- Email: [email protected]
Step-scan Fourier transform infrared absorption spectroscopy of liquid benzene cluster in a pulsed supersonic jet
Chemical Physics Letters 366 (2002) 28–33 www.elsevier.com/locate/cplett
Step-scan Fourier transform infrared absorption spectroscopy of liquid benzene cluster in a pulsed supersonic jet Shinichi Hirabayashi *, Yasuhiro Hirahara Division of Earth and Environmental Sciences, Graduate School of Environmental Studies, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan Received 8 July 2002; in final form 10 September 2002
Abstract We have obtained the infrared absorption spectrum of the jet-cooled C6 H6 in the 800–5000 cm1 region using stepscan Fourier transform infrared spectrometer combined with a pulsed supersonic jet system. The experimental spectrum was in good agreement with the convolution spectrum of the monomer band with the rotational temperature of 37 K and the broadbands of liquid C6 H6 , indicating the first spectroscopic evidence for the formation of supercooled liquid C6 H6 cluster in the supersonic jet. Ó 2002 Elsevier Science B.V. All rights reserved.
1. Introduction Adiabatic cooling in a supersonic jet provides significant simplification of faint and complex spectra of large molecules. The supersonic molecular beam can also produce super-saturation condition, and thus is a suitable tool for the production of clusters and condensed matters. Bartell et al. [1] have studied a variety of large clusters such as those of C2 H2 [2] and C6 H6 [3,4] in the supersonic jet by electron diffraction. They have indicated that the C6 H6 clusters existed as liquid under their experimental conditions using 1–22%
*
Corresponding author. Fax: +81-52-789-3013. E-mail address: [email protected] (S. Hirabayashi).
C6 H6 in He, Ne, and Ar at the stagnation pressures of 0.14–2.7 bar. Fourier transform infrared (FTIR) spectroscopy combined with the supersonic jet has been applied to the infrared absorption measurement of polyatomic molecules and clusters [5–10]. Snavely et al. [7] have used a continuous-scan FTIR spectrometer with a spectral resolution of 0.06 cm1 to record the broadband infrared absorption spectrum of C6 H6 in the 400–4000 cm1 region. In their experiment, the rotational temperature of C6 H6 monomer has been analyzed to be quite high (79 15 K) because of the continuous supersonic jet and low stagnation pressure, and no infrared band of C6 H6 clusters has been confirmed. Recently, we have successfully measured the infrared absorption bands of large C2 H2 cluster in the cubic phase of the solid, using FTIR spectroscopy [11]. In our experiments, the pulsed supersonic jet was
0009-2614/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 2 6 1 4 ( 0 2 ) 0 1 5 2 8 - 2
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
combined with step-scan FTIR spectrometer, the moving mirror of which was scanned stepwise. In this Letter, we present the infrared absorption spectrum of the jet-cooled C6 H6 between 800 and 5000 cm1 at a spectral resolution of 4 cm1 . From the analysis of three fundamental bands in the measured infrared spectrum, the production of liquid C6 H6 in the supersonic jet was confirmed for the first time by infrared spectroscopic study.
2. Experimental The apparatus employed in the study is a stepscan FTIR spectrometer (Bio-Rad Co. FTS-6000) combined with a pulsed nozzle (General Valve Co. 9-181-900) with 0.8 mm diameter orifice. Details of the experimental setup have been described elsewhere [11]. The resistance heaters were used between the reservoir and the nozzle head to obtain sufficient vapor pressure of C6 H6 . In order to prevent condensation, the nozzle was kept at a temperature higher than the sample reservoir temperature of 316 K, which corresponds to the vapor pressure of C6 H6 of 28 kPa. C6 H6 diluted in Ar was expanded into a vacuum chamber through a heated tube from the pulsed nozzle at a stagnation pressure of 300 kPa. The repetition rate of the nozzle operation was 10 Hz, and the jet pulse duration was approximately 1500 ls. The vacuum chamber was evacuated through a liquid nitrogen baffle by a 6 in. oil diffusion pump backed by a 32-l/s mechanical booster pump and a 5-l/s rotary pump. The pressure in the vacuum chamber was kept about 104 Torr during jet operation. In order to increase the absorption length, a multipass optics was incorporated in the vacuum chamber, allowing the infrared beam to travel seven times through the supersonic jet. The infrared spectrum of C6 H6 was recorded by the FTIR spectrometer equipped with step-scan mode. The A/D converter trigger control signal of FTIR provides a repetitive trigger to the external experiment cycling at each step. The pulsed nozzle was synchronized with this trigger signal using a timing controller and a nozzle driver. The FTIR spectrometer was run in the time-resolved stepscan mode to obtain the absorbance spectra. The
29
spectral resolution and time resolution used in the present experiment were 4 cm1 and 1000 ls, respectively, and the observable wavenumber region was 800–5000 cm1 with a ceramic lamp, a liquidN2 -cooled narrow band MCT detector, and a KBr beamsplitter. In the step-scan operation, the moving mirror of the interferometer does not translate at a constant speed, but is stationary at each optical path difference point while the signal is digitized during sampling time. The sampling rate of the A/D converter is 400 kHz, and the sampling points are co-added at each time point within a step to increase the signal-to-noise (S=N ) ratio. In the present experiment, data were recorded at 5 time points (0, 1000, 2000, 3000, and 4000 ls). After a short time (0.1 s), the mirror is moved to the next position and the process is repeated. The interferograms measured at each time are Fourier-transformed, and absorbance data are obtained using the spectra recorded at time points of 1000 and 2000 ls as sample and background, respectively.
