Direct synthesis of cyanate anion from dinitrogen catalysed by molybdenum complexes bearing pincer-type ligand

The study addresses the challenge of directly converting atmospheric dinitrogen (N₂) into valuable organonitrogen compounds under mild conditions, overcoming the thermodynamic and kinetic inertness of N₂. Conventional industrial nitrogen fixation via the Haber–Bosch process operates under harsh conditions and consumes significant fossil fuels; producing organonitrogen molecules typically requires additional multi-step transformations from ammonia. Prior molecular approaches have shown that transition-metal–bound dinitrogen or metal nitride intermediates can be functionalized by carbon-centered electrophiles to form N–C bonds under mild conditions, but catalytic, direct transformations to organonitrogen products have been lacking. Targeting cyanate (NCO⁻) and related isocyanates as thermodynamically favorable and industrially useful products, the authors propose a two-step, ambient-condition route from N₂ using molybdenum pincer complexes and chloroformate esters.

Previous work established that organonitrogen compounds can be synthesized from reactions of metal–dinitrogen or metal–nitride complexes with electrophiles, often via multi-step, stoichiometric sequences. For isocyanate derivatives from N₂, Kawaguchi and Schneider reported synthetic cycles in five and seven steps, respectively, using CO as a key reagent to convert nitride intermediates to isocyanate complexes, while Sita achieved a five-step cycle to trimethylsilyl isocyanate using CO₂ and a molybdenum silylimide; Hou reported a four-step cycle involving a titanium–dinitrogen complex and CO₂. These approaches rely on multiple steps under differing conditions and on CO/CO₂, complicating development of catalytic cycles and potentially inhibiting nitride formation. Separately, the authors’ group previously demonstrated catalytic ammonia formation from N₂ using molybdenum PNP-pincer complexes via direct N≡N bond cleavage, and showed stoichiometric N–C bond-forming reactivity of the same nitride complex with acyl chlorides to give amides. Chloroformate esters have also been used by others as carbonyl sources in transition-metal-catalyzed ketone synthesis, suggesting their potential as practical carbon electrophiles in N–C bond formation from N₂.

• Stoichiometric synthesis and characterization:

1. Carbamylation: The molybdenum nitride complex [Mo(N)I(PNP)] (1; PNP = 2,6-bis(di-tert-butylphosphinomethyl)pyridine) was reacted with 1.1 equiv phenyl chloroformate in THF at room temperature for 2 h to afford carbamate complex [Mo(NCO₂Ph)Cl(PNP)] (2) in 83% yield. Characterization by 1H/31P{1H} NMR, IR (νCO ≈ 1690 cm⁻¹), and X-ray crystallography confirmed structure.
2. Reductive conversion to NCO⁻: Reduction of 2 with 5 equiv SmI₂ in THF at rt for 2 h under 1 atm N₂ produced nitride 1 (74% NMR yield), samarium phenoxy complex [SmI₂(OPh)] (3a; 93% NMR yield), and samarium isocyanate species (formally [SmI₂(NCO)]). After hydrolysis, NCO⁻ was quantified by ion chromatography (IC) at 0.75 equiv per 2 (75% IC yield). NCO⁻ confirmed by IR (νNCO = 2170 cm⁻¹); 1 by ESI-MS (m/z 634.18). Samarium isocyanate analogue {[SmI(NCO)(18-crown-6)][I]} (3b) was independently synthesized (60% yield) to support NCO transfer and hydrolysis behavior.
3. Isotope labeling: Reduction of 15N-labeled 2 (2-15N) with SmI₂ under 1 atm 15N₂ yielded [Mo(15N)I(PNP)] (1) and 15NCO⁻ (62% IC yield), indicating formation of a dinitrogen-bridged dimolybdenum intermediate that undergoes N≡N cleavage to regenerate 1.
4. Identification of isocyanate intermediate: One-electron reduction of 2 with 1.0 equiv SmI₂ (THF, rt, 10 min, Ar) gave [Mo(NCO)ICl(PNP)] (4a) in 68% yield; IR νNCO = 2209 cm⁻¹ with 15N shift Δν = 14 cm⁻¹. Reduction of 2 with CoCp₂ gave [Mo(NCO)(OPh)Cl(PNP)] (4b) in 49% yield, highlighting Sm’s oxophilicity in removing phenoxy ligands. Reduction of 4a with 5 equiv SmI₂ under 1 atm N₂ furnished 1 (66% NMR yield) and NCO⁻ (75% IC yield), confirming 4a as an intermediate. Attempts to reduce 4b with CoCp₂/NaI under N₂ did not yield 1.

