Conformational preference in difluoroacetamide oligomers: probing the potential for foldamers with C–H⋯O hydrogen bonds

The C–H bond of a difluoroacetamide group, acidified by two adjacent fluorine atoms, could in principle provide conformational organisation for foldamers based on C–H⋯O hydrogen bonds. We find that in model oligomeric systems, this weak hydrogen bond leads only to partial organisation of the secondary structure, with the conformational preference of the difluoroacetamide groups being predominantly governed by dipole stabilisation.


General Information
All reactions were performed under a nitrogen atmosphere. Chemicals were purchased from Sigma Aldrich (Merck) and used without any further purification. Difluoroacetic anhydride was purchased from Manchester Organics. Deuterated solvents were purchased from Cambridge Isotope Laboratories or Sigma Aldrich (Merck). Anhydrous solvents were dispensed under nitrogen from a solvent purification system (Innovative Technologies PureSolve PS-MP-5) or were purchased from commercial suppliers.
High resolution spectrometry experiments (HRMS) were recorded on an Orbitrap Elite Thermo Scientific for electrospray ionisation experiments.

NMR Spectrometer Data
NMR experiments were performed on a Bruker Avance III HD 500 Cryo equipped with 5 mm DCH 13C-1H/D Cryo Probe (500 MHz). The temperature was controlled by Bruker BCU II unit. Additional 19 F experiments were performed on a Bruker Neo 600 MHz equipped with 5 mm TXO cryoprobe (cryo-enhanced for 19 F) at 298 K; and Bruker Avance 400. Solvent residual peak (7.26 ppm for 1 H and 77.0 ppm for 13 C in CDCl3; 2.05 for 1 H and 29.84 ppm for 13 C in acetone-d6) were used as references for 1 H and 13 C chemical shifts. 19 F NMR spectra were referenced to internal standard hexafluorobenzene (-161.64 ppm in CDCl3; -164.67 in acetone-d6) according to Togni et al. 1 The spectra were plotted in Mestrenova 14.2.1 or TopSpin 4.1.4 using standard processing methods.

NMR Pulse Sequences
Standard Bruker pulse sequences with the following parameters were used throughout the work.

N,N'-(Ethane-1,2-diyl)bis(N-benzyl-2,2-difluoroacetamide) (3)
To a solution of N,N'-dibenzylethylenediamine (120 mg, 0.5 mmol) and anhydrous pyridine (300 μL) in DCM (2 mL) at 0 °C was added difluoroacetic anhydride (350 μL, 2 mmol, 4 eq.) dropwise. A precipitate formed which dissolved upon complete addition of the reagent. The mixture was stirred at rt overnight (crude mixture 1 H NMR showed pyridinium NH proton at 15.34 ppm in CDCl3); then diluted with DCM and quenched with sat. aq. NaHCO3. The mixture was extracted with DCM (3 x 15 mL), washed with sat. aq. NaCl (15 mL) and sat. aq. KHSO4 (15 mL). The organic phase was separated and dried over Na2SO4. The solvents were evaporated under reduced pressure to give bis(amide) 3 as a colourless oil in quantitative yield as a mixture of rotamers which solidified to a white solid upon standing at rt.

Solvent Effects
Compound 3 was investigated by NMR in a range of solvents. Higher population of E,Z-3a was observed in solvents with higher polarity (acetone or methanol). Dichloromethane has higher polarity than chloroform (ε = 8.93 vs. 4.71, respectively). The populations remained constant in CDCl3 over the range of 278 -318 K.

NMR Exchange Studies of Rotational Barriers
Rotational barriers of difluoroacetamide group of 4 in different solvents were estimated by a series of 1D EXSY spectra with different mixing time (100 -500 ms) using initial rate approximation.   Exchange peak relative intensity vs. mixing time Figure 37. Plot of relative exchange peak intensities from 1 H 1D NOESY NMR spectra (500 MHz, CDCl3, mixing time 100 -500 ms) of 4 in different solvents at 298 K. Initial rate approximation was used to derive exchange rates as the slope of the line.

