Heterogeneous interaction and reaction of HONO on ice films between 173 and 230 K

Article abstract of DOI:10.1021/jp9937151

Both the heterogeneous interaction of HONO on the ice surface and the heterogeneous reaction of HONO with HBr on the ice surface have been investigated in a flow reactor interfaced with a differentially pumped quadrupole mass spectrometer. The surface uptake amount and initial uptake coefficient of HONO on the ice surface were determined as a function of the ice-film temperature between 173 and 205 K. The reaction probability of HONO over the HBr-treated ice surfaces has been determined as a function of HBr partial pressures at 190K, 200K, and 230K, respectively. The reaction mechanism is proposed and discussed. Kinetic analysis indicates that the heterogeneous reaction of HONO with HBr on ice surfaces follows the Eley-Rideal type.

Full text of DOI:10.1021/jp9937151

3150
J. Phys. Chem. A 2000, 104, 3150-3158
Heterogeneous Interaction and Reaction of HONO on Ice Films between 173 and 230 K
Liang Chu, Guowang Diao, and Liang T. Chu*
Department of EnVironmental Health and Toxicology, State UniVersity of New York at Albany and
Wadsworth Center, P.O. Box 509, Albany, New York 12201-0509
ReceiVed: October 18, 1999; In Final Form: January 11, 2000
Both the heterogeneous interaction of HONO on the ice surface and the heterogeneous reaction of HONO
with HBr on the ice surface have been investigated in a flow reactor interfaced with a differentially pumped
quadrupole mass spectrometer. The surface uptake amount and initial uptake coefficient of HONO on the ice
surface were determined as a function of the ice-film temperature between 173 and 205 K. The reaction
probability of HONO over the HBr-treated ice surfaces has been determined as a function of HBr partial
pressures at 190K, 200K, and 230K, respectively. The reaction mechanism is proposed and discussed. Kinetic
analysis indicates that the heterogeneous reaction of HONO with HBr on ice surfaces follows the Eley-
Rideal type.
I. Introduction
molecules react each other on the ice surface. HBr may
potentially form hydrate(s) near the ice surface at 190 and 200K
Nitrous acid is an important trace gas in the troposphere. The
gas-phase chemistry of HONO plays a role in the formation of
hydroxyl radicals (OH) because HONO photolyzes rapidly to
produce OH.1 Because of its rapid photolysis at daytime,
elevated concentrations of HONO have normally been observed
only at night ranging from a few ppbv at polluted sites to 70
11
in the fast-flow reactor. At 230 K, HBr could possibly be in
12
a liquid phase, according to the HBr phase diagram. What is
the difference in reaction mechanisms at 190 and 200K versus
at 230K? Also, for the purpose of atmospheric chemistry, a
reliable rate constant at atmospheric conditions provides a useful
tool to model complex atmospheric reactions. A reliable
heterogeneous reaction probability under atmospheric conditions
can be extrapolated from the mechanistic studies. The hetero-
geneous reaction of HONO + HBr may play a role in the
activation of bromine in the troposphere. This motivated us to
study the reaction of HONO + HBr(s) f BrNO + H2O(s) on
ice surfaces.
In this paper, we report the results of our experiments on the
uptake amount, uptake coefficient of HONO on the ice surface,
and the reaction probability for the reaction of HONO + HBr(s)
f BrNO + H2O(s) on ice surfaces. In the following sections,
we will briefly describe the experimental procedures used in
the determination of the uptake amount, the uptake coefficient,
and the reaction probability. We will present the results of the
uptake amount and uptake coefficient of HONO on the ice
surface as a function of ice-film temperatures. The reaction
probability of the HONO + HBr reaction is a function of partial
HBr pressures and ice-film temperatures. Finally, we will discuss
the nature of the interaction between HONO and ice, and a
reaction mechanism for the HONO + HBr reaction will be
proposed.
,2
3
,4
pptv in the Arctic. Recent observations reveal that daytime
steady-state HONO concentrations of 100-500 pptv have been
observed in the clean troposphere.2
The atmospheric chemistry of bromine species is character-
ized by their short lifetimes and their ready availability for gas-
phase catalytic cycles in the stratosphere.6-8 The longest-lived
reactive bromine species is HBr, but it constitutes only a small
,5
9
fraction of the total reactive bromine present in the atmosphere.
Even though it has a low concentration, HBr could be ac-
cumulated on the ice surface.10-13 Heterogeneous bromine
reactions play a role in converting bromine reservoir compounds,
9
BrONO2 and HBr, into photochemically reactive species. The
heterogeneous reaction of HBr on ice surfaces is limited to a
few studies. Chu and Chu studied the heterogeneous reaction
of HOCl + HBr on ice films.14 Abbatt and Allanic et al.
10
15
studied the heterogeneous reaction of HOBr + HBr on ice films.
Hanson and Ravishankara investigated the heterogeneous reac-
1
3
tions of ClONO2 + HBr and Cl2 + HBr on ice films. With
the higher HONO concentration in the troposphere and bromine
reservoir HBr, it is reasonable to speculate that HONO may
react with HBr on the ice surface. Recently, Seisel and Rossi
demonstrated the feasibility of this reaction.16
II. Experimental Section
Seisel and Rossi studied the reaction of HONO + HBr in
the temperature range between 180 and 200 K and reported the
The uptake coefficient is defined as the ratio of the number
of molecules that are taken by the surface to the total number
of molecules colliding on the ice surface. Three different
experiments, the uptake amount, uptake coefficient of HONO
on the ice surface, and the reaction probability of HONO +
HBr on the ice surface, were performed in a flow reactor coupled
to a differentially pumped quadrupole mass spectrometer (QMS).
Some of the apparatus details have been discussed in our
-
3
-2
reaction probability to be 1.0 × 10 to 2.2 × 10 , with almost
no temperature dependence.16 The rate of the reaction may be
slightly accelerated by increasing HBr concentration. However,
little is known about the mechanism of this reaction. From a
fundamental surface chemistry standpoint, it is important to
know which reactants adsorb on the ice surface and how
1
1,14,17
previous publications,
and we provide only a brief
*
To whom correspondence should be addressed. e-mail:
lchu@csc.albany.edu.
description and some modifications in this paper.
1
0.1021/jp9937151 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/22/2000

Interaction and Reaction of HONO on Ice Films
J. Phys. Chem. A, Vol. 104, No. 14, 2000 3151
Flow Reactor. The cylindrical flow reactor was constructed
of Pyrex glass. Its dimensions were 1.70 cm i.d. and 35 cm in
length. The temperature of the reactor was regulated by a liquid
nitrogen-cooled ethanol circulator (Neslab), and was measured
with a pair of J-type thermocouples located in the middle and
at the downstream end. During the experiment, the temperature
was maintained at 173, 180, 190, 200, and 230 K, and the
stability of the temperature was better than 0.3 K in every
experiment. The pressure of the reactor was controlled by a
downstream throttle valve (MKS Instrument, Model 651C), and
the stability of the pressure was better than 0.001 Torr in every
experiment. A double-capillary injector was used to admit both
HONO and HBr into the flow reactor during the reaction
probability measurement. The reactant HONO vapor was taken
from a nitrous acid solution. HONO vapor contains a small
amount of water vapor. To avoid the condensation of the water
vapor and reactants on the capillary wall at low temperatures,
room-temperature dry air was passed through the outside of the
capillary to keep it warm.
warmed to 230 K. Besides the extra ice film and additional water
vapor, the total pressure of the reactor was increased to 2.0 Torr,
to reduce the loss of the ice film.
