- Email: [email protected]
Ultrafast photodissociation dynamics of diphenylcyclopropenone studied by time-resolved impulsive stimulated Raman spectroscopy
Chemical Physics xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Chemical Physics journal homepage: www.elsevier.com/locate/chemphys
Ultrafast photodissociation dynamics of diphenylcyclopropenone studied by time-resolved impulsive stimulated Raman spectroscopy Hikaru Kuramochi a,b,c, Satoshi Takeuchi a,b, Tahei Tahara a,b,⇑ a
Molecular Spectroscopy Laboratory, RIKEN, 2-1 Hirosawa, Wako 351-0198, Japan Ultrafast Spectroscopy Research Team, RIKEN Center for Advanced Photonics (RAP), 2-1 Hirosawa, Wako 351-0198, Japan c PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi 332-0012, Japan b
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Time-resolved Raman spectroscopy Impulsive stimulated Raman Photodissociation Diphenylcyclopropenone
a b s t r a c t We studied ultrafast photodissociation reaction of diphenylcyclopropenone (DPCP) in solution by timeresolved impulsive stimulated Raman spectroscopy (TR-ISRS) using sub-6-fs pulses. The obtained femtosecond time-resolved Raman data of excited-state DPCP did not show any noticeable spectral change during its lifetime of 180 fs. This indicates that photoexcited DPCP is trapped at the excited-state potential energy minimum before undergoing the photodissociation, which is consistent with the predissociation picture of ultrafast photodissociation reaction of DPCP. The present study demonstrates the high capability of TR-ISRS for structural characterization of short-lived reactive transients that decay within only a few hundreds of femtoseconds. Ó 2018 Published by Elsevier B.V.
1. Introduction Chemical reactions consist of the formation and dissociation of chemical bonds, which involve various skeletal deformations in each elementary step. Revealing such nuclear rearrangements during chemical reactions is one of the most essential issues in fundamental studies of chemical reactions. Toward this goal, it is highly desirable to record a series of snapshot structures of a reacting molecule from the reactant, all the way down to the product with high time resolution. Nevertheless, this is impractical for thermally activated chemical reactions because it is generally difficult to start them by an external trigger with high temporal accuracy. On the other hand, in ultrafast photochemical reactions, the reactant molecules are instantaneously excited to an unstable region on the excited-state potential energy surface (PES), and they simultaneously start structural changes immediately. Therefore, we have a chance to observe continuous structural evolution of reacting molecules towards the photoproduct formation in this case. Observation of such a continuous structural change provides essential information about the nuclear motion responsible for facilitating the reaction (i.e., reaction coordinates), and thus enables us to clarify the relevant reactive PESs. Even for photochemical reactions, however, such observation has been limited so far to a few extre-
⇑ Corresponding author at: Molecular Spectroscopy Laboratory, RIKEN, 2-1 Hirosawa, Wako 351-0198, Japan. E-mail address: [email protected] (T. Tahara).
mely fast reaction systems like photoisomerization of the retinal chromophore in visual pigment rhodopsin and cis-stilbene [1,2]. Photodissociation reaction of diphenylcyclopropenone (DPCP) is one of the fastest photochemical reactions to date, and it involves a drastic structural change from the reactant to the product. Upon irradiation of the ultraviolet light, DPCP undergoes a photodissociation reaction, producing diphenylacetylene (DPA) and carbon monoxide [3], with a high quantum yield (1.00 ± 0.03 [4]). Previously, we investigated the excited-state dynamics of DPCP by femtosecond time-resolved absorption spectroscopy [5], and proposed a reaction scheme that is schematically illustrated as path I in Scheme 1. In this scheme, following photoexcitation to the optically allowed S2 state, excited-state DPCP undergoes a 200-fs decay and produces the S2 state of DPA and carbon monoxide. This scheme is based on the observation of the excited-state absorption (ESA) in the visible region upon the photoexcitation of DPCP, which is assignable to S2 DPA having a characteristic several-picosecond lifetime. Similar spectroscopic signatures have also been observed in the transient absorption studies of DPCP carried out by other groups [6,7]. In addition to this path I, another reaction scheme has been proposed, in which S2 DPCP directly generates DPA in the S0 state (path II) [8]. Therefore, further study is necessary to clarify the dynamics and mechanism of this, one of the fastest of all chemical reactions. As demonstrated in the above-mentioned transient absorption studies, the photodissociation reaction of DPCP is completed within only 500 fs. Because significant nuclear rearrangements take place on such a very short time scale, one would expect
https://doi.org/10.1016/j.chemphys.2018.02.023 0301-0104/Ó 2018 Published by Elsevier B.V.
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
2
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Scheme 1. Reaction scheme representing the ultrafast photodissociation of diphenylcyclopropenone (DPCP), which yields diphenylacetylene (DPA) and carbon monoxide as the products.
