Article pubs.acs.org/JPCA
Changes in Secondary Organic Aerosol Composition and Mass due to Photolysis: Relative Humidity Dependence Jenny P. S. Wong, Shouming Zhou, and Jonathan P. D. Abbatt* Department of Chemistry, University of Toronto, Toronto, Canada S Supporting Information *
ABSTRACT: This study is focused on the relative humidity (RH) dependence of watersoluble secondary organic aerosol (SOA) aging by photolysis. Particles containing αpinene SOA and ammonium sulfate, generated by atomization, were exposed to UV radiation in an environmental chamber at three RH conditions (5, 45, and 85%), and changes in chemical composition and mass were monitored using an aerosol mass spectrometer (AMS). Under all RH conditions, photolysis leads to substantial loss of SOA mass, where the rate of mass loss decreased with decreasing RH. For all RH conditions, the less oxidized components of SOA (e.g., carbonyls) exhibited the fastest photodegradation rates, which resulted in a more oxidized SOA after photolytic aging. The photolytic reactivity of SOA material exhibited a dependence on RH likely due to moisture-induced changes in SOA morphology or phase. The results suggest that the atmospheric lifetime of SOA with respect to photolysis is dependent on its RH cycle, and that photolysis may be an important sink for some SOA components occurring on an initial time scale of a few hours under ambient conditions.
1. INTRODUCTION Organic compounds are ubiquitous in ambient aerosol, yet their formation and transformation mechanisms in the atmosphere remain not fully characterized.1 These organic compounds are largely secondary in nature (i.e., secondary organic aerosol, SOA), produced from the condensation of oxidation products of volatile organic compounds. Away from sources, SOA is continually subjected to further chemical processing, or aging, that transforms the particulate physiochemical properties, which are important for climate, air quality, and human health. Current atmospheric models, with parametrizations based largely on gas- and gas-particle reactions studied in the laboratory, have difficulty predicting the mass and the degree of oxidation of ambient SOA, suggesting unidentified formation and aging mechanisms.2,3 Aging processes include heterogeneous oxidation, gas-phase oxidation of semivolatile components, and reactions in the condensed phase, such as photolysis, polymerization, and aqueous oxidation.3 Historically, the aging of SOA material has been investigated using SOA particles in environmental chambers or flow tube reactors, or particles collected on filters. Typically, these studies have been conducted under dry or low RH conditions. On the other hand, studies have also examined the aging of aqueous SOA solutions to investigate the role of cloud processing (e.g., saturated RH conditions). Collectively, these studies have shown that aging can affect the mass and the chemical composition of SOA, but the experimental conditions only represent a narrow range of RH conditions found in the atmosphere. Only a limited number of studies have investigated aging processes under a wide range of RH conditions where the reactivity of model organic particles was observed to be dependent on RH, as aging by ozone4 and OH radical5 were © XXXX American Chemical Society
enhanced under high RH conditions. For complex mixtures such as SOA, recent work by Zhou et al. observed that the kinetics of the heterogeneous ozonolysis of benzo[a]pyrene (BaP) within SOA material was suppressed under dry conditions, where the kinetic behavior was between that expected of solids and liquids. Additionally, it was observed that the reactivity of BaP was enhanced with increasing RH conditions. These results suggest that RH affects particle reactivity, as water uptake by hygroscopic SOA under high RH condition lowers the material’s viscosity, which allows for more rapid diffusion of reactants compared to dry conditions where SOA is semisolid.6 Indeed, there is growing experimental evidence that SOA can be an amorphous solid with high viscosity under dry/low RH conditions. With increasing RH, water acts as a plasticizer, decreasing the viscosity of the organic material (i.e., liquid SOA).7−10 In addition to the study by Zhou et al., other recent studies have shown that the phase/viscosity of SOA material can affect gas-particle partitioning11 and aging by heterogeneous OH oxidation.12 Given that past investigations have shown the importance of RH on the phase/viscosity of SOA, which has implications on particle reactivity, it is necessary to characterize how variations in RH conditions affect other aging processes. Of particular interest to the current study is the aging of SOA by photolysis. As with other SOA aging experiments, previous laboratory studies have focused on the photolysis of dry SOA particles, or Special Issue: Mario Molina Festschrift Received: July 10, 2014 Revised: September 4, 2014
A
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
2.3. Photolysis of SOA. All photolysis experiments were conducted in the University of Toronto Mobile Concentration and Aging (MOCA) chamber in batch mode. MOCA is a 1 m3 Teflon (FEP) bag mounted around a cubic Teflon-coated frame, surrounded by 24 UV lamps (Sylannia) on four of its six sides, cooled with multiple fans. The temperature and RH inside the chamber were continuously monitored (Vaisala). Following each experiment, the chamber was flushed with clean air and irradiated with UV light for a minimum of 12 h for cleaning. Prior to each experiment, the chamber was conditioned to a specific RH by flowing purified air through heated water bubblers into the chamber prior to particle introduction. To generate particles, ammonium sulfate (AS, 0.04 w/v %) was added to the SOA water extract to a total volume of 10 mL, resulting in a 3:2 inorganic-to-organic mass ratio. AS was chosen because its hygroscopic response to variations in RH is well characterized.16 This bulk solution was aerosolized into the chamber using a constant output atomizer (TSI 3076) for 20 min. From AMS measurements (described below) obtained using particle time-of-flight (PTOF) mode, the particles in the chamber had a typical mass mode of 230 nm prior to UV light exposure (size-resolved mass distribution shown in Figure S2, Supporting Information). Prior to photolysis, the typical particle mass loading in the chamber was 30 μg/m3. To study the photolysis of SOA material, the AS−SOA particles were mixed in the dark chamber for 20 min before they were exposed to UV irradiation for 33 min, following which the chamber was returned to dark conditions. Upon UV light exposure, the temperature inside the chamber increased by 1° (above room temperature), which resulted in a decrease of ∼5% in RH. It is important to note that the effects of UV light exposure can be due to volatilization (due to heating) and UV irradiation. Since this temperature increase was identical for all photolysis experiments, the contributions of volatilization to changes in SOA mass and composition are also identical. That being said, with the temperature increase so small, we believe the changes are largely due to photolysis. Two sets of experiments were conducted to characterize the photolysis of SOA material. For the first set of experiments, the photolysis of SOA by UVB light centered at 310 nm at three different RH conditions (85% (“high RH”), 45% (“mid RH”), and 5% (“low RH”)) was examined to elucidate the effects of RH. The polydisperse particles were fully deliquesced before entering the chamber, and then equilibrated to each RH condition in the chamber. Given that previous work has characterized the effects of RH on the particle phase and viscosity of SOA,9,10 we speculate that the SOA material is likely to be liquid at high RH and semisolid at low RH conditions, possibly having phase separated. At mid RH conditions, the particles were expected to have intermediate liquid water content, and the resulting phase and viscosity of the SOA material is likely to be between the values present under high and low RH conditions. For AS, it is deliquesced at high RH and effloresced at low RH conditions.16 The second set of experiments was conducted to investigate the UV wavelength dependence of SOA photolysis. Under the same high RH conditions, photolysis using UVA lamps centered at 360 nm was conducted, and compared to results obtained using UVB lamps centered at 310 nm. The wavelength dependence of the photon flux inside the chamber was measured using a spectroradiometer (StellaNet Inc.). The measured wavelength dependent photon fluxes from both UVB
SOA dissolved in bulk solutions, observing a reduction in SOA mass and significant photodissociation of compounds containing carbonyl and peroxide functional groups.13−15 Although there is growing evidence that photolysis may play an important role in the aging of SOA, the RH effects are still unknown. The objective of the current study is to investigate the photolysis of particles containing water-soluble α-pinene SOA material and ammonium sulfate under various RH conditions in an environmental chamber. The changes in SOA mass and chemical composition were measured using an Aerodyne aerosol mass spectrometer (AMS). The kinetics of SOA photolysis were examined in order to estimate the atmospheric importance of direct photochemical aging. The use of mixed SOA−ammonium sulfate particles was chosen for two reasons. In particular, the sulfate mass could be used to normalize the organic signal under the assumption of no sulfate photochemical reactivity. Also, the sulfate hygroscopic properties are well characterized ensuring that water was present in the particles at high relative humidity and not present at the lowest relative humidity.
