Oxidation of Carboxylic Acids Regenerates Hydroxyl Radicals in the

J. Phys. Chem. A 2010, 114, 6861–6869

6861

Oxidation of Carboxylic Acids Regenerates Hydroxyl Radicals in the Unpolluted and Nighttime Troposphere Gabriel da Silva* Department of Chemical and Biomolecular Engineering, The UniVersity of Melbourne. ParkVille 3010, Victoria, Australia ReceiVed: December 6, 2009; ReVised Manuscript ReceiVed: May 26, 2010

The hydroxyl radical (OH) controls the removal of organic compounds from the troposphere. Atmospheric chemistry models significantly under-predict OH levels in unpolluted environments, implying that they are regenerated via some unknown mechanism(s). This work uses computational chemistry to demonstrate that the photochemical oxidation of alkyl carboxylic acids can efficiently regenerate the hydroxyl radical via unimolecular decomposition of R-carboxyalkylperoxy radicals. For acetic acid and propanoic acid the proposed mechanism is predicted to dominate in the unpolluted lower troposphere, and it may also operate to some extent in the mid to upper troposphere. Alkyl carboxylic acids are also predicted to act as a new source of nighttime OH throughout the planetary boundary layer, where OH levels are also under-predicted. The thermodynamic requirements for reactions of this class are discussed, and some candidate OH-reforming molecules particularly relevant to aromatic photooxidation are identified. Adopting a broader perspective, the R-carboxyalkyl radical precursors that react with O2 to form the unstable R-carboxyalkylperoxy type radicals are also expected to form during combustion, in the interstellar medium, and from the γ-irradiation of glycine and related amino acids, and the potential importance of this new chemistry in these environments is discussed. Master equation simulations suggest that R-carboxyalkyl + O2 reactions provide a prompt OH source during the autoignition and combustion of biodiesel and other oxygenated biofuels, where carboxylic acids are formed as early stage oxidation products. Ketene combustion is also thought to proceed via these OH-reforming R-carboxyalkyl radicals. The in vivo formation of R-carboxyalkylperoxy radicals followed by oxidation to the highly reactive OH radical may induce oxidative stress in the human body, in a process initiated by γ-rays. Finally, the reaction of ketenes with OH to form R-carboxyalkyl radicals, followed by addition of NH2 or related species, is suggested as a new extraterrestrial pathway to amino acids. Introduction The hydroxyl radical (OH) controls the removal of volatile organic compounds (VOCs) and pollutants from the Earth’s lower atmosphere. The OH radical is highly reactive and shortlived and is primarily generated by UV radiation acting on ozone. The reaction of OH with VOCs produces peroxy radicals (RO2), either via a rapid addition mechanism (for unsaturated compounds) or slower abstraction reactions (for saturated hydrocarbons). In even moderately polluted environments these peroxy radicals react with NO to form alkoxyl radicals (+NO2), which typically generate HO2 along with oxygenated hydrocarbons. NO also reacts with HO2, forming OH (+NO2) in a process that helps polluted environments maintain high oxidative capacities. VOC oxidation is further aided by the reaction of HO2 with O3, which forms OH + 2O2. The NO2 produced in VOC oxidation is photolyzed in a process that forms ozone and regenerates NO. Thus, although VOCs are efficiently oxidized in polluted environments, the associated chemistry is responsible for elevated levels of tropospheric ozone, which is an environmental and health concern. The chemistry and HOx (OH + HO2) budget of the polluted, urban atmosphere is at present well understood1 and is described quite accurately by modern atmospheric chemistry models. It has emerged over the past decade that the chemistry of the unpolluted troposphere is not so well understood. This is * E-mail: [email protected].

particularly the case for forested environments in the daytime, where there are large biogenic sources of isoprene and other reactive, unsaturated VOCs. The standard interpretation of VOC oxidation chemistry in the unpolluted forest boundary layer is that organic peroxy radicals react predominantly with HO2 instead of NO, producing organic hydroperoxides.1 Models predict that biogenic VOC (BVOC) emissions from forests place a large burden on the troposphere’s oxidative capacity, because of the lack of OH regeneration that would otherwise arise from NO and O3 reacting with HO2. Over the past decade, however, several studies have suggested that unpolluted, forested environments somehow maintain significantly higher than predicted levels of OH.2-7 Airborne measurements over the Amazon rainforest in the Guyanas have demonstrated elevated levels of OH in the presence of low NO (ca. 20 ppt), and it was proposed that OH was somehow regenerated in the isoprene oxidation mechanism.6 Following this work, field trials in the Pearl River Delta of China observed OH levels that were several times larger than predicted when NO dropped below around 1 ppb, and this was attributed to a missing OH source of around 30 ppb/hour.7 A significant fraction of this missing OH may be explained by physical deficiencies in current atmospheric chemistry models,8 although it is likely that some new chemistry is still required to explain oxidative capacitates across a variety of unpolluted environments. The underpredicted oxidative capacity of the troposphere in the relative absence of NOx has major implications for our ability to predict global pollutant lifetimes and

10.1021/jp101279p  2010 American Chemical Society Published on Web 06/09/2010

6862

J. Phys. Chem. A, Vol. 114, No. 25, 2010

the effect of anthropogenic emissions on the Earth’s atmosphere. For instance, by underestimating the oxidative capacity of the pristine versus urban atmosphere we are somewhat masking the global warming (or cooling) potential of industrial emissions including VOCs, SOx, and NOx.9 There is at present a great need to describe reactions of organic peroxy radicals that can recycle OH in the atmosphere, and several candidate processes have been identified. It has been known for some time, for instance, that the acetylperoxy radical reacts with HO2 to yield OH,10 although it appears that the R-carbonyl functionality is required to achieve significant OH generation.10,11 Another novel process is the formation of dihydroxyepoxides, with OH regeneration, in the reaction of isoprene hydroxyhydroperoxides with OH,12 although this overall process does result in the net consumption of one OH molecule in converting isoprene into the epoxide. Recently, it has been shown that the β-hydroxyperoxy radicals formed in isoprene oxidation can undergo a unimolecular (thermal) decomposition reaction, forming OH along with the known lowNOx isoprene oxidation products HCHO + methacrolein (MACR) or methyl vinyl ketone (MVK).13,14 These reactions require activation barriers of around 20-22 kcal mol-1, resulting in peroxy radical lifetimes on the order of minutes to hours;13,14 subsequently, this chemistry is only expected to be of significance in remote pristine environments such as the ocean boundary layer, and perhaps also at nighttime (vide infra). No single process has yet been described that can account for the stability of OH in the unpolluted atmosphere, although given the emerging variety of reactions that can achieve this end it appears likely that some combination of new chemistry will be required. This study demonstrates that OH radicals are regenerated in the photochemical oxidation of n-alkyl carboxylic acids. The mechanism proposed here is predicted to dominate for NO and HO2 levels found in the low to mid regions of the unpolluted troposphere, whereas a chemically activated mechanism may result in prompt OH formation in the stratosphere. This novel process involves the rapid unimolecular decomposition of R-carboxyalkylperoxy radicals, which are produced by O2 addition to R-carboxyalkyl radicals. These R-carboxyalkyl radicals are known to form under tropospheric conditions from the reaction of OH with n-alkyl carboxylic acids, and as is discussed later in this manuscript are also important intermediates in combustion, biochemistry, and astrochemistry. Carboxylic acids are a major constituent of the troposphere and are responsible for most of the free acidity in water droplets at remote environments. They are emitted by biogenic and anthropogenic sources and are also VOC oxidation products, although their sources and sinks are not presently well accounted for. Alkyl carboxylic acids, especially acetic acid and propanoic acid, are found widely throughout the troposphere, generally at levels in the range of 0.1-2 ppb.15 Acetic acid has been detected at levels of around 1 ppb in the Amazonian forest boundary layer,16,17 and at up to 2 ppb in the upper troposphere.18 The reaction of acetic acid (CH3C(O)OH) with OH is relatively wellstudied and is known to proceed predominantly via abstraction of the hydroxyl H atom, forming CH3C(O)O.19,20 Abstraction of a weaker methyl H atom to yield the R-carboxymethyl radical, CH2C(O)OH, is also important, constituting around 25% of the reaction flux at ambient temperatures.19 The R-carboxymethyl radical will associate with O2 to form the R-carboxymethylperoxy radical, CH2(O2)C(O)OH, which is a focus of this study. Larger carboxylic acids react with OH to predominantly form R-carboxyalkyl radicals of the form RCHC(O)OH,21 due to a