• Proposed synthetic cycle and pathway: Two steps under ambient conditions: (i) carbamylation of 1 with chloroformate to 2; (ii) reduction of 2 with SmI₂ to generate an isocyanate complex (4a), chloride loss, dinitrogen coordination and formation of a dinitrogen-bridged dimolybdenum intermediate A, followed by direct N≡N cleavage to regenerate 1 and release NCO⁻ (transferred to Sm species).

• DFT calculations (B3LYP-D3, THF solvation): Step (i) carbamylation is highly exergonic (ΔG298 ≈ −36.6 kcal/mol) with low barrier (≈5.3 kcal/mol). Step (ii) one-electron-reduced 2 undergoes heterolytic C–OPh bond cleavage to 4a with very low barrier (1.6 kcal/mol) and is highly exergonic (−30.4 kcal/mol). For step (iii), after reduction of 4a, Cl⁻ dissociation is facile (ΔG298 ≈ 1.1 kcal/mol), N₂ coordination and dimerization to a Mo–N≡N–Mo unit are exergonic, and subsequent reductions weaken Mo–NCO bonds to release NCO⁻. Samarium(III) species thermodynamically favor transfer of PhO⁻ and NCO⁻ from Mo: ΔG298 ≈ −26.2 kcal/mol for PhO transfer and −10.5 kcal/mol for NCO transfer to [SmI₂(THF)₅].

• Catalytic studies (THF, rt, 1 atm N₂, 16 h): Reactions used SmI₂ (typically 36 equiv/Mo) and chloroformates (typically 12–24 equiv/Mo). Catalysts screened: Mo–PNP nitride 1; Mo–PPP nitride 5; Mo–PCP nitride 6. NCO⁻ quantified by ion chromatography. Slow syringe addition of chloroformate was used to suppress direct reaction with SmI₂ and improve turnover. Conditions varied by chloroformate substituent (Ph, Me, Et, iPr) and addition time (3–12 h) to optimize NCO⁻ production.

• Established a two-step synthetic cycle converting dinitrogen (N₂) to cyanate anion (NCO⁻) under ambient conditions mediated by molybdenum pincer complexes: (i) carbamylation of a Mo–nitride with chloroformate; (ii) SmI₂-promoted reduction releasing NCO⁻ and regenerating Mo–nitride via N≡N bond cleavage.
• Stoichiometric results: • Carbamate complex [Mo(NCO₂Ph)Cl(PNP)] (2) formed from [Mo(N)I(PNP)] (1) and phenyl chloroformate in 83% yield (THF, rt, 2 h). • Reduction of 2 with SmI₂ (5 equiv) under 1 atm N₂ produced 1 (74% NMR yield) and NCO⁻ (75% IC yield); [SmI₂(OPh)] formed in 93% NMR yield; samarium isocyanate species implicated. • One-electron reduction of 2 yielded isocyanate complex [Mo(NCO)ICl(PNP)] (4a) in 68% yield; reduction of 4a with SmI₂ under N₂ gave 1 (66% NMR yield) and NCO⁻ (75% IC yield), confirming 4a as an intermediate. • Isotope labeling showed formation of [Mo(15N)I(PNP)] and 15NCO⁻, supporting a dinitrogen-bridged dimolybdenum intermediate and direct N≡N cleavage.
• Catalytic formation of NCO⁻ from N₂ achieved: • With Mo–PCP catalyst 6, SmI₂ (36 equiv), phenyl chloroformate (12 equiv): 0.74 equiv NCO⁻/Mo (run 3). • Slow addition (3 h) of phenyl chloroformate increased to 1.43 ± 0.12 equiv NCO⁻/Mo (run 4). • Using methyl chloroformate and slow addition: 2.51 ± 0.20 equiv NCO⁻/Mo (run 5). • Extended slow addition of methyl chloroformate (24 equiv over 12 h) afforded 8.99 ± 0.15 equiv NCO⁻/Mo (75% yield based on SmI₂; run 12). • Controls showed necessity of catalyst, SmI₂, chloroformate, and N₂ (no NCO⁻ under Ar; negligible without SmI₂).
• Computational support: Low barriers and exergonic profiles for carbamylation and C–O bond cleavage; thermodynamic favorability for Sm(III)-assisted transfer of PhO⁻ and NCO⁻; mechanistic consistency with direct N≡N cleavage in dinitrogen-bridged Mo dimers.
• The cycle operates in only two steps, fewer than previously reported multi-step synthetic cycles for isocyanate derivatives from N₂.

