Abstract
The conformational equilibria of selectively halogenated cyclohexanes are explored both experimentally (VT-NMR) for 1,1,4,-trifluorocyclohexane 7 and by computational analysis (M06-2X/aug-cc-pVTZ level), with the latter approach extending to a wider range of more highly fluorinated cyclohexanes. Perhaps unexpectedly, 7ax is preferred over the 7eq conformation by ΔG = 1.06 kcal mol–1, contradicting the accepted norm for substituents on cyclohexanes. The axial preference is stronger again in 1,1,3,3,4,5,5,-heptafluorocyclohexane 9 (ΔG = 2.73 kcal mol–1) as the CF2 groups further polarize the isolated CH2 hydrogens. Theoretical decomposition of electrostatic and hyperconjugative effects by natural bond orbital analysis indicated that nonclassical hydrogen bonding (NCHB) between the C-4 fluorine and the diaxial hydrogens at C-2 and C-6 in cyclohexane 7 and 9 largely accounts for the observed bias. The study extended to changing fluorine (F) for chlorine (Cl) and bromine (Br) at the pseudoanomeric position in the cyclohexanes. Although these halogens do not become involved in NCHBs, they polarize the geminal −CHX– hydrogen at the pseudoanomeric position to a greater extent than fluorine, and consequent electrostatic interactions influence conformer stabilities.
1. Introduction
The C–F bond is the most polar in organic chemistry and selective fluorination can impart an unusual polarity, particularly in aliphatic compounds.1,2 Such polarity can induce contraintuitive behavior in organic molecules such as is described in the gauche preference for 1,2-difluoroethane (gauche effect)3 and the strong axial preference in 2-fluorotetrahydropyran (anomeric effect).4 The high electronegativity of fluorine significantly reduces the ability of carbon bound fluorine to enter into classical hydrogen bonding relative to nitrogen and oxygen;5 however, polarized C–F bonds can make stabilizing contacts to hydrogens through dipolar electrostatic interactions. This is evinced most dramatically in the equilibrium conformations of the 3-fluoropyridinium ring 1 as illustrated in Figure 1.6 The conformer with the fluorine axial is 5.4 kcal mol–1 more stable than that when the fluorine is equatorial with the C–F and the H–N+ running parallel with compensating dipoles. The F···H–N+ bonding angle here is 90°, which is notably not at all optimal for a classical hydrogen bond; however, the dipolar interaction between the axial C–F and the axial N(+)-H bonds is strong, and this contributes significantly to the stabilization observed. An “electrostatic gauche effect” of a similar magnitude is also observed in 2-fluoroethylammonium 2, where the 2-gauche conformer is significantly more stable than the 2-anti conformer by 5.8 kcal mol–1.7 For reference, these effects are much stronger than that observed in the classical gauche effect, which recognizes that the 3-gauche conformer of 1,2-difluoroethane is more stable that the 3-anti conformer by up to 0.8 kcal mol–1.3,8 In that case, the origin of the gauche effect in 1,2-difluoroethane and related neutral systems, as determined principally by electrostatics or hyperconjugation, remains an open discussion as both effects can make a significant contribution to the relatively small energy differences between the gauche and anti conformers.8,9
Figure 1.
Conformational energy preferences favoring axial energy for 1 and gauche energy for 2 and 3.6,7
In a tangential study, we recently reported that the introduction of fluorines into the 4-position of methoxycyclohexane 4 to generate 5 results in switching the relative energies of the axial and equatorial conformers as illustrated in Figure 2.10 Methoxycyclohexane 4 has an equatorial preference consistent with well-known effects of cyclohexanes, whereas difluoro derivative 5 has an axial preference. We attributed this largely to electrostatic effects. The outcomes were supported by a combination of experimental (VT-NMR) and DFT calculations in the NBO framework. The preference for the axial methoxyl in 5 is assigned to nonclassical hydrogen bonding (NCHB) stabilization between the methoxyl oxygen and the electropositive diaxial hydrogens at C-2 and C-5 of the cyclohexane ring as illustrated in the inset in Figure 2. These hydrogens are polarized by the vicinal fluorines, and a stabilizing 1,3-diaxial electrostatic interaction occurs when the OMe substituent is axial. Such transannular interactions are increasingly being discussed in the context of the anomeric effect.11 These interactions come into the category of nonclassical hydrogen bonds (NCHBs), with weak C–H hydrogen bonding donors and H···OR angles of 90°, a geometry not optimal for lone pair-antibonding orbital interactions in classical hydrogen bonding.12
Figure 2.
Conformational preferences for methoxycyclohexane 4 and difluorinated analogue 5 (values are calculated in gas phase calculations). The axial preference for 5 can be rationalized by a combination of NCHB (electrostatics), inductive effects, and hyperconjugation (see inset summary).10
In this paper, these effects are further explored by replacing the methoxyl group with fluorine to consider 1,1,4-trifluorocyclohexane 7. It is generally accepted that fluorine is a poorer hydrogen bonding acceptor than the oxygen in a methoxyl (-OMe) group; thus, the transannular interaction may be expected to weaken. However, this study reveals that there is a greater axial over equatorial preference observed for 7 than is found in 5. In the preparation of this manuscript, we located an abstract of the American Chemical Society (ACS) National Meeting in 1990, from Stolow and Kao, which indicated that cyclohexane 7 had been prepared and found to have an axial (10ax:1eq) preference in solution (FCCl3) at 170 K (ΔG0 = −0.77 ± 0.02 kcal mol–1).13 We cannot identify a paper disclosing further details; however, our conclusions certainly support the observation disclosed in the ACS Abstract.
2. Results and Discussion
The study aimed to combine and compare the outcomes from both computational calculations and VT-NMR experiments to explore the conformational preference of cyclohexane 7. In order to carry out VT-NMR analysis, a sample of 1,1,4-trifluorocyclohexane 7 was required. A direct approach by treatment of 1,1-difluorocyclohexanol with diethylaminosulfur trifluoride (DAST) failed giving only a complex mixture including elimination products; however, success was achieved by a highly efficient decarboxylative fluorination of carboxylic acid 6 following a photocatalytic protocol of MacMillan et al.14 as illustrated in Scheme 1.
Scheme 1. Synthesis of 7via Decarboxylative Fluorination of 6(14).
