Simulations of Smog-Chamber Experiments Using the Two

Article pubs.acs.org/est

Simulations of Smog-Chamber Experiments Using the TwoDimensional Volatility Basis Set: Linear Oxygenated Precursors Heber J. Chacon-Madrid,† Benjamin N. Murphy,† Spyros N. Pandis,†,‡ and Neil M. Donahue*,† †

Center for Atmospheric Particle Studies, Carnegie Mellon University, Pittsburgh, PA 15213, United States Department of Chemical Engineering, University of Patras, Patras, Greece

‡

S Supporting Information *

ABSTRACT: We use a two-dimensional volatility basis set (2DVBS) box model to simulate secondary organic aerosol (SOA) mass yields of linear oxygenated molecules: n-tridecanal, 2- and 7tridecanone, 2- and 7-tridecanol, and n-pentadecane. A hybrid model with explicit, a priori treatment of the first-generation products for each precursor molecule, followed by a generic 2DVBS mechanism for later-generation chemistry, results in excellent model-measurement agreement. This strongly confirms that the 2D-VBS mechanism is a predictive tool for SOA modeling but also suggests that certain important first-generation products for major primary SOA precursors should be treated explicitly for optimal SOA predictions.

■

INTRODUCTION Chemical transport models (CTMs) predict organic aerosol mass and its time evolution in an atmospheric context.1,2 Current three-dimensional CTMs, such as the Community Multiscale Air Quality Model3−5 (CMAQ) and the Comprehensive Air Quality Model6,7 (CAMx), have continually improved in the past decade regarding their predictions of particulate matter concentration (PM2.5). Nonetheless, there are still important aspects that need to be incorporated into models to achieve more accurate predictions. An important process not widely implemented in CTMs is extended chemical aging of organic aerosol.4 The typical residence time for a particle in the atmosphere is about a week,8,9 which is sufficient time for molecules to experience multiple generations of oxidation that can affect concentrations and composition of organic aerosol.10 Aging mechanisms have been implemented in some models such as the one presented in Carlton et al.,11 Lane et al.,12 and Murphy and Pandis.1 Murphy et al.13 included multiple-generation chemical aging mechanisms. They showed that it is important to understand the extent to which aging increases organic-aerosol concentrations. Specifically, aging pathways that increase aerosol concentrations compete in chemical aging mechanisms with pathways that either leave the concentrations unchanged or even deplete the aerosol.14−16 Accurate aging mechanisms may lead to better descriptions of total organic-aerosol levels, but they may also permit better prediction of other properties such as the organic mass to organic carbon (OM to OC) ratio and water solubility. This can be critical for accurate comparison with existing measurement networks that provide data on, for example, OC levels but not OM, or water-soluble organic carbon (WSOC). © 2012 American Chemical Society

A general challenge for models of organic-aerosol chemistry is the enormous number of chemical species comprising the aerosol.17,18 Models require some form of lumping, either representing the organics with a few surrogate molecules whose chemistry and phase partitioning can be described explicitly19 or representing condensable organics with pseudospecies characterized solely by their relevant physical properties (e.g., volatility).20,21 The lumping mechanisms based on pseudospecies are generally constrained by smog-chamber secondary organic aerosol (SOA) formation experiments, and so, the resulting parametrizations will represent the phase partitioning of the experimental conditions. Subsequent chemical evolution, or aging, is not straightforward to model without additional information.21 The two-dimensional volatility basis set (2D-VBS) framework22,23 enables a powerful prognostic tool for SOA evolution.13 The two dimensions are volatility and oxygenation, and so, the framework directly describes both OM and OC. A key hypothesis in the 2D-VBS aging mechanism is that the ensemble of molecules found in the atmosphere with a given volatility and degree of oxidation (i.e., one lumped species in the 2D-VBS) will have a common average chemical behavior. The 2D-VBS considers continuous chemical aging and includes fragmentation and functionalization paths as the aging occurs.4,24 Here, we use the 2D-VBS to model smog-chamber SOA production. Because individual precursors will always have Received: Revised: Accepted: Published: 11179

April 30, 2012 August 15, 2012 September 12, 2012 September 12, 2012 dx.doi.org/10.1021/es3017232 | Environ. Sci. Technol. 2012, 46, 11179−11186

Environmental Science & Technology

Article

to reactions that add oxygenated functionality to the carbon backbone without breaking it. Fragmentation refers to reactions that break the carbon backbone. Typically, at least one fragment is also a radical that adds new oxygenated functional groups. Both functionalization and fragmentation are described by generalized “kernels” that are then mapped onto specific C*−O/C bins of the 2D-VBS. This version of the mechanism treats oxidation by OH only; there is evidence that multifunctional oxygentated organics may have significant losses via photolysis,34,35 but that is not treated in this implementation. The functionalization kernel assumes that one generation of oxidation by OH results in products with C* lowered by between one and six decades (mostly two to four), and one to three added oxygens. The kernel is a 7 × 3 probability matrix that must be mapped into the 2D space because the kernel predicts the number of oxygen atoms added in a functionalization step, but the 2D-VBS requires the change in O/C. This is straightforward, and both the kernel and the mapping are described in Murphy et al.36 The fragmentation scheme assumes that C−C bonds cleave at a random position along the carbon backbone (on average), because fragmentation is presumed to be driven by the presence on functional groups on the backbone and these functional groups are in turn assumed to be randomly distributed on average for the myriad species in a given 2D-VBS bin. Fragmentation thus disperses products over a wide C* and O/C range. This kernel is again described in Murphy et al.36 The final requirement for the 2D-VBS mechanism is the branching ratio (β) between the functionalization and fragmentation pathways. We use O/C to determine the

