Intermolecular interactions in a phenol-substituted benzimidazole

The solid-state structure of 2-(4-hydroxyphenyl)-1-[(4-hydroxyphenyl)methyl]-5,6-dimethyl-1H-benzimidazole acetone disolvate exhibits O—H⋯N hydrogen bonds between benzimidazole units and O—H⋯O hydrogen bonds with one of the acetone solvate molecules as the acceptor. Density functional theory is used to estimate the strength of the interactions.


Chemical context
The formation of a gel rather than a crystalline solid depends on the ability of the dissolved gelator to self-assemble into a three-dimensional network structure incorporating the solvent via non-covalent interactions rather than self-assembly followed by crystallization. The study of the gelation properties of small organic compounds (organogelators) is of importance in soft-matter research because of possible biomedical applications (Lau & Kiick, 2015;Huynh et al., 2011;Ye et al., 2014), including potential use in tissue engineering (Xavier et al., 2015;Yan et al., 2015), drug delivery and diagnostics (Wu & Wang, 2016;Tibbitt et al., 2016), and medical implants (Liow et al., 2016;Yasmeen et al., 2014).
Our efforts in this area include the preparation, structural characterization and exploration of the intermolecular interactions in long-chain ester-substituted biphenyl derivatives (Geiger, Geiger, Moore et al., 2017;Geiger, Geiger, Roberts et al., 2018) and phenyphenol derivatives . We have also reported a novel long-chain estersubstituted benzimidazole gelator (Geiger, Zick et al., 2017).
In our continuing efforts to exploit benzimidazole as a gelator core, we synthesized 1-(4-hydroxybenzyl)-2-(4-hydroxyphenyl)-5,6-dimethyl-1H-benzimidazole in the hope of using it as a starting material to prepare derivatives with a propensity for gelation. This compound was isolated as the diacetone solvate, (1), and we report herein its structural char- ISSN 2056-9890 acterization and an exploration of its three-dimensional superstructure, including an examination of hydrogen-bond strengths.

Structural commentary
A view of the molecular structure of (1) with the atom-labeling scheme employed is seen in Fig. 1. The bond lengths and angles are all within the range reported for similar disubstituted benzimidazole derivatives (c.f. Geiger & DeStefano, 2016). The benzimidazole moiety is planar with the largest deviation for C7 [0.0344 (13) Å ]. The 2-(4-hydroxyphenyl) substituent is canted at an angle of 44.18 (7) from the benzimidazole plane and the N2-C7-C8-C13 torsion angle is À43.7 (2) .
In addition to the benzimidazole, the asymmetric unit of (1) contains two acetone molecules, one of which uses its carbonyl oxygen atom as acceptor in an O-HÁ Á ÁO hydrogen bond (see Table 1). The hydrogen-bonded acetone molecule exhibits a slightly longer C-O bond distance than the other acetone molecule [1.212 (3) Å versus 1.192 (3) Å ]. This observation is consistent with previous results (Ichikawa, 1979).  Table 1 show the hydrogen-bonding network exhibited by (1). Each of the phenol groups behaves as a donor in a hydrogen bond. The 2-(4-hydroxyphenyl) substituent participates in an O-HÁ Á ÁO interaction with one of the acetone solvate molecules as the acceptor. The 1-(4-hydroxyphenyl)methyl substituent forms an O-HÁ Á ÁN hydrogen bond in which an adjacent benzimidazole moiety serves as the acceptor. The result is a chain structure that runs parallel to [010]. Fig. 3 shows the Hirshfeld surface and fingerprint plot for the disubstituted benzimidazole moiety. The prinicpal hydrogen-bonding interactions are clearly visible. The surface coverages corresponding to HÁ Á ÁO and HÁ Á ÁN interactions are 16.0% and 5.9%, respectively. There are no significantinteractions observed. The fingerprint plot does, however, reveal a weak C-HÁ Á Á interaction that involves C16-H16 with the 2-(4-hydroxyphenyl) substituent ring system and C13-H13 with the benzene ring of the benzimidazole moiety (see Table 1 and Fig. 4). These interactions are between molecules translated along the a axis. The surface coverage corresponding to HÁ Á ÁC interactions is 24.7%.

