Silicon is a Frequent Component of Atmospheric Nanoparticles

Article pubs.acs.org/est

Silicon is a Frequent Component of Atmospheric Nanoparticles Bryan R. Bzdek,† Andrew J. Horan,† M. Ross Pennington,† Nathan J. Janechek,‡ Jaemeen Baek,§ Charles O. Stanier,‡,§ and Murray V. Johnston*,† †

Department of Chemistry and Biochemistry, University of Delaware, Newark, Delaware 19716, United States Department of Chemical and Biochemical Engineering, University of Iowa, Iowa City, Iowa 52242, United States § IIHR Hydroscience and Engineering, University of Iowa, Iowa City, Iowa 52242, United States ‡

S Supporting Information *

ABSTRACT: Nanoparticles are the largest fraction of aerosol loading by number. Knowledge of the chemical components present in nanoparticulate matter is needed to understand nanoparticle health and climatic impacts. In this work, we present field measurements using the Nano Aerosol Mass Spectrometer (NAMS), which provides quantitative elemental composition of nanoparticles around 20 nm diameter. NAMS measurements indicate that the element silicon (Si) is a frequent component of nanoparticles. Nanoparticulate Si is most abundant in locations heavily impacted by anthropogenic activities. Wind direction correlations suggest the sources of Si are diffuse, and diurnal trends suggest nanoparticulate Si may result from photochemical processing of gas phase Si-containing compounds, such as cyclic siloxanes. Atmospheric modeling of oxidized cyclic siloxanes is consistent with a diffuse photochemical source of aerosol Si. More broadly, these observations indicate a previously overlooked anthropogenic source of nanoaerosol mass. Further investigation is needed to fully resolve its atmospheric role.

■

INTRODUCTION

of Si suggests that an important anthropogenic contributor to nanoparticle mass has been overlooked.

■

Atmospheric nanoparticles (0.01 (%)

ref

Lewes, Delaware Pasadena, California Wilmington, Delaware Wilmington, Delaware Wilmington, Delaware Lewes, Delaware Hyytiälä, Finland

July−August 2012 May−June 2010 May 2006 July 2009 December 2009 October−November 2007 March−April 2011

rural/coastala suburban urban urban urban rural/coastala remote boreal forest

48 40 13 7.0 4.8 2.5 1.5

Bzdek et al.16 Pennington et al.20 Zordan et al.21 Klems et al.18 Klems et al.18 Bzdek et al.17 Pennington et al.19

In the summertime only, Lewes is highly populated.

according to 2010 U.S. census data using PopGrid, version 4.3. The emissions were merged with 2004 emissions processed by the Sparse Matrix Operator Kernel Emissions (SMOKE) model version 2.5. Oxidized cyclic siloxanes were represented in the model only as o-D4, o-D5, and o-D6. In other words, explicit molecular identification of products was not included. This is justified given the lack of knowledge about the distribution of products under atmospherically relevant oxidation conditions. Specific products have been identified in chamber studies using GC/MS analysis of filters, denuders, and extracts from reaction vessels.29,30 Studies also propose multistep mechanisms leading to a first generation of stable oxidation products. Sommerlade et al. (1993) detected D3TOH as a major product species from reaction of D4 using GC/MS.30 However, D3TOH only accounted for 46% of the products resolved by GC/MS, several other products were resolved, and still other products could not be resolved. Fragmentation in the GC/MS may occur as well. Whelan et al. (2004) modeled multigenerational oxidation leading to progressive substitution of methyl groups by alcohol groups, but experimental observations of atmospheric products to support their model are not available.38 Our approach of tracking the oxidized cyclic siloxane mass in the model is an upper limit, and further experimental refinement of the mechanism and product distribution is needed. Meteorological input was from the Weather Research and Forecasting (WRF) model, version 3.1.1, and represents the year 2004. While both meteorology and nonsiloxane emissions data represent the year 2004, the model output should be representative of summertime concentrations of siloxanederived atmospheric Si. Boundary conditions or emissions for non-U.S. countries are not included. Dry and wet deposition was not added to the model for the new cyclic siloxane species; therefore, the o-D4, o-D5, and o-D6 concentrations are upper bounds. Analysis of the model results was for the period 21 May to 10 June; however, the model simulation was begun on 15 May, and the six day spin up period of the model was used to reduce the influence of the zero initial condition used for the siloxanes.