3. Results and discussion The infrared spectrum of the jet-cooled C6 H6 obtained in the present study is shown in Fig. 1a. The m12 , m13 , and m14 modes in Herzberg notation [12] of the four infrared active fundamentals of C6 H6 were identified in the 3070, 1480, and 1035 cm1 regions, respectively. The m4 vibrational mode at 673 cm1 could not be measured because the observable wavenumber region was limited due to the use of the narrow band MCT detector and the KBr beamsplitter. In the m12 C–H stretching region, three peaks at 3049, 3078, and 3102 cm1 were assigned to the m12 , m13 þ m16 , and m2 þ m13 þ m18 modes, respectively. The infrared bands for m12 , m13 , and m14 modes had full-width at half maximums (FWHMs) of 16, 7, and 11 cm1 , respectively, showing sharper band contours than those at room temperature gas cell, as presented in Fig. 1. The measured bands could not be resolved to P-, Q-, and R-branches with the present spectral resolution. We analyzed the observed bands of C6 H6 by the simulation of the rotational band contours, and
30
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
Fig. 1. Experimental infrared absorption spectra of: (a) jet-cooled C6 H6 compared with (b) room-temperature C6 H6 in the 800–4000 cm1 region.
determined the rotational temperature. The observed spectra were fitted to the spectra calculated by using the spectroscopic constants [13–16], H€ onl–London factors [12], nuclear spin statistical weights [17], and Boltzmann factors. The calculated line positions and intensities were convoluted with rotational spectral lines having FWHMs of the resolution of the experimental spectrum. Figs. 2 and 3 show the simulated spectra of C6 H6 monomer in the 1480 and 3050 cm1 , respectively. The peak positions for m13 and m14 bands in the measured spectrum were shifted by 4 and 2 cm1 , respectively, to the high wavenumber compared with the simulated bands, as shown in Fig. 2. These observed bands cannot be reproduced with any rotational temperature in terms of peak position and band profile. On the other hand, the simulated m12 band in the 3050 cm1 region can represent the experimental spectrum except for the low-wavenumber side, and the peak position is in good agreement with the simulation. From the high-wavenumber band contour of the m12 band between 3050 and 3060 cm1 , the rotational temperature of C6 H6 monomer was determined to be 37 6 K, in the range where it is expected from the present pulsed jet experiment with the higher stagnation pressure and lower concentration compared with those in the previous continuous jet study (79 15 K) [7]. The result of the fitting
Fig. 2. Experimental infrared absorption spectrum for the fundamental m13 band of C6 H6 compared with the simulated monomer spectra with two rotational temperatures of 10 and 50 K.
and the fit residual are presented in Fig. 3a,b, respectively. This fit residual was reproduced by the use of a Gaussian component profile. Fig. 3c shows the convolution of the simulated monomer band with the rotational temperature of 37 K and Gaussian component. This unassigned broadband is relatively strong and is shifted to the red by 10 cm1 compared to the monomer band,
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
31
Fig. 4. The least-squares fits of the sum of the simulated monomer and cluster bands (lower trace) to the experimental bands (upper trace) in the (a) 1480 and (b) 1035 cm1 regions, respectively. The simulations assume a rotational temperature of 37 K.
