Atmospheric Chemistry of CF2BrH: Kinetics and Mechanism of Reaction with F and Cl Atoms and Fate of CF2BrO Radicals

Article abstract of DOI:10.1021/jp9530111

A pulse radiolysis technique was used to investigate the kinetics and products of the reaction of CF2BrH with fluorine atoms at 296 K.This reaction forms an adduct which is in dynamic equilibrium with CF2BrH and fluorine atoms.The UV absorption spectrum of the adduct was measured relative to the UV spectrum of the CH3O2 radical over the range 230-380 nm.At 280 nm, an absorption cross section of (1.3 +/- 0.3)E-17 cm² molecule^-1 was determined.From the absorption at 280 nm the equilibrium constant K₅ = /() was measured to be (1.59 +/- 0.13)E-17 cm³ molecule^-1.In 1 atm of SF6, the forward rate constant k₅ = (1.4 +/- 0.5)E-11 cm³ molecule^-1 s^-1 and the backward rate constant k_-5 = (8.8 +/- 3.0)E5 s^-1 were determined by monitoring the rate of formation and loss of the adduct.As part of the present work a relative rate technique was used to measure k(Cl + CF2BrH) = (5.8 +/- 1.0)E-15 cm³ molecule^-1 s^-1 at 296 K and 700 Torr of N2.The fate of the oxy radical, CF2BrO, in the atmosphere is bromine atom elimination and formation of COF2.

Full text of DOI:10.1021/jp9530111

7
050
J. Phys. Chem. 1996, 100, 7050-7059
Atmospheric Chemistry of CF2BrH: Kinetics and Mechanism of Reaction with F and Cl
Atoms and Fate of CF2BrO Radicals
Merete Bilde, Jens Sehested,* and Trine E. Møgelberg
Section for Chemical ReactiVity, EnVironmental Science and Technology Department, Risø National
Laboratory, DK-4000 Roskilde, Denmark
Timothy J. Wallington*
Ford Motor Company, Ford Research Laboratory, SRL-3083, P.O. Box 2053, Dearborn, Michigan 48121-2053
Ole J. Nielsen*
Ford Motor Company, Ford Forschungszentrum Aachen, Dennewartstrasse 25, D-52068 Aachen, Germany
X
ReceiVed: October 11, 1995; In Final Form: February 1, 1996
A pulse radiolysis technique was used to investigate the kinetics and products of the reaction of CF
fluorine atoms at 296 K. This reaction forms an adduct which is in dynamic equilibrium with CF
fluorine atoms. The UV absorption spectrum of the adduct was measured relative to the UV spectrum of the
2
BrH with
2
BrH and
-
17
CH
3
2
O
2
radical over the range 230-380 nm. At 280 nm, an absorption cross section of (1.3 ( 0.3) × 10
-1
cm molecule was determined. From the absorption at 280 nm the equilibrium constant K ) [adduct]/
5
-
17
3
-1
([F][CF
2
BrH]) was measured to be (1.59 ( 0.13) × 10 cm molecule . In 1 atm of SF
6
, the forward rate
and the backward rate constant k-5 ) (8.8 ( 3.0) ×
0 s were determined by monitoring the rate of formation and loss of the adduct. As part of the present
-
11
3
-1 -1
constant k
5
) (1.4 ( 0.5) × 10 cm molecule
s
5
-1
1
-15
3
-1
work a relative rate technique was used to measure k(Cl+CF
2
BrH) ) (5.8 ( 1.0) × 10 cm molecule
2
-1
s
at 296 K and 700 Torr of N
2
. The fate of the oxy radical, CF BrO, in the atmosphere is bromine atom
2
elimination and formation of COF .
1
. Introduction
Halons are used as fire-extinguishing agents. The active
2. Experimental Section
The two different experimental systems used in the present
10,11
species in halons is bromine. When heated to high temperatures,
halons decompose liberating bromine atoms, which trap the free
radicals that propagate combustion. However, bromine atoms
work have been described in detail previously
be discussed briefly here.
.1. Pulse Radiolysis System. Reactions were initiated by
and will only
2
1
are also very efficient in destroying ozone in the stratosphere.
the irradiation of SF6/CF2BrH and SF6/CF2BrH/CH4 mixtures
in a 1 L stainless steel reaction cell with a 30 ns pulse of 2
MeV electrons from a Febetron 705B field emission accelerator.
SF6 was always in great excess and was used to generate fluorine
atoms:
Bromine atoms are produced by photolysis of halons in the
stratosphere. CF2BrH has been proposed as a substitute for
conventional fully halogenated halons. Following release into
the atmosphere, hydrogen containing halogenated species react
with OH radicals to produce halogenated alkyl radicals which
2
will, in turn, react with O2 to give peroxy radicals. For
example, in the case of CF2BrH:
2 MeV e-
SF6
8 F + products
(3)
(4)
F + CF BrH f products
2
CF BrH + OH f CF Br + H O
(1)
(2)
2
2
2
CF Br + O + M f CF BrO + M
The radiolysis dose was varied by insertion of stainless steel
attenuators between the accelerator and the chemical reactor.
In this article we will refer to the radiolysis dose used in specific
experiments as a fraction of the maximum dose that is
achievable. The fluorine atom yield (required for quantification
of UV absorption spectra) was measured by monitoring the
transient absorption at 260 nm due to CH3O2 radicals produced
by radiolysis of SF6/CH4/O2 mixtures:
2
2
2
2
In the atmosphere the peroxy radical, CF2BrO2, will be
converted into the corresponding oxy radical, CF2BrO. As part
of a joint program between our two laboratories to survey the
atmospheric fate of halogenated compounds,3
-9
we have
conducted an experimental study of the atmospheric chemistry
of CF2BrH. A pulse radiolysis technique was used to determine
absolute rate constants for individual reactions and to investigate
the mechanism of the reaction of F atoms with CF2BrH. The
atmospheric fate of CF2BrO radicals was determined using a
FTIR spectrometer coupled to an atmospheric reactor. Results
are reported herein.
Using σ(CH3O2) ) 3.18 × 10- cm molecule , this
18
2
-1 12
1
5
absorption corresponds to a F atom yield of (3.2 ( 0.3) × 10
molecules cm- at full radiolysis dose and 1000 mbar of SF6.
3
13
The quoted uncertainty includes 10% uncertainty in σ(CH3O2)
and 2 standard deviations in the absorption measurements. The
uncertainties reported in this paper are 2 standard deviations
unless otherwise stated. Standard error propagation methods
are used to calculate combined uncertainties.
*
To whom correspondence should be addressed.
Abstract published in AdVance ACS Abstracts, April 1, 1996.
