.0 bar H₂-O₂ PEMFCs^[55]. Copyright 2024, Wiley-VCH; (E) E_1/2, J_k, Tafel, mass activity, and TOF at 0.87 V of the catalysts in 0.1 M KOH and (F) 0.5 M H₂SO₄^[79]. Copyright 2024, Wiley-VCH; (G) In situ ATR-SEIRAS spectra of ORR over Co₁/BNG in O₂-saturated 0.1 M KOH^[80]. Copyright 2024, Wiley-VCH; (H) The corresponding EXAFS R-space fitting curves of CoSAs/PSNC^[81]. Copyright 2023, Wiley-VCH; (I) O₂ absorption energies on the optimized model structures of SnN₄ and SnN₂O^[83]. Copyright 2024, Wiley-VCH. LSV: Linear sweep voltammetry; PEMFCs: proton exchange membrane fuel cells; TOF: turnover frequency; ATR-SEIRAS: attenuated total reflection-surface-enhanced infrared absorption spectroscopy; ORR: oxygen reduction reaction; EXAFS: extended X-ray absorption fine structure.

Electrochemical CO₂ reduction into fuels or value-added chemicals offers a promising route to mitigate energy and environmental challenges, garnering extensive interest from both academia and industry. Nevertheless, sluggish CO₂ activation kinetics and the scarcity of efficient electrocatalysts necessitate high energy inputs, impeding industrial deployment. Consequently, designing electrocatalysts with enhanced activity and selectivity is critical. Transition-metal SACs supported on N-doped porous carbon have demonstrated considerable advancement in CO₂ electro conversion, yet their symmetric atomic coordination constrains catalytic performance. Precisely tuning the atomic-level coordination environment to optimize intermediate adsorption and facilitate charge transfer to the single-atom active sites remains a formidable challenge. Recently, Huang et al. employed Mg(OH)₂ as a low-cost template, whose decomposition during pyrolysis generated water vapor for axial Ni-O coordination, while ammonia doping formed lateral Ni-N₄ sites on porous graphitic carbon^[84]. The resulting Ni SA-N-PGC catalyst, featuring asymmetric Ni-N₄-O coordination, achieved over 97% CO Faradaic efficiency (FE) at -0.76 V (9,097 h^-1) and maintained > 90% efficiency across -0.5 to -1.1 V, with negligible decay after 40 h. Mechanistic studies revealed that axial Ni–O coordination induced site asymmetry, optimized the local coordination environment, enhanced charge polarization, and lowered the Gibbs free energy of ^*COOH formation, thereby improving ORR kinetics and selectivity [Figure 6A and B]. Jin et al. designed an asymmetric Fe-S₁N₃ SAC, where one Fe atom is coordinated to one S and three N atoms, inducing pronounced geometric distortion^[85]. The Fe-S₁N₃ site dynamically adjusts its structure upon intermediate adsorption, decoupling ^*COOH and ^*CO binding energies and breaking their linear scaling relation. This structural adaptability yields a CO FE of 99.02% and a TOF of 7,804 h^-1, alongside robust stability. Furthermore, Wang et al. pointed out that main group SACs have the ability to prevent hydrogen evolution reactions (HER) and CO poisoning, making them promising for CO₂RR to produce CO^[86]. Based on this, the researchers reported an O-doping strategy that formed an asymmetric coordination structure (Ca-N₃O) around Ca atoms, which localized electrons on the p-orbital, thus enhancing CO₂ activation. Combining DFT calculations and experimental verification, they confirmed the advantages of the asymmetric coordination structure in CO₂ activation and ^*COOH generation. This catalyst exhibited excellent catalytic activity and selectivity in the CO₂ reduction to CO reaction. As shown in Figure 6C, the FE for CO was as high as 97.3% at -0.85 V [vs. reversible hydrogen electrode (RHE)], and in the flow cell, a high current density of 400 mA·cm^-2 was achieved while maintaining an FE of ≥ 90%. Liu et al. used a rapid pulsed discharge to fabricate a graphene-aerogel-supported Cu SA catalyst (CuN₁O₁-SAs/GAs)^[87]. The intense thermal shock and electromagnetic field drove in situ N-doping and atomic dispersion of Cu, yielding an asymmetric Cu-N₁O₁ coordination confirmed by atomic-level analysis. CuN₁O₁-SAs/GAs displayed high selectivity and activity for HCOOH production across a wide potential range in CO₂RR, with stable current densities over 10 h. In situ XAFS showed elongated Cu-N/O bonds that facilitate formate formation, and DFT calculations indicated a lower energy barrier for HCOOH vs. CO and a preference for HCOOH at unsaturated Cu-N₁O₁ sites [Figure 6D]. Notably, although bulk p-block metals (e.g., Pb, Bi) favor CO₂-to-HCOOH conversion, their single-atom analogs often exhibit enhanced CO selectivity due to fully occupied d-orbitals. Moreover, Zhou et al., guided by DFT calculations, designed a novel Pb SAC (Pb-N₂SV), in which Pb is coordinated to two N atoms, one S atom, and an adjacent vacancy, creating a highly asymmetric microenvironment^[88]. This symmetry breaking endowed Pb-N₂SV with superior CO₂ electroreduction activity at -0.47 V, achieving a CO FE of 97.3% [Figure 6E] and stable operation over 33 h, surpassing most p-block SACs. In situ IR and DFT analyses indicate that S doping enhances electron localization at the Pb center, promoting ^*COOH adsorption, while vacancy defects modulate electronic dispersion to facilitate ^*CO desorption [Figure 6F]. These synergistic effects optimize intermediate adsorption energies and accelerate CO₂RR.

