Kinetics and mechanisms for the reactions of CF3OCH3 and CF3OC(O)H with OH radicals using an environmental reaction chamber

Article abstract of DOI:10.1021/jp010137r

The atmospheric chemistry of CF₃OCH₃, a possible HCFCs/HFCs substitute was studied using a smog chamber/ FTIR technique. The ether was reacted with OH radicals prepared by photolysis of ozone in 100 Torr of an H₂O/O₂/He gas mixture in a 1 m³ temperature-controlled chamber. Using a relative rate method, k(OH + CF₃OCH₃) = (4.22 4 ± 0.84) × 10^-12 exp[(-1750 ± 350)/T] cm³ molecule^-1 s^-1 over the temperature range 268-308 K. The rate constant at 298 K was (1.19 ± 0.14) × 10^-14 cm³ molecule^-1 s^-1. The products of the OH radical initiated degradation of CF₃OCH₃ were investigated in 100-230 Torr of an O₂/He gas mixture at 298 K using in situ FTIR spectroscopy, k(OH + CF₃OC(O)H) = (1.68 ± 0.20) × 10^-14 cm³ molecule^-1 s^-1 at 298 K was determined. The major products of the oxidation of CF₃OC(O)H were COF₂ and CO₂. These results are discussed with respect to the atmospheric chemistry of CF₃OCH₃ and CF₃OC(O)H.

Full text of DOI:10.1021/jp010137r

10854
J. Phys. Chem. A 2001, 105, 10854-10859
Kinetics and Mechanisms for the Reactions of CF3OCH3 and CF3OC(O)H with OH
Radicals Using an Environmental Reaction Chamber
L. Chen,† S. Kutsuna,*,† K. Nohara,‡ K. Takeuchi,† and T. Ibusuki†
National Institute of AdVanced Industrial Science and Technology, 16-1 Onogawa, Tsukuba,
Ibaraki 305-8569, Japan, and Research Institute of InnoVatiVe Technology for the Earth,
16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan
ReceiVed: January 11, 2001; In Final Form: May 31, 2001
The atmospheric chemistry of CF3OCH3, a possible HCFCs/HFCs substitute was studied using a smog chamber/
FTIR technique. The ether was reacted with OH radicals prepared by photolysis of ozone in 100 Torr of an
H2O/O2/He gas mixture in a 1 m3 temperature-controlled chamber. Using a relative rate method, k(OH +
CF3OCH3) ) (4.22 ( 0.84) × 10-12 exp[(-1750 ( 350)/T] cm3 molecule-1 s-1 over the temperature range
268-308 K. The rate constant at 298 K was (1.19 ( 0.14) × 10-14 cm3 molecule-1 s-1. The products of the
OH radical initiated degradation of CF3OCH3 were investigated in 100-230 Torr of an O2/He gas mixture at
298 K using in situ FTIR spectroscopy. k(OH + CF3OC(O)H) ) (1.68 ( 0.20) × 10-14 cm3 molecule-1 s-1
at 298 K was determined. The major products of the oxidation of CF3OC(O)H were COF2 and CO2. These
results are discussed with respect to the atmospheric chemistry of CF3OCH3 and CF3OC(O)H.
