Photodissociation Dynamics of Allyl Chloride at 200 and 266 nm

May 28, 2014 - The photodissociation dynamics of allyl chloride at 200 and 266 nm has been studied by femtosecond time-resolved mass spectrometry ...
1 downloads 0 Views 1MB Size
Article pubs.acs.org/JPCA

Photodissociation Dynamics of Allyl Chloride at 200 and 266 nm Studied by Time-Resolved Mass Spectrometry and Photoelectron Imaging Huan Shen,*,†,‡ Jianjun Chen,†,‡ Linqiang Hua,§ and Bing Zhang*,§ †

College of Sciences, Huazhong Agricultural University, Wuhan 430070, P. R. China Institute of Applied Physics, Huazhong Agricultural University, Wuhan 430070, P. R. China § State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, P. R. China ‡

S Supporting Information *

ABSTRACT: The photodissociation dynamics of allyl chloride at 200 and 266 nm has been studied by femtosecond time-resolved mass spectrometry coupled with photoelectron imaging. The molecule was prepared to different excited states by selectively pumping with 400 or 266 nm pulse. The dissociated products were then probed by multiphoton ionization with 800 nm pulse. After absorbing two photons at 400 nm, several dissociation channels were directly observed from the mass spectrum. The two important channels, C−Cl fission and HCl elimination, were found to decay with multiexponential functions. For C−Cl fission, two time constants, 48 ± 1 fs and 85 ± 40 ps, were observed. The first one was due to the fast predissociation process on the repulsive nσ*/πσ* state. The second one could be ascribed to dissociation on the vibrationally excited ground state which is generated after internal conversion from the initially prepared ππ* state. HCl elimination, which is a typical example of a molecular elimination reaction, was found to proceed with two time constants, 600 ± 135 fs and 14 ± 2 ps. We assigned the first one to dissociation on the excited state and the second one to the internal conversion from the ππ* state to the ground state and then dissociation on the ground state. As we excited the molecule with 266 nm light, the transient signals decayed exponentially with a time constant of ∼48 fs, which is coincident with the time scale of C−halogen direct dissociation. Photoelectron images, which provided translational and angular distributions of the generated electron, were also recorded. Detailed analysis of the kinetic energy distribution strongly suggested that C3H4+ and C3H5+ were generated from ionization of the neutral radical. The present study reveals the dissociation dynamics of allyl chloride in a time-resolved way.



INTRODUCTION The photodissociation dynamics of allyl chloride has been studied intensively over the past few decades using various techniques.1−15 One of the foci of the studies was the dissociation pathways and the quantum yield of each channel. After UV excitation at around 200 nm, either C−Cl bond fission or HCl elimination can occur. The competition among different dissociation pathways and the complexity of the potential surface make its dissociation dynamics very complicated. Myers et al. measured the photofragment velocities and angular distributions of the photoproducts at 193 nm with a crossed laser−molecular beam apparatus and identified three competing channels occurring upon π → π* excitation.10 One channel for C−Cl bond fission and two pathways for HCl elimination were found. The branching ratio was found to be HCl/C−Cl = 0.12 ± 0.03 at 200 °C. Later, Morton et al. also investigated the photodissociation dynamics of allyl chloride at 193 nm using the crossed laser−molecular beam scattering apparatus.12 Different from the previous work, a tunable vacuum ultraviolet (VUV) laser was utilized to ionize © 2014 American Chemical Society

the neutral photofragments. Four significant primary reaction channels were found. Among them, two were C−Cl bond fission channels. One produced fast Cl atoms, and the other produced slow Cl atoms. Other than that, two different HCl elimination pathways were also observed. The measured branching of these primary reaction channels of [all C−Cl]/ [fast C−Cl]/[slow C−Cl]/[fast HCl]/[slow HCl]/[all HCl] was 1.00:0.971:0.029:0.291:0.167:0.458 (where fast referred to the high recoil kinetic energy channels). By taking the advantage of the velocity map imaging (VMI) technique, both Jung’s group14 and Butler’s group15 studied the photodissociation dynamics of allyl chloride at 235 nm. The kinetic energy distribution of Cl atoms was obtained from twodimensional photofragment velocity ion imaging. Both groups found that at least two channels were involved in the Received: January 15, 2014 Revised: May 27, 2014 Published: May 28, 2014 4444

