Size-Resolved Surface-Active Substances of Atmospheric Aerosol

Jul 26, 2018 - ACS eBooks; C&EN Global Enterprise .... Experimental Aerosol and Cloud Microphysics, Leibniz Institute for Tropospheric ... To assess t...
0 downloads 0 Views 1MB Size
Subscriber access provided by Kaohsiung Medical University

Environmental Processes

Size-resolved surface active substances of atmospheric aerosol: reconsideration of the impact on cloud droplet formation Ana Krofli#, Sanja Frka, Martin Simmel, Heike Wex, and Irena Grgi# Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b02381 • Publication Date (Web): 26 Jul 2018 Downloaded from http://pubs.acs.org on July 26, 2018

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 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 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.

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

Size-resolved surface active substances of atmospheric aerosol: reconsideration of the

2

impact on cloud droplet formation

3 Ana Kroflič†┴║, Sanja Frka†‡║*, Martin Simmel§, Heike Wex± and Irena Grgić†

4 5 †

6

Department of Analytical Chemistry, National Institute of Chemistry, Ljubljana, 1000,

7

Slovenia; ┴

8

Atmospheric Chemistry, Leibniz Institute for Tropospheric Research, Leipzig, 04318,

9 10

Germany; ‡

Division for Marine and Environmental Research, Ruđer Bošković Institute, Zagreb, 10000,

11 12

Croatia; §

Modelling of Atmospheric Processes, Leibniz Institute for Tropospheric Research, Leipzig,

13 14

04318, Germany; ±

Experimental Aerosol and Cloud Microphysics, Leibniz Institute for Tropospheric Research,

15

Leipzig, 04318, Germany

16 17



Co-first authors contributed equally to this work

18 19

*Corresponding author

20

Tel: +385 1 4561 185

21

Fax: +385 1 4680 242

22

Email address: [email protected]

23

1 ACS Paragon Plus Environment

Environmental Science & Technology

24

Abstract

25

Our current understanding of the importance of surface active substances (SAS) on

26

atmospheric aerosol cloud-forming efficiency is limited, as explicit data on the content of

27

size-resolved ambient aerosol SAS, which are responsible for lowering the surface tension (σ)

28

of activating droplets, are not available. We report on the first data comprising seasonal

29

variability of size-segregated SAS concentrations in ambient aerosol particulate matter (PM).

30

To assess the impact of SAS distribution within PM on cloud droplet activation and growth, a

31

concept of surfactant activity was adopted and a parametrization developed, i.e. surfactant

32

activity factor (SAF) was defined, which allowed translation of experimental data for use in

33

cloud parcel modeling. Results show that SAS-induced σ depression during cloud activation

34

may affect droplet number (Nd) as much as a two-fold increase in particle number, whereas by

35

considering also the size distribution of particulate SAS, Nd may increase for another 10%.

36

This study underscores the importance of size-resolved SAS perspective on cloud activation,

37

as data typically obtained from aqueous extracts of PM2.5 and PM10 may result in misleading

38

conclusions about droplet growth due to large mass fractions of supermicron particles with

39

SAS deficit and little or no influence on CCN and Nd.

40 41

Introduction

42

One of the major uncertainties in the current understanding of the Earth’s climate is the effect

43

that aerosol particles have on cloud formation processes and the resulting impact on the

44

radiative balance.1,2 The ability of an aerosol particle to act as a cloud condensation nucleus

45

(CCN) and to become a cloud droplet (i.e., its cloud-forming efficiency) is described by the

46

Köhler theory.3 The theory accounts for a competition between the Kelvin term, which

47

considers also the surface tension of particulate aqueous phase (σ), and the Raoult term,

48

which links the surrounding water vapor pressure to the solute content. The interplay between 2 ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29

Environmental Science & Technology

49

both effects results in a size- and composition-dependent growth of hydrated particles/droplets

50

in humid air and only a fraction of hygroscopic particles competing for the available water

51

vapor in the surrounding can grow beyond their critical sizes for spontaneous growth at a

52

given supersaturation. As larger particles nucleate easier than small ones, the effect of

53

chemical composition on the cloud-forming efficiency is most significant in the submicron

54

size-range particles. Coupling with inorganic components,4–6 both model and natural organic

55

surface active substances (SAS) or surfactants have been shown to lower σ of atmospheric

56

cloud, rain, and fog waters and aqueous extracts of atmospheric aerosol from various

57

sources.7–13 By influencing the Kelvin effect, σ depression caused by SAS could promote

58

cloud droplet formation and importantly affect cloud properties such as albedo.14 On the other

59

hand, it is believed that displacement of SAS from the droplet bulk to the particle/droplet

60

surface simultaneously attributes to a change in the Raoult term (opposing effect as from the

61

σ depression) which also has to be accounted for.15,16 In a recent publication, however, a

62

closure between chemical composition, growth factor measurements and CCN activity could

63

not be achieved under particular conditions which was explained by a lack in accounting for

64

the σ reduction of aerosol aqueous phase.17 Although diverse modeling efforts have been

65

made to assess the influence of the observed σ depression on cloud droplet nucleation and

66

growth, the magnitude and even the sign of the cumulative effect of SAS on cloud droplet

67

number (Nd) are still subjects of controversy.16,18–23 The direct experimental evidence for σ

68

depression in microscopic droplets remains elusive. Nozière et al. (2014) examined the

69

dynamic development of surface tension in droplets containing SAS, and demonstrated that

70

the effect of SAS on σ of growing droplets cannot be directly measured due to limitations of

71

state-of-the-art instrumentation for studying cloud droplet formation which is unable to detect

72

kinetically governed σ effects of aerosol SAS on CCN activity.24 By applying an alternative

73

approach, Ruehl et al. (2016) measured the size of haze droplets up to the point of cloud 3 ACS Paragon Plus Environment

