Formation and aging of secondary organic aerosol from toluene : changes in chemical composition , volatility , and hygroscopicity

Secondary organic aerosol (SOA) is transformed after its initial formation, but this chemical aging of SOA is poorly understood. Experiments were conducted in the Carnegie Mellon environmental chamber to form secondary organic aerosol (SOA) from the photo-oxidation of toluene and other small aromatic volatile organic compounds (VOCs) in the presence of NOx under different oxidizing conditions. The effects of the oxidizing condition on organic aerosol (OA) composition, mass yield, volatility, and hygroscopicity were explored. Higher exposure to the hydroxyl radical resulted in different OA composition, average carbon oxidation state (OSc), and mass yield. The OA oxidation state generally increased during photo-oxidation, and the final OA OSc ranged from −0.29 to 0.16 in the performed experiments. The volatility of OA formed in these different experiments varied by as much as a factor of 30, demonstrating that the OA formed under different oxidizing conditions can have a significantly different saturation concentration. There was no clear correlation between hygroscopicity and oxidation state for this relatively hygroscopic SOA.


15
Secondary organic aerosol (SOA) is produced when gas-phase precursors are oxidized, forming 16 lower volatility products that partition to the condensed phase. As SOA is estimated to account for 17 approximately 70% of total aerosol organic carbon mass (Hallquist et al., 2009), the influence of 18 SOA on aerosol composition and related properties is important and complex (Donahue et al., 19 We investigated the relationship between oxidation, volatility and hygroscopicity of SOA formed 1 from the photo-oxidation of toluene (methylbenzene) and other small aromatic volatile organic 2 compounds (VOCs) under a variety of oxidation conditions. Small aromatics are important 3 anthropogenic SOA precursors (Pandis et al., 1992;Vutukuru et al., 2006), and toluene serves as 4 a model system to study the formation of SOA from these compounds. A main objective of our 5 work is to connect the extent of oxidation and the changes in volatility of these experiments within 6 the 2D-VBS framework. The basic sequence of the experiments was to fill the chamber with clean air, inject the VOC and 13 nitrous acid (HONO), and turn on the UV lights to start formation of OH (from the photolysis of 14 HONO), oxidation of the VOC and formation of secondary organic aerosol (SOA). The number 15 of UV lights used, the initial VOC concentrations, and the number of HONO injections was varied 16 between experiments in order to create different oxidizing environments, as summarized in Table   17 1. The amount of SOA formed, the SOA oxidation state, its volatility and its hygroscopicity were 18 then measured as explained in more detail below. 19 Nitrous acid was produced immediately before injection by drop-wise addition of 12 ml 0.1 M 20 sodium nitrite solution to 24 ml 0.05 M sulfuric acid solution. Ammonium sulfate ((NH4)2SO4, 21 Sigma Aldrich, 99.99%)) seed particles were used in some experiments (Table 1) to provide 22 surface area onto which organics would condense as SOA. In the unseeded experiments, nucleation 23 of the organic vapors was observed. With the exception of Experiment 7 (Table 1), isotopically 24 labeled toluene was used ( 13 C-toluene, Cambridge Isotope Laboratories, 99%) as in a previous 25 study (Hildebrandt et al., 2011). All six ring carbons in the labeled toluene are 13 C-substituted, 26 leaving the methyl carbon unsubstituted. In two experiments (Expts. 6 and 8) other small aromatic 27 compounds were injected in addition to toluene as detailed in Table 1 in order to test whether these 28 VOCs behave similarly as toluene. Concentrations of the VOCs were monitored using a proton-  The AMS data were processed in Igor Pro 6.12 (Wavemetrics, Inc.) using the standard AMS data 12 analysis toolkits "Squirrel" version 1.51C for unit mass resolution (UMR) analysis and "Pika" 13 version 1.10C for high resolution (HR) analysis. HR analysis was performed using the W-mode 14 data since highest resolution is preferred to distinguish between isotopically labeled and unlabeled 15 ions. The lists of ions integrated in the HR analysis is similar to the list used previously 16 (Hildebrandt et al., 2011). Ion masses were fitted up to an m/z ratio of 105; above this the signal 17 was too noisy and/or the mass spectra were too crowded for reliable identification of ion atomic 18 composition. According to the UMR analysis more than 95% of the organic signal was below m/z 19 105, and the total organic mass was corrected based on this fraction calculated for each experiment 20 ( for hydrogen atoms (m/z = 1) formed in the fragmentation of H2O (Canagaratna et al., 2015).

