Kinetics of N2O5 hydrolysis on secondary organic aerosol and mixed ammonium bisulfate-secondary organic aerosol particles

Article abstract of DOI:10.1021/jp107721v

The kinetics of the hydrolysis reaction of N₂O₅ on secondary organic aerosol (SOA) produced through the ozonolysis of α-pinene and on mixed ammonium bisulfate-SOA particles was investigated using an entrained aerosol flow tube coupled to a chemical ionization mass spectrometer. We report room temperature uptake coefficients, γ, on ammonium bisulfate and SOA particles at 50% relative humidity of 1.5 × 10^-2 ± 1.5 × 10^-3 and 1.5 × 10 ^-4 ± 2 × 10^-5, respectively. For the mixed ammonium bisulfate-SOA particles, γ decreased from 2.6 × 10 ^-3 ± 4 × 10^-4 to 3.0 × 10^-4 ± 3 × 10^-5 as the SOA mass fraction increased from 9 to 79, indicating a strong suppression in γ with the addition of organic material. There is an order-of-magnitude reduction in the uptake coefficient with the smallest amount of SOA material present and smaller additional reductions with increasing aerosol organic content. This newly coated organic layer may either decrease the mass accommodation coefficient of N ₂O₅ onto the particle or hinder the dissolution and diffusion of N₂O₅ into the remainder of the aerosol after it has been accommodated onto the surface. The former corresponds to a surface effect and the latter to bulk processes. The low value of the uptake coefficient on pure SOA particles will likely make N₂O₅ hydrolysis insignificant on such an aerosol, but atmospheric chemistry models need to account for the role that organics may play in suppressing the kinetics of this reaction on mixed organic-inorganic particles.

Full text of DOI:10.1021/jp107721v

J. Phys. Chem. A 2010, 114, 13113–13121
13113
Kinetics of N O Hydrolysis on Secondary Organic Aerosol and Mixed Ammonium
2
5
Bisulfate-Secondary Organic Aerosol Particles
,
†
‡
†
Egda N. Escorcia,* Steven J. Sjostedt, and Jonathan P. D. Abbatt
Department of Chemistry, UniVersity of Toronto, Toronto, and Science and Technology Branch, EnVironment
Canada, North York, Ontario, Canada
ReceiVed: August 15, 2010; ReVised Manuscript ReceiVed: October 20, 2010
2 5
The kinetics of the hydrolysis reaction of N O on secondary organic aerosol (SOA) produced through the
ozonolysis of R-pinene and on mixed ammonium bisulfate-SOA particles was investigated using an entrained
aerosol flow tube coupled to a chemical ionization mass spectrometer. We report room temperature uptake
-2
coefficients, γ, on ammonium bisulfate and SOA particles at 50% relative humidity of 1.5 × 10 ( 1.5 ×
-
3
-4
-5
10
and 1.5 × 10 ( 2 × 10 , respectively. For the mixed ammonium bisulfate-SOA particles, γ decreased
-3 -4 -4 -5
from 2.6 × 10 ( 4 × 10 to 3.0 × 10 ( 3 × 10 as the SOA mass fraction increased from 9 to 79,
indicating a strong suppression in γ with the addition of organic material. There is an order-of-magnitude
reduction in the uptake coefficient with the smallest amount of SOA material present and smaller additional
reductions with increasing aerosol organic content. This newly coated organic layer may either decrease the
2 5 2 5
mass accommodation coefficient of N O onto the particle or hinder the dissolution and diffusion of N O
into the remainder of the aerosol after it has been accommodated onto the surface. The former corresponds
to a surface effect and the latter to bulk processes. The low value of the uptake coefficient on pure SOA
particles will likely make N O hydrolysis insignificant on such an aerosol, but atmospheric chemistry models
2 5
need to account for the role that organics may play in suppressing the kinetics of this reaction on mixed
organic-inorganic particles.
2
1
. Introduction
Nitric oxide and nitrogen dioxide, NO and NO
are primary reactants in the production of tropospheric O
harmful greenhouse gas and oxidant. Collectively referred to
as NO , these species are recycled when O is produced during
the day, but may ultimately be lost at night if they enter a
different pathway to form N , a nocturnal NO reservoir. N
can subsequently react with water on aerosol surfaces to yield
HNO
the general predictions of Dentener and Crutzen performed
more than a decade earlier.
2
, respectively,
, a
The uptake coefficient, γ, is the probability that a reaction
will occur between a N gas molecule and an aerosol surface
when these two species collide. Mozurkewich and Calvert
demonstrated in 1988 that N uptake on aqueous ammonium
bisulfate (NH HSO ) and sulfuric acid (H SO ) particles was
faster than through a homogeneous reaction under similar
atmospheric conditions. As such, to follow were a number of
studies which examined various organic and inorganic aqueous
systems at varying relative humidities (RHs), including, for
example, work performed on aqueous ammonium sulfate,
3
2 5
O
x
3
2
O
5
4
4
2
4
O
2 5
x
2 5
O
6
3
:
7
,8
N O + H O f 2HNO3
aerosols
(R1)
2
5(g)
2
9,10
sodium bisulfate and sodium nitrate, malonic acid and other
organic acids,1 and NaCl/artificial seawater
1,12
13-15
aerosol. The
on these
Given that NO
affected by the overall production, depletion, and chemistry of
, this heterogeneous reaction (R1) has been extensively
studied under a variety of conditions to determine its varying
x 3
, and consequently O , concentrations are
reported uptake coefficients for the hydrolysis of N
2
O
5
substrates vary from 0.0005 to 0.05 and are dependent on several
factors, including aerosol size, phase, and especially relative
humidity. The trend observed for most aqueous systems shows
an increase in γ with increasing relative humidity,
suppressions in γ occurring most notably as a result of organic
2 5
N O
1
-5
influence on NO
Jacob demonstrated that upon the inclusion of N
neous loss in the global chemical transport model, GEOS-
CHEM, NO and O concentrations decreased by 53.5% and
.4%, respectively, on a global average. By incorporating five
distinct aerosol substrates (dust, sulfate, organic carbon, black
carbon, and sea salt) and their diverse effects on N hydrolysis
into their model, the resulting O and NO distributions showed
x
and O
3
budgets. For example, Evans and
6-14
with
2
O
5
heteroge-
8,13,16-19
substrates or coatings,
and high concentrations of nitrate
9,10
x
3
present in the aerosol bulk.
1
9
Both of these effects have been investigated by several
researchers in an attempt to parametrize and express each in a
concise manner. Mentel et al. have described the ionic
mechanism of N O hydrolysis and related it to a “nitrate effect”,
5
essentially demonstrating that aerosols containing large amounts
O
2 5
9,10,20
3
x
2
better agreement with previously reported data than with
estimates from models that had not included heterogeneous
aerosol chemistry in their calculations. This study confirmed
-
of NO
3
, as well as small concentrations of water, can lead to
1
-
a suppression in γ; e.g., an increase in [NO
1
3
] from 13.4 to
58 mol/kg decreased γ from 0.018 to 0.0002.9 A recent study
,10
*
To whom correspondence should be addressed. Phone: (416) 946-7359.
by Bertram and Thornton outlines a systematic parametrization
of the influences of H O, nitrate, and chloride concentrations
on the uptake coefficient of N O for internally mixed particles
2 5
E-mail: egda.escorcia@utoronto.ca.
†
University of Toronto.
Environment Canada.
