Pd-catalyzed stereoselective tandem ring-opening amination/cyclization of vinyl γ-lactones: access to caprolactam diversity

A stereoselective amination/cyclization cascade process has been developed that allows for the preparation of a series of unsaturated and substituted caprolactam derivatives in good yields. This conceptually novel protocol takes advantage of the easy access and modular character of vinyl γ-lactones that can be prepared from simple precursors. Activation of the lactone substrate in the presence of a suitable Pd precursor and newly developed phosphoramidite ligand offers a stereocontrolled ring-opening/allylic amination manifold under ambient conditions. The intermediate (E)-configured ε-amino acid can be cyclized using a suitable dehydrating agent in an efficient one-pot, two-step sequence. This overall highly chemo-, stereo- and regio-selective transformation streamlines the production of a wide variety of modifiable and valuable caprolactam building blocks in an operationally attractive way.


S3. Procedure for the preparation of vinyl γ-lactones
General Procedure A: Under a N 2 atmosphere, to a separate flame-dried round-bottom flask equipped with a stirring bar was added the respective γ-oxobutanoic acid (10.0 mmol) and anhydrous THF (30 mL). The solution was cooled down to 0 ºC (ice/water), followed by dropwise addition of vinyl magnesium bromide in THF (1.0 M, 30.0 mL, 30.0 mmol). The reaction mixture was allowed to warm to room temperature and stirred for 16 h. The reaction mixture was quenched with saturated aqueous NH 4 Cl, then treated with HCl (4 M) until the pH was 3. The organic components were extracted with EtOAc (3 × 20 mL). Hereafter, the combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 , filtered, and then concentrated under reduced pressure. The residue was purified by flash chromatography to afford the corresponding lactone. According to a previously reported procedure, 3 to a solution of γ-oxobutanoic acid (10.0 mmol, 1.0 equiv) and tert-butyl alcohol (20.0 mmol, 2.0 equiv) in DCM (20 mL), DMAP (3.0 mmol, 0.3 equiv) was added. The resultant solution was cooled down to 0 °C and N,N'-dicyclohexylcarbodiimide (DCC, 12.0 mmol, 1.2 equiv) was added. The reaction mixture was stirred at room temperature for 12 h. The urea byproduct was filtered off and the organic layer was concentrated under vacuum. The crude residue was purified by a rapid flash chromatographic purification to give the corresponding ester product.
Under a N 2 atmosphere, to a separate flame-dried round-bottom flask equipped with a stirring bar was added the respective γ-oxobutanoic ester (5.0 mmol) and anhydrous THF (15 mL). The solution was cooled down to 0 ºC (ice/water), followed by dropwise addition of vinyl magnesium bromide in THF (1.0 M, 7.5 mL, 7.5 mmol). The reaction mixture was allowed to warm to room temperature and stirred for 16 h. The reaction mixture was quenched with saturated aqueous NH 4 Cl. The organic components were extracted with EtOAc (3 × 20 mL), and the combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 , filtered, and then concentrated under reduced pressure. The residue was purified by flash chromatography to afford the corresponding lactone.
The non-reported ligands L8-L12, L16-L17, L19 were prepared according to a reported procedure with slight modifications. 2d DCM, 0 ºC-rt To an oven-dried round bottom flask, distilled PCl 3 (180.0 µL, 2.06 mmol) was added to a solution of anhydrous Et 3 N (1.68 mL, 12.06 mmol) in DCM (15 mL) at 0 o C and the mixture stirred for 0.5 h at this temperature. Then, the respective amine (2.0 mmol) was added dropwise at 0 o C, and the reaction mixture stirred for 4 h at room temperature. The resultant solution was cooled to 0 o C, and then [1,1-biphenyl]-2,2-diol (372.4 mg, 2.0 mmol) or (±)-1,1′-binaphthalene-2,2′-diol (572.7 mg, 2.0 mmol) was added. The mixture was stirred for 16 h at room temperature, diluted with water and extracted with DCM (3 × 20 mL). The combined organic layers were dried over anhydrous Na 2 SO 4 , filtered, and then concentrated under reduced pressure. The residue was purified by flash chromatography to afford the respective phosphoramidite ligand. All purified ligands were fully characterized by NMR ( 1 H, 13

AcOH, DCM
To a stirred solution of the cyclopentanone (420.6 mg, 5.0 mmol) and cyclopentylamine (425.8 mg, 5.0 mmol) in DCM (15 mL) were added sodium triacetoxy borohydride (1.4836 g, 7.0 mmol) and acetic acid (300.0 mg, 5.0 mmol). The reaction mixture was stirred for 12 h at room temperature, and hereafter, 1 N aqueous NaOH was added. The resultant mixture was extracted with ester (3 × 20 mL). The combined organic layers were dried over anhydrous Na 2 SO 4 , filtered, and then concentrated under reduced pressure. The crude product could be directly used without further purification.