Environmental Science & Technology

74

droplet activation and beyond, for particles containing SAS.25 The observed droplet sizes at

75

activation being larger than expected by the Köhler theory are explained by organics

76

accumulated on the surface of haze droplets, forming a compressed film which reduces σ and

77

allows for larger droplets to form prior to their activation. Gérard et al. (2016) further

78

measured σ of isolated SAS from atmospheric aerosol samples at a range of dilutions,

79

discussing that atmospheric haze droplets prior to activation still have the σ clearly lower than

80

that of pure water.26 Nevertheless, a widely used parameterization in atmospheric models

81

known as the κ-Köhler theory27 describing CCN activation ignores the influence of SAS by

82

assuming that the σ of growing droplets is that of pure water, such that the experimentally

83

derived hygroscopicity parameter κ correctly describes particle activation when used together

84

with this assumption. Therefore, a more rigorous exploration of the content of ambient SAS

85

and the conditions under which they exert an effect on droplet activation are urgently needed

86

to elucidate their true impact on cloud-formation processes and global indirect aerosol

87

forcing. Here, the particular examination of SAS distribution in particles of submicron size

88

ranges, considered as critical for CCN activation, is explicitly addressed,17,26 because SAS

89

might occur enriched at these sizes, compared to the bulk samples.

90

Until now no explicit data on the size-resolved SAS in ambient particulate matter (PM) have

91

been published. Former efforts were limited to measuring SAS in various bulk aerosol water

92

extracts,28–36 including those that applied pre-concentrating solid-phase extraction (SPE)

93

protocols26,37,38 used at the same time for the isolation and concentration of aerosol HUmic-

94

LIke Substances (HULIS)e.g.,13. This is mainly due to the sensitivity limitations of classical

95

methods used for low-level SAS quantification in atmospheric samples. Notwithstanding, the

96

herein applied electrochemical method (phase sensitive alternating current (ac) voltammetry;

97

out of phase mode) has been proven to be sensitive and directly applicable for the qualitative

98

and quantitative characterization of SAS in various natural aqueous systems,e.g.,39–42 including 4 ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29

Environmental Science & Technology

99

bulk precipitation43,44 and aqueous aerosol extracts,34 serving as a platform for low-level SAS

100

determination in aqueous extracts of size-segregated aerosols.

101

We report the first data on size-segregated SAS concentrations in ambient aerosol particles

102

with aerodynamic diameters between 0.04 and 15.6 µm, divided into nine size bins. For cloud

103

modeling purposes, novel parametrization was developed which allowed us to account for σ

104

effects on droplet formation in a size-resolved manner. For this purpose, size-resolved

105

surfactant activity factors (SAF) were derived from the electrochemically determined size-

106

segregated SAS, which were originally considered dissolved within the adsorbed water layer

107

on the surface of the corresponding size-segregated particle population (i.e., within the water

108

shell of hydrated particles/droplets in the particular size bin). SAS distribution within PM was

109

obtained by combining size-resolved SAF with the measured PM mass size distribution. Size-

110

segregated proxy data for σ were estimated according to Gérard et al.,26 the only report on the

111

σ curves for the natural SAS of ambient PM2.5 samples. Finally, a cloud parcel model was

112

applied to inspect size-resolved σ effects of ambient SAS on the kinetics of droplet growth

113

and activation.

114 115

EXPERIMENTAL SECTION

116 117

Aerosol sampling

118

Seasonal size-segregated and PM10 aerosol samples were collected in parallel at an urban

119

background site of Ljubljana, Slovenia (approx. 279,000 inhabitants, 298 m a.s.l.). Size-

120

segregated aerosols were collected on pre-baked aluminum foils (500 °C for 24 h) with a 10-

121

stage low-pressure Berner cascade impactor (HAUKE, LPI 25/0,015/2) operated at ambient

122

temperature at a flow rate of 25.8 L min−1. 2–3 consecutive 48–72 h samples were combined

123

into each investigated sample set to ensure at least 90 µg of PM deposit for each size fraction 5 ACS Paragon Plus Environment

Environmental Science & Technology

124

as described elsewhere.45 In addition to ensure the sufficient amount of material for chemical

125

analyses, stages 1 and 2 of each sample set were processed together (called stage 2a herein).

126

In total, we examined four impactor sample sets (one for each season: winter (14/02–

127

20/02/2015), spring (15/05–25/05/2015), summer (10/08–21/08/2015) and autumn (05/11–

128

11/11/2015), each comprising nine size-segregated stages, representing the average of 6 to 12

129

days. In parallel, PM10 samples were collected onto prebaked glass fiber filters (0.47 mm;

130

Whatman) for 24 h at 2.3 m3 h−1 using a Leckel sampler. In each season (winter 02/2015;

131

spring 05/2015; summer 08/2015; autumn 11/2015), 6 daily PM10 samples were processed

132

together. The collected samples and the foil/filter blanks (no air drawn through the

133

impactor/sampler) were stored at −18 °C until analysis.

134 135

PM size distribution

136

The impactor comprises 10 stages of nominal aerosol size ranges expressed in aerodynamic

137

equivalent diameter, d; however, as already explained, stages 1 and 2 were combined (stage

138

2a: 0.038–0.104 µm; stage 3: 0.104–0.16 µm, stage 4: 0.16–0.305 µm, stage 5: 0.305–0.56

139

µm, stage 6: 0.56–1.01 µm, stage 7: 1.01–2.1 µm, stage 8: 2.1–3.99 µm; stage 9: 3.99–8.06

140

µm, stage 10: 8.06–15.6 µm). Aerosol mass of each stage of size-segregated aerosol was

141

determined by weighing the sample before and after the sampling; prior to weighing, all

142

samples were conditioned for at least 24 h at a relative humidity of 50±5% and a temperature

143

of 20±1 °C. For numerical modeling purposes (see below), discrete size-bin PM mass

144

distributions (Figure 1a) were inverted to continuous size distributions and the model was

145

initialized with the continuous PM size distributions of the respective seasons shown in

146

Figure 1b. For details about the applied numerical treatment see Supporting Information (SI),

147

Appendix 1.