da Silva lowering (ca. 5 kcal mol-1) of the C-H bond energies at secondary versus primary carbon atoms with β-carbonyl functionalities.21,22 Accordingly, unimolecular reactions of the R-carboxyalkylperoxy radicals formed from propanoic acid and the larger n-alkyl carboxylic acids assume increased significance over the corresponding process in acetic acid. This study also focuses on the R-carboxyethylperoxy radical that forms in the photochemical oxidation of propanoic acid, as a model for oxidation of these larger carboxylic acids. Computational Methods All reported energies are obtained with the composite G3SX method,23 which provides high accuracy for both thermochemistry and barrier heights. The G3SX model chemistry uses B3LYP/6-31G(2df,p) optimized structures, vibrational frequencies, and scaled zero point energies, along with higher-level single point wave function theory energies from HF through QCISD(T) theory, with empirical scaling corrections. The wavefunction theory energies are combined in such a way as to approximate the QCISD(T) energy with a large basis set. G3SX level optimized structures, vibrational frequencies, and energies are reported in the Supporting Information for all species involved in rate constant calculations. The G3SX method reproduces all experimental energies in the G3/99 test set with a mean absolute deviation of 0.95 kcal mol-1.23 For the more-representative subsets of substituted hydrocarbon and radical enthalpies of formation, this uncertainty is reduced to 0.72 and 0.67 kcal mol-1, respectively.23 With the DBH24/08 test set of barrier heights the G3SX method provides a mean absolute deviation of only 0.57 kcal mol-1,24 outperforming numerous more computationally expensive methods. In this study, critical transition states are also evaluated with the alternate (and arguably less-accurate)24 CBS-Q method,25 primarily to test the validity of B3LYP-optimized structures. Thermochemical properties are obtained for the most stable conformation of each molecule using the rigid-rotor-harmonicoscillator approximation with B3LYP/6-31G(2df,p) vibrational frequencies and moments of inertia, along with G3SX 298 K enthalpies. Standard enthalpies of formation (∆fH°298) are calculated from atomization energies, following the procedure reported recently,26 which with the G3SX method reproduces hydrocarbon enthalpies of formation to within experimental accuracy. Rate constants are calculated as a function of temperature from canonical transition state theory, with quantum mechanical tunnelling included as described previously.27 Electronic structure calculations were performed using Gaussian 09,28 with ChemRate29 used for thermochemical and transition state theory calculations. Energy surfaces are constructed according to the theory of intersecting harmonic parabolic wells30 and are semiquantitative. Results and Discussion As noted in the introduction, this study reveals that R-carboxyalkylperoxy radicals formed in carboxylic acid oxidation can undergo thermal decomposition in the atmosphere. The following section develops the mechanism of this process for the R-carboxymethylperoxy and R-carboxyethylperoxy radicals. Thermochemical properties and structures are provided for key species and transition states. Following this, transition state theory is used to obtain rate constants for these decomposition reactions, and their importance relative to conventional bimolecular processes is evaluated. This leads to a discussion of the importance of this new chemistry throughout the atmosphere,

OH Reforming in the Troposphere

J. Phys. Chem. A, Vol. 114, No. 25, 2010 6863

Figure 1. Energy surface for decomposition of the R-carboxymethylperoxy radical. Enthalpies are 298 K G3SX values.

TABLE 1: Thermochemical Properties for the r-Carboxymethyl and r-Carboxyethyl Radicals, and the Species Involved in the Reactions of Their Peroxy Radicalsa CH2C(O)OH CH3CHC(O)OH CH2(OO)C(O)OH CH3CH(OO)C(O)OH CH2OOH CH3CHOOH CO2 a

∆fH°298

S°298

Cp 300

Cp 400

Cp 500

Cp 600

Cp 800

Cp 1000

Cp 1500

Cp 2000

-56.6 -66.6 -81.3 -90.6 15.4 6.0 -95.6

68.157 76.071 80.536 88.143 65.980 74.712 51.040

16.688 21.170 21.733 27.454 14.952 19.817 8.796

19.888 25.854 26.086 33.231 17.156 23.512 9.770

22.428 29.886 29.719 38.182 19.014 26.824 10.554

24.431 33.218 32.630 42.234 20.524 29.615 11.194

27.370 38.283 36.851 48.269 22.813 33.945 12.167

29.461 41.933 39.726 52.497 24.516 37.146 12.846

32.747 47.561 43.925 58.815 27.323 42.214 13.794

34.513 50.484 147.965 62.016 28.883 44.896 14.229

Enthalpies of formation (∆fH°298) in kcal mol-1. Entropies (S°298) and heat capacities (Cp) in cal mol-1 K-1.

Figure 2. Optimized structure for the R-carboxymethylperoxy radical (left) and its intramolecular hydrogen shift transition state (right). Calculated at the B3LYP/6-31G(2df,p) level of theory.

under both day- and night-time conditions. In the section that follows this, the potential for prompt OH formation in the chemically activated reaction of R-carboxyalkyl radicals with O2 is determined, followed by a discussion of the importance of this mechanism in combustion systems. Finally, some attention is paid to the significance of R-carboxyalkyl radical chemistry in biochemical systems and in deep space. r-Carboxyalkylperoxy Radical Decomposition Mechanism. An energy surface for unimolecular reaction of the R-carboxymethylperoxy radical formed in the OH-initiated oxidation of acetic acid is presented in Figure 1, and the corresponding R-carboxyethylperoxy radical energy diagram for propanoic acid is available as Supporting Information. Thermochemical properties for all minima in both mechanisms are listed in Table 1, along with properties for the precursor R-carboxyalkyl radicals (transition state thermochemistry is provided as Supporting Information). From Figure 1 we see that both peroxy radicals can undergo a relatively low-energy intramolecular hydrogen shift from the carboxylic OH group to the peroxy radical moiety, with concerted C-C dissociation. In the case of acetic acid the barrier is 18.5 kcal mol-1, whereas

for propanoic acid the barrier is 16.5 kcal mol-1. At the CBS-Q level these respective barriers are 18.5 and 15.8 kcal mol-1, which are in excellent agreement with the G3SX results. The CBS-Q method is somewhat less-accurate than G3SX for both thermochemical properties and barrier heights, but importantly it does not feature B3LYP DFT structures and utilizes a different procedure to approximate the QCISD(T) energy approaching the complete basis set limit. Accordingly, the 1,5-hydrogen shift channel identified for the R-carboxyalkylperoxy radicals is unlikely to be an artifact of the theoretical protocol. The 1,5-hydrogen shift transition state structure for the R-carboxymethylperoxy radical is shown in Figure 2, compared to the optimized structure of the peroxy radical. Reaction proceeds via a six-membered ring transition state, in which C-CO2 dissociation occurs concurrently with H atom transfer. Intrinsic reaction coordinate scans confirm that the products are CO2 plus a hydroperoxyalkyl radical, and for both peroxy radicals considered here this first step is close to thermoneutral. In the transition state the forming O-H bond is almost complete, indicating that the H shift occurs early in the reaction coordinate, with C-C dissociation dominating in the later stages. Compared

