Significant Increase of Aromatics-Derived Secondary Organic Aerosol

Subscriber access provided by Binghamton University | Libraries

Article

Significant increase of aromatics-derived secondary organic aerosol during fall to winter in China Xiang Ding, Yu-Qing Zhang, Quanfu He, Qing-Qing Yu, JunQi Wang, Ru-Qin Shen, Wei Song, Yuesi Wang, and Xinming Wang Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 07 Jun 2017 Downloaded from http://pubs.acs.org on June 7, 2017

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

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

Page 1 of 24

Environmental Science & Technology

1

Significant increase of aromatics-derived secondary organic aerosol

2

during fall to winter in China

3 4

Xiang Ding †,*, Yu-Qing Zhang †, §, Quan-Fu He †, Qing-Qing Yu †, §, Jun-Qi Wang †, §,

5

Ru-Qin Shen †, Wei Song †, Yue-Si Wang ‡, , Xin-Ming Wang †,

‖

‖

6 7

† State Key Laboratory of Organic Geochemistry and Guangdong Provincial Key

8

Laboratory of Environmental Protection and Resources Utilization, Guangzhou

9

Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China

10

‡ State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric

11

Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing

12

100029, China

13

§ University of Chinese Academy of Sciences, Beijing 100049, China

14

ǁ Center for Excellence in Regional Atmospheric Environment, Institute of Urban

15

Environment, Chinese Academy of Sciences, Xiamen 361021, China

16 17 18

* Corresponding author

19

Dr. Xiang Ding

20

State Key Laboratory of Organic Geochemistry

21

Guangzhou Institute of Geochemistry, Chinese Academy of Sciences

22

511 Kehua Rd, Tianhe, Guangzhou 510640, China

23

Tel: +86-20-85290127; Fax: +86-20-85290706

24

E-mail: [email protected]

25 26

ACS Paragon Plus Environment


27

Page 2 of 24

Abstract

28

Human activities release large amounts of anthropogenic pollutants into the air,

29

and thereby produce substantial secondary organic aerosol (SOA). Aromatic

30

hydrocarbons (AHs) that mainly emitted from coal combustion, transportation,

31

solvent use and biofuel/biomass burning, are a major class of anthropogenic SOA

32

precursors. At present, there are few field studies focusing on AH-derived SOA

33

(SOAA) on a continental scale, especially in polluted regions of the world. In this

34

study,

35

2,3-dihydroxy-4-oxopentanoic acid (C5H8O5, DHOPA) was carried out at 12 sites

36

across 6 regions of China for the first time. The annual averages of DHOPA among

37

the 12 sites ranged from 1.23 to 8.83 ng m-3 with a mean of 3.48 ± 1.96 ng m-3. At all

38

observation sites, the concentrations of DHOPA from fall to spring were significantly

39

higher than those in summertime, and positive correlations were observed between

40

DHOPA and the biomass burning tracer (levoglucosan). This indicated that such a

41

nationwide increase of SOAA during the cold period was highly associated with the

42

enhancement of biomass burning emission. In the northern China, the highest levels

43

of DHOPA were observed in the coldest months during winter, probably due to the

44

enhancement of biofuel and coal consumption for household heating. In the southern

45

China, the highest levels of DHOPA were mostly observed in fall and spring, which

46

were associated with the enhancement of open biomass burning. The apparent

47

increases of DHOPA and levoglucosan levels during the cold period and the negative

48

correlations of visibility with DHOPA and levoglucosan imply that the reduction of

49

SOAA amount and biomass burning emission is an efficient way to reduce haze

50

pollution during fall to winter in China.