Figure 6. (A) Proposed reaction steps of electrocatalytic reduction of CO₂ to CO over the Ni-N₄-O catalytic site; (B) Reaction paths and free energy diagrams of CO₂ reduction to CO over Ni-N₄ and Ni-N₄-O catalytic sites. Inset shows the side view of the optimized configurations of intermediates over the Ni-N₄-O catalytic site^[84]. Copyright 2022, American Chemical Society; (C) FE_CO at different potentials in H-cell^[86]. Copyright 2023, Wiley-VCH; (D) The Gibbs free energy diagrams of pathways I (from CO₂ to HCOOH) and II (from CO₂ to CO)^[87]. Copyright 2024, Wiley-VCH; (E) Comparison of FE_CO at different potentials for NSC, Pb-N₄ and Pb-N₂SV in CO₂-saturated 0.5 M KHCO₃; (F) Gibbs free energy profiles^[88]. Copyright 2024, Wiley-VCH. FE_CO: CO Faradaic efficiency; NSC: sulfur- and nitrogen-co-doped carbon.

Multi-metal asymmetric coordination structures

In complex multistep reactions, the adsorption and desorption strengths of various intermediates at the active sites are often different. However, SACs feature only a single active center and therefore exhibit the same adsorption/desorption response for all intermediates. Therefore, multi-metal asymmetric coordination structures have been introduced into electrochemical systems with multiple reaction intermediates (e.g., ORR, OER, NRR, CO₂RR^{[40,67,68,73]}), enabling differentiated control over each intermediate’s adsorption and desorption behavior. Multi-metal asymmetric coordination structures can effectively regulate the charge state, orbital energy levels, d-band center, optimize electronic spin configurations, and influence the bonding and antibonding states of reactants, thus optimizing the adsorption and desorption of intermediates during the reaction. Moreover, multi-metal sites enable tandem catalysis during the reaction process. Different active site centers can achieve optimal adsorption and desorption capabilities for various intermediates, allowing the reaction to proceed smoothly^[89-91]. In this section, we will discuss the applications of multi-metal asymmetric coordination structures in ORR and CO₂RR.

