Glyoxal and Methylglyoxal Setschenow Salting ... - ACS Publications

Sep 3, 2015 - Glyoxal and Methylglyoxal Setschenow Salting Constants in Sulfate,. Nitrate, and Chloride Solutions: Measurements and Gibbs Energies...
0 downloads 0 Views 1MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Glyoxal and Methylglyoxal Setschenow Salting Constants in Sulfate, Nitrate and Chloride Solutions: Measurements and Gibbs Energies Eleanor M. Waxman, Jonas Elm, Theo Kurtén, Kurt V. Mikkelsen, Paul Jeffrey Ziemann, and Rainer Volkamer Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b02782 • Publication Date (Web): 03 Sep 2015 Downloaded from http://pubs.acs.org on September 15, 2015

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 29

Environmental Science & Technology

1

Glyoxal and Methylglyoxal Setschenow Salting Constants in Sulfate, Nitrate and Chloride

2

Solutions: Measurements and Gibbs Energies

3

Eleanor M. Waxman,1,2 Jonas Elm,3,4 Theo Kurtén,5 Kurt V. Mikkelsen,3 Paul J. Ziemann,1,2 and

4

Rainer Volkamer1,2*

5

1. University of Colorado Boulder, Department of Chemistry and Biochemistry, UCB 215,

6

Boulder, Colorado, United States

7

2. CIRES, University of Colorado, UBC 216, Boulder, Colorado, United States

8

3. University of Copenhagen, Department of Chemistry, Universitetsparken 5, 2100 København

9

Ø, Denmark

10

4. University of Helsinki, Department of Physics, P. O. Box 64, Finland

11

5. University of Helsinki, Department of Chemistry, P. O. Box 55, Finland

12 13

*Corresponding author. [email protected]. Telephone: +1 (303) 492-1843. Fax:

14

+1 (303) 492-5894

15 16

Abstract

17

Knowledge about Setschenow salting constants, KS, the exponential dependence of Henry’s Law

18

coefficients on salt concentration, is of particular importance to predict secondary organic

19

aerosol (SOA) formation from soluble species in atmospheric waters with high salt

20

concentrations, such as aerosols. We have measured KS of glyoxal and methylglyoxal for the

21

atmospherically relevant salts (NH4)2SO4, NH4NO3, NaNO3 and NaCl, and find that glyoxal

22

consistently ‘salts-in’ (KS of -0.16, -0.06, -0.065, -0.1 molality-1, respectively) while

23

methylglyoxal ‘salts-out’ (KS of +0.16, +0.075, +0.02, +0.06 molality-1). We show that KS values

1 ACS Paragon Plus Environment

Environmental Science & Technology

Page 2 of 29

24

for different salts are additive, and present an equation for use in atmospheric models.

25

Additionally, we have performed a series of quantum chemical calculations to determine the

26

interactions between glyoxal/methylglyoxal monohydrate with Cl-, NO3-, SO42-, Na+, and NH4+,

27

and find Gibbs free energies of water displacement of -10.9, -22.0, -22.9, 2.09, and 1.2 kJ/mol

28

for glyoxal monohydrate, and -3.1, -10.3, -7.91, 6.11, and 1.6 kJ/mol for methylglyoxal

29

monohydrate with uncertainties of 8 kJ/mol. The quantum chemical calculations support that

30

SO42-, NO3- and Cl- modify partitioning, while cations do not. Other factors such as ion charge or

31

partitioning volume effects likely need to be considered to fully explain salting effects.

32 33

1. Introduction

34

Most atmospheric aerosols contain an inorganic fraction.1,2 This fraction is typically

35

made up of salts containing sodium chloride (from sea spray) and ammonium, nitrate, and sulfate

36

that result from the gas-phase processing of anthropogenic precursor gases such as NOx, SO2,

37

and NH3. The hygroscopicity of these inorganic ions is responsible for the majority of the water

38

taken up by the aerosols and the formation of an aerosol aqueous phase.3–5 Salt concentrations

39

tend to be quite high in aerosols; between 3 M and 20 M.6

40

Organic aerosol formation by aqueous-phase processing is a topic of recent interest.7–19

41

Small, water-soluble species such as glyoxal, methylglyoxal, acetaldehyde, glycolaldehyde, or

42

isoprene epoxydiols (IEPOX) partition to the aerosol aqueous phase by Henry’s law. They can

43

then undergo further aqueous-phase reactions such as hydration, oligomerization, reactions with

44

nitrogen-containing species, reactions with sulfate ions, and aqueous-phase oxidation reactions in

45

the presence of OH radicals.20–24

2 ACS Paragon Plus Environment

Page 3 of 29

Environmental Science & Technology

46

The presence of salts in the aqueous phase can significantly impact the solubility and

47

activity of organics in the aerosol as well as modify bulk (as opposed to surface) reaction rate

48

constants. When salts increase the solubility of the organic, this is called “salting in” with

49

respect to the organic and when salts decrease the solubility of the organic, this is called “salting

50

out”. This effect was first published in 1889 by Setschenow who described it in terms of an

51

organic’s solubility in water:25

52

log ( S0 / S ) = KS Csalt

53

where S0 and S are the solubility of the organic in pure water and the salt solution respectively,

54

KS (here, per molarity) is the salting constant or Setschenow constant, and Csalt (molarity) is the

55

concentration of the salt solution, typically in molarity. Kampf et al.6 modified this equation to

56

represent Csalt based on molality, and to describe salting in terms of Henry’s law coefficients:6

57

 KH ,w log  K  H ,salt

58

where KH,w is the Henry’s law constant of the organic in pure water, KH,salt is the Henry’s law

59

constant in the salt solution, and csalt is the concentration of the salt in molality. Here, KS has the

60

units of per molality (m-1). Therefore, this effect changes the partitioning of these species

61

between the gas and aqueous phase. If the molecule salts in, partitioning towards the aqueous

62

phase is favored which increases the amount of precursor molecule available for further reactions

63

and SOA formation. If the molecule salts out, the reverse is true. The effective Henry’s law

64

constant thus becomes a function of environmental conditions.

(Eq. 1)

  = KS csalt 

(Eq. 2)

65

Recent measurements show that glyoxal salts in to ammonium sulfate aerosols.6 A series

66

of quantum chemical calculations suggests that this is because the hydrated forms of aqueous-

67

phase glyoxal, especially the mono-hydrate, bind to sulfate more strongly than water in the

3 ACS Paragon Plus Environment

Environmental Science & Technology

Page 4 of 29

68

sulfate hydration shell (water displacement).26 This results in the loss of free glyoxal in the

69

aqueous phase as part of the sulfate hydration shell. Therefore, additional glyoxal must partition

70

in to the aqueous phase to maintain equilibrium thus increasing the effective Henry’s law

71

constant. Here we present additional experimental data for glyoxal and methylglyoxal for the

72

atmospherically-relevant salts ammonium sulfate ((NH4)2SO4), ammonium nitrate (NH4NO3),

73

sodium nitrate (NaNO3), and sodium chloride (NaCl), as well as quantum mechanical

74

calculations for interactions with salt ions.

75

complicated than simply replacing the water in the anion hydration shell with exactly one

76

organic molecule.

