Article pubs.acs.org/JPCA
Experimental and Theoretical Studies of Intramolecular Hydrogen Bonding in 3‑Hydroxytetrahydropyran: Beyond AIM Analysis Daniela C. Solha, Thaís M. Barbosa, Renan V. Viesser, Roberto Rittner, and Cláudio F. Tormena* Chemistry Institute, University of Campinas - UNICAMP, P.O. Box 6154, 13083-970 Campinas, SP, Brazil S Supporting Information *
ABSTRACT: The conformational preferences of 3-hydroxytetrahydropyran (1) were evaluated using infrared and nuclear magnetic resonance spectroscopic data in solvents of different polarities. Theoretical calculations in the isolated phase and including the solvent effect were performed, showing that the most stable conformations for compound 1 are those containing the substituent in the axial and equatorial orientations. The axial conformation is more stable in the isolated phase and in a nonpolar solvent, while the equatorial conformation is more stable than the axial in polar media. The occurrence of intramolecular hydrogenbonded O−H···O in the axial conformer was detected from infrared spectra in a nonpolar solvent at different concentrations. Our attempt to evaluate this interaction using population natural bond orbital and topological quantum theory of atoms in molecules analyses failed, but topological noncovalent interaction analysis was capable of characterizing it.
■
authors23−25 have refuted the hyperconjugative contribution to the anomeric effect because the equatorial form is favored in polar solvents. An electrostatic model (Figure 1b) was introduced to explain the equatorial preference over axial for substituted tetrahydropyrans and sugars in polar solvents, but it fails to explain the geometrical parameters, such as the shorter C−O and larger C−X bond lengths in the axial orientation relative to the equatorial in the presence of electronegative substituents. Recently, some evidence has indicated that not only the electrostatic interaction (Figure 1b) but also hyperconjugation (Figure 1a and 1c) is dependent on the medium,29,30 providing further support that studies in this field are far from complete. Several studies have investigated 2-substituted tetrahydropyrans, and no consensus about the interactions responsible for the conformational preferences has been found. In contrast, conformational studies have neglected 3-substituted tetrahydropyrans. The importance of 3-substituted tetrahydropyrans is remarkable due to its involvement in synthetic intermediates,31−33 potential anticonvulsant agents,34 and the sex pheromones of male Tirathaba mundella.35 Although 3-substituted tetrahydropyrans are important in several areas, as previously mentioned, it is well known1,22 that some of these properties are due to the conformational preferences adopted by the molecular system. The simplest molecular system that could be used as a model compound to study the conformational preference in this series is 3hydroxytetrahydropyran (Figure 2). In the literature, only one study36 investigates the conformational preference of 3-
INTRODUCTION Studies on the conformation preferences of six-membered rings have provided the foundation for modern stereochemistry.1 However, the most important aspect of conformational analysis includes not only determining which conformation is most stable but also evaluating why one molecular arrangement is preferred over another. The evaluation of the driving force responsible for a specific molecular arrangement (conformational preference) improves our knowledge of the interactions involving atoms or functional groups responsible for stabilizing a conformation. In the last two decades, several studies have been conducted to describe the conformational preferences of several sixmembered ring systems, with cyclohexane systems being the most studied. In methylcyclohexane, the methyl group clearly preferentially adopts the equatorial orientation,1 but the reasons for this preference are still controversial.2−4 For more complex molecular systems, such as 1,2-disubstituted cyclohexanes,5−14 2-substituted cyclohexanones,15−20 and 2-substituted methylenecyclohexanes,21 the conformation equilibrium involves axial and equatorial conformers, and the preference for axial or equatorial is dictated by the balance between stereoelectronic interactions present in each molecular system. There is a controversy in the literature about which stereoelectronic effect is responsible for the conformational preference in such molecules as methylcyclohexane.2−4 A similar controversy has been observed in the literature introducing an explanation for the relative importance of stereoelectronic interactions (Figure 1) involving the anomeric effect for two-substituted tetrahydropyrans and sugars.22−28 The anomeric effect has been interpreted as being due to the hyperconjugative LP2(O)→σ*C−X interaction (Figure 1a), which stabilizes the axial conformation.22 However, other © 2014 American Chemical Society
Received: January 8, 2014 Revised: March 7, 2014 Published: March 31, 2014 2794
dx.doi.org/10.1021/jp500211y | J. Phys. Chem. A 2014, 118, 2794−2800
The Journal of Physical Chemistry A
Article
Figure 1. Three possible models to explain the anomeric effect: (a) hyperconjugative interaction designated as an endoanomeric effect, (b) electrostatic interaction (dipole−dipole), and (c) hyperconjugative interaction designated as an exoanomeric effect.
reduced density gradient (RDG) topological analyses,42 and the theoretical results were compared with experimental IR spectroscopic data for 3-hydroxytetrahydropyran (1) (Figure 2).