X
0022-3654/96/20100-7050$12.00/0 © 1996 American Chemical Society

Atmospheric Chemistry of CF2BrH
J. Phys. Chem., Vol. 100, No. 17, 1996 7051
Figure 2. Maximum transient absorbance at 280 nm following the
pulsed radiolysis of mixtures of 20 mbar (circles) and 2.5 mbar (squares)
of CF BrH, respectively, and 980 and 998 mbar of SF , respectively,
2 6
as a function of the radiolysis dose. The solid lines are linear regressions
to the low dose data (filled symbols).
Figure 1. Transient absorbance at 280 nm following the pulsed
radiolysis (11% of maximum dose) of 2.5 mbar of CF
mbar of SF . The UV path length was 120 cm. The smooth line is a
first-order fit to the experimental data.
2
BrH and 997.5
6
Transient absorptions were followed by multipassing the
output of a pulsed 150 W xenon arc lamp through the reaction
cell using internal White cell optics. The physical path length
of the cell is 10 cm. Total optical path lengths of 80 and 120
cm were used. After leaving the cell the light was guided
through a 1 m McPherson grating UV-vis monochromator and
detected with a Hamamatsu photomultiplier. All absorption
transients were produced in single-pulse experiments with no
signal averaging. The spectral resolution was 0.8 nm. Reagent
concentrations used were as follows: SF6, 895-995 mbar;
CF2BrH, 0-50 mbar; CH4, 2-40 mbar. All experiments were
performed at 296 K. SF6 (99.9%) was supplied by Gerling and
Holz, CF2BrH (>97%) was supplied by Fluorochem Ltd., and
CH4 (>99%) was supplied by Gerling and Holz. All reagents
were used as received. The partial pressures of the different
gases were measured with a Baratron absolute membrane
Figure 2 shows plots of the maximum absorbances for the
two different mixtures as functions of the radiolysis dose. The
maximum absorbances were linear functions of the radiolysis
dose up to 50% of the full dose. The deviations from linearity
at higher dose are ascribed to secondary radical-radical
reactions such as self-reaction of the absorbing species or
reactions between F atoms and the absorbing species. The solid
lines are linear least-squares fits to the low-dose data. The
slopes of the linear regressions in Figure 2 are 2.0 ( 0.1 for
mixtures containing 20 mbar of CF2BrH and 0.96 ( 0.04 for
mixtures containing 2.5 mbar of CF2BrH, respectively.
In all experiments the initial concentration of CF2BrH is in
great excess compared to that of fluorine atoms. At low doses
one would expect complete conversion of F atoms to products
according to reaction 4. As a consequence the absorbance
-
5
manometer with a detection limit of 10 bar.
F + CF BrH f products
(4)
2
A Princeton Applied Research OMA-II diode array was used
to measure the UV absorption spectrum of the CF2BrH-F
adduct. The diode array was installed at the exit slit of the
monochromator in place of the photomultiplier which was used
for measuring transient absorptions. The setup consisted of the
diode array, an image amplifier (type 1420-1024HQ), a control-
ler (type 1421) and a conventional personal computer used for
data acquisition, handling, and storage. Spectral calibration was
achieved using a Hg pen ray lamp.
should be independent of the initial concentration of CF2BrH.
It is clear from Figure 2 that this is not the case and the question
which arises is, “Why is the observed absorbance using an initial
concentration of 2.5 mbar CF2BrH approximately half of that
observed using 20 mbar of CF2BrH?” The data in Figure 2
show that F atoms do not react with CF2BrH via a simple
H-atom abstraction mechanism:
2
.2. FTIR Smog Chamber System. The FTIR system was
F + CF BrH f CF Br + HF
(4a)
2
2
interfaced to a 140 L Pyrex reactor. Radicals were generated
by the UV irradiation of mixtures of CF2BrH and Cl2, or F2, in
because, in this case the two dose plots would have the same
slopes. The data in Figure 2 can be explained if the species
which absorbs at 280 nm is an adduct in which a F atom is
weakly bound to CF2BrH and if the adduct is in dynamical
equilibrium with F atoms and CF2BrH:
7
(
00 Torr of N2, or air diluent, at 296 K using 22 blacklamps
760 Torr ) 1013 mbar). The loss of reactants and formation
of products were monitored by FTIR spectroscopy, using an
analyzing path length of 26 m and a resolution of 0.25 cm-
1
.
Infrared spectra were derived from 32 coadded spectra. Refer-
ence spectra were acquired by expanding known volumes of
reference materials into the reactor.
F + CF BrH h X
(5,-5)
2
The lower absorption observed for [CF2BrH] ) 2.5 mbar
compared to the absorption with [CF2BrH] ) 20 mbar can be
explained by a shift in the equilibrium toward the adduct at
higher concentrations of CF2BrH. In the following we shall
show that all the experimental evidence can be interpreted in
terms of an equilibrium between F, CF2BrH, and the adduct X.
In sections 3.2-3.6 we proceed on the assumption that reaction
4a makes a negligible contribution to the overall reaction. The
validity of this assumption is discussed in section 4.2.
3
. Pulse Radiolysis Results and Discussion
.1. Adduct Formation from the Reaction of F with
3
CF2BrH. Figure 1 shows the transient absorption at 280 nm
following the radiolysis (11% of maximum dose) of 2.5 mbar
of CF2BrH and 997.5 mbar SF6 using an optical path length of
1
20 cm. The maximum transient absorbances were measured
as a function of the radiolysis dose using mixtures of 980 mbar
of SF6 and either 2.5 or 20 mbar of CF2BrH.

7
052 J. Phys. Chem., Vol. 100, No. 17, 1996
Bilde et al.
2
Figure 3. Plot of maximum absorbance as a function of [CF BrH].
The UV path length was 120 cm. The relative radiolysis dose was 22%
of maximum dose.
Figure 4. Plot of k1st vs [CF
BrH]. See text for details.
2
to the parent species. This is reasonable because the decay of
the adduct is much slower than its formation. At any time the
concentration of F atoms will be given by [F] ) [F]0 - [X],
where [F]0 ) (7.0 ( 0.7) × 10 molecules cm . The system
is described as a first-order reversible reaction, and the dif-
ferential equations which describe this system can be solved
analytically. See, for example, ref 14, for details on the algebra.
The rate law for the formation of X is
3
.2. Determination of the Equilibrium Constant. Assum-
ing that all species behave as ideal gases the equilibrium constant
of (5) is defined as
1
4
-3
K ) k /k ) [X]/([F][CF BrH])
5
5
-5
2
where X is the CF2BrH-F adduct.
