Computational studies on the structure and vibrational spectra of 2-hydroxy-5-methyl-3-nitropyridine
Hari Ji Singh* & Priyanka Srivastava
Department of Chemistry, DDU Gorakhpur University, Gorakhpur 273 009
*E-mail: hari_singh81@hotmail.com
Received 21 July 2008; revised 13 January 2009; accepted 7 May 2009
The molecular structure and vibrational spectra of 2-hydroxy-5-methyl-3-nitropyridine have been investigated by Hartree-Fock and Density Functional Theory (DFT) using standard B3LYP functional and 6-311G(d) and 6-311G(3d,2p) basis sets. The results of the calculations are applied to simulate infrared spectra of the title compound which showed good agreement with the experimentally determined data. It has been found that both methods yield consistent data for the geometric parameters but DFT with a basis set of 6-311G(3d,2p) yielded vibrational frequencies much closer to the experimental data. Computed values at DFT(B3LYP)/6-311G(3d,2p) have been analyzed and their characterization was made with the help of Gaussview visualization program utilizing the data obtained from the Gaussian 03 calculation. A few of the discrepancies observed between the experimental and computed data of vibrational frequencies and their assignments have also been discussed.
Keywords: Vibrational analysis, Normal mode frequencies, 2-hydroxy-5-methyl-3-nitropyridine, Infrared spectra
1 Introduction
N-heterocycles, especially, pyridine ring system occurs in the structures of a wide variety of natural products, pharmaceuticals and agrochemical compounds. It plays a central role in the structure and properties of nucleic acids. Its biological importance is substantiated by the fact that pyridine ring plays a very important role in the fundamental metabolism in two ways: (i) as an oxidizing system by effective hydride abstraction in nicotinamide adenine dinucleotide (NAD+) and (ii) in dehydrogenase enzymes, and in transmination reactions, an important aspect of amino acid metabolism1-3, as vitamin B6.
Nitrogen in the pyridine ring has a lone pair of electrons which is not delocalized with the aromatic π-electron system and is easily available for protonation. The basicity becomes more pronounced if electron-donating groups are present on the ring at adjacent positions of N because of the increased electron density on the nitrogen atom4-6. Intensive studies on pyridine and substituted pyridines have been made in the past keeping in view their industrial importance both as a fundamental building block and as a solvent and reagent in organic syntheses7. The spectroscopic study of N-heterocyclic molecules including substituted pyridines has become quite interesting as they are the constituents8-11 of DNA and RNA.
Recently, a spectroscopic study of a substituted pyridine (2-hydroxy-5-methyl-3-nitropyridine) has been made12 and its various vibrational modes have been assigned using a simplified picture of the molecule with −CH3 and −NO2 groups as unit entities.
Due to existence of many contaminated vibrations, it would be difficult to assign the observed frequencies to a particular mode. The ab-initio and density functional theory13 (DFT) methods have become a powerful tool for the investigation of molecular structure and vibrational spectra. Supplemented by a visualization program, the assignments can accurately be made. Literature survey shows that no computational studies have been done so far on the titled compound. The present work has been performed with a view point of getting the vibrational frequencies on an optimized geometry of the titled compound. These frequencies are analyzed and compared with the experimental data12. Attempts have been made to find out an optimum method using a reasonable basis set to get a close agreement between the computed and the experimental data.
