Article pubs.acs.org/est
Molecular Transformations Accompanying the Aging of Laboratory Secondary Organic Aerosol Wiley A. Hall, IV,† M. Ross Pennington,‡ and Murray V. Johnston‡,* †
United States Department of Agriculture − Agricultural Research Service San Joaquin Valley Agricultural Center, Parlier, California 93648, United States ‡ Department of Chemistry and Biochemistry, University of Delaware, Newark, Delaware 19716, United States S Supporting Information *
ABSTRACT: The aging of fresh secondary organic aerosol (SOA), formed in a flow tube reactor by α-pinene ozonolysis, was studied by passing the fresh SOA into a second chamber for reaction with high levels of the hydroxyl radical. Two types of experiments were performed: (1) injection of a short plug of fresh SOA into the second chamber, where the particle mass and average O/C mole ratio were measured as a function of time after injection, and (2) injection of a continuous stream of fresh SOA into the second chamber, where particles were collected on a filter over a period of time for off line analysis by high performance mass spectrometry. These setups allowed the chemistry of SOA aging to be elucidated. The particle mass decreased and average O/C ratio increased with increasing aging time. Aged SOA showed an oligomer distribution shifted to lower molecular weight (fragmentation) and molecular formulas with higher O/C and lower H/C ratios (functionalization). Carbon oxidation states of individual molecules were higher for aged SOA, 0 to +2, than fresh SOA, −1 to 0. Tandem mass spectrometry of oligomers from fresh SOA showed small neutral losses associated with less oxidized functional groups such as aldehydes and ketones, while oligomers from aged SOA showed losses associated with more highly oxidized groups such as acids and peroxyacids. Product ion spectra of fresh SOA showed monomer building blocks with formulas corresponding to primary ozonolysis products such as pinic and pinonic acids, whereas aged SOA monomer building blocks corresponded to extremely oxidized products such as dimethyltricarballylic acid.
■
SOA is produced by mixing ozone with α-pinene in a flow tube reactor. This reaction is an important contributor to biogenic SOA in the troposphere1,11 and has been studied extensively in the laboratory using high performance mass spectrometry.6,10,12,13 When analyzed with the aerosol mass spectrometer (AMS),14 SOA produced by this reaction exhibits a similar O/ C ratio (≤0.5) to the portion of ambient oxygenated organic aerosol (OOA) originally designated at OOA-2,2,15,16 which has been considered to be freshly formed SOA because of its positive correlation with ambient VOC and oxidant levels.2,15,17,18 In contrast, there has been less success generating laboratory aerosol similar to the highly oxidized portion of ambient organic aerosol known as OOA-1, which typically exhibits an O/C ratio of ∼0.7.15 In this study, aged SOA is generated by passing the fresh SOA from α-pinene ozonolysis through a secondary PHoto-Assisted Reaction: Aerosol Oxidation by Hydroxyl Chamber (PhARAOH Chamber or PC) to simulate photo-oxidative aging in the atmosphere by reaction with hydroxyl radical. The aged aerosol
INTRODUCTION Secondary organic aerosol is produced when volatile organic compounds (VOC) in the atmosphere are oxidized to make particulate matter.1−3 This freshly formed, mildly oxidized aerosol can continue to react to form more highly oxidized, aged aerosol.2,4,5 Whereas the chemical reactions associated with initial oxidation of atmospheric VOC and subsequent formation of fresh SOA have been studied in detail,3,6,7 much less is known on a molecular level of the aging process. Oxidation of an organic compound leads to both functionalization (the addition of polar, oxygenated functional groups that decrease molecular volatility and increase the probability of aerosol formation) and fragmentation (the cleavage of C−C or C−O bonds which can result in the loss of small molecular species containing carbon and oxygen that increase molecular volatility and decrease the probability of aerosol formation).5,8 Functionalization is indicated by an increase in oxygen to carbon (O/C) ratio, or more precisely the oxidation state of carbon,9 as the aerosol ages. Fragmentation is indicated by a corresponding decrease in particulate mass.8 In this study, the SOA aging process is characterized at a molecular level using high performance mass spectrometry10 to determine elemental formulas and functional group information for individual organic compounds in the particle phase. Fresh © 2013 American Chemical Society
Received: Revised: Accepted: Published: 2230
September 27, 2012 December 31, 2012 February 7, 2013 February 7, 2013 dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
is characterized by several measurements including high performance mass spectrometry.
