Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

Investigation of solvent polarity effect on molecular structure and vibrational spectrum of xanthine with the aid of quantum chemical computations Turgay Polat a,⇑, Gurcan Yıldırım b a b

Kastamonu University, Department of Physics, Kastamonu 37100, Turkey Abant Izzet Baysal University, Department of Mechanical Engineering, Bolu 14280, Turkey

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Strong intra-molecular charge

transfer in the xanthine molecule.  Role of different medium polarity on

the geometric structure and vibrational spectra.  Determination of the assignments of fundamental vibrational modes belonging to the title molecule.  Improvement of the correlation between the theoretical and experimental geometric structure.

a r t i c l e

i n f o

Article history: Received 16 September 2013 Received in revised form 29 November 2013 Accepted 4 December 2013 Available online 17 December 2013 Keywords: Xanthine Tautomerism Density functional theory Solvent effect Infrared spectroscopy Hydrogen bonding

a b s t r a c t The main scope of this study is to determine the effects of 8 solvents on the geometric structure and vibrational spectra of the title compound, xanthine, by means of the DFT/B3LYP level of theory in the combination with the polarizable conductor continuum model (CPCM) for the first time. After determination of the most-steady state (favored structure) of the xanthine molecule, the role of the solvent polarity on the SCF energy (for the molecule stability), atomic charges (for charge distribution) and dipole moments (for molecular charge transfer) belonging to tautomer is discussed in detail. The results obtained indicate not only the presence of the hydrogen bonding and strong intra-molecular charge transfer (ICT) in the compound but the increment of the molecule stability with the solvent polarity, as well. Moreover, it is noted that the optimized geometric parameters and the theoretical vibrational frequencies are in good agreement with the available experimental results found in the literature. In fact, the correlations between the experimental and theoretical findings for the molecular structures improve with the enhancement of the solvent polarity. At the same time, the dimer forms of the xanthine compound are simulated to describe the effect of intermolecular hydrogen bonding on the molecular geometry and vibrational frequencies. It is found that the C@O and NAH stretching vibrations shift regularly to lower frequency value with higher IR intensity as the dielectric medium enhances systematically due to the intermolecular NAH  O hydrogen bonds. Theoretical vibrational spectra are also assigned based on the potential energy distribution (PED) using the VEDA 4 program. Ó 2013 Elsevier B.V. All rights reserved.

Introduction ⇑ Corresponding author. Tel.: +90 366 280 19 30; fax: +90 366 215 49 69. E-mail address: [email protected] (T. Polat). 1386-1425/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.saa.2013.12.035

Xanthines, being a class of purine bases, are common agrochemical and therapeutic agents, and easily oxidized to uric acid

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

and methyl uric acid [1]. Thus, they have pharmacologic properties stimulating the central nervous system and relax smooth muscles. Moreover, the xanthine compounds can be found in most human body tissues, fluids and other organisms. The unchecked level of xanthine in our human body may lead to kidney stone formation, urinary tract disease and muscle diseases. The treatments can be achieved avoiding the foods and drinks containing xanthine and its derivatives (coffee, tea and colas) [2], because a number of mild stimulants are derived from the xanthines such as caffeine, theophylline and theobromine. Even, the former (caffeine) is the most well-known natural psychoactive substance [3]. Xanthine derivatives are also occasionally found as the components of some pharmaceutical preparations such as formulations being used to alleviate neonatal apnea; or to treat the asthmatic manifestations and chronic obstructive pulmonary disease [4]. In the literature, the researchers have been interested in both experimental and theoretical studies on the xanthine and its derivatives for several years [5–13]. For example; Ucun et al. have calculated the molecular structures, vibrational frequencies and corresponding vibrational assignments of xanthine and its methyl derivatives [14]. In 2011, the vibrational wavenumbers, geometrical parameters, modes of vibrations and other thermodynamic parameters of xanthine molecule were investigated by the scientists, Arivazhagan and Jeyavijayan [2]. To the best of our knowledge, there have been no theoretical DFT calculations including the solvent effects on the molecular structure and vibrational spectra of xanthine and its dimers. Furthermore, solvent-induced vibrational frequency shifts have attracted interest for many years owing to the existence of significant information on chemical bonding and solute–solvent interactions. As well known, the intra-molecular frequency shifts are determined by the normal coordinate-dependent parts of the attractive and repulsive interactions between solute and solvent molecules [15]. At the same time, the tautomeric equilibria affecting the chemical and biological properties of the organic molecules are significantly sensitive to the environmental influences [16–20]. In the exhaustive work, we survey not only the tautomerism mechanism but also the solvent effects on the optimized molecular structure and vibrational spectra of the xanthene compound in detail. The changes of dipole moments and charges on atoms are also investigated both in the gaseous phase and in different solvents. The dimeric forms of xanthine are also analyzed to deduce the effect of intermolecular hydrogen bonding on the molecular geometry and vibrational frequencies. Further, the vibrational modes are assigned based on the potential energy distribution (PED) using the VEDA 4 program. Computational details Density functional theory (DFT) calculations lead to the prediction of more accurate molecular structure and vibrational frequencies than the conventional ab initio Hartree-Fock calculations [21–24]. The B3LYP method based on Becke’s three parameter hybrids functional combined with the Lee–Yang–Parr correlation functional (B3LYP) of DFT yields a good definition of harmonic vibrational frequencies for small and medium sized molecules [25–27]. All the calculations are performed using Gaussian 09 program [28] with GaussView 5 molecular visualization program package [29]. The optimized structure parameters and vibrational frequencies are calculated by using B3LYP method at 6-311++G (d,p) basis set. The values of the natural bond orbital (NBO) [30] atomic charges on the atoms and dipole moments are obtained with the B3LYP method in the Gaussian 09 package. The nonspecific solvent effects of the solvent medium are studied by means of the conductor-polarizable continuum model (CPCM) [31]. Force field scaling is exerted via the selective scaling in the natural internal coordinate representation based on SQM procedure [32–35]. Besides, VEDA-4 program [36] is preferred to determine

99

the transformation of the force field and subsequent normal coordinate analysis including the least square refinement of the scaling factors, calculations of the potential energy distribution (PED) and the prediction of IR and Raman intensities. At the same time, Raman activities (AR) are converted to the relative Raman intensities (IIR) by means of the following equation inferred from the basic theory Raman scattering [37,38],

Ii ¼

f ðm0  mi Þ4 Si mi ½1  expðhcmi =KTÞ

where h, c and k present the universal constants and f is properly chosen as common normalization factor for the intensities when m0 (cm1) denotes the exciting frequency and mi depicts the vibrational wave number of the normal mode. Results and discussion Energy analysis and tautomeric stability As well known, the different molecular structures belonging to a compound affect seriously its physical and chemical properties [39]. In the comprehensive study, the molecular energies of 28 tautomeric and conformer forms of the xanthine molecule are computed by means of the B3LYP/6-311++G (d,p) calculation level to find the favored conformation (most steady state) of the title compound as given in Fig. 1. The characteristic parameters such as the total energy (in Hartrees), ZPE corrected energy, relative energy (kcal-mol1), dipole moment (Debye) and imaginary frequency values are listed in Table 1. Of the 28 structures for xanthine molecule, the lowest energy value of EHartree = -562.607505 a.u. in gas phase is attributed to the tautomer-10. In forthcoming sections, the theoretical values belonging to the favored tautomer-10 will thus be taken for correlation and further discussions. Fig. 2 shows the molecular structure along with the atom-numbering scheme for the tautomer studied. Fig. 3 shows the calculated potential energy curves for the tautomers and conformations of xanthine. The shape of the potential energy as a function of the hydrogen atom transferring is illustrated in Fig. 3. Generally, the potential energy curve of a molecule is plotted as a function of the dihedral angle from 0° to 360° using a step of 10°. In this work, the main difference between the most stable two tautomers is a hydrogen atom transferring from N12 atom to N11 one, which increases the total energy value. Therefore, in order to investigate the most stable tautomer of xanthine, potential energy scans are performed without the dihedral angle. It is to be mentioned here that the differences between the energy parameters of the most stable two tautomers (tautomer-10 and tautomer-1) are found to be about 9.12 kcal mol1 for the total energy and 8.60 kcal mol1 for zero point corrected energy, respectively. The energy difference stems from a hydrogen atom transferring from N12 atom to N11 one. The main difference between the most stable and less stable tautomers is 0.058333 Hartree. At first glance, the main reason of the energy distribution seems to be the hydrogen atom orientation relative to the nearest group. For this reason, we investigated whether the most stable tautomer-10 result from formation of H  O contact (intramolecular hydrogen bonding-like interaction) or not. Because the calculated H  O distances are approximately 2.4 Å, we have concluded that hydrogen bonding-like interactions does not responsible for stability of tautomer-10 but hydrogen atom transferring and long-range electrostatic interactions can be responsible for it due to the fact. The dipole moment values are also different from each other as a consequence of the change of the charge distribution in the molecule is another interesting point. The dipole moment values of each tautomer studied in this work are tabulated in Table 1. It is

