Atmospheric chemistry of CF3CF double bond CF2: Kinetics and mechanism of its reactions with OH radicals, Cl atoms, and ozone

Article abstract of DOI:10.1021/jp000498r

Smog chamber/FTIR techniques were used to study the OH radical, the Cl atom, and ozone initiated oxidation of CF₃CF double bond CF₂ in 700 Torr of air at 296 K. Relative rate methods were used to measure k(OH+CF₃CF double bond CF₂) = (2.4±0.3)×10^-12 and k(Cl+CF₃CF double bond CF₂) = (2.7±0.3)×10^-11; absolute techniques were used to derive an upper limit of k(O₃+CF₃CF double bond CF₂)<3×10^-21 cm³ molecule^-1 s^-1. OH radical and Cl atom-initiated atmospheric oxidation of CF₃CF double bond F₂ gives COF₂ and CF₃C(O)F in molar yields of 100%. The atmospheric lifetime of CF₃CF double bond CF₂ is approximately 9 days with degradation proceeding via reaction with OH radicals to give trifluoroacetic acid in a molar yield of 100%. Results are discussed with respect to previous measurements of k(OH+CF₃CF double bond CF₂) and k(O₃+CF₃CF double bond CF₂) and the potential importance of CF₃CF double bond CF₂ as a source of trifluoroacetic acid.

Full text of DOI:10.1021/jp000498r

J. Phys. Chem. A 2000, 104, 7255-7260
7255
Atmospheric Chemistry of CF3CF)CF2: Kinetics and Mechanism of Its Reactions with OH
Radicals, Cl Atoms, and Ozone
M. Mashino, Y. Ninomiya, and M. Kawasaki
Department of Molecular Engineering, Kyoto UniVersity, Kyoto 606-8501, Japan
T. J. Wallington* and M. D. Hurley
Ford Research Laboratory, SRL-3083, Ford Motor Company, P.O. Box 2053, Dearborn, Michigan 48121-2053
ReceiVed: February 7, 2000; In Final Form: April 27, 2000
Smog chamber/FTIR techniques were used to study the OH radical, the Cl atom, and ozone initiated oxidation
of CF3CFdCF2 in 700 Torr of air at 296 K. Relative rate methods were used to measure k(OH + CF3CFd
CF2) ) (2.4 ( 0.3) × 10-12 and k(Cl + CF3CFdCF2) ) (2.7 ( 0.3) × 10-11; absolute techniques were used
to derive an upper limit of k(O3 + CF3CFdCF2) < 3 × 10-21 cm3 molecule-1 s-1. OH radical and Cl atom-
initiated atmospheric oxidation of CF3CFdCF2 gives COF2 and CF3C(O)F in molar yields of 100%. The
atmospheric lifetime of CF3CFdCF2 is approximately 9 days with degradation proceeding via reaction with
OH radicals to give trifluoroacetic acid in a molar yield of 100%. Results are discussed with respect to
previous measurements of k(OH + CF3CFdCF2) and k(O3 + CF3CFdCF2) and the potential importance of
CF3CFdCF2 as a source of trifluoroacetic acid.
1. Introduction
CF3CFdCF2, it seems reasonable to speculate that a similar
mechanism may be operative for CF3CFdCF2. If so, the
atmospheric oxidation of CF3CFdCF2 would yield oxalyl
fluoride (FC(O)C(O)F) instead of CF3C(O)OH. In the absence
of an experimental study of the atmospheric oxidation mech-
anism of CF3CFdCF2, it is difficult to assess its potential
contribution to the global CF3C(O)OH budget.
At the present time, the atmospheric oxidation mechanism
of CF3CFdCF2 is unclear. To remedy this situation, we have
used FTIR-smog chamber techniques to study the kinetics and
products of reactions 1, 2, and 3 at total pressures of 10-700
Torr of air diluent:
Trifluoroacetic acid, CF3COOH, has been detected in surface
waters (oceans, rivers, and lakes) and in fog, snow, and rainwater
samples around the globe1-5 and is a ubiquitous component of
the hydrosphere. Interestingly, while it is clearly established that
there is a substantial environmental burden of CF3C(O)OH
(estimated as 107-108 tonnes6), the sources of this compound
are unclear. No significant natural sources of trifluoroacetic acid
have been identified. Several man-made compounds, that is,
the anesthetics isoflurane (CF3CHClOCHF2) and halothane
(CF3CHClBr) and the CFC replacements HFC-134a (CF3CFH2)
and HCFC-123 (CF3CHCl2), are emitted into the environment
and produce CF3C(O)OH. However, the magnitude of these
industrial sources is several orders of magnitude too small to
account for the levels of CF3C(O)OH observed in the environ-
ment.6 At the present time, there is a significant research effort
to identify possible additional industrial and natural sources of
CF3C(O)OH and its precursors in the environment.
