Residential Coal Combustion as a Source of Levoglucosan in China

*Phone: +86-10-62752436. .... Source testing was performed in the laboratory of Residential Combustion Simulation at Peking University Shenzhen Gradua...
2 downloads 0 Views 2MB Size
Subscriber access provided by ECU Libraries

Article

Residential Coal Combustion as a Source of Levoglucosan in China Caiqing Yan, Mei Zheng, Amy P. Sullivan, Guofeng Shen, Yingjun Chen, Shuxiao Wang, Bin Zhao, Siyi Cai, Yury Desyaterik, Xiaoying Li, Tian Zhou, Örjan Gustafsson, and Jeffrey L Collett Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b05858 • Publication Date (Web): 15 Dec 2017 Downloaded from http://pubs.acs.org on December 16, 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 36

Environmental Science & Technology

Residential Coal Combustion as a Source of Levoglucosan in China

1 2 3

Caiqing Yan†, Mei Zheng†*, Amy P. Sullivan‡, Guofeng Shen∥,a, Yingjun Chen⊥,

4

Shuxiao Wang#, Bin Zhao#, Siyi Cai#, Yury Desyaterik‡, Xiaoying Li†, Tian Zhou†,

5

Örjan Gustafsson§, Jeffrey L. Collett Jr.‡

6 7 8 9 10 11 12



SKL-ESPC and BIC-ESAT, College of Environmental Sciences and Engineering,

Peking University, Beijing 100871, China ‡

Department of Atmospheric Science, Colorado State University, Fort Collins,

Colorado 80523, USA ∥

College of Urban and Environmental Sciences, Peking University, Beijing 100871,

China ⊥

Key Laboratory of Cities’ Mitigation and Adaption to Climate Change in Shanghai

13

(China Meteorological Administration), College of Environmental Science and

14

Engineering, Tongji University, Shanghai 200092, China.

15

#

16

School of Environment, Tsinghua University, Beijing 100084, China

17

State Key Joint Laboratory of Environment Simulation and Pollution Control,

§

Department of Environmental Science and Analytical Chemistry (ACES) and the

18

Bolin Centre for Climate Research, Stockholm University, 10691 Stockholm, Sweden

19

a

20

Laboratory (NRMRL), U.S. EPA, RTP, 200709, U.S.

21

*

presently ORISE postdoctoral fellow at National Risk Management Research

Corresponding author: [email protected]

1

ACS Paragon Plus Environment

Environmental Science & Technology

22

TOC

23

2

ACS Paragon Plus Environment

Page 2 of 36

Page 3 of 36

Environmental Science & Technology

24

ABSTRACT

25

Levoglucosan (LG) has been widely identified as a specific marker for biomass

26

burning (BB) sources, and frequently utilized in estimating the BB contribution to

27

atmospheric fine particles all over the world. However, this study provides direct

28

evidence to show that coal combustion (CC) is also a source of LG, especially in the

29

wintertime in Northern China based on both source testing and ambient measurement.

30

Our results show that low temperature residential CC could emit LG with emission

31

factors (EF) ranging from 0.3 to 15.9 mg kg-1. Ratios of LG to its isomers, mannosan

32

and galactosan, differ between CC and BB emissions, and the wintertime ratios in

33

Beijing ambient PM2.5 and source specific tracers including carbon isotopic signatures

34

all indicated a significant contribution from CC to ambient levoglucosan in winter in

35

Beijing. The results suggest that LG cannot be used as a distinct source marker for

36

biomass burning in special cases such as some cities in the northern China, where coal

37

is still widely used in the residential and industrial sectors. Biomass burning source

38

could be overestimated although such overestimation could vary spatially and

39

temporally.

3

ACS Paragon Plus Environment

Environmental Science & Technology

40

INTRODUCTION

41

Levoglucosan (LG, 1, 6-anhydro-β-D-glucopyranose, C6H10O5), a major pyrolysis

42

product of cellulose, has been proposed as a specific molecular marker of biomass

43

burning aerosols.