100

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

Fig. 1. Tautomeric and conformational forms of xanthine.

apparent from the table that the most stable structure (tautomer10) obtains a dipole moment value of 4.4354 D whereas the value of 7.4686 D is calculated for the unfavorable structure (tautomer1) at B3LYP/6-311++G(d,p) level of theory. On the other hand, in case of structure 14 the highest dipole moment is computed to be about 9.9759 D while the lowest dipolar one (0.6115 D) is attributed to the structure -13. As for the imaginary frequency, the absence of negative frequencies confirms that the stationary points are in accordance to the global minima on the potential energy surface (PES) for structures calculated [40,41]. Moreover,

no important change in the zero point corrections is observed for all the structures studied, nor does the stability order. The long and short of it is that the tautomer-10 obtains the most stable structure as compared to the other structures (Table 1). After this part of the paper, we are interested in the characterization of only tautomer-10. Solvent effect The solvent environment plays a very important role on the structure and function of the compounds [25]. Conductor-

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

101

Fig. 1 (continued)

Fig. 2. Molecular structure and atom numbering scheme adopted in this study for tautomer-10.

polarizable continuum (CPCM) and Onsager calculation models are two useful methods to investigate the environmental effects on the title compounds [31]. The former model where the cavity covering the compound is thought to be the union of a series of interlocking atomic spheres deals with a full quantum dynamics simulation when making the dipole moment computations. As a result, the CPCM induces free energy contributions from the interaction of the cavity and dispersion–repulsion [42,43]. As for the latter approach, the molecule is thought to place in a spherical cavity inside the homogenous polarizable medium of dielectric constant. Hence, a dipole moment of the opposite direction in the surrounding medium is induced by the molecule dipole as a consequence of the charge distribution in the solvent [44,45]. Based on the evidences given above, the former (CPCM) method is expected to provide a

more stable structure as compared to the latter (Onsager’s) model. Thereby, in the present study, the solvent effects on the self-consistent field (SCF) energies and dipole moments of the xanthine compound are examined at the B3LYP/6-311++G(d,p) calculation level by means of the CPCM model. The results obtained are numerically given in Table 2. It is obvious from the table that the SCF energy values tend to systematically decrease with the enhancement of the dielectric constant from 1 to 78.36 (gas phase to water) due to the increment in the stability of the molecule studied. In other words, the more solvent polarity the environment obtains, the more solvent-molecule interaction appears in the environment [46,47]. Moreover, one can see from Table 2 how the dipole moment changes against the solvent polarity. According to the results obtained, the dipole moment value regularly increases from 4.4354 D to 5.9558 D with the enhancement in the polarity of the environment as a consequence of the polar nature (non-uniform distribution of charges) of the title compound studied in the current work [25]. Furthermore, the molecular geometry findings (the bond lengths, bond angles and dihedrals) belonging to the xanthine molecule can be found in Table 3. It is visible from the table that all the bond distances are observed to be different from each other due to the presence of the electron engagements in the bond lengths and the conjugative effects in the title compound studied [48,49]. Thus, it is not wrong to say that the strong intra-molecular charge transfer (ICT) appears in the xanthine compound [50,51]. Similarly, the differentiation between the bond angle values in the present molecule is another evidence for the strong intra-molecular charge transfer in the molecule. Regardless, it is to be mentioned here that

102

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

-562,54

Theoretical Bond Lengths ( Α)

1.44

Total Energy (Hartree)

-562,55 -562,56 -562,57 -562,58 -562,59 -562,60

(a)

1.40 1.36 2

1.32

Gas Phase (R =0.96299) 2 Cyclohexane(R =0.96986) 2 Benzene(R =0.96948) 2 Toluene (R =0.96914) 2 Chloroform(R =0.97237) 2 Chlorobenzene(R =0.97237) 2 Dichloromethane(R =0.97379) 2 DMSO (R =0.97506) 2 Water (R =0.97431)

1.28 1.24

-562,61 1.20 1.20

-562,54

1.25

1.30

1.35

1.40

1.45

Experimental Bond Lengths (Α)

(b)

-562,56 0

Theoretical Bond Angles ( )

Total Energy (Hartree)

-562,55

-562,57 -562,58

Tautomer-10

-562,59 -562,60 -562,61

Fig. 3. Potential energy curves for tautomers and conformations of xanthine obtained at the B3LYP/6-311++G(d,p) level.

130 125 120 2

Gas Phase (R =0.98542) 2 Cyclohexane (R =0.98808) 2 Benzene (R =0.98848) 2 Toluene (R =0.98869) 2 Choloform (R =0.99013) 2 Chlorobenzene (R =0.99040) 2 Dichloromethane (R =0.99067) 2 DMSO (R =0.99091) 2 Water (R =0.99090)

115 110 105 105

110

115

120

125

130 0

Table 2 Calculated total energies and dipole moments of xanthine (tautomer 10) at B3LYP/6311++G(d,p) in the gas phase and solvents. tot

Medium

E

(hartree)

Gas phase Cyclohexane Benzene Toluene Chloroform Chlorobenzene Dichloromethane DMSO Water

562.607505 562.616998 562.618132 562.618529 562.623013 562.623833 562.625280 562.627398 562.627603

Dipole moment (Debye)

Dielectric constant (e)

4.4354 5.0894 5.1750 5.2052 5.5609 5.6291 5.7517 5.9375 5.9558

1 2.02 2.27 2.37 4.71 5.70 8.93 46.83 78.36

the experimental findings are in well agreement with the theoretical results apart from the very little differences [52]. This differentiation is related to the fact that the experimental findings belong to the solid phase of the title compound while the theoretical calculations are attributed to those of the gaseous and solution phase. Of the calculations, the maximum deviation is determined for the single bond length between C5 and N11 atoms due to the atomic positions in the molecule. As for the bond angle explorer, the largest differentiation between experimental and theoretical data is obtained for the N9AC4AN10 bond angle as a consequence of the different poison of the substituents, meaning that the bond angle of the N9AC4AN10 group might be more susceptible to intermolecular interactions as compared to those of the other groups [53,54]. So that we examine the effect of the solvent polarity on the molecular structure of the xanthine in more details, the correlation graphics between the experimental and calculated bond lengths and bond angles are displayed in Fig. 4a and b. Based on the findings obtained, the correlations of the bond lengths and bond angles increase with the enhancement of the dielectric

Experimental Bond Angles ( ) Fig. 4. Correlation graphics between the experimental and calculated optimized (a) bond lengths and (b) bond angles.