OH + CF3CFdCF2 f products
Cl + CF3CFdCF2 f products
O3 + CF3CFdCF2 f products
(1)
(2)
(3)
2. Experimental Section
Perfluorinated polymers are used widely in plastics, elas-
tomers, and water repellants. The global annual production of
these polymers was 40,000 tonnes in 1988.3 As noted by Jordan
and Frank,3 hexafluoropropene (CF3CFdCF2) is a pyrolysis
product of perfluoroalkyl ethers,7 polytetrafluoroethene (PTFE),8
and polyperfluoroethenepropene (FEP).8 It has been suggested
that incineration of municipal waste containing perfluorinated
polymers may result in the emission of CF3CFdCF2 into the
atmosphere and that this CF3CFdCF2 may undergo atmospheric
oxidation to give CF3C(O)OH.3
Studies of the atmospheric oxidation mechanism of
CF3OCFdCF2 have shown that HF elimination from the initially
formed excited [CF3OCF(•)CF2OH]* adduct is significant,9,10
leading to formation of oxalyl fluoride (FC(O)C(O)F). In light
of the obvious structural similarities between CF3OCFdCF2 and
All experiments were performed in a 140-L Pyrex reactor
interfaced to a Mattson Sirus 100 FTIR spectrometer.11 The
optical path length of the infrared beam was 27.7 m. The reactor
was surrounded by 22 fluorescent blacklamps (GE F15T8-BL),
which were used to photochemically initiate the experim-
ents. The oxidation of CF3CFdCF2 was initiated by reac-
tion with either OH radicals, Cl atoms, or O3 in 700 Torr total
pressure of O2/N2 diluent at 295 ( 2 K. Hydroxyl radicals
were generated by the UV irradiation of methylnitrite/NO/air
mixtures:
CH3ONO + hν f CH3O + NO
CH3O + O2 f CH2O + HO2
HO2 + NO f OH + NO2
* Corresponding author. Email: twalling@ford.com.
10.1021/jp000498r CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/14/2000

7256 J. Phys. Chem. A, Vol. 104, No. 31, 2000
Mashino et al.
Chlorine atoms were generated by the photolysis of molecular
chlorine:
Cl2 + hν f 2Cl
Loss of CF3CFdCF2 and the formation of products were
measured by FT-IR spectroscopy at a resolution of 0.25 cm-1
.
IR spectra were derived from 32 co-added interferograms.
Reference spectra were acquired by expanding known volumes
of authentic reference compounds into the chamber. All reagents
except CH3ONO were obtained from commercial sources at
purities >99%. Ultrahigh purity synthetic air was used as the
diluent gas in all experiments. CH3ONO was prepared by
dropwise addition of concentrated H2SO4 to a saturated solution
of NaNO2 in methanol and was devoid of any detectable
impurities using FTIR analysis. Reference spectra of CF3CFd
CF2, COF2, and CF3C(O)F were obtained by expansion of
calibrated volumes containing authentic samples of these
compounds into the reaction chamber. Unless otherwise stated,
all uncertainties quoted in the present manuscript are 2 standard
deviations from regression analyses.
In smog chamber experiments, unwanted loss of reactants
and products via photolysis, dark chemistry, and wall reactions
have to be considered. Control experiments were performed to
check for such unwanted losses of CF3CFdCF2, COF2, and
CF3C(O)F in the chamber; none were observed.
Three sets of experiments were performed. First, relative rate
techniques were used to determine rate constants for the
reactions of OH radicals and Cl atoms with CF3CFdCF2, using
C2H4, C2H2, and C2H5Cl as reference gases. Second, the
products of the OH radical- and Cl atom-initiated oxidation of
CF3CFdCF2 in 700 Torr of air were identified and quantified.
Third, the kinetics of the reaction of O3 with CF3CFdCF2 was
studied using an absolute technique.
Figure 1. Loss of CF3CFdCF2 versus the reference compounds
C2H5Cl, C2H2, and C2H4 in the presence of either Cl atoms (top panel)
or OH radicals (bottom panel). Experiments were performed at 296 K
in 700 Torr of either air or N2 diluent.
uncertainty reflects the accuracy of the measurements. There
are no literature data for k2 to compare with our results.