HBr-He mixtures. The HBr-He mixture was prepared by
mixing HBr (Matheson, 99.8%) and helium (Praxair, scientific
grade 99.9995%) in an all-glass manifold, which had been
-
6
previously evacuated to ∼10 Torr. The typical HBr-to-He
-
3
-5
mixing ratio was between 10 and 10 . HBr, along with
additional helium carrier gas, was introduced into the flow
reactor via the glass and PFA tubing, and the amount was
controlled by either a stainless steel flow controller (Teledyne-
Hastings) or Monel metering valve. All of the tubings and valves
were passivated by the HBr-He mixture to establish equilib-
rium, as monitored by the QMS prior to every experiment.
HONO Preparation and Calibration. Nitrous acid was
prepared by mixing a NaNO2 solution with an H2SO4 solu-
2
0,21
tion.
A 2.76 g quantity of NaNO2 (Aldrich) was dissolved
in 200 mL of distilled water; a few drops of 40% H2SO4 solution
were added while the NaNO2 solution was stirred. The acidity
of the solution was adjusted to pH 4.0, as monitored by a pH
meter (Corning). A colorless, clear HONO solution was
obtained. The prepared HONO solution was kept in the dark at
Ice-Film Preparation. The ice film was prepared as fol-
lows: Helium carrier gas was bubbled through a distilled-water
reservoir. Deionized water was purified with a Millipore Mili-Q
water system. The purified, distilled water had a resistivity g18
MΩ‚cm and was used in the reservoir. The reservoir was
maintained at 293.2 ( 0.1 K by a refrigerated circulator (Neslab
RTE-100LP). Helium saturated with the water vapor was
admitted to an inlet of the double-capillary injector. The ice
film was separated into two sections. The first section was ∼16.5
cm in length, and the second section was about 5 cm. The first
section was used to conduct the experiments, and the second
section, with extra thick ice (∼0.1 mm), was used to keep the
first section’s ice film from being evaporated in the reactor.
The two sections were separated by approximately 5-10 cm.
Note that the reactants, HONO and HBr, made contact with
the first section of the ice film only. During the course of ice
deposition, the double-capillary injector was slowly pulled out
at a constant rate, 2-3 cm/min, and a uniform ice film was
deposited on the inner surface of the reactor, which was between
2
73 K.
Helium carrier gas was bubbled through the HONO solution.
Both the HONO vapor and the water vapor from the HONO
solution were admitted into the reactor. The HONO vapor at
2
2
73 K was measured by an FTIR spectrometer (Mattson, RS-
) with a narrow band MCT detector to ensure the vapor-phase
composition. The IR spectra showed a few strong transitions
-
1
-1
of the N-O stretch at 1700 cm , the HON bend at 1263 cm ,
-
1
and the O-N stretch at 790 cm for the trans-HONO isomer.
-
1
For the cis-HONO isomer, the N-O stretch at 1641 cm and
-
1
the O-N stretch at 852 cm were observed. All of the
absorption frequencies in the IR spectrum agreed with the
2
2
standard database.
The gas-phase HONO concentration was determined by a
long-path IR absorption experiment based on the effective
2
3
HONO absorption cross-section. BaF2 windows were used in
the IR cell. The partial pressure of HONO near the outlet of
HONO bubbler was determined to be 0.019 and 0.013 Torr at
1
73 and 205 K. The amount of ice substrate deposited in the
first section was determined from the mass flow rate of the water
vapor and the deposition time. The average film thickness was
calculated by using the measured ice film geometric area, the
-
1
7
90 and 852 cm , respectively. The major gaseous impurities
were NO2 and NO, in addition to H2O. The NO2 and NO
impurities were approximately 6- and 13-fold higher than that
of the HONO concentration, as determined by the infrared
absorption.
3
mass of the ice, and the bulk density, 0.63 g/cm , of vapor-
deposited water ice.18 The average film thickness was about 30
µm at 190 and 200 K. After the uniform ice film was deposited,
the injector was rapidly moved to the second section, and the
extra thick ice film was deposited near the end of flow tube.
Because the injector and helium carrier gases could be a few
degrees warmer than the ice film, they would heat the thick ice
film at the end of flow tube. This provided a water vapor source
that was always slightly higher than or equal to the ice vapor
pressure at the ice film temperature in the flow reactor. This
water vapor was supplied to the first section of ice film to
prevent the ice film from evaporating during the experiment.
Additional water vapor from the HONO solution, which was
nearly in equilibrium with the vapor pressure of ice, was also
introduced to compensate for the ice-film loss from the
evacuation.
The HONO vapor pressure above the solution could be, in
principle, determined from Henry’s law. The concentration of
HONO in the solution was calculated from the following
equilibrium and Henry’s law:
-
+
NO
+ H h HONO
(1)
(2)
2
+
-
[
H ][NO ]
2
K )
a
[HONO]
+
-
where [H ] was determined by the pH meter and [NO2 ] was
the equilibrium concentration of the NaNO solutions. Using a
2
-
4
dissociation constant, K , of 5.6 × 10 at 298.15 K and an
a
2
4
Because the ice film sublimation rate was substantially higher
at 230 K,19 the loss of the ice film due to evacuation in the
flow reactor was noticeable if no additional water vapor was
introduced to the reactor. The ice film was difficult to grow at
enthalpy change of eq 1, the gas-phase HONO concentration
could be calculated from the following equation.
-
[
NO ]
2
T
2
30 K. For measurements conducted at 230 K, the ice film was
PHONO
)
(3)
+
K (1 + K /[H ])
vapor deposited at 200 K. The solid ice film was then rapidly
H
a

3152 J. Phys. Chem. A, Vol. 104, No. 14, 2000
Chu et al.
where KH is a Henry’s law constant. KH ) 49 M/atm at 298.15
K and was determined to be 218 M/atm at 273.15 K, based on
d(ln KH)/d(1/T) ) 4900.2
1,25
[NO2 ]T is the initial NaNO2
-
concentration or total N(III) concentration. We found that the
calculated gaseous HONO concentration was more than 2-fold
higher than that determined from the IR measurement. The exact
reason is unknown to us. One possible cause of this could be a
combination of HONO decomposition, the HONO loss on the
wall of transfer lines and cell, and perhaps the uncertainties in
equilibrium constants.26 In this paper, we used the mean gaseous
HONO concentration determined from the IR absorption because
it was closest to our experimental conditions.
In the nitrous acid solutions, HONO might be decomposed
5
,21,25,27
by the reactions
Figure 1. Plot of the HONO signal versus the reaction time (z/V) at
-
1
1
s
90 K. The pseudo first-order rate constant k ) 113 s and the radial
2HONO h NO + NO+ H O
(4)
(5)
and axial diffusion corrected rate constant k ) 121 s-1. The reaction
3
2
2
w
-
probability γ
w
) 6.5 × 10 . A signal for the product BrNO is also
2
NO + H O h HNO + HONO
2 2 3
shown in the figure.
in Figure 1. The pseudo first-order reaction rate constant, ks,
Gas-phase HONO, NO2, NO, and water vapor from the HONO
solutions were transferred via tubing and admitted into the
reactor. We were concerned that HONO might rapidly decom-
pose in the gas phase while being transferred to the reactor.