148 6 ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29

Environmental Science & Technology

149

Aerosol pretreatment

150

SAS of size-segregated aerosol and PM10 samples were concentrated by adopting a standard

151

solid phase extraction (SPE) protocol for the isolation of aerosol HULIS.26,46,47 Aqueous PM

152

extracts were first obtained by extraction of joined samples in 15–24 mL of high purity

153

reagent Milli-Q water (Millipore, Bedford, MA, USA) for 24 h at 4 ºC, followed by filtration

154

through a 0.22 µm pore size filter (Supelco, USA). Aqueous extracts were further treated

155

according to the SPE procedure steps on C18 columns (SEP-PAK VAC, 3 mL, 500 mg,

156

Waters; see SI, Appendix 3) to obtain MeOH isolates. Aqueous aerosol extracts and HULIS

157

extracts (obtained after evaporation of MeOH extracts to dryness and redissolution in Milli-Q

158

water) were further analyzed for their water-soluble organic carbon (WSOC and HULIS-C,

159

respectively) and SAS content. Blank filters were also subjected to the same processing steps

160

to ensure appropriate extraction procedure and control the purity of final extracts.

161

SAS extraction efficiency. SAS extraction yield was determined as a ratio between SAS

162

concentrations (in equiv. of model surfactant T-X-100) in aerosol aqueous extracts after and

163

before the C18 SPE procedure. The applied procedure enabled high SAS yields (nearly

164

100%), indicating their complete isolation from the aerosol water-soluble organic matter (SI,

165

Appendix 3).

166 167

Water-soluble organic carbon analysis

168

WSOC and HULIS-C were analyzed by sensitive Combustion TOC Analyzer (Teledyne,

169

Apollo 9000 HS) with a non-dispersive infrared gas detector (NDIR). In order to eliminate the

170

inorganic carbonates, each sample was acidified with 2 M HCl to pH 2–3 before analysis.

171

WSOC or HULIS-C of each sample was calculated as an average of three to five replicate

172

measurements.

173 7 ACS Paragon Plus Environment

Environmental Science & Technology

174

SAS analysis

175

Electrochemical measurements. The phase sensitive ac voltammetry was applied for SAS

176

analysis, being already used as a successful tool for the determination of SAS in a plethora of

177

environmental aquatic samples.39 Measurements (out-of-phase mode, frequency 77 Hz, and

178

amplitude 10 mV) were performed with µAutolab-type II (Eco Chemie B. V., The

179

Netherlands), GPES 4.6 software (Eco Chemie B. V., The Netherlands). A standard

180

polarographic Metrohm cell of 50 cm3 with a three-electrode system was used: working

181

electrode – hanging mercury drop electrode (HMDE; Metrohm, Switzerland; A = 0.01245

182

cm2), reference electrode – Ag/AgCl/3 mol L−1 KCl, auxiliary electrode – platinum coil.

183

Before each measurement, previously purified (450 °C for 5 h; charcoal organic residue

184

removal) saturated NaCl (Kemika, Croatia) solution was added to the sample to adjust the

185

electrolyte concentration to 0.55 mol L−1. In brief, because of their amphiphilic nature, SAS

186

have the capability of spontaneous adsorption at different phase boundaries. Due to the

187

hydrophobic character of the mercury electrode, its interface with aqueous electrolyte solution

188

serves as a good model for the study of SAS.48 SAS adsorption at the surface of the mercury

189

electrode changes its double layer capacitance, which is directly proportional to the measured

190

capacity current (IC) at the applied potential. At the optimal potential for the adsorption of

191

amphiphilic SAS (-0.6 V versus Ag/AgCl), IC decrease (compared to the capacity current of

192

the supporting electrolyte alone (IC0)) is linearly dependent on the adsorbate concentration in

193

aqueous solution, allowing for the precise quantification of low SAS concentrations.

194

Method sensitivity. Interfacial tensiometers and drop shape analyzers are commonly in use for

195

measurements of σ decrease due to the presence of SAS in aqueous solutions. However, due

196

to the insufficient sensitivity, their usefulness for the investigation of ambient PM samples

197

with low-level SAS is limited.e.g.,9,12,13 Taking tetraoctylphenolethoxylate (T-X-100) as typical

198

model SAS, we compared sensitivities of ac voltammetry and tensiometry obtained with a 8 ACS Paragon Plus Environment

Page 8 of 29

Page 9 of 29

Environmental Science & Technology

199

digital tensiometer (Kruss, K10T, GmbH, Hamburg). Ac voltammetry is shown to be two

200

orders of magnitude more sensitive than the conventional tensiometry (Figure S1b, for details

201

refer to SI, Appendix 2).

202

Calibration. Many scientists strive for the quantification of ambient aerosol surfactants,

203

although without knowing their chemical nature or strength (i.e., how surface active they

204

are).49 Traditional model surfactants are often used as surrogate substances for these versatile

205

species;8,50,51 however, particular model surfactants have already been shown to fail to

206

reproduce the overall surface-active properties of aerosol water-soluble fraction.e.g.,52 Some

207

authors have already reported on the potential of more realistic representation of aerosol SAS

208

deduced from real atmospheric samples.13,53 As we show that aerosol SAS fraction is

209

completely isolated within the HULIS material with joint water soluble and hydrophobic

210

properties, HULIS concentration can be regarded as a rough upper-limit estimate of aerosol

211

SAS. Thus, instead of using model surfactants as commonly applied today, calibration with

212

HULIS material isolated from the ambient PM10 samples collected in parallel to Berner

213

impactor samples was performed for accurate quantification of SAS in size-segregated

214

aerosols (details within SI, Appendices 3 and 4). Calibration plots ∆IC vs. concentration of

215

HULIS-C were constructed and are shown in Figure S2. Limits of SAS detection and

216

quantification are 0.1 and 0.16 mgC L−1 (equiv. of HULIS-C), respectively. Relative standard

217

deviation obtained from three to five repetitive analyses of the same solution containing 1

218

mgC L−1 is less than 5%.

