Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

Contents lists available at ScienceDirect

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

FT-IR, FT-Raman spectra and other molecular properties of 2,4- dichlorobenzonitrile: A interpretation by a DFT study D. Kattan a, M. Alcolea Palafox a,⇑, S. Kumar b, D. Manimaran c, Hubert Joe c, V.K. Rastogi d a

Departamento de Química-Física I, Facultad de Ciencias Químicas, Universidad Complutense, Madrid 28040, Spain Aryan Institute of Technology, 13th km Stone, NH 24, Jindal Nagar, Ghaziabad 201 002, India c Centre for Molecular and Biophysics Research, Department of Physics, Mar Ivanios College, Thiruvananthapuram 629 015, Kerala, India d R.D. Foundation Engineering College, NH-58, Kadrabad, Modinagar, Ghaziabad, India b

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

 The Raman and IR spectra of 2,4-

dichlorobenzonitrile were accurately simulated.  An accurate scaling procedure was used to improve the calculated wavenumbers.  Compared to the experimental the % error is very small in the majority of the bands.  The Potential Energy Distributions (PEDs) was calculated for each normal mode.  The energy gap HOMO–LUMO reflects the chemical activity of the molecule.

a r t i c l e

i n f o

Article history: Received 3 October 2013 Received in revised form 26 November 2013 Accepted 5 December 2013 Available online 18 December 2013 Keywords: 2,4-Dichlorobenzonitrile Dichlorobenzonitrile Scaling wavenumbers FT-Raman

a b s t r a c t FT-IR and FT-Raman spectra of 2,4-dichlorobenzonitrile at room temperature have been recorded in the regions 200–3500 cm1 and 0–3400 cm1, respectively. The observed vibrational bands were analyzed and assigned to different normal modes of the molecule according to the Wilson’s notation. Density functional calculations were performed to support our frequency assignments. Specific scale equations deduced from the benzene molecule were employed to improve the calculated values. For the majority of the normal modes, the deviations between the corresponding experimental and scaled theoretical wavenumbers are located in the expected range. A correct characterization of each normal mode is of vital importance in the assignment of the observed bands, and the same has been successfully done by the aid of Potential Energy Distributions (PEDs) calculated separately for each normal mode of 2,4dichlorobenzonitrile. The molecular structure was optimized and several thermodynamic parameters were determined. HOMO and LUMO orbital energy analysis were carried out. Ó 2013 Elsevier B.V. All rights reserved.

Introduction Benzonitriles (BNs) and its derivatives are widely used today in the manufacturing of polymers and anhydrous metallic salts and as well as intermediates for pharmaceuticals, agrochemicals, pesticides and other organic chemicals [1–3]. Several BN derivatives ⇑ Corresponding author. Tel.: +34 1913944272. E-mail address: [email protected] (M. Alcolea Palafox). 1386-1425/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.saa.2013.12.052

have important applications, e.g. 3-ethylbenzonitrile for the treatment of Urge Urinary Incontinence (UUI) [4], vamicamaide (a new anti-cholinergic agent synthesized from BN) also for treatment of UUI [5] and p-hydroxybenzonitrile has alpha-blocker properties on the cardiovascular system of rats [6]. Other derivatives such as 3,5-dichloro-4-hydroxybenzonitrile (chloroxynil), 3,5-dibromo4-hydroxybenzonitrile (bromoxynil), 3,5-diiodo-4-hydroxybenzonitrile (ioxynil) and 2,6-dichlorobenzonitrile (dichlobenil) are important herbicides, in special, 2,6-dichlorobenzonitrile [7]

90

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

inhibits the cellulose synthesis [8–10] and it is a potent nasal toxicant in rodents [11]. From the spectroscopic point of view, in recent years numerous experimental and theoretical studies have been made on the vibrational spectra of BN and its mono- and di-substituted derivatives [12–36]. In special, those with F- and Cl-atoms offer great interest and they have been studied by vibrational spectroscopy [26–36]. However, a spectroscopic analysis on 2,4-dichlorobenzonitrile molecule (2,4-DCBN), Fig. 1, has not been completely and rigorously carried out yet. Neither its molecular properties have been determined, although its vibrational spectra with poor force fields calculations have been reported earlier [34]. Since a correct characterization of the normal modes is of vital importance in the assignments of the observed bands, therefore the present paper represents a clear improvement of the already published results, correcting mislay in the assignments by using both the B3LYP DFT methods and an accurate scaling procedure. It provides more correct results than the previous ones. It follows our earlier works on the vibrational spectra and molecular properties of several of their chloro derivatives: 2,5-dichlorobenzonitrile [24], 3,5-dichlorobenzonitrile [37] and 2-amino-3,5-dichlorobenzonitrile [38]. Experimental 2,4-DCBN of spectral grade was obtained from M/s Aldrich Chemicals (Milwanke, WI, USA) and used as such without any further purification. The FT-IR spectrum of this compound in KBr pellet (with 1 mg sample per 300 mg KBr) and in Nujol mull were recorded with a Perkin Elmer FT-IR Model 1760 X in the 200– 3500 cm1 range. The resolution of the IR spectrometer was 2 cm1. The FT-Raman spectrum of 2,4-DCBN was recorded in powder form in the region of 0–3400 cm1 on Nicolet Raman 950 at room temperature. The sample was mounted in the sample illuminator

using an optical mount and no sample pretreatment was undertaken. The NIR output (1064 nm) of an Nd:YAG laser was used to excite probe. The instrument was equipped with InGaAs detector. The instrument was set at 250 mW and the spectrum was recorded over 400 scans at a fixed temperature at a resolution of 4 cm1.