Fig. 3. (a) Simulated C6 H6 monomer spectrum compared with the experimental spectrum in the 3050 cm1 region. (b) The residual obtained by subtraction of the simulated monomer band from the experimental spectrum. (c) The sum of the simulated monomer and cluster bands.
suggesting that this feature is most likely due to the large C6 H6 cluster. For m13 and m14 bands, we applied multi-component fitting to the observed band shapes using Gaussian components corresponding to the broad cluster bands together with C6 H6 monomer spectrum. The analysis was performed under 37 K rotational temperature condition. The results of the fitting are presented in Fig. 4. The determined peak wavenumbers and relative intensities for the simulated monomer and cluster bands are listed in Table 1. The wavenumbers of the large C6 H6 cluster bands for m12 , m13 , and m14 modes obtained from the fitting were 3041, 1479, and 1035 cm1 , respectively, which are red-shifted by 3–9 cm1 compared to those of the monomer bands. These values agree with those of infrared spectrum of
liquid C6 H6 reported by Bertie et al. [18] in Table 1. In addition, large C6 H6 clusters have been found to have liquid structure in the supersonic jets by the previous electron diffraction studies [3,4]. These data suggest that the clusters generated in the present supersonic jet are due to supercooled liquid droplets. Furthermore, the FWHMs and the relative intensity ratios support the suggested identification. The FWHMs of the broad cluster bands for m12 , m13 , and m14 bands are 9, 5.5, and 8 cm1 , respectively, which agree with those of liquid spectrum. The relative intensity ratios for m12 , m13 , and m14 bands in the simulated spectrum are also in reasonably good agreement with the infrared bands of liquid C6 H6 at 298 K, as shown in Table 1. Therefore, we conclude that this is the first infrared spectroscopic detection of liquid C6 H6 cluster in the supersonic jet. Finally, we briefly discuss the relation between the formation of C6 H6 clusters and the design of the nozzle. The nozzle used here is operated as pulsed, not continuous, and the diameter of the nozzle (0.8 mm) is quite larger than that of the previous electron diffraction studies (0.115 mm) [3,4], where only C6 H6 microdrops have been detected in a variety of expansion conditions. As mentioned in [4], high C6 H6 concentration, high stagnation pressure, and Ar carrier gas were
32
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
Table 1 Wavenumbers and relative intensities for the experimental and simulated C6 H6 spectra Band
ml4 ml3 m12 m13 þ m16 m2 þ m13 þ m18 a
Experimental
Experimentala
Simulated Monomer
Cluster
Liquid (298 K)
m ðcm1 Þ
Irel
m ðcm1 Þ
m ðcm1 Þ
Irel
m ðcm1 Þ
Irel
1035 1479 3050 3078 3102
0.35 0.97 1.00 0.31 0.62
1038 1484 3050 ... ...
1035 1479 3041 ... ...
0.55 1.48 1.00 ... ...
1036 1479 3036 3071 3091
0.45 1.40 1.00 0.41 0.56
Ref. [18].
suitable for generation of C6 H6 microdrops, which is similar to the jet conditions in the present experiment. In addition, the present nozzle geometry consists of the units with full length of 19 mm and circular holes of 3–5 mm diameter. This allows time for large clusters to assemble because of a greater number of collisions prior to expansion. The comparison indicates that our pulsed nozzle is more likely suitable for the formation of the large C6 H6 cluster, which has liquid structures.
4. Conclusion Step-scan FTIR absorption spectroscopy combined with the pulsed supersonic jet technique has been applied to the observation of the vibrational bands of C6 H6 . A total of seven infrared transitions, simplified owing to rotational cooling, were observed in the region from 800 to 5000 cm1 . The observed fundamental bands of C6 H6 could be explained by the calculated spectra of C6 H6 monomer with the rotational temperature 37 K and Gaussian components. The peak positions, bandwidths, and relative intensity ratios of the Gaussian components are in good agreement with those of liquid C6 H6 , providing the first spectroscopic evidence for the formation of liquid C6 H6 cluster in the supersonic jet. The present technique is demonstrated to be suitable for the investigation of infrared spectra of liquid and crystalline clusters, as well as the previous study of solid C2 H2 cluster [11].
Acknowledgements The present work was supported by Grantsin-Aid by the Ministry of Education, Culture, Sports, Science, and Technology of Japan (No. 12740120).