As shown in Figure 7A, Li et al. synthesized a Cu-Co/NC catalyst with outstanding bifunctional ORR/OER activity in alkaline media (ORR E_1/2 = 0.92 V), and maintained strong ORR performance in acidic (0.85 V) and neutral (0.74 V) electrolytes^[65]. When applied in zinc–air batteries, it delivered stable open-circuit voltage, polarization, and power density, and endured 1,000 charge-discharge cycles (510 h at 10 mA·cm^-2) without performance loss. DFT analysis revealed that Cu-Co dual sites induce asymmetric charge distribution, optimizing adsorption/desorption of oxygen intermediates and accelerating both ORR and OER kinetics. It is noteworthy that diatomic catalysts composed of two identical metal atoms likewise exhibit high catalytic activity. Li et al. synthesized a charge-asymmetric dual-atom catalyst Fe₂-S₁N₅/SNC using a two-step method^[92]. Fe₂-S₁N₅/SNC exhibited excellent electrocatalytic performance in acidic flow cells. In 0.1 mol·L^-1 HClO₄ solution, its half-wave potential reached 0.829 V, an improvement of 56 mV compared to the SAC Fe-N₄/NC. When loaded on the cathode of a hydrogen fuel cell, it demonstrated a high peak power density of 810 mW·cm^-2 [Figure 7B]. The authors attributed the catalyst’s excellent performance to the fact that charge-asymmetric dual-atom sites can more flexibly regulate the adsorption energy of point-charged intermediates at the active center, and the asymmetric electronic structure enhances the electronic transfer capability. In 2024, He et al. synthesized Co-Fe-SNC via a host–guest strategy to create bridging asymmetric Co-Fe coordination^[93]. Atomically dispersed Co-Fe sites, combined with sulfur-driven surface modulation, deliver superior ORR performance, achieving a half-wave potential of 0.92 V in alkaline media [Figure 7C]. DFT studies reveal that bimetallic sites strengthen oxygen-intermediate binding and lower activation barriers, while sulfur’s weak electronegativity fine-tunes the electronic structure for further activity enhancement. Employed as zinc–air battery cathodes, Co-Fe-SNC surpasses commercial Pt/C in power density, specific capacity, and durability. It is worth noting that Li et al. engineered a Fe-Ru dual-atom catalyst on N-doped carbon with moderate tensile strain, achieving record acidic ORR performance among Fe SACs^[94]. Strain-tuned d-band center yielded a half-wave potential of 0.86 V and a fuel-cell power density of 700 mW·cm^-2. After 50,000 cycles, the half-wave potential dropped by only 17 mV, demonstrating exceptional durability. In situ SEIRAS and DFT analyses reveal that tensile strain enhances d-electron penetration into antibonding orbitals, accelerating intermediate reduction and desorption. ATR-SEIRAS further confirmed strain-optimized Fe-O bonding and faster ORR kinetics in acid.

Figure 7. (A) Kinetic current density of electrocatalysts at 0.80 and 0.85 V^[65]. Copyright 2023, Wiley-VCH; (B) PEMFC polarization and power density curves of Fe₂-S₁N₅N₅/SNC^[92]. Copyright 2024, The Royal Society of Chemistry; (C) LSV polarization curves of the ORR in O₂-saturated 0.1 M KOH solution^[93]. Copyright 2024, The Royal Society of Chemistry; (D) Schematic illustration of the single-site catalytic mechanism for PdCu (left) and multi-site tandem mechanism for PdCu-Fe₃O₄ (right)^[95]. Copyright 2024, Wiley-VCH. PEMFC: Proton exchange membrane fuel cell; LSV: linear sweep voltammetry; ORR: oxygen reduction reaction.