N, DCE
To a stirred solution of cycloheptanone (1.0 g, 8.9 mmol) and NH 4 OAc (6.63 g, 86.0 mmol) in DCE (30 mL) were added sodium triacetoxy borohydride (2.65 g, 12.5 mmol) and Et 3 N (2.5 mL, 17.9 mmol). The reaction mixture was stirred for 48 h at room temperature, followed by addition of saturated aqueous NaHCO 3 . The resultant mixture was extracted with EtOAc (3 × 20 mL). The combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 , filtered, and then concentrated under reduced pressure. The residue was purified by flash chromatography to afford the desired amine as a yellow liquid.

S8
S8. Entry [a] Ligand Solvent E:Z-1 [c] 1/2 [c] Yield of 3 [%] [ Vinyl γ-lactone A (37.6 mg, 0.20 mmol, 1.0 equiv) was combined with Pd 2 (dba) 3 ·CHCl 3 , the ligand, aniline and the solvent at room temperature under air. The internal standard CH 2 Br 2 (1.0 equiv.) was added after the reaction mixture had been stirred at room temperature for 12 h, and then an aliquot of the mixture was taken for analysis allowing to determine the NMR yield of the amino acid intermediate, the Z/E ratio and the ratio 1/2 using signal integration. Hereafter, EDC (57.5 mg, 3.0 mmol, 1.5 equiv.) was added and the reaction mixture stirred for another 1 h, after which the yield of the targeted product 3 was determined by 1 H NMR spectroscopy. S10 S10. Typical procedure for the preparation of caprolactams

S211. Control reactions
The nucleophilic attack at the less hindered, terminal carbon is typically favored in Pdcatalyzed allylic substitution reactions. However, this preference can be modulated by the sterics and electronics of the π-allylmetals. For electronically related π-acceptor ligands (see: B. L. Feringa, J. F. Teichert, Angew. Chem. Int. Ed. 2010, 49, 2486) such as phosphoramidites and phosphites can increase the cationic character of the allyl unit, with an increased cationic character being more stable at the internal carbon center of the allyl group and thus directs (electronically) the nucleophilic addition to that position (see also: B. M. Trost, M. R. Machacek, A. Aponick, Acc. Chem. Res. 2006, 39, 747). According to our previous research based on Pd/phosphoramidite mediated regioselective allylic substitution (see refs. 14a and 17 of the main text), it showed significant potential for "branched amination" and thus providing a possible regio-selectivity issue when using phosphoramidite ligands (such as L8) in the formation of caprolactams in our tandem process, and potentially lowering the yield of the target caprolactam. Indeed, when we used vinyl cyclic carbonate as an allylic surrogate instead of a -vinyl lactone under the standard conditions, a mixture of regio-isomers were detected with a ratio of branched/(Z)-linear of 15:85 as determined by 1 H NMR integration. Note: branched allylic amine product (double doublet at 6.45 ppm), linear allylic amine product with a triplet at 6.18 ppm.

S217
The preparation of a vinyl -lactone substrate failed by using the standard method used to get access to the -lactones. After some optimization, it could be prepared though a two-step method from 5-benzoyl pentanoic acid: O Ph

Ph
The allylic amination step could be realized using modified conditions (55% NMR yield, E/Z = 94:6), while low yield (< 10%) of nine-membered lactam was detected from the crude reaction mixture by 1 H NMR after attempted cyclization.
Procedure: the vinyl -lactone (43.3 mg, 0.20 mmol, 1.0 equiv) was combined with Pd 2 (dba) 3 ·CHCl 3 (4.0 mg, 2.0 mol%), L8 (6.4 mg, 8.0 mol%) and aniline (28.5 mg, 0.30 S218 mmol) in DCM (0.30 mL) at rt in air. The reaction mixture was stirred at rt for 12 h, then an aliquot of the mixture was taken for NMR analysis, which showed the unsaturated amino acid intermediate (55% NMR yield) to have an E/Z ratio of 96:4. Hereafter, the mixture was diluted to 0.01 M and EDC (57.5 mg, 3.0 mmol) was added, and the reaction mixture stirred for another 1 h. Hereafter, only a trace amount of target product could be detected by 1 H NMR of the crude mixture.
The lower yield of the nine-membered lactam can be explained by the increasing entropic cost to pre-organize the substrate for intramolecular cyclization versus intermolecular "amide" bond formation.

S219
Analytical data for the seven-membered vinyl lactone:

S219. Characterization of byproduct 2 from bis-allylation of aniline
The isolation of completely pure byproduct 2 was unsuccessful. The characterization by 1 H NMR, IR and MS data of the purest sample of 2 is here provided to support its proposed structure.