To determine K5, the maximum transient absorption following
pulsed radiolysis of mixtures of SF6 and CF2BrH at 280 nm
was obtained from a double-exponential fit to the transient. The
absorbance was measured at low dose (22% of maximum) where
no secondary chemistry interferes with the measurements. A
plot of the absorbance as a function of the initial concentration
of CF2BrH is shown in Figure 3. The equilibrium constant K5
was derived from a two-parameter fit of the following expression
to the experimental data:
d[X]/dt ) k′ [F] - (k′ + k )[X], k′ ) k [CF BrH]
5
0
5
-5
5
5
2
The solution of this nonhomogeneous first-order linear dif-
ferential equation is
k′ [F]0
5
[X](t) )
(1 - exp(-(k′ + k )t))
5 -5
k′ + k
5
-5
The pseudo first order formation rate will be equal to k5[CF2BrH]
k-5. Hence, in Figure 4 the y-axis intercept gives k-5 and
A ) A K [CF BrH]/(1 + K [CF BrH])
x
5
2
5
2
+
-11
3
the slope gives k5. We obtain k5 ) (1.56 ( 0.31) × 10 cm
where A is the observed absorbance as a function of [CF2BrH]
and Ax is the absorbance when all F atoms are converted into
the adduct. The value obtained for the equilibrium constant
-1 -1
5
-1
molecule
s
and k-5 ) (8.3 ( 2.5) × 10 s . The
-17 3
equilibrium constant is K5 ) k5/k-5 ) (1.8 ( 0.8) × 10
cm
-1
molecule . This is in good agreement with the value of (1.59
-
17
3
-1
was K5 ) (1.59 ( 0.13) × 10
cm molecule . The value
-17
3
-1
(
0.13) × 10
cm molecule obtained in section 3.2.
obtained for Ax was 0.48 ( 0.01
3
.4. Kinetics of the CF2BrH-F Adduct in the Presence
To confirm this result and to determine the rate constants
for the forward and backward reactions in (5), k5 and k-5,
respectively, two additional sets of experiments were carried
out. They are described in the following subsections.
.3. Determination of k5 and k-5. Another set of informa-
tion can be extracted from the experiments described in section
.2. In all experiments the initial concentration of CF2BrH is
of CH4. To further confirm that the adduct X is formed in the
reaction of F atoms with CF2BrH (eq 5), a set of experiments
was performed with added CH4 to provide a loss mechanism
for F atoms (eq 6).
3
F + CH f CH + HF
(6)
4
3
3
at least 35 times the concentration of F atoms and hence the
loss of F atoms and the formation of products will follow
pseudo-first-order kinetics. The observed transient absorbances
In these experiments the concentration of CF2BrH was fixed
at 20 mbar, the concentration of CH4 was varied over the range
2-40 mbar, and SF6 was added so the total pressure remained
fixed at 1000 mbar. The transient absorption at 280 nm was
observed at different concentrations of CH4. The radiolysis dose
was 22% of maximum, and the optical path length was 120
cm. Figure 5 shows the observed variation of the maximum
absorbance as a function of the concentration ratio [CH4]/
[CF2BrH]. As shown in Figure 5, the addition of CH4 results
in a sharp decrease in the observed absorption at 280 nm.
The CH3 radical does not absorb at 280 nm and the reaction
1
st
were fit using the expression A(t) ) (Ainf - A0)[1 - exp(-k t)]
A0, where A(t) is the time-dependent absorbance, Ainf is the
+
1
st
absorbance at infinite time, k is the pseudo-first-order forma-
tion rate of X, and A0 is the extrapolated absorbance at t ) 0.
Figure 4 shows a plot of the pseudo-first-order rate constants
1st
for the formation of X, k , as a function of the concentration
of CF2BrH. The pseudo-first-order rate of formation of X
increased linearly with CF2BrH concentration. Linear least-
squares analysis of the data in Figure 4 gives a slope of (1.56
-11
3
-1
of F atoms with CH4 is fast, k6 ) 6.8 × 10
cm molecule
-11
3
-1 -1
-1 15
(
0.31) × 10
We assume that the formation kinetics for X are not
influenced by any loss mechanism apart from decomposition
cm molecule s .
s . CH4 is an efficient scavenger of F atoms. In contrast,
we do not anticipate that CH4 is an efficient scavenger of other
radicals potentially formed in the system. The species respon-

Atmospheric Chemistry of CF2BrH
J. Phys. Chem., Vol. 100, No. 17, 1996 7053
Figure 5. Plot of maximum absorption at 280 nm vs the concentration
Figure 6. Plot of kdecay vs fCH4. See text for details.
4 2
ratio [CH ]/[CF BrH]. See text for details.
and SF6 was added in an amount such that the total pressure
remained fixed at 1000 mbar. The observed decay of the
transient absorption was fitted using the first-order expression
given in section 3.3 with t ) 0 being the time for maximum
sible for the absorption at 280 nm must therefore be formed
from reaction of F atoms with CF2BrH, i.e., from reaction 5.
The competition between reactions 5 and 6 was used to
determine the rate constant for reaction 5. The solid line in
Figure 5 represents a two-parameter fit of the following
expression to the experimental data:
1
st
absorbance and k the decay rate of X. Figure 6 shows a plot
of kdecay versus fCH4. The solid line is a linear regression of the
5
-1
experimental data which gives k-5 ) (9.4 ( 1.5) × 10 s
5
-1
and an intercept of (1.3 ( 0.9) × 10 s .
-1
k6 [CH4]
The positive intercept indicates that even in the absence of
methane there is a significant loss of the adduct. The intercept
can be explained by radical-radical reactions such as the adduct
self reaction or the existence of a loss mechanism (eq 7) for
the adduct which does not involve the formation of F atoms:
Amax ) Aeq 1 +
(
)
k [CF BrH]
5
2
where Amax is the observed maximum absorbance as a function
of the concentration ratio [CH4]/[CF2BrH] and Aeq is the
absorbance with no CH4 present. Aeq and k6/k5 were simulta-
neously varied, and the best fit was achieved with Aeq ) (0.405
F + CF BrH h X
(5,-5)
2
(
0.02) and k6/k5 ) (5.42 ( 0.81). Using k6 ) (6.8 ( 1.4) ×
(ref 15) gives k5 ) (1.25 ( 0.25) × 10 which is
consistent with the value of k5 ) (1.56 ( 0.31) × 10
molecule-
determined in section 3.3. This consistency
X f products
(7)
-11
-11
1
0
-11
3
cm
The different values of K5, k5, and k-5 and a short summary
of how they were determined is given in Table 1. The three
values calculated for K5 show good agreement.
1
-1
s
strongly suggests that the species absorbing at 280 nm is formed
by the reaction of F atoms with CF2BrH. If reaction of another
species with CF2BrH gave rise to the observed absorbance at
We decide to cite a final value of K5 ) (1.59 ( 0.13) ×
-17
3
-1
1
0
cm molecule obtained using the method which gave
2
80 nm, it is highly unlikely that consistent values of k5 would
the greatest precision. For k5 and k-5 we choose to quote values
which are averages of the results in Table 1 with error limits
that encompass the extremes of the ranges; hence k5 ) (1.4 (
be obtained from the two different techniques.
