| Abstract
| - In a companion paper (Kroll, J. H.; Clarke, J. S.; Donahue, N. M.; Anderson, J. G.; Demerjian, K. L. J. Phys.Chem. A2001, 105, 1554) we present direct measurements of hydroxyl radical (OH) yields for the gas-phasereaction of ozone with a number of symmetric alkenes. Yields are strongly pressure-dependent, contrary tothe results of prior scavenger studies. Here we present a statistical-dynamical model of OH production fromthe reaction, utilizing RRKM/master equation calculations to determine the fate of the carbonyl oxideintermediate. This model agrees with our experimental results, in that both theory and observations indicatestrongly pressure-dependent OH yields. Our calculations also suggest that ethene ozonolysis produces OHvia a different channel than the substituted alkenes, though the identity of this channel is not clear. Thischannel may play a role in the ozonolysis of monosubstituted alkenes as well. Our time-dependent masterequation calculations show that the discrepancy between OH yields measured in our direct study and thosemeasured in prior scavenger studies may arise from differing experimental time scales; on short time scales,OH is formed only from the vibrationally excited carbonyl oxide intermediate, whereas on longer time scalesOH formation from thermal dissociation may be significant. To demonstrate this we present time-dependentmeasurements of OH yields at 10 Torr and 100 Torr; yields begin increasing after hundreds of milliseconds,an effect which is much more pronounced at 100 Torr. These results are entirely consistent with theoreticalpredictions. In the atmosphere, the thermalized carbonyl oxide may be susceptible to bimolecular reactionswhich, if fast enough, could prevent dissociation to OH; however there is little experimental evidence thatany such reactions are important. Thus we conclude that both mechanisms of OH formation (dissociation ofvibrationally excited carbonyl oxide and dissociation of thermalized carbonyl oxide) are significant in thetroposphere.
|
