183W NMR Spectroscopy Guides the Search for Tungsten Alkylidyne Catalysts for Alkyne Metathesis

Abstract Triarylsilanolates are privileged ancillary ligands for molybdenum alkylidyne catalysts for alkyne metathesis but lead to disappointing results and poor stability in the tungsten series. 1H,183W heteronuclear multiple bond correlation spectroscopy, exploiting a favorable 5 J‐coupling between the 183W center and the peripheral protons on the alkylidyne cap, revealed that these ligands upregulate the Lewis acidity to an extent that the tungstenacyclobutadiene formed in the initial [2+2] cycloaddition step is over‐stabilized and the catalytic turnover brought to a halt. Guided by the 183W NMR shifts as a proxy for the Lewis acidity of the central atom and by an accompanying chemical shift tensor analysis of the alkylidyne unit, the ligand design was revisited and a more strongly π‐donating all‐alkoxide ligand prepared. The new expanded chelate complex has a tempered Lewis acidity and outperforms the classical Schrock catalyst, carrying monodentate tert‐butoxy ligands, in terms of rate and functional‐group compatibility.


Ligand 15b
A one-neck round bottomed flask equipped with a stir bar was charged with silane S1 (222 mg, 0.393 mmol) and CH2Cl2 (5 mL). The resulting mixture was cooled to 0 °C. m-Chloroperbenzoic acid (291 mg, 1.30 mmol, 77% w/w) was added portionwise and the reaction mixture was allowed to warm to ambient temperature.
After 4 h, the mixture was carefully transferred to a separation funnel, diluted with CH2Cl2 (10 mL) and washed with sat. NaHCO3 (4 x 15 mL) and brine (3 x 10 mL).
The organic phase was then dried over MgSO4, filtered and concentrated in vacuo to give ligand 15b as a colorless solid (240 mg, 99%). 1   A two-necked, round-bottomed flask was equipped with a magnetic stir bar and a gas inlet connected to an argon-vacuum manifold. The flame-dried flask was filled with argon and charged with 1,3,4-tris-2'-bromophenylbenzene (14) (

Ligand 15c
A two-neck round bottomed flask equipped with a stir bar was charged with silane S2 (563 mg, 0.768 mmol) and CH2Cl2 (10 mL). The resulting mixture was cooled to 0 °C. m-Chloroperbenzoic acid (568 mg, 2.53 mmol, 77% w/w) was added in portions and the reaction mixture was allowed to warm to ambient temperature. After 4 h, the mixture was carefully transferred to a separation funnel, diluted with CH2Cl2 (10 mL) and washed with sat. NaHCO3 (4 x 15 mL) and brine (3 x 10 mL). The organic phase was then dried over MgSO4, filtered and concentrated in vacuo to give ligand 15c as a colorless solid (584 mg, 97%). 1

Complex S4
A 25 mL Schlenk flask was equipped with a magnetic stir bar and was flame dried under vacuum. The flask was filled with argon and W(CAr)Br3(dme) (Ar = 4-methoxyphenyl) (S3) 2 (503 mg, 0.795 mmol) was dissolved in THF (9 mL). Then a solution of NaOtBu (232 mg, 2.37 mmol) in THF (2 mL) was added dropwise at 23°C to the stirred solution. Stirring was continued for 14 h at ambient temperature before the solvent was removed in vacuo to obtain a dark brown solid. A second, flame dried 50 mL Schlenk flask was equipped with a magnetic stir bar and a Celite ® (2 cm) packed argon frit. The dark brown solid was suspended in n-pentane (4 x 5 mL) and was filtered through the Celite ® pad. The resulting filtrate was concentrated and the residue dried under vacuum (10 -3 mbar) to give complex S4 as a brown solid (376 mg, 91%) free of any residual THF. 1   The complex is still contaminated with starting complex W(CO6), which can't be removed due to its low solubility, but turned out not to be problematic in the next step. 1

Complex 16c
A 100 mL Schlenk flask was equipped with a magnetic stir bar and was flame dried under vacuum. The flask was filled with argon and charged with ligand 15c (296 mg, 0.379 mmol), which was azeotropically dried with benzene (3 x 5 mL) to remove any residual water. Toluene (29 mL) was added and the mixture vigorously stirred for 10 min to obtain a clear solution before a solution of complex 4b (199 mg, 0.382 mmol) in toluene (6 mL) was added to the vigorously stirred mixture. After 2 h stirring at ambient temperature, the solvent was removed in vacuo and the yellow/orange solid was extracted with n-pentane (4 x 5 mL). The solvent was removed in vacuo to give complex 16c as a yellow/orange powder (407 mg, quant.

Complex 16d
A 100 mL Schlenk flask was equipped with a magnetic stir bar and was flame dried under vacuum. The flask was filled with argon and charged with ligand 15d (291 mg, 0.471 mmol), which was azeotropically dried with benzene (3 x 5 mL) to remove any residual water. Toluene (31 mL) was added and the mixture vigorously stirred for 10 min to obtain a clear solution, before a solution of complex 4b (217 mg, 0.417mmol) in toluene (6 mL) was added dropwise to the vigorously stirred mixture. After 2 h stirring at ambient temperature, the solvent was removed in vacuo and the yellow/orange solid was extracted with n-pentane (4 x 5 mL  This solution was filtered via cannula and the filtrate was stored at -20°C for one week to obtain very sensitive, red crystals suitable for single-crystal X-ray diffraction.  because of the sensitivity of the material, a correct elemental analysis has not been obtained.

Complex 18
The NMR sample of complex 17 was stored for one week at ambient temperatures upon which time complex 18 was formed. Not all shifts of the new complex 18 in the 1 H and 13 C NMR NMR could be assigned due to signal broadening and overlaps with complex 17 (see attached spectra for more details).     Dilution Experiment and DOSY NMR:       Result from the line-shape simulation at 283K. Red: simulated spectrum obtained from DNMR after fitting; black: experimental spectrum; blue: difference between the experimental and the simulated spectrum.

COMPUTATIONAL DETAILS
All geometry optimizations were performed with the Gaussian09 package 13 using the PBE0 functional. 14 W was represented by the quasi-relativistic effective core potential (RECP) from the Stuttgart group and the associated basis sets. [15][16][17] The remaining atoms (H, C, O, Si) were represented by a double-ζ Def2-SVP basis set. 18 Geometry optimizations were performed using the GD3 dispersion correction 19 and the SMD model 20 to account for the solvent (toluene) NMR calculations were performed within the GIAO framework using ADF 2014 21 with the PBE0 functional and Slater-type basis sets of double-ζ quality (DZ). Relativistic effects were treated by the 2 component zeroth order regular approximation (ZORA). 22 Analysis of scalar-relativistic natural localized molecular orbitals were done with the NBO 6.0 program. 23 Calculated NMR shielding tensors were analyzed using these scalar-relativistic NLMO. 24,25 The 3D representation of the calculated shielding tensors were obtained as polar plots 26 of functions ∑ij ri ijrj.