drastic structural dynamics on the reactant- or product-state PES. Thus, it is intriguing to examine how the molecular structure evolves in the initially prepared S2 state of DPCP (or in photoproduct DPA), which is expected to provide not only a deeper understanding of the photodissociation of DPCP itself but also a wealth of fundamental knowledge about the reactive PESs in general. In this paper, we report on the ultrafast photodissociation dynamics of DPCP studied by time-resolved impulsive stimulated Raman spectroscopy (TR-ISRS) using sub-6-fs pulses [2,9–13]. This time-domain Raman approach allows us to track the change in the vibrational structure of the transient species with femtosecond temporal accuracy and a very wide frequency window ranging from the terahertz to >3000 cm1 region. Using TR-ISRS, we previously reported an attempt to track continuous structural evolution of cis-stilbene during its ultrafast photoisomerization [2]. More recently, we used TR-ISRS to investigate ultrafast photoreaction dynamics of several photo-responsive proteins [14,15], clarifying key structural events that drive their functions. These studies demonstrated the exquisite capability of TR-ISRS to study the structural dynamics occurring on the femto-to-picosecond time scale. In this study, we examined the structural dynamics of DPCP in the excited state, aiming at obtaining new insights into the mechanism of its ultrafast photodissociation reaction. 2. Experimental 2.1. Materials Diphenylcyclopropenone (98%) was purchased from Sigma Aldrich. It was used after the separation by column chromatography and subsequent recrystallization from cyclohexane. Spectroscopic-grade cyclohexane was purchased from Wako Pure Chemicals and used without further purification. To avoid the accumulation of product DPA during the measurement, a large amount of the sample solution (500 mL) was prepared and circulated. We estimated that 2% of DPCP was converted to DPA after the TR-ISRS measurement. Considering the much lower molar extinction coefficient of DPA at the actinic pump wavelength (e = 430 M1 cm1 at 315 nm) compared to DPCP (e = 15300 M1 cm1 at 315 nm), contribution of the accumulated DPA to the TRISRS data was estimated to be negligible. 2.2. Time-resolved impulsive stimulated Raman spectroscopy Details of the TR-ISRS setup have been described elsewhere [13]. Briefly, a Ti:sapphire regenerative amplifier (780 nm, 80 fs,
1 mJ, 1 kHz) pumped two noncollinear optical parametric amplifiers (NOPAs). The output of the first NOPA at 528 nm was sum-frequency mixed with the fundamental in a type-II BBO (b-barium borate) crystal, generating the actinic pump pulse (P1, 315 nm, 45 fs). The output of the second NOPA (460–640 nm, 5.8 fs) was divided into two, and they were used as the impulsive Raman pump pulse (P2) and the probe pulse (P3) to monitor the transient absorbance change. The pulse durations were evaluated at the sample position by self-diffraction frequency-resolved optical gating (SD-FROG [16]). The P1, P2 and P3 pulses were focused into a 300-lm-thick flow cell of the sample solution. The sample was flowed at a rate sufficient to replenish the sample between consecutive pulses. All the data were obtained with the P1 polarization parallel to the others. At the sample position, the energies of the P1, P2, and P3 pulses were typically 120, 95, and 4 nJ, with 160, 100, and 90 lm beam diameters, respectively. The TR-ISRS data were obtained by mechanically chopping every other P2 pulse, while the P2 pulse was physically blocked and the P1 pulse was chopped for the conventional pump-probe measurement. 2.3. Theoretical calculations Geometry optimizations of the ground state and S2 state of DPCP were carried out by the density functional theory (DFT) and time-dependent DFT methods, respectively, using the B3LYP functional and the 6-311 + G⁄⁄ basis set. All the theoretical calculations were performed with the Gaussian 09 suite program [17]. 3. Results and discussion We first discuss the pump-probe data of DPCP. Fig. 1A shows the steady-state and transient absorption spectra of DPCP in cyclohexane, as well as the spectra of the optical pulses used for the measurements. The actinic pump pulse (315 nm, 45 fs) was tuned to the red edge of the S2 S0 absorption band of DPCP, and the broadband probe pulse (460–640 nm, 5.8 fs) monitored the Sn S2 ESA of DPCP. (Note that the Sn S2 ESA band of product DPA is also within this probe spectral range, so that S2 DPA can be detected once it is generated by the photodissociation reaction [5].) The obtained pump-probe signal of DPCP is shown in Fig. 1B. The data show a rapid decay of the ESA signal with a time constant of 180 fs, which is in a good agreement with our previous pumpprobe measurements of DPCP [5]. Nevertheless, the present data only exhibits the complete decay of the ESA of DPCP, and we did not observe any noticeable residual transient absorption signal assignable to S2 DPA that is formed through the photodissociation reaction (path I). It is in sharp contrast to the result of the previous study in which the ESA of S2 DPA was clearly observed. The absence of the residual transient absorption signal implies that the formation of S2 DPA is negligible and that DPA is directly formed in its ground state under the present experimental condition (path II). We tentatively attribute this discrepancy with the previous pump-probe study to the difference in the actinic pump wavelength: 315 nm in the present study and 267/295 nm in the previous study. We discuss this point more in detail later. Next, ultrafast structural dynamics of S2 DPCP was investigated by TR-ISRS. In Fig. 2A, a schematic of TR-ISRS is shown. In TR-ISRS measurements, firstly, the actinic pump pulse (P1) prepares the excited-state population of the molecule, initiating its photoreaction. After the variable delay time DT, the ultrashort Raman pump pulse (P2) is introduced to induce coherent nuclear wavepacket motion in the excited state through the impulsive stimulated Raman scattering process. The ultrashort probe pulse (P3) records the induced coherent nuclear wavepacket motion as an oscillatory component of the P2-induced differential absorption signal by
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Fig. 1. (A) Spectral data of DPCP in cyclohexane. Red and magenta solid lines denote the steady-state absorption spectrum and transient absorption spectrum measured at 100 fs after photoexcitation (kexc = 315 nm), respectively. Spectra of the P1, P2, and P3 pulses used in the TR-ISRS measurements are also shown at the bottom. (B) Pump-probe signal of DPCP measured with the P1 and P3 pulses. The yellow broken line represents the best fit to the data using a single exponential function convoluted with the instrumental response.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
scanning the delay time s. Through the Fourier transformation of this time-domain vibrational data, a transient Raman spectrum at DT delay time is obtained. Importantly, in this technique, it is possible to define the timing to initiate the Raman transition with the femtosecond accuracy of the delay time (DT), so that the change in the vibrational structure can be tracked on the femtosecond time scale, which is basically impossible with conventional time-resolved spontaneous Raman spectroscopy. For the TR-ISRS measurement of DPCP, the 45-fs pulse at 315 nm was used as the actinic pump pulse (P1) to prepare DPCP in the S2 state. As the impulsive Raman pump and probe pulses (P2 and P3), we used the 5.8-fs pulses in the range of 460–640 nm, which are resonant with the Sn S2 ESA of DPCP, in order to selectively monitor the coherent nuclear wavepacket motion in the S2 state. Fig. 2B shows raw TR-ISRS data of DPCP obtained at various DT delay times, which correspond to the difference between transient absorption signals measured with and without the P2 pulse. These signals show the slowly-varying population dynamics as well as the oscillatory feature due to the coherent nuclear wavepacket motion. (The population component mainly originates from the bleach of S2 DPCP induced by the excitation to the upper excited state (Sn), and its temporal profile represents the recovery of the S2 state population from the Sn state.) Fig. 2C shows the oscillatory components extracted by the subtraction of the slowly-varying
3
