Combined experimental and theoretical study of long-range H–F interactions in α-fluoro amides

Abstract

A combined experimental and computational approach provided insight into the nature and conformational dependence of long-range ⁴J_HF couplings in α-fluoro amides. The dependence of ⁴J_HF on substituents and the solvent was investigated. H–F coupling constants determined by NMR spectroscopy are in agreement with DFT calculations. NBO analysis revealed that a favourable n_F→σ_NH* interaction correlates with the magnitude of ⁴J_HF.

---

The introduction of fluorine into organic compounds is a valuable strategy to modify their physico-chemical properties and improve, e.g., the pharmacokinetic profile of drugs.^1–5 One example are α-fluorinated amides that have been used to tune ligand–protein interactions^6–9 and modulate the activity of small molecules with, for example, anticonvulsant,¹⁰ antibacterial,¹¹ and cholesterol-lowering properties.¹² α-Fluorination has also enabled control over the backbone conformation of β-amino acids and β-peptides.^13–19 Previous studies showed that α-fluoropropionamides adopt mainly an antiperiplanar (ap) conformation, and to a lesser degree a gauche conformation (Fig. 1).^20–22 The driving force for the preferred ap conformation is the opposing dipoles of the vicinal C–F and C=O bonds.‡^22–24 Jaun and Seebach observed in α-fluoro β-peptides a conformation dependent NH–F (⁴J_HF) coupling between fluorine and the adjacent amide proton.^17–19 This long-range coupling is reminiscent of hydrogen bond-mediated fluorine couplings,§^25,26 as e.g. in ortho fluorinated anilides or benzamides.^27,28 Its presence suggests that the ap conformation is not only driven by electrostatic dipole–dipole interactions but might be supported by an additional weak interaction between fluorine and the amide group.¶^29–31

Fig. 1 Equilibrium between antiperiplanar (ap) and gauche conformers of α-fluoro amides.

Herein, we probed the nature of the long-range ⁴J_HF coupling constant by studying a series of α-fluoro amides bearing different substituents with computational and experimental tools. We investigated the α-fluoro amides in different solvents by NMR spectroscopy and in the solid state by X-ray crystal structure analysis. We show that the value of ⁴J_HF can be accurately predicted by DFT calculations and that a weak through-space stereoelectronic n→σ* interaction between the F and N–H moieties correlates with the magnitude of the ⁴J_HF coupling constant.

We started by synthesising α-fluoro amides 1–5 as model compounds for our experimental studies (Scheme 1). These propionamide and malonamide derivatives were readily accessed from the appropriate α-fluoro thioester by reaction with methyl amine, benzyl amine, and aniline. Notably, the thioester allowed even for effective amidation with the poor nucleophile aniline.³²

Scheme 1 Synthesis of α fluoro amides 1–5.

α-Fluoro amide 1 with methyl substituents at C(α) and at NH served as reference for our studies. Methyl ester derivative 4 was designed to probe the effect of an electron withdrawing group (EWG) at C(α). We envisioned that an EWG would lower the electron density at the fluorine atom and thereby weaken the putative NH–F interaction. Such a weakened interaction would be reflected in a smaller long-range ⁴J_HF coupling. Conversely, the phenyl moieties of anilides 3 and 5 were expected to increase the acidity of the N–H group³³ and thus strengthen the F⋯H–N interaction. Amide 2 with a benzyl group at the NH was used to explore the effect of an aromatic moiety that is not directly attached to the amide moiety.

Before starting the spectroscopic analysis, we performed DFT calculations with representative α-fluoro amides at the B3LYP/AUG-cc-pVTZ level of theory using the program Gaussian. For α-fluoro amide 1 restrained geometry optimizations (in steps of 15°) in vacuo were performed for a series of F–C–C=O dihedral angles (Θ). The ap conformer was found to be energetically favoured by 5 kcal mol⁻¹ over the gauche (Θ = −60°) conformer (Fig. 2, black curve). This is ∼1 kcal mol⁻¹ less than reported previously by Banks et al. and Tormena et al. who used the same functional but smaller basis sets.^21,22 In agreement with these reports, we also found that in vacuo the gauche conformer is not a local minimum on the potential energy surface. In contrast, the gauche conformer becomes a local minimum when the calculations were carried out with implicit solvation.²¹ Unrestrained geometry optimization converged to a local minimum of Θ = −55.3° with the polarizable continuum model (PCM) parametrized for CHCl₃ and Θ = −46.3° with PCM parametrized for DMSO. Frequency calculations in implicit solvent for both ap and gauche conformations of 1 led to thermally corrected free energy differences (ΔG_ap-g) of 2.8 kcal mol⁻¹ in CHCl₃ and 1.8 kcal mol⁻¹ in DMSO. This computational finding suggests that the ap conformer is favoured over the gauche conformer regardless of the solvent.

