Experimental Study of the Formation of Organosulfates from α-Pinene

Sep 9, 2016 - In the present study, quasi-static reactor and atmospheric simulation chamber experiments were performed to investigate the formation of...
0 downloads 0 Views 2MB Size
Subscriber access provided by SUNY DOWNSTATE

Article

Experimental Study of the Formation of Organosulfates from #-Pinene Oxidation. Part I: Product Identification, Formation Mechanisms and Effect of Relative Humidity Geoffroy Duporté, Pierre-Marie Flaud, Emmanuel Geneste, Sylvie Augagneur, Edouard Pangui, Housni Lamkaddam, Aline Gratien, Jean-Francois Doussin, Hélène Budzinski, Eric Villenave, and Emilie Perraudin J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.6b08504 • Publication Date (Web): 09 Sep 2016 Downloaded from http://pubs.acs.org on September 11, 2016

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.

The Journal of Physical Chemistry A 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 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Experimental Study of the Formation of Organosulfates from αPinene Oxidation. Part I: Product Identification, Formation Mechanisms and Effect of Relative Humidity

1 2 3 4 5 6

G. Duportéa, b, P.-M. Flauda, b, E. Genestea, b, S. Augagneura, b, E. Panguic, H. Lamkaddamc, A. Gratienc, J.-F. Doussinc, H. Budzinskia, b, E. Villenavea, b and E. Perraudina, b* a

7

b

8 9 10

Univ. Bordeaux, EPOC, UMR 5805, F-33405 Talence Cedex, France CNRS, EPOC, UMR 5805, 33405 Talence Cedex, France

c

Univ. Paris-Est-Créteil (UPEC) and Univ. Paris Diderot (UPD), LISA, UMR 7583, F-94010 Créteil, France

11 12

Abstract

13

In the present study, quasi-static reactor and atmospheric simulation chamber experiments were

14

performed to investigate the formation of α-pinene-derived organosulfates. Organosulfates (R-

15

OSO3H) were examined for the reactions between acidified ammonium sulfate particles exposed to

16

an individual gaseous volatile organic compound, such as α-pinene and oxidised products (α-pinene

17

oxide, isopinocampheol, pinanediol and myrtenal). Molecular structures were elucidated by liquid

18

chromatography interfaced to high-resolution quadrupole time-of-flight mass spectrometry

19

equipped with electrospray ionization (LC/ESI-HR-QTOFMS). New organosulfate products were

20

detected and identified for the first time in the present study. Reaction with α-pinene oxide was

21

found to be a favoured pathway for organosulfate formation (C10H18O5S) and to yield organosulfate

22

dimers (C20H34O6S and C20H34O9S2) and trimers (C30H50O10S2) under dry conditions (RH < 1 %) and high

23

particle acidity and precursor concentrations (1 ppm). The role of relative humidity on organosulfate

24

formation yields and product distribution was specifically examined. Organosulfate concentrations

25

were found to decrease with increasing relative humidity. Mechanistic pathways for organosulfate

26

formation from the reactions between α-pinene, α-pinene oxide, isopinocampheol or pinanediol

27

with acidified ammonium sulfate particles are proposed.

28 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

29

Introduction

30

Secondary organic aerosols (SOAs) are formed through atmospheric oxidation and

31

processing of volatile organic compounds (VOCs).1 It is now well recognized that SOA may

32

contribute to a significant fraction of the fine organic particulate matter, which is known to

33

play an important role in regional air quality and have adverse health impacts. 2 In addition,

34

they can also directly and indirectly affect regional and global climate.3-4 Despite continuing

35

improvements, knowledge about SOA precursors, and formation and evolution mechanisms

36

are still urgently needed for a better understanding and modelling of aerosol impacts. In

37

particular, to better evaluate these impacts, it is crucial to gain further insight into their

38

chemical composition, as well as into chemical processes describing SOA formation and

39

ageing in the atmosphere.

40

Organosulfates (R-OSO3H) are part of the highly complex organic fraction of atmospheric

41

aerosol. They have been measured and recently identified in ambient aerosols, rain and fog

42

water in Europe, America and Asia5-25 and were found to be of secondary origin.8, 26 These

43

studies showed that organosulfates may represent an important fraction of ambient organic

44

aerosols, estimated to contribute up to 30 % to PM 10 organic mass.26-27 Their highly

45

oxygenated and sulfated chemical structure suggests that their presence may modify aerosol

46

hygroscopic properties and have important climate impacts.18 Furthermore, organosulfates

47

may serve as good marker compounds for biogenic secondary organic aerosol (BSOA) that is

48

enhanced by the co-presence of anthropogenic pollution.28 As already demonstrated in