population component. The data obtained at the early positive DT delay times clearly show the rapidly-dephasing oscillatory feature due to the coherent nuclear wavepacket motion of S2 DPCP. The dephasing time of these oscillatory signals is comparable to the lifetime of S2 DPCP (180 fs), indicating that it is predominantly determined by the population lifetime of the S2 state. We note that, even after the dephasing of the vibrational coherence of S2 DPCP, a weak long-lived oscillatory feature is seen in all the data. These signals are assignable to the non-resonant ISRS signal of solvent cyclohexane. In TR-ISRS measurements with the open-band detection scheme, which is employed in this study, such non-resonant signals are generally not expected. This is because non-resonant ISRS induces the spectral oscillation of the probe, and ideally does not affect the total intensity of the probe pulse which is recorded in the open-band detection. Nevertheless, non-flat sensitivity of the detector or strong P1-induced transient absorption can potentially act as a color filter, making the non-resonant signals visible [13]. In the present study, we attribute the observed non-resonant ISRS signal to the former case, because its amplitude does not show noticeable dependence on the DT delay time (i.e., P1-induced transient absorption). Fig. 2D shows Fourier transform power spectra of the obtained time-domain vibrational data. A number of Raman bands are clearly observed in the entire frequency range with a high signalto-noise ratio. In the spectra, prominent bands due to S2 DPCP appear at 690, 971, 1170, 1349, 1572, 1669, and 1760 cm-1. This spectral pattern is significantly different from that of the S0 state which is shown in the bottom panel of Fig. 2D. In particular, the characteristic high-frequency C = O stretch band of the S0 state at 1853 cm1 is not seen in the same frequency region of the timeresolved Raman data. Fig. 3 shows the temporal profiles of the Fourier transform amplitudes of the selected bands. These bands show monotonic decays against the DT delay time, and all the temporal profiles perfectly follow the population decay of S2 DPCP (180 fs) which is indicated by the broken lines. These results demonstrate that the femtosecond time-resolved vibrational spectra of S2 DPCP, which has a lifetime as short as 180 fs, are certainly obtained by TR-ISRS. In the femtosecond time-resolved vibrational spectra obtained with TR-ISRS, any frequency shift of a particular band or change in the spectral shape is not noticeable, and all the observed bands monotonically decay with the 180-fs time constant. This result indicates that the structural change of S2 DPCP is not significant during its lifetime. Although definitive assignments for the observed transient Raman bands are difficult at the moment, we tentatively attribute 690, 971, 1170-cm1 bands to the vibrational modes of the phenyl rings. (Raman bands with a similar intensity pattern and frequencies were previously observed for S2 DPA, and they were assigned to the phenyl ring vibrations [13,18].) The characteristic, highest frequency band at 1760 cm1 is likely attributable to the C@O stretch vibration, which is 90 cm-1 down-shifted with respect to the ground state. Because the C@O stretch frequency of the cyclic ketone is known to be sensitive to the bond angle at the carbonyl part [19,20], the down-shifted frequency of the C@O stretch mode of S2 DPCP may reflect slight elongation of the C@C double bond of the three-membered ring, which increases the CA(C@O)AC bend angle. The photodissociation of S2 DPCP has been considered to be predissociation on the excited-state PES, which is generated from a bound PES and a dissociative PES with an avoided crossing. Thus, the S2 PES is expected to exhibit a potential barrier in the region where the two distinct types of PESs are nonadiabatically coupled. In this case, it is expected that the molecules are first excited onto the bound PES region and then move to dissociative PES while vibrating around the potential energy minimum (Fig. 4). Our TRISRS data accord well with this picture; the data indicated that
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
4
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Fig. 2. (A) Schematic of the TR-ISRS experiment. (B) Raw TR-ISRS data of DPCP in cyclohexane obtained at various DT delay times. (C) Oscillatory components of the TR-ISRS data obtained after subtraction of the slowly-varying population component. (D) Fourier transform power spectra of the oscillatory components of the TR-ISRS data. Magnified view is also given in the lower panel. Steady-state spontaneous Raman spectrum of DPCP in ethanol is shown for comparison, which is reprinted with the permission from Ref. [5] (Copyright 2004, AIP publishing).
Fig. 4. Schematic illustration of the potential energy surfaces relevant to the ultrafast photodissociation of diphenylcyclopropenone.
Fig. 3. Temporal profiles of the Fourier transform amplitudes of the prominent bands observed by TR-ISRS. Excited-state population dynamics, obtained by the pump-probe spectroscopy, is also shown with a blue broken line for comparison.
the photo-excited molecule is first trapped around the welldefined potential energy minimum of the S2 PES. Our TD-DFT calculation also supports the existence of the energy minimum for the PES of S2 DPCP. Fig. 5 depicts the molecular structure at the calculated S2 minimum, and its structure closely resembles the structure of the S0 state. Nevertheless, substantial elongation of the central C@C bond is recognized in the S2 structure, which is consistent with the down-shifted C@O stretch frequency observed in the present TR-ISRS measurement. The structural similarity between the S0 and S2 states predicted by the TD-DFT calculation suggests that DPCP only undergoes a minimal structural change from the Franck-Condon state to the S2 potential energy minimum upon photoexcitation. It is expected that such a small initial structural change is completed within a few tens of femtoseconds, considering the vibrational frequency of the relevant deformation mode of the three-membered ring (1608 cm1 in S0, vibrational period of
20 fs). Therefore, the corresponding spectral change was not observed in the present TR-ISRS measurements because of the limitation of the time resolution (45 fs). The dissociation of DPCP occurs from this relaxed S2 state by crossing the shallow potential energy barrier, and the height of this barrier determines the rate of this ultrafast dissociation reaction (180 fs). Although the main focus of this study is to elucidate the structural dynamics of reactant S2 DPCP, it is also intriguing to discuss how the molecular structure of the product relaxes immediately after the rapid formation. In our previous study carried out with 267-nm photoexcitation, it was found that the ESA spectrum of S2 DPA, formed by the dissociation, is remarkably different from the ESA spectrum of S2 DPA obtained by the direct photoexcitation of ground-state DPA. Based on this experimental result, we proposed that the dissociation of S2 DPCP generates DPA having a cis-bent structure, retaining the double-bond character at the central C@C bond [5]. If this is the case, TR-ISRS of product S2 DPA may allow us to track dynamic straightening of the molecular skeleton of DPA, as the triple bond is gradually formed. However, we did not clearly observe S2 DPA as the photoproduct in the present study, which indicates that the yield of S2 DPA is considerably smaller with the 315-nm excitation. In fact, the quantum yield of S2 DPA
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
5
new possibilities for exploration of structural dynamics of the molecules that have been proposed to undergo rapid structural change in the excited state, such as DNA bases [23], retinal proteins [24,25], myoglobin [26], and metal complexes [27–29]. We envision that TR-ISRS can shed new light on these ultrafast reactive systems and provide deeper understanding of reactive PESs. Acknowledgements This work was partly supported by JSPS KAKENHI Grant Numbers JP16H04102 to S. T. and JP25104005 to T. T. The Computations were performed using Research Center for Computational Science, Okazaki, Japan. H.K. was partly supported by RIKEN Special Postdoctoral Researchers (SPDR) program. References
Fig. 5. Optimized structures of the S0 and S2 states of DPCP. Bond lengths are denoted in the Ångström unit.
