Aerosol Formation from OH Oxidation of the Volatile Cyclic Methyl

Subscriber access provided by University of Newcastle, Australia

Article

Aerosol Formation from OH Oxidation of the Volatile Cyclic Methyl Siloxane (cVMS) Decamethylcyclopentasiloxane Yue Wu, and Murray V Johnston Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b00655 • Publication Date (Web): 23 Mar 2017 Downloaded from http://pubs.acs.org on March 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

Environmental Science & Technology

1

Aerosol Formation from OH Oxidation of the Volatile Cyclic Methyl Siloxane (cVMS)

2

Decamethylcyclopentasiloxane

3

Yue Wu and Murray V. Johnston*

4

Department of Chemistry and Biochemistry

5

University of Delaware, Newark DE, USA 19716

6

*Corresponding author: Phone 302-831-8014, Fax 302-831-6336, Email [email protected]

7 8

Abstract Aerosol

formation

from

OH oxidation

of decamethylcyclopentasiloxane (D5,

9

C10H30O5Si5), a cyclic volatile methyl siloxane (cVMS) found in consumer products, was studied

10

in a flow-through photo chamber with and without the presence of ammonium sulfate seed

11

aerosol. For the unseeded experiments, chemical characterization with high performance mass

12

spectrometry showed that the molecular composition changed substantially with aerosol mass

13

loading in the 1-12 µg/m3 range.

14

atoms/molecule) dominated the mass spectra of aerosols at higher mass loadings while ring

15

opened species (neither 5 nor 10 Si atoms/molecule) dominated the mass spectra of aerosols at

16

lower mass loadings.

17

area ratio suggest that nonvolatile ring opened species are formed in the gas phase and assist

18

particle formation through condensation, while dimers are formed by accretion reactions within

19

the particle phase as the particles grow.

20

presence of seed aerosol with similar siloxane aerosol mass loading but higher volume to surface

21

area ratio, where ring-opened species are much less prevalent than monomers or dimers and the 1

Monomers (5 Si atoms/molecule) and dimers (10 Si

Molecular signal intensity dependencies on the aerosol volume-to-surface

These conclusions are supported by experiments in the

ACS Paragon Plus Environment


Page 2 of 26

22

aerosol yield is higher.

Because of the importance of accretion chemistry, the aerosol yield

23

from D5 oxidation is likely to be strongly dependent on particle size and morphology.

24 25

Introduction Cyclic volatile methylsiloxanes (cVMS) are widely used in personal care products.1,2

26 27

Owing to high vapor pressures, more than 90% of cVMS emissions are into atmosphere.3

28

phase cVMS have been detected in both indoor and outdoor environments,4-8 with generally

29

much

30

decamethylcyclopentasiloxane (D5) and dodecamethylcyclopentasiloxane (D6) at levels up to

31

650 and 79 ng/m3 respectively in ambient air in Zurich, Switzerland.4

32

sources of VOCs in a university classroom and reported a D5 concentration of 60±32µg/m3,

33

which represented >90% of all VOCs detected.8 Silicon has been reported in ambient

34

nanoparticles in a variety of urban locations, and gas phase oxidation of cVMS was suggested as

35

a possible source.9

higher

concentrations

indoors.6-8

For

example,

Buser

et

al.

Gas

reported

Tang et al. examined the

36

The major degradation pathway of cVMS in the atmosphere is oxidation by hydroxyl

37

radical (OH),10 where generation of semivolatile and nonvolatile products can lead to aerosol

38

formation.11,12 An atmospheric transport model with built-in OH and D5 chemistry was used to

39

predict D5 concentrations in air at a rural site in Sweden, with good agreement between measured

40

and predicted concentrations.13 Janechek et al. modified the Community Multiscale Air Quality

41

(CMAQ) model by incorporating cVMS and their gas phase oxidation products, and found that

42

oxidized D5 concentrations could be up to 9 ng/m3 downwind of major urban areas.14 2 ACS Paragon Plus Environment

Page 3 of 26


43

In the work reported here, the molecular composition of secondary aerosol obtained from

44

D5 oxidation is studied both in the presence and absence of ammonium sulfate seed aerosol using

45

high performance mass spectrometry. Systematic changes in the molecular composition with

46

aerosol mass loading are observed and interpreted on the basis of estimated volatilities of the

47

individual molecular products, which give fundamental insight into the pathways15 involved in

48

secondary aerosol formation.

49

monomers react in the particle phase to give a nonvolatile dimer, are shown to have a substantial

50

impact on aerosol formation as the aerosol mass loading increases.

