Catalytic ammonia oxidation mediated by high-spin Fe(III) complex: combined experimental and DFT study

Abstract

Ammonia offers high hydrogen density and favorable transport properties, making it an appealing hydrogen carrier; yet conventional cracking methods for hydrogen release are energy-intensive. Molecular iron complexes offer a sustainable route for the homogeneous conversion of NH₃ to N₂ under mild conditions. Here, we describe a high-spin Fe^III-complex bearing a tetradentate N,N,O,O-donor trans-1,2-bis(2-hydroxy-3-methoxyphenyl-methaniminyl)cyclohexane ligand framework that catalyzes ammonia oxidation at room temperature. In combination with a triarylaminium oxidant and 2,4,6-collidine base, the catalyst produces up to 2.20 equivalents of N₂ per Fe center. Comprehensive characterization of the Fe^III-complex by FTIR, UV-vis, XPS, and X-ray diffraction, with Mössbauer and DFT analysis, confirmed its high-spin state. Moreover, DFT studies revealed that N–N bond formation in ammonia oxidation proceeds through nucleophilic attack followed by sequential proton- and electron-transfer steps. Together, these findings underscore the potential of high-spin Fe^III-complexes in ammonia oxidation catalysis and provide crucial mechanistic understanding of N–N bond formation.

---

Introduction

Global energy demand is rapidly increasing, intensifying concerns about carbon dioxide emissions associated with fossil fuel combustion.¹ Addressing this critical issue requires a transition toward renewable and sustainable energy systems.² Among various strategies, chemical energy storage stands out as highly promising, particularly if the stored energy can be efficiently utilized. In this regard, ammonia represents an attractive clean energy carrier, combining high energy density and carbon-free use with ease of liquefaction and a well-established worldwide infrastructure for production, storage, and distribution.^3–6 Despite these benefits, significant technical challenges remain, notably the issue of nitrogen oxide (NO_x) emissions when ammonia is combusted directly.⁷

Nevertheless, the intrinsic chemical energy stored in ammonia, particularly within its N–H bonds, can be efficiently harnessed to produce electricity using gas turbines, analogous to the energy contained in C–H and C–C bonds in conventional hydrocarbon fuels.^2–6 This utilization of ammonia as a sustainable fuel aligns closely with natural biological processes, such as anaerobic ammonia oxidation, which produces nitrite (NO₂⁻) and nitrate (NO₃⁻).^8–10 Through ammonia oxidation (AO), energy can be directly extracted as two ammonia molecules are oxidized to dinitrogen, generating six electrons and six protons. This transformation is hindered by the requirement for multiple proton- and electron-transfer steps to overcome the strong N–H bonds and facilitate N–N bond formation. Specifically, overcoming the high dissociation energy of ammonia's initial N–H bond (99.4 kcal mol⁻¹) is critical for advancing AO catalysis.^11,12

Historically, ammonia conversion to dinitrogen has primarily involved heterogeneous catalysis approaches, including electrochemical oxidation and thermal decomposition.^13–15 In contrast, homogeneous catalytic ammonia oxidation through coordination chemistry has been comparatively less explored, with only limited examples in the literature.^15b,16 Early studies in the 1980s by Meyer^17a and Collman^17b,c provided initial insights into homogeneous electrochemical AO; however, significant recent advancements by researchers such as Smith¹⁸ and Nakajima,¹⁹ employing ruthenium polypyridyl complexes, have renewed interest in molecular AO catalysis. Further expanding homogeneous AO catalysis, significant contributions from Bullock et al.,¹⁶ Chirik et al.,^20,21 and others²² have demonstrated catalytic homolytic N–H bond cleavage using organic radicals and metal complexes based on molybdenum,^{20,21,23–25} manganese,²⁶ and ruthenium.^27,28 Nevertheless, it is essential to investigate catalysts made of metals that are plentiful on Earth to improve sustainability and practical feasibility.

Motivated by biological nitrification in comammox bacteria²⁹ and the nitrogen-fixation role of the iron–molybdenum cofactor (FeMoco) in nitrogenase,^30,31 Peters and co-workers³² have recently reported iron-based molecular catalysts for AO. The results point to high-valent iron–amido species as precursors to N–N bond formation, emphasizing iron's potential as an accessible catalyst for ammonia oxidation.

Motivated by these developments, our research investigates explicitly iron complex featuring salen ligand, renowned for their robust catalytic performance in various oxidation reactions, including olefin epoxidation and benzylic oxidation.³³ Leveraging these established catalytic properties, we hypothesize that iron–salen complexes can efficiently catalyze ammonia oxidation under chemical oxidation conditions. Herein, we present our recent findings demonstrating successful ammonia oxidation catalysis using a high-spin iron complex, marking significant progress toward realizing full potential of ammonia as a sustainable and carbon-free energy carrier. In this study, ammonium triflate (NH₄OTf) was employed as a highly soluble and convenient ammonia source for homogeneous oxidation in non-aqueous media. To examine the effect of ammonia source on catalytic performance, ammonium nitrate and gaseous NH₃ (1 M MeCN solution) were also tested under comparable conditions, providing consistent results that confirm the intrinsic catalytic activity of the system.

Experimental section

Details of materials, characterization techniques, and DFT methods are provided in the SI. The Schiff base ligand was synthesized according to a reported procedure.³⁴

Synthesis of Fe^III-complex, [Fe^III(L)Cl] (1)

Schiff base ligand H₂L (1.93 g, 5 mmol) was dissolved in 30 mL of ethanol in a round bottom flask. A solution of FeCl₃·6H₂O (1.24 g, 5 mmol) in 15 mL of ethanol was added to the previous solution and stirred for 1–2 hour at room temperature. The colour of the solution changed from yellow to dark brown colour. The reaction mixture was filtered, and the filtrate was dried under vacuum. The crude product was dissolved in a small amount of dichloromethane and slowly layered with n-hexane. After 1 week in the air at room temperature, complex 1 was obtained as a needle shape dark brown crystalline solid (yield 2.13 g, 87.29%). Anal. calc. for [C₂₂H₂₄N₂O₄FeCl] (%) C, 56.01; H, 5.13; N, 5.94; found: C, 59.12; H, 5.67; N, 6.27. UV-vis (MeCN) [λ_max, nm]: 269, 305, 358, 518. FTIR (cm⁻¹): 480, 549, 1391, 1376, 1473, 1550, 1617, 2935, 2853, 3062.

Catalytic ammonia oxidation reactions

The catalytic oxidation of ammonia was evaluated using an Fe^III-complex under oxidative conditions, following the benchmark procedures reported by Nakajima et al.¹⁹ All manipulations were carried out under an inert atmosphere. Catalytic reactions were performed in a 50 mL borosilicate glass batch reactor equipped with gas inlet/outlet ports and a liquid injection port (Fig. S7). The gas inlet was connected to a mass flow controller (Bronkhorst) to regulate helium flow, while the outlet was directly connected to a gas chromatograph (GC) for on-line monitoring of gaseous products.

In a typical experiment, tris(4-bromophenyl)aminium hexachloroantimonate, [(p-BrC₆H₄)₃N˙]SbCl₆ (735 mg, 0.90 mmol), ammonium triflate (NH₄OTf, 501 mg, 3.0 mmol) and anhydrous acetonitrile (15 mL), previously degassed under helium were introduced into the reactor. In the case of gaseous ammonia, 1 M NH₃ solution in MeCN was injected after purging. The reactor was purged with helium using the mass flow controller until residual gases were completely removed. The reaction was initiated by injecting a degassed MeCN solution of the Fe^III-complex (0.01, 0.05, 0.25, 0.5 mmol), followed by 2,4,6-collidine (0.40 mL, 3.0 mmol), through the liquid injection port using a gas-tight syringe (Fig. S7). The mixture was stirred magnetically at room temperature for 1 h under helium atmosphere. The evolved N₂ was continuously monitored and quantified using a Shimadzu Nexis GC-2030 equipped with a BID detector and Micropacked ST column (2.0 m × 1.0 mm ID) with helium as the carrier gas. The N₂ yield was calculated by comparing the integrated peak area with that of a calibrated standard under identical conditions.