References [1] L.S. Bartell, L. Harsanyi, E.J. Valente, J. Phys. Chem. 93 (1989) 6201. [2] L.S. Bartell, L. Harsanyi, T.S. Dibble, P.J. Lennon, J. Phys. Chem. 94 (1990) 6009. [3] R.K. Heenan, E.J. Valente, L.S. Bartell, J. Chem. Phys. 78 (1983) 243. [4] E.J. Valente, L.S. Bartell, J. Chem. Phys. 80 (1984) 1451. [5] D.L. Snavely, S.D. Colson, K.B. Wiberg, J. Chem. Phys. 74 (1981) 6975. [6] D.L. Snavely, K.B. Wiberg, S.D. Colson, Chem. Phys. Lett. 96 (1983) 319. [7] D.L. Snavely, V.A. Walters, S.D. Colson, K.B. Wiberg, Chem. Phys. Lett. 103 (1984) 73. [8] A. Amrein, M. Quack, U. Schmitt, J. Phys. Chem. 92 (1988) 5455. [9] M. Quack, U. Schmitt, M.A. Suhm, Chem. Phys. Lett. 269 (1997) 29. [10] T. H€aber, U. Schmitt, M.A. Suhm, Phys. Chem. Chem. Phys. 1 (1999) 5573. [11] S. Hirabayashi, Y. Hirahara, Chem. Phys. Lett. 361 (2002) 265. [12] G. Herzberg, in: Infrared and Raman Spectra of Polyatomic Molecules, vol. II, Van Nostrand Reinhold, New York, 1945. [13] M.-L. Junttila, J.L. Domenech, G.T. Fraser, A.S. Pine, J. Mol. Spectrosc. 147 (1991) 513. [14] J.L. Domenech, M.-L. Junttila, A.S. Pine, J. Mol. Spectrosc. 149 (1991) 391.
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33 [15] M. Ishiguro, K. Harada, K. Tanaka, T. Tanaka, J. Mol. Spectrosc. 192 (1998) 235. [16] A. Uskola, F.J. Basterretxea, F. Castano, J. Mol. Spectrosc. 198 (1999) 429.
33
[17] E.B. Wilson, J. Chem. Phys. 3 (1935) 276. [18] J.E. Bertie, R.N. Jones, C.D. Keefe, Appl. Spectrosc. 47 (1993) 891.
Step-scan Fourier transform infrared absorption spectroscopy of liquid benzene cluster in a pulsed supersonic jet Shinichi Hirabayashi *, Yasuhiro Hirahara Division of Earth and Environmental Sciences, Graduate School of Environmental Studies, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan Received 8 July 2002; in final form 10 September 2002
Abstract We have obtained the infrared absorption spectrum of the jet-cooled C6 H6 in the 800–5000 cm1 region using stepscan Fourier transform infrared spectrometer combined with a pulsed supersonic jet system. The experimental spectrum was in good agreement with the convolution spectrum of the monomer band with the rotational temperature of 37 K and the broadbands of liquid C6 H6 , indicating the first spectroscopic evidence for the formation of supercooled liquid C6 H6 cluster in the supersonic jet. Ó 2002 Elsevier Science B.V. All rights reserved.
1. Introduction Adiabatic cooling in a supersonic jet provides significant simplification of faint and complex spectra of large molecules. The supersonic molecular beam can also produce super-saturation condition, and thus is a suitable tool for the production of clusters and condensed matters. Bartell et al. [1] have studied a variety of large clusters such as those of C2 H2 [2] and C6 H6 [3,4] in the supersonic jet by electron diffraction. They have indicated that the C6 H6 clusters existed as liquid under their experimental conditions using 1–22%
*
Corresponding author. Fax: +81-52-789-3013. E-mail address: [email protected] (S. Hirabayashi).