51 52

Introduction

a

one-year

concurrent

observation

of

the

SOAA

tracer,

53

Secondary organic aerosol (SOA) is formed by volatility-reducing homogenous

54

and heterogeneous reactions of volatile organic compounds (VOCs) in the air.1-3 As a

55

major but the least understood component of particulate matter (PM),4 SOA has

56

significant effects on global climate change and regional air quality.5 Since global

57

emissions of anthropogenic VOCs (AVOCs) are approximately one order of

58

magnitude lower than those of biogenic VOCs (BVOCs),6,

59

(ASOA) is thought to have minor contributions to global SOA.8 As a global chemical

60

transport modeling study predicted, ASOA formation might be as high as 10.9 Tg yr-1,


7

anthropogenic SOA

Page 3 of 24


61

accounting for 29% of global SOA.9 However, in urban areas and developing regions,

62

human activities emit large amounts of anthropogenic pollutants into the air, and

63

thereby produce substantial ASOA.10 Volkamer et al.

64

ASOA in the air of Mexico City. Based on ship and aircraft measurements, de Gouw

65

et al.

66

northeastern United States. Field observations also demonstrated that ASOA was

67

dominant over biogenic SOA (BSOA) at urban sites.13, 14 In addition, anthropogenic

68

emissions in urban regions could accelerate BVOCs oxidation and BSOA formation.15,

69

16

70

SOA

12

11

observed rapid formation of

found that SOA was mainly formed from AVOCs in urban plumes in the

Thus, in polluted regions, anthropogenic sources have important contributions to

71

Aromatic hydrocarbons (AHs) are typical AVOCs17 and a major class of ASOA

72

precursors. The formation of SOA from AHs (SOAA) depends on molecular structures

73

of AHs18 as well as types and concentrations of oxidants.19 Globally, the production of

74

SOAA was estimated within the range of 2 to 12 Tg yr-1.19 The observations of SOA

75

tracers in megacities showed that SOAA was dominant over the BSOA from isoprene,

76

monoterpenes and β-caryophyllene in Beijing20, Shanghai21, Guangzhou22 and

77

Jerusalem23. Moreover, the oxidation of AHs produces glyoxal24 which is an

78

important precursor for global SOA forming via irreversible uptake by aqueous

79

aerosols and clouds.25 Liu et al.

80

of glyoxal between satellite observations and model predictions in China was largely

81

due to the underestimation of current AHs emissions by a factor of 4 to10.

26

found that the disagreement in spatial distribution

82

After rapid urbanization and industrialization over the past three decades, China

83

is suffering heavily from severe PM pollution.27, 28 During wintertime haze events in

84

China, SOA contributed 44 to71% of organic aerosol (OA), and was estimated to be

85

mainly from fossil sources in the northern China and non-fossil sources in the

86

southern China.29,

87

combustion, transportation and solvent use)17 and non-fossil sources (e.g. biofuel and

88

biomass burning).31 Therefore, SOAA should be an important contributor to SOA in

89

China, especially in wintertime. A recent modeling study predicted that, compared

90

with BVOCs and other AVOCs, AHs were the predominant precursors of SOA during

91

springtime in China.32 Our previous ground-based observation across six regions of

92

China in summertime showed that SOAA was the single largest contributor to SOA in

93

North China.33 At present, due to the lack of large-scale and long-term field

94

measurements, it is still unclear what the role SOAA plays in PM and haze pollution

30

AHs could be emitted from both fossil sources (e.g. coal



95

over China. In this study, we carried out a one-year nationwide observation of the

96

SOAA tracer, 2,3-dihydroxy-4-oxopentanoic acid (C5H8O5, DHOPA), in order to

97

characterize spatial and seasonal variations of SOAA in China. We further check the

98

correlations between DHOPA and visibility to explore the role of SOAA in visibility

99

degradation over China.

