Subscriber access provided by - Access paid by the | UCSB Libraries
A: Environmental, Combustion, and Atmospheric Chemistry; Aerosol Processes, Geochemistry, and Astrochemistry
Effects of Gas-Particle Partitioning on Refractive Index and Chemical Composition of m-Xylene Secondary Organic Aerosol Kun Li, Junling Li, Weigang Wang, Jiangjun Li, Chao Peng, Dong Wang, and Maofa Ge J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.7b12792 • Publication Date (Web): 12 Mar 2018 Downloaded from http://pubs.acs.org on March 13, 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 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Effects of Gas-Particle Partitioning on Refractive Index and Chemical Composition of m-Xylene Secondary Organic Aerosol
1 2 3 4 5
Kun Li,†,‡,ǁ Junling Li,†,‡ Weigang Wang,*,†,‡ Jiangjun Li,‡,§ Chao Peng,†,‡ Dong
6
Wang,‡,§ and Maofa Ge*,†,‡
7 8 9 10 11 12 13
†
State Key Laboratory for Structural Chemistry of Unstable and Stable Species, CAS
Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China. ‡
University of Chinese Academy of Sciences, Beijing 100049, P. R. China.
§
Key Laboratory of Molecular Nanostructure and Nanotechnology, Institute of
Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China.
14 15 16
Abstract
17
The formation of secondary organic aerosol (SOA) contains partitioning processes of
18
the oxidation products between gas and particle phase, which could change the
19
particle-phase composition when particles grow. However, the effects of these
20
processes on the optical properties of SOA remain poorly understood. In this study,
21
we performed smog chamber experiments to investigate the effects of gas-particle
22
partitioning (GPP) on the refractive index (RI) and chemical composition of the
23
m-xylene SOA. Here, we show that the GPP processes, as organic mass increases, can
24
increase the proportions of semi- and intermediate volatile organic compounds
25
(SVOCs and IVOCs) in the particle phase and result in the decrease of SOA RI real
26
part for 0.09 ± 0.02 (without seeds) and 0.15 ± 0.02 (with seeds). This indicates that
27
the SOA optical properties are closely related to the total organic mass and
28
molecular-level composition. In addition, the presence of inorganic seeds promotes
29
the GPP to the particle phase and hence further decreases the RI real part for 0.05 ±
30
0.02. As pre-existing aerosols are ubiquitous in the ambient atmosphere, it is
31
suggested that there should be a certain correction when applying the SOA RI of
32
previous laboratory studies to air quality and climate models.
33
ACS Paragon Plus Environment
The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
34
Page 2 of 28
Introduction
35
As an important type of atmospheric fine particulate matter, secondary organic
36
aerosol (SOA) plays a vital role in visibility, climate change, and human health.1-2 The
37
formation and evolution of SOA were the foci of atmospheric chemistry in the past
38
few decades.3-8 During the formation of SOA, gas-particle partitioning (GPP, i.e.
39
condensation and evaporation between gas and particle phase) was found to be a
40
traditional and dominant approach.9-12 The partitioning between gas and particle phase
41
of a component i is given by: ∗
42
= 1 + ; = ∑
43
where is the partitioning coefficient, which means the fraction in particle phase,
44
∗ is the effective saturation concentration of the compound, is the
45
concentration of organic aerosol, and is concentration of the compound (in both
46
gas and particle phase).5,
47
increase as well, which will lead to the higher partitioning coefficient according to
48
the equation, so Equation (1) should be applied iteratively. One consequence of this
49
partitioning process is that the organic aerosol composition and oxidation state will be
50
changed as the total organic concentration increases, which has been proved by
51
laboratory chamber studies.13-14 It may have important impacts, because the changing
52
composition could influence the physical properties such as hygroscopic and optical
53
coefficients, which are important parameters related to visibility and aerosol climate
54
effects.
