Reports: B6

44764-B6 Computational Studies of the Atmospheric Chemistry of Alkene Ozonolysis Intermediates

Keith T. Kuwata, Macalester College

My laboratory has focused on two projects with PRF support.  Both have involved the use of quantum chemistry and statistical rate theory to predict quantities that can be tested against experiment.

           (1) The Ozonolysis of  Trans-2-Butene. My students and I used highly accurate electronic structure calculations, combined with master equation simulations, to predict the outcome of trans-2-butene ozonolysis in the atmosphere.  Figure 1 summarizes our results.

Figure 1.  Trans-2-butene ozonolysis mechanism treated by RRKM/master equation simulations.  The MCG3//QCISD/6-31G(d) relative energies (at 0 K, in kcal mol-1) are given in blue, and the 1-atm pseudo steady-state yields for the well and exit channels are given in purple.

The formation of the primary ozonide (16) was the entrance channel, and the initial energy of 16 was represented by a shifted thermal (298 K) distribution truncated at the zero-point corrected energy of transition structure TS 15.  The exit channels were the formation of syn acetaldehyde oxide 1 (and acetaldehyde, 18), formation of anti acetaldehyde oxide 4 (and acetaldehyde, 18), and reversion to trans-2-butene (13) and ozone (14).  There was no variation in pseudo steady-state yields from 1 Torr up to 1 atm; the 1-atm yields are shown in Figure 1.  None of the primary ozonide is predicted to revert to reactants or be collisionally stabilized. A large majority (66%) of 16 decomposes to the syn conformer of the Criegee intermediate.  This is in spite of the fact that the MCG3//QCISD/6-31G(d) barrier to forming the syn conformer is ~0.9 kcal mol-1 than the barrier to forming the anti conformer.  The preference for the syn conformer arises from the large number of states available to the chemically activated 16 at the syn transition state (TS 17), particularly from the active K-rotor of TS 17.

We then simulated possible reactions of the Criegee intermediates formed upon primary ozonide cycloreversion.  The parts of the mechanism that we considered, and the relative energetics of all participating species, are depicted in Figure 2, and master equation results are summarized in Figure 3.

Figure 2.  Acetaldehyde oxide mechanism treated by RRKM/master equation simulations.  The MCG3//QCISD/MG3 relative energies (at 0 K, in kcal mol-1) are given in blue.

Figure 3.  Product yields of the trans-2-butene ozonolysis reaction, as predicted by RRKM/master equation simulations.  Each simulation was run for 103 collisions to ensure that the pseudo steady state was achieved.  Trials were run at pressures of 1, 10, 50, 100, 200, 300, 400, 500, 600, 700, and 760 Torr.  Each pseudo steady-state yield reported is the average result of 105 Monte Carlo simulations. 

We predict that the branching between prompt isomerization (leading to 3 and 6) and thermalization (leading to 1 and 4) depends significantly on pressure.  At 1 atm, almost 70% of the Criegee intermediate formed in trans-2-butene ozonolysis is thermalized.  The subsequent reactivity of 1 and 4 was the subject of our other major project.

           (2) Transition State Theory Calculations on Criegee Intermediates. My students and I used canonical variational transition state theory (CVT), combined with the Eckart tunneling model, to predict thermal rate constants for Reactions 1, 2, 3, and 4:

Table 1 summarizes our predictions at the range of temperatures found in the troposphere.  All reported rate constants include the effect of tunneling.  We determined pseudo-first-order rate constants for hydration reactions 3 and 4 by assuming that the equilibrium vapor pressure of water was present at a given temperature.


TABLE 1.  First-Order and Pseudo-First-Order Rate Constants for Criegee Intermediate Reactions

CVT rate constant (s-1)

200 K

250 K

298 K

species 1 (R1 = H)

reaction 1

9.70 x 10-2

1.48 x 100

2.42 x 101

reaction 3

6.22 x 10-10

1.19 x 10-5

5.57 x 10-3

species 1 (R1 = syn-CHCH2)

reaction 1

9.17 x 10-1

1.12 x 101

1.46 x 102

reaction 3

1.41 x 10-8

9.56 x 10-5

2.64 x 10-2

species 1 (R1 = anti-CHCH2)

reaction 1

4.97 x 10-2

8.27 x 10-1

1.47 x 101

reaction 3

2.47 x 10-10

2.73 x 10-6

1.25 x 10-3

species 4

reaction 2

2.03 x 10-4

4.51 x 10-1

6.72 x 101

reaction 4

2.92 x 10-3

3.03 x 100

2.21 x 102

Syn-alkyl Criegee intermediates isomerize via a 1,4-hydrogen shift (Reaction 1) between 3 and 8 orders of magnitude faster than they react with water (Reaction 3).  The product of the isomerization, a vinyl hydroperoxide (3), falls apart promptly under atmospheric conditions to afford hydroxyl radical (•OH).  The significance of our finding is that the •OH yield from the ozonolysis of a given alkene should be determined only by the yield of syn-alkyl Criegee intermediate (such as what is shown in Figure 1 above).  For the case of trans-2-butene, we predict an •OH yield of 0.651, which agrees with the experimental consensus that the •OH yield lies between 0.52 and 0.67.

Anti-alkyl Criegee intermediates react far more rapidly with water (Reaction 4) than do the syn conformers, and the hydration reaction can be comparable or faster in rate than the isomerization (Reaction 2) to form dioxirane (6).  The product of hydration is a saturated hydroperoxide (10), which will not decompose under atmospheric conditions to form •OH.  Our results also suggest that other common Criegee intermediate scavengers, such as aldehydes and SO2, react almost exclusively with the anti conformer.  The yield of scavenger products (secondary ozonides for the aldehyde reaction, and H2SO4 for the SO2 reaction) is therefore an underestimate of the total yield of thermalized Criegee intermediate formed in an ozonolysis reaction.