unique behaviors with respect to ensembles of similar molecules (i.e., molecules with similar volatility and O/C), we expect that the unique features of the first oxidation step for any molecule (the first generation) will have to be represented explicitly. However, because many oxidation products will typically be formed even in the first generation of oxidation, we hypothesize that the average behavior assumed by the 2D-VBS will suffice for later generations of oxidation. This will hold whether the precursor is fully reduced (i.e., pentadecane) or somewhat oxidized (i.e tridecanal); the important trait is not the degree of oxidation of a precursor but the richness of the molecular ensemble in the experiment at any given time. Objectives of This Work. The goal of this work is to use a box model using the 2D-VBS mechanism to simulate SOA production and mass yields from smog-chamber experiments presented in Chacon-Madrid et al.,25 Presto et al.,26 ChaconMadrid and Donahue,15 and Chacon-Madrid et al.;27 the precursor species include n-pentadecane, n-tridecanal, 2- and 7tridecanone, and 2- and 7-tridecanol. These six species are alike in either carbon number or vapor pressure, and they all possess a straight-chain carbon backbone. The oxygenated species are representative first-generation products of hydrocarbons (such as n-pentadecane), so we are “walking through” a complete oxidation mechanism by studying later-generation products one by one. The most important goal of this work is to test whether only one generation of explicit chemistry of isomer species (relative to the starting point) is sufficient to acceptably predict SOA production and mass yields, with later-generation chemistry represented by a default 2D-VBS box model. Two-Dimensional Volatility Basis Set (2D-VBS) Framework as a Prognostic Tool. The details of the 2D-VBS are explained elsewhere;4,22−24 however, we will review a few key points in here. We expand on other details in the Supporting Information. The 2D-VBS is a discretized version of a two-dimensional space in which organic compounds are characterized by their saturation concentration (C* in μg m−3) and degree of oxidation, here O/C (molar ratio of oxygen to carbon atoms). We employ a 2D-VBS box model that predicts chemical aging of bulk organic aerosol, capturing average chemical behavior of an OA system. Other similar frameworks have been developed,18,28 but a unique characteristic of the 2D-VBS is that volatility (C*) is an attribute of the lumped species rather than being calculated from other properties.23 The chemical mechanism in the 2D-VBS describes how material from one bin (a specific C*, O/C pair) is transformed after reaction, with OH radicals here, into products in a number of other bins. The 2D-VBS tracks and conserves carbon, so with known C/H/O, it separately accounts for OC and OM. The nominal formula (CxHyOz) of a given bin is reasonably constrained,22 but the specific structurebranching, location of functional groups, and so forthis not just unknown, it is assumed to be a diverse mixture of different structural and functional isomers and analogues. Thus, the transformations in a 2D-VBS mechanism represent the average behavior of these isomers. This complements fully explicit models such as the Generator for Explicit Chemistry and Kinetics of Organics in the Atmosphere (GECKO-A)29−31 or the master chemical mechanism32,33 that include schemes with every compound and chemical reaction, potentially numbering 10 000 or more. The 2D-VBS splits an oxidation process into two pathways: functionalization and fragmentation.4,13 Functionalization refers

1

branching ratio, with β = (O/C) /6.4 For example, at an O/C = 0.5, β = 0.89, which means that ∼89% of carbon mass will fragment, moving most products into higher volatility bins. We chose this functional form because the probability of fragmentation rises sharply for low O/C, consistent with observations that any functional groups will greatly enhance the probability of fragmentation for even a large molecule14,15 but then levels off toward unity at O/C = 1. The exponent is a convenient parameter for sensitivity studies, as described in Donahue et al.24 Once a full distribution of species is determined, the total organic aerosol mass (COA) is determined based on equilibrium partitioning.21 In this work we assume an ideal solution, so all species of the 2D-VBS with the same C* are combined to yield a 1D-VBS representation of the volatility distribution.21−23 Iterative solution of the equilibrium partitioning is trival. The 2D-VBS mechanism is designed to describe the average behavior of organics, based on the assumption that oxidation of even one large precursor will rapidly lead to a very diverse set of product molecules, not to mention mixtures of precursors in the atmosphere. However, when considering OA formation from individual reactant molecules, it does not distinguish between functional-group and structural isomers. For example, functional-group isomers such as n-tridecanal and 2-tridecanone are both oxidized exactly in the same way in the 2D-VBS, regardless of which one is fragmented more in its firstgeneration chemistry.15 Also, structural isomers like 2- and 7tridecanone (or 2- and 7-tridecanol) would be treated identically. Thus, when considering specific precursor molecules, we do not expect the 2D-VBS to be accurate; however, we hypothesize that, by describing the first-generation chemistry of such isomers, either functional-group or structural, 11180

