Intact molecular ion formation of cyclohexane and 2,3-dimethyl-1,3-butadiene by excitation with a short, intense femtosecond laser pulse

Intact molecular ion formation of cyclohexane and 2,3-dimethyl-1,3-butadiene by excitation with a short, intense femtosecond laser pulse

Chemical Physics Letters 427 (2006) 255–258 Intact molecular ion formation of cyclohexane and 2,3-dimethyl-1,3-butadie...

194KB Sizes 0 Downloads 1 Views

Chemical Physics Letters 427 (2006) 255–258

Intact molecular ion formation of cyclohexane and 2,3-dimethyl-1,3-butadiene by excitation with a short, intense femtosecond laser pulse Michinori Tanaka, Subhasis Panja, Masanao Murakami, Tomoyuki Yatsuhashi, Nobuaki Nakashima * Department of Chemistry, Graduate School of Science, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi, Osaka 558-8585, Japan Received 15 May 2006; in final form 20 June 2006 Available online 27 June 2006

Abstract Cyclohexane and 2,3-dimethyl-1,3-butadiene (DMBD) were ionized and fragmented by 0.8 lm femtosecond pulses with a typical intensity of 4 · 1013 W cm2, which were resonant with the cation absorption of cyclohexane and DMBD. Their intact molecular ions were dominant at a pulse duration of 15 fs, whereas the molecules were heavily fragmented by excitation at 205–210 fs. Possible reasons for the formation of intact molecular ions by a short pulse are discussed in terms of the vibrational periods of excited precursor states.  2006 Elsevier B.V. All rights reserved.

1. Introduction An intact molecular ion forms from an organic molecule in response to excitation with femtosecond infrared laser pulses at a range of 1014 W cm2. However, the same molecule decomposes heavily under different excitation conditions, and some other molecules show practically no molecular ions but fragment ions [1]. Two important fields have emerged in relation to the formation/fragmentation of molecular ions. One is the control of chemical reactions: Selective bond dissociation has been demonstrated by optimally tailored femtosecond pulses [2,3]. The other is femtosecond laser mass spectrometry (FMLS): formation of intact molecular ions is crucial for trace analyses [4]. Many organic molecules, such as cyclohexane and 2,3dimethyl-1,3-butadiene (DMBD), show heavy fragmentation by 0.8 lm fs pulse irradiation. The mechanism for this reaction has been explained in terms of resonance with the


Corresponding author. Fax: +81 6 6605 2552. E-mail address: [email protected] (N. Nakashima).

0009-2614/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2006.06.065

cation electronic levels at 0.8 lm [1,5–9]. In fact, we have observed DMBD intact molecular ion by irradiation with pulses at 1.4 lm, where the cation electronic levels of DMBD are not resonant [1]. However, it is hard to achieve a similar condition for cyclohexane, because the non-resonant wavelengths are expected to be >2.2 lm, being outside of the region conventionally available. We have thus employed another excitation parameter; an ultra short fs pulse to produce intact molecular ion of cyclohexane. It is generally accepted that a fs pulse is more effective for this purpose than ns or ps pulses [5,10–16]: p-Nitroaniline fragmentation and C60 have been studied with 5 ps – 80 fs pulses [3] and with 5 ps–25 fs pulses [13,14], respectively. For C60, 25 fs pulse has produced Czþ 60 (z = 1, 2) without heavy fragmentation at 0.79 lm [13,14]. Hertel et al. recently detected a single peak of the singly charged molecular ion [15], where they irradiated C60 with a 9 fs pulse at 0.8 lm, but no explicit account has been made on the fragmentation mechanism in relation to the resonance with the cation electronic levels. Similar trends that a shorter excitation pulse results in less fragmentation have also been reported for metal carbonyls, typically Fe(CO)5 [5] and DMBD [16].