Computational methods The calculations were mainly carried out by using ab initio MP2 and Density Functional methods (DFT) [39], because they provide a very good overall description of medium-size molecules. For the wavenumber calculations [40–42] was only used DFT methods because they appear more accurate than HF and MP2 [43], and at lower computational cost. Among DFT methods, B3LYP is the most popular which uses a combination of the Becke’s three-parameter exchange functional (B3) [44] and the correlation functional LYP (Lee, Yang, Parr) [45]. The default fine integration grid was employed. MP2 calculations were mainly carried out to confirm the geometry structure and atomic charges calculated by B3LYP. These methods are implemented in the Gaussian 03 [46] program package. Several basis set differing in size and contraction were used for computation, but for simplicity and owing to the small improvement reached with it, only the results with 6-31G** basis set was discussed in the present study. With these methods and basis set we have previously studied the geometry, vibrational wavenumbers and thermodynamical parameters of a series of six difluoroBNs [25], as well as a detailed analysis of the vibrational spectra of 2,3-difluoroBN [35], 2,4-difluoroBN [33] and 2,5-difluoroBN [47]. Raman scattering activities (Si) calculated by Gaussian 03 W program were suitably converted to relative Raman intensities (Ii) using the following relationship derived from the basic theory of Raman scattering:

Ii ¼

f ð#o  #i Þ4 Si  i i #i 1  exp  hc# kT h

where #o is the exciting frequency (cm1), #i is the vibrational wavenumber of the ith normal mode, h, c and k are universal constants, and f is the suitable chosen common scaling factor for all the peak intensities.

Results and discussion Geometry optimization

Fig. 1. Label of the atoms in 2,4-DCBN molecule.

Optimized bond lengths, bond angles and torsional angles in 2,4-DCBN are listed in the 2nd and 3rd columns of Table 1 at the B3LYP and MP2 levels, respectively, while the labeling of the atoms is plotted in Fig. 1. For comparison purposes the calculated values in BN were collected in the last column. X-ray experimental data on 2,4-DCBN has not been reported yet. Nevertheless, the computed bond lengths and angles were, in general, very close and also in accordance with the microwave data reported in the molecule of BN [50], and with the calculated values in their derivatives [25]. The differences were in accordance with the average error reported [51] for these methods. However, several particular features are observed corresponding to the effect of replacement of the hydrogens by chlorine atoms on the benzonitrile moiety. Thus, the benzene ring appears slightly distorted with angles slightly out of the regular hexagonal structure. The broad features of this replacement in the geometry and NBO atomic charges [48,49] can be described as follows, as compared to BN:

91

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97 Table 1 Optimized geometrical parameters, bond lengths in Å and bond angles and torsional angles in degrees, calculated at the B3LYP/6-31G** and MP2/6-31G** levels in 2,4DCBN and in benzonitrile molecules. Parameters

2,4-DCBN

BN

B3LYP

MP2

B3LYP

Bond lengths C1AC2 C2AC3 C3AC4 C4AC5 C5AC6 C6AC1 C1AC7 C2ACL C4ACL C„N

1.408 1.393 1.394 1.396 1.389 1.406 1.431 1.743 1.748 1.163

1.405 1.393 1.394 1.397 1.389 1.404 1.432 1.729 1.733 1.184

1.405 1.392 1.397 1.397 1.392 1.405 1.435 – – 1.163

Bond angles C2AC1AC6 CAC2AC CAC3AC CAC4AC CAC5AC CAC6AC C6AC1AC7 C1AC2ACL C3AC4ACL CAC„N

118.7 120.8 119.0 121.5 118.9 121.0 119.4 120.4 118.9 178.0

119.3 120.4 119.3 121.2 119.1 120.7 119.5 120.5 119.1 178.4

120.1 119.7 120.2 120.2 120.2 119.7 120.0 – – 180.0

Torsional angles C6AC1AC7„N C1AC2AC3AC4

0.0 0.0

0.0 0.0

90.0 0.0

(i) The ring angle at the C2 and C4 substitution sites slightly increases. The effect with the chlorine atom is lower than with the fluorine atom in 2,4-difluoroBN [33]. (ii) The ring angles at the neighbor C3 and C5 positions decrease. The effect is also lower than with the fluorine atoms. (iii) The C1AC2 and C5AC6 bond lengths adjacent to the C1AC„ bond increase. (iv) The chlorine atoms produce a small shortening of the C3AC4 and C5AC6 (0.003 Å) bond lengths, while C1AC„ slightly decreases. (v) The chlorine atoms withdraw negative charge (ca. 0.2 e, where e is the charge of an electron) on the bonded atoms C2 and C4, whose negative charge is now very small, Table 2. As consequence, the adjacent C1, C3 and C5 atoms increment its negative charge ca. 0.02 e.

Table 2 Calculated natural NBO atomic charges at the B3LYP/6-31G** and MP2/6-31G** levels in 2,4-DCBN and in benzonitrile molecules. Atom

C1 C2 C3 C4 C5 C6 C7 „N CL11 CL12 H8 H9 H13

2,4-DCBN

BN

B3LYP

MP2

B3LYP

0.186 0.001 0.262 0.023 0.253 0.167 0.273 0.278 0.050 0.029 0.279 0.268 0.269

0.201 0.019 0.262 0.005 0.255 0.141 0.315 0.328 0.038 0.016 0.275 0.265 0.265

0.171 0.189 0.235 0.216 0.235 0.189 0.280 0.305 – – 0.249 0.247 0.257

These distortions are explained in terms of the change in hybridization affected by the substituent at the carbon site to which it is appended. Thus in the present case, the ring angles obtained can be reasonably explained by superposition of the ring angular distortions of ortho and para-dichlorobenzene and BN. It can be explained as follows: starting from the symmetrical structure of benzene, the ring angle slightly increases by nearly 1.2° at the site of chlorine substitution and decreases approximately by the same amount 1.2° in the adjacent positions. A similar but in opposite way takes place due to the C„N substitution. These features are because of the decrease in the negative charge on C2 and C4, with shortening of the nearest CAC bonds, and increment in the negative charge on C1. The increase in the negative charge on C1 leads to a lengthening of the nearest CAC bonds.

