A combined experimental and theoretical study on vibrational and electronic properties of (5-methoxy-1H-indol-1-yl)(5-methoxy-1H-indol-2-yl)methanone

2.1 General

Melting point was recorded on a Gallenkamp melting point instrument and is uncorrected. The FT-IR spectrum was recorded on a Perkin-Elmer 180 spectrometer in the range of 4000–50 cm⁻¹, with a spectral resolution of ±2 cm⁻¹. FT-Raman spectrum was recorded using the same instrument equipped with a FRA 106 Raman module (Nd:YAG laser source operating in the region 4000–100 cm⁻¹ at 1.064 μm line width, 200 mW power). Spectra were recorded with scanning speed of 30 cm⁻¹⋅min⁻¹ and spectral width of 2 cm⁻¹. The frequencies of all sharp bands are accurate to ± 1 cm⁻¹.

2.2 Synthesis

A suspension of (5-methoxy-2,3-dihydro-1-H-indol-1-yl) (5-methoxy-1H-indol-2-yl)methanone (0.50 mmol) and 2,3-dichloro-5,6-dicyanobenzo quinone (0.55 mmol) in ethyl acetate (25 mL) was stirred under reflux for 18 h. The reaction mixture was concentrated under reduced pressure and the precipitate collected by filtration. Column chromatography using silica gel and chloroform/methanol/ammonia (10/1:0/1) as an eluent furnished pure MIMIM (m.p. 178–179 °C) in almost quantitative yield. Spectroscopic data of MIMIM are consistent with those previously reported [13].

2.3 Quantum chemical calculations

Gaussian 03 W program was used to perform the entire quantum chemical calculations at DFT B3LYP and PBEPBE levels with 6-311++G (d, p) basis set [14]. No imaginary frequency modes were obtained at the optimized geometry of the title molecule, so that a true minimum was found on the potential energy surface. As a result, the unscaled calculated frequencies, reduced masses, force constants, infrared intensities, Raman activities and depolarization ratios were obtained. The calculated harmonic frequencies were scaled down to improve the calculated values. Wavenumbers calculated at PBEPBE level were scaled by 0.9067; wavenumbers calculated at B3LYP level were scaled by 0.958 (above 1700 cm⁻¹) or by 0.983 (below 1700 cm⁻¹) [15, 16]. After scaling, the deviation from experimental values was lower than 10 cm⁻¹, with few exceptions. The assignment of calculated normal modes was made based on the corresponding potential energy distributions (PEDs). PEDs were computed from vibrational frequencies using VEDA program [17]. Gaussview program was used to obtain visual animation and verification of the normal assignment modes [18].

The UV-Vis absorption spectrum of the molecule was calculated by time-dependent DFT (TD-DFT) at PBEPBE/6-311++G (d, p) level of theory. The solvent effect on the UV-Vis absorption spectrum of the molecule was also examined by applying the integral equation formalism-polarized continuum model (IEF-PCM). ¹H and ¹³C NMR chemical shifts of the title compound in CDCl₃ were calculated using the gauge-independent atomic orbital (GIAO) method. Ethical approval: The conducted research is not related to either human or animals use.

3.1 Molecular geometry

Molecular geometry was optimized using PBEPBE method with the atom numbering shown in Figure 1. By allowing the relaxation of all parameters, calculations converged to optimized geometries that correspond to true energy minima, as revealed by the lack of imaginary frequencies in the vibrational mode.

Figure 1

Optimized geometric structure and atom numbering of MIMIM (C₁₉H₁₆N₂O₃). Atom colors: carbon (violet), hydrogen (green), oxygen (red), nitrogen (blue).

3.2 Potential energy scan

Potential energy scan, i.e., the relationship between potential energy and molecular geometry, is a powerful approach to understanding molecular properties. The energy of MIMIM conformers was calculated by the AM1 theory. Two potential energy scans were carried out with the dihedral angles C15–N11–C10–C8 and C18–C17–O23–C24, corresponding to the link of the phenyl ring to the amide and the methoxy moieties, respectively. During scan, all the geometrical parameters were simultaneously relaxed and varied in steps of 30° from 0 to 360°. The potential energy was calculated for 121 conformers and the curve of potential energy as a function of dihedral angle is shown in Figure 2. Both DFT methods essentially provided the same value of global minimum energy (−1068 a.u. for PBEPBE and −1069 a.u. for B3LYP). The most stable conformer, corresponding to a global minimum energy, has two methoxy groups lying on the same plane.

Figure 2

The potential energy scan of the MIMIM molecule.