2. EXPERIMENTAL SECTION 2.1. SOA Generation and Collection. SOA was generated and collected separately from the photolysis experiments in order to remove the effects of gas-phase oxidants and volatile αpinene oxidation products. The ozonolysis of α-pinene was used to generate SOA material in a 1 m3 Telfon (FEP) dark chamber, under dry conditions (RH < 5%) and in a continuous flow mode. The terpene was introduced into the chamber via a 6 sccm flow from the headspace of a bubbler containing liquid α-pinene (Sigma-Aldrich), chilled to −10 °C and mixed with a 9 slpm flow of purified air. Ozone was generated by passing 0.5 slpm of purified air over a mercury lamp and introduced into the chamber to initiate SOA formation. The final mixing ratios were 750 ppb of ozone, as monitored by an ozone analyzer (Thermo Scientific) and 150 ppb of α-pinene, as measured by a proton-transfer-reaction mass spectrometer (Ionicon). The residence time in the chamber during SOA generation was 100 min. The resulting SOA particle size distribution was measured using a TSI Scanning Mobility Particle Sizer (SMPS: DMA TSI-3080 and CPC, TSI-3225). A sample volume distribution of the resulting SOA is shown in Figure S1 (Supporting Information). The SOA was collected on polytetrafluoroethylene filters (47 mm, 2 um pore size, Pall Corporation) at 9.5 slpm for 48 h after the O3 mixing ratio and SOA mass loading were stable. The amount of SOA collected on filter (∼9 mg) was determined from the difference in mass measured before and after collection. Immediately after collection, the water-soluble organic compounds in the SOA were extracted in 8.75 mL of purified water (18 mΩ) and stored in a freezer at −20 °C. 2.2. UV/vis Absorption Spectroscopy. The UV/vis absorption properties of SOA were measured by an UV/vis dual beam spectrometer (PerkinElmer). A bulk SOA solution was used, using a concentration identical to the solutions used for generating particles for photolysis experiments (described below). Purified water in a matching curvette (1 cm, quartz) was used as the reference. From the measured absorbance (base-10), the effective absorption cross section (base-e) of SOA was determined (see the Supporting Information for a detailed calculation). Since the UV lamps used in the current study to initiate photolysis have a sharp emission cutoff at 284 nm, the absorbance below this wavelength was not considered. B
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
and UVA lamps are shown in Figure 1, where the actinic flux at solar noon is provided for comparison. It is important to note
composition of particles due to wall loss were accounted for in order to isolate the effects due to UV irradiation (Figure S3, Supporting Information).
3. RESULTS AND DISCUSSION 3.1. Absorption Cross Section of SOA. While SOA is a complex mixture of many different compounds, the absorption cross section of the bulk SOA solution (Figure 2) provides
Figure 1. Measured photon flux in the chamber from UVB and UVA lamps compared to the actinic flux for a clear-sky summer day. The actinic flux was obtained from “Quick TUV Calculator”, available at http://cprm.acd.ucar.edu/Models/TUV/Interactive_TUV/ using the following parameters: SZA = 0, June 30, 2000, 300 Dobson overhead ozone, surface albedo of 0.1, and 0 km altitude.
Figure 2. Absorption cross section of bulk SOA solution (this work) and aqueous cis-pinonic acid.20
that the spectral flux of the UVB and UVA lamps is not fully representative of that in the atmosphere. Compared to ambient conditions, the UVB lamps are more intense at lower wavelengths while the UVA lamps are less intense for all wavelengths. 2.4. Particle Measurements. For particle mass and composition measurements, the particles were first dried using a silica gel diffusion dryer and then sampled online by the time-of-flight aerosol mass spectrometer (ToF-AMS, Aerodyne Research, Inc.). Experiments were conducted with the unit resolution AMS (cToF-AMS) unless otherwise stated. The operating and analysis procedure of the AMS has been previously reviewed in detail.17 Briefly, the AMS provides the size-resolved mass and chemical composition of the nonrefractory component of particles, including organics, sulfate, nitrate, and ammonium. Since the resulting mass spectra of the sampled particles include signals from major gases (N2, O2, H2O, Ar, and CO2), their contributions were accounted for by obtaining multiple filter measurements (e.g., zero particle signal) throughout each experiment. To understand the changes in organic particle mass due to photolysis, the loss of particle mass to the chamber walls needs to be accounted for. Since the chamber experiments were conducted in batch mode, where the particle mass concentration decreases over time due to deposition, the individual organic fragments and total organic mass were normalized to the total sulfate mass measured by the AMS in order to determine absolute changes due to photolysis, assuming sulfate is nonreactive. Additionally, the mass loading of α-pinene SOA has been shown to affect its chemical composition, as semivolatile organic compounds will increasingly partition to the particle at higher mass loadings.18,19 In the current study, where particle mass loading decreases over time due to wall loss, we expect the evaporation of semivolatile compounds over the course of the experiment. Control experiments were conducted where particles were not exposed to UV irradiation in the chamber, under each RH condition. These control experiments were conducted at the same temperature as typical photolysis experiments. The observed changes to the chemical
information about the functional groups that absorb light in the wavelength range emitted by the UVB and UVA lamps used in this study. The observed main absorption region corresponds to characteristic absorptions of carbonyl and peroxy functional groups, which are known products of SOA formed by α-pinene ozonolysis. The absorption cross section of a known carbonyl product, cis-pinonic acid,20 is also shown in Figure 2 for comparison. The absorption observed at higher wavelengths is likely due to the presence of peroxide functional groups, as absorption cross sections of peroxides tend to exhibit a broad tail into the actinic region.21 While it is ideal to measure the absorption cross section of SOA material in particles, this was not possible with the instrumentation available. However, a previous measurement of the mass absorption coefficient of SOA particles at RH 70% observed a similar spectral shape compared to the spectrum obtained in this study using bulk SOA solution, suggesting the absorptive properties of SOA material in bulk solutions are similar to those in particles under higher RH conditions.22 3.2. Photolytic Degradation of SOA. Absolute changes to the total organic mass, as indicated by the organic-to-sulfate (org:sulfate) mass ratio, due to exposure to UVB light under different RH conditions are shown in Figure 3. Org:sulfate ratios were normalized to pre-aging values to guide comparison of the changes for different experimental conditions. Under all RH conditions, decreases in total organic mass were observed due to UV light exposure, where the amount of SOA mass loss increases with RH. Absorption of UV light by organic compounds can initiate photolysis, where the fragmentation of their carbon skeleton results in the formation of smaller products (e.g., fewer carbons atoms) that are more volatile.18,23 We speculate that these volatile photolysis products evaporate from the particle, resulting in the observed loss of total organic mass. In fact, the formation of gas-phase products, such as CO, CH4, acetone, and other small VOCs, has been observed following the photodegradation of limonene−SOA on filters and in aqueous solutions.13,14 C
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 3. Time series profiles of the org:sulfate ratios measured by the AMS for the photolysis of SOA by UVB radiation under high RH (solid line), mid RH (dashed line), and low RH (dotted line) conditions. The shaded areas for each RH condition represent the variability (±1σ) between multiple experiments.
In addition to the total organic mass, the absolute changes in specific organic fragments provide additional insight into the photolysis of SOA. Due to the evaporation/ionization techniques used in the AMS, most compounds are subject to extensive fragmentation; however, specific fragments can be used as tracers for organic compounds with specific functional groups.17 The CHO+ fragment detected at m/z 29 arises from the fragmentation of compounds containing carbonyl functional groups,24 while two different fragments can contribute to the signal observed at m/z 43, with the C2H3O+ fragment originating from functionalities with a single oxygen atom (e.g., carbonyls, alcohols, and ethers) and the C3H7+ fragment originating from aliphatic molecules.25 Control experiments using the high-resolution ToF-AMS, which can separate different ions at the same nominal m/z, indicated that the C2H3O+ ion was the dominant signal at m/z 43 for this study. The organic ion at m/z 44 (CO2+) arises from the fragmentation of highly oxidized organic acids (e.g., carboxylic acids and esters)26 and peroxides.27 For all RH conditions, upon UVB irradiation, loss of the oxidized components of SOA was observed, where the decays of m/z 29 and m/z 43 (Figure 4a,b) were faster than that of m/z 44 (Figure 4c). Given that the absorption spectrum of SOA indicates the presence of carbonyl and peroxide function groups, both of which absorb UVB radiation, it is not surprising that we observed decays of m/z 29, 43, and 44, suggesting the photodegradation of these functional groups. To gain further insight of the wavelength dependence of SOA photolysis, experiments under high RH conditions were conducted using UVA lights. Compared to experiments using UVB radiation at the same RH condition, SOA exposed to UVA radiation exhibited slower decays for total organic mass (Figure 5a) and for all organic fragments considered (Figure 5b−d). In particular, the loss of m/z 29 and 43 is greatly suppressed for UVA lights, suggesting that the photolysis of less oxidized components (i.e., carbonyls) contributes significantly to the observed photochemical aging with the UVB light. It is important to note that the slower decay of m/z 44 compared to m/z 29 and 43 can also arise from secondary reactions of photolysis which result in the formation of highly oxidized organics. For example, photolysis of peroxides is known to generate OH radicals, which can oxidize other SOA components, forming highly oxidized components such as organic acids.28,29 Direct photolysis of aldehydes has also been observed to generate organic acids as well.13,20 Given that both peroxides and carboxylic acids contribute to the organic
Figure 4. Time series profiles of the sulfate normalized (a) m/z 29, (b) m/z 43, and (c) m/z 44 for the photolysis of mixed SOA−sulfate particles by UVB radiation at high RH (solid line), mid RH (dashed line), and low RH (dotted line) conditions. The shaded areas for each RH condition represent the variability (±1σ) between multiple experiments.
fragment at m/z 44, the combined result of photodissociation of peroxides and the formation of carboxylic acids may have resulted in the slower decay of m/z 44. The relative changes in the chemical composition of SOA due to photolysis can be illustrated using the framework developed by Ng et al. in Figure 6, where aging of SOA is described using the fraction of organic mass at m/z 43 vs 44 ( f43 and f44).26 Ambient SOA typically falls within the triangle denoted by the dashed lines in Figure 6c (inset), with less oxidized SOA mainly residing in the lower half of the triangle and highly oxidized SOA in the upper half. In general, studies have shown that SOA aging processes, such as OH oxidation, result in movement toward the upper left of the triangle space.30 The trends observed for this figure are also consistent with those observed for Figure 4, with, for example, a stronger decrease for m/z 43 compared to m/z 44, with the magnitude of the change depending on RH conditions. This suggests that enhanced loss of less oxidized particulate organics due to photolysis results in a more oxidized SOA. Note that the composition of the SOA used in the current study resides in the middle region of the triangle, suggesting that the organics are somewhat oxidized prior to photolysis. D
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 5. Time series profiles of the sulfate normalized (a) total organic mass (org:sulfate ratio, black), (b) m/z 29, (c) m/z 43, and (d) m/z 44 for the photolysis of mixed SOA−sulfate particles by UVB (solid lines) and UVA (dashed lines) radiation under high RH conditions. The shaded areas for each light source represent the variability (±1σ) between multiple experiments.
