EFFECT OF META SUBSTITUTION OF METHYL GROUP ON 2-HYDROXYPYRIDINE: SPECTROSCOPIC INVESTIGATION

1. Introduction

The nitrogen heterocyclic molecules are very important in different kind of biological processes as well as in industrial applications, thus it needs attention to understand the photophysical and photochemical properties of these molecules. However, the  fluorescence yield for these molecules is very low due to the  close-lying excited states. In these molecules, there is a possibility of a close proximity of having another excited state (nπ*), that arises due to the presence of lone pairs, with the ππ* excited state. In many heterocyclic molecules, these closely placed states result in strong mixing of the vibronic (vibrational and electronic) levels and thus affect the radiative as well as non-radiative energy decay properties of the  excited states. Bickel et al. [1] reported that the difference between the position of these two transition states (ππ* and nπ*) can be changed by substitution of a chemical group on any position of the molecule. Thus, the photophysical properties of these kinds of molecular systems depend on the  electronic states and can be changed by any suitable substitution. Hence this kind of a molecular system can be used as modelled molecules to study the  vibronic coupling of these transition states as well as the nature of molecular systems.

The methyl group can affect the  photophysical and photochemical properties of molecules and can also act as a probe for investigations. The coupling of the methyl group torsional motion with low frequency vibrations leads to the  concept of mixing of two closely spaced electronic states. Lim and team [2] studied the energy states nπ* and ππ* of nitrogenous heterocyclic molecules and they concluded for the molecules quinoline and isoquinoline that the lowest energy state is an nπ* state [2, 3]. Fischer [4, 5] worked on the spectroscopic study of isoquinoline vapour and found that the nπ* singlet state is slightly below the lowest energy ππ* singlet state. The  fluorescence excitation and dispersed fluorescence excitation spectroscopy of supersonic jet cooled isoquinoline and its methanol complex have been carried out by Felker and Zewail [6] and the coupling of ππ* and nπ* states was observed for the molecules and its complexes [3, 6, 7]. The methylisoquinolines were studied previously using laser induced fluorescence (LIF) and dipersed fluorescence (DF) excitation spectroscopic techniques [8]. The vibronic mixing of nπ* and ππ* excited states was observed in methylated isoquinoline molecules and it was also reported that the nπ* state possessed longer life time than the ππ* state.

The objective of this study is to understand the mechanism of the methyl torsional potential of the molecule in the ground electronic state as well as in the excited electronic states. However, the effect of methylation in the ground electronic state has been previously studied [9]. The focus is on the  study of close-lying electronic states and the interactions due to the methyl group substitution on different positions in these methylated 2-hydroxypyridines. However, the  interest is to understand the effect of different electronic environment on the  methyl torsional behaviour and thus on the barrier potential of the molecule.

2. Experimental section

The resonance enhanced multiphoton ionization (REMPI) experiment was performed using a tunable dye laser which was pumped by a second harmonic of a  pulsed Nd:YAG laser (Litron Nano Series Lasers, UK). For the  ionization of molecules, the output from the tunable dye laser was frequency doubled using another second harmonic crystal, while the  scanning of the  dye laser frequency was done by using a stepper motor which was controlled by the developed LabVIEW program. The angle tuning of the second harmonic crystal was done manually, and then this final output was made to cross through a  supersonic chamber, where the  interaction with the  sample takes place. The laser pulse and molecular jet were synchronized in time using two pulsed nozzle drivers (IOTA ONE Pulsed Driver from Parker Instrumentation) and a  self-made microsecond pulsed delay generator. The opening of the pulsed nozzle is optimized for the cold molecular beam and maximum ionization signal. A  convex lens of 8.0 cm focal length was used to focus the frequency doubled laser pulse inside the  chamber, where the interaction takes place. The ionization signal, which was reflected by a  plate termed as a  repulser plate (R), was collected in the  direction perpendicular to both laser and molecular beams with the  help of a  channel electron multiplier (DR.  SJUTS KBL 25RS). The  signal from the channel electron multiplier is averaged (typically 128 pulses) and digitized using an optically triggered digital oscilloscope (TDS 1012B), and stored in a  computer. The  molecules 3-mehtyl-2-hydroxypyridine and 5-methyl-2-hydroxypyri-dine were purchased from Sigma Aldrich and Alfa-Aesar chemical companies, respectively, and used without further purification. IR and Raman spectroscopy were performed using a  FTIR-Imaging System (3000 Hyperion Microscope with Vertex 80 FT-IR System, Bruker, Germany) with a spectral resolution of FT-IR 0.2  cm^–1 and a  HR800-UV confocal micro-Raman spectrometer (Horiba Jobin Yvon, France), respectively.