Introduction
OCH3 by OH radicals in the atmosphere. CF3OC(O)H has a
strong absorption in the wavenumber region of 1000-1300
cm-1,5 which overlaps the atmospheric window of 770-1430
Hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons
(HFCs) have been used as substitutes for chlorofluorocarbons
(CFCs) because they react with OH radicals in the troposphere
and have shorter atmospheric lifetimes than CFCs. However,
HCFCs have finite ozone depletion potentials because they
contain chlorine atoms. HCFCs and HFCs have strong absorp-
tion bands in the terrestrial infrared radiation range and, thus,
contribute to global warming.1
CF3OCH3 is one of the hydrofluoroethers (HFEs) that has
been developed to replace HCFCs/HFCs in such applications
as refrigerant and cleaning agents. Because the boiling point of
CF3OCH3 is 249.1 K and the vapor pressure is 4328 Torr at
298 K, it will be released into the atmosphere when used. As
with other fluorinated species, CF3OCH3 has been shown to
have little impact on stratospheric ozone and only has less
impact on global warming if it and its products are more rapidly
removed than HCFCs and HFCs.2
cm-1 6
. Therefore, CF3OC(O)H has the potential to be a
greenhouse gas, and it is necessary to investigate the atmospheric
chemistry of CF3OC(O)H. An upper limit for the rate constant
k(CF3OC(O)H + OH) was estimated to be around 9.8 × 10-15
cm3 molecule-1 s-1 based on the rate constant, k(CF3OC(O)H
+ Cl) ) 9.8 × 10-15 cm3 molecule-1 s-1;5 however, some
fluorinated organic compounds react faster with OH radicals
than with Cl atoms such as CHF2CF3 (HFC-125) and CH2FCF3
(HFC-134a).7
In this study, we employed a relative rate method to measure
k1 in the ambient temperature range of 268-308 K, which can
be used without extrapolation to estimate the atmospheric
lifetime of CF3OCH3 by scaling to a CH3CCl3 lifetime at 272
K.8 We monitored the concentration-time profiles of CF3OCH3
and its products in the reaction with OH radicals quantitatively
using in situ FTIR spectroscopy at 298 K. k(CF3OC(O)H +
OH) at 298 K was determined according to a reported method
by Meagher et al.9 The products of the reaction of CF3OCH3
with OH radicals observed in this experiment were also
compared with the products reported by Wallington et al.5 from
the reaction of CF3OCH3 with Cl atoms.
CF3OCH3 will be removed from the atmosphere by a reaction
with OH radicals as reaction 1:
CF3OCH3 + OH f CF3OCH2 + H2O
(1)
The rate constant for the reaction of CF3OCH3 with OH radicals,
k1, has been measured by several researchers2-4 at g 296 K.
Values of k1 at room temperature and below are needed to
evaluate the atmospheric lifetime of CF3OCH3. Although it is
possible to derive rate constants k1 below 296 K from extrapola-
tion of higher temperature data using the Arrhenius expression,
it is more reliable to make measurement at lower temperature.4
The chlorine atoms initiated oxidation of CF3OCH3 has been
investigated by Wallington et al.5 From their results of the
chlorine initiated oxidation, trifluoromethyl formate, CF3OC(O)H,
will be a major primary product from the oxidation of CF3-
Experimental Section
All experiments were performed by using a 1 m3 stainless
steel cylindrical chamber with an inner diameter of 1.0 m
interfaced to a Bomem DA8 FTIR spectroscope with a White-
optical multiple refection mirror system. The optical path length
of the infrared beam was 54 m. The inside wall of the chamber
was coated with Teflon to minimize the wall effects in the
reactions. Two 1 kW Xe short-arc lamps (USHIO Co., Japan)
were positioned on the top of the chamber to initiate the
photochemical experiments. The UV light was cut by an optical
filter (>260 nm; SHIMA QUARTZ, Co., Japan). The illumi-
nated part of the chamber was 18% in volume. The gas mixtures
* To whom correspondence should be addressed.
† National Institute of Advanced Industrial Science and Technology.
‡ Research Institute of Innovative Technology for the Earth.
10.1021/jp010137r CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/09/2001

Reactions of CF3OCH3 and CF3OC(O)H
J. Phys. Chem. A, Vol. 105, No. 48, 2001 10855
were continuously stirred using a fan with an diameter of 15
cm and a speed of 200 rpm, which was attached to the inside
of the chamber. The chemicals were separately added to the
chamber and pumped away after each complete experiment. The
chamber temperature can be controlled over the range 233-
308 K to (1 K using a temperature-control system that consists
of two refrigerators, a heater, and a coolant flow controlling
system.
OH radicals were produced by UV photolysis of O3 in the
presence of water vapor in 100 Torr of He as illustrated in the
following reaction sequence:
O3 + hV f O(1D) + O2
O(1D) + H2O f 2OH
(2)
(3)
Figure 1. Loss of CF3OCH3 versus the reference compounds of CH4
and CH3CCl3 in the presence of OH radicals. Experiments were
performed at 298 K in 100 Torr of He.
An O3/O2 (5%) gas mixture, which was generated from pure
O2 with a silent-discharge ozone generator (ECEA-1000,
EBARA JITSUGYO, Japan) was used in the experiments. The
concentration of OH radicals produced in the chamber was
estimated to be the order of (1-10) × 1010 molecule cm-3 from
the decay of the concentration of CH4 because of reaction with
OH radicals in the chamber.
