Vibrational Spectroscopic Characterization of 2-(2,4-Dinitrobenzyl

Article Cite This: J. Phys. Chem. A XXXX, XXX, XXX−XXX

pubs.acs.org/JPCA

Vibrational Spectroscopic Characterization of 2‑(2,4-Dinitrobenzyl)pyridine (α-DNBP) in Solution by Polarization-Resolved Spontaneous Raman Scattering and Broadband CARS Sebastian Küpper,† Vikas Kumar,† and Sebastian Schlücker* Fakultät für Chemie, Universität Duisburg-Essen, Universitätsstrasse 5, 45141 Essen, Germany

Downloaded via UNIV OF SOUTHERN INDIANA on July 22, 2019 at 07:10:42 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: The photochromic molecule 2-(2,4-dinitro-benzyl)-pyridine (α-DNBP) is characterized in solution by a combination of density functional theory employing a polarizable continuum model and polarization-resolved spontaneous and nonlinear Raman spectroscopies. By the comparison of theoretically predicted wavenumber positions and depolarization ratios with the experimental spectra acquired under electronically nonresonant conditions, polarized and depolarized Raman bands are assigned. Specifically, the symmetric stretching vibrations of the two nitro groups in ortho and para positions to the pyridine ring can be experimentally differentiated, mainly because of their different Raman depolarization ratios, which supports our prediction from theory. Compared to the polarization-resolved spontaneous Raman experiments, the vibrational spectroscopic differentiation of the two nitro groups is more pronounced in time-delayed polarization-resolved coherent anti-Stokes Raman scattering experiments. Overall, this linear and nonlinear vibrational spectroscopic characterization of the CH form paves the way for the interpretation of future time-resolved pump/nonlinear Raman probe studies on the ultrafast photoinduced intramolecular proton transfer in α-DNBP involving a nitro group as an intramolecular proton acceptor.

■

INTRODUCTION Over the past few decades, organic photochromic compounds have attracted significant attention as a promising class of materials in nanodevices for rewritable optical data storage and solar energy accumulators. Their photochromic properties are attributed to their ability to modify and switch physical properties by means of photoinduced inter- and/or intramolecular transformations. Photochromic compounds undergo an ultrafast (time-scale: 100−250 fs1−3) photoinduced intramolecular proton transfer (PIPT) because the translocation of a hydrogen atom does not require large structural changes. PIPT is not only observed in solution but also in amorphous or crystalline solids. Nitrobenzylpyridines are an important class of photochromic compounds: they exhibit photochromism in solution, in molten form, in powder form as well as in single crystals. In their pioneering work from 1925 Tschitschibabin and co-workers4 reported on the synthesis of 2-(2′,4′dinitrobenzyl)pyridine (α-DNBP) and its photochromic properties in the crystalline state. They observed that upon exposure to sunlight, the pale yellow color of α-DNBP crystals changes to dark blue, followed by decoloration in 24−48 h after storage in darkness. Later, Hardwick and co-workers5 showed that cooled solutions of α-DNBP in a variety of solvents also exhibit reversible photochromism. It is generally accepted that three different tautomers of α-DNBP are responsible for reversible photochromism via PIPT (Figure 1). Upon photoexcitation of the stable CH tautomer, the short-lived OH tautomer is generated, which can undergo a © XXXX American Chemical Society

Figure 1. Tautomeric forms of photochromic α-DNBP and their interconversion via PIPT.

second PIPT to yield the NH form. Many studies on the photochromic reaction of α-DNBP have been reported. Flash photolysis,6−8 UV−vis absorption spectroscopy,9,10 polarized optical absorption spectroscopy,11 femtosecond transient absorption spectroscopy,12,13 time-resolved resonance Raman spectroscopy,14−19 and resonance coherent anti-Stokes Raman scattering (CARS) spectroscopy18 methods have been implemented to study the photochromic reaction of αReceived: May 30, 2019 Revised: June 22, 2019 Published: June 25, 2019 A

DOI: 10.1021/acs.jpca.9b05142 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

constant of ε = 4.71 for simulating the solvent chloroform was employed. Figure 2 bottom additionally displays the corresponding depolarization ratio for each normal mode in this region of interest: totally symmetric modes have a depolarization ratio of ρ = 0, while depolarized modes exhibit a depolarization ratio of ρ = 0.75. For example, the normal mode at 852 cm−1 is a polarized mode because it has a depolarization ratio of ρ = 0.06: it involves mainly scissoring motions of the two nitro groups (Figure 3 top left) and