77

2. Experimental

78

2.1. Laboratory Salting Constant Measurements

79

Salting constant measurements for glyoxal and methylglyoxal were performed using two

80

different sets of instrumentation: (1) a novel inlet coupled with a quadrupole ion trap mass

81

spectrometer (ITMS measurements), and (2) gas chromatography with flame ionization detector

82

(GC-FID measurements).

83

unreported, the intercomparison between the two methods serves as method validation.

We also show that salting behavior is more

Since salting constant values for these species are previously

84

For both sets of measurements, 10.00 mL solutions of 0.1 M methylglyoxal (40% w/w

85

solution, Sigma Aldrich) or 0.2 M glyoxal (40% w/w solution, Sigma Aldrich) were prepared in

86

18 MΩ water (Millipore). Solution concentrations were chosen to minimize oligomerization27

87

but give sufficient gas-phase signal. In addition to glyoxal or methylglyoxal, the solutions also

88

contained 0, 0.5, 1.0, 1.5, 2.0., or 3.0 molarity salt solution (ammonium sulfate, ammonium

89

nitrate, sodium nitrate, or sodium chloride, all Sigma Aldrich).

All measurements were

4 ACS Paragon Plus Environment

Page 5 of 29

Environmental Science & Technology

90

performed at room temperature (22 ± 2°C). A list of experimental conditions and number of

91

repeats are listed in Table S1.

92

Notably, measurements of KS do not require knowledge of the absolute vapor pressures.

93

Instead both the ITMS and GC-FID setup rely on relative measurements only, i.e., the signal

94

measured by sampling the gas-phase above a solution containing both the organic and the salt is

95

compared to the signal from sampling the gas-phase above a solution that does not contain the

96

salt.

97

2.1.1 Quadrupole Ion Trap Measurements

98

A novel atmospheric pressure inlet was developed for these measurements. A diagram of

99

this inlet is shown in Figure 1, panel A. Solutions were added to a 10 mL graduated cylinder

100

outfitted to connect to a 0.5 inch UltraTorr. The system in front of the ion trap was then pumped

101

down to house vacuum pressures (nominally 60 hPa) to remove excess O2 and N2. The air in

102

equilibrium with the solution leaked through the pinhole (0.0007 inches, O’Keefe Controls).

103

This ensured that the system was in an effusion regime which means that the water molecules

104

diffused slightly faster than the organic solute molecules. The faster diffusion of the water

105

molecules helped to pull along the organic molecules, thus shortening the time scales under

106

which equilibrium was reached. The faster evaporation of water did not significantly alter the

107

salt molarity of the solution. The valve to the quadrupole ion trap mass spectrometer (Thermo-

108

Finnigan LCQ) was then opened, letting the sample in via the gas chromatography port. Data

109

was collected for thirty minutes at which point signal in the mass spectrometer had stabilized

110

(Figure S1). The last five minutes of the data was averaged and used in the analysis. No change

111

in solution volume was observed so no change in solution temperature due to evaporation was

112

expected to occur.

5 ACS Paragon Plus Environment

Environmental Science & Technology

Page 6 of 29

113

A blank (18 MΩ water, Millipore) was run between each methylglyoxal solution to

114

determine the extent of background methylglyoxal signal remaining in the ion trap. Signal was

115

normalized to m/z 19 (H3O+, the strongest ion in the spectrum and not a methylglyoxal fragment)

116

to account for any variability in ion trap signal. The water spectrum immediately preceding a

117

methylglyoxal spectrum was then subtracted from the methylglyoxal spectrum to remove any

118

residual background methylglyoxal, N2, O2, and instrument contaminants.

119

methylglyoxal in the blank water spectra was between 1 and 2% of the maximum signal of the

120

methyl glyoxl spectrum. A typical spectrum is shown in Figure 1, panel B. The standard NIST

121

EI spectrum is shown in black on the top and the blank-corrected methylglyoxal spectrum

122

measured in this experiment is shown on the bottom in red. There is excellent agreement

123

between the two spectra. The signals at m/z 32 and around m/z 16 and 18 are due to incomplete

124

water and air subtraction, and the extra signal at m/z 58 is due to background contamination from

125

the ion trap but overall this method does an excellent job of correcting for background signal.

126

To calculate salting constant values, we expand the Setschenow equation to include the

127

explicit definition of Henry’s law:

128

 Corg ,w  porg ,w log   Corg ,salt   porg ,salt

129

Typical residual

  =K c S salt   

(Eq. 3)

130

where Corg,w (molarity) is the concentration of the organic species in pure water, porg,w (atm) is

131

the vapor pressure of the organic species over that solution, Corg,salt (molarity) is the

132

concentration of the organic species in the salt solution with concentration csalt (here, molality

133

but analysis can also be done with molarity) and porg,salt is the vapor pressure of the organic

6 ACS Paragon Plus Environment

Page 7 of 29

Environmental Science & Technology

134

species over that salt solution. In this experiment, Corg,w and Corg,salt are the same (0.1 M

135

methylglyoxal or 0.2 M glyoxal), so this reduces to:

136

 porg ,salt log   porg , water 

  = K S csalt 

(Eq. 4)

137

The signal at m/z 43 from the ion trap is taken to be a proxy for the vapor pressure

138

because only relative measurements (rather than absolute pressure measurements) are necessary.

139

Thus the salting constants can be calculated solely from the ion trap signal.

140

2.1.2 Gas Chromatography Measurements

141

Methylglyoxal and glyoxal salting constant measurements were also made using gas

142

chromatography with flame ionization detection. GC-FID instruments are most sensitive to

143

reduced carbon and signal is proportional to the number of reduced carbons that elute from the

144

GC column at a given time.

145

Therefore, to increase the number of reduced carbons and thus the sensitivity of the GC as well

146

as lowering the vapor pressure of the molecules, they were derivatized with O-(2,3,4,5,6-

147

Pentafluorobenzyl)hydroxylamine hydrochloride (PFBHA, Sigma Aldrich).

148

reacts with carbonyl groups to form an oxime.

Methylglyoxal has one reduced carbon; glyoxal has none.

This molecule

149

Solutions of organics and salt in water were added to 22 mL Supelco vials with Mininert

150

valves with septa. A schematic of this is shown in Figure 1, Panel C. Solutions were allowed to

151

equilibrate for 45 minutes (at longer time scales, imidazoles were formed in ammonium

152

sulfate/glyoxal solutions and the aqueous phase concentration of glyoxal could no longer be

153

assumed to be the same as the initial concentration of glyoxal). A solid phase micro extraction

154

(SPME) fiber (65 µm poly(dimethylsiloxane)/divinylbenzene, Supelco) was first exposed to the

155

headspace above a continuously stirred 17 mg/mL solution of PFBHA for two minutes. The

7 ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 29

156

fiber was then exposed to the headspace above the solution of organics and salt for one minute

157

(methylglyoxal) or fifteen minutes (glyoxal). This method requires the assumptions that for salt

158

and pure water solutions (1) the fiber is identically coated with PFBHA every time, (2) the gas-

159

phase diffusion rates of the organic solute are the same, and (3) the derivatization reaction rates

160

are the same. These assumptions are difficult to test, but we have exposed the fiber to the

161

headspace above the PFBHA solution for the same amount of time to minimize variability in