was reported to be strongly dependent on the excitation wavelength; the quantum yield becomes larger as the excitation wavelength is swept towards the shorter wavelength side across the S2 S0 absorption band of DPCP [6]. In this study, we set the actinic pump wavelength at the red edge of the S2 S0 absorption band (315 nm) to avoid undesired direct excitation of product DPA. Under this experimental condition, it seems that S2 DPA is not efficiently formed and the direct formation of S0 DPA (path II) becomes the predominant relaxation pathway of S2 DPCP. (Formation of S0 DPA with the 315-nm photoexcitation of DPCP was previously confirmed by a time-resolved IR study [8].) We note that, if we tune the actinic pump wavelength to a shorter wavelength by which S2 DPA is generated more efficiently, we might have a chance to observe structural relaxation dynamics on the S2 PES of product DPA. Alternatively, the structural dynamics in the S0 state of DPA could be captured if we tune the probe pulse to the ultraviolet region, where hot S0 DPA can be probed immediately after the formation through the path II. In summary, we studied the ultrafast structural dynamics in the photodissociation of DPCP using TR-ISRS. The obtained femtosecond time-resolved Raman spectra of DPCP indicate that the structural change of the dissociative S2 state is not significant during its lifetime, suggesting the existence of a shallow potential energy barrier along the dissociation reaction coordinate. This is the first demonstration that the time-domain Raman probing can track the structural dynamics of transient molecules having a lifetime as short as 200 fs. We note that, although femtosecond stimulated Raman spectroscopy (FSRS) [21,22], a frequency-domain counterpart of TR-ISRS, can also provide time-resolved Raman spectra on the ultrafast time scale, the time-domain Raman approach has various advantages such as the high sensitivity and simplicity in interpreting obtained spectra. These high capabilities of TR-ISRS offer
[1] P. Kukura, D.W. McCamant, S. Yoon, D.B. Wandschneider, R.A. Mathies, Science 310 (2005) 1006. [2] S. Takeuchi, S. Ruhman, T. Tsuneda, M. Chiba, T. Taketsugu, T. Tahara, Science 322 (2008) 1073. [3] G. Quinkert, K. Opitz, W.W. Wiersdorff, J. Weinlich, Tetrahedron Lett. 4 (1963) 1863. [4] R.W. Fessenden, P.M. Carton, H. Shimamori, J.C. Scaiano, J. Phys. Chem. 86 (1982) 3803. [5] S. Takeuchi, T. Tahara, J. Chem. Phys. 120 (2004) 4768. [6] Y. Hirata, N. Mataga, Chem. Phys. Lett. 193 (1992) 287. [7] S.C. Doan, G. Kuzmanich, M.N. Gard, M.A. Garcia-Garibay, B.J. Schwartz, J. Phys. Chem. Lett. 3 (2011) 81. [8] H. Vennekate, Z. Phys. Chem. 225 (2011) 1089. [9] S. Fujiyoshi, S. Takeuchi, T. Tahara, J. Phys. Chem. A 107 (2003) 494. [10] G. Cerullo, L. Lüer, C. Manzoni, S. De Silvestri, O. Shoshana, S. Ruhman, J. Phys. Chem. A 107 (2003) 8339. [11] J.P. Kraack, A. Wand, T. Buckup, M. Motzkus, S. Ruhman, Phys. Chem. Chem. Phys. 15 (2013) 14487. [12] M. Liebel, P. Kukura, J. Phys Chem. Lett. 4 (2013) 1358. [13] H. Kuramochi, S. Takeuchi, T. Tahara, Rev. Sci. Instrum. 87 (2016) 043107. [14] T. Fujisawa, H. Kuramochi, H. Hosoi, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 138 (2016) 3942. [15] H. Kuramochi, S. Takeuchi, K. Yonezawa, H. Kamikubo, M. Kataoka, T. Tahara, Nat. Chem. 9 (2017) 660. [16] R. Trebino, K.W. DeLong, D.N. Fittinghoff, J.N. Sweetser, M.A. Krumbugel, B.A. Richman, D.J. Kane, Rev. Sci. Instrum. 68 (1997) 3277. [17] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. Montgomery, J. A., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N.J. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1. Gaussian, Inc., Wallingford CT, 2009. [18] T. Ishibashi, H. Hamaguchi, J. Phys. Chem. A 102 (1998) 2263. [19] R.M. Silverstein, F.X. Webster, D.J. Kiemle, D.L. Bryce, Spectrometric identification of organic compounds, John wiley & sons, 2014. [20] K. Iwata, H. Hamaguchi, Raman Spectroscopy, Kodansha Scientific, 2015. [21] M. Yoshizawa, M. Kurosawa, Phys. Rev. A 61 (1999) 013808. [22] P. Kukura, D.W. McCamant, R.A. Mathies, Annu. Rev. Phys. Chem. 58 (2007) 461. [23] C.E. Crespo-Hernández, B. Cohen, P.M. Hare, B. Kohler, Chem. Rev. 104 (2004) 1977. [24] R. Mathies, C. Brito Cruz, W. Pollard, C. Shank, Science 240 (1988) 777. [25] R.W. Schoenlein, L.A. Peteanu, R.A. Mathies, C.V. Shank, Science 254 (1991) 412. [26] L. Zhu, J. Sage, P. Champion, Science 266 (1994) 629. [27] M. Iwamura, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 129 (2007) 5248. [28] M. Iwamura, H. Watanabe, K. Ishii, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 133 (2011) 7728. [29] M. Iwamura, K. Nozaki, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 135 (2013) 538.