DNBP and its derivatives. X-ray diffraction studies have determined the crystal structure of the CH form (Figure 1 top),9,20 but not of the NH and OH isomers because of the instability of the NH form (Figure 1 bottom right) and the OH form (Figure 1 bottom left). Theoretical investigations such as semi-empirical (PM3) calculations,21 ab initio,22 and density functional theory (DFT) calculations11,23 were also performed. While the slow dynamics (nanoseconds to microseconds to milliseconds and even seconds) is well understood,6,14,15 the ultrafast PIPT dynamics (femtosecond to picosecond timescale) still largely remains unresolved.5,7,8,10,11,13,14,19,20,24−30 Vibrational spectroscopic techniques including infrared (IR) spectroscopy31 as well as various linear and nonlinear Raman spectroscopies32−34 offer the advantage that they provide chemical specificity by probing functional groups based on their intrinsic “molecular fingerprint”. IR spectroscopy has been performed on samples of αDNBP in potassium bromide wafer5 and sodium chloride pellets35 in dark and after irradiation. Andreev et al.22 claimed to provide the first nonresonant Raman spectroscopic characterization of the CH and NH forms of α-DNBP crystals using off-resonant near-IR laser excitation (Nd:YAG, FT-NIRRaman). In this contribution, we focus on the Raman spectroscopic characterization of the CH form of α-DNBP for reliable vibrational assignment of relevant normal modes potentially involved in the photochromism. Specifically, we were interested in unambiguously identifying the spectral signatures of the two nitro groups attached to the phenyl ring because the NO2 group in the ortho position is supposed to act as an intramolecular proton acceptor in the CH to OH conversion (Figure 1). To this end, we employ DFT calculations in combination with polarization-resolved spontaneous Raman36 and CARS37−41 of α-DNBP in solution. Previous studies have not been able to address this question because they did not employ a polarization-resolved detection. Our characterization paves the way for the interpretation of future optical fs-pump/ fs-probe experiments, in particular, fs-vibrational (IR/nonlinear Raman) spectroscopy, which may shed light on details of the underlying PIPT mechanism.

Figure 3. Calculated eigenvectors of selected normal modes of the CH form of α-DNBP.

therefore has only a moderate Raman activity (Figure 2 top). The two dominant Raman peaks at 1360 and 1379 cm−1 in Figure 2 top are both assigned to symmetric stretching vibrations of the two nitro groups: the normal mode at 1360 cm−1 involves mainly stretching motions located at the nitro group in the para position to the pyridine ring (Figure 3 top right), while the normal mode at 1379 cm−1 involves mainly stretching motions located at the other nitro group in the ortho position (Figure 3 bottom left). Interestingly, these two closely spaced and intense peaks of the two nitro stretching vibrations have a substantially different depolarization ratio of ρ = 0.43 (1360 cm−1) and ρ = 0.14 (1379 cm−1), respectively (Figure 2 bottom). The third intense Raman peak at 1642 cm−1 (Figure 2 top) is assigned to a phenyl ring mode with dominantly CC stretching motions (Figure 3 bottom right). Other Raman peaks at 1069, 1136, 1160, 1207, and 1230 cm−1 are assigned to C−H deformation modes, while the Raman peaks at 1013 and 1555 cm−1 belong to a phenyl ring mode and an asymmetric nitro stretching mode, respectively. A table with all normal modes of the CH form of α-DNBP together with their wavenumber values, IR and Raman activities as well as depolarization ratios can be found in the Supporting Information (Table S1). Polarization-resolved spontaneous Raman scattering experiments were performed for the vibrational assignment of prominent normal modes of the CH form of α-DNBP (Figure 4 top) because they provide information on peak positions, peak intensities as well as depolarization ratios. By recording the perpendicular (I⊥) and parallel (I∥) components of the Raman scattering intensity, the Raman depolarization ratio36 ρ = I⊥/I∥ was determined for dominant Raman peaks in the parallel spectrum (Figure 4 bottom). As mentioned above

■

RESULTS AND DISCUSSION Figure 2 top shows the calculated Raman spectrum of the CH form of α-DNBP in the fingerprint region from 800 to 1700 cm−1, calculated at the uB3LYP/6-311++G(d,p) level of theory. A polarizable continuum model with a dielectric

Figure 2. Calculated Raman spectrum (raw data) of the CH form of α-DNBP in solution. The positions of selected Raman peaks are indicated. B


Article


The spectral profiles of pump, Stokes, and probe pulses employed in the CARS experiments are depicted in Figure 5.

