The configuration of alkyl substituents about carbon-carbon unsaturated bonds exerts a controlling influence on the rate of the ozonolysis reaction. Alkyl substituents can increase (via the inductive effect) and decrease (via the steric effect) the activity of unsaturated bonds, and an accurate description of this information ought to correlate with the ozonolysis rate coefficient. A strong linear relationship is observed (R2 = 0.94), providing the basis of our SAR method. SAR estimates were tested against literature measurements of ozonolysis rate coefficients for 48 aliphatic alkenes and dialkenes, and were found to be accurate to within a factor of 2.3 of the measured value for the entire dataset. This represents a significant improvement over methods reported in the literature, where quoted predictions are at best accurate to within a factor of 6.5. Rates of gas-phase ozonolysis of alkenes and dialkenes can now be predicted with unprecedented accuracy using a simple SAR. The SAR was then validated against new experimental data. Absolute rate coefficients for the gas-phase reaction of ozone with a series of alkenes were determined in a simulation chamber at 295 +/- 2 K and atmospheric pressure by monitoring the loss of ozone in the presence of excess alkene. The rate coefficients (in units of 1 x 10(-18) cm3 molecule(-1) s(-1)) are: 5.12 +/- 0.93 for 1-pentene, 2,3-dimethyl; 406 +/- 49 for 2-pentene, 2-methyl; 151 +/- 5 for (E)-2-hexene, 14.5 +/- 1.0 for 1,5-hexadiene and 20.7 +/- 3.1 for 1,5-hexadiene, 2-methyl. There is good agreement between the experimental and predicted values and the adjustable parameters of the SAR are shown to be insensitive to the inclusion of the new data. The use of the SAR in atmospheric chemical modelling is investigated. Ozonolysis and OH radical rate coefficients are estimated for each alkene and dialkene present in the MCM v3.1. The effects of error within predicted rate coefficients upon modelled concentrations of a number of key species, including O3, OH, HO2, NO and NO2 were rather small and is not in itself a major cause of uncertainty in modelled concentrations.

No abstract

A generic method has been developed for the analysis of polar compounds and oligomers in secondary organic aerosol (SOA) formed during atmospheric simulation chamber experiments. The technique has been successfully applied to SOA formed in a variety of systems, ranging from ozonolysis of biogenic volatile organic compounds to aromatic photooxidation. An example application of the method is described for the SOA produced from the reaction of ozone with cis-3-hexenyl acetate, an important biogenic precursor. A range of solvents were tested as extraction media, and water was found to yield the highest recovery. Extracts were analyzed using reversed-phase liquid chromatography coupled to ion trap mass spectrometry. In order to determine correct molecular weight assignments and increase sensitivity for less polar species, a series of low-concentration mobile-phase additives were used (NaCl, LiBr, NH4OH). Lithium bromide produced better fragmentation patterns, with more structural information than in the other cases with no reduction in sensitivity. The main reaction products identified in the particle-phase were 3-acetoxypropanal, 3-acetoxypropanoic acid, and 3-acetoxypropane peroxoic acid and a series of dimers and trimers up to 500 Da. Structural identification of oligomers indicates the presence of linear polyesters possibly formed via esterfication reactions or decomposition of peroxyhemiacetals.

Heteroatomic unsaturated volatile organic compounds (HUVOCs) are common trace components of the atmosphere, yet their diverse chemical behaviour presents difficulties for predicting their oxidation kinetics using structure-activity relationships (SARs). An existing SAR is adapted to help meet this challenge, enabling the prediction of ozonolysis rates with unprecedented accuracy. The new SAR index, x(H), correlates strongly with available literature measurements of ozonolysis rate coefficients (R(2) = 0.87), a database representing 110 species. It was found that capturing the inductive effect rather than the steric effect is of primary importance in predicting the reactivity of these species, which is to be anticipated since HUVOCs can possess a variety of functional groups with a range of electron-withdrawing and donating tendencies. New experimental measurements of ozonolysis rate coefficients were conducted for 1-penten-3-ol, 3-methyl; ethene, 1,1-dimethoxy; E-2-pentenoic acid; E-1,2-dichloroethene; Z-1,2-dichloroethene; trichloroethene; tetrachloroethene; 1-butene, 3-chloro and 2-chloropropene, and were determined to be 5.15 × 10(-18), 4.82 × 10(-16), 3.07 × 10(-18), 8.05 × 10(-20), 4.88 × 10(-21), 6.04 × 10(-22), 1.56 × 10(-24), 2.26 × 10(-18) and 1.13 × 10(-19) cm(3) molecule(-1) s(-1), respectively. The index of the inductive effect, i(H), is compared with other indices of the electron-withdrawing capacity of a substitution, notably the Taft σ* constants and the rate of reaction of a given species with the hydroxyl radical, both of which are expected to be unaffected by steric factors. i(H) correlates strongly in both cases and suggests a universal response by olefinic species towards electrophilic addition.