Wavenumber calculation BN and their derivatives have been the subject of many experimental studies using the IR and Raman spectroscopic techniques. In these studies, the assignments proposed for the fundamental modes have been mainly based on correlation with normal modes in the benzene molecule and poor force field calculations. In the present work with 2,4-DCBN, the assignment of the experimental IR and Raman bands was based on DFT calculations and an accurate scaling procedure. For the assignment of the ring modes was followed the Varsanyi notation [52] for a 1-‘‘light’’-2,4-di‘‘heavy’’ tri-substituted benzene. This tri-substitution has been regarded by Varsanyi as the superposition of meta- and orthodisubstitutions. It is noted that a substituent is said to be light if the atom directly attached to a phenyl ring has an atomic mass less than 25 amu. Thus, m(CAX) stretching vibrations corresponding to the substituents are the normal modes 13, 7 (a and b), Table 3. The d(CAX) in-plane bending vibrations are the modes 15 and 9 (a and b), while the c(CAX) out-of-plane vibrations are the modes 17a and 10 (a and b).

Scaling the wavenumbers The reliable prediction of vibrational spectra is of considerable use in assigning the normal modes in a molecule. The selection of adequate quantum chemical methods and scaling procedures, remarkably reduce the risk in the assignment and can accurately determine the contribution of the different modes in an observed band. In addition to the error of the theoretical method used, the difference between the computed and experimental frequencies may be due to many different factors that are usually not even considered in the theory, such as anharmonicity, Fermi resonance and solvent effects. The introduction of scaling factors is capable of accounting for all these effects. To improve the computed wavenumbers in 2,4-DCBN molecule, several procedures can be carried out [42]. However, Table 4 shows the results obtained using only the scaling equation procedure, which employees the scaling equation calculated in benzene molecule. Figs. 2–5 show the experimental IR and Raman spectra, together with their simulated scaled spectra. The assignment of the

Table 3 Modes corresponding to the substituents in 2,4-DCBN. Modes

2,4-DCBN

Stretching In-plane bending Out-of-plane bending

7a, 7b, 13 9a, 9b, 15 10a, 10b, 17a

92

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

Table 4 Comparison of the calculated harmonic wavenumbers (x, cm1) at the B3LYP/6-31G** level, relative infrared intensities (A, %), relative Raman scattering activities (S, %), Raman depolarization ratios for plane (P) and unpolarized (U) incident light, reduced mass (l, amu), force constant (mdyn Å1, f), scaled wavenumbers (m, cm1) and experimental IR and FT-Raman wavenumbers obtained in 2,4-DCBN. Scaled m

Calculated values

x

A

S

P

U

l

f

3239 3232 3218 2354 1635 1595 1508 1419 1324 1284 1230 1171 1123 1074 971 887 844 839 717 705 624 594 589 503 431 407 361 347 244 198 171 132 85

1 0 0 21 100 23 91 30 0 1 2 8 58 29 0 17 64 30 0 14 21 1 9 5 1 5 3 1 1 1 0 5 5

17 29 12 100 33 1 2 1 1 3 9 4 4 1 0 0 1 1 0 2 0 0 1 1 0 2 1 1 0 0 0 1 0

0.35 0.26 0.87 0.41 0.66 0.36 0.28 0.42 0.88 0.38 0.23 0.19 0.15 0.24 1.00 1.00 0.15 1.00 1.00 0.22 0.87 0.44 1.00 1.00 1.00 0.37 0.84 0.38 1.00 0.99 1.00 0.98 1.00

0.48 0.38 0.92 0.54 0.77 0.49 0.41 0.55 0.93 0.52 0.34 0.29 0.23 0.36 1.00 1.00 0.23 1.00 1.00 0.33 0.92 0.58 1.00 1.00 1.00 0.51 0.90 0.52 1.00 0.99 1.00 0.99 1.00

1.09 1.10 1.09 12.68 6.35 7.66 2.62 3.26 11.30 1.38 3.10 1.42 3.16 3.26 1.31 1.37 8.19 1.38 4.84 7.32 8.53 7.76 4.31 7.73 3.67 12.33 10.04 11.01 7.94 19.19 6.11 14.96 10.19

0.16 0.16 0.16 1.00 0.24 0.28 0.08 0.09 0.28 0.03 0.07 0.03 0.06 0.05 0.02 0.02 0.08 0.01 0.04 0.05 0.05 0.03 0.02 0.03 0.01 0.03 0.02 0.02 0.01 0.01 0.00 0.00 0.00

3132 3125 3112 2281 1590 1551 1468 1382 1291 1252 1200 1144 1098 1050 951 871 829 824 707 696 618 589 584 501 432 409 365 351 252 208 182 145 100

IR KBr

Nujol

– 3070 m – 2230 s 1610 s 1580 s 1488 s 1382 s – 1265 w 1218 w 1132 m 1100 s 1075 m – 900 s 840 s 825 s – 686 s 620 s – 585 s 480 s 422 w 410 s 356 m 340 w 293 w, 250 m 218 s – – –

– – – – 1610 s 1576 s – – – 1252 w 1205 w 1135 m 1100 s 1067 m – 905 s, 870 m 832 s 815 s – 686 m 610 s – 570 s 500 s 422 w 406 m 360 sh – 290 s, 240 s 220 s – – –