100 101

Experimental section

102

Field sampling

103

This SOA study is a sub-project of a national aerosol campaign that primarily aims

104

to determine the size distribution of major components in PM over China.28 Sampling

105

was simultaneously carried out at 12 sites in China, including five urban sites, three

106

suburban sites, and four rural sites (Figure 1a). The 12 sites cover 6 regions of China,

107

including Hailun (HL) and Tongyu (TYU) in Northeast China, Beijing (BJ) and

108

Taiyuan (TY) in North China, Dunhuang (DH) and Shapotou (SPT) in Northwest

109

China, Hefei (HF), Wuxi (WX), and Qianyanzhou (QYZ) in East China, Kunming

110

(KM) and Xishuangbanna (BN) in Southwest China, and Sanya (SY) in South China.

111

The geographical dividing line between the northern and southern China is the Huai

112

River–Qin Mountains Line. This line approximates the 0 °C January isotherm in

113

China. There are central heating systems in urban areas of the northern China, but not

114

so in the southern China. The sites in the northern China include HL, TYU, BJ, TY,

115

DH, and SPT. The rest 6 sites (HF, WX, QYZ, KM, BN and SY) are located in the

116

southern China.

117

PM samples were collected with Anderson 9-stage cascade impactors equipped

118

with quartz fiber filter. One set of 9 size-fractionated filters was collected biweekly

119

over 48 hours at each site. A total of 294 sets of PM samples were collected from

120

October, 2012 to September 2013. Additionally, one set of field blanks was collected

121

at each site in the same way as ambient samples for 5 minutes when the sampler was

122

turned off.

123 124

Chemical analysis

125

Since not all size-fractionated filters had detectable levels of DHOPA, we

126

combined each set of nine filters into one sample for chemical analysis to compare

127

tracer levels at the 12 sites for the whole year.34, 35 The analytical procedure of SOA

128

tracers is described elsewhere.36-38 Isotope-labeled standard mixtures were spiked into


Page 4 of 24

Page 5 of 24


129

the samples as internal standards prior to solvent extraction. Specifically, dodecanoic

130

acid-d23 was used as the internal standard for DHOPA quantification. Samples were

131

extracted by sonication with different mixed solvents. The extracts of each sample

132

were combined, filtered and concentrated to ~2 mL. Then, the concentrated solution

133

was divided into two parts for methylation and silylation, respectively.

134

The silylated extract was analyzed for DHOPA with an Agilent 7890/5975C gas

135

chromatography/mass spectrometry equipped with an electron impact ionization

136

source and a 30 m HP-5 MS capillary column (i.d. 0.25 mm, 0.25 µm film thickness).

137

Due to the lack of a commercial standard, DHOPA was quantified using an isomer

138

surrogate, citramalic acid. Figure S1a in Supporting Information shows the retention

139

times and structures of these two isomers. Figure S1b presents the EI spectra of

140

silylated DHOPA. Table S1 summarizes the abundances of DHOPA at each sampling

141

site in China.

142

Daily temperature, relative humidity and the maximum solar radiation during

143

each sampling episode were downloaded from the China Meteorological Data Sharing

144

Service System (http://cdc.nmic.cn/home.do). Daily visibility during each sampling

145

episode at the airport close to each sampling site was downloaded from the website

146

(https://www.wunderground.com/). All these meteorological data are summarized in

147

Table S1.

148 149

Quality assurance and quality control

150

Field and laboratory blanks were analyzed in the same manner as the filter

151

samples. DHOPA was not detected in the field and laboratory blanks. The recovery of

152

the analytical method for citramalic acid was 91 ± 3% based on six spiked samples

153

(authentic standards spiked into solvent with pre-baked quartz filters). The method

154

detection limit (MDL) for citramalic acid was 0.11 ng m-3 with a total volume of 81.5

155

m3. The relative difference for DHOPA in samples collected in parallel (n=6) was less

156

than 10%. Considering the errors in field blanks, recovery, and surrogate

157

quantification, the uncertainty in DHOPA analysis was estimated to be 9% by the

158

method developed by Stone et al.39 The response factor of internal quantification for

159

citramalic acid (1.42) was consistent with that for ketopinic acid (1.27). The latter was

160

used as the surrogate for the quantification of all SOA tracers by Kleindienst et al.40