M. Tanaka et al. / Chemical Physics Letters 427 (2006) 255–258

We note here that there are parameters other than the pulse duration and excitation wavelength that affect formation of intact or fragmented ions: the focusing geometry of the laser pulse [17], the slit width in front of an ion detector of a time-of-flight (TOF) spectrometer [18]. The present paper reports on the effects of laser pulse duration on the formation of intact molecular ions under fixed focusing geometry and slit width. By excitation at 0.8 lm, which is resonant with the cation electronic levels for both cyclohexane and DMBD, we have observed dominant formation of their intact molecular ions in 15 fs durations, which is comparable with the vibrational periods of C–C bonds in molecules. 2. Experimental A TOF mass spectrometer (TOF-MS, KNTOF-1800, TOYAMA) and a Ti:sapphire laser system (Alpha 100/ XS, Thales Laser) [19] were used. The energy was reduced to 1 mJ/pulse, and the pulse duration was stretched to 205– 210 fs with an acousto-optic programmable dispersive filter (Dazzler, FASTLITE) with a positive chirp direction. A pulse of 15 fs was generated by a conventional method [20], as shown in Fig. 1. The observed pulse, sexp, was 1.3 times the transform-limited (TL) pulse of 12 fs, sTL, expected from the spectral width. The fundamental pulse was focused into a flowing Ar gas tube with a concave mirror of 2 m focal length and converted to a broad spectral pulse. A short pulse was reconstructed by a pair of chirp mirrors, and the pulse width was measured by a home-made autocorrelator after passing through an equivalent optics into the ionizing point in the TOF-MS tube. The polarization of the pulses was parallel to the TOFMS flight axis. A typical pulse energy was 0.3 mJ. The laser intensity was varied using ND filters (SIGMA KOKI). The beam was focused into the TOF chamber through a quartz window with a lens (f = 200 mm). The laser intensity at the focal point was estimated on the basis of the saturation intensity, Isat, of Xe [21]; Isat corresponds to the laser intensity where the probability of ionization is approximately constant. The Isat values of the Xe ion pulse were estimated to be 1.3 · 1014 and 0.93 · 1014 W cm2 for pulses of 15 and 200 fs, respectively [21,22].

Fig. 1. (a) A broadened spectrum of Ti:sapphire laser after an Ar tube, and (b) temporal profile measured by a single shot autocorrelator after compression by a pair of chirp mirrors.

The ions were extracted through a 0.5 mm slit at the extraction region. The TOF mass data were recorded on an oscilloscope (Wave Runner 6100, LeCroy). Cyclohexane (Aldrich, 99.5%) and DMBD (Aldrich, 98%) were used as received, and the pressures in the chamber were maintained at 5.0 · 105 Pa. 3. Results 3.1. Higher molecular ion intensity from shorter pulse excitation The mass spectra of cyclohexane with 205-, 32-, and 15 fs pulse excitation at 0.8 lm with an intensity of 3.8– 4.2 · 1013 W cm2 are shown in Fig. 2a. Singly charged molecular (M+) and many fragment ions were observed. The intact molecular ion intensity at 205 fs pulse excitation was very weak and barely detectable. The results reproduced the previous experiments by three groups at similar excitation intensity levels of 2 · 1013 23 · 1013 W cm2 and pulse durations of 44–200 fs [21,23–25]. The shortest pulse excitation of 15 fs showed the highest intensity of the intact molecular ion relative to other fragment ions, and C3 Hþ m showed the highest intensity among many fragments. The Coulomb explosion threshold has been reported to be 19.4 · 1013 W cm2 for 44 fs pulses [21]. The present irradiation intensities were far below this threshold. In fact, no signal broadening or splitting assignable to the explosion was observed. Shorter pulse excitation led to higher intensities of M+ and heavier fragment ions. The saturation intensities were measured to be 9.6 · 1013 W cm2 for the 15 fs pulse, 8.7 · 1013 W cm2 for the 32 fs pulse, and 6.3 · 1013 W cm2 for the 205 fs

Fig. 2. TOF mass spectra of (a) cyclohexane and (b) 2,3-dimethyl-1,3butadiene. Each excitation pulse width is indicated. The intensity was 3.8– 4.2 · 1013 for (a) and 2.3 · 1013 W cm2 (b). Molecular ion (M+) intensities are higher at shorter pulse duration.