measured photon flux in the chamber, the quantum yield for the loss of organics due to SOA photolysis can be estimated. Since the UV lamps used in the current study to initiate photolysis have a sharp emission cutoff at 284 nm, the absorbance below this wavelength was not considered. We note that this approach does not take into account the enhancement in photon flux that may occur inside wet particles relative to values in air.33 From the observed mass loss rates under high RH conditions for both UVB and UVA lights, the average effective photolysis quantum yield was determined to be 1.2 ± 0.2. This large quantum yield suggests that either the primary photolytic process efficiently gives rise to volatile products or secondary reactions (such as those driven by the photochemical formation of OH in the particle) may have also contributed to the observed total mass loss rates. The photolysis quantum yield for cis-pinonic acid, one of the identified compounds in α-pinene SOA, is 0.5 ± 0.3,20 i.e., also quite large and similar to our result. The factor of 2 difference could potentially be due to the presence of other more photolabile compounds in SOA material. At decreasing RH, the SOA becomes highly viscous and concentrated, where elevated concentrations can accelerate condensed phase reactions (e.g., polymerization, reactions with inorganics), thus potentially changing the absorptive properties of the material. Since the absorption cross section of SOA measured from bulk solutions may not be identical compared to particles under decreasing RH conditions, the current approach to determine the effective quantum yield for mid and low RH conditions was not pursued. Nevertheless, the suppression in mass loss rates with decreasing RH indicates that photolysis is dependent on RH conditions, where the mass loss rate under low RH is a factor of 2 slower than that at high RH. It is important to note that the observed mass loss rate at the low RH condition using UVB lights for the current study [(7.91 ± 0.13) × 10−5 s−1] is similar to the photolytic mass loss rate constant of dry α-pinene SOA−AS particles for UVA lights
Considering the experiments discussed above, the loss of organic components and changes in chemical composition due to SOA photolysis all show RH dependence, with enhanced loss observed with increasing RH conditions. We speculate that the observed RH dependence for SOA photolysis is likely due to the effects of viscosity on diffusion, as recent studies have shown that SOA exhibits liquid-like behavior under high RH and is semisolid/solid under low RH.9,10 In viscous semisolid/ solid material, the slow diffusion of SOA photodissociation products can favor their recombination, resulting in a reduction in the efficiency of photolysis. For example, the net quantum yields for the photolysis of certain amines and iodine were previously observed to be reduced in viscous media.31,32 Additionally, slow diffusion has been shown to kinetically hinder the evaporation of SOA.11 Thus, the effects of RH on diffusion can lead to the results observed in the current study, where, under lower RH conditions, slow diffusion of photolysis products resulted in less mass loss compared to higher RH conditions. 3.3. Kinetics of Photolysis. To further investigate the role of RH on the reactivity of SOA material on particles, the decay rates of particulate SOA mass loss due to photolysis were determined. The kinetic plot (shown in Figure S4, Supporting Information) indicated that the reaction exhibited first order kinetics at early times, as expected from direct photolysis. Linear fits to the first 10 min of data yield first-order mass loss rates (kSOA) for all RH and light conditions, listed in Table 1. Assuming that the initial observed mass loss is only due to the volatilization of small organic compounds from photolysis of SOA, the observed mass loss rate constants were equated to the effective photolysis rate constant: J=
∫λ σ(λ)ϕ(λ)F(λ) dλ
where σ(λ) is the absorption cross section of SOA, ϕ(λ) is the photolysis quantum yield, and F(λ) is the photon flux. Using the measured absorption cross section of SOA and the E
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
most rapid loss observed under high RH conditions. In particular, the less oxidized components of SOA (e.g., carbonyls) exhibited the fastest photodegradation, which resulted in a more oxidized SOA particle after photolytic aging. We hypothesize that the dependence of photolysis on RH conditions was due to moisture-induced viscosity changes of SOA material, affecting the diffusion of reactants/products. In this regard, despite a change in several orders of magnitude in the viscosity of the SOA material from high to low RH conditions, its photolysis was only slower by a factor of 2. This suggests that, even at such high viscosity values, photolyzed SOA material can nevertheless efficiently diffuse out of the particle and evaporate within the experimental time scale of the current study.8,10 To assess the importance of photolysis as an aging mechanism for SOA in the atmosphere, the atmospheric lifetime of SOA particles with respect to photolysis is estimated to be between that observed with the UVB lamps and that for the UVA lamps. In particular, using the experimentally determined absorption cross section and photolysis quantum yield, and calculated actinic fluxes, the initial photolysis lifetime is ∼3.7 h at Solar Noon for high RH conditions. Given that the average lifetime of particles in the atmosphere is approximately 1 week with respect to deposition, this very rough calculation indicates that photolysis can be an important sink of SOA material in the atmosphere due to volatilization. Clearly, there are uncertainties in this estimate, and this initial lifetime should not be equated to the overall lifetime for loss of SOA particles in the atmosphere. In particular, the contribution of secondary photolysis to the effective quantum yield was not constrained. Therefore, the average photolysis quantum yield determined in the current study is likely an upper limit to the primary photolysis process, corresponding to a lower limit for this calculated lifetime. Nevertheless, the effects of such secondary reactions of photolysis may be of importance to the fate of SOA. For example, it is known that photolysis of organic compounds in cloudwater is an important source of H2O2, which is an important oxidant.35 Also, we note that SOA reactivity may be dependent on its atmospheric age and the degree of prior processing. In particular, the slopes of the mass loss decay curves (e.g., Figure S4, Supporting Information) decreased with time, indicating depletion of the most photochemically active species (i.e., carbonyls and peroxides) and formation of more stable compounds, such as carboxylic acids. On the other hand, Henry and Donahue observed enhanced mass loss for the photolysis of SOA that was aged via OH oxidation, as oxidative aging by OH may lead to the formation of more carbonyl and peroxide compounds compared to photolytic aging.15 Additionally, brown carbon formed via condensed-phase reactions in aerosols has been observed to be photolabile.36 As such, it is important to understand how the reactivity of SOA changes with atmospheric age.
Figure 6. Aerosol composition described by the fraction of organic mass at m/z 43 and 44 (f43 vs f44) before (open circles), during (closed circles), and after (open circles) UV light exposure for (a) high RH, (b) mid RH, (c) and low RH conditions. The inset shows the zoomed-out view of the f43 vs f44 space (with dimensions of Figure 6a−c illustrated by the red box), where ambient SOA is distributed within the triangular area.
(6 × 10−5 s−1), reported by Henry and Donahue.15 In comparison, for dry isoprene SOA, Kroll et al.34 also observed mass loss [(1−5) × 10−5 s−1] due to irradiation with UVA lights. However, this mass loss rate might have arisen due to reaction with gas-phase OH radicals, due to the addition of H2O2 (both UV photons and OH radicals are present during irradiation).