■
COMPUTATIONAL DETAILS
In the search for the orientation of the O−H group relative to the tetrahydropyran ring for compound 1, the potential energy curves were scanned at the MP2/cc-pVDZ level by varying the H−O−C3−C2 dihedral angle from 0 to 360° in 10° steps (Figure 2) for the axial and equatorial conformations; for each step (fixed dihedral angle), the rest of the molecule was allowed to relax during geometries optimization. The geometries for each minimum in the curves were fully reoptimized and the frequencies were calculated, with zero-point energy (ZPE) corrections, using the MP2 theory and aug-cc-pVTZ and 6311++g(3df,3pd) basis sets (Table 1). A geometry optimization and single-point energy calculations at the MP2/aug-cc-pVQZ and CCSD/aug-cc-pVTZ level also have been performed, respectively, for the most stable conformers using the Gaussian 09 program.43 Theoretical values for the 3JHH coupling constants were calculated at the SOPPA(CCSD)44−46 level using optimized geometries with the Dalton 2.0 program47 employing the EPRIII48 basis set. Hyperconjugative interactions were evaluated using natural bond orbital (NBO 6.0) 49 analysis, as implemented in Gaussian 09, and the calculations were performed at the B3LYP/cc-pVTZ level. QTAIM and NCI topological analyses were performed using the resulting wave functions obtained from the MP2/aug-cc-pVTZ optimizations. QTAIM and NCI topological analyses were carried out with the AIMALL50 and NCIPLOT51 programs, respectively.
Figure 2. Equatorial and axial conformations for 3-hydroxytetrahydropyran (1).
hydroxytetrahydropyran using infrared (IR) spectroscopy. Two absorption frequencies for the hydroxyl group have been observed in CCl4 solution, one of which was assigned to an intramolecular hydrogen bond and the other to the free hydroxyl group. The presence of both free and bonded hydroxyl groups in carbon tetrachloride solutions of 3hydroxytetrahydropyran was interpreted as indicating the presence of an equilibrium between equatorial and axial conformations (Figure 2). However, the IR spectrum reported in the literature is of insufficient resolution and was only obtained in CCl 4 solution.36 Thus, the conformational preference of 3-hydroxytetrahydropyran in the isolated phase and in nonpolar and polar solvents remains unclear, inspiring us to conduct further studies to evaluate the conformational preference of this compound using nuclear magnetic resonance (NMR) and IR spectroscopy supported by theoretical calculations. To achieve this objective, experimental and theoretical 3JHH spin−spin nuclear coupling constants were used in the analysis of the conformational equilibrium of this molecule. The experimental data were supported by ab initio calculations using MP237,38 and CCSD39 approximations. The intramolecular hydrogen bond was evaluated using natural bond orbital (NBO) population,40 quantum theory of atoms in molecules (QTAIM),41 noncovalent interactions (NCI), and
Table 1. Energy (au), Energy Difference (kcal mol−1), Dipole Moment (Debye), and Theoretical Infrared Frequency (cm−1) for O−H Group for the Most Stable Hydroxyl Group Orientation for Compound 1 in the Axial and Equatorial Orientations Calculated Using MP2 Approach and Two Different Basis Sets basis set
a
aug-cc-pVTZ
6-311++g(3df,3pd)
conformer
energy
ΔE
μ
νO−H
Ia IIa IIIa Ie IIe IIIe
−346.208264 −346.204869 −346.208261 −346.207022 −346.206833 −346.206600
0.0 2.1 0.0 0.8 0.9 1.0
2.38 3.23 2.38 0.40 2.58 2.49
3783 3824 3783 3822 3827 3812
a
energy
ΔE
μ
νO−H
−346.340917 −346.337025 −346.340917 −346.339143 −346.338839 −346.338694
0.0 2.4 0.0 1.1 1.3 1.4
2.35 3.21 2.35 0.39 2.57 2.48
3828 3872 3828 3869 3878 3861
Experimental infrared frequencies for O−H group are: 3625, 3595, and 3477 cm−1. 2795
dx.doi.org/10.1021/jp500211y | J. Phys. Chem. A 2014, 118, 2794−2800
The Journal of Physical Chemistry A
Article
Figure 3. Potential energy curves for 3-hydroxytetrahydropyran (1) calculated at the MP2/cc-pVDZ level for the isolated molecule and including solvent (benzene and DMSO) effects: (a) for the axial conformation, highlighting three energy minima at Ia, IIa, and IIIa and (b) for the equatorial conformation, highlighting three energy minima at Ie, IIe, and IIIe.
