Heterogeneous Reaction HOCl + HBr → BrCl + H2O on Ice Films

Article abstract of DOI:10.1021/jp983407n

The heterogeneous reaction HOCl + HBr(s) → BrCl + H₂O(s) on the ice surface at 189 and 220 K has been investigated in a flow reactor interfaced with a differentially pumped quadrupole mass spectrometer. Pseudo first-order decay of HOCl over the HBr-treated ice surfaces has been determined under the condition of P_HOCI < P_HBr. For the HBr partial pressure in the range of 1.1 × 10^-7 to 6.6 × 10^-5 Torr, the reaction probability (y_g) was determined in the range of 0.06 to 0.38 at 189 K. The reaction probability is in the range of 0.01 to 0.07 at 220 K for HBr partial pressure from 7.2 × 10^-7 to 1.3 × 10^-5 Torr. The reaction probability was found to be strongly dependent on the ice surface temperature. The reaction probability is higher at the lower temperature than that at the warmer temperature and a mechanistic explanation is provided. The true reaction probabilities (y_t) of the reaction were calculated using a pore diffusion model. The kinetic analysis indicates that the heterogeneous reaction of HOCl + HBr may follow the Eley-Rideal type of mechanism. Also, the heat of uptake of HOCl on ice was determined to be about -8.5 ± 2 kcal/mol.

Full text of DOI:10.1021/jp983407n

J. Phys. Chem. A 1999, 103, 691-699
691
Heterogeneous Reaction HOCl + HBr f BrCl + H2O on Ice Films
Liang Chu 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: August 18, 1998; In Final Form: NoVember 23, 1998
2
The heterogeneous reaction HOCl + HBr(s) f BrCl + H O(s) on the ice surface at 189 and 220 K has been
investigated in a flow reactor interfaced with a differentially pumped quadrupole mass spectrometer. Pseudo
first-order decay of HOCl over the HBr-treated ice surfaces has been determined under the condition of
-
7
-5
P
HOCl < PHBr. For the HBr partial pressure in the range of 1.1 × 10 to 6.6 × 10 Torr, the reaction
probability (γ ) was determined in the range of 0.06 to 0.38 at 189 K. The reaction probability is in the range
g
-7
-5
of 0.01 to 0.07 at 220 K for HBr partial pressure from 7.2 × 10 to 1.3 × 10 Torr. The reaction probability
was found to be strongly dependent on the ice surface temperature. The reaction probability is higher at the
lower temperature than that at the warmer temperature and a mechanistic explanation is provided. The “true”
t
reaction probabilities (γ ) of the reaction were calculated using a pore diffusion model. The kinetic analysis
indicates that the heterogeneous reaction of HOCl + HBr may follow the Eley-Rideal type of mechanism.
Also, the heat of uptake of HOCl on ice was determined to be about -8.5 ( 2 kcal/mol.
I. Introduction
It is worth mentioning that atmospheric bromine species are
responsible for about 25% of polar ozone depletion and that
Heterogeneous reactions on the surface of polar stratospheric
clouds (PSCs) are critical to an understanding of the annual
appearance of the Antarctic ozone hole. The key chemical
processes that occur on polar stratospheric cloud particles
responsible for ozone depletion are heterogeneous chlorine
ozone destruction efficiency by bromine on a per atom basis is
about 50 times more than that of chlorine in the lower
2
stratosphere.
Characterizing the interaction of HOCl, HBr, and HCl with
ice is the first step toward revealing the reaction mechanism of
1
,2
reactions 1-2:
7
-14
reactions 2 and 3. While the uptake of HCl
ice
and HOCl on
were reported in the literature by several groups, the
uptake of HBr on ice films at the polar stratospheric temperature
1
3,15-17
ClONO (g) + HCl(s) f Cl (g) + HNO (s)
(1)
(2)
2
2
3
3
,18,19
HOCl(g) + HCl(s) f Cl (g) + H O(s)
was limited to a few studies.
It appears that the uptake of
2
2
HBr on ice surfaces is higher than that of HCl in general. The
higher uptake amount was explained by the formation of solid
hydrobromic acid hydrates during the uptake process.18 Another
possible explanation for the higher uptake is that a phase other
than HBr-in-ice solid solution is likely to have occurred in the
The atmospheric chemistry of bromine species is character-
ized by their short lifetimes and their ready availability for gas-
phase catalytic cycles. Even though bromine species have a low
concentration in the lower stratosphere, heterogeneous bromine
reactions are still important to rapidly convert bromine reservoir
compounds BrONO2 and HBr into photochemically reactive
species. This process is especially important when it is coupled
with the reservoir chlorine species. The following heterogeneous
reactions 3 and 4 that involve both bromine and chlorine are of
interest to study:3
1
9
experiments.
While HBr tends to form a hydrate, which is an ionic crystal,
-7
18,20
near the ice surface at ∼10 Torr and 188 K,
HCl is
21
dissociatively adsorbed on the ice surface. This difference may
alter the reaction HOCl + HBr f BrCl + H2O. It may behave
differently from the reaction HOCl + HCl f Cl2 + H2O. The
heterogeneous reaction HOCl + HCl f Cl2 + H2O on ice
surfaces has been investigated by a few groups,
HOBr + HCl f BrCl + H2O on ice has been studied by Abbatt
and Allanic et al. Abbatt also studied the heterogeneous
reactions HOBr + HBr f Br2 + H2O on ice. To the best of
our knowledge, there is no reported study for the reaction HOCl
,4
HOCl(g) + HBr(s) f BrCl + H O(s)
(3)
(4)
13,15,17,22
2
and
3
HOBr(g) + HCl(s) f BrCl + H O(s)
23
2
3
The significance of these reactions is reflected in several
aspects. The heterogeneous reactions can proceed in the dark
winter of the polar region and convert photochemically inactive
chlorine and bromine reservoir compounds into photochemically
reactive species such as Cl2 or BrCl. When spring comes and
light becomes available, Cl2 or BrCl is photolyzed to produce
Cl or Br radicals which then subsequently deplete polar ozone
through catalytic cycles.5
+
HBr f BrCl + H2O on ice at stratospheric temperature. With
the importance of understanding polar ozone depletion and
revealing the nature of chlorine and bromine species interacting
with the ice surface, we were motivated to study the heteroge-
neous reaction HOCl + HBr f BrCl + H2O and compare this
reaction to the HOCl + HCl f Cl2 + H2O reaction.
,6
In this paper, we report the first measurement of reaction
probability for the reaction HOCl + HBr f BrCl + H2O on
ice surfaces at 189 and 220 K. In the following sections, we
*
Author to whom correspondence should be addressed. E-mail: lchu@
cnsvax.albany.edu.