dx.doi.org/10.1021/es3017232 | Environ. Sci. Technol. 2012, 46, 11179−11186


Article

Figure 1. Product volatility distribution of generic 2D-VBS (gray) vs explict first-generation treatments (red) for different precursors (volatility indicated with a green circle). The red bars are slightly offset to reveal gray bars behind. These panels show the product distribution volatility of the expected first-generation products of OH oxidation for two different cases. The first case is the 2D-VBS simulation; this distribution is obtained from the average chemistry implied by the box model for bulk organic material.4,24 The second case, explict first generation, presents a distribution that is obtained from the literature hydrocarbon + OH chemistry for individual molecules.37−39 The main reason for the differences between both cases is that the 2D-VBS considers average chemistry of bulk organic material, which is quite different from that of linear mono-oxygenated molecules.

treatment of the first-generation chemistry improves 2D-VBS performance but also whether the generic mechanism used for later-generation aging performs well. We do not fit observed first-generation mass yields. Rather, the explicit first-generation products are based on known hydrocarbon plus OH chemistry.37−39 The stoichiometry of the products comes from the expected first-generation attacks on the different sites of the reactant molecule predicted by structure-reactivity relationships (Supporting Information, Figure S2);40 products of fragmentation are expected to come from OH attack on α carbons (side carbons to functional group) and also from attack on carbons where the alkoxy radicals can isomerize by 1,5-hydrogen-shift toward the α carbons. The decay of the parent reactant for the computer simulations, for both cases in Figure 1, is obtained from experimental data; this decay trace is also used to determine OH-radical concentrations in the system. OH-radical concentrations typically were roughly 107 molecules cm−3 for the first hour of oxidation, after which they dropped by about an order

and subsequently using the generic 2D-VBS box model, we can predict SOA formation to within the experimental error of the observations.

■

METHODS AND SIMULATIONS We perform paired simulations for each precursor with the 2DVBS box model. In the first case, there is no explicit firstgeneration chemistry included in the simulation. In the second case, the first step of chemistry is treated explicitly (meaning that the distribution of products over the 2D-VBS is treated based on the expected chemistry of the precursor). The first generation of each case is illustrated in a 1D-VBS format in Figure 1 for each precursor molecule studied here, as well as in more detail in the Supporting Information (Figures S1 and S2). The default VBS products are shown in gray and the explicit ones in red. In each case, the first-generation products are subsequently aged with the default 2D-VBS mechanism, assuming that this adequately describes the average oxidation of later-generation products. We can thus test whether explicit 11181



Article

Figure 2. Simulations vs experimental results. Each panel presents SOA mass yields for three sets of results: (i) 2D-VBS simulation (solid black); (ii) explicit first-generation simulation (solid violet); and (iii) experimental results.15,25,26 For all oxygenated species, there is excellent agreement between the experimental data and the explicit first-generation simulation. This means that establishing only the first-generation chemistry and then using the 2D-VBS box model is sufficient to generate acceptable agreement between simulations and experimental data.

ing to a carbonyl or ether functionality (dihydrofuran) in the multifunctional product. Saturation concentrations of the first-generation oxidation products must be estimated to place them in the 2D-VBS (Figure 1). Experimental data are very limited, so we use structure−activity relationships. There are multiple methods for vapor-pressure estimation;41−43 here, we use SIMPOL from Pankow and Asher.41 However, all methods present a degree of uncertainty in vapor-pressure calculation, adding to the uncertainty of our results for the explicit first-generation schemes.

of magnitude. It is important to mention that the experimental SOA mass yields presented (Figure 2) are the result of precursor and respective products photo-oxidation with the OH radical. Consequently, the experimental product distribution is not static; this is also the case for the 2D-VBS simulation. The importance of photo-oxidizing only the precursor versus the precursor and products is discussed in the Supporting Information in Figure S3. The first-generation products are presented in Figure S2 in the Supporting Information and also as red bars in Figure 1. The red species in Figure S2 of the Supporting Information represent products of fragmentation paths and the blue species those of functionalization. The stoichiometry has been rounded to the closest 0.05 due to the uncertainty in structure−reactivity relationships of Kwok and Atkinson.40 The oxygens in the fragmentation products (red species Figure S2 in the Supporting Information) represent a carbonyl or an ether functionality typical of dihydrofurans;39 both functionalities have a similar effect on vapor pressure.41 The functionalization products (blue species Figure S2 in the Supporting Information) possess two oxygens, one from the original oxygenated moiety in the reactant and the second correspond-

■

RESULTS AND DISCUSSION

In Figure 2, we present SOA mass yields as a function of total organic aerosol concentration (COA) for six different species: npentadecane, n-tridecanal, 2- and 7-tridecanone, and 2- and 7tridecanol. Experimental data come from Presto et al.26 for npentadecane, Chacon-Madrid and Donahue15 for carbonyls, and Chacon-Madrid et al.27 for alcohols (experimental conditions are presented in Table S1 of the Supporting 11182



