Vibrational excitations in chloromethyl radical formed by the photodissociation of chlorobromomethane Qianguang Li, Rongshu Zhu, Jinjun Lu, Xiu Zhang, and Bifeng Tang Citation: The Journal of Chemical Physics 140, 034303 (2014); doi: 10.1063/1.4861672 View online: http://dx.doi.org/10.1063/1.4861672 View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/140/3?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Alignment dependent chemisorption of vibrationally excited CH4(ν3) on Ni(100), Ni(110), and Ni(111) J. Chem. Phys. 135, 224703 (2011); 10.1063/1.3665136 Chloroacetone photodissociation at 193 nm and the subsequent dynamics of the CH3C(O)CH2 radical—an intermediate formed in the OH + allene reaction en route to CH3 + ketene J. Chem. Phys. 134, 054301 (2011); 10.1063/1.3525465 Coupled-surface investigation of the photodissociation of NH 3 ( A ̃ ) : Effect of exciting the symmetric and antisymmetric stretching modes J. Chem. Phys. 130, 234303 (2009); 10.1063/1.3132222 Site-dependent photodissociation of vibrationally excited CD 3 NH 2 J. Chem. Phys. 130, 164312 (2009); 10.1063/1.3122983 193-nm photodissociation of acryloyl chloride to probe the unimolecular dissociation of CH 2 CHCO radicals and CH 2 CCO J. Chem. Phys. 120, 4223 (2004); 10.1063/1.1644096

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

THE JOURNAL OF CHEMICAL PHYSICS 140, 034303 (2014)

Vibrational excitations in chloromethyl radical formed by the photodissociation of chlorobromomethane Qianguang Li,1 Rongshu Zhu,2 Jinjun Lu,1 Xiu Zhang,1 and Bifeng Tang1,a) 1

Department of Physics, Hubei Engineering University, Xiaogan City, Hubei Province 432000, People’s Republic of China 2 Environmental Science and Engineering Research Center, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen 518055, People’s Republic of China

(Received 27 August 2013; accepted 26 December 2013; published online 15 January 2014) Using velocity map ion imaging, the photodissociation of chlorobromomethane (CH2 BrCl) at 233– 234 nm has been studied. The total translational energy distributions and the anisotropy parameters have been determined from the ion images of the photofragments Br (2 P1/2 ) (denoted as Br∗ ) and Br (2 P3/2 ) (denoted as Br) for the dominant CH2 BrCl + hv → CH2 Cl + Br∗ and CH2 BrCl + hv → CH2 Cl + Br channels. Using an impulsive model invoking angular momentum conservation, the vibrational energy distributions of the chloromethyl radicals have been derived from the total translational energy distributions for the two channels. The study suggests that there are a number of vibrational modes of the chloromethyl radical to be excited in both of the two photodissociation channels. In the Br* channel, the CH2 s-stretch mode v1 has the most probability of excitation. While in the Br channel, the CH2 scissors mode ν 2 is attributed to the highest peak of the vibrational energy curve of the chloromethyl radical. The results further imply that, following absorption of one UV photon of 234 nm, other vibrational modes besides v5 (C–Br stretch mode) are also excited in the parent molecule. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4861672] I. INTRODUCTION

Vibrational state distributions for the products arising from photodissociation have been the subject of intense research,1–10 for the vibrational state distributions intrinsically reflect the forces and torques acting between recoiling fragments during their evolution into isolated species. Thus, they provide important information on the photodissociation dynamics. Butler and co-workers11 have studied photodissociation of 2-bromoethanol (BrCH2 CH2 OH) at 193 nm using velocity map imaging, separately resolving the C2 H4 OH + Br∗ and C2 H4 OH + Br channels. The nascent 2hydroxyethyl (C2 H4 OH) radical is subject to possible OH + C2 H4 dissociation reaction. Thus, they also measured the speed distribution of the final 2-hydroxyethyl radical. The amount of the nascent 2-hydroxyethyl radicals that undergo secondary dissociation process to form OH + C2 H4 depends on their internal vibrational energy distribution. They introduced an impulsive model,11–14 which we designate as Butler model, to characterize the partitioning of internal energy, and to determine the vibrational energy distribution of the nascent 2-hydroxyethyl radicals as a function of the total translational energy imparted in the photodissociation. The Butler model gives a nearly perfect prediction of the measured speed distribution of the final 2-hydroxyethyl radical. Chlorobromomethane (CH2 BrCl), a dihalomethane, when absorbing a UV photon, can be excited to different dissociative excited states, giving rise to the production of a) Author to whom correspondence should be addressed. Electronic mail:

[email protected]. 0021-9606/2014/140(3)/034303/7/$30.00

a halogen atom and a halomethyl radical. Considering that halomethyl radicals CH2 X can undergo further dissociation to produce a second halogen atom and a methylene radical, the internal vibrational energy distribution of the nascent halomethyl radical is an interesting problem. Tzeng et al.15 examined the photodissociation of chlorobromomethane at 248 and 193 nm using photofragment translational spectroscopy (PTS). There are four energetically allowed dissociation channels contributing in the wavelength region. CH2 BrCl + hv → CH2 Cl + Br∗ (2 P1/2 ),

(1)

CH2 BrCl + hv → CH2 Cl + Br(2 P3/2 ),

(2)

CH2 BrCl + hv → CH2 Br + Cl∗ (2 P1/2 ),

(3)

CH2 BrCl + hv → CH2 Br + Cl(2 P3/2 ).

(4)

Their results showed that the branching ratio of C–Br bond cleavage to C–Cl bond fission at 193 nm was 4.5 and at 248 nm only C–Br bond cleavage occurred. The absorption peak of chlorobromomethane at 203 nm16 has been assigned to the n (Br)→ σ ∗ (C–Br) transition. The n (Cl)→ σ ∗ (C–Cl) transition is expected to be an absorption peak near 164 nm, and the onset of the peak is near 194 nm.17, 18 North and co-workers17, 19 have studied the photodissociation dynamics of chlorobromomethane using resonance enhanced multiphoton ionization (REMPI) with timeof-flight mass spectrometry from 193 to 268 nm, which

140, 034303-1

© 2014 AIP Publishing LLC

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-2

Li et al.

mainly corresponds to the n (Br)→ σ ∗ (C–Br) transition. In order to characterize the non-adiabatic curve crossing dynamics, the wavelength dependent anisotropy parameters and product quantum yields were determined for both Br∗ and Br. Non-adiabatic transitions appeared to play a smaller role in chlorobromomethane dissociation than in its monohalogenated analogues, specifically CH3 Br. They suggested that this difference was the result of the intrinsic Cs symmetry and lower radial velocity of CH2 Cl-Br, and it was discussed in terms of the one-dimensional Landau–Zener model. Jung and co-workers20 have investigated the photodissociation dynamics of chlorobromomethane at 234 nm utilizing ion imaging system. They found that the total translational energy of the Br channel consisted of two parts which, they suggested, was caused by the interaction between σ ∗ (C–Br) and σ ∗ (C–Cl) surfaces. They interpreted the fast part in terms of a rigid impulsive model with no vibrational excitation and the slow part by a soft impulsive model with a little vibrational excitation. The energy gap was explained to be due to CH2 scissors mode v2 of the chloromethyl radical. Ng and coworkers21 have examined the photodissociation dynamics of chlorobromomethane using a time-sliced ion velocity imaging apparatus. Translational energy and angular distributions were measured for the Br∗ and Br channels systematically in the photodissociation wavelength range of 193–267 nm. Ion images for Cl (2 P3/2, 1/2 ) atoms have also been obtained at selected photodissociation wavelengths. Their results confirmed that CH2 Cl+ Br∗ /Br channels were the dominant primary channels in the whole A-band of chlorobromomethane absorption. Observed Cl (2 P3/2, 1/2 ) atoms were proposed to result from the secondary photodissociation process, CH2 Cl + hv → CH2 + Cl. They also found that the total translational energy distributions for the Br∗ channel could be fitted well with one Gaussian function, whereas the total translational energy distributions for the Br channel exhibited bimodal structures and could be decomposed into a slow and a fast Gaussian component. Ng and co-workers21 have also obtained the ion image of chloromethyl radicals at the nonresonant wavelength of 224.664 nm. They expected that the chloromethyl radicals formed by the UV photodissociation of chlorobromomethane were highly rotationally and vibrationally excited. However, they have not determined rotational and vibrational excitations in the chloromethyl radicals. The difficulty came from the different multiphoton ionization sampling efficiencies of the radicals in different internal rovibrational states. On the other hand, the nascent chloromethyl radicals in high vibrational excited states dissociated rapidly, which also made the measurements inaccurate. With the help of the newly built Butler model, we can get the vibrational energy distribution of a nascent radical formed by photodissociaton from the total translational energy distribution of the fragments. So, we decided to reinvestigate the photodissociation of chlorobromomethane, hoping to reveal more detailed dynamical features. This article reports a photodissociation study of chlorobromomethane at 233–234 nm employing the two-dimensional photofragment velocity map ion imaging technique. The (2+1) resonance-enhanced multiphoton ionization (REMPI) scheme was used to ionize state-selectively Br∗ and Br from the dominant CH2 BrCl + hv