162

PFBHA coating, and we have no reason to believe that the diffusion rate for either glyoxal or

163

methylglyoxal should vary based on the salt concentration since the salt is assumed to be non-

164

volatile, nor should this affect the derivatization rates. Thus it is important to expose the fibers

165

for a consistent period of time for a given organic solute. The fiber was then injected in to the

166

GC and measured immediately. The GC program was as follows: 40°C for 2 minutes, 10°C/min

167

to 110 °C, 5°C/min to 155 °C, 10°C/min to 280°C, 280°C for 5 minutes. The inlet was held at

168

250°C. The GC was a Hewlett-Packard model 6890 equipped with a 30 m × 0.32 mm Agilent

169

DB-1701 column with 1 µm thickness. PFBHA was always measured in significant excess to

170

the oxime products, indicating that the amount of oxime formed was limited by the amount of

171

gas-phase glyoxal or methylglyoxal, rather than the amount of PFBHA. Tests showed that after

172

a run, all PFBHA and oxime products had been volatilized off of the SPME fiber so no blanks

173

were run between measurements. For analysis, peak areas were used as the proxy for porg,w and

174

porg,salt. All measurements were repeated; the number of repeats are given in Table S1.

175

We additionally performed one set of measurements for glyoxal with both 1.6 molal

176

ammonium sulfate and 1.6 molal ammonium nitrate to study the effects of salting in a solution

177

with mixed salts.

178

2.2 Computational Methodology

8 ACS Paragon Plus Environment

Page 9 of 29

Environmental Science & Technology

179

Electrostatic interactions between the organics and salt ions were calculated using

180

quantum chemical calculations. All geometry optimizations and frequency calculations were

181

performed with Gaussian0928 using the M06-2X functional.29 The M06-2X functionality was

182

chosen on basis of recent benchmarks showing its adequate performance in describing sulfur

183

containing compounds and yielding reliable thermodynamics for sulfuric acid-water clusters.30–32

184

For the initial conformational sampling solvent effects are taken into account using a Polarizable

185

Continuum Model with the integral equation formalism variant (IEFPCM).33,34

186

Methylglyoxal can exist in three different forms in aqueous solution: unhydrated,

187

monohydrated and di-hydrated. Recently, we showed that glyoxal has an unfavorable interaction

188

with sulfate, but that partitioning into sulfate aerosol could occur through the hydrates leading to

189

a salting-in effect.26 Therefore only the methylglyoxal hydrates are further considered in this

190

study. Methylglyoxal monohydrate exists in two different forms since water can be added either

191

to the aldehyde or ketone moiety,35 which will be denoted monohydrate-a and monohydrate-k,

192

respectively. For details on the conformational sampling of molecules and clusters, see the

193

Supplemental Information. In Figure S2 the lowest identified free energy conformations are

194

shown optimized at the M06-2X/6-31+G(d) level of theory in water. The relative stability of the

195

monohydrates indicates that monohydrate-a is 17.9 kJ/mol more stable than monohydrate-k, and

196

therefore the second monohydrate would only exist at lower concentrations.

197

results from Nemet et al.35 show that the dihydrate can be present at significant concentrations

198

(about 40% of the total methylglyoxal), however as our calculations show that the monohydrate

199

is the most abundant species, we focus on the monohydrate for all further analysis.

200

formation of the dihydrate will occur from adding water to monohydrate-a through the following

201

reaction:

Experimental

The

9 ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 29

202

monohydrate-a + H2O ↔ dihydrate

203

The formation free energy of the dihydrate is found to be +7.32 kJ/mol, and once appropriate

204

water concentrations are accounted for, monohydrate-a and the dihydrate are expected to be

205

present in similar concentrations within uncertainties.

206

monohydrate is more prevalent than the dihydrate; therefore only the monohydrate-a

207

conformation needs to be taken into account when exploring the salting-in/out effect of

208

methylglyoxal. From here on in, we refer to monohydrate-a as simply monohydrate.

Experimental results show that the

209

To investigate potential salting-in/out effects of methylglyoxal with various ions, the

210

following reactions between the methylglyoxal monohydrate and the ions Cl–, NO3–, SO42−, Na+

211

and NH4+ are studied:

212

Monohydrate(aq) + Cl−(aq) ↔ (Cl−)(monohydrate)(aq) (1)

213

monohydrate(aq) + NO3−(aq) ↔ (NO3−)(monohydrate)(aq) (2)

214

monohydrate(aq) + SO42−(aq) ↔ (SO42-)(monohydrate)(aq) (3)

215

monohydrate(aq) + Na+(aq) ↔ (Na+)(monohydrate)(aq) (4)

216

monohydrate(aq) + NH4+(aq) ↔ (NH4+)(monohydrate)(aq) (5)

217

The dianions are known to be unstable in the gas-phase and prone to electron spill: the

218

exothermal and very rapid detachment of one electron to yield a radical monoanion and a free

219

electron.36

220

molecules significantly stabilize the dianion with respect to the monoanion and free electron, this

221

issue presents severe technical problems for the quantum chemical modelling of aqueous-phase

222

reactions of dianions. Continuum solvent models (e.g. IEFPCM), which describe most neutral

223

solvent-phase reactions qualitatively correctly, are not alone able to describe reactions involving

224

small dianions. In order to circumvent possible errors arising from electron detachment, as well

Though not directly relevant for solution-phase processes, where the solvent

10 ACS Paragon Plus Environment

Page 11 of 29

Environmental Science & Technology

225

as to improve the overall description of the hydration of the ions and ion-molecule clusters, we

226

have modeled hydration both implicitly (using the IEFPCM continuum solvent model) and

227

explicitly, by adding water molecules to the simulation. In the case of sulfate, it has been shown

228

that at least three water molecules are required to stabilize the dianion with respect to electron

229

detachment.37,38 The combination of both implicit and explicit hydration was recently shown to

230

provide a qualitatively correct description of salting out for glyoxal in sulfate solutions.26 By

231

utilizing a similar approach for methylglyoxal, the cluster reactions can be modeled using the

232

following displacement reactions:

233

monohydrate(aq) + (Cl−)(H2O)3(aq) ↔ (Cl−)(monohydrate)(H2O)2 + H2O (6)

234

monohydrate(aq) + (NO3−)(H2O)3(aq) ↔ (NO3−)(monohydrate)(H2O)2(aq) + H2O (7)

235

monohydrate(aq) + (SO42−)(H2O)6(aq) ↔ (SO42−)(monohydrate)(H2O)5(aq) + H2O (8)

236

monohydrate(aq) + (Na+)(H2O)3(aq) ↔ (Na+)(monohydrate)(H2O)2(aq) + H2O (9)

237

monohydrate(aq) + (NH4+)(H2O)3(aq) ↔ (NH4+)(monohydrate)(H2O)2(aq) + H2O (10)

238

Reactions 1-5 represent the fundamental chemical processes we wish to understand.

For

239

technical reasons we represent these as reactions 6-10 as model systems to study reactions 1-5.

240

Aside from the electron detachment issue, any potential errors in the entropy contribution are

241

simultaneously reduced by keeping the number of molecules fixed in the displacement reactions.