1
0.1021/jp983407n CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/22/1999

692 J. Phys. Chem. A, Vol. 103, No. 6, 1999
Chu and Chu
Figure 1. Schematic diagram of the experimental apparatus. The sliding injector with double capillary inlets was used to deposit water-ice on the
wall of the flow reactor and to admit HOCl, HCl, and HBr into the reactor. A QMS was employed to monitor HOCl, HCl, HBr, and BrCl concentrations
before, during, and after the reaction.
will briefly describe the experimental procedures used in reaction
probability measurements and present the experimental results.
These include the reaction probability as a function of partial
HBr pressures and ice film surface temperatures. We will briefly
discuss the reaction mechanism and the effect of temperature
on the reaction probability. Finally, we compare the results of
reaction 3 and reaction 2.
A double capillary injector was used to admit both HOCl and
HBr into the flow reactor during the reaction probability
measurement. Reactant HOCl vapor was taken from the HOCl
solution. HOCl vapor contained a small amount of water vapor.
To avoid water vapor condensing on the capillary wall at low
temperature, dry air was passed through the outside of the
capillary.
Ice-Film Preparation. The ice film was prepared as fol-
lows: helium carrier gas was bubbled through a constant-
temperature 293.2 ( 0.1 K water reservoir. Helium saturated
with water vapor was admitted to an inlet of the sliding Pyrex
injector. During the course of ice deposition, the sliding injector
was slowly pulled out at a constant speed and a uniform ice
film was deposited on the inner surface of the reactor, which
was at either 189 or 220 K. The amount of ice substrate
deposited was determined from the mass flow rate of the water
vapor and the deposition time. In a typical experiment at 189
K, the mass of the ice substrate deposited on the flow reactor
wall was about 30 mg. An average film thickness was calculated
by using the measured ice-film geometric area, the mass of ice,
II. Experimental Section
The reaction probability measurements were performed in a
flow reactor coupled to a differentially pumped quadrupole mass
spectrometer (QMS). The experimental apparatus is shown in
Figure 1. A part of the apparatus has been discussed in our
previous publications,18,24 and we provide only a brief descrip-
tion and some modifications in this paper.
Flow Reactor. The cylindrical flow reactor was constructed
of Pyrex glass. Its dimensions were 1.70 cm inner diameter and
3
5 cm length. The temperature of the reactor was regulated by
a liquid-nitrogen-cooled methanol circulator 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 either 189 or 220 K and the stability of the
temperature was better than 0.3 K in every experiment. The
pressure inside 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.
3
25
and the bulk density, 0.63 g/cm , of vapor-deposited water ice.
The typical average film thickness is 2.4 ( 0.2 µm at 189 K
and 42 ( 4 µm at 220 K. Note the ice film deposited at 220 K
was considerably thicker than the ice film deposited at 189 K.
At 220 K, the ice film sublimation rate was higher than that at
2
6
189 K. The loss of the ice film due to the evacuation in the

HOCl + HBr f BrCl + H2O on Ice Films
J. Phys. Chem. A, Vol. 103, No. 6, 1999 693
flow reactor would also be larger. Along with the higher total
pressure, we had to prepare a thicker film so that the film loss
was minimum when the ice was treated by HBr. During the
reaction, additional water vapor from the HOCl solution could
compensate for the loss of ice.
HBr-He and HCl-He Mixtures. The HBr-He mixture
was prepared by mixing HBr (Matheson, 99.8%) and helium
(
MG, Scientific grade 99.9995%) in an all-glass manifold, which
-
6
had been previously evacuated to 2 × 10 Torr. The typical
-
3
-5
HBr-to-He mixing ratio is 10 to 10 . HBr along with
additional helium carrier gas was introduced into the flow reactor
via the glass and Teflon PFA tubing, and the amount was
controlled by a stainless steel flow controller (Teledyne-
Hastings). The HCl-He mixture was prepared by a similar
method. The typical mixing ratio was nearly identical to the
HBr-to-He ratio.
Figure 2. Plot of the HOCl signal versus the reaction time at 189 K.
3
-1
HOCl Preparation and Calibration. The HOCl solution was
prepared by mixing an NaOCl solution with a MgSO4 solution.
The pseudo first-order rate constant k
s
) 1.25 × 10
s
and the
3
-1
27
corrected rate constant k
g
) 2.08 × 10 s . The reaction probability
γ
g
) 0.12. The total pressure ) 0.281 Torr and the flow velocity )
Forty grams of MgSO4‚7H2O was dissolved in 75 mL of
distilled water. The MgSO4 solution was then added to 75 mL
of NaOCl solution (6% active chlorine, Aldrich), drop by drop
in the dark. A white precipitate of Mg(OH)2 was formed during
the reaction. After completing the synthesis, the Mg(OH)2
precipitate was separated from the solution by decantation. A
2
9.1 m/s. A signal for product BrCl is also shown in the figure. In a
separate experiment, the formation of Br
time delay (see text for details).
2
was observed after a short
for a period of 10-15 min. The ice surface was not completely
saturated by HBr yet, but the concentration of HBr was higher
than that of HOCl. If we completely saturated the film, the HBr
hydrates would desorb from the surface and the entire film
would be evaporated. Finally, HOCl was admitted to the
reactor with continuing HBr flow in a separated capillary. The
gas-phase loss of HOCl was measured by the QMS at m/e )
52 as a function of the injector distance. For a pseudo first-
order reaction under the plug-flow condition, the following
equation holds for the reactant HOCl:
-
slightly yellowish, clear HOCl(OCl ) solution was obtained.
-
The concentration ratio of HOCl to OCl can be changed by
1
8
the acidity of the solution. Thus, it was important to control
the pH of the HOCl solution to achieve a higher HOCl yield.
In this study, the HOCl solution was monitored by a UV-visible
spectrometer (Beckman DU640). The relative absorption in-
-
tensity at 230 nm (HOCl) and 290 nm (OCl ) is an indication
of the HOCl concentration in the solution.28 The absorption
intensity at 230 nm was determined as a function of the pH in
the solution. We found that the most suitable pH value to
produce a higher concentration of HOCl was ∼6.5. However,
at a slightly higher pH, ∼7.5, gas-phase Cl2 can be reduced
significantly with the trade-off a lower HOCl yield.
ln[HOCl] ) -k (z/V) + ln[HOCl]
0
(5)
t
s
where z is the injector position, V is the flow velocity, [HOCl]t
is the gas-phase HOCl concentration measured by the QMS at
position z, and sub-0 is the initial reference injector position.