27
Second, the amount of H2O + attributed to organics was chosen so that the mass of water does not 28 correlate with the mass of organics (R < 0.01) as expected for these low relative humidity experiments. The ratios of organic H2O + to ( 13 CO2 + + CO2 + ) are provided in Table 2 Kostenidou et al. (2007) . 21 The amount of organic aerosol formed was quantified as the fractional aerosol mass yield (mass 22 of OA formed divided by mass of toluene reacted). The mass of OA formed was corrected for the 23 depositional loss of particles onto the chamber walls and for the condensational loss of organic 24 vapors to wall-deposited particles. The assumption was made that condensation of organic vapors 25 is not slowed by mass-transfer resistances, and that the wall-deposited particles are in equilibrium 26 with the organic vapors in suspension. Therefore, the total (corrected) concentration of OA can be 27 calculated by multiplying the OA/seed ratio by the initial seed concentration, as discussed in more In all of the experiments described here, the condensation sink after particle formation was 3 approximately 1 min -1 , while the initial condensation sink to ammonium sulfate seeds in seeded 4 experiments ranged between 0.6 and 1 min -1 . The timescale for sulfuric acid vapor wall loss in suspended aerosol population quickly grew to a size where this was also true. Thus, while wall 10 losses of semi-volatile vapors are a source of uncertainty, we have attempted to ensure that vapor- 11 particle interactions at least dominate over vapor-wall interactions. It is therefore reasonable to 12 expect that observed differences between experiments are not driven by interactions with the walls 13 but instead by chemical processing of the organic aerosol. Volatility data were collected for each experiment after the SOA had formed. During some 18 experiments, measurements were also made during the irradiation period (with the UV lights on) 19 to examine the volatility changes during photo-oxidation. TD data are analyzed in terms of Volume   Particle losses in the TD were also taken into account. These losses occur due to diffusion 3 (primarily of small particles), sedimentation (primarily of large particles), and thermophoresis; the 4 losses are therefore a function of sample flow rate, temperature, and particle size (Burtscher et al.,  The number losses for each TD temperature -residence time combination are calculated by 8 determining the losses over the size distribution measured by the SMPS. The number losses for 9 each size bin are then converted to a volume-based correction using the particle diameter of each 10 bin. This correction factor is applied to the calculated MFR values. The organic MFR was 11 calculated from AMS bypass and thermodenuder mass concentrations averaged over 6-9 minutes 12 for a given TD temperature and residence time. It is assumed that there are no significant changes 13 to composition and volatility over these averaging periods.  Table 1. Briefly, aerosol evaporation is simulated using experimental 18 inputs including TD temperature, residence time, particle mode diameter, mass concentration, and  The volatility of SOA formed in each experiment was determined assuming a fixed volatility 29 distribution shape using four saturation concentrations, 1, 10, 100, and 1000 µg m -3 . During the analysis the saturation concentrations are multiplied by a shifting factor, s. This practically shifts 1 the volatility distribution to lower or higher values assuming that the shape of the distribution does 2 not change. Differences in shifting factors can then be interpreted as differences in the saturation 3 vapor pressure of the OA formed in these experiments: for example, OA with shifting factor 10 is 4 ten times more volatile than the OA with shifting factor 1. The volatility distribution used is the The CCNc instrument calibration is used to determine the relationship between instantaneous 19 instrument flow rate and supersaturation as described in Moore et al. (2010). Ammonium sulfate 20 solution is atomized, dried using a silica gel diffusion dryer, charge-neutralized using Po-210, and 21 classified by a DMA. The flow is then introduced into both a CPC and a CCNc. The activation 22 ratio, or the ratio of CCN to total particles, is then plotted against the instantaneous flow rate to 23 yield data that are fit to a sigmoidal activation ratio function. The critical flow rate, Q*, is 24 determined, corresponding to where half of the total particles are activated and to a level of 25 supersaturation, s*, equal to the critical supersaturation of the classified aerosol (Section 2.4.2).