49

previous studies, α-pinene and more generally monoterpenes, are important precursors of

50

organosulfates in the atmosphere.6-7, 13, 19-20, 22-23, 26, 29-31 Moreover, organosulfates have been

51

shown to be generated in laboratory experiments from the oxidation of biogenic VOCs

2 ACS Paragon Plus Environment

Page 2 of 49

Page 3 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

52

(BVOCs)6-8, 26 and the reactive uptake of pinonaldehyde32, α-pinene oxide29-30, 33-34, glyoxal35-

53

36

54

(MBO)-derived epoxide40-41 in the presence of acidic sulfate particles. Although some

55

organosulfate formation channels have been proposed in these studies, such as

56

esterification of hydroxyl or keto groups8, 32, acid-catalyzed ring-opening of epoxides29, 42-44,

57

radical-initiated processes in wet aerosols36,

58

organonitrates47-48, the formation processes of monoterpene-derived organosulfates in the

59

atmosphere remain elusive.

60

SOA generated in simulation chambers from selected BVOCs have been shown to be

61

complex and dynamic mixtures.49-51 Characterizing their chemical composition at the

62

molecular level is therefore a challenging task. Unravelling the underlying multiphasic

63

chemistry from the time evolution of such a complex reaction mixture is even more difficult,

64

due to the number of compounds produced and possibly reacting during experiments. In this

65

context, organosulfate formation was investigated by studying the reactivity of individual

66

compounds in the presence of sulfate-containing particles. Four α-pinene oxidation products

67

or proxies (isopinocampheol, α-pinene oxide, myrtenal and pinanediol) presenting different

68

functional groups (alcohol, epoxide, aldehyde and diol, respectively) were selected, as well

69

as α-pinene. Heterogeneous reactions between these compounds and acidified sulfate

70

aerosol (ammonium sulfate/sulfuric acid) were investigated in highly controlled and

71

repeatable conditions, using a quasi-static reactor which has been developed in this work, to

72

study kinetics and mechanisms of a series of heterogeneous reactions. The reactivity of α-

73

pinene oxide was also examined in an atmospheric simulation chamber to confirm in more

74

atmospherically relevant conditions the results obtained from the quasi-static reactor study.

75

In both experimental approaches, the gaseous phase was characterized by proton transfer

, isoprene epoxydiols37-38, methacrylic acid epoxide (MAE)39 and 2-methyl-3-buten-2-ol

45-46

or nucleophilic substitution of

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

76

reactor time-of-flight mass spectrometer (PTR-TOF-MS) and particle-phase reaction products

77

were analysed by liquid chromatography combined with electrospray ionization (ESI) mass

78

spectrometry (LC/MS) or with by ESI high-resolution time-of-flight mass spectrometry

79

(LC/ESI-HR-QTOFMS) for structural elucidation. First, organosulfate structure identification is

80

reported for the five selected VOCs and corresponding formation pathways are proposed.

81

Then, organosulfate heterogeneous formation mechanism is presented in detail for α-pinene

82

oxide and more briefly for the other compounds with emphasis on the main features of

83

these mechanisms. In a companion article (proposed as Part II), the time-evolution of

84

organosulfates during their formation as well as the effect of particle acidity will be

85

reported.

86

Experimental section

87

Quasi-static reactor experiments

88

The reaction set-up developed in this work is presented here for the first time in Figure 1

89

and was inspired by previous studies examining elementary reactions52-54. Experiments are

90

based on the exposure of particles, deposited on a filter in a quasi-static reactor, to a

91

continuous flow of a single VOC under pseudo-first order kinetic conditions.

92

93

94

95

96

4 ACS Paragon Plus Environment

Page 4 of 49

Page 5 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

97 98 Dilution bulb

99 100 101 102 103

Off-line analysis (LC/MS, LC/QTOF/MS

104

105

106 107

Figure 1. Scheme of the new experimental set-up developed in this work for the study of heterogeneous organosulfate formation

108

109

The formation of organosulfates from α-pinene (Sigma Aldrich, 98 %), α-pinene oxide (Sigma

110

Aldrich, 97 %), isopinocampheol (Sigma Aldrich, 98 %), myrtenal (Sigma Aldrich, 97 %) and

111

pinanediol (Sigma Aldrich, 99 %) was studied. Table 1 presents the chemical structures of

112

these compounds. Their purities were verified using GC/MS analysis and the absence of

113

contamination was confirmed.

