FIGURE 6.4 Overall mechanisms of 03-alkene reactions in the gas and condensed phase, respectively.
In the liquid phase, the Criegee intermediates have been assumed to be zwitterions and hence the term "Criegee zwitterion" is commonly used. In the gas phase, the structure is usually written as a biradical (although it may really be more zwitterionic in character; e.g., see Cremer et al., 1993). Hence "Criegee biradical" is frequently used for this gaseous intermediate. Sander (1990), Bunnelle (1991), and Cremer et al. (1993) give a more detailed discussion of the structure and properties of the Criegee intermediate.
In any event, in the solvent cage in which they are formed in the liquid phase, or for higher molecular weight alkenes condensed on surfaces, the two fragments formed by decomposition of the primary ozonide are held in close proximity and recombine to form a secondary ozonide:
In addition to the effects of close proximity, the condensed phases act to remove excess energy in the fragments.
In the gas phase, however, two effects come into play. First, the two fragments formed on decomposition of the primary ozonide fly apart rapidly. As a result, the formation of secondary ozonides by recombination, reaction (44), does not occur to a significant extent in the gas phase [unless the mixture is doped with large quantities of an aldehyde or ketone to trap the Criegee intermediate (e.g., Neeb et al., f998a) or very high concentrations are used (e.g., Fajgar et al. (1996); Griesbaum et al. (1998)]. Second, there is no efficient mechanism for removal of excess energy from the carbonyl compound and the Criegee intermediate. The Criegee intermediate contains excess energy and either can be stabilized or decompose in a variety of ways. For example, for the two possible Criegee intermediates produced in the 03-propene reaction, the following paths are possible (where the branching ratios are the IUPAC evaluation recommendations (Atkinson et al., 1997a):
HCO + OH, (45b) CO + H20, (45c) -^C02 + H2, (45d) C02 + 2H, (45e)
Neeb et al. (1998a) report branching ratios for (45a) of 0.50, (45b) + (45c) of 0.23, (45d) + (45e) of 0.23, and (45f) of 0.04.
Table 6.10 gives the ranges of observed yields of the stabilized Criegee intermediates at 1 atm pressure in air and at room temperature. Clearly, significant decomposition of the intermediates occurs under typical tropospheric conditions.
The fraction of the primary ozonide decomposition that goes by path a in reaction (43) vs path b for unsymmetrical alkenes has been determined from the product yields in a number of studies (e.g., see series of papers by Grosjean and co-workers and Atkinson and co-workers). One might expect, based on thermochemi-cal arguments, that the decomposition giving the more stable, i.e., more highly substituted, biradical would be preferred, and this is indeed what has been observed (e.g., see Horie and Moortgat, 1991; Grosjean et al., 1994c; Atkinson et al., 1995c; and Grosjean and Grosjean, 1997, 1998a). Atkinson (1997a) recommends, based on the literature, the following branching ratios:
Was this article helpful?