J. Chem. Phys. 140, 034303 (2014)

→ CH2 Cl + Br∗ and CH2 BrCl + hv → CH2 Cl + Br channels. The speed distributions and the anisotropy parameters of the photoproducts (Br∗ and Br) were extracted from the ion image. Using the Butler model,11–14 which invokes angular momentum conservation, the vibrational energy distributions of the nascent chloromethyl radicals were determined from the total translational energy distributions of the photodissociation channels. Based on those, the excited vibrational states in the nascent chloromethyl radicals were tentatively determined, and detailed dynamical features for the photodissociation of chlorobromomethane were discussed at excited vibrational states level. II. EXPERIMENTAL

The experiments were performed using an ion velocity imaging apparatus, which was described in detail previously.22, 23 It is a modified time-of-flight (TOF) mass spectrometer equipped with an electrostatic ion lens similar to that reported by Eppink and Parker.24 The electrostatic ion lens consists of three 1 mm thick, 80 mm diameter stainless steel plates, the repeller, the extractor, and ground electrodes. Chlorobromomethane (99.9%) was purchased commercially without further purification. The gas mixture (10% sample seeded in 1.0 atm helium) was expanded into the vacuum through a pulsed nozzle with an orifice diameter of 0.6 mm. The pulsed nozzle was heated to a temperature of 40 ◦ C to keep the vibrational temperature of the sample unchangeable.11 The pulsed nozzle was driven by a valve driver (General Valve, IOTA-1). After skimmed by a 1 mm skimmer mounted ∼6 cm downstream, the molecular beam passed through a 5 mm hole in the repeller into the ion lens, where the molecular beam was intercepted by laser beam. Laser pulses were generated by a pulsed dye laser (Lambda Physik Scanmate 2E OG) pumped by the third harmonic (355 nm) of an Nd: YAG laser (YG 981 E 10). The laser pulses from the dye laser was then frequency-doubled by a BBO crystal combined with a compensator, and linearly polarized using a half-wave retardation plate, so they were vertically aligned and perpendicular to the molecular beam. Finally the laser pulses were focused into the photolysis region by using a 20 cm focal length lens. The laser wavelength was scanned over a range of ±0.8 cm−1 to cover the Doppler profile of the photofragments. The same laser pulse photodissociated the chlorobromomethane by single-photon absorption into the A-band, and ionized the photofragments by (2+1) resonance-enhanced multiphoton ionization (REMPI) at 233.95 nm (Br∗ ) or 232.98 nm (Br). The ions were extracted and accelerated by the electrostatic ion lens into a 50 cm long time-of-flight tube, and projected onto a 40 mm diameter two-dimensional position sensitive detector, which consists of a dual microchannel plate (MCP) coupled to a fast phosphor screen. To mass select a photofragment, the gain of the MCP was gated by applying a timed voltage (500 V) pulse (AVRH-3-C, PULSE GENERATOR) on the back plate, on which plate there was already a constant voltage (1000 V), while the front plate was fixed at the ground potential. The images of the ions on the phosphor screen were recorded using a charge-coupled device (CCD)

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-3

Li et al.

J. Chem. Phys. 140, 034303 (2014)

camera, and the REMPI time-of-flight mass spectra were acquired using a photomultiplier tube instead of CCD camera. The images were accumulated in a computer for 30 000 laser shots. The data acquisition was controlled using a commercial ion-imaging system (COD32/Video). The laser, molecular beam, and detection system were run at 10 Hz. Timing was synchronized using a delay pulse generator (Stanford Research System, DG 535). The speed, and thus the kinetic energy, of the bromine ions was calibrated using images produced following the photodissociation of Br2 at 265 nm.25 A magnification factor N = 1.306 was derived from the Br∗ or Br image by using the expression R = Nvt, where R is the ring radius, v the fragment recoil speed, and t the ion time-of-flight. We measured the speed of phohofragment Br formed by photodissociation of Br2 at 265 nm in the Br∗ +Br channel, determined that the speed value was 1650 m/s, and the half width at halfmaximum (HWHM) of the peak was 16 m/s. Then we ascertained that the relative error of speed is 0.01, and the relative error of translational energy is 0.02 in our apparatus. III. RESULTS

The raw images corresponding to Br∗ and Br generated after the photolysis of chlorobromomethane at 233.95 nm and 232.98 nm are shown in Figures 1(a) and 1(b), respectively. The shape of an image is dependent on the speed and angular distribution patterns of the fragments. By performing an inverse Abel transformation, a threedimensional velocity distribution can be reconstructed from the raw image. The angular distribution, P(θ ), can be obtained by integrating the reconstructed three-dimensional velocity distribution over a proper range of speed at each angle. The anisotropy parameter, β, is extracted by fitting P(θ ) with the standard formula, P (θ ) = (4π )−1 [1 + βP2 (cos θ )],

(5)

where θ indicates the angle between the recoil velocity of photofragments and the polarization axis of the photolysis laser. P(θ ) and P2 (cos θ ) denote the ion signal intensity and the second order Legendre polynomial, respectively. β varies from 2 (parallel transition) to –1 (perpendicular transition).