3. Theoretical background

Ab initio calculations have been used for the theoretical investigation of the molecules in the ground (S₀) as well as in the excited (S₁) electronic states. The  geometry optimization has been performed along with the  calculating barrier potential. The torsional potential in one dimension is given by $V\left(\tau \right)=\frac{V_3}{2}\left(1-\cos 3\tau \right)+\frac{V_6}{2}\left(1-\cos 6\tau \right)$ , where τ is the torsional angle. All the ab initio calculations for the  ground as well as the  excited electronic states were performed in Gaussian’09 [10] and the visualization of the molecular geometry and the molecular orbitals was carried out in the Gabedit software [11]. The procedure used for the theoretical study of the molecules in their ground (S₀) and excited (S₁) electronic states has been discussed in detail in our previous article [9, 12].

4. Results and discussion

4.1. REMPI spectroscopy

4.1.1. 3-Methyl-2-hydroxypyridine

The REMPI spectrum of supersonic jet-cooled 3-methyl-2-hydroxypyridine (3M2HP) is shown in Fig. 1. The spectrum was recorded after 30 min of the release of argon gas. Broad as well as sharp spectral bands were observed in a range of 33800 to 33500 cm^–1. The transition at 33830 cm^–1 is assigned to the origin of the spectrum as there is no more low frequency below this. There is no appreciable change in the intensity of the peak whenever pressure of the gas increases. The excitation spectrum for the parent molecule 2-hydroxypyridine arises due to the ππ* excitation [13] and no sign of nπ* transition states arises in the  spectrum. In a  similar manner we assign the  origin band at 33830 cm^–1 to ππ* transition. The  origin band is red-shifted by 2294  cm^–1 relative to the parent molecule indicating a change in excitation energy whenever the methyl group substitution takes place. There are several vibronic bands near the  origin showing intense low frequency transitions. On the other hand, in the case of 2-hydroxypyridine no such weak band is observed in this region [13].

The possible reason for appearing of many low frequency bands near the origin could be the methyl internal rotational transitions, as it shows low frequency bands. But all the  bands cannot be possible only because of the  torsional transitions. Nimlos et al. recorded the excitation spectra for 2-hydroxypyridine and observed a number of transitions which cannot be assigned to any vibrational band which is associated with ππ* transitions [14]. Such transitions were also reported by Felker et al. [6] and explained with the mixing of ππ* and nπ* vibrational states which do not have a spacing characteristic of ππ* transitions. The possible reason for appearing of many low frequency bands near the origin band could be the existence of a  nearby nπ* state, and the  vibronic coupling between the levels of ππ* and nπ*. Table 1 shows the observed bands and their assignment.

Close to the origin transition, a band at a separation of 21 cm^–1 is observed. Assigning this band to the 2e torsional transition of the methyl group proclaimed the 3a1 and 4e bands around 53 and 83 cm^–1, respectively. But there is no well-resolved band near the predicted value. Hence the second peak cannot be assigned to a torsional band. Also, assigning 21 cm^–1 to a 2e band produces a barrier of a very low magnitude. But the barrier potential in the ground as well as in the excited states will be of the same order as there is no splitting of 0a1, 0a1 and 1e, 1e transitions for the origin band. Similar bands were reported for 1 MPY [15] in which it was assumed that this could be some anomalous band associated with the nπ* transition.

The bands at 131 and 254 cm^–1 from the origin band in the spectrum may be assigned to 2e and 3a1 torsional transitions, respectively. The  potential parameters in the excited state can be obtained by fitting these bands. The  excited state torsional parameters (V′₃ = 464 cm^–1, V′₆ = 2 cm^–1, F′ = 5.3 cm^–1) were obtained by the best fit of these observed bands. The calculated as well as experimentally observed torsional frequencies are presented in Table 2. The bands at 349, 372, 409 and 456 cm^–1 can be assigned to 5e, 6a2, 6a1 and 7e torsional transition bands. There was no 4e band in the spectrum. The band at 278 cm^–1 from the origin band could be due to the  mixing of two vibrational modes. Hence, the band at 144 cm^–1 was assigned to the  second lowest vibrational mode (v¹₀) in the excited state and can explain the presence of the band at 278 cm^–1. There are many other vibrational modes in the spectrum due to the methyl torsional transitions which can be easily explained by the  combination of the  other vibrational modes, as listed in Table 2 as progression 2. The bands at 396, 501, 521, 553 and 601 cm^–1 are assigned to 3a1, 5e, 6a2, 6a1 and 7e torsion and vibration combination bands. The potential parameters used for the assignments of these combination bands are V′₃ = 472 cm^–1, V′₆ = 1 cm^–1 and F′ = 5.3 cm^–1.