C6 H6 in He, Ne, and Ar at the stagnation pressures of 0.14–2.7 bar. Fourier transform infrared (FTIR) spectroscopy combined with the supersonic jet has been applied to the infrared absorption measurement of polyatomic molecules and clusters [5–10]. Snavely et al. [7] have used a continuous-scan FTIR spectrometer with a spectral resolution of 0.06 cm1 to record the broadband infrared absorption spectrum of C6 H6 in the 400–4000 cm1 region. In their experiment, the rotational temperature of C6 H6 monomer has been analyzed to be quite high (79 15 K) because of the continuous supersonic jet and low stagnation pressure, and no infrared band of C6 H6 clusters has been confirmed. Recently, we have successfully measured the infrared absorption bands of large C2 H2 cluster in the cubic phase of the solid, using FTIR spectroscopy [11]. In our experiments, the pulsed supersonic jet was
0009-2614/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 2 6 1 4 ( 0 2 ) 0 1 5 2 8 - 2
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
combined with step-scan FTIR spectrometer, the moving mirror of which was scanned stepwise. In this Letter, we present the infrared absorption spectrum of the jet-cooled C6 H6 between 800 and 5000 cm1 at a spectral resolution of 4 cm1 . From the analysis of three fundamental bands in the measured infrared spectrum, the production of liquid C6 H6 in the supersonic jet was confirmed for the first time by infrared spectroscopic study.
2. Experimental The apparatus employed in the study is a stepscan FTIR spectrometer (Bio-Rad Co. FTS-6000) combined with a pulsed nozzle (General Valve Co. 9-181-900) with 0.8 mm diameter orifice. Details of the experimental setup have been described elsewhere [11]. The resistance heaters were used between the reservoir and the nozzle head to obtain sufficient vapor pressure of C6 H6 . In order to prevent condensation, the nozzle was kept at a temperature higher than the sample reservoir temperature of 316 K, which corresponds to the vapor pressure of C6 H6 of 28 kPa. C6 H6 diluted in Ar was expanded into a vacuum chamber through a heated tube from the pulsed nozzle at a stagnation pressure of 300 kPa. The repetition rate of the nozzle operation was 10 Hz, and the jet pulse duration was approximately 1500 ls. The vacuum chamber was evacuated through a liquid nitrogen baffle by a 6 in. oil diffusion pump backed by a 32-l/s mechanical booster pump and a 5-l/s rotary pump. The pressure in the vacuum chamber was kept about 104 Torr during jet operation. In order to increase the absorption length, a multipass optics was incorporated in the vacuum chamber, allowing the infrared beam to travel seven times through the supersonic jet. The infrared spectrum of C6 H6 was recorded by the FTIR spectrometer equipped with step-scan mode. The A/D converter trigger control signal of FTIR provides a repetitive trigger to the external experiment cycling at each step. The pulsed nozzle was synchronized with this trigger signal using a timing controller and a nozzle driver. The FTIR spectrometer was run in the time-resolved stepscan mode to obtain the absorbance spectra. The
29
spectral resolution and time resolution used in the present experiment were 4 cm1 and 1000 ls, respectively, and the observable wavenumber region was 800–5000 cm1 with a ceramic lamp, a liquidN2 -cooled narrow band MCT detector, and a KBr beamsplitter. In the step-scan operation, the moving mirror of the interferometer does not translate at a constant speed, but is stationary at each optical path difference point while the signal is digitized during sampling time. The sampling rate of the A/D converter is 400 kHz, and the sampling points are co-added at each time point within a step to increase the signal-to-noise (S=N ) ratio. In the present experiment, data were recorded at 5 time points (0, 1000, 2000, 3000, and 4000 ls). After a short time (0.1 s), the mirror is moved to the next position and the process is repeated. The interferograms measured at each time are Fourier-transformed, and absorbance data are obtained using the spectra recorded at time points of 1000 and 2000 ls as sample and background, respectively.
3. Results and discussion The infrared spectrum of the jet-cooled C6 H6 obtained in the present study is shown in Fig. 1a. The m12 , m13 , and m14 modes in Herzberg notation [12] of the four infrared active fundamentals of C6 H6 were identified in the 3070, 1480, and 1035 cm1 regions, respectively. The m4 vibrational mode at 673 cm1 could not be measured because the observable wavenumber region was limited due to the use of the narrow band MCT detector and the KBr beamsplitter. In the m12 C–H stretching region, three peaks at 3049, 3078, and 3102 cm1 were assigned to the m12 , m13 þ m16 , and m2 þ m13 þ m18 modes, respectively. The infrared bands for m12 , m13 , and m14 modes had full-width at half maximums (FWHMs) of 16, 7, and 11 cm1 , respectively, showing sharper band contours than those at room temperature gas cell, as presented in Fig. 1. The measured bands could not be resolved to P-, Q-, and R-branches with the present spectral resolution. We analyzed the observed bands of C6 H6 by the simulation of the rotational band contours, and
30
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
Fig. 1. Experimental infrared absorption spectra of: (a) jet-cooled C6 H6 compared with (b) room-temperature C6 H6 in the 800–4000 cm1 region.
