H-bonded reusable template assisted para-selective ketonisation using soft electrophilic vinyl ethers

In nature, enzymatic pathways generate Caryl−C(O) bonds in a site-selective fashion. Synthetically, Caryl−C(O) bonds are synthesised in organometallic reactions using prefunctionalized substrate materials. Electrophilic routes are largely limited to electron-rich systems, non-polar medium, and multiple product formations with a limited scope of general application. Herein we disclose a directed para-selective ketonisation technique of arenes, overriding electronic bias and structural congestion, in the presence of a polar protic solvent. The concept of hard–soft interaction along with in situ activation techniques is utilised to suppress the competitive routes. Mechanistic pathways are investigated both experimentally and computationally to establish the hypothesis. Synthetic utility of the protocol is highlighted in formal synthesis of drugs, drug cores, and bioactive molecules.


Optimizations: Supplementary
Key Observations: (a) N-Ac-Gly gives better yield and CBZ-Gly offers better selectivity. (b) Changing reaction time from 24 h to 36 h give incremental increase in yield.
(c) Addition of NaOAc gives slight increase in yield.

Supplementary Discussions: General Procedure A for para-Ketonisation:
In an oven-dried screw capped reaction tube was charged with magnetic stir-bar, benzylsilyl ether substrate (viscous benzylsilyl ether was weighed first), Pd(OAc)2 (10 mol%), ligand (N-CBZ-Gly or N-Ac-Gly; 20 mol%), Ag2CO3 (3 eq.) and NaOAc (2 eq.). 1.2 mL (for 0.1 mmol scale) of 1,1,1,3,3, was added followed by vinyl ether (3 eq.). The reaction tube was capped and stirred (900 rpm) on a preheated oil-bath at 80 °C for 24/36 h. Upon completion the mixture was cooled and diluted with EtOAc and filtered through a celite pad. The filtrate was evaporation under reduced pressure and the crude mixture was purified by column chromatography using silica (100-200 mesh size) and petroleum ether/ ethyl acetate as the eluent. The selectivity was monitored using 1 H-NMR signal in presence of 1,3,5-trimethoxybenzene as internal standard. The regio-selectivity was determined from 1 H-NMR signals of aromatic region and benzylic position.
[Note: The reaction is sensitive to the quality of HFIP. Freshly distilled solvent is recommended.]

Synthesis of Vinyl Ethers:
Procedure: In a three-neck round bottom flask (500 mL) dry toluene (60 mL) was taken under inert atmosphere and diethoxymethane or ethylal (20 mL) was added to it. Freshly distilled acetyl chloride (15.4 mL) was added to the mixture through dropping funnel over a period of 1 h. The mixture was stirred at room temperature for 6-7 h and triphenyl phosphine (60 g) was mixed and stirred for overnight. After initial clear solution, thick white solid precipitate formed. The reaction was filtered under suction and washed repeatedly with petroleum ether until a free-flowing solid forms. The solid was transferred to a round bottomed flask and suspended in toluene and warmed to 60 ºC for 3-4 h under reduced pressure (to remove moisture in the form of azeotrope). The wittig salt was then preserved in a air-tight container under nitrogen atmosphere.

Effect of D2-HFIP solvent:
Procedure: In a clean oven dried reaction tube all the reaction components were added and capped. The tube was purged with dried air using Schlenk line set up and dried d2-HFIP (by passing through activated neutral alumina) was added. The mixture was then stirred on a preheated oil-bath for 24 h. Upon completion the reaction composition was monitored by TLC and NMR.

Vinyl Ether -HFIP interaction:
Procedure: In a clean NMR tube ethyl vinyl ether (0.1 mmol) was taken along with 0.6 mL of CDCl3. The signal was recorded for the reference. In the following steps HFIP was added to the tube. The amount of HFIP added was increased from 0.5 eq. to 10 eq. and the signals were recorded following each addition. After the addition of 4 eq. of HFIP the changes in the NMR signal were prominent which further intensified upon heating. Finally, vinyl ether got consumed completely with the formation of multiple products. The products were also detected in GCMS.

Procedure:
The kinetic isotope effect was measured both from product distribution and rate constant ratio.

Product Distribution Study (PH/PD):
In a clean oven dried reaction tube 1:1 ratio mixture of the model substrate and deuterated variant (0.05 mmol each) was taken. All the remaining reagents were added according to the standard condition of a 0.1 mmol scale reaction. Upon completion of the reaction the product was isolated the product distribution was monitored by NMR. In this regard the signals corresponding to the −Me of acetyl group and the benzylic −CH2 were compared.

Computational Details and Discussions:
All calculations were performed with Gaussian 09. 2 The B3LYP 3,4 functional and a mixed basis set of SDD for Pd and 6-31G(d) for other atoms were used in geometry optimizations. Single-point energies were calculated using M06 5 and a mixed basis set of SDD for Pd and 6-311+G(d,p) for other atoms. Solvation energy corrections were calculated using the SMD 6 implicit solvation model. HFIP was used as the solvent. Since the solvent parameters for HFIP are not available in Gaussian 09, the solvent parameters of isopropanol were used and the dielectric constant of the solvent was modified to the dielectric constant of HFIP (ε = 16.7) by using the "scrf=(smd, solvent=2-propanol, read)" keywords in Gaussian 09. The 3D structures were generated using CYLView. 7 In the main manuscript (Figure 5a), only part of the catalytic cycle of the reaction of 1 with vinyl methyl ether was shown. The complete energy profile of the Pd-catalyzed parafunctionalization of 1 is shown below. As revealed in previous computational studies of Pd(II)catalyzed C−H activation with mono-N-protected amino acid (MPAA) ligands, 8 the para-C-H activation (s-TS1) occurred via the CMD mechanism with the N-acyl group on the MPAA ligand serving as a base. The energetics of the C−H activation pathway of 1 and the origin of para-selectivity was found to follow the same trend as observed during para-selective silylation reaction. 1 Following the C-H activation, intermediate s4 tautomerizes and rearranges to the more stable palladacycle 5 with the carboxylic group bound to the Pd center in an κ 2 fashion. The coordination of vinyl methyl ether to the Pd center in 5 leads to a slightly more stable intermediate 6, which undergoes migratory insertion via TS1 followed by β-H elimination (TS2). These two steps are both facile and require lower barriers than the C-H activation step. The resulting Pd(II) hydride 8 then undergoes reductive elimination to form the alkenyl ether product. Finally, oxidation of Pd(0) regenerates the Pd(II) active catalyst. The final reductive elimination and oxidation steps are expected to be facile and thus are not studied computationally. The C-H activation is irreversible and represents the rate-and selectivitydetermining step in the overall catalytic cycle.