FT-Raman

Characterization

3110 m 3072 vs 3035 m 2225 vs 1595 s 1568 w 1460 w 1370 vw 1275 w 1257 w 1215, 1189 m 1157, 1130 w 1070 vw 1020 vw 965 m 870 w 833 m 776 vw, br 742 vw 700 m – 600 w 575 m–s 475 m – 410 m 375 w 340 vw 240 w 218 m – 155 vs 105 m, 85 m

(100%) 2, m(C3AH) (100%) 20a, m(C5AH, C6AH) mainly in C5AH (100%) 20b, m(C6AH) (88%) m(C„N) + (11%) m(C1AC7) (93%) 8a, m(C@C) (97%) 8b, m(C@C) (92%) 19a, m(C@C) (91%) 19b, m(C@C) (96%) 14, m(C@C) (96%) 3, d(CAH) (84%) 18a, d(CAH) + (16%) m(C„N) (95%) 18b, d(CAH) (91%) 1, d(CCC) (92%) 12?, d(CCC,CH) (100%) 17b, c(C5AH, C6AH) (100%) 5, c(C3AH) (84%) 13, m(CAC7, CACL) + (16%) m(C„N) (97%) 11, c(C5AH, C6AH) (92%) 4, c(CCC) (67%)6b, d(CCC) + (17%) d(C„N) + (16%) m(CACL) (70%) 7b, m(CACL) + (30%) d(C„N) (64%) 6a, d(CCC) + (21%) d(C„N) + (15%) m(CACL) (85%) 16a, c(CCC) + (11%) c(C„N) (73%) 10a, c(CAC7) + (20%) c(C„N) (90%) 16b, c(CCC) + (10%) c(C„N) (90%) 7a, m(CACL) + (10%) d(C„N) (78%) 9b, d(C1AC7, CACL) + (22%) d(C„N) (70%) 15, d(CACL) + (30%) d(C„N) (33%) c(C„N) + (55%) c(CCC) +(12%) c(CACL) (65%) 9a, d(CACL) + (10%) d(C„N) + (25%) d(CCC) (32%) 10b, c(C2ACL) + (68%) c(CCC) (65%) d(C„N) + (25%) d(CACL) + (10%) c(CCC) (43%) 17a, c(C4ACL) + (35%) c(C„N) + (22%) c(CAH)

vs – Very strong; s – strong; m – medium; w – weak; vw – very weak; br – broad; m – stretching; d – in-plane bending; c – out-of-plane bending; s – torsion and x – wagging.

most important and intense Infrared/Raman bands was also included. Analysis of the vibrational wavenumbers Owing to its structure, the molecule of 2,4-DCBN belongs to the Cs symmetry point group. The 33 normal modes of vibration may be distributed between the two species (a0 and a00 ) of the Cs point group as: 23 a0 + 10 a00 . Table 4 collects the theoretical and experimental wavenumbers obtained by IR and Raman spectroscopy at the B3LYP/6-31G** level. The relatives IR and Raman intensities are obtained by normalizing the computed values to the intensity of the strongest line. Column 8th corresponds to the scaled values with the scaling equation: mscaled = 22.1 + 0.9543xcalculated. These values can be directly compared to the experimental IR bands obtained in KBr matrix, in Nujol and by FT-Raman spectroscopy. In the last column appears the characterization established by B3LYP for each calculated wavenumber in the ring modes and in the substituent modes. The % contribution of each mode, PED, to a computed wavenumber was also included. Contributions lower than 10% were not considered. Fig. 6 displays the atomic displacements corresponding to the ring normal modes with each computed wavenumber at the B3LYP/6-31G** level. These displacements are represented as xyz coordinates, in the standard orientation, that have been plotted to identify each vibration. At other levels of computations the differences observed in the plot have been small. The motions are only drawn in this Fig. 6 when the sum of the displacements on the X, Y and Z axis is higher than 0.07 on the carbon and nitrogen atoms, 0.04 on the chlorine atoms, and higher than 0.15 on

hydrogen atoms. Larger circles or arrows represent particular large displacements. The errors obtained in the predicted wavenumbers were very small, in general lower than 5%. These values are close to other BN derivatives studied by us earlier [25,37]. It demonstrates the good agreement between the scaled and the experimental wavenumbers. For a better discussion of the wavenumbers on the normal modes, the study has been divided into the following two sections, the phenyl ring modes and the CAX modes. The assignments for several of the phenyl ring modes are obvious and require no further discussion, therefore we concentrate here only on some important normal modes giving more specific information about 2,4-DCBN molecule. An analysis of the vibrations is as follows: Normal vibrations of the phenyl ring. CAH modes: CAH stretchings are attributed to the modes 2, 20a and 20b, and in these modes the displacement vector is mainly centered in one CAH bond. It is noted a close range (20 cm1) for the three calculated (scaled) CAH stretching wavenumbers, appearing that corresponding to the C3AH bond at higher wavenumber than to the C5AH and C6AH bonds. These modes are not coupled with other vibrations and thus they are easily identified. The range predicted for these modes (3000–3100 cm1) are in accordance to our scaled and experimental wavenumbers. In-plane bendings are the modes 18a, 18b and 3. Mode 18a appears slightly coupled with m(C„N) stretching vibration, increasing therefore its wavenumber to the scaled value of 1200 cm1, slightly out of the range established [52] for this

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

Fig. 2. A comparison of the theoretical and experimental IR spectra of 2,4-DCBN in the 4000–2000 cm1 spectral range. (a) The experimental FT-IR spectrum of the molecule in KBr. (b) The experimental FT-IR spectrum of the molecule in Nujol matrix. (c) The theoretical IR spectrum of the molecule, which was plotted over the scaled wavenumbers produced using the equation mscaled = 22.1 + 0.9543mcalculated.

