Cobalt-catalyzed deoxygenative triborylation of allylic ethers to access 1,1,3-triborylalkanes

Polyborylated organic compounds have been emerging as versatile building blocks in chemical synthesis. Here we report a selective cobalt-catalyzed deoxygenative 1,1,3-triborylation reaction of allylic ethers with pinacolborane to prepare 1,1,3-triborylalkane compounds. With naturally abundant and/or synthetic cinnamic methyl ethers as starting materials, we have achieved the synthesis of a variety of 1,1,3-triborylalkanes (25 examples). The synthetic utility of these 1,1,3-triborylalkanes is demonstrated through site-selective allylation, protodeborylation, and consecutive carbon-carbon bond-forming reactions. Mechanistic studies including deuterium-labeling and control experiments suggest that this 1,1,3-triborylation reaction proceeds through initial cobalt-catalyzed deoxygenative borylation of allylic ethers to form allylic boronates followed by cobalt-catalyzed 1,1-diborylation of the resulting allylic boronates.

1 H and 13 C spectra were recorded using Bruker 400 MHz, or 500 MHz NMR spectrometers and done in CDCl3 unless otherwise staed. 1 H NMR and 13 C NMR spectra were referenced to resonances of the residual signals of the deuterated solvents. Multiplicities are recorded as: s = singlet, d = doublet, t = triplet, dd = doublet of doublets, dt = doublet of triplets and m = multiplet. GC analysis was acquired on Agilent 6850 gas chromatograph equipped with a flame-ionization detector. GC-MS analysis was performed on Shimadzu GC-2010 gas chromatograph coupled to a Shimadzu QP2010 mass selective detector. HR-MS analyses were performed using Bruker micrOTOFQII (ESI).

General Procedure for Screening Reactions of Development of Catalytic 1,1,3-Triboronates Synthesis
In an Ar-filled glovebox, a 20-mL screw-capped vial was charged with cobalt(III) acetylacetonate (5.3 mg, 15.0 µmol), xantphos (8.7 mg, 15.0 µmol), norbornene, 2a (44.4 mg, 0.300 mmol), 1,3,5trimethoxybenzene (16.8 mg, 0.100 mmol), cyclohexane and a magnetic stirring bar. The solution was stirred and pincolborane was charged in the vial and sealed with a cap containing a PTFE septum. The vial was removed from the glovebox and stirred at the stated temperature for 2 h, after which, GC analysis was done.

Substrate Synthesis
Cinnamic methyl ether substrates were prepared from modified conditions of the literature. [5] (E)-Cinnamalcohol (1) was commercial available.

Supplementary Equation 1. Preparartion of (E)-Cinnamic Methyl Ether.
To a solution of triethyl phosphonoacetate (2.27 mL, 14.0 mmol) in THF (50 mL), NaH (60% dispersion in mineral oil, 0.56 g, 14.0 mmol) was added portion-wise at 0 ˚c. After 30 minutes, benzaldehyde (10 mmol) was added dropwise at 0 ˚c and stirred at room temperature for 1h. The reaction was quenched with saturated aq. NH4Cl solution and extracted with diethyl ether (3 x 50 mL). The organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude product was purified and isolated as a colourless oil using silica gel flash column chromatography with hexane/EA as eluent. The purified acrylate was diluted in Et2O (100 mL) and DIBAL-H (1.0 M in toluene, 30 mL, 30 mmol) was added dropwise at -78 ˚c and stirred for 2h. The reaction was quenched with a saturated aq. solution of Rochelle salt and stirred for 1 h. The aqueous phase was extracted with diethyl ether (3 x 50 mL) and the organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude allylic alcohol was engaged in the next step without further purification, dissolving in THF (50 mL) and cooled to 0 ˚c before NaH (60% dispersion in mineral oil, 0.48 g, 12.0 mmol) was added portion-wise. Methyl iodide (1.90 mL, 30 mmol) was added dropwise at 0 ˚c and stirred at room temperature for 18h. The reaction was quenched with saturated aq. NH4Cl solution and extracted with diethyl ether (3 x 50 mL). The organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude product was purified and isolated as a colourless oil using silica gel flash column chromatography with hexane/EA as eluent to give (E)-cinnamic methyl ethers. 1 H NMR (500 MHz, CDCl3) δ 7.64 -7.55 (m, 4H), 7.51 -7.41 (m, 4H), 7.38 -7.32 (m, 1H), 6.66 (d, J = 16.0 Hz, 1H), 6.34 (dt, J = 16.0, 6.0 Hz, 1H), 4.13 (dd, J = 6.0, 1.2 Hz, 2H), 3.42 (s, 3H). 13  (2-Methoxyethyl)triphenylphosphonium bromide (5.61 g, 14.0 mmol) was suspended in THF (100 mL) and NaH (60% dispersion in mineral oil, 0.56 g, 14.0 mmol) was added portion-wise at room temperature. After 1h, benzaldehyde (10 mmol) was added slowly and stirred for 24h at the same temperature. The reaction was quenched with saturated aq. NH4Cl solution and extracted with diethyl ether (3 x 50 mL). The organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude product was purified and isolated as a colourless oil using silica gel flash column chromatography with hexane/EA as eluent to give cinnamic methyl ethers of E/Z mixture.