The injector was typically pulled out 1 cm at a time. The first-
order HOCl decay for a typical experiment performed on the
ice film at 189 K is shown in Figure 2. The pseudo first-order
reaction rate constant, ks, was calculated from the least-squares
fit of eq 5 to the experimental data. ks was corrected for the
gas-phase axial and radial diffusion using a standard procedure,29
and the corrected rate constant is called kg. A diffusion
Helium gas was bubbled through the HOCl solution that was
maintained at 273.15 K. Both HOCl vapor and a small amount
of water vapor from the HOCl solution were admitted into the
reactor. The water vapor was used to prevent HOCl decomposi-
tion by the reaction of 2HOCl f Cl2O + H2O during the
transportation of HOCl into the flow reactor. The water vapor
pressure was always controlled to be about equal to the ice vapor
pressure at the ice-film temperature.
In the calibration experiment, the concentration of gas-phase
HOCl was determined by measuring the product Cl2 in reaction
with HCl in excess at 188 K. We assumed that (1) a 1:1
stoichiometric ratio for the formation of Cl2 from HOCl was
valid in the flow reactor, and (2) all Cl2 formed was released to
the gas-phase and quantitatively measured by the QMS. The
concentration of Cl2 was calibrated by using a pure Cl2 gas
2
-1
coefficient for HOCl in He of 240 Torr‚cm s at 202 K was
1
5
used in the calculation. The reaction probability γg was
2
2
calculated from kg using the following equation:
γ ) 2Rk /(ω + Rk )
2
(6)
g
g
g
where R is the radius of the flow reactor, and ω is the mean
HOCl molecular velocity at the ice-film temperature. In addition,
pore diffusion corrections can be made for the reaction prob-
ability γg and to provide the “true” reaction probability γt.3
(
Praxair, 99.5%) diluted in helium. The relative ionization cross
sections for HOCl-to-Cl2 and HOCl-to-HCl were then deter-
mined at different HCl and Cl2 pressures under the various QMS
operating conditions. We also determined the QMS output signal
as a function of the HOCl partial pressure and QMS operational
conditions. The uncertainty of HOCl concentration was esti-
mated to be in the range of 15-20%.
0,31
III. Results
Uptake of HOCl on Water-Ice. Figure 3 is a plot of the
HOCl signal versus the experimental time for the exposure of
HOCl on an ice surface. In this experiment, a 28 cm length of
the ice film was deposited on the wall of the flow reactor. The
gas-phase HOCl signal, as monitored by the QMS at m/e ) 52,
rapidly decreased when HOCl reached the ice surface. Within
2 min, the entire ice film was saturated. The uptake amount
Determination of the Reaction Probability. The reaction
probability of HOCl 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 film surface was treated
-7
-5
with HBr at pressures between 1.1 × 10 and 6.6 × 10 Torr

694 J. Phys. Chem. A, Vol. 103, No. 6, 1999
Chu and Chu
-
7
Figure 3. Uptake of HOCl on water-ice at PHOCl ) 1.0 × 10 Torr
and 189 K. (b) represents the HOCl signal. The uptake starts at t ) 0
Figure 5. Plot of the reaction probability γg versus partial HBr
pressures for the reaction of HOCl + HBr on the ice surface at 189
and at 220 K, respectively. Ice film thickness was 2.4 ( 0.2 µm at 189
K and 42 ( 4 µm at 220 K. The error bars represent one standard
2
min when the injector is pulled out. The entire ice film (150 cm ) is
saturated about 2 min. Desorption occurs at 10 min when the injector
is pushed in. This procedure can be repeated several times.
deviation ((σ) of a mean γ value. The solid curves were fitted to the
g
Eley-Rideal mechanism, and the dashed line was fitted to the
Langmuir-Hinshelwood mechanism.
-
5
6
.6 × 10 Torr. Two slightly different partial HOCl pressures
-7
were used in this study. PHOCl is about (5.4 ( 1.4) × 10 Torr
-
6
when PHBr g 1.1 × 10
Torr, and PHOCl is (1.5 ( 0.7) ×
-
7
-6
1
0
Torr when PHBr < 1.1 × 10 Torr. The concentration of
HBr was always greater than that of HOCl, thus the pseudo
first-order condition was satisfied. The reaction probability γg
as a function of PHBr at 189 and 220 K is presented in Figure
5
. The detailed experimental conditions are listed in Table 1.
All data listed in the table were averaged over 3 to 10
measurements. The errors in both Table 1 and Figure 5 represent
one standard deviation ((σ) of the average value. Figure 5
indicates that the reaction probability γg increased from 0.06
to 0.38 as PHBr increased and became less pressure dependent
-
5
at PHBr > 2 × 10 Torr at 189 K. At a warmer temperature,
2
20 K, the reaction probability γg showed a similar trend, that
is, it increased from 0.01 to 0.07 as PHBr increased. However,
the reaction probability at 220 K was lower than that at 189 K.
The ice-film roughness can also be taken into the consideration
by using a layer model to calculate the “true” reaction
Figure 4. Plot of the logarithm of HOCl 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, and PHOCl ) 6.0 × 10 Torr. The “heat of uptake” of
3
0,31
HOCl on ice was -8.5 ( 2 kcal/mol.
probability.
The corrected reaction probability γt using a
3
tortuosity factor τ ) 4 and true ice density Ft ) 0.925 g/cm is
was determined to be approximately 1 × 10 molecules/cm2
at 189 K. When the HOCl-saturated ice film was heated by the
injector, HOCl was desorbed and quantitatively recovered as
recorded by the QMS. The uptake and desorption procedure
could be repeated several times. The uptake amount Θ of HOCl
was almost identical to the desorbed HOCl amount. The uptake
and desorption process of HOCl on the ice film seems reversible.
The uptake of HOCl on the ice film was also measured at
different surface temperatures, and the result is shown in Figure
14
also listed in Table 1. γ was approximately a factor of 2-10
t
smaller than γ . At a higher γ , the reaction is completed before
g
g
HOCl reaches an internal ice surface and the correction for γt
is smaller. At a lower γg, the HOCl molecule has a chance to
enter an internal surface and so the γt correction is larger.
A reaction product, BrCl, was measured by the QMS at its
parent peak m/e ) 114. The increase of the BrCl signal was so
small that it was hardly detected by the QMS at 189 K (see
Figure 2), and in many circumstances the BrCl signal appeared
unchanged as the reaction proceeded. This seems to suggest
that BrCl was adsorbed on the surface if it was formed during
the reaction. We will discuss this in a later section.