Carboxylic acid 6 was completely consumed in the reaction, and the volatile cyclohexane 7 was extracted into diethyl ether, and then the solvent was carefully removed under vacuum. This preparation of 7 was used for subsequent variable-temperature nuclear magnetic resonance (VT-NMR) analysis.
2.1. VT-NMR Analysis
19F{1H}-NMR analysis was used to explore the conformational equilibria of 7. Ring inversion is too rapid at room temperature to resolve individual signals for 7ax and 7eq; however, these were readily resolved at lower temperature. Accordingly, 19F{1H}-NMR analysis of 7 was conducted at 195 K (−78 °C) in three different solvents (hexane (10% benzene-d6), dichloromethane-d2, and acetone-d6) and the outcomes are illustrated in Figure 3.
Figure 3.
Low temperature (195 K (−78 °C)) 19F{1H}-NMR spectra of 7, showing the signals for conformers 7ax (major) and 7eq (minor) in different solvents; (a) hexane (benzene-d6 10%); (b) dichloromethane-d2; (c) acetone-d6.
The assignments for each conformer were determined by applying proton coupling and recording 19F-NMR spectra.15 Each conformer has a signal with a large geminal 2JHF coupling (∼46 Hz) associated with the fluorine at C-4; however, 7ax also has a large 3JHF antiperiplanar coupling (∼46 Hz), whereas this is much reduced in 7eq (see Figure S4). In the 19F{1H}-NMR, the diastereotopic geminal CF2 group at C-1 constitutes an AB system and appears as a doublet of doublets (7ax, 2JFF = 234 Hz; 7eq,2JFF = 236 Hz), with the fluoromethylene (CHF) signal resonating upfield as a singlet. The major 7ax and minor 7eq conformers are clearly resolved as shown in Figure 3.
The 7ax conformer dominates over the 7eq in all cases and most significantly in hexane (12:1), although the ratios decrease as the polarity of the solvent increases, consistent with an electrostatic screening effect. In the case of hexane, it proved necessary to add-mix a deuterated solvent to provide an NMR “lock”. Thus, the hexane experiments were conducted with 10% benzene-d6 added, which will have a modest effect on the dielectric constant of the medium.
2.2. Theory Analysis
2.2.1. Cyclohexane 7
Theoretical calculations were carried out to complement the experimental observations, and they were extended to the additionally fluorinated cyclohexanes 9 and 10, as well as fluorocyclohexane 8, which was used as a reference compound. Calculations were run in the gas phase and using the IEFPCM implicit solvent model16 at the M06-2X/aug-cc-pVTZ level. Outcomes for 8 were close to that of a number of previous assessments which also marginally favor the 8eq conformer.17 The outcomes here for cyclohexanes 7–9 are illustrated in Figure 4.
Figure 4.
Conformational equilibria of fluorocyclohexanes 7 and 8 and analogues 9 and 10. Electrostatic interactions, equilibria energies (kcal mol–1) and atomic charges calculated in the gas-phase at the M06-2X/aug-cc-pVTZ theoretical level. Hydrogens in 1,3-relationships are chemically equivalent.
At the outset, the relative energies of 7ax and 7eq were explored. A significant energy difference of −1.06 kcal mol–1 was observed between 7ax and 7eq, favoring the axial conformer, in agreement with experiment. Notably, the difference in the axial preference for cyclohexane 7 (1.06 kcal mol–1) was larger when compared to methoxy-cyclohexane 5 (0.79 kcal mol–1). Natural bond orbital (NBO)17 analysis was used to deconvolute the total energy contributions from hyperconjugation, electrostatic interactions, and steric effects on the electronic energy between the axial and equatorial conformers of 7 (Table 1). This analysis revealed that electrostatic interactions (−ve values) govern the observed axial preference, whereas both hyperconjugative and steric effects (+ve values) drive the equilibrium toward the equatorial conformer.
Table 1. Gas-Phase Calculated Gibbs Free Energy (ΔG), Total Electronic Energy (ΔE) and NBO Analysis for Hyperconjugative (ΔENL), Electrostatic (ΔENCE) and Steric (ΔENSA) Contributions to the Total Electronic Energies of Compounds 5, 7, and 9–16 at the M06-2X/aug-cc-pVTZ Theoretical Level, in kcal mol–1b.
| Compound | ΔG | ΔE | (hyperconjugation) ΔENL | (electrostatics) ΔENCE | (sterics) ΔENSA |
|---|---|---|---|---|---|
| 5a | –0.79 | –1.08 | +2.65 | –3.50 | +3.38 |
| 7 | –1.06 | –1.10 | +1.41 | –3.24 | +1.93 |
| 8 | +0.12 | +0.08 | +1.61 | +3.76 | +2.00 |
| 9 | –2.73 | –2.81 | +8.39 | –20.29 | +1.01 |
| 10 | +2.00 | +2.16 | –3.67 | +5.16 | –0.75 |
| 11 | –0.70 | –0.93 | +0.18 | –5.81 | +2.58 |
| 12 | –0.67 | –0.94 | –0.56 | –6.14 | +2.69 |
| 13 | –1.41 | –1.66 | +2.66 | –10.38 | +2.18 |
| 14 | –0.92 | –1.23 | +1.02 | –8.00 | +3.28 |
| 15 | +3.70 | +3.64 | –1.40 | +3.21 | +0.28 |
| 16 | +4.04 | +3.87 | –0.78 | +2.94 | +0.73 |
Values obtained from ref (10b).
The Δ energies are considered as (ax–eq), thus negative energy values represent axial preference, and the positive ones represent equatorial preference
Interestingly for cyclohexane 7, a significant axial stabilization arises from the σCH → σ*CF hyperconjugative interactions, with a notable interaction energy of 5.81 kcal mol–1 in 7ax compared to 1.02 kcal mol–1 in 7eq. This finding aligns with the well-established hyperconjugative model used to elucidate the anomeric effect in different systems,19 where an antiperiplanar arrangement between an endo-cyclic donor group (σCH orbitals or 2p-type lone pair from oxygen, for example) and the antibonding orbital of an exo-cyclic C-X bond (nO → σ*CX or σCH → σ*CX interactions) is a driving force for the axial preference. Nevertheless, the cumulative effect of all hyperconjugative interactions in 7 still leads to favoring equatorial stabilization, but it is the electrostatic factor (ΔENCE), which tips the overall balance in favor of 7ax (detailed interactions addressed in the ESI).