Accretion reactions, specifically those where two semivolatile

51 52

Experimental Section

53

Secondary aerosol was generated in a Photo-oxidation chamber (PC) consisting of a 50L

54

(251 × 251 × 800 mm, WxHxL) rectangular bag composed of a perfluoroalkoxy copolymer

55

(Welch Fluorocarbon, Dover, New Hampshire). Before each experiment, the PC was cleaned by

56

exposure to 20 ppmv ozone in the presence of ultraviolet radiation for at least 48 hr and flushed

57

continuously with clean, dry, air for 4-5 days obtained from a Zero-Air generation system

58

(Model 737, Aadco, Cleves, Ohio). A Scanning Mobility Particle Sizer (SMPS; Model

59

3080/3078, TSI, St. Paul, Minnesota) was used to measure particle size distributions.

60

A syringe pump (Model No. 55-1199, Harvard Apparatus, Holliston, Massachusetts) was

61

used to feed precursor fluid (D5/MeOH as 1/5 by volume) (CAS No.541-02-6, Gelest,

62

Morrisville, Pennsylvania) at a rate of 0.1 µL/min into an airflow through a gently heated Teflon

63

fitting. This airflow was then diluted to adjust the initial concentration of gas phase D5 and a 3 ACS Paragon Plus Environment


Page 4 of 26

64

constant flow of the diluted air was sent into the PC. Ozone was generated by passing clean air

65

around a mercury lamp (Model No.81-1025-01, BHK Inc., Ontario, California).

66

laden air was then bubbled through deionized water and sent into the PC where interaction with

67

ultraviolet radiation produced OH. Using the method described by Hall and Johnston,16 the OH

68

mixing ratio for these experiments was estimated to be about 108 molecules/cm3.

69

combined and sent into the PC, the ozone, D5 and a third make-up air flow yielded a nominal

70

residence time of 15 min in the PC.

71

ammonium sulfate solution. The particles were then sent through a diffusion drier and entered

72

the PC through a portion of the makeup air flow. After mixing, the seed aerosol mass

73

concentration in the PC was 1.9 µg/m3. All experiments were performed with low relative

74

humidity (typically 8-10%) and a temperature of 27 oC.

75

that were performed, five each seeded and unseeded. Figure S1 shows representative particle

76

size distributions.

The ozone

When

Seed particles were generated by atomizing a 5mM

Table S1 summarizes the experiments

77

Particulate matter in the air flow exiting the PC was collected at a flow rate of ~2 L/min

78

for 20-24 hrs onto a Teflon coated, glass fiber filter (GF/D, CAT No.1823-025, Whatman, GE

79

Healthcare, Piscataway, NJ) for chemical analysis. Generally, ~0.2 mg of aerosol was collected

80

in each experiment for analysis and four separate filter collections were obtained for each

81

experiment in Table S1.

82

to identify and remove artifacts. After particle collection, each filter was sonicated for three

83

hours with 8 ml acetonitrile (ACN)/deionzied H2O 50/50 by volume (CAS No. 75-05-8, Fisher

84

Scientific, Pittsburgh, Pennsylvania) and the resulting solution was filtered through a 4

Background samples were collected prior to the injections of reactants


Page 5 of 26


85

polyvinylidenedifluoride (PVDF) filter (CAT No. 6747-2504, GE Healthcare, Piscataway, NJ).

86

The filtered solution was then concentrated nearly to dryness ( 60 µg/m3), where accurate mass measurements

104

and MS/MS spectra were used to characterize the molecular products of D5 oxidation.12

105

Although the mass spectra are complex, molecular products can be divided into three main types. 5 ACS Paragon Plus Environment


106

First are monomer products, defined here as those containing exactly 5 Si atoms, at least 5 O

107

atoms, and a molecular formula (and MS/MS spectra) consistent with one siloxane ring but no

108

other sites of unsaturation.

109

various combinations of functionalization where CH3 groups are replaced by OH and/or CH2OH

110

groups.

Relative to the D5 precursor molecule, monomer products contain

In Figure 1, monomer products are coded red.

111

Second are dimer products, defined here as those containing exactly 10 Si atoms, at least

112

10 O atoms, and a molecular formula (and MS/MS spectra) consistent with two siloxane rings

113

but no other sites of unsaturation.

114

functionalization where CH3 groups are replaced by OH and/or CH2OH, and the two siloxane

115

rings can be linked by O, CH2 or CH2CH2 groups.

Dimers also contain various combinations of

In Figure 1, dimer products are coded black.

Third are ring-opened products, defined here as those containing neither 5 nor 10 Si

116 117

atoms.

118

siloxane ring structure.