Figure 4. Polarization-resolved spontaneous Raman spectra of αDNBP in solution: perpendicular (I⊥) and parallel (I∥) component (top). Raman depolarization ratio ρ = I⊥/I∥ for selected Raman peaks (bottom).

in the context of the computed Raman spectrum (Figure 2 bottom), the depolarization ratio is an important additional spectroscopic parameter for the vibrational assignment. The computed (Figure 2) and the experimental (Figure 4) Raman spectra show a good agreement because characteristic Raman peaks can be identified based on both their wavenumber position and their depolarization ratio: polarized Raman peaks at 834 cm−1 (theo: 852 cm−1), 996 cm−1 (theo: 1013 cm−1) and 1050 cm−1 (theo: 1069 cm−1) as well as depolarized Raman peaks at 1124 cm−1 (theo: 1136 cm−1), 1149 cm−1 (theo: 1160 cm−1), 1198 cm−1 (theo: 1207 cm−1), 1214 cm−1 (theo: 1230 cm−1), 1536 cm−1 (theo: 1555 cm−1), and 1613 cm−1 (theo: 1642 cm−1). Theory predicts that the two nitrostretching peaks are spectrally separated by 19 cm−1 (Figure 2: 1360 vs 1379 cm−1) and have a substantially different depolarization ratio of ρ = 0.43 (1360 cm−1) and ρ = 0.14 (1379 cm−1). In the experimental polarization-resolved Raman spectrum (Figure 4 top), however, only a single peak with a symmetric line shape at 1354 cm−1 is observed. This suggests that the two symmetric nitro-stretching modes cannot be spectrally resolved, although the experimentally employed spectral resolution of ca. 4 cm−1 (monochromator with 55 cm focal length) should be sufficient. The same result was obtained also for the spontaneous Raman experiment without polarization control (see Figure S1). In Figure 4, the corresponding perpendicular (I⊥) component exhibits an asymmetric line shape. This suggests that the two nitro peaks may be discriminated because of their different depolarization ratios. We therefore calculated the Raman depolarization ratio for each wavenumber position across the NO2 stretching peak: there is a clear decrease in the depolarization ratio from ρ = 0.29 at 1347 cm−1 to ρ = 0.16 at 1363 cm−1. We therefore conclude that the single Raman peak at 1354 cm−1 is the superposition of the two nitrostretching vibrations: the contribution around 1347 cm−1 is assigned to the NO2 group in the para position (Figure 3 top right), the contribution around 1363 cm−1 to the NO2 group in the ortho position (Figure 3 bottom left). A reliable vibrational assignment of the nitro modes is important because they act as the proton acceptor during the intramolecular PIPT (OH form in Figure 1). This ultrafast process may potentially be resolved by UV pump/ultrafast nonlinear Raman probe experiments. We therefore also employed polarization-resolved CARS as a nonlinear Raman technique for the vibrational spectroscopic characterization of the CH form of α-DNBP in solution.

Figure 5. Spectral profiles of pump, Stokes, and probe pulses employed in the CARS experiments. The spectral width of the psprobe pulse is shown in the inset.

By using fs-pump and fs-Stokes pulses, the fingerprint region 620−1900 cm−1 is covered. The spectral resolution of the blueshifted CARS signal is determined by the spectral width of the ps-probe pulse (inset: ∼14 cm−1). Figure 6 bottom shows the CARS spectrum obtained when all three pulses overlap in time (Δt = 0). A strong background

Figure 6. CARS spectra α-DNBP in solution at different time delays of the ps-probe with respect to the fs-pump and fs-Stokes pulses. Solvent peaks are indicated by an asterisk.

in combination with disperse lineshapes is observed. This is because of the presence of the broad non-Raman resonant background (NRB) and its superposition with the narrow Raman-resonant contributions. The NRB is the electronic response of the system (χ(3) NR) and is essentially only present when the pulses overlap in time and space; it decays very fast compared with the vibrational coherence involving nuclear motions generated by the simultaneous excitation with the fspump and fs-Stokes pulse. Suppression of the NRB can be achieved, for example, by a delayed probe pulse with respect to the pump and Stokes pulses.42,43 The delayed probe pulse generates the vibrational CARS signal (χ(3) R ) after the NRB has already decayed in time. Figure 6 middle and top show the suppression of the NRB from α-DNBP in chloroform using C


