Effects of seed aerosols on the growth of secondary organic aerosols from the photooxidation of toluene.

Abstract

Hydroxyl radical (.OH)-initiated photooxidation reaction of toluene was carried out in a self-made smog chamber. Four individual seed aerosols such as ammonium sulfate, ammonium nitrate, sodium silicate and calcium chloride, were introduced into the chamber to assess their influence on the growth of secondary organic aerosols (SOA). It was found that the low concentration of seed aerosols might lead to high concentration of SOA particles. Seed aerosols would promote rates of SOA formation at the start of the reaction and inhibit its formation rate with prolonging the reaction time. In the case of ca. 9000 pt/cm3 seed aerosol load, the addition of sodium silicate induced a same effect on the SOA formation as ammonium nitrate. The influence of the four individual seed aerosols on the generation of SOA decreased in the order of calcium chloride>sodium silicate and ammonium nitrate>ammonium sulfate.

Full text

Page 1

Journal of Environmental Sciences 19(2007) 704–708
Effects of seed aerosols on the growth of secondary organic aerosols from the
photooxidation of toluene
HAO Li-qing, WANG Zhen-ya, HUANG Ming-qiang, FANG Li, ZHANG Wei-jun∗
Laboratory of Environmental Spectroscopy, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China.
E-mail: hlqing@aiofm.ac.cn
Received 6 July 2006; revised 16 August 2006; accepted 1 September 2006
Abstract
Hydroxyl radical (.OH)-initiated photooxidation reaction of toluene was carried out in a self-made smog chamber. Four individual
seed aerosols such as ammonium sulfate, ammonium nitrate, sodium silicate and calcium chloride, were introduced into the chamber
to assess their influence on the growth of secondary organic aerosols (SOA). It was found that the low concentration of seed aerosols
might lead to high concentration of SOA particles. Seed aerosols would promote rates of SOA formation at the start of the reaction and
inhibit its formation rate with prolonging the reaction time. In the case of ca. 9000 pt/cm3seed aerosol load, the addition of sodium
silicate induced a same effect on the SOA formation as ammonium nitrate. The influence of the four individual seed aerosols on the
generation of SOA decreased in the order of calcium chloride>sodium silicate and ammonium nitrate>ammonium sulfate.
Key words: toluene; photooxidation; secondary organic aerosol; seed aerosol
Introduction
Ambient aerosol can adversely affect atmospheric vis-
ibility, the climate and human health. Its source includes
the primary aerosol from direct emissions and secondary
aerosols through secondary processes. The secondary or-
ganic aerosol (SOA) is a key component in secondary
aerosols and accounts for as much as 50% of the total
aerosol mass. It is produced mainly through the.OH ini-
tiated photooxidation reaction, acid catalyzed reaction on
the surface of particle, and gas/particle partitioning or self-
nucleating of semi-volatile reaction products (Kleindienst
et al., 1999; Edney et al., 2005; Wang et al., 2005).
The SOA formation is related to the primary aerosol,
which is one of seed aerosols and can act as the absorption
or adsorption centers of SOA formation. The number size
distribution of these seed aerosols is the dominant factor
for condensation and coagulation process of the reaction
products, and the surface area of the particle can also play
an important role in the adsorption and catalytic reactions
(Oh and Andino, 2000). Over the last decade, laboratory
studieshavebeenconductedtoassesswhetherthepresence
of pre-existing submicron seed aerosols can affect the
SOA formation. Cocker’s studies indicated the presence of
ammonium sulfate seed aerosol with little relative humility
(RH) variation has a negative impact on the SOA yield
Project supported by the National Natural Science Foundation of China
(No. 20477043) and the Knowledge Innovation Program of Chinese
Academy of Sciences (No. KZCX2-SW-H08). *Corresponding author.
E-mail: wjzhang@aiofm.ac.cn.
(Cocker et al., 2001). His conclusion disagreed with the
results by Oh and Andino (2000) and by Koehler et al.
(2004). Under the condition of the high RH, Edney et al.
(2000) suggested that the presence of water on the seed
aerosol did not significantly affect the SOA yield, which
was also discrepant with the results of Cocker et al. (2001).
Besides, in the atmosphere, there are some seed aerosols
with different concentration and constituents. However,
studies as regards how these seed aerosols affect SOA
formation are less reported at present.
In this article, the effects of concentration and com-
position of four seed aerosols such as (NH4)2SO4,
NH4NO3, Na2SiO3 and CaCl2 on the SOA formation
were studied respectively during the UV-irradiation of
toluene/CH3ONO/NO/air mixtures in a self-made smog
chamber. The experiments were conducted using pre-
existed seed aerosols inside the smog chamber to simulate
the SOA formation. The aim of this study was to determine
the extent to which the SOA yields were affected by com-
positions and concentrations of seed aerosols as previously
mentioned.
1 Experimental
1.1 Materials
Toluene (>99%) was obtained from the Sigma-Aldrich
Chemistry Corporation, Germany; nitrogen oxide (99.9%)
was purchased from the Nanjing Special Gas Factory,
China. Purity of agents of (NH4)2SO4, NH4NO3, Na2SiO3