161



162

Results and Discussion

163

General marks

164

Annual average of the SOAA tracer, DHOPA, ranged from 1.23 to 8.83 ng m-3

165

among the 12 sites with a mean of 3.48 ± 1.96 ng m-3. The lowest level occurred at the

166

SPT site in Northwest China, and the highest concentration was observed at the BN

167

site in Southwest China (Figure 1a). Our measurements were consistent with the

168

values (1.20 to 17.8 ng m-3) reported in megacities in different regions of China.20-22,

169

41

170

The spatial distribution of DHOPA exhibited different characteristics in the

171

northern and southern regions of China. In the northern China, higher levels of

172

DHOPA were observed at the urban sites (DH and HL) and the lowest concentrations

173

occurred at the rural SPT site (Figure 1b). The paired sites within each northern region

174

exhibited synchronous variations in DHOPA levels and the correlations between

175

paired sites were significant (p0.05, Figure 1b).

179

The poor correlations of DHOPA between paired sites (p>0.05, Table 1 and Figure

180

S2b) implied a lack of spatial homogeneity of SOAA in the southern regions.

181 182

Significant increase of DHOPA from fall to spring

183

As shown in Figure 2, the concentrations of DHOPA at all sites were higher from

184

fall to spring and decreased in summer. The seasonal averages in fall, winter and

185

spring were all higher than that in summer at each site (Table S1), suggesting

186

nationwide increases in SOAA levels during the cold period in China. Seasonal trend

187

of DHOPA is determined by AHs emissions, atmospheric chemistry, gas-particle

188

partitioning, and planetary boundary layer (PBL). The Asia emissions of AHs in the INTEX-B inventories developed by Zhang et al.

189 190

42

191

transportation, and residential activities (biofuel and fossil fuel combustion, and

192

non-combustion sources). Among these sources, residential activities that contributed

193

about one third of AHs emissions17 exhibited strong seasonal variations, while power

194

plants, industrial and transportation sectors showed less seasonal changes.42 In

included major anthropogenic sectors, e.g. power plants, industry processes,


Page 6 of 24

Page 7 of 24


195

addition, open biomass burning (BB, e.g. forest fires and open burning of agriculture

196

residuals) that is excluded in the Asia emissions inventory42 also emits large amounts

197

of AHs.31,

198

https://firms.modaps.eosdis.nasa.gov/firemap/) recorded a large number of fire spots

199

and strong seasonality of fire activities in China during our campaign (Figure S3),

200

implying nationwide impact of open BB on air quality.

43-45

Fire Information for Resource Management System (FIRMS,

201

Previous field observations in China have demonstrated that biofuel/biomass

202

combustion were important sources of AHs in the atmosphere.46, 47 Both residential

203

biofuel combustion and open BB release substantial levoglucosan, a specific tracer for

204

cellulose in BB.48 Similar to DHOPA, high levels of levoglucosan were also observed

205

from fall to spring nationwide (Figure S4). Moreover, DHOPA was positively

206

correlated with levoglucosan at all sites (p0.05)

248

except HL, DH and SPT (Figure S6). Therefore, the nationwide increase of DHOPA

249

levels in wintertime was not mainly due to the drop of PBL height. In addition,

250

compared with winter, stronger wet deposition in summer will remove SOAA more

251

efficiently and may lead to lower DHOPA concentrations observed in summer

252

samples.

Archived

Meteorology

online

calculating

program

253 254

Major sources causing the increase of DHOPA

255

Our analysis suggested that the increase of DHOPA levels during the cold period

256

was mainly due to the enhancement of AHs emissions. There is a winter heating

257

season in the northern China, but not so in the southern China. During the cold period,

258

this is a significant difference in energy consumption between the northern and

259

southern China. Considering that DHOPA at the sites in the northern and southern

260

regions showed different behaviors in spatial homogeneity (Figure S2), we discuss the

261

major sources causing the increase of DHOPA in the northern and southern China,


Page 9 of 24


262

respectively.