6864

J. Phys. Chem. A, Vol. 114, No. 25, 2010

to the analogous process in the isoprene β-hydroxyperoxy radicals (which is around 20 kcal mol-1 endothermic),13 this intramolecular hydrogen shift is considerably more favored thermodynamically, which results from the stability of CO2. Formation of CO2 provides an added driving force for these reactions, which helps to reduce the thermodynamic component of the activation energy. Additionally, the added thermodynamic stability of the products means that the intrinsic barrier available for quantum mechanical tunnelling is actually larger, and this should further decrease lifetimes for these peroxy radicals. Both processes are, however, characterized by the formation of multiple thermodynamically favored CdO bonds, which enables them to overcome the penalty incurred by reformation of the notoriously high-energy OH radical (much of which is initially lost as thermal energy following the exothermic OH addition or abstraction reactions).31 When one considers the relatively high barrier required for decomposition of the β-hydroxyethylperoxy radical to HCHO + CH3CHO + OH,32 it appears that the formation of two new CdO bonds alone is not sufficient to overcome this energy penalty associated with OH production. In the case of isoprene, the ultimate formation of conjugated CdC-CdO type structures in MVK and MACR is required,13,14 whereas in carboxylic acid oxidation the regeneration of OH appears to be achievable only due to the added energy released by OdCdO formation (relative to simple aldehydes in both cases). The new insights into the energetics of OH reformation outlined above may be useful in identifying further like processes. This analysis indicates that in order for VOC oxidation to regenerate OH in the atmosphere, where little thermal energy is available to overcome reaction barriers, the formation of coproducts which gain thermodynamic stability greater than that associated with two rearrangements of the form C-OsO/H to CdO is required. This is in addition to the obvious need for a relatively low intrinsic activation energy. One possible mechanism involves the formation of multiple conjugated aldehydes/ketones, which could be achieved from the peroxy radicals formed by consecutive OH, O2 addition at the 2 and 3 positions in dicarbonyls of the form CH(O)CHd CHCH(O). These dicarbonyls are known to form in VOC oxidation, particularly from aromatic hydrocarbons where they are first-generation products.33 Interestingly, OH regeneration is observed experimentally in the reaction of toluene (and other aromatics) with OH.33 The hydroperoxyalkyl radical products seen here in R-carboxyalkylperoxy radical decomposition fall apart to an aldehyde + OH with barriers that lie just below the hydroperoxyalkyl radical energies at the G3SX level, in a considerably exothermic process. It is unclear if these radicals actually exist as stable ground-state species (as suggested by B3LYP structures), although it is apparent that one would not be able to isolate them at room temperature, and that they will promptly dissociate in the present systems given the considerable energy that they are formed with. Similar hydroperoxyalkyl radicals have been reported in previous studies, where they are also found to be transient intermediates.34 Accordingly, in the rate constant calculations presented here it is assumed that the intramolecular hydrogen shift proceeds directly to the three dissociated fragments. Interestingly, a conventional hydrogen shift reaction with significantly larger barrier (26.4 kcal mol-1 at the QCISD(T)/6-31++G(2df,2p) level), producing the hydroperoxyacetoxyl radical (CH2(OOH)C(O)O) has been previously reported for the R-carboxymethylperoxy radical,35 although the required transition state structure could not be located here.

da Silva TABLE 2: Lifetimes (in seconds) for the Acetic Acid and Propanoic Acid r-Carboxyalkylperoxy Radicals (CH2(OO)C(O)OH and CH3CH(OO)C(O)OH, Respectively) lifetime (s) temperature (K)

CH2(OO)C(O)OH

CH3CH(OO)C(O)OH

220 230 240 250 260 270 280 290 300 310 320

416 000 95 000 23 300 6160 1750 536 176 61.3 22.7 8.93 3.69

2200 654 204 66.7 23.1 8.45 3.26 1.32 0.565 0.252 0.117

Decomposition Kinetics. Rate constants have been calculated for thermal decomposition of the R-carboxymethylperoxy and R-carbxoyethylperoxy radicals, from canonical transition state theory. Lifetimes for these two radicals are listed in Table 2, at temperatures between 220 and 320 K (from calculated rate constants listed in the Supporting Information). So as to model these reactions across a wider temperature range, fits to the threeparameter form of the Arrhenius equation have also been obtained, between 300 and 2000 K. For decomposition of the CH2(OO)C(O)OH radical to CO2 + OH + HCHO, the expression k ) 3.45 × 108T1.051 exp(-8647/T) s-1 is obtained, whereas for the corresponding reaction in CH3CH(OO)C(O)OH we get k ) 1.44 × 108T1.164 exp(-7472/T) s-1. The main process competing with thermal decomposition of the R-carboxyalkylperoxy radicals will be O atom abstraction by NO. The OH yield from the R-carboxyalkylperoxy radicals as a function of NO concentration has been calculated at 300 K and is plotted in Figure 3. Here, it is conservatively assumed that the rate constant for both peroxy radicals reacting with NO is 1 × 10-11 cm3 molecule-1 s-1. For conditions typical of the unpolluted forest boundary layer,36 it is predicted that R-carboxyalkylperoxy radical lifetimes are short (tens of seconds and below) and that thermal decomposition is their dominant fate. For acetic acid the thermal decomposition mechanism begins to dominate for NO levels below around 180 ppt, whereas for propanoic acid this mechanism dominates at up to around 7 ppb (both at 300 K). This result suggests that OH regeneration in the oxidation of n-alkyl carboxylic acids should be a major reaction pathway in unpolluted regions of the planetary boundary layer, where NO levels are typically below 100 ppt (and often on the order of 20-40 ppt). Even if an uncertainty of (1.5 kcal mol-1 is assumed in the barrier heights, thermal decomposition remains the dominant fate of the R-carboxymethylperoxy radical at NO levels of ca. 20 ppt and below (700 ppt for R-carboxyethylperoxy). At low NO levels R-carboxyalkylperoxy radicals may react with HO2 to form a hydroperoxy carboxylic acid; thermal decomposition is still, however, expected to dominate, given that HO2 concentrations above pristine forested areas are generally around 50 ppt and below.37 Atmospheric Importance. Calculations presented above provide strong evidence for some degree of OH regeneration in the oxidation of n-alkyl carboxylic acids within the planetary boundary layer, provided sufficiently low levels of NO. Carboxylic acids are also found to be a major constituent of the upper regions of the troposphere. Temperatures in the tropopause can drop to ca. 220 K, with temperatures of around 260 K encountered at altitudes of several kilometres. Although peroxy radical decomposition is unlikely to significantly impact chem-

OH Reforming in the Troposphere

J. Phys. Chem. A, Vol. 114, No. 25, 2010 6865

Figure 3. Yield of OH from the R-carboxyalkylperoxy radicals formed in the photochemical oxidation of acetic acid and propanoic acid at 300 K, as a function of NO concentration.