■
EXPERIMENTAL SECTION Fresh SOA was generated in a flow tube reactor (FTR) as described elsewhere.19,20 Air flows in the FTR (set to give laminar flow through the reactor with a residence time of 23 s) were 100CPM (cm3/min) of α-pinene impregnated air and 675CPM of ozone (15−20 ppm, to ensure total reaction of αpinene within the FTR) for a total flow exiting the FTR of 775CPM. Whereas these conditions yield a rather high aerosol mass concentration, a previous high performance mass spectrometry study using similar reaction conditions but with a different apparatus6 has found that atmospherically unrealistic reactions, such as RO2−RO2 type reactions, do not contribute significantly to the product molecules formed. Furthermore, there has been little evidence of RO2−RO2 type reactions in a number of FTR studies using various chemical characterization methods.21−25 The aerosol flow from the FTR was sent into the PC along with an additional 775CPM flow of 15−20 ppm ozone that had been bubbled through deionized water. These flows gave a nominal residence time of 32 min for aerosol in the PC. The PC consisted of a 50L, box-shaped (251 × 251 × 800 mm, WxHxL) inner chamber made from 5mil (0.127 mm) thick perfluoroalkoxy copolymer (Welch Fluorocarbon, Dover, NH), suspended inside a larger outer chamber (419 × 610 × 978 mm, WxHxL). The inside walls of the outer chamber were coated with a reflective material, and four 36” (914 mm) long UV bulbs, coated to transmit radiation only around 254 nm (i.e., no 185 nm present, which would produce ozone as well), ran along the top length of the chamber. Four 4 × 4” (102 × 102 mm) square fans, two in the front of the chamber, and two in the back were located just below the lamps to keep the chamber temperature at ∼30 °C with lamps on. The inner chamber could be accessed by nine Teflon ports that connected to 1/4” ID stainless steel tubing through the outer chamber. Two ports were located in the upstream end the chamber (aerosol flow from FTR and ozone/water flow), 125.5 mm from the bottom, and 74.7 mm away from either side. Two ports were located on the side of the chamber (used for monitoring temperature); 74.7 mm from the front and the back, and 74.7 mm from the bottom. Five ports were located on the downstream end of the chamber (used for monitoring RH and ozone concentration, as well as exit flows for various methods of SOA collection and measurement), four located 49.2 mm away from the nearest side and bottom, and one in the center. In this work, SOA exiting the PC under dark conditions (no formation of OH radical) is referred to as fresh SOA because its chemical characteristics are essentially unchanged from aerosol exiting the FTR (confirmed by molecular analysis). SOA exiting the PC under light conditions (formation of OH radical) is referred to as aged SOA. Two experimental setups were used in this work: either a plug injection (part A of Figure 1) or a constant stream (part B of Figure 1) of particles from the FTR into the PC. Both setups were designed to maintain a higher aerosol flow rate entering the PC than what was needed for analysis, since this reduced the possibility of contamination. Excess flow from the FTR was vented to waste. For the setup in part A of Figure 1, the FTR was connected to the PC by a series of two 3-way valves, which could direct the flow from the FTR through a HEPA filter, or
Figure 1. Photo Assisted Reaction: Aerosol Oxidation by Hydroxyl (PhARAOH) Chamber experimental setups, (A) for the analysis of plug injections of aerosol, and (B) for the analysis of a constant aerosol stream. Circles represent 3-way valves.
directly into the chamber. A 0.5 mL injection of liquid α-pinene was made into the 100CPM air flow, where it was vaporized and sent into the FTR along with the ozone flow as described previously. A plug of aerosol exiting the FTR passed directly into the PC for a 5 min period, after which the FTR flow was directed through the HEPA filter to prevent any additional particles from entering the PC. A scanning mobility particle sizer (SMPS) and the nano aerosol mass spectrometer (NAMS), were used to characterize the aerosol exiting the PC. NAMS (described in detail elsewhere19,24,26−28) is used to quantitatively measure the elemental composition of nanoparticles in the 10−30 nm size range. For this work the NAMS conditions were optimized for analysis of 30 nm particles. For the setup in part B of Figure 1, α-pinene was fed continuously into the FTR airstream with a syringe pump set to a liquid flow rate of 1−3 μL/min. A 3-way valve directed the aerosol flow to the SMPS to monitor the aerosol concentration and size distribution until it reached the desired level (adjusted by changing syringe pump rate). The aerosol flow was then directed into the PC with the aerosol exiting the chamber monitored by SMPS at a rate of 0.3LPM (L/min) and collected for 24−48 hrs onto a Teflon coated, glass fiber filter (25 mm Fiberfilm, Pall Life Sciences, Port Washington, NY) for analysis at a rate of 1.0LPM. The filter was then extracted by sonicating for 30 min with 3 mL acetonitrile and brought to a final volume of 250 μL for analysis by ESI-FTMS with a 7T electrospray ionization Fourier transform ion cyclotron resonance (ESIFTICR) mass spectrometer (Bruker, Billerica, MA, Model: Apex-Qe) for accurate and precise mass measurements. Two 100 μL aliquots were taken from each extract and spiked with 33 μL of either a 4 μM solution of ammonium hydroxide, for negative ion mode, or DI water, for positive ion mode, prior to analysis. Peaks with a signal-to-noise ratio greater than 5 and an intensity relative to the base peak greater than 1% were characterized. Peak assignments were restricted to CHO compounds as (M+H)+ or (M+Na)+ ions in positive mode, or (M−H)− ions in negative mode. Allowed O/C ratios were equal to or less than 2.00, and allowed H/C ratios were between 0.4 and 2N+2 hydrogen atoms per carbon atom (N = number of carbon atoms). Formula assignments were only made if the error was less than 5 ppm ((assigned mass − measured mass)/measured mass). The remaining portion of each extract was later combined and the sample volume was 2231
