Experimental and theoretical studies of the reaction between CF3 and NO2 at 298 K

Article abstract of DOI:10.1016/S0009-2614(98)00074-8

The title reaction has been studied by pulse radiolysis combined with time-resolved infrared diode laser spectroscopy. The kinetics of CF₃ were analysed taking into account the two competing reactions CF₃+ CF₃+M → C₂F₆+M (1) and CF₃+NO₂ → CF₂O+FNO (2a). Based on studies of the yield and kinetics of CF₃ we determined values of the absorption cross section, σ(CF₃)=(1.96±0.20)×10^-17 cm² molecule^-1 at 1263.567 cm^-1 and the bimolecular rate constant, k₁=(1.04±0.12)×10^-11 cm³ molecule^-1 s^-1 with a bath gas pressure of 20 mbar at 298 K. Reaction (2a) was studied under pseudo-first-order conditions and the value of k_2a=(1.53±0.20)×10^-11 cm³ molecule^-1 s^-1 was found to be independent of bath gas pressure in the range 4-22 mbar. Based on the characteristic absorption spectra of the products, the branching ratio of reaction (2a) was found to be 0.95.

Full text of DOI:10.1016/S0009-2614(98)00074-8

3 April 1998
Ž
.
Chemical Physics Letters 286 1998 138–144
Experimental and theoretical studies of the reaction between CF3
and NO2 at 298 K
a
b
b
a
Palle Pagsberg , Jerzy T. Jodkowski , Emil Ratajczak , Alfred Sillesen
a
Chemical ReactiÕity Section, Plant Biology and Biogeochemistry Department, Risø National Laboratory, DK-4000 Roskilde, Denmark
b
Department of Physical Chemistry, Wrocław UniÕersity of Medicine, Pl. Nankiera 1, 50-140 Wrocław, Poland
Received 30 May 1997; in final form 19 January 1998
Abstract
The title reaction has been studied by pulse radiolysis combined with time-resolved infrared diode laser spectroscopy.
Ž .
The kinetics of CF3 were analysed taking into account the two competing reactions CF3q CF3qM ™ C2F6qM 1 and
Ž
.
CF3qNO2 ™ CF2OqFNO 2a . Based on studies of the yield and kinetics of CF3 we determined values of the absorption
y
17
cm2 moleculey1 at 1263.567 cmy1 and the bimolecular rate constant,
Ž
.
Ž
.
cross section, s CF3 s 1.96"0.20 =10
y
11
cm3 moleculey1 sy1 with a bath gas pressure of 20 mbar at 298 K. Reaction 2a was studied
Ž
.
Ž .
k1s 1.04"0.12 =10
under pseudo-first-order conditions and the value of k2as 1.53"0.20 =10
independent of bath gas pressure in the range 4–22 mbar. Based on the characteristic absorption spectra of the products, the
y
11
cm3 moleculey1 sy1 was found to be
Ž
.
Ž
.
branching ratio of reaction 2a was found to be 0.95. q 1998 Elsevier Science B.V.
1. Introduction
have been able to study the kinetics of CF3, CF2O
and FNO using time-resolved infrared diode laser
spectroscopy.
Ab initio electronic structure theory was em-
ployed to provide insight into the potential energy
surface of the structures which may play a role in the
reaction mechanism.
The role of CF3 radicals in stratospheric chem-
w
x
istry has been the subject of experimental 1,2 and
w
x
theoretical 3,4 studies. Kinetics and mechanisms of
the reactions of CF3 with O3, HO2 , NO and NO2 are
considered to be of special importance for the com-
plex chemistry of the ozone layer. Several products
have been identified, e.g. CF3O, CF3O2 and CF2O.
