Article pubs.acs.org/JPCA
Dynamics of H Atom Production from Photodissociation of Acetic Acid‑d1 Sung Man Park, Chan Ho Kwon,* and Hong Lae Kim*
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 28, 2015 | doi: 10.1021/acs.jpca.5b05241
Department of Chemistry, College of Natural Sciences and Institute for Molecular Science and Fusion Technology, Kangwon National University, Chuncheon 200-701, South Korea ABSTRACT: Detailed dissociation dynamics of H(D) from acetic acid-d1 (CH3COOD) has been investigated upon electronic excitation to the 1(n,π*), S1 state at 205 nm by measuring laser-induced fluorescence spectra of the fragment H(D) atoms. In addition, quantum yields for the H(D) atom dissociation channels, CH3COO + D and CH2COOD + H, were measured, which are 0.07 ± 0.03 and 0.17 ± 0.03, respectively. From the Doppler broadened spectra, the center-of-mass translational energy releases into products were obtained. To determine the detailed dissociation dynamics, twodimensional potential energy surfaces along the reaction coordinate including the coordinate directly coupled to the dissociation coordinate were examined by employing quantum chemical calculations. For the CH3COO + D channel, the coupled coordinate is the dihedral angle of D against the COO plane. The dissociation of D(H) from acetic acid should take place along the triplet surface via surface crossing from the initially excited S1 state. Along the triplet surface, an exit channel barrier exists, which originates from the structural difference between the T1 and the product asymptotes, especially the dihedral angle of D against the COO plane. The observed translational energy releases were successfully estimated by the barrier impulsive model based upon the calculated two-dimensional potential energy surfaces at the B3LYP/cc-pVDZ level of theory.
■
INTRODUCTION Photodissociation of small, saturated carboxylic acids such as formic (HCOOH) and acetic acid (CH3COOH) has been of great interest because of the acids’ atmospheric, combustion, and interstellar chemistry. Photodissociation dynamics of acetic acid has been thoroughly investigated because the molecule is large enough to show multiple dissociation pathways but small enough to probe underlying dissociation dynamics in detail.1−7 The UV absorption spectrum shows continuous absorption starting from 250 nm and a maximum near 210 nm, where this absorption band has been assigned as the n → π* transition.8,9 Upon electronic excitation to its 1(n,π*)(A″) state near 210 nm, the following primary decomposition processes can take place. CH3COOH(X̃ 1A′) + hν → CH4 + CO2 ̃2
may produce CH4 by the similar roaming mechanism. In any case, the quantum yield of reaction 1 was estimated to be around 0.1.12 Among the primary processes producing the two groundstate radical products, reaction 3 is forbidden according to, albeit simple, molecular orbital symmetry correlation arguments from the (n,π*)(A″) state.3 Reaction 2 has been thoroughly investigated by measuring rotationally resolved spectra of OH by laser-induced fluorescence spectroscopic techniques, from which the available energy, partitioned into product translation, rotation, and vibration, was measured. A large but similar amount of energy was found to be released into the product translation at various excitation energies, which led to the conclusion that there should be an energy barrier in the exit channel leading to the OH production.3 Upon excitation at 222 nm, the quantum yield for the OH production channel was measured to be 0.55−0.70, whereas the quantum yields of OH production from acid dimers were also measured to be less than 7% of the monomers.2 Then, considering the symmetry correlation while neglecting state-mixing, the quantum yields for the H atom production channels, reactions 4 and 5, would be at least 0.2−0.3 altogether. Although the OH production channels from acetic acid have been studied in detail, the H atom production channels, reactions 4 and 5, have never been studied. The H atom
(1) 2
CH3CO(X A′) + OH(X Π)
(2)
CH3(X̃ 2A″) + COOH(X̃ 2A′)
(3)
CH3COO(X̃ 2A′) + H(2S)
(4)
CH 2COOH(X̃ 2A′) + H(2S)
(5)
The molecular products (reaction 1) can be formed by H atom migration on the ground electronic state via internal conversion from the initially excited 1(n,π*) state. Recently, in the case of photodissociation of acetaldehyde, dissociation of H or CH3 followed by production of CH4 by a roaming mechanism was suggested.10,11 The H atom from reaction 4 © XXXX American Chemical Society
Received: June 2, 2015 Revised: August 20, 2015
A
DOI: 10.1021/acs.jpca.5b05241 J. Phys. Chem. A XXXX, XXX, XXX−XXX
Article
Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on August 31, 2015 | http://pubs.acs.org Publication Date (Web): August 28, 2015 | doi: 10.1021/acs.jpca.5b05241
The Journal of Physical Chemistry A
intensities. The diameter of the focused beam was chosen to be large enough for the fast H atoms not to escape from the viewing zone in a perpendicular direction with respect to the beam path in a given time interval, which ensures the correct shape of the Doppler profiles. This was checked by measuring the spectrum of H produced from impulsive dissociation of H2S upon UV absorption whose transition is known to be perpendicular at 205 nm, resulting in a distinct shape of the Doppler profiles under our experimental geometry in which the polarization direction of the light is perpendicular to the propagation direction.16 Incorrect Doppler profiles were noticed with tightly focused lenses where the center of the peak was lower than expected because of loss of the fast H atoms from the viewing zone as mentioned above. The spectra of H atoms were measured in a linear regime of the log−log plot of the fluorescence signal versus the laser power with a properly chosen lens (f = 400 mm). The linear plot was obtained from the laser power of 30 μJ/pulse up to 150 μJ/ pulse with the slope of 2.8 ± 0.1, which suggests that the dissociation of H and fluorescence should result from the threephoton processes. The laser power thus was kept as low as possible to avoid any saturation effect in the spectra, which was about 60 μJ/pulse. The line width of the laser light was measured from the gaseous I2 spectra near 615 nm, which was 0.07 cm−1 at fwhm. This laser line profile was deconvoluted from the measured Doppler profiles to deduce the translational energies of the H atoms.