11-12
(1)
If the organic concentration increases, will
55
Due to the complex compositions and various mixing states, the optical properties
56
of SOA and their effects on global radiative balance remain poorly quantified.1, 15-16
57
Refractive index (RI) is one of the most important factors of aerosol optical
58
properties.16 As other optical factors like scattering and extinction efficiencies can be
59
calculated using the RI, quantitatively understanding the RI of SOA is essential.15-17
60
In recent years, some laboratory studies have been carried out to determine the RIs of
61
SOA.17-28 The studied SOA precursors contain both biogenic (monoterpenes, isoprene
62
etc.) and anthropogenic (toluene, xylene, alkanes etc.) volatile organic compounds
63
(VOCs). Some of these studies find that the SOA RI changes as organic mass loading
64
increases (i.e. particles grow).17, 19 Kim et al. found that the RI of limonene SOA
65
increased as particles grow, while the RI of α-pinene SOA didn’t change obviously.19
ACS Paragon Plus Environment
Page 3 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
66
Li et al. found that SOA generated from m-xylene decreased dramatically among the
67
four studied aromatic precursors.17 However, the driving force of this kind of
68
changing in optical properties is still unclear.
69
In addition, when investigating the refractive index of SOA, previous studies
70
mostly focus on the SOA particles in the absence of aerosol seeds,15-26 though it has
71
been found that seeds could influence the GPP process and enhance the SOA
72
yields.29-31 Pre-existing aerosols can provide the surface for organic materials to
73
condense on, and increase the proportion of more volatile compounds in the particle
74
phase. They are ubiquitous in the ambient atmosphere; hence the impacts of
75
pre-existing aerosol seeds on SOA optical properties should be further studied. A
76
recent study investigated the effects of seeds on optical properties of SOA derived
77
from long-chain alkanes, a type of intermediate VOC (IVOC), attempting to relate the
78
chemical composition to RIs.27 However, as the more classical and widely studied
79
precursors of anthropogenic SOA, the aromatics, the effects of aerosol seeds on
80
optical properties of their SOA remain unknown.
81
In this study, we performed smog chamber experiments to investigate the effects
82
of the GPP process with and without inorganic seeds on the RI of SOA generated by
83
photo-oxidation
84
High-resolution electrospray ionization mass spectrometry was used to analyze the
85
molecular composition of the SOA particles. By combining the high-resolution
86
chemical analysis with optical detection, we try to find the relationship between
87
chemical composition and RI. We show the impacts of GPP and pre-existing
88
inorganic seeds on composition and RI of m-xylene SOA, and discuss their
89
atmospheric implications.
of
m-xylene,
a
typical
anthropogenic
aromatic
precursor.
90 91
Experimental
92
The experiments discussed here have been described in detail in a previous
93
publication.28 Briefly, the experiments were performed with a dual-reactor smog
94
chamber. The combined light sources including 340 nm broadband lamps and 365 nm
95
narrowband black lamps are used.32 The temperature was maintained at 293 ± 1K.
96
The RH in the reactor was controlled at < 5%. The size distribution of aerosol was
97
measured by the scanning mobility particle sizer (SMPS; Model 3936, TSI). The NOx
98
and O3 concentration were monitored by trace gas analyzers (T200UP and T400,
ACS Paragon Plus Environment
The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
99 100
Page 4 of 28
Teledyne API). The concentration of m-xylene was measured by the quadrupole proton transfer reaction mass spectrometry (PTR-MS; HS500, Ionicon).
101
Zero air was injected into the chamber before each experiment. NOx was
102
introduced into the chamber from a standard gas cylinder, and the concentration in the
103
reactor was 300 ± 50 ppb. A known volume of m-xylene (≥ 99%, Sigma-Aldrich) was
104
injected into a glass U-tube, and then it was introduced into the chamber with the zero
105
air flow. The concentration of m-xylene in the reactor was around 4 ppm. For the
106
seeded condition, the atomizer which contained inorganic salts (namely ammonium
107
sulfate, AS and sodium chloride, NaCl; ≥ 99%, Sigma-Aldrich) solution was used to
108
generate the inorganic aerosols. The seed particles were dried by a diffusion dryer and
109
then introduced into the reactors without size selecting. The seeds are solid under the
110
< 5% RH in the chamber, which can exclude the influence of liquid phase reactions
111
on compositions and optical properties.28 The number concentration of seeds in the
112
reactors is (1.6 ± 0.3) × 104 cm-3. The non-sizing particles have a mode of 63 ± 4 nm,
113
with geometric standard deviation (σg) of 1.8 ± 0.05. The UV lights were turned on 30
114
min after the NOx, m-xylene, seeds were introduced into the chamber to make sure
115
that they are well mixed. The representative reaction profiles are shown in Fig. S1.