to twenty-three nanometer mass normalized diameter particles were studied. Importantly, during all campaigns where ambient particles were sampled into NAMS, copper tubing was used in the inlet. No silicone tubing was used in the field setup. This tubing has been shown to emit contaminants that can impact particle composition.24,25 Additionally, Si was absent from the mass spectra of calibration aerosols at these sites, confirming that Si contamination did not arise from the measurement apparatus itself. Aerosol mass concentration measurements were accomplished with scanning mobility particle sizers (SMPS): electrostatic classifier model 3080 and condensation particle counter (CPC) model 3025a for Pasadena or CPC model 3788 for Lewes (TSI, Inc., St. Paul, Minnesota). Gas phase sulfuric acid in Lewes was measured using the Cluster Chemical Ionization Mass Spectrometer.26,27 Atmospheric Modeling. Model simulations of the cyclic siloxanes octamethylcyclotetrasiloxane (D4), decamethylcyclopentasiloxane (D5), and dodecamethylcyclohexasiloxane (D6), which are common organosilicon compounds present in personal care products, were performed using the chemical transport model Community Multiscale Air Quality (CMAQ). CMAQ, version 4.7.1, was used to model the contiguous U.S. at a horizontal resolution of 36 km for the period of 21 May to 10 June, matching the time of year of the CalNex measurements. Cyclic siloxanes are volatile, favor the gas phase, and react with hydroxyl (OH) radicals.28−30 The cyclic siloxane species, their reactions with the OH radical, and oxidized cyclic siloxanes (o-D4 , o-D5 , and o-D 6 ) were added to the cb05cl_ae5_aq mechanism within CMAQ. The individual cyclic siloxane OH reaction rate constants were from Atkinson 1991.28 An emission rate for D5 of 135.2 mg·person−1day−1 was used.31,32 This emission rate is based on 2008−2009 antiperspirant sales data and 2009 consumption data as previously described elsewhere.33 Since only a recent D5 emission rate was available, the D4 and D6 emission rates were calculated by multiplying the D5 emission rate by outdoor concentration ratios.34 Emissions ratios were more representative of current conditions given the lack of recent publicly available consumption or emissions data, coupled with ongoing formulation changes such as the substitution of D5 for D4 over time.35,36 This work uses the observed atmospheric concentration ratios (D4/D5 = 0.34, 10th and 90th percentiles of 0.18 and 0.53, respectively, and D6/D5 = 0.092, 10th and 90th percentiles of 0.036 and 0.16, respectively) directly without additional analysis to calculate emission ratios. In other words, no correction is made for the shift in ratios as photochemical aging proceeds.37 Sensitivity to this assumption is considered in the Results and Discussion section. Non-U.S. emissions were not included. Emission rates were spatially distributed

■

RESULTS AND DISCUSSION Table 1 presents a summary of seven field campaigns where nanoparticle composition was measured by NAMS, a single particle mass spectrometer that provides quantitative nanoparticle elemental composition. A surprising observation during many of these campaigns is that Si was found in a large fraction of ambient nanoparticles, which is indicated in Table 1 by the percentage of particles containing Si elemental mole fraction >0.01. Some potential sources of gaseous siloxanes that could be oxidized in the atmosphere to produce nanoparticulate Si include personal care products and polymers and coatings used 11138

dx.doi.org/10.1021/es5026933 | Environ. Sci. Technol. 2014, 48, 11137−11145

Environmental Science & Technology

Article

Figure 1. (a) Wind rose plot showing average Si mole fraction for particles analyzed during the Pasadena, California, campaign. Numbers in the plot indicate the number of 1 h time blocks when wind came from that direction and NAMS was fully operational. (b) Si, S, and C elemental mass concentrations in particulate matter between 20 and 25 nm diameter on 01−02 June 2010 during the Pasadena campaign. Mass concentrations were averaged to 1 h time blocks.

analyzed nanoparticles contained a substantial mole fraction of Si. Complicating the urban/remote interpretation is that the level of human activity onsite during the measurement period was also different. In Pasadena, several tens of researchers were consistently onsite during the campaign, which was located on a college campus with substantial human activity nearby, whereas only a handful of researchers or other personnel were consistently onsite in Hyytiälä. In both locations, many instruments were in use including a variety of mass spectrometers, particle collectors, sizers, and counters. In each case, “best practice” protocols for sensitive atmospheric measurements were in use to minimize cross-contamination and artifacts associated with colocation of instruments.

in or on commercial materials, such as building products, scientific equipment, and tubing.24,36,39,40 These sources are related to human activity and could be manifested on both local and regional scales. Whereas Si is a major component of windblown crustal matter, it is not expected to contribute to nanoparticulate mass. Several general characteristics of nanoparticulate Si are found in Table 1. First, Si is frequently observed in urban and suburban environments but rarely detected in remote environments. For example, in Pasadena, California (a densely populated suburban environment) 40% of all nanoparticles analyzed by NAMS contained a Si mole fraction >0.01, whereas in Hyytiälä, Finland (a remote environment) only 1.5% of the 11139



Article

Pasadena. These data are calculated by multiplying the NAMSmeasured elemental mass fraction by the SMPS-measured aerosol mass concentration in the 19−26 nm mobility diameter size channels. On these days, plumes of nanoparticles resulting from photochemical processing of rush hour motor vehicle emissions advected to the site from downtown Los Angeles.20 Such events were frequently observed during the campaign. The Si mass concentration is small during the morning (0.01 with no discernible wind direction dependence (see Figure S1 in the Supporting Information). Similar to Pasadena, the Si mass concentration generally increases during the daytime, concurrent with S mass concentration, although the two elements do not always exactly track each other, suggesting the existence of different sources or production pathways for each. Also during the Lewes summer campaign, nanoparticulate Si was more prevalent on days dominated by local aerosol than those dominated by regional events, suggesting that local source emissions (whether from activity on-site or nearby), were stronger than regional source emissions.16 Because the method of nanoparticle analysis provides only elemental composition, it is not possible to determine the molecular forms of Si in the nanoparticles. However, several studies have addressed the atmospheric concentrations and lifetimes of organosilicon compounds such as cyclic siloxanes, which are widely used in personal care products and industrial applications and which have the potential for long-range transport and bioaccumulation.36,39,40 Atmospheric siloxane concentrations have been shown to vary widely depending upon location and time of day from 1000 ng· m−3.33,34,43−48 Organosilicon compounds may undergo atmospheric oxidation by the OH radical28 and can be taken up onto aerosol.29,49−52 During the Lewes campaign, nanoparticle chemical composition was measured during new particle formation events, where particles grew rapidly to larger sizes.6,16 Based on knowledge of the nanoparticle composition and volume growth rate, it is possible to get a rough estimate of the concentration of condensable Si in the gas phase. The calculation is similar to that used to quantify the gas phase sulfuric acid concentration from the particle phase S mole fraction measurement during new particle formation.42 (The calculation is shown in detail in Supporting Information.) The goal is to determine whether or not it is feasible for organosiloxanes that have been previously detected in air to