219 220

Numerical modeling

221

The spectral microphysics model SPECS describes standard Köhler theory-based aerosol

222

particle growth as well as droplet nucleation and growth with respect to water content using a

223

2-dimensional grid.54 Thus, both, water and particulate mass were divided into discrete bins

9 ACS Paragon Plus Environment

Environmental Science & Technology

224

(360 for particulate mass with mass doubling every 8 bins, 264 for water mass with mass

225

doubling every 4 bins). Dry aerosol was assumed to be an internal mixture where 50% of the

226

particulate mass consisted of ammonium sulfate and 50% were insoluble substances, the latter

227

not contributing to the Raoult term at all. The number of moles of solute was kept constant

228

during the modeling. Surface tension (σ) was iteratively calculated for every time step from

229

the respective dry PM size distribution and the condensed water load (giving us time- and

230

size-resolved PM/water concentrations, c, see details below). Note that SAS bulk-surface

231

partitioning was not accounted for, which could lead to an overestimation of the effect of σ on

232

the modeled droplet growth.15,16 Nevertheless, the main focus of this work is to demonstrate

233

the influence of size-resolved SAS on droplet growth, as compared to assuming a size

234

independent σ. We do not want to elaborate on the absolute effect of σ, as that is still

235

controversial. Recent literature has reported contradictory results, ranging from a possibly

236

large effect of σ on atmospheric CCN concentrations (increase by a factor of 10)17 to cases for

237

which a range of results, including a negligible impact on CCN activation55, has been

238

reported.

239

The model was initialized with the continuous dry PM size distributions of the respective

240

seasons shown in Figure 1b. For the modeling purposes, d was converted into geometric

241

diameters, geo, via the formula: geo = water/AP with particle density AP = 1.5 g cm3

242

and water density water = 1.0 g cm3. Total aerosol number concentrations were 3.9×109,

243

5.4×109, 1.4×1010 and 1.5×1010 m-3 for spring, summer, autumn and winter, respectively.

244

Initially, equilibrium was assumed at a relative humidity (RH) of 99%, together with

245

accounting for a possible σ reduction due to the presence of SAS in the distinct aerosol size

246

fractions. The air parcel was lifted to a height of 250 m, starting at the 99% RH level, with a

247

constant updraft velocity of 0.5 m s−1 as a standard case. The only microphysical process

248

considered was water vapor condensation, i.e. droplet nucleation and further droplet growth 10 ACS Paragon Plus Environment

Page 10 of 29

Page 11 of 29

Environmental Science & Technology

249

by condensation. For the latter, a water accommodation coefficient of 0.5 was taken. The

250

growth calculation was interactively coupled to the vapor phase, meaning that e.g.,

251

supersaturation was explicitly calculated at each time step. Again, we stress that at each time

252

step, σ was also calculated from the condensed water load (see size-dependent σ evolution

253

during the modeling in SI, Appendix 11, Figure S11). Due to the 2D model configuration, the

254

underlying dry aerosol particle distribution remained perfectly conserved throughout the

255

model run while the spectrum referring to water mass evolved.

256

As the determination of the influence of SAS on the σ of aerosol water layer is beyond the

257

scope of this study, experimentally-obtained σ (mN m−1) dependence on PM/water

258

concentration (c in µg µL−1 aqueous solution) was taken from Gerard et al.26 and a model

259

function was fitted to the data (details in SI, Appendix 5, Eq. S2) to obtain the equation used

260

in the aerosol model (Case 1):

261

=

.  

  . 

+ 36.638

(1)

262

Note, cumulative SAS within PM2.5 was considered in the cited work; therefore, as we wanted

263

to also explicitly account for SAS distribution within PM, further data treatment was needed.

264

To account for a particular case of size-resolved SAS described below (Case 2), adapted PM

265

size distributions with considering experimentally-derived surfactant activity factors (SAF)

266

were initialized as well (Figure 1d; obtained according to SI Eq. S3, see Appendix 6) and used

267

only for the modeling of σ in Case 2, i.e. for the calculation of c (the modeled PM size

268

distribution remained unchanged).

269

SAF$ =

%&%' %&% ().* +,' +, ().*

(2)

270

where SASi and PMi are the experimentally determined SAS and PM concentrations of the

271

corresponding size bin i, respectively. Namely, SAF is a weight that corrects σ dependence on

272

c derived from Gerard et al.26 for the size-segregated SAS aspect brought-up in this work, 11 ACS Paragon Plus Environment

Environmental Science & Technology

273

describing the enrichment of SAS in the particular size bin: SAF values >1 represent the

274

fractions enriched with SAS, while values 1 at stages 3 and 4 for all seasons (Figure

395

1c), which resulted in more droplets formed in the case of SEG-SAS in comparison to EVEN-

396

SAS (except in winter). With this we show that averaged SAS data over a broader PM size

397

range (PM2.5 or PM10) are not necessarily representative for the cut-off size range and can

398

result in misleading conclusions about the droplet growth. As an extreme case, consideration

399

of the water-soluble size-segregated SAS distribution expressed on the surface area of particle

400

population (FILM-SAS case; details in SI, Appendix 7) resulted in a relatively low Nd

401

compared to EVEN-SAS and SEG-SAS (SI, Figure S6). Furthermore, a SAS-induced σ

402

decrease during droplet growth affected Nd about as much or even more than the factor-of-two

403

increase in CCN in WAT-2 (Figure 3a). Nenes et al.61 found numerous conditions where the