Furthermore, the driving force behind 7ax stabilization is identified as the formation of 1,3-diaxial CH···FC nonconventional hydrogen bonds (NCHBs), each providing ∼11.5 kcal mol–1 of electrostatic stabilization to 7ax as delineated in Table 2. The critical role of NCHBs in stabilizing the axial conformation of fluorocyclohexanes is underscored by fluorocyclohexane 8. In this case, the absence of CF2 groups results in less polarization of axial hydrogens, thus weakening the NCHBs (−10.5 kcal mol–1 each). As a result, electrostatic interactions, in addition to hyperconjugative and steric effects, become equatorial stabilizing, leading to a slight shift of the equilibrium favoring the equatorial conformer. Interestingly, each NCHB in this context was found to be weaker than the corresponding interactions in methoxycyclohexane 5ax, which rendered ∼16.8 kcal mol–1 of electrostatic stabilization for each CH···OC. This aligns with the general expectation that fluorine is a less effective hydrogen bond acceptor than oxygen.5 Importantly, the stronger axial preference in cyclohexane 7, compared to that of 5, cannot be assigned to stronger NCHBs. Instead, it is tied to an entropic penalty associated with the rotation of the C-OMe bond in the axial conformer of 5, similar to observations well-known in alkyl-substituted cyclohexanes.20 As illustrated in Figure 5a, there are two accessible conformers for 5ax, where the methoxyl group extends outside the ring, while in 5eq, due to the equatorial orientation of the methoxyl group, there are three possible conformers. Consequently, the axial conformer (5ax) in this case has less degrees of freedom, resulting in a relative entropic penalty (TΔS = 0.21 kcal mol–1), which marginally shifts the equilibrium toward the equatorial conformer (5eq). By contrast, compound 7 lacks such an entropic effect since rotation around the C–F bond does not induce any conformational change (Figure 5b). Consequently, there is no relative conformational entropy associated with the 7ax/7eq conformer equilibrium (TΔS = 0.02 kcal mol–1).
Table 2. Individual Interaction Energies in 5ax/5eq and 7ax/7eq Contributing to the Global ΔENL, ΔENCE, and ΔENSA Obtained at the M06-2X/aug-cc-pVTZ Theoretical Level, in kcal mol–1.
| ΔENL (hyperconjugation) | ΔENCE (electrostatics) | ΔENSA (sterics) | |
|---|---|---|---|
| 5ax | σC2H9 → σ*C1O = 5.53 | CH···OC NCHB – 16.8 | σC1O18/σC2H8 = 0.57 |
| σC2H9 → σ*C1H = <0.5 | σC1O18/σC2H9 = 4.29 | ||
| 5eq | σC2H9 → σ*C1O = 0.82 | C1H7···H11C3 + 4.1 | σC1O18/σC2H8 = 0.61 |
| σC2H9 → σ*C1H = 3.63 | σC1O18/σC2H9 = 0.80 | ||
| 7ax | σC2H9 → σ*C1F = 5.81 | CH···FC NCHB – 11.5 | σC1F18/σC2H8 = 0.77 |
| σC2H9 → σ*C1H = <0.5 | σC1F18/σC2H9 = 4.03 | ||
| 7eq | σC2H9 → σ*C1F = 1.02 | C1H7···H11C3 + 4.1 | σC1F18/σC2H8 = 0.66 |
| σC2H9 → σ*C1H = 3.30 | σC1F18/σC2H9 = 0.79 |
Figure 5.
(a) Rational for entropic destabilization of 5ax (OMe) relative to (b); the 7ax (F) conformer due to the additional conformers associated with the −OMe relative to the −F substituent.
Implicit solvation effects with increasing dielectric constants of the three solvents used experimentally (hexane, dichloromethane, and acetone) were then considered. In each case, this reduced the energy between the 7ax and 7eq conformers, consistent with electrostatic screening, and the consequent change in ratios (increase in 7eq population) observed by VT-NMR (Table 3).
Table 3. Comparison of the Experimentally Determined (VT-NMR) and Computationally Derived Conformer Ratios for 7 in Different Solvents.
|
7ax: 7eq ratios |
||
|---|---|---|
| NMR solvent | experiment | theory |
| hexane (10% benzene-d6) | 12:1 | 10.5:1 |
| DCM-d2 | 6.5:1 | 5.5:1 |
| acetone-d6 | 3.7:1 | 5.2:1 |
However, in all cases, the axial conformer 7ax was dominant, consistent with experiment. Other twist boat conformers were considered, but they were >5 kcal mol–1 higher than the axial conformers and were judged not to have any significant contribution to the conformer population in solution (Figure S6 in the SI).
Figure 6.
Calculated relative Gibbs free energies (ΔG) in kcal mol–1 and populations (Pop) in percentage for 7ax and 7eq in gas-phase and implicit solvents (hexane, dichloromethane, and acetone), at the M06-2X/aug-cc-pVTZ/IEFPCM level of theory.
2.2.2. Cyclohexanes 9 and 10
In order to explore these phenomena further in a theoretical context, the introduction of three CF2 groups into the ring was explored for cyclohexane 9. This resulted in a nearly identical NCHB interaction energy compared to 7 (−11.5 kcal mol–1). This similarity arises despite the higher polarization of the 2,6-diaxial hydrogens, presumably as the CF2 groups also withdraw electron density from the fluorine at position 4 of the ring, reducing its electron density. Nonetheless, the axial preference over the equatorial preference in cyclohexane 9 significantly increases, favoring 9ax by 2.73 kcal mol–1.
In the context of the fluorine-substituted compounds, where CH···FC NCHBs are weaker compared with their CH···OMe analogues, other electrostatic interactions also play a significant role in determining the conformational equilibria in 9. For example, in addition to stabilizing NCHBs, 1,3-Fax···Heq interactions greatly stabilize 9ax by −9.5 kcal mol–1. Meanwhile, 1,3-Hax···Hax electrostatic contacts notably destabilize the equatorial conformer 9eq (+5.7 kcal mol–1), substantially enhancing the observed axial preference. These electrostatic interactions are also reflected in the observed ring strains for compounds 9ax and 9eq. Specifically, when examining all C–C–C–C dihedral angles involving adjacent carbon atoms in the ring, as well as those encompassing the hydrogen and fluorine atoms at the monofluorinated carbon, the average calculated dihedral angle is −40.7 degrees for 9ax and −39.7 degrees for 9eq (Table 4). In the latter case, this represents a 2° difference in comparison to the unsubstituted cyclohexane’s dihedral angle of −41.7 degrees. The tighter average dihedral angle in 9eq indicates an increase in ring strain in the equatorial conformer, consequently resulting in its destabilization. This effect is not as pronounced in compounds 7 and 8 due to the weaker electrostatic interactions in these cases. In addition, the geminal F–C–H angle in the monofluorinated carbon is slightly wider in 9ax (109.9°) compared to that in 9eq (109.0°), highlighting the shorter CHax···FaxC contact in the former and longer 1,3-Hax···Hax contact in the latter.