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
Contents lists available at ScienceDirect
Chemical Physics journal homepage: www.elsevier.com/locate/chemphys
Ultrafast photodissociation dynamics of diphenylcyclopropenone studied by time-resolved impulsive stimulated Raman spectroscopy Hikaru Kuramochi a,b,c, Satoshi Takeuchi a,b, Tahei Tahara a,b,⇑ a
Molecular Spectroscopy Laboratory, RIKEN, 2-1 Hirosawa, Wako 351-0198, Japan Ultrafast Spectroscopy Research Team, RIKEN Center for Advanced Photonics (RAP), 2-1 Hirosawa, Wako 351-0198, Japan c PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi 332-0012, Japan b
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Time-resolved Raman spectroscopy Impulsive stimulated Raman Photodissociation Diphenylcyclopropenone
a b s t r a c t We studied ultrafast photodissociation reaction of diphenylcyclopropenone (DPCP) in solution by timeresolved impulsive stimulated Raman spectroscopy (TR-ISRS) using sub-6-fs pulses. The obtained femtosecond time-resolved Raman data of excited-state DPCP did not show any noticeable spectral change during its lifetime of 180 fs. This indicates that photoexcited DPCP is trapped at the excited-state potential energy minimum before undergoing the photodissociation, which is consistent with the predissociation picture of ultrafast photodissociation reaction of DPCP. The present study demonstrates the high capability of TR-ISRS for structural characterization of short-lived reactive transients that decay within only a few hundreds of femtoseconds. Ó 2018 Published by Elsevier B.V.
1. Introduction Chemical reactions consist of the formation and dissociation of chemical bonds, which involve various skeletal deformations in each elementary step. Revealing such nuclear rearrangements during chemical reactions is one of the most essential issues in fundamental studies of chemical reactions. Toward this goal, it is highly desirable to record a series of snapshot structures of a reacting molecule from the reactant, all the way down to the product with high time resolution. Nevertheless, this is impractical for thermally activated chemical reactions because it is generally difficult to start them by an external trigger with high temporal accuracy. On the other hand, in ultrafast photochemical reactions, the reactant molecules are instantaneously excited to an unstable region on the excited-state potential energy surface (PES), and they simultaneously start structural changes immediately. Therefore, we have a chance to observe continuous structural evolution of reacting molecules towards the photoproduct formation in this case. Observation of such a continuous structural change provides essential information about the nuclear motion responsible for facilitating the reaction (i.e., reaction coordinates), and thus enables us to clarify the relevant reactive PESs. Even for photochemical reactions, however, such observation has been limited so far to a few extre-
⇑ Corresponding author at: Molecular Spectroscopy Laboratory, RIKEN, 2-1 Hirosawa, Wako 351-0198, Japan. E-mail address: [email protected] (T. Tahara).
mely fast reaction systems like photoisomerization of the retinal chromophore in visual pigment rhodopsin and cis-stilbene [1,2]. Photodissociation reaction of diphenylcyclopropenone (DPCP) is one of the fastest photochemical reactions to date, and it involves a drastic structural change from the reactant to the product. Upon irradiation of the ultraviolet light, DPCP undergoes a photodissociation reaction, producing diphenylacetylene (DPA) and carbon monoxide [3], with a high quantum yield (1.00 ± 0.03 [4]). Previously, we investigated the excited-state dynamics of DPCP by femtosecond time-resolved absorption spectroscopy [5], and proposed a reaction scheme that is schematically illustrated as path I in Scheme 1. In this scheme, following photoexcitation to the optically allowed S2 state, excited-state DPCP undergoes a 200-fs decay and produces the S2 state of DPA and carbon monoxide. This scheme is based on the observation of the excited-state absorption (ESA) in the visible region upon the photoexcitation of DPCP, which is assignable to S2 DPA having a characteristic several-picosecond lifetime. Similar spectroscopic signatures have also been observed in the transient absorption studies of DPCP carried out by other groups [6,7]. In addition to this path I, another reaction scheme has been proposed, in which S2 DPCP directly generates DPA in the S0 state (path II) [8]. Therefore, further study is necessary to clarify the dynamics and mechanism of this, one of the fastest of all chemical reactions. As demonstrated in the above-mentioned transient absorption studies, the photodissociation reaction of DPCP is completed within only 500 fs. Because significant nuclear rearrangements take place on such a very short time scale, one would expect
https://doi.org/10.1016/j.chemphys.2018.02.023 0301-0104/Ó 2018 Published by Elsevier B.V.
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
2
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Scheme 1. Reaction scheme representing the ultrafast photodissociation of diphenylcyclopropenone (DPCP), which yields diphenylacetylene (DPA) and carbon monoxide as the products.