constant up to 46.83 after which both tend to decrease slightly. Thus, it would be more precise to say that the optimum molecular geometric values are obtained in the DMSO solution with the dielectric constant of 46.83. It is another interesting point deduced from Table 3 that the calculated dihedral angles remain unchanged in all the environment conditions as a consequence of the planarity of the xanthine molecule. In this part of the paper, we determine natural bond orbital (NBO) charges on the atoms using the Natural Population Analysis (NPA) for the equilibrium geometry of the xanthine compound in the different phases [30]. As well known, NBO analysis provides a convenient basis to investigate the effective atomic charges (the charge transfer) or conjugative interaction in molecular systems [55]. One can see the NBO charges calculated in the different phases of the xanthine molecule in Table 4. Before the discussion of the atomic charges, it is to be mentioned here that the charge distributions of the title compound xanthine alter significantly in the presence of a solvent reaction field. The similar results can be encountered in the literature [56–61]. In the xanthine molecule, all the magnitudes of nitrogen atoms are found to be negative value as a result of their high electronegativity. The magnitude belonging to the N12 atom tends to increase regularly as a consequence of the position in the molecule while the other atomic charges on the N9, N10 and N11 atoms decrease slightly with the increase of the solvent polarity due to the presence of the hydrogen bonding. Moreover, it is always agreed that among the nitrogen atomic charges the most negative value is obtained for the N9 atom at each computation performed in the gaseous and solution phases due to the existence of the resonance

103

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109 Table 4 NBO atomic charges of tautomer 10 based on B3LYP/6-311++G(d,p) method in the gas and solution phases.

a

Atoma

Gas phase (e = 1)

Cyclohexane (e = 2.02)

Benzene (e = 2.27)

Toluene (e = 2.37)

Chloroform (e = 4.71)

Chlorobenzene (e = 5.70)

Dichloromethane (e = 8.93)

DMSO (e = 46.83)

Water (e = 78.36)

C1 C2 C3 C4 C5 H6 H7 H8 N9 N10 N11 N12 O13 O14 H15

0.356 0.040 0.638 0.809 0.251 0.422 0.426 0.201 0.646 0.605 0.511 0.509 0.604 0.619 0.431

0.360 0.040 0.643 0.814 0.259 0.426 0.432 0.209 0.640 0.599 0.505 0.525 0.628 0.645 0.440

0.360 0.040 0.643 0.814 0.260 0.427 0.433 0.210 0.639 0.598 0.504 0.527 0.631 0.649 0.441

0.361 0.040 0.643 0.814 0.260 0.427 0.433 0.210 0.639 0.598 0.504 0.528 0.632 0.650 0.441

0.363 0.039 0.645 0.816 0.265 0.428 0.435 0.214 0.636 0.595 0.501 0.536 0.643 0.663 0.445

0.363 0.039 0.646 0.817 0.265 0.429 0.436 0.215 0.636 0.595 0.500 0.537 0.645 0.665 0.446

0.364 0.038 0.646 0.817 0.267 0.429 0.437 0.216 0.635 0.594 0.499 0.540 0.649 0.670 0.447

0.365 0.038 0.647 0.818 0.269 0.430 0.438 0.218 0.634 0.592 0.497 0.543 0.654 0.676 0.449

0.365 0.038 0.647 0.818 0.269 0.430 0.438 0.218 0.633 0.592 0.497 0.544 0.655 0.677 0.450

For numbering of atoms, see Fig. 2.

Table 5 Experimental IR, calculated frequencies, relative intensities and probable assignments of the tautomer 10. IIR intensities (km/mol); AR, Raman scattering activities (A4 =amu). Assignments, PEDb (%)

Frequencies (cm1) Experimentala FT-Raman

FT-IR

3137 (w)

3133 (ms)

2875 (ms)

2875 (w)

m(N10AH7) (55) + m(C5AH8) (32) m(O14@C4) (55) + m(O13@C3) (19) m(C4@N9) (47) + m(O14@C4) (22) m(C1@C2) (55)

2797 (ms) 1685 (ms) 1600 (ms) 1560 (ms)

2825 (ms) 1700 (ms) 1658 (vs) 1567 (ms)

1335 (vs)

1334 (ms)

m(C2AC3) (42) + m(N9AC3) (11) + m(N10AC4) (11) m(N12AC1) (20) + d(H8AC5@N12) (19) + d(N11AC5@N12) (10) m(N12AC1) (22) + m(N9AC3) (20) + m(N9AC4) (12) m(N11AC5) (56) + d(H15AN11AC2) (11) s(H8AC5@N12AC1) (74) + s(C1AN12@C5AN11) (13) d(N11AC5@N12) (58) m(N10AC4) (56) + d(C3AN9AC4) (10) x(C4AN9AC3@O13) (39) + x(C3AN9AC4AN10) (19) + s(C2@C1AN12@C5) (16) x(C4AN9AC3@O13) (34) + s(H6AN9AC3AC2) (22) + s(C2@C1AN12@C5) (22) x(C1AN12@C5) (55) x(C4AN9AC3@O13) (56) + x(N12AC1AN10AH7) (16) s(H15AN11AC2@C1) (88) s(H6AN9AC3AC2) (34) + s(C1AN12@C5AN11) (20) + x(C4AN9AC3@O13) (15) d(O14@C4AN10)(43) + d(N9AC4A N10)(10) + d(C3AN9AC4) (10) s(H7AN10AC1@C2) (40) + s(C1AN12@C5AN11) (22) d(N11AC2AC3) (29) + s(C1AN12@C5AN11) (19) + s(H7AN10AC1@C2) (18) + s(H15AN11AC2@C1) (17) + x(C1@C2AC3@O13) (14) m(N10AC1) (26) + m(N9AC3) (13) + d(C3AN9AC4) (13) d(N9AC4AN10) (43) + d(C4AN10AC1) (25) + d(O14@C4AN10) (18) d(C4AN10AC1) (38) + d(N9AC4AN10) (20) + m(N10AC4) (11) d(N10AC1AN12) (35) + s(C2@C1AN12@C5) (29) + x(N12AC1AN10AH7) (17) + s(C4AN10AC1@C2) (10) d(O14@C4AN10) (53) + m(N10AC1) (12) d(O13@C3AN9) (29) + d(N11AC5@N12) (25) + d(C4AN10AC1) (21) s(N9AC4AN10AC1) (76) s(C3AN9AC4AN10) (70) + x(C1@C2AC3@O13) (16) s(N9AC4AN10AC1) (71) + x(N12AC1AN10AH7) (15) Abbreviations used: m: stretching; d: bending; s: torsion; x: out of bending. a Taken from Ref. [70]. b PED less than 10% are not shown.

1268 (ms)

1258 (w)

1212 (ms)/ 1220 (ms) 1132 (w)

1215 (w)/ 1205 (w) 1120 (w)

1040 (w)

1032 (w) 958

853

547 (ms)

538 (s) 503

411 (w) 175 (ms)

AR activities

Symmetry

Calculated

m(N11AH15) (99) m(N10AH7) (96) m(N9AH6) (98)

d(C2@C1AN12) (28) + d(H7AN10AC1) (21) + m(N10AC1) (12) d(H15AN11AC2) (39) + d(H7AN10AC1) (22) d(H6AN9AC3) (38) + d(H15AN11AC2) (28) + m(O13@C3) (17) d(H8AC5@N12) (40) + m(N11AC5) (13) + m(N12@C5) (11) d(H15AN11AC2) (21) + d(C2@C1AN12) (16) + m(N9AC4) (15) m(N9AC4) (24) + d(C2@C1AN12) (14) + m(N9AC4) (10) m(N12AC1) (39) + m(C1@C2) (13) + d(H7AN10AC1) (12)