4.1.2. 5-Methyl-2-hydroxypyridine

The obtained REMPI spectrum of jet-cooled 5-methyl-2-hydroxypyridine (5M2HP) is shown in Fig. 2. The  spectrum is recorded after 30  min of the  release of the  gas. One intense peak with few other peaks was observed in a spectral range of 33900 to 34550  cm^–1. The  observed spectrum was then normalized using the  dye laser profile. The transition band at 34105 cm^–1 has been defined as the origin band of the molecule. This is the ππ* transition band. The characteristic of this band and the intensity pattern in the spectrum indicates an unchanged molecular conformation in the excited state compared to the ground state. The other bands are either vibrational bands or methyl torsional bands of the molecule. These bands are at 169, 267, 291, 316, 348 and 384 cm^–1 from the assigned origin (34105 cm^–1). The assignment of the torsional band transitions 169, 267 and 291 cm^–1 was identified as 3a1, 6a2 and 6a1 using the  potential parameters V₃ = 276 cm^–1, V₆ = –50 cm^–1 and F = 5.3 cm^–1. Meanwhile, the band at 316 cm^–1 matched with the band transitions for 3-methyl-2-hydroxypyridine. Hence, this band can be assigned to the vibration corresponding to the  3-methyl-2-hydroxypyridine molecule. The  other bands at 348 and 384 cm^–1 are assumed as some other vibration.

4.2. IR and Raman spectrum

4.2.1. 3-Methyl-2-hydroxypyridine

For the assignments of the obtained vibrational as well as torsional bands in the  ground electronic state, infrared (IR) and Raman spectroscopic techniques were analysed. The FT-IR and Raman spectrum of the molecule in the condensed phase is shown in Figs. 3 and 4, respectively. The spectral resolution in the  far IR region is very high compared to that in the mid IR region, thus there is a minimum discrepancy in few of the band positions and their relative intensity such as 490 and 557 cm^–1 in  3(a). However, the main interest is to observe the low frequency bands near the origin due to the  methyl torsion motion. The  calculated frequencies with their relative IR intensity and Raman activity are listed in Table ST1 as supplementary material (Appendix 1) under the harmonic approximation with the optimized geometry in the  B3LYP/TZVP level of theory. Few of the  calculated low frequency bands are in close proximity with the  experimentally observed bands in Raman and FT-IR. The band at 126 cm^–1 is the pure torsional band of the molecule in the ground state which can also be seen in the Raman spectrum at 107 cm^–1. While the bands at 150 and 271 cm^–1 in the Raman spectrum can also be seen in the  FT-IR spectrum at 162 and 272 cm^–1, respectively. These low frequency bands are obtained due to the torsional motion of the ring frame and close to the theoretically obtained bands at 163 and 273 cm^–1, respectively. The molecule has 15 atoms hence 39 fundamental frequencies are present. Among these 39 fundamental frequencies, 11 are in-plane and 28 are out-of-plane vibrational modes. For all the bands strong mixing of various vibrational modes was observed. However, the first three low frequency vibrations show torsional characteristics.

Figure 5 shows six low frequency vibrations in the ground electronic state of the molecule in which arrows represent the atomic displacements. The lowest frequency was at 126 cm^–1. This is because of the torsional motion of the methyl group. The next two calculated low frequencies are at 163 and 273 cm^–1. 163 cm^–1 is observed as a mixture of various torsional motions of the ring frame with the C₃C₈ torsion, while 273 cm^–1 is obtained because of the mixing of the torsional motion and the  bending/wagging motion of the  ring frame. These two frequencies are identified in the  observed bands in IR and in Raman spectra. However, there was no band at 349 cm^–1 in the Raman or the IR spectrum as it was present in the REMPI spectra of the molecule. There was no analogous frequency in the calculation. On the basis of these observations, this band was assigned to the  torsional ‘e’ progression (5e).

4.2.2. 5-Methyl-2-hydroxypyridine

The FT-IR and Raman spectrum of 5-methyl-2-hydroxypyridine is shown in Figs. 6 and 7, respectively. The  discrepancy, as discussed previously, is also observed in Fig .6(a) for 500 and 527 cm^–1 due to the  spectral resolution. The  B3LYP/ TZVP level of theory has been used to calculate the fundamental frequencies, their relative IR intensity and Raman activity under the  harmonic approximation with the optimized geometry, and it is listed in Table ST2 as supplementary material (Appendix 2).