3.2. Relative Rate Study of k(OH+CF3CFdCF2). The
kinetics of reaction 1 were measured relative to reactions 6
and 7:
CF3CFdCF2 + OH f products
C2H4 + OH f products
(1)
(6)
(7)
C2H2 + OH f products
3. Results
Initial concentrations were 6.9-10.4 mTorr of CF3CFdCF2,
50-102 mTorr of CH3ONO, 7.4-15 mTorr of NO, and 4.4-
10.1 mTorr of either C2H4 or C2H2 in 700 Torr of air or N2
diluent. The observed loss of CF3CFdCF2 versus those of
reference compounds in the presence of OH radicals is shown
in the bottom panel of Figure 1. Linear least-squares analysis
of the data gives k1/k6 ) 0.29 ( 0.02 and k1/k7 ) 2.65 ( 0.15.
Using k6 ) 8.53 × 10-12 and k7 ) 8.70 × 10-13 cm3 molecule-1
3.1. Relative Rate Study of k(Cl + CF3CFdCF2). The
kinetics of reaction 2 were measured relative to reactions 4
and 5:
CF3CFdCF2 + Cl f products
C2H4 + Cl f products
(2)
(4)
(5)
s-1 14
,
we derive k1 ) (2.47 ( 0.17) × 10-12 and (2.31 ( 0.13)
C2H5Cl + Cl f products
× 10-12 cm3 molecule-1 s-1. We estimate that potential
systematic errors associated with uncertainties in the reference
rate constants contribute an additional 10% uncertainty range
to k1. Propagating this additional uncertainty gives k1 ) (2.47
( 0.30) × 10-12 and (2.31 ( 0.27) × 10-12 cm3 molecule-1
s-1. We choose to cite a final value for k1 that is the average of
those determined using the two different reference compounds
together with error limits that encompass the extremes of the
individual determinations. Hence, k1 ) (2.4 ( 0.3) × 10-12
cm3 molecule-1 s-1. Quoted error reflects the accuracy of the
measurements. This result is in excellent agreement with the
previous measurements at 298 K of k1 ) (2.3 ( 0.1) × 10-12
by McIlroy and Tully,15 k1) (2.2 ( 0.1) × 10-12 cm3
molecule-1 s-1 by Orkin et al.,16 and k1 ) (2.1 ( 0.2) × 10-12
cm3 molecule-1 s-1 by Dubey et al.17
Reaction mixtures consisted of 6.4-9.8 mTorr of CF3CFdCF2,
73-83 mTorr of Cl2, and 3.4-76 mTorr of either C2H4 or
C2H5Cl in 700 Torr of air or N2 diluent. The rate constant
k2 was derived by observing the relative loss rates of
CF3CFdCF2 and the reference compounds; results are shown
in the top panel of Figure 1.
Linear least-squares analysis of the data in Figure 1 gives
k2/k4 ) 0.29 ( 0.02 and k2/k5 ) 3.26 ( 0.25. Using k4 ) 9.29
× 10-11 12
,
and k5 ) 8.04 × 10-12 13
,
we derive k2 ) (2.69 (
0.19) × 10-11 and (2.62 ( 0.20) × 10-11 cm3 molecule-1 s-1
.
We estimate that potential systematic errors associated with
uncertainties in the reference rate constants contribute an
additional 10% uncertainty to k2. Propagating this additional
uncertainty gives k2 ) (2.69 ( 0.32) × 10-11 and (2.62 ( 0.33)
× 10-11 cm3 molecule-1 s-1. We choose to cite a final value
for k2 that is the average of those determined using the two
different reference compounds together with error limits that
encompass the extremes of the individual determinations. Hence,
k2 ) (2.7 ( 0.3) × 10-11 cm3 molecule-1 s-1; the quoted
3.3. Products and Mechanism of Cl Atom Initiated
Oxidation of CF3CFdCF2. To investigate the products and
mechanism of the reaction of Cl atoms with CF3CFdCF2,
reaction mixtures consisting of 6.8 mTorr CF3CFdCF2, 7.5
mTorr Cl2, and 0-15 mTorr NO in 700 Torr air were introduced

Atmospheric Chemistry of CF3CFdCF2
J. Phys. Chem. A, Vol. 104, No. 31, 2000 7257
of CF3CFdCF2/Cl2 air mixtures in the presence (filled symbols)
or absence (open symbols) of NO. As seen from Figure 3, there
was no discernible difference in the COF2 and CF3C(O)F yields
between experiments performed with and without added NO.