2
‡
1
0.1021/jp107721v  2010 American Chemical Society
Published on Web 12/02/2010

1
3114 J. Phys. Chem. A, Vol. 114, No. 50, 2010
Escorcia et al.
containing these species.20 They assume the formation of a
through the ozonolysis of R-pinene. The researchers reported
a decrease in γ from 0.0187 to 0.0034 at 60% RH when an
SOA coating, approximately 15 nm thick, was added to the
8
+
protonated nitric acid intermediate (H
react with H
either promote or suppress N
Thornton concluded that either a water limitation exists in the
2
ONO
2
), which can then
2
O, nitrate, or chloride present in the aerosols to
reactivity. Bertram and
8
2
O
5
particles. Likewise, Badger et al. report decreases in γ by a
factor of 9, 3.8, and 2.3 at 25%, 50%, and 70% RH, respectively,
with the addition of only 6% humic acid by mass to ammonium
sulfate solutions, from which particles were formed by atomi-
zation. These substantial suppressions were interpreted as being
due to an effect on the mass accommodation coefficient of N
on these organic-containing particles.
The motivations for the research presented in this paper were
twofold. First was the need to supplement and expand the limited
+
reaction of H
modation of N
2
ONO
2
2
with water or nitrate or mass accom-
2
0
5
O is water dependent. In addition, they
31
illustrated that increasing nitrate concentrations in the particles
led to a decrease in the uptake coefficients measured and that
low water and high nitrate concentrations worked in synergy
to produce lower γ values. The researchers derived an expression
2 5
O
for γ that is dependent on H
tions, where the importance of each is regulated by the rate
2
O, nitrate, and chloride concentra-
2 5
literature data available on the heterogeneous reaction of N O
+
20
coefficients of the reaction of H
2
ONO
2
with each species.
with particles containing organics. Second was the desire to
better understand the mechanism of suppression, which occurs
upon the addition of an organic layer to aerosol, and how
The mechanism by which submonolayer-thick organic films
on particles decrease the uptake coefficient has yet to be
parametrized, but it is likely that they inhibit the mass
2 5
variable amounts of this new organic material affect N O
uptake. In particular, two mechanismssa surface-dependent
process and a bulk-dependent processsare possible. Given the
knowledge that tropospheric particles are commonly composed
of a mixture of inorganic and secondary organic material, we
investigated the reaction of N with ammonium bisulfate
aerosols mixed with SOA produced through the reaction of O
and R-pinene at 45-50% RH. R-Pinene was chosen since it is
the most common monoterpene, a class of compounds that
efficiently leads to SOA formation. Although SOA formed from
oxidation of monoterpenes in the laboratory is not as oxidized
as that found in the field, it remains the best model substance
we have for ambient SOA.
To contrast the behavior observed with these two-component
aerosols, we also present the uptake coefficients obtained for
single-component ammonium bisulfate at 50% RH and SOA
at 0%, 30%, and 50% RH. These experiments extend the work
accommodation of N
2
O
5
within the outer layers of the aerosol
particle.1
6-19
Thornton and Abbatt were the first to address the
effects of organic monolayers by coating artificial seawater with
3
2
hexanoic acid.13 The results indicated that the organic coating
2
O
5
suppressed γ by a factor of 3 from 0.024 to 0.008, but it was
undetermined whether the surfactant monolayers had the greatest
impact on the mass accommodation coefficient or the rate of
3
1
3
reaction at the interface. Other studies illustrate an equivalent
pattern, with the greatest suppressions in γ occurring for ordered
films composed of large organic molecules and high organic
1
8,19
19,21
coating concentrations.
demonstrated that a straight chain surfactant (1-octadecanol)
decreased the uptake coefficient of N onto 60 wt % aqueous
sulfuric acid to a significant degree, while a branched monolayer
phytanic acid) affected γ only slightly. Branched monolayers
For example, Cosman et al.
2 5
O
(
do not pack efficiently on an aerosol surface and thus do not
decrease the uptake coefficient as much as straight chain
monolayers because the spaces present between branched
8
performed by Folkers et al. but using a different experimental
technique, an aerosol flow tube coupled to a newly built
chemical ionization mass spectrometer (CIMS). We also employ
1
8,19,21
2 5
monolayers allow N O to diffuse more quickly.
-
the new CIMS N
2
O
5
detection method of I ·N
2
O
5
cluster
In general, organic monolayers have been shown to both
enhance and suppress uptake of gases to liquid. In particular,
researchers have monitored the entry of different gases, such
as HCl and HBr, onto sulfuric acid particles through 1-butanol
formation, as first demonstrated by Kercher et al., rather than
-
33
the more common approach of NO
3
detection.
22
and hexanol films. In these cases, the chemistry does not follow
that of N uptake in that the organic films actually enhance
rather than impede acidic uptake. It is believed that the hydroxyl
2. Experimental Methods
2
O
5
Figure 1 illustrates the overall experimental setup for the two-
component aerosol experiments. The layouts for each of the
single-component experiments were similar to that shown in
Figure 1, except that the SOA and ammonium bisulfate single-
2
2
groups on the organic layer help facilitate this uptake. On the
other hand, organic films are also known to suppress the uptake
of ammonia,23 water, nitric acid, and acetic acid.
24
25
26
3
component trials did not require the atomizer and O denuder,
2 5
Another class of N O studies have considered the role of
respectively. In addition, the single-component experiments only
required one mixing flow tube. We note that the overall
experimental approach is similar to that which we have used in
organics on the kinetics when present in large quantities in an
organic particle, much greater than a monolayer. In particular,
field work conducted by Bertram and Thornton at Boulder, CO,
and Seattle, WA, has indicated that a large organic presence
in/on aerosols decreases γ to a significant degree: the results
indicate that increasing organic-to-sulfate ratios affect γ by
decreasing the ability of the aerosols to take up water and by
11
a previous study. Details pertaining to the novel aspects of
this study are provided below.
2
.1. Chemical Ionization Mass Spectrometer. The chemical
ionization mass spectrometer (Figure 2) is a home-built instru-
ment constructed according to the designs of G. Huey and D.
Tanner from the Georgia Institute of Technology. The instru-
ment, along with all essential electronics and accessory items,
is mounted on a wheeled 104 cm × 61 cm × 114 cm aluminum
boxed frame. It is divided into four differentially pumped
chambers, the ion molecule region (IMR), the collisional
dissociation chamber (CDC), the intermediate chamber (IC),
and the multiplier chamber (MC). This instrument contains four
critical orifices, one each leading into the subsequent chamber,
two octupole ion guides (THS Instruments, LLC), one quad-
rupole (Extrel Core Mass Spectrometers, 9.5 mm diameter rods),
2
7
enhancing the strength of the nitrate effect. Brown et al. have
also indicated through ambient measurements obtained as part
of the New England Air Quality Study (NEAQS) that organics
2
8-30
may suppress reactivity.
From a laboratory perspective, two studies have examined
the effects of adding substantial quantities of organics to
8
inorganic substrates, i.e., the work of Folkers et al. and Badger
et al.31 In particular, Folkers, Mentel, and Wahner investigated
2 5
the reaction of N O with ammonium bisulfate aerosols coated
with secondary organic aerosol (SOA) material produced

Kinetics of N2O5 Hydrolysis
J. Phys. Chem. A, Vol. 114, No. 50, 2010 13115
2.2. Aerosol Generation. Ammonium bisulfate aerosols for
the single-component and two-component experiments were
produced from 20 and 2 wt % solutions, respectively, of
ammonium bisulfate (Sigma-Aldrich, 307602) using a constant
output atomizer (TSI 3076). A dump line was added from the
exit of the atomizer to vent excess particles for all ammonium
bisulfate experiments and to maintain the system at atmospheric
pressure. For the NH HSO single-component experiments only,
4 4
the aerosol passed through a silica gel dryer before entering a
mixing flow tube where both a wet (i.e., through a water
2
bubbler) and a dry flow of N were added to regulate the RH in
the system. These wet and dry flows were regulated by mass
flow controllers (MFCs), and varied from 600 to 800 and from
3
00 to 400 standard cubic centimeters per minute (sccm),
respectively, to achieve the desired relative humidity. The
resulting flow moved directly to the kinetics flow tube and
entered through a side inlet at the top.