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

fundamental in the first arm. The typical power used was 1 μJ for 266 nm, 3 μJ for 400 nm, and 30 μJ for 800 nm. Thus, the intensity was calculated to be 4.5 × 1012, 1.9 × 1012, and 1.4 × 1012 W/cm2 at the focusing region for each laser, respectively. The photoions and photoelectrons produced were analyzed by an ion time-of-flight mass spectrometer and by a VMI device, respectively. The corresponding signals were monitored as a function of the delay between the pump and probe pulses. The VMI device was also used for mapping the kinetic energy and the angular distributions of the photoelectrons resulting from the pump−probe process. Each image was accumulated over 40 000 laser shots, and the backgrounds were removed by subtracting each image collected under the same conditions. The whole setup had an energy resolution of 0.1 eV at 1 eV, which was calibrated by the photoelectron spectrum of O2 under the same conditions.28In order to minimize the effect of the earth’s magnetic field on the photoelectron trajectory, a double-layer μ-metal shield was installed along the axis of the ionization chamber.

dissociation. However, discrepancies were also found, although similar methodologies and the same wavelengths were utilized. Although many studies have been carried out as mentioned above, the dissociation dynamics have not been studied in a time-resolved way. Since dissociation rates are one of the key parameters that characterize a reaction, time-resolved studies will provide important information about the photodissociation dynamics. With the advances in femtosecond chemistry over the past few decades,16 researchers now can study the dissociation dynamics with femtosecond time resolution after the pioneering work done by Zewail and co-workers.17 With this advantage, Baklanov et al. studied the excited state dynamics of allyl iodide with 159 nm excitation.18 Only a lifetime of 290 ± 30 fs was observed, and many possible mechanisms were proposed. Farmanara et al. also studied the photodissocation dynamics of ethylene and vinyl chloride with 200 nm excitation.7 The lifetime for the ultrafast internal conversion process was determined. In the present work, the photodissociation dynamics of allyl chloride was studied by femtosecond time-resolved mass spectrometry coupled with photoelectron imaging technique. The molecule was selectively prepared in different excited states by two 400 nm or one 266 nm photon and then probed subsequently by multiphoton ionization with an 800 nm laser pulse. Different dissociation channels were monitored in a timeresolved way, and the generated photoelectron images were also acquired. Our results provide important information for the dissociation rate of each channel. Especially for the HCl elimination, which is a typical example molecule elimination reaction, the time scale was directly measured for different reaction pathways.



RESULTS AND DISCUSSION 1. Time-Resolved Ion Spectrum. Figure 1 displays the mass spectrum obtained with a 400 nm pump pulse and an 800



EXPERIMENTAL METHOD The experimental setup used here has been described in detail elsewhere.19 In brief, a mixture of 5% C3H5Cl seeded in He was expanded through a pulsed nozzle operating at 10 Hz. After passing through a skimmer, which separates the source chamber from the ionization chamber, the molecular beam interacted with the collinearly propagating pump and probe laser beams at a position midway between the repeller and extractor plates of the electrostatic lens. The electrostatic ion lenses installed in the ionization chamber consisted of a set of repeller, extractor, and ground electrodes, and the biased voltages were optimized with the aid of simulations using a standard ion optics trajectory program (Simion 6.0).27 The laser source employed here was a commercial regenerative amplified Ti:sapphire femtosecond laser system. A seed beam was generated by a Ti:sapphire oscillator pumped by a CW second harmonic of an Nd:YVO4 laser and then amplified by an Nd:YLF pumped regenerative amplifier to yield a 50 fs, 1 mJ pulse centered at 800 nm with a 1 kHz repetition rate. The fundamental laser beam was then divided into two arms by a beam splitter. One arm was frequency-doubled to 400 nm in a β barium borate crystal (BBO) and used as the pump beam. The other served as the probe beam though multiphoton ionization. Both the pump and probe beams were vertically polarized (parallel to the plane of the two-dimensional detector) and focused using a 25 cm focal-length lens. The temporal delay between them was precisely controlled using a motorized linear translation stage (PI, M-126.CG1) and driver (PI, C-862.00) under computer control. When pumped at 266 nm, the pump beam was generated in a BBO crystal by sum frequency mixing the second harmonic and the