Page 2

No. 6Effects of seed aerosols on the growth of secondary organic aerosols from the photooxidation of toluene705
and CaCl2(Shanghai Chemical and Medical Corporation,
China) was >99.7%. Methyl nitrite (CH3ONO) was syn-
thesized by dropping sulfuric acid into a methanol solution
of sodium nitrate (Hao et al., 2005; Wang et al., 2006) and
collected by a condenser of liquid nitrogen at 77 K.
1.2 Experimental methods
Experiments were performed in an 850-L sealed col-
lapsible polyethylene smog chamber. Its ratio of surface
to volume was 5.8 m−1. Around the chamber, there was
equipped with 12 sets of 40-W fluorescent black lamps
symmetrically that provide radiation in the 300–400 nm
region. The bag and lamps were housed inside a highly
reflective enclosure to enhance light distribution.
Prior to the start of experiment, the chamber was contin-
uously flushed with purified laboratory compressed air for
40 min. The compressed air was processed through three
consecutive packed-bed scrubbers, which contain activated
charcoal, silica gel and a Balston DFU?filter (Grade BX)
respectively, to remove trace hydrocarbon compounds,
moisture and particles. The four individual seed aerosols
were generated by aspirating 0.0067 mol/L aqueous solu-
tion through a stainless steel, constant rate atomizer (TSI
3076, USA), respectively. The mean diameter of each kind
of seed aerosols was approximately 80 nm. In all cases,
the seed aerosols were passed through a diffusion dryer
(TSI 3062) with exiting RH lower than 20%, followed
by an85Kr neutralizer (TSI 3012) before entering the
chamber. After obtaining the desired initial seed particles
concentration, 1.35 µl/L toluene, 24.0 µl/L CH3ONO, 2.0
µl/L NO were injected into the chamber, mixed with the
pre-existing seed aerosol. And again the chamber was
filled with the purified air to 850 L full volume. Turn on
four black lamps and initiate the photooxidation reaction.
Hydroxyl radicals will be generated by the photolysis
of methyl nitrate in air at wavelengths longer than 300
nm (Hao et al., 2005; Wang et al., 2006). The chemical
reactions leading to the formation of the.OH radical are as
follows:
CH3ONO + hν (> 300 nm) −→ CH3O.+ NO
CH3O.+ O2−→ HCHO + HO2.
HO2.+ NO −→.OH + NO2
.OH-initiated toluene photooxidation reaction will lead
to the formation of the SOA particles. During the course of
reaction, the number concentration of SOA particles was
detectedbyanaerodynamicparticlesizespectrometer(TSI
APS 3321) quickly.
(1)
(2)
(3)
1.3 Datum processing methods
Similar experimental procedures for the four individual
seed aerosols were designed to assess their influence on the
formation of SOA and the concentration of SOA particles
were measured by APS 3321directly. Figs.2a, 3a and 4a
show the growth of SOA particles as the functions of reac-
tion time. The model lines through the data were chosen
to fit the data points of measurement using Sigmoidal-
Fit method. The curves in Figs.2b, 3b and 4b represent
the incremental rate of SOA particles concentration as
functions of reaction time, and they were obtained by
making the first derivative of the corresponding curves in
Figs.2a, 3a and 4a, respectively.
2 Results
2.1 Size distribution of secondary organic aerosols
Fig.1 shows the size distributions of SOA in the pres-
ence/absence of the four individual seed aerosols. These
aerosol particles are distributed in the range of 0.5–1.4 µm
and belong to ultra fine particles, which can affect health
problemsduetohumans’inhalationofhighparticlematter.
Fig. 1 Size distribution of SOA particles from toluene photooxidation
reaction. Reaction time: 60 min; seed concentration: cv. 9000 pt/cm3;
dN/dlgDp displays differential or normalized particle size distribution,
normalized to one decade of particle size.
2.2 Effect of (NH4)2SO4seed aerosol on the SOA for-
mation
Fig.2 shows the influence of (NH4)2SO4seed aerosol
on the formation of SOA particles at the level of three
concentrations of particles. By comparison, the SOA for-
mation can be affected by (NH4)2SO4seed aerosol in three
different ways:
Firstly, the presence of seed aerosol shortens the time
to reach the gas-particle partitioning equilibrium. In the
absence of (NH4)2SO4 seed aerosol, it takes 197 min
to reach the gas-particle partitioning equilibrium between
gas/particle phases with the ultimate aerosol concentration
maintaining 2128 pt/cm3. In the presence of ca. 27000
pt/cm3seed aerosol, the period to reach such equilibrium
status is 135 min, and the lower the number concentra-
tion of seed aerosol is, the longer the time to reach an
equilibrium state is. Secondly, the presence of (NH4)2SO4
seed aerosol can accelerate the growth rate of SOA at
the initial stage of the photochemical reaction, followed
by decreasing the growth rate of SOA with prolonging
reaction time. In Fig.2b, the growth rate of SOA is faster
in the initial 45 min than that one in the absence of seed
aerosols, which indicates that there is a stimulating effect
of seed aerosol on the SOA formation. After 45 min, SOA
particle concentration grows at a lower rate in the presence
than in the absence of seed aerosol, which shows that
there is an inhibition effect of the seed aerosol on the