drastic structural dynamics on the reactant- or product-state PES. Thus, it is intriguing to examine how the molecular structure evolves in the initially prepared S2 state of DPCP (or in photoproduct DPA), which is expected to provide not only a deeper understanding of the photodissociation of DPCP itself but also a wealth of fundamental knowledge about the reactive PESs in general. In this paper, we report on the ultrafast photodissociation dynamics of DPCP studied by time-resolved impulsive stimulated Raman spectroscopy (TR-ISRS) using sub-6-fs pulses [2,9–13]. This time-domain Raman approach allows us to track the change in the vibrational structure of the transient species with femtosecond temporal accuracy and a very wide frequency window ranging from the terahertz to >3000 cm1 region. Using TR-ISRS, we previously reported an attempt to track continuous structural evolution of cis-stilbene during its ultrafast photoisomerization [2]. More recently, we used TR-ISRS to investigate ultrafast photoreaction dynamics of several photo-responsive proteins [14,15], clarifying key structural events that drive their functions. These studies demonstrated the exquisite capability of TR-ISRS to study the structural dynamics occurring on the femto-to-picosecond time scale. In this study, we examined the structural dynamics of DPCP in the excited state, aiming at obtaining new insights into the mechanism of its ultrafast photodissociation reaction. 2. Experimental 2.1. Materials Diphenylcyclopropenone (98%) was purchased from Sigma Aldrich. It was used after the separation by column chromatography and subsequent recrystallization from cyclohexane. Spectroscopic-grade cyclohexane was purchased from Wako Pure Chemicals and used without further purification. To avoid the accumulation of product DPA during the measurement, a large amount of the sample solution (500 mL) was prepared and circulated. We estimated that 2% of DPCP was converted to DPA after the TR-ISRS measurement. Considering the much lower molar extinction coefficient of DPA at the actinic pump wavelength (e = 430 M1 cm1 at 315 nm) compared to DPCP (e = 15300 M1 cm1 at 315 nm), contribution of the accumulated DPA to the TRISRS data was estimated to be negligible. 2.2. Time-resolved impulsive stimulated Raman spectroscopy Details of the TR-ISRS setup have been described elsewhere [13]. Briefly, a Ti:sapphire regenerative amplifier (780 nm, 80 fs,
1 mJ, 1 kHz) pumped two noncollinear optical parametric amplifiers (NOPAs). The output of the first NOPA at 528 nm was sum-frequency mixed with the fundamental in a type-II BBO (b-barium borate) crystal, generating the actinic pump pulse (P1, 315 nm, 45 fs). The output of the second NOPA (460–640 nm, 5.8 fs) was divided into two, and they were used as the impulsive Raman pump pulse (P2) and the probe pulse (P3) to monitor the transient absorbance change. The pulse durations were evaluated at the sample position by self-diffraction frequency-resolved optical gating (SD-FROG [16]). The P1, P2 and P3 pulses were focused into a 300-lm-thick flow cell of the sample solution. The sample was flowed at a rate sufficient to replenish the sample between consecutive pulses. All the data were obtained with the P1 polarization parallel to the others. At the sample position, the energies of the P1, P2, and P3 pulses were typically 120, 95, and 4 nJ, with 160, 100, and 90 lm beam diameters, respectively. The TR-ISRS data were obtained by mechanically chopping every other P2 pulse, while the P2 pulse was physically blocked and the P1 pulse was chopped for the conventional pump-probe measurement. 2.3. Theoretical calculations Geometry optimizations of the ground state and S2 state of DPCP were carried out by the density functional theory (DFT) and time-dependent DFT methods, respectively, using the B3LYP functional and the 6-311 + G⁄⁄ basis set. All the theoretical calculations were performed with the Gaussian 09 suite program [17]. 3. Results and discussion We first discuss the pump-probe data of DPCP. Fig. 1A shows the steady-state and transient absorption spectra of DPCP in cyclohexane, as well as the spectra of the optical pulses used for the measurements. The actinic pump pulse (315 nm, 45 fs) was tuned to the red edge of the S2 S0 absorption band of DPCP, and the broadband probe pulse (460–640 nm, 5.8 fs) monitored the Sn S2 ESA of DPCP. (Note that the Sn S2 ESA band of product DPA is also within this probe spectral range, so that S2 DPA can be detected once it is generated by the photodissociation reaction [5].) The obtained pump-probe signal of DPCP is shown in Fig. 1B. The data show a rapid decay of the ESA signal with a time constant of 180 fs, which is in a good agreement with our previous pumpprobe measurements of DPCP [5]. Nevertheless, the present data only exhibits the complete decay of the ESA of DPCP, and we did not observe any noticeable residual transient absorption signal assignable to S2 DPA that is formed through the photodissociation reaction (path I). It is in sharp contrast to the result of the previous study in which the ESA of S2 DPA was clearly observed. The absence of the residual transient absorption signal implies that the formation of S2 DPA is negligible and that DPA is directly formed in its ground state under the present experimental condition (path II). We tentatively attribute this discrepancy with the previous pump-probe study to the difference in the actinic pump wavelength: 315 nm in the present study and 267/295 nm in the previous study. We discuss this point more in detail later. Next, ultrafast structural dynamics of S2 DPCP was investigated by TR-ISRS. In Fig. 2A, a schematic of TR-ISRS is shown. In TR-ISRS measurements, firstly, the actinic pump pulse (P1) prepares the excited-state population of the molecule, initiating its photoreaction. After the variable delay time DT, the ultrashort Raman pump pulse (P2) is introduced to induce coherent nuclear wavepacket motion in the excited state through the impulsive stimulated Raman scattering process. The ultrashort probe pulse (P3) records the induced coherent nuclear wavepacket motion as an oscillatory component of the P2-induced differential absorption signal by
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Fig. 1. (A) Spectral data of DPCP in cyclohexane. Red and magenta solid lines denote the steady-state absorption spectrum and transient absorption spectrum measured at 100 fs after photoexcitation (kexc = 315 nm), respectively. Spectra of the P1, P2, and P3 pulses used in the TR-ISRS measurements are also shown at the bottom. (B) Pump-probe signal of DPCP measured with the P1 and P3 pulses. The yellow broken line represents the best fit to the data using a single exponential function convoluted with the instrumental response.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
scanning the delay time s. Through the Fourier transformation of this time-domain vibrational data, a transient Raman spectrum at DT delay time is obtained. Importantly, in this technique, it is possible to define the timing to initiate the Raman transition with the femtosecond accuracy of the delay time (DT), so that the change in the vibrational structure can be tracked on the femtosecond time scale, which is basically impossible with conventional time-resolved spontaneous Raman spectroscopy. For the TR-ISRS measurement of DPCP, the 45-fs pulse at 315 nm was used as the actinic pump pulse (P1) to prepare DPCP in the S2 state. As the impulsive Raman pump and probe pulses (P2 and P3), we used the 5.8-fs pulses in the range of 460–640 nm, which are resonant with the Sn S2 ESA of DPCP, in order to selectively monitor the coherent nuclear wavepacket motion in the S2 state. Fig. 2B shows raw TR-ISRS data of DPCP obtained at various DT delay times, which correspond to the difference between transient absorption signals measured with and without the P2 pulse. These signals show the slowly-varying population dynamics as well as the oscillatory feature due to the coherent nuclear wavepacket motion. (The population component mainly originates from the bleach of S2 DPCP induced by the excitation to the upper excited state (Sn), and its temporal profile represents the recovery of the S2 state population from the Sn state.) Fig. 2C shows the oscillatory components extracted by the subtraction of the slowly-varying
3