4
. The “heat of uptake” of HOCl on ice was calculated from
the slope of a plot of log Θ versus 1/T. The heat of uptake of
HOCl on ice was determined to be -8.5 ( 2 kcal/mol. This is
in good agreement with the value of -8.8 kcal/mol and -10.5
kcal/mol by quantum chemistry calculations1 as well as the
experimental values of -10.5 ( 2 kcal/mol and -14 ( 2 kcal/
mol.1
We also measured the reaction probability as a function of
6,32
-6
PHOCl while keeping PHBr as a constant, 3.20 × 10 Torr, at
189 K. The result is shown in Figure 6 and Table 2. The
experimental data indicate that the reaction probability was
nearly independent of the partial HOCl pressure in the range
3,15
HOCl + HBr f BrCl + H2O. The reaction probability for
the HOCl + HBr f BrCl + H2O reaction was determined by
observing the decay of gas-phase HOCl, monitored by the QMS,
as a function of the injection position. The reaction probabilities
were measured as a function of PHBr in the range 1.1 × 10- to
-
8
-7
3.5 × 10 to 8.5 × 10 Torr.
HOCl + HCl f Cl + H O. The reaction probability for
2
2
the reaction HOCl + HCl f Cl2 + H2O was measured in a
similar fashion. The thickness of the ice film was 2.5 ( 0.2
7

HOCl + HBr f BrCl + H2O on Ice Films
J. Phys. Chem. A, Vol. 103, No. 6, 1999 695
O on Ice Filmsa,b
(1/s)
TABLE 1: Reaction Probability for the Reaction HOCl + HBr f BrCl + H
2
c
temperature (K)
P
HBr (Torr)
V (m/s)
k
s
(1/s)
k
g
γ
g
γ
t
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
7
3
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
3
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
3
2
2
3
3
3
3
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
88.4 ( 0.1
88.9 ( 0.6
88.5 ( 0.1
88.8 ( 0.1
88.6 ( 0.2
89.0 ( 0.1
89.6 ( 0.1
89.1 ( 0.6
89.0 ( 0.4
88.8 ( 0.1
88.4 ( 0.3
88.8 ( 0.2
88.6 ( 0.1
88.2 ( 0.2
88.4 ( 0.2
88.7 ( 0.1
20.8 ( 0.7
19.8 ( 0.5
20.3 ( 0.3
20.3 ( 0.2
19.3 ( 1.2
20.3 ( 0.8
20.6 ( 0.5
20.0 ( 0.5
21.4 ( 0.1
20.5 ( 0.1
21.3 ( 0.1
19.3 ( 0.1
1.11 × 10
1.98 × 10
3.33 × 10
7.22 × 10
1.10 × 10
1.98 × 10
2.33 × 10
3.22 × 10
3.27 × 10
5.14 × 10
6.72 × 10
7.15 × 10
1.87 × 10
2.65 × 10
3.43 × 10
6.62 × 10
7.21 × 10
7.93 × 10
9.42 × 10
1.11 × 10
1.16 × 10
1.32 × 10
1.58 × 10
2.01 × 10
3.64 × 10
5.96 × 10
9.10 × 10
1.27 × 10
28.7
29.1
28.7
28.8
28.9
29.0
29.4
29.4
29.1
29.1
28.8
29.0
29.0
29.1
26.1
29.4
17.9
18.3
18.2
18.4
17.8
18.8
18.1
18.2
18.5
18.6
19.3
18.8
1.01 × 10
6.90 × 10
1.14 × 10
1.41 × 10
1.18 × 10
1.63 × 10
1.34 × 10
1.71 × 10
1.76 × 10
1.35 × 10
1.52 × 10
1.31 × 10
1.87 × 10
2.08 × 10
2.01 × 10
1.88 × 10
2.43 × 10
1.68 × 10
1.44 × 10
1.54 × 10
7.17 × 10
6.22 × 10
4.54 × 10
6.93 × 10
7.74 × 10
7.26 × 10
7.81 × 10
7.10 × 10
1.51 × 10
8.90 × 10
1.82 × 10
2.61 × 10
1.92 × 10
3.48 × 10
2.37 × 10
3.84 × 10
4.13 × 10
2.40 × 10
3.01 × 10
2.28 × 10
4.81 × 10
6.53 × 10
7.09 × 10
4.84 × 10
2.76 × 10
1.83 × 10
1.55 × 10
1.67 × 10
1.11 × 10
8.87 × 10
5.85 × 10
1.05 × 10
1.24 × 10
1.12 × 10
1.24 × 10
1.08 × 10
0.09
0.06
0.11
0.16
0.12
0.19
0.14
0.24
0.23
0.16
0.17
0.14
0.26
0.32
0.38
0.27
0.016
0.011
0.009
0.010
0.063
0.051
0.034
0.060
0.070
0.064
0.070
0.062
0.024
0.014
0.031
0.051
0.035
0.063
0.043
0.084
0.080
0.051
0.055
0.043
0.10
7
7
7
6
6
6
6
6
6
6
6
5
5
5
5
7
7
7
6
6
6
6
6
6
6
6
5
0.12
0.15
0.10
0.002
0.001
0.001
0.001
0.015
0.011
0.006
0.014
0.017
0.015
0.017
0.014
a
All data points are averages of 3 to 10 measurements. b Mean total pressure was 0.279 ( 0.008 Torr at 188.7 K and 0.506 ( 0.013 Torr at
x
2γg
2
20.3 K. Ice film thickness was 2.4 ( 0.2 µm at 188.7 K and 42 ( 4 µm at 220.3 K. c γt
)
where η is the
1/2
π{1 + η[2(N - 1) + (3/2) ]}
L
L L
effectiveness factor as defined in ref 30 and N is the number of granule layers. N is 5 and 17 for the thin and thick film, respectively.
depend on the temperature.33 Figure 5 shows that the reaction
probability strongly depends on the ice-film surface temperature.
This may be attributed to the following two factors. (1) A higher
HOCl uptake at 189 K (see Figure 4) indicates that a stronger
interaction between HOCl and the ice surface at the lower
temperature. The bond strength within a HOCl molecule could
be weaker under this circumstance. At a warmer temperature,
a weaker interaction between HOCl and ice is less favorable
for reaching the transition state and a higher reaction barrier is
expected. (2) the desorption of HBr on ice occurs at about 220
K in the fast flow reactor with the total pressure of 0.50 Torr.
This is illustrated in Figure 8a. As expected, the peak desorption
temperature Tp increases as the heating rate increases. The
activation energy for desorption Ea is calculated on the basis of
both the peak desorption temperature and the heating rate â
3
3
by
Figure 6. Plot of the reaction probability versus the partial HOCl
pressure at 188.5 K. The total pressure in the reactor was 0.278 Torr,
-
6
Ea
Ea
n-1
and PHBr ) 3.20 × 10 Torr. The reaction probability γ
g
was nearly
â
ln
) -
- ln
(7)
independent of PHOCl
.