Linear least-squares analysis of the composite data sets gives
molar yields of COF2 and CF3C(O)F of 104 ( 7% and 94 (
6%, respectively.
The reaction of Cl atoms with CF3CFdCF2 proceeds via
addition to give two different substituted alkyl radicals which,
in air, are expected to add O2 rapidly (within 1 µs) to give the
corresponding peroxy radicals:
Cl• + CF3CFdCF2 f CF3C(Cl)FCF2•
Cl• + CF3CFdCF2 f CF3C(•)FCF2Cl
CF3C(Cl)FCF2• + O2 f CF3C(Cl)FCF2O2•
CF3C(•)FCF2Cl + O2 f CF3C(O2•)FCF2Cl
(2a)
(2b)
There is no available information concerning the branching
ratio k2a/k2b, and we will proceed on the assumption that both
radicals are formed. Peroxy radicals react rapidly with NO,18
and for those experiments where NO was present, the sole fate
of the peroxy radicals will be reaction with NO. Such reactions
proceed via two channels, giving alkoxy radicals as major
products and alkyl nitrates as minor products, that is,
Figure 2. Infrared spectra acquired before (A) and after (B) a 60-s
irradiation (using 22 fluorescent lamps) of a mixture of 6.8 mTorr of
CF3CFdCF2 and 7.5 mTorr of Cl2 in 700 Torr of air. Panel (C) shows
the product spectrum obtained after subtracting features attributable to
CF3CFdCF2 (32% of the original amount) from panel (B). Panels (D)
and (E) are reference spectra of COF2 and CF3C(O)F.
CF3C(O2•)FCF2Cl + NO f CF3C(O•)FCF2Cl + NO2 (8a)
CF3C(O2•)FCF2Cl + NO f CF3C(ONO2)FCF2Cl (8b)
In the present work, there was no evidence of nitrate
formation showing that channel (8b) is of minor importance.
This observation is consistent with previous studies of haloge-
nated alkyl peroxy radicals that produce low nitrate yields in
their reactions with NO.18 Whether by peroxy radical self- or
cross-reactions or by peroxy radical and NO reactions, two
alkoxy radicals are produced: CF3C(Cl)FCF2O• and CF3C(O•)-
FCF2Cl. From the fact that the observed COF2 and CF3C(O)F
products account for 100% of the loss of CF3CFdCF2, we
conclude that the fate of CF3C(Cl)FCF2O• and CF3C(O•)-
FCF2Cl radicals is decomposition via C-C bond scission,
giving CF3C(Cl)F• radicals and CF2O, and CF3C(O)F and
•CF2Cl radicals, respectively. The atmospheric fate of CF3C(Cl)F•
and •CF2Cl radicals is addition of O2 to give a peroxy radical,
reaction with NO to give an alkoxy radical, and elimination of
a Cl atom to give either CF3C(O)F or COF2.19
3.4. Products and Mechanism of OH Radical-Initiated
Oxidation of CF3CFdCF2. To investigate the products and
mechanism of the reaction of OH radicals with CF3CFdCF2,
reaction mixtures consisting of 7.0-8.0 mTorr CF3CFdCF2,
50-114 mTorr CH3ONO, and 7.5-10.4 mTorr NO in either
10 or 700 Torr air were introduced into the reaction chamber
and subjected to UV irradiation. Typical consumptions of
CF3CFdCF2 were in the range 5-55%. Figure 4 shows typical
spectra acquired before (A) and after (B) 22-min irradiation
(using 22 fluorescent lamps) of a mixture containing 7.3 mTorr
of CF3CFdCF2, 51 mTorr of CH3ONO, and 7.5 mTorr of NO
in 700 Torr of air. Subtraction of IR features attributable to
CF3CFdCF2 and NO from (B) gives the product spectrum
shown in (C). Comparison with reference spectra of COF2 and
CF3C(O)F given in panels (D) and (E) shows these compounds
are products. There were no other carbon-containing products
detected. An upper limit of 5% was established for the molar
Figure 3. Formation of COF2 (circles) and CF3C(O)F (triangles) versus
loss of CF3CFdCF2 following UV irradiation of CF3CFdCF2/Cl2/air
mixtures in the presence (filled symbols) and absence (open symbols)
of NO.
into the reaction chamber and subjected to UV irradiation.