404

effect of chemical composition on Nd was comparable to the effect of the CCN number. Under

405

marine conditions, the influence of σ accounted for up to 50% of the Twomey effect, whereas

406

under urban conditions it exceeded it. Moreover, for urban aerosols, inclusion of the σ effect

407

exerted by water-soluble organic compounds may increase the CCN number concentration up

408

to 110% relative to the case in which only the inorganic aerosol composition is considered.60

409 410

Atmospheric relevance

411

To date, the influence of SAS on cloud formation has mainly been exploratory and partially

412

speculative, as explicit data on the abundance and size distribution of aerosol SAS,

413

responsible for lowering σ of activating droplets, have been unknown. Electrochemical

414

determination of low concentrations of SAS, together with the developed parametrization

415

enabled us to model droplet activation and growth with considering the size-resolved σ effects

416

on cloud formation. According to our results, SAS-induced σ depression during 18 ACS Paragon Plus Environment

Page 18 of 29

Page 19 of 29

Environmental Science & Technology

417

CCN activation (Case 1, without considering the influence of SAS partitioning between the

418

droplet surface and the bulk which likely overestimates the σ effect) may influence Nd to a

419

similar extent or even more than increasing particle number by a factor of two. The

420

investigated water-soluble SAS unlikely exert the observed surface-film effect on droplet

421

activation, as there is not enough SAS for lowering the σ of submicron particles substantially

422

(SI, Appendix 11). It should rather be volatile organic compounds62 or organics adsorbed at

423

the droplet surface from the gas phase,63 as proposed recently. Based on considering also the

424

size distribution of particulate SAS, we showed that Nd may increase for another 10% (up to

425

12% if moderate σ depression is considered), already when compared with a non-size

426

resolved treatment of SAS. This effect is especially pronounced in spring and summer, when

427

smaller particles (d ~ 100−250 nm) are enriched with SAS. The demonstrated influence of

428

size-segregated SAS on cloud formation is comparable to a 26% increase in apparent particle

429

hygroscopicity due to acetaldehyde adsorption to the aerosol-gas interface and the resulting σ

430

reduction.62 Although the observed 10% droplet number increase is not likely to result in

431

significant changes with respect to cloud lifetime and precipitation formation in urban

432

regions, there could be a special situation where such CCN number increase triggers the

433

precipitation. As the latter is most likely to happen under much cleaner than tested conditions

434

with low CCN numbers (where SAS effect on droplet number is expected to be larger),

435

studies should be repeated on aerosol samples from pristine environments (remote, maritime,

436

polar etc.). We believe that the abovementioned increase in Nd due to size-resolved SAS most

437

importantly influences the optical properties of clouds, which could have an indirect impact

438

on the climate forcing. If globally relevant, 10% increase in Nd changes the cloud albedo for

439

up to ~1%, corresponding to a radiative forcing in the order of −1 Wm−2.14,50 To better assess

440

the magnitude of the size-resolved SAS effect on cloud droplet formation, further large-scale

19 ACS Paragon Plus Environment

Environmental Science & Technology

441

field studies and climate predictions using the same approach are needed, especially in areas

442

that are critical for the climate response, such as oceanic regions.

443 444

Acknowledgments

445

We thank the Faculty of Chemistry and Chemical Technology University of Ljubljana,

446

Slovenia (UL FCCT) for the permission to install the Berner impactor in the area of the

447

faculty. The authors acknowledge Dr. Bojan Šarac from UL FCCT for the access to the

448

tensiometer and Dr. Janja Turšič from the Environment Agency of the Republic of Slovenia,

449

Ljubljana for providing PM10 samples and performing WSOC analyses. We gratefully

450

acknowledge Dr. Hartmut Herrmann from TROPOS for enabling this international

451

collaboration and Dr. Frank Stratmann also from TROPOS for a fruitful discussion. The

452

authors acknowledge the financial support from European Union Seventh Framework

453

Programme (FP7 2007-2013) Marie Curie FP7-PEOPLE-2011-COFUND (GA no291823)

454

through NEWFELPRO project (Contract no. 47) and the Slovenian Research Agency

455

(research core funding No. P1-0034).

456 457

Supporting Information

458

Detailed procedures of data manipulation: continuous PM and SAS distributions, sensitivity

459

of electrochemical determination of SAS, SAS isolation and extraction efficiency,

460

quantification of SAS, surface tension dependence on PM concentration, different

461

representations of SAS within aqueous aerosols, updraft velocity and lower-limit surface

462

tension impacts, drop number and cut-off diameter; surface tension evolution during droplet

463

nucleation; 1 Table, 11 Figures.

464 465

References

20 ACS Paragon Plus Environment

Page 20 of 29

Page 21 of 29

Environmental Science & Technology

466

(1)

Seinfeld, J. H.; Bretherton, C.; Carslaw, K. S.; Coe, H.; DeMott, P. J.; Dunlea, E. J.;

467

Feingold, G.; Ghan, S.; Guenther, A. B.; Kahn, R.; Kraucunas, I.; Kreidenweis, S. M.;

468

Molina, M. J.; Nenes, A.; Penner, J. E.; Prather, K. A.; Ramanathan, V.; Ramaswamy, V.;

469

Rasch, P. J.; Ravishankara, A. R.; Rosenfeld, D.; Stephens, G.; Wood, R. Improving our

470

fundamental understanding of the role of aerosol–cloud interactions in the climate system. P.

471

Natl. Acad. Sci. 2016, 113, 5781–5790.

472

(2)

473

Kerminen, V.-M.; Kondo, Y.; Liao, H.; Lohmann, U.; Rasch, P.; Satheesh, S.K.; Sherwood,

474

S.; Stevens, B.; Zhang, X. Y. Clouds and Aerosols. In Climate Change 2013: The Physical

475

Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the

476

Intergovernmental Panel on Climate Change, Cambridge University Press: Cambridge/New

477

York, 2013.