Article

irrelevant for SOA formation purposes), and a C13O2 species (corresponding to a dicarbonyl or carbonyl with a dihydrofuran). Both, the C12O and C13O2 species have one fewer oxygen than a typical isomerization reaction would predict.39,50,51 We chose such structures because isomerization reactions lead to dehydration under low relative humidity (our conditions for all experiments were of low relative humidity), expelling an oxygen from the carbon backbone and forming a dihydrofuran.52 However, the dihydrofuran can be in equilibrium with a hydroxy−carbonyl species with one more oxygen. Representing this equilibrium in the simulation was not straightforward, so we decided to represent the first-generation products with C12O, C13O2, and CO2. This would slightly underestimate experimental yields, consistent with Figure 2b. Other Precursors: Position of Functional Groups. The simulation pairs 2- and 7-tridecanone and 2- and 7-tridecanol present an instructive sequence to understand the effect of functional-group positioning on SOA formation potential. A major difference between the ketones and the alcohols is the order of magnitude difference in their vapor pressures; alcohols are less volatile. Nonetheless, both the alcohol and ketone pairs probe the influence of functional-group location. Consequently, 7-tridecanone and 7-tridecanol have systematically lower SOA yields than 2-tridecanone and 2-tridecanol, respectively. This is similar to results presented by Lim and Ziemann53,54 where substituted alkanes have lower SOA yields than analogous straight-chain alkanes. SOA yields are suppressed when the functional group is in the center due to creation of fragments (e.g., hexanal and heptanal for 7-tridecanone and 7-tridecanol) of higher vapor pressure when compared to the case where the functional group is on the side. For example, a C10O species (expected product from 2-tridecanone and 2-tridecanol) can contribute to later-generation SOA more effectively than C6O and C7O species. The position dependence of substituents is evident in Figure 1: low C* yields for functional groups in the 2-position (Figure 1c, e) are much larger than yields with the functional group in the 7-position (Figure 1d, f). The small discrepancies between experimental data and the explicit first-generation box model can also be attributed to uncertainty in the structure−reactivity relationships (SRRs) from Kwok and Atkinson.40 SRRs are used to determine the stoichiometry of products of fragmentation and functionalization in here (Figure S2 in the Supporting Information). The SRR can be instructive in general to capture the likelihood of fragmentation but with a relatively high degree of uncertainty. Level of Oxygenation Intercomparison. In order to better assess the performance of the 2D-VBS box model, we compare experimental O/C with that from the explicit firstgeneration box model. Experimental data exists for npentadecane, n-tridecanal, and 2-tridecanol. The intercomparison can be seen in Figure S4 in the Supporting Information. The agreement is acceptable considering the uncertainty in estimating O/C. Oxygenation is estimated using unit mass resolution data measured by a quadrupole AMS as described in Aiken et al.55 Both experimental data and box-model predictions reach O/C levels of roughly 0.4, which is very consistent with levels of oxygenation typically observed in smog chambers.56 Environmental Significance. The 2D-VBS box model is an effective prognostic tool for simulating SOA formation and evolution. Even though the box model was created for bulk organic aerosol with average chemistry, it is capable, as shown

Information); however, here, the SOA mass yields have been corrected for a typical OA density of 1.4 g cm−3.10,44 In each panel of Figure 2, we present the following: (i) the default 2D-VBS box model simulation (black solid line); (ii) the explicit first-generation followed by 2D-VBS box model simulation (violet solid line); and (iii) the experimental SOA mass yields (each species has a different color for experimental data). The simulations were carried out for only the range of precursor consumed in the experiments (from zero to the maximum value), so the limit of curves (the maximum COA) for the simulations can also be compared with the data and with each other. All experiments presented here were performed under high-NOx conditions (maximum of 2 ppb C/ppb NOx), while the 2D-VBS box model simulations are for generic conditions.24 High-NOx conditions lead to organic nitrate (−ONO2) formation that can reduce vapor pressure by ∼2.5 orders of magnitude.41 Low-NOx conditions favor hydroperoxide (−OOH) formation,45−48 which can lower vapor pressures by a similar factor. Therefore, the vapor pressure of products, and SOA mass yields, of high- and low-NOx cases might not be very different, unless very UV-sensitive products are formed.34,35 For this reason, the explicit first generation followed by the 2D-VBS can reasonably predict high-NOx experiments even when its scheme does not consider NOx conditions. In most cases, the default 2D-VBS box model overestimates the experimental SOA mass yields. This is because the default model was not designed for individual molecules but for bulk organic aerosol material, capturing average chemistry, that includes species that easily fragment and those that functionalize more effectively. However, the predictive ability of SOA mass yields from the box model with explicit first-generation chemistry is acceptable and within the experimental uncertainty for all of the species studied here. Experimental SOA yields have a minimum uncertainty of ±20%15,49 as seen in the error bars in Figure 2, and the explicit model falls within that range of the data for most of the species over most of the concentration ranges presented in Figure 2. We must emphasize that the firstgeneration mechanism is based on a priori consideration of the chemistry using structure−reactivity relations and contains no fitting to the data. The model-measurement agreement for the explicit case is thus a strong confirmation of the underlying mechanism. n-Pentadecane. The two simulations (explicit and generic) roughly bound the observations, with the explicit firstgeneration mechanism resulting in better agreement at low COA and the generic mechanism more closely matching the maximum COA formation. Essentially all of the SOA formed in the simulation comes from later-generation products, especially with the explicit first-generation mechanism, and so, the differences between the simulations arise from how much of “head start” the products in the generic mechanism get over the products in the explicit mechanism, as seen in Figure 1a. n-Tridecanal. The explicit first-generation simulation in Figure 2b captures the fragmentation path that n-tridecanal undergoes in its first-generation chemistry25 as opposed to the default 2D-VBS scheme, which severely overestimates SOA formation over the full COA range. The modest underestimation of the explicit simulation may be the assignment of products for the first-generation chemistry (Figure S2 in the Supporting Information). Our assumption is that first-generation photooxidation of n-tridecanal forms three different species. They are a C12O species (corresponding to n-dodecanal), CO2 (which is 11183