263

In the northern China, ambient temperature dropped below zero from November

264

to February, and the highest levels of DHOPA were all observed in the coldest months

265

during winter (Figure 2 a-f). There are central heating systems in urban areas of the

266

northern China. Domestic heating supply in urban areas usually starts in

267

mid-November and lasts until the next mid-March. In 2012, heating consumed ~7%

268

of coal usage nationwide.52 In addition, biofuel combustion accounted for

269

approximately two thirds of the total energy consumption for cooking and heating in

270

rural areas of China.53-55 A recent study found that residential emissions, mostly from

271

heating and cooking with coal and biofuel, contributed more than 70% of annual OC

272

emission in the North China Plain, and more fuel was burned for heating during

273

January and February.56 Thus, the significant increase of DHOPA levels during winter

274

in the northern China should be largely due to the enhancement of coal and biofuel

275

consumption for heating. Since open BB activities declined during winter in the

276

northern part of China (Figure S3b), the positive correlations between DHOPA and

277

levoglcusoan (Figure 3a) indicated that the elevated DHOPA levels were highly

278

associated with the enhancement of biofuel combustion. Without the measurement of

279

specific tracers for coal combustion, we could not directly evaluate the impact of coal

280

combustion on DHOPA levels. As Figure 3 presented, the correlations between

281

DHOPA and levoglcusoan at the northern sites were stronger (r values) and more

282

significant (p values) than those at the southern sites. Considering the concurrent

283

increases of coal and biofuel consumption for heating during the cold period, such

284

strong and significant correlations at the northern sites might partly result from the

285

enhancement of coal combustion for heating. The energy consumption for heating

286

explains the observed regional characteristics of DHOPA in the northern China.

287

Higher population density in urban areas demands more fuel for heating than rural

288

areas. Hence, higher levels of DHOPA were observed at urban sites in the northern

289

China. Since space heating starts simultaneously within a region, it is expected that

290

paired sites within a region exhibit synchronous increases in DHOPA levels. This

291

probably explains the significant correlation of DHOPA between the DH-SPT paired

292

sites with a distance of ~940 km (Table 1), since other anthropogenic emissions of

293

AHs between the urban DH site and the rural SPT site should be quiet different on

294

such a large geographical scale.



295

In the southern China, although ambient temperature indeed decreased during

296

winter, the highest levels of DHOPA did not occur in the coldest months but were

297

mostly observed in fall and spring (Figure 2 g-l). Ambient temperature in wintertime

298

was above zero at the southern sites during our sampling period, especially in

299

Southwest China and South China (Table S1). Due to relatively high temperature in

300

wintertime, there is no heating supply in the southern regions of China. Thus, the

301

enhancement of energy consumption for household heating during winter in the

302

southern China should be not as significant as that in the northern China. On the other

303

hand, as the FIRMS recorded, the IndoChina Peninsula, Indian Peninsula and

304

southern China witnessed a huge increase of fire spots in winter and spring (Figure S3

305

b-c), suggesting a significant enhancement of open BB there. Moreover, DHOPA was

306

positively correlated with levoglucosan at all sites in the southern China (Figure 3b).

307

These factors indicated that the significant increase of DHOPA levels from fall to

308

spring in the southern China was associated with the enhancement of open BB. This

309

also explains the observed regional characteristics of DHOPA in the southern China.

310

Both forest fires and open burning of agriculture residuals always happen in rural and

311

sub-urban areas, and the frequency and intensity of open BB vary from place to place.

312

If rural and sub-urban areas undergo intensive open BB, DHOPA levels there might be

313

higher than those at urban sites, and paired sites within a region would not show

314

synchronous variations of DHOPA levels. This might explain the poor correlation of

315

DHOPA between the WX-HF paired sites within a distance of ~280 km.