Fig. 2 DFT calculated ⁴J_HF coupling constant and relative B3LYP/AUG-cc-pVTZ energies as functions of the dihedral angle F–C–C=O (Θ) in 1.

Next, we computed the dependence of the ⁴J_HF coupling constant on the dihedral angle Θ for α-fluoro amide 1 (Fig. 2, blue curve).³⁴ The calculated ⁴J_HF coupling curve shows a sharp minimum (strongest coupling) at the ap conformation where NH and F are closest in space (⁴J_HF = −4.3 Hz, Θ = 180°). For the gauche conformation the coupling strength is significantly attenuated (⁴J_HF = −0.8 Hz, Θ = 300°). The calculated dependence of ⁴J_HF on Θ is in agreement with experimental observations by Jaun and Seebach for 2-fluoro-β³-alanine in peptides.¹⁷ For Θ close to 180° the absolute value of ⁴J_HF was reported to be 3.7 Hz whereas for Θ close to 90° no resolved coupling could be detected (|⁴J_HF| < 0.5).

We also performed unrestrained geometry optimization in vacuo of the ap conformations for 1, 3, and the N-methyl analogue of 4 using the same basis set and functional. The resulting minimum energy conformations were then used to predict the values of ⁴J_HF. These calculated ⁴J_HF coupling constants of 1, 3, and 4 differ significantly from each other. The absolute value for anilide 3 is greater (⁴J_HF = −7.0 Hz) and that of 4 with an electron withdrawing ester group next to the C–F bond is smaller (⁴J_HF = −3.3 Hz) compared to the one of reference compound 1 (⁴J_HF = −4.3 Hz, Table 1, entry 1).

Table 1 Experimental and calculated values of ⁴J_HF (Hz) for 1–5^a

		^4 J HF [Hz]
Entry	Solvent	1	2	3	4	5
a Determined at 50 mM. b The calculations were performed on the N-methyl analogue of 4.
1	Calc. (ap)	−4.3	n.d.	−7.0	−3.3^b	n.d.
2	CDCl3	−4.4	−4.6	−6.8	−3.3	−4.5
3	CD2Cl2	−4.3	−4.3	−6.5	−2.7	−4.2
4	d6-Acetone	−2.9	−2.8	−3.9	−1.5	−1.8
5	CD3OH	−2.4	>−2	n.d.	>−2	n.d.
6	d6-DMSO	−1.7	−1.7	−1.8	−0.7	>−1

---

Next, we determined absolute values of the ⁴J_HF coupling constants experimentally by recording ¹H and ¹⁹F NMR spectra and started by using CDCl₃ as a solvent. As suggested by the calculations, the sign of ⁴J_HF was found to be negative by analysis of the cross-peak fine-structure in a constant-time ¹³C-HMBC spectrum.|| Reassuringly, the experimental ⁴J_HF values in CDCl₃ do not deviate by more than 0.2 Hz from the predicted values (Table 1, entries 1 and 2). Such a good agreement is remarkable, in particular in light of the well-known difficulties of DFT calculations to predict coupling constants involving fluorine.^35,36

Since amide moieties can form intermolecular H-bonds that could disturb the conformational properties of α-fluoro amides, we recorded NMR spectra of 1–5 at different concentrations in CD₂Cl₂ (0.8–200 mM). These studies showed that the experimental values of ⁴J_HF hardly change upon increasing the concentration.|| In contrast, the NH chemical shift has a significant concentration dependence, which suggests that intermolecular interactions between the amides occur (K_d ∼ 0.1–1 M).|| This finding shows that possible intermolecular interactions between the amides do not affect the ⁴J_HF coupling constant to a significant extent.