116
The light extinction and RIs of aerosols are detected through a custom-built
117
cavity ring-down spectrometer (CRDS), see previous work for details.33 The CRDS
118
provides the decay time τ directly, and the RI is gained by programs based on Bohren
119
and Huffman.34 The detailed process and method are introduced elsewhere.17 Briefly,
120
The RI of each diameter is calculated by minimizing the following merit function (χ): = (, − ,!"!" (#, $))&
(2)
121
where n is the real part of RI, k is the imaginary part of RI, and Qext is the extinction
122
efficiency. Qext,measured is determined using the following equation:
123
, = (
'
+
( −, )
)*) !" ,
-
(3)
124
where Stot is the total surface area that measured by SMPS, L is the length of the
125
cavity, c is the velocity of light, l is the length filled with aerosol and τ0 is the decay
126
time when the cavity is filled with zero air. This paper and a previous publication28
127
share the same dataset of SOA optical properties. While that paper focuses on effects
128
of multiphase reactions, this study mainly investigates the effects of GPP process.
ACS Paragon Plus Environment
Page 5 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
129
The morphology of m-xylene SOA particles was determined via atomic force
130
microscope (AFM). The aerosol particles were collected by impacting on silicon
131
wafers and then were analyzed by a tapping mode AFM (NT-MDT Ntegra) at room
132
temperature (~298 K) and ambient RH (~30%). The SOA particles were also collected
133
by PTFE membrane filters with a pore size of 0.22 µm (FGLP047, Millipore), and
134
then were dissolved in 5 mL methanol. The solution was then analyzed by
135
ultra-high-resolution electrospray ionization time-of-flight mass spectrometry
136
(UHR-ESI-TOF-MS; Impact HD, Bruker) in positive ion mode to get the mass
137
spectra.
138 139
Results and discussions
140
The morphology and structure of aerosols in the presence of inorganic seeds are
141
important to retrieve the RI, as particle shape and mixing state can influence the
142
application and calculation of Mie theory. The studied inorganic salts are not spherical,
143
which may introduce some deviation when applying Mie theory. Hence, we at first
144
investigate the morphology and mixing state of SOA in the presence of inorganic
145
seeds.
146
The AFM image of SOA particles in the presence of AS seeds is illustrated in Fig.
147
1. It is indicated that the particles are in the core-shell structure, of which the organic
148
coating is very soft and was impacted to be a very thin layer on the silicon wafer
149
(height < 50 nm). As a result, we speculate that the organic aerosols are likely liquid,
150
and they have an approximately spherical shape when suspended in the air.17, 35 This
151
finding suggests that there is little deviation when using the Mie theory to calculate RI
152
of seeded SOA, which is similar to our previous study about the non-seeded aromatic
153
SOA.17
154
ACS Paragon Plus Environment
The Journal of Physical Chemistry
c Height (nm)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 28
200
section 1
section 2
100
0 0
155
300
600
900
1200
1500
0
300
600
900
1200
1500
Width (nm)
156
Figure 1 AFM images of the m-xylene SOA sample in the presence of AS seeds. a. Height image
157
with two cross sections; b. phase image, in which core-shell can be observed; c. two cross sections
158
in the height image.
159 160
Same as our previous studies, negligible light absorption was found for m-xylene
161
SOA at 532 nm, so only the real part of RI are considered here.17, 28 For brevity, “RI”
162
represents the real part in the following text. The RIs of seeded experiments are
163
retrieved by Mie theory, assuming that the particles are in the core-shell structure.
164
However, as the seed is much smaller than the whole particle, there seems to be little
165
differences between core-shell assumption and homogeneous assumption (see Fig.
166
S2). The RIs of m-xylene SOAs under non-seeded and seeded conditions are shown in
167
Fig. 2. It is found that the SOA is generated earlier in the presence of seeds, which
168
agrees with previous studies.29 The RIs decrease as time goes on, which is the same
169
under both non-seeded and seeded conditions. In addition, at the same reaction time,
170
the RIs are smaller for seeded experiments compared with non-seeded experiments. It
171
should be noted that the RI data in the companion paper28 are from the period when
172
RIs are nearly stable, in this case after 200 min.