119

consistent with CH3 replacement by OH and/or CH2OH, and sometimes indicate an additional

120

site of unsaturation, for example, Si=O.

121

Formation of these products necessarily requires fragmentation of the original D5 Molecular formulas and MS/MS spectra of these products are

In Figure 1, ring-opened products are coded blue.

Scheme 1 shows candidate structures for three prominent ions of each type, which are

122

starred in the mass spectra of Figure 1.

While not proven, candidate structures are consistent

123

with molecular formulas and MS/MS spectra. Figure S2 shows MS/MS spectra for the three

124

starred ions in Figure 1. As discussed in our previous work,12 accurate mass and MS/MS spectra

125

often permit identification of specific functional groups and dimer linkages, but they generally

126

cannot determine functional group locations. 6 ACS Paragon Plus Environment

Page 6 of 26

Page 7 of 26


127

In our previous study with a high aerosol mass loading (> 60 µg/m3),12 ring-opened

128

products had very low signal intensities that could be explained in part by molecular

129

decomposition within the ESI source.

130

only briefly (referred to as fragmentation products in ref. 12). In contrast, the ring-opened

131

products in Figure 1 have very high signal intensities that cannot be explained simply by ESI

132

decomposition.

133

(dissolved in ACN/H2O 50/50 by volume) for 3 hours under the same experimental condition.

134

Figure S3 shows that the ESI mass spectrum obtained after sonication was generally devoid of

135

ring-opened species, indicating that the high abundance of these products in the filter samples

136

was not from sonochemistry.

137

in plasma polymerization experiments, 17 so it is not unreasonable to expect they can be formed

138

by OH oxidation.

Therefore, those ring-opened products were discussed

Additionally, a control experiment was performed by sonicating 1 mg/ml D5

We note that ring-opened products have been reported previously

139

The mass loading dependence of the three product types are explored further in Figure 2a

140

where the ratio of the summed signal intensities of dimers to ring-opened products is plotted as a

141

function of aerosol volume to surface area ratio. As shown in Figure S4, the volume to surface

142

area ratio of the unseeded aerosols studied in this work scale approximately linearly with mass

143

loading.

144

Figure 2a shows that the ratio of dimers to ring-opened products increases approximately

145

linearly with increasing volume to surface area ratio.

As will be discussed later, all dimers and

146

most ring-opened products are expected to be extremely nonvolatile with predicted saturation

147

concentrations (C*) that are orders of magnitude lower than the aerosol mass loadings in these 7 ACS Paragon Plus Environment


148

experiments.

149

mechanisms for these two types of products.

150

particle formation and/or early growth since they are most prevalent at low volume to surface

151

area ratio.

152

when they strike the particle surface.

153

directly at the particle surface.

154

determined by the available surface area.

155

The approximately linear relationship in Figure 2a suggests different formation Ring-opened products appear to be necessary for

These species are likely to be formed directly in the gas phase and then condense It is also possible that ring-opened products are formed

In either case, particle growth via ring-opened species is

Our earlier work, performed with a high aerosol mass loading, speculated that dimers

156

were formed in the gas phase.12

157

study shows that dimers must be formed by accretion reactions in the particle phase, since they

158

are nonvolatile but strongly represented only in higher mass loadings where the aerosol volume

159

is larger.

160

previously. 18

However, the mass loading dependence observed in the current

The possibility of condensed phase formation of oligomers has been reported

161

A linear dependence on volume to surface area ratio would be expected for a

162

volume-limited process such as accretion chemistry relative to a surface-limited process such as

163

condensation.

164

group for accretion reaction products in biogenic secondary organic aerosol (SOA) as a function

165

of particle size in the 35-110 nm range.19

166

particle of size d is 2d/3, making the range of volume to surface area ratios where accretion

167

chemistry becomes important similar for the siloxane secondary aerosol in Figure 2a and the

168

biogenic SOA studied previously.19

The linear relationship in Figure 2a is similar to that recently reported by our

Note that the volume to surface area ratio for a

8 ACS Paragon Plus Environment

Page 8 of 26

Page 9 of 26

169


Figure 2b shows ratio of monomer to dimer mass-weighted signal intensities.

In

170

contrast to ring-opened products, the monomer to dimer ratio is relatively invariant with volume

171

to surface area ratio as would be expected for a quasi-equilibrium process where monomers react

172

to form dimers. Note that monomer partitioning between the gas and particle phases like

173

accretion chemistry is a volume-limited process.