SCHEME 1: Proposed Mechanism for the Photochemical Oxidation of Acetic Acid in Low-NOx Environments

istry at the tropopause, it may be of some importance in the mid- to upper troposphere, particularly for propanoic acid, where the R-carboxyethylperoxy radical lifetime is predicted to be under 10 min at 230 K. Further work is required to characterize the potential importance of R-carboxyalkylperoxy radical decomposition in the tropopause, although for the low- to midtroposphere it appears to be the dominant reaction pathway, outside of major pollution events. In the following section it is also revealed that prompt OH regeneration in the R-carboxyalkyl radical + O2 reactions may take place to a significant extent in the stratosphere, now via a chemically activated reaction mechanism. The proposed mechanism for low-NOx oxidation of acetic acid in the low- to mid-troposphere is depicted in Scheme 1. The main reaction channel is abstraction from the carboxylic OH site, which ultimately leads to CH3 + CO2. The CH3 radical is converted to HCHO + HO2 in the troposphere, either by reaction with HO2 (forming CH3OOH) or NO (forming CH3O). Photolysis of CH3OOH regenerates OH, although this hydroperoxide can also further consume OH by reaction back to CH3O2 (+H2O). As revealed here, abstraction of a methyl H atom in acetic acid leads to HCHO + OH + CO2, thus ensuring OH regeneration while avoiding the production of HO2. Because of the abundance of acetic acid and other carboxylic acids throughout the troposphere, this chemistry is expected to significantly impact the HO2 to OH ratio. Furthermore, these reactions are of increased importance for propanoic acid and larger carboxylic acids, due to the shorter peroxy radical lifetimes and larger rate constants for H atom abstraction from the secondary carbon atoms. This is expected to shift the HOx composition further away from HO2 toward OH.

Although this study has focused on saturated alkyl carboxylic acids, the mechanism described here may also play a role in the photochemical oxidation of unsaturated vinyl carboxylic acids. Being unsaturated organics, it is expected that their reactions with OH will be faster than for saturated alkyl carboxylic acids, making them relatively short-lived species. Acrylic acid, for example, should react with OH via addition to the terminal methylene site, forming the CH2(OH)CHC(O)OH radical. Subsequent reaction of this species with O2 will form the R-carboxyperoxy radical CH2(OH)CH(O2)C(O)OH. The barrier for thermal decomposition of this peroxy radical is calculated as 16.5 kcal mol-1 (at the G3SX level) and this mechanism is therefore expected to become significant at NO levels similar to those predicted for the R-carboxyethylperoxy radical (calculated decomposition rate constants are listed in the Supporting Information). This mechanism is also expected to play a role in the oxidation of other unsaturated carboxylic acids, such as the methyloxobutenoic acids formed in isoprene photooxidation.38 The recent discovery of thermally unstable peroxy radicals that can regenerate OH radicals in the troposphere may have implications for OH reactivity in forested environments with large biogenic emission sources. The OH reactivity, with units of s-1, is equivalent to the inverse of the OH lifetime and reflects the abundance of all compounds that react with OH and their rate constants in these reactions. Measurements of OH reactivity in a Michigan forest found that a large component of OH reactivity could not be accounted for by the measured VOCs and their oxidation products.39 Interestingly, the missing component of the OH reactivity increased with increasing temperature. It was proposed that this missing reactivity was due to

6866

J. Phys. Chem. A, Vol. 114, No. 25, 2010

the temperature-dependent emission of unaccounted-for BVOCs.39 Recent modeling work of VOC oxidation above a section of the Amazon rainforest has identified that oxygenated VOC (OVOC) production is significantly under-predicted,40 and it was suggested that missing OVOCs could account for much of the missing OH reactivity in other forests. As the photochemical age of an airmass increases, the OH burden is shifted from BVOCs like isoprene toward OVOCs like aldehydes and ketones. Because OH-initiated oxidation of BVOCs can increase the number of reactive compounds present (e.g., isoprene f HCHO + MVK/MACR), OH reactivity can increase with photochemical age, which is a function of time and OH concentration. If OVOCs are predominantly responsible for the missing OH reactivity measured by di Carlo et al.,39 it is not clear how a temperature dependence would arise. A proposal put forward here is that this temperature-dependent missing OH reactivity could be a result of OH-regenerating peroxy radical decomposition. By increasing OH levels, these reactions increase the photochemical age and therefore OH reactivity of an air mass. The missing OH reactivity component measured by di Carlo39 was found to follow an inverse-log plot, equivalent to an Arrhenius plot, where their reported slope corresponds to an effective activation energy of 19 kcal mol-1. This value is similar to that reported for the thermal decomposition of isoprene β-hydroxyperoxy13,14 and carboxylic acid R-carboxyperoxy radicals, and it is at around the level required for a unimolecular reaction to become important in the daytime unpolluted boundary layer. The observations of di Carlo et al.39 may therefore be an indication that OH-regenerating decomposition of peroxy radicals (whether they are from isoprene, carboxylic acids, or other unidentified precursors) is responsible for a significant component of missing OH reactivity, in which case they would also be recycling significant amounts of OH. Finally, some discussion is warranted on the potential importance of R-carboxyalkyl radical + O2 reactions to nighttime chemistry. In the nighttime troposphere, when photochemistry is absent, OH levels are found to be significantly below those encountered in the daytime. However, it is now wellaccepted that OH is still present in appreciable quantities throughout the remote41 and urban42 planetary boundary layer at night, although the sources are not well quantified. This study suggests that the oxidation of carboxylic acids provides a new nighttime OH source. Oxidation of carboxylic acids at night will be initiated by the low levels of OH that are present, whereas R-carboxyalkyl radical formation may also occur via reaction with NO3 and other radicals. Organic peroxy radicals become relatively long-lived at night, due to depressed HO2 levels and the lack of NO2 photolysis, and as such thermal decomposition increases in significance relative to bimolecular reactions with HO2 and NO. Furthermore, this chemistry is able to play a role throughout the planetary boundary layer (even in polluted areas), due to the near-quantitative conversion of NO to NO2. When carboxylic acid oxidation via the R-carboxyalkyl radical pathway is initiated by OH, recycling of this initial OH radical over the course of the night can help maintain oxidative capacities, as the process is HOx neutral. If this process is initiated by another radical source, such as NO3, then it becomes OH generating. Similar reactions are likely to be occurring with isoprene oxidation, where β-hydroxyperoxy radicals decompose to OH + HCHO + MACR/MVK.13,14 In this case, however, H atom abstraction by NO3 (for example) can not produce the required isoprene-OH radical precursors, and this process can therefore only act to maintain OH levels.