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
experiments is to determine the end point of SOA aging rather than to simulate a typical atmospheric exposure. Control experiments were performed to study the role of gas phase species in aerosol formation. Plug injections of fresh SOA were made by passing the aerosol through a HEPA filter before entering the PC. The HEPA filter removed particles from the air stream but allowed volatile and semivolatile gas phase species to pass directly into the PC. With the UVC lights off (no OH in the PC), a small amount of aerosol was formed presumably due to reaction of residual α-pinene and ozone. However, when the UVC lights were turned on, all of this aerosol disappeared and, importantly, no new aerosol mass was produced. Evidently, the OH concentration in the PC is so high that gas phase oxidation occurs very quickly, which leads to the formation highly volatile, low molecular weight products that do not create aerosol. However, aerosol mass is observed to exit the PC when fresh SOA particles are allowed to enter the PC (i.e., HEPA filter removed). Therefore, the aged aerosol characterized in this work should be regarded as being produced by particle phase reaction of OH with nonvolatile oligomers23 in the fresh SOA.
adjusted to 3 mL for FTMS-MS analysis. The combined sample was split into two aliquots and spiked (25% of aliquot volume) for positive or negative ion analysis as above. Each odd m/z peak in the mass spectrum was then isolated and, if the peak intensity was sufficient, fragmented by infrared multiphoton dissociation (IRMPD). The fragmentation spectra were then combined and analyzed as previously described6 to assign molecular formulas, and identify neutral losses. The hydroxyl radical concentration in the PC was estimated by feeding a known, excess concentration of SO2 in air into the chamber, along with ozone and water vapor as in part B of Figure 1 (with the SO2 cylinder replacing the FTR because no SOA was involved in this experiment) and using the SMPS to monitor the rate at which aerosol mass is formed. Hydroxyl radical reacts with SO2 to form H2SO4 with a second-order rate constant29 of 9.0 × 10−13 cm3molecule−1s−1. By feeding SO2 in excess relative to hydroxyl, and assuming all of the sulfuric acid formed aerosol with negligible wall loss, the concentration of hydroxyl radical is given by: d[H 2SO4 ] = k′[OH·] dt
■
RESULTS AND DISCUSSION Time Dependence of SOA Aging. The plug injection setup in part A of Figure 1 was used to investigate several timedependent characteristics of SOA aging in the PC. Three plugs of aerosol were analyzed after passing through the PC with the lamps off (no OH production and hence no aging of fresh SOA), and five plugs were analyzed with the lamps on (OH causes fresh SOA to age). Each set of plugs was averaged to give the data discussed below. Figure S2 of the Supporting Information shows a plot of particle mass concentration versus time after injection into the PC with the lamps off, which provides a measure of the aerosol residence time inside the PC. Essentially, the entire particle mass has exited the PC after 60 min. When the lights are turned on, the mass concentration decreases more quickly with time owing to fragmentation that occurs during the aging process. Part A of Figure 2 shows the fraction of particle mass remaining (mass concentration with lights on divided by the mass concentration with lights off at the same time point) as a function of time. As the particle number size distribution was essentially identical for both fresh and aged SOA (∼55−60 nm geometric mean), it is assumed that wall losses were nearly identical for the two types of SOA, though this assumption was not investigated further. Aged SOA mass is lost at a slower rate at the beginning of the process than late in the process. For example, it takes about 10 min to lose 20% of the particle mass near the beginning of the experiment whereas it takes only about 5 min to lose the last 20% near the end of the experiment. After 42 min, nearly all of the SOA mass has reacted away. Concurrent with mass loss by fragmentation, functionalization also occurs as the aging process proceeds. Part B of Figure 2 shows the average O/C ratio of particles analyzed by NAMS (total moles O/total moles C for each two minute period, as determined by the deconvolution method described elsewhere27) as a function of time. With the lights off, the O/C ratio remains constant around 0.5. With the lights on, the O/C ratio steadily increases. As the NAMS measures the total amount of O and C in a particle, water could potentially cause an artifact, but previous experiments27 have shown that very little particulate water survives the aerosol sampling process.