242

We have chosen to hydrate the singly charged ions by three water molecules and the doubly

243

charged ions with six water molecules to be certain that there are no errors due to electron

244

detachment. The structure of the ion-water clusters have in several cases previously been

245

identified and were extracted from the following publications: (Cl–)(H2O)n from Nadykto et al.39,

246

(NO3−)(H2O)n from Liu et al.38, (NH4+)(H2O)n from Pickard et al.40 and (SO42−)(H2O)n from

247

Lambrecht et al.41 In case of the (Na+)(H2O)3 cluster, we were unable to find any literature

11 ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 29

248

structure and it was therefore manually constructed.

For details of the construction and

249

configurational sampling of the monohydrate-ion-water clusters, the SI and the references cited

250

therein.

251

3. Results and Discussion

252

The results of the salting constant measurements are shown in Figure 2, where the slopes

253

of the lines correspond to the KS values. Salting constants for (NH4)2SO4 are shown in panel A,

254

NH4NO3 in panel B, NaNO3 in panel C, and NaCl in panel D.

255

measurements are shown with open circles in brighter colors.

256

designated with solid lines. Methylglyoxal GC-FID measurements are shown in darker colors

257

with open squares. Fits to the methylglyoxal GC-FID measurements are given with dotted lines.

258

Glyoxal GC-FID measurements are shown in darker colors with solid diamonds. Fits to the

259

glyoxal data are given with dashed lines. The grey lines are the average of the slopes of the GC-

260

FID and ITMS measurements for methylglyoxal.

Methylglyoxal ITMS

Fits to the ITMS data are

261

Salting constants for glyoxal and averages of methylglyoxal values are presented in Table

262

1 along with the smallest, most water-soluble species measured in Wang et al.42 and Endo et al.43

263

(converted to molality). All salting constant values are presented in Table S2 in both molarity

264

and molality, with ITMS and GC-FID measurements for methylglyoxal separated for clarity.

265

Excellent agreement is observed between the ITMS measurements and the GC-FID

266

measurements for all the salts, except for ammonium nitrate, where measurements only agree

267

marginally within error. The source of this discrepancy is presently unclear. Generally the good

268

agreement for other salts confirms that both methods gave consistent results.

12 ACS Paragon Plus Environment

Page 13 of 29

Environmental Science & Technology

269

Our results in Table 1 are presented in molality, as this is the more relevant and more

270

easily calculated value for atmospheric aerosols.6 For consistency in Table 1, we have converted

271

the values from Wang et al.42 from molarity to molality (see Table S1 for molarity units).

272

Additionally, good agreement is observed between the glyoxal-(NH4)2SO4 salting

273

constant as measured in Kampf et al.6 and in this work, given the very different salt

274

concentration regimes (this work: 0 to 3.8 m ammonium sulfate in bulk solutions; Kampf et al:6

275

4-15 m ammonium sulfate in metastable aerosols) and different methods of measuring both the

276

gas and aqueous phase concentrations.

277

Glyoxal very clearly salts in to aqueous salts, whereas methylglyoxal salts out. This

278

indicates that molecular structure is critical for determining the salting behavior of a species, as

279

glyoxal and methylglyoxal are fairly similar except for the methyl group. They are both small α-

280

dicarbonyl species, have very high Henry’s law constants, have the ability to form hydrates in

281

water, and have very high O/C ratios. However, the addition one methyl group converts an

282

aldehyde group to a ketone; this change in structure switches the species from salting in to

283

salting out. Additionally, even the smallest, most water-soluble species (2-butoxyethanol) tested

284

in Wang et al.42 also salts out – more strongly than methylglyoxal. All of the species measured

285

in Wang et al.42 and Endo et al.43 as well as methylglyoxal salt out; this suggests that molecules

286

that salt in are likely a small subset of the molecules that occur in the atmosphere. Additionally,

287

there is a large difference in the magnitude of the salting constants for methylglyoxal and for the

288

species from Wang et al.42, where the larger species in Wang et al.42 were found to salt out of

289

(NH4)2SO4 solutions much more strongly than methylglyoxal. Figure 3 shows that there is a

290

clear trend with increasing salting out constants with carbon number for a series of 2-ketones (2-

291

hexanone to 2-decanone were measured) and aldehydes.42 Interestingly, there seems to be an

13 ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 29

292

inverse correlation with expected number of OH groups and magnitude of the salting out

293

constant measured in Wang et al.42 In particular, 1-hexanol (1 OH group), 2-hexanone (0 OH

294

groups as hydration is not expected), and heptanal (2 OH groups expected due to hydration) salt

295

out with KS values of 0.364 m-1, 0.38 m-1 and 0.35 m-1, respectively. Despite the longer carbon

296

chain heptanal has the lowest value. This is consistent with work by Kurtén et al.26 who found

297

that the salting in from glyoxal is due to the interaction of the OH groups in glyoxal

298

monohydrate and glyoxal dihydrate with the sulfate anion.26 Figure 3 also shows calculations for

299

salting constants predicted using a poly-parameter linear free energy relationship (ppLFER; for

300

details see Wang et al.42) for glyoxal and methylglyoxal with open triangles and open squares,

301

respectively. ppLFER predictions of salting constants are shown for the unhydrated,

302

monohydrate, and dihydrate forms. The values for the unhydrated and monohydrate forms seem

303

to bracket the value backextrapolated from the ketone fit. The best estimate for the salting

304

constants calculated by the ppLFER method are shown as a solid triangle (glyoxal) and square

305

(methylglyoxal), accounting for the hydration equilibrium. The best estimate value for glyoxal

306

corresponds to that of the dihydrate form. For methylglyoxal, the best KS value represents

307

approximately 60% monohydrate and 40% dihydrate35 and falls in between both forms. The KS

308

measurements for glyoxal and methylglyoxal are given by the black triangle and square,

309

respectively. The best prediction from the ppLFER method shows a reasonable agreement for

310

methylglyoxal, but overestimates salting out by about 20%. For glyoxal, the ppLFER estimation

311

suggest salting out, while salting in is observed. Glyoxal is significantly smaller, more water-

312

soluble, and more highly oxygenated (i.e., has a larger number of OH groups) than the molecules

313

used during fitting of the ppLFER parameters; it is thus not expected to be well-represented by

314

this equation. This further underscores the importance of molecular structure, as the

14 ACS Paragon Plus Environment

Page 15 of 29

Environmental Science & Technology

315

measurements are significantly lower (less salting out) than predicted by the correlation, in good

316

consistency with the salt interactions of glyoxal and methylglyoxal hydrates. This highlights the

317

need for additional measurements of water-soluble SOA precursors.

318

Our results are consistent with the Hofmeister series, which was initially developed to

319

explain the strength of ions on the salting out of proteins. This series is listed as CO32- > SO42- >

320

S2O32- > H2PO4- > F- > Cl- > Br- > NO3- > I- > ClO4- > SCN- where CO32- has the greatest effect

321

on protein salting and SCN- has the least effect.44

322

(NH4)2SO4 has the strongest effect, followed by NaCl. The nitrate salts have the weakest effect

323

on salting, but a non-zero effect. Moreover, Endo et al.43 measured salting constants for a

324

number of molecules in NaCl and Wang et al.42 measured salting constants for the same species

325

in (NH4)2SO4. In all cases, the molecules salted out of (NH4)2SO4 more strongly than NaCl.