114

115

116

117

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

118 119

Page 6 of 49

Table 1. Molecular structures and vapor pressures of biogenic volatile organic compounds studied in the present work α-pinene

α-pinene oxide

Isopinocampheol

Pinanediol

OH

Molecular structure Molecular formula Molecular weight -1 (g.mol ) Vapour pressure (Torr)

O

OH

O

Myrtenal

OH

C10H16

C10H16O

C10H18O

C10H18O2

C10H14O

136.23

152.23

154.25

170.25

150.22

0.823 at 273 Ka

0.13 at 273 K b

0.03 at 295 K b

0.007 at 295 K b

0.44 at 273 K b

120 121 122 123

a

124

VOC concentration was adjusted by dilution of this primary flow at VOC vapour pressure in

125

O2/N2 mixture, of which the ratio was adapted to get 20 % O2 in the final mixture. O2

126

(99.9990 % purity, Linde Gas SA) and N2 (99.9990 % purity, Linde Gas SA) flows were

127

controlled using mass flow controllers. The reaction cell is a 120 mm long glass tube with a

128

24 mm internal diameter and gas inlet and outlet are placed opposite to each other to allow

129

a proper gas circulation within the cell. Sulfate aerosols were generated by atomizing

130

aqueous solutions containing seed aerosol components (summarized in Table 2) and are

131

deposited on 47-mm PTFE filters (Millipore, FluoropeTM, 0.2 µm FG). VOC concentrations

132

were monitored on-line by proton transfer reactor time-of-flight mass spectrometer (PTR-

133

TOF-MS) (Kore Technology) and measured on the range 0.1-10 ppm. Blank samples were

134

prepared by exposing inorganic seed aerosol, deposited on a PTFE filter, to N 2/O2 gas

135

mixture during 15 min for quality control.

136

The validation of this new experimental approach was carefully performed to ensure well

137

defined and repeatable conditions. VOC losses on the reactor walls were characterized by

138

comparing VOC concentrations measured at the outlet of the bubbler with that at the exit of

Hawking and Armstrong, 1954 55 ; b Calculated using SPARC model56

VOC flow was generated by flushing nitrogen (N2) into a bubbler of liquid compound. The

6 ACS Paragon Plus Environment

Page 7 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

139

the set-up and was considered as negligible (< 2 %). Similarly, VOC losses on a simple filter

140

were also determined to be negligible (< 2 %). Gas-phase residence times in the reactor cell

141

were calculated to be fast enough (< 1 s) to ensure a constant concentration of VOC along

142

experiments. Calibrations of the PTR-TOF-MS were performed before and after each series

143

of experiments to check the linearity of the instrument and the absence of any response

144

factor drift. Details on PTR-TOF-MS calibration are presented in supplementary information.

145

Experiments were performed under highly dry conditions (RH < 1 %) where pH calculation

146

are not relevant. Nevertheless, the pH of the particle was calculated to be < 0, as an upper

147

limit, using E-AIM model II57-58 and assuming a 10 % RH. The pH of the particles was also

148

calculated to be < 0 for the chamber experiments performed at 20 and 50 % RH.

149

Operating mode

150

In this study, reactions were performed under darkness, at atmospheric pressure and room

151

temperature (around (295 ± 2) K). Experimental conditions are detailed in Table 2.

152

Approximately about 4 mg of inorganic seed aerosol were generated during 15 min,

153

deposited on a PTFE filter and weighed using a daily calibrated microbalance (TR-64, Denver

154

Instrument Company). Before each experiment, O2/N2/VOC gas mixture was introduced in

155

the reference cell to monitor the VOC concentration. During this operation, the filter was put

156

in the reactor. The gaseous flow was then allowed to pass through the reaction cell and the

157

gas phase was monitored by PTR-TOF-MS throughout experiment periods. For each VOC, at

158

least 6 experiments were carried out at different times of reaction. At the end of the

159

reaction, O2 and VOC gas flows were stopped whereas N2 flow was maintained during 5 min

160

to purge the reactor cell. Particles were then recovered and weighed. Particle losses in the

161

reactor were usually lower than the precision of the balance (0.0001 g). Then, particles were 7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

162

directly introduced into a polypropylene tube (Falcon TM, BD Biosciences) in 3 mL of

163

acetonitrile (ACN). An internal standard aliquot (Camphor sulfonic acid, Sigma Aldrich, 98 %),

164

controlled by gravimetry, was added to the mixture. Aerosol samples were finally kept at -18

165

°C until extraction.