where the change in sulfuric acid concentration with time is given by the measured change in aerosol mass with time, and k′ is equal to the second-order rate constant multiplied by the SO2 concentration. The aerosol mass was measured with the SMPS assuming a particle density of 1.84g/mL. A diffusion dryer was inserted inline before the SMPS inlet to remove water from the aerosol. Part A of Figure S1 of the Supporting Information shows plots of aerosol mass concentration exiting the PC (as determined by SMPS) versus time after the onset of SO2 flow into the chamber for a variety of reaction conditions (varying the relative humidty and UV intensity while maintaining the same starting concentration of O3). Three of the experiments were performed with all four UVC lamps on (high light intensity) and the relative humidity set to 10, 25, and 5%, respectively. One experiment was performed with only two of the UVC lamps on (low light intensity) and 22% RH. The plots are linear with slopes related to the hydroxyl concentration for each set of conditions. Part B of Figure S1 of the Supporting Information shows a plot of decrease of measured ozone concentration in the PC versus hydroxyl concentration. The plot is linear suggesting that with this experimental setup the hydroxyl concentration can be inferred from the drop in ozone concentration. On the basis of the measured drop in ozone concentration (10 ppmv), the hydroxyl concentration in SOA aging experiments was maintained at 6 × 1010 molecules/cm3. The hydroxyl radical concentration measurement is based on the assumption that the aerosol is 100% un-neutralized H2SO4. If, for example, the background ammonia concentration was large enough to produce fully neutralized (NH4)2SO4, the calculated hydroxyl concentration would be 4.5 × 1010 molecules/cm3. The 10ppmv drop in [O3] could be achieved in several ways, for example by setting the relative humidity to 10% with all four lights on full strength (low RH, high UV−C conditions) or by setting the relative humidity to 25% with only two of the four lights on and the power reduced by 50% with a variable auto transformer (Powerstat, Superior Electric Co., Bristol Ct., #PN116B). In an aging experiment with a residence time of 10 min in the PC, the total hydroxyl radical exposure would be ∼3.6 × 1013 molecules-s/cm3, which is higher than what has been used in other work,2,29,30 but the goal of these 2232
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
particulate matter having the same time point for aging to be collected on the filter over time. For the conditions used in the continuous injection experiments, the mass concentration of aged SOA (lights on) was reduced to about 50% of the corresponding value for fresh SOA (lights off). A total of 9 aged SOA samples (6 at 10% RH, 3 at 25%RH as described in the experimental section) and 6 fresh SOA samples (3 each at 10% and 25% RH) were collected. Each sample was analyzed individually by ESI-FT-ICR-MS and molecular formulas were assigned to the detected m/z peaks using a method described previously.6,33 The O/C and H/C ratios for each assigned formula were determined and the mass and intensity weighted averages33 of these values were calculated for the each sample. A plot of the average O/C and H/C ratios from the negative ion spectra of each sample is given in Figure 3. Aged SOA samples showed significantly
Figure 3. Average H/C and O/C ratios (mass and intensity weighted) from the negative ion mass spectra for each sample analyzed. Square markers are aged SOA samples and triangles are fresh SOA samples. A linear fit to all of the SOA samples is shown.
higher O/C ratios and lower H/C ratios than fresh SOA samples. Although all aged SOA samples were obtained with the same nominal hydroxyl concentration in the PC (experimental section), the samples aged under higher RH tended to give higher O/C ratios (and lower H/C ratios) than those aged under lower RH indicating that aqueous reactions may be of some importance. High vs low RH averages are given in Table S1 of the Supporting Information. The average O/C ratios observed for aged samples (generally 0.6 − 0.7) suggest an effective reaction time of ∼10−15 min based on the plot in part B of Figure 2. The O/C ratios measured in this work are somewhat lower than those observed for the reaction of αpinene in potential aerosol mass (PAM) experiments.29,30 At first glance, this may seem surprising considering that the OH radical concentration in the PC is approximately 1 order of magnitude higher than typically used in PAM experiments. However, it should be emphasized that aerosol in a PAM experiment is produced by reaction of OH with gas phase species. In the PC experiments described here, gas phase species do not lead to aerosol formation (experimental section), presumably because gas-phase oxidation is so fast that only highly volatile, low molecular weight products are formed. Because aerosol aging in the PC occurs in the particle phase, the lower O/C ratios measured here relative to PAM experiments implies that particle phase oxidation proceeds at a much slower rate than gas phase oxidation. The change in O/C and H/C ratios between a reactant and product can be used to infer what reaction has taken place. For
Figure 2. (A) Total suspended particle mass (TSP) of aged SOA relative to fresh SOA as a function of residence time in the PhARAOH Chamber. (B) O/C ratio for aged (squares) and fresh (triangles) SOA as a function of residence time. (C) O/C ratio vs % of particle mass lost from the data in parts (A) and (B) for aged SOA.