The kinetics of the reaction of CF3OCH3 with OH radicals
was measured using the relative rate method, which has been
described in several previous publications.10,11 Absolute rate
constants for the reaction of CH4 and CH3CCl3 have been
measured accurately, and the values of these two rate constants
are comparable to that of CF3OCH3.7 Therefore, CH4 and CH3-
CCl3 were used as the reference compounds in this study. Cl
and CF3O radicals are produced in the reaction of OH with CH3-
CCl3 and OH with CF3OC(O)H, respectively. Both of these
species may contribute to the loss of CF3OCH3, CH3CCl3, and
CH4, but the relative rate plots obtained in this work are quite
linear suggesting that Cl and CF3O formed in secondary
reactions do not complicate the kinetic studies.
All experiments were performed in the temperature range of
268-308 K in 100 Torr He. The typical initial concentrations
of CF3OCH3, CH4 (or CH3CCl3), H2O, O3 , and O2 were 1.3 ×
1014, 2.5 × 1014 (or 1.2 × 1014), 1.3 × 1017, 1.2 × 1016, and
2.4 × 1017 molecule cm-3 in 100 Torr of He, respectively. Each
experiment was performed with either CH4 or CH3CCl3 as the
reference compounds. The loss of CF3OCH3, CH4, and CH3-
CCl3 was measured with the FTIR spectrometer at a resolution
of 0.5 cm-1. The concentrations of CF3OCH3, CH4, and CH3-
CCl3 were determined with their absorption cross sections (base
10) of 1.11 × 10-18, 2.04 × 10-20, and 6.17 × 10-19 cm2
molecule-1 at 1167, 3149, and 728 cm-1, respectively.
Under various combinations of gas mixtures and with or
without irradiation, the concentration of CF3OCH3, CH4, and
CH3CCl3 were monitored for 6 h. A linear least-squares analysis
of the concentrations of CF3OCH3, CH4, and CH3CCl3 gives
<2% changes of these compounds with irradiation condition
at 308 K. The changes in concentrations of CF3OCH3, CH4,
and CH3CCl3 were obtained to be <2% in the presence of ozone
with or without irradiation conditions for 6 h at 308 K,
respectively. The heterogeneous reactions of CF3OCH3, CH4,
and CH3CCl3 with H2O were not observed in the presence of
H2O without irradiation conditions at 298 K. Therefore, the
losses of CF3OCH3, CH4, and CH3CCl3 via photolysis or dark
chemistry were confirmed to be insignificant in this chamber.
Estimation of the rate constant for the reaction of CF3OC(O)H
with OH radicals was performed in three runs. The initial
concentrations were CF3OCH3 (2.5 × 1014 molecule cm-3)/
H2O (1.3 × 1017 molecule cm-3) at 298 K in 100 Torr of He.
In these runs, an O3/O2 (5%) gas mixture was continuously
added to the system at a flow rate of 0.3-0.5 mL min-1 to
maintain a concentration of O3 between 2 × 1015 and 3 × 1015
molecule cm-3 during the UV irradiation. The UV irradiation
was continued for 50-70 h and the total pressure increased to
160-230 Torr at the end of the runs. The fractional loss of
CF3OCH3 was 90% for a 50-70 h photolysis. The primary
oxidation product CF3OC(O)H increased initially with time and
was subsequently removed in secondary reactions forming COF2
which was identified and quantified by IR spectra. The
absorption cross section, ꢀCOF2, of 6.3 × 10-19 cm2 molecule-1
at 1928 cm-1 was obtained from an artificial COF2/N2 standard.
The reagents used were CH4 (99.7%) and CH3CCl3 (99%;
both from GL Science, Japan), COF2/N2 standard (85%) and
He (99.99995%; both from Takachiho Chemical Industry, Co.,
Japan), and pure O2 (99.99%, Nippon Sanso, Corp., Japan). CF3-
OCH3 (99%) was obtained from the Research Institute of
Innovative Technology for the Earth (RITE).
Results and Discussion
Kinetics of the Reaction of CF3OCH3 with OH Radicals.