M. Tanaka et al. / Chemical Physics Letters 427 (2006) 255–258

pulse. The saturation intensities were in line with the previous value of 8.5 · 1013 W cm2 for a 44 fs pulse [21]. Femtosecond ionization of DMBD was studied in our laboratory with pulses of 1000, 300, and 35 fs durations [16]. The present results at 40 fs, shown Fig. 2, were essentially equal to that at 35 fs [16]. The signal intensity of the molecular ion at the shortest pulse width of 15 fs was the highest among the three pulse durations. The saturation intensities were 5.9 · 1013 W cm2 for 15 fs pulse, 4.1 · 1013 W cm2 for 40 fs, and 2.7 · 1013 W m2 for 210 fs. 3.2. Signal intensities of molecular ions The molecular ion intensities in each spectrum of Fig. 2 were evaluated as a ratio to the total ion intensity (M+ ion + fragment ions). The ratios of the M+ intensity relative to the total ion intensity, measured at different pulse durations and intensities, are plotted in Figs. 3a,b. A higher intensity led to a lower ratio, i.e. enhancement of fragmentation even below the threshold of Coulomb explosion (19 · 1013 W cm2 for cyclohexane [21)]. The present results of the pulse-duration dependence from 210 to 15 fs suggest that excitation with a pulse of <10 fs could produce fragment-free molecular ions from hydrocarbons like cyclohexane and DMBD in the intensity region of Isat. 4. Discussion There are three conceivable pathways of fragmentation with fs laser pulse irradiation, under the assumption that a cation is an intermediate [7,26]: (i) The potential curve of the cation is distorted considerably by a strong laser field, leading to dissociation. The structure of the cation is changed by the field and, as a result, excitation laser wavelengths can resonate with a dissociative potential. (ii) The cation absorbs laser energy, and an excited state of the ion is generated by internal conversion to vibrational energy, followed by intramolecular vibrational redistribution. Finally the cation dissociates with a statistical rate. (iii) The cation is directly excited to a repulsive state to dissociate. þ Examples for pathway (i) are the following: Hþ 2 and D2 14 2 dissociation at 2 · 10 W cm is suggested to occur on


the time scale of the vibrational motion, 10 fs, under strong non-resonant fields [27]. The dissociative ionization of ethanol has been studied with a chirped laser pulse [28] in the region of 1015 W cm2, where Coulomb explosion occurs. Many fragment ions with high intensities were observed at chirped pulses compared to the TL pulse of 32 fs. They have explained the observations in terms of the nuclear dynamics on the ethanol monocation surface. The C–C dissociation was dominant with the TL pulse, while the C–O dissociation was enhanced several folds with elongated pulses, typically 720 fs. The nuclear dynamics of Hþ 2 and ethanol monocation have been well supported by quantum mechanical wave packet approaches [29]. Pathways (ii) and (iii) have been used to explain many organic compounds including cyclohexane and DMBD at 35–1000 fs pulse excitation. If molecules are irradiated with a pulse at a resonant wavelength, heavy fragmentation is observed at 1014 W cm2, as demonstrated for cyclohexane [21,23–25], DMBD and 1,4-cyclohexadiene [7], benzene [6], anthracene [9], and metal complexes [5]. Strong laser fields induce a large broadening of electronic levels, but the resonance effects between the excitation wavelength and cation electronic levels still seem to be effective. As shown in Fig. 3, fragmentation of cyclohexane and DMBD are both suppressed by 15 fs pulse excitation. Furthermore, our observations suggest that the tendency toward intact molecular ion formation is more pronounced by excitation with <15 fs. For a typical C–C stretching vibration of neutral cyclohexane, 1000 cm1; the dissociation time is estimated to be 10 fs. The present results thus indicate that fragmentation can be suppressed when the excitation pulse duration is shorter than a half of the vibrational period, where the nuclear motion stays essentially fixed. This explanation is based on the extension of pathway (i). One could also explain the present results in line with pathways (ii) and (iii). Ionization occurs near the laser peak intensity; therefore, more energy in the trailing part of longer pulse is likely to be exploited for cation excitation at the same peak intensity. The cation absorption spectra of the two molecules have quite different features; that of cyclohexane cation is very broad from UV to >1.3 lm [30], whereas that of DMBD has an absorption edge at the present laser wavelength, 0.8 lm [31]. Heavy fragmentation in cyclohexane can be interpreted as that energy deposition by resonance in this molecule is more effective than that in DMBD. Acknowledgement This work was financially supported in part by a Grantin-Aid (No 14077214) from the Ministry of Education, Culture, Sports, Science and Technology Japan to N.N. References

Fig. 3. Ratios of molecular ion yields to total ion yields of (a) cyclohexane and (b) 2,3-dimethyl-1,3-butadiene. The ratios of molecular ions with a 15 fs laser pulse are the highest.