Part C of Figure 2 combines the data in parts A and B of Figure 2 by plotting O/C ratio vs percentage of mass lost during aging. The percentage of mass lost at a specific time is given by: (TSPagedSOA − TSPfreshSOA )/TSPfreshSOA
where TSP is the total suspended particle mass at each time point as measured by the SMPS. The O/C ratio increases almost linearly with percentage of mass lost to a value of about 1.5, followed by a discontinuous jump toward 2.0 right at 100% mass lost. An O/C ratio of 1.5 could correspond to glyoxal oligomers,31 whereas an O/C ratio of 2.0 could correspond to oxalate32 that subsequently decomposes to the oxidation end point of CO2. The plots in Figure 2 illustrate concurrent functionalization and fragmentation of fresh SOA as it ages. Molecular Characterization of Aged SOA. Because of the need to collect sample over a long period of time, molecular characterization experiments were performed with the continuous injection setup in part B of Figure 1. This setup allowed 2233
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
instance, replacing a hydrogen on an aliphatic carbon with a hydroxyl or peroxide group results in a slope (ΔH/C)/(ΔO/ C) of zero, replacing it with a carboxylic acid group yields a slope of −1.0, and replacing it with a ketone or aldehyde gives a slope of −2.0 (all examples of functionalization reactions). For fragmentation reactions (where a portion of the compound is lost and replaced by an aldehyde), the loss of a ketone gives a slope of −1.33 and the loss of a carboxylic acid gives a slope of +0.57. For the negative ion data in Figure 3, H/C and O/C were strongly anticorrelated (correlation coefficient = −0.951) and found to have a slope of −1.2, which is somewhat larger than a slope of −1 reported previously using the Aerosol Mass Spectrometer (AMS).5 This difference may be a consequence of compound dependent detection by ESI. If positive ion ESI data were included along with negative ion data in Figure 3, a linear regression of the combined data sets would give a slope close to −1, because the positive ion data show little correlation between H/C and O/C ratios (correlation coefficient = −0.526). A large slope is observed in negative ion ESI because this mode favors the detection of compounds containing acid functional groups, which are being formed during the aging process; in contrast, hardly any change is observed in the positive ion mode as it favors detection of compounds containing only aldehyde and ketone functionalities, which disappear as aging proceeds.6,10 Figures 4−6 provide molecular insight into the bulk chemical changes shown in Figures 2 and 3. Parts A and B of Figure 4
Figure 5. van Krevelen plots for (A) fresh and (B) aged SOA. Marker size is proportional to relative intensity of the corresponding ion. Blue and red highlights are drawn as an aid to the eye. The red line has a slope of −1.2 (Figure 3).
show representative negative ion mode mass spectra, for fresh and aged SOA samples respectively. A complete list of assigned molecular formulas and relative ion intensities is included in Table S2 of the Supporting Information. The fresh SOA spectrum in part A of Figure 4 is similar to those discussed previously for α-pinene ozonolysis.12,21,33 Ions corresponding to monomer, dimer, and trimer oxidation products are clearly evident in the spectrum. The low oxygen content of these compounds is indicated by color coding of the ions based on O/C ratio of the assigned molecular formula. The aged SOA spectrum in part B of Figure 4 is completely different. Monomer and oligomer regions have coalesced, and the O/C ratios of the assigned molecular formulas have increased. Functionalization during aging is indicated by the appearance of molecular products with high oxygen contents. Fragmentation is indicated by the decrease of oligomers relative to monomerlike products. Not surprisingly, the most intense highly oxidized compounds reside in the monomer region of the aged SOA spectrum, which presumably have been produced by a combination of functionalization and fragmentation of oligomers. Two molecular formulas are of particular note: C5H5O6− and C10H13O7− were detected in all aged SOA samples with average relative intensities of ∼20% and 30% respectively and have been reported as prominent ambient negative ions in Hyytiälä, Finland.34 These and other ions reported by Ehn et al.34 in both field and laboratory studies are listed in Table S3 of the Supporting Information along with
Figure 4. Example negative ion mode mass spectra for (A) fresh SOA (average O/C = 0.44, H/C = 1.41), and (B) aged SOA (average O/C = 0.61, H/C = 1.20). 2234
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