A value of k-5 can be obtained by observing the decay of X
in the presence of CH4. The following reactions are considered:
-11
3
-1 -1
5
0
s .
.5) × 10
cm molecule
s
and k-5 ) (8.8 ( 3.0) × 10
-
1
F + CF BrH h X
(5,-5)
2
3
.5. Decay of the Adduct. The decay of X was studied by
radiolysis of mixtures of 20 mbar of CF2BrH and 980 mbar of
SF6. The decay of the transient absorptions were observed at
F + CH f CH + HF
(6)
4
3
2
80 nm with the radiolysis dose varied over an order of
Every time X decomposes to F atoms and CF2BrH the F
atoms can reform X via reaction 5 or be consumed by reaction
magnitude. Figure 7A shows the transient absorption at 280
nm following the radiolysis at full radiolysis dose. The optical
path length was 120 cm. The decay was fitted using a first
order expression. Figure 8 displays a plot of kdecay as a function
of the maximum transient absorbance. Using the steady-state
approximation for X, the concentration ratio [X]/[F] is constant.
In the limit of zero dose and hence zero F atoms, no secondary
chemistry takes place and the only way that the concentration
of X can decrease is by decomposition of X, i.e., reaction 7.
The intercept in Figure 8 can therefore be interpreted as the
first-order decay rate constant for reaction 7. This gives k7 )
6
. The fraction of F atoms reacting with methane is given by
-1
k [CF BrH]
5
2
fCH4 ) 1 +
(
)
k6
[CH4]
If all the F atoms produced from the decomposition of X are
removed by reaction with methane, the decay rate of X will be
k-5. However, if only a fraction of F atoms is removed through
this channel, the rate of decay for X is given as kdecay ) fCH k-5.
4
4
-1
This means that if we plot kdecay as a function of fCH , the slope
(7.41 ( 0.72) × 10 s .
4
is k-5. Decay rates were obtained from experiments where the
concentration of CF2BrH was held fixed at 20 mbar, the
concentration of CH4 was varied over the range 2-40 mbar,
Using the diode array the radiolysis of mixtures of 20 mbar
of CF2BrH and 980 mbar of SF6 was followed. The UV path
length was 80 cm, and the resolution was 0.8 nm. Figure 9

7
054 J. Phys. Chem., Vol. 100, No. 17, 1996
Bilde et al.
TABLE 1
3
[cm molecule s-1
3
-1
k ]
-5 [s-1
method
as a function of [CF
formation as a function of [CF
figure
K
5
[cm molecule-1]
k
5
]
-17
A
k
x
2
BrH]
3
4
5
6
(1.59 ( 0.13) × 10
-17
-11
5
5
2
BrH]
(1.8 ( 0.8) × 10
(1.3 ( 0.3) × 10
(1.56 ( 0.31) × 10
(1.25 ( 0.25) × 10
(8.3 ( 2.5) × 10
(9.4 ( 1.5) × 10
-11
ratio plot with CH
decay of X in the presence of CH
4
-17
4
Figure 9. Absorbance following radiolysis of 20 mbar of CF
80 mbar of SF
electron pulse. The sharp absorption lines are due to CF
2
BrH and
9
6
. The spectrum was recorded 100-101 µs after the
2
radicals.
Figure 7. Transient absorbance following the pulsed radiolysis (A)
53% of maximum dose) of a mixture of 20 mbar of CF BrH and 980
mbar of SF . The UV path length was 120 cm. (B) Transient absorbance
following the pulsed radiolysis (42% of maximum dose) at 249 nm of
a mixture of 20 mbar of CF BrH and 980 mbar of SF . The UV path
length was 80 cm and the resolution was 0.16 nm.
(
2
6
2
6
Figure 10. Transient absorbance at 249 nm 50 µs after the electron
pulse versus the relative dose. Mixtures of 5 mbar of CF BrH and 980
mbar of SF were radiolysed.
2
6
to assume that CF2 is formed from decomposition of the adduct
via reaction 7.
Figure 7B displays the transient absorption at 249 nm
following the radiolysis of a mixture of 995 mbar SF6 and 5
mbar of CF2BrH. The dose was 42% of maximum, and the
optical path length was 80 cm. At 249 nm CF2 radicals absorb
strongly, and it is seen in Figure 7B that the absorption does
not fall to zero within the observed time interval, this is ascribed
to the formation of CF2 radicals. CF2 radicals are relatively
unreactive and are expected to be persistent in the reaction cell
over a time scale of the order of 10- s. A series of experiments
was performed where the radiolysis dose was varied over an
order of magnitude. As shown in Figure 7A, at 50 µs after the
radiolysis pulse the loss of adduct is essentially complete.
Figure 10 shows a plot of the absorbance 50 µs after the electron
pulse versus the relative dose. The solid line is a linear least
squares fit to the low dose data. The slope is â ) (0.61 (
0.04). The yield of CF2 relative to the amount of [F]0 was
4
Figure 8. kdecay versus maximum transient absorbance. See text for
details.
shows the spectrum obtained 100 µs after the electron pulse.
The characteristic spectrum of CF2 was observed. At this time
there is no adduct left (see Figure 7A), and it seems reasonable

Atmospheric Chemistry of CF2BrH
J. Phys. Chem., Vol. 100, No. 17, 1996 7055
TABLE 2: Selected UV Absorption Cross Sections for the
CF
wavelength
nm)
2
BrH-F Adduct
2
0
20
10 σ
wavelength
(nm)
10 σ
2
(
(cm molecule-1)
2
(cm molecule-1)
2
30
40
50
60
70
80
90
00
466
795
985
310
320
330
340
350
360
370
380
1012
682
437
266
163
94
2
2
2
2
2
2
3
977
1051
1290
1482
1334
53
18
Figure 11. UV spectrum of the CF
2
BrH- -F adduct. The spectrum of
the CH Br- -F adduct is shown for comparison.
3
calculated from the expression [CF2]/[F0] ) â2.303/(σ(CF2)-
1
5
-1
[
F]0), where [F]0 ) 3.2 × 10 molecule cm and l ) 80 cm is
the optical path length. Unfortunately, the literature values of
σ(CF2) differ considerably: σ249nm(CF2) ) (2.9 ( 0.4) ×
-
17 16
-18 17
1
)
0
,
σ249nm(CF2) ) (8.0 ( 1.4) × 10
-1 18
,
and σ249nm(CF2)
-
18
2
(8.4 ( 0.4) × 10
cm molecule . From these values
the yield of CF2 was calculated to be 19%, 69%, or 65%,
respectively. Quantification of the CF2 yield is further com-
plicated by the fact that σ249nm(CF2) is sensitive to the internal
excitation of the CF2 radical.18 The internal excitation of the
CF2 radicals observed here is unknown. Therefore, we are
unable to provide a precise quantification of the substantial CF2
yield.