The optimized bond lengths and bond angles (°) of the most stable conformer were calculated using PBEPBE and B3LYP methods; selected data are listed in Table S1. The optimized geometrical parameters of MIMIM were compared with the experimental values of its crystal structure [19].

The relation between experimental and theoretical values of bond lengths and bond angles of MIMIM are described by the following equations (see Figure S1 for a graphic representation):

d_cal. = 0.39059d_exp. + 0.3648 (R² = 0.99313) for bond lengths by PBEPBE method

δ_cal. = −0.21469δ_exp. +1.00311 (R² = 0.96407) for bond angles by PBEPBE method

d_cal. = 0.73235d_exp. + 0.37812 (R² = 0.99256) for bond lengths by B3LYP method

δ_cal. = 1.01124 δ_exp. −1.20579 (R² = 0.95927) for bond angles by B3LYP method

Good linearity (R² >0.99) was observed between calculated and experimental values of bond lengths using both PBEPBE and B3LYP methods. C–H bond lengths predicted by B3LYP method are systematically too long: average C–H bond length in the MIMIM crystal is 0.96 Å, whereas theoretically calculated values are all >1 Å. The deviation from experimental values may arise from low scattering of hydrogen atoms in the X-ray diffraction experiment. Both B3LYP and PBEPBE methods predict C–N bond lengths in the range of experimental values (1.37–1.41, 1.38–1.41 and 1.36–1.41 Å, respectively). Good agreement is also observed between calculated (B3LYP and PBEPBE) and experimental C–C bond lengths (1.38–1.43, 1.39–1.43 and 1.37–1.42 Å).

Only modest linearity (R² >0.95) was obtained between calculated and experimental bond angles. Some experimental values were closer to those calculated by B3LYP method, others were better predicted by PBEPBE method (Table S1). Overall, calculated values showed no significant variation from the experimental ones, except for two bond angles (C3–C4–N7 and C19–C15–N11).

3.3 Natural bond orbital analysis

Natural bond orbital analysis was performed to elucidate hyper-conjugation, re-hybridization and delocalization of electron density within the molecule. Analysis highlighted the following intramolecular interactions and energetic contributions: (a) Conjugation of π(C1–C6) to π*(C2–C3) and π*(C4–C5) (17.5 and 18.0 kcal/mol, respectively); (b) Hyper-conjugation of σ(C1–O21) to σ*(C1–C2), (C1–C6), (C2–C3) and (C5–C6) (1.40, 1.42, 1.44 and 1.42 kcal/mol, respectively); (c) Hyper-conjugation of σ(C4–N7) to σ*(C2–C3), (C3–C4), (C4–C5), (N7–C8) and (N7–H28) (1.35, 1.34, 1.33, 1.16 and 1.17 kcal/mol, respectively); (d) Delocalization of the lone pair (LP) of N11 to C10–O20 (0.89 kcal/mol); (e) Delocalization of LP of O20 to C8–C10 and C10–N11 (1.10 and 1.02 kcal/mol, respectively); (f) Delocalization of LP of O21 to C1–C2, C1–C6, C22–H35 and C22–H36 (1.08, 1.10, 0.93 and 0.92 kcal/mol, respectively); and (g) LP of O23 to C16–C17, C17–C18 and C24–H38 (1.09, 1.10 and 0.92 kcal/mol, respectively).

3.4 Mulliken charge distribution

Mulliken population analysis was carried out by PBEPBE and B3LYP methods to estimate partial atomic charges of the title molecule (Table 1 and Figure 3). The two methods provided partially contrasting results, as they predict charges of opposite sign on most aromatic carbons. However, according to both methods, most negative charge is localized on the electronegative oxygen and nitrogen atoms; little negative charge is also located on the methoxy carbons as well as the aromatic carbons C12 and C18. Conversely, positive charge is distributed over hydrogen atoms (charge = 0.14–0.17 by PBEPBE method and 0.08–0.28 by B3LYP method).

Figure 3

Mulliken charge distribution of MIMIM calculated by B3LYP and PBEPBE methods.

3.5 Frontier molecular orbital analysis

Analysis of frontier molecular orbitals is crucial to understand the physicochemical properties of molecules, because the energy gap between HOMO and LUMO determines the kinetic stability, chemical reactivity, optical polarizability and chemical hardness–softness of the molecule [20][21, 22].