166

Chamber experiments

167

Organosulfate formation from α-pinene oxide reactions was also studied in CESAM chamber

168

(French acronym for Experimental Multiphasic Atmospheric Simulation Chamber) at LISA

169

(Créteil, France). CESAM chamber has been described in detail elsewhere 59. Briefly, this

170

facility consists of an evacuable cylindrical 4.2 m3 stainless steel chamber. In this study,

171

experiments were performed at atmospheric pressure under dark conditions. Initial

172

experimental conditions are described in Table 2. Between each experiment, the chamber

173

was cleaned by maintaining a secondary vacuum better than 6 x 10-4 mbar overnight. A

174

complementary manual cleaning of the inner walls was performed using purified water and

175

ethanol when needed. The chamber was then filled with synthetic air produced from a

176

mixture of 200 hPa of O2 (Linde, 5.0) and 800 hPa of N2 generated from the evaporation of

177

pressurized liquid nitrogen (Messer). Inorganic seed aerosols were introduced into the

178

chamber by atomizing a solution containing the seed aerosol constituents summarized in

179

Table 2. Particle size distributions (20-980 nm in diameter) were measured with a scanning

180

mobility particle sizer (SMPS) composed of a TSI 3080 differential mobility analyser (DMA)

181

and a TSI 3010 condensation particle counter (CPC). A technical issue with SMPS system did

182

not allow us to follow the evolution of aerosol mass for the last two experiments (C5 and

183

C6). After the introduction of inorganic aerosols into the chamber, injection of α-pinene

184

oxide (Sigma-Aldrich, 97 % purity) was performed by introducing a precisely known partial

8 ACS Paragon Plus Environment

Page 8 of 49

Page 9 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

185

pressure of the compound (0.20 ± 0.001 mbar) prepared in a bulb of a known volume (V =

186

2.90516 L) from a frozen pure standard solution using a vacuum gas manifold. The bulb

187

content was then immediately flushed with N2 into the chamber. The concentration of α-

188

pinene oxide was monitored using a Fourier-transform infrared spectrometer (FTIR) from

189

BrukerTM Tensor 37® GMbH. The total optical path length for the in-situ FTIR measurement

190

was set to 192 m. In addition to FTIR, gas-phase constituents were also monitored by PTR-

191

TOF-MS. Each reaction was studied for 3 hours.

192

Organosulfate formation was monitored by filter samples collected over the time of the

193

experiment (Ftot). Aerosol sampling was achieved using 47-mm PTFE filters (Millipore,

194

FluoropeTM, 0.2 µm FG) at a flow rate of 3 L min-1. The filter holders were installed

195

downstream of an activated charcoal denuder, used to trap reactive gases and reduce

196

positive sampling artefacts. Furthermore, inorganic seed aerosols were collected before the

197

injection of VOC for 25 min (blank filter). After collection, filter samples were introduced in a

198

flask with 3 mL of ACN and a known amount of internal standard and kept at -18 °C through

199

to LC-ESI-MS and LC/ESI-HR-QTOFMS analyses.

200

Sample preparation and analysis

201

Sample preparation

202

Filter samples were extracted with internal standard in 3 mL of ACN (HPLC Grade, JT Baker)

203

during 15 min of ultrasonic agitation. Camphor sulfonic acid was used as internal standard

204

for quantification of organosulfates, due to the lack of commercially available authentic

205

standards for organosulfates. Although the use of this internal standard adds some

206

uncertainty to the absolute quantification of organosulfates, it provides relative but yet

207

reliable information and it allows to compare quantitatively the organosulfate formation 9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 49

208

between experiments. Then, extracts were filtered by centrifugation using PTFE filters

209

(Ultrafree-MC, PTFE Membrane, 0.22 µm) and concentrated to about 50 µL under gentle N 2

210

(g) stream (99.995 % purity, Linde Gas SA) at 40 °C. Syringe standard (octanoic acid

211

Isotec, 99 %) was added to the samples for quantification of the internal standard. This

212

double quantification allows for the calculation of internal standard recovery yields and

213

hence to check that internal standards, and therefore organosulfates, are not lost along the

214

analytical procedure. Internal standard recovery yields were higher than 80 % for all samples

215

analysed in the present work. Finally, two aliquots were prepared for each sample: one

216

consisted in the dilution of 10 µL of the extract in 90 µL of milli-Q water, as it corresponds to

217

the eluent initial composition in the chromatographic separation method, the other in 100 %

218

ACN to prevent products from hydrolysis.

219

The calibration solutions were also used to check and monitor the conditions of the

220

chromatographic and detection systems. These solutions were injected before each analysis

221

sequence and about every 10 samples in order to calculate the internal standard response

222

factors with respect to the syringe standards. Analyses were carried out only if conditions

223

required for the analyses (absence of contamination in blank samples, sensitivity and

224

response factors within 15 % of the optimized conditions) were fulfilled.