Table 4. Selected 3-Atom Angles and 4-Atom Dihedrals (Degrees) from the Gas-Phase Optimized Structures of Unsubstituted Cyclohexane and Compounds 7–10, Obtained at the M06-2X/aug-cc-pVTZ Theoretical Level.
| dihedral (ω) | cyclohexane | 7ax | 7eq | 8ax | 8eq | 9ax | 9eq | 10ax | 10eq |
|---|---|---|---|---|---|---|---|---|---|
| 1–2–3–4 | 55.7 | 53.5 | 54.5 | 54.8 | 55.9 | 51.4 | 51.4 | 50.3 | 53.3 |
| 2–3–4–5 | –55.7 | –55.0 | –53.8 | –56.8 | –55.6 | –50.3 | –50.5 | –49.8 | –48.9 |
| 3–4–5–6 | 55.7 | 55.0 | 53.8 | 56.7 | 55.7 | 50.2 | 50.5 | 49.8 | 48.9 |
| 7–1–2–3 | –178.5 | –178.9 | –178.6 | –177.9 | –178 | –175.7 | –174.6 | –171.5 | –176.4 |
| 7–1–2–9 | –58.3 | –58.0 | –58.1 | –57.2 | –57.8 | –52.5 | –51.1 | –52.0 | –57.1 |
| 18–1–2–3 | 64.4 | 64.9 | 64.7 | 66.3 | 65.5 | 64.4 | 66.4 | 72.1 | 66.1 |
| 18–1–2–9 | –175.4 | –174.2 | –174.9 | –173 | –174.3 | –172.3 | –170.1 | –168.4 | –174.5 |
| average | –41.7 | –41.8 | –41.8 | –41.0 | –41.2 | –40.7 | –39.7 | –38.5 | –41.2 |
| angle (∠) | cyclohexane | 7ax | 7eq | 8ax | 8eq | 9ax | 9eq | 10ax | 10eq |
|---|---|---|---|---|---|---|---|---|---|
| 1–2–3 | 111.1 | 111.3 | 110.3 | 111.5 | 110.3 | 113.0 | 112.7 | 112.9 | 110.3 |
| 2–1–6 | 111.1 | 112.5 | 111.7 | 112.6 | 112.0 | 112.0 | 112.2 | 112.8 | 111.4 |
| 2–1–7 | 110.4 | 110.7 | 109.0 | 110.8 | 109.1 | 111.3 | 111.7 | 114.0 | 114.1 |
| 2–1–18 | 109.0 | 108.2 | 110.1 | 108.1 | 110.0 | 110.4 | 109.8 | 109.3 | 108.5 |
| 2–3–4 | 111.1 | 110.4 | 110.5 | 111.0 | 111.1 | 107.3 | 108.2 | 109.3 | 110.5 |
| 3–4–5 | 111.1 | 113.8 | 113.9 | 110.7 | 111.0 | 114.2 | 113.8 | 111.9 | 111.7 |
| 7–1–18 | 106.9 | 106.4 | 106.7 | 106.1 | 106.5 | 109.9 | 109.0 | 106.7 | 107.4 |
| average | 110.1 | 110.5 | 110.3 | 110.1 | 110.0 | 111.2 | 111.1 | 111.0 | 110.6 |
It is worth noting that the primary factor for axial stabilization in 9ax remains the 1,3-diaxial Fax···Hax NCHBs. This is evident when the stabilizing CHax···Fax interactions are replaced by destabilizing CFax···Fax contacts, as illustrated in pentafluorocyclohexane 10. In this case, the emergence of F···F repulsions (+16.1 kcal mol–1 each) in 10ax leads to a preference for the 10eq conformer by 2.00 kcal mol–1 and results in an elevated torsional strain in 10ax, illustrated by its considerably tighter average dihedral angle (−38.5°), 3.2° higher than the average dihedral angle of an unsubstituted cyclohexane (−41.7°). Furthermore, the stability of 10eq is reinforced by electrostatic transannular NCHBs, each contributing −8.4 kcal mol–1 in further stability to 10eq. Similar to 9, the F–C–H angle in the monofluorinated carbon is slightly wider in 10eq (107.4°) in order to favor NCHB formation compared to 10ax (106.7°). Again, the increase in solvent polarity attenuates the electrostatic NCHB interactions, resulting in compounds 8ax, 9ax, and 10eq becoming progressively less stable going from hexane to dichloromethane and acetone (see Table S3), consistent with an electrostatic screening.
2.2.3. Comparison with Other Halogens in Cyclohexanes 11–16
Given the significant impact of electrostatic interactions, particularly NCHBs, on the axial stabilization observed for fluorine, this study was extended also to chlorinated and brominated cyclohexane analogues to explore how these interactions evolve with different halogens. These halogens do not make strong hydrogen bond acceptors when bound to carbon.21 The relative energies and contributions to stability were considered for cyclohexanes 11-16. Interestingly, electrostatic interactions continue to dictate the conformational equilibria in these cases; however, owing to the lower electronegativity of Cl and Br, electrostatic interactions other than NCHBs emerge as the predominant factors. For instance, in compounds 11 and 12, the 1,3-diaxial NCHBs to chlorine and bromine exhibit only a weak axial stabilization (−2.7 and −1.2 kcal mol–1, respectively), as shown in Figure 7. Instead, the axial preference is primarily influenced by destabilizing 1,3-δ+Hax···δ+Hax interactions in the equatorial conformer, with each contributing approximately 5.5 kcal mol–1 in destabilizing energy. It is noteworthy that the hydrogen atom geminal to the halogen substituent in 11(Cl) and 12(Br) is more positively charged compared to that of the fluorinated analogue 7, despite the higher electronegativity of fluorine. This counterintuitive observation arises from the better superposition between nF lone pairs and σ*CH antibonding orbitals, resulting in a more effective charge transfer back to the hydrogen atom through hyperconjugation, compared to the same effect for 11(Cl) and 12(Br) (Figure 8a). Consequently, the axial preferences diminish in comparison to 7 (F), reducing to 0.70 kcal mol–1 in 11 (Cl) and 0.67 kcal mol–1 in 12 (Br).
