J. Phys. Chem. A 2010, 114, 7275–7283
7275
Photodissociation of Gaseous Acetyl Chloride at 248 nm by Time-Resolved Fourier-Transform Infrared Spectroscopy: The HCl, CO, and CH2 Product Channels Yu-Ting Liu, Ming-Tsang Tsai, Chia-Yun Liu, Po-Yu Tsai, and King-Chuen Lin* Department of Chemistry, National Taiwan UniVersity, Taipei 106, Taiwan, and Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106, Taiwan
Y. H. Shih and A. H. H. Chang Department of Chemistry, National Dong Hwa UniVersity, Shoufeng, Hualien 974, Taiwan ReceiVed: April 5, 2010; ReVised Manuscript ReceiVed: May 25, 2010
In one-photon dissociation of gaseous acetyl chloride at 248 nm, time-resolved Fourier-transform infrared emission spectroscopy is used to detect the fragments of HCl, CO, and CH2 in the presence of Ar or O2. The high-resolution spectra of HCl and CO are analyzed to yield the corresponding internal energy deposition of 8.9 ( 1.1 and 6.2 ( 0.9 kcal/mol. The presence of the CH2 fragment is verified by detecting the CO2 product resulting from the reaction of CH2 and the added O2. The probability of the HCl formation via a hot Cl reaction with the precursor is examined to be negligible by performing two experiments, the CH3COCl pressure dependence and the measurement of Br2 with Cl reaction. The HCl elimination channel under the Ar addition is verified to be slowed by 2 orders of magnitude, as compared to the Cl elimination channel. The observed fragments are proposed to dissociate on the hot ground electronic state via collision-induced internal conversion. A two-body dissociation channel is favored leading to HCl and CH2CO, followed by secondary dissociation. 1. Introduction UV photodissociation of acyl chlorides in the gas phase is dominated by the C-Cl bond rupture,1-5 although R C-C and C-Cl bonds have almost equivalent bonding energies. In the photodissociation of acetyl chloride, Butler and co-workers observed the release of highly anisotropic angular distributions of Cl atoms upon 1(n, π*)CO excitation at 248 nm.1,6,7 The C-Cl bond broke through the diabatic interaction with 1(nCl, σC-Cl*) repulsive configuration. The cleavage lifetime within 200 fs was found using photofragment ion imaging with femtosecond realtime clocking.8 A fraction of resultant CH3CO fragment could further undergo secondary decomposition with a barrier of 14.2 ( 0.5 kcal/mol.9 Given a shorter photolyzing wavelength at 236 nm, the relative yield of Cl(2P3/2) was measured as 0.6 with the average translational energy of 6.5 kcal/mol.2 The secondary product of CO was found to lie in the ground vibrational level with rotational population up to J ) 30.4 In contrast to the secondary production of CO, photodissociation of gaseous formaldehyde gave rise to CO on the hot electronic ground state via internal conversion.10 When acetyl chloride was dissociated at 248 or 266 nm with higher laser fluence, Kong and co-workers obtained the IR emission profile of CO (V e 8) with temperature of 6900 K.11 The CO was fragmented, as the acetyl radical continued absorbing one additional photon. When the photodissociation condition is changed to condensed phase, a distinctly different mechanism takes place. CH3COCl was dissociated to HCl exclusively forming the HCl-ketene complex with a T-shaped structure in an Ar-matrixisolated photodissociation.12 The HCl product was initiated from the hot ground state acetyl chloride via internal conversion but not from a secondary H-abstraction by the eliminated Cl atom. * To whom correspondence should be addressed:
[email protected]; fax, 886-2-23621483.