4. CONCLUSIONS AND ATMOSPHERIC IMPLICATIONS In this work, the RH dependence for the photolysis of a chemically complex mixture, water-soluble SOA, was illustrated. Photolysis resulted in substantial organic mass loss, with the
Table 1. Total Organic Mass Loss Rate Constants for Different RH Conditionsa kSOA (s−1) light source UVB UVA a
high RH
mid RH −4
low RH −4
(1.64 ± 0.09) × 10
(1.67 ± 0.20) × 10 (4.69 ± 0.14) × 10−5
(7.91 ± 0.13) × 10−5
The uncertainties in kSOA represent variability (±1σ) between multiple experiments. F
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Diffusion in Secondary Organic Aerosol. Faraday Discuss. 2013, 165, 391−406. (7) Virtanen, A.; Joutsensaari, J.; Koop, T.; Kannosto, J.; Yli-Pirilä, P.; Leskinen, J.; Mäkelä, J. M.; Holopainen, J. K.; Pöschl, U.; Kulmala, M.; et al. An Amorphous Solid State of Biogenic Secondary Organic Aerosol Particles. Nature 2010, 467, 824−827. (8) Koop, T.; Bookhold, J.; Shiraiwa, M.; Pöschl, U. Glass Transition and Phase State of Organic Compounds: Dependency on Molecular Properties and Implications for Secondary Organic Aerosols in the Atmosphere. Phys. Chem. Chem. Phys. 2011, 13, 19238−19255. (9) Saukko, E.; Lambe, A. T.; Massoli, P.; Koop, T.; Wright, J. P.; Croasdale, D. R.; Pedernera, D. A.; Onasch, T. B.; Laaksonen, A.; Davidovits, P.; et al. Humidity-Dependent Phase State of SOA Particles from Biogenic and Anthropogenic Precursors. Atmos. Chem. Phys. 2012, 12, 7517−7529. (10) Renbaum-Wolff, L.; Grayson, J. W.; Bateman, A. P.; Kuwata, M.; Sellier, M.; Murray, B. J.; Shilling, J. E.; Martin, S. T.; Bertram, A. K. Viscosity of A-Pinene Secondary Organic Material and Implications for Particle Growth and Reactivity. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 8014−8019. (11) Cappa, C. D.; Wilson, K. R. Evolution of Organic Aerosol Mass Spectra upon Heating: Implications for OA Phase and Partitioning Behavior. Atmos. Chem. Phys. 2011, 11, 1895−1911. (12) Kolesar, K. R.; Buffaloe, G.; Wilson, K. R.; Cappa, C. D. OHInitiated Heterogeneous Oxidation of Internally-Mixed Squalane and Secondary Organic Aerosol. Environ. Sci. Technol. 2014, 48, 3196− 3202. (13) Mang, S. A.; Henricksen, D. K.; Bateman, A. P.; Andersen, M. P. S.; Blake, D. R.; Nizkorodov, S. A. Contribution of Carbonyl Photochemistry to Aging of Atmospheric Secondary Organic Aerosol. J. Phys. Chem. A 2008, 112, 8337−8344. (14) Bateman, A. P.; Nizkorodov, S. A.; Laskin, J.; Laskin, A. Photolytic Processing of Secondary Organic Aerosols Dissolved in Cloud Droplets. Phys. Chem. Chem. Phys. 2011, 13, 12199−12212. (15) Henry, K. M.; Donahue, N. M. Photochemical Aging of APinene Secondary Organic Aerosol: Effects of OH Radical Sources and Photolysis. J. Phys. Chem. A 2012, 116, 5932−5940. (16) Tang, I. N.; Fung, K. H.; Imre, D. G.; Munkelwitz, H. R. Phase Transformation and Metastability of Hygroscopic Microparticles. Aerosol Sci. Technol. 1995, 23, 443−453. (17) Canagaratna, M. R.; Jayne, J.; Jimenez, J. L.; Allan, J. D.; Alfarra, M. R.; Zhang, Q.; Onasch, T. b.; Drewnick, F.; Coe, H.; Middlebrook, A.; et al. Chemical and Microphysical Characterization of Ambient Aerosols with the Aerodyne Aerosol Mass Spectrometer. Mass Spectrom. Rev. 2007, 26, 185−222. (18) 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. (19) Shilling, J. E.; Chen, Q.; King, S. M.; Rosenoern, T.; Kroll, J. H.; Worsnop, D. R.; DeCarlo, P. F.; Aiken, A. C.; Sueper, D.; Jimenez, J. L.; et al. Loading-Dependent Elemental Composition of A-Pinene SOA Particles. Atmos. Chem. Phys. 2009, 9, 771−782. (20) Lignell, H.; Epstein, S. A.; Marvin, M. R.; Shemesh, D.; Gerber, B.; Nizkorodov, S. Experimental and Theoretical Study of Aqueous Cis-Pinonic Acid Photolysis. J. Phys. Chem. A 2013, 117, 12930− 12945. (21) Mang, S. A.; Walser, M. L.; Pan, X.; Xing, J-H; Bateman, A. P.; Underwood, J. S.; Gomez, A. L.; Park, J.; Nizkorodov, S. A. Photochemistry of Secondary Organic Aerosol Formed from Oxidation of Monoterpenes. In Atmospheric Aerosols; ACS Symposium Series; American Chemical Society: Washington, DC, 2009; Vol. 1005, pp 91−109. (22) Zhong, M.; Jang, M. Light Absorption Coefficient Measurement of SOA Using a UV−Visible Spectrometer Connected with an Integrating Sphere. Atmos. Environ. 2011, 45, 4263−4271. (23) Calvert, J. G.; Pitts, J. N. J. Photochemistry; John Wiley & Sons, Inc.: New York, 1966. (24) Lee, A. K. Y.; Hayden, K. L.; Herckes, P.; Leaitch, W. R.; Liggio, J.; Macdonald, A. M.; Abbatt, J. P. D. Characterization of Aerosol and
Also, we point out that this work has implications to photooxidation SOA chamber and flow tube studies, given that many of these reactors use UV lamps similar to those in this experiment. Thus, the changes in SOA mass and chemical composition due to photolytic aging observed in this study may also occur in other experiments. Finally, the formation mechanism of the mixed SOA−AS is not necessarily the same as that in the atmosphere, probably resembling the process by which particles are formed by cloud processing more than that from condensation of terpene oxidation products on an ammonium sulfate particle. Overall, this is one of the first studies to quantify the changes in SOA that occur upon exposure to ultraviolet light, and the first study to look at the effect of RH on this process. This RH dependence needs to be further investigated with other aging processes, SOA types, and various conditions (e.g., peroxides are more likely to be formed under low NOx conditions during SOA formation), in order to understand the processing of such species under a wide range of atmospherically relevant conditions.
■
ASSOCIATED CONTENT
S Supporting Information *
Details of absorption cross section calculations, mass distribution of SOA generated during filter collection, mass distribution of AS−SOA particles during photolysis experiments, org:sulfate ratio time trace from control experiment, and kinetic plots for total organic mass loss. 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 The authors thank Natural Sciences and Engineering Research Council and Canada Foundation for Innovation for funding.
■
REFERENCES
(1) 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.; et al. Ubiquity and Dominance of Oxygenated Species in Organic Aerosols in Anthropogenically-Influenced Northern Hemisphere Midlatitudes. Geophys. Res. Lett. 2007, 34, L13801. (2) Heald, C. L.; Ridley, D. A.; Kreidenweis, S. M.; Drury, E. E. Satellite Observations Cap the Atmospheric Organic Aerosol Budget. Geophys. Res. Lett. 2010, 37, L24808. (3) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.; Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.; Goldstein, A. H.; et al. The Formation, Properties and Impact of Secondary Organic Aerosol: Current and Emerging Issues. Atmos. Chem. Phys. 2009, 9, 5155−5236. (4) Lee, J. W. L.; Carrascón, V.; Gallimore, P. J.; Fuller, S. J.; Björkegren, A.; Spring, D. R.; Pope, F. D.; Kalberer, M. The Effect of Humidity on the Ozonolysis of Unsaturated Compounds in Aerosol Particles. Phys. Chem. Chem. Phys. 2012, 14, 8023−8031. (5) Nguyen, T. B.; Coggon, M. M.; Flagan, R. C.; Seinfeld, J. H. Reactive Uptake and Photo-Fenton Oxidation of Glycolaldehyde in Aerosol Liquid Water. Environ. Sci. Technol. 2013, 47, 4307−4316. (6) Zhou, S.; Shiraiwa, M.; McWhinney, R. D.; Pöschl, U.; Abbatt, J. P. D. Kinetic Limitations in Gas-Particle Reactions Arising from Slow G
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Cloud Water at a Mountain Site during WACS 2010: Secondary Organic Aerosol Formation through Oxidative Cloud Processing. Atmos. Chem. Phys. 2012, 12, 7103−7116. (25) Zhang, Q.; Alfarra, M. R.; Worsnop, D. R.; Allan, J. D.; Coe, H.; Canagaratna, M. R.; Jimenez, J. L. Deconvolution and Quantification of Hydrocarbon-like and Oxygenated Organic Aerosols Based on Aerosol Mass Spectrometry. Environ. Sci. Technol. 2005, 39, 4938− 4952. (26) 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.; et al. Organic Aerosol Components Observed in Northern Hemispheric Datasets from Aerosol Mass Spectrometry. Atmos. Chem. Phys. 2010, 10, 4625−4641. (27) Aiken, A. C.; DeCarlo, P. F.; Jimenez, J. L. Elemental Analysis of Organic Species with Electron Ionization High-Resolution Mass Spectrometry. Anal. Chem. 2007, 79, 8350−8358. (28) Lim, Y. B.; Tan, Y.; Perri, M. J.; Seitzinger, S. P.; Turpin, B. J. Aqueous Chemistry and Its Role in Secondary Organic Aerosol (SOA)Formation. Atmos. Chem. Phys. 2010, 10, 10521−10539. (29) 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. 2011, 11, 6465−6474. (30) Slowik, J. G.; Wong, J. P. S.; Abbatt, J. P. D. Real-Time, Controlled OH-Initiated Oxidation of Biogenic Secondary Organic Aerosol. Atmos. Chem. Phys. 2012, 12, 9775−9790. (31) Ratcliff, M. A.; Kochi, J. K. Cage Effects and the Viscosity Dependence of the Photolysis of Dibenzylamine and Tribenzylamine. J. Org. Chem. 1972, 37, 3275−3281. (32) Otto, B.; Schroeder, J.; Troe, J. Photolytic Cage Effect and Atom Recombination of Iodine in Compressed Gases and Liquids: Experiments and Simple Models. J. Chem. Phys. 1984, 81, 202−213. (33) Nissenson, P.; Knox, C. J. H.; Finlayson-Pitts, B. J.; Phillips, L. F.; Dabdub, D. Enhanced Photolysis in Aerosols: Evidence for Important Surface Effects. Phys. Chem. Chem. Phys. 2006, 8, 4700− 4710. (34) Kroll, J. H.; Ng, N. L.; Murphy, S. M.; Flagan, R. C.; Seinfeld, J. H. Secondary Organic Aerosol Formation from Isoprene Photooxidation. Environ. Sci. Technol. 2006, 40, 1869−1877. (35) Faust, B. C. Photochemistry of Clouds, Fogs, and Aerosols. Environ. Sci. Technol. 1994, 28, 216A−222A. (36) Sareen, N.; Moussa, S. G.; McNeill, V. F. Photochemical Aging of Light-Absorbing Secondary Organic Aerosol Material. J. Phys. Chem. A 2013, 117, 2987−2996.