Figure 12. Loss of CF BrH versus loss of methane when CF BrH/
2
2
methane mixtures were exposed to Cl atoms in 700 Torr of air diluent.
using the FTIR system to investigate the kinetics and mechanism
of the reaction of Cl atoms with CF2BrH. The techniques used
are described in detail elsewhere. Photolysis of molecular
chlorine was used as a source of Cl atoms:
3
.6. UV Absorption Spectrum of the Adduct X. The
2
0
absolute cross section of the adduct X was determined at 280
nm from the experiments described in section 3.1. The
absorption cross section was determined from the expression σ
2.303R(1 + K5[CF2BrH])/(lK5[CF2BrH][F]0), where R is the
slope of a linear regression fit to the data shown in Figure 2, l
)
Cl + hν f 2Cl
(8)
(9)
2
15
-3
)
120 cm, and [F]0 ) 3.2 × 10 molecules cm is the F atom
Cl + CF BrH f CF Br + HCl
2 2
yield at full radiolysis dose. The results were σ ) (1.35 (
-17
2
-1
0
)
.15) × 10 cm molecule from experiments with [CF2BrH]
The kinetics of reaction 9 were measured relative to reactions
0 and 11. The observed losses of CF2BrH versus CD4 and
-
17
20 mbar and σ ) (1.16 ( 0.13) × 10
from experiments
1
with [CF2BrH] ) 2.5 mbar.
The UV absorbance spectrum of the adduct X following the
pulsed irradiation of a mixture of 20 mbar of CF2BrH and 980
mbar of SF6 was measured using a diode array. The conditions
were as follows: spectral resolution ) 3 nm, wavelength )
Cl + CD f CD + DCl
(10)
(11)
4
3
Cl + CH f CH + HCl
4
3
2
30-380 nm, gate ) 1-2 µs, 53% of full radiolysis dose. To
CH4 following the UV irradiation of CF2BrH/CD4/Cl2 and
CF2BrH/CH4/Cl2 mixtures, respectively, in 700 Torr total
pressure of air diluent are shown in Figure 12. Linear least-
squares analysis gives k9/k10 ) 0.89 ( 0.05 and k9/k11 ) 0.061
place the observed UV absorption of the adduct on an absolute
basis the absorptions were scaled to that at 280 nm and
converted into absolute absorption cross sections using σ(280nm)
-
17
2
-1
)
1.3 × 10 cm molecule . The spectrum of the adduct is
-15
(
0.005. Using k10 ) 6.1 × 10
(ref 20) and k11 ) 1.0 ×
shown in Figure 11, and the absorption cross sections are given
in Table 2. The overall uncertainty in the absorption cross
sections is estimated to be 25% and arises from uncertainty in
the F atom yield, the equilibrium constant, and the stability in
the lamp pulse used to record the spectrum. In Figure 11 a
spectrum of the adduct formed in the reaction of F atoms with
CH3Br is shown for comparison. As seen from Figure 11,
the spectra of the two adducts are almost identical suggesting
that they have similar structure.
-13
-15
1
0
0
(ref 21) gives k9 ) (5.4 ( 0.3) × 10
and k9 ) (6.1 (
-15
3
-1 -1
.5) × 10
cm molecule s , respectively. We choose to
cite a final value for k9 which is the average of the two
experimental determinations with error limits which encompass
the extremes of the determinations. Hence, k9 ) (5.8 ( 0.8) ×
-15
3
-1 -1
10
cm molecule s . We estimate that potential systematic
1
9
errors associated with uncertainties in the reference rate
constants could add an additional 10% to the uncertainty range.
Propagating this additional 10% uncertainty gives k9 ) (5.8 (
-15
3
-1 -1
1
.0) × 10
cm molecule s . While there is no literature
4
. FTIR Results and Discussion
.1. FTIR Study of the Reaction of Cl Atoms with
CF2BrH. A series of relative rate experiments was performed
available for k9 for direct comparison, our result is of the order
4
of magnitude expected for reaction of Cl atoms with a
2
2
halomethane containing two F substituents.

7
056 J. Phys. Chem., Vol. 100, No. 17, 1996
Bilde et al.
Figure 13. Formation of COF
following the Cl atom (circles) or F atom (triangles) initiated oxidation
of CF BrH in 700 Torr of air diluent.
2 2
versus the loss of CF BrH observed
Figure 14. Loss of CF
2
BrH versus loss of CF
3 2 3 3
CFH and CF CH when
CF BrH/CF CFH and CF
2
3
2
2
BrH/CF CH mixtures were exposed to F
3
3
2
atoms in 700 Torr of either air (filled symbols) or N
diluent.
2
(open symbols)
Figure 13 shows a plot of the observed formation of COF2
versus the loss of CF2BrH following UV irradiation of a mixture
of 9.2 mTorr of CF2BrH and 174 mTorr of Cl2 in 700 Torr of
air. Linear least-squares analysis of the data in Figure 13 gives
a molar yield of COF2 of 100 ( 4%. COF2 was the only
carbon-containing product observed. By analogy to the behavior
of other peroxy radicals22 it is expected that the self-reaction
of CF2BrO2 radicals proceeds to give the corresponding alkoxy
radicals, which will then eliminate a Br atom to give COF2:
-1
23
to be 72 and 104 kcal mol , respectively. The bond strengths
-1
in Br-Cl and H-Cl are 52 and 103 kcal mol , respectively.
Hence, H atom abstraction is the only thermodynamically
allowed channel of the reaction of Cl atoms with CF2BrH. The
experimental results discussed above are consistent with the
expectation that CF2Br radicals are formed in 100% yield from
reaction 9. In light of the adduct formation between F atoms
and CF2BrH, it seems likely that Cl atoms also form an adduct
with CF2BrH. From the present work we cannot discern
whether formation of CF2Br radicals in reaction 9 occurs via
concerted H-atom abstraction or via the intermediacy of a
CF2BrH-Cl adduct which eliminates HCl.
CF BrO + CF BrO f CF BrO + CF BrO + O
2
(12)
(13)
2
2
2
2
2
2
CF BrO + M f CF O + Br
2
2
4
.2. FTIR Study of the Reaction of F Atoms with CF2BrH.