For the reaction between CF3 and NO2 the major
products are CF2O and FNO. However, details of the
reaction mechanism and branching ratios for the
proposed reactions are still being discussed. Re-
2. Experimental studies
2.1. Experimental
The experimental technique, based on pulse radi-
olysis combined with time-resolved infrared diode
laser spectroscopy, has previously been described in
ported values for the overall rate constant are in the
y11
Ž
.
range of 0.27–3.2 =10
cm3 moleculey1 sy1
w
x
w x
5–11 . In the present experimental investigation we
details 12 . CF3 radicals were produced by radioly-
0009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S0009-2614 98 00074-8

(
)
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
139
r
Ž
.
sis of ArrSF6rCF3I mixtures and the kinetics of CF3
was studied by monitoring the transient absorption of
strong rotational transitions within the n3 band of
position of R7 27 was the same as reported by
Yamada and Hirota within an estimated uncertainty
of "10y3 cmy1. With the diode laser tuned to the
center of this line we observed a simple second order
decay in accordance with the combination reaction
w
x
CF3. 13 . The product FNO was identified by
recording the characteristic rotational pattern of the
n1 band at 1844 cmy1. For kinetic studies we have
q
Ž
.
chosen a strong R-branch transition, R5 17 ob-
CF qCF qM ™ C F qM .
3Ž 6 Ž
1
.
.
Ž .
served at 1857.324 cmy1 14 . The other main prod-
3
2
w
x
uct, CF2O was identified by scanning the n4 band at
1243.3 cmy1 15 and for kinetic studies we have
selected a strong P-branch transition at 1231.02
cmy1. The experimental kinetic features were ana-
lyzed by computer modeling using the CHEM-
The initial yield of F-atoms was determined by
measurement of the amount of CH4 consumed in the
titration reaction, F q CH4 ™ HF q CH3. Four
strong CH4 lines in the range of 1235.925–1236.029
cmy1 were scanned before and after electron pulse
w
x
w
x
SIMUL program 16 . High-purity gases, Ar)
99.998%, SF6 and NO2 )98% were supplied by
L’Air Liquide and CF3I)97% was provided by
PCR.
irradiation and the initial yield of F-atoms was found
14
w x w x Ž .
to be
molecules cmy3
The kinetics of CF3 shown in Fig. 1 were anal-
w x w x
ysed using the value of CF3 max s F combined
F
s D CH 4 s 1.65 " 0.17 = 10
.
0
2.2. Kinetics of CF3
0
Ž
.
with the observed decay half-life of ts 290"5
w
x
CF3 radicals were produced by the reaction Fq
CF3I ™ FIqCF3 which was initiated by pulse radi-
olysis of ArrSF6rCF3I mixtures. The n3 band of
CF3 has been recorded in the range of 1233 to 1277
mss1r2k1 CF3 max . From the maximum of the
transient absorption signal recorded with an optical
path length of 320 cm we have derived an absorption
y17
Ž
.
Ž
.
cross section of s CF3 s 1.96"0.20 =10
r
cmy1 by Yamada and Hirota 13 . To avoid interfer-
ence with the strong SF6 band at around 1255 cmy1
we have selected a strong R-branch transition at
1263.567 cmy1. The spectrum shown as an inset in
Fig. 1 was obtained by fast scan of the diode laser in
the range of 1263.45 –1263.60 cmy1. N2O was used
for wavelength calibration and the calculated line
cm2 moleculey1 at the line center of the n3, R7 27
Ž .
w
x
transition at 1263.567 cmy1. The line-strength corre-
sponding to a Doppler width of 9.44=10y4 cmy1
was found to be Ss3.94=10y20 cm moleculey1
.
The value of k1 s 1.04"0.12 =10
cm3 mole-
y11
Ž
.
culey1 sy1, obtained with a total pressure of pt s20
Ž
.
mbar may be compared with a value of 2.2"0.5
=10y12 cm3 moleculey1 sy1 derived from a sec-
ond-order plot of CF3 radicals following photolysis
Ž
.
w x
of CF3 2CO at 0.79 mbar 17 . Experimental evi-
dence for a pronounced pressure dependence of reac-
Ž .
tion 1 was presented in the pioneering work by
Pimentel et al., where the kinetics of CF3 was stud-
w
x
ied by rapid scan infrared spectroscopy 18 . A value
of k1 s9.8=10y12 cm3 moleculey1 sy1 was ob-
tained with a total pressure of 132 mbar at 298 K.