The results obtained at 298 K based on the two reference
compounds of CH4 and CH3CCl3 are shown in Figure 1, and
eq 4 is used to determine the rate constant:10,11
[CF3OCH3]0
[CF3OCH3]t
k
[reference]0
[reference]t
ln
)
1 ln
(4)
(
)
(
)
kr
where [CF3OCH3]0 and [reference]0 represent the initial con-
centrations of the reactant and reference compounds, [CF3-
OCH3]t and [reference]t represent the concentrations of reactant
and reference compounds at the reaction time t, and k1 and kr
are the rate constants for the reaction of OH radicals with CF3-
OCH3 and reference compounds, respectively. The plots of ln-
([CF3OCH3]0/[CF3OCH3]t) versus ln([Reference]0/[Reference]t)
gave straight lines, which intersected the origin for the two
references of CH4 and CH3CCl3. The slopes from the linear
least-squares analysis of the data in Figure 1 give k1/kr ) 1.91
( 0.12 and 1.18 ( 0.08 for CH4 and CH3CCl3, respectively.
The errors reported are (2 standard deviation and represent
precision only. The k1 (298 K) values were estimated to be (1.20
( 0.14) × 10-14 and (1.18 ( 0.12) × 10-14 cm3 molecule-1
s-1 from the rate constants of the reactions of CH4 and CH3-

10856 J. Phys. Chem. A, Vol. 105, No. 48, 2001
Chen et al.
TABLE 1: Measured Rate Constant, k1/kr and k1, as a
Function of Temperature
a
k1/kr
k1
T (K)
CH4
CH3CCl3
CH4
CH3CCl3
268
278
281
288
290
298
308
1.07 ( 0.12
1.22 ( 0.07
0.593(0.07
1.85 ( 0.23
2.08 ( 0.17
0.764(0.10
1.07(0.09
0.875(0.05
1.30 ( 0.26
1.18 ( 0.08
1.11 ( 0.06
1.12(0.22
1.18(0.08
1.30(0.07
1.91 ( 0.12
2.01 ( 0.28
1.20(.0.08
1.55(0.22
a The unit is 10-14 cm3 molecule-1 s-1
.
CCl3 with OH radicals at 298 K, 6.3 × 10-15 and 1.0 × 10-14
,
respectively,7 and the ratio of k1/kr were determined using eq
4. The values of k1 (298 K) obtained using CH4 and CH3CCl3
as reference compounds were the same within experimental
uncertainty. We estimate that the potential systematic errors
associated with uncertainties in the reference rate constants add
a further 10% uncertainty in the values of k1. The values at 298
K of k1 of 1.30 × 10-14 and (1.0 ( 0.07) × 10-14 cm3
molecule-1 s-1 reported by Orkin et al.3 and DeMore et al.,4
respectively, are in agreement with the data obtained in this
study within experimental uncertainty. The absolute rate mea-
surement value at 296 K of k1 of (2.14 ( 0.15) × 10-14 cm3
molecule-1 s-1 measured at a total pressure of 35 Torr by Zhang
et al.2 is a factor of about 2 higher than the present measurement,
possibly because the presence of impurities in the sample.
The values of k1 at different temperatures were estimated from
k1/kr and the rate constant of k(CH4) ) 2.45 × 10-12 exp[-
(1775 ( 100)/T] and k(CH3CCl3) ) 1.8 × 10-12 exp[-(1550
( 150)/T] cm3 molecule-1 s-1.7 The result of k1 obtained are
summarized in Table 1. The temperature dependence of k1 is
shown in Figure 2. The rate constant Arrhenius expression of
k1 was derived to be (4.22 ( 0.84) × 10-12 exp[-(1750 ( 350)/
T] cm3 molecule-1 s-1 from a linear least-squares fits to the
data in Figure 2. The uncertainty in the Arrhenius expression
was calculated by considering the random errors of k1.
Figure 2. Arrhenius plot of kinetics data obtained by a relative rate
method for CF3OCH3 reaction with OH radicals.
The value of k1 calculated at 272 K, the average troposphere
temperature,8 from the Arrhenius expression is used to estimate
the atmospheric lifetime and GWP of CF3OCH3. The atmo-
spheric lifetime of CF3OCH3 with respect to loss by reaction
with OH radicals was estimated from eq 5 to be 4.9 years:
k
CH3CCl3
τCF OCH
)
τCH CCl ,
3
(5)
3
3
3
k
CF3OCH3
where τCF3OCH3 and τCH3CCl3 represent the tropospheric lifetime
of CF3OCH3 and CH3CCl3 through the reaction with OH
radicals. τCH3CCl3 was estimated to be 5.5 years by Spivakovaky
et al.8 kCF3OCH3 ) 6.8 × 10-15 and kCH3CCl3 ) 6.0 × 10-15 cm3
molecule-1 s-1 represent the reaction rate constants for the
reactions of CF3OCH3 and CH3CCl3 with OH radicals at 272
K.