The observed values are β(Br∗ ) = 1.45 ± 0.11 and β(Br) = 0.84 ± 0.06, which are in accord with the early results,20, 21 taking into consideration of the experimental errors. In the present experiment the photodissociation process of chlorobromomethane is governed by the five excited electronic states, [2A , 1A ], 3A and [4A , 2A ]. The [2A , 1A ] and the [4A , 2A ] states come from the doubly degenerate 3 Q1 and 1 Q1 states splitting when the symmetry lowers from C3v to Cs , respectively. In addition, the excited 3 Q0 state in C3v symmetry becomes the 3A states in Cs symmetry. Ab initio calculations17, 19 have shown that the n → σ ∗ (C–Br) transition in chlorobromomethane retained their local character. The dipole moment for transition to the 3A state was parallel to the C–Br bond, and the dipole moments for transitions to the 2A , 1A , 4A , and 2A states were perpendicular to the C–Br bond. The 3A state is adiabatically correlated to the Br channel and the 4A state adiabatically relates with the Br∗ channel. An avoid crossing occurs between the 3A and 4A states, resulting in a diabatic correlation of the 3A state to the Br∗ channel and the 4A state to Br channel. The remaining states 2A , 1A , and 2A are correlated adiabatically to the Br channel. The previous studies of North and co-workers17, 19 have found that the [2A , 1A ] states play a prominent role in photodissociation dynamics at wavelength longer than 250 nm, although they only make a very small contribution to the overall absorption of the A-band. Moreover, they also found that the 3A and [4A , 2A ] states contribute nearly equally to the A-band absorption and the 3A state also makes a significant participation in the red-wing of the A-band absorption due to its slightly broader absorption. Considering those findings, of β(Br∗ ) = 1.45 ± 0.11 and β(Br) = 0.84 ± 0.06, we predict that the Br∗ photofragments mainly come from the 3A state with minor contribution from 4A state and Br channel originates predominantly from the 3A and [4A , 2A ] states. The speed distribution can be extracted by integrating the reconstructed three-dimensional velocity distribution over all angles at each speed. The speed distributions of Br∗ and Br are presented in Figures 2(a) and 2(b), respectively. The speed distributions of the photofragments were then used to calculate the total translational energy distributions of the Br∗ and Br channels using the equations, P (ET ) = P (v)

ET =

FIG. 1. Raw ion images of Br∗ (a) and Br (b) fragments from the photolysis of chlorobromomethane at 233.95 nm and 232.98 nm, respectively. The arrow indicates the laser polarization direction.

dv , dET

1 mBr 2 (mBr + mCH2 Cl ) × v . 2 mCH2 Cl Br

(6)

(7)

The total translational energy distributions of Br∗ and Br channels are presented in Figures 3(a) and 3(b), respectively. As mentioned above, the relative error of total translational energy is 0.02, and the maximum error is determined as about 2.5 kJ/mol. We repeated the ion imaging experiments of Br∗ and Br for three times, and estimated that the relative error of the signal intensities is 0.01. We added the error bars of the signal intensities in Figure 3. From Figure 3 we can infer that the structures in the total translational energy distributions are reliable.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-4

Li et al.

J. Chem. Phys. 140, 034303 (2014)

FIG. 2. Speed distributions (blue curves) of fragments Br∗ (a) and Br (b) from the raw ion images and β parameters at each speed (red curves).

FIG. 3. Total translational energy distributions (blue curves) for Br∗ (a) and Br (b) channels. The red plus and minus signs represent the error bars of the signal intensities.

IV. ANALYSIS AND DISCUSSION

chloromethyl radicals. As Butler11–14 mentioned, at each measured total translational energy ET , one may use energy conservation to calculate the internal energy of the chloromethyl radical, Eint (CH2 Cl):

