Facile conversion of ammonia to a nitride in a rhenium system that cleaves dinitrogen

Rhenium complexes with aliphatic PNP pincer ligands have been shown to be capable of reductive N2 splitting to nitride complexes. However, the conversion of the resulting nitride to ammonia has not been observed. Here, the thermodynamics and mechanism of the hypothetical N–H bond forming steps are evaluated through the reverse reaction, conversion of ammonia to the nitride complex. Depending on the conditions, treatment of a rhenium(iii) precursor with ammonia gives either a bis(amine) complex [(PNP)Re(NH2)2Cl]+, or results in dehydrohalogenation to the rhenium(iii) amido complex, (PNP)Re(NH2)Cl. The N–H hydrogen atoms in this amido complex can be abstracted by PCET reagents which implies that they are quite weak. Calorimetric measurements show that the average bond dissociation enthalpy of the two amido N–H bonds is 57 kcal mol−1, while DFT computations indicate a substantially weaker N–H bond of the putative rhenium(iv)-imide intermediate (BDE = 38 kcal mol−1). Our analysis demonstrates that addition of the first H atom to the nitride complex is a thermochemical bottleneck for NH3 generation.


S -5
H2 quantification from 1eoxidation of (PNP)Re(NH2)(Cl) (3): Under N2, 54 mg FcPF6 (0.16 mmol, 1.1 equiv) was slurried in 2 mL THF in a 7 mL scintillation vial. The vial was sealed with a septum cap, and a solution of 89 mg (PNP)Re(NH2)(Cl) (2, 0.15 mmol, 1.0 equiv) in 2 mL THF was injected into the vial via syringe. An immediate color change from deep blue to orange occurred, as well as formation of a brown precipitate. The reaction was stirred for 10 min at ambient temperature, then chilled for 5 min in a dry ice/isopropanol bath to remove THF vapor from the headspace. 50 μL CH4 gas was added via airtight syringe into the headspace of the vial as an internal standard, and 50 μL of the headspace of the vial was removed via airtight syringe for H2 quantification on the GC. No H2 peak was observed in the resulting chromatograph, indicating less than 1% yield of H2. The reaction was repeated, with identical results.
Removal of the volatile materials under vacuum gave a sticky, dark red-orange residue.
The residue was extracted with 2 x 5 mL THF and filtered through Celite. Removal of THF from the orange-brown filtrate under vacuum yielded a dirty orange solid. The solid was redissolved in ~10 mL THF, from which a ~2 mL aliquot was taken. The aliquot was dried under vacuum, dissolved in 0.7 mL THF-d8, and analyzed using NMR spectroscopy vs. a 1,3,5trismethoxybenzene (TMB) internal standard to quantify the Re products. Analysis of the 1 H NMR spectra showed 52% yield of nitride 5 and 17% yield of dichloride 1 for the first run, and 47% yield of nitride 5 and 20% yield of dichloride 1 for the second run.  F 4] (0.12 mmol) was dissolved in 2 mL THF in a 7 mL scintillation vial. The vial was sealed with a septum cap, and a solution of 46 mg Cp*2Co (0.14 mmol, 1.1 equiv) dissolved in 2 mL THF was injected into the vial via syringe. An immediate color change from light tan to dark brown-gold occurred, as well as formation of a yellow S -6 precipitate. The reaction was stirred for 10 min at ambient temperature, then chilled for 5 min in a dry ice/isopropanol bath to remove THF vapor from the headspace. 50 μL CH4 gas was added via airtight syringe into the headspace of the vial as an internal standard, and 50 μL of the headspace of the vial was removed via airtight syringe for H2 quantification on the GC. Gas chromatography used a ThermoFisher Trace 1300 GC apparatus with a thermal conductivity detector and a Supelco fused silica capillary column (5 Å molecular sieves, 30 m x 0.53 mm), and N2 as a carrier gas. No H2 peak was observed in the resulting chromatograph, indicating less than 1 ppm (1% yield) of H2. The reaction was repeated, with identical results.
Removal of the volatile materials under vacuum gave a sticky, dark brown residue. The residue was extracted with 2 x 5 mL THF and filtered through Celite. Removal of THF from the brown filtrate under vacuum yielded a sticky brown solid. The solid was redissolved in ~10 mL THF, from which a ~2 mL aliquot was taken. The aliquot was dried under vacuum, dissolved in 0.7 mL THF-d8, and analyzed using NMR spectroscopy vs. a 1,3,5-trimethoxybenzene (TMB) internal standard to quantify the Re products. Analysis of the 1 H NMR spectra showed formation of 3 in 73% for the first run and 61% for the second run. Decamethylcobaltocenium was also observed, though the peak was broadened and shifted from rapid self-exchange with Cp*2Co ( Figure S5B).   Figure S4. 1 H (400 MHz, THF-d8) and 31 P{ 1 H} (162 MHz, THF-d8) NMR spectra of 4 in THF-d8 before (bottom, maroon) and after (top, teal) the addition of 1.05 equiv KHMDS. Complex 3 is formed in 61% spectroscopic yield vs. SPMe3 in a capillary. Figure S5A. 31 P{ 1 H} NMR spectra (162 MHz, THF-d8) of 4 in THF-d8 prior to (maroon) and after addition of 1.2 equiv CoCp2*. Under Ar, the reaction gave 3 in 60% spectroscopic yield (green). Under N2, the reaction gave 3 in 60% spectroscopic yield, along with other unidentified minor products (blue). Spectroscopic yields measured vs. SPMe3 in a capillary. For comparison, a spectrum of authentically prepared 3 is included (purple). Figure S5B. 1 H NMR spectra (400 MHz, THF-d8) of 4 in THF-d8 prior to (maroon) and after addition of 1.2 equiv CoCp2*. Under Ar, the reaction gave 3 in 60% spectroscopic yield (green). Under N2, the reaction gave 3 in 60% spectroscopic yield, along with other unidentified minor products (blue). Spectroscopic yields measured vs. SPMe3 in a capillary. For comparison, a spectrum of authentically prepared 3 is included (purple). The broad peaks at 9-11 ppm are presumed to be from Cp*2Co and Cp*2Co + in fast exchange, and integrate to 2.9 Co per mol 3 in the N2 spectrum and 2.5 Co per mol 3 in the Ar spectrum.     CVs measured at 100 mV/s with a glassy carbon disc working electrode, Pt wire auxiliary electrode, and Ag wire psuedoreference. Figure 4 in the text shows that the anodic response is pseudoreversible at faster scan rates. Potentials referenced to Cp2Fe +/0 after the experiments. Bottom left: Comparison of cyclic voltammograms of 3 (1.7 mM) prior to electrolysis, measured in 0.2 M tetrabutylammonium hexafluorophosphate solution in THF containing excess 2,6-lutidine (34 mM) under Ar, using a glassy carbon (black) or carbon paper (red) working electrode, Pt mesh counter electrode, and Ag wire pseudoreference electrode. The electrolysis potential (+0.66 V vs. open circuit potential) is noted in blue. Right: Charge passed vs. time, and current vs. time, during the electrolysis of 3. The electrolysis was stopped after 2.2 equiv of charge per Re was passed. The continued increase of charge passed is likely due to degradation and contributes to a low yield.
The experimental conditions for the titration of (PNP)Re(NH2)(Cl) (3) with t Bu3PhO • in THF are summarized in Table S1. Three separate runs were performed, and all measurements were taken at ambient temperature (298 K).  Figure S18 shows a representative thermogram and integrated titration curve for the titration of 3 with t Bu3PhO • in THF. Greyed out points are not taken into account in the fitting process. The first point is generally neglected due to the dilution effect within the pre-experiment equilibration time. The mean heat value was determined for the titration steps before the equivalence point. From the experimental reaction enthalpy of the reaction of amide 3 with TBP (2 equiv.) to nitride 2 in THF at 298 K (-50.9 ± 0.5 kcal/mol), the average BDEN-H of 3 was calculated to be 55.4 ± 1 kcal/mol according to the following equations. The BDEO-H of t Bu3PhO • in THF (80.8 ± 0.3 kcal/mol) was estimated from the experimental BDFEO-H, 3 assuming negligible differential entropies of solution of t Bu3PhO • and t Bu3PhOH and using TS 0 (H•) = 6.4 kcal mol -1 as entropy of solvation of the hydrogen atom based on the entropy of solvation of H2.