The above-mentioned primary dissociation pathway of the C-Cl bond rupture in the gas phase was not found. On the basis of the theoretical prediction in the matrix isolation conditions, the energy barrier to the C-Cl bond fission was enlarged such that the dissociation lifetime was slowed to 10-7 s and the HCl elimination may become the dominant channel.13 In this work, we find a new channel leading to HCl, CO, and CH2 elimination in the 248 nm photolysis of gas phase CH3COCl in the presence of Ar or O2 gases using time-resolved FT-IR emission spectroscopy. The addition of Ar or O2 helps regulate the relative yields of these fragments. Time-dependent rotational and vibrational distributions of HCl and CO are analyzed to determine their individual ro-vibrational energies. A further experiment on the reaction between Br2 and Cl is carried out to clarify whether the HCl product is contributed from the hot Cl reaction. Since the Cl atomic fragment cannot be detected by FT-IR spectroscopy, CH4 is added in reaction with Cl to help evaluate the elimination rate constants between Cl and HCl. The CH2 appearance is verified by the detection of CO2 in the reaction of CH2 and O2. The obtained CO product results from a mechanism different from those reported.3,4,9,11 Finally, with the aid of theoretical calculations, a plausible reaction mechanism is discussed. 2. Experimental Section The apparatus contains a step-scan FT-IR emission spectroscopy for spectral detection of the fragments and a six-way stainless steel cube as reaction chamber in which a multipass optical system is housed. The multipass optical system, composed of two pairs of 2 in. diameter gold-coated curved mirrors, was positioned perpendicular to the laser propagation direction. Like a Welsh cell, the optical unit worked to enhance the IR emission signal-to-noise ratio. The signal was delayed by three to four passes within the optical unit, equivalent to 3-5 ns.
10.1021/jp1030653 2010 American Chemical Society Published on Web 06/22/2010
7276
J. Phys. Chem. A, Vol. 114, No. 27, 2010
The enhanced signal was guided with a 2 in. diameter CaF2 lens sitting inside the reaction chamber, followed by an IR telescope to focus onto the entrance of FT-IR spectroscopy. The telescope situated outside was flushed with the N2 gas to prevent interference from the CO2 and moisture. A 20-30 ns pulsed KrF excimer laser (Lambda Physik, COMPex-102) with a repetition rate of 15-20 Hz was used as the radiation source emitting at 248 nm. The incident beam was controlled in the energy range of 15-60 mJ/pulse prior to focusing with a spherical lens to reduce its beam size from 25 × 10 to 3 × 1 mm2. The sample CH3COCl (purity, 99.95%) was used after several freeze-pump-thaw cycles, but Ar and O2 (purity >99.999%) were used without further purification. The precursor, along an inlet intruding from the top of the chamber, was injected at a pressure of ∼1-3 Torr through an effusive nozzle with a 2.5 mm orifice. The background pressure was adjusted to less than 50 mTorr by controlling the valve to the vacuum pump. Thus the flow beam was under approximately room temperature conditions. The molecular flow was photolyzed at a distance ∼1 cm downstream from the nozzle. The Ar or O2 partial pressure was controlled up to 1200 mTorr in the chamber. Each gas pressure was adjusted by an individual needle valve. The emission signals of HCl or CO fragments were temporally resolved with a step-scan FT-IR spectrometer (IFS 66v/S, Bruker), which allowed for evacuation of background air and moisture. The FT-IR was performed in a step-scan mode, with which the movable mirror of the interferometer can be moved step-by-step.14-17 The distance of each step is a multiple of onehalf the HeNe reference laser wavelength (0.316 µm). In this work the digitized signals were repeatedly collected for 30 laser shots. The delay time adopted to resolve two successive interferograms was set at 5 µs. The obtained interferograms were finally Fourier transformed to give rise to a series of timeresolved spectra. The high- and low-resolution spectra were acquired with spectral resolution at 0.25 (or 0.6) and 12 cm-1, respectively. The parameter adjustment and fast Fourier transformation were performed with version 8.0 of OPUS software. Appropriate cutoff filters were applied to increase the signalto-noise ratio of the IR spectra. The transmitted signal was monitored with an InSb detector cooled at 77 K, followed by a preamplifier. The detector and preamplifier had the response time of ∼700 and ∼200 ns, respectively. The output was then fed into a 200 kHz 16-bit transient digitizer for signal processing, which restricted the temporal resolution of IR signal at 5 µs with uncertainty of