The two-component experiments did not require a dryer, as
the flow from the atomizer, in addition to a single wet flow of
N
2
(∼750 sccm, regulated by a needle valve and rotometer)
added before the kinetics flow tube, was employed to fix the
RH to a value of 50% (Figure 1). Two preparatory flow tubes
were required for these experiments: the first for the mixing of
ammonium bisulfate aerosol and R-pinene gas and the second
for the mixing of the contents of the first flow tube with O
The flow exiting the second flow tube passed through an O
Figure 1. Schematic depicting the overall experimental setup for the
two-component aerosol experiments for N
2
O
5
hydrolysis. Both the
.
3
ammonium bisulfate and SOA single-component experiments required
only one mixing flow tube and additional flows to regulate the RH.
3
denuder before entering the kinetics flow tube. R-Pinene gas
was produced by passing a small (up to 30 sccm) dry N flow
through a fritted glass bubbler containing R-pinene liquid (Fluka,
9%) held at room temperature. A N dilution flow was also
2
9
2
added after the bubbler (500-700 sccm) to produce a range of
R-pinene concentrations. Ozone was produced by passing a flow
of O
2
(up to 30 sccm) over a 22.9 cm long (mercury) Hg Pen-
to O at a wavelength
Ray lamp (UVP, LLC) to photolyze O
2
3
of 185 nm. Considering experimental procedures and gas flows
similar to those outlined by George and Abbatt,34 we estimate
that O and R-pinene mixing ratios for these experiments were
3
up to 20 and 50 ppm, respectively.
The particles produced for the SOA single-component experi-
ments employed only one flow tube for the mixing of R-pinene
3 3
and O . SOA exiting this flow tube also passed through an O
denuder before entering the kinetics flow tube. In this case, the
dump line to the atmosphere was positioned at the top of the
kinetics flow tube to vent excess aerosol. All flows for these
experiments were regulated by MFCs, and the concentrations
were slightly higher (by no more than a factor of 2) than those
Figure 2. Schematic of the CIMS displaying the four differentially
pumped chambers: ion molecule region (IMR), collisional dissociation
chamber (CDC), intermediate chamber (IC), and multiplier chamber
2
of the mixed experiments. Additional wet and dry N flows were
(
MC).
added to the main flow just before the kinetics flow tube to
regulate the RH. These flows were added in various ratios (total
flow of 2 slpm) to produce conditions of 0%, 30%, and 50%
RH.
and an on-axis electron multiplier detector (ITT Power Solutions,
Inc., 7550 M channel electron multiplier). The pressures, when
-4
-6
It should be noted that with the exception of minor changes
to gas flows the experimental conditions leading to the produc-
tion of the SOA were held constant throughout the study in an
attempt to also keep the composition of the SOA constant. As
such, the change in reactivity is a result of the changing mass
fraction of the SOA component and/or RH, rather than a result
of changing SOA composition. Also, we assume that all the
SOA is formed by gas-phase reactions followed by gas-to-
particle conversion. Given the low solubility of monoterpenes,
we do not anticipate that much organic oxidation is occurring
in the condensed phase.
in operational mode, are 45, 0.1, 1 × 10 , and 1 × 10 Torr
in the IMR, CDC, IC, and MC, respectively. Ions entering the
CIMS follow a trajectory from the inlet into the IMR, CDC,
IC, and last MC, where they are detected. The CIMS samples
a total of 7-8 standard liters per minute (slpm) through the
2
IMR, with 4 slpm N pertaining to the flow required to transport
the reagent ion and the other half corresponding to sample flow.
The sample flow for these experiments was not as large as 4
2 5
slpm, so a N O -free makeup flow was added. The reagent ion
flow enters the IMR through a side cavity positioned after the
first critical orifice, whereas the sample flow enters the IMR
via an inlet fixed with Teflon tubing and then proceeds to the
first aperture.
All aerosol surface areas were characterized by means of a
scanning mobility particle sizer (SMPS) system consisting of a

1
3116 J. Phys. Chem. A, Vol. 114, No. 50, 2010
Escorcia et al.
differential mobility analyzer (DMA; TSI 3080-3081) and
condensation particle counter (CPC; TSI 3025A). The SMPS
system sampled from a tee located immediately before the top
of the kinetics flow tube.
2
2 5 2 5
.3. N O Generation and Detection. N O was produced
in situ in a 50 cm long glass flow tube through the dark reaction
of O
up to 4 sccm) flowing over a small Hg Pen-Ray lamp. NO
was supplied from a NO
BOC Gases consisting of 2000 ppm NO
3
and NO
2
. O
3
was produced through the photolysis of O
2
(
2
2
cylinder custom-ordered from Linde
in N . NO was added
2
2
2
in excess (∼13-27 ppm in the kinetics flow tube) to ensure
that the equilibrium shifted toward the product (N
than toward the reactants (NO and NO ). The [N
2
O
O
5
) rather
]/[NO
3
2
2
5
3
]
4
ratio was approximated as being roughly 10 , thus indicating
that N concentrations were much larger than NO concentra-
tions. Once formed, N then proceeded to the kinetics flow
O
2 5
3
2
O
5
tube, entered via a 6 mm o.d. (∼4.5 mm i.d.) injector at the
top, and exited along with the main flow at the bottom toward
the IMR of the CIMS.
Figure 3. Surface area size distributions of ammonium bisulfate
particles and SOA combined (dotted line) and ammonium bisulfate
particles without the addition of SOA (dashed line). Given that
approximately 20% of the total surface area for the two-component
experiments was not measured by the SMPS system, the distributions
were corrected using log-normal fits (solid lines).
-
We use I as the CIMS reagent ion, produced from methyl
3
iodide, CH I (Sigma Aldrich, 99%), following exposure to a
33
polonium-210 radioactive source. It is delivered from a home-
built permeation tube (6.4 cm long) that is composed of Teflon
tubing (PFA, 3 mm i.d., 5 mm o.d.) plugged at the ends with a
the interior walls with halocarbon wax. It was also covered with
1
cm long Teflon rod (PTFE, 0.32 cm diameter) and crimped
aluminum foil to eliminate the possibility of N
and fitted with a glass injector, 6 mm o.d. (∼4.5 mm i.d.) and
20 cm long. The relative humidity was monitored at the top
2 5
O photolysis
shut with stainless steel tubing (0.56 cm o.d., 0.52 cm i.d.).
The permeation tube is placed within an aluminum block and
heated to 70 °C by a temperature controller (Omega Engineering
1
entrance to the kinetics flow tube with a hygrometer (VWR).
Kinetics measurements were performed by varying the
position of the injector inside the flow tube. It was pulled out
systematically in 15 cm increments every 3 min while the ion
signal and aerosol surface area were monitored. These surface
areas were determined by averaging seven distributions recorded
sequentially during the time of the kinetics measurements. Given
that several distributions for the two-component and ammonium
bisulfate single-component experiments were not fully measured
by the SMPS, the missing portions were assessed using log-
normal fits (see Figure 3). The ammonium bisulfate to SOA
mass fractions were determined from the SMPS volume
measurements, in particular by measuring the volume of the
ammonium bisulfate aerosol with and without SOA present.