326

Gordon and Thorne45 measured salting constants of naphthalene in sodium sulfate as well as a

327

number of chloride salts and found that sulfate caused naphthalene to salt out more strongly than

328

any of the chloride salts. Görgényi et al.46 measured salting constants for chloroform, benzene,

329

chlorobenzene, and anisole in a wide variety of nitrate, chloride, and sulfate salts.

330

chloroform, chlorobenzene, and anisole, sulfate salting constants were significantly higher than

331

the chloride salting constants which were higher than the nitrate salting constants. For benzene,

332

they found that sulfate had the strongest effect but nitrate and chloride salts had similar effects.46

For both glyoxal and methylglyoxal,

For

333

We also find a decrease in vapor pressure upon going from 1.6 m (NH4)2SO4 to 1.6 m

334

(NH4)2SO4 plus 1.6 m NH4NO3 with a set of six replicate measurements, which indicates that the

335

addition of NH4NO3 causes additional salting in beyond that caused by (NH4)2SO4. The decrease

336

is consistent with that observed upon going from 0 m NH4NO3 to 1.6 m NH4NO3. This suggests

337

that salting constants are additive since this additional decrease is what would be predicted based

15 ACS Paragon Plus Environment

Environmental Science & Technology

Page 16 of 29

338

on the NH4NO3 salting constant. Therefore, the concentration of glyoxal that partitions to the

339

aerosol will be dependent on the concentration of all salts. We propose that the salting behavior

340

of glyoxal in a mixed (NH4)2SO4/NH4NO3 system can be calculated using:

341

 K H ,w log  K  H ,salt

342

This equation is consistent with our data, but the additivity of salting constants merits further

343

research.

  = −0.24 × c( NH4 )2 SO4 − 0.07 × cNH4 NO3 

(Eq. 5)

344

The fact that salting constants appear to be additive is consistent with the literature.

345

Gordon and Thorne45 performed a systematic series of salting constant measurements on

346

naphthalene in mixed salt solutions. They also found that the effects of salts are additive for

347

mixtures of sodium chloride with magnesium chloride, sodium sulfate, cesium bromide, calcium

348

bromide, and potassium bromide. Our results are thus in good agreement with this study since

349

we also find that mixture of (NH4)2SO4 and NH4NO3 is additive. However, this warrants further

350

investigation.

351

Results of quantum chemical calculations are given in Table 2. This lists the ∆G values

352

for the displacement of one water molecule with either a glyoxal monohydrate molecule or a

353

methylglyoxal monohydrate molecule. These values are all negative for anions and positive for

354

cations. This means that these monohydrates are expected to replace water in the hydration

355

shell, and in the case of glyoxal, replace the water molecules in the hydration shell surrounding

356

NO3- and SO42- quite significantly. We use the equation ∆G=-RTlnK to calculate the equilibrium

357

constant for the displacement of a water molecule with an organic, where R is the ideal gas

358

constant in J/mol K, T is the temperature in Kelvin, and K is the dimensionless equilibrium

359

constant. The ∆G values listed in Table 2 result in equilibrium constants between 88 (Cl-) and

16 ACS Paragon Plus Environment

Page 17 of 29

Environmental Science & Technology

360

12000 (SO42-) for glyoxal and between 4 (Cl-) and 67 (NO3-) for methylglyoxal. However, the

361

organics we have studied do not significantly displace water molecules in the hydration shells of

362

the cations, as the equilibrium constants for these replacements are around 0.5 or less, which

363

means that only water molecules in the hydration shell is favored. The fact that the quantum

364

calculations predict replacement of water in the hydration shell for both glyoxal and

365

methylglyoxal, while both molecules behave quite different in terms of their overall salt

366

interaction indicates that other factors affect the interaction between the organics and the salt.

367

However, we note that the uncertainty on the quantum calculations is likely ±8 kJ/mol. In the

368

case of glyoxal, this results in an equilibrium constant greater than 1 within the entire range of

369

the ∆G calculations and thus glyoxal is always predicted to salt in. In the case of methylglyoxal,

370

the equilibrium constant can be slightly less than 1 for both Cl- and SO42- at the far end of the ∆G

371

range, allowing for the possibility of both salting in (all salts) and salting out (for Cl- and SO42-).

372

Thus weak salting in is within the error bars of the salting constant predictions for interactions

373

between methylglyoxal and these two anions. However, the trends predicted by the quantum

374

calculations also differ from those observed experimentally. This indicates that the reactions

375

represented by the calculations as well as other factors that are not yet well understood determine

376

the overall salting behavior of organic molecules. Other known factors that have been associated

377

to explain salting are ion charge and size effects47 as well as the dielectric properties and size of

378

the organic molecule.48

379

Quantum calculations can help assess whether salting in and out is likely to occur, but

380

should be used in combination with additional factors, such as the ion charge, and volume effects

381

that also need to be considered. Additional measurements need to be made for other water-

382

soluble organic molecules such as IEPOX and methyl tetrol, and further work on mixed salt

17 ACS Paragon Plus Environment

Environmental Science & Technology

Page 18 of 29

383

solutions should be performed to confirm whether Eq (5) presents a good approximation over a

384

wider parameter space of mixed salt solutions.

385 386

4. Atmospheric Implications

387

The measured salting constants for glyoxal and methylglyoxal have implications for SOA

388

formation: at the high salt concentrations found in atmospheric aerosols, more glyoxal will

389

partition in to the aerosols than expected by just Henry’s law, while less methylglyoxal will

390

partition to the aerosols. This effect acts on top of the ~13 times higher effective Henry’s law

391

constant for glyoxal, and makes glyoxal relatively more available for further processing and

392

aerosol formation, but methylglyoxal less so.

393

Additionally, there is some evidence that salting constants are additive, both for benzene

394

in a variety of salts49 and for glyoxal in (NH4)2SO4 and NH4NO3. Glyoxal’s behavior in mixed

395

(NH4)2SO4/NH4NO3 aerosol can be calculated from Equation 5. This has significant implications

396

for predicting salting behavior in ambient aerosols as those typically contain a mixture of salts.1

397

For atmospheric modeling purposes we recommend using Eq. (5) to represent salting behavior

398

based on the molality of sulfate, nitrate, and chloride.

399 400

Acknowledgements

401

This work was supported by NSF-EAGER grant AGS-1452317 (RV). EMW is the recipient of a

402

CIRES Graduate Research Fellowship. JE and KVM thank the Danish Center for Scientific

403

Computing for computational resources and the Center for Exploitation of Solar Energy,

404

Department of Chemistry, University of Copenhagen, Denmark for financial support.

405

Furthermore, JE thanks and the Carlsberg Foundation for financial support.TK thanks the

18 ACS Paragon Plus Environment

Page 19 of 29

Environmental Science & Technology

406

Academy of Finland for funding and the CSC IT Centre for Science for computer time. PJZ was

407

supported by NSF grant AGS-1219508. We would like to thank A. Ranney and M. Claflin for

408

technical assistance, and an anonymous reviewer for providing Abraham solvation parameters

409

among other helpful comments.