478

(3)

479

Soc. 1936, 32, 1152−1161.

480

(4)

481

Hallberg, A. Hygroscopic Growth of Aerosol Particles in the PoValley. Tellus Ser. B 1992,

482

44B, 556–569.

483

(5)

484

A.; Colvile, R. Hygroscopic Growth of Aerosol-Particles and Its Influence on Nucleation

485

Scavenging in Cloud – Experimental Results from Kleiner-Feldberg. J. Atmos. Chem. 1994,

486

19, 129–152.

487

(6)

488

and chemistry of atmospheric aerosol for Grand Canyon, Arizona, in winter 1990. Atmos.

489

Environ. 1994, 28, 827–839.

Boucher, O.; Randall, D.; Artaxo, P.; Bretherton, C.; Feingold, G.; Forster, P.;

Köhler, H. The nucleus in and the growth of hygroscopic droplets. Trans. Faraday

Svenningsson, B.; Hansson, H.-C.; Wiedensohler, A.; Ogren, J. A.; Noone, K. J.;

Svenningsson, B.; Hansson, H. C.; Wiedensohler, A.; Noone, K.; Ogren, J.; Hallberg,

Pitchford, M. L.; McMurry, P. H. Relationship between measured water vapor growth

21 ACS Paragon Plus Environment

Environmental Science & Technology

490

(7)

Capel, P. D.; Gunde, R.; Zuercher, F.; Giger, W. Carbon speciation and surface

491

tension of fog. Environ. Sci. Technol. 1990, 24, 722–727.

492

(8)

493

Dissolution behavior and surface tension effects of organic compounds in nucleating cloud

494

droplets. Geophys. Res. Lett. 1996, 23(3), 277–280.

495

(9)

496

atmospheric wet aerosol and cloud/fog droplets in relation to their organic carbon content and

497

chemical composition. Atmos. Environ. 2000, 34, 4853–4857.

498

(10)

499

tension of Rax cloud water and its relation to the concentration of organic material. J.

500

Geophys. Res. 2002, 107(D24), 4752.

501

(11)

502

atmospheric water-soluble organic compounds. Atmos. Environ. 2004, 38, 6135–6138.

503

(12)

504

water-soluble organic component of size-segregated aerosol, cloud water and wet depositions

505

from Jeju Island during ACE-Asia. Atmos. Environ. 2005, 39, 211–222.

506

(13)

507

substances in the aqueous extract of tropospheric fine aerosol. J. Atmos. Chem. 2005, 50, 279–

508

294.

509

(14)

510

surface-active organic solutes in growing droplets. Nature 1999, 401(6750), 257–25.

511

(15)

512

A. The role of surfactants in Köhler theory reconsidered. Atmos. Chem. Phys. 2004, 4, 2107–

513

2117.

Shulman, M. L.; Jacobson, M. C.; Carlson, R. J.; Synovec R. E.; Young T. E.

Facchini, M. C.; Decesari, S.; Mircea, M.; Fuzzi, S.; Loglio, G. Surface tension of

Hitzenberger, R.; Berner, A.; Kasper-Giebl, A.; Loflund, M.; Puxbaum, H. Surface

Tuckermann, R.; Cammenga H. K. The surface tension of aqueous solutions of some

Decesari, S.; Facchini, M. C.; Fuzzi, S.; McFiggans, G. B.; Coe, H.; Bower, K. N. The

Kiss, G.; Tombacz, E.; Hansson, H. C. Surface tension effects of humic-like

Facchini, M. C.; Mircea, M.; Fuzzi S.; Charlson R. J. Cloud albedo enhancement by

Sorjamaa, R.; Svenningsson, B.; Raatikainen, T.; Henning, S.; Bilde, M.; Laaksonen,

22 ACS Paragon Plus Environment

Page 22 of 29

Page 23 of 29

Environmental Science & Technology

514

(16)

Sorjamaa. R.; Laaksonen, A. The influence of surfactant properties on critical

515

supersaturations of cloud condensation nuclei. J. Aerosol Sci. 2006, 37, 1730–1736.

516

(17)

517

D.; Decesari, S.; Rinaldi, M.; Hodas, N.; Facchini, M. C.; Seinfeld, J.; O’ Dowd, C. Surface

518

tension prevails over solute effect in organic-influenced cloud droplet activation. Nature

519

2017, 546, 637–641.

520

(18)

521

of organic acids on the basis of surface tension and osmolality measurements. Atmos. Chem.

522

Phys. 2007, 7, 4601–4611.

523

(19)

524

and inorganic salts on cloud droplet activation. Atmos. Chem. Phys. 2011, 11, 3895–3911

525

(20)

526

activation – a model investigation considering the volatility, water-solubility and surface

527

activity of organic matter. J. Geophys. Res. 2002, 107, 4662.

528

(21)

529

organic aerosol in cloud droplet formation. Atmos. Chem. Phys. 2011, 11, 4073–4083.

530

(22)

531

carbon on cloud drop number concentration. J. Geophys. Res. 2005, 110, D18211.

532

(23)

533

on droplet Growth: Implications for Cloud Microphysical Processes and Climate. J. Atmos.

534

Sci. 2002, 59, 2006–2018.

535

(24)

536

aerosol surfactants reveals new aspects of cloud activation. Nat. Commun. 2014, 5, 4335.

537

(25)

Ovadnevaite, J.; Zuend, A.; Laaksonen, A.; Sanchez, A. K.; Roberts, G.; Ceburnis,

Varga, Z.; Kiss, G.; Hansson, H. C. Modelling the cloud condensation nucleus activity

Frosch, M.; Prisle, N. L.; Bilde, M.; Varga, Z.; Kiss, G. Joint effect of organic acids

Anttila, T.; Kerminen, V.-M. Influence of organic compounds on cloud droplet

Prisle, N. L.; Maso, M. D.; Kokkola, H. A simple representation of surface active

Ervens, B.; Feingold, G.; Kreidenweis, S. M. Influence of water soluble organic

Feingold, G.; Chuang, P. Y. Analysis of the Influence of Film-Forming Compounds

Nozière, B.; Baduel, C.; Jaffrezo, J. L. The dynamic surface tension of atmospheric

Ruehl, C. R.; Davies, J. F.; Wilson, K. R. Science 2016, 351, 1447.