As shown in Figure 3, there is just one peak, centered at 79 ± 2 kJ/mol, in the total translational energy distribution of Br∗ channel, and there are three peaks in the total translational energy distribution of Br channel, centered at 92 ± 2 kJ/mol, 83 ± 2 kJ/mol, and 75 ± 2 kJ/mol. In the previous experiments, Jung and co-workers20 found that there were two Gaussian curves in the total translational energy distribution of the Br channel, the energy gap of the two peaks was 25 kJ mol−1 . They attributed the energy gap to the CH2 scissors mode v2 of the chloromethyl radical. However, the energy of this vibrational mode is 16.648 kJ/mol,26 this difference of energy value puts the assignment to be suspicious. Compared with the previous results, in the present experiment the fast Gaussian component was split into two peaks in the total translational energy distribution of the Br channel. The energy gaps between the largest translational peak and the two smaller translational peaks are 9 and 17 kJ/mol, respectively. The energy gaps of 9 and 17kJ/mol seem equal to the energies of the C–Cl stretch mode v3 and the CH2 scissors mode v2 26 of the chloromethyl radical, but we could not assign like that. The internal energy of the chloromethyl radical includes rotational, vibrational and electronic energies, where the electronic energy is zero (ground electronic state). So, the total translational energy gap does not have to be equal to the vibrational energy gap. We need employ the Butler model to determine the vibrational energy distribution of the nascent

Eint (CH2 Cl) = Ehν + Eint (CH2 BrCl) − D0 (C − Br) − Eint (Br(2 PJ )) − ET .

(8)

Here Ehν is the energy of photodissociation photon, 511.5690 kJ/mol for Br∗ channel, and 513.6975 kJ/mol for Br channel. D0 (C–Br) is the bond dissociation energy of the C–Br bond, experimentally determined as 266.40 ± 0.80 kJ/mol.27 Eint (Br(2 PJ )) is the internal energy in each Br spin-orbit state, 44.12 kJ/mol for Br∗ and 0 kJ/mol for Br.28 ET is the total translational energy for each photodissociation channel. Eint (CH2 BrCl) is the internal energy of the parent molecules. We assume that the supersonic expansion rotationally cools the parent chlorobromomethane molecules, and that they have a thermal distribution of vibrational energy at the nozzle temperature, 40 ◦ C. Using the harmonic vibrational frequencies calculated at the B3LYP/6-311++G (3df, 2p)29 level and scaled by 0.9854, we calculate the average vibrational energy of the parent molecules and take it as the internal energy, Eint (CH2 BrCl) = 2.70 kJ/mol. The internal energy of the chloromethyl radical, Eint (CH2 Cl), is equal to the sum of the rotational energy Erot (CH2 Cl) and the vibrational energy Evib (CH2 Cl).

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-5

Li et al.

J. Chem. Phys. 140, 034303 (2014)

The Butler model for the rotational energy partitioning to nascent radicals begins with a classical model for angular momentum conservation. The model neglects both the angular momentum of the photon and the electronic angular momentum of the products, and assumes that the supersonic expansion rotationally cools the chlorobromomethane molecule, so that its initial angular momentum is zero. Conservation of angular momentum then requires that the classical orbital angular momentum of the recoiling photofragments is equal in magnitude and opposite in direction to the rotational angular momentum of the chloromethyl radical product. This gives the usual classical expression: μ |vrel | b = I ω.

(9)

Here, b is the impact parameter, vrel is the relative velocity between the Br and CH2 Cl co-fragments and |vrel | is its magnitude, μ is the reduced mass of the CH2 Cl + Br system, ω is the angular speed of the rotating chloromethyl radicals, and I is the moment of inertia about the axis of rotation of the chloromethyl radical. In a simple impulsive model where the repulsive force acts along the direction of the C–Br bond, the axis of rotation is perpendicular to the plane containing the center of mass of the chloromethyl radical and the C–Br bond. Equation (9) leads to the prediction below for the rotational energy, Erot , imparted to the chloromethyl radical as a function of the total translational energy ET : Erot =

μb2 ET . I

(10)

We substitute Eq. (10) into Eq. (8) and rearrange to get Evib (CH2 Cl) = Ehν + Eint (CH2 BrCl) − D0 (C − Br)   μb2 2 ET . (11) − Eint (Br[ PJ ]) − 1 + I Now, we use a simplified Butler model, and just calculate the μb2 /I factor from the equilibrium geometry of the parent molecule, which comes from the Gaussian output of the B3LYP/6-311++G (3df, 2p),29 determining the μb2 /I factor to be 1.16. Substituting the value into Eq. (11) and using the experimental total translational energy distributions, we get the vibrational energy distributions of the nascent chloromethyl radicals for the two channels. Figure 4 presents the vibrational energy distribution of the chloromethyl radical from CH2 Cl + Br∗ channel and Figure 5 presents that from CH2 Cl + Br channel. As shown in Figures 4 and 5, there is just one peak in the vibrational energy distribution of the chloromethyl radical from CH2 Cl + Br∗ channel, and three peaks in the vibrational energy distribution of the chloromethyl radical from CH2 Cl + Br channel. We list the vibrational energies of these peaks in Table I, the errors correspond to the half width at halfmaximum (HWHM) of each peak. The vibrational normal modes of the chloromethyl radical26, 29, 30 are also listed in Table I for comparison. As shown in Table I, the only one peak in the vibrational energy distribution of the chloromethyl radical from CH2 Cl + Br∗ channel is centered at 33 kJ/mol, the half width at half-maximum (HWHM) of the peak is 20 kJ/mol. Due