Figure 1. Mass spectrum obtained when allyl chloride was pumped at 400 nm and probed at 800 nm. The delay between the pump and probe pulses was set to zero.

nm probe pulse. The time delay between them was set to zero. Five ion peaks were observed, and they can be assigned to CH2+, C2H3+, C3H4+, C3H5+, and C3H5Cl+ using their respective time-of-flight. The power of the pump and probe pulses was set to 3 and 30 μJ, respectively, where the ionization of the molecule with pump or probe only was as low as possible, but the two-color signal was strong enough to make a time-resolved measurement. With the mass spectra obtained at different pump−probe delays, the time-resolved mass yield can be obtained after integrating the mass signal over its respective range. The results are shown in Figure 2. For C3H5Cl+, C3H5+, and C3H4+, the decay traces should be fitted by multiexponential functions, while for C2H3+ and CH2+, a single exponential decay was observed. With careful data fitting, the time-dependent traces of C2H3+ and CH2+ could be well fitted with one time constant, 70 ± 2 and 68 ± 3 fs, respectively. For C3H5+, three time constants, 68 ± 3 fs, 48 ± 1 fs, and 85 ± 40 ps are needed, while for C3H4+, 71 ± 4 fs, 600 ± 135 fs, and 14 ± 2 ps were needed. The lifetime constants longer than 10 ps have larger uncertainties due to our limited observing window, which is less than 15 ps. 4445

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

Figure 2. Ion yield for C3H5Cl+ and its fragments as a function of the pump−probe delay. The intensity of the ion signal was normalized in each panel.

It was quite surprising for us to find that the time-dependent trace for the C3H5Cl+ parent ion showed an “inverse” feature with the pump−probe delay. This indicated that the 800 nm pulse acted as the pump while the 400 nm pulse acted as the probe. Actually, in pump−probe experiment, either pulse can be the pump or the probe. The role of the pulse is determined by the absorption cross section of molecule at a particular wavelength and the power of the laser. It seemed the 800 nm pulse was resonant with certain energy levels of C3H5Cl after multiphoton absorption and thus populated the molecule to an excited state before the 400 nm pulse (see the following paragraphs). Data fitting found that two time constants, 70 ± 1 fs and 105 ± 51 ps, were needed to reproduce the observed time-dependent trace. The steady state absorption of allyl chloride was measured by Worrell in the 1970s, and they found that the first absorption peak centered at around 170 nm.20 This band was assigned to the π → π*CC transition with the assistance of calculations. In the short wavelength region between 140 and 160 nm, there were three bumps also assigned to the π → π* transition. In the long wavelength region, the absorption trace showed a long tail that extended beyond 200 nm. Photoemission measurements at 199 nm found the electronic character to be an admixture of π*CC and σ*C−Cl character.5 Thus, the long tail was mostly due to n → σ*C−Cl transitions. Moreover, Myers et al. pointed out that the broken symmetry in the gauche conformer permitted extensive coupling between with the ππ* and nσ* electronic configurations due to an avoided crossing in the Franck−Condon (FC) region.10 Hence, besides those two transitions, there was also a πσ* character in the FC region. This information suggests that exciting allyl chloride with a pulse at around 200 nm will trigger many possible dissociation pathways. The dissociation dynamics is revealed by the observed time constants in the present time-resolved study. C2H3+ or CH2+ is a good start, since it has only one time constant, which is ∼70 fs. This indicates that the ions are produced only when the pump and probe are temporally overlapped, since our pulses have a width of ∼50 fs. It also worth noticing that the transient ion signal of each species includes a similar fast decay, which is ∼70 fs. This time constant can be assigned to the instrument