26
The Q* and s* are determined for a range of aerosol sizes, yielding, for the flow rate range (0.1-27 0.9 L min -1 ) and temperature gradient (∆T=6°C) in the CCNc column, supersaturations ranging 28 from approximately 0.1 to 0.5%.
where A=(4Mwσw)/(RTρw); Mw, σw, and ρw are the molar mass, surface tension, and density of water, 8 respectively. R is the universal gas constant, T is CCNc mid-column temperature, and dp is the dry 9 particle diameter selected by the DMA prior to the CCNc. concentrations are also shown for Expt. 9; these data were not available for Expt. 2. There were 16 two photo-oxidation periods ("lights on") during each experiment, and HONO was injected every 17 time before lights were turned on. During experiment 2, the OA was alternatively passed through 18 the bypass and the TD throughout the experiment (only the bypass data are shown in Fig 2). The

19
TD was held at the same temperature during the photo-oxidation periods to observe changes in 20 volatility during this period, and the temperature in the TD was varied during the dark period to 21 obtain a thermogram. During experiment 9, the OA was passed only through the bypass during 22 photo-oxidation, and it was alternated between bypass and TD (at different temperatures) during 23 the dark period. The OA concentrations increased during the oxidation periods (lights on) as 24 toluene was oxidized to form SOA. Toluene concentrations decreased by approximately 50% 25 during the first photo-oxidation period in Expt. 9 and decreased an additional 20% during the 26 second photo-oxidation period. Thus, gas-phase toluene is always present in the system and fresh toluene SOA is expected to form at the same time as the previously formed toluene SOA is aged 1 photo-chemically. 2 A moderate increase in the OA oxidation state was observed during the "lights on" period in both 3 experiments. Two competing effects can influence the OA oxidation state: First, according to 4 partitioning theory, species of increasingly higher volatility will partition to the particle phase as work, suggesting that the aging effect dominated.

11
In an attempt to produce highly oxidized OA, photo-oxidation of the OA was continued for over 12 24 hours during experiments 3 and 5. Figure 3 shows the time series of oxidation state and 13 elemental ratios (O:C and H:C) for Expt. 3. As before, HONO was injected before every irradiation 14 period. The OA oxidation state increased during the first few hours of irradiation but significantly 15 decreased after longer irradiation. Plausible explanations for this decrease in oxidation state (which 16 was also observed after long irradiation in experiment 5) include the condensation of less oxidized 17 vapors, photolysis of OA components and their fragmentation after continued oxidation with OH.

18
Fragmented products may have a high oxidation state but high volatility (due to their smaller size) 19 and evaporate from the OA, decreasing the OA average oxidation state. Photolysis of organic 20 compounds is expected to occur throughout the experiment (e.g. Surratt et al., 2006), but as long 21 as OH reactions dominate the oxidation state of the bulk OA increases. Future experiments should 22 aim to isolate OH from photolysis reactions by, for example, using a dark OH source. This would 23 help to constrain these effects and eventually represent them in chemical transport models.
24 experiments (number 2 and 4). The aim in the design of these experiments was to create different 10 photochemical conditions. Therefore, fewer HONO injections were performed, more toluene was 11 injected, fewer lights were used (resulting in lower UV intensity), and the irradiation period was 12 shorter in Expt. 7 compared to Expt. 9 (10 minutes and 3 hours, respectively). The decay of toluene, 13 monitored by the PTR-MS, was used to estimate the OH exposure of the OA during irradiation.
14 Total OH exposure during Expt. 7 was 7-8 times lower than during Expt. 9. Figure 4 shows the 15 OA mass yields for experiments 7 and 9 as a function of the corrected OA concentration in the 16 system; only the first irradiation period was used for Expt. 9 as uncertainties due to wall losses 17 increase over the course of an experiment. The OA yields are higher for Expt. 9, which exhibited 18 higher OH exposure. Considering only the first irradiation period, OH exposure in Expt. 9 was 19 five times higher than in Expt. 9. The OA formed in Expt. 9 also exhibited significantly higher 20 oxidation state (~0, Table 2) than the OA formed in Expt. 7 (~-0.3), and its volatility was 21 approximately a factor of seven lower than the volatility of the OA formed in Expt. 7 (Table 2).
Previous work has suggested that higher OH exposures help to reduce wall losses (Kroll et al., (around zero), the volatility of the OA does not correlate significantly with bulk oxidation state. 5 The volatility and oxidation state of OA formed in all experiments is analyzed further below.  The 15 and 25 s residence time datasets were independently modeled, resulting in two estimates 20 of volatility reduction for each experiment. These were quite similar ( MFRs is shown in Figure S3.