of tandem mass spectrometry.6,35 Accurate mass measurement of precursor and product ions by FTMS permits the molecular formulas of neutral losses to be determined. These neutral losses can then be associated with specific functional groups. Tandem mass spectrometry experiments were performed as described previously6 for 75 high intensity precursor ions in the fresh SOA spectrum and 50 high intensity precursor ions in the aged SOA spectrum. Part A of Figure 6 shows the percentage of precursor ions from each spectrum that showed a specific neutral loss. Precursor ions from fresh SOA were more likely to lose low O/C ratio molecular fragments than aged SOA, and precursor ions from aged SOA were more likely to lose high O/ C ratio molecular fragments than fresh SOA. The results suggest that fresh SOA contains aliphatic and carbonyl functionalities (indicated by loss of CH2, C2H4, CH2O, and C3H6O) that are converted to acids (indicated by losses of CO2, CH2O2, and C2H4O2) and peroxy acids (indicated by loss of CH2O3) upon aging. Accurate mass measurements of product ions also give insight into the composition of oligomer building blocks in fresh vs aged SOA. Part B of Figure 6 shows the fraction of precursor ions in the fresh and aged SOA that gave specific product ions. Many precursor ions in the fresh SOA spectrum gave product ions indicating the presence of known monomer compounds such as terpenylic acid (C8H11O4−), norpinonic acid (C9H13O3−), pinic acid (C9H13O4−), and pinonic acid (C10H15O3−). Each is a significant product of α-pinene ozonolysis and a known or expected building block for oligomers.6 However, these product ions were rarely detected in the spectra of precursor ions from aged SOA. Instead, product ions from the aged SOA spectra tended to have molecular formulas with a larger number of oxygen atoms and fewer number of carbon atoms. One noteworthy product ion corresponds to α,α-dimethyltricarballylic acid (C8H11O6−), which has been reported as an aging product of cis-pinonic acid1,2,36 indicating that the monomer building blocks for oligomers in aged SOA are themselves products of both functionalization and fragmentation of monomer building blocks in fresh SOA. Parts A and B of Figure 6 show the most relevant neutral losses and product ions. All neutral losses and product ions are summarized in Figure S4 of the Supporting Information. An important feature of the experiments described here is the very high OH concentration used for aging (∼6 × 1010 molecules/cm3). Donahue et al.36 have pointed out that gas phase oxidation typically proceeds at a rate of 2−5 times faster than condensed phase oxidation. Together, the high OH concentration and fast gas-phase reaction rate cause rapid oxidation of gas-phase precursors to highly volatile, low molecular weight species that do not contribute to aerosol formation. Therefore, the aged aerosol characterized in this work is mainly the result of particle phase oxidation of nonvolatile oligomers. A remarkable aspect of oligomer aging in these experiments is the slow rate at which oxidation proceeds. For example, the continuous flow experiment corresponds to about 5000 h of exposure at an atmospherically relevant OH concentration of 2 × 106 cm−3, which is much longer than the atmospheric lifetime of SOA. It is possible that the slow oxidation rate in these experiments reflects hindered diffusion in the particle phase. If so, important topics for future investigation are the roles of particle phase composition and morphology in determining condensed phase reaction rates,
Figure 6. Percentage of precursor ions that give (A) a small molecule neutral loss, or (B) a monomer size product ion, for fresh (white, n = 75) and aged (black, n = 50) SOA.
how many aged SOA samples they were detected in and their average intensity. There was little to no difference in the intensity of these peaks between SOA aged under low and high RH conditions. In all, 75% of the monomer and 33% of the dimer molecular formulas reported in Ehn et al. had matches in the aged SOA spectrum, with lower MW monomers and dimers more likely to have a match. Parts A and B of Figure 5 show the molecular products of fresh and aged SOA as van Krevelen plots. For fresh SOA, most of the signal intensity resides in the blue highlighted region of the plot that encompasses molecular formulas of known monomer compounds and their expected oligomer products.6,7 For aged SOA, almost all of the intensity is gone from blue region and new ions appear with assigned formulas having higher O/C and lower H/C ratios (red highlighted region). The red line is identical on each graph and shows the slope of −1.2 from Figure 3 passing through the largest point in part A of Figure 5 (m/z 357.154, assigned as C17H25O8−) and shows movement from the blue area to the red. These new products