response function of our apparatus. Hence, C2H3+ and CH2+ are believed to come from dissociative ionization of the parent molecule or ionization of its natural radical when two pulses are overlapped temporally. The C3H5Cl+ parent ion was generated when the 800 nm pulse acted as the pump and the 400 nm pulse acted as the probe. This indicates that after absorbing several 800 nm photons, the molecule reaches an energy level where the absorption cross-section is much larger than that at 200 nm. The steady state absorption of allyl chloride shows its first absorption peak at around 170 nm. This band is at least 1 order of magnitude stronger than its absorption at 200 nm.20 Hence, it is most likely that the molecule absorbs 5 photons at 800 nm and is resonant with a ππ* state. The state on resonance should have a very long lifetime, since the lifetime constant was 105 ± 51 ps in the fitting. The C3H5+ can be generated by dissociative ionization of the parent ion or ionization of the C3H5 radical. Since C3H5+ was observed when pump and probe were not temporally overlapped and the intensity for the pump pulse was kept pretty low to avoid multiphoton ionization of the C3H5Cl molecule, we believe C3H5+ was produced by ionization of the C3H5 radical. Under the current conditions, the power was designed to absorb two photons at 400 nm, and the dissociated products were then ionized by an 800 nm pulse. Especially for the pump pulse, it cannot absorb one photon, since no absorption was found there. It is also hard to absorb more than two photons because we used 3 μJ and a long focal-length 25 cm lens. This was true, since we hardly saw any C3H5+ with the 400 nm pulse alone. Since the potential surfaces of the nσ* and πσ* electronic states are supposed to be repulsive23 and the dissociation process is fast, we assigned the 48 ± 1 fs component to the direct dissociation on the nσ* and/or πσ* potential surface. It is worth noticing that the reliability of this lifetime constant is limited by our pulse duration (∼50 fs). We believe there is a component there and its lifetime is beyond our detection limit. The 85 ± 40 ps component is related to a slow process, and it is most probably due to ground state dissociation. This type of slow-dissociation channel was also found in the photodissociation of vinyl chloride22 and 2chloropropene,21 where a long time constant was assigned to 4446

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

internal conversion from the ππ* electronic state to the ground state and subsequently dissociated on the ground state. In the case of vinyl chloride, Farmanara et al. observed a slow channel with a lifetime of 110 ps.7 On the basis of the similarity of electronic states between vinyl chloride and allyl chloride and the similarity of the observed time scale, we assign the observed 85 ± 40 ps component to the internal conversion from the ππ* electronic state to the vibrationally excited ground state and dissociation on the ground state. The information we learned from C3H4+ is quite important in the present study. Since C3H4 is produced along with HCl elimination and C3H5 is produced from C−Cl fission, detecting these two radicals provides an opportunity to study the competition between these two photodissociation pathways. In addition, it provides the time constant of HCl elimination for the first time. As found in previous studies by Myers et al., three different HCl elimination channels are open after excitation with 193 nm light.10 H 2CCHCH 2Cl + 193 nm → H 2CCCH 2 + HCl Eavail = 124 kcal/mol

Figure 3. Qualitative surface diagram for the photofragmentation process upon the excitation to the ππ* state.

(1)

H 2CCHCH 2Cl + 193 nm → H 2CCHC:H + HCl

well fitted by a single-exponential decay function with a time constant of ∼50 fs. Since all the fragments essentially have the same decay time with the parent ion, we believe all the fragment ions result predominantly from dissociative ionization of C3H5Cl+. Previously, the Suzuki group23 studied C−Cl bond breaking in ultraviolet photodissociation of vinyl chloride at 193 and 210 nm using photofragment ion imaging technique. With increasing wavelength, the fraction of the high kinetic energy component of the Cl fragment increases. In other words, the repulsive state plays a more important role when red-shifting the wavelength. Lee et al. have also investigated the photodissociation of propargyl bromide at 193 and 248 nm using an angle-resolved beam apparatus.6 They found that propargyl bromide undergoes exclusively a simple C−Br bond fission when excited at 248 nm. Additionally, the time scale of C−halogen direct bond fission on the repulsive potential surface has been measured to be 40 fs by Zewail et al. using femtosecond time-resolved mass spectrometry.24 The transient signals can be fitted by a fast decay curve with a lifetime of 48 fs here, which is consistent with the time scale of C−halogen direct dissociation. Therefore, the excited state that was reached by 266 nm excitation might exhibit a totally repulsive character in allyl chloride, most probably nσ* excited state. Although, as shown in ref 20, the absorption at 266 nm should be weak, the present study suggests it has a certain intensity. On the other hand, the parent ion will have long-lived component if the molecule is pumped by five 800 nm photons and C3H5+ and C3H4+ will have long-lived component if it is pumped by two 400 nm photons. Given the power of 1 μJ for 266 nm laser, one photon absorption is the most possible scenario. 2. Photoelectron Kinetic Energy Distribution. With the photoelectron images obtained, we can extract the kinetic energy distribution of the photoelectron and thus understand the dynamics in a coincident way. To achieve a good signal-tonoise ratio in photoelectron imaging, it is necessary to avoid contributions from one-color signals. In the present work, we took three photoelectron images at each time delay, one with both lasers on and two with each laser on. Each raw electron image provided by the VMI device is a two-dimensional