IIR intensities

428 (w) –

3525 3512 3482

113.89 107.67 86.42

95.36 93.95 83.62

A0 A0 A0

3147 1733 1707 1569 1540 1419 1409 1374 1360 1284 1254

2.45 825.96 856.67 65.19 126.05 26.18 37.19 163.43 2.68 26.68 41.40

115.32 32.25 76.69 33.19 25.69 3.90 49.40 19.90 10.18 59.75 9.24

A0 A0 A0 A0 A0 A0 A0 A0 A0 A0 A0

1225

10.07

15.20

A0

1165 1086 1073 957 932 820 819 731 718 691 668 623 603 602 510

62.37 98.84 2.34 17.66 0.52 5.22 12.06 0.30 53.58 20.18 13.33 11.54 1.48 85.78 164.86

5.08 1.86 11.92 1.18 7.08 0.76 0.53 0.53 0.03 0.42 0.75 0.04 23.23 0.62 0.44

A0 A0 A0 A0 A0 A0 A00 A00 A00 A00 A0 A00 A0 A00 A00

26.06 498 461 366

2.53 3.25 13.30 5.35

A0 0.71 4.14 2.24

A00 A0 A0

311 287 173 141 103

1.02 23.82 0.36 4.54 0.37

0.63 1.13 0.25 0.22 0.02

A00 A0 A00 A00 A00

104

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

IR Intensity (%)

nature [62]. Similar to the nitrogen atomic charges, the magnitudes of the oxygen atomic charges are calculated to be negative because of the acceptor feature (the excess electron density) of the oxygen atoms. Of the oxygen atoms, the O14 obtains far more negative value owing to higher positively charged (low electronegative) carbon atom. Even, the difference between the bond length values (C4@O14 < C3@O13) confirms the various charge magnitude values on the oxygen atoms. Besides, it is customary to say that the magnitudes of the oxygen atomic charges decrease monotonously with the enhancement in the dielectric constant values as a consequence of the effect of a solvent reaction field. As for the magnitudes of the hydrogen atomic charges, each has only positive values (acceptor feature) at the different solvent polarities. This is related to the fact that the charge transfer carries out from the hydrogen to the other atoms. The charge magnitudes on the H6, H7 and H15 atoms are found to be close to each other, but different from that on the H8 atom, stemming from the atomic positions in the title compound. It should be noted here that the highest magnitude is calculated for the H15 atom whereas the smallest one ascribes to the H8 atom. The hydrogen atomic magnitudes tend to enhance with increasing the solvent polarity value. Additionally, that the carbon atoms have positive magnitude except for the C2 atom is the evidence for the presence of the resonance effect among the atoms. The difference in the magnitudes is an evidence for the strong intra-molecular charge transfer (ICT) in the title

in gas phase in cyclohexane in benzene in toluene in chloroform in chlorobenzene in dichloromethane in dmso in water

1850

1800

1750

1700

1650

Wavenumbers (cm-1) Fig. 5. The calculated IR spectra of xanthine using B3LYP/6-311++G(d,p) level in solvents.

molecule [63]. It is another probable result that the solvent polarity generally leads to increase of the charge magnitude. The second-order Fock matrix was used to evaluate the donor– acceptor interactions in the NBO basis. Strong electron delocalisation in the Lewis structure also shows up as donor–acceptor interactions. This bonding–anti bonding interaction can be quantitatively described in terms of the NBO approach that is expressed by means of second-order perturbation interaction energy E(2) [64,65]. The stabilization energy E(2) associated with i(donor) ? j(acceptor) delocalization is estimated from the second-order perturbation approach [66] as given below

Eð2Þ ¼ qi

F 2 ði; jÞ ej  ei

where qi is the donor orbital occupancy, ei and ej are diagonal elements (orbital energies) and F(i,j) is the off-diagonal Fock matrix element. The second order perturbation analysis of Fock matrix of xanthine are given in Table 7. The most important interactions are from lone pair orbitals to the antibonding orbitals which give rise to the stabilization of the molecule. The largest second-order perturbation energy is obtained to be 54.76 kcal/mol for the n (LP (1) N10) ? p* (C4AO14) hyperconjugative interaction of tautomer-10. Likewise, n (LP (1) N11) ? r* (C5AN12) interaction between the nitrogen lone pair and the C5AN12 antibonding orbital is 54.47 kcal/mol. B3LYP calculations show that the most important interactions have occurred from the electron donating LP1 (N9), (N10) and (N11) to the antibonding acceptor p*(C3AO13), p*(C4AO14), r* (C1AC2), p* (C1AC2) and r* (C5AN12) with stabilization energy from 30.16 to 54.76 kcal mol1 by DFT level. This is a evidence that elongation and the weaking the bonds (C1AC2), (C4AO14) and (C3AO13). The n (LP (2) O13) ? r* (C3AN9) stabilisation energy of lone pair of electrons present in the oxygen atom (O13) to the antibonding orbital r* of (C3AN9) is equal to 28.69 kcal/mol. The bond pair donor orbital, p (C1AC2) ? p* (C3AO13) interaction between the (C1AC2) bond pair and the (C3AO13) antibonding orbital give more stabilisation of 27.66 kcal mol1 while between the (C5AN12) bond pair and the (C1AC2) antibonding orbital has 26.77 kcal mol1. According to our results, the intramolecular charge transfer (n (LP) ? p*, n (LP) ? r*, p ? p*, p ? r*) occurs in xanthine molecule. Assignments of fundamentals of the xanthine molecule are also determined with respect to the point group CS including an identity (E) and a reflection (rh) element. The xanthine with the 15 atoms is described by vibrational modes of 27 A0 + 12 A00 Here, A0 presents the symmetric (in plane) modes while A00 denotes the anti-symmetric (out of plane) modes [67,70]. As well known from the three Cartesian displacements a molecule provides 3N internal

Table 6 Comparison of the calculated C@O stretching vibrations (cm1) of xanthine using B3LYP/6-311++G(d,p) level in gas phase and solvents.

m(C@O) Gas phase Cyclohexane Benzene Toluene Chloroform Chlorobenzene Dichloromethane DMSO Water Exp. b IR (cm1) (solid state)

m(C@O)

e

Unscaled

Scaleda

IIR

AR

Unscaled

Scaleda

IIR

AR

Dielectric constant

1765.04 1736.38 1732.59 1731.25 1715.42 1712.38 1707.02 1698.72 1697.91

1706.79 1679.08 1675.41 1674.12 1658.81 1655.87 1650.69 1642.66 1641.88 m(C@O) 1658

856.7 1318.0 1382.6 1405.6 1675.5 1726.6 1821.1 1955.1 1968.0

76.7 109.0 112.9 114.2 129.7 132.6 134.6 141.4 142.1

1792.12 1763.09 1759.54 1758.31 1744.26 1741.66 1736.68 1729.76 1729.09

1732.98 1704.91 1701.48 1700.29 1686.70 1684.19 1679.40 1672.68 1672.03 m(C@O) 1700

826.0 989.9 1013.2 1021.8 1134.8 1159.4 1200.1 1276.9 1284.9

32.3 72.6 80.2 83.0 121.2 129.6 148.8 175.6 178.4

1 2.02 2.27 2.37 4.71 5.70 8.93 46.83 78.36

IIR and AR: calculated infrared intensities and Raman activities. a The harmonic C@O stretching frequencies were calculated by B3LYP/6-311++G(d,p) level of theory and then scaled by 0.967. b Taken from Ref. [70].

105

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

Table 7 Second order perturbation theory analysis of Fock matrix of xanthine using NBO analysis using B3LYP/6-311++G(d,p) method.