Page 3

706HAO Li-qing et al. Vol. 19
Fig. 2 Impact of ammonium sulfate seed aerosol on the formation of secondary organic aerosol. (a) growth curve; (b) increment curve.
SOA formation. Similar results for the ca. 18000 pt/cm3
seed aerosol experiments are shown in curve (3) of Fig.2b.
The SOA particles grow faster initially, then undergo a
35 min period of growth and ultimately decrease. For the
case of ca. 27000 pt/cm3seed aerosol, the stimulating time
is only 25 min as shown in curve (4). Lastly, the finial
equilibrium number concentration of SOA particle in the
presence of seed aerosol is lower than that in the absence
of seed aerosol. The finial SOA particle concentrations
will become lower as the number concentrations of seed
aerosols are higher, which indicates that the presence
of (NH4)2SO4 seed aerosol will reduce the amount of
SOA and the degree of reduction is dependent on the
concentrations of seed aerosols.
2.3 Effect of CaCl2seed aerosol on the SOA formation
Fig.3 shows the influence of CaCl2 seed aerosols on
the SOA formation. It can be seen that in the present
experiments, the presence of CaCl2seed shorten the time
to reach the gas/particle equilibrium state, and the CaCl2
seed aerosol also acts as an accelerator in the process
of SOA formation at the initial stage of the reaction and
inhibits the formation of SOA particles while the reaction
keeps on. When the number concentration of CaCl2seed
aerosol is low, its stimulating effect on the SOA formation
will be much larger. These results are in good agreement
with those of (NH4)2SO4seed.
2.4 Potentials of four individual seed aerosols in the
SOA formation
The effects of four individual seed aerosols on the
SOA formation are shown in Fig.4. In each curve, the
particle number concentration is increased as a function
of reaction time in the presence and absence of seed
aerosol, respectively. The primary number concentration
of each kind of the seed aerosol was ca. 9000 pt/cm3.
It can be seen that CaCl2 seed has the most prominent
stimulating effect on the SOA formation. The effect of
sodium silicate on the SOA formation was the same as the
ammonium nitrate seed aerosol. The stimulatory effect of
seed aerosol on the SOA formation increased in the order
of CaCl2>NH4NO3/Na2SiO3>(NH4)2SO4.
3 Discussion
3.1 Dependence of the formation of SOA on seed
aerosols
The experimental results show that all the four seed
aerosols have effects on the formation of SOA from the
oxidation products of toluene. These can be interpreted
using the expression for the overall SOA yields from the
.OH initiated photooxidation reaction of toluene and some
subsequent reactions are as follows (Odum et al., 1997):
YSOA=
?
i
Yi= Mo
?
i
(
αiKi,om
1 + Ki,omMo) (4)
Where, Yi and YSOA are the yield of an individual
product and the total SOA yield respectively; αi is the
proportionality constant relating the concentration of ROG
that reacts to the concentration of product i (Ci) that is
formed; Ki,om is the gas-particle partitioning coefficient,
which is inversely proportional to the compounds vapor
Fig. 3 Impact of calcium chloride seed aerosol on the formation of secondary organic aerosol. (a) growth curve; (b) increment curve.