H
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
Changes in Secondary Organic Aerosol Composition and Mass due to Photolysis: Relative Humidity Dependence Jenny P. S. Wong, Shouming Zhou, and Jonathan P. D. Abbatt* Department of Chemistry, University of Toronto, Toronto, Canada S Supporting Information *
ABSTRACT: This study is focused on the relative humidity (RH) dependence of watersoluble secondary organic aerosol (SOA) aging by photolysis. Particles containing αpinene SOA and ammonium sulfate, generated by atomization, were exposed to UV radiation in an environmental chamber at three RH conditions (5, 45, and 85%), and changes in chemical composition and mass were monitored using an aerosol mass spectrometer (AMS). Under all RH conditions, photolysis leads to substantial loss of SOA mass, where the rate of mass loss decreased with decreasing RH. For all RH conditions, the less oxidized components of SOA (e.g., carbonyls) exhibited the fastest photodegradation rates, which resulted in a more oxidized SOA after photolytic aging. The photolytic reactivity of SOA material exhibited a dependence on RH likely due to moisture-induced changes in SOA morphology or phase. The results suggest that the atmospheric lifetime of SOA with respect to photolysis is dependent on its RH cycle, and that photolysis may be an important sink for some SOA components occurring on an initial time scale of a few hours under ambient conditions.
1. INTRODUCTION Organic compounds are ubiquitous in ambient aerosol, yet their formation and transformation mechanisms in the atmosphere remain not fully characterized.1 These organic compounds are largely secondary in nature (i.e., secondary organic aerosol, SOA), produced from the condensation of oxidation products of volatile organic compounds. Away from sources, SOA is continually subjected to further chemical processing, or aging, that transforms the particulate physiochemical properties, which are important for climate, air quality, and human health. Current atmospheric models, with parametrizations based largely on gas- and gas-particle reactions studied in the laboratory, have difficulty predicting the mass and the degree of oxidation of ambient SOA, suggesting unidentified formation and aging mechanisms.2,3 Aging processes include heterogeneous oxidation, gas-phase oxidation of semivolatile components, and reactions in the condensed phase, such as photolysis, polymerization, and aqueous oxidation.3 Historically, the aging of SOA material has been investigated using SOA particles in environmental chambers or flow tube reactors, or particles collected on filters. Typically, these studies have been conducted under dry or low RH conditions. On the other hand, studies have also examined the aging of aqueous SOA solutions to investigate the role of cloud processing (e.g., saturated RH conditions). Collectively, these studies have shown that aging can affect the mass and the chemical composition of SOA, but the experimental conditions only represent a narrow range of RH conditions found in the atmosphere. Only a limited number of studies have investigated aging processes under a wide range of RH conditions where the reactivity of model organic particles was observed to be dependent on RH, as aging by ozone4 and OH radical5 were © XXXX American Chemical Society
enhanced under high RH conditions. For complex mixtures such as SOA, recent work by Zhou et al. observed that the kinetics of the heterogeneous ozonolysis of benzo[a]pyrene (BaP) within SOA material was suppressed under dry conditions, where the kinetic behavior was between that expected of solids and liquids. Additionally, it was observed that the reactivity of BaP was enhanced with increasing RH conditions. These results suggest that RH affects particle reactivity, as water uptake by hygroscopic SOA under high RH condition lowers the material’s viscosity, which allows for more rapid diffusion of reactants compared to dry conditions where SOA is semisolid.6 Indeed, there is growing experimental evidence that SOA can be an amorphous solid with high viscosity under dry/low RH conditions. With increasing RH, water acts as a plasticizer, decreasing the viscosity of the organic material (i.e., liquid SOA).7−10 In addition to the study by Zhou et al., other recent studies have shown that the phase/viscosity of SOA material can affect gas-particle partitioning11 and aging by heterogeneous OH oxidation.12 Given that past investigations have shown the importance of RH on the phase/viscosity of SOA, which has implications on particle reactivity, it is necessary to characterize how variations in RH conditions affect other aging processes. Of particular interest to the current study is the aging of SOA by photolysis. As with other SOA aging experiments, previous laboratory studies have focused on the photolysis of dry SOA particles, or Special Issue: Mario Molina Festschrift Received: July 10, 2014 Revised: September 4, 2014
A
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
2.3. Photolysis of SOA. All photolysis experiments were conducted in the University of Toronto Mobile Concentration and Aging (MOCA) chamber in batch mode. MOCA is a 1 m3 Teflon (FEP) bag mounted around a cubic Teflon-coated frame, surrounded by 24 UV lamps (Sylannia) on four of its six sides, cooled with multiple fans. The temperature and RH inside the chamber were continuously monitored (Vaisala). Following each experiment, the chamber was flushed with clean air and irradiated with UV light for a minimum of 12 h for cleaning. Prior to each experiment, the chamber was conditioned to a specific RH by flowing purified air through heated water bubblers into the chamber prior to particle introduction. To generate particles, ammonium sulfate (AS, 0.04 w/v %) was added to the SOA water extract to a total volume of 10 mL, resulting in a 3:2 inorganic-to-organic mass ratio. AS was chosen because its hygroscopic response to variations in RH is well characterized.16 This bulk solution was aerosolized into the chamber using a constant output atomizer (TSI 3076) for 20 min. From AMS measurements (described below) obtained using particle time-of-flight (PTOF) mode, the particles in the chamber had a typical mass mode of 230 nm prior to UV light exposure (size-resolved mass distribution shown in Figure S2, Supporting Information). Prior to photolysis, the typical particle mass loading in the chamber was 30 μg/m3. To study the photolysis of SOA material, the AS−SOA particles were mixed in the dark chamber for 20 min before they were exposed to UV irradiation for 33 min, following which the chamber was returned to dark conditions. Upon UV light exposure, the temperature inside the chamber increased by 1° (above room temperature), which resulted in a decrease of ∼5% in RH. It is important to note that the effects of UV light exposure can be due to volatilization (due to heating) and UV irradiation. Since this temperature increase was identical for all photolysis experiments, the contributions of volatilization to changes in SOA mass and composition are also identical. That being said, with the temperature increase so small, we believe the changes are largely due to photolysis. Two sets of experiments were conducted to characterize the photolysis of SOA material. For the first set of experiments, the photolysis of SOA by UVB light centered at 310 nm at three different RH conditions (85% (“high RH”), 45% (“mid RH”), and 5% (“low RH”)) was examined to elucidate the effects of RH. The polydisperse particles were fully deliquesced before entering the chamber, and then equilibrated to each RH condition in the chamber. Given that previous work has characterized the effects of RH on the particle phase and viscosity of SOA,9,10 we speculate that the SOA material is likely to be liquid at high RH and semisolid at low RH conditions, possibly having phase separated. At mid RH conditions, the particles were expected to have intermediate liquid water content, and the resulting phase and viscosity of the SOA material is likely to be between the values present under high and low RH conditions. For AS, it is deliquesced at high RH and effloresced at low RH conditions.16 The second set of experiments was conducted to investigate the UV wavelength dependence of SOA photolysis. Under the same high RH conditions, photolysis using UVA lamps centered at 360 nm was conducted, and compared to results obtained using UVB lamps centered at 310 nm. The wavelength dependence of the photon flux inside the chamber was measured using a spectroradiometer (StellaNet Inc.). The measured wavelength dependent photon fluxes from both UVB
SOA dissolved in bulk solutions, observing a reduction in SOA mass and significant photodissociation of compounds containing carbonyl and peroxide functional groups.13−15 Although there is growing evidence that photolysis may play an important role in the aging of SOA, the RH effects are still unknown. The objective of the current study is to investigate the photolysis of particles containing water-soluble α-pinene SOA material and ammonium sulfate under various RH conditions in an environmental chamber. The changes in SOA mass and chemical composition were measured using an Aerodyne aerosol mass spectrometer (AMS). The kinetics of SOA photolysis were examined in order to estimate the atmospheric importance of direct photochemical aging. The use of mixed SOA−ammonium sulfate particles was chosen for two reasons. In particular, the sulfate mass could be used to normalize the organic signal under the assumption of no sulfate photochemical reactivity. Also, the sulfate hygroscopic properties are well characterized ensuring that water was present in the particles at high relative humidity and not present at the lowest relative humidity.
2. EXPERIMENTAL SECTION 2.1. SOA Generation and Collection. SOA was generated and collected separately from the photolysis experiments in order to remove the effects of gas-phase oxidants and volatile αpinene oxidation products. The ozonolysis of α-pinene was used to generate SOA material in a 1 m3 Telfon (FEP) dark chamber, under dry conditions (RH < 5%) and in a continuous flow mode. The terpene was introduced into the chamber via a 6 sccm flow from the headspace of a bubbler containing liquid α-pinene (Sigma-Aldrich), chilled to −10 °C and mixed with a 9 slpm flow of purified air. Ozone was generated by passing 0.5 slpm of purified air over a mercury lamp and introduced into the chamber to initiate SOA formation. The final mixing ratios were 750 ppb of ozone, as monitored by an ozone analyzer (Thermo Scientific) and 150 ppb of α-pinene, as measured by a proton-transfer-reaction mass spectrometer (Ionicon). The residence time in the chamber during SOA generation was 100 min. The resulting SOA particle size distribution was measured using a TSI Scanning Mobility Particle Sizer (SMPS: DMA TSI-3080 and CPC, TSI-3225). A sample volume distribution of the resulting SOA is shown in Figure S1 (Supporting Information). The SOA was collected on polytetrafluoroethylene filters (47 mm, 2 um pore size, Pall Corporation) at 9.5 slpm for 48 h after the O3 mixing ratio and SOA mass loading were stable. The amount of SOA collected on filter (∼9 mg) was determined from the difference in mass measured before and after collection. Immediately after collection, the water-soluble organic compounds in the SOA were extracted in 8.75 mL of purified water (18 mΩ) and stored in a freezer at −20 °C. 2.2. UV/vis Absorption Spectroscopy. The UV/vis absorption properties of SOA were measured by an UV/vis dual beam spectrometer (PerkinElmer). A bulk SOA solution was used, using a concentration identical to the solutions used for generating particles for photolysis experiments (described below). Purified water in a matching curvette (1 cm, quartz) was used as the reference. From the measured absorbance (base-10), the effective absorption cross section (base-e) of SOA was determined (see the Supporting Information for a detailed calculation). Since the UV lamps used in the current study to initiate photolysis have a sharp emission cutoff at 284 nm, the absorbance below this wavelength was not considered. B
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
and UVA lamps are shown in Figure 1, where the actinic flux at solar noon is provided for comparison. It is important to note
composition of particles due to wall loss were accounted for in order to isolate the effects due to UV irradiation (Figure S3, Supporting Information).