Reported values of the limiting high-pressure rate
Ž
.
constant are in the range of krec.inf s 2.3"1.0 =
10y11 cm3 moleculey1 sy1 at Ts300–1500 K 19
w
x
y12
Ž
.
and 3.9"1.3 =10
cm3 moleculey1 sy1 at 296
w
x
K 20 . However, despite of obvious uncertainties
with respect to the fall-off curve we think that our
value of k1 obtained at pt s20 mbar and MsSF6
at 298 K may be taken with confidence.
Ž
.
Ž
.
Fig. 1. Kinetics of the reaction CF3 qCF3 qM ™ C2 F6 qM
monitored at 1263.567 cmy1. p M s p SF6 s20 mbar at 298
Ž
.
Ž
.
K.

(
)
140
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
NO2 the half-life was reduced by a factor of ten
Ž
.
lower curve and the time profile could be fitted
with a single exponential curve taking into account a
minor contribution from the absorption of a stable
product which was later identified as CF2O. Fig. 2
shows the decay rates of CF3 observed observed
with varying partial pressures of NO2 and total
pressures in the range of 4–22 mbar at 298 K.
Values of the bimolecular rate constant are presented
in Table 1a. Within the experimental uncertainties
y11
cm3 mole-
Ž
.
the value of k2 s 1.53"0.20 =10
culey1 sy1 was found to be independent of the total
pressure.
2.3. Kinetics of product formation
Two main products were identified by the charac-
teristic features of the n4 band of CF2O at 1243.0
cmy1 and the n1 band of FNO at 1844.1 cmy1. The
spectrum of CF2O shown in Fig. 3 was observed
after irradiation of a gas mixture containing 0.2 mbar
NO2 , 0.5 mbar CF3I and 19.3 mbar SF6. Line posi-
tions and relative intensities of the P-branch transi-
tions are in perfect agreement with the high resolu-
tion FTIR spectrum reported by Camy-Peyret et al.
Fig. 2. Kinetics of CF3. The insert shows the decay of CF3 in the
Ž
.
absence of NO2 upper curve and in the presence of 0.1 mbar
Ž
.
NO2 lower curve . Values of the pseudo-first-order rate constant
k
) rsy1 were determined with varying partial pressures of NO2
Ž
.
.
Ž
.
and total pressures of p ArqCF3IqNO2 , 22 mbar open circles ,
Ž
Ž
.
11 mbar open triangles and 4 mbar filled triangles .
For the reaction between CF3 and NO2 we have
considered the following channels:
CF3 qNO2 ™ CF2OqFNO,
CF3 qNO2 ™ CF3OqNO,
CF3 qNO2 ™ CF2OqFqNO,
CF3 qNO2 qM ™ CF3NO2 qM.
2a
2b
2c
2d
Ž
Ž
Ž
Ž
.
.
.
.
Thus, the initial decay rate of CF3 is described by
the expression
)
w
x
w
x
w
x
ydln CF3 rdtsk sk1 CF3 qk2 NO2
,
0
0
where k2 s k2a q k2b q k2c q k2d represents the
overall rate constant. The last term becomes domi-
nating at high NO2 concentrations, i.e. under
pseudo-first order conditions where the decay can be
expressed by a single exponential function,
CF3 s CF3 exp yk) t .
w
x
w
x
Ž
.
0
The transient absorption of CF3 observed with a total
Ž
.
pressure of p ArqCF3I s22.0 mbar is shown in
Fig. 2. The self-reaction was found to take place
with a half-life of about 300 ms as shown as an inset
Fig. 3. Characteristic P-branch transitions of the product CF2O.
The insert shows the formation kinetics monitored at 1226.815
cmy1
.
Ž
.
upper curve in Fig. 2. In the presence of 0.1 mbar

(
)
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
141
Table 1
Experimental conditions and results on the reaction between CF3 and NO2 at 298 K
)
y1
3
p total
mbar
p NO2 mbar
10y4 k)
No. of exp.
1012 kbi cm moleculey1 sy1
Notes
Ž
.
Ž
.
Ž
. Ž
.
Ž
s
.
Ž
.
Ž .