vibration 1120–1170 cm1. Mode 18b is computed with higher IR intensity and lower Raman activity than mode 18a. This fact is in agreement with the experimental IR bands with medium intensity at 1132 cm1 (18b) and weak value at 1218 cm1 (18a), and with the Raman bands with weak intensity at 1157 cm1 (18b) and medium value at 1215 cm1 (18a). The IR bands at 1252 and 1265 cm1 and the Raman line at 1257 cm1 were well assigned to mode 3, scaled at 1252 cm1. In 2,4-dichlorotoluene this mode was assigned to the experimental band at 1278 cm1, in accordance to our results. The three out-of-plane bending modes 17b, 5 and 11 are something insensitive to substitution and thus they can be easily characterized and assigned with very low error. They appear identified at the 800–990 cm1 range, as almost pure modes, and in good accordance with our results. Mode 5 corresponds to C3AH, while modes 17b and 11 are associated with C5AH and C6AH bonds, respectively. These assignments find further support from the literature values [37]. C@C modes: In general they appear strongly coupled with the CAH modes. The obtained PED values have confirmed that the CC bond stretchings of the phenyl ring are generally in strong coupling with its CH bending vibrations. According to the same theoretical data, the tangential vibrations 8a, 8b, 19a, 19b and 14 belong to the stretching group. In 1,2,4-trisubstitution the frequency of 8b component is reported [52] to be higher than that of 8a, while in other cases it is reverse. Similarly, it appears with modes 19a and 19b. However, according to the displacement vectors of the atoms

93

Fig. 3. A comparison of the theoretical and experimental IR spectra of 2,4-DCBN in the 2000–400 cm1 spectral range. (a) The experimental FT-IR spectrum of the molecule in KBr. (b) The experimental FT-IR spectrum of the molecule in Nujol matrix. (c) The theoretical IR spectrum of the molecule, which was plotted over the scaled wavenumbers produced using the equation mscaled = 22.1 + 0.9543mcalculated from benzene molecule.

Fig. 4. A comparison of the theoretical and experimental Raman spectra of 2,4DCBN in the 3500–2000 cm1 spectral range. (a) The experimental FT-Raman spectrum of the molecule in solid phase. (b) The theoretical Raman spectrum of the molecule, which was plotted over the scaled wavenumbers produced using the equation mscaled = 22.1 + 0.9543mcalculated.

obtained by B3LYP, and with the characterization of the normal modes reported by the same authors [52], we have reversed this assignment with higher wavenumber in the ‘‘a’’ than ‘‘b’’ modes,

94

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

In Wilson notation [53] the out-of-plane skeletal vibrations correspond to 4, 16a and 16b of benzene. The frequency of the normal mode 4 is rather insensitive to substitution, which facilitate its identification. Thus, it was calculated at 717 cm1 in 2,4-DCBN and at 703 cm1 in 2,4-DFBN vs. 718 cm1 in benzene molecule. Modes 16a and 16b appear coupled with out-of-plane C„N vibrations that cause a frequency decrease as compared to those calculated in 2,4-DFBN at 632 cm1 and 454 cm1, respectively. The range predicted for these modes is in accordance to our scaled and experimental wavenumbers.

Fig. 5. A comparison of the theoretical and experimental Raman spectra of 2,4DCBN in the 2000–0 cm1 spectral range. (a) The experimental FT-Raman spectrum of the molecule in solid phase. (b) The theoretical Raman spectrum of the molecule, which was plotted over the scaled wavenumbers produced using the equation mscaled = 22.1 + 0.9543mcalculated.

Table 4. The frequency of these pairs ‘‘8’’ and ‘‘19’’ is decreased with increment of the halogen atom masses. e.g. mode 8a is calculated at 1672 cm1 in 2,4-DFBN vs. 1635 cm1 in 2,4-DCBN; 1635 cm1 (8b) vs. 1595 cm1, 1547 cm1 (19a) vs. 1508 cm1, and 1482 cm1 (19b) vs. 1419 cm1. Similar decrease is observed in mode 14 with the increment of the halogen masses. According to Scherer [52] modes 19a and 19b couples very strongly with vibrational pair 18, with similar directions of the motions of the hydrogen atoms in both modes. The very strong IR intensity calculated for mode 19a (the second strongest of the spectrum) is in accordance to that observed experimentally. The assignment of mode 14, which is also known as the Kekule ring stretching mode, is usually very difficult as one of the CAH bending vibrations appears in its vicinity. Radial skeletal vibrations correspond to 1, 12, 6a and 6b of benzene. Modes 1 and 12 have been characterized as pure modes with strong IR intensity. The ring breathing (mode 1) and the trigonal planar ring bending (mode 12) are the most widely discussed modes in the literature [32,33,37]. These modes are drastically affected in magnitudes upon substitution. CAX stretching vibrations couple strongly with modes 1,6a and 6b. Thus, for a 2,4-di-‘‘heavy’’ substituent modes 1 and 12 are expected in the 1020–1110 and 500–650 cm1 ranges, respectively, instead of the reported ranges at 630–740 and 685–800 cm1 for a 2,4-di-‘‘light’’ substituent. Mode 1 was calculated at 1223 cm1 (scaled at 1098 cm1) in excellent accordance to the strong IR band in KBr matrix and in Nujol at 1100 cm1, to the Raman line at 1070 cm1, and as well to the range reported for this mode [52]. We have doubts in the assignment of mode 12 to the scaled wavenumber at 1050 cm1. The displacement vectors observed correspond to this mode 12, strongly coupled with CAH bending modes, but its wavenumber is out-of the range reported for this mode. CACl stretching and C„N bending vibrations couple with modes 6a and 6b. Due to this coupling, modes 6a and 6b appear characterized out of the range reported for these modes [52]. Mode 6a was scaled at 589 cm1 in accordance to the Raman line at 600 cm1, while mode 6b was scaled at 696 cm1 and related to the IR band at 686 cm1.