The energy and shape of MIMIM molecular orbitals HOMO–1, HOMO, LUMO and LUMO+1 was calculated using B3LYP method (Table 2 and Figure 4). It is clear from Figure 4 that HOMO is delocalized over almost the whole molecule, whereas LUMO is mainly localized on the rings. The energy difference between HOMO and LUMO in several solvents (acetonitrile, chloroform, cyclohexane, DMSO, ethanol and methanol) varies between 2.55 and 2.60 eV, highlighting a limited solvent effect on molecular energetic profile.

Figure 4

Molecular orbitals and energies of MIMIM (PBEPBE method).

Chemical hardness and electronegativity of MIMIM are 1.27–1.30 eV and 3.64–4.05 eV, respectively.

Dipole moment, being related to the charge distribution within a molecule, determines the intensity of intermolecular interactions, and is therefore an important molecular property. The highest values (>4.0 D) of dipole moment of MIMIM were obtained in highly polar or strong H-bond donor solvents (acetonitrile, DMSO, ethanol and methanol), which are able to stabilize MIMIM charge separation by forming strong intermolecular bonds.

In order to evaluate the energetic behavior of the title compound, we calculated the excitation energies (E), absorption wavelengths (λ) and oscillator strengths (f) in acetonitrile, chloroform, cyclohexane, DMSO, ethanol and methanol (Table 3). Calculations predicted maximum intensities of electronic transitions at 464, 428, 417, 344 and 327 nm in cyclohexane.

3.6 Molecular electrostatic potential and total electron density

The molecular electrostatic potential (MEP) map is a graphic representation of the electrostatic potential of molecules as a function of their constant electron density (ED). MEP illustrates the charge distribution of molecules three-dimensionally in terms of color grading: red < orange < yellow < green < blue. Red color indicates negatively charged areas (with a nucleophilic character), whereas blue color indicates positively charged areas (electrophilic in nature). The negative regions are usually associated with lone pair of electrons on electronegative atoms; positive regions are those having electropositive atoms such as hydrogen atoms. MEP diagram helps understand the relationship between molecular structure and physicochemical properties. MEP and ED of MIMIM are illustrated in Figure 5. The MEP values of MIMIM are in the range of +0.0467 to −0.0467 a.u. The most electronrich – and hence nucleophilic – regions in the molecule correspond to the oxygen atoms of the methoxy and carbonyl groups.

Figure 5

Molecular electrostatic potential (MEP) map and electron density (ED) of MIMIM.

3.7 Vibrational analysis

The maximum number of potentially active (i.e., observable) fundamental vibrations of a non-linear molecule containing N atoms is 3N–6, apart from three translational and three rotational degrees of freedom. MIMIM has 40 atoms and therefore 114 normal modes of vibration.

All vibrations are active in Raman and infrared absorptions. The detailed vibrational assignment of experimental wavenumbers is based on normal mode analysis and comparison with theoretically scaled wavenumbers with PED by B3LYP and PBEPBE methods. Experimental and simulated (PBEPBE method) infrared and Raman spectra of MIMIM are shown in Figures 6 and 7, respectively. Simulations by B3LYP method are reported in Figures S2 and S3. The observed and scaled theoretical frequencies using PBEPBE and B3LYP methods with 6-311++G (d, p) basis set along with their PEDs are listed in Table S3. Generally, PBEPBE was more accurate than B3LYP in the prediction of IR and Raman wavenumbers.

Figure 6

Experimental and simulated (PBEPBE) FT-IR spectra of MIMIM.

Figure 7

Experimental and simulated (PBEPBE) FR-Raman spectra of MIMIM.

3.8 NMR analysis

¹H and ¹³C NMR chemical shifts of MIMIM were calculated and compared with experimental ones (Table 4). Experimental and theoretical chemical shift values were in good agreement, with a maximum discrepancy of 4.7 ppm for C1 in ¹³C NMR spectrum and 1.21 ppm for H28 in ¹H NMR spectrum.

In ¹H NMR spectrum, the most deshielded protons are those belonging to the methoxy groups, due to the electron-withdrawing effect of the oxygen atom. The aromatic ring protons were experimentally observed in the range of 6.99–8.37 ppm, with a substantial overlap to the predicted range (6.59–8.24 ppm).

In ¹³C NMR spectrum, aromatic carbons were observed in the expected range (100–160 ppm) in two regions: 102–117 ppm (C2, C3, C6, C16, C18 and C19) and 127–156 ppm (C1, C4, C5, C14, C15 and C17). Aromatic carbons attached to oxygen (C1 and C17) were the most deshielded, whereas carbons adjacent to nitrogen (C4, C8, C12 and C15) were the most shielded.