Troposphere alkene reactions

The generation of OH in 03-alkene reactions has important implications for tropospheric chemistry. Thus the 03-alkene reactions could be important free radical sources at dusk and during the night when pho-tolytic sources of OH are minimal (e.g., Paulson and Orlando, 1996 Bey et al., 1997 Paulson et al., 1998). For example, Paulson and Orlando (1996) predicted that 10-15% of the total radical production may be from 03 alkene reactions in a typical rural area in the southeastern United States. As seen in Fig. 6.6, this reaction is expected to be most important at night. 

In contrast to the water phase the HO radicals can have a much longer lifetime in gaseous media, i.e. up to 1 s for the OH and 60 s for the HO radical, respectively (Fabian, 1989). Despite the low concentration of OH radicals of about 10 molecules per cm in the sunlit troposphere (Ehhalt, 1999) they play an important role in controlling the removal of many organic natural and manmade compounds from the atmosphere (Eisele et al., 1997, Eisele and Bradshaw, 1993). Even in indoor environments, the formation of hydroxyl radicals is possible by ozone/alkene reactions (Atkinson et al., 1995). Steady-state indoor hydroxyl radical concentrations of about 6.7x10 ppb equivalent to 1.7x10 molecules cm were calculated at an ozone concentration of 20 ppb (Weschler and Shields, 1996). 

Alkenes are removed from the troposphere via reaction with 0(3P) atoms, OH radicals, N03 radicals, and 03. Of these, reaction with 0(3P) atoms is essentially negligible under atmospheric conditions. 

The results of LACTOZ have provided an extended kinetic data base for the following classes of reactions reactions of OH with VOCs, reactions of NO3 with VOCs and peroxy radicals, reactions of O3 with alkenes, reactions of peroxy radicals (self reactions, reaction with HO2, other RO2, NO, NO2), reactions of alkoxy radicals (reactions with O2, decomposition, isomerisation), thermal decomposition of peroxynitrates. Photolysis parameters (absorption cross-section, quantum yields) have been refined or obtained for the first time for species which photolyse in the troposphere. Significantly new mechanistic information has also been obtained for the oxidation of aromatic compounds and biogenic compounds (especially isoprene). These different data allow the rates of the processes involved to be modelled, especially the ozone production from the oxidation of hydrocarbons. The data from LACTOZ are summarised in the tables given in this report and have been used in evaluations of chemical data for atmospheric chemistry conducted by international evaluation groups of NASA and lUPAC. 

All carbonyl oxides proved to be highly photolabile, and on photolysis yield dioxiranes 3 or split off oxygen atoms to produce ketones. Oxygen atoms are also formed thermally from vibrationally excited 1. Thus, if the large exothermicity of the ozonolysis reaction is taken into account, 1 might be a source of O atoms and OH radicals in the troposphere. The role of dioxiranes has not yet been discussed in context with atmospheric chemistry, although the formation of these species in contrast to the isomeric carbonyl oxides - in ozone/alkene reactions has been unequivocally demonstrated [13]. 

To pare the list of VOC oxidations down to the most important processes, we can calculate the effective lifetimes of organics with respect to reactions with each of the oxidants listed in the previous section. Since these natural lifetimes are defined as r = 1 / [X], we also need to assume an average concentration for the oxidant, [X]. We can therefore take a typical organic from each of the major classes (alkane, alkene, aromatic, etc.) and compare the individual lifetimes for reaction with OH, 03, N03, etc. Those reactions having very long lifetimes are insignificant with respect to their contribution to tropospheric chemistry and hence can be ignored for the purposes of this discussion. 

The room temperature rate constants for the reactions of 03 with some alkenes are given in Table 6.9. While the values are many orders of magnitude smaller than those for the corresponding OH reactions, the fact that tropospheric ozone concentrations are so much larger makes these reactions a significant removal process for the alkenes. 

Tropospheric chemistry models have to take into account a significant number of chemical reactions required to simulate correctly tropospheric chemistry. In the global background marine troposphere, it seems reasonable to consider a simplified chemistry scheme based on O3/ NOx/ CH, and CO photochemical reactions. However, natural emissions of organic compounds from oceans (mainly alkenes and dimethyl sulphide-DMS) might significantly affect the marine boundary layer chemistry and in particular OH concentrations. Over continental areas both under clean and polluted conditions,... 

Reactions with ozone are competitive with the daytime OH radical reactions and the nighttime NO-, radial reactions as a tropospheric loss process for the alkenes. These reactions have been shown to proceed via initial O, addition to the olefinic double bond, followed by rapid decomposition of the resulting molozonide (Atkinson, 1990) ... 

As the product molecule AB becomes more complex, the value of k, decreases because the combination energy is distributed among more and more vibrational modes. The concentration of the third body, [M], is usually related directly to the pressure since in the atmosphere M is the sum of N2 and 02. The concentration of M at which the reaction rate behavior changes from third-order to second-order is lower the more complex the product molecule. Combination of two hydrogen atoms to form H2 is third-order all the way up to 104atm. On the other hand, addition of the OH radical to the alkene, 1-butene, C4H8, is second-order at all tropospheric pressures. 