1.868 Å

1.868 Å

Dimer A

Donor (i) ? Acceptor (j)

E(2) a (kcal mol1)

E(j)–E(i) (a.u.)

p(C1AC2) ? r (N11AH 15) p(C1AC2) ? p (C3AO13) p(C1AC2) ? r (C5AN12) p(C1AN10) ? p(C4AO14) p(C1AN12) ? r (C5AH8) p(C2AC3) ? p (C1AC2) p(C3AO13) ? p (C1AC2) p(C5AN12) ? r (C1AC2) n(LP(1) N9) ? p(C3AO13) n(LP(1) N9) ? p(C4AO14) n(LP (1) N10) ? r (C1AC2) n(LP (1) N10) ? p

4.13 27.66 13.08 2.17 4.20 3.95 4.53 26.77 50.40 53.68 44.97 54.76

1.11 0.30 0.27 1.46 1.21 1.27 0.38 0.33 0.28 0.28 0.29 0.28

0.061 0.082 0.055 0.050 0.064 0.063 0.042 0.090 0.107 0.110 0.104 0.112

30.16 54.47

0.30 0.27

0.084 0.109

6.00 6.86

0.94 0.83

0.068 0.068

17.78 28.69 26.27 25.73

0.73 0.65 0.65 0.66

0.104 0.123 0.119 0.119

(C4AO14) n(LP (1) N11) ? p (C1AC2) n(LP (1) N11) ? r (C5AN12) n(LP (1) N12) ? r (C1AC2) n(LP (1) N12) ? r (C5AN11) n(LP (2) O13) ? r (C2AC3) n(LP (2) O13) ? r (C3AN9) n(LP (2) O14) ? r (C4AN9) n(LP (2) O14) ? r (C4AN10)

1.871 Å

1.871 Å

b

F(i,j) c (a.u.)

LP – lone pair. a Stabilisation (delocalisation) energy. b Energy difference between i (donor) and j (acceptor) NBO orbitals. c Fock matrix element i and j NBO orbitals.

Dimer B

1.818 Å

1.818 Å

Dimer C Fig. 6. Three views of the calculated dimer forms of xanthine at the B3LYP/6311++G(d,p) level.

modes. 6-Mode of them is associated with the translation and rotational modes. Roughly,

Cinternal ¼ 30A0 þ 15A00 Based on the character table belonging to the CS point group,

Ctransition ¼ 2A0 þ A00 and Crotational ¼ A0 þ 2A00 ; we will thus find the vibration modes for the xanthine compound as given below.

Cv ibretional ¼ C45  Ctransition  Crotational ¼ 27A0 þ 12A00

The available experimental and theoretical vibrational frequencies with their respective dominant normal modes, infrared intensities, Raman activities and related potential energy distribution (PED’s) values are listed in Table 5, in which the wavenumbers are given after scaled by a scaling factor of 0.967 [68,69]. That the observed and calculated vibrational frequencies are found to be in good agreement with each others is obvious in the table. According to the experiment results obtained from [70], the CAH stretching band of nitrogen heterocydic aromatic compound is appeared in the frequency of 2825 cm1 in the FTIR spectra and 2797 cm1 in the FT Raman spectra, respectively. The related mode is calculated to be about 3147 cm1 at B3LYP/6-311++G(d,p) level of theory. The shift between the experimental and theoretical frequencies results from the intra-molecular and inter-molecular hydrogen bonding effects [71]. The corresponding IR band intensity is found to be about 2.45 km/mol. As for the bending vibrations, the IR bands experimentally identified in 958 and 853 cm1 are attributed to the torsion and deformation vibrations whereas the theoretical results are noted to be in a range of 819–957 cm1 with their intensities of 12.06 and 17.66 km/mol, respectively. Moreover, the vibrational bands to be observed at 2875–3133 cm1 in the FTIR spectra and 2875–3137 cm1 in the FT Raman spectra of the title compound are ascribed to the NAH stretching modes while the bands are computed to be in a range of 3147–3525 cm1 at B3LYP/6-311++G(d,p) basis set. It is another important point obtained from Table 5 that among the NAH stretching vibrations, the greatest IR intensity value is determined to be about 113.89 km/mol for the N11AH15 stretching mode. Additionally, the C@O stretching bands of two carbonyl group vibrations are assigned to the strong bands at 1658 and 1700 cm1 in FTIR spectra, and 1600 and 1685 cm1 in FT Raman spectra. On the other hand, the C@O stretching vibrations are calculated to be in the range from 1707 cm1 to 1733 cm1 at B3LYP/ 6-311++G(d,p) calculation level. Based on the results, the experimental evidences obtained from the solid form of the xanthine

106

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

molecule are noticed to be slighter than the calculations performed in the gasoeus phase. The calculated band at 1707 cm1 with the assignment of C14@O24 stretching mode obtains the highest IR intensity value of 856.67 km/mol among the fundamental vibrational modes studied. It is necessary to underline that the C@O stretching vibration modes in the FTIR spectrum even appear at 1725 with ±65 cm1 [72]. The other important mode for the identification of the xanthine compound is the CAN stretching vibrations. These vibration groups are observed at 1032–1334 cm1 and 1040–1335 cm1 in the FTIR and FT Raman spectrum of the title compound. Theoretically, the characteristic CAN stretching bands are generally calculated around 1073–1360 cm1. Similar to CAN stretching modes, the C@N stretching band is experimentally observed at around 1658 cm1 in both spectra whereas the bands are computed at 1374 and 1658 cm1. The considerable variation of the wavenumber between the C@N and CAN modes may correspond to the existence of the different electron engagement in the modes. Besides, the characteristic mode of the C@C stretching band is determined to be about 1567 cm1 in the FTIR and 1560 cm1 in FT Raman

spectrum. According to the PED analysis in Table 5, the bands observed at 1254 and 1569 cm1 are assigned to the C@C stretching vibration. The IR intensities are found to be middle level such as 65.19 and 41.40 km/mol. This stretching mode is also known as the degree of conjugation through the p-electron chain in the structure [73]. Thus, it is not wrong to say that the charge transfer interaction between the donor and acceptor groups appears throughout the p-system [74]. Additionally, the CAC stretching bands experimentally lie at 1215 and 1212 cm1 in the FTIR and FT Raman spectrum, respectively. The stretching band is theoretically found to be about 1225 cm1 at B3LYP/6-311++G(d,p) level of theory. As for the skeletal vibrations of the title compound, the corresponding wavenumbers being assigned to the in- and out-of plane IR modes are calculated in the range between 103 and 1540 cm1. Of the modes, the maximum IR intensity value of 126.05 km/mol is observed at 1540 cm1 for the mixture bending modes of C2@C1AN12 and H7AN10AC1. Moreover, the effect of the different solvents on the wavenumbers of the xanthine molecule is determined by means of the DFT/ B3LYP/6-311++G (d,p) with CPCM model. Here, the ˆ (C@O)

Table 8 Selected (strong IR intensity) vibrational frequencies (cm1) of Dimer C at B3LYP/6-311++G(d,p) level. Assignments, PEDa (%)