Eavail = 95 kcal/mol (2)

H 2CCHCH 2Cl + 193 nm → HC·CHC·H 2 + HCl Eavail = 73 kcal/mol (3)

The fast HCl elimination with a high translational energy distribution was assigned to channel (1), and the one with slow translational energy was assigned to channel (2) or channel (3).10 One possible explanation for the slow HCl elimination channel is that the coupling between the ππ* state and πσ* state leads to efficient internal conversion from the πσ* state to the ground state. Thus, the long-lived component with lifetime of 14 ± 2 ps can be regarded as the lifetime for the slow HCl elimination process on the ground state in channel (2) and/or channel (3). On the other hand, reaction (1) is due to the excited state mechanism as suggested in ref 10. Ab initio calculations for vinyl chloride also revealed that the πσ* state was also a repulsive state.23 Thus, a very short time constant is expected. In the present measurement, a lifetime of 600 fs was found. However, in vinyl chloride, a much shorter lifetime, 40 ± 10 fs, was observed and it was attributed to the nσ* state or πσ* state dissociation.7 One thing we should keep mind is that allyl chloride has longer carbon chain than vinyl chloride. Especially in channel (1), the carbon chain in H2CCCH2 radical is linear while in H2CCH−CH2Cl it is bent. This indicates a large geometry change is expected in allyl chloride before HCl elimination. That might be the reason why a longer lifetime constant was observed. However, more theoretical work is needed to more deeply understand the complicated excited-state surface. The overall relaxation dynamics are summarized in Figure 3. The photodissociation dynamics of allyl chloride at 266 nm has also been studied. The transient intensity change of the parent ion and its fragments are displayed in Figure 4. The observed dynamics is totally different from those found with a 400 nm pulse as pump, where the multiexponential decays were observed. We found that all the time-dependent traces can be 4447

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

Figure 4. Ion yield for C3H5Cl+ and its fragments as a function of the delay time when pumped at 266 nm and probed at 800 nm. The intensity of the signal was normalized in each panel.

Supporting Information). Also, these peaks were located at the same positions when using different pump wavelengths. The observed peaks can be assigned using conservation of energy,

projection of the actual three-dimensional (3D) electron distribution. Therefore, the velocity and angular distributions of the photoelectrons can be obtained from these images. Given the fact that the 3D distribution has cylindrical symmetry around the polarization axis of the laser, a full 3D photoelectron image can be reconstructed by using the basis-set expansion method (BASEX).25 The raw and reconstructed 3D electron images at zero time delay are presented in Figure 1S in the Supporting Information. The kinetic energy distribution of photoelectrons P(υe−) was obtained from the angular integration of the reconstructed image as a function of the radial distance from the center, as shown in Figure 5. Five

Ee = n1hv1 + n2hv2 + E0 − Ei

(4)

where n1 and n2 stand for the numbers of photon involved in the pump−probe process. hν is the photon energy. E0 is the internal energy of parent molecule, and it can be neglected in the cold supersonic beam conditions. Ei is the internal energy of ionic parent at the ith level, and Ee is the kinetic energy of the outgoing electrons. Multiphoton ionization with a (2 + 3′)/(1′ + 5) scheme for 400 + 800 nm and a (1 + 4′) scheme for 266 + 800 nm was proposed. Since the peak positions of the kinetic energy distribution are the same in these two cases, we take the 400 nm pump and 800 nm probe as the example. After absorbing two photons at 400 nm and three photons at 800 nm simultaneously, the molecule will obtain an energy of 10.87 eV. The vertical ionization potential of C3H5Cl is 10.20 eV.20 Thus, the photoelectron kinetic energy should be distributed in the 0−0.67 eV region. Within this region, there are two peaks centered at 0.36 and 0.67 eV. Apparently, the 0.67 eV peak comes from the C3H5Cl parent ion. The other peak, centered at the 0.36 eV, cannot be assigned to the C3H5Cl parent ion, since its second, third, and fourth vertical ionization potentials are located at 11.17, 11.27, and 12.28 eV, respectively.20 Even by absorbing one or two more 800 nm photons, it cannot yield a photoelectron energy peaked at 0.36 eV. Thus, this peak mostly comes from the ionization of the CH2 radical (its ionization potential is 10.4 eV 26) which is generated after absorbing two 400 nm photons. Since it needs seven photons to reach 10.84 eV, we believe some intermediate states should be on resonance.