Although the hygroscopic growth factor at 50% RH for the
Inc., CN1A series). As the permeation tube is heated, CH
3
I
diffuses through the Teflon walls and is carried by a 4 slpm N
2
flow toward the 2 Po source and then to the IMR of the CIMS.
10
2 5
N O was detected by monitoring the reagent ion cluster,
-
-
33
(R2)
(R3)
(R4)
I ·N
2
O
5
, rather than NO
3
, as shown in the reactions below:
-
-
I + N O f NO + INO2
2
5
3
-
-
I + H O + M f I ·H O + M
2
2
-
-
I + N O + M f I ·N O + M
2
5
2
5
35
-
-
ozonolysis of R-pinene is essentially 1.0, it is ∼1.22 ( 0.02
I ·H O + N O f I ·N O + H O
(R5)
2
2
5
2
5
2
36-38
for ammonium bisulfate
under these conditions. As such,
the volumes measured for the ammonium bisulfate particles
were reduced by a factor of 1.22 (see ref 36-38) to account
for the presence of water. These corrected volumes were then
used to convert to mass fractions assuming a density of 1.78
and 1.41 g/cm for ammonium bisulfate and SOA produced
through the ozonolysis of R-pinene, respectively.
-
Dissociative charge transfer (reaction R2) is problematic, as I
-
3 3
can also react with HNO to yield an equivalent NO ion, thus
requiring the subtraction of a chemical background from all
measurements. Ion cluster formation (reaction R4 and R5),33
however, is accompanied by minimal background.
3
39
40
Calibrations were performed as outlined by Thornton et al.,11
This approach of determining the inorganic mass fraction
assumes that each component in the aerosol has an associated
amount of water, a quantity which is not altered by the mixing
by producing N
and then trapping it in an ethanol/liquid N
The trap was then warmed and the N signal was monitored
2
O
5
through the reaction of excess NO
2 3
and O
2
slush bath (163 K).
4
1,42
2
O
5
of the components; i.e., they independently take up water.
while a small, known flow passed through the trap and into a
Also known as the Zdanovskii-Stokes-Robinson (ZSR) as-
sumption, it describes the final hygroscopic growth factor of
the mixed component aerosol as a linear combination of the
hygroscopic growth factors and volume fractions of the indi-
vidual components.4 The ZSR method has been experimen-
larger dilution flow. The resulting sensitivity and detection limit
-
of this instrument for the I ·N
2
O
5
cluster are 0.024 Hz/pptv
-
and 100 pptv (S/N ) 2), respectively, for an I signal level of
1,42
1
1
25 kHz. The lowest reagent ion signals used in the work were
5 kHz.
43
44
tally validated by, for example, Badger et al., Brooks et al.,
45
2
.4. Kinetics Measurements and Analysis Approach. The
kinetics flow tube, mounted vertically, is 90 cm long with an
inner radius of 3.0 cm. Wall losses of N were minimized
by maintaining the flow tube at atmospheric pressure and coating
and Gysel et al. These researchers investigated the hygroscopic
4
3,44
growths of mixed humic acid-ammonium sulfate
particles
45
2
O
5
and aged air masses using tandem differential mobility analysis
(TDMA) to demonstrate the validity of this technique.

Kinetics of N2O5 Hydrolysis
J. Phys. Chem. A, Vol. 114, No. 50, 2010 13117
γ
ω
SA
N O N O
2 5
II
2
5
k )
(E1)
4
where ωN O is the mean molecular speed of N
2
O
5
and SA is
2
5
the total aerosol surface area. For the magnitude of the uptake
coefficients measured in this work, the gas-phase diffusion
47
limitation, as assessed using the Fuchs-Sutugin corrections,
is no more than 3%.
3
. Results
Figure 4. Corrected (i.e., background-subtracted) decay plots of the
-
I ·N
2
O
5
signal (cps) vs injector position (cm) at 50% RH in the absence
The surface area distributions for the SOA single-component
(
triangles, dashed line) and presence (solid lines) of aerosols. The decays
experiments were log-normal distributions that peaked around
00 nm and produced very large surface areas. The ammonium
-
4
-2
correspond to an aerosol surface area of 1.2 × 10 and 1.5 × 10
4
2
3
cm /cm for ammonium bisulfate (squares) and SOA (circles) produced
bisulfate single-component distributions were wider, peaked at
a smaller diameter, roughly 300 nm, and corresponded to smaller
overall surface areas. Figure 3 shows distributions obtained for
two-component system experiments, both for ammonium bisul-
fate particles with SOA added (dotted line) and for the
ammonium bisulfate particles without SOA present (dashed
line); i.e., the SOA was removed by turning off ozone and
R-pinene in the SOA-generation system while the carrier flows
through the ozonolysis of R-pinene, respectively.
Background control runs were required to determine wall
losses of N O and thus were conducted in the absence of
2 5
aerosols, but at relative humidity conditions equivalent to those
during experimental runs. These background runs were usually
performed at the beginning and end of a set of kinetic runs.
2
(N ) remained unchanged. The log-normal fits are shown as solid
-
Since N O
2 5
was monitored via ion cluster formation (I ·N
2 5
O ),
lines. As can be seen, the ammonium bisulfate distribution peaks
at 325 nm, whereas the distribution for the ammonium bisulfate
to SOA (41:59 AB-SOA by mass) two-component particle
peaks at 375 nm. Upon the addition of SOA to the ammonium
bisulfate particles, the aerosol distributions became wider and
shifted slightly to larger sizes. We estimate that errors in the
total aerosol surface area measurements as determined by the
SMPS system are on the order of (25%, which includes
possible errors from the CPC, possible losses within the kinetics
flow tube, and the errors associated with estimating (through
log-normal fits) the amount of surface area that we did not
directly measure.
the background signal subtracted at this mass-to-charge ratio
corresponded to the signal of a small chemical background and
multiplier noise present during sampling times. That is, despite
having nitric acid present in the flow tube as a product of the
reaction, it gave rise to no background at m/z 235. The
background values we observed via this detection method ranged
2 5
from 7 to 25 cps or from 1.2 to 4.2 ppb N O , respectively, for
any kinetic run. Using essentially the same experimental setup
as Thornton et al.,11 the losses of aerosol in the flow tube are
estimated to be less than 10%. Also, as in that previous study,11
measurements were not taken within 15 cm of the end of the
2 5
Figure 5 displays the first-order rate constants for N O on
flow tube to ensure an adequate mixing length of N
flow, as well as to establish laminar flow. The N
2
O
5
into the
the particles versus the total aerosol surface area for the AB
and SOA single-component decays, where the rate constants
have been corrected for wall loss and laminar flow effects by
the approach of Brown.46 Similar plots of the mixed aerosol
experiments are not included in this graph, but were determined
using the same procedure as with the single-component systems.
The uptake coefficients for the two-component runs, however,
were determined from a two-point relationship, one measure-
ment corresponding to the wall-loss background rate constant
with zero aerosol surface area and the second to the rate constant
associated with the two-component aerosol surface area. We
used this approach because it was not practical to maintain a
constant ratio of AB to SOA within the particle over a wide
range of total aerosol surface areas. Nevertheless, the kinetics
obtained on the single-component species (Figure 5) verify that
our overall kinetics approach was valid, with a highly linear
relation between the magnitude of the first-order decay constant
and aerosol surface area. Therefore, we feel confident that this
relationship also holds in the runs with the mixed AB-SOA
particles.
2
O mixing
5
ratios normally present during experiments varied from 20 to
50, from 350 to 750, and from 70 to 330 ppb for the
1
ammonium bisulfate (AB), SOA, and AB-SOA experiments,
respectively.