Therefore, the difference in Si content between Pasadena and Hyytiälä is more likely due to human activity, whether it is regional or adjacent to the site, than instrumentation. Second, the fraction of nanoparticles containing Si changes across different campaigns in the same location. The most obvious example is the two campaigns in Lewes, Delaware. In a campaign there in fall 2007 very few particles contained Si, whereas in summer 2012 nearly half of all analyzed particles contained Si. The local environment was very different during each campaign: (1) because of seaside tourism, the population in Lewes is very large during summer but small during fall, winter, and spring, (2) human activity onsite was substantial during the summer campaign but very small during the fall campaign, and (3) a wide range of instrumentation was in use during the summer campaign but only NAMS was present during the fall campaign. As discussed above for Pasadena and Hyytiälä, the difference between Lewes in the summer and fall is more likely human activity, whether regional or adjacent to the site, than instrumentation. In urban Wilmington, Delaware, campaign-to-campaign differences are also evident, with more particles containing Si analyzed during campaigns in spring and summer, and fewer particles containing Si analyzed during winter. Among the field sites studied, Wilmington is unique in that a substantial fraction of measured nanoparticles arise directly from motor vehicle emissions within a few tens to hundreds of meters from the sampling location.18,41 These particles are sampled within seconds of emission, leaving very little time for condensation of secondary species to occur. The campaign-to-campaign variation at this site most likely arises from the seasonal change in the fraction of particles from motor vehicles, which was much higher in the winter than summer. Finally, we note qualitatively that across all campaigns, Si is more abundant in periods of higher temperature (Pasadena, Lewes summer, Wilmington spring/summer) than lower temperature (Lewes fall, Wilmington winter, Hyytiäl ä spring). The higher abundance of Si in Wilmington spring than Wilmington summer is not readily explained by differences in temperature or photochemical activity, and may reflect different precursor source strengths during the two measurement periods. We now focus in more detail on measurements performed in Pasadena because (1) Si was observed frequently in ambient nanoparticles in this location, (2) this environment is of substantial interest with respect to air quality, and (3) mass concentrations in the nanoparticle size regime were much larger here than in other locations. Figure 1a presents a wind rose plot showing the 1 h average Si mole fraction as a function of wind direction at the site. The numbers inside the plot indicate the number of 1 h time blocks during which the wind came from that direction and NAMS was fully operational. The average Si mole fractions range from 0.02 to 0.06 depending upon wind direction. However, there is no clear wind direction dependence. Note that the wind came from the northwest during only 8 of the 320 h time-blocks, and the Si mole fraction varied substantially among these time-blocks. In fact, the standard deviation of each wind direction average is of the same magnitude as the average itself. In other words, the Si mole fraction from each wind direction is highly variable, and there is no evidence for dominant stationary sources. Together, these observations argue for a diffuse source of the Si around the site. Figure 1b presents time dependent elemental mass concentrations for 20−25 nm mass normalized diameter nanoparticles for two typical days during the campaign in 11140



Article

Table 2. Estimationa of Gas Phase o-D5 Concentration during New Particle Formation Events in Lewes event date

volume growth rateb (nm·h−1)

Si mole fractionc

S mole fractionc

[o-D5]a (molecules·cm−3)

[H2SO4]d (molecules·cm−3)

12 August 2012 13 August 2012 21 August 2012

8.9 ± 0.8 7.1 ± 1.0 9.8 ± 0.3

0.005 ± 0.0005 0.007 ± 0.0007 0.006 ± 0.0006

0.04 ± 0.004 0.04 ± 0.004 0.06 ± 0.006

1.3 × 10 1.5 × 106 1.6 × 106

1.9 ± 1.0 × 107 2.2 ± 1.1 × 107 1.3 ± 0.7 × 107

6

a

Estimated values depend in part on the chemical form of this species, which for the purpose of this calculation is assumed to include 5 Si atoms per molecule. The estimation also depends on the physical properties of o-D5; the uncertainty associated with physical properties is expected to be on the order of ±50% of the tabulated concentration (see Supporting Information). bCalculated from the SMPS measurement. cMeasured directly by NAMS. dMeasured directly by the Cluster Chemical Ionization Mass Spectrometer.