HONO, NO2, NO, and water vapor were assumed to be in
equilibrium over the nitrous acid solutions in the bubbler. The
HONO concentration was governed by Henry’s law. When the
helium gas was bubbled through the HONO solutions, HONO
could be decomposed in the transfer lines. We measured the
HONO signal using the QMS, after HONO flowed through a
was calculated from the least-squares fit of the experimental
-
1
data to eq 6. ks ) 113 s at 190 K was obtained and is shown
in the figure. The value of ks was then corrected for gas-phase
axial and radial diffusion using a standard procedure,29 and the
corrected rate constant was called kw. A diffusion coefficient
for HONO in helium was estimated using the Fuller equation,
3
0
and was expressed as
-
2
1.75
D ) 2.3853 × 10
T
/P
(7)
6-in. length of tubing and an 80-in. length of tubing, respectively.
where T is the temperature in Kelvin and P is the total pressure
of the reactor in Torr. The reaction probability γw was calculated
from kw using the following equation.
There was no noticeable difference in the HONO signal by the
-
QMS at m/e ) 47 between the two cases. This demonstrated
that a change in HONO concentration was not detectable during
the transfer to the flow reactor under the fast-flow condition.
The uncertainty of HONO concentration was estimated to be
about 20-30%, as determined by the FTIR. From our measure-
ments, there were no detectable reactions for NO2 + HBr and
NO + HBr over the ice surface at 190 K. The gas-phase
γ ) 2R k /(ω + R k )
(8)
w
w
w
where R is the radius of the flow reactor (0.85 cm) and ω is the
mean HONO molecular velocity at the ice-film temperature.
A layered pore-diffusion model was employed to correct the
ice-film porosity to obtain the “true” reaction probability γt.3
γt is equivalent to a reaction probability determined as if the
ice film were a “homogeneous” layer over the glass surface.
This is useful to model the surface reaction mechanism. On the
basis of previous studies conducted under nearly identical
2
8
reactions between NOx and HBr were negligibly small.
1-34
Determination of the Uptake Coefficient and Reaction
Probability. The procedure for the uptake-coefficient measure-
ment was nearly identical to the reaction-probability measure-
ments. We use the reaction probability measurements to illustrate
the experimental procedures. The reaction probability γw of
HONO with HBr on the ice film was determined as follows:
First, an ice film was vapor-deposited on the inner wall of the
flow reactor. Second, the first section of the film surface (see
ice-film preparation section) was pretreated with HBr at
3
2,18,34,35
conditions,
it is known that H2O ice films can be
approximately treated as hexagonally close-packed spherical
32
granules stacked in layers. The “true” reaction probability, γt,
is related to the value, γw, by
-
8
-5
pressures between 2.5 × 10 and 6.5 × 10 Torr for ∼10
min. The ice-film surface was not completely saturated by HBr.
Finally, HONO was admitted to the reactor with continuing HBr
flow in a separated capillary. The gas-phase loss of HONO was
x
3γw
γt )
(9)
1
/2
π{1 + η[2(N - 1) + (3/2) ]}
L
-
measured by the QMS at m/e ) 47 as a function of the injector
where the effectiveness factor η is the fraction of the film surface
position z. In all of the experiments, the concentration of HBr
was always larger than that of HONO. For a pseudo first-order
reaction under the plug-flow condition, the following equation
holds for the reactant HONO:
32
that participates in the reaction, and NL is the number of
36
granule layers. Approximately, this model can be considered
as the deconvolution of the reaction probability as if the reaction
occurred on the top layer of the surface. The detailed calculations
for these parameters can be found in ref 32. A tortuosity factor
ln[HONO] ) -k (z/V) + ln[HONO]
(6)
3
z
s
0
τ ) 4 and true ice density Ft ) 0.925 g/cm were used in the
above calculation. This was based on a treatment of the ice film
that was vapor-deposited on the flow tube at nearly identical
conditions.
where z is the injector position, V is the mean flow velocity,
HONO]z is the gas-phase HONO concentration measured by
3
1,32
[
the QMS at position z, and the subscript zero is the initial
reference injector position. For a typical experiment performed
on the ice film at 190 K, the first-order HONO decay is shown
The reaction probability of HONO + HBr can also be
determined by the formation of the product BrNO. The reaction
product, BrNO, was measured by the QMS at its parent peak

Interaction and Reaction of HONO on Ice Films
J. Phys. Chem. A, Vol. 104, No. 14, 2000 3153
-
7
Figure 3. Uptake of HONO on water-ice at PHONO)1.9 × 10 Torr
and 190 K. (b) represents the HONO signal. The uptake starts at t )
2
0
min when the injector was pulled out. The entire ice film (150 cm )
was saturated within 3 min. Desorption occurred when the injector was
pushed in. This procedure can be repeated a few times.
For purposes of comparison, Figure 2b shows the increase of
the BrNO signal, based on the direct measurement of the BrNO
parent signal. The pseudo first-order reaction rate constant was
-1
calculated to be 53 s . The two rate constants are in very good
agreement. It is important to point out that the HBr surface
concentration was assumed to be nearly constant in the
calculation of rate constant. This assumption can be justified
as the ice film was pretreated by HBr prior to the reaction
probability measurement. During the measurement, the loss of
HBr to the ice surface was significantly lower than the amount
HBr adsorbed on the ice surface. However, one would still
observe some gaseous HBr loss, which would not affect the
measurements significantly. A sound approach was to saturate
the ice surface by HBr. However, this would form hydrate(s)
Figure 2. (a) Plot of the fragment Br signal (∆) (S79 - 0.40S80) from
-
BrNO versus the reaction time. The signals from m/e ) 79(9) and
-
m/e ) 80 (b) were shown in the figure. The Br fragment signal from
BrNO was replotted in the form of ([BrNO]′
the solid triangles. The pseudo first-order reaction rate constant was
calculated from the Br signal using eq 10 to be 45 s . (b) Plot of the
BrNO signal versus the reaction time. The pseudo first-order reaction
rate constant was calculated from this data set to be 53 s , which is
shown as a dashed line.
∞
- [BrNO]′
z
), shown as
-
1
11,12
and erupt the HBr-treated ice film.
-
1
The uptake amount was determined by measuring the total
amount of HONO lost onto the ice surface till the surface was
saturated as monitored by the QMS. The experimental procedure
-
m/e ) 109. The increase in the BrNO signal was observed at
1
1,12,37
a high HBr partial pressure (See Figure 1). ks was calculated
was identical to our previous study.
The detailed proce-
from the following equation
dures can be found from these references.
ln([BrNO] - [BrNO] ) ) -k (z/V) + ln[BrNO]
(10)
∞
z
s
∞
Results
Uptake of HONO on Water-Ice. Figure 3 is a plot of the
gas-phase HONO signal versus the experimental time for the
exposure of HONO on an ice surface. In this experiment, an
ice film was deposited on the wall of the flow reactor. The gas-
where [BrNO]∞ was the BrNO concentration when all HONO
molecules were converted to BrNO, and [BrNO]z was the BrNO
concentration at the position z. By using BrNO data, ks was
-
1
calculated to be 96 s , which was very close to the ks value of
-
-1
phase HONO signal, as monitored by the QMS at m/e ) 47,
1
13 s determined from the HONO decay data shown in Figure
decreased rapidly when HONO reached the entire ice surface.
The entire ice film was saturated in less than 3 min at 190 K.
1.
Because the parent peak signal of BrNO was approximately
14
3% of the Br fragment in a BrNO mass spectrum,1 the parent
6
The uptake amount Θ was determined to be 1 × 10 molecules/
1
2
cm at 190 K, based on the geometric surface area of the flow
BrNO signal was difficult to detect by the QMS at a lower HBr
partial pressure. Instead of monitoring the parent peak BrNO,
reactor. For cases in which the HONO-saturated ice film was
“heated” by the injector (i.e., the injector was pushed back),
HONO was desorbed and quantitatively recovered as recorded
by the QMS. The uptake amount of HONO was almost identical
to the desorbed HONO amount. The uptake and desorption
procedure could be repeated several times. The uptake amount
of HONO on the ice film was measured at different surface
temperatures and is shown in Figure 4. The “heat of uptake”
∆Hupt of HONO on ice was determined to be -8.1 ( 2.0 kcal/
mol from the slope of a plot of log Θ versus 1/T (see Figure 4)
using eq 11.