Figure 7.
Conformational equilibria of compounds 11–16. Electrostatic interaction energies (kcal mol–1) and atomic charges were calculated in Gas-Phase at the M06-2X/aug-cc-pVTZ theoretical level. Hydrogens in 1,3-relationships are chemically equivalent.
Figure 8.
Outcomes at the M06-2X/aug-cc-pVTZ theory level: (a) nX → σ*CH hyperconjugative interaction that decreases the atomic charge of the geminal H in compounds 7eq, 11eq, and 12eq. (b) Sum of nF/nX steric repulsion energies that destabilize the 15ax and 16ax conformers.
Calculations extended to cyclohexanes 13 (Cl) and 14 (Br), which have more CF2 groups in the ring. This further polarizes the hydrogen atoms, and the halogen substituents shift toward virtually neutral (−0.01 au in Cl) or positive (+0.06 au in Br) atomic charges. Consequently, 1,3-Hax···Xax electrostatic interactions become very weak (−0.4 kcal mol–1) in 13 (Cl) or repulsive (+1.8 kcal mol–1) in 14 (Br). In both cases, the axial preference is once again assigned to stabilizing 1,3-δ-Fax···δ+Heq interactions in the axial conformers (∼12 kcal mol–1 each) and destabilizing 1,3-δ+Hax···δ+Hax electrostatic contacts (∼7 kcal mol–1 each) in the equatorial counterparts. Notably, the higher axial preference in chlorinated analogue 13 (1.41 kcal mol–1) compared to brominated compound 14 (0.92 kcal mol–1) is attributed, among other smaller electrostatic interactions, to the modestly stabilizing 1,3-δ+Hax···δ-Clax interactions as opposed to similarly destabilizing 1,3-δ+Hax···δ+Brax interactions.
Despite the absence of nonconventional hydrogen bonds involving chlorine and bromine in the cyclohexanes studied, the introduction of CF2 groups at positions 3 and 5 of the ring, as observed in compounds 15 (Cl) and 16 (Br), induces a substantial shift in the equilibria toward the equatorial conformers by 3.70 and 4.04 kcal mol–1, respectively. This is assisted by the formation of strong 1,3-Fax···Hax NCHBs in the equatorial conformers, contributing −10.6 kcal mol–1 in 15 (Cl) and −10.7 kcal mol–1 in 16 (Br) in stabilizing electrostatic energy, respectively. Notably, despite the comparable values for these NCHBs in both cases, the global difference in the magnitude of the equatorial preference can be attributed to higher steric hindrance in 16ax (Br) resulting from more repulsive nBr/nF orbital interactions, compared to similar nCl/nF interactions in 15 (Cl) (Figure 8b).
3. Conclusions
In summary, a clear role for electrostatic nonclassical CF···HC hydrogen bonding interactions is demonstrated in influencing the conformational equilibrium of selectively halogenated cyclohexanes. Notably, a pseudoanomeric effect is observed for 1,1,4-trifluorocyclohexane 7 perhaps unexpectedly showing a bias for the 7ax over the 7eq conformer. A comparison is made between the experimental data in solution and in different solvents and that established by theory in the gas phase. Conformer populations established by experiment and theory for 7 are relatively close as summarized in Table 3. The theory approach was able to deconvolute electrostatic from hyperconjugative contributions to the relative stabilization of the halogenated cyclohexanes. Electrostatic contributions dominate the preference for 7ax over 7eq. Notably too, the energy difference between the 7ax/7eq conformer is greater than that found for the methoxyl derivative 5. Stabilizing nonclassical hydrogen bonding (NCHB) CO···HC interactions makes a greater contribution in 5 than those involving fluorine in 7, but attaining the 5ax conformer requires a significantly larger entropy penalty than for 7ax, and thus globally 7 shows the higher axial preference. Additional CF2 groups were introduced around the cyclohexane ring and depending on their placement the cyclohexanes adopt preferred axial (eg9) or equatorial (eg10) conformers, influenced largely by electrostatic (NCHB) CF···HC interactions.
Theory comparisons extended to exploring the analogous chloro- and bromocyclohexanes 11–16 with CF2’s similarly placed in the cyclohexane rings. In these cases, NCHBs between CCl···HC and CBr···HC did not offer significant stabilizing interactions. Instead, it was found that the geminal hydrogens in CHX (X = Cl, Br) were more electropositive than that found for CHF, and (NCHB) CF···HCX electrostatic interactions involving fluorine to these geminal hydrogens contributed significantly to the conformer populations of 11–16.
4. Experimental Section
4.1. Computational Methods
The optimization of axial and equatorial conformers for compounds 7–10 and 11–16 employed Truhlar’s hybrid meta-GGA functional M06-2X,22 coupled with Dunning’s correlation consistent triple-ζ basis set augmented with diffuse functions aug-cc-pVTZ.23 The choice of the M06-2X/aug-cc-pVTZ theoretical level was based on its proven accuracy in previous studies involving analogous fluorocyclohexanes.10 Harmonic frequency calculations at the same theoretical level were carried out in order to identify each geometry as a true energy minimum, showing no imaginary frequency. Thermal corrections to the electronic energy within the ideal gas-rigid rotor-harmonic oscillator model were derived from these frequency calculations, providing the ring interconversion ΔG energy for the equilibria of compounds 7–10 and 11–16. NBO calculations were performed using the NBO7.0 program18 at the M06-2X/aug-cc-pVTZ level of theory including the NBO energetic analysis, natural Coulomb electrostatics,24 and natural steric analysis25 (LEWIS, NCE, and STERIC keywords, respectively). The Gibbs free energies in solution were determined using the M06-2X/aug-cc-pVTZ theoretical level employing the integral equation formalism variant of the polarazible continuum model (IEFPCM).16 All calculations were performed using Gaussian 16 Rev C.01 program.26
Acknowledgments
M.H. acknowledges support from the Chinese Scholarship Council for a studentship. R.A.C. and B.A.P. thanks FAPESP for a Young Researcher Award (#2018/03910-1) and a scholarship (#2023/14064-2), respectively. FAEPEX is acknowledged by a fellowship (#3472/23). CENAPAD-SP and CESUP are also acknowledged for the computational resources used in theory calculations.