formed in aged SOA are evenly distributed across the entire m/ z range of the spectrum. Recently, it has been suggested that the carbon oxidation state is a better indicator of SOA chemistry than O/C ratio.9 As shown in Figure S3 of the Supporting Information, the carbon oxidation states for assigned formulas in fresh SOA generally range between −1 and 0, whereas those for aged SOA range between 0 and +2. Whereas there is no systematic variation of carbon oxidation state with product molecular weight for fresh SOA, the highest oxidation states for aged SOA tend to be associated with the smallest molecules (∼10 or fewer carbon atoms). The functional groups associated with this change in carbon oxidation state during aging can be investigated through the use 2235
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
(8) 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 (36), 8005−8014. (9) 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 (2), 133−139. (10) Nizkorodov, S. A.; Laskin, J.; Laskin, A. Molecular chemistry of organic aerosols through the application of high resolution mass spectrometry. Phys. Chem. Chem. Phys. 2011, 13 (9), 3612−3629. (11) Guenther, A.; Hewitt, C. N.; Erickson, D.; Fall, R.; Geron, C.; Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.; McKay, W. A.; Pierce, T.; Scholes, B.; Steinbrecher, R.; Tallamraju, R.; Taylor, J.; Zimmerman, P. A global model of natural volatile organic compound emissions. J. Geophys. Res. 1995, 100 (D5), 8873−8892. (12) Heaton, K. J.; Sleighter, R. L.; Hatcher, P. G.; Hall, W. A., IV; Johnston, M. V. Composition domains in monoterpene secondary organic aerosol. Environ. Sci. Technol. 2009, 43 (20), 7797−7802. (13) Kalberer, M.; Sax, M.; Samburova, V. Molecular size evolution of oligomers in organic aerosols collected in urban atmospheres and generated in a smog chamber. Environ. Sci. Technol. 2006, 40 (19), 5917−5922. (14) Canagaratna, M. R.; Jayne, J. T.; Jimenez, J. L.; Allan, J. D.; Alfarra, M. R.; Zhang, Q.; Onasch, T. B.; Drewnick, F.; Coe, H.; Middlebrook, A.; Delia, A.; Williams, L. R.; Trimborn, A. M.; Northway, M. J.; DeCarlo, P. F.; Kolb, C. E.; Davidovits, P.; Worsnop, D. R. Chemical and microphysical characterization of ambient aerosols with the aerodyne aerosol mass spectrometer. Mass Spectrom. Rev. 2007, 26 (2), 185−222. (15) 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 (10), 4625−4641. (16) Zhang, Q.; Worsnop, D. R.; Canagaratna, M. R.; Jimenez, J. L. Hydrocarbon-like and oxygenated organic aerosols in pittsburgh: Insights into sources and processes of organic aerosols. Atmos. Chem. Phys. 2005, 5 (12), 3289−3311. (17) Volkamer, R.; Jimenez, J. L.; San Martini, F.; Dzepina, K.; Zhang, Q.; Salcedo, D.; Molina, L. T.; Worsnop, D. R.; Molina, M. J. Secondary organic aerosol formation from anthropogenic air pollution: Rapid and higher than expected. Geophys. Res. Lett. 2006, 33 (17), L17811. (18) Zhang, Q.; Jimenez, J. L.; Canagaratna, M. R.; Allan, J. D.; Coe, H.; Ulbrich, I.; Alfarra, M. R.; Takami, A.; Middlebrook, A. M.; Sun, Y. L.; Dzepina, K.; Dunlea, E.; Docherty, K.; DeCarlo, P. F.; Salcedo, D.; Onasch, T.; Jayne, J. T.; Miyoshi, T.; Shimono, A.; Hatakeyama, S.; Takegawa, N.; Konodo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S.; Weimer, S.; Demejian, K.; Williams, P.; Bower, K.; Bahreini, R.; Cottrell, L.; Griffin, R. J.; Rautiainen, J.; Sun, J. Y.; Zhang, Y. M.; Worsnop, D. R. Ubiquity and dominance of oxygenated species in organic aerosols in anthropogenically-influenced northern hemisphere midlatitudes. Geophys. Res. Lett. 2007, 34, L13801. (19) Tolocka, M. P.; Heaton, K. J.; Dreyfus, M. A.; Wang, S.; Zordan, C. A.; Saul, T. D.; Johnston, M. V. Chemistry of particle inception and growth during α-pinene ozonolysis. Environ. Sci. Technol. 2006, 40 (6), 1843−1848. (20) Tolocka, M. P.; Saul, T. D.; Johnston, M. V. Reactive uptake of nitric acid into aqueous sodium chloride droplets using real-time single-particle mass spectrometry. J. Phys. Chem. A 2004, 108 (14), 2659−2665. (21) Gao, Y.; Hall, W. A.; Johnston, M. V. Molecular composition of monoterpene secondary organic aerosol at low mass loading. Environ. Sci. Technol. 2010, 44 (20), 7897−7902.
and whether or not the physical characteristics of laboratory aerosol match those of ambient aerosol.
■
ASSOCIATED CONTENT
S Supporting Information *
Four figures and three tables. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
DISCLAIMER. Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendations or endorsement by the U.S. Department of Agriculture. The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by the National Science Foundation under grant number CHE1110554. The authors thank Derrick Allen for his contributions to the experimental apparatus.