Vibrations of Nitrile (C„N) group. The C„N group involves three vibrations, namely, m(C„N) stretching, d(C„N) in-plane, and c(C„N) out-of-plane bending. The geometry of the cyano group is affected insignificantly by new substituents on the phenyl ring. Hence, the vibrational wavenumbers of the cyano group remains almost unchanged or very little from BN molecule [25]. The stretching mode is highly localized on the C„N bond with a Potential Energy Distribution (PED) of 88%. The characteristic experimental wavenumber of this mode in BNs falls ca. 2200– 2300 cm1 with IR intensity which varies from medium-weak to strong depending on the substituents. Electron-withdrawing groups, such as chlorine atoms, decrease the IR band intensity and increase the wavenumber value to the higher limit of the characteristic spectral region. In 2,4-DCBN this mode is calculated at 2354 cm1 (vs. 2349 cm1 in BN) with medium IR intensity and very strong Raman activity, the highest of the spectra. These results are in good agreement with the experimental spectra, and with the data reported in others BNs [36], because of the intensity is enhanced by the conjugation of the aromatic ring. These features appear very well reproduced in Figs. 2 and 4, where m(C„N) mode is clearly identified. Contributions of the C„N bending in-plane mode are observed in the calculated bands at 705, 624, 594 cm1, and in many of the bands lower than 450 cm1, while the out-of-plane modes are identified in the bands at 589, 503, 431 and 297 cm1. The main contribution for the d(C„N) mode is observed in the scaled band at 145 cm1 with medium IR intensity and weak Raman intensity, and closely related to the very strong Raman line at 155 cm1. The main contribution for c(C„N) mode is observed in the scaled bands at 252 and 100 cm1, well related to the Raman lines at 240 and 105 cm1. Electron-withdrawing groups, such as chlorine atoms, decrease the wavenumber of the bending modes. Thus, the in-plane mode is calculated at 132 cm1 vs. 165 cm1 in BN, while the out-of-plane mode is calculated at 85 cm1 vs. 147 cm1 in BN. CAX vibrations. In 2,4-DCBN the m(CAX) stretching vibrations corresponding to the substituents are the normal modes 13, 7a and 7b, Table 3. Their wavenumbers depend on the mass and bond strength of the substituents. When the substituent is ‘‘light’’, the stretching vibrations may be practically localized to the carbon atom carrying the substituent and in a lesser degree, to the neighboring ones. Thus, mode 13 was characterized as m(C1AC7) and identified tentatively in the scaled wavenumber at 696 cm1. Some contribution of this mode is observed in the scaled value at 1200 cm1. In Varsanyi’s book this mode is reported in the 1100–1260 cm1 range. CACl stretchings correspond to 7a and 7b modes and they were calculated below 500 cm1. Mode 7b is reported in the 280–470 cm1 range and 7a in the 200–470 cm1 range [52]. Both have a strong CCC contribution. d(CAX) in-plane bending vibrations correspond to the modes 9a, 9b and 15. Mode 9b was characterized as d(C1AC7) while modes 9a and 9b as d(CACl). Due to the strong coupling with CCC and C„N modes, their identification was not clear. They appear reported [45] in the 150–310, 200–440 and 120–280 cm1

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

95

Fig. 6. Characterization of the phenyl ring normal modes computed in 2,4-DCBN molecule at the B3LYP/6-31G** level, with the main assignment and the wavenumber in which they were computed.

ranges corresponding to modes 9a, 9b and 15, respectively. Modes 9a and 9b were assigned to the IR bands at 218 and 356 cm1, respectively, and mode 15 to the band at 340 cm1, in accordance to our scaled values. Mode 15 is characterized as an in-plane rocking motion of the ring and due to the strong coupling with CCC and C„N bending modes its frequency appears out-of the range established for this mode. CAX out-of-plane vibrations correspond to modes 10a, 10b and 17a, Table 3. Modes 10b and 17a were related to CACl substituent, and thus they were scaled below 200 cm1, while mode 10a appears related to C1AC7„, and scaled at 501 cm1. These modes are also strongly coupled with the CH, CCC and C„N out-of-plane

vibrations, making its identification difficult. However, the scaling equation used in the present work, permit us to assign the experimental bands with high accuracy. The 200–350, 100–250 and 80– 160 cm1 ranges reported [52] for the modes 10a, 10b and 17a, respectively, are in accordance to our scaled values at 501, 182 and 100 cm1, respectively, and as well to the experimental values. Other molecular properties The values of the NBO atomic charges with the B3LYP and MP2 methods are listed in Table 2. A general accordance is observed between both methods, with some specific differences. It is noted

96

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97

Table 5 Theoretical computed total energies and Gibbs Free energy (AU), rotational constants (GHz), entropies (cal mol1 K1) and dipole moments (Debyes) at the B3LYP and MP2/ 6-31G** level in 2,4-DCBN molecule. Parameters

B3LYP

MP2

BNd

Total energy + ZPE (AU) Gibbs Free energy (AU) Rotational constants (GHz)

.601464a .636080a 1.87 0.60 0.45

.475399b .510218b 1.89 0.60 0.46

.400771c .431048c 5.65 1.54 1.21

92.74 41.32 30.82 20.60 3.57

93.78 41.32 30.81 21.66 3.93

78.56 39.8 27.8 10.9 4.55

Entropy (cal mol1 K1) Total Translational Rotational Vibrational Dipole moments (Debyes) a b c d

1243. 1241. (without ZPE). 324. In benzonitrile molecule at MP2 level.