The explanation of a siloxane photochemical source of nanoparticulate Si was further investigated through 3-D modeling with the CMAQ chemical transport model. CMAQ modeling permitted visualization of the expected spatial distribution of gas phase cyclic siloxanes and oxidized reaction products. While numerous previous modeling studies of cyclic siloxanes have been conducted, none has included gas phase reaction and transport of both parent siloxanes and products. Previous modeling has included 3-D atmospheric modeling of D5 at hemispheric scales,33,45,47 global modeling of D5 using a multimedia mass balance model at 15° resolution,45,53 box modeling of D4 and D5,52 global multimedia fate and transport modeling using a zonally averaged global model,54,55 and box modeling (including partitioning to cloud droplets and aerosols) over the U.S. of D4 including proposed multiple generations of reaction products.38 Calculated partitioning properties and subsequent multimedia partitioning behavior for cyclic siloxanes and oxidized reaction products based on structure activity relationships have also been investigated.56−58 In this work, the three most common cyclic siloxanes used in personal care products (D4, D5, and D6) as well as the corresponding oxidized reaction products were modeled for the continental U.S. using CMAQ. Model results are summarized in Figure 2. Averages in the model surface layer are mapped for the period of 21 May to 10 June coinciding with CalNex. Figure 2a presents the modeled U.S. cyclic siloxanes (summed D4, D5, and D6). The spatial distribution of cyclic siloxanes shows enhancement in populated areas agreeing with recent Midwestern measurements that found peak concentrations in Chicago.34 Similarly, Figure 2b shows enhanced levels of the reaction products of cyclic siloxanes in Southern California and the Eastern U.S., precisely the locations (Pasadena, Lewes) found by NAMS to have prevalent nanoparticulate Si. Model-observation comparison of the S/Si mole ratio yields insight about the nature of nanoparticle photochemical Si as well (Figure 2c). The S/Si mole ratio is directly measured by NAMS and can be inferred from CMAQ. If gas phase oxidation of the precursors controls production of condensable S and Si, then particle composition should be proportional to the gas phase SO2/cyclic siloxane mole ratio multiplied by the appropriate oxidation rates. Figure 2c plots the SO2/cyclic siloxane mole ratio. In areas with a high SO2/cyclic siloxane mole ratio (denoted by warm colors) high S/Si mole ratios are expected in photochemically produced nanoparticles. Conversely, in areas with a low SO2/cyclic siloxane mole ratio (cool colors), a relative enhancement of nanoparticulate Si is expected. Locations such as Los Angeles, Chicago, and the East Coast display some of the lowest SO2/cyclic siloxane mole ratios due to increased cyclic siloxane concentrations. On the other hand, rural locations have higher SO2/cyclic siloxane mole ratios. The high SO2/cyclic siloxane mole ratios present

provide the requisite growth of nanoparticles during these events. Note that any nanoparticulate Si measured during new particle formation necessarily must arise from regional gas-toparticle conversion. For this assessment, the molecular precursor is assumed to be D5, a cyclic siloxane whose gas phase concentration has been reported several times.33,34,36,43−47 The main pathway of D5 oxidation is by OH28 leading to formation of one or more condensable oxidized products we will refer to as o-D5. For the purpose of estimating the gas phase concentration of o-D5, we assume its physical properties are similar to the D5 precursor except that the equilibrium vapor pressure has decreased substantially. Measured particle phase Si mole fractions and estimated gas phase o-D5 concentrations are given in Table 2 for three particle formation events. Also shown are corresponding measured particle phase S mole fractions and measured gas phase sulfuric acid concentrations for the same events. As discussed elsewhere,16 the sulfuric acid concentrations calculated from the S mole fractions agree quantitatively with the measured sulfuric acid concentrations. It should be noted that the estimated o-D5 concentrations represent lower limits since they assume condensational growth rather than equilibrium partitioning to the particle phase. For these regional particle formation events, the o-D5 concentrations required to grow the particles are on the order of 1 × 106 molecules·cm−3 assuming 5 Si atoms per molecule. (The gas phase concentration would increase proportionally for fewer Si atoms per molecule.) Except for remote sites,33,46,47 gas phase D5 concentrations are generally on the order of several tens to hundreds of ng·m−3 or 107−108 molecules·cm−3.34,43−45 If gas phase D5 concentrations in Lewes were at a similar level during the particle formation events, then photooxidation of D5 to o-D5 could potentially explain the measured particle phase Si mole fractions. Particulate, noncrustal Si has not been reported in much detail before, with the exception of a large source of nanoparticulate Si observed in a study in Houston, Texas,11 and much smaller amounts in Pittsburgh, Pennsylvania,10 and Atlanta, Georgia.12 If photochemical processes are the source of low volatility Si-containing products that condense onto particles, the impact will be more apparent on a mole fraction basis in nanoparticles than in larger (e.g., accumulation mode) particles because of the larger surface to volume ratio for nanoparticles, and also because the Kelvin effect restricts the range of semivolatile compounds able to partition into nanoparticles. Additionally, measurements of particulate Si at larger sizes may be skewed by the presence of crustal Si, which would not be present in nanoparticles. While no Si contamination was observed from the measurement apparatus itself (see Methods Section), we cannot rule out that siloxane emissions from containers and other equipment adjacent to the measurement site influenced the results. 11141



Article

the D5 emission rate used in this study. As mentioned in the Methods Section, the D4 and D6 emission estimates are not well-known. For this work, we used calculated emission ratios, but the observational data set was small (n = 15), lacked spatial and temporal coverage (3 measurement sites, all in the Midwest U.S. during the summer months), and collocated measurements (e.g., NO and NOy) for determination of photochemical age were not available. However, the direction and magnitude of the adjustment for photochemical aging can be estimated. The D4/D5 ratio should be lowest in fresh emissions and then increase due to differential reaction rates with OH. The D6/D5 ratio should behave in the opposite fashion. The highest concentration samples of siloxanes did have the lowest D4/D5 ratios, and using the 10th percentile value instead of the mean would reduce D4 concentrations by up to 47% and reduce total siloxanes by up to 11%. An increase in D6 emissions to the 90th percentile of the observed ratios would increase D 6 concentrations by up to 74% and increases total siloxanes by up to 5%. Large changes to the emitted ratios of the cyclic siloxanes would be required to change the spatial patterns of Figure 2. Sensitivity to neglected deposition in o-D4, o-D5, and o-D6 was evaluated using a representative depositional lifetime of 5 days.59 This is a typical lifetime for sulfate, which is extremely water-soluble and expected to be more soluble than the oxidized cyclic siloxanes. Deposition removal probability increases with atmospheric age of the species. For species with an age of 1 day the expected depositional loss is approximately 18%, increasing to 63% at 5 days. Atmospheric age of oxidized siloxanes is lowest near sources (urban areas) and we argue that depositional loss should have low influence on modeled concentrations in these locations. The model will overestimate oxidized siloxanes in locations well downwind of source regions due to neglected deposition. In conclusion, the element Si has been observed in measurements of nanoparticle elemental composition in several locations. Si was more frequently detected in locations where anthropogenic impacts are expected to be greater, whereas Si was rarely observed in areas with little anthropogenic impact. Si was shown to have no obvious wind direction dependence, suggesting its sources are diffuse. The Si mass concentration tended to increase during the daytime, suggesting its presence in nanoparticles may arise from photochemical processing of gas-phase precursors containing Si. An initial estimate of the gas phase concentration for the condensing species suggests that oxidation of cyclic siloxanes such as D5 could explain the presence of nanoparticulate Si, but much more work is required to show this rigorously. Similarly, preliminary modeling of cyclic siloxanes and their oxidation products are consistent with the NAMS observations, corroborating the interpretation of the NAMS field measurements. Future work should address the sources, oxidation pathways, partitioning behavior, hygroscopicity, and molecular forms of atmospheric particulate Si. Such knowledge would better constrain the lifetimes of Si-containing compounds in the atmosphere as well as the mechanism of human exposure (e.g., gas or aerosol phase). Future measurement campaigns would benefit from simultaneous gas and particle phase siloxane measurements performed across a wide range of locations. The presence of photochemically produced Si in ambient nanoparticles indicates that photochemically produced Si must also be a component of larger particles, and its quantitative contribution to accumulation mode mass should be inves-