Figure 4 displays N
dashed line) and in the presence (solid lines) of aerosol at
5-50% RH. First-order kinetics were observed in all experi-
2 5
O decays both in the absence (triangles,
4
ments. The ammonium bisulfate and SOA single-component
decays shown correspond to a surface area (of each respective
-
4
-2
2
3
aerosol type) of 1.2 × 10 and 1.5 × 10 cm /cm ,
respectively. As can be seen, the decay of ammonium bisulfate
(
squares) is similar to that of SOA (circles), but for two very
distinct surface areas.
The flow in the kinetics flow tube was held constant during
each experimental run by a critical orifice in a Teflon disk
situated within an UltraTorr fitting placed at the bottom of the
flow tube. Depending on the pinhole used, the total flow varied
from 1000 to 1600 sccm. The slopes of the decays shown in
A total of six different AB-SOA mixed aerosol compositions
were studied, in addition to AB (50% RH) and SOA single-
component (0%, 30%, and 50% RH) particles, as indicated in
Table 1. γ values and the RH conditions for each respective
aerosol type are also summarized in Table 1, along with the
uncertainties in the resulting uptake coefficients. The uncertain-
ties quoted for the single-component kinetics arise from an
unweighted least-squares fit to the data in Figure 5. For the
I
Figure 4 are converted into pseudo-first-order rate constants (k )
using the bulk flow velocity in the flow tube, which varied from
-1
I
0
.64 to 1.02 cm s . k values are then corrected to account for
nonplug flow conditions using the method as described by
Brown.46 The second-order rate constant (k ) corresponds to
II
I
the slope of a graph of k vs aerosol surface area, thus leading
to the determination of γ using the following expression:

1
3118 J. Phys. Chem. A, Vol. 114, No. 50, 2010
Escorcia et al.
Figure 6. Uptake coefficients for N
2
O
5
hydrolysis vs the inorganic
mass fraction for this study at 50% RH (solid squares), ammonium
3
1
sulfate and humic acid aerosols at 50% RH, Badger et al. (open
triangles), SOA-coated ammonium bisulfate aerosols at 60% RH,
8
Folkers et al. (plus signs), ambient aerosols measured in Seattle, WA,
2
7
at 70% RH, Bertram et al. (solid circles), and the results of the
2
0
parametrization (open squares) developed by Bertram and Thornton
as applied to our data.
increasing particulate organic mass fraction. The results of
Figure 5. Pseudo-first-order rate constant vs aerosol surface area for
the hydrolysis of N on ammonium bisulfate (A) and SOA (B)
31
Badger et al. for humic acid and ammonium sulfate aerosol,
2
O
5
8
Folkers et al. for SOA and ammonium bisulfate aerosol, and
particles at 50% RH. Ammonium bisulfate aerosols were generated
from a 20 wt % solution using an atomizer, and the SOA particles
were produced through the ozonolysis of R-pinene. The rate constants
Bertram et al.27 for their field study investigating the organic to
sulfate ratio in ambient aerosol in Seattle, WA, are also included
in Figure 6 for easy comparison. Note that we obtained the
4
6
have been corrected for wall loss using the Brown approach.
8
values from Folkers et al. through the same approach outlined
2 5
TABLE 1: Uptake Coefficients, γ, for N O
Hydrolysis on
above, assuming inorganic fractions of 5% and 7% by volume,
Single-Component AB and SOA Particles as well as
8
as stated by the authors. We received the data of Bertram et
a
Two-Component AB-SOA (by Mass Fraction) Particles
27
al. directly from the researchers. In addition, we have also
included the results of the parametrization developed by Bertram
and Thornton20 as applied to our data.
aerosol
RH (%)
104γ
AB
50
47
48
45
46
47
46
50
30
0
146 ( 15
26 ( 4
9
7
7
4
4
2
1:9 AB-SOA
8:22 AB-SOA
7:23 AB-SOA
8:52 AB-SOA
1:59 AB-SOA
1:79 AB-SOA
18 ( 0.5
20 ( 3
4. Discussion
The ability of an organic component to suppress N
2 5
O
6.2 ( 0.3
6.6 ( 1.4
3.0 ( 0.3
1.5 ( 0.2
1.7 ( 0.3
0.6 ( 0.2
hydrolysis when added to a soluble inorganic has been previ-
ously addressed by other researchers.8
,13,16-19,21
Our results
SOA
SOA
SOA
indicate that N uptake is slower on aerosol containing an
2 5
O
organic fraction, with a large suppression in γ occurring even
for aerosol with a small SOA fraction. It is evident from Figures
4 and 5 that only a small quantity of ammonium bisulfate aerosol
is required to induce a large decay and thus lead to a large uptake
coefficient. Conversely, large aerosol surface areas were neces-
sary to measure the small γ in the SOA single-component
experiments.
a
All experiments were conducted at approximately equivalent
relative humidities, except for those with pure SOA at 0% and 30%
RH.
single-point decays with the mixed composition particles, the
uncertainties were calculated by propagating errors in the first-
order rate constants through to the calculated slope of the kI
versus surface area plot. In addition, we believe the systematic
and precision errors associated with the calculated inorganic
mass fraction to be on the order of (8% and (12%, respec-
tively. The former corresponds to errors pertaining to the value
for the hygroscopic growth factor of the mixed aerosol, whereas
the latter correlates to variability in the volume measurements
of the ammonium bisulfate single-component and mixed aerosol
experiments determined by the SMPS. We estimate that our
systematic uncertainties in the uptake coefficients are generally
larger, being on the order of (30%. This estimate arises
primarily from uncertainties in the aerosol surface areas, as well
as from errors in the flow velocities and Brown correction
procedure.46
These results indicate that aqueous ammonium bisulfate is
-2
-3
highly reactive, producing a γ ) 1.5 × 10 ( 1.5 × 10 ,
1-2 orders of magnitude larger than all other γ values obtained
for this study. This uptake coefficient is in agreement with other
aqueous systems, both AB and non-AB systems. In fact,
researchers8
,48-50
who have investigated ammonium bisulfate
particles at relative humidity conditions of 48-60% have
reported a very narrow range of γ ) 0.016-0.018, whereas
only two investigations have reported higher values: those of
20
Bertram and Thornton (γ ) 0.028, 60% RH) and Mozurkewich
6
and Calvert (γ ) 0.05, 50% RH). We consider these values as
2 5
upper thresholds to the uptake coefficient of N O on aqueous
AB particles. The Bertram and Thornton result may be higher
due to a low nitrate effect, arising from the very low N O
5
2
20
mixing ratios used. If we compare our result against those for
non-AB aqueous systems under similar wet conditions, we find
The effects of the organic fraction on γ are represented in
Figure 6, which shows the suppression in γ as a function of
7
,8,31,48,49
2 5
that N O hydrolysis conducted on ammonium sulfate,

Kinetics of N2O5 Hydrolysis
J. Phys. Chem. A, Vol. 114, No. 50, 2010 13119
9
,49
NaCl/artificial seawater,13-15 and malonic
acid aerosols produces γ values of 0.010-0.044, 0.00042-
.0032, 0.0102-0.035, and 0.018, respectively. Thus, with the
-4
sodium nitrate,
fraction and 5.4 × 10 (+0.017/-0.012) for 9:91 AB-SOA.