Furthermore, asymmetric catalysts utilizing metal oxides as supports have likewise demonstrated superior ORR performance. For example, Li et al. successfully synthesized a PdCu-Fe₃O₄ in-plane heterostructure through sequential reduction of different metal precursors^[95]. This structure combines the advantages of Pd and Fe₃O₄, achieving synergistic catalysis. The authors pointed out that this method ensures the uniformity and stability of the heterostructure. ORR catalytic performance evaluation of the prepared PdCu-Fe₃O₄ in-plane heterostructure showed that it maintained close to 4e^- electron transfer numbers at different potentials, indicating its high catalytic activity. Notably, the study found that the proton-coupled electron transfer (PCET) steps during ORR occur in an orderly manner at Pd and Fe sites. The initial PCET steps occur at Fe sites, while the last two PCET steps take place at the optimal Pd sites [Figure 7D]. This reveals the tandem catalysis mechanism of ORR on the PdCu-Fe₃O₄ in-plane heterostructure, providing a new research paradigm for understanding ORR mechanisms in complex systems.

Remarkably, catalysts featuring multi-metal asymmetric coordination structures also find extensive applications in the CO₂RR. For example, Zhu et al. constructed a quasi-covalently bonded Ni–Cu dual-atom electrocatalyst (Ni/Cu-N-C) exhibiting exceptional CO₂ reduction performance^[69]. It delivers peak current densities and the lowest onset potential across the full potential window, achieving 13.7 mA·cm^-2 at -0.7 V (vs. RHE) - 3.8 and 5.9 times those of Ni-N-C and Cu-N-C, respectively [Figure 8A]. Ni/Cu-N-C maintains ≥ 97.7 % FE throughout, with j_(co) of 24.8 mA·cm^-2 at -0.8 V and an unprecedented TOF of 20,695 h^-1 at -0.6 V [Figure 8B]. It shows stable current and efficiency over 60 h. In situ XAFS indicates that Cu accelerates CO₂-CO conversion and prevents Ni over-reduction, boosting site stability. DFT/NBO analyses reveal that Ni–C antibonding states in NiCuN₆ sites raise ^*COOH adsorption via Ni 3d-C 2p hybridization and stronger Ni-Cσ bonding, underpinning superior intrinsic activity [Figure 8C and D]. As shown in Figure 8E, Chen et al. designed a Cu atom fixed on an N, S-doped carbon matrix for CO₂RR, Cu₂-NC^[96]. The results showed that Cu₂-SNC exhibited extraordinary and stable electrocatalytic performance for CO₂-ethanol conversion (with FE of 62.6% at -0.8 V vs. RHE), far outperforming Cu-NC and Cu₂-NC. DFT calculations revealed the reaction mechanism for the high selectivity of ethanol and were further confirmed by in situ attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy. The asymmetric electronic cloud distribution around Cu dual-atom sites in CuN₂-CuNS destabilizes ethylene reaction intermediates, thus favoring ethanol formation. Similarly, Sun et al. fabricated a sulfur-bridged asymmetric CuNi dual-atom site catalyst, Cu-S-Ni/SNC, which exhibited excellent CO₂RR performance in an H-type cell, with a CO Faradaic efficiency (FE_CO) of 98.1% at -0.65 V^[75]. In a flow cell, the industrial current density reached 550 mA·cm^-2 [Figure 8F and G]. The high performance of Cu-S-Ni/SNC is due to the electronic tuning effect between Cu and Ni through the sulfur bridge, where electrons transfer from Cu to Ni. In situ spectroscopy and DFT calculations showed that Cu, as the primary adsorption site, is doubly regulated by S and Ni atoms, with Ni and S atoms enhancing Cu’s CO₂ activation and the formation of the ^*COOH intermediate. The study showed that adjusting the type of bridging atom can effectively modulate the charge state of asymmetric sites, which is beneficial for CO₂RR development.