44

contribution of biomass burning sources, and assess the impacts of wood burning,

45

agricultural waste burning and wildfire emissions on ambient air quality in urban,

46

rural background and remote areas. 5-8

1-4

It has been widely utilized in numerous studies to identify the

47

Previous studies have reported high-levels of atmospheric LG in China, with

48

maxima in cold seasons (fall and winter) and a minimum in the warm season

49

(summer), at urban, suburban and background.

50

studies

51

usually occurred in fall followed by winter, and were about 2-5 times higher than in

52

summer. The highest LG concentrations in fall in China could be closely associated

53

with the burning of agricultural crop residues in farmland during the autumn

54

post-harvest period. 17,18 In winter high ambient LG levels were primarily attributed to

55

residential biomass/biofuel (e.g., straws and wood branches) burning for household

56

cooking and heating, especially in rural and suburban regions, coupled with more

57

stagnant air and other unfavorable meteorological conditions. 16,19

9,10,15

9-16

For example, some year-long

at urban sites in Beijing reported that the highest LG concentrations

58

A recent study based on positive matrix factorization modeling implied that the

59

biomass burning source factor only explained 84% of the total variation of LG in

60

Barcelona, Spain, 20 suggesting that there might be other potential sources of ambient

61

LG. It is known that LG could be produced from combustion and pyrolysis of

62

cellulose,

63

burning, as well as charcoal broiling.

64

of other materials (e.g., fossil fuels), biodegradation or hydrolysis of cellulose do not

4

and unquestionable that it could be emitted from crop straw and wood 21

Previous studies have stated that combustion

4

ACS Paragon Plus Environment

Page 4 of 36

Page 5 of 36

Environmental Science & Technology

22,23

65

produce any LG.

66

emissions, as coal is formed through geologic processes from dead plant matter and

67

other organic matter living millions of years ago. A few studies have mentioned that

68

combustion of some low quality coals, such as peat, lignite and brown coal could also

69

be an additional source of LG in regions where these materials are utilized as major

70

household fuels.

71

and residential coal combustions in China, but the measured LG was attributed to

72

emissions from combustion of the ignition materials, such as woods or paper, whose

73

influence was not anticipated and removed in those earlier studies. 8,26

[24,25]

However, in theory, LG might exist in coal combustion

LG was previously reported in the emissions from industrial

74

Coal and biomass are two major sources of energy in China. An overwhelming

75

majority of China’s residents (including most rural and suburban areas) still use coal

76

and biomass (collectively referred to as solid fuels) for domestic cooking and heating

77

purposes, 27 especially during the winter heating season. Particularly residents in rural

78

areas in China may be inclined to choose low-ranking coal for residential heating and

79

cooking as cost is still a primary concern. Raw coals with low coalification and

80

without efficient coal washing are widely used. Based on the statistics of coal usage in

81

People’s Republic of China by International Energy Agency, the most important coal

82

used in the residential sectors is bituminous coal.

83

electricity plants, coke is used besides bituminous coal, and in the industry sector,

84

more types of coal, including bituminous coal, coke, and anthracite are used, but still

85

with bituminous coal dominated. 28 Wang et al. indicate that coal used by residents in

86

Beijing-Tianjin-Hebei area in the northern China are mainly low ranked bituminous

87

with a small amount of anthracite.

88

sources of LG and the relative contribution of each source type is urgently needed to

89

reduce uncertainties in source apportionment studies, to accurately estimate the

29

28

While in heating plants and

Therefore, a better understanding of potential

5

ACS Paragon Plus Environment

Environmental Science & Technology

90

impact of biomass burning on air quality, and to formulate effective haze control

91

strategies.

92

The primary goal of this study is to identify if coal combustion is a significant

93

source of LG in Northern China. To achieve this, fine particle (PM2.5, particles with

94

aerodynamic diameter less than or equal to 2.5 µm) filter samples from source tests

95

(e.g., crop straw, wood and coal burning) conducted in the laboratory using typical

96

biomass-fired cooking stoves and coal-fired residential stoves were collected.

97

Simultaneously, offline filter-based fine particle samples were also collected during

98

winter in Beijing. Saccharides (i.e., anhydrosugars, monosaccharides, disaccharides,

99

and alcohol-sugars) were examined in both the ambient and source samples. Source

100

specific diagnostic indicators were compared for the different source types and

101

ambient samples.