4.0
11.0
22.0
a CF3 decay kinetics
a
b
b
0.00–0.20
0.00–0.15
0.00–0.20
0.09–7.65
0.24–5.60
0.16–7.82
9
6
8
15.3
15.4
15.2
Ž .
b CF2O formation kinetics
b
b
b
a
11.0
22.0
44.0
88.0
0.10–0.40
0.11–0.40
0.10–0.35
0.16–0.35
2.90–7.00
5.00–11.0
7.00–14.0
14.0–20.0
7
8
6
5
5.10
8.43
11.5
13.6
Ž .
5.5
11.0
22.0
44.0
c FNO formation kinetics
a
b
b
a
0.10–0.28
0.04–0.32
0.10–0.41
0.10–0.42
1.47–2.17
2.30–4.72
3.99–7.71
4.61–11.8
8
8
15
9
1.74
3.67
6.03
9.22
)
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
p total s p Ar qp CF3I qp NO2 and p total rp CF3I s4.4.
a Estimated uncertainties of kbi are in the range 10–15%.
b Estimated uncertainties of kbi are in the range 5–10%.
)
)
w
x
Ž
.
Ž
.
15 . An example of the formation kinetics of CF2O
sion from the excited products CF2O and FNO
w
x
is shown as an inset in Fig. 3. The time profiles of
CF2O absorption could be reproduced within the
experimental signal-to-noise ratio by curve fitting of
have been observed 11 .
2.4. Branching ratios of Õarious product channels
Ž
.
a simple pseudo-first order expression, A CF2O s
)
w
Ž
.x
Ž .
Amax 1yexp yk t . The apparent bimolecular rate
The yield of CF2O produced via reaction 2a was
derived from the product spectrum shown in Fig. 3
using the line strengths obtained from the reference
spectrum 15 . The yield of CF2O was found to be
nearly as high as the yield of CF3 produced under
constants were derived from the slope of k) versus
Ž
.
p NO2 at constant total pressure. The experimental
conditions and results summarized in Table 1b show
a strong pressure dependence of the apparent bi-
molecular rate constant. Similar results obtained by
studies of FNO formation kinetics are summarised in
Table 1c. The difference in rate constants of CF3
decay and product formation may be explained by
relaxation of vibrationally excited products:
)
w
x
w
x
Ž
the same experimental conditions, CF2O max s 1.6
CF qNO ™ CF O ) q FNO
,
2a’
Ž
.
Ž
.
Ž
.
3
2
2
)
CF O
qM ™ CF O qM ,
3
Ž
. Ž
.
Ž
.
Ž .
2
2
)
FNO
qM ™ FNO qM .
4
Ž .
Ž
. Ž
.
Ž
.
The rates of the vibrational relaxation were found to
increase with increasing pressures of the bath gas as
shown in Fig. 4. At increasing bath gas pressures,
the rate constants for the formation of CF2O and
FNO approach the rate constant for the decay of
CF3. This interpretation is supported by recent exper-
Fig. 4. Pressure dependence of the formation kinetics of FNO
Ž
.
Ž
.
circles and CF2O squares compared with the decay kinetics of
Ž .
Ž
.
imental studies of reaction 2 where infrared emis-
CF3 triangles .

(
)
142
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
Table 2
Enthalpies of formation at 298 K calculated at the G2 and G2MP2 levels
Ž
.
Ž
.
w
x Ž
.
Molecular system
G2MP2 kcalrmol
G2 kcalrmol
Exp. 25 kcalrmol
CF3
CF3O
CF2O
CF3NO2
CF3ONO
y115.0
y156.2
y150.5
y170.7
y186.7
y115.4
y156.6
y150.4
y112"1
y157"2
y153"2
14
.
Ž
.
Ž
.
Ž
.
"0.1 =10
molecules cmy3 s 0.95"0.04 =
Fock RHF self-consistent field SCF wavefunc-
tions were used for the closed-shell molecules
w
x
Ž .
CF3 o. Thus, reaction 2a is by far the predominant
reaction under the experimental conditions em-
Ž
.
whereas unrestricted Hartree–Fock UHF SCF
wavefunctions were used for the open-shell species.
The electron correlation energy was calculated using
second-order Møller–Plesset pertubation theory
ployed.