23 ACS Paragon Plus Environment

Environmental Science & Technology

538

(26)

Gérard, V.; Nozière, B.; Baduel, C.; Fine, L.; Frossard, A. A.; Cohen, R. C. Anionic,

539

Cationic, and Nonionic Surfactants in Atmospheric Aerosols from the Baltic Coast at Askö,

540

Sweden: Implications for Cloud Droplet Activation. Environ. Sci. Tech. 2016, 50, 2974−2982.

541

(27)

542

growth and cloud condensation nucleus activity. Atmos. Chem. Phys. 2007, 7, 1961–1971

543

(28)

544

396–397.

545

(29)

546

surfactants in marine aerosol formed during particular weather conditions. Nuovo Cimento

547

Soc. Ital. Fis. C-Geophys. Space Phys. 1985, 8, 704–713.

548

(30)

549

damage by surfactant-polluted seaspray on Pinus pinea and Pinus halepensis. Environ. Mon.

550

Ass. 2005, 105, 175–191.

551

(31)

552

Mediterranean pines to simulated sea aerosol polluted with an anionic surfactant: prospects

553

for biomonitoring. Ann. Forest Sci. 2005, 62, 351–360.

554

(32)

555

Technol. 2004, 38(24), 6501–6506.

556

(33)

557

M.; Mohamed, C. A. R.; Latif, M. T. Surfactants in the sea-surface microlayer and their

558

contribution to atmospheric aerosols around coastal areas of the Malaysian peninsula. Mar.

559

Pollut. Bull. 2010, 60(9), 1584–1590.

560

(34)

561

substances in atmospheric aerosol: an electrochemical approach. Tellus, Ser. B 2012, 64, 12.

Petters, M. D.; Kreidenweis, S. M. A single parameter representation of hygroscopic

Blanchard, D. C. Sea-to-Air Transport of Surface Active Material. Science 1964, 146,

Loglio, G.; Tesei, U.; Mori, G.; Cini, R.; Pantani, F. Enrichment and transport of

Nicolotti, G.; Rettori, A.; Paoletti, E.; Gullino, M. L. Morphological and physiological

Rettori, A.; Paoletti, E.; Nicolotti, G.; Gullino, M. L. Ecophysiological responses of

Latif, M. T.; Brimblecombe, P. Surfactants in Atmospheric Aerosols. Environ. Sci.

Roslan, R. N.; Hanif, N. M.; Othman, M. R.; Azmi, W. N. F. W.; Yan, X. X.; Ali, M.

Frka, S.; Dautović, J.; Kozarac, Z.; Ćosović, B.; Hoffer, A.; Kiss, G. Surface-active

24 ACS Paragon Plus Environment

Page 24 of 29

Page 25 of 29

Environmental Science & Technology

562

(35)

Jaafar, S. A.; Latif, M. T.; Chian, C. W.; Han, W. S.; Wahid, N. B. A.; Razak, I. S.;

563

Khan, M. F.; Tahir, N. M. Surfactants in the sea-surface microlayer and atmospheric aerosol

564

around the southern region of Peninsular Malaysia. Mar. Pollut. Bull. 2014, 84(1−2), 35−43.

565

(36)

566

marine aerosols at different locations along the Malacca Straits. Environ. Sci. Pollut. Res.

567

2014, 21(10), 6590−6602.

568

(37)

569

Artaxo, P. A possible role of ground-based microorganisms on cloud formation in the

570

atmosphere. Biogeosciences 2010, 7(1), 387−394.

571

(38)

572

aerosols from Grenoble, France. Atmos. Environ. 2012, 47(0), 413−420.

573

(39)

574

surface-active substances in sea-water. Limno.l Oceanogr. 1982, 27, 361−368.

575

(40)

576

characterization of natural and ex-situ reconstructed sea-surface microlayers. J. Colloid

577

Interf. Sci. 1998, 208, 191−202.

578

(41)

579

active substances in the natural sea surface microlayers of the coastal Middle Adriatic

580

stations. Estuar. Coast Shelf S. 2009, 85, 555–564.

581

(42)

582

substances in the sea-surface microlayer and water column. Mar. Chem. 2009, 115, 1−9.

583

(43)

584

organic matter (DOC) in atmospheric precipitation of urban area of Croatia (Zagreb).

585

Water Air Soil Poll. 2004, 158, 295–310.

Mustaffa, N.; Latif, M.; Ali, M.; Khan, M. Source apportionment of surfactants in

Ekström, S.; Nozière, B.; Hultberg, M.; Alsberg, T.; Magnér, J.; Nilsson, E. D.;

Baduel, C.; Nozière, B.; Jaffrezo, J.-L. Summer/winter variability of the surfactants in

Ćosović, B.; Vojvodić, V. The application of ac polarography to the determination of

Gašparović, B.; Kozarac, Z.; Saliot, A., Ćosović, B.; Möbius, D.. Physicochemical

Frka, S.; Kozarac, K.; Ćosović, B. Characterization and seasonal variations of surface

Wurl, O.; Miller, L.; Ruttgers, R.; Vagle, S. The distribution and fate of surface-active

Leko, P. O., Kozarac, Z.; Ćosović, B.. Surface active substances (SAS) and dissolved

25 ACS Paragon Plus Environment

Environmental Science & Technology

586

(44)

Orlović-Leko, P.; Kozarac, Z.; Ćosović, B.; Strmečki, S.; Plavšić, M. Characterization

587

of atmospheric surfactants i n the bulk precipitation by electrochemical tools. J. Atmos. Chem.