Article

(2) Kanakidou, M.; Seinfeld, J. H.; Pandis, S. N.; Barnes, I.; Dentener, F. J.; Facchini, M. C.; Dingenen, R. V.; B. Ervens, A. N.; Nielsen, C. J.; Swietlicki, E.; Putaud, J. P.; Balkanski, Y.; Fuzzi, S.; Horth, J.; Moortgat, G. K.; Winterhalter, R.; Myhre, C. E. L.; Tsigaridis, K.; Vignati, E.; Stephanou, E. G.; Wilson, J. Organic aerosol and global climate modelling: a review. Atmos. Chem. Phys. 2005, 5, 1053−1123. (3) Appel, K. W.; Bhave, P. V.; Gilliland, A. B.; Sarwar, G.; Roselle, S. J. Evaluation of the community multiscale air quality (CMAQ) model version 4.5: sensitivities impacting model performance; Part IIparticulate matter. Atmos. Environ. 2008, 42, 6057−6066. (4) Jimenez, J. L.; Canagaratna, M. R.; Donahue, N. M.; Prevot, A. S. H.; Zhang, Q.; Kroll, J. H.; DeCarlo, P. F.; Allan, J. D.; Coe, H.; Ng, N. L.; Aiken, A. C.; Docherty, K. S.; Ulbrich, I. M.; Grieshop, A. P.; Robinson, A. L.; Duplissy, J.; Smith, J. D.; Wilson, K. R.; Lanz, V. A.; Hueglin, C.; Sun, Y. L.; Tian, J.; Laaksonen, A.; Raatikainen, T.; Rautiainen, J.; Vaattovaara, P.; Ehn, M.; Kulmala, M.; Tomlinson, J. M.; Collins, D. R.; Cubison, M. J.; E; Dunlea, J.; Huffman, J. A.; Onasch, T. B.; Alfarra, M. R.; Williams, P. I.; Bower, K.; Kondo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S.; Weimer, S.; Demerjian, K.; Salcedo, D.; Cottrell, L.; Griffin, R.; Takami, A.; Miyoshi, T.; Hatakeyama, S.; Shimono, A.; Sun, J. Y.; Zhang, Y. M.; Dzepina, K.; Kimmel, J. R.; Sueper, D.; Jayne, J. T.; Herndon, S. C.; Trimborn, A. M.; Williams, L. R.; Wood, E. C.; Middlebrook, A. M.; Kolb, C. E.; Baltensperger, U.; Worsnop, D. R. Evolution of organic aerosols in the atmosphere. Science 2009, 326, 1525−1529. (5) Daewon, B.; Kenneth, L. S. Review of the governing equations, computational algorithms, and other components of the Models-3 Community Multiscale Air Quality (CMAQ) modeling system. Appl. Mech. Rev. 2006, 59, 51−77. (6) Morris, R. E.; Koo, B.; Guenther, A.; Yarwood, G.; McNally, D.; Tesche, T. W.; Tonnesen, G.; Boylan, J.; Brewer, P. Model sensitivity evaluation for organic carbon using two multi-pollutant air quality models that simulate regional haze in the southeastern United States. Atmos. Environ. 2006, 40, 4960−4972. (7) Schell, B.; Ackermann, I. J.; Hass, H.; Binkowski, F. S.; Ebel, A. Modeling the formation of secondary organic aerosol within a comprehensive air quality model system. J. Geophys. Res. 2001, 106, 28,275−228,293. (8) Wagstrom, K.; Pandis, S. Determination of the age distribution of primary and secondary aerosol species using a chemical transport model. J. Geophys. Res. 2009, 114, 1−12. (9) Balkanski, Y. J.; Jacob, D. J.; Gardner, G. M.; Graustein, W. C.; Turekian, K. K. Transport and residence times of tropospheric aerosols inferred from a global three-dimensional simulation of 210Pb. J. Geophys. Res. 1993, 98, 20573−20586. (10) Rudich, Y.; Donahue, N. M.; Mentel, T. F. Aging of organic aerosol: bridging the gap between laboratory and field studies. Annu. Rev. Phys. Chem. 2007, 58, 321−352. (11) Carlton, A. G.; Bhave, P. V.; Napelenok, S. L.; Edney, E. D.; Sarwar, G.; Pinder, R. W.; Pouliot, G. A.; Houyoux, M. Model representation of secondary organic aerosol in CMAQv4.7. Environ. Sci. Technol. 2010, 44, 8553−8560. (12) Lane, T. E.; Donahue, N. M.; Pandis, S. N. Simulating secondary organic aerosol formation using the volatility basis-set approach in a chemical transport model. Atmos. Environ. 2008, 42, 7439−7451. (13) Murphy, B. N.; Donahue, N. M.; Fountoukis, C.; Pandis, S. N. Simulating the oxygen content of ambient organic aerosol with the 2D volatility basis set. Atmos. Chem. Phys. 2011, 11, 7859−7873. (14) Kroll, J. H.; Smith, J. D.; Che, D. L.; Kessler, S. H.; Worsnop, D. R.; Wilson, K. R. Measurement of fragmentation and functionalization pathways in the heterogeneous oxidation of oxidized organic aerosol. Phys. Chem. Chem. Phys. 2009, 11, 8005−8014. (15) Chacon-Madrid, H. J.; Donahue, N. M. Fragmentation vs. functionalization: chemical aging and organic aerosol formation. Atmos. Chem. Phys. 2011, 11, 10553−10563. (16) Kroll, J. H.; Seinfeld, J. H. Chemistry of secondary organic aerosol: formation and evolution of low-volatility organics in the atmosphere. Atmos. Environ. 2008, 42, 3593−3624.