Ž
.
Ž .
The branching ratios of the reactions 2b and 2c
have also been investigated by monitoring the forma-
tion of NO at 1835.6 cmy1
Ž
. w
x
.
MP2 22 in the ‘‘frozen core’’ approximation with
RHF and UHF wavefunctions as references. The
CF3 qNO2 ™ CF3OqNO,
2b’
2c’
Ž
Ž
.
.
equilibrium geometries of the minimum energy
structures were optimized using analytical gradients
CF3 qNO2 ™ CF2OqFqNO.
)
Ž .
Ž .
at both U HF and U MP2 levels with the 6–31G
basis set. Total energies for open-shell species were
calculated using a projection method included in the
GAUSSIAN program. The effects of spin contamina-
tion were examined by the next highest spin compo-
nent to the UHF wavefunctions. The total energies of
the molecular systems were also examined using the
GAUSSIAN-2 G2 method 23 . Details of the theo-
retical calculations will be presented elsewhere. The
optimized geometries, rotational constants, vibra-
tional levels as well as the total electronic energies
were obtained at different levels of theory. Consider-
ing the larger molecules, CF3NO2 and CF3ONO,
calculations of the total energies were performed
using the less sophisticated G2MP2 approach 24 .
Tables 2 and 3 25 summarize the values of en-
thalpies of formation D Hf, of the reactants and
products as well as the reaction enthalpies D Hr, 298
calculated at G2 and G2MP2 levels. Fig. 5 shows the
energy diagram for the reaction proceeding via a
The relative yield of NO was found to be low,
x w
w
x
.
Ž
.
NO r CF3 0 s 1.0"0.5 %, indicating that the re-
Ž
Ž .
.
actions 2b and 2c are of minor importance. Re-
Ž
cently reaction 2d has been reported to take place
with a branching ratio of about 30% at total pres-
w
x
sures in the range of 300–600 Torr 10 . However, in
the present investigation we have not been able to
observe the infrared spectrum of the adduct, CF3NO2.
It seems most likely that the observation of CF3NO2
at higher pressures may be ascribed to stabilisation
of the adduct by collisions with the bath gas.
Ž
.
w x
3. Theoretical studies
w
x
w
x
3.1. Computational methodology and results.
All calculations were carried out using the
298
w
x
GAUSSIAN 94 program 21 . Restricted Hartree–
Table 3
Enthalpies of reaction at 298 K calculated at the G2 and G2MP2 levels
Ž
.
Ž
.
w
x Ž
.
Reaction
G2MP2 kcalrmol
G2 kcalrmol
Exp. 25 kcalrmol
CF3 qNO2 ™ CF2OqFNO
CF3 qNO2 ™ CF3NO2
CF3 qNO2 ™ CF3ONO
CF3ONO ™ F3OqNO
CF3O ™ CF2OqF
y64.6
y58.9
y74.9
49.0
y64.5
y65"1
24.9
25.0
23"4

(
)
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
143
time profiles of CF3 and products by time-resolved
infrared diode laser spectroscopy. The decay rate of
CF3 in the presence of NO2 was found to be inde-
pendent of bath gas pressures in the range of 4–22
Ž
.
mbar at 298 K. Our value of k2 s 1.5"0.2 =
10y11 cm3 moleculey1 sy1 may be compared with
the results of previous studies, 1011k2rcm3 mole-
culey1 sy1 s1.0 8 , 1.6 and 2.1 9 , 2.4 11 , 2.5
6,7 and 3.2 10 . The formation of CF2O and FNO
with an estimated yield of 95"4 % may be ex-
w x
w x
w x
w
x
w x
Ž
.
plained in terms of a four-center transition state, as
w x
originally proposed by Sugawara et al. 6 and re-
cently supported by the time-resolved infrared emis-
w
x
sion studies by Oum and Hancock 11 . The reac-
Ž
.
Ž .
tions 2b and 2c were found to be of minor
Ž
importance based on the observed NO-yield of 1"
0.5 %. Our result is in serious disagreement with a
.
previous investigation where the branching ratio of
Ž
.
