[3 + 2] Cycloadditions of Tertiary Amine N-Oxides and Silyl Imines as an Innovative Route to 1,2-Diamines

We have developed a one-pot synthetic method for producing 1,2-diamines from easily prepared and commercially available precursors through a formal umpolung process. Our method utilizes an efficient [3 + 2] cycloaddition as the key step in forming substituted 1,2-diamines in moderate to high yields. These resulting compounds can undergo subsequent transformations, demonstrating their utility as synthetic building blocks for more complex scaffolds. Finally, we propose a reasonable mechanism for this transformation using density functional theory modeling, justifying the experimental observations.

Our initial conditions involved mixing the 4-methoxyphenyl silyl imine 2a and TMAO together in a 1:1 ratio with three equivalents of base in THF. Excitingly, following aqueous workup we generated diamine 4a in good (70%) yield after a three-hour reaction time (Entry 1). Decreasing the reaction time (Entry 2) to two hours gave a mild improvement in yield, (75% yield). Further decreasing the reaction time had a negative impact on the yield, as did using fewer than three equivalents of LDA (Entries [3][4][5]. This agrees with our prior work, where we noted that two equivalents of LDA are required to doubly deprotonate the system, and a third is required to stabilize the transition state by complexing to it.
Diamine was prepared according to the general procedure using TMAO (184 mg, 2.

N,N-dimethylaniline-N-oxide (SI-3)
Dimethylaniline (1.26 mL, 10 mmol, 1.0 equiv.) was added to a round-bottom flask and dissolved in chloroform (10 mL, 1.0 M). The reaction was cooled to 0 °C in an ice bath, and 75% metachloroperoxybenzoic acid (2.36 g, 10 mmol, 1.0 equiv.) was added portion-wise over a period of 10 min. After the addition was complete, the reaction was warmed to RT and let to stir for 1 h. Potassium carbonate was then added and the reaction was stirred vigorously for 1 h. Solids were filtered off and the filtrate was dried over sodium sulfate. The sodium sulfate was filtered off and the filtrate was concentrated resulting in the pure N-oxide as a yellow solid. (91% yield,1.243 g, 9.1 mmol).

Procedure for the di-tosylation of 4a (5)
Under ambient conditions, a RBF flask was charged with 4a (31mg, 0.17 mmol, 1.0 equiv.), diisopropylethylamine (0.073 mL, 0.43 mmol, 2.5 equiv.) and DCM (1.7 mL, 0.1M). p-Toluenesulfonylchloride (67 mg, 0.35 mmol, 2.05 equiv.) was added all at once and the solution was stirred at RT while being monitored by TLC until consumption of starting material. After 2h, the organic layer was washed with DI H2O and the aqueous layers were rinsed with DCM (x2). The organic layers were combined, dried with MgSO4, and concentrated in vacu.

Procedure for the mono-tosylation of 4a (6)
Under ambient conditions, a RBF flask was charged with 4a (33mg, 0.18 mmol, 1.0 equiv.), diisopropylethylamine (0.061 mL. 0.36 mmol, 2.0 equiv.) and DCM (3.4 mL, 0.1M). The solution was cooled to -78 ⁰ C while p-tolulenesulfonylchloride (34mg, 0.18 mmol 1.0 equiv.) was dissolved in DCM (0.5 mL) in a separate flask. The sulfonylchloride/DCM suspension was added dropwise to the 4a solution, via syringe, over the course of 10 minutes at -78 ⁰ C. After addition, the solution was allowed to slowly warm to room temperature and was monitored by TLC until consumption of starting material. After 2h, the organic layer was washed with DI H2O and the aqueous layers were rinsed with DCM (x2). The organic layers were combined, dried with MgSO4, and concentrated in vacu. The compound was purified via flash column chromatography (5% MeOH in DCM) to afford a white solid. (72% yield, 43 mg, 0.13 mmol.) (Rf = 0.54 in 10:1 DCM/MeOH) 1

Procedure for the mono-nosylation of 4a (7)
Under ambient conditions, a RBF flask was charged with 4a (54 mg, 0.3 mmol, 1.0 equiv.), triethylamine (0.080 mL, 0.6 mmol, 2.0 equiv.) and DCM (3.0 mL, 0.1M). The solution was cooled to -78 ⁰ C while 2-nitrobenzenesulfonylchloride (66 mg, 0.3 mmol, 1.0 equiv.) was dissolved in DCM (0.5 mL) in a separate flask. The sulfonylchloride/DCM suspension was added dropwise to the 4a solution, via syringe, over the course of 10 minutes at -78 ⁰ C. After addition, the solution was allowed to slowly warm to room temperature and was monitored by TLC until consumption of starting material. After 2h, the organic layer was washed with DI H2O and the aqueous layers were rinsed with DCM (x2). The organic layers were combined, dried with MgSO4, and concentrated in vacu.

Procedure for the pivaloyl mono-functional of 4a (8)
Under ambient conditions, a RBF flask was charged with 4a (50 mg, 0.28 mmol, 1.0 equiv.), triethylamine (0.074 mL, 0.56 mmol, 2.0 equiv.) and DCM (2.8 mL, 0.1M). The solution was cooled to -78 ⁰ C while 2,2-dimethylpropanoyl chloride (pivaloyl chloride) (0.034 mL, 0.28 mmol, 1.0 equiv.) was dissolved in DCM (0.5 mL) in a separate flask. The pivaloyl chloride/DCM suspension was added dropwise to the 4a solution, via syringe, over the course of 10 minutes at -78 ⁰ C. After addition, the solution was allowed to slowly warm to room temperature and was monitored by TLC until consumption of starting material. After 2h, the organic layer was washed with DI H2O and the aqueous layers were rinsed with DCM (x2). The organic layers were combined, dried with MgSO4, and concentrated in vacu.

Computational Section General Information
Quantum chemistry methods of meta-hybrid density functional theory (DFT) 14 and second-order Møller-Plesset perturbation theory 15 were carried out at the Center for Computational Sciences (CCS) at Duquesne University using Gaussian 16. The M06-2X functional 16 with Dunning's jul-cc-pVDZ basis set 17 was used primarily to calculate electronic, enthalpic and free energies for both ground and transition structures. M06-2X, developed by Truhlar and co-workers, has been reported to be accurate to within 1.2 kcal/mol for reaction barriers and within 0.37 kcal/mol of non-covalent interaction energies. 16 To verify the accuracy of M06-2X/jul-cc-pDVZ calculations, a subset of activation enthalpies were calculated using expanded basis sets and/or higher levels of theory. Using the M06-2X functional the basis set was increased to a triple zeta (jul-cc-pVTZ) and more diffuse functions (aug-cc-pVDZ). In our previous published paper we verified our theory level, while also using jul-cc-pVDZ, with second-order Møller-Plesset calculations on the first deprotonation transition structure. 18 Vibrational frequency calculations were used to confirm all stationary points as either minima or transition structures and to provide thermodynamic corrections for enthalpies and free energies.

S-66
Sum of electronic and thermal Free Energies= -1312.163577