2 Computational Details
Geometry optimization and vibrational frequencies of 2-hydroxy-5-methyl-3-nitropyridine were calculated at the Hartree-Fock14 and DFT levels with B3LYP (Becke-Lee-Yang-Parr three parameters) hybrid
Table 1 — Optimized geometrical parameters of 2-hydroxy-5-methyl-3-nitropyridine obtained by HF and DFT calculation
Parameters HF DFT
6-311G(d) 6-311G(3d,2p) 6-311G(d) 6-311G(3d,2p) Bond length (Ǻ)
N1-C2 1.318 1.317 1.331 1.330
N1-C6 1.308 1.307 1.326 1.326
C2-C3 1.401 1.399 1.418 1.417
C2-O11 1.312 1.309 1.328 1.324
C3-N10 1.447 1.444 1.456 1.451
C3-C4 1.392 1.389 1.395 1.392
C4-C5 1.371 1.370 1.385 1.384
C4-H9 1.071 1.070 1.083 1.080
C5-C6 1.397 1.395 1.405 1.403
C5-C8 1.509 1.508 1.507 1.507
C6-H7 1.076 1.076 1.089 1.085
C8-H15 1.083 1.081 1.092 1.089
C8-H16 1.085 1.083 1.095 1.092
C8-H17 1.085 1.083 1.092 1.089
N10-O13 1.199 1.199 1.244 1.244
N10-O14 1.182 1.182 1.218 1.218
O11-H12 0.945 0.947 0.979 0.982
Bond angle (degree)
C6-N1-C2 119.637 119.485 119.089 118.954
N1-C2-C3 119.903 120.001 120.193 120.284
N1-C2-O11 114.966 115.243 115.736 116.174
C3-C2-O11 125.131 124.755 124.071 123.542
C2-C3-C4 119.668 119.694 119.741 119.791
C2-C3-N10 122.294 122.109 121.985 121.661
C4-C3-N10 118.038 118.196 118.273 118.547
C3-C4-C5 119.983 119.939 119.810 119.718
C3-C4-H9 118.546 118.481 118.353 118.271
C5-C4-H9 121.471 121.579 121.836 122.011
C4-C5-C6 115.325 115.367 115.764 115.847
C4-C5-C8 123.244 123.170 122.610 122.535
C6-C5-C8 121.431 121.462 121.620 121.613
C5-C6-N1 125.483 125.513 125.402 125.405
C5-C6-H7 119.443 119.281 119.467 119.274
N1-C6-H7 115.074 115.207 115.131 115.321
C2-O11-H12 111.327 109.881 108.006 106.738
C3-N10-O13 117.971 118.277 117.863 118.095
C3-N10-O14 118.349 118.309 118.957 119.034
O13-N10-O14 123.679 123.414 123.180 122.872
C5-C8-H15 110.985 110.764 111.510 111.306
C5-C8-H16 111.377 111.267 111.357 111.260
C5-C8-H17 111.377 111.268 111.374 111.106
H15-C8-H16 107.631 107.817 107.365 107.509
H16-C8-H17 107.656 107.750 107.277 107.482
H17-C8-H115 107.631 107.817 107.740 107.997
harmonic vibrational frequency calculations. The symmetries of the vibrational modes were determined using the standard procedure18. Utilizing Gaussview program19 and symmetry considerations, assignment of the vibrational frequencies was made with a high degree of confidence. Analysis of the computed values shows significant variation in the data obtained at different level of theories. Percentage deviations of the computed frequencies from the experimental data are plotted in Fig. 2 for HF and DFT methods using two different basis sets. Results show that there is a less variation in the case of DFT as compared to HF values. Fig. 2 also shows that the variations are more pronounced at lower frequencies. These low frequency data may be due to hindered rotation and may not represent the true vibrations. This trend is the same for both the methods used in the present study.
Results plotted in Fig. 2 further reveal that DFT calculations using a basis set incorporating polarized functions yielded results that are in a better agreement with the experimental data. Therefore, a detailed analysis of the vibrational frequencies has been performed with the computed data obtained at DFT (B3LYP)/6-311G(3d,2p) and the results are presented in Table 2. The relative intensities of IR and Raman active vibrations are calculated from the computed data and these are also listed in Table 2. Results show the absence of any imaginary frequency inferring that the optimized structure is a stable minimum on its potential energy surface (PES).
The computed values have been utilized to simulate IR spectra as shown in Fig. 3. The result shows an agreement with the experimentally observed spectra of the titled compound. A comparison of the observed spectra and the spectra based on the computational data reveals the agreement for the prominent peaks having high relative intensities. There is an agreement with the assignments for these prominent peaks.
