The Onset of H + Ketene Products from Vinoxy Radicals Prepared by Photodissociation of Chloroacetaldehyde at 157 nm

We investigate the unimolecular dissociation of the vinoxy radical (CH₂CHO) prepared with high internal energy imparted from the photodissociation of chloroacetaldehyde (CH₂ClCHO) at 157 nm. Using a velocity map imaging apparatus, we measured the speed distribution of the recoiling chlorine atoms, Cl­(²P_3/2) and Cl­(²P_1/2), and derived from this the resulting distribution of kinetic energy, <i>P</i>(<i>E</i>_T), imparted to the Cl + vinoxy fragments upon dissociation. Using conservation of energy, the distribution of kinetic energy was used to determine the total internal energy distribution in the radical. The <i>P</i>(<i>E</i>_T) derived for the C–Cl bond fission presented in this work suggests the vinoxy radicals are mostly formed in the à state. We also took ion images at <i>m</i>/<i>z</i> = 42 and <i>m</i>/<i>z</i> = 15 to characterize the branching between the unimolecular dissociation channels of the vinoxy radical to H + ketene and methyl + CO products. Our results show a marked change in the branching ratio between the two channels from the previous study on the photodissociation of chloroacetaldehyde at 193 nm by Miller et al. (<i>J. Chem. Phys.</i>, <b>2004</b>, 121, 1830) in that the production of ketene is now favored over the production of methyl. To help analyze the data, we developed a model for the branching between the two channels that takes into account how the change in rotational energy en route to the products affects the vibrational energy available to surmount the barriers to the channels. The model predicts the portion of the C–Cl bond fission <i>P</i>(<i>E</i>_T) that produces dissociative vinoxy radicals, then predicts the branching ratio between the H + ketene and CH₃ + CO product channels at each <i>E</i>_T. The model uses Rice–Ramsperger–Kassel–Marcus rate constants at the correct sums and densities of vibrational states while accounting for angular momentum conservation. We find that the predicted portion of the <i>P</i>(<i>E</i>_T) that produces H + ketene products best fits the experimental portion (that we derive by taking advantage of conservation of momentum) if we use a barrier height for the H + ketene channel that is 4.0 ± 0.5 kcal/mol higher than the isomerization barrier en route to CH₃ + CO products. Using the G4 computed isomerization barrier of 40.6 kcal/mol, this gives an experimentally determined barrier to the H + ketene channel of 44.6 kcal/mol. From these calculations, we also predict the branching ratio between the H + ketene and methyl + CO channels to be ∼2.1:1