m(N11AH15, N29AH26) (99) m(N9AH6, N20AH18) (97) m(N21AH19, N10AH7) (98) m(C25AH27, C5AH8) (96) m(O23@C16, O13@C3) (56) + m(O14@C4, O22@C17) (21) m(O14@C4, O22@C17) (48) + m(O23@C16, O13@C3) (17) m(C1@C2, C24@C30) (54) + m(C24AN21, C1AN10) (41) d(C2@C1AN12, C30@C24AN28) (29) + d(H7AN10AC1, H19AN21AC24) (23) + m(N10AC1, N21AC24) (14) d(H7AN10AC1, H19AN21AC24) (42) + d(H15AN11AC2, H26AN29AC30) (18) d(H8AC5AN12, H27AC25AN28) (38) + m(N11AC5, N29AC25) (14) + m(N12AC5, N28AC25) (12) d(H15AN11AC2, H26AN29AC30) (30) + m(N11AC5, N29AC25) (19) + d(C2AN11AC5, C30AN29AC25) (18) d(H6AN9AC3, H18AN20AC16) (44) + d(H15AN11AC2, H26AN29AC30) (15) + m(O13@C3, O23@C16) (11) m(N12AC1, N28AC24) (35) + m(C1@C2, C24@C30) (14) + d(H7AN10AC1, H19AN21AC24) (13) d(H7AN10AC1, H19AN21AC24) (43) + m(C1AN12, C24AN28) (12) + m(C2AN11, C30AN29) (12) m(N12AC1, N28AC24) (22) + d(H8AC5AN12, H27AC25AN28) (21) + d(N11AC5AN12, N29AC25AN28) (11) d(H8AC5AN12, H27AC25AN28) (21) + d(C2AN11AH15, C30AN29AH26) (18) + m(N11AC2, N29AC30) (10) m(N9AC3, N20AC16) (23) + m(N11AC5, N29AC25) (20) + m(N9AC4, N20AC17) (13) d(H15AN11AC2, H26AN29AC30) (35) + d(H8AC5AN12, H27AC25AN28) (34) m(N10AC4, N21AC17) (22) + m(N9AC4, N20AC17) (22) + d(H6AN9AC3, H18AN20AC16) (15) + d(H7AN10AC1, H19AN21AC24) (15) d(N11AC5AN12, N29AC25AN28) (65) [(H7AN10AC1@C2, H19AN21AC24@C30) (85) d(C5AN12AC1, C25AN28AC24) (26) + d(C1AN10AC4, C24AN21AC17) (24) x(H8AC5AN12AC1, H27AC25AN28AC24) (96) x(C4AN9AC3@O13, C17AN20AC16@O23) (36) + x(C2AC1AN10AH7, C30AC24AN21AH19) (32) x(C4AN9AC3@O13, C17AN20AC16@O23) (32) + s(H6AN9AC3AC2, H18AN20AC16AC30) (20) + s(C2@C1AN12AC5, C30@C24AN28AC25) (20) s(H6AN9AC3AC2, H18AN20AC16AC30) (34) + s(C1AN12@C5AN11, C24AN28AC25AN29) (20) + x(C4AN9AC3@O13, C17AN20AC16@O23) (18) d(O14@C4AN10, O22@C17AN21) (35) + d(N9AC3AO13, N20AC16AO23) (34) + d(C3AN9AH6, C16AN20AH18) (11) s(H6AN9AC3AC2, H18AN20AC16AC30) (36) + s(H8AC5AN11AH15, H27AC25AN29AH26) (22) d(C2AC3AN9, C30AC16AN20) (45) + d(C3AN9AC4, C17AN20AC16) (17) + d(C3AC2AC1, C16AC30AC24) (15) s(H6AN9AC3AC2, H18AN20AC16AC30) (42) + s(H8AC5AN11AH15, H27AC25AN29AH26) (22) d(N9AC4AN10, N20AC17AN21) (43) + d(C2AC1AN10, C30AC24AN21) (24) + d(O14@C4AN10, O22@C17AN21) (16) s(H15AN11AC2@C1, H26AN29AC30@C24) (76) d(C4AN10AC1, C17AN21AC24) (37) + d(N9AC3AC2, N20AC16AC30) (26) d(O14@C4AN10, O22@C17AN21) (55) + d(O13@C3AN9, O23@C16AN20) (16) s(C4AN10AC1AN12, C17AN21AC24AN28) (46) + s(N9AC4AN10AC1, N20AC17AN21AC24) (26) d(O13@C3AN9, O23@C16AN20) (30) + d(C3AC2AN11, C16AC30AN29) (25) s(N9AC4AN10AC1, N20AC17AN21AC24) (74) s(O13@C3AC2@C1, O23@C16AC30@C24) (45) + x(C3AC2AN11AC5, C16AC30AN29AC25) (15) s(O14@C4AN10AC1, O22@C17AN21AC24) (35) + s(N10AC1AC2AN11, N21AC24AC30AN29) (24) d(N21AH19  O14@C4, N10AH7  O22@C17) (76) s(H19  O14@C4AN9, H7  O22@C17AN20) (75) x(C4@O14  H19AN21, C17@O22  H7AN10) (76) Abbreviations used: m: stretching; d: bending; s: torsion; x: out of bending. a PED less than 10% are not shown.

B3LYP/6-311++G(d,p) Unscaled freq.

Scaled freq.

IIR

3648 3601 3288 3251 1773 1746 1618 1598 1494 1466 1429 1400 1350 1334 1279 1205 1122 1107 1020 963 856 854 847 752 743

3528 3482 3179 3144 1714 1688 1565 1545 1445 1418 1382 1354 1305 1290 1237 1165 1085 1070 986 931 828 826 819 727 718

239.96 157.42 2894.60 4.27 1100.86 3117.44 113.38 289.88 28.26 117.35 303.59 37.34 64.43 82.43 3.83 99.00 284.91 0.23 28.78 6.30 179.38 110.22 29.63 16.31 77.17

715

691

20.23

700 646 626 623 537 515 487 404 334 300 186 150 124 56 26 15

677 625 605 602 519 498 471 391 323 290 180 145 120 54 25 15

32.38 32.30 1.27 92.43 53.03 186.19 105.37 20.42 4.50 57.15 2.65 12.83 0.60 10.22 0.34 2.75

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

stretching mode is chosen to be reference vibration. The theoretical IR spectra obtained are given graphically in Fig. 5. It is obvious from the evidences (Figs. 2 and 5, and Table 6) that the xanthine compound studied in this work have two strong C@O bonds appearing in the region of 1697–1792 cm1 at the different solvent polarities (dielectric constant). It is understood that the stretching C@O vibrations monotonously shift to lower frequency value with higher IR intensity as the dielectric medium increases steeply. This is association with the elongation of the C@O bond length as consequence of the bond weakening (Table 6). In other words, the electronegativity of the oxygen atom (negatively charged) tends to enhance with the solvent polarity. More negative charge focuses on the oxygen atoms in the xanthine molecule at higher polarity, meaning the formation of the hydrogen bond between the C@O group in the molecule and the suitable groups of the solvents. The long and short of it is that the environment polarity affects considerably the vibrational spectrum due to the effect of the solvent reaction field on the compound studied. Fig. 6 shows three stable dimer forms belonging to the xanthine compound in the planar symmetry. The computations are performed at B3LYP/6-311++G (d,p) level of theory to obtain the most favored (stable) one. The calculated energy values of three dimers indicate that dimer C (the most stable) has the lowest energy value of -1125.23647143 a.u. due to the far stronger H-bonds in the structure as compared to the H-bonds in the others. The energy difference between dimer C and dimer B (dimer A) is 2.82 (2.78) kcal mol1. Consequently, it can be noted that the dimer B is the least stable (most reactive) form among the forms studied. The intermolecular NAH  O bond lengths are, in this respect, found to be about 1.868 Å for dimer A and 1.871 Å for dimer B, respectively. The NAH  O bond length of dimer C is calculated to be about 1.818 Å. Table 8 shows selected (strong IR intensity) vibrational frequencies (cm1) of Dimer C at B3LYP/6-311++G(d,p) level. In dimer C form, the C4@O14 and especially N10AH7 stretching vibrations corresponding to the intermolecular hydrogen bond formation have a significant variation. The former vibration mode appears at 1792.12 cm1 in the monomer form (in gas phase), but starts at 1745.54 cm1 and at 1742.86 cm1 in the dimer form with two bands. Likewise, the latter vibration mode is calculated to be about 3631.44 (scaled 3512) cm1 in the monomer form (in gas phase), but appears with two bands at 3287.78 (scaled 3179) and 3247.14 cm1 in the dimer form. It is another important point that

Fig. 7. The total electron density isosurface mapped with molecular electrostatic potential of xanthine.

107

the initial band (3179 cm1) appears with null Raman activity in the presence of very strong IR intensity and the other band (3247.14 cm1) is vice versa. In this part of the exhaustive study, we examine the influence of the solvent polarity on the molecular structure and vibrational spectra of the dimer C with the aid of DFT calculations performed at B3LYP/6-311++G (d,p) level with CPCM model in the different medium polarity. For the equilibrium geometry of the dimer C, it is realized that the C4@O14 bond length expands significantly while the N10AH7 bond length contracts drastically with the enhancement of the solvent polarity. Furthermore, the NAH  O

Fig. 8. The contour map of molecular electrostatic potential surface of xanthine.