Figure 5. Photoelectron spectra extracted from the photoelectron images. The pump−probe delay was set to zero.

peaks, centered at 0.36, 0.67, 0.97, 1.67, and 3.32 eV, were observed, as shown in Table 1. The peak positions did not change with pump−probe delay (as shown in Figure 2S in the Table 1. Observed peaks of photoelectron energy distribution Energy (eV)

ε1

ε2

ε3

ε4

ε5

Peak Range

0.36 0.07−0.44

0.67 0.58−0.71

0.97 0.91−1.10

1.67 1.37−1.90

3.23 3.20−3.44

4448

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

ionization. It is expected that the initially populated excited state exhibits a totally repulsive character. The transient signals of the parent ion and its fragments can be fitted to a single exponential decay with a time constant of ∼48 fs, which is coincident with the time scale of C−halogen direct dissociation on the nσ* state potential surface. The photoelectron energy distribution verified that the fragment ion came from ionization of neutral radicals.

The third and the forth peaks are centered at 0.97 and 1.67 eV. They are less intense than the first and the second peaks, especially with a 400 nm pump and a 800 nm probe. We assign the 0.97 eV peak to the ionization of the C2H3 radical (its ionization potential is 8.25 eV 26) or the C3H5 radical (the ionization potential is 8.18 eV 26) after absorbing six 800 nm (9.3 eV) photons. The 1.61 eV peak is hard to assign based on the information we have. One possibility is the ionization of the H2CCHC:H or the HC·CHC·H2 radical because they are the only species for which the ionization potential is not currently available. With this assumption, one of them should have an ionization potential of 7.4 eV. The present analysis agrees well with the previous section, where C3H5+ and C3H4+ are assigned to be ionization products of the neutral radical but not dissociation products of the parent ion. The last peak, ε5 = 3.22 eV, should come from the same species as that at ε4 = 1.67 eV provided that the peak positions of the two are ∼1.55 eV away. That implies the latter was generated by absorbing one more photon. The time-resolved photoelectron spectra extracted from the electron images are also shown in Figure 2S in the Supporting Information. The photoelectron spectra for each delay time were not normalized to emphasize their evolution. More importantly, the energy positions of the observed structures in the spectra do not change with delay time. Only their relative intensities change. As mentioned above, our instrument response function has a width of ∼70 fs when pumped with 400 nm and probed with 800 nm. Therefore, the time-resolved photoelectron spectra within 110 fs is too short to observe the population transfer. The intensity of four main peaks at a variety of pump−probe time delays is also shown in Figure 2S in the Supporting Information. With increasing delay time, all peaks decrease similarly. The decrease of the intensity of these peaks is faster when excited with 266 nm photon than with 400 nm photons. This is coincident with the shorter lifetime observed at 266 nm.



ASSOCIATED CONTENT

S Supporting Information *

(a) Photoelectron images obtained using VMI device with different pump wavelengths; (b) time-dependent electron spectra extracted from the photoelectron images at variety of pump−probe delays. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Authors

*H.S.: e-mail: [email protected]. *B.Z.: e-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors gratefully acknowledge the support from the Fundamental Research Funds for the Central Universities (No. 2662013BQ046), and the National Science Foundation of China (No. 21203242).