Results and discussion

Synthesis and structural analysis

The Schiff base ligand H₂L, trans-1,2-bis(2-hydroxy-3-methoxyphenyl-methaniminyl)cycrohexane, was synthesized following a previously reported method.³⁴ The ethanolic solution of the free base ligand was slowly mixed with FeCl₃·6H₂O solution in ethanol over 30 minutes. The resulting mixture was stirred for 1–2 hours at 60 °C, changing the solution color from yellow to a dark brown. The solvent was then evaporated under vacuum, yielding the isolated penta-coordinated Fe^III-complex (1).

The IR spectra of the free Schiff base ligand H₂L, and Fe^III-complex (1) were recorded in the solid state (Fig. S1). The spectra reveal bands at 3011 cm⁻¹, and 3062 cm⁻¹, respectively, corresponding to sp² C–H stretching vibrations.³⁴ Another set of bands appeared at 2939 cm⁻¹ (H₂L), and 2935 cm⁻¹ (1) attributed to sp³ C–H stretching vibrations. The characteristic imine stretching vibration, ν(C=N), was observed at 1624 cm⁻¹ in the free ligand (H₂L). Upon coordination with Fe^III, this band shifted to 1617 cm⁻¹ in Fe^III-complex (1), indicating coordination through the imine nitrogen (–C=N).³⁴ The ν(C=C) stretching vibrations appeared in the range of 1550–1474 cm⁻¹. Additionally, the B3LYP/DFT approach was used to calculate the IR spectra and identify vibrational bands (Table S2). The calculated vibrational frequencies closely match the experimental data, with slight differences due to DFT being performed on isolated gas-phase molecules while the experimental data reflect solid-state material. Moreover, the spectrum of Fe^III-complex (1), in comparison to that of the free ligand, displays the emergence of two bands at 540 and 480 cm⁻¹, corresponding to Fe–N and Fe–O stretching vibrations, respectively, thereby confirming coordination of the ligand to the metal center.

The electronic absorption properties of both the Schiff base ligand precursor (H₂L) and its Fe^III-complex (1) were examined using UV–Visible spectroscopy. Measurements were conducted in acetonitrile (3.0 μM) at 298 K, with absorbance recorded over the 200–800 nm range (Fig. 1). The spectra obtained for both H₂L and Fe^III-complex (1) are consistent with literature-reported data for related systems.³⁵ The UV-vis spectrum of Fe^III-complex (1) exhibits two strong absorption bands at 269 nm and 305 nm, which are also observed in the free ligand spectrum at 272 nm and 334 nm. These bands are attributed to π → π* and intra-ligand (IL) transitions within the aromatic rings and azomethine (C=N) units of the Schiff base framework. These transitions, arising from the delocalized π-electron system, are only slightly shifted upon metal coordination, indicating a minor electronic perturbation due to complexation. Importantly, two additional low-intensity bands appear at 358 nm and 518 nm in the spectrum of Fe^III-complex (1), which are absent in the free ligand. These are assigned to ligand-to-metal charge transfer (LMCT) transitions, specifically from the non-bonding orbitals of the phenolate oxygen atoms to the partially vacant d-orbitals of the high-spin Fe(III) center.³⁵ Time-dependent density functional theory (TDDFT) calculations further support the spectral assignments by providing a frontier molecular orbital (FMO) interpretation consistent with the experimental data. The computed spectrum closely matches the experimental transitions in both energy and intensity. Notably, the LMCT bands at 358 nm and 518 nm are attributed to HOMO−1 → LUMO and HOMO → LUMO transitions, respectively, involving Fe-based donor orbitals (d_z² and d_x²−y²) and ligand-centered π* acceptor orbitals. The higher-energy bands at 269 nm and 305 nm are similarly attributed to π → π* and IL transitions, also corresponding to HOMO−1 → LUMO excitations within the ligand framework. Thus, the combined UV-vis spectroscopic and TDDFT analyses confirm successful metal coordination and are consistent with a five-coordinate Fe(III) center in a square pyramidal geometry.

Fig. 1 Electronic absorption spectra of the ligand H₂L (red line) and the Fe^III-complex (1) (blue line) in acetonitrile at 298 K, with TDDFT-calculated oscillator strengths for Fe^III-complex (1) shown as vertical blue lines (right axis).

The XPS spectrum of the Fe^III-complex (1) shows distinct peaks corresponding to Fe 2p orbitals, characteristic of the Fe^(III) oxidation state (Fig. 2A). The spectrum reveals Fe 2p_3/2 and Fe 2p_1/2 peaks at binding energies consistent with Fe(III), typically found at 711.91 eV for Fe 2p_3/2 and 723.47 eV for Fe 2p_1/2. Additionally, intense satellite peaks at 715.40 and 729.10 eV accompany the main Fe 2p signals, arising from shake-up transitions and ligand to metal charge transfer (LMCT), indicative of strong Fe–N/O covalency and a high-spin electronic configuration. These findings are consistent with literature-reported XPS features for high-spin Fe^III in nitrogen- and oxygen-donor ligand environments.³⁶ The spectral data thus validate the successful formation and oxidation state integrity of the Fe^III center in the synthesized complex.

Fig. 2 Spectroscopic characterization of Fe^III-complex (1): (A) XPS spectrum highlighting the Fe 2p core-level region; (B) Mössbauer spectrum recorded at 295 K.

Mössbauer spectroscopy provides direct evidence for the electronic configuration of the Fe center. The spectrum of Fe^III-complex (1) (Fig. 2B) exhibits a well-defined quadrupole-split doublet with parameters δ = 0.35 mm s⁻¹ and ΔEq = 0.80 mm s⁻¹. These values are characteristic of a high-spin Fe(III) (S = 5/2) center, consistent with the presence of five unpaired electrons occupying the majority-spin d orbitals. The relatively large isomer shift reflects the higher electron density at the iron nucleus typical of Fe(III), while the moderate quadrupole splitting indicates an asymmetric ligand field around the metal center. The clear doublet without additional hyperfine features confirms a single electronic environment, supporting the assignment of a uniform high-spin state. These results are in line with other reported high-spin Fe(III) Schiff-base complexes, further validating the oxidation state and spin configuration of Fe^III-complex (1).³⁷

Crystallographic characterization

Single crystals of the Fe^III-complex (1) suitable for X-ray crystallographic analysis were successfully grown by the slow diffusion of n-hexane vapor into a dichloromethane solution of the complex at ambient temperature under aerobic conditions. The crystal structure was unambiguously determined by single-crystal X-ray diffraction analysis. The Fe^III-complex (1) crystallizes in the triclinic space group P1̄, possessing unit cell parameters a = 12.1450(8) Å, b = 15.8095(10) Å, c = 16.2008(10) Å, α = 87.8740(10)°, β = 82.5000(10)°, γ = 68.3490(10)°, a unit cell volume (V) of 2164.19(11) ų with Z = 4 molecules per unit cell, measured at T = 100 K.

The asymmetric unit of the Fe^III-complex (1) comprises a central Fe^III ion coordinated by a planar tetradentate di-anionic Schiff base ligand (L²⁻) with an N₂O₂ donor atom and an apical chloride ligand as illustrated in Fig. 3. The coordination environment around the Fe^III center can be best described as a distorted square pyramidal (sqp) geometry, confirmed by the geometric parameter τ, calculated using the relationship τ = |β − α|/60, where β and α are the largest and second largest coordination angles, respectively. For this complex, the angles β = N1–Fe1–O2 and α = N2–Fe1–O1 yield a τ value of 0.34, indicative of a moderate distortion from ideal square pyramidal geometry (where τ = 0 corresponds to ideal square pyramidal, and τ = 1 represents ideal trigonal bipyramidal geometry).