Typical consumptions of CF3CFdCF2 were in the range
5-90%. Figure 2 shows typical spectra acquired before (A) and
after (B) a 60-s irradiation (using 22 fluorescent lamps) of a
mixture containing 6.8 mTorr of CF3CFdCF2 and 7.5 mTorr
of Cl2 in 700 Torr of air. Subtraction of IR features attributable
to CF3CFdCF2 from (B) gives the product spectrum shown in
(C). Comparison of the IR features in panel (C) with reference
spectra of COF2 and CF3C(O)F shown in panels (D) and (E)
shows the formation of these compounds. After subtraction of
features attributable to COF2 and CF3C(O)F, residual IR features
at 788, 964, 1240, 1302, and 1763 cm-1 were observed due to
one or more unidentified additional products.
Figure 3 shows a plot of the observed formation of COF2
and CF3C(O)F versus loss of CF3CFdCF2 following irradiation

7258 J. Phys. Chem. A, Vol. 104, No. 31, 2000
Mashino et al.
Figure 5. Formation of COF2 (circles) and CF3C(O)F (triangles) versus
loss of CF3CFdCF2 following the UV irradiation of CF3CFdCF2/CH3-
ONO/NO in 700 (open symbols) or 10 (filled symbols) Torr of air
diluent at 296 K.
In the present work, there was no evidence of nitrate for-
mation, showing that channel (9b) is of minor importance. This
observation is consistent with previous studies of fluorinated
alkyl peroxy radicals that produce low nitrate yields in their
reactions with NO.18 By peroxy radical and NO reaction, two
alkoxy radicals are produced: CF3C(OH)FCF2O• and CF3C(O•)-
FCF2OH. The observed COF2 and CF3C(O)F products account
for 100% of the loss of CF3CFdCF2; the fate of both alkoxy
radicals is decomposition via C-C bond scission. The resulting
R-hydroxy alkyl radicals react rapidly with O2 to give the
corresponding carbonyl and an HO2 radical:
Figure 4. Infrared spectra acquired before (A) and after (B) a 22-min
irradiation of a mixture of 7.7 mTorr of CF3CFdCF2, 51 mTorr of
CH3ONO, and 7.5 mTorr of NO in 700 Torr of air. Panel (C) shows
the product spectrum obtained after subtracting features attributable to
CF3CFdCF2 and NO (57% of the original amount) from panel (B).
Panels (D) and (E) are reference spectra of COF2 and CF3OC(O)F.
yield of oxalyl fluoride, showing that the OH-initiated atmo-
spheric oxidation of CF3CFdCF2 does not proceed via the same
mechanism as CF3OCFdCF2.10
The observed yields of COF2 (circles) and CF3C(O)F
(triangles) are plotted versus the loss of CF3CFdCF2 in Figure
5 for experiments conducted in either 700 (open symbols) or
10 (filled symbols) Torr of air diluent. As seen from Figure 5,
there was no discernible effect of total pressure on the observed
product yields. Linear least-squares analysis of the composite
data set gives molar yields of COF2 and CF3C(O)F of 98 (
7% and 90 ( 6%. Quoted errors represent statistical uncertain-
ties (2 standard deviations); we estimate that systematic un-
certainties associated with calibration of the reference spectra
contribute an additional 10% uncertainty.
CF3C(•)(OH)F + O2 f CF3C(O)F + HO2
•CF2OH + O2 f COF2 + HO2
3.5. Study of k(O3 + CF3CFdCF2). The kinetics of reaction
3 were investigated by filling the smog chamber with 700 Torr
of ozonized O2 containing 899-963 mTorr of O3, then adding
1.5-8.1 mTorr of CF3CFdCF2 and monitoring the decay of
CF3CFdCF2 and O3 for 60-70 min:
The reaction of OH radicals with CF3CFdCF2 proceeds via
addition to give two different substituted alkyl radicals which,
in air, will add O2 rapidly (within 1 µs) to give the corresponding
peroxy radicals:
CF3CFdCF2 + O3 f products
(3)
During the 60-70 min of each experiment, there was a
slow but discernible loss of ozone in the chamber, corresponding
to a loss of approximately 2% in 60 min. We attribute the
slow loss of ozone mainly to loss via decomposition on the
chamber walls. Loss of ozone via reaction 3 is of minor im-
portance because ozone is present in great excess compared to
CF3CFdCF2.