Urea Donor/Acceptor System
When a stronger H-bond donor such as arylurea 6 is present, it completely forces the preference of the amide to an H-bond acceptor (6a) in CDCl3. However, a weak exchange crosspeak is visible in 2D NOESY spectrum where the urea is an H-bond acceptor (6b). The urea NH chemical shift changes from δH 7.83 (intramolecular H-bond) to 6.23 ppm (no intramolecular N-H-O bond) in CDCl3. The computed chemical shift of the C-H does not change upon acting as the H-bond donor. Interestingly, the NBO analysis suggested that the orbital interactions of the H-bond are quite comparable (2.7 for urea donor vs. 0.9 kcal/mol for C-H donor) but certainly other factors including entropy influence the H-bond properties. However, in acetone-d6, a conformer was detected, which, based on a HOE crosspeak, could be assigned as 6b. Again, the change in solvent polarity can result in minor conformation to be populated (δNH change from 8.25 to 7.78 ppm, where urea binds acetone).

Crystallography
X-ray diffraction experiments on 3 were carried out at 100(2) K on a Bruker D8 Venture using Mo-Kα radiation (λ = 0.71073 Å) radiation. Intensities were integrated in SAINT 1 and absorption corrections based on equivalent reflections were applied using SADABS. 2 The structure was solved using ShelXT 3 all refined by full matrix least squares against F 2 in ShelXL 4,5 using Olex2 6 . All of the non-hydrogen atoms were refined anisotropically. While all of the hydrogen atoms were located geometrically and refined using a riding model. Crystal structure and refinement data are given in Table 1

Geometry Optimization
Density functional theory (DFT) calculations were performed with ORCA 5.0.2. 7 The geometries were optimized using the hybrid B3LYP functional 8,9 with D3(BJ) dispersion correction 10 and the triple-zeta plus polarization def2-TZVP basis set. 11 All DFT calculations were conducted with the Defgrid2 integration grid, TightSCF convergence criteria, and Rijcosx approximation. Frequency calculations were performed at the same level of theory as for geometry optimizations to verify the stationary points as minima (no imaginary frequencies); or transition states (one imaginary frequency) as well as to obtain thermal Gibbs free energy corrections at 298 K. Grimme's entropy corrections using quasi-rigid rotor harmonic approximation (qRRHO) 12 were applied to all frequencies below 100 cm -1 .

Single-Point Calculations
To refine the computed energy, single point calculations were performed in ORCA 5.0.2 at B3LYP-D3(BJ)/def2-TZVP/SMD(CH2Cl2) level of theory. SMD implicit solvation model for chloroform was used. 14

Gibbs Free Energies
The ΔG value was obtained by adding the corresponding free energy corrections at 298 K and SMD(chloroform) solvation correction (sum of CPCM dielectric and GCDS contributions) calculated at the B3LYP-D3(BJ)/def2-TZVP level, to ΔE, calculated at the single-point calculation at DLPNO-CCSD(T1) level of theory.

NMR Calculations
NMR scalar coupling constants were computed in ORCA 5.0.2 at DSD-PBEP86 17,18 /pcSseg-3 19 level of theory using AutoAux option, VeryTightSCF settings, and SMD solvation model for chloroform. The rotational barrier between structures 5_SM and 5_TS1 is 73.25 kJ·mol -1 . The methyl groups are puckered towards the carbonyl oxygen in the transition state. The structure 5_TS2 has a different Npyramidalization with methyl groups puckered towards the fluorine atoms and is higher in energy. Nearly identical reaction barrier was computed from the conformer with fluorine atom synclinal to the C=O bond (72.50 kJ·mol -1 ). The structures are depicted in the chapter Coordinates of Optimized Structures. The X-ray structure (conformer 3b') is not the most populated structure in solution and accounts for ~5 % of the population. The X-ray structure contains antiperiplanar N-CH2CH2-N, while the more populated conformers of 3b in solution have gauche arrangement.