225

Two complementary off-line techniques were used to study the chemical composition of

226

aerosols after exposure to VOCs. LC/ESI-HR-QTOFMS allowed to determine chemical

227

structures of organosulfates formed in the present study, whereas LC/ESI-MS was used to

228

validate extraction procedure and quantify products generated during experiments. The

229

analytical method (using LC/ESI-HR-QTOFMS) allowed to quantify organosulfates with a limit

230

of quantification of 10 pg and a standard deviation equal to ± 15 %.

10 ACS Paragon Plus Environment

13

C,

Page 11 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

231

Liquid chromatography coupled to electrospray ionization mass spectrometry (LC/(-)ESI-

232

MS)

233

Filter extracts were analysed by LC/ESI-MS (Agilent Technology, Serie 1100 LC/MSD). Two

234

separation methods were used in this work and details are provided in Table S1. The first

235

one was developed in order to quantify organosulfate monomers and the second to improve

236

dimer separation. The eluent composition was (A) 0.2 % formic acid in Milli-Q grade water

237

and (B) 0.2 % formic acid in acetonitrile. The injected volume was 5 µL. The following

238

parameters were selected for the operation of the (-)ESI-MS after optimization to 3.0 kV for

239

capillary voltage, 100 V for fragment, 350 °C for drying gas temperature, 11 L min-1 for drying

240

gas flow and 30 psig for nebulizer pressure, to minimize fragmentation of molecular ion and

241

improve the sensitivity of MS signal detection. All products were detected as their

242

deprotonated ions (i. e. [M – H]-).

243

Liquid chromatography coupled to electrospray ionization quadrupole time-of-flight mass

244

spectrometry (LC/(-)ESI-HR-QTOFMS)

245

The same chromatographic separation methods were used as described above. The HPLC

246

system (Agilent Series 1290, Agilent Technologies) was coupled to a high resolution

247

quadrupole time-of-flight mass spectrometer Model 6540 Agilent (Agilent Technologies)

248

equipped with Dual ESI, operating in negative mode using the following operation

249

parameters: capillary voltage - 3000 V; nebulizer pressure - 30 psig; drying gas flow - 11 L

250

min-1; sheath gas temperature - 350 °C; fragmentor voltage - 100 V; skimmer voltage - 65 V;

251

octopole RF - 750 V. LC/ESI-HR-QTOFMS accurate mass spectra were recorded across the

252

range 100-1000 m/z at 2 GHz. The recorded data were processed with Mass Hunter

253

Qualitative software (version B.05.00 Agilent Technologies, Santa Clara, CA, USA). Accurate

254

mass measurements of each peak from the extracted ion chromatograms (EICs) were 11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 49

255

obtained by means of a calibrant solution delivered by an external quaternary pump. This

256

solution contains the internal reference masses HP-921 = hexakis (1H, 1H, 3H-

257

tetrafluoropropoxy)-phosphazine [M+HCO2]- at m/z = 966.000725. Stability of mass accuracy

258

was checked daily. The instrument was operated in MS/MS mode for identification of

259

selected compounds. Structure analysis of single molecular ions in the mass spectra from

260

reactivity experiments was performed by mass-selecting the ion of interest, which was in

261

turn submitted to 10-60 eV collision with N2 in the collision cell. Organosulfate functional

262

groups were identified from the potential loss of SO3- (m/z = 79.9574), SO4- radical (m/z =

263

95.9523) or HSO4- (m/z = 96.9601). For all experiments conducted in the present work and

264

presented in Table 2, blank samples were performed to check for any contamination or

265

memory

effects.

No

organosulfates

were

detected

12 ACS Paragon Plus Environment

in

blank

samples.

Page 13 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

266

The Journal of Physical Chemistry

Table 2. Summary of the experimental conditions in reactor and chamber experiments and quantification of organosulfates

Quasi-static reactor experiments

# experiment series

VOC (concentration in ppb)

R1

α-pinene oxide (1000 ± 20)

R2

number of experimentsa

Particle mass (mg)

RH (%)

0.03/0.05

n=8

4.1

α-pinene (1000 ± 20)

0.03/0.05

n=7

R3

isopinocampheol (1000 ± 20)

0.03/0.05

R4

pinanediol (1000 ± 20)

R5

myrtenal (1000 ± 20)

267 268

a

m/z 231.0697b

m/z 233.0853 b

m/z 249.0802 b

m/z 401.2003 b

m/z 481.1571 b

Sum of organosulfates

OS 231

OS 233 (1)

OS 233 (2)

OS 233 (3)

OS 249 (1)

OS 249 CAc

OS 249 (2)

ƩDiOS 401

ƩDiOS 481