Figure 6 shows the observed decay of CF3CFdCF2 as a
function of time when CF3CFdCF2/O3/O2 mixtures were
allowed to stand in the dark in the chamber. Control experiments
using CF3CFdCF2/O2 mixtures established that CF3CFdCF2
loss was negligible (<2% in 60 min) in the absence of O3. We
conclude that the CF3CFdCF2 loss shown in Figure 6 is caused
by reaction with ozone. It is interesting to note that whereas
the initial ozone concentrations used in the experiments shown
in Figure 6 varied by less than 7%, the rates of decay of
CF3CFdCF2 varied by 36% (pseudo first-order rates of 0.9-
1.4 × 10-4 s-1). Furthermore, close inspection of the data in
Figure 6 reveals a systematic increase in the pseudo first-order
CF3CFdCF2 loss rate with the initial concentration of
CF3CFdCF2. The simplest interpretation of this observation is
that there is a complication in the system caused by radicals
formed from the O3 + CF3CFdCF2 reaction that react with
OH• + CF3CFdCF2 f CF3C(OH)FCF2•
OH• + CF3CFdCF2 f CF3C(•)FCF2OH
CF3C(OH)FCF2• + O2 f CF3C(OH)FCF2O2•
CF3C(•)FCF2OH + O2 f CF3C(O2•)FCF2OH
(1a)
(1b)
There is no available information concerning the branching
ratio k1a/k1b, and we will proceed on the assumption that both
radicals are formed. Peroxy radicals react rapidly with NO,18
and for those experiments where NO was present, the sole fate
of the peroxy radicals will be reaction with NO. Such reactions
generally proceed via two channels, giving alkoxy radicals as
major products and alkyl nitrates as minor products, that is,
CF3C(O2•)FCF2OH + NO f CF3C(O•)FCF2OH + NO2
(9a)
CF3C(O2•)FCF2OH + NO f CF3C(ONO2)FCF2OH (9b)

Atmospheric Chemistry of CF3CFdCF2
J. Phys. Chem. A, Vol. 104, No. 31, 2000 7259
reaction with ozone of 8 years. Current estimates suggest that
typical marine boundary layer levels of Cl atoms are around
104 cm-3 23-25
with lower levels over continental areas. Using
,
[Cl] ) 104 cm-3 in conjunction with the value k2 ) 2.7 ×
10-11 cm3 molecule-1 s-1 reported in this study gives a life-
time of 43 days for CF3CFdCF2 via reaction with Cl atoms.
As for other alkenes, photolysis of CF3CFdCF2 in the atmo-
sphere will not be of any significance.26 The atmospheric loss
of CF3CFdCF2 is dominated by reaction with OH radicals,
giving COF2 and CF3C(O)F in molar yields of 100%. The
atmospheric fate of COF2 and CF3C(O)F is incorporation into
water droplets (within 5-15 days) followed by hydrolysis to
give either CO2 and HF or CF3C(O)OH and HF.19
With regard to the second aim, we show that, following
release into the atmosphere, CF3CFdCF2 is converted within a
week or two into CF3C(O)OH. There are two atmospheric fates
of CF3C(O)OH; rainout and reaction with OH. Reaction with
OH is slow; approximately 95% of the CF3C(O)OH produced
from the atmospheric oxidation of CF3CFdCF2 will be rained
out and will enter the hydrosphere.27 To assess the importance
of CF3CFdCF2 as a source of CF3C(O)OH, we need to compare
it with other sources and with the environmental burden of
CF3C(O)OH. Table 1 provides a list of the known sources of
CF3C(O)OH in the atmosphere. We have shown that any
CF3CFdCF2 that is emitted into the atmosphere is converted
into CF3C(O)OH. The importance of CF3CFdCF2 as a
CF3C(O)OH source depends on the amount of CF3CFdCF2
emitted. There are no published data concerning the global
industrial production of CF3CFdCF2. However, based upon the
published fluoropolymer production in 1997, it has been
estimated that the global industrial production of CF3CFdCF2
lies in the range 10-15 ktonnes/year.28 It is reasonable to expect
that essentially all of this CF3CFdCF2 will be incorporated into
chemical products and that direct emission of CF3CFdCF2
from the point of production will be minimal. It is possible
that CF3CFdCF2 is released during incineration of fluoro-
polymers. The global production of fluoropolymers is large
(40,000 tonnes in 1988); if even a small fraction is converted
into CF3CFdCF2, this would be significant. To the best of our
knowledge, there have been no measurements of CF3CFdCF2
emissions from incineration plants. Jordan and Frank3 assume
that 2500 tonnes/yr of fluoropolymer waste is incinerated in
Europe and that 10% of the fluoropolymer is released to the
atmosphere as CF3CFdCF2. Emission measurements are re-
quired to determine if this assumption is reasonable and to
provide a clearer picture of the importance of this potential
source of CF3CFdCF2.