The experimental observation that Cl-initiated oxidation of
CF2BrH in our system gives 100% yield of COF2 is entirely
consistent with the expected behavior. In addition to the
experiments performed in air, two experiments were performed
to investigate the products following the UV irradiation of
CF2BrH/Cl2/N2 mixtures. The initial CF2BrH concentration was
To investigate the kinetics and mechanism of the reaction of F
atoms with CF2BrH, a series of relative rate experiments was
performed using the FTIR system. Photolysis of molecular
fluorine was used as a source of F atoms:
F2 + hν f 2F
(16)
(4)
1
2.3 mTorr, the Cl2 concentration was either 2.7 or 10.4 Torr,
and N2 was added to give a total pressure of 700 Torr.
CF2BrCl was the major product following UV irradiation of
CF2BrH/Cl2 mixtures, in yields of 88 ( 4% and 94 ( 4% from
experiments employing 2.7 and 10.4 Torr of Cl2, respectively.
In addition to CF2BrCl, one or more unknown products were
observed with infrared features at 793, 831, 923, 1093, 1153,
F + CF BrH f products
2
The kinetics of reaction 4 were measured relative to reactions
7 and 18. The observed losses of CF2BrH versus CF3CFH2
1
-1
F + CF CFH f CF CFH + HF
(17)
(18)
and 1161 cm . Trace amounts of COF2 were also detected
3
2
3
-1
by virtue of its characteristic feature at 774 cm . The observed
products can be rationalized in terms of the following reactions:
F + CF CH f CF CH + HF
3
3
3
2
CF BrH + Cl f CF Br + HCl
(9)
(14)
(15)
and CF3CH3 following the UV irradiation of CF2BrH/CF3CFH2/
F2 and CF2BrH/CF3CH3/F2 mixtures, respectively, in 700 Torr
total pressure of N2 or air diluent are shown in Figure 14. Linear
least-squares analysis gives k4/k17 ) 0.67 ( 0.06 and k4/k18 )
0.40 ( 0.04. The reactivity of both CF3CFH2 and CF3CH3
toward F atoms has been the subject of recent investigations in
2
2
CF Br + Cl f CF BrCl + Cl
2
2
2
CF Br f f products
2
1
5,24
-12
Reaction 15 represents an unknown loss mechanism for CF2Br
radicals. The increase in CF2BrCl yield with increasing Cl2
concentration reflects a competition between reactions 14 and
our laboratories.
Using k17 ) 1.3 × 10
(ref 15) and k18
-12
-13
) 2.3 × 10
(ref 24) gives k4 ) (8.7 ( 0.8) × 10
and k4
-
13
3
-1 -1
) (9.2 ( 0.9) × 10
cm molecule s , respectively. We
1
5 for the available CF2Br radicals. The trace of COF2 reflects
choose to cite a final value for k4 which is the average of the
two experimental determinations with error limits which en-
compass the extremes of the determinations. Hence, k4 ) (9.0
the presence of a small amount of O2 in the 700 Torr of N2
diluent, and so reactions 2, 12, and 13 play a small role.
At this point it is germane to consider the thermochemistry
of reaction 9. By comparison to structurally similar molecules,
we estimate the C-Br bond and C-H bond strengths in CF2BrH
-13
3
-1 -1
( 1.1) × 10 cm molecule s . We estimate that potential
systematic errors associated with uncertainties in the reference
rate constants could add an additional 20% to the uncertainty

Atmospheric Chemistry of CF2BrH
J. Phys. Chem., Vol. 100, No. 17, 1996 7057
TABLE 3: Observed Product Yields (percent) following
Radiation of F
2
/CF
2 2
BrH/N Mixtures
1
1.5 mTorr
11.5 mTorr
1.92 Torr
11.5mTorr
3.8 Torr
[CF
2
BrH][F
2
]
0.444 Torr
[
[
[
[
COF
2
]
53 ( 4
9 ( 1
8 ( 2
5 ( 1
80 ( 9
14 ( 4
64 ( 6
16 ( 3
<2
15 ( 3
70 ( 6
16 ( 6
<2
CF
CF
CF
4
3
3
]
Br]
CF
O
3
3
]
total C
94 ( 13
101 ( 11
range. Propagating this additional 20% uncertainty gives k4 )
-
13
3
-1 -1
(
9.0 ( 2.1) × 10
cm molecule s .
While there are no literature data to compare with this result,
-13
3
we can compare the value of k4 ) (9.0 ( 2.1) × 10
cm
molecule-
1
s
-1
obtained using the FTIR system with the value
expected on the basis of results from the pulse radiolysis
experiments. As discussed in sections 3.4 and 3.5, the pulse
radiolysis results show that the reaction proceeds via the
formation of an adduct which is in dynamical equilibrium with
F atoms and CF2BrH. In addition to decomposition into
reactants, the adduct is also lost via a process which follows
pseudo-first-order kinetics to give products other than F atoms
and CF2BrH:
F + CF BrH h X
(5,-5)
2
X f products
(7)
The reactivity of F atoms toward CF2BrH determined by the
FTIR technique was derived by comparing the loss of CF2BrH
to that of a reference compound whose reactivity toward F atoms
is known. Hence, the FTIR technique is “blind” to any reaction
channel which leads to reformation of CF2BrH within the
experimental time scale (minutes). The “effective” rate constant
measured by the FTIR method is then K5k7. Using the values
Figure 15. Product yields of CF
4
(O), CF
3
Br(2), COF
2 3 3
(b), and CF O -
CF (0) following irradiation of CF
3
2
BrH/F
2
/N mixtures. The experi-
2
mental details are given in the text and in Table 3.
Four carbon-containing products were observed following UV
irradiation of CF2BrH/F2/N2 mixtures; CF4, CF3Br, COF2, and
CF O CF . In all product studies care must be taken to ensure
-
17
3
-1
of K5 ) (1.59 ( 0.13) × 10
0.7) × 10 s from the pulse radiolysis study, it would be
predicted that the “effective” rate constant measured by the FTIR
cm molecule and k7 ) (7.4
3
3
3
4
-1
that the reactant compound is lost only via the reaction of interest
and that the products are not lost via secondary reactions. To
assess possible heterogeneous and photolytic loss in the
chamber, CF BrH/air and CF Br/air mixtures were prepared,
(
-
12
3
-1 -1
method would be (1.2 ( 0.2) × 10 cm molecule s . The
-
13
3
value measured by the FTIR, (9.0 ( 2.1) × 10
cm
2
3
molecule- s , is in excellent agreement with the value
1
-1
allowed to stand for 20 min, then irradiated for 20 min; no loss
<1%) of CF2BrH or CF3Br was discernible. It has been
(
predicted by the pulse radiolysis experiments. This agreement
supports the assumption made in section 3.1 that direct hydrogen
abstraction makes a negligible contribution to the reaction of F
atoms with CF2BrH. The kinetic data obtained using the pulse
radiolysis and FTIR techniques provide a consistent picture of
the reaction of F atoms with CF2BrH proceeding essentially
via the formation of an adduct. Under the present experi-
mental conditions, 1 atm at 296 K, the predominant fate of the
CF2BrH- -F adduct is decomposition to reactants. A small
fraction (8%) of the adduct decomposes into products.