Fig. 3 Molecular structure (at 100 K) of (A) Fe^III-complex (1). The thermal ellipsoids of the ORTEP representation are set at 50% probability. The solvated water molecule has been omitted for clarity.

In the distorted square pyramidal arrangement, the Fe^III ion is displaced out of the mean basal plane formed by C₁₄N₂O₂ ligand core (Δ₁₈^Fe) by 0.64 Å towards the axial chloride ligand, reinforcing the square pyramidal assignment. The average bond lengths within the coordination sphere: Fe–N = 2.0925(3) Å, Fe–O = 1.8995(2) Å, and Fe–Cl = 2.2273(8) Å, align well with expected values for Fe^III complexes in analogous coordination environments.³⁸ Additionally, the asymmetric unit contains a discrete water molecule, which resides within an open compartment formed by the coordinated ligand framework. This water molecule participates in stabilizing hydrogen-bonding interactions with the methoxy (–OCH₃) oxygen atoms of the ligand, contributing to the overall crystal packing stability.

The structural representation of Fe^III-complex (1) is presented in Fig. 3, showing displacement ellipsoids drawn at the 50% probability level. Molecular arrangements within the crystal lattice and detailed atom-numbering schemes are provided in Fig. S2 and S3, respectively. Comprehensive crystal data, including full data collection parameters and bond lengths and angles, are tabulated in Tables S1 and S3–S5, respectively.

Chemical ammonia oxidation

At the onset of this research, we evaluated the ammonia oxidation activity of an Fe^III-complex under oxidative conditions following the standard procedure reported by Nakajima et al.¹⁹ Encouragingly, we discovered that the Fe^III-complex (1) exhibited ammonia oxidation potential at room temperature (Fig. 4, Table 1). The reactions were conducted using ammonium trifluoromethanesulfonate (NH₄OTf, 3.0 mmol) as the ammonia source, tris(4-bromophenyl)ammoniumyl hexachloroantimonate (0.90 mmol) as the oxidant, and 2,4,6-collidine (3.0 mmol) as a base. Varying catalytic amounts of the Fe^III-complex (0.010, 0.05, 0.25, and 0.50 mmol) were tested in 15 mL of acetonitrile at room temperature for 2 hours (Fig. 4, Table 1). The experiments were conducted in a 50 mL reactor equipped with an inlet helium gas port connected to a mass flow controller (Bronkhorst), a silicone rubber septum port for 2,4,6-collidine injection, and an outlet gas port connected to a gas chromatograph (Shimadzu Nexis GC-2030) with a BID detector for on-line nitrogen gas analysis.

Fig. 4 Ammonia oxidation (AO) activity under varying conditions. (A) Time-dependent N₂ production rates at different catalyst loadings (0.5, 0.25, 0.05, and 0.01 mmol). (B) Total amount of N₂ evolved at each catalyst loading. (C) N₂ production rates using different ammonia sources/solvents (NH₄NO₃, NH₃, and MeOH). (D) Total amount of N₂ evolved for each nitrogen source/solvent. In part A and C, the N₂ production rate is normalized to the catalyst mass.

Table 1 Catalytic ammonia oxidation with Fe^III-complex (1)

Entry	Catalyst amount (mmol)	NH3 source	Solvent	N2^a (equiv.)	N2^b (%)
Reactions of ammonium triflate (3.0 mmol), oxidant (0.9 mmol), and base (3.0 mmol) were carried out in the presence of iron catalyst under an He atmosphere at room temperature for 2 h.a Equivalents are produced N2 per Fe atom.b Yields are based on oxidant. Selectivity for N2 >99%, determined by on-line GC (BID detector).
1	0.5	NH4OTf	MeCN	0.05	17.83
2	0.25	NH4OTf	MeCN	0.18	31.20
3	0.05	NH4OTf	MeCN	2.20	73.3
4	0.01	NH4OTf	MeCN	3.01	20.05
5	0.05	NH4OTf	MeOH	0.48	16.05
6	0.05	NH4NO3	MeCN	0.61	20.53
7	0.05	NH3(MeCN)	MeCN	0.73	24.36
8	0.00	NH4OTf	MeCN	0.00	0.00

---

Prior to initiating the catalytic investigation, the reactor's headspace was purged with helium to eliminate air. Subsequently, the 2,4,6-collidine was injected, and the resultant N₂ was quantified by directly analyzing the gas stream at a flow rate of 20 mL min⁻¹. Under optimum conditions (0.05 mmol catalyst), the reaction yielded 0.109 mmol N₂, equating to 2.20 equivalents per Fe atom and attaining a 73% yield relative to the oxidant. At higher catalyst loadings, decreased N₂ production is attributed to possible catalyst–catalyst interactions such as aggregation or dimerization under oxidative conditions, which reduce the number of active sites.^39–41

Subsequent experiments demonstrated considerable effects resulting from changes in ammonia supplies and solvents. Substituting NH₄OTf with NH₄NO₃ led to significantly diminished N₂ production (0.61 equivalents and a 20.53% yield, Table 1), underscoring the pivotal impact of ammonia source selection on catalytic efficacy. NH₄OTf presumably provided superior solubility and enhanced contact with the Fe(III) catalyst relative to other investigated ammonium salts, hence augmenting overall catalytic efficacy. The selection of solvent significantly affected the results; acetonitrile was much more effective than methanol, as seen by a considerable reduction in catalytic activity with methanol (0.48 equivalents and a 16.05% yield). The higher performance with acetonitrile may stem from its advantageous dielectric constant, better catalyst solubility, and optimum stability of reactive intermediates.

Furthermore, trials conducted in the absence of a base resulted in negligible N₂ generation, so unequivocally demonstrating the essential function of the base in promoting the ammonia oxidation process, presumably by efficiently deprotonating intermediates. These thorough investigations emphasize the ideal catalytic conditions attained with NH₄OTf as the ammonia source, acetonitrile as the solvent, and a catalyst concentration of 0.05 mmol, effectively balancing high catalytic efficiency and significant yield, thereby underscoring the potential applicability of the Fe(III) complex in practical ammonia oxidation processes. Gas-phase selectivity, showed N₂ as the sole nitrogen-containing product with no detectable N₂O, NO, or NO₂, giving >99% selectivity toward N₂ under all conditions.

For comparison, Nakajima and co-workers reported Ru-based catalysts capable of oxidizing ammonia to N₂ with up to 11.9 equiv. per Ru atom (79% yield), and mechanistic studies supported a nitride–nitride coupling pathway.¹⁹ More recently, Toda and co-workers demonstrated Mn-salen complexes as the first earth-abundant homogeneous systems for ammonia oxidation, producing up to 17.1 equiv. of N₂ per Mn atom under chemical conditions.⁴² Against this backdrop, the observed activity of the Fe(III) complex underscores its potential as a sustainable alternative for ammonia oxidation catalysis.

DFT-based electronic structure and mechanistic analysis of ammonia oxidation

Density functional theory (DFT) calculations were performed to elucidate both the electronic structure of the Fe^III-complex (1) and the mechanistic pathway of ammonia oxidation to dinitrogen. Geometry optimizations for the doublet (S = 1/2), quartet (S = 3/2), and sextet (S = 5/2) spin states, followed by frequency analyses, confirmed that all optimized structures correspond to true minima, with computed geometries closely matching experimental data. Spin-density analysis revealed predominant localization of unpaired electrons on the Fe center, confirming its +III oxidation state and high-spin (sextet) configuration (Fig. S4, Table S3). To further investigate the catalytic mechanism of ammonia oxidation to dinitrogen, we performed density functional theory (DFT) calculations at the B3LYP level, incorporating Grimme's D3 dispersion correction and the SMD solvation model with MeCN as the solvent. The reaction was modelled using an iron-based catalyst and evaluated through two alternative mechanistic routes, designated pathway A and pathway B. Each pathway comprises a sequence of oxidation (electron transfer) and deprotonation (proton transfer) steps, and the thermodynamic feasibility of each transformation was assessed by calculating the corresponding free energy changes (ΔG). The computed catalytic cycles are depicted in Fig. 5A and B, corresponding to pathways A and B, respectively, with the free energy changes (ΔG_rxn) annotated for each step. Among the two, pathway B (the red pathway, which proceeds without ligand exchange) is thermodynamically more favorable, as shown in Fig. 6.

