Fig. 2: Data-driven optimization of the chiral ligand. | Nature Catalysis

Fig. 2: Data-driven optimization of the chiral ligand.

From: Enantioselective C–H annulations enabled by either nickel- or cobalt-electrocatalysed C–H activation for catalyst-controlled chemodivergence

Fig. 2: Data-driven optimization of the chiral ligand.

a, Initial experimental ligand screening. Reaction conditions: 1 (0.2 mmol), 2 (0.3 mmol), Ni(OAc)2·4H2O (0.04 mmol), ligand (0.06 mmol), BmimPF6 (0.05 M), 1,4-dioxane/DMA (4:1, v/v, 5 ml). Undivided cell. GF, graphite felt anode; Pt, platinum plate cathode. Stirring rate, 600 rpm. Constant current electrolysis (CCE) at 2.0 mA. 100 °C, 18 h reaction time. Q, 8-quinolinyl; Bmim, 1-butyl-3-methylimidazolium; cat., catalytic; n.d., not detected. b, The general workflow for data-driven ligand optimization with the final top three feature space. ML, machine learning; Exp, experimental. c, Respective visualizations of the NCIs (π–π and CH–π) in TS1 with L7 are made with the NCIPLOT program. Selected distances and angles are shown for TS1 and TS1′. Red represents strong repulsive interactions in the plotted surfaces, whereas green and blue represent weak and strong attractive interactions, respectively. Energies are with reference to the Int0L7 in the ligand comparison plot. ΔΔG, Gibbs free energy difference between the major and minor product isomers; ΔG‡, Gibbs free energy difference between reactant (Int0L7) and transition state.

Back to article page