Ž
.
reaction 2b was estimated to be 30"12 %, based
w x
on NO2 consumption monitored at 400 nm 9 . The
simultaneous formation of F-atoms and NO via reac-
Ž
.
tion 2c has recently been investigated and the
Ž
.
.
branching ratio was reported to be 1.5"0.2 % while
the contribution from reaction 2b was found to be
negligible 11 . The association channel 2d , which
leads to the formation of CF3NO2 has been reported
to take place with a relative yield of about 30% at
Ž
Fig. 5. Energy diagram and structure of the four-center transition
Ž
.
.
state TS for the strongly exothermic reaction CF3 qNO2 l
w
x
Ž .
)
)
)
Ž
Ž
.
Ž
.
CF3ONO
™ CF2O q FNO .
Ž
.
non-planar transition state TS . The molecular pa-
rameters have been used to calculate the rate con-
stants in terms of unimolecular rate theory. The
high-pressure limiting rate constants for the two
addition reactions were determined using one of the
versions of the statistical adiabatic channel model -
the maximum free energy method of Quack and Troe
w
x
total pressure in the range of 100–600 Torr 10 .
However, at total pressures in the range of 4–88
mbar at 298 K employed in the present investigation
we found no evidence for the formation of CF3NO2.
Thus, it appears that the branching ratio of reaction
Ž
.
2d increases with increasing bath gas pressures.
The complex reaction mechanism has been analysed
by detailed theoretical calculations of molecular
structures and energetics to obtain the potential en-
ergy surface shown in Fig. 5. The overall reaction
w
x
26 . The reaction enthalpies at 0 K and vibrational
y1
Ž
.
Ž
.
frequencies, n C–N s1512 cm
CF3NO2 and
y1
Ž
.
Ž
.
n C–O s1284 cm
CF3ONO , representing the
reaction coordinates, were used for evaluation of the
internal model parameters, with a ‘‘standard value’’
of arbs0.5 27 . High-pressure limiting values of
9.1x10y12 and 1.4x10y11 cm3 moleculey1 sy1 at
298 K were found for the CF3NO2 and CF3ONO
formation reactions.
)
)
)
Ž
.
Ž
.
Ž
.
CF3 q NO2 l CF3ONO ™ CF2O q FNO
is strongly exothermic and the product molecules are
produced in vibrationally excited states. The ob-
served pressure dependence of the formation kinetics
w
x
of CF2O and FNO is ascribed to relaxation of vibra-
)
Ž
.
tionally excited product molecules, CF2O
and
)
Ž
.
FNO which have also been detected by infrared
w
x
chemiluminescence 7,9,11 . The calculated high-
pressure limiting rate constant for the formation of
CF3ONO of 1.4=10y11 cm3 moleculey1 sy1 is
4. Discussion
Kinetics and mechanism of the reaction between
Ž
.
CF3 q NO2 have been studied by monitoring the
close to our experimental value of k2 s 1.5"0.2

(
)
144
P. Pagsberg et al.rChemical Physics Letters 286 1998 138–144
=10y11 cm3 moleculey1 sy1. Thus, the results of
our theoretical studies provide a strong support for
our interpretation of the experimental results.
w
w
x
Ž
.
13 C. Yamada, E.J. Hirota, J. Chem. Phys. 78 983 1703.
x
14 P. Pagsberg, A. Sillesen, J.T. Jodkowski, E. Ratajczak, Chem.
Ž
.
Phys. Lett. 249 1995 358.
w
w
x
15 C. Campy-Peyret, J.-M. Flaud, A. Goldman, F.J. Murcray,
R.D. Blatherwick, F.S. Bonomo, D.G. Murcray, C.P. Rins-
Ž
.
land, J. Mol. Spectrosc. 149 1991 481.
Acknowledgements
x
16 O.L. Rasmussen, E. Bjergbakke, B. Lynggaard, P. Pagsberg,
P. Kirkegaard, CHEMSIMUL Version 9306, Risø National
Laboratory, 1993.