Fig. 5 Proposed catalytic cycle for ammonia oxidation by (A) the complex via a ligand exchange pathway and (B) without ligand exchange. Free energy changes (ΔG) for individual steps are given in kcal mol⁻¹. Oxidation steps are highlighted in red and deprotonation steps in blue.

Fig. 6 Calculated free energy profile (kcal mol⁻¹) for the conversion of NH₃ to N₂ (A to O) via the catalytic cycle in Fig. 6, mediated by the Fe^III-complex (1). The ligand exchange pathway is shown in green, and the pathway without ligand exchange is shown in red. Free energies for individual steps were computed in SMD-MeCN continuum solvent.

This preference arises from its avoidance of high-energy intermediates and a more accessible overall energy landscape. Accordingly, the following mechanistic discussion focuses primarily on pathway B. For each intermediate, geometry optimizations were carried out by considering three possible spin states of the Fe(III) center in low-spin, intermediate-spin, and high-spin, and the reported free energy values correspond to the most stable (lowest-energy) spin configuration. Furthermore, detailed electronic structure analyses were performed for all key intermediates (⁶I to ⁶XV) involved in the multistep proton and electron transfer events of mechanistic pathway B (Fig. 7).

Fig. 7 DFT-computed mechanism for Fe^III-complex (1) catalyzed ammonia oxidation, showing key intermediates (A to O), spin states, spin density distributions, and free energy changes (in kcal mol⁻¹). The calculations highlight electron and proton transfer steps and the evolution of metal- and ligand-centered radical character throughout the catalytic cycle. All intermediates shown were obtained from DFT calculations.

Additionally, a possible bimetallic N–N coupling pathway was evaluated. Four spin states: singlet, triplet, septet, and 11-et were examined, with the S = 11 state identified as the most stable configuration for the bimetallic intermediate. In this high-spin configuration, both Fe centres display parallel spins, resulting in an overall high-spin bimetallic species. As shown in the Fig. S5, the nucleophilic attack of NH₃ on the Fe(IV)–NH· species is highly exergonic (ΔG = −32.39 kcal mol⁻¹), whereas the bimetallic amido-coupling pathway is considerably less favorable (ΔG = 4.73 kcal mol⁻¹). These findings indicate that the mononuclear nucleophilic attack mechanism is energetically preferred, while the bimetallic pathway is not favoured under the studied conditions.

Coordination of NH₃ to the catalyst (steps A to B). Ammonia coordinates to the Fe(III) center of the complex [Fe^III(L)Cl], forming the ammine complex [Fe^III(L)(NH₃)Cl]. This is a straightforward ligand coordination event without any change in oxidation state or spin multiplicity. Spin-density analysis confirms that the Fe center remains in a high-spin Fe(III) state, with the spin density slightly increasing from +4.07 in A to +4.12 in B, consistent with a sextet (S = 5/2) ground state. The process is thermodynamically favorable, with a computed free-energy change of −3.71 kcal mol⁻¹, suggesting spontaneous NH₃ binding under standard conditions. This coordination marks the first step in substrate activation and sets the stage for subsequent redox and bond-forming events in the catalytic pathway.

First oxidation/deprotonation (steps B to D). The first electron/proton transfer step begins with the oxidation of [Fe^III(L)(NH₃)Cl], a paramagnetic d⁵ complex. In the step from B → C, the complex [Fe^III(L)(NH₃)Cl] undergoes a one-electron oxidation, forming [Fe^III(L⁺˙)(NH₃)Cl]⁺. Spin-density analysis reveals that the oxidation is ligand-centered, as evidenced by the emergence of significant negative spin density on the ligand backbone (−0.25), indicating formation of a ligand π–cation radical (L⁺˙) with β-spin character. The iron center retains a high positive spin density, decreasing only slightly from +4.12 (⁶II) to +3.81 (⁵III), confirming that it remains in the Fe(III) oxidation state. This shift in spin distribution alters the magnetic properties of the intermediate (⁵III). Unlike ⁶II, which is a sextet (S = 5/2) due to high-spin Fe(III), intermediate ⁵III exhibits a quintet (S = 2) ground state. This arises from antiferromagnetic coupling between the high-spin Fe(III) center (S = 5/2, α-spin) and the ligand radical (S = 1/2, β-spin), resulting in a net spin of S = 2. The NH₃ ligand remains essentially spin-inactive, with minimal spin density (0.12), confirming it does not participate in the redox event. The overall oxidation is exergonic, with a free-energy change of −3.80 kcal mol⁻¹, suggesting a spontaneous nature under standard conditions. This step represents a key early redox activation of the ligand framework, setting up subsequent proton and electron transfer events critical for N–N bond formation in the catalytic cycle.

In the C → D step, the first deprotonation of the formally [Fe^III(L⁺˙)(NH₃)Cl]⁺ complex by 2,4,6-collidine is calculated to be endergonic, with a free-energy change of +19.80 kcal mol⁻¹, indicating that this step requires external driving force under reaction conditions. The proton abstraction from the coordinated NH₃ generates an Fe(III)-aminyl radical species [Fe^III(L)(NH₂˙)Cl], depending on the resulting electronic distribution. Spin-density analysis of the lowest-energy intermediate (³IV) reveals significant unpaired electron density on both the Fe center (+1.83) and the NH₂ ligand (+0.17), with nearly zero spin density on the ligand backbone, confirming that the ligand radical character (L⁺˙) has been neutralized. This occurs through internal electron redistribution, where electron density from the more electron-rich NH₂ group compensates for the ligand radical, restoring the ligand to a closed-shell (L) state. As a result, the radical character shifts from the ligand backbone to the NH₂ group, forming an aminyl radical. These features clearly support an electronic structure best described as an Fe^(III)–aminyl radical, where the iron remains in the Fe(III) oxidation state. The ground state of this intermediate is a triplet (S = 1), arising from ferromagnetic coupling between the low-spin Fe(III) center (S = 1/2) and the aminyl radical (S = 1/2) on the NH₂ ligand. This configuration reflects a substantial reorganization of spin density relative to intermediate ⁵III and highlights the role of the deprotonation step in converting a ligand-centered radical into a metal–ligand radical pair, setting the stage for subsequent redox and bond-forming processes in the catalytic pathway.

Second oxidation/deprotonation (steps D to F). The second electron/proton transfer step begins with the oxidation of Fe(III)-aminyl radical, [Fe^III(L)(NH₂˙)Cl], a low spin d⁵ complex. The calculated free energy change for this oxidation (step D → E in Fig. 4) is exergonic by −4.30 kcal mol⁻¹. Spin density analysis (Fig. 5a) indicates that the oxidation is primarily ligand-centered, leading to the formation of a [Fe^III(L⁺˙)(NH₂˙)Cl]⁺ species. The lowest-energy electronic configuration corresponds to an overall doublet spin state (2S + 1 = 2), with the Fe^III center adopting a low-spin configuration (S = 1/2). The resulting spin distribution suggests ferromagnetic coupling between the aminyl radical (NH₂˙) and the low-spin Fe^III fragment. The overall electron transfer process is exergonic, with a computed free-energy change of −4.30 kcal mol⁻¹, underscoring its thermodynamic feasibility and electronic driving force.