Article


compared with normal CARS at time zero (Figure 6 bottom). By using a combination of both time delay and polarization control even the small residual NRB can be completely suppressed. Already at Δt = 400 fs, no disturbing NRB contributions are observed anymore (Figure 8 middle). Larger delay times (Δt = 700 fs, Figure 8 top) lead to decreasing CARS signals. At both delay times in polarization- and timeresolved CARS, we observe more Raman peaks as compared to time-delayed CARS spectra without polarization control (Figure 6 top). Raman peaks at 837 cm−1 (scissoring motions of the two nitro groups), 996 cm−1 (phenyl ring mode) 1050, 1151 and 1210 cm−1 (C−H deformation modes), 1356 cm−1 (symmetric stretching vibrations of the nitro groups), 1540 cm−1 (asymmetric stretching vibrational mode of nitro groups), and 1611 cm−1 (phenyl ring mode) can be identified. Again, the agreement with the DFT calculations and the spontaneous Raman spectra is good. The CARS spectra in Figure 8 exhibit only a single peak in the nitro-stretching region at 1356 cm−1, that is, the two nitro peaks can unfortunately not be spectrally resolved. Similar to the polarization-resolved spontaneous Raman experiments in Figure 4, it would be helpful for future fs time-resolved pump/ nonlinear Raman probe experiments to differentiate between the NO2 groups in ortho and para positions. We therefore performed CARS experiments with both time delay for NRB suppression and polarization control for distinguishing modes with different depolarization ratios. In addition to NRB suppression at a fixed analyzer position at θ = −60° (Figure 7 top), polarization-resolved CARS experiments at additional analyzer positions (êA) can distinguish between Ramanresonant contributions based on their different polarization directions originating from different depolarization ratios. The concept of polarization-resolved CARS for distinguishing Raman-resonant contributions because of differences in depolarization ratios is shown in Figure 7 bottom: the CARS signal of totally symmetric vibrations with ρ = 0 occurs in the direction of φ = 52.7°. Thus, it is suppressed when the analyzer is placed at an angle of θ = −37.3°. Figure 9 shows timedelayed (Δt = 700 fs) CARS spectra with varying analyzer positions of θ = −37°, −39°, and −43°. Interestingly, at θ = −37° two peaks around 1356 cm−1 can be clearly observed: the contribution at ca. 1347 cm−1 is more intense than the contribution at 1370 cm−1. Upon rotating the analyzer to −39° and −43°, the intensity ratio I(1347 cm−1)/I(1370 cm−1)

time delays of 400 and 700 fs, respectively. At larger delay times (Δt = 700 fs), the NRB suppression is already quite efficient and symmetric, nondispersive lines shapes are observed, indicating that no interference with the NRB occurs anymore. At this delay, we were still able to detect Raman peaks at 839 cm−1 (scissoring motions of the two nitro groups), 1002 cm−1 (phenyl ring mode), 1053 and 1157 cm−1 (C−H deformation modes), and 1356 cm−1 (symmetric stretching vibrations of the nitro groups). These observations are all in good agreement with the spontaneous Raman results (Figure 4). The relative peak intensities in the CARS spectrum cannot be directly compared to the corresponding intensities in the spontaneous Raman spectrum because the intensity profile of the pump/Stokes pulses is not flat but Gaussian shaped (cf. Figure 5, blue curve: recorded intensity profile of a nonresonant four-wave-mixing anti-Stokes signal from a glass plate). Here, we present raw CARS data, that is, no recalibration with the spectral profiles of the pump/Stokes pulses was performed. Figure 7 illustrates the concept of polarization-resolved CARS for suppression of the NRB and identifying vibrational

Figure 7. Concept of polarization-resolved CARS for suppression of the NRB (top) and for identifying vibrational CARS peaks based on their different depolarization ratio (bottom).