Figure 2. Surface layer CMAQ model results averaged from 21 May to 10 June. Species plotted include (a) the sum of D4, D5, and D6 cyclic siloxanes, (b) the sum of the oxidized cyclic siloxane species (o-D4, oD5, and o-D6), and (c) the SO2/cyclic siloxane mole ratio.

over the Pacific, in Mexico, and in Canada are likely the result of zero boundary conditions for siloxanes. These modeled results suggest that Los Angeles and the East Coast are both locations where the Si contribution to nanoparticle mass would be most significant, which corresponds well with NAMS field observations in these locations. The modeled results also suggest for Los Angeles that o-D5 concentrations should peak in the mid to late afternoon, which coincides with the measured peak mass concentrations of Si in Figure 1b. Cyclic siloxane emissions are uncertain but our conclusions are likely robust relative to these uncertainties. A recent observational emission estimate32 was within a factor of 1.4 of 11142



Article

(6) Zhang, R.; Khalizov, A.; Wang, L.; Hu, M.; Xu, W. Nucleation and growth of nanoparticles in the atmosphere. Chem. Rev. 2012, 112 (3), 1957−2011. (7) Bzdek, B. R.; Johnston, M. V. New particle formation and growth in the troposphere. Anal. Chem. 2010, 82 (19), 7871−7878. (8) Ehn, M.; Thornton, J. A.; Kleist, E.; Sipila, M.; Junninen, H.; Pullinen, I.; Springer, M.; Rubach, F.; Tillmann, R.; Lee, B.; LopezHilfiker, F.; Andres, S.; Acir, I.-H.; Rissanen, M.; Jokinen, T.; Schobesberger, S.; Kangasluoma, J.; Kontkanen, J.; Nieminen, T.; Kurten, T.; Nielsen, L. B.; Jorgensen, S.; Kjaergaard, H. G.; Canagaratna, M.; Maso, M. D.; Berndt, T.; Petaja, T.; Wahner, A.; Kerminen, V.-M.; Kulmala, M.; Worsnop, D. R.; Wildt, J.; Mentel, T. F. A large source of low-volatility secondary organic aerosol. Nature 2014, 506 (7489), 476−479. (9) Zhao, J.; Ortega, J.; Chen, M.; McMurry, P. H.; Smith, J. N. Dependence of particle nucleation and growth on high-molecularweight gas-phase products during ozonolysis of alpha-pinene. Atmos. Chem. Phys. 2013, 13 (15), 7631−7644. (10) Bein, K. J.; Zhao, Y. J.; Wexler, A. S.; Johnston, M. V. Speciation of size-resolved individual ultrafine particles in Pittsburgh, Pennsylvania. J. Geophys. Res.: Atmos. 2005, 110 (D7), No. D07s05, DOI: 10.1029/2004jd004708. (11) Phares, D. J.; Rhoads, K. P.; Johnston, M. V.; Wexler, A. S. Sizeresolved ultrafine particle composition analysis2. Houston. J. Geophys. Res.: Atmos. 2003, 108 (D7), 8420 DOI: 10.1029/ 2001jd001212. (12) Rhoads, K. P.; Phares, D. J.; Wexler, A. S.; Johnston, M. V. Sizeresolved ultrafine particle composition analysis, 1. Atlanta. J. Geophys. Res.: Atmos. 2003, 108 (D7), 8418 DOI: 10.1029/2001jd001211. (13) Pennington, M. R.; Johnston, M. V. Trapping charged nanoparticles in the nano aerosol mass spectrometer (NAMS). Int. J. Mass Spectrom. 2012, 311 (1), 64−71. (14) Wang, S. Y.; Johnston, M. V. Airborne nanoparticle characterization with a digital ion trap-reflectron time of flight mass spectrometer. Int. J. Mass Spectrom. 2006, 258 (1−3), 50−57. (15) Wang, S. Y.; Zordan, C. A.; Johnston, M. V. Chemical characterization of individual, airborne sub-10-nm particles and molecules. Anal. Chem. 2006, 78 (6), 1750−1754. (16) Bzdek, B. R.; Horan, A. J.; Pennington, M. R.; DePalma, J. W.; Zhao, J.; Jen, C. N.; Hanson, D.; Smith, J. N.; McMurry, P. H.; Johnston, M. V. Quantitative and time-resolved nanoparticle composition measurements during new particle formation. Faraday Discuss. 2013, 165 (1), 25−43. (17) Bzdek, B. R.; Zordan, C. A.; Luther, G. W.; Johnston, M. V. Nanoparticle chemical composition during new particle formation. Aerosol Sci. Technol. 2011, 45 (8), 1041−1048. (18) Klems, J. P.; Pennington, M. R.; Zordan, C. A.; McFadden, L.; Johnston, M. V. Apportionment of motor vehicle emissions from fast changes in number concentration and chemical composition of ultrafine particles near a roadway intersection. Environ. Sci. Technol. 2011, 45 (13), 5637−5643. (19) Pennington, M. R.; Bzdek, B. R.; DePalma, J. W.; Smith, J. N.; Kortelainen, A.-M.; Hildebrandt Ruiz, L.; Petaja, T.; Kulmala, M.; Worsnop, D. R.; Johnston, M. V. Identification and quantification of particle growth channels during new particle formation. Atmos. Chem. Phys. 2013, 13 (20), 10215−10225. (20) Pennington, M. R.; Klems, J. P.; Bzdek, B. R.; Johnston, M. V. Nanoparticle chemical composition and diurnal dependence at the CalNex Los Angeles Ground Site. J. Geophys. Res.: Atmos. 2012, 117 (D00), No. D00V10, DOI: 10.1029/2011JD017061. (21) Zordan, C. A.; Wang, S.; Johnston, M. V. Time-resolved chemical composition of individual nanoparticles in urban air. Environ. Sci. Technol. 2008, 42 (17), 6631−6636. (22) Bzdek, B. R.; Lawler, M. J.; Horan, A. J.; Pennington, M. R.; DePalma, J. W.; Zhao, J.; Smith, J. N.; Johnston, M. V. Molecular constraints on particle growth during new particle formation. Geophys. Res. Lett. 2014, DOI: 10.1002/2014gl060160. (23) Ryerson, T. B.; Andrews, A. E.; Angevine, W. M.; Bates, T. S.; Brock, C. A.; Cairns, B.; Cohen, R. C.; Cooper, O. R.; de Gouw, J. A.;