Upon deprotonation of [Fe^III(L⁺˙)(NH₂˙)Cl]⁺ (step D → E in Fig. 4), the electronic structure undergoes an intramolecular electron redistribution process. When a proton is removed from the NH₂ ligand, the resulting electron redistribution triggers internal electron transfer from the Fe–NH₂˙ moiety to the ligand radical (L⁺˙). This internal electron transfer neutralizes the ligand radical cation, converting L⁺˙ into a neutral ligand (L). Concurrently, the iron center is oxidized from Fe(III) to Fe(IV), and the deprotonated nitrogen ligand is transformed into an iminyl radical (NH˙) species. Consequently, the final resulting complex, [Fe^IV(L)(NH˙)Cl], is best described as a low-spin Fe(IV) species coupled antiferromagnetically with an iminyl radical (NH˙) ligand, exhibiting an overall doublet (2S + 1 = 2) ground state. This interpretation aligns closely with DFT-computed spin densities, which show a moderate spin density localized at iron (+1.83), spin density at nitrogen (0.17), and essentially no residual radical character on the ligand backbone (L). The overall deprotonation and electron transfer process is endergonic, with a computed free-energy change of 11.56 kcal mol⁻¹, underscoring its thermodynamic feasibility and electronic driving force.

Third oxidation/deprotonation, and nucleophilic attack of NH₃ on iminyl nitrogen (steps F–I). Upon examining the nucleophilic attack step (F → G), intermediate ²VI ([Fe^IV(L)(NH˙)Cl]), initially in a low-spin doublet ground state, undergoes radical coupling via direct nucleophilic attack by ammonia (NH₃). The iminyl nitrogen ligand (spin density 0.17 at NH˙) in intermediate ²VI acts as an electrophilic site, facilitating the nucleophilic addition and formation of a stable N–N bond, yielding intermediate ⁶VII ([Fe^III(L)(NH–NH₃)Cl]). This coupling event transfers electron density from ammonia to the iminyl nitrogen ligand, neutralizing its radical character. The resultant electron redistribution reduces the Fe(IV) center to Fe(III), the iron transitions from a low-spin (S = 1) to a more favorable high-spin (S = 5/2, sextet) ground state, clearly indicated by the substantial increase in spin density at iron (from +1.13 to +4.06). The overall radical coupling and electron transfer process is highly exergonic, with a computed free-energy change of −32.39 kcal mol⁻¹, underscoring its strong thermodynamic and electronic driving force.

In the third oxidation step (G → H), the complex undergoes a one-electron oxidation from ([Fe^III(L)(NH–NH₃)Cl]) to ([Fe^IV(L)(NH–NH₃)Cl]⁺) which is indicated by the spin-density redistribution and electronic structural changes. Initially, intermediate ⁶VII is in a high-spin Fe(III) sextet state (spin density at Fe: +4.06), reflecting an electron-rich and stable Fe(III) center coordinated to the hydrazine-type ligand (NH–NH₃). Upon oxidation, intermediate ³VIII is formed, and the spin density at iron significantly decreases (from +4.06 to +2.42), indicating oxidation of Fe(III) to Fe(IV). The ligand remains non-radical, as evidenced by negligible spin density on nitrogen atoms (0.05–0.06). This oxidation step therefore clearly occurs at the iron center, converting it from a high-spin Fe(III) (d⁵) into a low-spin Fe(IV) (d⁴) species. Consequently, intermediate ³VIII adopts a triplet (S = 1) ground state with the Fe(IV) oxidation state. This one-electron oxidation is exergonic, with a computed free-energy change of −2.70 kcal mol⁻¹, highlighting that the step is thermodynamically feasible.

In the third deprotonation step from intermediate ³VIII ([Fe^IV(L)(NH–NH₃)Cl]⁺) to ³IX ([Fe^IV(L)(NH–NH₂)Cl]), the proton is abstracted from the hydrazine-type ligand (NH–NH₃), forming an Fe(IV)-hydrazido species. The spin density distribution shows minimal changes at the iron center (spin density slightly decreases from +2.42 in ³VIII to +1.17 in ³IX), confirming that the oxidation state (Fe(IV)) and spin delocalized over the Fe(IV) center and NH–NH₂ (spin density 0.14) moiety during this step. The nitrogen atoms also show negligible spin density (0.02), confirming no radical character arises on the ligand upon proton removal. Thus, the step H → I primarily represent a ligand-centered deprotonation without significant electron-transfer processes or spin-state transitions at the metal center. The resultant species (³IX) remains in a low-spin Fe(IV) triplet (S = 1) ground state. This step is highly thermodynamically favorable, as indicated by the large negative free-energy change (−23.90 kcal mol⁻¹), clearly demonstrating that deprotonation of the hydrazine ligand is energetically favorable and readily occurs under reaction conditions.

Fourth oxidation/deprotonation (step I to K). In the fourth oxidation step from intermediate ³IX ([Fe^IV(L)(NH–NH₂)Cl]) to intermediate ²X ([Fe^IV(L⁺˙)(NH–NH₂˙)Cl]⁺), a one-electron oxidation occurs that introduces significant radical character simultaneously on the ligand backbone and the hydrazido ligand. Spin-density analysis reveals three distinct spin centers: positive spin density on the iron center (+1.04, α-spin), substantial positive spin density on the hydrazido ligand (+0.85, α-spin), and negative spin density on the ligand backbone (−0.62, β-spin). The positive spin densities at iron and hydrazido ligand indicate a strong ferromagnetic coupling between Fe(IV) and the hydrazido radical (NH–NH₂˙). However, this combined Fe–hydrazido fragment is antiferromagnetically coupled with the ligand π–cation radical (L⁺˙), carrying opposite β-spin, resulting in partial spin cancellation and stabilization of the intermediate in a doublet (S = 1/2) ground state. The one-electron oxidation step is exergonic (−33.12 kcal mol⁻¹), remains thermodynamic feasible.

In the fourth deprotonation step (J → K), intermediate ²X ([Fe^IV(L⁺˙)(NH–NH₂˙)Cl]⁺) undergoes proton removal from the hydrazido radical ligand (NH–NH₂˙), forming intermediate ²XI ([Fe^III(L)(N–NH₂)Cl]). Spin-density analysis clearly indicates a significant redistribution of spin density, with the iron center having a positive spin density of +1.09 (α-spin), and essentially negligible spin density on both the ligand backbone (L) and the diazenido ligand (N=NH₂), indicating that the radical character previously localized on these ligands is effectively quenched upon proton loss. Thus, this deprotonation step is accompanied by internal electron transfer, which reduces the ligand radical (L⁺˙) and hydrazido radical (NH–NH₂˙) back to their closed-shell states, stabilizing a purely Fe(III)-centered species. The iron center, therefore, remains in a low-spin Fe(III) oxidation state, and the intermediate ²XI maintains a doublet (S = 1/2) ground state, now originating solely from a single unpaired electron localized primarily on Fe. The free-energy change for this deprotonation and electron redistribution step is moderately endergonic (11.29 kcal mol⁻¹), yet remains feasible under reaction conditions.

Fifth oxidation/deprotonation (step K to M). In the fifth oxidation step (K → L), intermediate ²XI ([Fe^III(L)(N–NH₂)Cl]), initially a doublet species (Fe spin density = +1.09), undergoes one-electron oxidation to form intermediate ³XII ([Fe^III(L⁺˙)(N–NH₂)Cl]⁺). Spin-density analysis of intermediate ³XII shows substantial spin density localized on the iron center (+1.24) and on the ligand backbone (+0.53), with negligible spin density on the diazenido ligand (N–NH₂). These data clearly indicate that oxidation occurs primarily on the ligand backbone, forming a ligand π–cation radical (L⁺˙), while the Fe center maintains a low-spin Fe(IV) state. Crucially, intermediate ³XII now adopts a triplet (S = 1) ground state, resulting from ferromagnetic coupling between the Fe(III) center (α-spin) and ligand radical (also α-spin). Thus, the spins on Fe and ligand align in parallel, increasing the overall spin multiplicity to a triplet. The computed free-energy change for this ligand-centered oxidation step is significantly exergonic (−8.08 kcal mol⁻¹), identifying it as thermodynamically feasible under reaction conditions.