■
EXPERIMENTAL SECTION Nuclear Magnetic Resonance Experiments. The solvents were commercially available and used without further purification. 1H NMR spectra were recorded on a spectrometer operating at 600 MHz for 1H. Measurements were carried out at a probe temperature of 25 °C using solutions of ca. 10 mg cm−3 in different solvents. The 1H spectra were based on TMS reference. Typical conditions for the 1H spectra were 16 transients, a spectral width of 5.4 kHz, and 64k data points, giving an acquisition time of 6.1 s and zero-filled to 128k to give a digital resolution of 0.08 Hz/point. Compound 1 was fully characterized using 1D 1H and 13C spectra and 2D COSY and HSQC contour plots. Infrared Experiments. The IR spectra for compound 1 were recorded on a Shimadzu FT-IR Prestige 21 spectrometer with 1 cm−1 resolution and 32 scans at a concentration of 0.01, 0.02, and 0.04 mol/L in carbon tetrachloride solution using a 1.00 cm quartz cell for the hydroxyl group region (3700 to 3200 cm−1). Syntheses: 3-Hydroxytetrahydropyran (1).52 A solution of 3,4-dihydro-2H-pyran (9.1 g, 0.1mol) in anhydrous THF (80 mL) was cooled to 0 °C, and 50 mL of 1 mol/L of a solution of borane in THF was added dropwise. The reaction was stirred at 0 °C for 30 min and then for 3 h until reaching room temperature. After this period, the reaction mixture was oxidized with 33 mL of 3 mol/L NaOH solution and 16 mL of 30% hydrogen peroxide. The THF was removed under reduced pressure, and 5 g of sodium chloride was added to the aqueous solution, which was extracted with diethyl ether and dried over sodium sulfate. The solvent was removed under reduced pressure, and the desired product was distilled (80−84 °C/8 mmHg), yielding 5.59 g (45% yield) of pure 3hydroxytetrahydropyran.
coupling constants for individual conformers weighted by its molar fraction. The experimental values previously listed show that for compound 1 the 3JH2H3 coupling increases when solvent polarity increases, suggesting that the equilibrium is shifted toward the equatorial conformer (Figure 2), which presents a higher vicinal 3JH2aH3a coupling constant. In nonpolar solvent (CDCl3), the experimental coupling constant value (3JH2H3 = 5.88 Hz) suggests a preference for the axial conformer, which presents a lower 3JH2eH3e coupling constant. The experimental data were corroborated by theoretical calculations performed to search for the most stable conformers of the isolated molecule (gas phase) and the including solvent effects. As previously mentioned, the hydroxyl group for compound 1 can adopt either an axial or an equatorial orientation, in principle, but there is no information about the energetic differences between them or even the orientation of the hydroxyl hydrogen in relation to the six-membered ring. To evaluate the most stable arrangement for OH substituents, the dihedral angle H−O−C3−C2 (Figure 2) for compound 1 was scanned from 0 to 360° (Figure 3) for the substituent in the axial and equatorial orientations. The calculations were performed at the MP2/cc-pVDZ level in the isolated phase and included solvent effect using the SMD model.53 As can be observed in Figure 3, there are three minima for the hydroxyl hydrogen orientation for each conformation, which differ in stability and the interconversion barrier. For the axial conformer (Figure 3a), the most stable conformation in the isolated molecule and when the solvent effect was added is conformation Ia (Figure 4), followed by conformation IIa. The minimum for conformation IIIa was not well characterized in the potential energy curve (Figure 3a). For the equatorial conformer (Figure 3b), similarly to the axial, there are three minima for the hydroxyl hydrogen orientation, but in this case, all of them have almost equal energies. In addition to the energy difference between hydroxyl hydrogen orientations, the most important is the difference in the interconversion barrier (Figure 3a) between Ia to IIa, which is ∼4.2 kcal mol−1for the axial conformer but