-
-
both m/e ) 79 (Br) and m/e ) 80 (HBr) signals were recorded
-
by the QMS. The difference between the total m/e ) 79 signal
and Br fragment signal of HBr was the Br fragment signal
present in BrNO molecules. The ratio of the Br fragment signal
from the HBr molecule to HBr parent signal was determined
to be 40:100 by the QMS using pure HBr at an ionization
voltage of 66 eV condition. This ratio is in very good agreement
22
with the literature value of 39:100. Figure 2a illustrates the
-
total m/e ) 79 signal (S79), the HBr signal (S80), and the
difference (S79 - 0.40S80), which was proportional to the
concentration of BrNO. [BrNO]′) (S79 - 0.40S80) was re-
plotted in the form of ([BrNO]′∞ - [BrNO]′z), and the pseudo
∆H
RT
ln Θ ) -
+ C
(11)
-
1
first-order reaction rate constant was determined to be 45 s .

3154 J. Phys. Chem. A, Vol. 104, No. 14, 2000
Chu et al.
t
Figure 6. Plot of the “true” reaction probability, γ , versus partial HBr
pressures for the reaction of HONO + HBr on the ice surface at 190
K (b), 200 K (2), and 230 K (9). Ice-film thickness was 30 ( 2.2 µm
at 190 K, 30 ( 2.5 µm at 200 K, and 33 ( 3.1 µm at 230 K. The solid
lines were fitted to eq 21 at 190 and 200 K, and to eq 20 at 230 K.
Figure 4. Plot of the logarithm of HONO surface density versus 1/T.
The solid line is the least-squares fit to the experimental data, and the
dashed lines represent the 95% confidence level. The total pressure
-
7
was 0.50 Torr, PHONO ) 1.9 × 10 Torr and the film thickness was
3
0 ( 1 µm. The “heat of uptake” of HONO on ice was -8.1 ( 2.0
Figure 5 are one standard deviation ((σ) of the mean value.
The results indicate that the fraction of HONO molecules
sticking on the ice surface is very low. This is consistent with
the lower uptake amount at the same temperature range.
HONO + HBr f BrNO + H2O. The reaction probability
for the HONO + HBr f BrNO + H2O reaction was determined
by observing the decay of gas-phase HONO over the HBr-
treated ice surface as a function of the injection position as
shown in Figure 1. In the gas phase, there was no observable
reaction between HONO and HBr and between NOx and HBr.
The HONO loss was attributed to the HBr-treated ice surface.
The reaction probability γw was determined as a function of
PHBr and the ice-film temperature. The “true” reaction prob-
kcal/mol.
TABLE 1: Uptake Coefficient of HONO on the Ice Surface
between 178 and 200 Ka,b,c
T (K)
k
s
(s-1)
k
w
(s-1
)
γ
w
γ
t
-
-
-
-
-
-
3
3
3
3
4
4
-4
-5
-5
-5
-5
-5
1
1
1
1
1
2
78.2
81.7
88.9
92.0
96.9
00.2
56 ( 6
43 ( 5
35 ( 5
25 ( 4
14 ( 3
11 ( 3
58
45
37
26
14
11
3.7 × 10
2.6 × 10
2.2 × 10
1.6 × 10
8.0 × 10
6.4 × 10
1.4 × 10
8.1 × 10
6.4 × 10
3.9 × 10
1.7 × 10
1.3 × 10
a
total ) 0.500 ( 0.02 Torr. b Mean flow velocity ) 15.5 ( 0.3
P
c
m/s. Ice film thickness ) 30 ( 1.8 µm.
ability γt was calculated by using a layer model in which the
ice-film roughness was taken into the consideration.3
1,32
The
“
true” reaction probability γt as a function of PHBr at 190 K,
2
00 K, and 230 K is presented in Figure 6. The detailed
experimental conditions, γw, and γt are listed in Table 2. All
data listed in the table were averaged over 2-5 measurements.
The errors in Table 2 and Figure 6 represent one standard
deviation ((σ) of an average value. The surface concentration
of HBr used in this study was always greater than that of HONO,
thus the pseudo first-order condition was always satisfied.
Figure 6 shows that the “true” reaction probability γt increased
-
5
-3
-5
from 3.0 × 10 to 2.2 × 10 at 190 K and from 1.1 × 10
-
3
to 1.9 × 10 at 200 K, as PHBr increased. At a warmer
temperature, 230 K, the reaction probability γt showed a similar
Figure 5. Initial uptake coefficient γ of HONO on water-ice at 178
t
-
5
-4
to 200 K. The solid line is the least-squares fit of the experimental
data using eq 15.
trend, that is, it increased from 1.5 × 10 to 1.9 × 10 as
PHBr increased. However, γt is smaller than that of 190 and 200
K.
Uptake Coefficient of HONO on the Ice Film. The uptake
coefficient of HONO on the ice film was determined by
observing the loss of gas-phase HONO over the ice film surface
as a function of the injection position z. For every measurement,
the ice film was freshly prepared. The measured γw represents
the initial uptake coefficient. The uptake coefficients were
measured as a function of ice-film temperatures, and the results
are tabulated in Table 1. The table includes both γw and γt, and
the data were averaged over 2-6 measurements. Data analysis
was outlined in the Experimental Section. The “true” uptake
coefficient was calculated using eq 9, which was based on the
layered pore diffusion model.32 γt as a function of the ice film
temperature is shown in Figure 5. Figure 5 shows that γt
Table 2 shows that γ is ∼8- to 50-fold lower than γ . At a
t
w
-
2
higher γ (i.e., 10 ), the reaction is nearly completed at the
w
external surface before HONO reaches an internal ice surface,
and the correction for γ is smaller. At a lower γ , these HONO
t
w
molecules, which did not react with adsorbed HBr on the
external surface, have an opportunity to diffuse into the internal
surfaces and then react with HBr. The pore diffusion model
takes this into consideration, and the correction for γ is larger.