tigated. Additionally, if cyclic siloxanes are indeed an important source of atmospheric nanoparticulate Si, these observations suggest studies of nanoparticle composition in indoor environments are warranted, as indoor siloxane concentrations can be much larger than outdoor concentrations34,44 and indoor concentrations of oxidizing species can be of the same order of magnitude as outdoor concentrations.60,61 More generally, these observations indicate that nanoparticulate Si is widespread, has been previously overlooked, and requires further study to elucidate its sources and atmospheric fate.

■

ASSOCIATED CONTENT

■

AUTHOR INFORMATION

S Supporting Information *

One figure and a description of the estimated gas phase o-D5 concentration calculation. This material is available free of charge via the Internet at http://pubs.acs.org. Corresponding Author

*Telephone: (302) 831-8014. Fax: (302) 831-6335. E-mail: [email protected]. Notes

The views expressed are solely those of the authors, and the U.S. EPA does not endorse any products or commercial services mentioned in this publication. The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS M.V.J. acknowledges support of NSF grant number AGS1205304. C.O.S. acknowledges support of NSF grant number ATM-0748602. Funding for a portion of the CalNex campaign was provided by the California Air Resources Board. B.R.B. acknowledges a STAR Graduate Fellowship (FP-91731501) awarded by the U.S. Environmental Protection Agency (EPA). We thank Joost A. de Gouw, Jose-Luis Jimenez, John H. Seinfeld, and Jochen Stutz for organization of the CalNex ground site and the NOAA Earth System Research Laboratory for provision of meteorological data at the site. We acknowledge Jun Zhao for a nanoparticle growth rate calculation in Lewes and Joseph W. DePalma for computational modeling of the D5 molecule. We thank George W. Luther, III, for providing access to the Lewes site and facilitating the measurement campaign, Peter H. McMurry for constructive comments in the preparation of this manuscript, and Roger van Egmond for providing the D5 per capita emission rate estimate used in ref 33.

■

REFERENCES

(1) Gong, H.; Linn, W. S.; Clark, K. W.; Anderson, K. R.; Sioutas, C.; Alexis, N. E.; Cascio, W. E.; Devlin, R. B. Exposures of healthy and asthmatic volunteers to concentrated ambient ultrafine particles in Los Angeles. Inhal. Toxicol. 2008, 20 (6), 533−545. (2) Maudgalya, T.; Genaidy, A.; Weckman, G.; Shell, R.; Karwowski, W.; Wallace, S. A critical appraisal of epidemiological studies investigating the effects of ultrafine particles on human health. Hum. Fact. Ergon. Man. 2008, 18 (3), 358−373. (3) Oberdorster, G.; Oberdorster, E.; Oberdorster, J. Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 2005, 113 (7), 823−839. (4) Charlson, R. J.; Schwartz, S. E.; Hales, J. M.; Cess, R. D.; Coakley, J. A.; Hansen, J. E.; Hofmann, D. J. Climate forcing by anthropogenic aerosols. Science 1992, 255 (5043), 423−430. (5) Lohmann, U.; Feichter, J. Global indirect aerosol effects: A review. Atmos. Chem. Phys. 2005, 5 (3), 715−737. 11143