determined the rotational temperature. The observed spectra were fitted to the spectra calculated by using the spectroscopic constants [13–16], H€ onl–London factors [12], nuclear spin statistical weights [17], and Boltzmann factors. The calculated line positions and intensities were convoluted with rotational spectral lines having FWHMs of the resolution of the experimental spectrum. Figs. 2 and 3 show the simulated spectra of C6 H6 monomer in the 1480 and 3050 cm1 , respectively. The peak positions for m13 and m14 bands in the measured spectrum were shifted by 4 and 2 cm1 , respectively, to the high wavenumber compared with the simulated bands, as shown in Fig. 2. These observed bands cannot be reproduced with any rotational temperature in terms of peak position and band profile. On the other hand, the simulated m12 band in the 3050 cm1 region can represent the experimental spectrum except for the low-wavenumber side, and the peak position is in good agreement with the simulation. From the high-wavenumber band contour of the m12 band between 3050 and 3060 cm1 , the rotational temperature of C6 H6 monomer was determined to be 37 6 K, in the range where it is expected from the present pulsed jet experiment with the higher stagnation pressure and lower concentration compared with those in the previous continuous jet study (79 15 K) [7]. The result of the fitting
Fig. 2. Experimental infrared absorption spectrum for the fundamental m13 band of C6 H6 compared with the simulated monomer spectra with two rotational temperatures of 10 and 50 K.
and the fit residual are presented in Fig. 3a,b, respectively. This fit residual was reproduced by the use of a Gaussian component profile. Fig. 3c shows the convolution of the simulated monomer band with the rotational temperature of 37 K and Gaussian component. This unassigned broadband is relatively strong and is shifted to the red by 10 cm1 compared to the monomer band,
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
31
Fig. 4. The least-squares fits of the sum of the simulated monomer and cluster bands (lower trace) to the experimental bands (upper trace) in the (a) 1480 and (b) 1035 cm1 regions, respectively. The simulations assume a rotational temperature of 37 K.
Fig. 3. (a) Simulated C6 H6 monomer spectrum compared with the experimental spectrum in the 3050 cm1 region. (b) The residual obtained by subtraction of the simulated monomer band from the experimental spectrum. (c) The sum of the simulated monomer and cluster bands.
suggesting that this feature is most likely due to the large C6 H6 cluster. For m13 and m14 bands, we applied multi-component fitting to the observed band shapes using Gaussian components corresponding to the broad cluster bands together with C6 H6 monomer spectrum. The analysis was performed under 37 K rotational temperature condition. The results of the fitting are presented in Fig. 4. The determined peak wavenumbers and relative intensities for the simulated monomer and cluster bands are listed in Table 1. The wavenumbers of the large C6 H6 cluster bands for m12 , m13 , and m14 modes obtained from the fitting were 3041, 1479, and 1035 cm1 , respectively, which are red-shifted by 3–9 cm1 compared to those of the monomer bands. These values agree with those of infrared spectrum of
liquid C6 H6 reported by Bertie et al. [18] in Table 1. In addition, large C6 H6 clusters have been found to have liquid structure in the supersonic jets by the previous electron diffraction studies [3,4]. These data suggest that the clusters generated in the present supersonic jet are due to supercooled liquid droplets. Furthermore, the FWHMs and the relative intensity ratios support the suggested identification. The FWHMs of the broad cluster bands for m12 , m13 , and m14 bands are 9, 5.5, and 8 cm1 , respectively, which agree with those of liquid spectrum. The relative intensity ratios for m12 , m13 , and m14 bands in the simulated spectrum are also in reasonably good agreement with the infrared bands of liquid C6 H6 at 298 K, as shown in Table 1. Therefore, we conclude that this is the first infrared spectroscopic detection of liquid C6 H6 cluster in the supersonic jet. Finally, we briefly discuss the relation between the formation of C6 H6 clusters and the design of the nozzle. The nozzle used here is operated as pulsed, not continuous, and the diameter of the nozzle (0.8 mm) is quite larger than that of the previous electron diffraction studies (0.115 mm) [3,4], where only C6 H6 microdrops have been detected in a variety of expansion conditions. As mentioned in [4], high C6 H6 concentration, high stagnation pressure, and Ar carrier gas were
32
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33
Table 1 Wavenumbers and relative intensities for the experimental and simulated C6 H6 spectra Band
ml4 ml3 m12 m13 þ m16 m2 þ m13 þ m18 a
Experimental
Experimentala
Simulated Monomer
Cluster
Liquid (298 K)
m ðcm1 Þ
Irel
m ðcm1 Þ
m ðcm1 Þ
Irel
m ðcm1 Þ
Irel
1035 1479 3050 3078 3102
0.35 0.97 1.00 0.31 0.62
1038 1484 3050 ... ...