2
p
(
)
k T
(
)
k T
B
p
k θ
B
0
0
µm, and the temperature was 188.4 ( 0.2 K. The reaction
probability measured as a function of PHCl at 188 K is presented
in Figure 7. Also, both γg and γt are tabulated in Table 3 along
with detailed experimental conditions. For comparison purposes,
we choose to use γg in all plots, since γg is widely used in the
atmospheric chemistry literature. Figure 7 indicates that the
reaction probability increased slightly at lower HCl pressure,
where θ0 is the coverage at the start of the thermal desorption
curve. k0 is the preexponential factor. kB is the Boltzmann
constant. Ea was determined to be 11.1 ( 3.5 kcal/mol (see
Figure 8b). The peak desorption temperature of 220 K implies
that the residence time of HBr on the surface decreases
dramatically at the desorption temperature and HBr is not readily
available for the reaction. Thus, a reduced reaction rate is
expected near or above the desorption temperature and the
above. It is useful to point that double desorption peaks,
separated by ∼5 K, were observed in some higher surface
coverage experiments. This indicates the formation of a “new”
adsorbed species and is consistent with the formation of a
hydrate. Likely, HBr desorption is a main reason to explain the
-6
then became less pressure dependent at PHCl > 1.5 × 10 Torr.
IV. Discussion
Effect of Temperature on the Reaction HOCl + HBr f
BrCl + H2O. Usually surface reactions are the nonactivated
type and the rate constant of a surface reaction does not strongly

696 J. Phys. Chem. A, Vol. 103, No. 6, 1999
Chu and Chu
TABLE 2: Reaction Probability for the Reaction HOCl + HBr f BrCl + H
2
O on Ice Filmsa,b
(1/s)
temperature (K)
P
HOCl (Torr)
thickness (µm)
V (m/s)
k
s
k
g
(1/s)
γ
g
-
-
-
-
-
-
-
-
-
-
7
7
7
7
7
7
7
8
8
8
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
1
89.2 ( 0.1
89.1 ( 0.1
89.1 ( 0.2
89.4 ( 0.1
88.1 ( 1.1
87.8 ( 0.1
88.0 ( 0.6
88.4 ( 0.1
88.4 ( 0.1
87.7 ( 0.1
(8.5 ( 0.3) × 10
(6.0 ( 0.2) × 10
(4.6 ( 0.2) × 10
(3.5 ( 0.1) × 10
(1.7 ( 0.2) × 10
(1.2 ( 0.1) × 10
(1.0 ( 0.1) × 10
(7.5 ( 0.1) × 10
(6.2 ( 0.4) × 10
(3.5 ( 0.1) × 10
2.4
2.5
2.6
2.4
2.7
2.6
2.5
2.7
2.4
2.5
29.3
29.1
29.1
29.1
28.9
29.2
29.2
29.1
28.6
29.7
1.34 × 10
1.85 × 10
2.05 × 10
1.48 × 10
1.66 × 10
1.49 × 10
1.35 × 10
1.51 × 10
1.25 × 10
1.33 × 10
2.78 × 10
5.43 × 10
5.75 × 10
2.76 × 10
4.02 × 10
2.88 × 10
2.61 × 10
2.97 × 10
2.35 × 10
2.28 × 10
0.15
0.28
0.30
0.16
0.22
0.16
0.15
0.17
0.13
0.13
a
Data are mean values. b Mean total pressure was 0.278 ( 0.004 Torr. Mean temperature was 188.5 ( 0.6 K. Ice film thickness was 2.5 ( 0.2
-
6
µm. PHBr )(3.20 ( 0.30) × 10 Torr.
(a)
Figure 7. Plot of the reaction probability versus the partial HCl pressure
for the reaction of HOCl with HCl on ice at 188 K. The total pressure
-
7
in the reactor was 0.277 Torr and PHOCl ) 7.2 × 10 Torr. The error
bars represent one standard deviation of the mean value. The solid curve
was fitted to the Eley-Rideal mechanism. Partial HCl pressure was
always larger than partial HOCl pressure.
(b)
TABLE 3: Reaction Probability for the Reaction HOCl +
HCl f Cl
2
+ H
2
O on Ice Films at 188 Ka
temperature
V
(K)
PHCl (Torr) (m/s)
ks (1/s)
kg (1/s)
γg
γt
-
-
-
-
-
-
-
-
-
7
7
6
6
6
6
6
6
6
2
2
1
1
1
1
1
1
1
1
1
88.5 ( 0.1 6.04 × 10
88.4 ( 0.1 9.39 × 10
88.5 ( 0.1 1.45 × 10
88.0 ( 0.2 1.83 × 10
88.6 ( 0.1 2.31 × 10
88.3 ( 0.1 3.20 × 10
88.5 ( 0.1 4.90 × 10
88.2 ( 0.1 6.59 × 10
88.6 ( 0.1 8.26 × 10
28.4 5.24 × 10 6.35 × 10 0.04 0.01
2 3
28.9 7.58 × 10 1.01 × 10 0.06 0.014
3
3
28.5 1.35 × 10 2.44 × 10 0.14 0.043
3
3
28.8 1.44 × 10 2.74 × 10 0.16 0.051
3
3
28.5 1.33 × 10 2.38 × 10 0.14 0.043
3
3
28.2 1.21 × 10 2.02 × 10 0.12 0.035
3
3
28.4 1.39 × 10 2.58 × 10 0.15 0.047
3
3
28.4 1.19 × 10 1.97 × 10 0.12 0.035
3
3
28.6 1.33 × 10 2.37 × 10 0.14 0.043
a
Total pressure was 0.277 ( 0.006 Torr. Ice film thickness was 2.5
0.2 µm. PHOCl ) (7.2 ( 1.2) × 10 Torr.
Figure 8. (a) The HBr desorption temperature profile on the ice surface.
The solid line indicates a desorption peak temperature T
of ∼221 K
with a heating rate â ) 1.7 K/min. The dashed line shows T
∼ 224K
with a heating rate â ) 2.6 K/min. The ice thickness was about 3 µm,
-
7
(
p
p
temperature dependence of γg. These results show that the
heterogeneous reaction of HOCl + HBr on ice is largely
influenced by temperature in the 189 to 220 K range. This
finding may play an important role in understanding the
heterogeneous chemistry that occurs near the PSC surface.
Reaction Mechanism. The reaction mechanism is a way to
express the nature of the chemical reaction. Two types of surface
reactions are commonly found involving two reactants. They
are (1) the Langmuir-Hinshelwood mechanism in which both
molecules are adsorbed on the surface and then react with each
other and (2) the Eley-Rideal mechanism in which one
molecule is adsorbed and the other gas-phase molecule reacts
with the adsorbed molecule. The chemical reactions occurring
near the ice surface at about 190-200 K may not exactly follow
these ideal surface reaction types. Either HBr or HOCl molecules
2
p p
and the total pressure was 0.5 Torr. (b) Plot of ln(â/T ) versus 1/T to
obtain the activation energy for desorption. The error bars represent
one standard deviation ((σ) of the measurements. One horizontal error
bar represents a typical error of the temperature about (1 K. All data
in the plot have the similar horizontal error bar, which is not shown in
the figure for a better viewing purpose.