FIG. 4. Vibrational energy distribution (blue curve) of the chloromethyl radical from CH2 Cl + Br∗ channel. The red plus sing indicates the location of the peak.

to the limit of the resolution, we cannot assign the vibrational energy distribution to the vibrational energy levels of the chloromethyl radical. We can just determine that there are really some vibrational modes excited in the chloromethyl radical, and that the CH2 s-stretch mode v1 (36.56383 kJ/mol) of the chloromethyl radical has the most probability of excitation, considering the experimental errors. There are three peaks in the vibrational energy curve of the CH2 Cl + Br channel. The energy value of the first peak is 51 ± 4 kJ/mol, which is close to the energies of several combined modes in the chloromethyl radical, such as, v1 + v2 , v1 + v6 , v5 + v3 , and v5 + v6 . So, the peak cannot be assigned exactly. The second peak locates at 70 ± 3 kJ/mol, and the energy gap between the second peak and the first peak is 19 ± 3 kJ/mol, which can be attributed to the CH2 scissors mode v2 of the chloromethyl radical. The energy value of the third peak is 88 ± 6 kJ/mol. The energy gap between the third peak and the first peak is 37 ± 6 kJ/mol. This energy gap may be attributed to three vibrational modes, the

FIG. 5. Vibrational energy distribution (blue curve) of the chloromethyl radical from CH2 Cl + Br channel. The red plus sings indicate the locations of the peaks.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-6

Li et al.

J. Chem. Phys. 140, 034303 (2014)

TABLE I. Comparison between the vibrational normal modes and the peak energies of the vibrational energy distributions of the chloromethyl radical. Vibrational normal modes of chloromethyl radical

P(Evib )’s peaks Br∗

Frequency Mode

Approximate description

v1 (a1 ) v2 (a1 ) v3 (a1 ) v4 (b1 ) v5 (b2 ) v6 (b2 )

CH2 s-stretch CH2 scissors CCl stretch CH2 wag CH2 a-stretch CH2 rock

a b c

channel

Br channel

cm−1 (expt.)

cm−1 (calc.)c

kJ mol−1

Evib (kJ mol−1 )

Evib (kJ mol−1 )

Evib (kJ mol−1 )

3055.08a 1391b 827b 395b

3057 1383 833 386 3276 984

36.56383a 16.648b 9.896b 4.726b 39.214c 11.784c

33 ± 20

51 ± 4 70 ± 3 88 ± 6

19 ± 3 37 ± 6

Reference 30. Reference 26. Calculated using Gaussian 03 at B3LYP/6-311++G(3df, 3pd) level,29 scaled by 0.9854.

CH2 s-stretch mode v1 , the CH2 a-stretch mode v5 and the overtone of the v2 mode. We cannot determine which mode is excited; maybe the three modes all have probability for excitation. According to the study of Ng and co-workers,21 in the ∗ Br channel at 224.609, 231.962, 234.021, and 264.921 nm, all the total translational energy distributions could be fitted well by one Gaussian function, and the observed anisotropy parameters of 1.29–1.51 almost didn’t changed with the speed of the Br∗ fragments at each wavelength. They predicted that the formation of the Br∗ channel originated mostly from 3A state. From Figure 2(a) we can find, the present result is consonant with the previous conclusion. Ng and co-workers also found that, in the Br channel at 224.834, 229.200, 233.681, 250.953, 264.209, and 266.634 nm, all the total translational energy distributions could be decomposed into a slow and a fast Gaussian component. In 250.95–266.63 nm wavelength range, the anisotropy parameters of the fast Gaussian components were higher than that of the slow components. So, the slow and fast Gaussian components were ascribed to the initial excitation of the [2A , 1A ] and 3A states, respectively. In the wavelength range of 193.3–233.68 nm, they found that the anisotropy parameters of the fast Gaussian components were similar to that of the slow components. Comparing Figure 2(b) with the speed distribution of Br at 233.681 nm in the experiment of Ng and co-workers,21 we can find that there is one peak added to the speed distribution of Br at 232.98 nm in our experiment, which is the highest peak in the middle of the curve. If without this peak, the curve of the speed distribution in the present study would be comparable to the curves in the previous study. Moreover, as shown in Figure 2(b), the anisotropy parameters distribution (anisotropy parameters at each speed) in the present study would also be similar to that in the study of Ng and co-workers, if the anisotropy parameter distribution relative to the highest peak is removed from the curve. As mentioned above, we have repeated the experiments for three times, and the structure in the speed distribution of Br is reproducible. We can infer that the present experimental data are reliable. Even though, the discrepancy still needs to be interpreted. To explain the above results, we have calculated geometries of the chlorobromomethane molecule in the ground state and three excited states, 3A , 4A , and 2A , using the Gaus-