174

the lowest volume to surface area ratio arises from a slight change in the distribution of

175

monomers – compare the mass spectrum in Figure 1a to those in 1b and 1c. As a group, the

176

monomers in Figure 1a are less volatile than those in Figures 1b and 1c based on calculated

177

saturation concentrations of candidate molecular structures (see below).

The slight increase in monomers to dimers at

178 179 180

Aerosol Formation in Seeded Experiments Secondary aerosol formation in the presence of ammonium sulfate seed was found to

181

give different product distributions than those observed in the unseeded experiments.

182

compares the mass spectra of secondary aerosol from seeded and unseeded experiments where

183

the aerosol mass loadings after subtraction of the seed are the same.

184

between the two is the very low contribution of ring-opened products to the seeded spectrum.

185

In fact, the mass spectra obtained from seeded experiments show no significant change with

186

aerosol mass loading.

187

Figure 3

The main difference

These results are consistent with the above conclusions for unseeded aerosol.

In seeded

188

experiments, formation and/or growth of new particles is less important since secondary aerosol

189

is able to grow on the preexisting seeds.

The volume to surface area ratio of the seed aerosol 9



Page 10 of 26

190

was 18 nm, which then increased to as high as 32 nm depending on the amount of secondary

191

aerosol produced.

192

represent only a minor contribution to aerosol mass at such high values of volume to surface area

193

ratio.

194

the presence of seed particles), secondary aerosol formation proceeds mainly by a combination

195

of monomer partitioning and particle phase accretion reactions to produce dimers.

Based on the plot in Figure 2a, ring-opened species would be expected to

Once particle formation and early growth has been accomplished (or replaced through

196 197

Product Volatility Estimation

198

The difference in composition between the seeded and unseeded aerosols in Figure 3 is

199

further examined on the basis of molecular volatility as given by the saturation concentration

200

(C*). Measuring or estimating values of C* for individual molecular components is important

201

since in combination with the aerosol mass loading it determines the fraction of that component

202

that exists in the gas vs. particle phase at equilibrium.20, 21 Several volatility estimation methods

203

have been reported in the literature with varying degrees of complexity.22-24 The method

204

introduced by Nannoolal et al. is most relevant to the current study since it considers group

205

contributions involving silicon.24, 25 Beginning with a candidate molecular structure, the method

206

calculates a normal boiling point based on group contributions, a group interaction term and

207

finally the vapor pressure (saturation concentration).

208

We tested this approach by comparing measured and calculated boiling points and

209

saturation concentrations for cyclic siloxanes D3 to D7.

The results, shown in Table S2,

210

demonstrate agreement between measured and calculated C* within a factor of 5 or less. 10 ACS Paragon Plus Environment

To

Page 11 of 26


211

further test the method, we took the molecular formula for the ring-opened product in Scheme 1

212

and calculated C* values for eight candidate structures including the one shown in Scheme 1.

213

The results, shown in Table S3, give predicted C* values within a factor of 2-4 of each other.

214

The relatively close agreement for different candidate structures arises in large part from similar

215

Si-O scaffolding and only minor perturbation due to type and location of functional groups.

216

Applying this method to candidate structures for all molecular products leads values spanning

217

many orders of magnitude in C*, whereas the uncertainty associated with any individual

218

prediction is likely to be much less than one order of magnitude.

219

formulas and predicted C* values is given in Table S4.

220

molecular formula vs. log C* is shown in Figure S5. All dimers and most ring-opened products

221

are relatively nonvolatile, having predicted log C* values below -2.

222

semivolatile or volatile, though a few can be considered nonvolatile.

223

A complete list of molecular

A summary plot of O/Si ratio of the

Most monomers are

Figure 4 summarizes the mass spectra in Figure 3 using a volatility basis set

224

representation,21,

26

225

fraction) vs. log C*.

226

mass fraction.

227

the signal intensity of the species multiplied by its molecular mass, then divided by the

228

summation of mass-weighted intensities over all species:

which normally involves a plot of aerosol mass concentration (or mass In Figure 4, the mass-weighted signal intensity is used as a surrogate for

The mass-weighted signal intensity (SImw) of a molecular species is defined as

, =

∑

229

SImw is an approximate measure of the mass fraction of a species, but it is not precise because

230

different species can have different ESI detection efficiencies. 11 ACS Paragon Plus Environment

Indeed, we find for the mass


Page 12 of 26

231

spectrometer used in this study that ESI detection efficiencies for different cyclic siloxanes can

232

vary by up to a factor of 3.

233

ESI has been shown to be able to accurately determine average molecular weights for polymeric

234

samples.27, 28

235

A smaller variation is likely for linear polydimethyl siloxanes since

Figure 4 provides insight into the contributions of different volatility species to seeded vs.