1
1
Interestingly, the comparison to the mixed ammonium sulfate-
31
0
humic acid particle studies by Badger et al. shows qualitatively
the same behavior. In particular, we note that the uptake
coefficients for ammonium bisulfate (this study) and ammonium
sulfate31 aerosols decreased by a factor of 5.6 and 4 upon the
addition of less than 7% by mass fraction SOA and humic acid,
respectively. In both cases, the decreases observed in γ thereafter
occur in smaller degrees than the original decline and last level
off at values almost identical to those of their single-component
organic counterparts. It is interesting that similar kinetics arise
from systems where the organics are either coated onto an
inorganic substrate, as in our study, or initially mixed with the
exception of the sodium nitrate experiments, all γ values of
both the AB and non-AB aqueous systems lie within the same
order of magnitude and are relatively similar. The sodium nitrate
results are at the lower end of the spectrum, as the uptake
coefficients have been suppressed by large nitrate concentrations
9
,49
present in the aerosols.
generally validate our experimental technique.
We believe these comparisons
The uptake coefficients we report for the SOA single-
component experiments at 0%, 30%, and 50% RH are 6.0 ×
-
5
-5
-4
-5
-4
1
0
( 2.0 × 10 , 1.7 × 10 ( 3.0 × 10 , and 1.5 × 10
-5
31
(
2.0 × 10 , respectively. These very low values are not
inorganics, as with Badger et al.
dissimilar to those reported by other researchers who have
studied the hydrolysis reaction on pure complex organic
We now consider two separate models to describe the
suppression in the kinetics when organics are present. First, the
initial large decline of the uptake coefficient may be due to a
mass accommodation effect whereby the N O5 is not as
effectively taken up by the aerosol surface as it is on a purely
8
substrates. In particular, we note that Folkers et al. obtained a
-4
value of 4.5 × 10 (+0.015/-0.009) for SOA produced
2
through the reaction of O and R-pinene at 58% RH. To our
3
knowledge, this is the only other uptake coefficient reported in
the literature for SOA formed from the ozonolysis of R-pinene.
The factor of 3 disagreement between the value reported here
for SOA at 50% RH and that of Folkers et al. is possibly due
to some systematic difference in experimental techniques
between flow tube and chamber experiments, such as different
compositions of SOA given that we form ours on a very short
time scale at high precursor concentrations. However, we have
recently compared the aerosol mass spectrometer (AMS)
inorganic aqueous surface. The surfactant-like properties of the
individual components of the SOA and humic acid mixtures
will give rise to this behavior. In this regard, we are in agreement
31
13,16-19,21
with the conclusions of Badger et al. Other researchers
have also attributed the decreases in N O uptake to this effect,
2
5
particularly those who have examined this process through
organic monolayers. We also note that the subsequent, more
gradual decline in reactivity with increasing organic fraction is
possibly due to progressively lower water content on the surface.
A second, alternate model of reactivity, recently presented, for
example, by Anttila et al., is that the organic component in the
particle is present as a phase-separated liquidlike layer on an
3
4
spectrum of the SOA formed in this manner and found that it
is similar to that of particles generated in chamber studies;51
the O:C ratios of the SOA inferred from these investigations
34
were similar: 0.26-0.34 for the flow tube approach and about
52
inorganic aqueous spherical core. As such, the particle consists
5
1
0
.25 for the chamber study. Indeed these values indicate that
of two phases, each of which has an associated partitioning
the SOA examined in this study, similar to that of George et
(
solubility) coefficient and diffusion coefficient. A mass ac-
commodation coefficient is also associated with the organic
coating. From this perspective, N hydrolysis occurs by mass
3
4
al., was fresh SOA that did not undergo oxidation and that it
34,51
was similar in composition to both studies.
Thus, at this time
2 5
O
we cannot fully explain the difference between the reported
uptake coefficients for our SOA single-component study at 50%
accommodation into the particle surface, dissolution onto the
organic coating, and then diffusion to the aqueous core. Anttila
et al. tested their parametrization with aerosol chamber experi-
ments using aqueous sulfate particles coated with monoterpene
8
RH and the Folkers et al. result.
We also observe a similarity between our values at 30% and
3
1
52
5
0% RH and the ones reported by Badger et al. for reaction
reaction products. The researchers discovered that their model
on a commercial (Aldrich) humic acid aerosol at 25% and 50%
held most true to their empirical results for the cases in which
the diffusion coefficient and solubility coefficient for the organic
phase were at least 1 order of magnitude smaller than those of
the aqueous phase. According to this model, the mass accom-
modation coefficient of the organic coating is not sufficiently
-4
-4
-4
-4
RH, γ ) 1 × 10 ( 1 × 10 and 3 × 10 ( 1 × 10 ,
respectively. While certainly not the same complex mixture as
3
that comprising R-pinene/O SOA, it is noteworthy that the
kinetics with humic acid particles are so comparable. Perhaps
this arises because humic acid, as SOA, contains a wide range
of oxygenated organic functional groups, such as carboxylic
52
reduced by the coat to account for the suppression in γ. Rather,
the decreases in the uptake coefficient are due to the small
diffusion and solubility coefficients of the organic coating. Given
35
acids. In addition, the hygroscopic growth factors of SOA and
humic acid43 at 50% RH are similarly small: 1.0 and 1.07,
respectively.
8
the similarity of the Folkers et al. results and those presented
by Anttila et al.52 to our own, this model and aerosol particle
reactivity may well be an accurate description of the behavior
seen in our experiments. At this point, we cannot unequivocally
state whether the organic coating has its greatest effect in
reducing γ by decreasing the mass accommodation coefficient
In general, we note that very low γ values arise for these
organic systems that exhibit low water content. Particularly, the
1
1
reported uptake coefficient on solid malonic acid particles is
less than 0.001 for an RH of 10-50%. However, when an
organic system takes up water, as in the case of aqueous malonic
acid, the uptake coefficient is similar to that reported on aqueous
of N
and diffusion of N
2
O
5
into the organic phase or by hindering the dissolution
into the organic layer and aqueous layer,
2
O
5
11
inorganic systems and increases to γ ) 0.025 at 50% RH.
respectively, as a result of the inherently small solubility and
diffusion coefficients of the organic coating.
The results for the mixed AB-SOA aerosol obtained in this
work are qualitatively in agreement with those of the two other
studies of comparable systems (Figure 6). Specifically, the
uptake coefficient for AB-SOA particles that were studied in
We note that Bertram and Thornton have demonstrated that
the water concentration in particles is a driving factor in the
reactivity of mixed component aqueous systems.20 To test this
hypothesis, we applied the parametrization of Bertram and
8
the Folkers et al. study was higher on the more inorganic-rich
-3
20
system: 3.4 × 10 (+0.15/-0.12) for 96:4 AB-SOA by mass
Thornton to our mixed AB-SOA particle compositions to

1
3120 J. Phys. Chem. A, Vol. 114, No. 50, 2010
Escorcia et al.
model the behavior of γ as a function of nitrate and water
concentrations. In particular, we assumed a nitrate concentration
of 2 M, on the basis of the previous work of Thornton et al.11
Alternatively, Anttila et al.52 present a model suggesting that
organic coatings form on inorganic particle cores, a model which
may also describe our results.
