RESUMO
The reactions of SO3 with H, O, and OH radicals have been investigated by ab initio calculations. For the SO3 + H reaction (1), the lowest energy pathway involves initial formation of HSO3 and rearrangement to HOSO2, followed by dissociation to OH + SO2. The reaction is fast, with k(1) = 8.4 x 10(9)T(1.22) exp(-13.9 kJ mol(-1)/RT) cm(3) mol(-1) s(-1) (700-2000 K). The SO3 + O --> SO2 + O2 reaction (2) may proceed on both the triplet and singlet surfaces, but due to a high barrier the reaction is predicted to be slow. The rate constant can be described as k(2) = 2.8 x 10(4)T(2.57) exp(-122.3 kJ mol(-1)/RT) cm(3) mol(-1) s(-1) for T > 1000 K. The SO3 + OH reaction to form SO2 + HO2 (3) proceeds by direct abstraction but is comparatively slow, with k(3) = 4.8 x 10(4)T(2.46) exp(-114.1 kJ mol(-) 1/RT) cm(3) mol(-1) s(-1) (800-2000 K). The revised rate constants and detailed reaction mechanism are consistent with experimental data from batch reactors, flow reactors, and laminar flames on oxidation of SO2 to SO3. The SO3 + O reaction is found to be insignificant during most conditions of interest; even in lean flames, SO3 + H is the major consumption reaction for SO3.
RESUMO
The thermal dissociation of SO3 has been studied for the first time in the 1000-1400 K range. The experiments were conducted in a laminar flow reactor at atmospheric pressure, with nitrogen as the bath gas. On the basis of the flow reactor data, a rate constant for SO3 + N2 --> SO2 + O + N2 (R1b) of 5.7 x 10(17) exp(-40000/T) cm3/(mol s) is derived for the temperature range 1273-1348 K. The estimated uncertainty is a factor of 2. The rate constant corresponds to a value of the reverse reaction of k1 approximately 1.8 x 10(15) cm6 mol(-2) s(-1). The reaction is in the fall-off region under the investigated conditions. The temperature and pressure dependence of SO2 + O (+N2) was estimated from the extrapolation of low temperature results for the reaction, together with an estimated broadening parameter and the high-pressure limit determined recently by Naidoo, Goumri, and Marshall (Proc. Combust. Inst. 2005, 30, 1219-1225). The theoretical rate constant is in good agreement with the experimental results. The improved accuracy in k(1) allows a reassessment of the rate constant for SO3 + O --> SO2 + O2 (R2) based on the data of Smith, Tseregounis, and Wang (Int. J. Chem. Kinet. 1982, 14, 679-697), who conducted experiments on a low-pressure CO/O2/Ar flame doped with SO2. At the location in the flame where the net SO3 formation rate is zero, k2 = k1[SO2][M]/[SO3]. A value of 6.9 x 10(10) cm3 mol(-1) s(-1) is obtained for k2 at 1269 K with an uncertainty a factor of 3. A recommended rate constant k2 = 7.8 x 10(11) exp(-3065/T) cm3 mol(-1) s(-1) is consistent with other flame results as well as the present flow reactor data.