236

unseeded aerosol.

Figure 4a shows that the majority of nonvolatile matter (log C* < -2) in

237

unseeded aerosol consists of ring-opened products with a minor contribution from dimers, while

238

the opposite is true for seeded aerosol in Figure 4b.

239

ring-opened species are needed for particle formation and/or early growth in an unseeded

240

experiment but not in the presence of seeds.

241

is higher in the seeded experiment, which suggests that dimer formation in the particle phase is at

242

least partially reversible.

243

decomposition would have occurred during the sample preparation process or in the reactor itself

244

when the aerosol was formed.

245

then the higher fraction of intermediate volatility monomers in Figure 4b would suggest that they

246

can be trapped within the particle phase owing to hindered diffusion29-31 and/or phase

247

separation.32-35

This observation is not surprising since

The fraction of intermediate volatility monomers

However, it is not clear from the experiments performed whether such

If reversibility occurred on the timescale of aerosol formation,

248 249 250 251

Aerosol Yield and Atmospheric Implications Adding seed particles had the effect of significantly increasing the aerosol yield.

For

example, the unseeded experiment in Figure 1a had identical reaction conditions to the seeded 12 ACS Paragon Plus Environment

Page 13 of 26


252

experiment in Figure 3b, yet the secondary aerosol mass concentration was a factor of 2 greater

253

for the seeded experiment.

254

reflection on the kinetics of aerosol formation.

255

was slow since it relied on formation of ring-opened species that represented a small fraction of

256

the D5 oxidation product yield.

257

significant amount of time needed to reach a volume to surface area ratio that became favorable

258

for dimer formation.

259

level much faster in the presence of seed particles, enhancing the formation of dimers through

260

accretion chemistry and hence the amount of secondary aerosol produced within the residence

261

time of the reactor.

262

smaller impact due to vapor wall loss than in the unseeded experiments.

The higher aerosol yield in the seeded experiment is likely a In the unseeded experiment, particle formation

The consequence of slow particle formation would be a

In contrast, the aerosol volume to surface area ratio reached a favorable

The higher aerosol yields of the seeded experiments may also reflect a

263

Overall, we found the aerosol yield to decrease with decreasing reacted D5 mass

264

concentration, reaching values of 0.15 and 0.08 respectively for seeded and unseeded

265

experiments at the lowest secondary aerosol mass concentrations studied.

266

values for aerosol yields are not based on direct measurement of the change in D5 concentration,

267

but calculated from the OH mixing ratio which itself was estimated from a different set of

268

measurements whose error was difficult to assess.

We caution that these

269

This work shows that condensation, partitioning and particle phase chemistry (in this case

270

accretion reactions) all contribute to secondary aerosol formation by gas phase D5 oxidation.

271

Condensation appears to be driven mainly by the formation of ring-opened products in the gas

272

phase, though a few monomer oxidation products are also nonvolatile. 13 ACS Paragon Plus Environment

Partitioning is


273

dominated by monomer oxidation products, most of which are semivolatile.

274

nonvolatile but formed mainly by accretion reactions within the particle phase.

Page 14 of 26

Dimers are

275

Future experiments should focus on accurate and precise measurements of aerosol yield.

276

Since aerosol formation from D5 oxidation is strongly influenced by accretion reactions, work is

277

also needed to study this chemistry under a wider range of experimental conditions and particle

278

phase morphologies that might be encountered in ambient aerosol.

279 280

Supporting Information

281

Four tables (S1-S4) and five figures (S1-S5).

282 283 284

Acknowledgement This research was supported by the National Science Foundation under grant number

285

CHE-1408455.

The Orbitrap mass spectrometer used in this study was purchased under grant

286

number S10 OD016267-01 and supported by grant number 1 P30 GM110758-01, both from the

287

National Institutes of Health.

288 289


Page 15 of 26


290

References

291

1. Horii, Y.; Kannan, K., Survey of organosilicone compounds, including cyclic and linear

292

siloxanes, in personal-care and household products. Arch. Environ. Con. Tox. 2008, 55, (4),

293

701-710.

294

2. Wang, R.; Moody, R. P.; Koniecki, D.; Zhu, J. P., Low molecular weight cyclic volatile

295

methylsiloxanes in cosmetic products sold in Canada: Implication for dermal exposure.

296

Environ. Int. 2009, 35, (6), 900-904.

297

3. Genualdi, S.; Harner, T.; Cheng, Y.; MacLeod, M.; Hansen, K. M.; van Egmond, R.; Shoeib,

298

M.; Lee, S. C., Global Distribution of Linear and Cyclic Volatile Methyl Siloxanes in Air.