population component. The data obtained at the early positive DT delay times clearly show the rapidly-dephasing oscillatory feature due to the coherent nuclear wavepacket motion of S2 DPCP. The dephasing time of these oscillatory signals is comparable to the lifetime of S2 DPCP (180 fs), indicating that it is predominantly determined by the population lifetime of the S2 state. We note that, even after the dephasing of the vibrational coherence of S2 DPCP, a weak long-lived oscillatory feature is seen in all the data. These signals are assignable to the non-resonant ISRS signal of solvent cyclohexane. In TR-ISRS measurements with the open-band detection scheme, which is employed in this study, such non-resonant signals are generally not expected. This is because non-resonant ISRS induces the spectral oscillation of the probe, and ideally does not affect the total intensity of the probe pulse which is recorded in the open-band detection. Nevertheless, non-flat sensitivity of the detector or strong P1-induced transient absorption can potentially act as a color filter, making the non-resonant signals visible [13]. In the present study, we attribute the observed non-resonant ISRS signal to the former case, because its amplitude does not show noticeable dependence on the DT delay time (i.e., P1-induced transient absorption). Fig. 2D shows Fourier transform power spectra of the obtained time-domain vibrational data. A number of Raman bands are clearly observed in the entire frequency range with a high signalto-noise ratio. In the spectra, prominent bands due to S2 DPCP appear at 690, 971, 1170, 1349, 1572, 1669, and 1760 cm-1. This spectral pattern is significantly different from that of the S0 state which is shown in the bottom panel of Fig. 2D. In particular, the characteristic high-frequency C = O stretch band of the S0 state at 1853 cm1 is not seen in the same frequency region of the timeresolved Raman data. Fig. 3 shows the temporal profiles of the Fourier transform amplitudes of the selected bands. These bands show monotonic decays against the DT delay time, and all the temporal profiles perfectly follow the population decay of S2 DPCP (180 fs) which is indicated by the broken lines. These results demonstrate that the femtosecond time-resolved vibrational spectra of S2 DPCP, which has a lifetime as short as 180 fs, are certainly obtained by TR-ISRS. In the femtosecond time-resolved vibrational spectra obtained with TR-ISRS, any frequency shift of a particular band or change in the spectral shape is not noticeable, and all the observed bands monotonically decay with the 180-fs time constant. This result indicates that the structural change of S2 DPCP is not significant during its lifetime. Although definitive assignments for the observed transient Raman bands are difficult at the moment, we tentatively attribute 690, 971, 1170-cm1 bands to the vibrational modes of the phenyl rings. (Raman bands with a similar intensity pattern and frequencies were previously observed for S2 DPA, and they were assigned to the phenyl ring vibrations [13,18].) The characteristic, highest frequency band at 1760 cm1 is likely attributable to the C@O stretch vibration, which is 90 cm-1 down-shifted with respect to the ground state. Because the C@O stretch frequency of the cyclic ketone is known to be sensitive to the bond angle at the carbonyl part [19,20], the down-shifted frequency of the C@O stretch mode of S2 DPCP may reflect slight elongation of the C@C double bond of the three-membered ring, which increases the CA(C@O)AC bend angle. The photodissociation of S2 DPCP has been considered to be predissociation on the excited-state PES, which is generated from a bound PES and a dissociative PES with an avoided crossing. Thus, the S2 PES is expected to exhibit a potential barrier in the region where the two distinct types of PESs are nonadiabatically coupled. In this case, it is expected that the molecules are first excited onto the bound PES region and then move to dissociative PES while vibrating around the potential energy minimum (Fig. 4). Our TRISRS data accord well with this picture; the data indicated that
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
4
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
Fig. 2. (A) Schematic of the TR-ISRS experiment. (B) Raw TR-ISRS data of DPCP in cyclohexane obtained at various DT delay times. (C) Oscillatory components of the TR-ISRS data obtained after subtraction of the slowly-varying population component. (D) Fourier transform power spectra of the oscillatory components of the TR-ISRS data. Magnified view is also given in the lower panel. Steady-state spontaneous Raman spectrum of DPCP in ethanol is shown for comparison, which is reprinted with the permission from Ref. [5] (Copyright 2004, AIP publishing).
Fig. 4. Schematic illustration of the potential energy surfaces relevant to the ultrafast photodissociation of diphenylcyclopropenone.
Fig. 3. Temporal profiles of the Fourier transform amplitudes of the prominent bands observed by TR-ISRS. Excited-state population dynamics, obtained by the pump-probe spectroscopy, is also shown with a blue broken line for comparison.
the photo-excited molecule is first trapped around the welldefined potential energy minimum of the S2 PES. Our TD-DFT calculation also supports the existence of the energy minimum for the PES of S2 DPCP. Fig. 5 depicts the molecular structure at the calculated S2 minimum, and its structure closely resembles the structure of the S0 state. Nevertheless, substantial elongation of the central C@C bond is recognized in the S2 structure, which is consistent with the down-shifted C@O stretch frequency observed in the present TR-ISRS measurement. The structural similarity between the S0 and S2 states predicted by the TD-DFT calculation suggests that DPCP only undergoes a minimal structural change from the Franck-Condon state to the S2 potential energy minimum upon photoexcitation. It is expected that such a small initial structural change is completed within a few tens of femtoseconds, considering the vibrational frequency of the relevant deformation mode of the three-membered ring (1608 cm1 in S0, vibrational period of
20 fs). Therefore, the corresponding spectral change was not observed in the present TR-ISRS measurements because of the limitation of the time resolution (45 fs). The dissociation of DPCP occurs from this relaxed S2 state by crossing the shallow potential energy barrier, and the height of this barrier determines the rate of this ultrafast dissociation reaction (180 fs). Although the main focus of this study is to elucidate the structural dynamics of reactant S2 DPCP, it is also intriguing to discuss how the molecular structure of the product relaxes immediately after the rapid formation. In our previous study carried out with 267-nm photoexcitation, it was found that the ESA spectrum of S2 DPA, formed by the dissociation, is remarkably different from the ESA spectrum of S2 DPA obtained by the direct photoexcitation of ground-state DPA. Based on this experimental result, we proposed that the dissociation of S2 DPCP generates DPA having a cis-bent structure, retaining the double-bond character at the central C@C bond [5]. If this is the case, TR-ISRS of product S2 DPA may allow us to track dynamic straightening of the molecular skeleton of DPA, as the triple bond is gradually formed. However, we did not clearly observe S2 DPA as the photoproduct in the present study, which indicates that the yield of S2 DPA is considerably smaller with the 315-nm excitation. In fact, the quantum yield of S2 DPA
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023
H. Kuramochi et al. / Chemical Physics xxx (2018) xxx–xxx
5
new possibilities for exploration of structural dynamics of the molecules that have been proposed to undergo rapid structural change in the excited state, such as DNA bases [23], retinal proteins [24,25], myoglobin [26], and metal complexes [27–29]. We envision that TR-ISRS can shed new light on these ultrafast reactive systems and provide deeper understanding of reactive PESs. Acknowledgements This work was partly supported by JSPS KAKENHI Grant Numbers JP16H04102 to S. T. and JP25104005 to T. T. The Computations were performed using Research Center for Computational Science, Okazaki, Japan. H.K. was partly supported by RIKEN Special Postdoctoral Researchers (SPDR) program. References
Fig. 5. Optimized structures of the S0 and S2 states of DPCP. Bond lengths are denoted in the Ångström unit.