da Silva r-Carboxyalkyl Radical + O2 Kinetics. In this section, the kinetics and products of O2 addition to the R-carboxymethyl and R-carboxyethyl radicals are considered. Standard heats of formation for the CH2C(O)OH and CH3CHC(O)OH free radicals are calculated to be -56.6 and -66.6 kcal mol-1, respectively. This makes O2 addition to these resonantly stabilized R-carboxyalkyl radicals exothermic by 24.7 and 24.0 kcal mol-1, respectively. Peroxy radicals formed with around this amount of excess vibrational energy typically undergo collisional deactivation at atmospheric temperatures and pressures, unless further reaction pathways are available with barriers well below that of the entrance channel.43,44 Accordingly, in the troposphere the only process that needs to be considered for R-carboxyalkyl radical + O2 reactions is formation of the corresponding R-carboxyalkylperoxy radical, with subsequent bimolecular and unimolecular reactions. If we wish to include this new chemistry in kinetic models for combustion, however, we need to understand the reaction products over a wider range of temperature and pressure conditions. Although this is not the focus of this study, a preliminary analysis of the R-carboxyalkyl + O2 reaction kinetics is provided here. Master equation simulations have been performed for the two R-carboxyalkyl + O2 reaction processes, using QRRK theory for k(E).45 Densities of states for the activated peroxy radicals are described using a reduced set of three frequencies,a obtained from fits of the thermochemical properties reported in this study. This approach is useful here as it allows for the use of estimated rate constant parameters for O2 addition/dissociation, eliminating the need for variational transition state theory calculations. Collisional energy transfer is modeled with a 〈∆Edown〉 value of 1000 cm-1, where argon is the collider. Lennard-Jones parameters for the C2H3O4 radical are 3.5 Å and 400 K, whereas for the C3H4O5 radical they are 4.5 Å and 500 K. An empirical estimate of 10-11 cm3 molecule-1 s-1 is used for both O2 association rate constants, which assumes that they are barrierless. High-pressure limit rate constant expressions for reverse dissociation of the peroxy radicals are estimated as 5 × 1012 exp(-12430/T) and 5 × 1012 exp(-11930/T) s-1 for the respective R-carboxymethyl and R-carboxyethyl radicals. Decomposition of these peroxy radicals to CO2 + OH + the corresponding aldehyde is described using the fitted rate constant expressions reported above. Rate constants in the chemically activated R-carboxyalkyl radical + O2 reaction mechanism, as a function of temperature (300-2000 K) at 10 atm are shown in Figure 4. A similar figure pertaining to the R-carboxymethyl radical + O2 reaction is provided as Supporting Information. Considered pathways are collisional deactivation of the peroxy radical, reverse reaction to the R-carboxyalkyl radical + O2, or forward reaction to OH plus other products. For relevant combustion temperatures (>800 K) both R-hydroxyalkyl + O2 processes proceed almost entirely to the forward dissociated products, with little collisional stabilization of the peroxy radicals above 1000 K, and only a minor amount of reverse dissociation at ca. 2000 K. Accordingly, in combustion systems this chemistry can be included as a formally direct process from the R-carboxyalkyl radicals (+O2), with rate constant at around the high-pressure limit for O2 association. It should be emphasized, though, that at high temperatures the thermal isomerization and/or decomposition of the R-carboxyalkyl radicals may become important, and will also need to be included in kinetic models. At 1 atm and 300 K both radical + O2 mechanisms essentially result in only quenching of the R-carboxyalkylperoxy radicals. For the R-carboxymethyl radical + O2 system, branching to the

OH Reforming in the Troposphere

J. Phys. Chem. A, Vol. 114, No. 25, 2010 6867

Figure 4. Rate constants to different product sets in the chemically activated CH3CHC(O)OH + O2 reaction, at 10 atm.

CO2 + OH + HCHO product set, in a direct process, is only 0.9% of the total reaction flux, whereas for the corresponding process in the R-carboxyethyl radical + O2 reaction it is 0.6%. This process can therefore be neglected, and the fate of R-carboxyalkyl radicals in the lower troposphere will only be association with O2 to form an RO2 radical. At around the tropopause, however, it appears that these R-carboxyalkyl + O2 processes proceed to new, dissociated products in significant quantities. At 200 K and 0.1 atm the R-carboxymethyl + O2 reaction is predicted to yield 2.2% CO2 + OH + HCHO, whereas for the same conditions the R-carboxyethyl + O2 reaction forms 1.2% CO2 + OH + CH3CHO. This result is due to the reduced collisional energy transfer taking place at these low pressures, coupled with tunnelling in the decomposition reactions which assists in maintaining appreciable reaction rates at low temperatures. As pressures decrease to be on the order of hundredths of atm in the stratosphere, with temperatures from ca. 200-300 K, formally direct decomposition with OH regeneration actually becomes the dominant fate of these R-carboxyalkyl radicals. For example, at 0.01 atm and 300 K (characteristic of the mid to upper stratosphere), 47% of the R-carboxymethyl radical directly produces CO2 + OH + HCHO upon reaction with O2, whereas 37% of the R-carboxyethyl radical is predicted to follow its corresponding reaction. The direct formation of these products therefore needs to be considered in OH-initiated oxidation of carboxylic acids in the stratosphere. Although the regeneration of OH in a formally direct process upon reaction of R-carboxyalkyl radicals with O2 appears to be relatively unimportant in the troposphere, a similar process may play a role in isoprene photooxidation. Recent studies have identified low-barrier isomerization and decomposition reactions in the isoprene β-hydroxperoxy12,13 and Z conformer δ-hydroxyperoxy radicals,13 similar to those studied here. Peeters et al.13 suggested that for conditions typical of the pristine forest boundary layer the δ channel would dominate (resulting in HO2 production) because of the smaller reaction barrier, even though O2 addition to the isoprene-OH radical adducts favors the β-hydroxyperoxy radicals. Repeated O2 addition/dissociation cycles were found to channel the reaction flux through the δ-hydroxyperoxy radicals, due to the smaller barrier for their further reaction compared to the β-hydroxyperoxy radicals. da Silva et al.12 suggested that repeated O2 addition/dissociation

cycles could result in an appreciable prompt (chemically activated) OH source, as demonstrated above for the R-carboxyalkyl + O2 reactions. Although branching to OH-regenerating products will be small in the lower troposphere (below 1% for the R-carboxyalkyl radicals, and expected to be even less for the isoprene-OH adducts), it could amount to a significant process if O2 addition/dissociation takes place hundreds or even thousands of times. This is plausible considering that calculated lifetimes for the isoprene peroxy radicals are subsecond, although a better understanding of O2 addition and dissociation kinetics is required to evaluate the true role of this mechanism in the photochemical oxidation of isoprene. Combustion Chemistry. In addition to its role in atmospheric chemistry, the mechanism described here is expected to play a role in combustion systems. Carboxylic acids are not generally considered to be important combustion intermediates, although they may play a significant role in the combustion of oxygenated biofuels such as the methyl and ethyl esters (biodiesel) and ethanol. Osswald et al.46 observed acetic acid as a minor intermediate in the combustion of methyl acetate and ethyl formate, two biodiesel model compounds. Interestingly, acetic acid production appears to peak at moderate combustion temperatures (ca. 1500 K), indicating that carboxylic acids may be involved in the autoignition and early stage combustion of biofuels. The reaction of carboxylic acids with OH and other radicals at relevant combustion temperatures is expected to proceed via abstraction of the weak H atoms, forming RCHC(O)OH type radicals. The reaction of these radicals with O2 should proceed directly to OH, CO2, and an aldehyde, as described here. It therefore appears that biofuel oxidation can generate carboxylic acids at low to moderate temperatures and that these carboxylic acids will react further in an exothermic process that either regenerates or produces OH (depending on the initiating radical). Another pathway to R-carboxyalkyl radicals in combustion systems (and elsewhere) is via OH addition to ketenes, compounds of the form R(R′)CdCdO. Ketenes, especially the parent ethenone (CH2CO, often referred to as ketene), are a major class of combustion intermediate, again thought to be of particular importance to the oxidation of biodiesel and other oxygenated biofuels.47 The ketene + OH reaction appears to be in the high-pressure limit at moderately high temperatures and low pressures,48 which has been interpreted as indicating