Estimate of the Kinetics of the Reaction of CF3OC(O)H
with OH Radicals. Figure 3 shows IR spectra obtained from
the experiment of the OH radicals initiated photooxidation of
CF3OCH3.
The observed products of the reaction were CF3OC(O)H,
COF2, CO2, and traces of CO. CF3OC(O)H was identified from
its reported spectrum.5 Therefore, making the reasonable as-
sumption that CF3OC(O)H and COF2 account for all of the
reaction products containing both carbon and fluorine, then the
absorption cross-section of CF3OC(O)H, ꢀCF3OC(O)H, was deter-
Figure 3. IR spectra observed before (A) and after (B) 15 h irradiation
of a gas mixture of CF3OCH3 (2.5 × 1014)/H2O (2.5 × 1017 molecule
cm-3) at 298 K in 100 Torr of He. O3/O2 (5%) gas mixture was
introduced into the chamber at a flow of 0.5 mL/min continuously
during the UV irradiation. (C) A spectrum of B after subtraction of
unreacted CF3OCH3, (D) a spectrum of C after subtraction of COF2,
and (E) a reference spectrum of COF2.
mined from the material balance ∆CF3OC(O)H]t ) ∆[CF3-
OCH3]t - [COF2]t, where ∆[CF3OCH3]t is ([CF3OCH3]0 -
[CF3OCH3]t), in the initial 4.2 h period. On the basis of a blank
experiment of COF2 in the system, about 12% of the COF2 will
be lost from the chamber by wall reaction in the initial 4.2 h
period. Because the yield of COF2 after a 4.2 h photolysis
experiment is only around 10%, the error in the calculation of

Reactions of CF3OCH3 and CF3OC(O)H
J. Phys. Chem. A, Vol. 105, No. 48, 2001 10857
Figure 5. Plot of [CF3OC(O)H]t/[CF3OCH3]0 versus [CF3OCH3]t/[CF3-
OCH3]0. The curve is a fit of eq 7 in the text to the data of [CF3OC-
(O)H]t/[CF3OCH3]0.
tainties in k1 add a further 20% uncertainty range for k6. Thus,
CF3OC(O)H reacts faster with OH radicals than with Cl atoms
by a factor 1.7 at 298 K.5
Figure 4. Plot of the [CF3OC(O)H]t and [COF2]t versus ∆[CF3OCH3]t.
The data were obtained from the experiment presented in Figure 3.
Long period photoillumination (50-70 h) was carried out in
which the primary product CF3OC(O)H was reacted in second-
ary reaction. These conditions were needed to determine the
parameters R and k6/k1 accurately by fitting the plot of
[CF3OC(O)H]t/[CF3OCH3]0 versus ∆[CF3OCH3]t/[CF3OCH3]0.
Because such relatively long photolysis reactions might be
accompanied by some artifact reactions, the dark reactions of
CF3OC(O)H and COF2 were examined in the experimental
system. Irradiation of CF3OCH3 (2.5 × 1014)/O3 (1.25 × 1016)/
O2 (2.4 × 1017)/H2O (1.3 × 1017 molecule cm-3) mixtures was
carried out at 298 K in 100 Torr total pressure of He. After
CF3OC(O)H and COF2 were produced by UV irradiation for 6
h, the gas mixture was stored in the chamber for 24 h without
further UV irradiation. The decays of CF3OC(O)H and COF2
were monitored and have apparent first-order rate constants of
(5.1 ( 1.0) × 10-7 and (7.8 ( 0.2) × 10-6 s-1, respectively.
The dark reaction of CF3OC(O)H was considered to be
insignificant in the kinetic experiments, in which small amounts
of COF2 were shown to react with water on the wall of the
chamber heterogeneously. In addition, photolysis of CF3OC(O)H
because of UV irradiation (>260 nm) was also shown to be
negligible.