was reported to be strongly dependent on the excitation wavelength; the quantum yield becomes larger as the excitation wavelength is swept towards the shorter wavelength side across the S2 S0 absorption band of DPCP [6]. In this study, we set the actinic pump wavelength at the red edge of the S2 S0 absorption band (315 nm) to avoid undesired direct excitation of product DPA. Under this experimental condition, it seems that S2 DPA is not efficiently formed and the direct formation of S0 DPA (path II) becomes the predominant relaxation pathway of S2 DPCP. (Formation of S0 DPA with the 315-nm photoexcitation of DPCP was previously confirmed by a time-resolved IR study [8].) We note that, if we tune the actinic pump wavelength to a shorter wavelength by which S2 DPA is generated more efficiently, we might have a chance to observe structural relaxation dynamics on the S2 PES of product DPA. Alternatively, the structural dynamics in the S0 state of DPA could be captured if we tune the probe pulse to the ultraviolet region, where hot S0 DPA can be probed immediately after the formation through the path II. In summary, we studied the ultrafast structural dynamics in the photodissociation of DPCP using TR-ISRS. The obtained femtosecond time-resolved Raman spectra of DPCP indicate that the structural change of the dissociative S2 state is not significant during its lifetime, suggesting the existence of a shallow potential energy barrier along the dissociation reaction coordinate. This is the first demonstration that the time-domain Raman probing can track the structural dynamics of transient molecules having a lifetime as short as 200 fs. We note that, although femtosecond stimulated Raman spectroscopy (FSRS) [21,22], a frequency-domain counterpart of TR-ISRS, can also provide time-resolved Raman spectra on the ultrafast time scale, the time-domain Raman approach has various advantages such as the high sensitivity and simplicity in interpreting obtained spectra. These high capabilities of TR-ISRS offer
[1] P. Kukura, D.W. McCamant, S. Yoon, D.B. Wandschneider, R.A. Mathies, Science 310 (2005) 1006. [2] S. Takeuchi, S. Ruhman, T. Tsuneda, M. Chiba, T. Taketsugu, T. Tahara, Science 322 (2008) 1073. [3] G. Quinkert, K. Opitz, W.W. Wiersdorff, J. Weinlich, Tetrahedron Lett. 4 (1963) 1863. [4] R.W. Fessenden, P.M. Carton, H. Shimamori, J.C. Scaiano, J. Phys. Chem. 86 (1982) 3803. [5] S. Takeuchi, T. Tahara, J. Chem. Phys. 120 (2004) 4768. [6] Y. Hirata, N. Mataga, Chem. Phys. Lett. 193 (1992) 287. [7] S.C. Doan, G. Kuzmanich, M.N. Gard, M.A. Garcia-Garibay, B.J. Schwartz, J. Phys. Chem. Lett. 3 (2011) 81. [8] H. Vennekate, Z. Phys. Chem. 225 (2011) 1089. [9] S. Fujiyoshi, S. Takeuchi, T. Tahara, J. Phys. Chem. A 107 (2003) 494. [10] G. Cerullo, L. Lüer, C. Manzoni, S. De Silvestri, O. Shoshana, S. Ruhman, J. Phys. Chem. A 107 (2003) 8339. [11] J.P. Kraack, A. Wand, T. Buckup, M. Motzkus, S. Ruhman, Phys. Chem. Chem. Phys. 15 (2013) 14487. [12] M. Liebel, P. Kukura, J. Phys Chem. Lett. 4 (2013) 1358. [13] H. Kuramochi, S. Takeuchi, T. Tahara, Rev. Sci. Instrum. 87 (2016) 043107. [14] T. Fujisawa, H. Kuramochi, H. Hosoi, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 138 (2016) 3942. [15] H. Kuramochi, S. Takeuchi, K. Yonezawa, H. Kamikubo, M. Kataoka, T. Tahara, Nat. Chem. 9 (2017) 660. [16] R. Trebino, K.W. DeLong, D.N. Fittinghoff, J.N. Sweetser, M.A. Krumbugel, B.A. Richman, D.J. Kane, Rev. Sci. Instrum. 68 (1997) 3277. [17] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. Montgomery, J. A., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N.J. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1. Gaussian, Inc., Wallingford CT, 2009. [18] T. Ishibashi, H. Hamaguchi, J. Phys. Chem. A 102 (1998) 2263. [19] R.M. Silverstein, F.X. Webster, D.J. Kiemle, D.L. Bryce, Spectrometric identification of organic compounds, John wiley & sons, 2014. [20] K. Iwata, H. Hamaguchi, Raman Spectroscopy, Kodansha Scientific, 2015. [21] M. Yoshizawa, M. Kurosawa, Phys. Rev. A 61 (1999) 013808. [22] P. Kukura, D.W. McCamant, R.A. Mathies, Annu. Rev. Phys. Chem. 58 (2007) 461. [23] C.E. Crespo-Hernández, B. Cohen, P.M. Hare, B. Kohler, Chem. Rev. 104 (2004) 1977. [24] R. Mathies, C. Brito Cruz, W. Pollard, C. Shank, Science 240 (1988) 777. [25] R.W. Schoenlein, L.A. Peteanu, R.A. Mathies, C.V. Shank, Science 254 (1991) 412. [26] L. Zhu, J. Sage, P. Champion, Science 266 (1994) 629. [27] M. Iwamura, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 129 (2007) 5248. [28] M. Iwamura, H. Watanabe, K. Ishii, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 133 (2011) 7728. [29] M. Iwamura, K. Nozaki, S. Takeuchi, T. Tahara, J. Am. Chem. Soc. 135 (2013) 538.
Please cite this article in press as: H. Kuramochi et al., Chem. Phys. (2018), https://doi.org/10.1016/j.chemphys.2018.02.023