In the deprotonation step (L → M), intermediate ³XII ([Fe^III(L⁺˙)(N–NH₂)Cl]⁺), initially in a triplet ground state (S = 1), undergoes proton removal from the diazenido ligand (N–NH₂), forming intermediate ⁵XIII ([Fe^III(L)(N=NH˙)Cl]). Spin-density analysis clearly shows a major electronic rearrangement: spin density at the iron center significantly increases from +1.24 in ³XII to +2.70 in ⁵XIII, indicating a spin switching of low-spin Fe(III) to intermediate spin (S = 3/2) Fe(III). Concurrently, the nitrogen ligand develops notable radical character (spin density = +0.98), corresponding to the formation of a diazenyl radical (N=NH˙). Negligible spin density remains on the ligand backbone (L), demonstrating the ligand π–cation radical (L⁺˙) is fully neutralized to its closed-shell form during this process. As a consequence, intermediate ⁵XIII adopts an intermediate-spin Fe(III) quintet ground state (S = 2), resulting from the ferromagnetic coupling between the Fe(III) center and the diazenyl radical ligand. The computed free-energy change for this step is significantly exergonic (−28.56 kcal mol⁻¹), highlighting the strong thermodynamic driving force favoring the formation of intermediate ⁵XIII under experimental conditions.

Sixth oxidation/deprotonation (steps M to O) and N₂ dissociation (catalytic cycle restart). In the sixth oxidation step (M → N), intermediate ⁵XIII ([Fe^III(L)(N=NH˙)Cl]), initially in a quintet ground state (S = 2), undergoes a one-electron oxidation to form intermediate ⁴XIV ([Fe^III(L⁺˙)(N=NH˙)Cl]⁺). Spin-density analysis clearly indicates oxidation occurring predominantly on the ligand backbone, reflected by the emergence of significant negative (β-spin) density on ligand L (−0.56). The iron spin density decreases slightly from +2.70 (⁵XIII) to +2.47 (⁵XIV), while spin density on the diazenyl radical ligand slightly increases from +0.98 to +1.10, consistent with a persistent radical character. Thus, intermediate ⁴XIV features three distinct spin centers: the high-spin Fe(III) (α-spin), the diazenyl radical ligand (N=NH˙, α-spin), and the ligand π–cation radical (L⁺˙, β-spin). The quartet ground state (S = 3/2) in intermediate ⁴XIV arises from strong ferromagnetic coupling between the Fe(III) center and diazenyl radical ligand (both α-spin), forming a combined high-spin fragment, which is further coupled antiferromagnetically to the ligand radical cation (β-spin), resulting the net quartet (S = 3/2) spin state. This oxidation step is nearly thermoneutral, having only a minor free-energy increase of +0.59 kcal mol⁻¹, making it energetically feasible under typical catalytic conditions.

In the sixth deprotonation step (N → O), intermediate ⁴XIV ([Fe^III(L⁺˙)(N=NH˙)Cl]⁺), initially in a quartet ground state (S = 3/2), undergoes proton removal from the diazenyl radical ligand (N=NH˙), forming intermediate ⁶XV ([Fe^III(L)(N≡N)Cl]). This proton abstraction triggers internal electron transfer, where electron density previously localized as radical character on the ligand backbone (L⁺˙) and diazenyl ligand (N=NH˙) is redistributed back onto the iron center. Consequently, both ligand radicals become closed-shell species, losing their radical nature, as evidenced by negligible spin densities on these fragments. Simultaneously, the iron center is formally switch from intermediate spin (3/2) Fe(III) to high-spin Fe(III), confirmed by a significantly increased spin density (+3.97). Intermediate ⁶XV thus adopts a high-spin sextet (S = 5/2) ground state arising from the d⁵ electronic configuration of Fe(III). The computed free-energy change for this internal electron transfer and deprotonation step is highly exergonic (−92.42 kcal mol⁻¹), underscoring a strong thermodynamic driving force stabilizing the formation of intermediate ⁶XV. This high exergonicity can be attributed to the formation of the N₂ triple bond. In the final step, the loss of dinitrogen (N₂) is exergonic, by −2.55 kcal mol⁻¹ for ⁶XV intermediate. The catalytic loop can be restarted by the detachment of the N₂ to form intermediate ⁶I.

Hence, in the overall ammonia oxidation catalytic cycle, the spin-state transition from low-spin to intermediate-spin at the Fe centre plays a crucial role in facilitating N–N bond formation by enhancing both metal–ligand orbital interactions and electronic flexibility. In the low-spin state, Fe d-orbitals are mostly paired, which restricts σ/π overlap with the nitrogen donors. Upon transition to an intermediate-spin configuration, partial population of the antibonding eg* orbitals increase Fe–N/O covalency and allows spin density delocalization onto the –NH_x ligands. This delocalization promotes radical-type coupling between Fe–NH· intermediates or nucleophilic attack of NH₃ on electrophilic Fe–NH species, thereby lowering the barrier for N–N bond formation.^43,44 Furthermore, Zott and Peters have demonstrated that spin-state flexibility is crucial for efficient ammonia oxidation catalysis, as it allows the Fe center to access multiple oxidation and spin states during turnover. This flexibility enables the Fe center to mediate multielectron transformations with reduced reorganization energy, thereby promoting faster and more efficient N–N bond formation.³² Accordingly, the low-spin to intermediate-spin transition in our system provides an optimal balance between redox flexibility, Fe–N covalency, and spin delocalization, all of which are essential for effective Fe-mediated ammonia oxidation.

Conclusions

In conclusion, we report a high-spin Fe^III-complex with a trans-1,2-bis(2-hydroxy-3-methoxyphenylmethaniminyl)cyclohexane salen-type ligand that efficiently catalyzes ammonia oxidation to dinitrogen under oxidative conditions. In the presence of a triarylaminium oxidant and 2,4,6-collidine base, the catalyst delivers up to 2.20 equivalents of N₂ per iron center. Integrated spectroscopic, crystallographic, and computational analyses establish the structural and electronic features of the complex and reveal a nucleophilic attack pathway for N–N bond formation. Together, these findings advance molecular ammonia oxidation catalysis with earth-abundant iron, offering key mechanistic insight for the rational design of future catalysts and reinforcing the potential of ammonia as a practical carbon-free energy carrier.

Conflicts of interest

The Authors declare no competing financial interest.

Data availability

The related data have been provided in the form of supplementary information (SI). Supplementary information: the additional, experimental details, characterizations and computational details are available in the SI. See DOI: https://doi.org/10.1039/d5cy01107h.

CCDC 2384217 contains the supplementary crystallographic data for this paper.⁴⁵

Acknowledgements

This research was supported by the King Fahd University of Petroleum and Minerals (KFUPM). The authors gratefully acknowledge the support provided by the Deanship of Research (DR) at King Fahd University of Petroleum & Minerals (KFUPM) for funding this work through project No.: INHT2410.

