Enantioseparation of planar chiral ferrocenes on cellulose-based chiral stationary phases: Benzoate versus carbamate pendant groups

In this study, the enantioseparation of 14 planar chiral ferrocenes containing halogen atoms, and methyl, iodoethynyl, phenyl, and 2-naphthyl groups, as substituents, was explored with a cellulose tris(4-methylbenzoate) (CMB)-based chiral column under multimodal elution conditions. n-Hexane/2-propanol (2-PrOH) 95:5 v/v, pure methanol (MeOH), and MeOH/water 90:10 v/v were used as mobile phases (MPs). With CMB, baseline enantioseparations were achieved for nine analytes with separation factors (α) ranging from 1.24 to 1.77, whereas only three analytes could be enantioseparated with 1.14 ≤ α ≤ 1.51 on a cellulose tris(3,5-dimethylphenylcarbamate) (CDMPC)-based column, used as a reference for comparison, under the same elution conditions. Pendant group–dependent reversal of the enantiomer elution order was observed in several cases by changing CMB to CDMPC. The impact of analyte and chiral stationary phase (CSP) structure, and MP polarity on the enantioseparation, was evaluated. The two cellulose-based CSPs featured by different pendant groups were also compared in terms of thermodynamics. For this purpose, enthalpy (ΔΔH°), entropy (ΔΔS°) and free energy (ΔΔG°) differences, isoenantioselective temperatures (T_iso), and enthalpy/entropy ratios (Q), associated with the enantioseparations, were derived from van ’t Hoff plots by using n-hexane/2-PrOH 95:5 v/v and methanol/water 90:10 v/v as MPs. With the aim to disclose the functions of the different substituents in mechanisms and noncovalent interactions underlying analyte–selector complex formation at molecular level, electrostatic potential (V) analysis and molecular dynamics simulations were used as computational techniques. On this basis, enantioseparations and related mechanisms were investigated by integrating theoretical and experimental data.