Data Availability Statement
The data underlying this study are available in the published article and its Supporting Information.
Supporting Information Available
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.joc.3c02868.
Experimental protocols for the synthesis of 7 and outlined computational methods and associated data (PDF)
The authors declare no competing financial interest.
Supplementary Material
References
- a Hutchet Q. A.; Kuhn B.; Wagner B.; Fischer H.; Kan-say M.; Zimmerli D.; Carreira E. M.; Műller K. On the Polarity of Partially Fluorinated Methyl Groups. J. Fluor. Chem. 2013, 152, 119–128. 10.1016/j.jfluchem.2013.02.023. [DOI] [Google Scholar]; b Smart B. E. Fluorine Substituent Effects (on Bioactivity). J. Fluor. Chem. 2001, 109, 3–11. 10.1016/S0022-1139(01)00375-X. [DOI] [Google Scholar]
- a Troup R. I.; Jeffries B.; Saudain R. E.-B.; Georgiou E.; Fish J.; Scott J. S.; Chiarparin E.; Fallan C.; Linclau B. Skipped Fluorination Motifs: Synthesis of Building Blocks and Comparison of Lipophilicity Trends with Vicinal and Isolated Fluorination Motifs. J. Org. Chem. 2021, 86, 1882–1900. 10.1021/acs.joc.0c02810. [DOI] [PubMed] [Google Scholar]; b Jeffries B. F. J.; Wang Z.; Felstead H. R.; Le Questel J.-Y.; Scott J.; Chiarparin E.; Graton J.; Linclau B. Systematic Investigation of Lipophilicity Modulation by Aliphatic Fluorination Motifs. J. Med. Chem. 2020, 63, 1002–1031. 10.1021/acs.jmedchem.9b01172. [DOI] [PubMed] [Google Scholar]; c Linclau B.; Wang Z.; Compain G.; Paumelle V.; Fontenelle C. Q.; Wells N.; Weymouth-Wilson A. Investigating the Influence of (Deoxy)fluorination on the Lipophilicity of Non-UV-active Fluorinated Alkanols and Carbohydrates by a New log P Determination Method. Angew. Chemie. Int. Ed. 2016, 55, 674–678. 10.1002/anie.201509460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- a Thiehoff C.; Rey Y. P.; Gilmour R. The Fluorine Gauche Effect: A Brief History. Isr. J. Chem. 2017, 57, 92–100. 10.1002/ijch.201600038. [DOI] [Google Scholar]; b Wolfe S. Gauche effect. Stereochemical consequences of adjacent electron pairs and polar bonds. Acc. Chem. Res. 1972, 5, 102–111. 10.1021/ar50051a003. [DOI] [Google Scholar]
- a Wang C.; Ying F.; Wu W.; Mo Y. How Solvent Influences the Anomeric Effect: Roles of Hyperconjugative versus Steric Interactions on the Conformational Preference. J. Org. Chem. 2014, 79, 1571–1581. 10.1021/jo402306e. [DOI] [PubMed] [Google Scholar]; b Mo Y. Computational evidence that hyperconjugative interactions are not responsible for the anomeric effect. Nature. Chem. 2010, 2, 666–671. 10.1038/nchem.721. [DOI] [PubMed] [Google Scholar]; c Lemieux R. U. Effects of unshared pairs of electrons and their solvation on conformational equilibria. Pure. Appl. Chem. 1971, 25, 527–548. 10.1351/pac197125030527. [DOI] [Google Scholar]; d Edward J. T. Stability of glycosides to acid hydrolysis. A conformational analysis. Chem. Ind. 1955, 36, 1102–1104. [Google Scholar]
- a Vulpetti A.; Dalvit C. Hydrogen Bond Acceptor Propensity of Different Fluorine Atom Types: an Analysis of Experimentally and Computationally Derived Parameters. Chem. - Eur. J. 2021, 27, 8764–8773. 10.1002/chem.202100301. [DOI] [PubMed] [Google Scholar]; b Schneider H.-J. Hydrogen Bonds with Fluorine. Studies in Solution, in Gas Phase and by Computations, Conflicting Conclusions from Crystallographic Analyses. Chem. Sci. 2012, 3, 1381–1394. 10.1039/c2sc00764a. [DOI] [Google Scholar]; c Dunitz J. D.; Taylor R. Organic Fluorine Hardly Ever Accepts Hydrogen Bonds. Chem. - Eur. J. 1997, 3, 89–98. 10.1002/chem.19970030115. [DOI] [Google Scholar]; d Howard J. A.; Hoy V. J.; O’Hagan D.; Smith G. T. How Good is Fluorine as a Hydrogen Bond Acceptor?. Tetrahedron 1996, 52, 12613–12622. 10.1016/0040-4020(96)00749-1. [DOI] [Google Scholar]
- a Nairoukh Z.; Strieth-Kalthoff F.; Bergander K.; Glorius F. Understanding the Conformational Behavior of Fluorinated Piperidines: The Origin of the Axial-F Preference. Chem. - Eur. J. 2020, 26, 6141–6146. 10.1002/chem.202001355. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Sun A.; Lankin D. C.; Hardcastle K.; Snyder J. P. 3-Fluoropiperidines and N-methyl-3-fluoropiperidinium Salts: The Persistence of Axial Fluorine. Chem. - Eur. J. 2005, 11, 1579–1591. 10.1002/chem.200400835. [DOI] [PubMed] [Google Scholar]; c Lankin D. C.; Grunewald G. L.; Romero F. A.; Oren I. Y.; Snyder J. P. The NH---FC Dipole Orientation Effect for Pendant Exocyclic CH2F. Org. Lett. 2002, 4, 3557–3560. 10.1021/ol026358c. [DOI] [PubMed] [Google Scholar]; d Snyder J. P.; Chandrakumar N. S.; Sato H.; Lankin D. C. The Unexpected Diaxial Orientation of cis-3,5-Difluoropiperidine in Water: a Potent CF---NH Charge-dipole Effect. J. Am. Chem. Soc. 2000, 122, 544–545. 10.1021/ja9934504. [DOI] [Google Scholar]