3
3
may diffuse into the ice bulk to a small extent because the
-
12
-15
2
diffusion coefficient is on the order of 10 -10
cm /s.
However, it may be convenient to describe the reaction in terms
of the ideal surface reaction empirically. This kind of mechanism
can provide a simple picture as how the HOCl and HBr
molecules react with each other to form products.
Both the Eley-Rideal and Langmuir-Hinshelwood mech-
anisms were being used to test the experimental data for the

HOCl + HBr f BrCl + H2O on Ice Films
J. Phys. Chem. A, Vol. 103, No. 6, 1999 697
reaction HOCl + HBr f BrCl + H2O. In the case of the Eley-
Rideal mechanism, the reactant HBr chemisorbs onto the
surface. Adsorbed HBr can then be “solvated” near the surface
and reacts with an incoming gas-phase HOCl molecule to form
an HOCl-HBr complex. The rearrangement of bonds within
the complex then forms the products. This is illustrated by the
following reactions:
in Figure 5 as the solid line. γo was obtained from the fit to be
0.32 and 0.14 at 189 and 220 K, respectively. By the same token,
one can use the Langmuir-Hinshelwood mechanism, R )
ksθHOClθHBr, to fit the experimental results. The result of that
fit is also shown in Figure 5 as the dashed line; however, the
fit is not as good as that of the Eley-Rideal type. Equation 11
also explains that the measured reaction probability is indepen-
dent of the partial HOCl pressure as indicated in Figure 6. It
appears that the reaction follows the Eley-Rideal mechanism
experimentally. In addition, the desorption temperature of HOCl
k1
+
-
HBr + S y\ z HBr(ad) T H (H O)(ad) + Br (H O)(ad)
2 2
k2
(8a)
17,37
on ice is about 173-175 K
and the HOCl surface residence
time at 189 K is about milliseconds, estimated on the basis of
the heat of HOCl uptake. This fact also supports the idea that
the reaction resembles the Eley-Rideal type. The fact seems
contrary to the HOCl uptake experiment, and this issue will be
discussed in a following paragraph.
ks
98
+
-
H Br (ad) + HOCl f HOCl‚‚‚BrH(ad)
BrCl(ad) + H O(ad) (8b)
2
k5
BrCl(ad) y
\
z BrCl + S
(8c)
The Eley-Rideal mechanism can also be used to explain the
HOCl + HCl reaction. The solid line in Figure 7 is the fit to eq
k6
1/2
1/2
Note we expressed HBr in an ionic form, which can either
be a dissociative or hydrate form on the surface. We will discuss
this intermediate at the end of this section. We expect that
reaction 8b is the rate-determining step. The reasons are (1)
the sticking probability of HBr on the ice surface at 200 K is
11 where bHBrP
is replaced by bHClP
(a dissociative
HBr
HCl
isotherm). The curve represents the experimental data well. The
mechanism shows that adsorbed HCl is ionized near the ice
surface and then reacts with incoming HOCl molecules to form
Cl2.
3
4-36
higher than 0.12.
The uptake of HBr is on the order of
The kinetic analysis indicates that the HOCl + HBr reaction
follows the Eley-Rideal mechanism. This is a steady-state
analysis and does not reveal the reaction time scale. From the
gas-surface dynamics standpoint, the desorption temperature
means that the distribution of desorbed HOCl molecules is
centered at about 173-175 K. A very small fraction of HOCl
molecules (tails of the distribution) has the probability to adsorb
on the surface at 180 Kelvin. When the ice film is porous and
has a large surface area, a measurable uptake may be obtained.
This is the case for the HOCl uptake experiment where the entire
ice film was exposed to HOCl molecules at once (see Figure
3) and the uptake amount is low.
15
2
1
0
molecules/cm and HBr is in excess; (2) HOCl is the
limiting agent in this study; and (3) BrCl adsorbed on the surface
did not affect the measured γg’s. Both 8a and 8c are unlikely
to be the rate-limiting step. The rate of the reaction R is
proportional to the HBr surface coverage, θHBr, and PHOCl:
dPHOCl
R ) -
) ksθHBrPHOCl
(9)
dt
Reaction probability, γ, is given by R/φHOCl, where φHOCl )
PHOCl/ 2πmk T, m is the mass of HOCl. γ can be expressed
x
B
by eq 10:
While the HOCl molecule has a finite lifetime (thousands
vibrational cycles) on the surface at 189 K, HOCl may hop many
times on the surface before it either reacts with HBr or is
k k
1
s
P
1/2
HBr
c
S0
38
R
γ )
k2
desorbed from the surface. If this is the case, the HOCl +
)
)
φHOCl
k1
k2
k6
k5
HBr reaction can be described as a trapping-mediated Lang-
muir-Hinshelwood type at a molecular level. Because HOCl
physically remains on the surface before it desorbs from the
surface, the term “trapping-mediated” is used to describe the
process. The trapping-mediated process increases the reaction
probability dramatically as to the “single collision” Eley-Rideal
type. This explains the high reaction probability (γ > 0.1) very
well. The conclusion of this discussion is that the true rate law
in terms of actual surface concentration is probably the trapping-
mediated Langmuir-Hinshelwood type but the observable rate
law is Eley-Rideal.
1/2
HBr
1
+
P
+
PBrCl
1
HBr
/2
bHBr
P
γo
(10)
1/2
1
+ bHBr
P
+ bBrClPBrCl
HBr
where c ) 2πmk T, bHBr ) k1/k2, and bBrCl )k6/k5. In order
x
B
to express eq 10 as a Langmuir form, we collected all constants
together and called γo, which is cksS0, where S0 is the total
number of sites on the surface. The experimental results suggest
that both HBr and BrCl are adsorbed on the surface. bHBr and
bBrCl (adsorption equilibrium constant) are expected to be on a
similar order of magnitude. We have PHBr > PHOCl and so PHBr
We want to comment on an intermediate of the reaction. At
1
89 K or higher, the dynamic nature of the ice surface is
expected to “solvate” HBr molecules near the surface or HBr
might ionize even as it adsorbs atop the surface.39 HBr may
also form hydrates under our experimental conditions and
/2
-5
>
PBrCl. Therefore, bHBrPHBr1 . bBrClPBrCl when PHBr < 10
Torr. The term, bBrClPBrCl, can be ignored from eq 10 and eq
1
0 can be simplified to
2
0,35,40,41
hydrobromic hydrates are ionic crystal.
Either case
-
1
HBr
/2
indicates that HBr is in an ionic form, Br , near the ice surface.