Fig. 2 — Plot of percentage deviations of computed vibrational frequencies of 2-hydroxy-5-methyl-3-nitropyridine at different level of theories — Experimental data [Ref.12]; •, HF method;
ο, DFT (B3LYP) Theory a-6-311G (d); b-6-311G (3d,2p)
However, the assignments of some of the peaks made by Yadav et al.12 are in disagreement with the computed data. During the present study, visualization of frequencies in 3D has been made by Gassview. With this program, assignments of vibrations can be made with a high degree of certainity. The calculations show that 2-hydroxy- 5-methyl-3-nitropyridine has a planar structure of C1
point group symmetry and hence, all the calculated frequency transforming to the same symmetry species (A). The existence of C1 point group shows that no symmetry exists in the title molecule. The molecule consists of 17 atoms and expected to have 45 normal modes of fundamental vibrations out of which 16 vibrational modes are of stretching type and 29 are of bending type. The vibrational frequencies calculated using DFT methods are known to be overestimated
Fig. 1 — Optimized geometry of 2-hydroxy-5-methyl- 3-nitropyridine obtained at DFT (B3LYP)/6-311G(3d,2p)
probably because of the neglect of anharmonicity of vibrations in the real systems. Accepted values of scaling factors for DFT is 0.96 and it has been used to correct the frequency values20. Scaled frequencies are recorded in Table 2. The relative values of IR
intensities and Raman scattering activities corresponding to various vibrational frequencies were calculated from the absolute values obtained during computation. Absolute and relative values are also listed in Table 2. The assignment of the computed
ν5 2960(2911) 13 5 79 32 υasym(CH3)
ν6 2910(2885) 21 8 245 100 υsym(CH3)
ν7 1597 139 55 12 5 υ(CCC)+β(CNC)
ν8 1547(1550) 85 33 23 9 υ(C-C)+ β(O-H)
ν9 1534(1540) 242 96 28 11 υasym(NO2)
ν10 1447 251 100 5 2 β(C-H)+υ(C-OH)
ν11 1446 15 6 6 2 τ(CH3)
ν12 1435 8 3 10 4 ρ(CH3)
ν13 1392 184 73 5 2 β(O-H)+β(C-H)
ν14 1369 2 1 17 7 ω(CH3)
ν15 1351 49 20 7 9 β(C-H)+β(O-H)
ν16 1320 32 13 5 2 β(C-H)+β(C-N)
ν17 1274 139 55 121 49 υ(C-NO2)+β(C-H)
ν18 1236 98 39 5 2 β(C-H)+β(O-H)
ν19 1211 135 54 105 43 υ(C-CH3)+υ(C-NO2)
ν20 1145(1159) 32 13 2 1 β(C-H)+β(O-H)
ν21 1063(1077) 10 4 0 0 β(C-H)+υ(C-NO2)
ν22 1030 6 2 1 0 ω(CH3)
ν23 981 3 1 1 0 t(CH3)
ν24 950 0 0 0 0 γ(C-H)
ν25 932 22 9 10 4 υ(C-CH3)+υ(C-NO2)
ν26 897(829) 6 2 0 0 γ(C-H)
ν27 821 17 7 27 11 s(NO2)
ν28 780 72 29 1 0 γ(O-H)
ν29 763 20 8 1 0 γ(O-H)
ν30 760 18 7 6 2 υ(C-CH3)+υ(C-NO2)
ν31 723 0 0 0 0 γ(C-NO2)+ γ(C-N)
ν32 665 2 1 6 2 υ(C-N)+υ(C-H)
ν33 566 5 2 2 1 β(C-N)+β(O-H)
ν34 533 7 3 0 0 γ(C-H)
ν35 466 2 1 9 4 υ(C-CH3)
ν36 428 3 1 4 2 β(O-H)+β(N-O)
ν37 423 3 1 1 0 γ(C-H)+ γ(C-N)
ν38 384 1 0 2 1 υ(C-H)+υ(C-NO2)
ν39 335(494) 1 0 1 0 γ(C-C)+ γ(CH3)
ν40 328 8 3 1 0 β(O-H)+β(CH3)
ν41 221 2 1 3 1 β(CH3)+β(NO2)
ν42 160 1 0 0 0 γ(C-C)+ γ(CH3)
ν43 127 1 0 1 0 γ(O-H)+ γ(C-H)
ν44 77(83) 1 0 0 0 t(NO2)
ν45 23 1 0 0 0 ρ(CH3)
aUnits of frequency are cm−1, bUnits of IR intensity are Km/mole, cUnits of Raman scattering activity are Ǻ4/amu.
υ-stretching; υasym-asymmetrical stretching; υsym-symmetrical stretching; β-in-plane-bending; γ-out-of-plane bending;
ω-wagging; ρ-rocking; τ-torsion; t-twisting; s-scissor. Values in bracket refer to the experimental data12.
frequencies is done by visualizing the vibrations in 3D using Gaussview (Table 2). A few of the computed vibrational frequencies showing high relative intensities in IR spectra and the discrepancies observed between experimental and computational data are discussed here.
