Regioselective Monoborylation of Spirocyclobutenes

We present a strategy for the synthesis of spirocyclic cyclobutanes with modulable exit vectors based on the regioselective monoborylation of spirocyclobutenes. Using an inexpensive copper salt and a commercially available bidentate phosphine, a broad variety of borylated spirocycles have been prepared with complete regiocontrol. The boryl moiety provides a synthetic handled for further functionalization, allowing access to a wide array of spirocyclic building blocks from a common intermediate.

After being stirred for 12 h at room temperature, water (2.5 mL/mmol of alcohol) and CH2Cl2 (2.5 mL/mmol of alcohol) were added to the reaction mixture. Then, the layers were separated and the aqueous phase was extracted with CH2Cl2 (3x). The combined organic phases were dried (MgSO4), filtered and the solvent evaporated under reduced pressure. The residue was purified by flash column chromatography (SiO2; cyclohexane/EtOAc 80:20) to afford the desired product SI-3 (9.42 g, 25.6 mmol) in 91% as a white solid. The spectral data matched with those previously reported. 1
The reaction mixture was redissolved in THF (2 mL/mmol) and cooled to 0 °C. Then, 1M aqueous solution of NaOH (1 mL/mmol) and H2O2 (0.2 mL/mmol) were successively added dropwise. The mixture was stirred at room temperature for 1h. After this time, water was added and the phases were separated. The aqueous phase was extracted with EtOAc (3x), the combined organic layers dried (MgSO4) and the solvent removed in vacuo. The residue was purified by flash column chromatography (SiO2, CHCl3/i-PrOH 95:5, flash column chromatography repeated twice to separate from traces of oxidation product of 2j) to afford alcohol SI-4 (265 mg, 1.17 mmol) in 83% yield as a colourless oil. A solution of DMSO (165 μL, 2.33 mmol, 2 equiv) in CH2Cl2 (1 mL) was added to a solution of oxalylchloride (200 μL, 2.33 mmol, 2 equiv) in CH2Cl2 (4 mL) at -78 °C, under an argon atmosphere. The mixture was stirred for 15 min at -78 °C and a solution of alcohol SI-4 (265 mg, 1.17 mmol) in CH2Cl2 (1 mL) was added dropwise. After stirring for 30 min, Et3N (652 μL, 4.68 mmol, 4 equiv) was added to the mixture and then the reaction was allowed to warm to 0 °C. The reaction was stirred for further 45 min. Then, aqueous NaHCO3 was added and the mixture was extracted with CH2Cl2 (3x), dried (MgSO4) and concentrated under reduced pressure. The residue was purified by flash column chromatography (SiO2; hexane/EtOAc 80:20) to afford the corresponding aldehyde 4 (215 mg, 0.95 mmol) in 82% yield as a colourless oil.
In order to carry out the dehydration, the aldol product was dissolved in toluene (0.7 M) and a catalytic amount of p-TsOH·H2O (69.4 mg, 0.365 mmol, 0.5 equiv) was added. The reaction was refluxed (oil bath) for 3 h. Once the reaction was finished, it was diluted with EtOAc (5 mL), washed with a saturated solution of NaHCO3, dried over MgSO4, and the solvent removed under reduced pressure. The crude was used in the next step without further purification.

Vinylation Synthesis of tert-butyl 2-vinyl-7-azaspiro[3.5]nonane-7-carboxylate, 7
An oven-dried vial was charged with boronic ester 2a (70.3 mg, 0.2 mmol, 1 equiv) and was dissolved in anhydrous THF (1 mL). 2 Then, vinyl magnesium bromide (0.8 mmol, 1.0 M in THF, 4.0 equiv) was added dropwise at a 0 °C, and the resulting solution was stirred for 30 min at room temperature. After cooling the reaction mixture at -78 °C (dry ice/acetone), a solution of iodine (203.0 mg, 0.8 mmol, 4.0 equiv) in anhydrous THF (1.7 mL) was added dropwise followed by stirring for 20 min. After this time, the solution was warmed at 0 °C, and a suspension of NaOMe (86.4 mg, 1.6 mmol, 8.0 equiv) in methanol (1 mL) was added in a single portion and was stirred for a further 30 min at that temperature. Then, a saturated aqueous solution of Na2S2O3 (7 mL) was added, followed by DCM (10 mL). The phases were separated and the aqueous phase was extracted with DCM (2 × 10 mL). The combined organic phases were dried (MgSO4) and concentrated under reduced pressure. The crude residue was purified by flash column chromatography (SiO2; cyclohexane/EtOAc 90:10 to 80:20) to afford the desired alkene 7 (25.2 mg, 0.100 mmol) in 50% yield as colorless oil.