6868

J. Phys. Chem. A, Vol. 114, No. 25, 2010

that OH addition does not yield stable C2H3O2 isomers (like the R-carboxymethyl radical). Current kinetic models instead include only direct formation of dissociated products like CO2 + CH3.49 Addition of OH to the central C atom in CH2CO, forming CH2C(O)OH, should be exothermic by around 54.4 kcal mol-1, using heats of formation reported here and in the literature.47 This is considerably exothermic, and OH addition to CH2CO will form an activated [CH2C(O)OH]* adduct with considerable excess vibrational energy. However, isomerization of this species to CH3CO2 (which readily dissociates to CH3 + CO2) also requires a significant barrier and is predicted to dominate at only high temperatures and low pressures.50 The apparent absence of pressure-dependence in the CH2CO + OH reaction can then be explained by the lack of a reverse dissociation channel under most tested experimental conditions, given the large barrier for this process. At energies where this channel can be accessed, the somewhat lower-energy pathway to CH3 + CO2 also becomes available (ca. 7 kcal mol-1 below the CH2CO + OH energy).50 Although further work is required to characterize the kinetics and products of the CH2CO + OH reaction across a wide range of temperatures and pressures, it appears that quenched CH2C(O)OH will be a significant product in engines at moderate temperatures and below. Inclusion of this chemistry in kinetic models, along with further reaction with O2 to regenerate OH, should lead to improved modeling of biodiesel ignition chemistry, and the combustion of other fuels. Biochemistry. The R-carboxyalkyl radicals studied here are known to be important in certain biochemical systems, in addition to their involvement in the atmosphere and in combustion outlined above. Gamma irradiation of amino acids has long been known to produce an R-carboxyalkyl radical + NH2. For the simplest amino acid, glycine, the R-carboxyalkyl radical CH2C(O)OH is the irradiation product.51 There is significant interest in understanding the radicals that form when DNA and proteins are exposed to radiation, as well as in the further chemistry of these radicals.52 In the human body, the R-carboxyalkyl radicals that are produced when amino acids are exposed to γ-rays will be available to react with dissolved O2, producing R-carboxyalkylperoxy radicals that may then be able to decompose to OH (plus other products) on biologically relevant time scales. The formation of highly reactive OH in vivo would be significant, due to its ability to induce oxidative stress, and may point to a new mechanism of radiationinduced oxidative stress. Astrochemistry. The final role for R-carboxyalkyl radicals that is considered in this work is in astrochemistry. Ketenes are found throughout the interstellar medium (ISM),53 and reaction with OH is predicted to be a major sink,48 where it appears that collisional stabilization of R-carboxyalkyl radicals could be an important process. Although these radicals may not be formed at the very low pressures associated with the ISM, they should be produced when reaction occurs on surfaces such as ice grains and dust particles. Interstellar R-carboxyalkyl radicals will not react with O2 but can associate with any number of radicals and ions that have been detected in space. Because these radical-radical and radical-ion reactions are typically barrierless, they are rapid at the low temperatures encountered in the ISM (tens to hundreds of K). For example, reaction of the R-carboxymethyl radical with NH2, a known interstellar radical,54 should produce glycine. Reaction of other R-carboxyalkyl radicals with NH2 and structurally related radicals and ions should be able to achieve the interstellar synthesis of a wide variety of amino acids. This chemistry is predicted to be of particular importance when these reactions take place on particle surfaces, not for any catalytic purpose, but purely to remove excess energy

da Silva so as to avoid decomposition. A diverse range of extraterrestrial amino acids have been detected in meteorites such as that found near Murchison,55 and glycine may have been detected toward the Sagittarius B2 star forming region.56 The reactions leading to amino acid synthesis are not presently well accounted for, and the ketene + OH/R-carboxyalkyl + NH2•/+ sequence of reactions provides a new mechanism for their production. Conclusions A theoretical study of the R-carboxymethylperoxy and R-carboxyethylperoxy radicals formed in carboxylic acid oxidation demonstrates that they thermally decompose to an aldehyde + CO2 + OH at ambient temperatures. The rapid decomposition of R-carboxyalkylperoxy radicals provides a new mechanism for OH regeneration in the pristine forest boundary layer and in the unpolluted free troposphere, as well as throughout the planetary boundary layer at night. In the forest boundary layer carboxylic acids are present at high levels, and the mechanism revealed here should go some way toward describing the elevated levels of OH recently observed in these environments. The R-carboxyalkyl radicals considered here will also form in combustion systems, from carboxylic acid oxidation and in ketene + OH reactions, and here their reaction with O2 is expected to lead directly to OH, CO2, and an aldehdye. The γ-irradiation of amino acids is also known to form R-carboxyalkyl type radicals, and their subsequent reaction with O2 in vivo, forming OH, may induce oxidative stress. In the ISM, the reaction of ketenes with OH, followed by addition of NH2 and other radicals and ions, is proposed as a new pathway to extraterrestrial amino acids. Acknowledgment. Computational facilities provided by the Victorian Partnership for Advanced Computing (VPAC). Supporting Information Available: B3LYP/6-31G(2df,p) structures and vibrational frequencies; G3SX energies; transition state thermochemistry; R-carboxyethylperoxy radical energy surface; rate constants for decomposition of the acetic acid, propanoic acid, and acrylic acid R-carboxyperoxy radicals; and R-carboxyethyl + O2 QRRK rate constant plot. This information is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Atkinson, R. Atmos. EnViron. 2000, 34, 2063–2101. (2) Tan, D.; Faloona, I.; Simpas, J. B.; Brune, W.; Shepson, P. B.; Couch, T. L.; Sumner, A. L.; Carroll, M. A.; Thornberry, T.; Apel, E.; Riemer, D.; Stockwell, W. J. Geophys. Res. 2001, D20, 24407–24427. (3) Carslaw, N.; Creasey, D. J.; Harrison, D.; Heard, D. E.; Hunter, M. C.; Jacobs, P. J.; Jenkin, M. E.; Lee, J. D.; Lewis, A. C.; Pilling, M. J.; Saunders, S. M.; Seakins, P. W. Atmos. EnViron. 2001, 35, 4725–4737. (4) Kuhn, U.; Andreae, M. O.; Ammann, C.; Araujo, A. C.; Brancaleoni, E.; Ciccioli, P.; Dindorf, T.; Frattoni, M.; Gatti, L. V.; Ganzeveld, L.; Kruijt, B.; Lelieveld, J.; Lloyd, J.; Meixner, F. X.; Nobre, A. D.; Poschl, U.; Spirig, C.; Stefani, P.; Thielmann, A.; Valentini, R.; Kesselmeier, J. Atmos. Chem. Phys. 2007, 7, 2855–2879. (5) Ren, X.; Olson, J. R.; Crawford, J. H.; Brune, W. H.; Mao, J.; Long, R. B.; Chen, Z.; Chen, G.; Avery, M. A.; Sachse, G. W.; Barrick, J. D.; Diskin, G. S.; Huey, L. G.; Fried, A.; Cohen, R. C.; Heikes, B.; Wennberg, P. O.; Singh, H. B.; Blake, D. R.; Shetter, R. E. J. Geophys. Res. 2008, 113, D05310. (6) Lelieveld, J.; Butler, T. M.; Crowley, J. N.; Dillon, T. J.; Fischer, H.; Ganzeveld, L.; Harder, H.; Lawrence, M. G.; Martinez, M.; Taraborrelli, D.; Williams, J. Nature 2008, 452, 737–740. (7) Hofzumahaus, A.; Rohrer, F.; Lu, K.; Bohn, B.; Brauers, T.; Chang, C.-C.; Fuchs, H.; Holland, F.; Kita, K.; Kondo, Y.; Li, X.; Lou, S.; Shao, M.; Zeng, L.; Wahner, A.; Zhang, Y. Science 2009, 324, 1702–1704. (8) Pugh, T. A. M.; MacKenzie, A. R.; Hewitt, C. N.; Langford, B.; Edwards, P. M.; Furneaux, K. L.; Heard, D. E.; Hopkins, J. R.; Jones, C. E.; Karunaharan, A.; Lee, J.; Gills, G.; Misztal, P.; Moller, S.; Monks, P. S.; Whalley, L. K. Atmos. Chem. Phys. Discuss. 2009, 9, 19243.

