Comments on the linear free-energy correlation between O3 and OH addition reactions reported in “rate constants for the gas-phase reactions of O3 with a series of carbonyls at 296 K”

✍ Scribed by J. S. Gaffney; S. Z. Levine

• Publisher: John Wiley and Sons
• Year: 1982
• Tongue: English
• Weight: 152 KB
• Volume: 14
• Category: Article
• ISSN: 0538-8066
• DOI: 10.1002/kin.550141111

✦ Synopsis

Recently, Atkinson et al. [l] reported measurements on the reaction of ozone with a series of carbonyls. In that study a correlation between ozone addition and hydroxyl-radical addition reactions was employed to predict OH addition coefficients for acrolein and crotonaldehyde of approximately 2 X and 5 X cm3 molecule-l s-l, respectively. These estimates, as pointed out by the authors, are in disagreement (factor of 2-3 lower) with the rate coefficients previously predicted by ourselves for the same OH addition reactions using linear correlations with both ionization potentials and O(3P) rate data [2]. It was also suggested in that paper that this discrepancy was probably due to the likelihood that O(3P) atoms react significantly with carbonyls via an abstraction mechanism, that is, O(3P) rate data could not be appropriately correlated with OH addition reaction data.

We believe that another, more probable, explanation exists for the above-mentioned discrepancy in rate constant estimates, and that this explanation involves the manner in which the correlation method is handled. Because this method of evaluating rate constants represents a potentially important predictive tool for chemical modelers, we feel it is necessary to reconcile the "apparent" disagreement in the OH-acrolein and OH-crotonaldehyde addition reaction coefficients estimated from 0 3 correlations as compared to ionization potential and O(3P) correlations. In doing so, we will also demonstrate the necessity of employing linear correlations in a consistent manner.

Atkinson et al. have themselves noted the existence of precedents for linear correlations of rate data in cases where reaction mechanisms are equivalent. However, in handling their data of log [ 0 3 rate constant] versus log [OH radical addition rate constant] for a series of alkenes, these authors appear to have arbitrarily fitted these data to a curve. If one performs a