Aminofluorination: transition-metal-free N–F bond insertion into diazocarbonyl compounds

Gem-aminofluorination of diazocarbonyl compounds has been achieved for the first time.


General information
All reactions were carried out with standard Schlenk techniques under argon. All reagents were used as received from commercial suppliers unless otherwise stated. All solvents were purified by distillation following standard procedures. Reaction progress was monitored by thin layer chromatography (TLC) and components were visualized by observation under UV light at254nm.Flash column chromatography was performed using silica gel 60 (200-300 mesh). All 1 H NMR, 13 C NMR and 19 F NMR spectra were recorded on Bruker AV-III 400 in CDCl 3 . Chemical shifts were reported in parts per million (ppm, δ). Proton nuclear magnetic resonance ( 1 H NMR) spectra is referenced to the peak of tetramethylsilane (δ = 0.00) and reported as follows: chemical shift (ppm), multiplicity (s = singlet, t = triplet, q = quartet, m = multiplet) and coupling constant (Hz). Carbon-13 nuclear magnetic resonance ( 13 C NMR) spectra is referenced tothe solvent center peak of CDCl 3 (δ = 77.0). CAUTION! Even though we have noted no explosive tendencies of the diazo compounds, it is strongly recommended that they should be handled with great care and proper protection. According to a known procedure [1][2][3] , to the solution of ethyl phenylacetate (5mmol) and 4-methylbenzenesulfonyl azide (TsN 3 ) (1.24 g, 6mmol) in anhydrous CH 3 CN or THF (40 mL) was added 1,8-diazabicyclo-[5.4.0]-undec-7-ene (DBU) (1.14 g, 7.5 mmol) slowly at room temperature. Then the reaction mixture was stirred at room temperature for 15 hours. After water (40 mL) was added, the resulting mixture was extracted with diethyl ether (3 × 40 mL). The combined organic layer was washed with brine (20 mL) and dried over anhydrous Na 2 SO 4 . After removing the solvent under reduced pressure, the residual was purified by column chromatography on silicagel to give 1a-1k, 1m, 1o-1q, 1s-1v.

Nitrogen evolution
The solution of diazoester 1a ( (1-bromo-4-fluorobenzene). All reaction was monitored to 0~20 % yield of 2. Initial rates were obtained by linear fit of the concentration-time plot ( Figure S2 and Figure S3). A plot of ln(initial rate) vs. ln ([1a]) or In([NFSI]) showed that the rate for the germinal aminofluorination of diazoesters is first-order in both 1a and NFSI ( Figure S4 and Figure S5, equation 1).

Rate = k[1a][NFSI]
(1) Effect of temperature. The effect of temperature on initial rate for the reaction between diazoesters (0.15 M) and NFSI (0.10 M) in DCE were studied from 313 to 353 K ( Figure S6). The activation parameters were obtained from the plot of ln(initial rate/T) vs 1/T according to Eyring equation. ΔH ‡ and ΔS ‡ were found to be 17.1 kcal mol -1 and 13.0 cal mol -1 K -1 , respectively. The negative ∆S ‡ value suggests that a bimolecular transition state involving NFSI and 1a might be generated.

Computational method
All the calculations were carried out by using ORCA program package. 5 Full geometry optimization and frequency calculation were performed by using B3LYP functional 6,7 coupled with def2-SVP 8 basis set for all atoms. A larger basis set of def2-TZVPP4 was employed for single point energy corrections. To improve computational speed, the RIJCOSX approximation [9][10][11] in combination with def2-SVP/J and def2-TZVPP/J 12 auxiliary basis sets was applied. Dispersion effects were computed by using the well-established dispersion corrections D3 with Becke-Johnson damping scheme. 13,14 Solvation effects were taken into account by the universal solvation model based on solute electron density (SMD) 15 with the conductor-like screening model (COSMO). 16

Calculated potential energy surface
As shown in Scheme S1, four possible transformation pathways from reactant 1a to product 2a were calculated to explore the reaction mechanism. The calculated barriers of the first step in each pathway are collected in Table S1. Scheme S1. The first step of the four possible transformation pathways from reactant 1a to product 2a. The major geometric differences were marked in red.
As can be seen from Table S1, pathway D is the lowest energy pathway where a fluorine transfer transition state (TS D ) is located with an enthalpy barrier of 16.2 kcal mol −1 , which is in good agreement with experimental value of 17.1 kcal mol −1 . The C−F and N−F bond lengths in TS D are 1.772 Å and 1.920 Å, respectively ( Figure S8).
The following intermediate Int 1 after C−F bond formation and C−N bond breaking is exothermic by 25.6 kcal mol −1 ( Figure S9). Though the C-N bond has not been formed yet, these two atoms are quite close with a distance of 2.858 Å ( Figure S8e), indicating the intermediate Int 1 is an ionic pair. All attempts to find other possible mechanisms excluding those shown in Scheme S1 failed as the obtained energy barriers were higher than 32 kcal mol −1 which were not possible in real experiment.
The second step in pathway D was found to proceed in no barrier fashion with N 2 leaving. Thus, the corresponding intermediate Int 2 is exothermic by 12.8 kcal mol −1 .