t
A reaction product, BrNO, was determined by the QMS using
two methods. (1) We took direct measurements, by the QMS,
-
at its parent peak m/e ) 109. The increase of the BrNO signal
was observed at 190 K (see Figures 1 and 2b) at a high
concentration of HBr. (2) We also took measurements of the
Br fragment signal of BrNO. The Br fragment signal was
observed to increase during the reaction even at a low
concentration of HBr (Figure 2a). Results from both of the
-
4
-5
decreased from 1.4 × 10 to 1.3 × 10 as the temperature
increased from 178 K to 200 K. γw showed a similar trend as
one could read from Table 1. The errors listed in Table 1 and

Interaction and Reaction of HONO on Ice Films
J. Phys. Chem. A, Vol. 104, No. 14, 2000 3155
TABLE 2: Reaction Probability for the Reaction of HONO + HBr f BrNO + H
2
O on Ice Films between 190 and 230 Ka,b,c,d
(s-1)
k
(s-1
)
γ
γ
w(BrNO)
e
γ
T (K)
P
HBr (Torr)
V (m/s)
k
s
w
w(HONO)
t(HONO)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
8
8
7
7
7
7
6
6
6
5
5
5
5
8
8
7
7
6
6
6
5
5
5
7
6
6
6
6
5
5
5
-3
-3
-3
-3
-3
-3
-3
-3
-2
-2
-2
-2
-2
-4
-4
-4
-4
-3
-3
-3
-2
-2
-2
-4
-4
-3
-3
-3
-3
-3
-3
-5
-5
-5
-5
-4
-4
-4
-4
-3
-3
-3
-3
-3
-5
-5
-5
-5
-5
-5
-4
-3
-3
-3
-5
-5
-5
-5
-5
-5
-5
-4
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
2
1
2
2
2
2
1
1
2
2
2
2
2
2
2
2
90.9 ( 0.3
90.0 ( 1.8
89.9 ( 1.1
90.2 ( 0.3
91.7 ( 0.7
91.4 ( 1.4
91.5 ( 0.7
91.3 ( 0.8
92.2 ( 1.0
92.7 ( 0.8
91.5 ( 1.2
92.0 ( 0.3
91.2 ( 1.5
00.2 ( 0.3
99.3 ( 1.2
00.3 ( 0.3
99.8 ( 0.5
00.5 ( 0.3
00.4 ( 0.3
00.7 ( 1.5
00.0 ( 0.3
99.8 ( 0.3
99.7 ( 0.4
29.8 ( 0.3
29.2 ( 0.3
30.0 ( 0.3
29.6 ( 0.3
28.3 ( 0.3
30.2 ( 0.5
29.2 ( 0.8
29.4 ( 0.8
2.9 × 10
7.4 × 10
1.1 × 10
2.2 × 10
4.7 × 10
9.5 × 10
2.0 × 10
3.1 × 10
6.0 × 10
2.2 × 10
3.2 × 10
4.9 × 10
6.5 × 10
2.5 × 10
7.4 × 10
2.3 × 10
7.3 × 10
1.6 × 10
3.0 × 10
6.6 × 10
2.9 × 10
4.5 × 10
6.5 × 10
5.1 × 10
1.0 × 10
1.9 × 10
3.2 × 10
5.7 × 10
1.1 × 10
2.0 × 10
3.4 × 10
3.8
3.8
3.8
26 ( 6
22 ( 5
28
23
23
24
54
99
92
138
221
314
307
306
304
10
1.6 ( 0.2 × 10
1.3 ( 0.3 × 10
1.3 ( 0.4 × 10
1.4 ( 0.1 × 10
3.1 ( 0.3 × 10
4.8 ( 2.7 × 10
4.1 ( 0.9 × 10
8.0 ( 3.4 × 10
1.3 ( 0.4 × 10
1.8 ( 0.4 × 10
1.8 ( 0.7 × 10
1.8 ( 0.3 × 10
1.7 ( 0.8 × 10
5.7 ( 4.0 × 10
5.9 ( 3.0 × 10
8.3 ( 5.0 × 10
9.2 ( 2.4 × 10
1.8 ( 0.2 × 10
2.1 ( 1.0 × 10
6.1 ( 2.2 × 10
1.3 ( 0.4 × 10
1.6 ( 0.2 × 10
1.6 ( 0.7 × 10
8.7 ( 1.6 × 10
7.6 ( 3.0 × 10
1.5 ( 0.9 × 10
1.7 ( 0.3 × 10
1.6 ( 1.3 × 10
1.7 ( 0.5 × 10
2.8 ( 0.4 × 10
4.5 ( 1.1 × 10
3.8 × 10
3.0 × 10
3.0 × 10
3.2 × 10
1.0 × 10
2.2 × 10
1.7 × 10
5.4 × 10
1.2 × 10
2.2 × 10
2.1 × 10
2.1 × 10
1.9 × 10
1.1 × 10
1.1 × 10
1.7 × 10
1.9 × 10
4.5 × 10
5.5 × 10
3.4 × 10
1.3 × 10
1.7 × 10
1.9 × 10
1.7 × 10
1.5 × 10
3.4 × 10
3.9 × 10
3.8 × 10
4.1 × 10
8.8 × 10
1.9 × 10
-3f
-4f
-3f
-3
1.1 × 10
8.7 × 10
1.5 × 10
2.6 × 10
22 ( 6
15.3
15.5
15.4
16.2
15.8
16.1
15.8
15.6
15.6
15.9
3.8
24 ( 3
52 ( 5
93 ( 45
88 ( 12
128 ( 62
198 ( 30
269 ( 60
262 ( 100
263 ( 58
255 ( 121
10 ( 7
-2
-2
1.6 × 10
1.4 × 10
-4f
-4f
3.8
3.8
10 ( 5
10
15
16
32
4.9 × 10
5.8 × 10
15 ( 9
15.3
15.5
15.4
15.7
16.0
15.6
15.7
9.7
16 ( 3
31 ( 5
34 ( 11
116 ( 42
205 ( 32
251 ( 31
255 ( 112
16 ( 3
36
127
232
295
303
17
15
29
32
31
33
54
-2
-2
1.2 × 10
1.1 × 10
9.9
15 ( 5
10.2
10.6
10.0
9.9
10.3
10.6
28 ( 21
31 ( 10
30 ( 9
32 ( 12
52 ( 9
79 ( 24
85
a
-7
The total pressure was 0.501 ( 0.01 Torr at 190 and 200 K (except PHBr < 2 × 10 Torr, where the total pressure was 2.00 ( 0.02 Torr).
b
c
The total pressure was 2.00 ( 0.02 Torr at 230 K. Ice film thickness was 30 ( 2.2 µm, 30 ( 2.5 µm, and 33 ( 3.1 µm at 190, 200, and 230
d
-7
-7
-8
K, respectively.
Torr.
text for details.)
P
HONO ) (1.9 ( 1.0) × 10 Torr with the exception of when PHBr is less than 7 × 10 Torr where PHONO ) (1.1 ( 0.9) × 10
e
f
γ
w(BrNO) was calculated from the BrNO signal. Means that γw(BrNO) was calculated from the Br fragment signal of the BrNO molecule. (See
methods indicated that BrNO was formed during the reaction,
and that the measured reaction probabilities were consistent with
each other, as shown both in Figure 2 and in Table 2.
represents the migration of HONO from the weakly bound state
to the adsorption site. From 178 to 200 K, the uptake amount
of HONO on the ice surface is relatively low, approximately 1
1
4
2
×
10 molecules/cm . This agrees with the idea that HONO
IV. Discussion
molecules trap onto a weakly bound state and also desorb to
the gas phase because of thermal energy. With the suggestion
of a lower γt or γw, a small fraction of HONO molecules are
able to move to the adsorption state between 178 and 200 K.
This fraction of molecules may control the uptake rate, and it
is reasonable to assume that the initial uptake coefficient is
limited by the rate of reaction 13. This can be expressed as
1
. Uptake of HONO on Ice Film. Figure 3 showed the
HONO loss on the ice surface as a function of the time. The
HONO loss rate onto the ice surface was approximately constant
at t ≈ 0-0.5 min and then decreased nonlinearly until it reached
saturation coverage. This suggested that the uptake coefficient
varied slightly at the lower coverage and then decreased to zero
at saturation coverage. This trend cannot be explained by
assuming that the adsorbate binds to a series of identical surface
sites, which predicts that the uptake coefficient varies linearly
with the coverage.38
The behavior of the HONO loss on the ice surfaces may be
explained by the following: A gas-phase HONO molecule
impinges on an occupied site that could not be adsorbed.