1035 1479 3041 ... ...
0.55 1.48 1.00 ... ...
1036 1479 3036 3071 3091
0.45 1.40 1.00 0.41 0.56
Ref. [18].
suitable for generation of C6 H6 microdrops, which is similar to the jet conditions in the present experiment. In addition, the present nozzle geometry consists of the units with full length of 19 mm and circular holes of 3–5 mm diameter. This allows time for large clusters to assemble because of a greater number of collisions prior to expansion. The comparison indicates that our pulsed nozzle is more likely suitable for the formation of the large C6 H6 cluster, which has liquid structures.
4. Conclusion Step-scan FTIR absorption spectroscopy combined with the pulsed supersonic jet technique has been applied to the observation of the vibrational bands of C6 H6 . A total of seven infrared transitions, simplified owing to rotational cooling, were observed in the region from 800 to 5000 cm1 . The observed fundamental bands of C6 H6 could be explained by the calculated spectra of C6 H6 monomer with the rotational temperature 37 K and Gaussian components. The peak positions, bandwidths, and relative intensity ratios of the Gaussian components are in good agreement with those of liquid C6 H6 , providing the first spectroscopic evidence for the formation of liquid C6 H6 cluster in the supersonic jet. The present technique is demonstrated to be suitable for the investigation of infrared spectra of liquid and crystalline clusters, as well as the previous study of solid C2 H2 cluster [11].
Acknowledgements The present work was supported by Grantsin-Aid by the Ministry of Education, Culture, Sports, Science, and Technology of Japan (No. 12740120).
References [1] L.S. Bartell, L. Harsanyi, E.J. Valente, J. Phys. Chem. 93 (1989) 6201. [2] L.S. Bartell, L. Harsanyi, T.S. Dibble, P.J. Lennon, J. Phys. Chem. 94 (1990) 6009. [3] R.K. Heenan, E.J. Valente, L.S. Bartell, J. Chem. Phys. 78 (1983) 243. [4] E.J. Valente, L.S. Bartell, J. Chem. Phys. 80 (1984) 1451. [5] D.L. Snavely, S.D. Colson, K.B. Wiberg, J. Chem. Phys. 74 (1981) 6975. [6] D.L. Snavely, K.B. Wiberg, S.D. Colson, Chem. Phys. Lett. 96 (1983) 319. [7] D.L. Snavely, V.A. Walters, S.D. Colson, K.B. Wiberg, Chem. Phys. Lett. 103 (1984) 73. [8] A. Amrein, M. Quack, U. Schmitt, J. Phys. Chem. 92 (1988) 5455. [9] M. Quack, U. Schmitt, M.A. Suhm, Chem. Phys. Lett. 269 (1997) 29. [10] T. H€aber, U. Schmitt, M.A. Suhm, Phys. Chem. Chem. Phys. 1 (1999) 5573. [11] S. Hirabayashi, Y. Hirahara, Chem. Phys. Lett. 361 (2002) 265. [12] G. Herzberg, in: Infrared and Raman Spectra of Polyatomic Molecules, vol. II, Van Nostrand Reinhold, New York, 1945. [13] M.-L. Junttila, J.L. Domenech, G.T. Fraser, A.S. Pine, J. Mol. Spectrosc. 147 (1991) 513. [14] J.L. Domenech, M.-L. Junttila, A.S. Pine, J. Mol. Spectrosc. 149 (1991) 391.
S. Hirabayashi, Y. Hirahara / Chemical Physics Letters 366 (2002) 28–33 [15] M. Ishiguro, K. Harada, K. Tanaka, T. Tanaka, J. Mol. Spectrosc. 192 (1998) 235. [16] A. Uskola, F.J. Basterretxea, F. Castano, J. Mol. Spectrosc. 198 (1999) 429.
33
[17] E.B. Wilson, J. Chem. Phys. 3 (1935) 276. [18] J.E. Bertie, R.N. Jones, C.D. Keefe, Appl. Spectrosc. 47 (1993) 891.