sian 03 package.29 The geometries of the ground state and the three excited states were optimized at B3LYP/6-311++G (3df, 3pd) level and UCIS/6-311++G (3df, 3pd) level, respectively. The calculation results exhibit that, comparing with the ground state, in all the three excited states the length of C–Br bond is increased obviously, the H–C–H angle is enlarged, and the length of C–H bond is decreased. These changes would excite the CH2 s-stretch mode v1 , the CH2 scissors mode v2 , and the C–Br stretch mode v5 in the parent molecule31 when the n (Br) → σ ∗ (C–Br) transition happens. The calculations imply that, following absorption of one UV photon of 234 nm by the parent molecule, the CH2 s-stretch mode v1 , the CH2 scissors mode ν 2 , and the C–Br stretch mode v5 , are excited in the parent molecule. Then, the CH2 s-stretch mode v1 of the parent molecule relaxes to the CH2 s-stretch mode v1 of the chloromethyl radical, and the CH2 scissors mode ν 2 of the parent molecule to the CH2 scissors mode ν 2 of the chloromethyl radical. The calculation also reveals that when the parent molecule transits from the ground state to the [4A , 2A ] states the change of the H–C–H angle is comparatively larger than to the 3A state. The Br∗ photo-fragments mainly come from the 3A state, while Br photo-fragments partly originate from the [4A , 2A ] states, so the CH2 scissors mode ν 2 is excited more easily in the Br channel than in the Br∗ channel. This can explain why there is an obvious peak of the CH2 scissors mode ν 2 in the vibrational energy curve of the chloromethyl radical from the Br channel. The anisotropy parameters relative to the ν 2 peak (shown in Figure 2(b)) is obviously smaller than that at the other parts, is about 0.45. This observation is consistent with the assignment that the ν 2 peak comes from the excitation of the CH2 scissors mode ν 2 of the parent molecule at the [4A , 2A ] states. As mentioned in the experimental part, the vibrational temperature of the sample molecules was kept at 40 ◦ C in our experiment, which was higher than that in the experiment of Ng and co-workers.21 We presume that, the Frank-condon factor might favor the transition from some one vibrational excited level of the ground state to the vibrational excited levels ν 2 of the [4A , 2A ] states. So, a peak of the CH2 scissors mode ν 2 emerged in the total translational energy distribution of Br channel at 232.98 nm in our experiment, but did not in the experiment of Ng and co-workers. For proving the

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

034303-7

Li et al.

guess, we have run a comparison experiment for Br channel at 232.98 nm, in which the pulsed nozzle is not heated and the reservoir storing the liquid sample is cooled with ice, just as Ng and co-workers did.21 The result indicates that the highest peak disappears and the curve of the total translational energy distribution is similar to that of Ng and co-workers.21 V. CONCLUSION

Based on our study, the following conclusions regarding the photodissociation of chlorobromomethane at 234 nm can be reached: (1) In the Br∗ channel, there are a number of vibrational normal modes in the chloromethyl radical to be excited, and of these modes, the CH2 s-stretch mode v1 has the most probability of excitation. (2) In the Br channel, there are three distinct peaks in the vibrational energy curve of the chloromethyl radical. The highest peak can be attributed to the CH2 scissors mode ν 2 of the chloromethyl radical, but the other two peaks cannot be assigned exactly. (3) The analysis of the excited vibrational modes in the chloromethyl radicals has provided strong evidence that, following absorption of one UV photon of 234 nm by the parent molecule, other vibrational modes besides the C–Br stretch mode v5 are also excited. ACKNOWLEDGMENTS

All the authors gratefully acknowledge support from the Key Foundation of Chinese Ministry of Education (Grant no. 211117). And all the authors gratefully acknowledge help from Professor Laurie J. Butler and Dr. Caroline C. Womack (The University of Chicago, USA). 1 Z.

Yu, M. Cheng, X. Xu, D. Yu, Y. Du, and Q. Zhu, Chem. Phys. Lett. 488, 158 (2010). 2 M.-T. Tsai, Y.-T. Liu, C.-Y. Liu, P.-Y. Tsai, and K.-C. Lin, Chem. Phys. 376, 1 (2010). 3 J. C. Poveda, A. Guerrero, I. Álvarez, and C. Cisneros, J. Photochem. Photobiol., A 220, 31 (2011).