The reaction between HOCl and HBr is then expected to be
ionic in nature. This was reflected in eq 8b. Mechanistically, it
is not possible with our current equipment to establish whether
γ0bHBrP
γ ) γ0θHBr
)
(11)
1
HBr
/2
1
+ bHBr
P
-
where bHBr is constant at a given temperature. γo is the reaction
probability on an ice surface at its saturation capacity for HBr.
The nonlinear least-squares fitting was used to fit the experi-
mental data to eq 11 (Eley-Rideal type), and the result is shown
the reaction occurs directly between HOCl + Br or via the
intermediate H2OCl followed by reaction with Br as an
analogue to the HOCl + HCl reaction on ice. A general
mechanism proposed by Eigen and Kustin is that the reaction
+
-
4
2

698 J. Phys. Chem. A, Vol. 103, No. 6, 1999
Chu and Chu
may occur via the following pathway at a higher pH:43
Reaction 14 is consistent with our previous discussion that
BrCl may adsorb on the ice surface. If reaction 14 is faster than
reaction 8c, BrCl will then be converted into Br2. We examined
-
-
19
HOCl + Br f [HOCl]Br
(12)
(13)
the product Br2 at m/e ) 158. We indeed observed Br2 formation
after a short time delay during the experiment (see Figure 2);
however, the amount of Br2 formed is below the stoichiometric
ratio by a factor of 3-5. Hanson and Ravishankara observed a
similar phenomenon for the ClONO2 + HBr reaction on ice.19
Comparison to Previous Results. For the reaction HOCl +
HCl f Cl2 + H2O, we may compare our study to previous
measurements. Abbatt and Molina reported that the reaction
-
+
[
HOCl]Br + H f BrCl + H O
2
-
In this reaction pathway, hydrated Br initiates a nucleophilic
attack on the Cl end of the HOCl molecule, similar to the
hydrolysis of ClONO2.44 The orientation of the HOCl molecule
is with the H-end down the surface.32 If we assume a hydrate,
say HBr‚3H2O, is formed near the ice surface, both HBr‚3H2O
+
-
41
15
and HBr‚2H2O have a H5O2 Br group, and a six-membered
probability γg is 0.16 ( 0.10 at 202 K and 0.24 at 195 K.
Hanson and Ravishankara measured γg > 0.3 at 200 K.13 Chu
et al. determined the reaction probability is about 0.3 at 188
K. Oppliger et al. reported γg is 0.15 at 200 K. Our result is
ring intermediate
2
2
17
-
6
-7
0
1
.14 at PHCl ) 4.9 × 10 Torr, PHOCl ) 7.2 × 10 Torr, and
89 K (see Figure 7). Within uncertainties of measurement, this
study is in good agreement with the earlier studies.
For the reaction HOCl + HBr f BrCl + H2O, Abbatt and
Nowak reported that the uptake coefficient is from 0.02 to 0.24
is likely formed. Then a proton transfer follows from a
neighboring water molecule to form the products H2O and BrCl.
If the Br-end of BrCl is still bonded to the H2O surface and the
bond is strong, it is possible that BrCl may not desorb to the
gas phase. This may be a reason why we did not observe a
strong BrCl signal increase in the experiment. Donaldson et al.
-
6
-5
for PHBr ) 1.5 × 10 - 7.3 × 10 Torr in 69.3 wt % sulfuric
4
8
acid solution at 228 K. We obtained a reaction probability γg
-
7
)
×
0.06-0.38 on the ice surface for PHBr ) 1.1 × 10 - 6.6
-
5
10 Torr at 189 K. The uptake coefficients are comparable
for both systems, and the reactive uptake efficiency is very high.
Provided the relatively low photolysis rates of HBr (j ) 4.9 ×
applied a similar mechanism to the HOCl + HCl reaction in
-
6
49
-4
50
sulfuric acid.45 Another possible intermediate is a four-centered
10
1/s) and HOCl (j ) 3 × 10 1/s) in the lower
stratosphere, this implies that reaction 3 is efficient to convert
reservoir Br and Cl, into photochemically active BrCl both on
the PSC surface in the dark winter of the polar region and in
sulfuric acid aerosols. However, this reaction rapidly consumes
ring
2
all available HBr (∼2pptv or less) in a day or so. By then, the
reaction is insignificant. The reaction is a source to maintain
4
low concentration of HBr in the polar region. Overall, reaction
in either hydrated HBr or ionized HBr. Richardson et al. showed
that the four-centered intermediate is energetically favorable in
3
changes the partitioning of both Br and Cl species and may
play a role in bromine heterogeneous chemistry in the lower
stratosphere.
-
46
the HOCl + HCl reaction with the assistance of Cl . It
involves a minimum amount of bond breaking and formation.
A similar situation may apply to the HOCl + HBr reaction.
Both pathways can explain the formation of BrCl. Further
experimental and theoretical studies are required to fully resolve
the reaction mechanism.
V. Conclusion
HOCl saturates the water-ice surface rapidly, and the uptake
of HOCl with water-ice is likely a reversible process. The
reaction probability γ for the HOCl + HBr f BrCl + H O
Comparison of the Reaction HOCl + HBr f BrCl + H2O
to HOCl + HCl f Cl2 + H2O. The reaction probability for
the reaction HOCl + HBr f BrCl + H2O (eq 3) at 189 K is in
the range 0.06-0.38. The reaction probability for the reaction
HOCl + HCl f Cl2 + H2O (eq 2) at 188 K is in the range
g
2
reaction ranges from 0.06 to 0.38 at 189 K and from 0.01 to
0.07 at 220 K. The experimental results suggest that the observed
rate law for the heterogeneous reaction of HOCl + HBr is the
Eley-Rideal type. The results obtained from this study can be
extrapolated to stratospheric conditions and can be used in
atmospheric chemistry modeling.
0
.04-0.16. The slightly higher reaction probability γg for
reaction 3 can be explained in terms of the nucleophilicity of
the reagent Br . Since reactions 2 and 3 are in an identical ice
-
environment, the more nucleophilic reagent has the higher
Acknowledgment. The authors thank two anonymous
reviewers for helpful suggestions and comments on the manu-
script. This work was supported by the National Science
Foundation under grant ATM-953065.
-
-
reaction probability. Cl is a hard Lewis base; Br is a softer
Lewis base and thus a better nucleophile to donate its unshared
electron pair to HOCl. This results in a higher nucleophilic
constant and higher reaction rate.47 This picture is also consistent
with the six-membered ring intermediate as discussed in the
previous paragraph.
For reaction 2, the product Cl2 is always detected in our study.
Unlike reaction 2, the gas-phase product BrCl in reaction 3 is
produced in small amounts such that the BrCl signal change is
nearly undetectable by QMS (see Figure 2). A possible reason
is that the product may further proceed through the following
heterogeneous sequence in excess HBr:
References and Notes
(
1) Solomon, S.; Garcia. R. R.; Rowland, F. S.; Wuebbles, D. J. Nature
986, 321, 755.