Fig. 9. Electrostatic potential surface of xanthine.

108

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109

distance in water (1.825 Å) is found to longer than that in gas phase (1.818 Å) as a consequence of the interaction between solvent and the molecular system. As for the vibrational analyzes in the water (the greatest medium), the C4@O14 and N10AH7 stretching vibrations are calculated at 1693.52, 1683.04 cm1; and 3262.61, 3237.94 cm1, respectively. For the mode of t(C4@O14), the first band appears with null IR intensity in the strong Raman activity whereas the second band (at 1683.04 cm1) takes place with null Raman activity in the strong IR intensity. On the other hand, for t(N10AH7) vibrational mode, the first band (at 3262.61 cm1) is observed with null Raman activity in the strong IR intensity whereas the second band (at 3237.94 cm1) is vice versa. The molecular electrostatic potential surface (MESP) which is mapping electrostatic potential onto the iso-electron density surface simultaneously displays electrostatic potential (electron + nuclei) distribution, molecular shape, size and dipole moments of the molecule and it provides a visual method to understand the relative polarity of the compounds [75]. The total electron density mapped with electrostatic potential surface of xanthine constructed by B3LYP/6-311++G(d,p) method is shown in Fig. 7. The electrostatic potential contour map and the surface for positive and negative potentials are shown in Figs. 8 and 9, respectively. The color scheme for the MESP surface is red, electron rich, partially negative charge; blue, electron deficient, partially positive charge; light blue, slightly electron deficient region; yellow, slightly electron rich region; green, neutral; respectively [76]. It is obviously from the Figs. 7 and 8 that the region around oxygen atoms associated with carbon through double bond represents the most negative potential region (red). The hydrogen atoms attached to the nitrogen atoms and an oxygen atom posses the maximum positive charge. The predominance of green region in the ring surfaces corresponds to a potential halfway between the two extremes red and dark blue color [76]. Thus, the total electron density surface mapped with electrostatic potential clearly reveals the presence of high negative charge on the carbonyl oxygen while more positive charge around the hydrogen atoms region.

Conclusion In this comprehensive study, we investigate the effects of 8 solvents on the geometric structure and vibrational spectra of the xanthine compound with the help of the density functional theory method at the standard B3LYP/6-311++G(d,p) calculation level with the polarizable conductor continuum model for the first time. Theoretical vibrational spectra are also assigned by means of the PED analyses determined from the VEDA 4 program. We find the favored tautomer belonging to the xanthine molecule in terms of the SCF energy values, and then examine the atomic charges and dipole moments with respect to the solvent polarity in detail. Moreover, the dimer forms of the xanthine compound are simulated to explain the influence of intermolecular hydrogen bonding on the molecular geometry and vibrational frequencies, and the major conclusions to be inferred from this work are the followings:  The molecular energy values of 36 tautomeric and conformer forms of the xanthine molecule display that the global minimum energy of the most favorable compound is predicted to be about 562.607505 au.  The researches of the solvent polarity effect on the SCF energy and dipole moment indicate that the molecule stability improves with the enhancement of the polarity due to the reduction of the SCF energy value. As for the dipole moment examination, there is a systematic increase in the dipole moment value. This is attributed to the polar nature (nonuniform distribution of charges) of the title compound.

 Medium polarity also affects considerably the atomic charge magnitudes (especially oxygen atomic charge) as a result of the solvent reaction field. Even, the difference in the charge magnitudes points out the strong intra-molecular charge transfer in the compound.  Although the optimized geometric parameters are shown to have a good agreement with the available experimental evidences in the literature, there are little deviations on the parameters because of the strong intra-molecular charge transfer in the xanthine molecule.  Similar to the molecular geometry findings, the theoretical and experimental vibrational modes are noted to be in good agreement with each other. The shift between the experimental and theoretical vibrations stems from the intra-molecular and intermolecular hydrogen bonding effects.  Moreover, the C@O and NAH stretching vibrations systematically shift to lower frequencies with higher IR intensity values as a consequence of the bond weakening when the solvent polarity increases regularly, confirming the existence of the intermolecular NAH  O hydrogen bonds in the compound. It is another probable result that the variation of the stretching vibration verifies the charge transfer interaction between the donor and acceptor groups through the p-system in the compound. Hence, the hydrogen bond forms between the C@O groups in the molecule and the suitable groups of the solvents.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.saa.2013.12.035. References [1] K. Oettl, G. Reibnegger, Biochim. Biophys. Acta 1430 (1999) 387–395. [2] M. Arivazhagan, S. Jeyavijayan, Spectrochim. Acta Part A 79 (2011) 161– 168. [3] T. Umemura, K. Ueda, K. Nishioka, T. Hidaka, H. Takemoto, S. Nakamura, D. Jitsuiki, J. Soga, C. Goto, K. Chayama, M. Yoshizumi, Y. Higashi, J. Am. Coll. Cardiol. 98 (2006) 1538–1541. [4] F. Crea, C. Stefano, G. Manfredi, S. Sammartano, J. Mol. Liq. 178 (2013) 37–43. [5] R.E. Wing, C.L. Swanson, W.M. Doane, C.R. Russell, J. Water Pollut. Control Fed. 46 (1974) 2043–2047. [6] B. Giese, Angew. Chem. Int. Ed. Engl. 24 (1985) 553–565. [7] R.M. Bricka, Investigation and Evaluation of the Performance of Solidified Cellulose and Starch Xanthate Heavy Metal Sludges. US Army Corps of Engineers Waterway Experiment Station Technical Report EL-88-5, Vicksburg, 1988. [8] I. Ryu, N. Sonoda, Chem. Int. Ed. Engl. 35 (1996) 1050–1066. [9] R.K. Kasim, Synth. Met. 85 (1997) 1291–1297. [10] L. Jean-Baptiste, S. Yemets, R. Legay, T. Lequeux, J. Org. Chem. 71 (2006) 2352– 2359. [11] F.L. Fu, R.M. Chen, Y. Xiong, Sep. Purif. Techonol. 52 (2006) 383–393. [12] C.F. Huang, R. Nicolay, Y. Kwak, F.C. Chang, K. Matyjaszewski, Macromolecules 42 (2009) 8198–8210. [13] S. Shahzadi, S. Ali, R. Jabeen, M.K. Khosa, Turk. J. Chem. 33 (2009) 307–312. [14] F. Ucun, A. Sag˘lam, V. Güçlü, Spectrochim. Acta Part A 67 (2007) 342–349. [15] A.A. Stolov, W.A. Herrebout, B.J. Van der Veken, J. Mol. Struct. 480–481 (1999) 499–503. [16] S.J. Cho, A.A. Mohamed, S.A.K. Elroby, Int. J. Quantum. Chem. 107 (2007) 63– 71. [17] J. Leis, K.D. Klika, M. Karelson, Tetrahedron 54 (1998) 7497–7504. [18] O.G. Parchment, D.V.S. Green, P.J. Taylor, I.H. Hiller, J. Am. Chem. Soc. 115 (1993) 2352–2356. [19] P. Cieplak, P. Bash, U.C. Singh, P.A. Kollman, J. Am. Chem. Soc. 109 (1987) 6283–6289. [20] C.J. Cramer, D.G. Truhlar, J. Am. Chem. Soc. 115 (1993) 8810–8817. [21] M.A. Palafox, G. Tardajos, A.G. Martines, V.K. Rastogi, D. Mishra, S.P. Ojha, W. Kiefer, Chem. Phys. 340 (2007) 17–31. [22] D. Avci, Y. Atalay, M. Sekerci, M. Dincer, Spectrochim. Acta A 73 (2009) 212– 217. [23] F. Ucun, V. Guclu, A. Saglam, Spectrochim. Acta A 70 (2008) 524–531. [24] H. Gokce, S. Bahceli, Spectrochim. Acta Part A 78 (2011) 803–808. [25] F. Karaboga, U. Soykan, M. Dogruer, B. Ozturk, G. Yildirim, S. Cetin, C. Terzioglu, Spectrochim. Acta Part A 113 (2013) 80–91.