here, of predicting OA mass formation of individual molecules. This can be done by simply establishing the first-generation chemistry of the individual molecule. Evidently, the firstgeneration product distribution for all of the systems studied here is sufficiently diverse for the generic mechanism to accurately describe the subsequent chemistry. When applied to simulations of the real atmosphere, the averaging nature of the 2D-VBS mechanism is an asset, as it is highly unlikely that one or even a handful of unique organic compounds dominate the OA mass at any location or time. On the other hand, the results from this work suggest that it may be important to treat freshly emitted OA mass with aging chemistry that is more representative of the major compounds from particular sources (if known). It is expected though that, as oxidation reactions continue to transform this OA mass, the OA constituents will become more varied and treating their chemistry with the generic 2D-VBS mechanism will be sufficient. Transport models that simulate SOA aging typically employ a parametrization of SOA formation based on observed mass yields from smog-chamber experiments carried out over a few hours. The approach used in the current work is analogous, but there are subtle differences. The first-generation yields used here represent the first stable products from the initial attack of an oxidant on the precursor gas. Traditional chamber yields take into account any number of generations of chemistry occurring in an experiment. The implications of this difference, especially in three-dimensional photochemical transport models, are not clear. The precursor molecules studied here illustrate some of the subtleties of applying the 2D-VBS to modeling the aging of specific molecules; however, these compounds are not the most relevant species when addressing SOA formation in the atmosphere. It will be important for future work to apply this same approach to experiments with compounds such as toluene and terpenes to determine if only one explicit step of chemistry is necessary before treatment with the 2D-VBS is adequate.

■

ASSOCIATED CONTENT

S Supporting Information *

Conditions of all experiments performed; figures showing default 2D-VBS photo-oxidation simulations, explicit firstgeneration 2D-VBS simulations, effects of continuous aging on SOA mass yields, and oxygen-to-carbon ratio intercomparison; and further details of the 2D-VBS box model. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This research was supported by grant AGS1136479 from the U.S. National Science Foundation. REFERENCES

(1) Murphy, B. N.; Pandis, S. N. Simulating the formation of semivolatile primary and secondary organic aerosol in a regional chemical transport model. Environ. Sci. Technol. 2009, 43, 4722−4728. 11184