299

Environ. Sci. Technol. 2011, 45, (8), 3349-3354.

300

4. Buser, A. M.; Kierkegaard, A.; Bogdal, C.; MacLeod, M.; Scheringer, M.; Hungerbuhler, K.,

301

Concentrations in Ambient Air and Emissions of Cyclic Volatile Methylsiloxanes in Zurich,

302

Switzerland. Environ. Sci. Technol. 2013, 47, (13), 7045-7051.

303

5. Krogseth, I. S.; Kierkegaard, A.; McLachlan, M. S.; Breivik, K.; Hansen, K. M.; Schlabach,

304

M., Occurrence and Seasonality of Cyclic Volatile Methyl Siloxanes in Arctic Air. Environ.

305

Sci. Technol. 2013, 47, (1), 502-509.

306

6. Zhu, J. P.; Wong, S. L.; Cakmak, S., Nationally Representative Levels of Selected Volatile

307

Organic Compounds in Canadian Residential Indoor Air: Population-Based Survey. Environ.

308

Sci. Technol. 2013, 47, (23), 13276-13283.



Page 16 of 26

309

7. Shields, H. C.; Fleischer, D. M.; Weschler, C. J., Comparisons among VOCs measured in

310

three types of US commercial buildings with different occupant densities. Indoor Air 1996, 6,

311

(1), 2-17.

312

8. Tang, X. C.; Misztal, P. K.; Nazaroff, W. W.; Goldstein, A. H., Siloxanes Are the Most

313

Abundant Volatile Organic Compound Emitted from Engineering Students in a Classroom.

314

Environ. Sci. Technol. Lett. 2015, 2, (11), 303-307.

315

9. Bzdek, B. R.; Horan, A. J.; Pennington, M. R.; Janechek, N. J.; Baek, J.; Stanier, C. O.;

316

Johnston, M. V., Silicon is a Frequent Component of Atmospheric Nanoparticles. Environ.

317

Sci. Technol. 2014, 48, (19), 11137-11145.

318

10. Atkinson, R., Kinetics of the Gas-Phase Reactions of a Series of Organosilicon Compounds

319

with OH and NO3 Radicals and O3 at 297 ± 2 K. Environ. Sci. Technol. 1991, 25, (5),

320

863-866.

321

11. Sommerlade, R.; Parlar, H.; Wrobel, D.; Kochs, P., Product Analysis and Kinetics of the

322

Gas-Phase Reactions of Selected Organosilicon Compounds with OH Radicals Using a Smog

323

Chamber - Mass-Spectrometer System. Environ. Sci. Technol. 1993, 27, (12), 2435-2440.

324

12. Wu, Y.; Johnston, M. V., Molecular Characterization of Secondary Aerosol from Oxidation

325

of Cyclic Methylsiloxanes. J. Amer. Soc. Mass Spectrom. 2016, 27, (3), 402-409.

326

13. McLachlan, M. S.; Kierkegaard, A.; Hansen, K. M.; van Egmond, R.; Christensen, J. H.;

327

Skjoth, C. A., Concentrations and Fate of Decamethylcyclopentasiloxane (D5) in the

328

Atmosphere. Environ. Sci. Technol. 2010, 44, (14), 5365-5370.


Page 17 of 26


329

14. Janechek, J.N.; Hansen, M.K.; Stanier, O.C., Comprehensive Atmopheric Modelling of

330

Reactive Cyclic Siloxanes and Their Oxidation Products. Atmos. Chem. Phys. Discuss. 2017,

331

doi:10.5194/acp-2016-1159, in review.

332

15. Kroll, J. H.; Seinfeld, J. H., Chemistry of secondary organic aerosol: Formation and

333

evolution of low-volatility organics in the atmosphere. Atmos Environ 2008, 42, (16),

334

3593-3624.

335

16. Hall, W. A.; Pennington, M. R.; Johnston, M. V., Molecular Transformations Accompanying

336

the Aging of Laboratory Secondary Organic Aerosol. Environ. Sci. Technol. 2013, 47, (5),

337

2230-2237.

338

17. Fouquet, T.; Petersen, J.; Bomfim, J. A. S.; Bour, J.; Ziarelli, F.; Ruch, D.; Charles, L.,

339

Electrospray tandem mass spectrometry combined with authentic compound synthesis for

340

structural characterization of an octamethylcyclotetrasiloxane plasma polymer. Int J Mass

341

Spectrom. 2012, 313, 58-67.