Supramolecular features
The interaction energies were calculated using density functional theory with the CE-B3LYP/6-31G(d,p) functional/ basis set combination (see Section 7 for details). The results of the calculations are reported in Table 2. As expected, the electrostatic component is the primary contributor to the traditional hydrogen-bonding interactions and the dispersive Table 1 Hydrogen-bond geometry (Å , ).

Figure 1
View of the molecular structure of (1) showing the atom-labeling scheme. Displacement ellipsoids for non-hydrogen atoms are drawn at the 30% probability level.

Figure 3
Hirshfeld surface (left) and fingerprint plot (right) for the benzimidazole moiety of (1). component dominates for the C-HÁ Á Á interactions. The C-HÁ Á Á interactions appear to reinforce each other with the sum of their contributions exceeding that of the traditional hydrogen-bond energies. The M06 suite of density functionals are reported to outperform B3LYP for dispersion and ionic hydrogenbonding interactions (Walker et al., 2013;Zhao & Truhlar, 2008) and so the M06-2X/6-31G(d,p) functional/basis set combination was also used to calculate the interaction energies. The results are found in Table 2. The value obtained using the M06-2X functional compares favorably with the CE-B3LYP functional result for the phenolÁ Á Áacetone hydrogen bond, but the values are decidedly less for the C-HÁ Á Á and the inter-benzimidazole O-HÁ Á ÁN hydrogen bonds. The calculations employing the M06-2X functional were performed in the gas phase; however, in the solid state, intermolecular interactions do not occur in isolation, which may account for the difference in results.
Energies are in kJ mol À1 and are corrected for BSSE.

Figure 4
Partial packing diagram of (1)  Single crystals of (1) were obtained by slow evaporation of a dilute acetone solution of the product.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 3. A refined extinction coefficient [0.006 (2)] was employed to calculate the correction factor applied to the structure-factor data. H atoms bonded to C were refined using a riding model with C-H = 0.95 Å for H bonded to aromatic C atoms, 0.99 Å for methylene H atoms, and 0.98 Å for the methyl H atoms. U iso (H) = kU eq (C), where k = 1.2 for H atoms bonded to aromatic and methylene C atoms and 1.5 for H atoms bonded to methyl C atoms. H atoms bonded to oxygen were refined freely, including isotropic displacement parameters.

Hirshfeld surface, fingerprint plots, interaction energy calculations
Hirshfeld surfaces, fingerprint plots, and interaction energies were calculated using CrystalExplorer17 (Turner et al., 2017), in which the C-H bond lengths were converted to normalized values based on neutron diffraction results (Allen et al., 2004). Interaction energies were calculated employing the CE-B3LYP/6-31G(d,p) functional/basis set combination and are corrected for basis set superposition energy (BSSE) using the counterpoise (CP) method (Boys & Bernardi, 1970). The interaction energy is broken down as where the k values are scale factors, E 0 ele represents the electrostatic component, E 0 pol the polarization energy, E 0 dis the dispersion energy, and E 0 rep the exchange-repulsion energy (Turner et al., 2014;Mackenzie et al., 2017).
Interaction energy calculations were also performed on molecules in the gas phase using SPARTAN'16 (Wavefunction, 2016). DFT calculations using the M06-2X (Zhao & Truhlar, 2008) functional with a 6-31G(d,p) basis set were employed for the determination of interaction energies, which were corrected for BSSE employing the CP method (Boys & Bernardi, 1970). Atomic coordinates obtained from the crystallographic analysis were used for all non-H atoms. Because bond lengths obtained for H atoms from X-ray crystallographic analyses are unreliable, the positions of the H atoms were optimized to their energy minima using the M06-2X/6-31G(d,p) functional/basis set combination.

Funding information
This work was supported by a Congressionally directed grant from the US Department of Education for the X-ray diffractometer (award No. P116Z100020) and a grant from the Geneseo Foundation.

Special details
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.