CARS peaks based on their different depolarization ratio. Pump and Stokes beams have a parallel polarization. The polarization of the probe beam is rotated by 60° clockwise. In an optically transparent and isotropic medium, the NRB occurs at an angle φ = 30° (ρ = 1/3) (Figure 7 top). Thus, suppression of the NRB is achieved by rotating the analyzer (êA) to θ = −60°. However, at this analyzer position also the CARS signal of modes with ρ = 1/3 is suppressed as well. Figure 8 bottom shows the efficient suppression of the NRB at time zero (Δt = 0) using polarization-controlled CARS

Figure 9. CARS spectra of α-DNBP in solution with both probe time delay and polarization control for resolving the spectrally overlapping NO2 stretching motions around 1356 cm−1. Solvent peaks are indicated by an asterisk.

Figure 8. Time-delayed and polarization-controlled CARS spectra of α-DNBP in solution at different ps-probe pulse delays. Solvent peaks are indicated by an asterisk. D


Article


pulses were spatially and temporally synchronized on the sample kept in the cuvette employing a folded BOXCARS geometry. The laser power at the sample was kept low to avoid sample degradation and higher-order nonlinear effects (pump: ca. 0.8 mW, Stokes: ca. 0.8 mW, probe: ca. 0.9 mW). The simultaneous action of the fs-pump/Stokes pulses generates broadband vibrational coherence in the sample, while the psprobe pulse subsequently interacts to produce the CARS signal with sufficient spectral resolution. The CARS spectra were then dispersed in a monochromator with 30 cm focal length (HORIBA Scientific iHR320) and recorded by a CCD. For time-resolved CARS measurements, the probe pulse was delayed by a motorized optical delay line. For polarizationresolved CARS a combination of half-waveplate and polarizer each in pump, Stokes, and probe beams was inserted before the sample to get the polarization geometry in excitation as shown in Figure 7. An additional polarizer was placed in the pathway of the generated CARS signal to act as a variable analyzer. Time-delayed and polarization-resolved CARS spectra were acquired at various analyzer positions as shown in Figure 9.

decreases significantly. This trend can be understood by the different Raman depolarization ratios of the symmetric stretching vibrations of the two NO2 groups, as revealed by the DFT calculation (Figure 2) and the polarization-resolved spontaneous Raman experiment (Figure 4): the peak at 1347 cm−1 from the nitro group in the para position has a higher depolarization ratio than the peak at 1370 cm−1 from the nitro group in the ortho position. Upon rotating the analyzer from θ = −37° to −43° we favor the detection of the stronger depolarized modes (Figure 7) and therefore the CARS intensity ratio I(1347 cm−1)/I(1370 cm−1) decreases. Even though the NRB has already decayed at Δt = 700 fs and most peak profiles are symmetric, while experimentally observed lineshapes of the two overlapping NO2 peaks are asymmetric; this is due to the coherent superposition of the corresponding two χ(3) R contributions, which differ in intensity, phase, and polarization behavior. Overall, both polarizationresolved spontaneous Raman and CARS experiments enable the differentiation of the two nitro groups as key players in the photoinduced intramolecular proton transfer of α-DNBP. Moreover, other Raman peaks at 837 cm−1 (scissoring motions of the two nitro groups) 1002 cm−1 (phenyl ring mode), 1053, 1151 and 1216 cm−1 (C−H deformation modes), 1543 cm−1 (asymmetric stretching vibrations of nitro groups), and 1615 cm−1 (phenyl ring mode) are more pronounced.

■

CONCLUSIONS The photochromic molecule α-DNBP was characterized for the first time by both polarization-resolved spontaneous Raman scattering and CARS. The agreement between experimentally determined and theoretically predicted peak positions and depolarization ratios is good. The contributions of the two nitro groups in ortho and para positions to the pyridine ring, which are postulated to play an important role as a proton acceptor in the PIPT, can be spectroscopically differentiated by their different polarization behavior. In the future, we plan to perform fs-UV pump/fs-nonlinear Raman probe experiments on α-DNBP in chloroform for studying the ultrafast PIPT. The central aim of these future studies is to identify the involved transient species based on their intrinsic Raman spectroscopic fingerprint. The present vibrational spectroscopic assignment of the CH form of α-DNBP in solution provides the basis for clearly differentiating the CH ground-state species from the transient species in the excited electronic states.