OH Reforming in the Troposphere (9) Shindell, D. T.; Faluvegi, G.; Koch, D. M.; Schmidt, G. A.; Unger, N.; Bauer, S. E. Science 2009, 326, 716–718. (10) Hasson, A. S.; Tyndall, G. S.; Orlando, J. J. J. Phys. Chem. A 2004, 108, 5979–5989. (11) Dillon, T. J.; Crowley, J. N. Atmos. Chem. Phys. 2008, 8, 4877–4889. (12) Paulot, F.; Crounse, J. D.; Kjaergaard, H. G.; Ku¨rten, A.; St. Clair, J. M.; Seinfeld, J. H.; Wennberg, P. O. Science 2009, 325, 730–733. (13) da Silva, G.; Graham, C.; Wang, Z.-F. EnViron. Sci. Technol. 2010, 44, 250. (14) Peeters, J.; Nguyen, T. L.; Vereecken, L. Phys. Chem. Chem. Phys. 2009, 11, 5935–5939. (15) Chebbi, A.; Carlier, P. Atmos. EnViron. 1996, 30, 4233–4249. (16) Andreae, M. O.; Talbot, R. W.; Andreae, T. W.; Harross, R. C. J. Geophys. Res. 1988, 93, 1616–1624. (17) Talbot, R. W.; Andreae, M. O.; Berresheim, H.; Jacob, D. J.; Beecher, K. M. J. Geophys. Res. 1990, 95, 16799–16811. (18) Jacob, D. J.; Heikes, E. G.; Fan, S.-M.; Logan, J. A.; Mauzerall, D. L.; Bradshaw, J. D.; Singh, H. B.; Gregory, G. L.; Talbot, R. W.; Blake, D. R.; Sachse, G. W. J. Geophys. Res. 1996, 101, 24235–24250. (19) Butkovskaya, N. I.; Kukui, A.; Pouvesle, N.; Le Bras, G. J. Phys. Chem. A 2004, 108, 7021–7026. (20) (a) Vimal, D.; Stevens, P. S. J. Phys. Chem. A 2006, 110, 11509. (b) Huang, Y.; Dransfield, T. J.; Miller, J. D.; Rojas, R. D.; Castillo, X. G.; Anderson, J. G. J. Phys. Chem. A 2009, 113, 423. (c) Szabo, E.; Tarmoul, J.; Tomas, A.; Fittschen, C.; Dobe, S.; Coddeville, P. React. Kinet. Catal. Lett. 2009, 96, 299. (21) Singleton, D. L.; Paraskevopoulos, G.; Irwin, R. S. J. Am. Chem. Soc. 1989, 111, 5248–5251. (22) da Silva, G.; Bozzelli, J. W. J. Phys. Chem. A 2006, 110, 13058. (23) Curtiss, L. A.; Redfern, P. C.; Raghavachari, K.; Pople, J. A. J. Chem. Phys. 2001, 114, 108–117. (24) Zheng, J.; Zhao, Y.; Truhlar, D. G. J. Chem. Theory. Comput. 2009, 5, 808. (25) Ochterski, J. W.; Petersson, G. A.; Montgomery, J. A., Jr. J. Chem. Phys. 1996, 104, 2598. (26) da Silva, G.; Moore, E. E.; Bozzelli, J. W. J. Phys. Chem. A 2009, 113, 10264. (27) (a) da Silva, G.; Cole, J. A.; Bozzelli, J. W. J. Phys. Chem. A 2009, 113, 6111. (b) da Silva, G.; Bozzelli, J. W.; Asatryan, R. J. Phys. Chem. A 2009, 113, 8596. (c) da Silva, G. Chem. Phys. Lett. 2009, 474, 13. (28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin; A. J.; Cammi; R.; Pomelli; C.; Ochterski; J. W.; Ayala; P. Y.; Morokuma; K.; Voth; G. A.; Salvador; P.; Dannenberg; J. J.; Zakrzewski; V. G.; Dapprich; S.; Daniels; A. D.; Strain; M. C.; Farkas; O.; Malick; D. K.; Rabuck; A. D.; Raghavachari; K.; Foresman; J. B.; Ortiz; J. V.; Cui; Q.; Baboul; A. G.; Clifford; S.; Cioslowski; J.; Stefanov, B. B.; Liu; G.; Liashenko; A.; Piskorz; P.; Komaromi; I.; Martin; R. L.; Fox; D. J.; Keith; T.; Al-Laham; M. A.; Peng; C. Y.; Nanayakkara; A.; Challacombe; M.; Gill; P. M. W.; Johnson; B.; Chen; W.; Wong; M. W.; Gonzalez; C.; and Pople, J. A. Gaussian 09, reVision A.02; Gaussian, Inc.: Wallingford CT, 2009. (29) Mokrushin, V.; Bedanov, V.; Tsang, W.; Zachariah, M.; Knyazev, V. ChemRate, Version 1.5.8; National Institute of Standards and Testing: Gaithersburg, MD, 2009. (30) da Silva, G.; Kennedy, E. M.; Dlugogorski, B. Z. J. Phys. Org. Chem. 2007, 20, 167–179. (31) Interestingly, the isoprene hydroxyhydroperoxide + OH f dihydroxyepoxide + OH reactions identified in ref 12 make use of the hydroxyl radical’s potential energy before it can be lost to the surrounds through collisional deactivation. Additional stability (around 20 kcal mol-1) is inferred by liberation of a weak OsOH bond (replaced by CsO), in spite of added ring strain. Utilization of this energy embedded in hydroperoxides is inherent in the Q•OOH + O2 mechanism of low-temperature alkane ignition, and may provide another pathway to OH reformation in the atmosphere, although production of 2OH would be necessary to achieve an OH-neutral process. (32) (a) Vereecken, L.; Peeters, J. J. Phys. Chem. A 1999, 103, 1768. (b) Olivella, S.; Sole´, A. J. Phys. Chem. A 2004, 108, 11651. (c) Kuwata, K. T.; Dibble, T. S.; Sliz, E.; Petersen, E. B. J. Phys. Chem. A 2007, 111, 5032. (d) da Silva, G.; Bozzelli, J. W.; Liang, L.; Farrell, J. T. 5th US Combustion Meeting, San Diego, March 25-28, 2007, paper C27. (e) Za´dor, J.; Fernandes, R. X.; Georgievskii, Y.; Meloni, G.; Taatjes, C. A.; Miller, J. A. Proc. Comb. Inst. 2009, 32, 271. (33) (a) Wagner, V.; Jenkin, M. E.; Saunders, S. M.; Stanton, J.; Wirtz, K.; Pilling, M. J. Atmos. Chem. Phys. 2003, 3, 89. 31. (b) Bloss, C.; Wagner, V.; Bonzanini, A.; Jenkin, M. E.; Wirtz, K.; Martin-Reviejo, M.; Pilling, M. J. Atmos. Chem. Phys. 2005, 5, 623. (c) Bloss, C.; Wagner, V.; Jenkin,