ꢀCF3OC(O)H due to correction for wall loss of COF2 will be
minimal. A value for ꢀCF3OC(O)H was obtained ) 4.0 × 10-19
cm2 molecule-1 at the absorption band of 1814 cm-1 with an
estimated error of 2%.
Plots of the [CF3OC(O)H]t and [COF2]t against loss of
CF3OCH3 in Figure 4 indicate that CF3OC(O)H and COF2 are
the primary and secondary products against loss of CF3OCH3,
respectively. CF3OC(O)H is an intermediate in the consecutive
reactions 1 and 6:
CF3OCH3 + OH fff RCF3OC(O)H + products (1)
CF3OC(O)H + OH f products
(6)
The parameter R is the yield of CF3OC(O)H from CF3OCH3 (0
e R e 1). The following eq 7 can be derived:9
R
6
1
y )
(1 - x)[(1 - x){k /k - 1} - 1]
(7)
k6
k1
1 -
∆[CF3OCH3]t
Because CF3OC(O)H reacts with OH radicals via H abstrac-
tion from the -OC(O)H group, the value of k6 should decrease
with a decrease in temperature. A lower limit for the lifetime
of CF3OC(O)H of 2.0 years can be estimated from k6 (298 K)
using eq 5. This lifetime is long enough for CF3OC(O)H to
contribute to global warming because CF3OC(O)H shows a
x )
y )
[CF3OCH3]0
[CF3OC(O)H]t
[CF3OCH3]0
strong absorption in the wavenumber region 1000-1300 cm-1
.
where x and y are the ratio of CF3OCH3 reacted and CF3OC(O)H
formed to the initial concentration of CF3OCH3 at reaction time
t and k1 and k6 are the reaction rate constants of reactions 1 and
6, respectively. A fit of the eq 7 to the data in Figure 5 of
[CF3OC(O)H]t/[CF3OCH3]0 versus ∆[CF3OCH3]t/[CF3OCH3]0
gives the values of R and k6/k1. The obtained values of R and
k6/k1 were 1.04 ( 0.01 and 1.40 ( 0.04, respectively, as the
simple average of three runs. A value of R ) 1 shows that the
yield of CF3OC(O)H is close to unity for the reaction of
CF3OCH3 with OH radicals. Using k1 (298 K) ) (1.2 ( 0.14)
× 10-14 cm3 molecule-1 s-1 from this work, k6 (298 K) was
estimated to be (1.68 ( 0.20) × 10-14 cm3 molecule-1 s-1. We
estimate that potential systematic errors associated with uncer-
However, CF3OC(O)H is expected to hydrolyze easily by
analogy with esters reference. Further study on dissolution into
clouds and wet deposition of CF3OC(O)H is important in order
to evaluate the contribution of CF3OCH3 to global warming.
Mechanism of the Reaction of CF3OCH3 with OH Radi-
cals. According to the product distribution yields, the following
mechanism is proposed for the reaction of CF3OCH3 with OH
radicals under the conditions of the present studies:
CF3OCH3 + OH f CF3OCH2 + H2O
(1)
(8)
CF3OCH2 + O2 + M f CF3OCH2O2 + M

10858 J. Phys. Chem. A, Vol. 105, No. 48, 2001
Chen et al.
The peroxy radical formed in reaction 8 can react with HO2
and itself:5,3,14
or H2O to produce CF3OH as in eqs 16-18 in the gas phase or
on the chamber walls:16-18
CF3O + CF3OCH3 f CF3OH + OH
CF3O + CF3OC(O)H f CF3OH + OH
CF3O + H2O f CF3OH + OH
(16)
(17)
(18)
CF3OCH2O2 + HO2 f CF3OCH2OOH + O2 (9a)
CF3OCH2O2 + HO2 f CF3OC(O)H + H2O + O2
2CF3OCH2O2 f 2CF3OCH2O + O2
(9b)
(10a)
2CF3OCH2O2 f CF3OC(O)H + CF3OCH2OH + O2 (10b)
Because of the high concentration of H2O of 1.3 × 1017
molecule cm-3 in the chamber, the fate of CF3O radicals is
considered to react with H2O in this reaction system. CF3OH
formed from a reactions (16-18) can rapidly decompose
heterogeneously to COF2 and HF on the chamber walls:16,19,20
However, CF3OCH2OOH and CF3OCH2OH were not observed.
