60th Annual Report on Research 2015

Ozonolysis has long been known to be an important mechanism in the tropospheric oxidation of alkenes from biogenic and anthropogenic sources, including emission from petrochemical sources. Additionally, these reactions are a non-photolytic source of OH radicals, accounting for ca. 30% of tropospheric OH radicals in the daytime and essentially all of the OH radicals at night. Yet there is still much unknown about this class of reactions, in particular the atmospheric fate of the carbonyl oxide species R1R2COO, known as Criegee intermediates, produced in the reaction. Ozonolysis of linear internal alkenes, such as trans-2-butene, will generate a methyl-substituted Criegee intermediate CH3CHOO. Ozonolysis of branched internal alkenes with a (CH3)2C=C structural unit, such as 2,3-dimethyl-2-butene, will yield a dimethyl-substituted Criegee intermediate (CH3)2COO. The OH yield from ozonolysis changes substantially with alkene structure, increasing dramatically (approaching 100%) for reactions that proceed through alkyl-substituted Criegee intermediates with alpha-hydrogens. With partial support from PRF, this laboratory has generated the methyl-, dimethyl-, and ethyl-substituted Criegee intermediates by an alternate synthetic route, characterized their UV absorption spectra, and investigated the unimolecular decay dynamics of energized Criegee intermediates to OH products.

Recently, cold methyl- or dimethyl-substituted Criegee intermediates have been vibrationally activated in the CH stretch overtone region to drive the 1,4 hydrogen transfer reaction from an adjacent methyl group to terminal oxygen, followed by dissociation to OH radical products. The infrared ‘action’ spectra of cold syn-CH3CHOO and (CH3)2COO were obtained in the CH stretch overtone region near 6000 cm-1 with sensitive detection of OH products. Many vibrational features were observed, arising from overtones or combinations of the various CH stretches of the Criegee intermediates, most of which appear to be mixed vibrational states. Some of the vibrational features exhibit distinguishable rotational contours enabling preliminary assignments, yet display extensive homogeneous broadening due to rapid intramolecular vibrational energy redistribution. Most importantly, for both CH3CHOO and (CH3)2COO, the lowest energy features at ca. 5600 cm-1 provide an upper limit for the effective barrier height for the critical hydrogen transfer step. Remarkable similarities are found between the IR spectra, effective barrier heights, and OH (v=0) product state distributions for syn-CH3CHOO and (CH3)2COO.

In both syn-CH3CHOO and (CH3)2COO, the experimentally determined effective barrier height was found to be 1-2 kcal mol-1 lower than recent theoretical predictions for the transition state (including zero-point corrections) separating the Criegee intermediate from vinyl hydroperoxide and OH + vinoxy products. This suggests that other alkyl-substituted Criegee intermediates with alpha-hydrogens are likely to be more facile in producing OH radicals than anticipated from current models of alkene ozonolysis in the troposphere. Complementary electronic structure calculations have been used to map out the intrinsic reaction coordinate (IRC) from each Criegee intermediate to the five-membered, ring-like transition state for 1,4 hydrogen transfer. In particular, the IRC revealed the geometric distortions required to reach the transition state, specifically internal rotation of the syn-methyl group, extension of the CH bond involving the alpha-hydrogen, and a heavy atom backbone motion to close the HCCOO ring. A related vibrational analysis was used to identify the minimum number of quanta required in specific modes to reach the transition state, most notably the CCOO ring closure mode. Finally, perturbation theory was used to show that the various CH stretch modes are coupled together by higher-order anharmonic coupling terms in the potential. Similar theoretical results were obtained for both the methyl- and dimethyl-substituted Criegee intermediates.