Figure 8. (A) LSV curves acquired in CO₂-saturated 0.5 M KHCO₃ solution on a rotating disc electrode at a rotating speed of 1,600 rpm. The inset highlights the LSV curves in the potential range from -0.1 to -0.7 V; (B) FE for CO production at various applied potentials; (C and D) COHP analysis of the Ni–C bond on NiN₄ and NiCuN₆; note that -COHP is used as the a-axis^[69]. Copyright 2022, American Chemical Society; (E) FE value at different potentials by Cu₂-SNC catalysts^[96]. Copyright 2024, Wiley-VCH; (F) FE_CO at different potentials of Cu-S-Ni/SNC and references; (G) LSV curves of Cu-S-Ni/SNC in the flow cell^[75]. Copyright 2024, Wiley-VCH. LSV: Linear sweep voltammetry; FE: Faradaic efficiency; COHP: crystal orbital Hamilton population; SNC: sulfur- and nitrogen-co-doped carbon.

Aberration-corrected scanning transmission electron microscopy with electron energy loss spectroscopy

Electron microscopy techniques, including scanning electron microscopy (SEM), transmission electron microscopy (TEM), and HRTEM, have been widely used for morphology characterization of catalysts. However, they are limited by insufficient resolution and thus cannot accurately characterize individual atoms. To address this limitation, aberration-corrected scanning transmission electron microscopy (AC-STEM) has been developed to observe single atoms. In particular, spherical aberration-corrected HAADF-STEM (AC-HAADF-STEM) can display larger metal atoms as prominent bright spots. This allows for the observation of single-atom sites and dual-atom sites in catalyst materials. Sun et al. used AC-HAADF-STEM to characterize the Cu-S-Ni/SNC catalyst they prepared, as shown in Figure 9A, where the yellow circles highlight isolated bimetallic sites, and the red circles encircle individual Cu or Ni single atoms^[75]. Figure 9B shows the 3D model of the dual-atom site at position 1 in Figure 9A. The authors statistically analyzed most of the sites in the image, and the results showed that bimetallic sites accounted for more than 70% [Figure 9C], confirming that Cu-Ni bimetallic sites dominate in the catalyst. Similarly, Xie et al. used AC-HAADF-STEM to confirm that most of the Zn-Fe metals in their prepared Fe-Zn@SNC catalyst exist as isolated dual-atom sites (as shown in the yellow circle in Figure 9D)^[97]. Furthermore, the authors statistically analyzed the distance between the dual-atom sites, and the results Figure 9E showed that the distance between dual-atom sites in Fe-Zn@SNC catalyst is primarily concentrated between 2.8 to 3.3 Å.

Figure 9. (A) HAADF-STEM image of Cu-S-Ni/SNC. The dual-atom sites and single sites were circled with dotted yellow line and red line, respectively; (B) 3D model of Cu-S-Ni sites along the region of 1 in (A); (C) Density profiles of Cu-S-Ni and single-atom sites in the HAADF-STEM^[75]. Copyright 2024, Wiley-VCH; (D) HAADF-STEM image of Fe-Zn@SNC; (E) Kernel method of frequency statistic and Gauss fitting of diatomic distances obtained from all the yellow circles in (D)^[97]. Copyright 2023, Wiley-VCH. HAADF-STEM: High-angle annular dark-field scanning transmission electron microscopy; SNC: sulfur- and nitrogen-co-doped carbon.

