Efficient catalytic alkyne metathesis with a fluoroalkoxy-supported ditungsten(III) complex

The molybdenum and tungsten complexes M2(OR)6 (Mo2F6, M = Mo, R = C(CF3)2Me; W2F3, M = W, R = OC(CF3)Me2) were synthesized as bimetallic congeners of the highly active alkyne metathesis catalysts [MesC≡M{OC(CF3)nMe3−n}] (MoF6, M = Mo, n = 2; WF3, M = W, n = 1; Mes = 2,4,6-trimethylphenyl). The corresponding benzylidyne complex [PhC≡W{OC(CF3)Me2}] (WPhF3) was prepared by cleaving the W≡W bond in W2F3 with 1-phenyl-1-propyne. The catalytic alkyne metathesis activity of these metal complexes was determined in the self-metathesis, ring-closing alkyne metathesis and cross-metathesis of internal and terminal alkynes, revealing an almost equally high metathesis activity for the bimetallic tungsten complex W2F3 and the alkylidyne complex WPhF3. In contrast, Mo2F6 displayed no significant activity in alkyne metathesis.


Experimental procedures
[Mo 2 {OC(CF 3 ) 2 Me} 6 ] (Mo2F6) [1] A solution of 400 mg (0.68 mmol) [Mo 2 Cl 6 (dme) 2 ] [1] in 25 mL CH 2 Cl 2 was treated with 933 mg (4.08 mmol) NaOC(CF 3 [7] 4.65 g (40 mmol) 1-Methyl-4-ethynylbenzene dissolved in 140 mL THF were treated with 50 mL (80 mmol, 1.6 M in hexane) n-BuLi at −20 °C. The reaction mixture was stirred for one hour at low temperature and 7.2 mL (14.2 g, 100 mmol) methyl iodide were added dropwise at −20 °C. The solution was warmed to room temperature and stirred for six hours. The reaction was quenched with a saturated NH 4 Cl solution (100 mL) and the aqueous phase was extracted with ethyl acetate (3 × 100 mL). The combined organic phases were dried over Na 2 SO 4 and the solvent was removed in vacuo. The colorless liquid was distilled at 3 mbar and 65 °C (3.38 g, 26 mmol, 65%).  [7] 2.0 g (11.1 mmol) 1-Bromo-4-ethynylbenzene dissolved in 60 mL THF were treated with 6.9 mL (11.1 mmol, 1.6 M in hexane) n-BuLi at −78 °C. The reaction mixture was stirred for one hour at low temperature and 1.6 mL (3.14 g, 22.1 mmol) methyl iodide were added dropwise at −78 °C. The solution was warmed to room temperature and stirred for three hours. The reaction was quenched with a saturated NH 4 Cl solution (50 mL) and the aqueous phase was extracted with CH 2 Cl 2 (3 × 80 mL). The combined organic phases were dried over MgSO 4 and the solvent was removed in vacuo. After flash chromatography (hexane) the product was isolated as a colorless oil (1.53 g, 7.8 mmol, 71%).  [7] 500 mg (3.78 mmol) 1-Methoxy-4-ethynylbenzene dissolved in 15 mL THF were treated with 2.4 mL (3.78 mmol, 1.6 M in hexane) n-BuLi at −20 °C. The reaction mixture was stirred for one hour at low temperature and 0.6 mL (7.56 mmol) methyl iodide were added dropwise at −20 °C. The solution was warmed to room temperature and stirred for 16 hours. The reaction was quenched with a saturated NH 4 Cl solution (30 mL) and the aqueous phase was extracted with CH 2 Cl 2 (3 × 70 mL). The combined organic phases were dried over MgSO 4 and the solvent was removed in vacuo. After flash chromatography (hexane) the product was isolated as a colorless oil (475 mg, 3.25 mmol, 86%      General procedure for self-metathesis. Under an argon atmosphere, a flask was charged with the substrate (0.5 mmol), molecular sieves 5 Å (500 mg) and toluene (internal alkynes: 2.5 mL, terminal alkynes: 24 mL). Then the catalyst (0.5 mol % W2F3, 1 mol % W Ph F3) was added and the mixture was stirred for 2 h at room temperature. The catalyst and the molecular sieves were removed by filtration through alumina and the solvent was evaporated. The crude reaction product was purified by flash chromatography on silica gel with ethyl acetate-hexane (1:8).

4-Methoxy-1-propinylbenzene
General procedure for RCAM. Under an argon atmosphere, a flask was charged with the substrate (0.5 mmol), molecular sieves 5 Å (1.0 g) and toluene (24 mL). Then the catalyst (1 mol % W2F3, 2 mol % W Ph F3) was added and the mixture was stirred for 2 h at room temperature. The catalyst and the molecular sieves were removed by filtration through alumina and the solvent was evaporated. The crude reaction product was purified by flash chromatography on silica gel with ethyl acetate-hexane (1:8).
General procedure for ACM. Under an argon atmosphere, a flask was charged with the substrate (0.5 mmol), TMS-propyne or TMS-acetylene (1.0 mmol), molecular sieves 5 Å (500 mg) and toluene (internal alkynes: 2.5 mL, terminal alkynes: 24 mL). Then the catalyst (1 mol % or 2 mol % W2F3) was added and the mixture was stirred for 2 h at room temperature. The catalyst and the molecular sieves were removed by filtration through alumina and the solvent was evaporated. The crude reaction product was purified by flash chromatography on silica gel with ethyl acetate-hexane (1:10).

X-ray structure analysis
Numerical data are collected in table S1. Single crystals were mounted on glass fibers (W2F3·NHMe 2 and W Ph F3) or on top of a human hair (W2F3) in perfluorinated inert oil. Measurements were performed on Oxford Diffraction Xcalibur diffractometers using monochromated Mo Kα radiation. Absorption corrections were applied on the basis of multi-scans and additional absorption correction based on face indexing and integration on a Gaussian grid. Data reduction was performed with Crysa-lisPro. [13] The structures were solved by intrinsic phasing with SHELXT-2014/5 [14] and refined on F 2 using the program SHELXL-2017/1 [15] . The Hydrogen H1 on the nitrogen atom in W2F3·NHMe 2 was refined freely. All other H atoms in all the reported crystal structures were placed in idealized positions and refined using a riding model. for each tungsten atom are given in Table S1.  (4) This structure only confirms the connectivity of this molecule and discussion of any bond length is not meaningful. This kind of disorder is an inherent problem in M 2 X 6 systems resulting from a pseudo octahedral ligand arrangement and an internal flip mechanism for the metal atoms. [16] The different positions of the metal atoms on the disordered metal-metal vector occupy the edges of an imaginary cuboid.
W Ph F3. Two alkoxide ligands were refined with a disorder model comprising two or three different positions, respectively. The simulated precession images of this dataset (see Figures S11 and S12) show very weak satellite peaks in the 0kl and hk0 planes for low diffraction angles only. Meaningful integration of these peaks was not possible. We assume that these satellite peaks result from a modulation of the crystal structure and that the refined disorder model is the most eligible approximation. S12 Figure S11: Simulated precession image of the 0kl plane showing weak satellite peaks for small diffraction angles. [13] Figure S12: Simulated precession image of the hk0 plane showing weak satellite peaks for small diffraction angles. [13] Complete data have been deposited with the Cambridge Crystallographic Data Centre under the CCDC numbers 1850924−1850926. These data can be obtained free of charge from http://www.ccdc.cam.ac.uk/.