energy values are found to be: HOMO = 7.38 eV; MO = 1.99 eV; HOMO–LUMO energy gap = 5.39 eV

LU-

Summary and conclusions The most important findings of the present work are the following: (1) The equilibrium geometry, the harmonic wavenumbers and thermodynamical data for 2,4-DCBN have been obtained from MP2 ab initio method and B3LYP hybrid type DFT method. (2) The small distortions of the benzene ring are explained in terms of the change in hybridization affected by the substituents at the carbon site to which they are appended. (3) Present paper clearly represents an improvement of the already published results, correcting mislay in the assignments by using B3LYP DFT methods and an accurate scaling procedure. (4) To achieve a reliable prediction and improve the accuracy in the assignment of the wavenumbers was used a scaling equation procedure. Thus, very low errors were obtained with the scaled values. Moreover, the assignments of the observed bands in IR and Raman spectra find support from calculations and are consistent with the established ranges of the vibrations. Therefore, our assignments seem to be correct.

Acknowledgements

Fig. 7. HOMO and LUMO plots of 2,4-DCBN at B3LYP/6-31G** level.

that the negative charge is mainly localized on the nitrogen atom, which produces the largest positive charge on the C7 atom. Chlorine atoms have a very small positive charge, while on the appended C2 and C4 atoms is the negative charge very small. Thermodynamical parameters for 2,4-DCBN were also calculated and collected in Table 5. Scale factors suggested by us earlier have been used without any modification [40,41] for an accurate prediction in determining the Zero-Point Vibration Energies (ZPVE), and the entropy, Svib(T), Table 5. The rotational constants are very small as compared to BN. Because of symmetry of the molecule, and the small charge on C2, C4 and the chlorine atoms, the dipole moment is mainly along the C1AC7„N axis. Its value is slightly smaller than that calculated in 2,4-DFBN (4.00 D) and in BN (4.55 D), as well as the experimental value reported in BN (4.53 D) [54]. As is known from the literature, frontier orbital electron densities of atoms can be used as an efficient tool in a detailed characterization of donor–acceptor interactions [55]. HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) energy analysis of 2,4-DCBN has been carried out by density functional theory method at B3LYP/6-31G** level. The orbital energy analysis provides insights into the nature of this transition within the orbitals, these energy values of LUMO, HOMO and their energy gap reflects the chemical activity of the molecule [56]. These orbital plots (43a ? 44a) are generated by Gaussian cube files containing the HOMO and LUMO electron densities (isovalue = 0.02) and are shown in Fig. 7. The HOMO, LUMO orbital

The authors, V.K.R. and H.J., are grateful to UGC, New Delhi, India for the award of MRP (Grant No. F. No. 35-2/2008). One of the authors (V.K.R.) is also grateful to Prof. Rakesh Kumar, HoD Physics and Sri V.C. Goel, Vice Chancellor, CCS University, Meerut for providing necessary facilities and encouragement during the course of this work. Thanks are also due to Prof. O.F. Nielsen, Department of Chemistry, University of Denmark, for recording the FT-IR and FT-Raman spectra of the compound. References [1] M. Alcolea Palafox, V.K. Rastogi, L. Mittal, Int. J. Quantum Chem. 94 (2003) 189–204. [2] M. Alcolea Palafox, V.K. Rastogi, L. Mittal, Int. J. Quantum Chem. 106 (2006) 1885–1901. [3] M. Alcolea Palafox, V.K. Rastogi, Asian J. Phys. 22 (3) (2013) 43–73. [4] J.A. Butera, M.M. Antane, S.A. Antane, T.M. Argentieri, C. Freeden, R.F. Graceffa, B.H. Hirth, D. Jenkins, J.R. Lennox, E. Matelan, N.W. Norton, D. Qluagliato, J.H. Sheldon, W. Spinelli, D. Warga, A. Wojdan, M. Woods, J. Med. Chem. 43 (2000) 1187–1202. [5] K. Kagara, S. Goto, M. Ichihara, J. Synt, Org. Chem. Jpn. 57 (1999) 415. [6] S.L. Cheav, S. Kirkiacharian, F. Pieri, D. Poisson, Ann. Pharm. Fr. 56 (1998) 205. [7] F. Xie, X. Zhou, M. Behr, C. Fang, Y. Horii, J. Gu, K. Kannan, X. Ding, Toxicol. Appl. Pharm. 249 (1) (2010) 101–106. [8] J.R. Anderson, W.S. Barnes, P. Bedinger, J. Plant Physiol. 159 (1) (2002) 61. [9] N. Orologas, S.G. Delivopoulos, A. Dimopoulou, I. Tsekos, Phycologia 44 (5) (2005) 465–476. [10] L. Peng, L. Zhang, X. Cheng, L.-S. Fan, H.-Q. Hao, Plant Biol. 15 (2) (2013) 405– 414. [11] F. Xie, J. D’Agostino, X. Zhou, X. Ding, Chem. Res. Toxicol. 26 (3) (2013) 388– 398. [12] S. Dheivamalar, V. Silambarasan, Spectrochim. Acta A 96 (2012) 480–484. [13] V. Arjunan, K. Carthigayan, S. Periandy, K. Balamurugan, S. Mohan, Spectrochim. Acta A 98 (2012) 156–169. [14] V. Krishnakumar, N. Surumbarkuzhali, S. Muthunatesan, Spectrochim. Acta A 71 (5) (2009) 1810–1813. [15] V.K. Rastogi, H.P. Mital, S.N. Sharma, S. Chattopadhyay, Ind. J. Phys. 65B (1991) 356. [16] V.K. Rastogi, D.K. Jain, Y.C. Sharma, Asian J. Chem. 3 (1991) 113. [17] G. Bottura, S. Arora, J.K. Gupta, V.K. Rastogi, Asian J. Phys. 1 (1992) 58. [18] N. Misra, O. Prasad, L. Sinha, A. Pandey, J. Molec, Struct. (Theochem) 822 (1–3) (2007) 45–47.