3. RESULTS AND DISCUSSION 3.1. Absorption Cross Section of SOA. While SOA is a complex mixture of many different compounds, the absorption cross section of the bulk SOA solution (Figure 2) provides
Figure 1. Measured photon flux in the chamber from UVB and UVA lamps compared to the actinic flux for a clear-sky summer day. The actinic flux was obtained from “Quick TUV Calculator”, available at http://cprm.acd.ucar.edu/Models/TUV/Interactive_TUV/ using the following parameters: SZA = 0, June 30, 2000, 300 Dobson overhead ozone, surface albedo of 0.1, and 0 km altitude.
Figure 2. Absorption cross section of bulk SOA solution (this work) and aqueous cis-pinonic acid.20
that the spectral flux of the UVB and UVA lamps is not fully representative of that in the atmosphere. Compared to ambient conditions, the UVB lamps are more intense at lower wavelengths while the UVA lamps are less intense for all wavelengths. 2.4. Particle Measurements. For particle mass and composition measurements, the particles were first dried using a silica gel diffusion dryer and then sampled online by the time-of-flight aerosol mass spectrometer (ToF-AMS, Aerodyne Research, Inc.). Experiments were conducted with the unit resolution AMS (cToF-AMS) unless otherwise stated. The operating and analysis procedure of the AMS has been previously reviewed in detail.17 Briefly, the AMS provides the size-resolved mass and chemical composition of the nonrefractory component of particles, including organics, sulfate, nitrate, and ammonium. Since the resulting mass spectra of the sampled particles include signals from major gases (N2, O2, H2O, Ar, and CO2), their contributions were accounted for by obtaining multiple filter measurements (e.g., zero particle signal) throughout each experiment. To understand the changes in organic particle mass due to photolysis, the loss of particle mass to the chamber walls needs to be accounted for. Since the chamber experiments were conducted in batch mode, where the particle mass concentration decreases over time due to deposition, the individual organic fragments and total organic mass were normalized to the total sulfate mass measured by the AMS in order to determine absolute changes due to photolysis, assuming sulfate is nonreactive. Additionally, the mass loading of α-pinene SOA has been shown to affect its chemical composition, as semivolatile organic compounds will increasingly partition to the particle at higher mass loadings.18,19 In the current study, where particle mass loading decreases over time due to wall loss, we expect the evaporation of semivolatile compounds over the course of the experiment. Control experiments were conducted where particles were not exposed to UV irradiation in the chamber, under each RH condition. These control experiments were conducted at the same temperature as typical photolysis experiments. The observed changes to the chemical
information about the functional groups that absorb light in the wavelength range emitted by the UVB and UVA lamps used in this study. The observed main absorption region corresponds to characteristic absorptions of carbonyl and peroxy functional groups, which are known products of SOA formed by α-pinene ozonolysis. The absorption cross section of a known carbonyl product, cis-pinonic acid,20 is also shown in Figure 2 for comparison. The absorption observed at higher wavelengths is likely due to the presence of peroxide functional groups, as absorption cross sections of peroxides tend to exhibit a broad tail into the actinic region.21 While it is ideal to measure the absorption cross section of SOA material in particles, this was not possible with the instrumentation available. However, a previous measurement of the mass absorption coefficient of SOA particles at RH 70% observed a similar spectral shape compared to the spectrum obtained in this study using bulk SOA solution, suggesting the absorptive properties of SOA material in bulk solutions are similar to those in particles under higher RH conditions.22 3.2. Photolytic Degradation of SOA. Absolute changes to the total organic mass, as indicated by the organic-to-sulfate (org:sulfate) mass ratio, due to exposure to UVB light under different RH conditions are shown in Figure 3. Org:sulfate ratios were normalized to pre-aging values to guide comparison of the changes for different experimental conditions. Under all RH conditions, decreases in total organic mass were observed due to UV light exposure, where the amount of SOA mass loss increases with RH. Absorption of UV light by organic compounds can initiate photolysis, where the fragmentation of their carbon skeleton results in the formation of smaller products (e.g., fewer carbons atoms) that are more volatile.18,23 We speculate that these volatile photolysis products evaporate from the particle, resulting in the observed loss of total organic mass. In fact, the formation of gas-phase products, such as CO, CH4, acetone, and other small VOCs, has been observed following the photodegradation of limonene−SOA on filters and in aqueous solutions.13,14 C
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 3. Time series profiles of the org:sulfate ratios measured by the AMS for the photolysis of SOA by UVB radiation under high RH (solid line), mid RH (dashed line), and low RH (dotted line) conditions. The shaded areas for each RH condition represent the variability (±1σ) between multiple experiments.
In addition to the total organic mass, the absolute changes in specific organic fragments provide additional insight into the photolysis of SOA. Due to the evaporation/ionization techniques used in the AMS, most compounds are subject to extensive fragmentation; however, specific fragments can be used as tracers for organic compounds with specific functional groups.17 The CHO+ fragment detected at m/z 29 arises from the fragmentation of compounds containing carbonyl functional groups,24 while two different fragments can contribute to the signal observed at m/z 43, with the C2H3O+ fragment originating from functionalities with a single oxygen atom (e.g., carbonyls, alcohols, and ethers) and the C3H7+ fragment originating from aliphatic molecules.25 Control experiments using the high-resolution ToF-AMS, which can separate different ions at the same nominal m/z, indicated that the C2H3O+ ion was the dominant signal at m/z 43 for this study. The organic ion at m/z 44 (CO2+) arises from the fragmentation of highly oxidized organic acids (e.g., carboxylic acids and esters)26 and peroxides.27 For all RH conditions, upon UVB irradiation, loss of the oxidized components of SOA was observed, where the decays of m/z 29 and m/z 43 (Figure 4a,b) were faster than that of m/z 44 (Figure 4c). Given that the absorption spectrum of SOA indicates the presence of carbonyl and peroxide function groups, both of which absorb UVB radiation, it is not surprising that we observed decays of m/z 29, 43, and 44, suggesting the photodegradation of these functional groups. To gain further insight of the wavelength dependence of SOA photolysis, experiments under high RH conditions were conducted using UVA lights. Compared to experiments using UVB radiation at the same RH condition, SOA exposed to UVA radiation exhibited slower decays for total organic mass (Figure 5a) and for all organic fragments considered (Figure 5b−d). In particular, the loss of m/z 29 and 43 is greatly suppressed for UVA lights, suggesting that the photolysis of less oxidized components (i.e., carbonyls) contributes significantly to the observed photochemical aging with the UVB light. It is important to note that the slower decay of m/z 44 compared to m/z 29 and 43 can also arise from secondary reactions of photolysis which result in the formation of highly oxidized organics. For example, photolysis of peroxides is known to generate OH radicals, which can oxidize other SOA components, forming highly oxidized components such as organic acids.28,29 Direct photolysis of aldehydes has also been observed to generate organic acids as well.13,20 Given that both peroxides and carboxylic acids contribute to the organic
Figure 4. Time series profiles of the sulfate normalized (a) m/z 29, (b) m/z 43, and (c) m/z 44 for the photolysis of mixed SOA−sulfate particles by UVB radiation at high RH (solid line), mid RH (dashed line), and low RH (dotted line) conditions. The shaded areas for each RH condition represent the variability (±1σ) between multiple experiments.
fragment at m/z 44, the combined result of photodissociation of peroxides and the formation of carboxylic acids may have resulted in the slower decay of m/z 44. The relative changes in the chemical composition of SOA due to photolysis can be illustrated using the framework developed by Ng et al. in Figure 6, where aging of SOA is described using the fraction of organic mass at m/z 43 vs 44 ( f43 and f44).26 Ambient SOA typically falls within the triangle denoted by the dashed lines in Figure 6c (inset), with less oxidized SOA mainly residing in the lower half of the triangle and highly oxidized SOA in the upper half. In general, studies have shown that SOA aging processes, such as OH oxidation, result in movement toward the upper left of the triangle space.30 The trends observed for this figure are also consistent with those observed for Figure 4, with, for example, a stronger decrease for m/z 43 compared to m/z 44, with the magnitude of the change depending on RH conditions. This suggests that enhanced loss of less oxidized particulate organics due to photolysis results in a more oxidized SOA. Note that the composition of the SOA used in the current study resides in the middle region of the triangle, suggesting that the organics are somewhat oxidized prior to photolysis. D
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 5. Time series profiles of the sulfate normalized (a) total organic mass (org:sulfate ratio, black), (b) m/z 29, (c) m/z 43, and (d) m/z 44 for the photolysis of mixed SOA−sulfate particles by UVB (solid lines) and UVA (dashed lines) radiation under high RH conditions. The shaded areas for each light source represent the variability (±1σ) between multiple experiments.