342 343

18. Cypryk, M.; Apeloig, Y., Mechanism of the acid-catalyzed Si-O bond cleavage in siloxanes and siloxanols. A theoretical study. Organometallics. 2002, 21, (11), 2165-2175.

344

19. Tu, P.; Johnston, M. V., Particle Size Dependence of Biogenic Secondary Organic Aerosol

345

Molecular Composition, Atmos. Chem. Phys. Discuss. 2017, doi:10.5194/acp-2017-53, in

346

review.

347 348

20. Donahue, N. M.; Robinson, A. L.; Trump, E. R.; Riipinen, I.; Kroll, J. H., Volatility and Aging of Atmospheric Organic Aerosol. Top. Curr. Chem. 2014, 339, 97-143.



Page 18 of 26

349

21. Donahue, N. M.; Robinson, A. L.; Stanier, C. O.; Pandis, S. N., Coupled partitioning,

350

dilution, and chemical aging of semivolatile organics. Environ. Sci. Technol. 2006, 40, (8),

351

2635-2643.

352

22. Donahue, N. M.; Epstein, S. A.; Pandis, S. N.; Robinson, A. L., A two-dimensional volatility

353

basis set: 1. organic-aerosol mixing thermodynamics. Atmos. Chem. Phys. 2011, 11, (7),

354

3303-3318.

355

23. Donahue, N. M.; Kroll, J. H.; Pandis, S. N.; Robinson, A. L., A two-dimensional volatility

356

basis set - Part 2: Diagnostics of organic-aerosol evolution. Atmos. Chem. Phys. 2012, 12, (2),

357

615-634.

358

24. Nannoolal, Y.; Rarey, J.; Ramjugernath, D., Estimation of pure component properties. Part 2.

359

Estimation of critical property data by group contribution. Fluid Phase Equilibr. 2007, 252,

360

(1-2), 1-27.

361

25. Nannoolal, Y.; Rarey, J.; Ramjugernath, D.; Cordes, W., Estimation of pure component

362

properties Part 1. Estimation of the normal boiling point of non-electrolyte organic

363

compounds via group contributions and group interactions. Fluid Phase Equilibr. 2004, 226,

364

45-63.

365

26. Robinson, A. L.; Donahue, N. M.; Shrivastava, M. K.; Weitkamp, E. A.; Sage, A. M.;

366

Grieshop, A. P.; Lane, T. E.; Pierce, J. R.; Pandis, S. N., Rethinking organic aerosols:

367

Semivolatile emissions and photochemical aging. Science 2007, 315, (5816), 1259-1262.


Page 19 of 26


368

27. Yan, W. Y.; Gardella, J. A.; Wood, T. D., Quantitative analysis of technical polymer

369

mixtures by matrix assisted laser desorption/ionization time of flight mass spectrometry. J.

370

Amer. Soc. Mass Spectrom. 2002, 13, (8), 914-920.

371

28. Liu, X. M.; Maziarz, E. P.; Heiler, D. J.; Grobe, G. L., Comparative studies of poly(dimethyl

372

siloxanes) using automated GPC-MALDI-TOF MS and on-line GPC-ESI-TOF MS. J. Amer.

373

Soc. Mass Spectrom. 2003, 14, (3), 195-202.

374

29. Faulhaber, A. E.; Thomas, B. M.; Jimenez, J. L.; Jayne, J. T.; Worsnop, D. R.; Ziemann, P. J.,

375

Characterization of a thermodenuder-particle beam mass spectrometer system for the study

376

of organic aerosol volatility and composition. Atmos. Meas. Technol. 2009, 2, (1), 15-31.

377

30. Grayson, J. W.; Zhang, Y.; Mutzel, A.; Renbaum-Wolff, L.; Boge, O.; Kamal, S.; Herrmann,

378

H.; Martin, S. T.; Bertram, A. K., Effect of varying experimental conditions on the viscosity

379

of alpha-pinene derived secondary organic material. Atmos. Chem. Phys. 2016, 16, (10),

380

6027-6040.

381

31. Renbaum-Wolff, L.; Grayson, J. W.; Bateman, A. P.; Kuwata, M.; Sellier, M.; Murray, B. J.;

382

Shilling, J. E.; Martin, S. T.; Bertram, A. K., Viscosity of alpha-pinene secondary organic

383

material and implications for particle growth and reactivity. Proc. Nat. Acad. Sci. 2013, 110,

384

(20), 8014-8019.

385

32. Laskina, O.; Morris, H. S.; Grandquist, J. R.; Qiu, Z.; Stone, E. A.; Tivanski, A. V.; Grassian,

386

V. H., Size Matters in the Water Uptake and Hygroscopic Growth of Atmospherically

387

Relevant Multicomponent Aerosol Particles. J. Phys. Chem. A 2015, 119, (19), 4489-4497.