However, the molecule could be trapped into a weakly bound
state. The molecule can then diffuse around the surface and find
a site to adsorb. This can be summarized by the following
equations:
Rate ) k [HONO(p)]
(14)
3
[HONO(p)] can be calculated from reaction 12 using the steady-
state approximation. The uptake coefficient can be written in
terms of eq 14 as follows:
Rate
R
k2/k
γt )
)
)
(
1/4)[HONO(g)]ω(A/V)
3
R
(15)
ν /ν exp(-(E - E )/k T)
2
3
2
3
B
where ω is the molecular velocity of HONO, A/V is the surface-
to-volume ratio, R ) 4k1V/ωA, νi is the preexponential factor,
and kB is the Boltzmann constant. The “true” uptake coefficients
were fitted to eq 15, and the result is shown in Figure 5 as a
solid line. This line fit the experimental data very well from
k1
HONO(g) y
\
z HONO(p)
(12)
(13)
k2
k3
HONO(p)
98 HONO(ad)
1
78 to 200 K. E2 - E3 ) 6.5 kcal/mol was obtained from the
Reaction 12 represents the gas-phase HONO molecules entering
and desorbing out of the weakly bound state. Reaction 13
least-squares fit. E2 is the activation energy of molecules from
the weakly bound state to the gas phase. E3 is the activation

3156 J. Phys. Chem. A, Vol. 104, No. 14, 2000
Chu et al.
proposed by the following reactions:
k1
+
-
HBr(g) + 2S y
\
z HBr(ad) T H (H O)(ad) + Br (H O)(ad)
2
2
k2
(16)
H+
ks
-
-
Br (ad) + HONO 8 HONO‚‚‚Br (ad) H+8
BrNO(g) + H O(ad) (17)
2
We expect that reaction 17 is the rate-determining step. The
reasons are as follows: The initial uptake coefficient of HBr
-
6
on the ice surface (γw ) 0.12 at 190 K, PHBr ) 1.0 × 10
Torr) is always higher than the reaction probability of HONO
-
6
+
HBr reaction (γw ) 0.0048 at 190 K, PHBr ) 1.0 × 10
Figure 7. Initial uptake coefficient of HBr on water-ice between 181
-
6
Torr). HBr was in excess and HONO was the limiting agent in
this study. This suggests that reaction 16 cannot be the rate-
limiting step. The rate of the reaction is proportional to the HBr
surface concentration, and the partial pressure of HONO, PHONO
and 227 K. The total pressure was 0.50 Torr, PHBr ) 1.0 × 10 Torr,
and the film thickness was 28 ( 2.1 µm. (b) was the uptake coefficient
γ
w
t
of HBr on water-ice. (2) was the “true” uptake coefficient γ of
HBr on water-ice as corrected by the pore diffusion model. See details
in the text.
d[HONO]
R ) -
) k [HBr] P
ad HONO
(18)
energy of molecules from the weakly bound state to the
adsorption state. E2 - E3 ) 6.5 kcal/mol indicates the barrier
between the weakly bound state and the adsorption state. The
relatively high barrier is reflected in the small uptake amount
and lower uptake coefficient. We may compare this E2 - E3
value to a few similar systems. E2 - E3 ) 4.1 kcal/mol for
s
dt
The HBr surface concentration in eq 18 can be derived from
the steady-state approximation and adsorption equilibrium of
eq 16.
k1 1/2
1/2
1/2
ClONO2 hydrolysis on the ice surface and 5.6 kcal/mol for the
-
[
HBr](ad) ) [Br ]
)
PHBr S ) bHBrPHBr S (19)
HOBr uptake on the ice surface.3
9,40
These values are slightly
(ad) ( )
k
2
lower than E2 - E3 of HONO/ice and, thus, have higher uptake
coefficients.
where bHBr ) (k1/k2)1/2, So ) S + [Br ]ad + [H ]ad) S +
-
+
2
[HBr]ad, and So and S are the total number of sites on the
2
. HONO + HBr Reaction. First, we will examine some
surface and available site, respectively.
Reaction probability, γ, is given by R/(φHONO A/V), where
φHONO is the total HONO flux colliding on the surface (φHONO
experimental observations for HBr and HONO on the ice surface
and then draw some conclusions about the reaction mechanism.
A. HBr on Ice. The initial “true” uptake coefficient γt of HBr
on pure ice varied from 0.06 to 0.007 for cases in which the
temperature increased from 181 to 227 K, as shown in Figure
) PHONO/
2πmk T, where m is the molecular weight of
HONO), A/V is the surface-to-volume ratio, and is expressed
x
B
7
. The solid lines in Figure 7 are used to guide the reader’s
by eq 20
eye. γw is also shown in Figure 7 in the same scale. It is clear
that γt at 230 K is smaller than those found between 190 and
1
HBr
/2
1/2
b P
HBr HBr
cbHBr
k
P
S
R
s
0
V
A
γt )
)
) γo
2
00 K. It is important to note that the HBr uptake coefficient is
1/2 ( )
1/2
φHONO(A/V)
1 + 2bHBr
P
1 + 2bHBrP
HBr
HBr
higher than the reaction probability of HONO + HBr at the
same temperature and PHBr. The desorption temperature of HBr
(20)
-
6
was ∼225-230 K for PHBr ) 1 × 10 Torr, and the total
pressure of the flow reactor was 0.5 Torr.14 This suggested that
HBr adsorbed on the ice surface during the reaction.
where c ) 2πmk T, γ0 ) c(V/A) ks So, and bHBr is a constant
x
B
1/2
at a given temperature. The term bHBrPHBr represents the HBr
molecules dissociatively adsorbed on the ice surface based on
eq 16.
B. HONO on Ice. Figures 3 and 4 indicated that the HONO
uptake rate by the ice surface was low, and HONO was easy to
desorb by the warm (about 10 degree higher) injector. By
comparing the “heat of uptake”, ∆Hupt, of HONO to ∆Hupt of
HOCl on ice, we hope to draw some conclusions about the
interaction between HONO and water. The desorption temper-
For the experimental data at 190 K, the fitted result by using
eq 20 is shown in Figure 8 as the dashed line. Apparently, the
data were not well fitted. When HBr adsorbed on the ice surface
between 190 and 200 K, hydrates were likely to be formed near
11
the ice surface. Previously, we have shown the isothems for
4
1,42
ature of HOCl on ice is between ∼170-175 K.
The mean
f
HBr on the ice surface can be written as θ ) K P , where f )
“
heat of uptake” of HOCl on the ice surface was -10 ( 2 kcal/
11,12
0.83.
In terms of a mathematical expression of the isothems,
43
mol. ∆Hupt of HONO on ice (-8.1 ( 2.0 kcal/mol) is slightly
lower than that of HOCl. This indicates that the interaction
between HONO and ice was weaker than that of HOCl with
ice. We expect that the desorption temperature of HONO on
ice is likely lower than 175 K. The surface residence time τ of
HONO on ice at 190 and 200 K can be estimated from ∆Hupt
on the basis of τ ) τoexp(-∆Hupt /kBT) to be about milliseconds.
This indicates that the chance for HONO molecules to desorb
is fairly high, at 190 K or higher.
this is close to the Langmuir isotherm. If we allow HBr to form
hydrate(s), the expression for [HBr]ad in eq 20 must change.
We can write eq 20 as
k k
1
s
c
P S
HBr 0
R
k2
V
A
bHBrPHBr
γt )
)
) γo
(21)
( )
φHONO(A/V)
k1
1 + bHBrPHBr
1
+
PHBr
k2
C. Reaction Mechanism. On the basis of the above analysis,
HBr is adsorbed on the ice surface and HONO is mainly in the
gas phase between 190 and 230 K. A possible mechanism is
where we approximately treated the hydrate(s) as associative
forms, and the Langmuir isotherm was used. The chemical

Interaction and Reaction of HONO on Ice Films
J. Phys. Chem. A, Vol. 104, No. 14, 2000 3157
1
4
on ice occurs between ∼225-230 K in the fast-flow reactor.