Page 4

No. 6Effects of seed aerosols on the growth of secondary organic aerosols from the photooxidation of toluene707
Fig. 4 Impact of four individual seed aerosols on the formation of secondary organic aerosols. (a) growth curve; (b) increment curve.
and pressure, and Mois the absorbing organic aerosol mass
concentration, including the organic/inorganic material
and associated water present in the aerosol phase.
For model Equation (1), it has two limiting conditions
based on the value for Ki,omMo. Firstly, it is suggested
that for low organic mass concentration and for products
that have relatively small partitioning coefficients, that is
Ki,omMo?1, then the SOA yield will be directly propor-
tional to the total aerosol organic mass concentration Mo:
YSOA= Mo
?
i
(αiKi,om) (5)
secondly, for very non-volatile products and/or for large
organic mass concentration , that is Ki,omMo?1, the indi-
vidual product yields will be independent of the organic
mass concentration and will be equal to αi, then,
YSOA=
?
i
αi
(6)
In the absence of seed, Mois very small at the initial
stage of the photooxidation reaction of toluene, that is,
Ki,omMo?1, the low-yield of SOA will lead to the low
number concentration of SOA particle which can be pre-
dicted from the Equation (2). As the reaction keeps on,
the formed SOA particles can also act as the absorbing
material Mo. The SOA yield will increase with Mo and
the number concentration will become much high. When
it comes to the case of Ki,omMo?1, the SOA yield is only
related to αias described by Equation (3), so the SOA yield
did not change and the finial number concentration of SOA
particle will keep constant.
Inthepresenceofseedaerosols,thevalueof Moislarger
than that in the absence of seed aerosols at the initial stage
of the oxidation reaction. According to the Equation (2),
YSOAis higher than that in the absence of seed aerosols,
which shows that the SOA number concentration is larger.
As a result of the higher concentration SOA particles,
the situation of Ki,omMo?1 will appear soon during a
shorter period due to the rapid increase of Mo. Thus, the
reaction system will account for a shorter time to maintain
a gas/particle partitioning equilibrium state and the SOA
number concentration will remain constant.
3.2 Stimulating effect of CaCl2seed aerosol on the SOA
formation
Among the four individual seed aerosols, the CaCl2
aerosol shows the strongest hygroscopic behavior, and the
most significant promotion effect on the formation of SOA.
For example, at 50% RH, the liquid water amount of the
existing seed aerosol CaCl2 mol−1is about three times
greater than that within (NH4)2SO4(Cocker et al., 2001).
At 75%–80% RH, the level of the relative humidity is
above the deliquescence point of calcium chloride. The
retention of water on the CaCl2aerosol surface increased
dramatically which results in a highly hydrated aerosol.
For the hydrophilic oxidant compounds, more gas-phase
products will dissolve in the aerosol water according to the
Henry’s Law, therefore, the SOA yield is increased.
4 Conclusions
Secondary organic aerosol is one of the important
members in the atmospheric particle matters. Its formation
will make a main contribution to the ambient aerosol
concentration. Based on our experimental results, it can
be inferred that SOA might be a minor source of the at-
mospheric aerosol in the regions with high concentrations
of background aerosols. These results might have potential
applications to the air pollution control strategies.
References
Cocker D R, Clegg S L, Flagan R C et al., 2001. The effect
of water on gas-particle partitioning of secondary organic
aerosol. Part I: α-pinene/ozone system[J]. Atmos Environ,
35: 6049–6072.
Edney E O, Driscoll D J, Speer R E et al., 2000. Impact of aerosol
liquid water on secondary organic aerosol yields of irra-
diated toluene/propylene/NOx/(NH4)2SO4/air mixtures[J].
Atmos Environ, 34: 3907–3919.
Edney E O, Kleindienst T E, Jaoui M et al., 2005. For-
mation of 2-methyl tertrols and 2-methylglyceric acid
in secondary organic aerosol from laboratory irradiated
isoprene/NOx/SO2/air mixtures and their detection in ambi-
ent PM2.5samples collected in the eastern United States[J].
Atmos Environ, 39: 5281–5289.
Hao L Q, Wang Z Y, Huang M Q et al., 2005. Size distribution of
the secondary organic aerosol particles from the photooxi-
dation of toluene[J]. J Environ Sci, 17(5): 912–916.

Page 5

708HAO Li-qing et al. Vol. 19
Kleindienst T E, Smith D F, Li W et al., 1999. Secondary
organic aerosol formation from the oxidation of aromatic
hydrocarbons in the presence of dry submicron ammonium
sulfate aerosol[J]. Atmos Environ, 33: 3669–3681.
Koehler C A, Fillo J D, Ries K A et al., 2004. Formation
of secondary organic aerosol by reactive condensation of
furandiones, aldehydes, and water vapor onto inorganic
aerosol seed particles[J]. Environ Sci Technol, 38: 5064–
5072.
Odum J R, Jungkamp T P W, Griffin R J et al., 1997. Aromatics,
reformulated gasoline, and atmospheric organic aerosol
formation[J]. Environ Sci Technol, 31: 1890–1897.
Oh S, Andino J M, 2000. Effects of ammonium sulfate aerosols
on the gas-phase reactions of the hydroxyl radical with
organic compounds[J]. Atmos Environ, 34: 2901–2908.
Wang Z Y, Hao L Q, Zhang W J, 2005. Chemical process on
the formation of secondary organic aerosol[J]. Prog Chem,
17(4): 732–739.
Wang Z Y, Hao L Q, Zhou L Z et al., 2006. Real-time detec-
tion of individual secondary organic aerosol particle from
photooxidation of toluene using aerosol time of flight mass
spectrometer[J]. Sci China Ser B, 49(3): 267–272.