316

Other anthropogenic sources, e.g. power plants, industry processes, and

317

transportation also affect regional variability of DHOPA. However, these

318

anthropogenic sectors exhibited little seasonal changes in the Asia emission

319

inventories.42 Moreover, the distribution of these anthropogenic emissions was not

320

spatial homogeneity (Figure S7). Synchronous variations of these anthropogenic

321

emissions might be expected within a small geographical scale but not so within a

322

large scale. If these anthropogenic sources determined regional variability of DHOPA,

323

we would see more significant correlations of DHOPA between closer paired sites. In

324

fact, as Table 1 showed, the DH-SPT paired sites with the farthest distance of ~940

325

km showed a significant correlation of DHOPA (p0.05). This

327

suggested that the observed regional variability of DHOPA might be not mainly

328

caused by these local anthropogenic sources.


Page 10 of 24

Page 11 of 24


329

It is interesting to note that, although the rural BN site in Southwest China had

330

low anthropogenic emissions of AHs (Figure S7), DHOPA levels at BN were the

331

highest among the 12 sites (Figure 1). To check the origins of SOAA at the BN site,

332

we analyzed the clusters of air masses there using the HYSPLIT4 model.57 The cluster

333

analysis was performed based on 48-hr backward trajectories during each sampling

334

episode, considering that the lifetimes of AHs in ambient air are 1−2 days except

335

benzene (Table S4). As Figure 4a showed, prior to arriving the BN site, air masses

336

mainly passed through Thailand (47%), Burma (42%) and Vietnam (8%) in the

337

IndoChina Peninsula. Only 3% of air masses transported from China inland. From fall

338

to spring, the significant enhancement of open BB in Southwest China and the

339

IndoChina Peninsula (Figure S3 a-c) will release large amounts of AHs, and hence

340

form substantial SOAA. During the same period, air masses at BN were all originated

341

from the intensive open BB regions in the IndoChina Peninsula (Figure 4b). And

342

DHOPA levels peaked in spring. In summer when open BB activities decreased in

343

these areas (Figure S3 d), DHOPA levels at BN were significantly shrunk. The

344

positive correlation between DHOPA and levoglucosan at the BN site (Figure 3b)

345

demonstrated the significant impact of open BB on DHOPA levels there. In addition,

346

Vietnam, Thailand and Burma in the IndoChina Peninsula were also high emission

347

areas of anthropogenic AHs in Asia (Figure S7). AHs emitted from anthropogenic

348

sources in these developing countries could form large amounts of SOAA and

349

transport to BN. All these emissions would lead to high DHOPA levels observed at the

350

rural BN site during our campaign. Located in the intensive open BB regions, the KM

351

site also showed a significant correlation between DHOPA and levoglucosan (Figure

352

3b). However, air masses at KM mainly passed through low emission regions of

353

anthropogenic AHs in Yunnan province (Figure S8). Thus, the KM site showed lower

354

levels of DHOPA compared with the BN site.

355 356 357

Implication in haze pollution control in China From fall 2012 to winter 2013, China suffered from severe haze events.58,

59

358

During an extreme winter haze episode in January 2013, OM at urban locations in

359

Beijing, Shanghai, Guangzhou and Xi’an constituted a major fraction (30–50%) of

360

PM2.5, and SOA contributes 44–71% of OM.29 Our year-round sampling from 2012 to

361

2013 overlapped the severe haze periods. Besides DHOPA and levoglucosan, BSOA



362

tracers from isoprene, monoterpenes and β-caryophyllene at the 12 sites were reported

363

in our previous studies.34,

364

average of BSOA tracers at the 12 sites was the lowest in wintertime and increased in

365

summertime (Figure S4). To probe the impact of SOA and BB on visibility

366

degradation over China, we checked the correlations between visibility and these

367

tracers. Visibility was negatively correlated with DHOPA and levoglcusoan at 7 of the

368

12 sites (p

Significant Increase of Aromatics-Derived Secondary Organic Aerosol

Recommend Documents