Binding of Nitriles and Isonitriles to V(III) and Mo(III) Complexes: Ligand vs Metal Controlled Mechanism

The synthesis and structures of nitrile complexes of V(N[tBu]Ar)3, 2 (Ar = 3,5-Me2C6H3), are described. Thermochemical and kinetic data for their formation were determined by variable temperature Fourier transform infrared (FTIR), calorimetry, and stopped-flow techniques. The extent of back-bonding from metal to coordinated nitrile indicates that electron donation from the metal to the nitrile plays a less prominent role for 2 than for the related complex Mo(N[tBu]Ar)3, 1. Kinetic studies reveal similar rate constants for nitrile binding to 2, but the activation parameters depend critically on the nature of R in RCN. Activation enthalpies range from 2.9 to 7.2 kcal·mol–1, and activation entropies from −9 to −28 cal·mol–1·K–1 in an opposing manner. Density functional theory (DFT) calculations provide a plausible explanation supporting the formation of a π-stacking interaction between a pendant arene of the metal anilide of 2 and the arene substituent on the incoming nitrile in favorable cases. Data for ligand binding to 1 do not exhibit this range of activation parameters and are clustered in a small area centered at ΔH‡ = 5.0 kcal·mol–1 and ΔS‡ = −26 cal·mol–1·K–1. Computational studies are in agreement with the experimental data and indicate a stronger dependence on electronic factors associated with the change in spin state upon ligand binding to 1.

S6 prepared and cooled to -35 °C. The solution of PhCN and 2 were subsequently added to the stirring suspension of FcBAr F 4. The reaction mixture was allowed to warm to room temperature while stirring. The color of the solution changed from purple to dark green as the reaction progressed.
After stirring at room temperature for 2 h, the reaction mixture was filtered through a plug of Celite on a glass fiber filter. The filter cake was washed with diethyl ether (1 mL). The filtrate solution was concentrated to a total volume of 2 mL and stored at -35 °C for 24 h. Dark green crystals could be isolated by decanting the mother liquor away, washing with n-pentane, and then drying under reduced pressure. Yield: 0.179 g (1546.14 g/mol, 0.11 mmol, 40%). 1  FTIR study of Binding of DFBN to 2 in Toluene Solution. All studies were performed using standard inert atmosphere techniques. Toluene was dried by distillation from sodium benzophenone ketyl into flame dried glassware and stored under Argon. Solutions were prepared in a Vacuum-Atmospheres glove box, filtered, and loaded into a gas tight Hamilton syringe fitted with a syringe valve. Infrared Spectra were obtained on a Perkin Elmer series 2000 FTIR spectrometer with an MCT detector. The temperature of the FTIR cell was controlled by a water bath which was circulated through the jacketed cell, but also through copper tubing surrounding the cell which was housed in a small chamber fitted with CaF2 windows. The temperature of the cell was measured by a precision thermistor obtained from Omega Engineering and cemented to the body of the FTIR cell. In addition, the temperature of the Argon atmosphere surrounding the cell in the small boxed area was measured with a Pt RTD. Spectroscopic data for nitrile binding are summarized in Table S1. a Some data for RCN-1 have been previously reported; 7 b Binding of Me2NCN to 1 yields a sideon adduct. 8 Representative FTIR data from one series of experiments is shown in Figure S1 and shows the increase in Absorbance of the band at 2216 cm −1 assigned to DFBN-2 with decreasing temperature. Figure S1. Variable temperature FTIR data for binding of DFBN to 2 in toluene. The broad nature of the band as well as an apparent shoulder may be due to different conformational isomers which may also be temperature dependent.