To provide further insight into the mechanism of reaction 5,
experiments were performed using the FTIR smog chamber in
which CF2BrH/F2 mixtures in 700 Torr of either air or N2 diluent
were subject to UV irradiation. COF2 was the only carbon-
containing product observed following UV irradiation of
CF2BrH/F2 mixtures in air diluent. The formation of COF2 is
plotted versus the loss of CF2BrH in Figure 13. The results
obtained using F atom initiated oxidation of CF2BrH in air are
indistinguishable from those obtained using Cl atom initiation,
with COF2 formed in essentially 100% yield. Three experiments
were performed in which F2/CF2BrH mixtures at a total pressure
of 700 Torr made up with N2 diluent were irradiated and the
loss of CF2BrH and the formation of products were monitored
by FTIR spectroscopy. Experimental conditions are given in
Table 3. Figure 15 shows the product yields with linear least-
squares analysis.
established previously that COF2 and CF3O3CF3 are not subject
to loss via heterogeneous reactions, photolysis, or reaction with
25
F atoms. Finally, CF2BrH/CF3Br/F2/air mixtures were subject
to UV irradiation to check for reaction of F atoms with CF3Br.
In such experiments loss of CF2BrH was observed but there
was no loss (<1%) of CF3Br, showing that the CF3Br is not
consumed in this system via reaction with F atoms (F atoms
may react with CF3Br, but if so, the adduct must decompose to
reform CF3Br resulting in no overall loss of this compound).
As seen from Table 3, the yields of CF4 and CF3Br increased,
while those of COF2 and CF3O3CF3 decreased, with increasing
[F2]0. CF3Br and COF2 formation can be rationalized by the
formation of CF2Br radicals via overall hydrogen atom abstrac-
tion from CF2BrH followed by reaction either with F2 or with
O2 impurity in the chamber to give CF3Br or CF2BrO2 radicals.
The CF2BrO2 radicals are then converted into COF2 via reactions
1
2 and 13:
CF BrH + F f CF Br + HF
(4a)
(19)
(2)
2
2
CF Br + F f CF Br + F
2
2
3
CF Br + O f CF BrO
2
2
2
2
The formation of CF4 and CF3O3CF3 shows that a substantial
fraction of the reaction of F atoms with CF2BrH proceeds via

7
058 J. Phys. Chem., Vol. 100, No. 17, 1996
Bilde et al.
a mechanism other than simple hydrogen abstraction. There is
only one reaction that can form CF3O3CF3 in the chamber,
namely
radiolysis experiments shows that decomposition to reactants
(pathway -5) is the main fate of the adduct. The question then
is which of the remaining possibilities account for that fraction
of the adduct loss which is not simple decomposition into
reactants. The results from the FTIR study shed some light on
CF O + CF O f CF O CF
3
(20)
3
3
2
3
3
1
5
this issue. CF3H is relatively unreactive toward F atoms and,
if formed in the FTIR experiments, will not be lost via secondary
reactions. The absence of any observable CF3H shows that
pathway 27 is of no importance. For those experiments
employing the largest [F2]0, CF4 was the dominant product
accounting for up to 70% of the CF2BrH loss. Pathways 23
and 25 lead indirectly (via the CF2Br + F2 reaction) or directly
to CF3Br and not CF4, so it appears these pathways are, at most,
of minor importance. Pathway 26 leads to CF4 formation via
reaction of the CF3 radical with F2 and so could be important.
In the presence of 700 Torr of air, CF3 radicals are converted
CF3O2 and CF3O radicals are formed following reaction of CF3
radicals with O2. The detection of CF3O3CF3 as a minor product
shows that CF3 radicals are formed. The CF4 product may also
reflect the formation of CF3 radicals in the chamber and their
subsequent reaction with F2:
CF + F f CF + F
(21)
(22)
(ref
3
2
4
CF + O + M f CF O + M
3
2
3
2
-14
At 296 K and 700 Torr total pressure k21 ) 7.0 × 10
2
5
-
12
3
-1 -1 27
essentially quantitatively into CF3O3CF3. The observation of
00% yield of COF2 in CF2BrH/F2/air experiments shows that
2
6) and k22 ) 3.5 × 10
cm molecule s . Hence, k22/
1
k21 ) 50. For the conditions of experiment no. 1 (see Table 3)
and assuming all the CF4 is attributable to reaction 21, then the
O2 impurity is present at a concentration which is 2.2% of the
F2, i.e., 9.9 mTorr. The presence of such a small amount of O2
from a combination of incomplete evacuation, leaks of air into
the chamber, and/or the presence of O2 in the reactant gases is
not unreasonable.
in 700 Torr of air diluent pathway 26 is not a significant loss
of the adduct, i.e., if reaction 26 is responsible for the observed
CF4 in the N2 experiments, then there must be a reaction of the
adduct with O2 to give directly, or indirectly, COF2 in the
experiments conducted in air diluent.
In the pulse radiolysis experiments we have shown that the
decomposition of the adduct gives a substantial (20-72%) yield
of CF2 radicals. Hence, we conclude that pathway 24, although
slightly endothermic, is of importance. There is insufficient
kinetic data concerning the chemistry of CF2 radicals to predict
their behavior in the FTIR smog chamber experiments. It is
conceivable that in the presence of 2-4 Torr of F2 the CF2
radicals react to give CF4, but in the presence of trace amounts
of air they react to give mainly COF2 and in 700 Torr of air
they react to give COF2 exclusively.
5
. Discussion
The pulse radiolysis experiments described in section 3
demonstrate that the reaction of F atoms with CF2BrH proceeds
substantially via the formation of an adduct, CF2BrH- -F, which
at 296 K decomposes rapidly to regenerate CF2BrH and F atoms
and to give other products. Let us consider how the results
obtained using the FTIR smog chamber system fit in with this
picture. Several consistent points emerge. First, the observation
of loss of CF2BrH following UV irradiation of CF2BrH/F2
mixtures shows that either not all of the reaction gives the
adduct, or not all of the adduct decomposes to regenerate the
initial reactants, or both. Second, the observation of CF4 and
CF3O3CF3 products cannot be explained if the reaction of F
atoms with CF2BrH proceeded via a simple hydrogen abstraction
mechanism. There are several thermodynamically feasible
decomposition pathways for the adduct:
6
. Conclusion
Halogen atom adducts have been observed in solution; see,
for example, refs 29-31. These adducts are described as
3
2
charge-transfer complexes according to the Mulliken theory.