This implies that the surface residence time of HBr is reduced
dramatically above the desorption temperature, and thus, the
adsorbed HBr amount is reduced (i.e., increased k2 in eq 16 or
decreased bHBr). HBr is not readily available for the reaction.
A reduction in the reaction rate is expected near the desorption
temperature and above.
3
. Comparison to Previous Study. For HONO uptake on
the ice surface, Fenter and Rossi reported that the uptake
-
3
44
coefficient γw of HONO was 1 × 10 at 180 K. We
determined that the uptake coefficient, γw, of HONO on water-
-
3
ice was 2.3 × 10 at 180 K. Within the uncertainties of
measurement, this study is in very good agreement with the
earlier study.
t
Figure 8. Plot of the reaction probability, γ , versus partial HBr
pressures for the reaction of HONO + HBr on the ice surface at 190
K (b). The solid line was fitted to eq 21, and the dashed line was
fitted to eq 20 for Eley-Rideal mechanism. The dotted line was fitted
to eq 22 if the reaction follows Langmuir-Hinshelwood mechanism.
The uptake amount of HONO was not studied previously.
The uptake amount of HONO was determined to be between 3
1
4
13
2
×
10 and 1 × 10 molecules/cm between 173 and 205 K in
this study. It is close to the uptake amount of HOCl, between
(See text for details.)
1
4
13
2
1
K.
× 10 and 5 × 10 molecules/cm between 189 and 220
14
nature for HBr‚nH2O is ionic within the molecular crystal. The
experimental data at 190 K were fitted by eq 21, and the result
is shown in Figure 8 as a solid line, which fit the experimental
data very well. Both solid lines at 190 and 200 K in Figure 6
were fitted by eq 21. However, for cases in which the data at
Seisel and Rossi reported that the average uptake coefficient
γw of HBr on ice was 0.32 ( 0.12 between 180 and 200 K and
showed a slightly negative temperature dependence from 180
16
to 200 K. Fl u¨ ckiger et al. reported a value between 0.2-0.34
2
30 K were fitted to eq 21, the result was unsatisfactory. The
45
from 190 to 210 K. Hanson and Ravishankara reported the
experimental data at 230 K could be fitted by eq 20 very well,
and the result is shown in Figure 6 as the solid line. The
conclusion drawn from this exercise is that at 190 and 200 K,
HBr may form hydrates, and the hydrate is adsorbed near the
ice surface. The adsorbed HBr reacts with incoming HONO
molecules via the Eley-Rideal mechanism to form BrNO,
which subsequently desorbs from the surface. The analysis
showed that HBr behaves like a dissociative adsorption on the
ice surface at 230 K. We speculate that either HBr is dissocia-
tively adsorbed on the ice surface or HBr is in a quasi-liquid
46
uptake coefficient was larger than 0.2 at 200 K. This study
determined the initial uptake coefficient γw to be between 0.17
and 0.04, which decreased as the temperature increased from
1
81 to 227K. The results of this study are slightly lower than
previous reports, but they show the same trend of the negative
temperature dependence. Within the uncertainties of the experi-
ments, we would consider our results in agreement with previous
results.
For the reaction HONO + HBr f BrNO + H2O, we may
compare our study to the previous measurement. Seisel and
+
layer that results in a higher ion mobility for H (H2O)n and
-3
Rossi reported that the reaction probability γw was 8.8 × 10
-
Br . This suggestion is supported by the HBr-ice-phase
-2
-4
-2
-
2.2 × 10 at 180 K, 1.7 × 10 - 2.2 × 10 at 190 K, and
-3 -3 16
diagram at the low temperature,12 in which HBr is either in the
metastable “ice” phase or in the liquid phase, depending on the
partial HBr pressure. Both states are believed to have free ions,
1
.3 × 10 - 9.7 × 10 at 200 K. Our results suggest that γw
-3 -2
is between 1.3 × 10 and 1.8 × 10 at 190 K and between
-4
-2
5
.7 × 10 and 1.6 × 10 at 200 K. Within the uncertainties
-
+
namely Br and H (H2O). In terms of the isotherm, it behaves
just like a dissociative adsorption and has the functional form
of eq 20.
of the measurement, this study is in excellent agreement with
the earlier study.
4. Atmospheric Implications. The heterogeneous formation
The above discussion was based on the experimental fact that
HONO was unlikely to be adsorbed on the ice surface at 190 K
and above. If we assume that both HONO and HBr were
adsorbed on the ice between 190 and 230 K, the reaction follows
a Langmuir-Hinshelwood mechanism. The reaction probability
can be expressed as
of photochemically active BrNO may play a role in the
partitioning of bromine compounds in the lower stratosphere
and in the troposphere. This study indicates that the bromine-
activation efficiency depends on the HBr concentration and
aerosol temperatures. The typical HBr concentration in the
47
stratosphere is ∼2pptv, which is ∼50-fold lower than the HBr
concentration used in our measurements. We expect that the
reaction probability, γw, for the HONO + HBr reaction would
1/2
bHONObHBrPHBr
γt ) γo
(22)
1/2
2
-3
be approximately 10 . If we compare this reaction to some
(1 + 2bHBr
P
+ bHONOPHONO)
HBr
important heterogeneous atmospheric processes with higher
4
8
reaction probabilities, e.g., γw ) 0.3, it is easily concluded
that bromine activation via the HONO + HBr is not a critical
process in the lower stratosphere. However, a small fraction of
BrNO can be formed in the dark night or polar winter time
over the PSC surfaces. This will change the bromine partitioning
in the lower stratosphere slightly.
where bHONO is an adsorption constant. An attempt was made
to fit the experimental data at 190 K by eq 22 at a constant
PHONO condition, and the result is shown in Figure 8 as the
dotted line. Clearly, it does not represent the experimental data
very well. This evidence further indicates that the reaction is
not likely to be Langmuir-Hinshelwood type.
D. Effect of the Temperature on the Reaction Probability.
The reaction probability decreases as the temperature increases.
The reaction probabilities at 190 and 200 K are similar, but the
reaction probability at 230 K is lower than that of 190 and 200
K. This may be attributed to the fact that the desorption of HBr
At the simulated polar tropospheric temperature of 230 K,
-4
the reaction probability γw is estimated at ∼10 . The photolysis
-
9 8
rate of HBr at the Arctic ground level is <10 . HBr is a good
reservoir compound in the troposphere. With the 70 pptv HONO
4
concentration in the Arctic, we can speculate that this reaction

3158 J. Phys. Chem. A, Vol. 104, No. 14, 2000
Chu et al.
may occur over Arctic snow/ice surfaces. If we assume the
photolysis rate of BrNO is similar to that of ClNO in the
troposphere, the photochemical lifetime of BrNO is on the order
of 1 h. This is a potential pathway to activate HBr in the
(20) Zhang, R.; Leu, M.-T.; Keyser L. F. J. Phys. Chem. 1996, 100,
3
39.
21) Park, J.-Y.; Lee, Y.-N. J. Phys. Chem. 1988, 92, 6294.
22) NIST Standard Reference Database Number 69. http://webbook.nist.
(
(
gov/chemistry/, November 1998 Release.
