Light Absorption Properties and Radiative Effects of Primary Organic

Subscriber access provided by SUNY DOWNSTATE

Article

Light Absorption Properties and Radiative Effects of Primary Organic Aerosol Emissions Zifeng Lu, David G. Streets, Ekbordin Winijkul, Fang Yan, Yanju Chen, Tami C. Bond, Yan Feng, Manvendra K. Dubey, Shang Liu, Joseph P. Pinto, and Gregory R. Carmichael Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b00211 • Publication Date (Web): 26 Mar 2015 Downloaded from http://pubs.acs.org on March 29, 2015

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

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

Page 1 of 38

Environmental Science & Technology

1

Light Absorption Properties and Radiative Effects of Primary Organic

2

Aerosol Emissions

3

Zifeng Lu,*,† David G. Streets,† Ekbordin Winijkul,† Fang Yan,† Yanju Chen,‡ Tami C.

4

Bond,‡ Yan Feng,§ Manvendra K. Dubey,║ Shang Liu,║ Joseph P. Pinto,┴ and Gregory R.

5

Carmichael# †

6 7

‡

8 9 §

10 11 12 13

Energy Systems Division, Argonne National Laboratory, Argonne, Illinois 60439, United States

║

Department of Civil and Environmental Engineering, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, United States

Environmental Science Division, Argonne National Laboratory, Argonne, Illinois 60439, United States

Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States

14 15

┴

16 17

#

National Center for Environmental Assessment, U.S. EPA, Research Triangle Park, North Carolina 27709, United States

Department of Chemical and Biochemical Engineering, The University of Iowa, Iowa City, Iowa 52242, United States

18 19

Revised Manuscript Submitted to

20


21

March 20, 2015

22 23

The submitted manuscript has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory (“Argonne”). Argonne, a U.S. Department of Energy Office of Science laboratory, is operated under Contract No. DE-AC02-06CH11357. The U.S. Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.

* Corresponding Author: E-mail: [email protected]; phone: (630) 252-9853; fax (630) 252-8007.

1 ACS Paragon Plus Environment


24

ABSTRACT

25 26

Organic aerosols (OA) in the atmosphere affect Earth’s energy budget by not only scattering but

27

also absorbing solar radiation due to the presence of the so-called “brown carbon” (BrC)

28

component. However, the absorptivities of OA are not or poorly represented in current climate

29

and chemical transport models. In this study, we provide a method to constrain the BrC

30

absorptivity at the emission inventory level using recent laboratory and field observations. We

31

review available measurements of the light-absorbing primary OA (POA), and quantify the

32

wavelength-dependent imaginary refractive indices (kOA, the fundamental optical parameter

33

determining the particle’s absorptivity) and their uncertainties of the bulk POA emitted from

34

biomass/biofuel, lignite, propane, and oil combustion sources. In particular, we parameterize the

35

kOA of biomass/biofuel combustion sources as a function of the black carbon (BC)-to-OA ratio,


Page 2 of 38

Page 3 of 38


36

indicating that the absorptive properties of POA depend strongly on burning conditions. The

37

derived fuel-type-based kOA profiles are incorporated into a global carbonaceous aerosol

38

emission inventory, and the integrated kOA values of sectoral and total POA emissions are

39

presented. Results of a simple radiative transfer model show that the POA absorptivity warms

40

the atmosphere significantly and leads to ~27% reduction in the amount of the net global average

41

POA cooling compared to results using the non-absorbing assumption.

42 43 44

INTRODUCTION Atmospheric carbonaceous aerosols (i.e., black carbon, BC, and organic aerosol, OA)

45

affect Earth’s energy budget by scattering and absorbing solar radiation. While BC is commonly

46

considered to be the second most important global warming agent in the atmosphere (after only

47

CO2),1 OA is typically treated as a cooling agent that purely scatters solar radiation. However,

48

recent studies showed that OA components can also contribute substantially to light absorption.

49

In contrast to BC, which absorbs light throughout the ultraviolet (UV)-visible spectrum, the so-

50

called “brown carbon” (BrC) component in OA absorbs mostly in the near-UV and blue

51

wavelength and exhibits stronger wavelength-dependence.2 BrC has been widely observed in

52

biomass/biofuel3-18 and fossil-fuel14,19-23 combustion sources, and including its contribution in

53

climate and chemical transport models has been shown to improve the simulations of aerosol

54

light absorption significantly.24,25 BrC has been reported to be responsible for from ~20% to

55

>50% of the aerosol light absorption at UV wavelengths23-29 and thus could perturb the radiative

56

balance and influence tropospheric photochemistry.