References

1. A. Grinberg Dana, O. Elishav, A. Bardow, G. E. Shter and G. S. Grader, Nitrogen-based fuels: a power-to-fuel-to-power analysis, Angew. Chem., Int. Ed., 2016, 55, 8798–8805.
2. O. Elishav, B. Mosevitzky Lis, E. M. Miller, D. J. Arent, A. Valera-Medina, A. Grinberg Dana, G. E. Shter and G. S. Grader, Progress and prospective of nitrogen-based alternative fuels, Chem. Rev., 2020, 120, 5352–5436.
3. K. Morita, Ammonia − a fuel of the future?, available at: https://www.euractiv.com/section/energy-environment/opinion/ammonia-a-fuelof-the-future/, (accessed 10 February 2021).
4. A. H. Tullo, Is ammonia the fuel of the future? Industry sees the agricultural chemical as a convenient means to transport hydrogen, available at: https://cen.acs.org/business/petrochemicals/ammonia-fuel-future/99/i8, (accessed 2 September 2025).
5. P. L. Dunn, B. J. Cook, S. I. Johnson, A. M. Appel and R. M. Bullock, Oxidation of ammonia with molecular complexes, J. Am. Chem. Soc., 2020, 142, 17845–17858.
6. J. G. Chen, R. M. Crooks, L. C. Seefeldt, K. L. Bren, R. M. Bullock, M. Y. Darensbourg, P. L. Holland, B. Hoffman, M. J. Janik, A. K. Jones, M. G. Kanatzidis, P. King, K. M. Lancaster, S. V. Lymar, P. Pfromm, W. F. Schneider and R. R. Schrock, Beyond fossil fuel–driven nitrogen transformations, Science, 2018, 360, 6611.
7. X. Zhu, J. Du, Z. Yu, Y.-B. Cheng and Y. Wang, NOx emission and control in ammonia combustion: state-of-the-art review and future perspectives, Energy Fuels, 2024, 38, 43–60.
8. N. Gruber and J. N. Galloway, An Earth-system perspective of the global nitrogen cycle, Nature, 2008, 451, 293–296.
9. C. Ferousi, S. H. Majer, I. M. DiMucci and K. M. Lancaster, Biological and bioinspired inorganic N–N bond-forming reactions, Chem. Rev., 2020, 120, 5252–5307.
10. X. Zhang, B. B. Ward and D. M. Sigman, Global nitrogen cycle: critical enzymes, organisms, and processes for nitrogen budgets and dynamics, Chem. Rev., 2020, 120, 5308–5351.
11. J. J. Warren, T. A. Tronic and J. M. Mayer, Thermochemistry of proton-coupled electron transfer reagents and its implications, Chem. Rev., 2010, 110, 6961–7001.
12. J. Zhao, A. S. Goldman and J. F. Hartwig, Oxidative addition of ammonia to form a stable monomeric amido hydride complex, Science, 2005, 307, 1080–1082.
13. N. V. Rees and R. G. Compton, Carbon-free energy: a review of ammonia- and hydrazine-based electrochemical fuel cells, Energy Environ. Sci., 2011, 4, 1255–1260.
14. E. K. Dann, E. K. Gibson, R. H. Blackmore, C. R. A. Catlow, P. Collier, A. Chutia, T. E. Erden, C. Hardacre, A. Kroner, M. Nachtegaal, A. Raj, S. M. Rogers, S. F. R. Taylor, P. Thompson, G. F. Tierney, C. D. Zeinalipour-Yazdi, A. Goguet and P. P. Wells, Mechanistic insights into ammonia oxidation over transition-metal catalysts, Nat. Catal., 2019, 2, 157–163.
15. (a) M. Usman, A. Ali, Z. H. Yamani and M. N. Shaikh, Catalytic pathways for efficient ammonia-to-hydrogen conversion towards a sustainable energy future, Sustainable Energy Fuels, 2024, 8, 5329–5351; (b) M. Usman, A. Ali, A. Jedidi, A. Ajeebi, M. M. Hossain, K. M. Yau, H. Alghamdi, M. A. Aziz and M. N. Shaikh, Rare-earth-metal promoters (La, Ce, Nd, Sm) on nickel-supported Al₂O₃ catalysts for ammonia decomposition, Fuel, 2025, 396, 135272.
16. (a) P. L. Dunn, B. J. Cook, S. I. Johnson, A. M. Appel and R. M. Bullock, Oxidation of ammonia with molecular complexes, J. Am. Chem. Soc., 2020, 142, 17845–17858; (b) J. Li, F. Zhang, H. Xiong, Y. Cai and B. Zhang, Molecular catalysts for electrocatalytic ammonia oxidation, Sci. China: Chem., 2024, 67, 3976–3993.
17. (a) M. S. Thompson and T. J. Meyer, Oxidation of coordinated ammonia, J. Am. Chem. Soc., 1981, 103, 5577–5579; (b) J. P. Collman, J. E. Hutchison, M. S. Ennis, M. A. Lopez and R. Guilard, Reduced nitrogen hydride complexes of a cofacial metallodiporphyrin and their oxidative interconversion, J. Am. Chem. Soc., 1992, 114, 8074–8080; (c) J. P. Collman, J. E. Hutchison, M. S. Ennis, M. A. Lopez and R. Guilard, Reduced nitrogen hydride complexes of a cofacial metallodiporphyrin and their oxidative interconversion: an analysis of ammonia oxidation and prospects for a dinitrogen electroreduction catalyst based on cofacial metallodiporphyrins, J. Am. Chem. Soc., 1992, 114, 8074–8080.
18. F. Habibzadeh, S. L. Miller, T. W. Hamann and M. R. Smith III, Homogeneous electrocatalytic oxidation of ammonia to N₂ under mild conditions, Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 2849–2853.
19. K. Nakajima, H. Toda, K. Sakata and Y. Nishibayashi, Ruthenium-catalysed oxidative conversion of ammonia into dinitrogen, Nat. Chem., 2019, 11, 702–709.
20. G. W. Margulieux, Z. R. Turner and P. J. Chirik, Synthesis and ligand modification chemistry of a molybdenum dinitrogen complex: redox and chemical activity of a bis(imino)pyridine ligand, Angew. Chem., Int. Ed., 2014, 53, 14211–14215.
21. M. J. Bezdek, S. Guo and P. J. Chirik, Coordination-induced weakening of ammonia, water and hydrazine X–H bonds in a molybdenum complex, Science, 2016, 354, 730–733.
22. (a) J. A. Gutiérrez-Orgaz, M. Nicaso, A. M. Beiler, M. Martínez-Belmonte, J. Benet-Buchholz, M. Segado-Centellas, X. Sala, C. Bo and A. Llobet, Exploring copper β-diketiminate complexes for heterogeneous ammonia oxidation anchored on graphitic surfaces via CH–π and π–π interactions, Inorg. Chem. Front., 2025, 12, 7234–7249; (b) H. Xiong, J. Yang, J. Li, Y. Cai, F. Zhang, J.-Y. Chen, R.-Z. Liao, L. Sun and B. Zhang, Electrochemical Ammonia Oxidation Catalyzed by a Ruthenium Complex with a Dangling Sulfonate Group, ACS Catal., 2025, 15, 8633–8642; (c) T. Gupta, Sanyam, S. Saraswat, A. Mondal and B. Mondal, Electrochemical ammonia oxidation with a homogeneous molecular redox mediator, Chem. Sci., 2025, 16, 14377–14383; (d) J. Li, X. Shi, F. Zhang, X. Lu, Y. Zhang, R. Liao and B. Zhang, Electrocatalytic Ammonia Oxidation by a Ruthenium Complex Bearing a 2,6-Pyridinedicarboxylate Ligand, JACS Au, 2025, 5, 1812–1821.
23. G. W. Margulieux, M. J. Bezdek, Z. R. Turner and P. J. Chirik, Ammonia activation, H₂ evolution and nitride formation from a molybdenum complex with a chemically and redox non-innocent ligand, J. Am. Chem. Soc., 2017, 139, 6110–6113.
24. P. Bhattacharya, Z. M. Heiden, E. S. Wiedner, S. Raugei, N. A. Piro, W. S. Kassel, R. M. Bullock and M. T. Mock, Ammonia oxidation by abstraction of three hydrogen atoms from a Mo–NH₃ complex, J. Am. Chem. Soc., 2017, 139, 2916–2919.
25. M. J. Bezdek and P. J. Chirik, Interconversion of molybdenum imido and amido complexes by proton-coupled electron transfer, Angew. Chem., Int. Ed., 2018, 57, 2224–2228.