■

MATERIALS AND METHODS DFT calculations were performed with the GAUSSIAN 2016 program package44 with the uB3LYP functional and the 6311++G(d,p) basis set. The effect of the solvent chloroform was simulated using a polarizable continuum model (dielectric constant ε = 4.71). Solid α-DNBP was purchased from TCI Deutschland GmbH. Chloroform (spectroscopic grade) was obtained from Fisher Scientific. Both chemicals were used without any further purification. A 2 M solution of α-DNBP in chloroform was prepared in dark. All linear and nonlinear Raman experiments were performed on a 2 M solution of α-DNBP in a 2 mm quartz cuvette. Spontaneous Raman measurements were performed with a HeNe laser and a home-built Raman microspectrometer comprising an inverted microscope (Nikon Eclipse Ti) and monochromator with 55 cm focal length (HORIBA Scientific iHR550) equipped with a CCD. For polarization-resolved linear Raman experiments, the parallel and perpendicular components were recorded sequentially employing a set of polarizers in the same spectroscopy setup. The laser power at the sample was ca. 7 mW. For time- and polarization-resolved CARS experiments, fspump/Stokes, and ps-probe pulses were generated from outputs of two nonlinear optical parametric amplifiers (NOPAs; Orpheus-F, Light Conversion) synchronously pumped by an amplified mode-locked Yb-laser (PHAROS, Light Conversion). Both NOPAs provide ∼50 fs pulses (200 mW average powers) independently tunable in the wavelength range 640−920 nm at a 20 kHz repetition rate. For CARS experiments reported here, the two NOPA outputs were set at wavelengths centered at 670 and 731 nm serving as fs-pump and fs-Stokes pulses, respectively, thereby covering the entire fingerprint region (620−1900 cm−1). A part of the first NOPA output was spectrally filtered by a narrowband filter (Alluxa) to synthesize ps-probe pulses at 667.7 nm with ∼14.1 cm−1 spectral width (Figure 5). The pump, Stokes, and probe

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.9b05142.

■

Calculated vibrational spectrum and spontaneous Raman spectrum (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Sebastian Schlücker: 0000-0003-4790-4616 Author Contributions †

S.K. and V.K. contributed equally.

Funding

This work is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)Projekt nummer 278162697SFB 1242 “non-equilibrium dynamics of condensed matter in the time domain” (project A04). E