Supplementary Equation 3. Preparartion of 2-Aryl-allylmethyl Ether.
To a solution of N,2-Dimethoxy-N-methylacetamide (1.60 g, 12.0 mmol) synthesized by the literature reported method in THF (50 mL), Grignard solution (10 mmol) was added portion-wise at 0 ˚c. The reaction was stirred at room temperature and monitored with TLC for the completion of reaction. The reaction was quenched with saturated aq. NH4Cl solution and extracted with diethyl ether (3 x 50 mL). The organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude product was purified and isolated as a colourless oil using silica gel flash column chromatography with hexane/EA as eluent and used for the next step. The ketone was then dissolved in THF (10 mL) and slowly added to a stirring mixture of methyltriphenylphosphonium bromide (5.00 g, 14.0 mmol) and potassium tert-butoxide (1.57 g, 14.0 mmol) in THF (20 mL) at 0 ˚c. The mixture was warm to room temperature and stirred for three hours. The reaction was quenched with saturated aq. NH4Cl solution and extracted with diethyl ether (3 x 50 mL). The organic phases were combined and washed with brine and dried over Na2SO4, after which, the solvent was removed under reduce pressure. The crude product was purified and isolated as a colourless oil using silica gel flash column chromatography with hexane/EA as eluent to give the 2aryl-allylmethyl ether.  Ar OMe

2-(1-(4-methoxyphenyl)-5-methyl-1-phenylhex-5-en-3-yl)thiophene (16)
To a solution of 15 (40.6 mg, 0.100 mmol) in THF (1 mL) was added 2-thienyllithium (0.5 mL, 1 M in THF/Hexene, 0.500 mmol) dropwise at -78 ˚c and stirred for 2 h before the solution was warmed to room temperature. N-Bromosuccinimide (NBS, 20.9 mg, 0.120 mmol, 0.1 M in MeOH) was added dropwise. The reaction was stirred at room temperature for another 1 h, after which, a saturated solution of Na2S2O3 (2 mL) was added and the mixture was diluted with Et2O (10 mL). The organic layer was separated and the aqueous layer was extracted with Et2O. The combined organic layers were washed with brine and dried over Na2SO4. The solvent was removed under reduce pressure. 16 was isolated (26.4 mg, 0.073 mmol, 73%, dr 4.0:1) was purified and isolated as a yellowish oil using flash chromatography on silica gel (100:0 to 100:1 hexane/EtOAc). Rf = 0.8 (hexane/EtOAc 10:1). 1   In using sodium tert-butoxide accompanied by MeOH as the protic source, 11 underwent protodeborylation at the gem-bisboron containing carbon center to form 1,3-bisboron compound 14 in an anti-diastereoselective fashion. This diastereoselectivity can be explained by the proposed carbanion model in the presences of Bronsted base. The chelated model A, with the phenyl-substituent on the equatorial position anti to the substituent on the carbanion minimizes steric strain and allows approach from small electrophile to form anti-diastereoselective 14.