388 389

Page 20 of 26

33. Veghte, D. P.; Altaf, M. B.; Freedman, M. A., Size Dependence of the Structure of Organic Aerosol. J. Amer. Chem. Soc. 2013, 135, (43), 16046-16049.

390

34. Virtanen, A.; Kannosto, J.; Kuuluvainen, H.; Arffman, A.; Joutsensaari, J.; Saukko, E.; Hao,

391

L.; Yli-Pirila, P.; Tiitta, P.; Holopainen, J. K.; Keskinen, J.; Worsnop, D. R.; Smith, J. N.;

392

Laaksonen, A., Bounce behavior of freshly nucleated biogenic secondary organic aerosol

393

particles. Atmos. Chem. Phys. 2011, 11, (16), 8759-8766.

394

35. Werner, J.; Dalirian, M.; Walz, M. M.; Ekholm, V.; Wideqvist, U.; Lowe, S. J.; Ohrwal, G.;

395

Persson, I.; Riipinen, I.; Bjorneholm, O., Surface Partitioning in Organic-Inorganic Mixtures

396

Contributes to the Size-Dependence of the Phase-State of Atmospheric Nanoparticles.

397

Environ. Sci. Technol. 2016, 50, (14), 7434-7442.


Page 21 of 26


3

Relative Intenisty (%)

a) 1.2 µg/m

3

b) 5.6 µg/m

3

c) 12 µg/m

200

400

398

600 m/z (+)

800

1000

399

Figure 1. ESI mass spectra in positive ion mode of secondary aerosol from D5 oxidation under

400

various aerosol mass loadings.

401

measurements from four different samples. Peaks considered as ring opened products are coded

402

blue, dimers are coded black, and monomers are coded red. Candidate structures for starred ions

403

are given in Scheme 1.

Ion signal intensities are averaged over four separate

404



Page 22 of 26

.

Monomer ( )

Ring opened product ( )

Dimer ( )

405 406

Scheme 1. Candidate structures for the three types of reaction products found in siloxane

407

secondary aerosol. These structures are consistent with molecular formulas and MS/MS spectra

408

of the starred ions in Figure 1.

409 410 411



SI Ratio of Dimers/Ring-opened

Page 23 of 26

0.6

a)

0.4

0.2

0.0 10

20

30

40

SI Ratio of Monomers/Dimers

Volume/Surface Area (nm) 5

b)

4 3 2 1 0 10

20

30

40

Volume/Surface Area (nm)

412 413

Figure 2. Signal intensity ratio of a) dimers to ring-opened products and b) monomers to dimers

414

vs. volume/surface area ratio of the secondary aerosol.

415

deviation.

Error bars represent one standard

Error bars smaller than the symbols are not shown.



Page 24 of 26

Relative Intenisty (%)

a) Unseeded

b) Seeded

200

400

600 m/z (+)

800

1000

416 417

Figure 3. ESI mass spectra in positive ion mode for a) unseeded and b) seeded aerosol produced

418

by OH oxidation of D5 under conditions that give similar mass loadings (after subtraction of seed

419

aerosol) of ~3 µg/m3.

420

four different samples.

421

coded black, and monomers are coded red.

Ion signal intensities are averaged over four separate measurements from Peaks considered as ring opened products are coded blue, dimers are

422



Mass-weighted Signal Intensities (%)

Mass-weighted Signal Intensities (%)

Page 25 of 26

423

0.8

NVOC

LVOC

SVOC

IVOC

(a) Unseeded

0.6 0.4 0.2 0.0 2

3

logC (saturation conc., µg/m ) 0.8

NVOC

LVOC

SVOC

IVOC

(b) Seeded

0.6 0.4 0.2 0.0 2

3

logC (saturation conc., µg/m )

424

Figure 4. Mass-weighted signal intensities summed over the three types of ions in a) unseeded

425

and b) seeded aerosols vs. log C* for candidate structures that correspond to individual ions.

426

These plots are for assigned ions in the mass spectra of Figures 3a and 3b, respectively.

427

Ring-opened products are coded blue, dimers are coded black, and monomers are coded red. 25 ACS Paragon Plus Environment


TOC Graphic

428

200

400

600

800

1000

m/z (+)

NVOC

LVOC

SVOC

IVOC

Ring-opened Dimers Monomers

2

logC (µ µg/m )

430


Page 26 of 26

Aerosol Formation from OH Oxidation of the Volatile Cyclic Methyl

Recommend Documents