102 103



MATERIALS AND METHODS

104

Ambient PM2.5 sampling. Ambient PM2.5 samples (over a 24-hr period on a daily

105

basis) were collected at an urban site located on the campus of Peking University in

106

Beijing, using a high volume sampler (VFC-PM2.5, Thermo Fisher Scientific Co., U.S.,

107

with a flow rate of 1.13 m3 min-1) and a co-located four-channel sampler (TH-16A,

108

Tianhong, China, 16.7 L min-1, 47 mm i.d. Teflon and quartz filters) from 11 to 31

109

January 2013. Quartz filters (8×10 inch, 2500 QAT-UP, Pall Corp., NY, USA) were

110

prebaked at 550℃ for 6 h in a muffle furnace and wrapped in pre-combusted

111

aluminum foil. All filter samples were stored in triple plastic bags in a refrigerator at

112

-20℃ after sampling until analysis. Field blanks were collected at the beginning and

113

end of the sampling period with the same operation procedure but without pump on.

114

Source testing and source sample collection. Source testing was performed in the 6

ACS Paragon Plus Environment

Page 6 of 36

Page 7 of 36

Environmental Science & Technology

115

laboratory of Residential Combustion Simulation at Peking University Shenzhen

116

Graduate School. The burning simulation system (see Figure 1) in the laboratory has

117

been described in previous studies. 30,31 The difference here is the combustion pan has

118

been changed to biomass-fired cooking stoves or coal-stoves respectively.

119

combustion device was equipped with a vertical chimney. Smoke emissions from the

120

chimney were then collected by a fume hood, and went through a flue duct, where

121

emission exhaust was then isokinetically withdrawn by a sampling probe and mixed

122

with zero air (15 times for crop straws, and 5 times for coals and woods) in a dilution

123

system. PM2.5 from the mixing chamber was collected on quartz (pre-baked at 550 ℃

124

in muffle furnace) and Teflon filters through four PM2.5 cyclones, each with a flow

125

rate of 16.7 L/min. Altogether, 5 types of crop straws, 8 types of wood logs, and 6

126

types of domestic coals with different geological maturity (e.g., high-volatile

127

bituminous coal (HVB), medium-volatile bituminous coal (MVB), semi-anthracite

128

(SA)) were collected from some agricultural provinces and the main coal-mining

129

regions in China, such as Shandong, Shanxi, Heilongjiang, and Hunan provinces. The

130

main characteristics of these solid fuels are presented in Table S1, and the details of

131

the proximate and ultimate analysis results of coals were shown in Table S2. Crop

132

straws, woods and coals were burned in commonly used biomass-fired cooking stoves

133

(Figure S1a) and residential coal-stoves in China (Figure S1b for bitumenite, and

134

Figure S1c for honeycomb briquette), respectively. It is noted that these residential

135

stoves were not equipped with any emission control technologies.

32

The

136

Approximately 2.35±0.26 kg crop residue, 1.94±0.39 kg wood and 1.62±0.31 kg

137

coal were weighed and burned during each testing cycle (each sample) to provide

138

enough heat to boil 15 kg water. The pre-weighed crop straws were ignited directly in

139

the stove, while woods and coals were ignited outside of the laboratory with 7

ACS Paragon Plus Environment

Environmental Science & Technology

140

alcohols/biomass and woods/burned coals, respectively. After ignition, the burned

141

woods or coals were moved to ignite the pre-weighed woods or coals in the stove,

142

respectively, in order to remove the influence of ignition materials. Each fuel was

143

tested at least twice to ensure repeatability. Before each combustion test, one 30-min

144

system blank operation was conducted, which was similar to the operation of source

145

testing but without fuel burning, to reduce the risk of cross contamination between

146

samples. A negligible level of particulate matter was found in the blank samples. All

147

data presented below were blank corrected.