26. B. J. Cook, S. I. Johnson, G. M. Chambers, W. Kaminsky and R. M. Bullock, Triple hydrogen atom abstraction from Mn–NH₃ complexes results in cyclophosphazenium cations, Chem. Commun., 2019, 55, 14058–14061.
27. P. Bhattacharya, Z. M. Heiden, G. M. Chambers, S. I. Johnson, R. M. Bullock and M. T. Mock, Catalytic ammonia oxidation to dinitrogen by hydrogen atom abstraction, Angew. Chem., Int. Ed., 2019, 58, 11618–11624.
28. P. L. Dunn, S. I. Johnson, W. Kaminsky and R. M. Bullock, Diversion of catalytic C–N bond formation to catalytic oxidation of NH₃ through modification of the hydrogen atom abstractor, J. Am. Chem. Soc., 2020, 142, 3361–3365.
29. (a) B. Kartal and J. T. Keltjens, Anammox biochemistry: a tale of heme c proteins, Trends Biochem. Sci., 2016, 41, 998–1011; (b) K. M. Lancaster, J. D. Caranto, S. H. Majer and M. A. Smith, Alternative bioenergy: updates to and challenges in nitrification metalloenzymology, Joule, 2018, 2, 421–441.
30. B. M. Hoffman, D. Lukoyanov, Z.-Y. Yang, D. R. Dean and L. C. Seefeldt, Mechanism of nitrogen fixation by nitrogenase: the next stage, Chem. Rev., 2014, 114, 4041–4062.
31. D. Lukoyanov, N. Khadka, Z.-Y. Yang, D. R. Dean, L. C. Seefeldt and B. M. Hoffman, Reductive elimination of H₂ activates nitrogenase to reduce the N≡N triple bond: characterization of the E₄(4H) Janus intermediate in wild-type enzyme, J. Am. Chem. Soc., 2016, 138, 10674–10683.
32. (a) M. D. Zott, P. Garrido-Barros and J. C. Peters, Electrocatalytic ammonia oxidation mediated by a polypyridyl iron catalyst, ACS Catal., 2019, 9, 10101–10108; (b) M. D. Zott and J. C. Peters, Enhanced ammonia oxidation catalysis by a low-spin iron complex featuring cis coordination sites, J. Am. Chem. Soc., 2021, 143, 7612–7616; (c) M. D. Zott and J. C. Peters, Improving Molecular Iron Ammonia Oxidation Electrocatalysts via Substituent Effects that Modulate Standard Potential and Stability, ACS Catal., 2023, 13, 14052–14057.
33. (a) C. Baleizão and H. Garcia, Chiral salen complexes: an overview to recoverable and reusable homogeneous and heterogeneous catalysts, Chem. Rev., 2006, 106, 3987–4043; (b) G. Olivo, O. Lanzalunga and S. Di Stefano, Non-heme imine-based iron complexes as catalysts for oxidative processes, Adv. Synth. Catal., 2016, 358, 843–863; (c) W. Sun and Q. Sun, Bioinspired manganese and iron complexes for enantioselective oxidation reactions: ligand design, catalytic activity, and beyond, Acc. Chem. Res., 2019, 52, 2370–2381; (d) B. Stanje, P. Traar, J. A. Schachner, F. Belaj and N. C. Mösch-Zanetti, Iron-catalyzed oxidation of benzylic alcohols to benzoic acids, Dalton Trans., 2018, 47, 6412–6420; (e) C. B. Provis-Evans, S. Lau, V. Krewald and R. L. Webster, Regioselective alkyne cyclotrimerization with an in situ-generated [Fe(II)H(salen)]·Bpin catalyst, ACS Catal., 2020, 10, 10157–10168.
34. M. Usman, S. Tabassum, F. Arjmand, R. A. Khan, M. S. Ali, H. A. Al-Lohedan, A. Alsalme, M. A. Farah, K. M. Al-Anazi and M. Ahmad, CuII–NaI heteronuclear complex as anticancer entity against human breast cancer cell lines: DNA binding, cleavage, and computational studies, Inorg. Chim. Acta, 2018, 479, 229–239.
35. V. K. Sivasubramanian, M. Ganesan, S. Rajagopal and R. Ramaraj, Photoinduced electron transfer reactions of an oxovanadium(IV) Schiff base complex with aromatic amines, J. Org. Chem., 2002, 67, 1506–1510.
36. (a) J. Du, G. Liu, F. Li, Y. Zhu and L. Sun, Iron–salen complex and Co²⁺ ion-derived cobalt–iron hydroxide/carbon nanohybrid as an efficient oxygen evolution electrocatalyst, Adv. Sci., 2019, 6, 1900117; (b) M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson and R. S. C. Smart, Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni, Appl. Surf. Sci., 2011, 257, 2717–2730; (c) S. J. Gerber and E. Erasmus, Electronic effects of metal hexacyanoferrates: An XPS and FTIR study, Mater. Chem. Phys., 2018, 203, 73–81; (d) A. P. Grosvenor, B. A. Kobe, M. C. Biesinger and N. S. McIntyre, Investigation of multiplet splitting of Fe 2p XPS spectra and bonding in iron compounds, Surf. Interface Anal., 2004, 36, 1564–1574.
37. T. Kurahashi, Y. Kobayashi, S. Nagatomo, T. Tosha, T. Kitagawa and H. Fujii, A functional model for intradiol-cleaving dioxygenases: iron–salen complexes, Inorg. Chem., 2005, 44, 8156–8166.
38. (a) Y.-X. Zhou, X.-F. Zheng, D. Han, H.-Y. Zhang, X.-Q. Shen, C.-Y. Niu, P.-K. Chen, H.-W. Hou and Y. Zhu, Syntheses and crystal structures of Fe(III) and Mn(III) complexes with salen ligands: [Fe(salen)Cl]·CH₃CN and [Mn(salen)(SCN)(CH₃OH)]·CH₃CN, Synth. React. Inorg., Met.-Org., Nano-Met. Chem., 2006, 36, 693–699; (b) E. D. Portakal, Y. Kaya, E. Demirayak, E. K. Yeldir, A. Erçağ and İ. Kaya, Synthesis, characterization, and investigation of some properties of the new symmetrical bisimine Ni(II), Zn(II), and Fe(III) complexes derived from the monoimine ligand, Appl. Organomet. Chem., 2021, 35, e6265; (c) M. Cozzolino, V. Leo, C. Tedesco, M. Mazzeo and M. Lamberti, Salen, salan and salalen iron(III) complexes as catalysts for CO₂/epoxide reactions and ROP of cyclic esters, Dalton Trans., 2018, 47, 13229–13238; (d) H.-L. Shyu, H.-H. Wei, G.-H. Lee and Y. Wang, Structure, magnetic properties, and epoxidation activity of iron(III) salicylaldimine complexes, J. Chem. Soc., Dalton Trans., 2000, 915–918.
39. H. Toda, K. Kuroki, R. Kanega, S. Kuriyama, K. Nakajima, Y. Himeda, K. Sakata and Y. Nishibayashi, Manganese-catalyzed ammonia oxidation into dinitrogen under chemical or electrochemical conditions, ChemPlusChem, 2021, 86, 1511–1516.
40. W. Yang, G. A. Filonenko and E. A. Pidko, Performance of homogeneous catalysts viewed in dynamics, Chem. Commun., 2023, 59, 1757–1768.
41. R. H. Crabtree, G. Ertl, H. Knözinger, F. Schüth and J. Weitkamp, Deactivation in Homogeneous Transition Metal Catalysis: Causes, Avoidance, and Cure, Chem. Rev., 2015, 115, 127–150.
42. Kinetic study and reaction mechanisms in homogeneous catalysis, in Comprehensive Coordination Chemistry III, ed. A. S. Borovik, Elsevier, Oxford, 2021, vol. 10, pp. 249–278.
43. Y. Li, J.-Y. Chen, Q. Miao, X. Yu, L. Feng, R.-Z. Liao, S. Ye, C.-H. Tung and W. Wang, A Parent Iron Amido Complex in Catalysis of Ammonia Oxidation, J. Am. Chem. Soc., 2022, 144, 4365–4375.
44. L. Nurdin, Y. Yang, P. G. N. Neate, W. E. Piers, L. Maron, M. L. Neidig, J.-B. Lin and B. S. Gelfand, Activation of ammonia and hydrazine by electron rich Fe (ii) complexes supported by a dianionic pentadentate ligand platform through a common terminal Fe (iii) amido intermediate, Chem. Sci., 2021, 12, 2231–2241.
45. CCDC 2384217: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2l0z7v.

---