Article

Fehsenfeld, F. C.; Ferrare, R. A.; Fischer, M. L.; Flagan, R. C.; Goldstein, A. H.; Hair, J. W.; Hardesty, R. M.; Hostetler, C. A.; Jimenez, J. L.; Langford, A. O.; McCauley, E.; McKeen, S. A.; Molina, L. T.; Nenes, A.; Oltmans, S. J.; Parrish, D. D.; Pederson, J. R.; Pierce, R. B.; Prather, K.; Quinn, P. K.; Seinfeld, J. H.; Senff, C. J.; Sorooshian, A.; Stutz, J.; Surratt, J. D.; Trainer, M.; Volkamer, R.; Williams, E. J.; Wofsy, S. C. The 2010 California research at the nexus of air quality and climate change (CalNex) field study. J. Geophys. Res.: Atmos. 2013, 118 (11), 5830−5866, DOI: 10.1002/jgrd.50331. (24) Timko, M. T.; Yu, Z. H.; Kroll, J. H.; Jayne, J. T.; Worsnop, D. R.; Miake-Lye, R. C.; Onasch, T. B.; Liscinsky, D.; Kirchstetter, T. W.; Destaillats, H.; Holder, A. L.; Smith, J. D.; Wilson, K. R. Sampling artifacts from conductive silicone tubing. Aerosol Sci. Technol. 2009, 43 (9), 855−865. (25) Yu, Y.; Alexander, M. L.; Perraud, V.; Bruns, E. A.; Johnson, S. N.; Ezell, M. J.; Finlayson-Pitts, B. J. Contamination from electrically conductive silicone tubing during aerosol chemical analysis. Atmos. Environ. 2009, 43 (17), 2836−2839. (26) Zhao, J.; Eisele, F. L.; Titcombe, M.; Kuang, C.; McMurry, P. H. Chemical ionization mass spectrometric measurements of atmospheric neutral clusters using the cluster-CIMS. J. Geophys. Res.: Atmos. 2010, 115 (D8), No. D08205, DOI: 10.1029/2009JD012606. (27) Zhao, J.; Smith, J. N.; Eisele, F. L.; Chen, M.; Kuang, C.; McMurry, P. H. Observation of neutral sulfuric acid-amine containing clusters in laboratory and ambient measurements. Atmos. Chem. Phys. 2011, 11 (21), 10823−10836. (28) Atkinson, R. Kinetics of the gas phase reactions of a series of organosilicon compounds with OH and NO3 radicals and O3 at 297 ± 2 K. Environ. Sci. Technol. 1991, 25 (5), 863−866. (29) Chandramouli, B.; Kamens, R. M. The photochemical formation and gas-particle partitioning of oxidation products of decamethyl cyclopentasiloxane and decamethyl tetrasiloxane in the atmosphere. Atmos. Environ. 2001, 35 (1), 87−95. (30) Sommerlade, R.; Parlar, H.; Wrobel, D.; Kochs, P. Product analysis and kinetics of the gas-phase reactions of selected organosilicon compounds with OH radicals using a smog chamber-mass spectrometer system. Environ. Sci. Technol. 1993, 27 (12), 2435−2440. (31) Van Egmond, R. Personal Communication to C. O. Stanier, 16 October 2013. (32) Buser, A. M.; Bogdal, C.; MacLeod, M.; Scheringer, M. Emissions of decamethylcyclopentasiloxane from Chicago. Chemosphere 2014, 107 (1), 473−475. (33) McLachlan, M. S.; Kierkegaard, A.; Hansen, K. M.; van Egmond, R.; Christensen, J. H.; Skjoth, C. A. Concentrations and fate of decamethylcyclopentasiloxane (D5) in the atmosphere. Environ. Sci. Technol. 2010, 44 (14), 5365−5370. (34) Yucuis, R. A.; Stanier, C. O.; Hornbuckle, K. C. Cyclic siloxanes in air, including identification of high levels in Chicago and distinct diurnal variation. Chemosphere 2013, 92 (8), 905−910. (35) Reisch, M. S. Storm over silicones. Chem. Eng. News 2011, 89 (18), 10−13. (36) Wang, D. G.; Norwood, W.; Alaee, M.; Byer, J. D.; Brimble, S. Review of recent advances in research on the toxicity, detection, occurrence and fate of cyclic volatile methyl siloxanes in the environment. Chemosphere 2013, 93 (5), 711−725. (37) Warneke, C.; McKeen, S. A.; de Gouw, J. A.; Goldan, P. D.; Kuster, W. C.; Holloway, J. S.; Williams, E. J.; Lerner, B. M.; Parrish, D. D.; Trainer, M.; Fehsenfeld, F. C.; Kato, S.; Atlas, E. L.; Baker, A.; Blake, D. R. Determination of urban volatile organic compound emission ratios and comparison with an emissions database. J. Geophys. Res.: Atmos. 2007, 112 (D10), No. D10S47, DOI: 10.1029/ 2006jd007930. (38) Whelan, M. J.; Estrada, E.; van Egmond, R. A modelling assessment of the atmospheric fate of volatile methyl siloxanes and their reaction products. Chemosphere 2004, 57 (10), 1427−1437. (39) Balducci, C.; Perilli, M.; Romagnoli, P.; Cecinato, A. New developments on emerging organic pollutants in the atmosphere. Environ. Sci. Pollut. Res. 2012, 19 (6), 1875−1884.