As these researchers calculated an upper limit for [HNO
3
] of
From an atmospheric modeling perspective, the models
1
1
0
.35 M for 30 ppb N
2
O
5
,
and since the average N
2
O
5
2 5
demonstrate that their results are sensitive to the N O uptake
-
3
concentration for our mixed aerosol experiments was ∼180 ppb,
we estimated an average nitrate concentration which scaled
equivalently by a factor of 6. A nitrate concentration of 2 M is
also the concentration that leads to a parametrized γ value
equivalent to that of the empirical result for ammonium bisulfate
at 50% RH. As with Bertram and Thornton, we calculated
coefficient down to values of at least 10 and perhaps even
lower,1 so deciphering the role of organics in suppressing the
-5
-
2
room temperature uptake coefficient from values above 10 ,
as present with soluble inorganic and small organics, is very
important. It does appear, however, that the reactivity on a purely
biogenic organic particle, which demonstrates uptake coefficients
-4
[
H
2
O
(l)] using the aerosol inorganics model (AIM II), where
of 10 , is unlikely to significantly impact atmosphere processes
or influence model calculations to a high degree. Given that
ambient measurements do not directly match the behavior
demonstrated in the few laboratory studies of mixed composition
we assume that the organics do not contribute significant water
to the particles.38,53
The resulting uptake coefficients of the parametrization
increased from 0.004 to 0.016 while the [H
2
O(l)] increased from
2 5
particles, the mechanism by which organics suppress N O
7
.2 to 27.1 M as the inorganic fraction increased. The results
heterogeneous reactivity within the atmosphere needs to be
better determined before it can be included in atmospheric
models.
shown in Figure 6 (open squares) illustrate no sharp initial
decline as with the empirical data, with relatively minor
subsequent decreases compared to the laboratory results. This
supports the contention that the kinetics suppression is due to
a mechanism different from the processes parametrized by this
model, probably driven by the different species present or by
different particle morphologies.
Acknowledgment. This project was financially supported by
the NSERC. The building of the new CIMS instrument was
supported by the NSERC and the Ontario Ministry of the
Environment. Such support does not indicate endorsement by
the Ministry of the contents of this material. We acknowledge
the support of G. Huey and D. Tanner (Georgia Institute of
Technology) for their designs for the CIMS, as well as Glenn
Wolfe (University of Washington) for the data acquisition
program for the CIMS and Joel Thornton for discussions.
Likewise, we note that direct measurements of the reactivity
2 5
of N O on ambient aerosol do not fully match the observations
made in the laboratory. In particular, the results of Bertram and
co-workers27 illustrate that the uptake coefficient does indeed
show a dependence on the organic mass fraction (Figure 6) at
70% RH for measurements obtained in Seattle, WA. The decline
in the observed uptake coefficient with increasing organic mass
fraction is not nearly as strong as that observed in the laboratory.
The uptake coefficient drops from 0.031 to 0.0016 for inorganic:
organic mass ratios ranging from 30:70 to 7:93, and the γ values
References and Notes
(
1) Evans, M. J.; Jacob, D. J. Geophys. Res. Lett. [Online] 2005, 32,
DOI 10.1029/2005GL022469. http://www.agu.org.myaccess.library.
utoronto.ca/journals/gl/gl0509/2005GL022469/2005GL022469.pdf, accessed
July 25, 2010.
are consistently much higher than all other laboratory results
(
2) Dentener, F. J.; Crutzen, P. J. J. Geophys. Res. 1993, 98, 7149–
163.
(3) Davis, J. M.; Bhave, P. V.; Foley, K. M. Atmos. Chem. Phys. 2008,
, 5295–5311.
4) Riemer, N.; Vogel, H.; Vogel, B.; Anttila, T.; Kiendler-Scharr, A.;
of mixed particles.8
,27,31
This could arise for a number of reasons.
7
For example, the particles in Seattle may be externally mixed,
such that a fraction of inorganic-rich particles keep the reactivity
high, even in the presence of a substantial number of low-
reactivity organic-rich particles, or there may be soluble
8
(
Mentel, T. F. J. Geophys. Res. [Online] 2009, 114, DOI 10.1029/
2008JD011369. http://www.agu.org.myaccess.library.utoronto.ca/journals/
jd/jd0917/2008JD011369/2008JD011369.pdf, accessed July 25, 2010.
(5) Liao, H.; Seinfeld, J. H. J. Geophys. Res. [Online] 2005, 110, DOI
inorganics present on the surface of the field particles. Although
the researchers27 note that the inorganic fraction is not entirely
1
0.1029/2005JD005907. http://www.agu.org.myaccess.library.utoronto.ca/
composed of ammonium sulfate and therefore may contain other
inorganic material, it is unlikely that higher N O reactivity
2 5
journals/jd/jd0518/2005JD005907/2005JD005907.pdf, accessed October 13,
2010.
(
6) Mozurkewich, M.; Calvert, J. G. J. Geophys. Res. 1988, 93, 15889–
occurs as a result of these unknown inorganics, since ammonium
sulfates react as rapidly as any other inorganic substrate. Finally,
the organics present in the field may be different from those in
the laboratory, perhaps having been oxidatively processed
at the surface, which will have increased their ability to take
up water.
1
5896.
(
7) Hu, J. H.; Abbatt, J. P. D. J. Phys. Chem. A 1997, 101, 871–878.
(8) Folkers, M.; Mentel, T. F.; Wahner, A. Geophys. Res. Lett. [Online]
003, 30, DOI 10.1029/2003GL017168. http://www.agu.org.myaccess.
2
library.utoronto.ca/journals/gl/gl0312/2003GL017168/2003GL017168.pdf, accessed
March 3, 2010.
(
9) Mentel, T. F.; Sohn, M.; Wahner, A. Phys. Chem. Chem. Phys.
999, 1, 5451–5457.
10) Wahner, A.; Mentel, T. F.; Sohn, M.; Stier, J. J. Geophys. Res.
1998, 103, 31103–31112.
11) Thornton, J. A.; Braban, C. F.; Abbatt, J. P. D. Phys. Chem. Chem.
1
(
5
. Conclusion and Atmospheric Implications
(
We report laboratory measurements of the uptake coefficient
Phys. 2003, 5, 4593–4603.
of N
2
O
5
hydrolysis on ammonium bisulfate and SOA single-
(12) Griffiths, P. T.; Badger, C. L.; Cox, R. A.; Folkers, M.; Henk, H. H.;
Mentel, T. F. J. Phys. Chem. A 2009, 113, 5082–5090.
component particles as well as ammonium bisulfate-SOA
mixed particles. Our values for the AB and SOA single-
component studies are in agreement with previous work for both
(
13) Thornton, J. A.; Abbatt, J. P. D. J. Phys. Chem. A 2005, 109, 10004–
0012.
(14) Behnke, W.; George, C.; Scheer, V.; Zetzsch, C. J. Geophys. Res.
1
aqueous AB8
,48-50
7-9,11,13-15,31,48,49
1997, 102, 3795–3804.
and aqueous non-AB
systems
8
,31
(15) Stewart, D. J.; Griffiths, P. T.; Cox, R. A. Atmos. Chem. Phys.
004, 4, 1381–1388.
(16) McNeill, V. F.; Wolfe, G. M.; Thornton, J. A. J. Phys. Chem. A
and other organic experiments, respectively. We observe a
strong suppression in γ upon the addition of increasingly larger
organic fractions, with the most notable decrease occurring
between pure AB particles and the addition of a small organic
fraction. This initial decline suggests that mass accommodation
is possibly inhibited with the inclusion of organics in the particle.
2
2007, 111, 1073–1083.
17) McNeill, V. F.; Patterson, J.; Wolfe, G. M.; Thornton, J. A. Atmos.
Chem. Phys. 2006, 6, 1635–1644.
18) Park, S. C.; Burden, D. K.; Nathanson, G. M. J. Phys. Chem. A
2007, 111, 2921–2929.
(
(

Kinetics of N2O5 Hydrolysis
J. Phys. Chem. A, Vol. 114, No. 50, 2010 13121
(
19) Cosman, L. M.; Bertram, A. K. J. Phys. Chem. A 2008, 112, 4625–
(33) Kercher, J. P.; Riedel, T. P.; Thornton, J. A. Atmos. Meas. Tech.