28
Sensitivity runs were performed in order to examine the effects of the accommodation coefficient 29 and enthalpy of vaporization parameters. In summary, the analysis revealed that changing the mass accommodation coefficients between 0.01, 0.1 and 1 for Expts. 7 and 9, which exhibited quite 1 different experimental conditions, changes neither the relative volatility reduction nor the 2 goodness of fit (represented by the sum of squared residuals, SSR) by more than 15%. In addition,  reduction. This is consistent with the assumption of a constant volatility distribution shape shifting. 13 Each data point represents a single experiment in terms of volatility reduction or the change in log 14 C* and the corresponding OSc. In general, the more oxidized organic aerosol is less volatile. This 15 is consistent with functionalization reactions decreasing the volatility of the OA as it is oxidized. 16 Using a least squares fit, a straight line is fit to the dataset giving a relation of (OSc) = 0.284 17 (∆log10 C*) -0.245. This suggests that an increase of the oxidation state by approximately 0.3 18 units corresponds to a reduction of the average volatility by an order of magnitude for the toluene 19 SOA system examined here. However, as discussed in section 3.2, the volatility of individual 20 species composing the OA is not always correlated to their oxidation state. O:C (Fig 9 top, left) or between κorg and OSc ( Fig. 9 top, right) across all experimental conditions. This is counter to the conventional view that oxidative aging of aerosol generally increases its hygroscopicity and oxidation suggests that there may be another process, aside from bulk oxidation 8 changes, causing changes in the measured hygroscopicity. Sareen et al. (2013) showed that gas-9 phase compounds such as methylglyoxal can act as surfactants, which depress surface tension and 10 enhance CCN-activity (and hygroscopicity). As methylglyoxal as well as other gas-phase surface- yields, more oxidized SOA, and reduced SOA volatility but only modest differences in hygroscopicity. Volatility varied by a factor of 30 for different OH exposure, and a ten-fold 1 decrease in volatility was associated with a 0.3 increase in carbon oxidation state. The SOA was 2 relatively hygroscopic for organic material, with 0.1 < κ < 0.2 and if anything a slightly negative 3 relationship between kappa and oxidation state was observed, suggesting a possible role for 4 surfactants or oligomeric compounds. The relationship between hygroscopicity, oxidation state 5 and volatility may be modulated by gas-phase compounds.

6
While individual experiments with different OH exposure showed clear aging effects as different 7 oxidation states and OA volatility, these effects were not evident within every single experiment. 8 This suggests that a complex interplay exists between gas-phase processes, including oxidation 9 reactions that both functionalize and fragment condensable organic species as well as photolysis

Experiment 9
Toluene / 6 reactor was dark are shown with white background while the periods with UV-lights are shaded 1 yellow. Also shown is the OA oxidation state (right axis) and concentrations of toluene during 2 Expt. 9 (toluene concentrations have been divided by 6 on the figure for easier readability).      Table 1.  Expt 4 Non-denuded Expt 4 Thermally-denuded Expt 6 Non-denuded Expt 6 Thermally-denuded Expt 5 Non-denuded Expt 8 Non-denuded Expt 9 Non-denuded