Article

The Journal of Physical Chemistry A Notes

(19) Yokoyama, K.; Kobayashi, T. Photochromism of 2-(2’,4’d1n1trobenzyl)pyridine in solution studied by time-resolved resonance raman spectroscopy. Chem. Phys. Lett. 1982, 85, 175−178. (20) Seff, K.; Trueblood, K. N. The Crystal and Molecular Structure of 2-(2’,4’-Dinitrobenzyl)Pyridine. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. 1968, 24, 1406−1415. (21) Khatib, S.; Tal, S.; Godsi, O.; Peskin, U.; Eichen, Y. Site Selective Processes: A Combined Theoretical and Experimental Investigation of Thermally Activated Tautomerization Processes in 2(2,4-Dinitrobenzyl) Pyridine Derivatives. Tetrahedron 2000, 56, 6753−6761. (22) Andreev, G. N.; Schrader, B.; Hristozova, D. A.; Delchev, V. B.; Petrov, J. S.; Rademacher, P. Vibrational spectra and structure of the photochromic 2-(2′,4′-dinitrobenzyl)pyridine. J. Mol. Struct. 2003, 645, 77−87. (23) Floare, C.; Astilean, S.; Bogdan, M.; O, C. DFT Structure Optimization and Vibrational Spectra of the Photochromic 2-(2′, 4′Dinitrobenzyl) Pyridine. Rom. J. Phys. 2005, 50, 815−829. (24) Naumov, P. Photochromism of Ortho-Nitrobenzylpyridines: A Brief Overview. J. Mol. Struct. 2006, 783, 1−8. (25) Sousa, J. A.; Loconti, J. D. Photoisomerism: A Colorless Photoinduced Intermediate of O-Nitrobenzylpyridine. Science 1964, 146, 397. (26) Sousa, J. A.; Weinstein, J. Photoisomerization of 2-(2,4Dinitrobenzyl)Pyridine and 2-(2-Nitro-4-Cyanobenzyl)Pyridine. J. Org. Chem. 1962, 27, 3155−3159. (27) Tinland, B.; Décoret, C. A molecular orbital study of the phototropism of 2 - (2′,4′ - dinitrobenzyl) pyridine. Tetrahedron Lett. 1971, 12, 2467−2470. (28) Wettermark, G. Light-Induced Isomerization of O-Nitrotoluene in Water Solution. J. Phys. Chem. 1962, 66, 2560−2562. (29) Wettermark, G.; Ricci, R. General Acid Catalysis of the Fading of Photoisomerized 2,4-Dinitrotoluene. J. Chem. Phys. 1963, 39, 1218−1223. (30) Ziane, O.; Casalegno, R.; Corval, A. Multistep photoinduced proton transfer in crystalline 2-(2′,4′-dinitrobenzyl)pyridine. Chem. Phys. 1999, 250, 199−206. (31) McDonald, R. S. Review: Infrared Spectrometry. Anal. Chem. 1986, 58, 1906−1925. (32) Opilik, L.; Schmid, T.; Zenobi, R. Modern Raman Imaging: Vibrational Spectroscopy on the Micrometer and Nanometer Scales. Annu. Rev. Anal. Chem. 2013, 6, 379−398. (33) Evans, C. L.; Xie, X. S. Coherent Anti-Stokes Raman Scattering Microscopy: Chemical Imaging for Biology and Medicine. Annu. Rev. Anal. Chem. 2008, 1, 883−909. (34) Polli, D.; Kumar, V.; Valensise, C. M.; Marangoni, M.; Cerullo, G. Broadband Coherent Raman Scattering Microscopy. Laser Photonics Rev. 2018, 12, 1800020. (35) Corval, A.; Kuldová, K.; Eichen, Y.; Pikramenou, Z.; Lehn, J. M.; Trommsdorff, H. P. Photochromism and Thermochromism Driven by Intramolecular Proton Transfer in Dinitrobenzylpyridine Compounds. J. Phys. Chem. 1996, 100, 19315−19320. (36) Proffitt, W.; Porto, S. P. S. Depolarization Ratio in Raman Spectroscopy as a Function of Frequency. J. Opt. Soc. Am. 1973, 63, 77−80. (37) Nissum, M.; Funk, J.-M.; Kiefer, W. Polarization-Sensitive Resonance CARS Spectroscopy of Magnesium Octaethylporphine. J. Raman Spectrosc. 1999, 30, 605−610. (38) Oudar, J.-L.; Smith, R. W.; Shen, Y. R. Polarization-Sensitive Coherent Anti-Stokes Raman Spectroscopy. Appl. Phys. Lett. 1979, 34, 758−760. (39) Brakel, R.; Mudogo, V.; Schneider, F. W. Polarization Sensitive Resonance CARS Spectroscopy. J. Chem. Phys. 1986, 84, 2451−2458. (40) Fishman, A. I.; Klimovitskii, A. E.; Skvortsov, A. I. CARS in Conformational Analysis. Polarization CARS Spectra of Chlorocyclohexane. J. Raman Spectrosc. 1997, 28, 623−625. (41) Saito, Y.; Ishibashi, T. a.; Hamaguchi, H. o. Polarizationresolved coherent anti-Stokes Raman scattering (CARS) spectroscopy: a new probe of molecular symmetry through accurate

The authors declare no competing financial interest.