Next, we explored the effect of substituents at the α-fluoro amide on the value of the ⁴J_HF coupling. The NMR spectra of anilides 3 and 5 in CDCl₃ revealed stronger ⁴J_HF couplings (−6.8 Hz and −4.5 Hz) compared to those of the analogous benzyl amides 2 and 4 (−4.6 Hz and −3.3 Hz). This observation is in agreement with the expected higher N–H acidity of 3 and 5 in comparison to 2 and 4, respectively, and indicates a stronger F⋯H–N interaction. Conversely, the absolute ⁴J_HF values are lower in case of 4 and 5 that bear ester moieties compared to 2 and 3 with methyl groups at Cα. This indicates that an electron withdrawing group (ester) at Cα lowers the electron density at fluorine and weakens the F⋯H–N interaction. Thus, in accordance with our expectations, changes in molecular structure designed to modulate the strength of the putative F⋯H–N hydrogen bond are mirrored in the measured values of the ⁴J_HF coupling constant.

We then investigated the influence of the solvent on the value of the ⁴J_HF coupling constant by recording ¹H and ¹⁹F NMR spectra of 1–5 in CD₂Cl₂, d₆-acetone, CD₃OH, and d₆-DMSO. In the more polar solvents the strength of the ⁴J_HF coupling in 1–5 decreases (Table 1, entries 4–6). Considering the significant differences in the calculated ⁴J_HF coupling constants of gauche and ap conformations (Fig. 2), these data show that the ap conformer becomes less populated in polar solvents, which is expected based on the higher dipole moment of the gauche compared to the ap conformation.

Compounds 2, 3, and 4 crystallized and allowed for their conformational analysis by X-ray crystallography (Fig. 3).²² α-Fluoro amides 2 and 3 are in ap conformations in the crystalline state. Their interatomic NH–F distances of approximately 2.2 Å are significantly shorter than the sum of the van der Waals radii (2.5–2.7 Å). The F⋯H–N angles are 102° and 105° for 2 and 3, respectively. Both, the inter-atom distances and the bond angles, match the criteria for H–F hydrogen bonds as defined by Dunitz and Taylor (F–H distance < 2.3 Å, F⋯H–X angle > 90°).³⁷ In contrast, the α-fluoro amido moiety of 4 adopts a near gauche conformation in the solid state (Θ = 42°). The crystal packing of 2, 3, and 4 reveals that the amide groups are involved in intermolecular chain-type N–H–O interactions. In 4, these intermolecular hydrogen bonds in the solid state override the favourable ap orientation of the F–C–C=O moiety observed in solution.³⁸

Fig. 3 X-ray crystal structures of 2–4.

Finally, we investigated the origin of the observed ⁴J_HF long-range coupling constant by NBO calculations.³⁹ Second order perturbation analysis revealed the presence of a favourable electron donation from the fluorine lone pair (n_F) to the antibonding N–H orbital (σ_NH*) (Fig. 4). This finding suggests that this weak through-space interaction gives rise to the observed ⁴J_HF coupling. The calculations predict a contribution of this n_F→σ_NH* interaction of 0.8 kcal mol⁻¹ for 1, 1.3 kcal mol⁻¹ for 3, and 0.6 kcal mol⁻¹ for 4 to the stability of the ap conformer of these α-fluoro amides.** Thus, the strength of this interaction increases progressively from the N-benzyl amides 4 and 1 to the N-phenyl amide 3, in agreement with the observed increase of the ⁴J_HF coupling constants.

Fig. 4 Graphical representation of the n_F–σ_NH* interaction in 1. (iso value = 0.02 e Å⁻³).††

In conclusion, our results show that the ⁴J_HF coupling in α-fluoro amides depends on the substituents and the solvent. Calculated ⁴J_HF couplings for the ap conformations of a series of α-fluoro amides match the experimental values. The ⁴J_HF coupling is strongest in the ap conformation when F and HN are closest in space. In this conformation the strengths of the observed ⁴J_HF couplings and predicted n_F→σ_NH* interactions correlate. This suggests that the ⁴J_HF coupling is a genuine through-space coupling concomitant with a F⋯H–N hydrogen bond.