However, AC-STEM cannot provide information on the chemical composition and elemental identification of the sample, which is complemented by EELS. In recent years, EELS has offered unparalleled advantages for the characterization of SACs and DACs. By measuring the energy-loss spectrum of a high-energy electron beam transmitted through a specimen, EELS yields rich information about elemental composition, chemical bonding, oxidation state and coordination environment^[98]. When combined with aberration-corrected high-resolution scanning transmission electron microscopy (STEM), it enables sub-angstrom localization of individual atoms, while the energy-loss spectrum confirms elemental identity and discriminates impurity atoms. This structural–chemical dual-criterion approach permits both qualitative and quantitative determination of atomic positions and oxidation states, thereby effectively eliminating misidentification^[99-101]. The STEM–EELS tandem thus provides significant advantages in active-site identification of SA catalysts, elemental qualitative–quantitative analysis, microstructural characterization and in situ dynamic observation. Zhang et al. used HAADF-STEM to characterize the FeCu bimetallic catalyst they prepared (the schematic structure of which is shown in Figure 10A)^[78]. Figure 10B shows a large number of bright spots dispersed on the carbon support, which are considered to be metal sites. The researchers pointed out that the proportion of metal atom pairs is about 70%. Additionally, the authors performed EELS analysis on the atomic pair within the red box in Figure 10B, and the results, shown in Figure 10C, confirmed that the two bright spots within the red box are Fe-Cu atomic pairs. As shown in Figure 10D, Yang et al. undertook a two-phase method to create Cu-N-C SACs, featuring a uniform and distinctly defined Cu²⁺-N₄ configuration^[102]. It is also confirmed the dispersion of Cu atoms by observing HAADF-STEM images [Figure 10E], and the clear signals of C, N, and Cu in EELS, combined with elemental mapping, further corroborated the N/C coordination [Figure 10F]. Despite the single-atom sensitivity of AC-STEM with EELS and its capability to obtain rich physicochemical properties, the requirement for sample stability under the electron beam necessitates the combination of other characterization techniques for analysis^[98].

Figure 10. (A) The local structure of the M¹M²-DAC; (B) HAADF-STEM image of FeCu-DAC; (C) EELS of an Fe-Cu atomic pair as marked in the red rectangle in the HAADF-STEM image^[78]. Copyright 2023, American Chemical Society; (D) The structure of Cu-N-C SAC; (E) Atomic-resolution STEM-HAADF image; (F) The EELS extracted from the yellow rectangle in STEM-HAADF image^[102]. Copyright 2021, American Chemical Society. DAC: Dual-atom catalyst; HAADF-STEM: high-angle annular dark-field scanning transmission electron microscopy; EELS: electron energy-loss spectroscopy; SAC: single-atom catalyst.

XAS

Synchrotron radiation XAS plays a crucial role in the study of asymmetric coordination structures. XAS includes extended XANES and extended XAFS (EXAFS). XANES can provide information on the atomic electronic orbitals and oxidation states in SACs, as well as the electron density distribution, oxidation state changes, and the localization of electrons in single atoms. This is particularly important for analyzing the electron transfer processes in catalytic reactions. On the other hand, EXAFS can reveal the local structure of single atoms in detail, including coordination environment, bond lengths, and coordination numbers, which can help researchers understand the arrangement of surrounding atoms in SACs. Furthermore, by utilizing the high temporal resolution capabilities of synchrotron radiation XAS, researchers can monitor the dynamic changes of SACs in real-time during catalytic reactions, capturing the evolution of their structure and electronic states at different stages of the reaction^[103-107]. Additionally, the high temporal resolution of synchrotron radiation XAS allows researchers to monitor the dynamic changes of SACs during catalytic reactions in real time, capturing the evolution of their structure and electronic states at different stages of the reaction.

As shown in Figure 11A, Pei et al. investigated the local coordination environment of Co-S_xN_4-x SACs using XANES and FT-EXAFS techniques^[108]. The XANES spectra revealed that, compared to Co foil and Co₃O₄, the Co in SACs carries a positive charge and exhibits a lower valence state, which decreases as the S content increases. FT-EXAFS analysis [Figure 11B] confirmed the single-atom structure of Co and unveiled a new Co-S scattering peak that intensifies with increasing S content. Through EXAFS fitting [Figure 11C-F], researchers extracted structural parameters, finding that an increase in Co-S coordination number leads to a decrease in Co-N coordination number, indicating the substitution of N by S.

Figure 11. (A) XANES and (B) FT-EXAFS spectra of the Co-S_xN_4-x SACs; (C) Co-N₄, (D) Co-S₁N₃, (E) Co-S₂N₂ and (F) Co-S₃N₁^[108]. Copyright 2021, Springer Nature. XANES: X-ray absorption near edge structure; FT-EXAFS: Fourier transform extended X-ray absorption fine structure; SACs: single-atom catalysts.