REFERENCES

(1) Kawasaki, M.; Kasatani, K.; Sato, H.; Shinohara, H.; Nishi, N. Photodissociation of Molecular Beams of Halogenated Hydrocarbons at 193 nm. Chem. Phys. 1984, 88, 135−142. (2) Umemoto, M.; Seki, K.; Shinohara, H.; Nagashima, U.; Nishi, N.; Kinoshita, M.; Shimada, R. Photofragmentation of Mono- and Dichloroethylenes: Translational Energy Measurements of Recoiling Cl and HCl Fragments. J. Chem. Phys. 1985, 83, 1657−1666. (3) Sato, K.; Tsunashima, S.; Takayanagi, T.; Fujisawa, G.; Yokoyama, A. Translational Energy Distributions of the Products of the 193 and 157 nm Photodissociation of Chloroethylenes. J. Chem. Phys. 1997, 106, 10123−10133. (4) Katayanagi, H.; Yonekuira, N.; Suzuki, T. C−Br Bond Rupture in 193 nm Photodissociation of Vinyl Bromide. Chem. Phys. 1997, 231, 345−353. (5) Browning, P. W.; Kitchen, D. C.; Arendt, M. F.; Butler, L. J. Investigating the C−Cl Antibonding Character in the ππ* Excited State of Vinyl, Allyl, and Propargyl Chloride: Emission Spectra and ab Initio Calculations. J. Phys. Chem. 1996, 100, 7765−7771. (6) Lee, Y. R.; Lin, S. M. Photodissociation of CHCCH2X (X=Br and Cl) by Translational Spectroscopy. J. Chem. Phys. 1998, 108, 134− 141. (7) Farmanara, P.; Stert, V.; Radloff, W. Ultrafast Internal Conversion and Fragmentation in Electronically Excited C2H4 and C2H3Cl Molecules. Chem. Phys. Lett. 1998, 288, 518−522. (8) Farmanara, P.; Steinkellner, O.; Wick, M. T.; Wittmann, M.; Korn, G.; Stert, V.; Radloff, W. Ultrafast Internal Conversion and Photodissociation of Molecules Excited by Femtosecond 155 nm Laser Pulses. J. Chem. Phys. 1999, 111, 6264−6270. (9) Fan, H.; Pratt, S. T. Photodissociation of Propargyl Bromide and Photoionization of the Propargyl Radical. J. Chem. Phys. 2006, 124, 144313_1−144313_8. (10) Myers, T. L.; Kitchen, D. C.; Hu, B.; Butler, L. J. Investigating Conformation Dependence and Non-Adiabatic Effects in the Photo-



CONCLUSION The photodissociation dynamics of allyl chloride was explored using femtosecond time-resolved mass spectrum and photoelectron imaging technique in the UV region. When 400 and 800 nm lasers were employed, two different pump−probe schemes were observed. For the parent ion, it absorbed five 800 nm photons and produced a stable excited parent neutral that was subsequently ionized by one 400 nm probe photon. Radicals were generated after absorbing two 400 nm photons to the ππ* state and then dissociating with different reaction channels. The lifetime for two different C−Cl bond fission channels were found to be 48 ± 1 fs and 85 ± 40 ps, respectively. They were attributed to the predissociation process on the repulsive nσ*/πσ* state and a slow dissociation after internal conversion to the vibrationally excited ground state. The time constants of two different HCl elimination channels were measured to be 600 ± 135 fs and 14 ± 2 ps. The slow decay with a lifetime of 14 ± 2 ps suggests that the slow HCl elimination channel comes from efficient internal conversion from the ππ*/πσ* state to the ground state and dissociation on the ground state. The relatively fast decay of 600 ± 135 fs was proposed to dissociate on the excited state surface. When the pump wavelength was shift to 266 nm, the transient signals displayed a signal exponential decay. They were generated after one 266 nm photon excitation followed by four 800 nm photons, which induce nonresonant multiphoton 4449