Evidence for epoxide formation from NO3 reactions with alkenes has been reported (Hjorth, Schindler). These results show that oxirane formation occurs via a radical adduct, in which rotation about the C-C bond can occur before elimination of NO2. Reactions of NO3 with isoprene and 2-butene at low pressure and at low O2 concentrations gave mainly oxiranes, whereas in air at atmospheric pressure, oxirane yields were negligible. However, even at atmospheric pressures, the reaction of NO3 with 2,3-dimethyl-2-butene gave an oxirane yield of around 20 %. Thus, it is apparent that, at least with some alkenes, oxirane formation may be important under tropospheric conditions. 

Combination of the available rate constant data with estimates of the atmospheric concentrations of O3, NO3 and OH indicates that the reaction of NO3 with substituted alkenes and terpenes may be the dominant tropospheric oxidation process for these species. Formation of carbonyl nitrates in the reactions could be important in the long-range transport of odd nitrogen. 

The alkene yields measured in the experiments are shown in Table 1. The yields have been corrected for loss due to reaction with OH radicals. For all the alkanes investigated formation of alkenes was observed, but the yields were always well below 1 %. The results show that the reaction of alkyl radicals with O2 to form alkenes is not a significant process under normal tropospheric conditions. 

The fast reaction of NO3 radicals with alkenes was found to give substantial yields of bifunctional carbonyl-nitroxy products and the mechanism of their formation is expected to be relevant also for the conditions of the troposphere. Studies performed by other workers [12] have shown that such organic nitrates may contribute to the atmospheric transport of NOx by acting as reservoirs, releasing NOx by their degradation in the atmosphere, but bifiinctional nitrates may also be subject to fast removal by wet deposition. 

The objective of present research was to provide a better understanding of the chemical processes involved in production and loss of ozone in the troposphere. This was achieved by providing kinetic and mechanistic data for several reactions of peroxy radicals involved in the photo-oxidation of volatile organic compounds (VOC). Additional aims were to determine the product quantum yields in the photolysis of carbonyl compounds, and to investigate the mechanism in the ozonolysis of alkenes, especially in the presence of water vapour. 

In a second part of the work, a Structure-Activity Relationship has been developed for the addition of OH to (poly)alkenes. It was shown that the total rate constants can be expressed in very good approximation by a sum of partial, site-specific rate constants for addition to a given (double-bonded) carbon atom, the values of these partial rate constants depending solely on the stability-type of the ensuing p-hydroxy radical. The SAR is particularly useful in that it also allows the prediction of the detailed primary product distributions of (poly)alkene + OH reactions. Therefore, the SAR is a powerful tool in the modeling of the tropospheric OH-initiated oxidation of biogenic VOC. 

Ozone plays a major role in the degradation of unsaturated VOCs in the troposphere, especially during night-time. The rate constants of the ozonolysis of a variety of alkenes have been reported [1]. However, in most instances the fate of the primary products of the ozonolysis is unknown, although the secondary reaction products are of crucial importance for the overall understanding of the alkene/ozone chemistry. The classical Criegee mechanism of the ozonolysis reaction involves the primary ozonide (POZ, 1,2,3-trioxolane), which cleaves to the Criegee intermediate (carbonyl O oxide) and a carbonyl compound [2, 3]. The secondary ozonide (SOZ, 1,2,4-trioxolane) is formed from these components in a [l,3]-dipolar cycloaddition reaction. 

The Criegee intermediate has been claimed to be of importance in tropospheric chemistry [4] but never been observed by direct spectroscopic methods in the ozonolysis reaction. The aims of our research were therefore (i) to provide spectroscopic (IR, UVA is) data of a variety of substituted carbonyl O oxides, (ii) to develop a theoretical model which allows the prediction of the spectra of carbonyl O oxides which are not accessible by laboratory studies, but might be of importance to tropospheric chemistry, and (iii) to elucidate the mechanism of the ozonolysis of alkenes and investigate the role of carbonyl O oxides in these reactions. 

The reaction of ozone and alkenes is sufficiently fast that it can compete with other removal processes and provide sinks for both ozone and alkenes in the troposphere. While kinetic data for a series of alkene/ozone reactions have been reported, not much is known about details of the reaction mechanisms, the role that carbonyl O oxides play, and the role that free radicals play in these processes. Our laboratory experiments provide the spectroscopic data (both infrared and UV/visible) that are important for the spectroscopic identification of Criegee intermediates in the troposphere. In addition, we were able to characterize secondary partially oxidized products (aldehydes, peroxides etc.) that are produced during the gas-phase ozonolysis. These products might lead to a net increase of ozone, if oxygen atoms are formed during their decomposition. 

The oxidation of alkanes, alkenes and simple aromatics at 293 K under NOx rich tropospheric conditions has been studied using laser pulse initiation combined with cw laser long path absorption/LIF for the detection of OH and NO2. In the case of aliphatic hydrocarbons the absolute yield and the kinetics of the formation of these products have been found to be sensitive indicators for the reaction behaviour of the oxy radicals RO. In combination with mechanistic simulations rate constants for individual reactions as well as branching ratios have been derived, which permit the evaluation of the compound specific NO/NO2 conversion factors (NOCON - factors) for the first oxidation steps. In the case of benzene and toluene oxidation the results indicate that reaction of the primary formed X cyclohexa-dienyl radical (X = Cl, OH) with O2 is the dominant pathway, although the rate coefficients were found to be lower than 2 x 10" cmVs. 

---