D. Kattan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 89–97 [19] A.P. Kumar, G.R. Rao, Spectrochim. Acta A 53 (1997). 2033, 2041, 2049. [20] H. Kamoto, H. Inishi, Y. Nakamura, S. Kohtani, R. Nakayaki, J. Chem. Phys. 260 (2000) 193. [21] V.K. Rastogi, M. Alcolea Palafox, B. Lal, V. Jain, Ind. J. Pure Appl. Phys. 38 (2000) 564–569. [22] N.P. Singh, R.A. Yadav, Ind. J. Phys. 75B (2001) 347. [23] M. Alcolea Palafox, V.K. Rastogi, J.K. Vats, J. Raman Spectrosc. 37 (2006) 85–99. [24] V.K. Rastogi, S. Singhal, G.R. Rao, M. Alcolea Palafox, Natl. Acad. Sci. Lett. 29 (2006) 445–454. [25] V.K. Rastogi, M. Alcolea Palafox, S. Singhal, S.P. Ojha, W. Kiefer, Int. J. Quantum Chem. 7 (2007) 1099–1114. [26] V. Krishnakumar, S. Dheivamalar, Spectrochim. Acta A 71 (2) (2008) 465–470. [27] N. Sundaraganesan, C. Meganathan, B.D. Joshua, P. Mani, A. Jayaprakash, Spectrochim. Acta A 71 (3) (2008) 1134–1139. [28] S. Sudha, N. Sundaraganesan, M. Kurt, M. Cinar, M. Karabacak, J. Mol. Struct. 985 (2–3) (2011) 148–156. [29] V. Kumar, Y. Panikar, M. Alcolea Palafox, J.K. Vats, I. Kostova, V.K. Rastogi, Ind. J. Pure Appl. Phys. 48 (2) (2010) 85–94. [30] N. Sundaraganesan, G. Elango, S. Sebastian, P. Subramani, Ind. J. Pure Appl. Phys. 47 (7) (2009) 481–490. [31] A.R. Krishnan, H. Saleem, S. Subashchandrabose, N. Sundaraganesan, S. Sebastain, Spectrochim. Acta A 78 (2) (2011) 582–589. [32] V.K. Rastogi, C.B. Arora, S.K. Singhal, D.N. Singh, R.A. Yadav, Spectrochim. Acta A 53 (1997) 2505. [33] V.K. Rastogi, M. Alcolea Palafox, S. Singh, S.K. Singhal, Asian J. Phys. 7 (1998) 229–249. [34] V.K. Rastogi, V. Jain, M. Alcolea Palafox, D.N. Singh, R.A. Yadav, Spectrochim. Acta A 57 (2001) 209–216. [35] M. Alcolea Palafox, V.K. Rastogi, C. Singh, R.P. Tanwar, Spectrochim. Acta A 57 (2001) 2373–2389. [36] V.K. Rastogi, M. Alcolea Palafox, R.P. Tanwar, L. Mittal, Spectrochim. Acta A 58 (2002) 1987–2004. [37] M. Alcolea Palafox, V. Bena Jothy, S. Singhal, I. Hubert Joe, S. Kumar, V.K. Rastogi, Spectrochim. Acta A 116 (2013) 509–517. [38] V.K. Rastogi, M. Alcolea Palafox, R. Tomar, Upama Singh, Spectrochim. Acta A 110 (2013) 458–470.

97

[39] J.M. Seminario, P. Politzer (Eds.), Modern Density Functional Theory: A Tool for Chemistry, vol. 2, Elsevier, Amsterdam, 1995. [40] M. Alcolea Palafox, Recent Res. Dev. Phys. Chem. 2 (1998) 213–232. [41] M. Alcolea Palafox, Int. J. Quantum Chem. 77 (2000) 661–684. [42] M. Alcolea Palafox, V.K. Rastogi, Spectrochim. Acta A 58 (2002) 411–440. [43] W.J. Hehre, L. Radom, P.v.R. Schleyer, J.A. Pople, Ab initio Molecular Orbital Theory, Wiley & Sons, New York, USA, 1986. [44] A.D. Becke, Phys. Rev. A 38 (1988) 3098. [45] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B37 (1988) 785. [46] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision B.04 Inc, Pittsburgh PA, 2003. [47] V.K. Rastogi, S. Singhal, M. Alcolea Palafox, G.R. Rao, Ind. J. Phys. 84 (2010) 151. [48] J.E. Carpenter, F. Weinhold, J. Mol. Struct. (Theochem) 169 (1988) 41. [49] A.E. Reed, L.A. Curtiss, F. Weinhold, Chem. Rev. 88 (1988) 899. [50] J. Casado, L. Nygaard, G.O. Sorensen, J. Mol. Struct. 8 (1971) 211. [51] J.W. Finley, P.J. Stephens, J. Mol. Struct. (Theochem) 357 (1995) 225. [52] G. Varsanyi, Assignments for Vibrational Spectra of Seven Hundred Benzene Derivatives, vol. 1, Adam Hilger, London, 1974, p. 280. [53] E.B. Wilson Jr., Phys. Rev. 45 (1934) 706. [54] M.F. Vuks, Zh. Fiz. Khim. 52 (1978) 134. [55] A.R. Katritzky, V.S. Lobanov, M. Karelson, Chem. Soc. Rev. 24 (1995) 279–287. [56] D.F.V. Lewis, C. Ioannides, D.V. Parke, Xenobiotica 24 (1994) 401–408.