We are grateful to the Swiss National Science Foundation (grant 200020_169423) and the Laboratory of Organic Chemistry of ETH Zurich for financial support. We thank Rainer Frankenstein and Michael Solar for recording NMR spectra and X-ray crystal structures. This publication is dedicated to Prof. Bernhard Jaun on the occasion of his 70th birthday.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. E. P. Gillis, K. J. Eastman, M. D. Hill, D. J. Donnelly and N. A. Meanwell, J. Med. Chem., 2015, 58, 8315–8359.
2. S. Purser, P. R. Moore, S. Swallow and V. Gouverneur, Chem. Soc. Rev., 2008, 37, 320–330.
3. K. Muller, C. Faeh and F. Diederich, Science, 2007, 317, 1881–1886.
4. A. A. Berger, J. S. Voller, N. Budisa and B. Koksch, Acc. Chem. Res., 2017, 50, 2093–2103.
5. C. Thiehoff, Y. P. Rey and R. Gilmour, Isr. J. Chem., 2017, 57, 92–100.
6. L. Unione, M. Alcala, B. Echeverria, S. Serna, A. Arda, A. Franconetti, F. J. Canada, T. Diercks, N. Reichardt and J. Jimenez-Barbero, Chem. – Eur. J., 2017, 23, 3957–3965.
7. M. Ettaoussi, A. Sabaouni, M. Rami, J. A. Boutin, P. Delagrange, P. Renard, M. Spedding, D. H. Caignard, P. Berthelot and S. Yous, Eur. J. Med. Chem., 2012, 49, 310–323.
8. D. P. Zlotos, N. M. Riad, M. B. Osman, B. R. Dodda and P. A. Witt-Enderby, MedChemComm, 2015, 6, 1340–1344.
9. M. Winkler, T. Moraux, H. A. Khairy, R. H. Scott, A. M. Slawin and D. O'Hagan, ChemBioChem, 2009, 10, 823–828.
10. N. Pessah, M. Bialer, B. Wlodarczyk, R. H. Finnell and B. Yagen, J. Med. Chem., 2009, 52, 2233–2242.
11. G. O. Danelon and O. A. Mascaretti, J. Fluorine Chem., 1992, 56, 109–140.
12. J. W. Clader, D. A. Burnett, M. A. Caplen, M. S. Domalski, S. Dugar, W. Vaccaro, R. Sher, M. E. Browne, H. Zhao, R. E. Burrier, B. Salisbury and H. R. Davis, Jr., J. Med. Chem., 1996, 39, 3684–3693.
13. M. Salwiczek, E. K. Nyakatura, U. I. Gerling, S. Ye and B. Koksch, Chem. Soc. Rev., 2012, 41, 2135–2171.
14. T. L. March, M. R. Johnston, P. J. Duggan and J. Gardiner, Chem. Biodiversity, 2012, 9, 2410–2441.
15. V. Peddie, R. J. Butcher, W. T. Robinson, M. C. Wilce, D. A. Traore and A. D. Abell, Chem. – Eur. J., 2012, 18, 6655–6662.
16. E. Cosimi, O. D. Engl, J. Saadi, M. O. Ebert and H. Wennemers, Angew. Chem., Int. Ed., 2016, 55, 13127–13131.
17. B. Jaun, D. Seebach and R. I. Mathad, Helv. Chim. Acta, 2011, 94, 355–361.
18. R. I. Mathad, F. Gessier, D. Seebach and B. Jaun, Helv. Chim. Acta, 2005, 88, 266–280.
19. R. I. Mathad, B. Jaun, O. Flögel, J. Gardiner, M. Löweneck, J. D. C. Codée, P. H. Seeberger, D. Seebach, M. K. Edmonds, F. H. M. Graichen and A. D. Abell, Helv. Chim. Acta, 2007, 90, 2251–2273.
20. C. R. S. Briggs, D. O'Hagan, J. A. K. Howard and D. S. Yufit, J. Fluorine Chem., 2003, 119, 9–13.
21. C. F. Tormena, N. S. Amadeu, R. Ritter and R. J. Abraham, Perkin Trans. 2, 2002, 773–778.