Trifluoroborate salt Synthesis of Potassium Trifluoroborate Salt of 2m, 9
An oven-dried vial was charged with boronic ester 2m (240 mg, 0.742 mmol, 1.0 equiv) and was dissolved methanol (13 mL). Then, KHF2 (4.5 M aqueous solution, 0.97 mL) was added dropwise, and the resulting solution was stirred for 30 min at room temperature. 9 After this time, the residue was re-dissolved in methanol (6 mL) and water (4 mL). Again, all volatiles were removed under reduced pressure. This cycle was repeated (x10) to remove all pinacol. The solid that was obtained was triturated with acetone (3 x 5 mL) and filtered through a plug of Celite®. The solvent was evaporated to yield the trifluoroborate salt 9 (192 mg, 0.633 mmol) in 85% yield as a white solid. mp = 226-228 °C.

Comparison of the Copper-Catalyzed Borylation of Cyclobutene 1a and Bromide SI-6
An alternative approach to prepare monoborylated spirocycles, as those describe in this work, could be through copper-catalyzed borylation of the corresponding bromide. To compare both approaches we prepared the spirocyclobutyl bromide SI-6. 5 In our hands, the preparation of the starting bromide compared to the cyclobutene is less convenient as it requires refluxing in CH3CN for 3 days to obtain the product in only 66% yield. Regarding the borylation, when we used the conditions optimized for our cyclobutenes (0.5 equiv of KOt-Bu) the borylated product 2a was obtained in only 39% yield. Increasing the amount of base to 1 equiv (as Ito published for alkyl bromides) 6 the yield of the borylated product increased to 67% yield. Although they are both valid approaches, perhaps the cyclobutene is more convenient as it doesn't require a stochiometric amount of base and opens the door to explore the difunctionalization of the double bond.

Additional Computational Results and Computational Details
All the calculations reported in this paper were performed with the Gaussian 09 suite of programs. 7 Electron correlation was partially taken into account using the hybrid functional usually denoted as B3LYP 8 in conjunction with the D3 dispersion correction suggested by Grimme et al. 9 using the standard double- quality def2-SVP 10 basis set for all atoms. Solvent effects (solvent = tetrahydrofuran) were taken into account by means of the Polarization Continuum Model (PCM) 11 method. This level is denoted PCM(tetrahydrofuran)-B3LYP-D3/def2-SVP. Geometries were fully optimized in solution without any geometry or symmetry constraints. Reactants, intermediates, and products were characterized by frequency calculations, 12 and have positive definite Hessian matrices. Transition structures (TS's) show only one negative eigenvalue in their diagonalized force constant matrices, and their associated eigenvectors were confirmed to correspond to the motion along the reaction coordinate under consideration using the Intrinsic Reaction Coordinate (IRC) method.

S21
Cartesian coordinates (in Å) and free energies (in a.u.) of all the stationary points discussed in the text. All calculations have been performed at the PCM(tetrahydrofuran)-B3LYP-D3/def2-SVP level.

X-RAY Data
Crystal Structure Report for compound 2l 13 Compound 2l was recrystallized (hexane) in order to obtain appropriate crystals for X-ray analysis.
A single crystal of compound 2l was covered with a layer of an inert mineral oil, mounted on a MiTeGen micromount with the aid of a microscope, and placed under a low temperature nitrogen stream. The intensity data set was collected at 250 K on a Bruker Kappa Apex II diffractometer equipped with a Mo sealed tube and graphite monocromator. The dataset was integrated with SAINT, the structure was solved with SHELXS-97 and the model refined by a least-squares method against F2 with SHELXL-2014. Data were corrected for absorption effects using the multi-scan method SADABS. All the hydrogen atoms were positioned geometrically and refined using a riding model, while the non-hydrogen atoms were refined anisotropically. 14 Figure S2: X-ray crystallography of compound 2l.

Data collection details for 2l
A clear colourless prismatic-like specimen of C19H34BNO4, approximate dimensions 0.049 mm x 0.056 mm x 0.229 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The integration of the data using a triclinic unit cell yielded a total of 30136 reflections to a maximum θ angle of 25.34° (0.83 Å resolution), of which 3765 were independent (average redundancy 8.004, completeness = 99.4%, Rint = 5.28%, Rsig = 3.33%) and 2546 (67.62%) were greater than 2σ(F 2 ). The final cell constants of a = 6.1650(4) Å, b = 11.7928(7) Å, c = 14.2810(8) Å, α = 86.510(3)°, β = 88.691(3)°, γ = 86.448(3)°, volume = 1034.16(11) Å 3 , are based upon the refinement of the XYZ-centroids of reflections above 20 σ(I). The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.9830 and 0.9960. The final anisotropic full-matrix least-squares refinement on F 2 with 233 variables converged at R1 = 5.65%, for the observed data and wR2 = 18.74% for all data. The goodness-of-fit was 1.007. The largest peak in the final difference electron density synthesis was 0.427 e -/Å 3 and the largest hole was -0.262 e -/Å 3 with an RMS deviation of 0.059 e -/Å 3 . On the basis of the final model, the calculated density was 1.128 g/cm 3 and F(000), 384 e -.  Table 3. Atomic coordinates and equivalent isotropic atomic displacement parameters (Å 2 ) for 2l U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.