57

Knowledge of the optical properties of primary OA (POA) emissions is essential in

58

modeling their climate effects. The complex refractive index (m=n−ik) is the fundamental

59

parameter determining the particles’ optical properties, where the real (n) and the imaginary (k)



60

parts represent the scattering and absorbing components, respectively. For example, the mass

61

absorption and scattering cross-sections (MAC and MSC) of spherical particles—the two most

62

important optical properties—can be derived from m and the particles’ size distribution through

63

Mie theory.30 In current climate and chemical transport models, light absorption properties of

64

POA are poorly represented. Most models still treat POA as non-absorbing (i.e., k values of OA,

65

kOA=0) and underestimate the total warming effect of carbonaceous aerosols. For the limited

66

studies that took into account the BrC effects, the treatments were also too simplistic to reflect

67

the variety of reported kOA values, which typically span two orders of magnitude. For example,

68

Park et al.24 calculated the light absorption coefficients of BrC over East Asia in the GEOS-

69

Chem model as a product of the simulated BC concentration, a BrC-to-BC ratio of one, and a

70

fixed MAC of 3.8 m2 g−1. Feng et al.26 treated 66% of the total POA from biofuel/biomass

71

emissions as BrC, whereas Lin et al.31 assumed that all POA from biomass, biofuel, and

72

residential coal burning were BrC. Both studies used spectrally dependent kOA from Kirchstetter

73

et al.3 and Chen and Bond7 as high- and low-absorbing scenarios in their models to evaluate the

74

influence of BrC on the global distribution of radiative forcing (RF). Clearly, better

75

parameterization of the light-absorbing properties of POA is urgently needed in models.

76

Since BrC is a complex mixture encompassing a large group of organic compounds with

77

various absorptivities and lacking a formal analytical definition,1,2 we focus on the light-

78

absorbing properties of the bulk POA instead of treating the BrC separately as a POA subset. We

79

first review the currently available kOA-related measurements and develop fuel-type-based kOA

80

spectral profiles, as well as their uncertainties, for POA emitted from both biofuel/biomass and

81

fossil-fuel combustion sources. Then these wavelength-dependent kOA profiles are incorporated

82

into a global POA emission inventory and the integrated light absorption properties of sectoral


Page 4 of 38

Page 5 of 38


83

and total POA emissions are estimated. We also show how the POA radiative effect changes

84

when accounting for the POA absorption.

85

METHODS AND DATA

86

Data Sets of Biomass/Biofuel Combustion Sources. Currently, the majority of light-

87

absorbing POA studies have focused on the biomass/biofuel combustion sources, because POA

88

emissions from these sources are highly absorptive2 and contribute ~90% of the global POA

89

budget.1,32 However, reported kOA values varied substantially among different studies due to the

90

complex and poorly understood chemical nature of the light-absorbing compounds comprising

91

POA. Recently, Saleh et al.14 found that kOA of biomass burning emissions can be parameterized

92

as a function of the BC-to-OA ratio. Following the same idea, we reviewed the available

93

measurements of absorbing OA from not only biomass burning but also biofuel combustion

94

sources and examined whether the absorption properties of OA from these sources can be linked

95

to the BC-to-OA ratio.

96

Data sets of biomass/biofuel combustion considered in this work3-18 are shown in Figure 1.

97

We used laboratory OA measurements of fresh and chemically aged emissions and field OA

98

measurements near intensive biomass/biofuel emission sources. Ambient and remote sensing

99

measurements were discarded since they may be influenced by emissions from a variety of

100

sources and by the presence of secondary organic aerosols (SOA). We also included a series of

101

well-controlled FLAME (Fire Lab at Missoula Experiment) measurements of biomass

102

smoke.12,13,17 Since the reported optical properties of these measurements were for the bulk

103

carbonaceous aerosols, only experiments with the OA fraction >95% of the total aerosol mass

104

were included. These data were utilized in Figure 1c only to show the wavelength dependence of

105

the light-absorbing OA in a wide range of BC-to-OA ratios.

106

Figure 1



Page 6 of 38

We assumed that the kOA has a power-law wavelength dependence14,33,34 and can be

107 108

expressed as:

109

kOA,  k OA,550  550 /  

110

where kOA,550 is kOA at the wavelength (λ) of 550 nm and w is the wavelength dependence of kOA.