Figure 6. Decay of CF3CFdCF2 in experiments employing mix-
tures of 1.5 mTorr CF3CFdCF2 and 927 mTorr O3 (open circles), 2.8
mTorr CF3CFdCF2 and 963 mTorr O3 (filled circles), 4.3 mTorr
CF3CFdCF2 and 899 mTorr O3 (filled squares), and 8.1 mTorr
CF3CFdCF2 and 920 mTorr O3 (filled triangles) in 700 Torr of O2
diluent at 296 K.
CF3CFdCF2, leading to an enhanced loss of CF3CFdCF2. This
effect will be more important at higher initial concentrations of
CF3CFdCF2, leading to the observed behavior. Our conclusion
is that the lowest value of k(CF3CFdCF2 + O3) derived from
the data in Figure 6 should be treated as an upper limit. Hence,
we report k(CF3CFdCF2 + O3) < 3 × 10-21 cm3 molecule-1
s-1. This result is approximately 104 times lower than the
value of k(CF3CFdCF2 + O3) ) 2.2 × 10-17 cm3 molecule-1
s-1 reported by Heicklen.20 The initial concentrations of
CF3CFdCF2 used in the study by Heicklen were 100-1000
times greater than those used herein. It seems reasonable to
suspect that secondary loss of CF3CFdCF2 was a major problem
in the previous work and that the value of k3 reported by
Heicklen20 is erroneously large.
4. Conclusions
The motivation behind the present study was 2-fold; first, to
provide an understanding of the atmospheric chemistry of
CF3CFdCF2 and second, to assess its potential contribution to
the environmental burden of trifluoroacetic acid.
With regard to the first aim, we present a substantial body
of kinetic and mechanistic data pertaining to the atmospheric
chemistry of CF3CFdCF2. The atmospheric lifetime of
CF3CFdCF2 with respect to reaction with OH can be esti-
mated using three pieces of information: the value of k1 ) 2.4
× 10-12 measured herein, k(OH + CCl3CH3) ) 1.0 × 10-14
cm3 molecule-1 s-1 at 296 K,21 and the atmospheric lifetime of
CCl3CH3 with respect to reaction with OH of 5.9 years.22 This
approach gives an atmospheric lifetime of CF3CFdCF2 with
respect to reaction with OH of 9 days. The atmospheric lifetime
of CF3CFdCF2 with respect to reaction with ozone can be
estimated by combining a typical tropospheric ozone concentra-
tion of 50 ppb with k3 < 3 × 10-21 cm3 molecule-1 s-1, giving
a lower limit for the lifetime of CF3CFdCF2 with respect to
At this point, we should contrast the environmental burden
of CF3C(O)OH, which has been estimated at 107 - 108 tonnes,6
with the potential source from atmospheric oxidation of
CF3CFdCF2, which has been estimated at 250 tonnes.3 There
is a very large difference between the environmental burden of
CF3C(O)OH and its potential source from CF3CFdCF2. It is
possible that atmospheric oxidation of CF3CFdCF2 may
TABLE 1: Compounds Known to Produce CF3COOH in the Atmosphere
compound
common name
molar CF3COOH yield
atmospheric lifetime
estimated TFA flux (tonne/yr)
CF3CHClBr
CF3CHClOCHF2
CF3CHCl2
CF3CHFCl
CF3CH2F
Halothane
Isoflurane
HCFC-123
HCFC-124
HFC-134a
0.6
0.6
0.6
1.0
0.13
1.0
1.2 years
5 years
1.5 years
6.0 years
14.6 years
9 days
520
280
<760
<320
1200a
250?b
CF3CFdCF2
hexafluoropropene
a Assuming [HFC-134a] ) 10 pptv and 5 pptv in N. and S. hemispheres, respectively.29-31 b Estimate from ref 3.

7260 J. Phys. Chem. A, Vol. 104, No. 31, 2000
Mashino et al.
(14) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr,
J. A.; Rossi, M. J.; Troe, J. J. Phys. Chem. Ref. Data 1997, 26, 521.
(15) McIlroy, A.; Tully, F. P. J. Phys. Chem. 1993, 97, 610.
(16) Orkin, V. L.; Huie, R. E.; Kurylo, M. J. J. Phys. Chem. A 1997,
101, 9118.
constitute a significant fraction of the man-made CF3C(O)OH
in the environment, but it does not appear to contribute a
significant fraction of the levels of CF3C(O)OH observed in
the global environment.