2009, 2, 193–204.
4
8
2
635.
(20) Bertram, T. H.; Thornton, J. A. Atmos. Chem. Phys. 2009, 9, 8351–
(34) George, I. J.; Abbatt, J. P. D. Atmos. Chem. Phys. 2010, 10, 5551–
363.
5563.
(
21) Cosman, L. M.; Knopf, D. A.; Bertram, A. K. J. Phys. Chem. A
(35) Wex, H.; Petters, M. D.; Carrico, C. M.; Hallbauer, E.; Massling,
A.; McMeeking, G. R.; Poulain, L.; Wu, Z.; Kreidenweis, S. M.; Stratmann,
F. Atmos. Chem. Phys. 2009, 9, 3987–3997.
(36) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics:
From Air Pollution to Climate Change; John Wiley & Sons, Inc.: New
York, 1998.
(37) Tang, I. N.; Munkelwitz, H. R. J. Geophys. Res. 1994, 99, 18801–
18808.
(38) Clegg, S. L.; Brimblecombe, P. J. Phys. Chem. A 1998, 102, 2137–
2154.
(39) CRC Handbook of Chemistry and Physics, 90th ed.; Lide, D. R.,
Ed.; CRC Press/Taylor and Francis: Boca Raton, FL, 2010.
(40) Kostenidou, E.; Pathak, R. K.; Pandis, S. N. Aerosol Sci. Technol.
2007, 41, 1002–1010.
(41) Stokes, R. H.; Robinson, R. A. J. Phys. Chem. 1966, 70, 2126–
2130.
(42) Zdanovskii, A. Zh. Fiz. Khim. 1948, 22, 1475–1485.
(43) Badger, C. L.; George, I.; Griffiths, P. T.; Braban, C. F.; Cox, R. A.;
Abbatt, J. P. D. Atmos. Chem. Phys. 2006, 6, 755–768.
(44) Brooks, S. D.; DeMott, P. J.; Kreidenweis, S. M. Atmos. EnViron.
2004, 38, 1859–1868.
(45) Gysel, M.; Crosier, J.; Topping, D. O.; Whitehead, J. D.; Bower,
K. N.; Cubison, M. J.; Williams, P. I.; Flynn, M. J.; McFiggans, G. B.;
Coe, H. Atmos. Chem. Phys. 2007, 7, 6131–6144.
(46) Brown, R. L. J. Res. Natl. Bur. Stand. 1978, 83, 1–8.
(47) Fuchs, N. A.; Sutugin, A. G. Highly Dispersed Aerosols; Ann Arbor
Science Publishers: Ann Arbor, MI, 1970.
008, 112, 2386–2396.
(
Phys. Chem. A 2005, 109, 7458–7465.
(
15–325.
(
005, 109, 7449–7457.
(
22) Lawrence, J. R.; Glass, S. V.; Park, S. C.; Nathanson, G. M. J.
23) D a¨ umer, B.; Niessner, R.; Klockow, D. J. Aerosol Sci. 1992, 23,
3
24) Lawrence, J. R.; Glass, S. V.; Nathanson, G. M. J. Phys. Chem. A
2
25) Stemmler, K.; Vlasenko, A.; Guimbaud, C.; Ammann, M. Atmos.
Chem. Phys. 2008, 8, 5127–5141.
(
26) Gilman, J. B.; Vaida, V. J. Phys. Chem. A 2006, 110, 7581–7587.
(27) Bertram, T. H.; Thornton, J. A.; Riedel, T. P.; Middlebrook, A. M.;
Bahreini, R.; Bates, T. S.; Quinn, P. K.; Coffman, D. J. Geophys. Res. Lett.
Online] 2009, 36, DOI 10.1029/2009GL040248. http://www.agu.org.
myaccess.library.utoronto.ca/journals/gl/gl0919/2009GL040248/2009GL040248.
pdf, accessed March 3, 2010.
[
(28) Brown, S. S.; Ryerson, T. B.; Wollny, A. G.; Brock, C. A.; Peltier,
R.; Sullivan, A. P.; Weber, R. J.; Dub e´ , W. P.; Trainer, M.; Meagher, J. F.;
Fehsenfeld, F. C.; Ravishankara, A. R. Science 2006, 311, 67–70.
(
Ravishankara, A. R. ReV. Sci. Instrum. 2002, 73, 3291–3301.
(
Whitlow, S.; Williams, E. J.; Lerner, B. M.; Jakoubek, R.; Middlebrook,
A. M.; DeGouw, J. A.; Warneke, C.; Goldan, P. D.; Kuster, W. C.;
Angevine, W. M.; Sueper, D. T.; Quinn, P. K.; Bates, T. S.; Meagher, J. F.;
Fehsenfeld, F. C.; Ravishankara, A. R. Geophys. Res. Lett. [Online] 2004,
29) Brown, S. S.; Stark, H.; Ciciora, S. J.; McLaughlin, R. J.;
30) Brown, S. S.; Dibb, J. E.; Stark, H.; Aldener, M.; Vozella, M.;
3
1, DOI 10.1029/2004GL019412. http://www.agu.org.myaccess.library.
utoronto.ca/journals/gl/gl0407/2004GL019412/2004GL019412.pdf, accessed
October 13, 2010.
(48) Kane, S. M.; Caloz, F.; Leu, M. T. J. Phys. Chem. A 2001, 105,
6465–6470.
(
31) Badger, C. L.; Griffiths, P. T.; George, I.; Abbatt, J. P. D.; Cox,
R. A. J. Phys. Chem. A 2006, 110, 6986–6994.
32) Zhang, Q.; Jimenez, J. L.; Canagaratna, M. R.; Allan, J. D.; Coe,
(
49) Hallquist, M.; Stewart, D. J.; Stephenson, S. K.; Cox, R. A. Phys.
Chem. Chem. Phys. 2003, 5, 3453–3463.
(
(
(
50) Griffiths, P. T.; Cox, R. A. Atmos. Sci. Lett. 2009, 10, 159–163.
H.; Ulbrich, I.; Alfarra, M. R.; Takami, A.; Middlebrook, A. M.; Sun, Y. L.;
Dzepina, K.; Dunlea, E.; Docherty, K.; DeCarlo, P. F.; Salcedo, D.; Onasch,
T.; Jayne, J. T.; Miyoshi, T.; Shimono, A.; Hatakeyama, S.; Takegawa, N.;
Kondo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S.; Weimer, S.;
Demerjian, K.; Williams, P.; Bower, K.; Bahreini, R.; Cottrell, L.; Griffin,
R. J.; Rautiainen, J.; Sun, J. Y.; Zhang, Y. M.; Worsnop, D. R. Geophys.
Res. Lett. [Online] 2007, 34, DOI 10.1029/2007GL029979. http://
www.agu.org.myaccess.library.utoronto.ca/journals/gl/gl0713/2007GL029979/
51) Bahreini, R.; Keywood, M. D.; Ng, N. L.; Varutbangkul, V.; Gao,
S.; Flagan, R. C.; Seinfeld, J. H.; Worsnop, D. R.; Jimenez, J. L. EnViron.
Sci. Technol. 2005, 39, 5674–5688.
(
52) Anttila, T.; Kiendler-Scharr, A.; Tillmann, R.; Mentel, T. F. J. Phys
Chem. A 2006, 110, 10435–10443.
53) Carslaw, K. S.; Clegg, S. L.; Brimblecombe, P. J. Phys. Chem.
995, 99, 11557–11574.
(
1
2
007GL029979.pdf, accessed April 21, 2010.
JP107721V