Reports: AC6
46303-AC6 Product Imaging of the Dissociation of Combustion Radicals
Objectives Studies of the photodissociation dynamics of hydrocarbon radicals provides a wealth of thermochemical, kinetic, and mechanistic data that is of use in modeling combustion chemistry. Allyl radicals are important in a variety of combustion processes including the reaction of propene with O2 and the thermal decomposition of butene. In collaboration with groups at Emory University and the University of Rome Sapienza, we are investigating the energy pathways and products for dissociation of the allyl radical, CH2CHCH2. Experimentally, the radical is made by photolysis or thermal decomposition of allyl iodide and is then excited with light at 248 nm. H and CH3 products are detected using standard rotating source molecular beam techniques or product imaging with velocity mapping. Theoretically, we are performing classical trajectory calculations on a potential energy surface calculated by the Bowman group using an aug-cc-pVTZ basis set. Results The competition between rearrangement of the excited allyl radical via a 1,3 sigmatropic shift versus sequential 1,2 shifts has been observed and characterized using isotopic substitution, laser excitation, and molecular beam techniques. Both rearrangements produce a 1-propenyl radical that subsequently dissociates to methyl plus acetylene. The 1,3 shift and 1,2 shift mechanisms are equally probable for CH2CHCH2, whereas the 1,3 shift is favored by a factor of 1.6 in CH2CDCH2. The translational energy distributions for the methyl and acetylene products of these two mechanisms are substantially different. Both of these allyl dissociation channels are minor pathways compared to hydrogen atom loss.
Ab initio calculations have been performed, using the GAUSSIAN03 package, on the energies and vibrational frequencies of the stable and transition state structures in the reaction path. These structures were optimized at the B3LYP/ccpVDZ level, followed by single point electronic energy calculations at the QCISD.T./cc-pVTZ level and zero point energy corrections by a B3LYP/cc-pVDZ anharmonic frequency analysis. The energies, geometries, and vibrational frequencies obtained from the ab initio calculations were then used as input for RRKM calculations to investigate the microcanonical reaction rates. The rates calculated in this way allowed the branching ratios to be calculated by a full integration of the forward rate equations, including the H-loss reactions. Results from this calculation were in reasonable agreement with the measured branching ratios when the error of 2-3 kcal/mole in the barrier heights was taken into account.
Trajectory calculations provide good agreement with the measured kinetic energy distribution for the H elimination channel. However, the source of the different distributions for the two methyl elimination channels has not been identified. The trajectory calculations give good agreement with the distribution for the 1,3 hydrogen shift channel, but not for the 1,2 hydrogen shift channel.
Outlook Promising areas of future research involve measurement of the interenal energy distribution of the methyl fragments by product imaging and the high-resolution kinetic energy distribution of the H atom by Rydberg tagging. In parallel, classical trajectory calculations can help to interpret these results. Of particular interest will be determining the reverse barriers to CH3+HCCH, H+allene, and H+propyne formation.