A theoretical study of the ligand-controlled palladium-catalysed regiodivergent synthesis of dibenzosilepin derivatives

Abstract

Palladium-catalysed ligand-controlled 1,n-palladium migration of silicon-tethering substrates provides a regiodivergent synthesis strategy for constructing silicon-bridged π-conjugated compounds possessing a 6,6-fused or a 5,7-fused scaffold. Density functional theory (DFT) calculations were carried out to elucidate the detailed mechanism of this 1,n-palladium migration involving syn- or anti-carbopalladation. The computational results suggest that alkyne insertion is the regioselectivity-determining step. Upon catalysis without the BINAP ligand, the 1,2-insertion of an alkyne into the Pd–aryl bond leads to the formation of 6,6-fused benzophenanthrosiline, which is more favorable than the 2,1-insertion of alkyne by 4.2 kcal mol⁻¹. The selective formation of 5,7-fused benzofluorenosilepins via the 2,1-alkyne insertion is facilitated by the BINAP ligand. The 1,2-alkyne insertion with the BINAP ligand is disfavoured due to the steric repulsion between the phenyl group of the substrate and the naphthalene group of the BINAP ligand. The 2,1-alkyne insertion with the BINAP ligand orients the ligand away from the phenyl group of the substrate, which can avoid steric repulsion.