410

Supplemental Information: Supplemental information is available and contains two figures:

411

one of representative ITMS signals for methylglyoxal measurements and one of the lowest

412

energy methylglyoxal conformations and two tables: one for measurement conditions and one

413

for salting constant values in molarity as well as molality, and additional details about the

414

quantum chemical calculations. This information is available free of charge via the Internet at

415

http://pubs.acs.org/ .

416

References

417 418 419 420

(1)

Zhang, Q.; Jimenez, J. L.; Canagaratna, M. R.; Allan, J. D.; Coe, H.; Ulbrich, I.; Alfarra, M. R.; Takami, A.; Middlebrook, A. M.; Sun, Y. L.; et al. Ubiquity and Dominance of Oxygenated Species in Organic Aerosols in Anthropogenically-Influenced Northern Hemisphere Midlatitudes. Geophys. Res. Lett. 2007, 34 (13), L13801.

421 422 423 424

(2)

Philip, S.; Martin, R. V; van Donkelaar, A.; Lo, J. W.-H.; Wang, Y.; Chen, D.; Zhang, L.; Kasibhatla, P. S.; Wang, S.; Zhang, Q.; et al. Global Chemical Composition of Ambient Fine Particulate Matter for Exposure Assessment. Environ. Sci. Technol. 2014, 48 (22), 13060–13068.

425 426 427

(3)

Nguyen, T. K. V.; Petters, M. D.; Suda, S. R.; Guo, H.; Weber, R. J.; Carlton, A. G. Trends in Particle-Phase Liquid Water during the Southern Oxidant and Aerosol Study. Atmos. Chem. Phys. 2014, 14 (20), 10911–10930.

428 429 430

(4)

Wang, X.; Ye, X.; Chen, H.; Chen, J.; Yang, X.; Gross, D. S. Online Hygroscopicity and Chemical Measurement of Urban Aerosol in Shanghai, China. Atmos. Environ. 2014, 95, 318–326.

431 432 433

(5)

Herrmann, H.; Schaefer, T.; Tilgner, A.; Styler, S. A.; Weller, C.; Teich, M.; Otto, T. Tropospheric Aqueous-Phase Chemistry: Kinetics, Mechanisms, and Its Coupling to a Changing Gas Phase. Chem. Rev. 2015, 115 (10), 4259–4334.

19 ACS Paragon Plus Environment

Environmental Science & Technology

Page 20 of 29

434 435 436 437

(6)

Kampf, C. J.; Waxman, E. M.; Slowik, J. G.; Dommen, J.; Pfaffenberger, L.; Praplan, A. P.; Prevot, A. S. H.; Baltensperger, U.; Hoffmann, T.; Volkamer, R. Effective Henry’s Law Partitioning and the Salting Constant of Glyoxal in Aerosols Containing Sulfate. Environ. Sci. Technol. 2013, 47 (9), 4236–4244.

438 439 440 441

(7)

Waxman, E. M.; Dzepina, K.; Ervens, B.; Lee-Taylor, J.; Aumont, B.; Jimenez, J. L.; Madronich, S.; Volkamer, R. Secondary Organic Aerosol Formation from Semi- and Intermediate-Volatility Organic Compounds and Glyoxal: Relevance of O/C as a Tracer for Aqueous Multiphase Chemistry. Geophys. Res. Lett. 2013, 40 (5), 978–982.

442 443 444 445

(8)

De Haan, D. O.; Corrigan, A. L.; Smith, K. W.; Stroik, D. R.; Turley, J. J.; Lee, F. E.; Tolbert, M. A.; Jimenez, J. L.; Cordova, K. E.; Ferrell, G. R. Secondary Organic AerosolForming Reactions of Glyoxal with Amino Acids. Environ. Sci. Technol. 2009, 43 (8), 2818–2824.

446 447 448

(9)

De Haan, D. O.; Corrigan, A. L.; Tolbert, M. A.; Jimenez, J. L.; Wood, S. E.; Turley, J. J. Secondary Organic Aerosol Formation by Self-Reactions of Methylglyoxal and Glyoxal in Evaporating Droplets. Environ. Sci. Technol. 2009, 43 (21), 8184–8190.

449 450 451 452

(10)

De Haan, D. O.; Hawkins, L. N.; Kononenko, J. A.; Turley, J. J.; Corrigan, A. L.; Tolbert, M. A.; Jimenez, J. L. Formation of Nitrogen-Containing Oligomers by Methylglyoxal and Amines in Simulated Evaporating Cloud Droplets. Environ. Sci. Technol. 2010, 45 (3), 984–991.

453 454 455

(11)

Hawkins, L. N.; Baril, M. J.; Sedehi, N.; Galloway, M. M.; De Haan, D. O.; Schill, G. P.; Tolbert, M. A. Formation of Semisolid, Oligomerized Aqueous SOA: Lab Simulations of Cloud Processing. Environ. Sci. Technol. 2014, 48, 2273–2280.

456 457 458

(12)

Powelson, M. H.; Espelien, B. M.; Hawkins, L. N.; Galloway, M. M.; De Haan, D. O. Brown Carbon Formation by Aqueous-Phase Carbonyl Compound Reactions with Amines and Ammonium Sulfate. Environ. Sci. Technol. 2013, 48 (2), 985–993.

459 460 461 462

(13)

Galloway, M. M.; Chhabra, P. S.; Chan, A. W. H.; Surratt, J. D.; Flagan, R. C.; Seinfeld, J. H.; Keutsch, F. N. Glyoxal Uptake on Ammonium Sulphate Seed Aerosol: Reaction Products and Reversibility of Uptake under Dark and Irradiated Conditions. Atmos. Chem. Phys. 2009, 9 (10), 3331–3345.

463 464

(14)

Woo, J. L.; Kim, D. D.; Schwier, A. N.; Li, R.; McNeill, V. F. Aqueous Aerosol SOA Formation: Impact on Aerosol Physical Properties. Faraday Discuss. 2013, 165, 357–367.

465 466 467

(15)

Sareen, N.; Schwier, A. N.; Shapiro, E. L.; Mitroo, D.; McNeill, V. F. Secondary Organic Material Formed by Methylglyoxal in Aqueous Aerosol Mimics. Atmos. Chem. Phys. 2010, 10 (3), 997–1016.

20 ACS Paragon Plus Environment

Page 21 of 29

Environmental Science & Technology

468 469 470

(16)

Shapiro, E. L.; Szprengiel, J.; Sareen, N.; Jen, C. N.; Giordano, M. R.; McNeill, V. F. Light-Absorbing Secondary Organic Material Formed by Glyoxal in Aqueous Aerosol Mimics. Atmos. Chem. Phys. 2009, 9 (7), 2289–2300.

471 472 473

(17)

Dzepina, K.; Cappa, C. D.; Volkamer, R. M.; Madronich, S.; DeCarlo, P. F.; Zaveri, R. A.; Jimenez, J. L. Modeling the Multiday Evolution and Aging of Secondary Organic Aerosol During MILAGRO 2006. Environ. Sci. Technol. 2011, 45 (8), 3496–3503.

474 475 476

(18)

Volkamer, R. M.; Martini, F. S.; Molina, L. T.; Salcedo, D.; Jimenez, J. L.; Molina, M. J. A Missing Sink for Gas-Phase Glyoxal in Mexico City: Formation of Secondary Organic Aerosol. Geophys. Res. Lett. 2007, 34 (19), L19807.