- a Gooseman N. E.; O’Hagan D.; Peach M. J.; Slawin A. M.; Tozer D. J.; Young R. J. An Electrostatic Gauche Effect in β-Fluoro- and β-Hydroxy-N-ethylpyridinium Cations. Angew. Chemie. Int. Ed. 2007, 46, 5904–5908. 10.1002/anie.200700714. [DOI] [PubMed] [Google Scholar]; b Briggs C. R.; Allen M. J.; O’Hagan D.; Tozer D. J.; Slawin A. M.; Goeta A. E.; Howard J. A. The Observation of a Large Gauche Preference When 2-Fluoroethylamine and 2-Fluoroethanol Become Protonated. Org. Biomol. Chem. 2004, 2, 732–740. 10.1039/b312188g. [DOI] [PubMed] [Google Scholar]
- a Port V. C.; Cormanich R. A. There and Back Again: The Role of Hyperconjugation in the Fluorine Gauche Effect. Phys. Chem. Chem. Phys. 2021, 23, 17329–17337. 10.1039/D1CP02806E. [DOI] [PubMed] [Google Scholar]; b Aufiero M.; Gilmour R. Informing Molecular Design by Stereoelectronic Theory: The Fluorine Gauche Effect in Catalysis. Acc. Chem. Res. 2018, 51, 1701–1710. 10.1021/acs.accounts.8b00192. [DOI] [PubMed] [Google Scholar]; c Yamamoto I.; Jordan M. J.; Gavande N.; Doddareddy M. R.; Chebib M.; Hunter L. The Enantiomers of Syn-2, 3-difluoro-4-aminobutyric Acid Elicit Opposite Responses at the GABA C Receptor. Chem. Commun. 2012, 48, 829–831. 10.1039/C1CC15816C. [DOI] [PubMed] [Google Scholar]; d Tavasli M.; O’Hagan D.; Pearson C.; Petty M. C. The Fluorine Gauche Effect. Langmuir Isotherms Report the Relative Conformational Stability of (±)-Erythro- and (±)-Threo-9,10-difluorostearic acids. Chem. Commun. 2002, 1226–1227. 10.1039/b202891c. [DOI] [PubMed] [Google Scholar]
- a Alabugin I. V.; Kuhn L.; Krivoshchapov N. V.; Mehaffy P.; Medvedev M. G. Anomeric Effect, Hyperconjugation and Electrostatics: Lessons from Complexity in a Classic Stereoelectronic Phenomenon. Chem. Soc. Rev. 2021, 50, 10212–10252. 10.1039/D1CS00564B. [DOI] [PubMed] [Google Scholar]; b Díaz N.; Jiménez-Grávalos F.; Suárez D.; Francisco E.; Martín-Pendás Á. Fluorine Conformational Effects Characterized by Energy Decomposition Analysis. Phys. Chem. Chem. Phys. 2019, 21, 25258–25275. 10.1039/C9CP05009D. [DOI] [PubMed] [Google Scholar]; c Thacker J. C.; Popelier P. L. Fluorine Gauche Effect Explained by Electrostatic Polarization Instead of Hyperconjugation: An Interacting Quantum Atoms (IQA) and Relative Energy Gradient (REG) Study. J. Phys. Chem. A 2018, 122, 1439–1450. 10.1021/acs.jpca.7b11881. [DOI] [PubMed] [Google Scholar]
- a Piscelli B. A.; O’Hagan D.; Cormanich R. A. Contribution of Hyperconjugation and Inductive Effects to the Pseudo-anomeric Effect in 4-Substituted Methoxycyclohexanes. J. Phys. Chem. A 2023, 127, 724–728. 10.1021/acs.jpca.2c08090. [DOI] [PubMed] [Google Scholar]; b Piscelli B. A.; O’Hagan D.; Cormanich R. A. The Contribution of Non-classical CHax···OC Hydrogen Bonds to the Anomeric Effect in Fluoro and Oxa-methoxycyclohexanes. Phys. Chem. Chem. Phys. 2021, 23, 5845–5851. 10.1039/D0CP06646J. [DOI] [PubMed] [Google Scholar]
- a Wiberg K. B.; Bailey W. F.; Lambert K. M.; Stempel Z. D. The Anomeric Effect: It’s Complicated. J. Org. Chem. 2018, 83, 5242–5255. 10.1021/acs.joc.8b00707. [DOI] [PubMed] [Google Scholar]; b Takahashi O.; Yamasaki K.; Kohno Y.; Ueda K.; Suezawa H.; Nishio M. The Origin of the Generalized Anomeric Effect: Possibility of CH/n and CH/π Hydrogen Bonds. Carbohydr. Res. 2009, 344, 1225–1229. 10.1016/j.carres.2009.04.011. [DOI] [PubMed] [Google Scholar]
- a Ajitha M. J.; Huang K.-W. Non-classical C–H···X Hydrogen Bonding and Its Role in Asymmetric Organocatalysis. Synthesis 2016, 48, 3449–3458. 10.1055/s-0035-1562475. [DOI] [Google Scholar]; b Johnston R. C.; Cohen D. T.; Eichman C. C.; Scheidt K. A.; Cheong P. H.-Y. Catalytic Kinetic Resolution of a Dynamic Racemate: Highly Stereoselective β-lactone Formation by N-heterocyclic Carbene Catalysis. Chem. Sci. 2014, 5, 1974–1982. 10.1039/C4SC00317A. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stolow R.; Kao J. Y. Conformational Studies of 1,1,4-Trifluorocyclohexane. Abstr. Pap. Am. Chem. Soc. 1990, 284. [Google Scholar]
- Ventre S.; Petronijevic F. R.; MacMillan D. W. Decarboxylative Fluorination of Aliphatic Carboxylic Acids via Photoredox Catalysis. J. Am. Chem. Soc. 2015, 137, 5654–5657. 10.1021/jacs.5b02244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dolbier W. R., Jr.Guide to Fluorine NMR for Organic Chemists; Wiley, J.; Inc. S.2009, Hoboken, NJ. [Google Scholar]
- Menucci B.; Cammi R.; Tomasi J. Excited states and solvatochromic shifts within a nonequilibrium solvation approach: A new formulation of the integral equation formalism method at the self-consistent field, configuration interaction, and multiconfiguration self-consistent field level. J. Chem. Phys. 1998, 109, 2798. 10.1063/1.476878. [DOI] [Google Scholar]