148

Chemical analysis. Organic carbon (OC) and elemental carbon (EC) were

149

analyzed by the Desert Research Institute’s (DRI) Thermal/Reflectance Optical

150

Carbon Analyzer (Model 2001 A) following the thermal/optical reflectance (TOR)

151

IMPROVE A protocol. Total water-soluble organic carbon (WSOC) as well as

152

water-soluble saccharides such as anhydrosugars (LG, mannosan and galactosan),

153

monosaccharides (arabinose, galactose, glucose, mannose and xylose) and

154

alcohols/polyol- sugars (glycerol, inositol, threitol and mannitol) were measured for

155

all ambient and source samples. The details of the analytical protocol are available in

156

Sullivan et al.

157

quartz filter sample was extracted with 10 mL deionized water (DI water, >18 MΩ) by

158

sonication for 30 min, then filtered using a 0.2 µm PTFE membrane filter (Whatman

159

Inc.). A 1 mL aliquot was used for the WSOC measurement (Sievers Model 800 Turbo

160

Total Organic Carbon (TOC) Analyzer). Another 1 mL aliquot was used to analyze for

161

saccharides using a Dionex DX-500 series ion chromatograph with an ED-50

162

electrochemical detector operating in integrating amperometric mode using waveform

163

A and a GP-50 gradient pump. It should be noted that all the carbonaceous component

164

analyses for each sample were performed with the same quartz filter, and

33,34

Briefly, one 47-mm punch of each ambient and source testing

8

ACS Paragon Plus Environment

Page 8 of 36

Page 9 of 36

Environmental Science & Technology

165

water-soluble species analyses were based on the same water extractions.

166

For ambient samples, an aliquot of the water extracts was used for water-soluble

167

ion (e.g., K+, NH4+, Mg2+, Ca2+, NO3-, SO42-) analysis by ion chromatography

168

(DIONEX, ICS-2500) following the method described by Guo et al. 35 Water extracts

169

of eight ambient samples were further treated for

170

method described previously. 36,37 Briefly, the water extracts were freeze-dried and

171

re-dissolved in 150 µl hydrochloric acid to be de-carbonated, and then transferred into

172

pre-combusted silver capsules and evaporated and dried in an oven at 60 ℃. The

173

WSOC isolates were then sent to the U.S. National Ocean Sciences Accelerator Mass

174

Spectrometry (NOSAMS) facility at Woods Hole Oceanographic Institution (MA,

175

USA) for high-precision

176

fossil and non-fossil source WSOC (WSOC_F & WSOC_NF) could be found in

177

previous studies. 36, 37 Briefly, the fraction modern (Fm) was determined by comparing

178

the measured 14C/12C ratios in the blank, ambient samples, and modern standard (i.e.,

179

NBS Oxalic Acid I in AD 1950), and then converted into the fraction of the

180

contemporary carbon (Fc) by normalization with a conversion factor of 1.06, to

181

correct the ambient 14C excess due to the atmospheric nuclear bomb tests in the 1950s.

182

Finally, the fractional contributions of contemporary biomass/biogenic sources versus

183

radiocarbon-extinct fossil fuel sources were determined based on the isotopic mass

184

balance.

14

14

C analysis of WSOC with the

C-AMS analysis. The detailed calculation processes for

185

In addition, ambient samples were digested by 4.5 mL nitric acid, 1.5 mL

186

hydrochloric acid, and 0.2 mL hydrofluoric acid at 190 ℃ for 40 min, and then

187

diluted to 10 mL with DI-water after digestion and cooling, and finally 23 trace metals

188

(e.g., Al, Fe, Mn, Zn, As, Cu, Co, Se, Mo, Cr) were analyzed by inductively coupled

189

plasma mass spectrometry (ICP-MS). Details of this method could be found in 9

ACS Paragon Plus Environment

Environmental Science & Technology

190

Page 10 of 36

previous study. 38

191 192



RESULTS AND DISCUSSION

193

Direct evidence of LG from residential coal combustion emissions. In this study,

194

it was found that LG was detectable in all the tested coal combustion source samples

195

and in all crop straw and wood burning source samples. As the ignition procedures

196

utilized in the study ensured that the emissions were solely from the combustion of

197

the pure tested fuels (e.g., coal), this study provides direct evidence that residential

198

coal combustion could also be a contributor of LG.