measured photon flux in the chamber, the quantum yield for the loss of organics due to SOA photolysis can be estimated. Since the UV lamps used in the current study to initiate photolysis have a sharp emission cutoff at 284 nm, the absorbance below this wavelength was not considered. We note that this approach does not take into account the enhancement in photon flux that may occur inside wet particles relative to values in air.33 From the observed mass loss rates under high RH conditions for both UVB and UVA lights, the average effective photolysis quantum yield was determined to be 1.2 ± 0.2. This large quantum yield suggests that either the primary photolytic process efficiently gives rise to volatile products or secondary reactions (such as those driven by the photochemical formation of OH in the particle) may have also contributed to the observed total mass loss rates. The photolysis quantum yield for cis-pinonic acid, one of the identified compounds in α-pinene SOA, is 0.5 ± 0.3,20 i.e., also quite large and similar to our result. The factor of 2 difference could potentially be due to the presence of other more photolabile compounds in SOA material. At decreasing RH, the SOA becomes highly viscous and concentrated, where elevated concentrations can accelerate condensed phase reactions (e.g., polymerization, reactions with inorganics), thus potentially changing the absorptive properties of the material. Since the absorption cross section of SOA measured from bulk solutions may not be identical compared to particles under decreasing RH conditions, the current approach to determine the effective quantum yield for mid and low RH conditions was not pursued. Nevertheless, the suppression in mass loss rates with decreasing RH indicates that photolysis is dependent on RH conditions, where the mass loss rate under low RH is a factor of 2 slower than that at high RH. It is important to note that the observed mass loss rate at the low RH condition using UVB lights for the current study [(7.91 ± 0.13) × 10−5 s−1] is similar to the photolytic mass loss rate constant of dry α-pinene SOA−AS particles for UVA lights
Considering the experiments discussed above, the loss of organic components and changes in chemical composition due to SOA photolysis all show RH dependence, with enhanced loss observed with increasing RH conditions. We speculate that the observed RH dependence for SOA photolysis is likely due to the effects of viscosity on diffusion, as recent studies have shown that SOA exhibits liquid-like behavior under high RH and is semisolid/solid under low RH.9,10 In viscous semisolid/ solid material, the slow diffusion of SOA photodissociation products can favor their recombination, resulting in a reduction in the efficiency of photolysis. For example, the net quantum yields for the photolysis of certain amines and iodine were previously observed to be reduced in viscous media.31,32 Additionally, slow diffusion has been shown to kinetically hinder the evaporation of SOA.11 Thus, the effects of RH on diffusion can lead to the results observed in the current study, where, under lower RH conditions, slow diffusion of photolysis products resulted in less mass loss compared to higher RH conditions. 3.3. Kinetics of Photolysis. To further investigate the role of RH on the reactivity of SOA material on particles, the decay rates of particulate SOA mass loss due to photolysis were determined. The kinetic plot (shown in Figure S4, Supporting Information) indicated that the reaction exhibited first order kinetics at early times, as expected from direct photolysis. Linear fits to the first 10 min of data yield first-order mass loss rates (kSOA) for all RH and light conditions, listed in Table 1. Assuming that the initial observed mass loss is only due to the volatilization of small organic compounds from photolysis of SOA, the observed mass loss rate constants were equated to the effective photolysis rate constant: J=
∫λ σ(λ)ϕ(λ)F(λ) dλ
where σ(λ) is the absorption cross section of SOA, ϕ(λ) is the photolysis quantum yield, and F(λ) is the photon flux. Using the measured absorption cross section of SOA and the E
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
most rapid loss observed under high RH conditions. In particular, the less oxidized components of SOA (e.g., carbonyls) exhibited the fastest photodegradation, which resulted in a more oxidized SOA particle after photolytic aging. We hypothesize that the dependence of photolysis on RH conditions was due to moisture-induced viscosity changes of SOA material, affecting the diffusion of reactants/products. In this regard, despite a change in several orders of magnitude in the viscosity of the SOA material from high to low RH conditions, its photolysis was only slower by a factor of 2. This suggests that, even at such high viscosity values, photolyzed SOA material can nevertheless efficiently diffuse out of the particle and evaporate within the experimental time scale of the current study.8,10 To assess the importance of photolysis as an aging mechanism for SOA in the atmosphere, the atmospheric lifetime of SOA particles with respect to photolysis is estimated to be between that observed with the UVB lamps and that for the UVA lamps. In particular, using the experimentally determined absorption cross section and photolysis quantum yield, and calculated actinic fluxes, the initial photolysis lifetime is ∼3.7 h at Solar Noon for high RH conditions. Given that the average lifetime of particles in the atmosphere is approximately 1 week with respect to deposition, this very rough calculation indicates that photolysis can be an important sink of SOA material in the atmosphere due to volatilization. Clearly, there are uncertainties in this estimate, and this initial lifetime should not be equated to the overall lifetime for loss of SOA particles in the atmosphere. In particular, the contribution of secondary photolysis to the effective quantum yield was not constrained. Therefore, the average photolysis quantum yield determined in the current study is likely an upper limit to the primary photolysis process, corresponding to a lower limit for this calculated lifetime. Nevertheless, the effects of such secondary reactions of photolysis may be of importance to the fate of SOA. For example, it is known that photolysis of organic compounds in cloudwater is an important source of H2O2, which is an important oxidant.35 Also, we note that SOA reactivity may be dependent on its atmospheric age and the degree of prior processing. In particular, the slopes of the mass loss decay curves (e.g., Figure S4, Supporting Information) decreased with time, indicating depletion of the most photochemically active species (i.e., carbonyls and peroxides) and formation of more stable compounds, such as carboxylic acids. On the other hand, Henry and Donahue observed enhanced mass loss for the photolysis of SOA that was aged via OH oxidation, as oxidative aging by OH may lead to the formation of more carbonyl and peroxide compounds compared to photolytic aging.15 Additionally, brown carbon formed via condensed-phase reactions in aerosols has been observed to be photolabile.36 As such, it is important to understand how the reactivity of SOA changes with atmospheric age.
Figure 6. Aerosol composition described by the fraction of organic mass at m/z 43 and 44 (f43 vs f44) before (open circles), during (closed circles), and after (open circles) UV light exposure for (a) high RH, (b) mid RH, (c) and low RH conditions. The inset shows the zoomed-out view of the f43 vs f44 space (with dimensions of Figure 6a−c illustrated by the red box), where ambient SOA is distributed within the triangular area.
(6 × 10−5 s−1), reported by Henry and Donahue.15 In comparison, for dry isoprene SOA, Kroll et al.34 also observed mass loss [(1−5) × 10−5 s−1] due to irradiation with UVA lights. However, this mass loss rate might have arisen due to reaction with gas-phase OH radicals, due to the addition of H2O2 (both UV photons and OH radicals are present during irradiation).