J. Chem. Phys. 140, 034303 (2014) 4 C.

Escure, T. Leininger, and B. Lepetit, Chem. Phys. Lett. 480, 62 (2009). 5 G. Altinay and R. B. Metz, J. Am. Soc. Mass Spectrom. 21, 750 (2010). 6 A. Portnov, M. Epshtein, S. Rosenwaks, and I. Bar, J. Chem. Phys. 136, 024313 (2012). 7 Z. Chen, F. Liu, B. Jiang, X. Yang, and D. H. Parker, J. Phys. Chem. Lett. 1, 1861 (2010). 8 G. Gitzinger, M. E. Corrales, V. Loriot, R. de Nalda, and L. Bañares, J. Chem. Phys. 136, 074303 (2012). 9 J. Li, C. Zhang, Q. Zhang, Y. Chen, C. Huang, and X. Yang, J. Chem. Phys. 135, 116102 (2011). 10 X. Tang, X. Zhou, M. Wu, S. Liu, F. Liu, X. Shan, and L. Sheng, J. Chem. Phys. 136, 034304 (2012). 11 B. J. Ratliff, C. C. Womack, X. N. Tang, W. L. Landau, L. J. Butler, and D. E. Szpunar, J. Phys. Chem. A 114, 4934 (2010). 12 C. C. Womack, W.-H. Fang, D. B. Straus, and L. J. Butler, J. Phys. Chem. A 114, 13005 (2010). 13 B. J. Ratliff, B. W. Alligood, L. J. Butler, S.-H. Lee, and J. J.-M. Lin, J. Phys. Chem. A 115, 9097 (2011). 14 C. C. Womack, R. S. Booth, M. D. Brynteson, L. J. Butler, and D. Szpunar, J. Phys. Chem. A 115, 14559 (2011). 15 W. B. Tzeng, Y. R. Lee, and S. M. Lin, Chem. Phys. Lett. 227, 467 (1994). 16 P. Cadman and J. P. Simsons, Trans. Faraday Soc. 62, 631 (1966). 17 P. Zou, W. S. McGivern, and S. W. North, Phys. Chem. Chem. Phys. 2, 3785 (2000). 18 J. Doucet, R. Gilbert, P. Sauvageau, and C. Sandorfy, J. Chem. Phys. 62, 366 (1975). 19 W. S. McGivern, T. Li, P. Zou, and S. W. North, J. Chem. Phys. 111, 5771 (1999). 20 S. H. Lee, Y. J. Jung, and K. H. Jung, Chem. Phys. 260, 143 (2000). 21 J. G. Zhou, K. C. Lau, E. Hassanein, H. F. Xu, S. X. Tian, B. Jones, and C. Y. Ng, J. Chem. Phys. 124, 034309 (2006). 22 R. Zhu, B. Tang, X. Zhang, and B. Zhang, J. Phys. Chem. A 114, 6188 (2010). 23 X. Zhang, D. Wu, and B. Tang, Acta Phys.-Chim. Sin. 28, 1045 (2012). 24 A. T. J. B. Eppink and D. H. Parker, Rev. Sci. Instrum. 68, 3477 (1997). 25 Y.-J. Jee, Y.-J. Jung, and K.-H. Jung, J. Chem. Phys. 115, 9739 (2001). 26 T. D. Fridgen, X. K. Zhang, J. M. Parnis, and R. E. March, J. Phys. Chem. A 104, 3487 (2000). 27 J. Li, J. Yang, Y. Mo, K. C. Lau, X. M. Qian, Y. Song, J. Liu, and C. Y. Ng, J. Chem. Phys. 126, 184304 (2007). 28 S. Arepalli, N. Presser, D. Robie, and R. J. Gordon, Chem. Phys. Lett. 117, 64 (1985). 29 M. J. Frisch, G. W. Trucks, H. B. Schlegel et al., Gaussian 03, Revision A.01 Gaussian, Inc., Pittsburgh, PA, 2003. 30 E. S. Whitney, T. Haeber, M. D. Schuder, A. C. Blair, and D. J. Nesbitt, J. Chem. Phys. 125, 054303 (2006). 31 T. Shimanouchi, Tables of Molecular Vibrational Frequencies Consolidated (National Bureau of Standards, Gaithersburg, MD, 1972), Vol. I, pp. 1–160.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7 On: Thu, 04 Dec 2014 14:46:10

Vibrational excitations in chloromethyl radical formed by the photodissociation of chlorobromomethane.

Using velocity map ion imaging, the photodissociation of chlorobromomethane (CH2BrCl) at 233-234 nm has been studied. The total translational energy d...
915KB Sizes 0 Downloads 5 Views