477 478 479

(19)

Volkamer, R. M.; Ziemann, P. J.; Molina, M. J. Secondary Organic Aerosol Formation from Acetylene (C2H2): Seed Effect on SOA Yields due to Organic Photochemistry in the Aerosol Aqueous Phase. Atmos. Chem. Phys. 2009, 9 (6), 1907–1928.

480 481 482

(20)

Ervens, B.; Volkamer, R. Glyoxal Processing by Aerosol Multiphase Chemistry: Towards a Kinetic Modeling Framework of Secondary Organic Aerosol Formation in Aqueous Particles. Atmos. Chem. Phys. 2010, 10 (17), 8219–8244.

483 484 485

(21)

Carlton, A. G.; Turpin, B. J.; Altieri, K. E.; Seitzinger, S.; Reff, A.; Lim, H.-J.; Ervens, B. Atmospheric Oxalic Acid and SOA Production from Glyoxal: Results of Aqueous Photooxidation Experiments. Atmos. Environ. 2007, 41 (35), 7588–7602.

486 487 488 489

(22)

Ortiz-Montalvo, D. L.; HäkkinenS.A.K.; Schwier, A. N.; Lim, Y. B.; McNeill, V. F.; Turpin, B. J. Ammonium Addition (and Aerosol pH) Has a Dramatic Impact on the Volatility and Yield of Glyoxal Secondary Organic Aerosol. Environ. Sci. Technol. 2014, 48, 255–262.

490 491 492 493

(23)

Surratt, J. D.; Chan, A. W. H.; Eddingsaas, N. C.; Chan, M.; Loza, C. L.; Kwan, A. J.; Hersey, S. P.; Flagan, R. C.; Wennberg, P. O.; Seinfeld, J. H. Reactive Intermediates Revealed in Secondary Organic Aerosol Formation from Isoprene. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 6640–6645.

494 495 496 497

(24)

Xu, L.; Guo, H.; Boyd, C. M.; Klein, M.; Bougiatioti, A.; Cerully, K. M.; Hite, J. R.; Isaacman-VanWertz, G.; Kreisberg, N. M.; Knote, C.; et al. Effects of Anthropogenic Emissions on Aerosol Formation from Isoprene and Monoterpenes in the Southeastern United States. Proc. Natl. Acad. Sci. U. S. A. 2015, 112 (1), 37–42.

498 499

(25)

Setschenow, J. Über Die Konstitution Der Salzlösungen Auf Grund Ihres Verhaltens Zu Kohlensäure. Z. Phys. Chem. 1889, Vierter Ba (1), 117–125.

500 501 502

(26)

Kurtén, T.; Elm, J.; Prisle, N. L.; Mikkelsen, K. V; Kampf, C. J.; Waxman, E. M.; Volkamer, R. Computational Study of the Effect of Glyoxal-Sulfate Clustering on the Henry’s Law Coefficient of Glyoxal. J. Phys. Chem. A 2014, 119 (19), 4509–4514.

21 ACS Paragon Plus Environment

Environmental Science & Technology

Page 22 of 29

503 504

(27)

Fratzke, A. R.; Reilly, P. J. Thermodynamic and Kinetic Analysis of the Dimerization of Aqueous Glyoxal. Int. J. Chem. Kinet. 1986, 18 (7), 775–789.

505 506 507

(28)

Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A. Gaussian 09, revision B. 01; Gaussian Inc: Wallingford, CT, 2010.

508 509 510 511

(29)

Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States, and Transition Elements: Two New Functionals and Systematic Testing of Four M06Class Functionals and 12 Other Function. Theor. Chem. Acc. 2007, 120 (1), 215–241.

512 513 514

(30)

Elm, J.; Bilde, M.; Mikkelsen, K. V. Assessment of Density Functional Theory in Predicting Structures and Free Energies of Reaction of Atmospheric Prenucleation Clusters. J. Chem. Theory Comput. 2012, 8 (6), 2071–2077.

515 516

(31)

Elm, J.; Bilde, M.; Mikkelsen, K. V. Assessment of Binding Energies of Atmospherically Relevant Clusters. Phys. Chem. Chem. Phys. 2013, 15 (39), 16442–16445.

517 518 519

(32)

Leverentz, H. R.; Siepmann, J. I.; Truhlar, D. G.; Loukonen, V.; Vehkamäki, H. Energetics of Atmospherically Implicated Clusters Made of Sulfuric Acid, Ammonia, and Dimethyl Amine. J. Phys. Chem. A 2013, 117 (18), 3819–3825.

520 521 522

(33)

Miertuš, S.; Scrocco, E.; Tomasi, J. Electrostatic Interaction of a Solute with a Continuum. A Direct Utilizaion of AB Initio Molecular Potentials for the Prevision of Solvent Effects. Chem. Phys. 1981, 55 (1), 117–129.

523 524

(34)

Tomasi, J.; Mennucci, B.; Cammi, R. Quantum Mechanical Continuum Solvation Models. Chem. Rev. 2005, 105 (8), 2999–3093.

525 526

(35)

Nemet, I.; Vikić-Topić, D.; Varga-Defterdarović, L. Spectroscopic Studies of Methylglyoxal in Water and Dimethylsulfoxide. Bioorg. Chem. 2004, 32 (6), 560–570.

527

(36)

Simons, J. Molecular Anions. J. Phys. Chem. A 2008, 112 (29), 6401–6511.

528 529

(37)

Wang, X.-B.; Nicholas, J. B.; Wang, L.-S. Electronic Instability of Isolated SO42- and Its Solvation Stabilization. J. Chem. Phys. 2000, 113 (24), 10837–10840.

530 531 532 533

(38)

Liu, Y.-R.; Wen, H.; Huang, T.; Lin, X.-X.; Gai, Y.-B.; Hu, C.-J.; Zhang, W.-J.; Huang, W. Structural Exploration of Water, Nitrate/water, and Oxalate/water Clusters with BasinHopping Method Using a Compressed Sampling Technique. J. Phys. Chem. A 2014, 118 (2), 508–516.

534 535 536

(39)

Nadykto, A. B.; Yu, F.; Natsheh, A. Al. Anomalously Strong Effect of the Ion Sign on the Thermochemistry of Hydrogen Bonded Aqueous Clusters of Identical Chemical Composition. Int. J. Mol. Spec. 2009, 10, 507–517. 22 ACS Paragon Plus Environment

Page 23 of 29

Environmental Science & Technology

537 538 539 540

(40)

Pickard, F. C.; Dunn, M. E.; Shields, G. C. Comparison of Model Chemistry and Density Functional Theory Thermochemical Predictions with Experiment for Formation of Ionic Clusters of the Ammonium Cation Complexed with Water and Ammonia; Atmospheric Implications. J. Phys. Chem. A 2005, 109 (22), 4905–4910.

541 542 543

(41)

Lambrecht, D. S.; Clark, G. N. I.; Head-Gordon, T.; Head-Gordon, M. Exploring the Rich Energy Landscape of Sulfate-Water Clusters SO4(2-) (H2O)(n=3-7): An Electronic Structure Approach. J. Phys. Chem. A 2011, 115 (41), 11438–11454.

