Interception of Transient Anti-Allyl–ir Species with Cu–Azomethine Ylide: Mechanistic Origins of Z-Selectivity in an Asymmetric Allylation Reaction

Hui Xu,Xinxin Niu,Yanfeng Dang
DOI: https://doi.org/10.1021/acscatal.4c03771
IF: 12.9
2024-01-01
ACS Catalysis
Abstract:Comprehensive computational studies have been applied to explore the mechanism of Z-retentive asymmetric allylation reactions via dual Ir/Cu catalysis. In contrast to the classical syn-pi-allyl-Ir(III) species that the substituent on allyl orients to the opposite direction of a phosphorus ligand, mechanistic studies reveal that the active anti-pi-allyl-Ir(III) complex involves a distinct structure, wherein the terminal/C1 of allyl points away from the chiral phosphorus ligand and the substituent on allyl lies adjacent to COD/P ligands. Then, a chiral Cu(I)-azomethine ylide undergoes facile alpha-allylation with stereopure anti-pi-allyl-Ir to linearly selectively give enantioenriched Z-olefin products. The nucleophilic attack is more kinetically favored than anti-to-syn-allyl pi-sigma-pi isomerization, thus attaining Z-selectivity. Transition state models and distortion/interaction analyses unveil that the enantioinduction stemming from the Cu(I)-ylide species can face-selectively interact with the terminal/C1 of anti-pi-allyl-Ir species using attractive dispersion effects. Contrastively, the pathway giving its enantiomer is blocked by sterically caused distortions. Because the internal/C3 site of anti-pi-allyl-Ir is shielded by the cyclometalated moiety and the COD ligand, the branched pathways are inhibited, clarifying the linear selectivity to achieve Z-retentive allylation. Owing to the reverse active site in pi-allyl-Ir, the allylation between syn-allyl-Ir and Cu(I)-ylide provides a branch-selective product bearing vicinal stereocenters. The studies demonstrate on a molecular level how ligand-casted chiral scaffolding can be subtly applied to orient analogue in stereoinduction and how the mode of regioselection is changed by different active sites in syn-/anti-pi-allyl-Ir(III) intermediates.
What problem does this paper attempt to address?