[1] N. Nakashima, T. Yatsuhashi, M. Murakami, R. Mizoguchi, Y. Shimada, in: S.H. Lin, A.A. Villaeys, Y. Fujimura (Eds.), Advances


[2] [3]

[4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

M. Tanaka et al. / Chemical Physics Letters 427 (2006) 255–258 in Multi-photon Processes and Spectroscopy, vol. 17, World Scientific, Singapore (Chapter 3). R.J. Levis, G.M. Menkir, H. Rabitz, Science. 292 (2001) 709. N.P. Moore, G.M. Menkir, A.N. Markevitch, P. Graham, R.J. Levis, in: A.D. Bandrauk, Y. Fujimura, R.J. Gordon (Eds.), ACS Symposium, Laser Control and Manipulation of Molecules, vol. 821, ACS, Washington, DC, 2002 (Chapter 14). D.J. Smith et al., Rapid Commun. Mass Spectrom. 12 (1998) 813. S.A. Trushin, W. Fuß, W.E. Schmid, J. Phys. B 37 (2004) 3987. R. Itakura, J. Watanabe, A. Hishikawa, K. Yamanouchi, J. Chem. Phys. 114 (2001) 5598. H. Harada, S. Shimizu, T. Yatsuhashi, S. Sakabe, Y. Izawa, N. Nakashima, Chem. Phys. Lett. 342 (2001) 563. H. Harada et al., J. Phys. Chem. A 107 (2003) 6580. M. Murakami, R. Mizoguchi, Y. Shimada, T. Yatsuhashi, N. Nakashima, Chem. Phys. Lett. 403 (2005) 238. R. Weinkauf, P. Aicher, G. Wesley, J. Grotemeyer, E.W. Schlag, J. Phys. Chem. 98 (1994) 8381. K.W.D. Ledingham et al., Rapid Commun. Mass Spectrom. 9 (1995) 1522. J. Matsumoto, C.-H . Lin, T. Imasaka, Anal. Chem. 69 (1997) 4524. E.E.B. Campbell, K. Hansen, K. Hoffmann, G. Korn, M. Tchaplyguine, M. Wittmann, I.V. Hertel, Phys. Rev. Lett. 84 (2000) 2128. I.V. Hertel, T. Laarmann, C.P. Schultz: Ultrafast Excitation, Ionization and Fragmentation of C60, Elsevier, Amsterdam, in press. I.V. Hertel, Private communication, 2005.

[16] R. Mizoguchi, Y. Shimada, H. Shinohara, T. Yatsuhashi, N. Nakashima, Rev. Laser Eng. 32 (2004) 717 (in Japanese). [17] X.P. Tang et al., Phys. Rev. A 71 (2005) 045401. [18] L. Robson et al., J. Am. Soc. Mass Spectrom. 16 (2005) 82. [19] T. Yatsuhashi, N. Nakashima, J. Phys. Chem. A 109 (2005) 9414. [20] S.A. Trushin, S. Panja, K. Kosma, W.E. Schmid, W. Fuß, Appl. Phys. B 80 (2005) 399. [21] S.M. Hankin, D.M. Villeneuve, P.B. Corkum, D.M. Rayner, Phys. Rev. A 64 (2001) 013405. [22] T. Yatsuhashi, T. Obayashi, M. Tanaka, M. Murakami, N. Nakashima, J. Phys. Chem. A 110 (2006) 7763. [23] M.J. DeWitt, R.J. Levis, J. Chem. Phys. 108 (1998) 7045. [24] M. Castillejo, S. Couris, E. Koudoumas, M. Martı´n, Chem. Phys. Lett. 289 (1998) 303. [25] M. Castillejo, S. Couris, E. Koudoumas, M. Martı´n, Chem. Phys. Lett. 308 (1999) 373. [26] A.N. Markevitch, D.A. Romanov, S.M. Smith, H.B. Schlegel, M.Yu. Ivanov, R.J. Levis, Phys. Rev. A 69 (2004) 013401. [27] H. Niikura, P.B. Corkum, D.M. Villeneuve, Phys. Rev. Lett. 90 (2003) 203601. [28] R. Itakura, K. Yamanouchi, T. Tanabe, T. Okamoto, F. Kannari, J. Chem. Phys. 119 (2003) 4179. [29] H. Kono, Y. Sato, M. Kanno, K. Nakai, T. Kato, Bull. Chem. Soc. Jpn. 79 (2006) 196. [30] T. Shida, Y. Takemura, Radiat. Phys. Chem. 21 (1983) 157. [31] T. Shida, Electronic Absorption Spectra of Radical Ions, Elsevier, New York, 1988.