J. Phys. Chem. A, Vol. 114, No. 25, 2010 6869 M. E.; Volkamer, R.; Bloss, W. J.; Lee, J. D.; Heard, D. E.; wirtz, K.; Martin-Reviejo, M.; Rea, G.; Wenger, J. C.; Pilling, M. J. Atmos. Chem. Phys. 2005, 5, 641. (d) Baltaretu, C. O.; Lichtman, E. I.; Hadler, A. B.; Elrod, M. J. J. Phys. Chem. A 2009, 113, 221. (34) (a) Bozzelli, J. W.; Dean, A. M. J. Phys. Chem. A 1990, 94, 3313. (b) Venkatesh, P. K.; Dean, A. M.; Cohen, M. H.; Carr, R. W. J. Chem. Phys. 1999, 111, 8313. (c) Rienstra-Kiracofe, J. C.; Allen, W. D.; Schaefer, H. F. J. Phys. Chem. A 2000, 104, 9823. (d) Vereecken, L.; Nguyen, T. L.; Hermans, I.; Peeters, J. Chem. Phys. Lett. 2004, 393, 432. (e) Kuwata, K. T.; Hasson, A. S.; Dickinson, R. V.; Peterson, E. B.; Valin, L. C. J. Phys. Chem. A 2005, 109, 2514. (35) Rosado-Reyes, C. M.; Francisco, J. S. J. Phys. Chem. A 2006, 110, 4419–4433. (36) Daytime temperatures typically greater than 300 K and NO levels of around 100 ppt and below. (37) Given that RO2 radicals react with HO2 at around the same rate as NO. The self- and cross-reactions of RO2 radicals are expected to assume even less importance, with lower concentrations and rate constants expected. (38) Paulot, F.; Crounse, J. D.; Kjaergaard, H. G.; Kroll, J. H.; Seinfeld, J. H.; Wennberg, P. O. Atmos. Chem. Phys. 2009, 9, 1479–1501. (39) Di Carlo, P.; Brune, W. H.; Martinez, M.; Harder, H.; Lesher, R.; Ren, X.; Thornberry, T.; Carroll, M. A.; Young, V.; Shepson, P. B.; Riemer, D.; Apel, E.; Campbell, C. Science 2004, 304, 722. (40) Karl, T.; Guenther, A.; Turnipseed, A.; Tyndall, G.; Artaxo, P.; Martin, S. Atmos. Chem. Phys. 2009, 9, 7753. (41) (a) Faloona, I.; Tan, D.; Brune, W.; Hurst, J.; Barket, D., Jr.; Couch, T. L.; Shepson, P.; Apel, E.; Riemer, D.; Thornberry, T.; Carroll, M. A.; Sillman, S.; Keeler, G. J.; Sagady, J.; Hooper, D.; Paterson, K. J. Geophys. Res. 2001, 106, 24315. (b) Sillman, S.; Carroll, M. A.; Thornberry, T.; Lamb, B. K.; Westberg, H.; Brune, W. H.; Faloona, I.; Tan, D.; Shepson, P. B.; Sumner, A. L.; Hastie, D. R.; Mihele, C. M.; Apel, E. C.; Riemer, D. D.; Zika, R. G. J. Geophys. Res. 2002, 107, 4043. (42) (a) Ren, X.; Harder, H.; Martinez, M.; Lesher, R. L.; Oliger, A.; Simpas, J. B.; Brune, W. H.; Schwab, J. J.; Demerjian, K. L.; He, Y.; Zhou, X.; Gao, H. Atmos. EnViron. 2003, 37, 3639. (b) Martinez, M.; Harder, H.; Kovacs, T. A.; Simpas, J. B.; Bassis, J.; Lesher, R.; Brune, W. H.; Frost, G. J.; Williams, E. J.; Stroud, C. A.; Jobson, B. T.; Roberts, J. M.; Hall, S. R.; Shetter, R. E.; Wert, B.; Fried, A.; Alicke, B.; Stutz, J.; Young, V. L.; White, A. B.; Zamora, R. J. J. Geophys. Res. 2003, 108, 4617. (43) For example, the unstable R-hydroxyethylperoxy radical (CH3CH(OO)OH), which is formed with around 37 kcal mol-1 of excess energy, and decomposes to acetaldehyde + HO2 with a barrier of less than 14 kcal mol-1, requires a pressure of around 100 atm in order to be collisionally deactivated at 300 K. (44) da Silva, G.; Bozzelli, J. W.; Liang, L.; Farrell, J. T. J. Phys. Chem. A 2009, 113, 8923. (45) (a) Bozzelli, J. W.; Chang, A. Y.; Dean, A. M. Int. J. Chem. Kinet. 1997, 29, 161. (b) Chang, A. Y.; Bozzelli, J. W.; Dean, A. M. Z. Phys. Chem. 2000, 214, 1533. (c) Sheng, C. Y.; Bozzelli, J. W.; Dean, A. M.; Chang, A. Y. J. Phys. Chem. A 2002, 106, 7276. (46) Osswald, P.; Struckmeier, U.; Kasper, T.; Kohse-Ho¨inghaus, K.; Wang, J.; Cool, T. A.; Hansen, N.; Westmoreland, P. R. J. Phys. Chem. A 2007, 111, 4093. (47) Simmie, J. M.; Metcalfe, W. K.; Curran, H. J. ChemPhysChem 2008, 9, 700. (48) Brown, A. C.; Canosa-Mas, C. E.; Parr, A. D.; Wayne, R. P. Chem. Phys. Lett. 1989, 161, 491. (49) Faravelli, T.; Goldaniga, A.; Zappella, L.; Ranzi, E.; Dagaut, P.; Cathonnet, M. Proc. Comb. Inst. 2000, 28, 2601. (50) Hou, H.; Wang, B.; Gu, Y. Phys. Chem. Chem. Phys. 2000, 2, 2329. (51) Teslenko, V. V.; Gromovoi, Y. S.; Krivenko, V. G. Mol. Phys. 1975, 30, 425. (52) For example, see: (a) Croft, A. K.; Easton, C. J.; Radom, L. J. Am. Chem. Soc. 2003, 125, 4119. (b) Vivekananda, S.; Sadilek, M.; Chen, X.; Turecek, F. J. Am. Soc. Mass. Spectrom. 2004, 15, 1055. (c) Hopkinson, A. C. Mass Spectrom. ReVs. 2009, 28, 655. (53) (a) Macdonald, G. H.; Gibb, A. G.; Habing, R. J.; Millar, T. J. Astron. Astrophys. 1996, 119, 333. (b) Nummelin, A.; Bergman, P.; Hjalmarson, A.; Friberg, P.; Irvine, W. M.; Millar, T. J.; Ohishi, M.; Saito, S. Astrophys. J. Supp. Ser. 2000, 128, 213. (54) van Dishoeck, E. F.; Jansen, D. J.; Schilke, P.; Phillips, T. G. Astrophys. J. 1993, 416, L83. (55) (a) Kvenvolden, K.; Lawless, J.; Pering, K.; Peterson, E.; Flores, J.; Ponnamperuma, C.; Kaplan, I. R.; Moore, C. Nature 1970, 228, 923. (b) Kvenvolden, K. A.; Lawless, J. G.; Ponnamperuma, C. Proc. Natl. Acad. Sci. U.S.A. 1971, 68, 486. (c) Engel, M. H.; Macko, S. A.; Silfer, J. A. Nature 1990, 348, 47. (d) Engel, M. H.; Macko, S. A. Nature 1997, 389, 265. (56) Kuan, Y.-J.; Charnley, S. B.; Huang, H.-C.; Tseng, W.-L.; Kisiel, Z. Astrophys. J. 2003, 593, 848.

JP101279P