588

2010, 66, 11−26.

589

(45)

590

carbon in humic-like matter of ambient size-segregated water soluble organic aerosols from

591

urban background environment. Atmos. Environ. 2018, 173, 239-247.

592

(46)

593

Pio, C.; Preunkert, S.; Legrand, M. Determination of water and alkaline extractable

594

atmospheric humic-like substances with the TU Vienna HULIS-analyser in samples from six

595

background sites in Europe. J. Geophys. Res. 2007, 112, D23S10.

596

(47)

597

organic matter from atmospheric aerosol. Talanta 2001, 55, 561–572.

598

(48)

599

natural organic matter in marine and freshwater systems. A plea for use of mercury for

600

scientific purposes. Electroanalysis. 2010, 17-18(22), 1994–2000.

601

(49)

602

Aerosols. Top. Curr. Chem. 2014, 339, 20–259.

603

(50)

604

Atmos. Environ. 2000, 34, 4917–4932.

605

(51)

606

atmospheric water-soluble organic compounds. Atmos. Environ. 2004, 38, 6135–6138.

607

(52)

608

Investigation of molar volume and surfactant characteristics of water-soluble organic

609

compounds in biomass burning aerosol. Atmos. Chem. Phys. 2008, 8, 799–812.

Frka, S.; Grgić, I.; Turšič, J.; Gini, M. I.; Eleftheriadas. K. Seasonal variability of

Feczko, T.; Puxbaum, H.; Kasper-Giebl, A.; Handler, M.; Limbeck, A.; Gelencsér, A.;

Varga, B.; Kiss, G.; Ganszky, I.; Gelencser, A.; Krivacsy, Z. Isolation of water-soluble

Ćosović, B.; Kozarac, Z.; Frka, S.; Vojvodić, V. Electrochemical adsorption study of

McNeill, V.F.; Sareen, N.; Schwier, A. N. Surface-Active Organics in Atmospheric

Seidl, W. Model for a surface film of fatty acids on rainwater and aerosol particles.

Tuckermann, R.; Cammenga, H. K. The surface tension of aqueous solutions of some

Asa-Awuku, A.; Sullivan, A. P.; Hennigan, C. J.; Weber, R. J.; Nenes, A.

26 ACS Paragon Plus Environment

Page 26 of 29

Page 27 of 29

Environmental Science & Technology

610

(53)

Salma, I.; Ocskay, R.; Chi, X.; Maenhaut, W. Sampling artefacts, concentrations and

611

chemical composition of fine water soluble organic carbon and humic-like substances in a

612

continental urban atmospheric environment. Atmos. Environ. 2007, 41, 4106–4118.

613

(54)

614

microphysical models based on the linear discrete method. Atmos. Res. 2006, 80, 218–236.

615

(55)

616

H.; Cappa C. D. Establishing the impact of model surfactants on cloud condensation nuclei

617

activity of sea spray aerosols, Atmos. Chem. Phys. Discuss. 2018, https://doi.org/10.5194/acp-

618

2018-207, in review.

619

(56)

620

molecular weight of humic-like substances isolated from fine atmospheric aerosol. Atmos

621

Environ 2003, 37, 3783–3794.

622

(57)

623

aerosol. Sci. World J. 2002, 2, 1138–1146.

624

(58)

625

Jalba, A.; Despiau, S.; Dayan, U.; Temara, A. MBAS (methylene blue active substances) and

626

LAS (linear alkylbenzene sulphonates) in Mediterranean coastal aerosols: sources and

627

transport processes. Atmos. Environ. 2011, 45, 6788–6801.

628

(59)

Twomey, S. Pollution and the planetary albedo. Atmos. Environ. 1974, 8, 1251–1256.

629

(60)

Mircea, M.; Facchini, M. C.; Decesari, S.; Fuzzi, S.; Charlson, R. J. The influence of

630

the organic aerosol component on CCN supersaturation spectra for different aerosol types,

631

Tellus Ser B – Chem. Phys. Meteor. 2002, 54, 74–81.

632

(61)

633

H. Can Chemical Effects on Cloud Droplet Number Rival the First Indirect Effect? Geophys.

634

Res. Lett, 2002, 24(17), 1848.

Simmel, M.; Wurzler, S. Condensation and nucleation in sectional cloud

Forestieri, S. D.; Staudt, S. M.; Kuborn, T. M.; Faber, K.; Ruehl, C. R.; Bertram, T.

Kiss, G.; Tombacz, E.; Varga, B.; Alsberg, T.; Persson, L, Estimation of the average

Sukhapan, J.; Brimblecombe, P. Ionic surface-active compound in atmospheric

Becagli, S.; Ghedini, C.; Peeters, S.; Rottiers, A.; Traversi, R.; Udisti, R.; Chiari, M.;

Nenes, A.; Charlson, R. J.; Facchini, M. C.; Kulmala, M.; Laaksonen, A.; Seinfeld, J.

27 ACS Paragon Plus Environment

Environmental Science & Technology

635

(62)

Sareen, N.; Schwier, A. N.; Nenes, A.; McNeill, V. F. Surfactants from the gas phase

636

may promote cloud droplet formation. P. Natl. Acad. Sci. 2013, 110, 2723–2728.

637

(63)

638

Growth of Submicrometer Particles at High Relative Humidity. J. Phys. Chem. A 2014, 118,

639

3952–3966.

Ruehl, C. R.; Wilson, K. R. Surface Organic Monolayers Control the Hygroscopic

640 641 642 643 644 645 646 647

648 649

TOC/Abstract Art

650 28 ACS Paragon Plus Environment

Page 28 of 29

Aerosol size

Page 29 Environmental of 29 Science & Technology

ACS Paragon Plus Environment Surfactant activity factor = Surfactant enrichment