We are most grateful to Dr. Claude Camy-Peyret
for sending a list of line positions and intensities for
w
w
w
x
17 C.E. Brown, J.J. Orlando, J. Reid, D.R. Smith, Chem. Phys.
w
x
the n4 band of carbonyl fluoride 15 . Numerical
calculations were carried out in the Wrocław Center
of Networking and Supercomputing.
Ž
.
Lett. 142 1987 213.
x
18 T. Ogawa, G.A. Carlson, G.C. Pimentel, J. Phys. Chem. 74
Ž
.
1970 2090.
x
Ž
.
19 K. Glanzer, M. Maier, J. Troe, J. Phys. Chem. 84 1980
¨
1681.
w
w
x
Ž
.
20 A.B. Vakhtin, Int. J. Chem. Kinet. 28 1996 443.
x
References
21 M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Rob, J.R. Cheeseman, T. Keith, G.A. Peters-
son, J.A. Montgomery, M.A. Al-Laham, V.G. Zakrzewski,
J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B. Stefanov, A.
Nanayakkara, M.Challacombe, C.Y. Peng, P.Y. Ayala, W.
Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R. Gom-
perts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. DeFrees, J.
Baker, J.P. Stewart, M. Head-Gordon, C. Gonzales, J.A.
w x
1
F. Caralp, R. Lesclaux, A.M. Dognon, Chem. Phys. Lett. 129
Ž
.
1986 433.
w x
2
Ž
.
R.C. Clemitshaw, J.R. Sodeau, J. Phys. Chem. 93 1989
3552.
w x
3
J.S. Francisco, Z. Li, I.H. Williams, Chem. Phys. Lett. 140
Ž
.
1987 531.
w x
4
Ž
.
J.S. Francisco, I.H. Williams, Int. J. Chem. Kinet. 20 1988
455.
Ž
.
Pople Eds. , in: Gaussian 94, Revision D.3, Gaussian Inc.,
w x
5
M.J. Rossi, J.R. Barker, D.M. Golden, J. Chem. Phys. 71
Pittsburgh PA, 1995.
Ž
.
1979 65.
w
w
x
x
Ž
.
22 C. Møller, M.S. Plesset, Phys. Rev. 46 1934 618.
23 L.A. Curtiss, K. Raghavachari, G.W. Trucks, J.A. Pople, J.
w x
6
K. Sugawara, T. Nakanaga, H. Takeo, C. Matsumura, J.
Phys. Chem. 93 1989 1894.
Ž
.
Ž
.
Chem. Phys. 94 1991 7221.
w x
Ž
.
7
w x
8
J.S. Francisco, Z. Li, Chem. Phys. Lett. 162 1989 528.
T.J. Bevilacqua, D.R. Hanson, C.J. Howard, J. Phys. Chem.
w
w
x
24 L.A. Curtiss, K. Raghavachari, J.A. Pople, J. Chem. Phys. 98
Ž
.
1993 1293.
Ž
.
97 1993 3750.
x
25 W.B. DeMore, S.P. Sander, D.M. Golden, R.F. Hampson,
M.J. Kurylo, C.J. Howard, A.R. Ravishankara, C.E. Kolb,
M.J. Molina, Chemical Kinetics and Photochemical Data for
Use in Stratospheric Modeling, Evaluation No. 11, JPL
Publication 94-26, 1994, p.194.
w x
9
J. Sehested, O.J. Nielsen, C.A. Rinaldi, S.I. Lane, J.C.
Ferrero, Int. J. Chem. Kinet. 28 1996 579.
Ž
.
w
w
w
x
Ž
.
Ž
10 A.B. Vakhtin, Int. J. Chem. Kinetics 29 1997 203.
x
.
11 K.W. Oum, G. Hancock, J. Phys. Chem. 101 1997 2634.
x
w
x
Ž
.
12 P. Pagsberg, E. Ratajczak, A. Sillesen, in: R.G. Compton, G.
26 M. Quack, J. Troe, Ber. Bunsenges. Physik. Chem. 81 1977
Ž
.
Hancock Eds. , Research in Chemical Kinetics, vol. 1,
Elsevier, Amsterdam, 1993, p. 65.
329.
w
x
Ž
.
27 C.J. Cobos, J. Troe, J. Chem. Phys. 83 1985 1010.