Article

(34) Henry, K. M.; Donahue, N. M. Photochemical aging of α-pinene secondary organic aerosol: effects of OH radical sources and photolysis. J. Phys. Chem. A 2012, 116, 5932−5940. (35) Chacon-Madrid, H. J.; Henry, K. M.; Donahue, N. M. Photooxidation of pinonaldehyde at low NOx: from chemistry to organic aerosol formation. Atmos. Chem. Phys. Discuss. 2012, 12, 7727−7752. (36) Murphy, B. N.; Donahue, N. M.; Fountoukis, C.; Dall’Osto, M.; O’Dowd, C.; Kiendler-Scharr, A.; Pandis, S. N. Functionalization and fragmentation during ambient organic aerosol aging: application of the 2-D volatility basis set to field studies. Atmos. Chem. Phys. Discuss. 2012, 12, 9857−9901. (37) Atkinson, R.; Arey, J. Gas-phase tropospheric chemistry of biogenic volatile organic compounds: a review. Atmos. Environ. 2003, 37 (Supplement 2), 197−219. (38) Atkinson, R. Atmospheric chemistry of VOCs and NOx. Atmos. Environ. 2000, 34, 2063−2101. (39) Lim, Y. B.; Ziemann, P. J. Products and mechanism of secondary organic aerosol formation from reactions of n-alkanes with OH radicals in the presence of NOx. Environ. Sci. Technol. 2005, 39, 9229−9236. (40) Kwok, E. S. C.; Atkinson, R. Estimation of hydroxyl radical reaction rate constants for gas-phase organic compounds using a structure-reactivity relationship: an update. Atmos. Environ. 1995, 29, 1685−1695. (41) Pankow, J. F.; Asher, W. E. SIMPOL.1: a simple group contribution method for predicting vapor pressures and enthalpies of vaporization of multifunctional organic compounds. Atmos. Chem. Phys. 2008, 8, 2773−2796. (42) Myrdal, P. B.; Yalkowsky, S. H. Estimating pure component vapor pressures of complex organic molecules. Ind. Eng. Chem. Res. 1997, 36, 2394−2499. (43) Nannoolal, Y.; Rarey, J.; Ramjugernath, D. Estimation of pure component properties: Part 3. Estimation of the vapor pressure of non-electrolyte organic compounds via group contributions and group interactions. Fluid Phase Equilib. 2008, 269, 117−133. (44) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.; Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.; Goldstein, A. H.; Hamilton, J. F.; Herrmann, H.; Hoffmann, T.; Iinuma, Y.; Jang, M.; Jenkin, M. E.; Jimenez, J. L.; Kiendler-Scharr, A.; Maenhaut, W.; McFiggans, G.; Mentel, T. F.; Monod, A.; Prévôt, A. S. H.; Seinfeld, J. H.; Surratt, J. D.; Szmigielski, R.; Wildt, J. The formation, properties and impact of secondary organic aerosol: current and emerging issues. Atmos. Chem. Phys. 2009, 9, 5155−5236. (45) Logan, J. A.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B. Tropospheric chemistry: a global perspective. J. Geophys. Res. 1981, 86, 7210−7254. (46) Donahue, N. M.; Prinn, R. G. In situ nonmethane hydrocarbon measurements on SAGA 3. J. Geophys. Res. 1993, 98, 16915−16932. (47) Maksymiuk, C. S.; Gayahtri, C.; Gil, R. R.; Donahue, N. M. Secondary organic aerosol formation from multiphase oxidation of limonene by ozone: mechanistic constraints via two-dimensional heteronuclear NMR spectroscopy. Phys. Chem. Chem. Phys. 2009, 11, 7810−7818. (48) Docherty, K. S.; Wu, W.; Lim, Y. B.; Ziemann, P. J. Contributions of organic peroxides to secondary aerosol formed from reactions of monoterpenes with O3. Environ. Sci. Technol. 2005, 39, 4049−4059. (49) Hildebrandt, L.; Donahue, N. M.; Pandis, S. N. High formation of secondary organic aerosol from the photo-oxidation of toluene. Atmos. Chem. Phys. 2009, 9, 2973−2986. (50) Atkinson, R.; Arey, J.; Aschmann, S. M. Atmospheric chemistry of alkanes: review and recent develpments. Atmos. Environ. 2008, 42, 5859−5871. (51) Atkinson, R. Rate constants for the atmospheric reactions of alkoxy radicals: an updated estimation method. Atmos. Environ. 2007, 41, 8468−8485. (52) Holt, T.; Atkinson, R.; Arey, J. Effect of water vapor concentration on the conversion of a series of 1,4-hydroxycarbonyls to dihydrofurans. J. Photochem. Photobiol., A 2005, 176, 231−237.