544 545

(42)

Wang, C.; Lei, Y. D.; Endo, S.; Wania, F. Measuring and Modeling the Salting-out Effect in Ammonium Sulfate Solutions. Environ. Sci. Technol. 2014, 48 (22), 13238–13245.

546 547 548

(43)

Endo, S.; Pfennigsdorff, A.; Goss, K. U. Salting-Out Effect in Aqueous NaCl Solutions Increases with Size and Decreases with Polarities of Solute Molecule. Environ. Sci. Technol. 2012, 46 (3), 1496–1503.

549 550

(44)

Zhang, Y.; Cremer, P. S. Interactions between Macromolecules and Ions: The Hofmeister Series. Curr. Opin. Chem. Biol. 2006, 10 (6), 658–663.

551 552 553

(45)

Gordon, J. E.; Thorne, R. L. Salt Effects on the Activity Coefficient of Naphthalene in Mixed Aqueous Electrolyte Solutions. I. Mixtures of Two Salts. J. Phys. Chem. 1967, 71 (13), 4390–4399.

554 555

(46)

Görgényi, M.; Dewulf, J.; Van Langenhove, H.; Héberger, K. Aqueous Salting-out Effect of Inorganic Cations and Anions on Non-Electrolytes. Chemosphere 2006, 65, 802–810.

556 557

(47)

Debye, P.; McAulay, J. Das Elektrische Feld Der Ionen Und Die Neutralsalzwirkung. Phys. Zeitschrift 1925, 26, 22–29.

558 559

(48)

McDevit, W. F.; Long, F. A. The Activity Coefficient of Benzene in Aqueous Salt Solutions. Jour. Am. Chem. Soc. 1952, 74 (7), 1773–1777.

560 561 562

(49)

Gordon, J. E.; Thorne, R. L. Salt Effects on the Activity Coefficient of Naphthalene in Mixed Aqueous Electrolyte Solutions. I. Mixtures of Two Salts. J. Phys. Chem. 1967, 71 (13), 4390–4399.

563 564

23 ACS Paragon Plus Environment

Environmental Science & Technology

Molecule

KS (NH4)2SO4

KS NH4NO3

Glyoxal

-0.24±0.02 M-1 -0.16±0.02 M-1

--0.06±0.02 M-1

Methylglyoxal

0.16±0.03 M-1

0.075±0.03 M-1

KS NaNO3

Page 24 of 29

KS NaCl

Reference

---0.065±0.006 M-1 -0.10±0.02M-1 0.02±0.02 M-1

0.06±0.02 M-1

Kampf (2013)6 this work this work

-1

-1

2-hexanone

0.38±0.01 M --

---

---

0.18±0.02 M Wang (2014)42 -1 0.198±0.004 M Endo (2012)43

heptanal

0.35±0.03 M-1 --

---

---

1-hexanol

0.364±0.006 M-1 --

---

---

0.21±0.01 M-1 Wang (2014)42 0.221±0.004 M-1 Endo (2012)43

2-butoxyethanol

0.31±0.02 M-1 --

---

---

-Wang (2014)42 -1 0.211±0.009 M Endo (2012)43

-0.24±0.01 M-1

Wang (2014)42 Endo (2012)43

565

Table 1: Compilation of KS values for small oxygenated species in aerosol-relevant salts (all in

566

molality-1). Good agreement is observed between the literature value and the value measured in

567

this work for glyoxal in (NH4)2SO4 despite the different salt regimes and different methods used

568

for measuring the gas and aqueous phase concentrations. All species except for glyoxal are

569

determined to salt out of solution.

570

24 ACS Paragon Plus Environment

Page 25 of 29

Environmental Science & Technology

Ion

∆G (kJ/mol)

ClNO3SO42Na+ NH4+

-10.9 -22.0 -22.9 2.1 1.2

ClNO3SO42Na+ NH4+

-3.1 -10.3 -7.91 6.11 1.6

∆G range (kJ/mol) Glyoxal -2.6 to -19.3 -13.6 to -30.4 -14.6 to -31.3 10.5 to -6.28 9.54 to -7.20 Methylglyoxal 5.23 to -11.5 -1.9 to -18.6 0.46 to -16.3 14.5 to -2.3 9.96 to -6.78

K

K range

88 8400 12000 0.42 0.62

2.8 to 2700 270 to 260000 390 to 380000 0.014 to 13 0.020 to 19

3.6 67 26 0.081 0.52

0.12 to 110 2.2 to 2100 0.83 to 800 0.0026 to 2.5 0.017 to 16

571 572

Table 2: Calculated ∆G, ∆G range assuming ±8 kJ/mol, K, K range assuming ±8 kJ/mol for

573

glyoxal and methylglyoxal. ∆G values are calculated for the replacement of a water molecule in

574

the hydration shell of the anion in the Ion column by either a glyoxal monohydrate molecule or a

575

methylglyoxal monohydrate molecule. The corresponding K values are calculated from the

576

equation ∆G=-RTlnK.

577

25 ACS Paragon Plus Environment

Environmental Science & Technology

Page 26 of 29

Figure 1: A) Salting constant inlet for mass spectrometer. Solutions of known organic and salt concentration are added to graduated cylinder. System is pumped down to remove oxygen and nitrogen. Water and organics leak through the pinhole to create an effusion regime and then enter the mass spectrometer. B) Mass spectrometric proof of methyl glyoxal detection in EI mode. The standard NIST EI spectrum of methyl glyoxal is shown on the top in black and the spectrum recorded in the lab is shown on the bottom in red with air and water subtracted out. The very good agreement between the spectra is proof of methyl glyoxal detection. C) Schematic of SPME and GC-FID measurements. SPME fiber samples the headspace above the solution of organic and salt, and results in a two-peak chromatogram. 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 27 of 29

Environmental Science & Technology

Figure 2: Salting constants for glyoxal (solid diamonds, dashed fit) and methylglyoxal calculated from both ITMS data (open circles, solid fit) and GC-FID data (open squares, dotted fit). Slopes of data correspond to KS values, which are given in Table 1 and Table S1. Solid grey lines: average of KS from ITMS and GC-FID data for methylglyoxal. Excellent agreement is observed between both methylglyoxal measurements except for NH4NO3. The positive slopes for methylglyoxal indicate that it salts out. The negative slopes for glyoxal indicate that it salts in. 100x110mm (220 x 220 DPI)

ACS Paragon Plus Environment

Environmental Science & Technology

Figure 3: Relationship between KS values for (NH4)2SO4 and carbon number. Measurements from Wang et al.42 are given by red circles (2-ketones) and blue circles (aldehydes); fit the 2-ketone data given by the red dashed line. KS values for glyoxal (squares) and methylglyoxal (triangles) measured in this work (solid black symbols) are compared with predictions from the poly-parameter linear free energy relationship (ppLFER, Wang et al.42) for unhydrated, monohydrate, and dihydrate forms, respectively (open green symbols). The best estimations from the ppLFER are the solid green symbols. For glyoxal, this corresponds to the dihyrate value; for methylglyoxal to a linear combination of 60% monohydrate and 40% dihydrate. 116x102mm (220 x 220 DPI)

ACS Paragon Plus Environment

Page 28 of 29

Page 29 of 29

Environmental Science & Technology

254x190mm (96 x 96 DPI)

ACS Paragon Plus Environment