111

For those studies that measure the OA absorption coefficients/efficiencies, w is often described

112

using the absorption Ångström exponent (AAE), where AAE=w+1 for small particles.14 We

113

made the following corrections/calculations to reported results not giving kOA,550 and/or w

114

directly:

115



w

(1)

for studies reporting the kOA (or absorption coefficients/efficiencies) at multiple

116

wavelengths, w (or AAE) was determined from the linear regression fit of log(kOA) (or

117

log(absorption coefficients/efficiencies)) versus log(λ);

118



reported or calculated AAE values were converted to w by subtracting one;

119



for kOA values reported at wavelengths other than 550 nm, kOA,550 values were calculated

120 121

using Eq (1); 

for studies measuring the density-normalized absorption coefficient (σliquid=α/ρ) from

122

organic liquid (e.g., methanol and acetone) extracts of POA, kOA was calculated by:34

123

kOA,     liquid,   / 4

124

where ρ is the density of the dissolved compounds and was fixed to 1.2 g cm−3.35 Note

125

that σliquid is different from the particulate MAC although they have the same units of m2

126

g−1;34

127 128



(2)

kOA,550 and w of Kirchstetter et al.’s3 results were corrected to 0.022 and 3.3, respectively (S1 in the Supporting Information, SI);


Page 7 of 38

129 130 131




when converting the mass of OC (organic carbon) to OA, an OA-to-OC ratio of 2.0 was used, as suggested by Turpin and Lim35 for biofuel/biomass burning sources. Monte Carlo Simulations of Ångström Attribution of POA Absorption for Lignite

132

Combustion Sources. Although most of the reported kOA values were focused on

133

biomass/biofuel burning sources, some studies did show that POA emissions from

134

coal,19,20,23,36,37 oil,14,21 and propane22 combustion contain light-absorbing components. We focus

135

on coal in this section and show the treatments of others in the next section.

136

Bond et al. measured the light-absorbing properties of carbonaceous aerosols emitted from

137

a small industrial boiler burning lignite19,37 and residential stoves burning both bituminous coal

138

and lignite.20 They found that light absorption by particles from these low-temperature coal

139

combustion sources exhibited a strong spectral dependence with AAE up to 2.9. They

140

hypothesized that a spectrally dependent k of co-emitted OA is the most plausible explanation of

141

this phenomenon.36 Unfortunately, they did not provide absorption properties of OA, but only of

142

the bulk carbonaceous aerosols; and to our knowledge, there is currently no study reporting the

143

kOA directly for coal combustion sources. Here, we applied the widely-used absorption

144

attribution method3,11,27,38 to the measurements of Bond et al. for the lignite-burning industrial

145

boiler19, in order to separate the light absorption by OA from the total aerosol absorption, and

146

retrieved the kOA values through the Mie calculation. This particular dataset was chosen because

147

the emissions were well characterized, and most of the relevant information can be found in a

148

series of subsequent publications.37,39-41

149

The experiments of Bond et al. were conducted at a lignite-burning plant in Leipzig,

150

Germany. MACs for the bulk submicrometer particles were given at four wavelengths (450, 550,

151

750, and 1000 nm). We assumed that BC and POA were the only light absorbers in the emitted

152

particles. Therefore, the measured MAC at λ can be expressed as 7 ACS Paragon Plus Environment


Page 8 of 38

153

MAC  f BC  MACBC,  f POA  MACPOA,

154

where fBC and fPOA are the mass fractions of BC and POA in the total submicrometer particles,

155

respectively. The light absorption was assumed to be attributed solely to (“encapsulated”) BC at

156

1000 nm,

157

MAC1000  f BC  MACBC,1000

158

Assuming that the spectral absorption of (“encapsulated”) BC follows the power law relationship

159

(  AAE BC )38,42, MACBC at a shorter wavelength is given by:

160

MACBC,      MACBC,1000  1000 

161

Combining Eq (3)−(5), MACPOA at λ was determined as 1

(3)

(4)

 AAEBC

(5)

 AAEBC       MAC    MAC1000    1000   

162

MACPOA, 

163

We then performed Mie calculations and optimized the kOA,λ until the Mie-predicted MACPOA,λ

164

matched the one estimated by Eq (6). Finally, w was determined from the linear regression fit of

165

log(kOA,λ) versus log(λ), and kOA over the whole spectrum was derived from Eq (1).

166

f POA

(6)

We used a Monte Carlo approach to simulate the above processes and determine the

167

uncertainties. The Monte Carlo method is a problem-solving technique that approximates the

168

probability of certain outcomes by running multiple simulations with random variables. In this

169

work, we assigned appropriate probability distributions to all input parameters and performed

170

10,000 simulations. Unless specified otherwise, the term “uncertainty” in this article refers to

171

one standard deviation (±1σ) or the coefficient of variation (CV, σ divided by the mean)

172

expressed as a percentage. We applied normal distributions for MAC measurements with

173

uncertainties of 30% provided by Bond et al.19 As a key input to the Mie model, POA size


Page 9 of 38


174

distribution was assumed to be similar to the bulk emission with a count mean diameter (CMD)

175

of 50±5 nm and a geometric standard deviation (GSD) of 1.45±0.05.39-41 We ignored both the

176

ultrafine (CMD ~4 nm) and the accumulation mode (CMD ~350 nm) detected in the

177

experiments, since particles in these two modes together accounted for

Light Absorption Properties and Radiative Effects of Primary Organic

Recommend Documents