39 fundamental frequencies were obtained for the  molecule in which mixing of various vibrational bands was observed as in 3M2HP. There were no such low frequency vibrational bands as in 3-methyl-2-hydroxypyridine. The  first two calculated frequencies were obtained as mixing of the  vibrational motions. A  low frequency at 74 cm^–1 was observed as mixing of the torsional characteristic of the methyl group and the out-of-plane wagging/bending motion of the ring frame, while the band at 141 cm^–1 was due to the mixing of the torsional motion of the ring as well as its out-of-plane wagging/bending motion.

4.3. Theoretical study: Ab initio calculations

4.3.1. 3-Methyl-2-hydroxypyridine

The intensity pattern observed from the  REMPI spectrum suggests the change in the methyl group position upon excitation from the  ground electronic state to the  excited electronic state (ππ*). The torsional potential due to the methyl group internal motion was estimated with the observed 5e band. Further investigation for the conformational study and the torsional potential in the excited state was carried out using the  B3LYP/TZVP level of theory. The optimized geometry in the excited state is shown in Fig. 8(a). In the obtained minimum energy conformation of the  molecule, the  ring frame is always planar and the internal rotation of the methyl group was observed. The torsional angle dependence of the  potential energy curve for the excited state (S₁) is shown in Fig. 8(b).

The curve was extracted from the ground state potential energy and the  calculated excitation energies for every torsional angle. The  calculated value of the three-fold potential term V₃ was found to be 464 cm^–1 and matched with the experimentally observed value. It can also be seen from the potential curve that the potential minimum has been shifted by 60° in the excited state relative to the ground state. This is in agreement with the result obtained from the optimized geometry and from the observed REMPI spectrum. The  change in the  methyl group conformation and the magnitude of the torsional potential barrier in 3-methyl-2-hydroxypyridine upon excitation (S₀ → S₁) were explained using the π*–σ* hyperconjugation [15]. The π*–σ* hyperconjugation in the LUMO at the top of the barrier conformation stabilizes these in the  ground as well as in the  excited states. The  orbital contour diagram for 3-methyl-2-hydroxypyridine in the  HOMO and LUMO in the minimum and top of the barrier conformation is shown in Fig. 9. The π*–σ* hyperconjugation was observed only in the 0 degree conformation in the  LUMO. This signifies the change in conformation of the methyl group in the  excited state, i.e. rotation of the  methyl group upon excitation from the ground electronic state to the excited electronic state (S₀ → S₁).

4.3.2. 5-Methyl-2-hydroxypyridine

As observed from the  obtained REMPI spectrum, there is no such change in the  conformation of the  methyl group obtained in 5M2HP as in 3M2HP when excitation took place from the  ground electronic state to the  excited electronic state. The B3LYP/TZVP level of theory has been taken into account for the further investigation of the conformation and torsional potential of the molecule in the excited state (S₁). The optimized geometry of 5-methyl-2-hydroxypyridine in the S₁ state is shown in Fig. 10(a). The ring frame is always planar whenever geometry optimization in the minimum energy conformation has been performed. The torsional angle dependence of the potential energy curve for the excited state (S₁) is shown in Fig. 10(b). The  calculated three-fold potential term V₃ was found to be 92 cm^–1. The obtained potential curve for the optimized geometry is in agreement with the REMPI spectrum of the molecule.

As discussed previously, the observed π*–σ* hyperconjugation explained the changes in the methyl group conformation and in the magnitude of the torsional potential barrier in 3M2HP upon excitation (S₀ → S₁). But in the case of 5M2HP, there is no change of methyl group conformation (as in the REMPI spectrum) hence no π*–σ* hyperconjugation is involved. The orbital contour diagram for 5-methyl-2-hydroxypyridine in the  HOMO and LUMO in the  minimum energy conformation and top of the barrier energy conformation is shown in Fig. 11.

5. Conclusions

The obtained REMPI spectrum shows an internal rotation of the  methyl group in 3-methyl-2-hydroxypyridine upon excitation from the ground state to the  excited state which can also be observed from the obtained results from the ab initio calculations as well as from the intensity pattern calculation. The band at 33830 and 34105 cm^–1 was assigned as the ππ* transition state and named as the origin band for 3M2HP and 5M2HP, respectively. Many low frequency vibrational bands were near the origin band which could be because of the vibronic coupling of the nπ* and ππ* transition states in the  3-methyl-2-hydroxypyridine molecule. The π*–σ* hyperconjugation is responsible for the  conformational change in the  methyl group in 3-methyl-2-hydroxypyridine, while there was no hyperconjugation in 5-methyl-2-hydroxypyridine.

Acknowledgements

Authors would like to thank the Department of Science and Technology, India for providing the financial support for this work. We would also like to thank Prof. T. Kundu from the Indian Institute of Technology Bombay and Dr. Rajeev K. Sinha from Manipal University, India for their valuable suggestions and continuous discussion.

Appendix 1

Appendix 2

References