C-H vibrations — There are two hydrogen atoms that are attached to the ring carbons. The calculation shows two C-H stretching vibrations at 3078 and 3010 cm−1 and these are in close agreement with the experimental values of 3068 and 3024 cm−1 as reported by Yadav et al12. The C-H in-plane-bending vibrations are visualized at 1447, 1392, 1351, 1320, 1274, 1236, 1145 and 1063 cm−1. Visualization of these frequencies reveals that these vibrations are contaminated by C-N, C-NO2 and C-OH stretchings.
The corresponding experimental values for C-H in-plane-bending12 have been reported at 1132 and 1077 cm−1. The computed values for the C-H out-of- plane bending vibrations are at 950, 897 and 533 cm−1. The experimental values12 have been reported at 829 and 807 cm−1.
C-N vibrations — The identification of C-N vibrations in the heterocyclic ring is found to be a difficult task because of the mixing of several vibrations. Using Gaussview visualization program we find that the C-N ring stretching corresponds to the frequency of 1320 cm−1 whereas Yadav et al.12 assigned a value of 1211 cm−1 for the above C-N stretching during their experimental studies. Our calculation followed by Gaussview visualization showed the 1211 cm−1 frequency corresponding to a mixed mode stretching frequencies due to C-NO2 and
C-CH3. Other C-N out-of-plane bending vibrations are found to be at 723 and 423 cm−1 during computation while the experimentally assigned value12 is at 342 cm−1.
Vibrations of NO2 group — The nitro group is substituted at the third position in the heterocyclic ring of the titled compound. The computational results listed in Table 2 show that C-NO2 stretching is mixed with other vibrations and are found at 1274, 1211, 1063, 932, 760 and 384 cm−1. Visualizing these vibrations we find that these are contaminated by υ(C-CH3), υ(C-H) and β(C-H) vibrations. The asymmetrical stretching and scissor of NO2 have been calculated at 1534 cm−1 and 821 cm−1, respectively.
The experimentally observed value of NO2
asymmetrical stretching is found at 1540 cm−1 whereas there is no assignment is made for scissor NO2 in the experimental data12. C-NO2 out-of-plane bending is computationally assigned at 723 cm−1 while the experimental value is reported at 216 cm−1 whereas a computational value of 221 cm−1 is visualized and assigned as in-plane-bending. On the other hand, the experimental value of 687 cm−1 is assigned to this particular bending. These two bending are quite contradictory in their assigned values during the experiment and computation12. A more reliable assignment may be attributed to the computational data because of the visualization of these vibrations in 3D by Gassview program. NO2
torsional frequencies are assigned as 77 (computated) and 83 (experiment) which are in close agreement to each other.
C-C vibrations — C-C stretching of the ring carbons atoms are prominent. This is reflected by their high relative intensities as presented in Table 2.
These vibrations are found at 1597 and 1547 cm−1 along with in-plane-bending due to β(C-C-N) and β(O-H) while the experimental values are at 1670, 1550 and 1505 cm−1 as observed by Yadav et al12. The calculated frequencies are on lower side as compared to the experimental values. This may be due to contamination with other coupled vibrations.
Computationally, the C-C out-of-plane bending is found at 335 and 160 cm−1 while experimental values for these vibrations12 have been given at 494, 521 and 747 cm−1. In experimental observation bending in-plane shows at 1016 cm−1 and 605 cm−1. In their study, Yadav et al.12 have assigned 953 and 936 cm−1 vibrations to (C-C-C) trigonal bending. This is in contrast to the computational data. The nearby computed frequencies in this region are at 952 and 932. Visualization of the these two frequencies shows that 952 belongs to C-H out-of-plane bending whereas the 932 frequency belongs to a mixed vibration
Fig. 3 — Normalized IR intensities of 2-hydroxy-5-methyl- 3-nitropyridine calculated at DFT (B3LYP)/6-311G(3d,2p)
in-plane is found at 1547, 1392, 1351, 1236, 1145, 566, 428 and 328 cm−1. Visualization of these vibrations shows that these are highly contaminated by υ(C-C), β(C-H), β(C-N), β(N-O) and β(C-CH3) vibrations. Experimentally O-H bending in-plane in this molecule has been assigned12 at 1159 and 669 cm−1. Computational data for O-H out-of-plane bending are obtained at 780, 763 and 127 whereas experimentally observed value12 is at 243 cm−1. Vibrations of CH3 group — The-CH3 group is substituted at the fifth position of the titled molecule.