S8
Due to broadening of the nitrile band with increasing temperature, equilibrium constants were computed based on peak areas rather than heights. FTIR studies were made in the overall temperature range of 9 to 49 o C. Plots of ln(Keq) versus 1/T are shown in Figure S2. The derived enthalpies and entropies of binding for the three individual runs are collected in Table   S1 and the average value was adopted. Table S2. Thermochemical data for binding of DFBN to 2 derived from van't Hoff plots in Figure  S2. (0.0203 g, 0.126 mmol) was used as the limiting reagent and was analytically pure as verified by NMR spectroscopy. The calorimeter cell was sealed, taken from the glovebox, and loaded into the S9 Setaram C-80 calorimeter. Following temperature equilibration, the reaction was initiated and the calorimeter rotated to achieve mixing. Following return to baseline, the calorimeter cell was taken into the glovebox, opened, and 1mL of the solution loaded into an NMR tube. NMR spectra of both the stock solution and the calorimetry solution were then acquired and the reaction was confirmed as quantitative. and all remaining quantities derived from the kinetic data are reported with their standard deviations. All concentrations are reported after mixing in the stopped-flow cell.    AdCN (1 mM) at −62 °C, acquired over 9 s. Selected traces shown for clarity. Initial spectrum is shown in black and final spectrum in red. S12 Figure S8. Time-resolved spectral changes accompanying the reaction between 2 (0.3 mM) and AdNC (1 mM) at −62 °C, acquired over 9 s. Selected traces shown for clarity. Initial spectrum is shown in black and final spectrum in red.           X-ray Crystallography. Low-temperature diffraction data were collected on a Siemens Platform three-circle diffractometer equipped with a Bruker APEX CCD, using graphite S24 monochromated Mo K  radiation ( = 0.71073 Å) from a water-cooled sealed tube. The structures were solved by dual-space methods using SHELXT 9 and refined against F 2 on all data by fullmatrix least squares with SHELXL-2017 10 following established refinement strategies. 11 All nonhydrogen atoms were refined anisotropically. All hydrogen atoms were included into 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 they are linked to (1.5 times for methyl groups).
Compound DFBN-2 crystallizes in the triclinic centrosymmetric space group P-1 with one molecule of DFBN-2 per asymmetric unit. Structure determination was straightforward and no restraints were applied.
Compound Me2NCN-2 crystallizes in the cubic centrosymmetric space group Pa-3 with one third molecule of Me2NCN-2 per asymmetric unit. The triple-bonded NMe2 moiety was refined as six-fold disordered in a fashion where two disorder components are crystallographically independent, the other four are generated by the same crystallographic threefold axis that completes the full molecule from the third contained in the asymmetric unit (see Figure S19). The ratio of the two independent disorder components was refined freely and converged at 0.50(4).
Thus, each disorder component is occupied to approximately one sixth. This makes discussing the geometry of the NMe2 moiety difficult; especially the C-N-C angles cannot be determined with confidence. The disorder was refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. This could be modelled for four of them. Additionally, one of the three nitrogen bound dimethylphenyl rings was modelled as disordered over two positions. The disorders were refined with the help of similarity restraints on 1-2 and 1-3 distances and displacement parameters as well as rigid bond restraints for anisotropic displacement parameters. All disorder ratios were refined freely; for the dimethyl-phenyl ring it converged at 0.622 (7).
Details of the data quality and a summary of the residual values of the refinement of DFBN-2, Me2NCN-2 and [PhCN-2][BAr F 4] are listed in Tables S11-S13. Tables S14-S16 give all bond lengths and angles for the structures.      Tables S19 and S20. Excellent agreement is observed between the computed results and experimental structural data obtained by X-ray crystallography as shown in Table S18. Likewise, good agreement between computed and experimental νCN IR wavenumbers is generally also observed (see Table S21). Finally, thermochemical data for ligand binding to 1 and 2 computed S41 by DFT calculations are collected in Table S22. Due to the weak computed binding for many ligands, as well as experimental difficulties, limited comparison can be made between computed thermochemical and experimental data. In spite of that, a reasonable correlation is seen between calculated and available experimental enthalpies of nitrile binding with differences generally lower than 2.5 kcal·mol −1 as shown in Table S23. Moreover, the thermochemical values computed for MeCN binding to 1 (ΔH = −5.5 kcal·mol −1 ) are in good agreement with those determined experimentally for other alkyl nitriles (ΔH = −6 ± 2 kcal·mol −1 for AdCN binding to 1 4 ). However, as can be seen in Table S23, larger differences between DFT-calculated and experimentally The values obtained using this procedure are collected between brackets in Table S23 and a good agreement is observed now between the thermochemical values obtained by DFT calculations and experimentally in this work based on variable temperature FTIR studies or solution calorimetric data.

S42
Finally, Table S17 supports the reliability of the DFT-computed values for the stabilization energy due to π-stacking or C-H··· π interactions as those described in the structure of DFBN-2 ( Figure 1 in the main text) and the interaction of benzene solvent with complex 2 in the B configuration ( Figure 7, main text). The interaction energy between benzene and different arenes were computed and the results are collected in Table S17 along with available data previously computed at high-level ab-initio calculations (CCSD(T)/CBS level). 22 The        Scheme S2. Reaction of the most stable structure from Figure 7 and L containing a π-stacking or C-H··· π interactions with a benzene molecule to yield complex L-2 and the most stable tilted Tshape benzene dimer. 22    Tables S26 and S27 respectively.   Tables S26 and S27 respectively. Figure S28. SOMOS of the optimized structure of the minima for MeNC binding to 1' in the quartet state (structure shown in Figure S25) at the PBE0-D3(BJ)/Def2-SV(P) level of theory. Hydrogen atoms omitted for clarity. Isovalue = 0.04. Figure S29. SOMOS of the optimized structure of the MECP between the quartet and doublet states in MeCN binding to 1' (structure shown in Figure S27 (right)) at the PBE0-D3(BJ)/Def2-SV(P) level of theory. Hydrogen atoms omitted for clarity. Isovalue = 0.04.