■
REFERENCES
(1) 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.; Prevot, 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. (2) 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.; Dunlea, E. 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 (5959), 1525−1529. (3) Johnson, D.; Marston, G. The gas-phase ozonolysis of unsaturated volatile organic compounds in the troposphere. Chem. Soc. Rev. 2008, 37 (4), 699−716. (4) 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. (5) Ng, N. L.; Canagaratna, M. R.; Jimenez, J. L.; Chhabra, P. S.; Seinfeld, J. H.; Worsnop, D. R. Changes in organic aerosol composition with aging inferred from aerosol mass spectra. Atmos. Chem. Phys. Discuss 2011, 11 (3), 7095−7112. (6) Hall, W.; Johnston, M. Oligomer formation pathways in secondary organic aerosol from ms and ms/ms measurements with high mass accuracy and resolving power. J. Am. Soc. Mass Spectrom. 2012, 23 (6), 1097−1108. (7) Yu, J.; Cocker, D. R.; Griffin, R. J.; Flagan, R. C.; Seinfeld, J. H. Gas-phase ozone oxidation of monoterpenes: Gaseous and particulate products. J. Atmos. Chem. 1999, 34, 207−258. 2236
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
Environmental Science & Technology
Article
(22) Gao, Y.; Johnston, M. V. Online deposition of nano-aerosol for matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2009, 23 (24), 3963−3968. (23) Hall, W. A.; Johnston, M. V. The thermal-stability of oligomers in alpha-pinene secondary organic aerosol. Aerosol Sci. Technol. 2012, 46 (9), 983−989. (24) Heaton, K. J.; Dreyfus, M. A.; Wang, S.; Johnston, M. V. Oligomers in the early stage of biogenic secondary organic aerosol formation and growth. Environ. Sci. Technol. 2007, 41 (17), 6129− 6136. (25) Tolocka, M. P.; Jang, M.; Ginter, J. M.; Cox, F. J.; Kamens, R. M.; Johnston, M. V. Formation of oligomers in secondary organic aerosol. Environ. Sci. Technol. 2004, 38 (5), 1428−1434. (26) Wang, S.; Zordan, C. A.; Johnston, M. V. Chemical characterization of individual, airborne sub-10-nm particles and molecules. Anal. Chem. 2006, 78 (6), 1750−1754. (27) Zordan, C. A.; Pennington, M. R.; Johnston, M. V. Elemental composition of nanoparticles with the nano aerosol mass spectrometer†. Anal. Chem. 2010, 82 (19), 8034−8038. (28) Zordan, C. A.; Wang, S.; Johnston, M. V. Time-resolved chemical composition of individual nanoparticles in urban air. Environ. Sci. Technol. 2008, 42 (17), 6631−6636. (29) Kang, E.; Root, M. J.; Toohey, D. W.; Brune, W. H. Introducing the concept of potential aerosol mass (pam). Atmos. Chem. Phys. 2007, 7 (22), 5727−5744. (30) Massoli, P.; Lambe, A. T.; Ahern, A. T.; Williams, L. R.; Ehn, M.; Mikkilä, J.; Canagaratna, M. R.; Brune, W. H.; Onasch, T. B.; Jayne, J. T.; Petäjä, T.; Kulmala, M.; Laaksonen, A.; Kolb, C. E.; Davidovits, P.; Worsnop, D. R. Relationship between aerosol oxidation level and hygroscopic properties of laboratory generated secondary organic aerosol (soa) particles. Geophys. Res. Lett. 2010, 37 (24), L24801. (31) Hastings, W. P.; Koehler, C. A.; Bailey, E. L.; De Haan, D. O. Secondary organic aerosol formation by glyoxal hydration and oligomer formation: Humidity effects and equilibrium shifts during analysis. Environ. Sci. Technol. 2005, 39 (22), 8728−8735. (32) Tan, Y.; Carlton, A. G.; Seitzinger, S. P.; Turpin, B. J. Soa from methylglyoxal in clouds and wet aerosols: Measurement and prediction of key products. Atmos. Environ. 2010, 44 (39), 5218−5226. (33) Hall, W. A.; Johnston, M. V. Oligomer content of α-pinene secondary organic aerosol. Aerosol Sci. Technol. 2011, 45 (1), 37−45. (34) Ehn, M.; Kleist, E.; Junninen, H.; Petäjä, T.; Lönn, G.; Schobesberger, S.; Dal Maso, M.; Trimborn, A.; Kulmala, M.; Worsnop, D. R.; Wahner, A.; Wildt, J.; Mentel, T. F. Gas phase formation of extremely oxidized pinene reaction products in chamber and ambient air. Atmos. Chem. Phys. 2012, 12 (11), 5113−5127. (35) LeClair, J. P.; Collett, J. L.; Mazzoleni, L. R. Fragmentation analysis of water-soluble atmospheric organic matter using ultrahighresolution ft-icr mass spectrometry. Environ. Sci. Technol. 2012, 46 (8), 4312−4322. (36) 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 (34), 13503−13508.
2237
dx.doi.org/10.1021/es303891q | Environ. Sci. Technol. 2013, 47, 2230−2237