The absence of CF3OCH2OOH was supported by the measure-
ment of k9a/(k9a + k9b) ) 0.2 by Wallington et al.5 In addition,
hydroperoxides and alcohols are generally very reactive toward
OH radicals, k(OH + CH3OOH) ) 7.4 × 10-12, k(OH +
CH3OH) ) 8.9 × 10-13, and k(OH + C2H5OH) ) 3.18 × 10-12
cm3 molecule-1 s-1 at 298 K,7 although the presence of fluorine
in CF3OCH2OOH and CF3OCH2OH would be expected to
reduce their reactivity. Therefore, it is possible that CF3OCH2OOH
and CF3OCH2OH would react with OH radicals to regenerate
CF3OCH2O2 and CF3OCH2O radicals in the chamber.
CF3OH f COF2 + HF
(19)
COF2 is a secondary product of the reaction of CF3OCH3 with
OH radicals via CF3OC(O)H and CF3OH. This mechanism is
consistent with the observation of COF2 in Figure 4.
The CF3OCH2O radical produced in reaction 10a can react
Summary
with O2 or decompose by carbon-oxygen bond fission:
The rate constant of the reaction of CF3OCH3 with OH
radicals is k1(T) ) (4.22 ( 0.84) × 10-12 exp[(-1750 ( 350)/
T] cm3 molecule-1 s-1, and the atmospheric lifetime is estimated
to be 4.9 years using the rate constant at 272 K. The rate constant
of the reaction of CF3OC(O)H with OH radicals at 298 K is
(1.68 ( 0.20) × 10-14 cm3 molecule-1 s-1, and the lower limit
for the lifetime of CF3OC(O)H is estimated to be 2.0 years in
this study. CF3OCH3 has less impact on global warming because
it and its oxidation products are more rapidly removed from
the atmosphere than typical refrigerants of HCFC-22 and HFC-
134a.7
In the atmosphere, the CF3OCH2 radical produced in the
reaction of CF3OCH3 with OH radicals will rapidly react with
O2 to form the peroxy radical CF3OCH2O2. The CF3OCH2O2
radicals will react with HO2, NO2, NO, and other peroxy
radicals. The reaction of the CF3OCH2O2 radical with HO2 has
been known to produce CF3OC(O)H within a yield of 80 (
11%.5 CF3OCH2O2NO2 formed from the reaction of CF3OCH2O2
with NO2 will decompose to CF3OCH2O2 and NO2 due to its
thermally instability. The reaction of CF3OCH2O2 with NO will
produce the CF3OCH2O radical. The fate of the CF3OCH2O
radical has been shown to form CF3OC(O)H by reaction with
O2. CF3OC(O)H will also be oxidized by OH radicals in the
atmosphere. The CF3OC(O) radical produced from the reaction
of CF3OC(O)H with OH radicals is known to react with O2 to
form CF3OC(O)O2 or decompose to the CF3O radical and CO
in this study. The formation of the CF3OC(O)O2 radical is likely
to be the main processes in the atmospheric condition. As with
other peroxy radicals, the CF3OC(O)O2 radicals also will react
with HO2, NO2, NO, and other peroxy radicals in the atmo-
sphere, but the atmospheric chemistry of the CF3OC(O)O2
radicals has not been known, it is necessary to study these
reactions in detail. On the other hand, CF3OC(O)H is predicted
to be hydrolyze easily; therefore, it is important to investigate
the removal of CF3OC(O)H from the atmosphere by dissolution
into clouds and wet deposition.
CF3OCH2O + O2 f CF3OC(O)H + HO2
(11a)
CF3OCH2O + M f CF3O + HC(O)H + M (11b)
Because the yield of CF3OC(O)H is close to unity, decomposi-
tion via reaction 11b was considered to be insignificant in this
reaction system. This result is consistent with the measurement
by Wallington et al.,5 who reported the yield of CF3OC(O)H
as unity over the partial pressure range 6-700 Torr of O2.5
Variation of the pressure in this work from 100 to 230 Torr by
adding O3/O2 to the gas mixtures did not impact the yield of
CF3OC(O)H supporting this observation.