(17) Dubey, M. K.; Hanisco, T. F.; Wennberg, P. O.; Anderson, J. G.
Geophys. Res. Lett. 1996, 23, 3215.
(18) Wallington, T. J.; Dagaut, P.; Kurylo, M. J.; Chem. ReV. 1992, 92,
667.
(19) Wallington, T. J.; Schneider, W. F.; Worsnop, D. R.; Nielsen, O.
J.; Sehested, J.; DeBruyn, W.; Shorter, J. A. EnViron. Sci. Tech. 1994, 28,
320A.
Acknowledgment. We thank Archie McCulloch (ICI Chemi-
cals & Polymers Ltd.) and James Franklin (Solvay Research
and Technology) for helpful discussions and the Japanese
Government for a NEDO grant that made this collaborative
research project possible.
(20) Heicklen, J. J. Phys. Chem. 1966, 70, 477.
(21) 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 Publication No. 97-4; NASA Jet Propulsion Laboratory, Pasadena,
CA, 1997.
(22) Intergovernmental Panel on Climate Change (IPCC). The Science
of Climate Change; Cambridge University Press: New York, 1995.
(23) Pszenny, A. A. P.; Keene, W. C.; Jacob, D. J.; Fan, S.; Maben, J.
R.; Zetwo, M. P.; Springer-Young, M.; Galloway, J. N. Geophys. Res. Lett.
1993, 20, 699.
References and Notes
(1) Zehavi, D.; Seiber, J. N. Anal. Chem. 1996, 68, 3450.
(2) Frank, H.; Klein, A.; Renschen, D. Nature 1996, 382, 34.
(3) Jordan, A.; Frank, H. EnViron. Sci. Technol. 1999, 33, 522.
(4) Wujcik, C. E.; Zehavi, D.; Seiber, J. N. Chemosphere 1998, 36,
1233.
(5) Scott, B. F.; Muir, D. C. G.; Spencer, C.; MacDonald, R.; Witter,
A.; Fisk, A. Poster presented at the Atmospheric Reactive Substances
Symposium, Bayreuth, April, 1999.
(24) Singh, H. B.; Thakur, A. N.; Chen, Y. E.; Kanakidou, M. Geophys.
Res. Lett. 1996, 23, 1529.
(6) Kim B. R.; Suidan, M. T.; Wallington, T. J.; Du, X. EnViron. Eng.
Sci., submitted for publication.
(25) Vogt, R.; Crutzen, P. J.; Sander, R. Nature 1996, 383, 327.
(26) Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.; Moortgat,
G. K.; Wallington, T. J.; Yarwood, G. Mechanisms of the Atmospheric
Oxidation of the Alkenes; Oxford University Press: New York, 2000.
(27) Kanakidou, M.; Dentener, F. J.; Crutzen, P. J. J. Geophys. Res.
1995, 100, 18781.
(28) Franklin, J.; McCulloch, A. Private communication, 1999.
(29) Montzka, S. A.; Myers, R. C.; Butler, J. H.; Elkins, J. W.; Lock,
L. T.; Clarke, A. D. Geophys. Res. Lett. 1996, 23, 169.
(30) Oram, D. E.; Reeves, C. E.; Sturges, W. T.; Penkett, S. A.; Fraser,
P. J.; Langenfelds, R. L. Geophys. Res. Lett. 1996, 23, 1949.
(31) Shirai, T.; Makide, Y. Chem. Lett. 1998, No. 4, 357.
(7) Eapen, K. C.; Chen, L. S.; Chen, G. J. J. Fluorine Chem. 1997,
81, 143.
(8) Atkinson, B.; Atkinson, V. A. J. Chem. Soc. 1957, 32, 2086.
(9) Li, Z.; Tao, Z.; Naik, V.; Good, D. A.; Hansen, J.; Jeong, G. R.;
Francisco, J. S.; Jain, A.; Wuebbles, D. J. J. Geophys. Res. 1999, in press.
(10) Mashino, M.; Kawasaki, M.; Wallington, T. J.; Hurley, M. D. J.
Phys. Chem. A 2000, 104, 2925.
(11) Wallington, T. J.; Japar, S. M. J. Atmos. Chem. 1989, 9, 399.
(12) 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. Jet Propulsion Laboratory Publication 97-4; Pasadena, CA, 1997.
(13) Wine, P. H.; Semmes, D. H. J. Phys. Chem. 1983, 87, 3572.