199

In general, LG accounted for the majority of the measured sugar compound mass in

200

all kinds of source emissions (Coal: 54.7±7.8%; Crop straw: 74.2±11.1%; Wood:

201

77.4±13.3%),

202

6-anhydro-β-mannopyranose)

203

6-anhydro-β-D-galactopyranose), which are produced from pyrolysis of hemicellulose.

204

However, the saccharide profiles observed in the emissions from different solid fuel

205

combustions were distinct (see Figure S2). For example, unlike biomass burning

206

emissions, arabinose was not detectable in most of the coal combustion emissions.

207

Furthermore, the percentages of glycerol and monosaccharides (galactose, glucose,

208

mannose and xylose) were much higher in coal combustion samples compared to

209

biomass burning samples.

followed

by

its

two and

isomers,

mannosan

galactosan

(MN,

(GA,

1, 1,

210

Emission factors of source-specific LG. Emission factor (EF), expressed in

211

emitted pollutant mass per kg of fuel consumed, is calculated based on equations (1)

212

and (2).

EFf ,i = 213

m f ,i ∆mdry , f

×

Q0 × DR1 × DR2 Qt 10

ACS Paragon Plus Environment

Page 11 of 36

Environmental Science & Technology

214

(1)

Q0 = v0π ( 215 216

D0 2 ) 2

(2) Where

217

EFf ,i

m f ,i

represents EFs of species i from fuel f;

∆mdry , f

is the collected mass of

is the mass of burned fuels f (dry basis); DR1 and

218

species i from fuel f;

219

DR2

220

Qt

221

dilution tunnel and determined by equation (2);

222

and smoke velocity (m s-1) of the first dilution tunnel, respectively.

represent the dilution ratios in the first and second dilution tunnel, respectively;

is the sampling volume during the sampling time t;

D0

Q0

is the flow rate in the first

v0

represent the diameter (m)

and

223

Figure 2 shows the LG emission factors (EFLG, mg/Kg-1) for the different fuels

224

(i.e. crop straw, wood, coal). The coefficient of variation (CV) based on the repeated

225

tests for each type of fuel was within the range of 0.044-0.49, much lower than that

226

for the different fuel types (CV: 0.82-1.61), indicating a larger variation of EFLG due

227

to the diversities in the fuels themselves. The EFLG was within the range of 0.3-15.9

228

mg Kg-1 for coals, 11.7-429 mg Kg-1 for crop straws, and 5.2-147 mg Kg-1 for woods.

229

The averaged EFLG for different solid fuels ranks from high to low as crop straw (Ave.

230

± Std.: 133±125 mg Kg-1) > wood (Ave. ± Std.: 46.7±41.6 mg Kg-1) > coal (Ave. ±

231

Std.: 3.1±4.9 mg Kg-1) (see Figure 2A). That is, biomass burning sources emit more

232

LG per unit fuel weight, about 1-3 orders of magnitude higher than coal combustion

233

emissions (see Figure 2B).

234

Of different biomass fuels, rice straws collected in north China exhibited the

235

highest EFLG with an average of 326±91.6 mg Kg-1, followed by wheat straw

236

(173±59.7 mg Kg-1) and rice straw collected in south China (121±37.0 mg Kg-1), 11

ACS Paragon Plus Environment

Environmental Science & Technology

237

respectively, which were comparable to the reported EFLG of 150±130 mg Kg-1 for the

238

simulated open wheat burning conducted in the U.S. EPA burn test facility by

239

Dhammapala et al.

240

with the highest EFLG of 92.8-147 mg Kg-1 for oak and the lowest EFLG of ~5.2 mg

241

Kg-1 for locust; the average EFLG for the other woods was 29.0±5.2 Kg-1. It is noted

242

that, the two semi-anthracite

243

Shanxi province exhibited the highest EFLG for coals, with averaged EFLG value of

244

12.7±4.6 mg Kg-1 and 4.0±0.2 mg Kg-1, respectively, followed by the honeycomb

245

briquette made by anthracite from Jaingmen, Guangdong province (JM-SA, 0.63 mg

246

Kg-1). They were all higher than those from high-volatile bituminous coals mined in

247

Yanzhou, Shandong province (YZ-HVB, 0.43 mg Kg-1), and medium-volatile

248

bituminous coals mined in Jixi, Heilongjiang province (JX-MVB, 0.44 mg Kg-1) and

249

Baotou, Inner Mongolia (BT-MVB, 0.35±0.06 mg Kg-1). It is reported in previous

250

studies that LG was more likely from combustion of low-rank (low geologic maturity

251

coals.