(17) Goldstein, A. H.; Galbally, I. E. Known and unexplored organic constituents in the Earth’s atmosphere. Environ. Sci. Technol. 2007, 41, 1514−1521. (18) Kroll, J. H.; Donahue, N. M.; Jimenez, J. L.; Kessler, S. H.; Canagaratna, M. R.; Wilson, K. R.; Altieri, K. E.; Mazzoleni, L. R.; Wozniak, A. S.; Bluhm, H.; Mysak, E. R.; Smith, J. D.; Kolb, C. E.; Worsnop, D. R. Carbon oxidation state as a metric for describing the chemistry of atmospheric organic aerosol. Nat. Chem. 2011, 3, 133− 139. (19) Griffin, R. J.; Nguyen, K.; Dabdub, D.; Seinfeld, J. H. A coupled hydrophobic-hydrophilic model for predicting secondary organic aerosol formation. J. Atmos. Chem. 2003, 44, 171−190. (20) Odum, J. R.; Hoffmann, T.; Bowman, F.; Collins, D.; Flagan, R. C.; Seinfeld, J. H. Gas/particle partitioning and secondary organic aerosol yields. Environ. Sci. Technol. 1996, 30, 2580−2585. (21) Donahue, N. M.; Robinson, A. L.; Stanier, C. O.; Pandis, S. N. Coupled partitioning, dilution, and chemical aging of semivolatile organics. Environ. Sci. Technol. 2006, 40, 2635−2643. (22) Donahue, N. M.; Epstein, S. A.; Pandis, S. N.; Robinson, A. L. A two-dimensional volatility basis set: 1. organic-aerosol mixing thermodynamics. Atmos. Chem. Phys. 2011, 10, 3303−3318. (23) Donahue, N. M.; Kroll, J. H.; Pandis, S. N.; Robinson, A. L. A two-dimensional volatility basis setPart 2: Diagnostics of organicaerosol evolution. Atmos. Chem. Phys. 2012, 12, 615−634. (24) Donahue, N. M.; Henry, K. M.; Mentel, T. F.; Kiendler-Scharr, A.; Spindler, C.; Bohn, B.; Brauers, T.; Dorn, H. P.; Fuchs, H.; Tillmann, R.; Wahner, A.; Saathoff, H.; Naumann, K.-H.; Möhler, O.; Leisner, T.; Müller, L.; Reinnig, M.-C.; Hoffmann, T.; Salo, K.; Hallquist, M.; Frosch, M.; Bilde, M.; Tritscher, T.; Barmet, P.; Praplan, A. P.; DeCarlo, P. F.; Dommen, J.; Prévôt, A. S. H.; Baltensperger, U. Aging of biogenic secondary organic aerosol via gas-phase OH radical reactions. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 13503−13508. (25) Chacon-Madrid, H. J.; Presto, A. A.; Donahue, N. M. Functionalization vs. fragmentation: n-aldehyde oxidation mechanisms and secondary organic aerosol formation. Phys. Chem. Chem. Phys. 2010, 12, 13975−13982. (26) Presto, A. A.; Miracolo, M. A.; Donahue, N. M.; Robinson, A. L. Secondary organic aerosol formation from high-NOx photo-oxidation of low volatility precursors: n-alkanes. Environ. Sci. Technol. 2010, 44, 2029−2034. (27) Chacon-Madrid, H. J.; Valorso, R.; Aumont, B.; Donahue, N. M. Simulations of smog-chamber experiments using GECKO-A: linear oxygenated precursors. Aerosol Sci. Technol. 2012, submitted. (28) Pankow, J. F.; Barsanti, K. C. The carbon number-polarity grid: a means to manage the complexity of the mix of organic compounds when modeling atmospheric organic particulate matter. Atmos. Environ. 2009, 43, 2829−2835. (29) Aumont, B.; Szopa, S.; Madronich, S. Modelling the evolution of organic carbon during its gas-phase tropospheric oxidation: development of an explicit model based on a self generating approach. Atmos. Chem. Phys. 2005, 5, 2497−2517. (30) Camredon, M.; Aumont, B.; Lee-Taylor, J.; Madronich, S. The SOA/VOC/NOx system: an explicit model of secondary organic aerosol formation. Atmos. Chem. Phys. 2007, 7, 5599−5610. (31) Valorso, R.; Aumont, B.; Camredon, M.; Raventos-Duran, T.; Mouchel-Vallon, C.; Ng, N. L.; Seinfeld, J. H.; Lee-Taylor, J.; Madronich, S. Explicit modelling of SOA formation from α-pinene photooxidation: sensitivity tovapour pressure estimation. Atmos. Chem. Phys. 2011, 11, 6895−6910. (32) Saunders, S. M.; Jenkin, M. E.; Derwent, R. G.; Pilling, M. J. Protocol for the development of the Master Chemical Mechanism, MCM v3 (Part A): tropospheric degradation of non-aromatic volatile organic compounds. Atmos. Chem. Phys. 2003, 3, 161−180. (33) Chen, Q.; Liu, Y.; Donahue, N. M.; Shilling, J. E.; Martin, S. T. Particle-phase chemistry of secondary organic material: model and measurement of O:C and H:C elemental ratios provide constraints. Environ. Sci. Technol. 2011, 45, 4763−4770. 11185



Article

(53) Lim, Y. B.; Ziemann, P. J. Effects of molecular structure on aerosol yields from OH radical-initiated reactions of linear, branched, and cyclic alkanes in the presence of NOx. Environ. Sci. Technol. 2009, 43, 2328−2334. (54) Lim, Y. B.; Ziemann, P. J. Chemistry of secondary organic aerosol formation from OH radical-initiated reactions of linear, branched, and cyclic alkanes in the presence of NOx. Aerosol Sci. Technol. 2009, 43, 604−619. (55) Aiken, A. C.; DeCarlo, P. F.; Kroll, J. H.; Worsnop, D. R.; Huffman, J. A.; Docherty, K. S.; Ulbrich, I. M.; Mohr, C.; Kimmel, J. R.; Sueper, D.; Sun, Y.; Zhang, Q.; Trimborn, A.; Northway, M.; Ziemann, P. J.; Canagaratna, M. R.; Onasch, T. B.; Alfarra, M. R.; Prevot, A. S. H.; Dommen, J.; Duplissy, J.; Metzger, A.; Baltensperger, U.; Jimenez, J. L. O/C and OM/OC ratios of primary, secondary, and ambient organic aerosols with high-resolution time-of-flight aerosol mass spectrometry. Environ. Sci. Technol. 2008, 42, 4478−4485. (56) Ng, N. L.; Canagaratna, M. R.; Zhang, Q.; Jimenez, J. L.; Tian, J.; Ulbrich, I. M.; Kroll, J. H.; Docherty, K. S.; Chhabra, P. S.; Bahreini, R.; Murphy, S. M.; Seinfeld, J. H.; Hildebrandt, L.; Donahue, N. M.; DeCarlo, P. F.; Lanz, V. A.; Prévôt, A. S. H.; Dinar, E.; Rudich, Y.; Worsnop, D. R. Organic aerosol components observed in Northern Hemispheric datasets from aerosol mass spectrometry. Atmos. Chem. Phys. 2010, 10, 4625−4641.

11186


Simulations of Smog-Chamber Experiments Using the Two

Recommend Documents