2) WMO. Scientific Assessment of Ozone Depletion: 1994, Report
1
(
37; WMO: Geneva, 1995; Chapters 3, 10.
(3) Abbatt, J. P. D. Geophys. Res. Lett. 1994, 21, 665.
(4) Lary, D. J.; Chipperfield, M. P.; Toumi, R.; Lenton, T. J. Geophys.
Res. 1996, 101, 1489.
(
(
5) Molina, L. T.; Molina, M. J. J. Phys. Chem. 1987, 91, 433.
6) McElroy, M. B.; Salawitch, R. J.; Wofsy, S. C.; Logan, J. A. Nature
BrCl(s) + HBr(s) f Br + HCl(s)
(14)
2
1986, 321, 759.

HOCl + HBr f BrCl + H2O on Ice Films
J. Phys. Chem. A, Vol. 103, No. 6, 1999 699
(
7) Abbatt, J. P. D.; Beyer, K. D.; Fucaloro, A. F.; McMahon, J. R.;
Wooldridge, J. R.; Zhang, R.; Molina, M. J. J. Geophys. Res. 1992, 97,
5819.
(32) Brown, A. R.; Doren, D. J. J. Phys. Chem. B 1997, 101, 6308.
(33) Masel, R. I. Principles of Adsorption and Reaction on Solid
Surfaces, Wiley: New York, 1996; Chapter 7.
1
(
(
8) Wofsy, S. C.; Molina, M. J. J. Geophys. Res. 1988, 93, 2442.
9) Hanson, D. R.; Mauersberger, K. J. Phys. Chem. 1992, 96, 7079.
(34) Breil, M.; Behr, P.; Zellner, R. PSI-Proc. 1997, 97-02, 35. (γHBr
) 0.12, abstract is available on the CA online.)
(
10) Kroes, G.-J.; Clary, D. C. J. Phys. Chem. 1990, 94, 4700.
11) Koehler, B. G.; McNeill, L. S.; Middlebrook, A. M.; Tolbert, M.
A. J. Geophys. Res. 1993, 98, 10563.
(35) Rieley, H.; Aslin, H. D. J. Chem. Soc., Faraday Trans. 1995, 91,
2349.
(
(36) Hanson, D. R.; Ravishankara, A. R. in The Tropospheric Chemistry
(
(
(
12) Gertner, B. J.; Hynes, J. T. Science 1996, 271, 1563.
13) Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem. 1992, 96, 2682.
14) Chu, L. T.; Leu, M.-T.; Keyser, L. F. J. Phys. Chem. 1993, 97,
of Ozone in the Polar Regions, NATO ASI Series, Niki, H.; Becker, K. H.,
Eds.; Springer-Verlag: New York , 1993;, pp 281-290.
(37) Bahnam, S. F.; Horn, A. B.; Koch, T. G.; Sodeau, J. R. Faraday
7
779.
Discuss. 1995, 100, 321.
38) Weinberg, W. H. in Dynamics of Gas-Surface Interactions Rettner,
C. T., Ashfold, M. N. R., Eds.; RSC: Cambridge, 1991; Chapter 5.
39) Bianco, R.; Gertner, B. J.; Hynes, J. T. Ber. Bunsen-Ges. Phys.
Chem. 1998, 102, 518.
40) Delzeit, L.; Rowland, B.; Devlin, J. P. J. Phys. Chem. 1993, 97,
0312.
(
15) Abbatt, J. P. D.; Molina, M. J. Geophys. Res. Lett. 1992, 19, 461.
16) Geiger, F. M.; Hicks, J. M.; Dios, A. C. J. Phys. Chem. 1998, 102,
(
(
1
1
514.
(
(
17) Oppliger, R.; Allanic, A.; Rossi, M. J. J. Phys. Chem. A 1997, 101,
903.
(
(
(
(
(
18) Chu, L. T.; Heron, J. W. Geophys. Res. Lett. 1995, 22, 3211.
1
19) Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem. 1992, 96, 9441.
20) Chu, L. T.; Chu, L. J. Phys. Chem. A 1999, 103, 384.
21) Horn, A. B.; Chesters, M. A.; McCoustra, M. R. S.; Sodeau, J. R.
(
(
41) Lundgren, J.-O. Acta Cryst. 1970, B26, 1893.
42) Banham, S. F.; Horn, A. B.; Koch, T. G.; Sodeau, J. R. Faraday
Discuss. 1995, 100, 321.
J. Chem. Soc., Faraday Trans. 1992, 88, 1077.
22) Chu, L. T.; Leu, M.-T.; Keyser, L. F. J. Phys. Chem. 1993, 97,
2798.
23) Allanic, A.; Oppliger, R.; Rossi, M. J. J. Geophys. Res. 1997, 102,
3529.
(43) Eigen, M.; Kustin, K. J. Am. Chem. Soc. 1962, 84, 1335.
(44) Bianco, R.; Hynes, J. T. J. Phys. Chem. 1998, 102, 309.
(45) Donaldson, D. J.; Ravishankara, A. R.; Hanson, D. R. J. Phys.
(
1
2
(
Chem. 1997, 101, 4717.
46) Richardson, S. L.; Francisco, J. S.; Mebel, M. A.; Morokuma, K.
(
(
(
(
24) Chu, L. J. Vac. Sci. Technol. A. 1997, 15, 201.
25) Keyser, L. F.; Leu, M.-T. J. Colloid Interface Sci. 1993, 155, 137.
26) Haynes, D. R.; Tro, N. J.; George, S. M. J. Phys. Chem. 1992, 96,
Chem. Phys. Lett. 1997, 270, 395.
(47) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 2nd ed.; Harper & Row: New York, 1981; pp 331-339.
(48) Abbatt, J. P. D.; Nowak, J. B. J. Phys. Chem. A 1997, 101, 2131.
(49) Yung, Y. L.; Pinto, J. P.; Watson, R. T.; Sander, S. P. J. Atm. Sci.
1980, 37, 339.
(50) 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, Evaluation 12; JPL: Pasadena, CA, 1997; pp 265.
8
502.
(
(
(
(
27) D’Ans, J.; Freund, H. E. Z. Elektrochem. 1957, 61, 10.
28) Molina, L. T.; Molina, M. J. J. Phys. Chem. 1978, 82, 2410.
29) Brown, R. L. J. Res. Natl. Bur. Stand. (U.S.) 1978, 81, 1.
30) Keyser, L. F.; Moore, S. B.; Leu, M.-T. J. Phys. Chem. 1991, 95,
5
2
496.
(
31) Keyser, L. F.; Leu, M.-T.; Moore, S. B. J. Phys. Chem. 1993, 97,
800.