Similarly, Sun et al. studied the local coordination structure of the Cu-S-Ni asymmetric sites in Cu-S-Ni/SNC catalysts using XANES and EXAFS techniques [Figure 12A]^[75]. Figure 12B and C shows the normalized Cu and Ni K-edge XANES spectra of Cu-S-Ni/SNC and comparative samples. The results show that Cu in Cu-S-Ni/SNC has a higher oxidation state compared to Cu foil, presenting a positive oxidation state, but its oxidation state lies between that of CuNi/NC and Cu/SNC, indicating that the oxidation state of Cu is influenced by Ni and S elements. Similarly, the Ni K-edge XANES spectrum shows that Ni in Cu-S-Ni/SNC also exhibits a positive oxidation state, with its oxidation state between that of CuNi/NC and Ni/SNC, which is mainly due to the effect of the coordination environment and the doping elements [Figure 12C]. These observations suggest that the electronic oxidation states of Cu and Ni are both between 0 and +2, and are influenced by the S atoms and the second metal.

Figure 12. (A) Atomic structure model of Cu-S-Ni/SNC; (B) The Cu K edge XANES spectra of Cu-S-Ni/SNC; (C) The Ni K edge XANES spectra of Cu-S-Ni/SNC; (D) The Cu K edge FT-EXAFS spectra of Cu-S-Ni/SNC and the references; (E) The Ni K edge FT-EXAFS spectra of Cu-S-Ni/SNC and the references; (F) The Cu K-edge XANES spectra of Cu-S-Ni/SNC at various potentials during CO₂RR; (G) The Ni K-edge XANES spectra of Cu-S-Ni/SNC at various potentials during CO₂RR; (H) The Cu K-edge FT-EXAFS at open circuit, -0.4, -0.7, and -1.0 V vs. RHE; (I) The Ni K-edge FT-EXAFS at open circuit, -0.4, -0.7, and -1.0 V vs. RHE^[75]. Copyright 2024, Wiley-VCH. SNC: Sulfur- and nitrogen-co-doped carbon; XANES: X-ray absorption near edge structure; FT-EXAFS: Fourier transform extended X-ray absorption fine structure; RHE: reversible hydrogen electrode.

EXAFS spectra elucidate local atomic coordination in Cu–S–Ni/SNC. Cu K-edge data [Figure 12D] feature a main Cu–N peak at 1.65 Å, a Cu–S shoulder at 1.85 Å, and a minor Cu–Ni signal at 2.62 Å - longer than the 2.47 Å direct Cu–Ni bond. Ni K-edge spectra [Figure 12E] show analogous peaks at 1.72 Å (Ni-N) and 1.92 Å (Ni-S), plus a Ni-Cu scattering at ~2.5 Å. The metallic scattering signatures confirm Cu-Ni interactions and the modulatory effect of S ligands on electronic structure, providing a structural basis for the catalyst’s synergistic effects and enhanced performance.

To elucidate the true active sites and their dynamic evolution, the authors employed in situ XAS spectroscopy to monitor the catalytic behavior of Cu-S-Ni/SNC [Figure 12F-I]. Upon exposure to CO₂-saturated KHCO₃, the Cu K-edge XANES shifts +1.05 eV, indicating CO₂ adsorption–induced charge transfer and oxidation of Cu sites, while the Ni K-edge shifts -0.48 eV, suggesting Ni acts as an electron acceptor to modulate Cu’s oxidation state. Under applied potential, both Cu and Ni pre-edge peaks shift to lower binding energies, indicating progressive reduction; in Cu K-edge FT-EXAFS, the Cu–N bond contracts from 1.62 to 1.55 Å and Cu-S from 1.85 to 1.78 Å, with analogous shifts observed in the Ni EXAFS. From these observations, the authors conclude that Cu atoms serve as the principal active sites for CO₂ reduction, and that S-derived electrons, coupled into the Cu/Ni 3d orbitals, synergistically enhance Cu’s adsorption and catalytic performance.

