Synthesis of 2-Oxazolines from Ring Opening Isomerization of 3-Amido-2-Phenyl Azetidines

Chiral 2-oxazolines are valuable building blocks and famous ligands for asymmetric catalysis. The most common synthesis involves the reaction of an amino alcohol with a carboxylic acid. In this paper, an efficient synthesis of 2-oxazolines has been achieved via the stereospecific isomerization of 3-amido-2-phenyl azetidines. The reactions were studied in the presence of both Brønsted and Lewis acids, and Cu(OTf)2 was found to be the most effective.

Recently, we reported the synthesis of chiral 4,5-dihydrothiazol-2-amines and 4,5dihydrooxazol-2-amines through an unexpected ring opening reaction of azetidines [19]. Based on this strategy, we now extend our research work and report herein the asymmetric synthesis of 2-oxazolines via the stereospecific isomerization of 3-amido-2-phenyl azetidines in the presence of acid (Scheme 1).
Recently, we reported the synthesis of chiral 4,5-dihydrothiazol-2-amines and 4,5dihydrooxazol-2-amines through an unexpected ring opening reaction of azetidines [19]. Based on this strategy, we now extend our research work and report herein the asymmetric synthesis of 2-oxazolines via the stereospecific isomerization of 3-amido-2-phenyl azetidines in the presence of acid (Scheme 1).
After optimizing the reaction conditions, we extended the substrate scope and different amides were examined (Scheme 4, Table 2). The amides with aryl, heteroaryl and alkyl groups were successfully isomerized to 2-oxazolines in the presence of Cu(OTf)2 in high yields ( Table 2, entries 1-9 and 13-16). However, those substrates with 2-hydroxyl or 2amino groups could not be isomerized, presumably due to their coordination to Cu(OTf)2. Nevertheless, these substrates could be isomerized to 2-oxazolines in the presence of CF3COOH in good yields ( A proposed mechanism for this transformation is shown below (Scheme 5). The isomerization of amides 3 occurred regiospecifically by presumably an SN2 nucleophilic attack at the more active C2 but not the less-hindered C4 of the azetidine ring, thus the stereochemistry of 2-oxazolines 4 was shown to be cis [21]. This is also supported by comparison of the coupling constant (10 Hz) between H4 and H5 with reported cis-2-oxazolines [22,23]. In addition, all the structures of 2-oxazolines were established by 1 H and 13 C NMR, high resolution mass spectra (HRMS), IR. Furthermore, the absolute configuration of 4l was further confirmed by X-ray analysis (Figure 1).  A proposed mechanism for this transformation is shown below (Scheme 5). The isomerization of amides 3 occurred regiospecifically by presumably an SN 2 nucleophilic attack at the more active C2 but not the less-hindered C4 of the azetidine ring, thus the stereochemistry of 2-oxazolines 4 was shown to be cis [21]. This is also supported by comparison of the coupling constant (10 Hz) between H4 and H5 with reported cis-2oxazolines [22,23]. In addition, all the structures of 2-oxazolines were established by 1 H and 13 C NMR, high resolution mass spectra (HRMS), IR. Furthermore, the absolute configuration of 4l was further confirmed by X-ray analysis (Figure 1).

General Information
All reactants and reagents were commercially available and were used without fur-

General Information
All reactants and reagents were commercially available and were used without further purification. 1 H and 13 C NMR spectra were recorded on a Bruker Advance III 400 MHz spectrometer (Billerica, MA, USA). Chemical shifts are reported in δ values (ppm) relative to an internal reference (0.03% v/v) of tetramethylsilane (TMS) for 1 H NMR or the solvent signal, chloroform (CDCl3), for 13 C NMR. IR data was obtained with an IRAffinity-1 spectrometer (Shimadzu, Kyoto, Japan). High resolution mass spectrometry (HRMS) was conducted with a high-resolution LCT Premier XE mass spectrometer in positive ESI mode (Waters, MA, USA). Melting points were measured on a digital melting point apparatus and are uncorrected.

General Information
All reactants and reagents were commercially available and were used without further purification. 1 H and 13 C NMR spectra were recorded on a Bruker Advance III 400 MHz spectrometer (Billerica, MA, USA). Chemical shifts are reported in δ values (ppm) relative to an internal reference (0.03% v/v) of tetramethylsilane (TMS) for 1 H NMR or the solvent signal, chloroform (CDCl 3 ), for 13 C NMR. IR data was obtained with an IRAffinity-1 spectrometer (Shimadzu, Kyoto, Japan). High resolution mass spectrometry (HRMS) was conducted with a high-resolution LCT Premier XE mass spectrometer in positive ESI mode (Waters, MA, USA). Melting points were measured on a digital melting point apparatus and are uncorrected.