252

maturity and LG emission, which could be examined in more details in future study.

24, 25

39

EFLG for different wood fuels had a relatively small variation,

mined in Yangquan (YQ-SA) and Jincheng (JC-SA),

This study suggests that it is not a simple relationship between coal

253

Estimated total emissions of LG from residential solid fuels in northern China.

254

Although the EFLG for coal (EFLG_coal) is the lowest among solid fuels in this study,

255

the total emission of LG from coal combustion maybe substantial as there is large coal

256

combustion in China. China is the leading coal producer in the world. 40 According to

257

the China Statistical Yearbook, coals account for 20% of the total residential energy in

258

the residential sectors in 2013. 41 Based on the same activity data for biomass and coal

259

in domestic sectors described in Ma et al.

260

measured in this study, LG emissions from domestic solid fuels were roughly

261

estimated (see Text S1). Given that most of the solid fuels tested in this study were

42

and the EFLG_coal, EFLG_crop and EFLG_wood

12

ACS Paragon Plus Environment

Page 12 of 36

Page 13 of 36

Environmental Science & Technology

262

mainly collected and used in northern China, especially coals were mined from

263

several typical mines in 4 provinces (e.g., Shanxi, Inner Mongolia, Heilongjiang, and

264

Shandong) in northern China, estimations here were only performed in northern China

265

(see Figure 3).

266

LG emissions from residential coal combustion were lower compared to those from

267

residential biomass burning in all provinces in northern China (see Figure 3A). Total

268

LG emissions from domestic coal combustion in northern China estimated here were

269

approximately 5.26±4.31 (0.42-14.39) t yr-1 vs. 1.18±0.896 (0.046-2.89) kt yr-1 for

270

domestic biomass burning emissions. That is, on the average, coal combustion

271

contributed approximately only ~1 % of the total LG emissions from the domestic

272

sector in northern China in the year of 2013. However, it should be noted that in

273

Beijing, the capital of China, the LG emission from residential coal combustion in

274

2013 totaled about 3.23 t yr-1, accounting for about 7% of the total LG emissions from

275

residential sectors, much higher than the percentages in other provinces (see Figure

276

3B and Table S3). This reflects the relatively high usage of coal (432×104 tce yr-1,

277

where tce is ton of standard coal equivalent) compared to biomass (37×104 tce yr-1)

278

in residential sectors in Beijing.42 That is, the coal usage in the residential sectors in

279

Beijing was about 12 times of the biomass, which was similar to the cases in Neimeng

280

and Shanxi provinces, where coal usage was larger than that of biomass. Besides, it is

281

noteworthy that the percentage of coal combustion contribution to total LG emissions

282

in cities in northern China (e.g., Beijing) could be much higher during the wintertime

283

due to increased coal consumption for heating and cooking.

284

Diagnostic indicators for solid fuel emissions. Ratios of LG to PM2.5 and other

285

co-emitted carbonaceous components were compared among the different solid fuels

286

and were listed in Table 1. In general, LG constituted 2.0±1.5% and 2.0±1.6% of 13

ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 36

287

emitted PM2.5 mass in crop straws and woods, respectively, consistent with previously

288

reported values ranging from 2 to 10%.

289

0.3±0.4% of the PM2.5 mass from coal combustion emissions.

290

3,39,48,49

By contrast, LG only accounted for

Ratios of LG to carbonaceous components. The LG/OC ratio, a ratio used for 6,50

291

estimating contributions of biomass burning in many previous studies,

292

0.018±0.012

293

(0.00003-0.025) (carbon/carbon basis, µgC µg-1C) in this study for crop straw, wood

294

and coal combustion products, respectively. LG/OC ratios were significantly different

295

between coal and biomass (i.e., crop straw and wood) burning emissions based on a

296

t-test (p