22. J. W. Banks, A. S. Batsanov, J. A. K. Howard, D. O'Hagan, H. S. Rzepa and S. Martin-Santamaria, J. Chem. Soc., Perkin Trans. 2, 1999, 2409–2411.
23. D. O'Hagan, Chem. Soc. Rev., 2008, 37, 308–319.
24. D. C. Lankin, N. S. Chandrakumar, S. N. Rao, D. P. Spangler and J. P. Snyder, J. Am. Chem. Soc., 1993, 115, 3356–3357.
25. (a) P. A. Champagne, J. Desroches and J. F. Paquin, Synthesis, 2015, 306–322; (b) J. C. Hierso, Chem. Rev., 2014, 114, 48384867; (c) H. J. Schneider, Chem. Sci., 2012, 3, 1381–1394.
26. (a) G. T. Giuffredi, V. Gouverneur and B. Bernet, Angew. Chem., Int. Ed., 2013, 52, 10524–10528; (b) B. Linclau, F. Peron, E. Bogdan, N. Wells, Z. Wang, G. Compain, C. Q. Fontenelle, N. Galland, J. Y. Le Questel and J. Graton, Chem. – Eur. J., 2015, 21, 17808–17816.
27. I. Alkorta, J. Elguero, H. H. Limbach, I. G. Shenderovich and T. Winkler, Magn. Reson. Chem., 2009, 47, 585–592.
28. C. Li, S. F. Ren, J. L. Hou, H. P. Yi, S. Z. Zhu, X. K. Jiang and Z. T. Li, Angew. Chem., Int. Ed., 2005, 44, 5725–5729.
29. S. J. Grabowski, Chem. Rev., 2011, 111, 2597–2625.
30. E. Arunan, G. R. Desiraju, R. A. Klein, J. Sadlej, S. Scheiner, I. Alkorta, D. C. Clary, R. H. Crabtree, J. J. Dannenberg, P. Hobza, H. G. Kjaergaard, A. C. Legon, B. Mennucci and D. J. Nesbitt, Pure Appl. Chem., 2011, 83, 1619–1636.
31. R. W. Newberry and R. T. Raines, Nat. Chem. Biol., 2016, 12, 1084–1088.
32. T. Kanzian, T. A. Nigst, A. Maier, S. Pichl and H. Mayr, Eur. J. Org. Chem., 2009, 6379–6385.
33. F. G. Bordwell and G. Z. Ji, J. Am. Chem. Soc., 1991, 113, 8398.
34. W. Deng, J. R. Cheeseman and M. J. Frisch, J. Chem. Theory Comput., 2006, 2, 1028–1037.
35. T. Helgaker, M. Jaszuński and M. Pecul, Prog. Nucl. Magn. Reson. Spectrosc., 2008, 53, 249–268.
36. J. Vaara, Phys. Chem. Chem. Phys., 2007, 9, 5399–5418.
37. J. D. Dunitz and R. Taylor, Chem. – Eur. J., 1997, 3, 89–98.
38. C. R. Jones, P. K. Baruah, A. L. Thompson, S. Scheiner and M. D. Smith, J. Am. Chem. Soc., 2012, 134, 12064–12071.
39. E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A. Bohmann, C. M. Morales and F. Weinhold, NBO 5.9, Theoretical Chemistry Institute, University of Wisconsin, Madison WI, 2012.

---

Footnotes

† Electronic supplementary information (ESI) available: Details on the NMR spectroscopic analyses, DFT calculations, X-ray crystallography, and analytical data of 1–5. CCDC 1884973–1884976. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c8cc09987a

‡ For a similar charge-dipole effect, see ref. 24.

§ For related examples on F⋯H–O bonds, see ref. 26.

¶ Such a through space interaction is reminiscent of intraresidual hydrogen bonds within β-sheets formed by peptides, ref. 31.

|| For details, see the ESI.†

** In the gauche conformation of 1 this stabilizing contribution is absent.

†† Note, the relative signs of the n and σ* orbitals have no physical relevance.

---