Supplementary Equation 14. Reaction of 2a-D with HBpin.
In an Ar-filled glovebox, a 20-mL screw-capped vial was charged with cobalt(III) acetylacetonate (5.3 mg, 15.0 µmol), xantphos (8.7 mg, 15.0 µmol), nbe (141.2 mg, 1.500 mmol), 2a-D (0.300 mmol), cyclohexane (10 mL) and a magnetic stirring bar. The solution was stirred and pincolborane (261.2 µL, 1.800 mmol) was charged in the vial and sealed with a cap containing a PTFE septum. The vial was removed from the glovebox and stirred at 100 ˚c for 2 h, after which, the crude product was purified using silica gel flash column chromatography (column I.D. 13.4 mm) with hexane/EtOAc. Next, chloroform-d1 (99.8 atom % D, 2 equivalent) was added and dissolved in chloroform (0.6 mL) for 2 H NMR analysis. The sample was then dried for 1 H NMR and quantitative 13 C NMR analysis. In an Ar-filled glovebox, a 20-mL screw-capped vial was charged with cobalt(III) acetylacetonate (5.3 mg, 15.0 µmol), xantphos (8.7 mg, 15.0 µmol), nbe (141.2 mg, 1.500 mmol), 2a (0.300 mmol), cyclohexane (10 mL) and a magnetic stirring bar. The solution was stirred and DBpin (1.800 mmol) was charged in the vial and sealed with a cap containing a PTFE septum. The vial was removed from the glovebox and stirred at 100 ˚c for 2 h, after which, the crude product was purified using silica gel flash column chromatography (column I.D. 13.4 mm) with hexane/EtOAc (100:1 to 20:1). Next, chloroform-d1 (99.8 atom % D, 1 equivalent) was added and dissolved in chloroform (0.6 mL) for 2 H NMR analysis. The sample was then dried for 1 H NMR and quantitative 13 C NMR analysis. The title compound was synthesized according to the known procedures. [14] The title compound was isolated as a colorless oil after chromatography on silica gel (100:1 hexane/EtOAc). 1  In an Ar-filled glovebox, a 10-mL screw-capped vial was charged with cobalt(III) acetylacetonate (1.8 mg, 5.0 µmol), xantphos (2.9 mg, 5.0 µmol), nbe (47.1 mg, 0.500 mmol), 17 (0.100 mmol), cyclohexane (3 mL) and a magnetic stirring bar. The solution was stirred and HBpin (1.200 mmol) was charged in the vial and sealed with a cap containing a PTFE septum. The vial was removed from the glovebox and stirred at 100 ˚c for 2 h, after which, the crude product was determined by GCMS with internal standard to give a ratio of 24:76 (7a:8a).

Supplementary Equation 18. Preparation of compound 19.
The title compound was synthesized according to the known procedures. [15,16] 11B NMR of crude products Supplementary Equation 20. Reaction of 2a with HBpin.
In an Ar-filled glovebox, a 20-mL screw-capped vial was charged with cobalt(III) acetylacetonate (5.3 mg, 15.0 µmol), xantphos (8.7 mg, 15.0 µmol), nbe (141.2 mg, 1.500 mmol), 2a (0.300 mmol), cyclohexane (10 mL) and a magnetic stirring bar. The solution was stirred and HBpin (1.800 mmol) was charged in the vial and sealed with a cap containing a PTFE septum. The vial was removed from the glovebox and stirred at 100 ˚c for 2 h, after which, the crude product was dissolved in chloroform-d1 (0.6 mL) for 11 B NMR analysis. The chemical shift of Bpin-OMe (V, 22.3 ppm) is matched with the known literature. [18] Supplementary Figure 10. 11 B NMR of the product from the reaction of 2a with HBpin.

Reactions of 6 with HBpin
Supplementary Equation 21. Reaction of 6 with HBpin.
Compound 6 was treated under standard conditions, which resulted in no conversion of 6.

Details of Single-Crystal X-ray Diffraction Analysis of 8g
Supplementary Figure 11. X-ray structure of 8g.
A specimen of C27H44B3FO6, approximate dimensions 0.524 mm x 0.559 mm x 0.637 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured (λ = 0.71073 Å).
The structure was solved and refined using the Bruker SHELXTL Software Package, with Z = 4 for the formula unit, C27H44B3FO6. The final anisotropic full-matrix least-squares refinement on F 2 with 405 variables converged at R1 = 6.28%, for the observed data and wR2 = 15.94% for all data. The goodness-of-fit was 1.093. The largest peak in the final difference electron density synthesis was 0.523 e -/Å 3 and the largest hole was -0.483 e -/Å 3 with an RMS deviation of 0.053 e -/Å 3 . On the basis of the final model, the calculated density was 1.149 g/cm 3 and F(000), 1112 e -.