Computational details
Density Functional Theory (DFT) calculations were performed using Gaussian 09 (revision D.01). 7 Geometries were optimized using the hybrid functional B3LYP, with basis set def2-TZVP used for all atoms and an ultrafine integration grid. Unless otherwise noted, all calculations include the GD3BJ version of Grimme's dispersion correction and a conductor-like polarizable continuum model (CPCM) for THF solvent. Frequency calculations were performed to confirm that the optimized structures were minima; the optimized structures for the complexes reported showed no negative frequencies.   Figure S20. Overlay of crystal structure (yellow) and calculated structure (blue) of 3. Hydrogen atoms (other than amide hydrogens) are omitted for clarity. Average 51.0

Redox potential computation:
The

Crystallographic details
Crystals were isolated in an Ar glovebox by decanting the mother liquor from the crystals before transferring and submerging them in high-viscosity petroleum oil on a microscope slide. Single crystals suitable for X-ray diffraction were identified under a polarizing microscope and mounted on 200 μm MiTeGen Dual-Thickness MicroLoops in a mixture of high-viscosity petroleum, then mounted on the diffractometer instrument. Low-temperature diffraction data (ωscans) were collected on a Rigaku MicroMax-007HF diffractometer coupled to a Dectris Pilatus3R detector with Mo Kα (λ = 0.71073 Å) for the structure of 007c-17049 (3). Low-temperature diffraction data (ω-scans) were collected on a Rigaku MicroMax-007HF diffractometer coupled to a Saturn994+ CCD detector with Cu Kα (λ = 1.54178 Å) for the structure of 007b-17124 (4). The diffraction images were processed and scaled using either Rigaku CrystalClear software or Rigaku Oxford Diffraction software. The structure was solved with SHELXT and was refined against F 2 on all data by full-matrix least squares with SHELXL.   S -39

[(PNP)Re(NH3)2(Cl)][BAr F 4] (4).
All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in the model at geometrically calculated positions and refined using a riding model. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the U value of the atoms to which they are linked (1.5 times for methyl groups; 1.2 for NH3 groups). The pentane and BAr F sites are disordered. Four of the CF3 groups of the BAr  (12), respectively. The CF3 groups were restrained to have similar 1,2 C-C and C-F distances for chemically identical bonds. The 1,3 F-F distance were also restrained to be similar. The thermal parameters of the CF3 groups were retained to behave as a rigid group. The pentane was disordered across the crystallographic inversion center. The special position constraints were suppressed and the site occupancies were fixed at 0.5. The pentane was retained to behave as a rigid group. The thermal parameter of C3A was restrained to be identical to that of its neighbour due to its proximity to the crystallographic inversion center. Figure S19. The complete numbering scheme of 4 with 50% thermal ellipsoid probability levels. The hydrogen atoms are omitted for clarity.