This group contains symmetrical as well as asymmetrical stretching frequencies. The computed values of two asymmetrical stretching vibrations of CH3 are found at 2983 and 2960 cm−1 while experimental values are reported at 2970 and 2911 cm−1. Computed CH3 symmetrical stretching frequency is found to be at 2910 cm−1 while experimental value12 is at 2885 cm−1. Computed values for other CH3 vibrations are assigned as torsion (1446),twisting (981), rocking (1435) and wagging at 1369 and 1030 cm−1. Experimentally rocking12 has been shown to occur at 1065 cm−1. The other two frequencies observed experimentally in this region have been assigned to be bending in-plane at 1456 (βasymCH3) and 1331cm−1 (βsymCH3). Visualizing C-CH3 stretching, it is found that it is contaminated by various other vibrational frequencies resulting from υ(C-NO2) and β(C-H) vibrations. Computed values of 1211, 932, 760 and 466 have been assigned to be on account of such vibrations. The experimental value for C-CH3 stretching12 has been shown to occur at 1295 cm−1. Two other C-CH3 vibrations have been assigned to be as bending in-plane found computationally at 328 and 221 cm−1 while the assignment made for the experimental data12 is at 440 cm−1.
4 Conclusions
The normal mode frequencies and corresponding vibrational assignment of title compound are examined theoretically using the Gaussian 03
during the experimental study were found to be contradictory to the computed data. However, the assignments made during the present investigation can be put on a greater confidence level because these are visualized in three dimensions using a Gaussview program.
References
1 Medhi R N, Barman R, Medhi K C & Jois S S, Spectrochimica Acta, Part A, 56 (2001) 1523.
2 Krishnakumar V & Xavier R J, Spectrochimica Acta, Part A, 61 (2005) 253.
3 Joule J A & Smith G F, Heterocyclic Chemistry (University Press, Cambridge), (1978) 44.
4 Krishnakumar V, Keresztary G, Sundius T & Xavier R J, Spectrochimica Acta, Part A, 61 (2005) 261.
5 Yadav B S, Kumar V, Singh V & Sembwal B S, Indian J Pure & Appl Phys, 37 (1999) 34.
6 Krishnakumar V & Ramasamy R, Indian J Pure & Appl Phys, (40) (2002) 252.
7 Sherman A R “Pyridine” in e-EROS (Encyclopedia of Reagents for Organic Synthesis) (Ed: L Paquette) 2004, (J Wiley, New York).
8 Sundaraganesan N, Saleem H & Mohan S, Spectrochimica Acta, 59 A (2003) 1113.
9 Arivazhagan M & Krishnakumar V, Indian J Pure & Appl Phys, 41 (2003) 341.
10 Abdulla H I & El-Bermani F M, Spectrochimica Acta, 57A (2001) 2659.
11 Topacle A & Bayari S, Spectrochimica Acta, Part A, 57 (2001) 1385.
12 Yadav B S, Ali Israt, Kumar P & Yadav P, Indian J Pure &
Appl Phys, 45 (2007) 979.
13 Parr R G & Yang W, Density-functional Theory of Atoms and Molecules (Oxford University Press, Oxford), 1989.
14 Roothaan C C J, Revs Mod Phys, 23 (1951) 69.
15 Becke A D, J Chem Phys, 98 (1993) 5648.
16 Lee C, Yang W & Parr R G, Phys Rev B, 37 (1988) 785.
17 Frisch M J, Trucks G W, Schlegel H B, Scuseria G E et al., Gaussian 03, Revision C 02 (Inc., Wallingford CT) 2004.
18 Cotton F A, Chemical Applications of Group Theory (Wiley Interscience, New York), 1971.
19 Frisch A, Nielsen A B & Holder A J, Gaussview Users Manual, Gaussian Inc, 2000.
20 Pople J A, Schlegel H B, Krishnan R, Defrees D J, Binkley J S, Frish M J, Whiteside R A, Hout R H & Hehre W J, Int J Quantum Chem, Quantum Chem Symp, 15 (1981) 269.