production channels from photodissociation of formic acid in the (n,π*)1,3 state at 205 nm were studied by measuring Doppler broadened H atom spectra.13 From the spectra, the translational energy releases in the H atom production channels were measured. The observed translational energy releases were successfully explained by the barrier impulsive model. It has been concluded that the dissociation of H should take place on the triplet state with an exit channel barrier from the observed translational energy releases and quantum chemical potential energy surface calculations. In the present article, we report the laser-induced fluorescence spectra of H(D) atoms produced from photodissociation of acetic acid-d1 upon photon absorption at 205 nm, from which the translational energies released into products are measured by Doppler profile analyses. Channels 4 and 5 are distinguished from each other by isotopic substitution. Thus, the primary dissociation channels are CH3COOD(X̃ 1 A′) + hν → CH3COO(X̃ 2 A′) + D(2S) ̃2
2
CH 2COOD(X A′) + H( S)
(4′) (5′)
From the spectra, quantum yields for channels 4′ and 5′ were also measured relative to that of H2S at 205 nm. The observed translational energy releases are interpreted based upon potential energy surfaces obtained by quantum chemical calculations, and the detailed dissociation dynamics will be discussed.
■
■
EXPERIMENTS Experiments were performed in a low-pressure flow-cell which was evacuated to a pressure of 10−3 Torr by a mechanical pump. The gaseous sample was slowly flowed through the cell at a pressure of about 50 mTorr from a reservoir that contained the liquid and the vapor in equilibrium at ambient temperature. The sample pressures were controlled by needle valves. The liquid acetic acid-d1 and trifluoroacetic acid were purchased from Aldrich (>98% purity) and used without further purification. The purity and isotopic substitution were checked by IR and NMR spectra. Spectra of H atoms produced from photodissociation were measured by laser-induced fluorescence. The 205.14 nm light (205.09 nm for D atoms) induces the 1s → 3s,3d transition by two-photon absorption, and fluorescence from the 3s → 2p transition at 656.6 nm (Balmer-α line) was detected.14,15 The 205.14 nm light was generated by mixing of the fundamental and the doubled output of a dye laser (Continuum ND-6000) in a BBO crystal, which was pumped by the second harmonic of an Nd:YAG laser (Continuum Surelite III). The conversion efficiency of mixing was maximized by placing a half-wave plate in the beam path, and the resulting output was horizontally polarized. The pulse width of the light was 7 ns, and the wavelength was measured by a wavelength calibrator. Within the same light pulse, the parent molecule absorbs a photon of energy high enough to be dissociated, producing H atoms, which results in the so-called one-color experiment. The fluorescence was detected by a photomultiplier tube (Hamamatsu R-928) through a bandpass filter centered at 656 nm, and the measured signal was recorded by a digital storage oscilloscope. The integrated signals were fed to a PC, and the spectra were corrected by variation of the laser powers. To induce the fluorescence from the H atoms upon twophoton absorption, use of lenses of various focal lengths was attempted to focus the beam and hence to increase the light
RESULTS AND DISCUSSION
Translational Energy Releases. The spectra of H and D atoms produced from photodissociation of acetic acid-d1 are presented in Figure 1. The Doppler broadened spectra are symmetrically extended from the center frequency, ν0, which is 97 494 cm−1 in the present experiment, from −υν0/c to +υν0/c, where c is the speed of light and υ is the magnitude of the velocity vector. Although other experimental geometries such as εd ∥ kp (polarization direction of the light parallel to the propagation direction) could not be applied in this experiment, the Gaussian-like Doppler profiles were observed for the present, perpendicular geometry, which implies that the isotropic velocity distribution of the H atom fragments would be assumed. The observed one-dimensional Doppler profile was fitted by a single Gaussian speed distribution with the laserline deconvolution. Thus, the average kinetic energy of the two photofragments in the center-of-mass frame can be obtained from the second moment of the excitation spectra using the following formula: ⟨E T⟩ = (mRH /mR )⟨(Δν /ν0)2 ⟩3mHc 2/2
where mH, mRH, and mR are the mass of the H atom, molecule being photodissociated, and of the sibling fragment of the H atom, respectively. The spin−orbit couplings in the 3s and 3d state are neglected in obtaining the translational energies because the experimental second moment is not accurate enough to justify the very small correction of the fine structure splitting (