- a Luo Q.; Randall K. R.; Schaefer H. F. Easy Chairs: The Conformational Preferences of Polyfluorocyclohexanes. RSC Adv. 2013, 3, 6572–6585. 10.1039/c3ra40538a. [DOI] [Google Scholar]; b Durig J. R.; Ward R. M.; Nelson K. G.; Gounev T. K. Conformational Stability from Variable Temperature Infrared Spectra of Xenon Solution, r0 Structural Parameters, and Vibrational Assignment of Fluorocyclohexane. J. Mol. Struct. 2010, 976, 150–160. 10.1016/j.molstruc.2010.01.058. [DOI] [Google Scholar]; c Weldon A. J.; Vickrey T. L.; Tschumper G. S. Intrinsic Conformational Preferences of Substituted Cyclohexanes and Tetrahydropyrans Evaluated at the CCSD (T) Complete Basis Set Limit: Implications for the Anomeric Effect. J. Phys. Chem. A 2005, 109, 11073–11079. 10.1021/jp0550311. [DOI] [PubMed] [Google Scholar]
- NBO 7.0, E. D.; Glendening J. K.; Badenhoop A. E.; Reed J. E.; Carpenter J. A.; Bohmann C. M.; Morales P.; Karafiloglou C. R.; Landis; Weinhold F.. Theoretical Chemistry Institute, University of Wisconsin, Madison, 2018. [Google Scholar]
- a Juaristi E.; Notario R. Theoretical Evidence for the Relevance of n(F) → σ*(C–X) (X = H, C, O, S) Stereoelectronic Interactions. J. Org. Chem. 2016, 81, 1192–1197. 10.1021/acs.joc.5b02718. [DOI] [PubMed] [Google Scholar]; b Freitas M. P. The anomeric effect on the basis of natural bond orbital analysis. Org. Biomol. Chem. 2013, 11, 2885–2890. 10.1039/c3ob40187a. [DOI] [PubMed] [Google Scholar]; c Hasanzadeh N.; Nori-Shargh D.; Farzipour M.; Ahmadi B. The origin of the anomeric effect: probing the impacts of stereoelectronic interactions. Org. Biomol. Chem. 2015, 13, 6965–6976. 10.1039/C5OB00585J. [DOI] [PubMed] [Google Scholar]; d Juaristi E.; Notario R. Density Functional Theory Computational Reexamination of the Anomeric Effect in 2-Methoxy- and 2-Cyano-1,3-dioxanes and 1,3-Dithianes. Stereoelectronic Interactions Involving the Cyano (C≡N:) Group Revealed by Natural Bond Orbital (NBO) Analysis. J. Org. Chem. 2018, 83, 10326–10333. 10.1021/acs.joc.8b01458. [DOI] [PubMed] [Google Scholar]
- Booth H.; Everett J. R. The Experimental Determination of the Conformational Free Energy, Enthalpy, and Entropy Differences for Alkyl Groups in Alkylcyclohexanes by Low Temperature Carbon-13 Magnetic Resonance Spectroscopy. J. Chem. Soc., Perkin Trans. 1980, 2, 255–259. 10.1039/p29800000255. [DOI] [Google Scholar]
- Aullón G.; Bellamy D.; Orpen A. G.; Brammer L.; Bruton E. A. Metal-bound Chlorine Often Accepts Hydrogen Bonds. Chem. Commun. 1998, 653–654. 10.1039/A709014E. [DOI] [Google Scholar]
- Zhao Y.; Truhlar D. G. The M06 Suite of Density Functionals for Main Group Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States, and Transition Elements: Two New Functionals and Systematic Testing of Four M06-Class Functionals and 12 Other Functionals. Theor. Chem. Acc. 2008, 120, 215–241. 10.1007/s00214-007-0310-x. [DOI] [Google Scholar]
- Kendall R. A.; Dunning T. H. Jr.; Harrison R. J. Electron affinities of the first-row atoms revisited. Systematic basis sets and wave functions. J. Chem. Phys. 1992, 96, 6796–6806. 10.1063/1.462569. [DOI] [Google Scholar]
- Weinhold F.; Landis C. R.. Discovering Chemistry with Natural Bond Orbitals; Wiley-VCH, 2012; pp 145ff. [Google Scholar]
- a Badenhoop J. K.; Weinhold F. Natural bond orbital analysis of steric interactions. J. Chem. Phys. 1997, 107, 5406–5421. 10.1063/1.474248. [DOI] [Google Scholar]; b Badenhoop J. K.; Weinhold F. Natural steric analysis of internal rotation barriers. Int. J. Quantum Chem. 1999, 72, 269–280. . [DOI] [Google Scholar]
- Gaussian 16, Revision C.01, Frisch M. J.; Trucks G. W.; Schlegel H. B.; Scuseria G. E.; Robb M. A.; Cheeseman J. R.; Scalmani G.; Barone V.; Petersson G. A.; Nakatsuji H.; Li X.; Caricato M.; Marenich A. V.; Bloino J.; Janesko B. G.; Gomperts R.; Mennucci B.; Hratchian H. P.; Ortiz J. V.; Izmaylov A. F.; Sonnenberg J. L.; Williams-Young D.; Ding F.; Lipparini F.; Egidi F.; Goings J.; Peng B.; Petrone A.; Henderson T.; Ranasinghe D.; Zakrzewski V. G.; Gao J.; Rega N.; Zheng G.; Liang W.; Hada M.; Ehara M.; Toyota K.; Fukuda R.; Hasegawa J.; Ishida M.; Nakajima T.; Honda Y.; Kitao O.; Nakai H.; Vreven T.; Throssell K.; Montgomery J. A. Jr.; Peralta J. E.; Ogliaro F.; Bearpark M. J.; Heyd J. J.; Brothers E. N.; Kudin K. N.; Staroverov V. N.; Keith T. A.; Kobayashi R.; Normand J.; Raghavachari K.; Rendell A. P.; Burant J. C.; Iyengar S. S.; Tomasi J.; Cossi M.; Millam J. M.; Klene M.; Adamo C.; Cammi R.; Ochterski J. W.; Martin R. L.; Morokuma K.; Farkas O.; Foresman J. B.; Fox D. J.. Gaussian, Inc.: Wallingford CT, 2016. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this study are available in the published article and its Supporting Information.