The findings demonstrate that direct conversion of N₂ to cyanate under ambient conditions is feasible via a minimal two-step sequence mediated by molybdenum pincer complexes. By coupling chloroformate carbamylation of a Mo–nitride with SmI₂-driven reduction that expels phenoxy, chloro, and isocyanate ligands to samarium and regenerates the nitride through dinitrogen cleavage, the authors close a productive cycle that produces NCO⁻. Isotope labeling and DFT calculations substantiate the proposed intermediates and energetics, including low barriers for carbamylation and C–OPh bond cleavage, facile chloride dissociation after reduction, and formation of a dinitrogen-bridged dimolybdenum species that undergoes direct N≡N bond scission. Catalytically, careful control of electrophile addition and choice of chloroformate (less hindered, less reactive toward SmI₂) improved performance, achieving up to ~9 turnovers per Mo under 1 atm N₂ at room temperature. This approach avoids the need for CO/CO₂ as in prior multi-step cycles and operates under uniform, mild conditions, marking a significant advance toward practical N₂-to-organonitrogen transformations. The role of SmI₂ as both reductant and acceptor of leaving groups is critical to drive the cycle forward and suppress off-cycle Mo–OPh or Mo–NCO species, in line with computed thermodynamics.

The work establishes a two-step, ambient-condition synthetic cycle for producing cyanate anion (NCO⁻) directly from dinitrogen, mediated by molybdenum complexes bearing pincer-type ligands. Key steps are carbamylation of a Mo–nitride with chloroformate to form a Mo–carbamate and SmI₂-promoted reduction that generates a dinitrogen-bridged dimolybdenum intermediate, undergoes direct N≡N cleavage, regenerates the nitride, and releases NCO⁻ to samarium. The cycle was translated to catalysis, delivering up to 8.99 equiv of NCO⁻ per Mo catalyst under 1 atm N₂ at room temperature with optimized slow addition of methyl chloroformate. DFT analyses corroborated the low barriers and exergonic nature of the critical steps and rationalized the beneficial role of Sm(III) in ligand transfer. Future directions include increasing catalyst turnover numbers and rates, broadening electrophile scope beyond chloroformates, optimizing ligand frameworks for improved activity and stability, exploring alternative, more sustainable reductants than SmI₂, and isolating/characterizing transient samarium isocyanate species to further refine the mechanism.

• The catalytic turnovers, while superstoichiometric, remain modest and rely on large excesses of SmI₂ and controlled slow addition of chloroformates.
• The process depends critically on SmI₂ as a strong, stoichiometric reductant and oxophilic group acceptor; alternative reductants were ineffective in key steps (e.g., CoCp*₂ failed to regenerate nitride 1 from 4b).
• The proposed samarium isocyanate species formed in situ were not structurally characterized; only an analogue (3b) was prepared independently to support NCO transfer behavior.
• Product formation is sensitive to the chloroformate substituent (bulk hinders efficiency) and to reaction atmosphere (no NCO⁻ under Ar).
• The isocyanate Mo intermediate 4a showed ligand disorder in crystallography, limiting precise structural metrics around NCO coordination.