ToF-SIMS

ToF-SIMS is an advanced analytical technique that bombards a sample with high-energy ions to sputter secondary ions, which are then separated based on their mass-to-charge ratios. ToF-SIMS, as a static SIMS technique, allows for re-analysis of the sample surface without causing destruction. The mass spectra obtained not only encompass atomic ions but also include molecular ions unique to the original surface, providing richer molecular information compared to XPS and boasting higher resolution^[109].

Koshy et al. synthesized Ni-N-doped carbon materials, which are designed as electrocatalysts for CO₂ reduction^[110]. The critical information regarding the presence and distribution of NiN_xC_y fragments gathered through the ToF-SIMS provided definitive evidence for the existence of isolated, nitrogen-coordinated single nickel atom sites. Specifically, within a Ni–N–C material derived from pyrolyzed polyacrylonitrile (NiPACN), a clear peak corresponding to the ⁵⁸NiN₂C₂^- fragment was detected at 109.94 amu. Notably, no peak appeared within the same mass range in the metal-free PACN control, providing compelling evidence that the NiN₂C₂^- fragment originates from NiPACN [Figure 13A]. Similarly, a peak for NiPACN, corresponding to the anticipated and confirmed presence of the ⁶⁰Ni isotope fragment ⁶⁰NiN₂C₂^-, was observed at approximately 111.94 amu, which was absent in the PACN sample [Figure 13B].

Figure 13. Comparison of ToF-SIMS intensity of NiPACN-3.5 wt% and metal-free PACN: (A) at m/z = 109.94 (⁵⁸NiN₂C₂ weight); (B) at m/z = 111.94 (⁶⁰NiN₂C₂ weight)^[110]. Copyright 2020, Wiley-VCH; ToF-SIMS positive spectrum results of (C) Co Pc and (D) Cu Pc before and after constant potential measurements^[111]. Copyright 2023, The Royal Society of Chemistry; The ToF-SIMS spectra of FeCu-DAC and Fe/Cu-SAC. Ionized fragments of (E) [FeCuN₄O₂]^-, (F) [FeN₄]^-, and (G) [CuN₄]^- are compared for FeCu-DAC and Fe/Cu-SAC^[78]. Copyright 2023, American Chemical Society. ToF-SIMS: Time-of-flight secondary ion mass spectrometry; PACN: DAC: dual-atom catalyst; SAC: single-atom catalyst.

The study by Mukadam et al. also employed ToF-SIMS as a key analytical technique, in which ToF-SIMS was used to investigate the surface composition and stability of phthalocyanine-based single-atom Co/CuPc catalysts during the electrochemical reduction of furfural^[111]. By analyzing the changes in the total ion counts of Co Pc and Cu Pc before and after the constant potential testing, it was found that the normalized count of Co Pc decreased by 16% [Figure 13C], indicating the stability of Co Pc towards furfural reduction. In contrast, for Cu Pc, the normalized counts of Cu⁺ and Cu Pc⁺ decreased by 67% and 69%, respectively [Figure 13D], suggesting that Cu Pc was removed. Therefore, Cu Pc exhibits less stability towards furfural reduction compared to Co Pc. Zhang et al. employed ToF-SIMS to probe FeCu-DAC by analyzing ion-beam–ejected fragments^[78]. A reference Fe/Cu-SAC was prepared via ZIF-8 pyrolysis with co-impregnated Fe and Cu nitrates. The diatomic [FeCuN₄O₂]^- fragment exhibits markedly higher intensity in FeCu-DAC than in Fe/Cu-SAC [Figure 13E], whereas the mononuclear [FeN₄]^- and [CuN₄]^- fragments display the opposite trend [Figure 13F and G]. This unambiguously demonstrates that FeCu-DAC preserves a [FeCuN₄O₂] ensemble akin to its precursor, while Fe/Cu-SAC consists predominantly of discrete [FeN₄] and [CuN₄] motifs.