(40) Graiver, D.; Farminer, K. W.; Narayan, R. A review of the fate and effects of silicones in the environment. J. Polym. Environ. 2003, 11 (4), 129−136. (41) Klems, J. P.; Pennington, M. R.; Zordan, C. A.; Johnston, M. V. Ultrafine particles near a roadway intersection: Origin and apportionment of fast changes in concentration. Environ. Sci. Technol. 2010, 44 (20), 7903−7907. (42) Bzdek, B. R.; Zordan, C. A.; Pennington, M. R.; Luther, G. W.; Johnston, M. V. Quantitative assessment of the sulfuric acid contribution to new particle growth. Environ. Sci. Technol. 2012, 46 (8), 4365−4373. (43) Buser, A. M.; Kierkegaard, A.; Bogdal, C.; MacLeod, M.; Scheringer, M.; Hungerbuhler, K. Concentrations in ambient air and emissions of cyclic volatile methylsiloxanes in Zurich, Switzerland. Environ. Sci. Technol. 2013, 47 (13), 7045−7051. (44) Companioni-Damas, E. Y.; Santos, F. J.; Galceran, M. T. Linear and cyclic methylsiloxanes in air by concurrent solvent recondensation-large volume injection-gas chromatography-mass spectrometry. Talanta 2014, 118 (1), 245−252. (45) Genualdi, S.; Harner, T.; Cheng, Y.; MacLeod, M.; Hansen, K. M.; van Egmond, R.; Shoeib, M.; Lee, S. C. Global distribution of linear and cyclic volatile methyl siloxanes in air. Environ. Sci. Technol. 2011, 45 (8), 3349−3354. (46) Kierkegaard, A.; McLachlan, M. S. Determination of linear and cyclic volatile methylsiloxanes in air at a regional background site in Sweden. Atmos. Environ. 2013, 80 (1), 322−329. (47) Krogseth, I. S.; Kierkegaard, A.; McLachlan, M. S.; Breivik, K.; Hansen, K. M.; Schlabach, M. Occurrence and seasonality of cyclic volatile methyl siloxanes in Arctic air. Environ. Sci. Technol. 2013, 47 (1), 502−509. (48) Wang, X. M.; Lee, S. C.; Sheng, G. Y.; Chan, L. Y.; Fu, J. M.; Li, X. D.; Min, Y. S.; Chan, C. Y. Cyclic organosilicon compounds in ambient air in Guangzhou, Macau and Nanhai, Pearl River Delta. Appl. Geochem. 2001, 16 (11−12), 1447−1454. (49) Latimer, H. K.; Kamens, R. M.; Chandra, G. The atmospheric partitioning of decamethylcyclopentasiloxane (D5) and 1-hydroxynonamethylcyclopentasiloxane (D4TOH) on different types of atmospheric particles. Chemosphere 1998, 36 (10), 2401−2414. (50) Navea, J. G.; Xu, S. H.; Stanier, C. O.; Young, M. A.; Grassian, V. H. Heterogeneous uptake of octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5) onto mineral dust aerosol under variable RH conditions. Atmos. Environ. 2009, 43 (26), 4060−4069. (51) Navea, J. G.; Xu, S. H.; Stanier, C. O.; Young, M. A.; Grassian, V. H. Effect of ozone and relative humidity on the heterogeneous uptake of octamethylcyclotetrasiloxane and decamethylcyclopentasiloxane on model mineral dust aerosol components. J. Phys. Chem. A 2009, 113 (25), 7030−7038. (52) Navea, J. G.; Young, M. A.; Xu, S. H.; Grassian, V. H.; Stanier, C. O. The atmospheric lifetimes and concentrations of cyclic methylsiloxanes octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5) and the influence of heterogeneous uptake. Atmos. Environ. 2011, 45 (18), 3181−3191. (53) MacLeod, M.; von Waldow, H.; Tay, P.; Armitage, J. M.; Wohrnschimmel, H.; Riley, W. J.; McKone, T. E.; Hungerbuhler, K. BETR globalA geographically-explicit global-scale multimedia contaminant fate model. Environ. Pollut. 2011, 159 (5), 1442−1445. (54) Wania, F. Assessing the potential of persistent organic chemicals for long-range transport and accumulation in polar regions. Environ. Sci. Technol. 2003, 37 (7), 1344−1351. (55) Xu, S. H.; Wania, F. Chemical fate, latitudinal distribution and long-range transport of cyclic volatile methylsiloxanes in the global environment: A modeling assessment. Chemosphere 2013, 93 (5), 835−843. (56) Buser, A. M.; Schenker, S.; Scheringer, M.; Hungerbuhler, K. Comparing the performance of computational estimation methods for physicochemical properties of dimethylsiloxanes and selected siloxanols. J. Chem. Eng. Data 2013, 58 (11), 3170−3178. (57) Hughes, L.; Mackay, D.; Powell, D. E.; Kim, J. An updated state of the science EQC model for evaluating chemical fate in the 11144



Article

environment: Application to D5 (decamethylcyclopentasiloxane). Chemosphere 2012, 87 (2), 118−124. (58) Kim, J.; Powell, D. E.; Hughes, L.; Mackay, D. Uncertainty analysis using a fugacity-based multimedia mass-balance model: Application of the updated EQC model to decamethylcyclopentasiloxane (D5). Chemosphere 2013, 93 (5), 819−829. (59) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, 2nd ed. ed.; Wiley: Hoboken, NJ, 2006. (60) Alvarez, E. G.; Amedro, D.; Afif, C.; Gligorovski, S.; Schoemacker, C.; Fittschen, C.; Doussin, J. F.; Wortham, H. Unexpectedly high indoor hydroxyl radical concentrations associated with nitrous acid. Proc. Natl. Acad. Sci. U.S.A. 2013, 110 (33), 13294− 13299. (61) Gligorovski, S.; Weschler, C. J. The oxidative capacity of indoor atmospheres. Environ. Sci. Technol. 2013, 47 (24), 13905−13906.

11145


Silicon is a Frequent Component of Atmospheric Nanoparticles

Recommend Documents