CF3OC(O)H formed can react with OH radicals in the
chamber in reaction 6:
CF3OC(O)H + OH f CF3OC(O) + H2O
(6)
CF3OC(O) radicals will rapidly react with O2 to produce
CF3OC(O)O2 radicals, which will produce CF3OC(O)O radicals
via reaction with peroxy radicals (RO2):
CF3OC(O) + O2 + M f CF3OC(O)O2 + M (12)
CF3OC(O)O2 + RO2 f CF3OC(O)O + RO + O2
(13)
The small amounts of CO seen in the IR spectra of the products
in Figure 3 shows that CF3OC(O) radicals also will decompose
to CF3O radicals and CO in competition with reaction 12:
CF3OC(O) f CF3O + CO
(14)
The CF3OC(O)O radical may decompose to CF3O radical and
CO2:15
Acknowledgment. The authors are grateful to Professor
Hiroshi Bandow at Osaka Prefecture University for his helpful
suggestions. This research is supported in part by the New
Energy and Industrial Technology Development Organization
CF3OC(O)O + M f CF3O + CO2
(15)
A CF3O radical is expected to react with CF3OCH3, CF3OC(O)H,

Reactions of CF3OCH3 and CF3OC(O)H
J. Phys. Chem. A, Vol. 105, No. 48, 2001 10859
(9) Meagher, R. J.; Mcintosh, M. E.; Hurley, M. D.; Wallington, T. J.
Int. J. Chem. Kinet. 1997, 29, 619.
(10) Atkinson, R. Chem. ReV. 1986, 86, 69.
(NEDO) and the Research Institute of Innovative Technology
for the Earth (RITE).
(11) Finlayson-Pitts, B. J.; Hernandez, S. K.; Berko, H. N. J. Phys. Chem.
1993, 97, 1172.
References and Notes
(12) Ko, M.; Shia, R.-L.; Sze, N.-D.; Magid, H.; Bray, R. G. J. Geophys.
Res. 1999, 104, 8173.
(13) Lightfoot, P. D.; Cox, R. A.; Crowley, J. N.; Destriau, M.; Hayman,
G. D.; Jenkin, M. E.; Moortgat, G. K.; Zabel, F. Atmos. EnViron. 1992,
26A, 1085.
(1) UNEP/WMO. Global Ozone Research and Monitoring Project;
Scientific Assessment of Ozone Depletion; Report No. 37; 1994.
(2) Zhang, Z.; Saini, R. D.; Kuryol, M. J.; Huie, R. E. J. Phys. Chem.
1992, 96, 9301.
(3) Orkin, W. L.; Khamaganov, V. G.; Guschin, A. G.; Huie, R. E.;
Kurylo, M. J. The 13th International Symposium on Gas Kinetics; The
University College, Dublin: Dublin, Ireland, 1994.
(14) Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92,
667.
(4) Hsu, K.-J.; DeMore, W. B. J. Phys. Chem. 1995, 99, 11141.
(5) Christensen, L. K.; Wallington, T. J.; Guschin, A.; Hurley, M. D.
J. Phys. Chem. A 1999, 103, 4202.
(6) Hanel, R. A.; Conrath, B. J.; Kunde, V. G.; Prabhakara, C.; Revah,
I.; Salomonson, V. V.; Wolford, G. J. Geophys. Res. 1972, 77, 2629.
(7) 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. JPL Publ. 1997, 97-4.
(8) Spivakovsky, C. M.; Logan, J. A.; Montzka, S. A.; Balkanski, Y.
J.; Foreman-Fowler, M.; Jones, D. B. A.; Horowitz, L. W.; Fusco, A. C.;
Brenninkmeijer, C. A. M.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B. J.
Geophys. Res. 2000, 105, 8931.
(15) Atkinson, R. J. Phys. Chem. Ref. Data 1994, 20, 459.
(16) Sehested, J.; Wallington, T. J. EnViron. Sci. Technol. 1993, 27,
146.
(17) Wallington, T. J.; Schneider, W. F. EnViron. Sci. Technol. 1994,
28, 1198.
(18) Turnipseed, A. A.; Barone, S. B.; Jensen, N. R.; Hanson, D. R.;
Howard, C. J.; Ravishankara, A. R. J. Phys. Chem. 1995, 99, 6000.
(19) Huey, L. G.; Hanson, D. R.; Lovejoy, E. R. J. Geophys. Res. 1995,
100, 18771.
(20) Lovejoy, E. R.; Huey, L. G.; Hanson, D. R. J. Geophys. Res. 1995,
100, 18775.