4. CONCLUSIONS AND ATMOSPHERIC IMPLICATIONS In this work, the RH dependence for the photolysis of a chemically complex mixture, water-soluble SOA, was illustrated. Photolysis resulted in substantial organic mass loss, with the
Table 1. Total Organic Mass Loss Rate Constants for Different RH Conditionsa kSOA (s−1) light source UVB UVA a
high RH
mid RH −4
low RH −4
(1.64 ± 0.09) × 10
(1.67 ± 0.20) × 10 (4.69 ± 0.14) × 10−5
(7.91 ± 0.13) × 10−5
The uncertainties in kSOA represent variability (±1σ) between multiple experiments. F
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Diffusion in Secondary Organic Aerosol. Faraday Discuss. 2013, 165, 391−406. (7) Virtanen, A.; Joutsensaari, J.; Koop, T.; Kannosto, J.; Yli-Pirilä, P.; Leskinen, J.; Mäkelä, J. M.; Holopainen, J. K.; Pöschl, U.; Kulmala, M.; et al. An Amorphous Solid State of Biogenic Secondary Organic Aerosol Particles. Nature 2010, 467, 824−827. (8) Koop, T.; Bookhold, J.; Shiraiwa, M.; Pöschl, U. Glass Transition and Phase State of Organic Compounds: Dependency on Molecular Properties and Implications for Secondary Organic Aerosols in the Atmosphere. Phys. Chem. Chem. Phys. 2011, 13, 19238−19255. (9) Saukko, E.; Lambe, A. T.; Massoli, P.; Koop, T.; Wright, J. P.; Croasdale, D. R.; Pedernera, D. A.; Onasch, T. B.; Laaksonen, A.; Davidovits, P.; et al. Humidity-Dependent Phase State of SOA Particles from Biogenic and Anthropogenic Precursors. Atmos. Chem. Phys. 2012, 12, 7517−7529. (10) Renbaum-Wolff, L.; Grayson, J. W.; Bateman, A. P.; Kuwata, M.; Sellier, M.; Murray, B. J.; Shilling, J. E.; Martin, S. T.; Bertram, A. K. Viscosity of A-Pinene Secondary Organic Material and Implications for Particle Growth and Reactivity. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 8014−8019. (11) Cappa, C. D.; Wilson, K. R. Evolution of Organic Aerosol Mass Spectra upon Heating: Implications for OA Phase and Partitioning Behavior. Atmos. Chem. Phys. 2011, 11, 1895−1911. (12) Kolesar, K. R.; Buffaloe, G.; Wilson, K. R.; Cappa, C. D. OHInitiated Heterogeneous Oxidation of Internally-Mixed Squalane and Secondary Organic Aerosol. Environ. Sci. Technol. 2014, 48, 3196− 3202. (13) Mang, S. A.; Henricksen, D. K.; Bateman, A. P.; Andersen, M. P. S.; Blake, D. R.; Nizkorodov, S. A. Contribution of Carbonyl Photochemistry to Aging of Atmospheric Secondary Organic Aerosol. J. Phys. Chem. A 2008, 112, 8337−8344. (14) Bateman, A. P.; Nizkorodov, S. A.; Laskin, J.; Laskin, A. Photolytic Processing of Secondary Organic Aerosols Dissolved in Cloud Droplets. Phys. Chem. Chem. Phys. 2011, 13, 12199−12212. (15) Henry, K. M.; Donahue, N. M. Photochemical Aging of APinene Secondary Organic Aerosol: Effects of OH Radical Sources and Photolysis. J. Phys. Chem. A 2012, 116, 5932−5940. (16) Tang, I. N.; Fung, K. H.; Imre, D. G.; Munkelwitz, H. R. Phase Transformation and Metastability of Hygroscopic Microparticles. Aerosol Sci. Technol. 1995, 23, 443−453. (17) Canagaratna, M. R.; Jayne, J.; Jimenez, J. L.; Allan, J. D.; Alfarra, M. R.; Zhang, Q.; Onasch, T. b.; Drewnick, F.; Coe, H.; Middlebrook, A.; et al. Chemical and Microphysical Characterization of Ambient Aerosols with the Aerodyne Aerosol Mass Spectrometer. Mass Spectrom. Rev. 2007, 26, 185−222. (18) 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. (19) Shilling, J. E.; Chen, Q.; King, S. M.; Rosenoern, T.; Kroll, J. H.; Worsnop, D. R.; DeCarlo, P. F.; Aiken, A. C.; Sueper, D.; Jimenez, J. L.; et al. Loading-Dependent Elemental Composition of A-Pinene SOA Particles. Atmos. Chem. Phys. 2009, 9, 771−782. (20) Lignell, H.; Epstein, S. A.; Marvin, M. R.; Shemesh, D.; Gerber, B.; Nizkorodov, S. Experimental and Theoretical Study of Aqueous Cis-Pinonic Acid Photolysis. J. Phys. Chem. A 2013, 117, 12930− 12945. (21) Mang, S. A.; Walser, M. L.; Pan, X.; Xing, J-H; Bateman, A. P.; Underwood, J. S.; Gomez, A. L.; Park, J.; Nizkorodov, S. A. Photochemistry of Secondary Organic Aerosol Formed from Oxidation of Monoterpenes. In Atmospheric Aerosols; ACS Symposium Series; American Chemical Society: Washington, DC, 2009; Vol. 1005, pp 91−109. (22) Zhong, M.; Jang, M. Light Absorption Coefficient Measurement of SOA Using a UV−Visible Spectrometer Connected with an Integrating Sphere. Atmos. Environ. 2011, 45, 4263−4271. (23) Calvert, J. G.; Pitts, J. N. J. Photochemistry; John Wiley & Sons, Inc.: New York, 1966. (24) Lee, A. K. Y.; Hayden, K. L.; Herckes, P.; Leaitch, W. R.; Liggio, J.; Macdonald, A. M.; Abbatt, J. P. D. Characterization of Aerosol and
Also, we point out that this work has implications to photooxidation SOA chamber and flow tube studies, given that many of these reactors use UV lamps similar to those in this experiment. Thus, the changes in SOA mass and chemical composition due to photolytic aging observed in this study may also occur in other experiments. Finally, the formation mechanism of the mixed SOA−AS is not necessarily the same as that in the atmosphere, probably resembling the process by which particles are formed by cloud processing more than that from condensation of terpene oxidation products on an ammonium sulfate particle. Overall, this is one of the first studies to quantify the changes in SOA that occur upon exposure to ultraviolet light, and the first study to look at the effect of RH on this process. This RH dependence needs to be further investigated with other aging processes, SOA types, and various conditions (e.g., peroxides are more likely to be formed under low NOx conditions during SOA formation), in order to understand the processing of such species under a wide range of atmospherically relevant conditions.
■
ASSOCIATED CONTENT
S Supporting Information *
Details of absorption cross section calculations, mass distribution of SOA generated during filter collection, mass distribution of AS−SOA particles during photolysis experiments, org:sulfate ratio time trace from control experiment, and kinetic plots for total organic mass loss. 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 The authors thank Natural Sciences and Engineering Research Council and Canada Foundation for Innovation for funding.
■
REFERENCES
(1) 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.; et al. Ubiquity and Dominance of Oxygenated Species in Organic Aerosols in Anthropogenically-Influenced Northern Hemisphere Midlatitudes. Geophys. Res. Lett. 2007, 34, L13801. (2) Heald, C. L.; Ridley, D. A.; Kreidenweis, S. M.; Drury, E. E. Satellite Observations Cap the Atmospheric Organic Aerosol Budget. Geophys. Res. Lett. 2010, 37, L24808. (3) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.; Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.; Goldstein, A. H.; et al. The Formation, Properties and Impact of Secondary Organic Aerosol: Current and Emerging Issues. Atmos. Chem. Phys. 2009, 9, 5155−5236. (4) Lee, J. W. L.; Carrascón, V.; Gallimore, P. J.; Fuller, S. J.; Björkegren, A.; Spring, D. R.; Pope, F. D.; Kalberer, M. The Effect of Humidity on the Ozonolysis of Unsaturated Compounds in Aerosol Particles. Phys. Chem. Chem. Phys. 2012, 14, 8023−8031. (5) Nguyen, T. B.; Coggon, M. M.; Flagan, R. C.; Seinfeld, J. H. Reactive Uptake and Photo-Fenton Oxidation of Glycolaldehyde in Aerosol Liquid Water. Environ. Sci. Technol. 2013, 47, 4307−4316. (6) Zhou, S.; Shiraiwa, M.; McWhinney, R. D.; Pöschl, U.; Abbatt, J. P. D. Kinetic Limitations in Gas-Particle Reactions Arising from Slow G
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Cloud Water at a Mountain Site during WACS 2010: Secondary Organic Aerosol Formation through Oxidative Cloud Processing. Atmos. Chem. Phys. 2012, 12, 7103−7116. (25) Zhang, Q.; Alfarra, M. R.; Worsnop, D. R.; Allan, J. D.; Coe, H.; Canagaratna, M. R.; Jimenez, J. L. Deconvolution and Quantification of Hydrocarbon-like and Oxygenated Organic Aerosols Based on Aerosol Mass Spectrometry. Environ. Sci. Technol. 2005, 39, 4938− 4952. (26) 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.; et al. Organic Aerosol Components Observed in Northern Hemispheric Datasets from Aerosol Mass Spectrometry. Atmos. Chem. Phys. 2010, 10, 4625−4641. (27) Aiken, A. C.; DeCarlo, P. F.; Jimenez, J. L. Elemental Analysis of Organic Species with Electron Ionization High-Resolution Mass Spectrometry. Anal. Chem. 2007, 79, 8350−8358. (28) Lim, Y. B.; Tan, Y.; Perri, M. J.; Seitzinger, S. P.; Turpin, B. J. Aqueous Chemistry and Its Role in Secondary Organic Aerosol (SOA)Formation. Atmos. Chem. Phys. 2010, 10, 10521−10539. (29) 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. 2011, 11, 6465−6474. (30) Slowik, J. G.; Wong, J. P. S.; Abbatt, J. P. D. Real-Time, Controlled OH-Initiated Oxidation of Biogenic Secondary Organic Aerosol. Atmos. Chem. Phys. 2012, 12, 9775−9790. (31) Ratcliff, M. A.; Kochi, J. K. Cage Effects and the Viscosity Dependence of the Photolysis of Dibenzylamine and Tribenzylamine. J. Org. Chem. 1972, 37, 3275−3281. (32) Otto, B.; Schroeder, J.; Troe, J. Photolytic Cage Effect and Atom Recombination of Iodine in Compressed Gases and Liquids: Experiments and Simple Models. J. Chem. Phys. 1984, 81, 202−213. (33) Nissenson, P.; Knox, C. J. H.; Finlayson-Pitts, B. J.; Phillips, L. F.; Dabdub, D. Enhanced Photolysis in Aerosols: Evidence for Important Surface Effects. Phys. Chem. Chem. Phys. 2006, 8, 4700− 4710. (34) Kroll, J. H.; Ng, N. L.; Murphy, S. M.; Flagan, R. C.; Seinfeld, J. H. Secondary Organic Aerosol Formation from Isoprene Photooxidation. Environ. Sci. Technol. 2006, 40, 1869−1877. (35) Faust, B. C. Photochemistry of Clouds, Fogs, and Aerosols. Environ. Sci. Technol. 1994, 28, 216A−222A. (36) Sareen, N.; Moussa, S. G.; McNeill, V. F. Photochemical Aging of Light-Absorbing Secondary Organic Aerosol Material. J. Phys. Chem. A 2013, 117, 2987−2996.
H
dx.doi.org/10.1021/jp506898c | J. Phys. Chem. A XXXX, XXX, XXX−XXX