We present here a large body of self-consistent kinetic and
mechanistic data which shows that the reaction of F atoms with
CF2BrH gives an adduct in the gas phase. The adduct is in
dynamic equilibrium with CF2BrH and F atoms. The adduct
can also decompose to CF2 radicals and other products. Further
work is needed to assess if this novel reaction mechanism is
important in reaction of F atoms with other brominated
compounds.
CF BrH- -F f CF BrH + F
(-5)
2
2
-
1
CF BrH- -F f CF Br + HF
∆H ) -32 kcal mol
2
2
(23)
-
1
Acknowledgment. We thank Steve Japar (Ford Motor Co.)
for a critical reading of the manuscript.
CF BrH- -F f CF + Br + HF
∆H ) +3 kcal mol
2
2
(24)
-
-
-
1
1
1
References and Notes
CF BrH- -F f CF Br + H
∆H ) -13 kcal mol
∆H ) -30 kcal mol
∆H ) -49 kcal mol
(25)
2
3
(1) World Meteorological Organization, Global Ozone Research and
Monitoring Project-Report No. 20; Scientific Assessment of Stratospheric
Ozone, appendix; AFEAS Report, 1989; Vol. 2, Chapter 6.
CF BrH- -F f CF + HBr
(26)
(27)
2
3
(2) Atkinson, R. J. Phys. Chem. Ref. Data 1989, Monograph No. 1.
(3) Wallington, T. J.; Nielsen O. J. Chem. Phys. Lett. 1991, 187, 33.
(4) Wallington, T. J.; Ball, J. C.; Nielsen, O. J.; Bartkiewicz, E. J.
CF BrH- -F f CF H + Br
2
3
Phys. Chem. 1992, 96, 1241.
5) Nielsen, O. J.; Ellermann, T.; Bartkiewicz, E, Wallington, T. J.;
Hurley, M. D. Chem. Phys. Lett. 1992, 192, 82.
(6) Nielsen O. J.; Ellermann, T; Sehested, J.; Bartkiewicz, E.; Wall-
ington, T. J.; Hurley, M. D. Int. J. Chem. Kinet. 1992, 24, 1009.
The heats of reaction above were derived using ∆Hf(CF3) )
(
-1 28
-1
-
111 kcal mol , ∆Hf(CF2BrH) ) -109 kcal mol , ∆Hf(CF2-
-
1
Br) ) -57 kcal mol (estimated by interpolation of literature
2
3
data for CF3H and CBr3H, and CF3 and CBr3 ), and
(
7) Sehested, J.; Wallington T. J. EnViron. Sci. Technol. 1993, 27, 146.
1
∆
Hf(CF2BrH--F) ) -90 kcal mol- (assumed, for the sake of
(8) Nielsen, O. J.; Ellermann, T.; Sehested, J.; Wallington, T. J. J. Phys.
Chem. 1992, 96, 10875.
argument, equal to ∆Hf(CF2BrH) + ∆Hf(F)). Let us consider
these possibilities in turn and assess their likely importance
based upon the experimental observations presented herein. As
discussed in sections 3.2-3.5, the evidence from the pulse
(
9) Sehested, J.; Ellermann, T.; Nielsen, O. J.; Wallington, T. J.; Hurley,
M. D. Int. J. Chem. Kinet. 1993, 25, 701.
10) Wallington, T. J.; Japar, S. M. J. Atmos. Chem. 1989, 9, 399.
(11) Nielsen, O. J. Risø-R-480, 1984.
(

Atmospheric Chemistry of CF2BrH
J. Phys. Chem., Vol. 100, No. 17, 1996 7059
(
12) Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92,
67.
13) Møgelberg, T. E.; Sehested, J.; Bilde, M.; Wallington, T. J.; Nielsen,
O. J. J. Phys. Chem., in press.
14) Capellos, C.; Bielski, B. H. J. Kinetic Systems; Wiley & Sons: New
York, 1972.
(23) Lias, S. G.; Liebman, J. F.; Levin, R. D.; Kafafi, S. A. NIST
Standard Reference Database; Structures and Properties, Version 2.01, Jan
1994.
(24) Nielsen, O. J.; Gamborg, E.; Sehested, J.; Wallington, T. J.; Hurley,
M. J. J. Phys. Chem. 1994, 98, 9518.
6
(
(
(25) Nielsen, O. J.; Ellermann, T.;Sehested, J.; Bartkiewicz, E.; Wall-
(15) Wallington, T. J.; Hurley, M. D.; Shi, J.; Maricq, M. M.; Sehested,
ington, T. J.; Hurley, M. D. Int. J. Chem. Kinet. 1992, 24, 1009.
J.; Nielsen, O. J.; Ellermann, T. Int. J. Chem. Kinet. 1993, 25, 651.
(
1.
26) Plumb, I. C.; Ryan, K. R. Ber. Bunsen-Ges Phys. Chem. 1986, 6,
(16) Sharpe, S. Hartnett,B.; Sethi, H. S.; Sethi, D. S. J. Photochem. 1987,
1
3
8, 1.
(
27) Kaiser, E. W.; Wallington, T. J.; Hurley, M. D. Int. J. Chem. Kinet.
(
(
(
17) Duperrex, R.; Bergh, H. J. Chem. Phys. 1979, 71, 3613.
18) Chaudhury, P. K. J. Phys. Chem. 1995, 99, 12084.
19) Sehested, J.; Bilde, M.; Møgelberg, T.; Wallington. T. J.; Nielsen,
1
995, 27, 205.
(28) Schneider, W. F.; Wallington, T. J. J. Phys. Chem. 1993, 97, 12783.
O. J. J. Phys. Chem., submitted.
(29) Gover, T. A.; Porter, G. Proc. R. Soc. 1961, A262, 476.
(30) B u¨ hler, R. E. J. Phys. Chem. 1972, 76, 22.
(20) Wallington, T. J.; Hurley, M. D. Chem. Phys. Lett. 1992, 189, 437.
(
21) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
(31) Fornier, P. V.; Bonneau, R.; Joussout-Dubien, J. Chem. Phys. Lett.
1973, 19, 2.
Kurylo, M. J.; Howard, C. J.; Ravishankara A. R.; Kolb, C. E.; Molina, M.
J. Jet Propulsion Laboratory Publication 94-26, Pasadena, CA, 1994.
(32) Mulliken, R. S.; Person, W. B. Molecular Complexes; Wiley &
(22) Mallard, W. G.; Westley, F.; Herron, J. T.; Hampson, R. P. NIST
Sons: New York, 1969.
Chemical Kinetics Database, version 6.0, NIST Standard Reference Data,
Gaithersburg, MD, 1994.
JP9530111