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450

The Journal of Physical Chemistry A

Article

dissociation of Allyl Chloride at 193 nm. J. Chem. Phys. 1996, 104, 5446−5456. (11) Mueller, J. A.; Parsons, B. F.; Butler, L. J.; Qi, F.; Sorkhabi, O.; Suits, A. G. Competing Isomeric Product Channels in the 193 nm Photodissociation of 2-Chloropropene and in the Unimolecular Dissociation of the 2-Propenyl Radical. J. Chem. Phys. 2001, 114, 4505−4521. (12) Morton, M. L.; Butler, L. J.; Stephenson, T. A.; Qi, F. C−Cl Bond Fission, HCl Elimination, and Secondary Radical Decomposition in the 193 nm Photodissociation of Allyl Chloride. J. Chem. Phys. 2002, 116, 2763−2775. (13) Fan, H.; Pratt, S. T.; Miller, J. A. Secondary Decomposition of C3H5 Radicals Formed by the Photodissociation of 2-Bromopropene. J. Chem. Phys. 2007, 127, 144301_1−144301_8. (14) Park, M. S.; Lee, K. W.; Jung, K. H. Br(2Pj) and Cl(2Pj) Atom Formation Dynamics of Allyl Bromide and Chloride at 234 nm. J. Chem. Phys. 2001, 114, 10368−10374. (15) Liu, Y.; Bulter, L. J. C−Cl Bond Fission Dynamics and Angular Momentum Recoupling in the 235 nm Photodissociation of Allyl Chloride. J. Chem. Phys. 2004, 121, 11016−11022. (16) Zewail, A. H. Femtochemistry: Atomic-Scale Dynamics of the Chemical Bond. J. Phys. Chem. A 2000, 104, 5660−5694. (17) Dantus, M.; Rosker, M. J.; Zewail, A. H. Real-Time Femtosecond Probing of “Transition States” in Chemical Reactions. J. Chem. Phys. 1987, 87, 2395−2397. (18) Baklanov, A. V.; Maltsev, V. P.; Karlsson, L.; Sassenberg, U.; Persson, A. Pump−Probe Femtosecond-Laser VUV REMPI Technique Applied to the Study of Highly Excited States of Allyl Iodide. J. Chem. Soc., Faraday Trans. 1996, 92, 1681−1682. (19) Eppink, A. T. J. B.; Parker, D. H. Velocity Map Imaging of Ions and Electrons Using Electrostatic Lenses: Application in Photoelectron and Photofragment Ion Imaging of Molecular Oxygen. Rev. Sci. Instrum. 1997, 68, 3477−3484. (20) Worrell, C. W. The Photoelectron and Absorption Spectra of Allyl Halides. J. Electron Spectrosc. 1974, 3, 359−367. (21) Buff, R. D.; Parr, A. C.; Jason, A. J. The Photoionization of Allyl Chloride from Onset to 20 eV. Int. J. Mass Spectrom. Ion Phys. 1981, 40, 31−34. (22) Blank, D. A.; Sun, W.; Suits, A. G.; Lee, Y. T.; North, S. W.; Hall, G. E. Primary and Secondary Processes in the 193 nm Photodissociation of Vinyl Chloride. J. Chem. Phys. 1998, 108, 5414−5425. (23) Tonokura, K.; Daniels, L. B.; Suzuki, T.; Yamashita, K. C−Cl Bond Rupture in Ultraviolet Photodissociation of Vinyl Chloride. J. Phys. Chem. A 1997, 101, 7754−7764. (24) Zhong, D.; Zewail, A. H. Femtosecond Real-Time Probing of Reactions. 23. Studies of Temporal, Velocity, Angular, and State Dynamics from Transition States to Final Products by FemtosecondResolved Mass Spectrometry. J. Phys. Chem. A 1998, 102, 4031−4058. (25) Dribinski, V.; Ossadtchi, A.; Mandelshtam, V. A.; Reisler, H. Reconstruction of Abel-Transformable Images: The Gaussian Basis-Set Expansion Abel Transform Method. Rev. Sci. Instrum. 2002, 73, 2634− 2642. (26) http://cccbdb.nist.gov/. (27) Dahl, D. A.; Delmore, J. E.; Appelhans, A. D. SIMION PC/PS2 Electrostatic Lens Design Program. Rev. Sci. Instrum. 1990, 61, 607− 609. (28) Eppink, A. T. J. B.; Parker, D. H. Photoelectron and Photofragment Velocity Map Imaging of State-Selected Molecular Oxygen Dissociation/Ionization Dynamics. J. Chem. Phys. 1997, 107, 2357−2362.

4450

dx.doi.org/10.1021/jp500495b | J. Phys. Chem. A 2014, 118, 4444−4450