(23) Barney, W. S.; Wingen, L. M.; Lakin, M. J.; Brauers, T.; Stutz, J.;
troposphere and may contribute to an understanding of the Arctic
boundary ozone loss.49
Finlayson-Pitts, B. J. J. Phys. Chem. A 2000, 104, 1692.
th
(24) Lide, D. R. CRC Handbook of Chemistry and Physics, 78 edition;
CRC Press: New York, 1998; pp 8-44.
25) Becker, K. H.; Kleffmann, J.; Kurtenbach, R.; Wiesen, P. J. Phys.
Chem. A 1996, 100, 14 984.
26) Becker, K. H.; Kleffmann, J.; Kurtenbach, R.; Wiesen, P.; Febo,
A.; Gherarhi, M.; Sparapani, R. Geophys. Res. Lett. 1995, 22, 2485.
(27) Wiesen, P.; Kleffmann, J.; Kurtenbach R.; Becker K. H. Faraday
Discuss. 1995, 100, 121.
V. Conclusion
(
This study shows that HONO is rapidly saturated on the
water-ice surface between 173 and 230 K. The “true” uptake
coefficient of HONO on ice film was determined to be between
(
-
4
-5
1
.4 × 10 and 1.3 × 10 from 178 to 200 K. The “heat of
uptake” of HONO on ice was determined to be approximately
8.1 ( 2.0 kcal/mol. The reaction probability of HONO over
the HBr treated ice surface increases as PHBr increases over
(28) Rosser, W. A., Jr.; Wise, H. J. Phys. Chem. 1960, 64, 602.
(29) Brown, R. L. J. Res. Natl. Bur. Stand. (U. S.) 1978, 83, 1.
(30) Cussler, E. L. Diffusion, Mass Transfer in Fluid Systems; Cambridge
-
University Press: New York, 1984; Chapter 5.
31) Keyser, L. F.; Moore, S. B.; Leu, M.-T. J. Phys. Chem. 1991, 95,
496.
32) Keyser, L. F.; Leu, M.-T.; Moore, S. B. J. Phys. Chem. 1993, 97,
2800.
(
1
90-230 K. Kinetic analysis indicated that the heterogeneous
5
reaction of HONO with HBr on ice surfaces followed the Eley-
Rideal type.
(
(
33) Chu, L. T.; Leu, M.-T.; Keyser, L. F. J. Phys. Chem. 1993, 97,
7
1
779.
Acknowledgment. The authors would like to thank Professor
Barbara Finlayson-Pitts, who provided the infrared absorption
cross section data of HONO and a preprint of reference 23,
and two anonymous reviewers for helpful suggestions. The
authors also acknowledge financial support by the National
Science Foundation under Grant No. ATM-9530659.
(34) Chu, L. T.; Leu, M.-T.; Keyser, L. F. J. Phys. Chem. 1993, 97,
2 798.
(
35) Keyser, L. F.; Leu, M.-T. Micros. Res. Tech. 1993, 25, 343.
b
(36) NL) a x + N log(x + c) where a ) -0.1037, b ) 4.6841 × 10
-
15
,
N ) 10.6290, and c ) 0.8690. x () 0.5-35) is the film thickness in
µm. The parameters were derived from references 32 and 33. This expression
is valid for a thin ice film at about 200 K.
(
37) Chu, L. T.; Chu, L. J. Phys. Chem. B 1997, 101, 6271.
References and Notes
(38) Masel, R. I. Principles of Adsorption and Reaction on Solid
Surfaces; Wiley: New York, 1996, Chapters 5 and 7.
(
1) Calvert J. G.; Yarwood, G.; Dunker, A. M. Res. Chem. Intermed.
(39) Berland, B. S.; Tolbert, M. A.; George, S. M. J. Phys. Chem. A
1
994, 20, 463.
1
997, 101, 9954.
(
(
2) Lammel, G.; Cape, J. N. Chem. Soc. ReV. 1996, 25, 361.
3) Appel, B. R.; Winer, A. M.; Tokiwa, Y.; Biermann, H. W. Atmos.
(40) Chu, L.; Chu, L. T. J. Phys. Chem. A 1999, 103, 8640.
(41) Oppliger, R.; Allanic, A.; Rossi, M. J. J. Phys. Chem. 1997, 101,
EnViron. 1990, 24A, 611.
1
903.
42) Bahnam, S. F.; Horn, A. B.; Koch, T. G.; Sodeau, J. R. Faraday
Discuss. 1995, 100, 321.
43) Calculated on the basis of the results presented in reference 14.
(
(
(
4) Li, S.-M. J. Geophys. Res. 1994, 99, 25 469.
5) Vecera Z.; Dasgupta P. K. EnViron. Sci. Technol. 1991, 25, 255.
6) McElroy, M.B.; Salawitch, R. J.; Wofsy, S. C.; Logan, J. A. Nature,
(
(
1
986, 321, 759.
Hanson, D. R.; Ravishankara, A. R., J. Phys. Chem. 1992, 96, 2628. Abbatt,
J. P. D.; Molina, M. J. Geophys. Res. Lett. 1992, 19, 461. Brown, A. R.;
Doren, D. J. J. Phys. Chem. B. 1997, 101, 6308. Geoger, F. M.; Hicks, J.
M.; Dios, A. C. J. Phys. Chem. A 1998, 102, 1514.
(
(
7) Lary, D. J. J. Geophys. Res. 1996, 101, 1505.
8) Yung, Y. L.; Pinto, J. P.; Watson, R. T.; Sander, S. P. J. Atmos.
Sci. 1980, 37, 339.
(9) Lary, D. J.; Chipperfield, M. P.; Toumi, R.; Lenton, T. J. Geophys.
(
(
44) Fenter, F. F.; Rossi, M. J. J. Phys. Chem. 1996, 100, 13765.
45) Fl u¨ ckiger, B.; Thielmann, A.; Gutzwiller, L.; Rossi, M. J. Ber.
Res. 1996, 101, 1489.
(
(
(
(
(
(
10) Abbatt, J. P. D. Geophys. Res. Lett. 1994, 21, 665.
Bunsen-Ges. Phys. Chem. 1998, 102, 915.
46) Hanson, D. R.; Ravishankara, A. R. In The Tropospheric Chemistry
11) Chu, L. T.; Heron, J. W. Geophys. Res. Lett. 1995, 22, 3211.
12) Chu, L. T.; Chu, L. J. Phys. Chem. A 1999, 103, 384.
13) Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem. 1992, 96, 9441.
14) Chu, L.; Chu, L. T. J. Phys. Chem. A 1999, 103, 691.
15) Allanic, A.; Oppliger, R.; Rossi, M. J. J. Geophys. Res. 1997, 102,
(
of Ozone in the Polar Regions; Niki, H., Becker, K. H., Eds.; NATO ASI
Series; Verlag: NY.; 1993, pp 281-290.
(47) Johnson, D. G.; Traub, W. A.; Chance, K. V.; Jucks, K. W.
Geophys. Res. Lett. 1995, 22, 1373.
(48) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling, EVaulation 12; JPL: Pasadena, 1997; p 233.
(49) Barrie, L, A.; Bottenheim, J. W.; Schnell, R. C.; Crutzen, P. J.;
Rasmussen, R. A. Nature 1988, 334, 138.
2
9
3 529.
(16) Seisel, S.; Rossi, M. J. Ber. Bunsen-Ges. Phys. Chem. 1997, 101,
43.
(
(
(
17) Chu, L. T. J. Vac. Sci. Technol., A 1997, 15, 201.
18) Keyser, L. F.; Leu, M.-T. J. Colloid Interface Sci. 1993, 155, 137.
19) Haynes, D. R.; Tro, N. J.; George, S. M. J. Phys. Chem. 1992, 96,
8
502.