173
These findings may be due to the following reasons. Particle nucleation occurs in
174
the absence of seed, and according to the Köhler theory,36 the gaseous components for
175
nucleation need relative lower volatility than heterogeneous condensation on the
176
surface of seeds. Under non-seeded condition, longer gas-phase oxidation period is
177
needed to form lower volatile organic compounds that nucleate to form new particles,
178
so the particles are formed later. Previous studies show that the presence of
179
pre-existing seeds promotes the condensation of organic vapors.29-30 Therefore, at the
180
same reaction time, the particle diameter under the seeded condition is throughout
181
larger than the diameters under the non-seeded condition and the aerosol mass is also
182
larger (see the color map in Fig. 2). Hence, products with higher volatility could
183
condense on the particles under the seeded condition. Likewise, as reaction goes on,
ACS Paragon Plus Environment
Page 7 of 28
184
total organic concentration increases and particles getting larger, so more products
185
with higher volatility will condense on the particles. The higher volatility compounds
186
possibly have lower RIs, which lead to the lower bulk RIs of SOA under the seeded
187
condition, and the decrease of RIs as diameter increases. In addition to the reason
188
above, the ongoing oxidation and aging process may change the chemical
189
composition of the particles as well, which is another possible explanation for the
190
decreased RI with increasing photochemical time.
Non-seeded 1# Non-seeded 2# Seeded 1# Seeded 2# Non-seeded fit Seeded fit
1.60
Real part of SOA RI
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
1.55
1.50
Dsm (nm) 420
400.0
280
280.0
140
160.0
1.45
1.40 100
150
200
250
300
191
t (min)
192
Figure 2 Refractive indices (RIs) of SOA under non-seeded and seeded conditions. The X-axis is
193
the time after lights were turned on. Particle surface mean diameters (Dsm) are in the range of
194
140-420 nm and are shown in the color map.
195 196
To verify the hypothesis above, the SOA particles are collected by filters and are
197
analyzed by ultra-high-resolution electrospray ionization time-of-flight mass
198
spectrometry (UHR-ESI-TOF-MS). The particles are collected at two different times
199
for both seed conditions: the first one is when SOA RIs are just detectable and
200
particles are relatively small (~110 min for seeded experiments; ~170 min for
201
non-seeded experiments); another one is at the end of an experiment when the
202
particles stop growth (~300 min). The mass spectra of these four SOAs are shown in
203
Fig. 3. In general, the mass spectra are similar at different sizes, no matter for
204
non-seeded or seeded experiments. However, if carefully analyzing the mass spectra
205
(Fig. S3), one will find that the bigger particles have more abundant species with
206
lower molecular weight (mostly < 200 Da). In addition, the mass spectra are very
207
different between non-seeded and seeded experiments, especially for the products
ACS Paragon Plus Environment
The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
208
with higher molecular weight (> 400 Da). The non-seeded SOA contains more species
209
with molecular weight of > 400 Da, which are likely the highly oxygenated
210
multifunctional compounds (HOMs) formed during new particle formation.37-38
211
Though HOMs have been observed mainly from the oxidation of biogenic VOCs,37-40
212
recent studies show that they could also be formed from aromatics.41-42 It is noticed
213
from Fig. 2 and Fig. S1 that the particles grow about 1 h later under non-seeded
214
conditions, during when the HOMs may be generated by multi-step gas phase OH
215
oxidation (i.e. autoxidation).41, 43 For seeded experiments, the large molecular weight
216
products of > 400 Da are much less than those of non-seeded experiments, even
217
negligible. The reason may be that the semi- or low volatile compounds mostly
218
condensed into particle phase in the presence of seeds before they could be oxidized
219
to HOMs in the gas phase.
220
221 222
Figure 3 Mass spectra of SOA under non-seeded and seeded conditions with different diameters. a.
223
the non-seeded condition, of which the black and red sticks are at the onset of particles and the
224
end of the experiments, namely ~120 nm and ~250 nm. b. the seeded condition, of which the
ACS Paragon Plus Environment
Page 8 of 28
Page 9 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
225
black and red sticks are at the onset of particles and the end of the experiments, namely ~150 nm
226
and ~390 nm. The enlarged view of this figure is shown in Fig. S3.
227 228
To clearly see the change of chemical compositions when particles grow or when
229
with pre-existing seeds, the subtraction plots of mass spectra are shown in Fig. 4.
230
These three MS are the differences between large particles and small particles under
231
non-seeded condition (a); the differences between large particles and small particles
232
under seeded condition (b); and the differences between seeded and non-seeded
233
particles at the end of the experiments (c). As discussed above, seeded SOA has more
234
low molecular weight products and less high molecular weight products (Fig. 4c). It is
235
also found that the most obvious difference when particles grow is the molecules of
236