CARS spectroscopy of O2(1Δg) from the Hartley band photodissociation of O3: Dynamics of the dissociation

Abstract

Rotationally and vibrationally resolved CARS spectra of the O2(1Δg) photofragment produced by the photodissociation of O3 at 17 wavelengths between 230 and 311 nm are reported. The spectra are taken under collision-free conditions, therefore, they reveal the nascent rotational and vibrational state distributions of the O2(1Δg) photofragment. At all photolysis wavelengths studied the vibrational distribution peaks very sharply at v=0, although all energetically allowed vibrational states are observed. The rotational state distributions are narrow, and peak typically at high J. The rotational distribution shifts to lower J as the photolysis wavelength increases. These observations imply vibrationally adiabatic, rotationally impulsive energy release in the dissociation. The shape and width of the rotational distributions can be completely accounted for by the spread in the O3 thermal rotation and zero-point vibration contributions to the O2(1Δg) photofragment angular momentum. The most striking observation about the O2(1Δg) photofragment quantum state distribution is an apparent propensity for even-J states. Experiments with 18O enriched ozone indicate that this propensity is observed only for 16O16O, not for 18O16O, and by implication not for 17O16O. We show that this is the consequence of a selective depletion of only odd-J rotational states of 16O16O(1Δg) by a curve crossing to O2(3Σg), but an equal depletion of both even-J and odd-J rotational states of 18O16O and 17O16O(1Δg) by the curve crossing. The odd-J selectivity for 16O16O is a consequence of the restriction of 3Σg to only odd-J states, due to the requirement of even nuclear exchange symmetry for this homonuclear species with spin-zero nuclei. As a result of the different curve crossing behavior, the quantum yield for 3Σg is twice as great for 18O16O and 17O16O as it is for 16O16O, and this imposes a mass-independent isotopic fractionation in the photodissociation: the O2(1Δg) fragments are depleted of 17O and 18O, while the O2(3Σg) fragments are enriched in these isotopes.

Photodissociation of ozone at 193 nm by high-resolution photofragment translational spectroscopy

“…The vibrational state distribution-is thus determined by Franck-Condon factors at the crossing seam. The branching fraction obtained is 16% for this channel, slightly higher than reported elsewhere in the Hartley band, [11,12] but consistent with a nonadiabatic transition coupled with a higher velocity through the crossing region.…”

“…The dominant photochemical events there are fairly well understood as a result of several molecular beam experiments, [10,11] an extensive series of CARS experiments performed by Valentini and coworkers, [12], and resonance Raman experiments by Kinsey and others. [13,14] It is known that, in the Hartley band, the initially prepared "B" state (1 B 2 in C 2 v, iJ in Cs) correlates to singlet products, so these dominate (85-90%).…”

“…This is because the preserved 0-0 bond length is little changed on going from the 0 3 equilibrium configuration (1.271 A) to the product 0 2 (a 1~9 ) (1.215 A). [12] If in this case, too, the 0 2 ( a 1 ~9 ) product is predominantly in the ground vibrational level, then the rotational energy peaks at 21 kcal/mol, corresponding to J = 70. This (37%) is a larger fraction of energy in rotation than is seen at the longer wavelengths, and significantly larger than is expected based on a simple rotationally impulsive model (23%).…”

“…[4,47,45] If we assume a· rotationally impulsive dissociation, the fraction of available energy appearing in rotation is variously given in the range of 15 to 23%. [12,47] For the maximum of the peak, the translational energy release corresponds to 68 kcal/mol. If this represents 77 to 85% of the avalable energy (not including vibration) then the energy in rotation is in the range 12 to 20 kcaljmol, corresponding to J = 53 -69 Taking the intermediate value can thus infer the most probable vibrational level to be roughly v=19, with a very broad range of levels populated.…”

CO2+O(1D) isotopic exchange: Laboratory and modeling studies

“…Increasing the quantum yield of the singlet channel to 100% and decreasing it to 80% was found to have no significant effect on the model-calculated isotopic compositions. Valentini et al, [1987] showed that the UV photolysis of 03 can produce mass independently enriched 02 through an isotopically selective curve crossing from the singlet surface to the triplet surface. This effect is not relevant to the present study because the process does not influence which 03 bond is broken, only whether the photolysis leads to singlet or triplet products.…”