T. Polat, G. Yıldırım / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 98–109 [26] N.C. Handy, P.E. Masley, R.D. Amos, J.S. Andrews, C.W. Murray, G. Laming, Chem. Phys. Lett. 197 (1992) 506–515. [27] G. Bakalarski, B. Lesyng, Pol. J. Chem. 72 (1998) 1792–1797. [28] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, et al. Gaussian 09, Revision B.01, Gaussian Inc., Wallingford, CT, 2010. [29] R.D. Dennington, T.A. Keith, J.M. Millam, GaussView 5, Gaussian Inc., 2008. [30] E.D. Glendening, A.E. Reed, J.E. Carpenter, F. Weinhold, in: NBO Version 3.1 Programme Manual, University of Wisconsin, Madison, 1996. [31] V. Barone, M. Cossi, J. Phys. Chem. A 102 (1998) 1995–2001. [32] G. Fogarasi, P. Pulay, In Vibrational Spectra and Structure, Elsevier, Amsterdam, 1985. [33] P. Pulay, G. Fogarasi, G. Pongor, J.E. Boggs, A. Vargha, J. Am. Chem. Soc. 105 (1983) 7037–7047. [34] G. Fogarasi, X. Zhou, P.W. Taylor, P. Pulay, J. Am. Chem. Soc. 114 (1992) 8191– 8201. [35] G. Rauhut, P. Pulay, J. Phys. Chem. 99 (1995) 3093–3100. [36] M.H. Jamroz, Vibrational Energy Distribution Analysis, VEDA 4 Program, Warsaw, Poland, 2004. [37] P.L. Polavarapu, J. Phys. Chem. 94 (1990) 8106–8112. [38] G. Keresztury, S. Holly, J. Varga, G. Besenyei, A.Y. Wang, J.R. Durig, Spectrochim. Acta A 49 (1993) 2007–2026. [39] L. Padmaja, M. Amalanathan, C. Ravikumar, I.H. Joe, Spectrochim. Acta A 74 (2009) 349–356. [40] S. Yanase, T. Oi, Z. Naturforsch. A 56 (2001) 297–306. [41] A.S. Tiwary, P.S. Sengupta, A.K. Mukherjee, J. Theor. Comput. Chem. 7 (2008) 331–346. [42] E. Canses, J. Thomasi, B. Mennucci, J. Thomasi, J. Chem. Phys. 107 (1997) 3032– 3042. [43] A.J. Barnes, M.A. Majid, M.A. Stuckey, P. Gregory, C.V. Stead, Spectrochim. Acta A 41 (1985) 629–635. [44] J.P. Schermann, Spectroscopy and Modeling of Biomolecular Building Blocks, Elsevier Science, France, 2007. [45] A.A. Ahmed, Computational and Theoretical Chemistry 999 (2012) 251–258. [46] G. Yildirim, S.D. Senol, M. Dogruera, O. Ozturk, A. Senol, A.T. Tasci, C. Terzioglu, Spectrochim. Acta A 85 (2012) 271–282. [47] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 105 (2005) 2999–3093. [48] S. Subashchandrabose, H. Saleem, Y. Erdogdu, O. Dereli, V. Thanikachalam, J. Jayabharathi, Spectrochim. Acta A 86 (2012) 231–241. [49] D.A. Dhas, I.H. Joe, S.D.D. Roy, S. Balachandran, Spectrochim. Acta A 79 (2011) 993–1003. [50] X.J. Peng, F.L. Song, E. Lu, Y.N. Wang, W. Zhou, J.L. Fan, Y.L. Gao, J. Am. Chem. Soc. 127 (2005) 4170–4181.

109

[51] G. Qian, Z. Zhong, M. Luo, D. Yu, Z. Zhang, D. Ma, Z.Y. Wang, J. Phys. Chem. C 113 (2009) 1589–1595. [52] M. Nonella, G. Hanggi, E. Dubler, J. Mol Struct. THEOCHEM 279 (1993) 173– 190. [53] X. Ma, J. Jiao, J. Yang, X. Huang, Y. Cheng, C. Zhu, Polymer 53 (2012) 3894– 3899. [54] C.H. Choi, M. Kertesz, J. Phys. Chem. A 101 (1997) 3823–3831. [55] M.T. Gulluoglu, Y. Erdogdu, J. Karpagam, N. Sundaraganesan, S. Yurdakul, J. Mol. Struct. 990 (2011) 14–20. [56] M. Koca, G. Yildirim, C. Kirilmis, F. Karaboga, Arab. J. Sci. Eng. 37 (2012) 1283– 1291. [57] A.N. Chermahini, A. Teimouri, A.S. Beni, Struct. Chem. 22 (2011) 175–181. [58] S. Cetin, G. Yildirim, C. Parlak, M. Akdogan, C. Terzioglu, Spectrochim. Acta A 79 (2011) 1024–1033. [59] L. Xiao-Hong, T. Zheng-Xin, Z.J. Xian-Zhou, Mol. Struct. THEOCHEM 900 (2009) 50–54. [60] C. Moller, M.S. Plesset, Phys. Rev. 46 (1934) 618–622. [61] S. Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 55 (1981) 117–129. [62] V. Arjunana, S.T. Govindarajab, S. Sakiladevic, M. Kalaivania, S. Mohand, Spectrochim. Acta A 84 (2011) 196–209. [63] U. Soykan, S. Cetin, B. Ozturk, F. Karaboga, Y. Zalaoglu, M. Dogruer, G. Yildirim, C. Terzioglu, J. Mol. Struct. 1049 (2013) 479–487. [64] I.E. Reed, F. Weinhold, J. Chem. Phys. 78 (1983) 4066–4073. [65] K. Fukui, Science 218 (1982) 747–754. [66] J.P. Foster, F. Weinhold, J. Am. Chem. Soc. 102 (1980) 7211–7218. [67] H. Buyukuslua, M. Akdogan, G. Yildirim, C. Parlak, Spectrochim. Acta A 75 (2010) 1362–1369. [68] O. Dereli, Y. Erdogdu, M.T. Gulluoglu, E. Türkkan, A. Ozmen, N.J. Sundaraganesan, Mol. Struct. 1012 (2012) 168–176. [69] G. Yildirim, Y. Zalaoglua, C. Kirilmis, M. Koca, C. Terzioglu, Spectrochim. Acta Part A 81 (2011) 104–110. [70] S. Gunasekaran, G. Sankari, S. Ponnusamy, Spectrochim. Acta A 61 (2005) 117– 127. [71] N.P.G. Roeges, A Guide to Complete Interpretation of Infrared Spectra of Organic Structures, first ed., John Wiley & Sons Inc., New York, 1994. [72] M. Govindarajan, K. Ganasan, S. Periandy, M. Karabacak, Spectrochim. Acta Part A 79 (2011) 646–653. [73] L. Rimai, M.E. Heyde, D. Gill, J. Am. Chem. Soc. 95 (1973) 4493–4501. [74] M. Snehalatha, C. Ravikumar, N. Sekhar, V.S. Jayakumar, I.H. Joe, J. Raman Spectrosc. 39 (2008) 928–936. [75] J.M. Seminario, Recent Developments and Applications of Modern Density Functional Theory, vol. 4, Elsevier, Amsterdam, 1996. pp. 800–806. [76] V. Arjunan et al., Spectrochim. Acta Part A 104 (2013) 182–196.