■

REFERENCES

(1) Laermer, F.; Elsaesser, T.; Kaiser, W. Femtosecond spectroscopy of excited-state proton transfer in 2-(2′-hydroxyphenyl)benzothiazole. Chem. Phys. Lett. 1988, 148, 119−124. (2) Schwartz, B. J.; Peteanu, L. A.; Harris, C. B. Direct Observation of Fast Proton Transfer: Femtosecond Photophysics of 3-Hydroxyflavone. J. Phys. Chem. 1992, 96, 3591−3598. (3) Smith, T. P.; Zaklika, K. A.; Thakur, K.; Walker, G. C.; Tominaga, K.; Barbara, P. F. Spectroscopic Studies of Excited-State Intramolecular Proton Transfer in 1-(Acylamino)Anthraquinones. J. Phys. Chem. 1991, 95, 10465−10475. (4) Tschitschibabin, A. E.; Kuindzhi, B. M.; Benewolenskaja, S. W. Ü ber Nitrierungsprodukte des α- und des γ-Benzyl-Pyridins. Ber. Dtsch. Chem. Ges. 1925, 58, 1580−1587. (5) Hardwick, R.; Mosher, H. S.; Passailaigue, P. Photochromotropic behaviour of 2-(2', 4'-dinitrobenzyl)-pyridine. Trans. Faraday Soc. 1960, 56, 44−50. (6) Klemm, E.; Klemm, D.; Graness, A.; Kleinschmidt, J. Investigation of photochromism of 2-(2′,4′-dinitrobenzyl)-pyridine by nanosecond laser absorption spectroscopy. Chem. Phys. Lett. 1978, 55, 113−115. (7) Mosher, H. S.; Hardwick, E. R.; Hur, D. B. Flash-Photolysis Studies of α- and γ-(2,4-Dinitrobenzyl)pyridine. J. Chem. Phys. 1962, 37, 904−909. (8) Mosher, H. S.; Souers, C.; Hardwick, R. Photochromotropism of gamma-(2,4-Dinitrobenzyl)-Pyridine in Solution. J. Chem. Phys. 1960, 32, 1888−1889. (9) Scherl, M.; Haarer, D.; Fischer, J.; DeCian, A.; Lehn, J.-M.; Eichen, Y. Proton-Transfer Processes in Well-Defined Media: Experimental Investigation of Photoinduced and Thermal ProtonTransfer Processes in Single Crystals of 2-(2,4-Dinitrobenzyl)pyridine Derivatives. J. Phys. Chem. 1996, 100, 16175−16186. (10) Sixl, H.; Warta, R. Reaction mechanism of photochromic 2(2′,4′-dinitrobenzyl) pyridine (DNBP) single crystals. Chem. Phys. 1985, 94, 147−155. (11) Naumov, P.; Sakurai, K.; Ishikawa, T.; Takahashi, J.; Koshihara, S.-y.; Ohashi, Y. Intramolecular Nitro-Assisted Proton Transfer in Photoirradiated 2-(2’,4’-Dinitrobenzyl)pyridine: Polarized Optical Spectroscopic Study and Electronic Structure Calculations. J. Phys. Chem. A 2005, 109, 7264−7275. (12) Mitra, S.; Ito, H.; Tamai, N. Transient behavior of 2-(2′,4′dinitrobenzyl)pyridine photochromism studied by ultrafast laser spectroscopy. Chem. Phys. 2004, 306, 185−189. (13) Takeda, J.; Chung, D. D.; Zhou, J.; Nelson, K. A. Intramolecular proton transfer of a 2-(2′,4′-dinitrobenzyl)pyridine studied by femtosecond transient absorption spectroscopy. Chem. Phys. Lett. 1998, 290, 341−348. (14) Kimura, H.; Sakata, K.; Takahashi, H. Time-Resolved Resonance Raman and Absorption Studies of Orthonitrobenzyl Compounds: Primary Process of Photochromic Reaction. J. Mol. Struct. 1993, 293, 1−4. (15) Takahashi, H.; Hirukawa, S.; Suzuki, S.; Torii, Y.; Isaka, H. Time-resolved resonance Raman spectra of transient species of photochromic 2-(2′,4′-dinitrobenzyl) pyridine. J. Mol. Struct. 1986, 146, 91−99. (16) Takahashi, H.; Kobayashi, Y.; Kaneko, N.; Igarashi, T.; Yagi, I. Resonance Raman spectra of short-lived species of photochromic 2and 4-(2′,4′-dinitrobenzyl)pyridine. J. Raman Spectrosc. 1982, 12, 125−129. (17) Takahashi, H.; Watanabe, Y.; Sakai, M.; Tachikawa, M. Photoinduced Intramolecular Hydrogen Transfer Reaction of Ortho Nitrobenzyl Compounds. Laser Chem. 1999, 19, 357−362. (18) Takahashi, H.; Werncke, W.; Pfeiffer, M.; Lau, A.; Johr, T. Structural Changes in Excited States and Photoisomers Determined by Time-Resolved Raman Scattering. Appl. Phys. B 1994, 59, 403− 414. F


Article

The Journal of Physical Chemistry A determination of the Raman depolarization ratio. J. Raman Spectrosc. 2000, 31, 725−730. (42) Kamga, F. M.; Sceats, M. G. Pulse-Sequenced Coherent AntiStokes Raman Scattering Spectroscopy: A Method for Suppression of the Nonresonant Background. Opt. Lett. 1980, 5, 126−128. (43) Kumar, V.; Osellame, R.; Ramponi, R.; Cerullo, G.; Marangoni, M. Background-Free Broadband CARS Spectroscopy from a 1-MHz Ytterbium Laser. Opt. Express 2011, 19, 15143−15148. (44) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H. et al. Gaussian 16, Rev. B.01, Wallingford, CT, 2016.

G


Vibrational Spectroscopic Characterization of 2-(2,4-Dinitrobenzyl

Recommend Documents