Kinetic and mechanistic studies for reactions of CF3CH2CHF2 (HFC-245fa) initiated by H-atom abstraction using atomic chlorine

Article abstract of DOI:10.1021/jp963735s

Kinetics and mechanisms of Cl atom initiated oxidation reactions of CF₃CH₂CHF₂ (HFC-245fa) were investigated by long-path FTIR spectroscopic methods at 297 ± 2 K. The Cl atoms initiated the reaction mainly via H-atom abstraction from the terminal carbon (≥95%) to produce a CF₃CH₂CF₂ radical. Subsequent reactions in 700 Torr of air produced CF₃CHO and CF₂O as the primary products. Secondary reactions of CF₃CHO with Cl atoms led to the formation of CF₃C(O)OH, CF₃OH, CO₂, and CF₃OOOCF₃ in the experimental system. However, under atmospheric conditions, CF₃CHO would undergo a series of rapid oxidation and decomposition reactions ultimately leading to HF and CO₂, and no long-lived organic decomposition products would be produced.

Full text of DOI:10.1021/jp963735s

2648
J. Phys. Chem. A 1997, 101, 2648-2653
Kinetic and Mechanistic Studies for Reactions of CF3CH2CHF2 (HFC-245fa) Initiated by
H-Atom Abstraction Using Atomic Chlorine
†
‡
§
Junyi Chen, Valerie Young, and Hiromi Niki
Centre for Atmospheric Chemistry and Department of Chemistry, York UniVersity, 4700 Keele Street,
North York, Ontario, Canada M3J 1P3
Hillel Magid*
Buffalo Research Laboratory, AlliedSignal, Inc., 20 Peabody Street, Buffalo, New York 14210
X
ReceiVed: NoVember 8, 1996; In Final Form: January 28, 1997
Kinetics and mechanisms of Cl atom initiated oxidation reactions of CF
investigated by long-path FTIR spectroscopic methods at 297 ( 2 K. The Cl atoms initiated the reaction
mainly via H-atom abstraction from the terminal carbon (g95%) to produce a CF CH CF radical. Subsequent
reactions in 700 Torr of air produced CF CHO and CF O as the primary products. Secondary reactions of
CF CHO with Cl atoms led to the formation of CF C(O)OH, CF OH, CO , and CF OOOCF in the experimental
system. However, under atmospheric conditions, CF CHO would undergo a series of rapid oxidation and
, and no long-lived organic decomposition products
3 2 2
CH CHF (HFC-245fa) were
3
2
2
3
2
3
3
3
2
3
3
3
decomposition reactions ultimately leading to HF and CO
would be produced.
2
Introduction
glass cylinder (5 cm diameter, 50 cm long, 100 cm path length)
surrounded by 6 UV fluorescent lamps (GE F30T8/BLB; 300
nm < λ < 400 nm) was used as the IR cell and photochemical
reactor for experiments carried out in He and N2; another Pyrex
glass cylinder (30 cm diameter, 2 m long, 92 m path length)
surrounded by 26 UV fluorescent lamps (GE F40T12/BLB; 300
nm < λ < 400 nm) was used as the IR cell and photochemical
reactor for experiments carried out in the presence of O2.
Spectra were collected over the frequency range 500-4000
Destruction of stratospheric ozone by chlorofluorocarbons
1
(
CFCs) has led to an international agreement to replace CFCs
with atmospherically acceptable alternatives. Hydrochlorof-
luorocarbons (HCFCs) and hydrofluorocarbons (HFCs) have
been developed as CFC alternatives because they can, to a large
extent, be removed by tropospheric OH radicals. Thus, the
alternatives have shorter lifetimes in the troposphere as well
as lower ozone depletion potential (ODP) and global warming
potential (GWP) values than those of CFCs. However, the
tropospheric degradation of the alternatives, after initial hydro-
gen abstraction by OH radicals, will lead to new halogen-
containing products, the atmospheric behavior of which may
not be well-known. Thus, to assess the atmospheric accept-
ability of the alternatives, we need to understand their degrada-
tion mechanisms and product distributions.
-
1
-1
cm in 70 s at 0.125 cm resolution by co-adding 16 scans,
using a Mattson FTIR spectrometer (Galaxy 4326C) with a KBr
beamsplitter and a liquid N2 cooled Hg-Cd-Te detector.
Cl atoms were generated by the UV irradiation (300 nm < λ
400 nm) of Cl2 (Aldrich, 99+%). N2 (O2 < 3 ppm, total
<
hydrocarbon < 1 ppm), He (O2 < 0.5 ppm) and synthetic air
total hydrocarbon < 0.1 ppm) were obtained from Liquid
(
Carbonic Inc. Reactant (CF3CH2CHF2, 99%) was supplied by
the Buffalo Research Laboratory of AlliedSignal. No IR peaks
of impurities were observed in the reactant. CF3CHO was
synthesized by hydrolyzing CF3CH(OH)OCH3 (PCR, 99%) in
a mixture of H2SO4 (Aldrich, 95.5%)/P2O5 (Aldrich, 97%).
CF3C(O)OH (99.9%) and CF3CH2CClF2 (97%) were obtained
from PCR and the Buffalo Research Laboratory of AlliedSignal,
respectively.
There have been many studies of the atmospheric reaction
mechanisms of HCFCs and HFCs after initial abstraction of
2
-5
hydrogen atoms.
However, most of the studies were
concerned with the degradation of HCFCs and HFCs containing
one or two carbons. There are few experimental data available
for larger HCFCs and HFCs. Presented here is the long-path
FTIR spectroscopic kinetic and mechanistic study of the Cl atom
initiated oxidation of CF3CH2CHF2 (HFC-245fa) at 297 ( 2K
in 700 Torr of He, N2, and air diluents. This molecule is being
considered as a potential non-ozone-depleting replacement
for CH3CCl2F (HCFC-141b) as a blowing agent for foam
applications.6
Reactant and product species reported in this work, unless
otherwise specified, were identified and quantified by reference
to quantitative IR spectra of pure compounds previously
collected in our laboratory. The concentration uncertainties of
these reference spectra are estimated to be less than 5%, except
for CF3C(O)OH. Because of the formation of the dimer, (CF3C-
a-c
Experimental Section
(
O)OH)2, and because the necessity of separating the dimer adds
The details of the experimental facility and procedures have
been described in our previous publications. In brief, a Pyrex
uncertainty to the reference spectrum, the concentration uncer-
tainty of the reference spectrum of CF3C(O)OH is estimated to
be 10%.
7
*
Corresponding author. E-mail: hillel.magid@alliedsignal.com.
Present address: NU-Pharm Inc, 380 Elgin Mills East, Richmond Hill,
†
Ontario, L4C 5H2, Canada.
Results and Discussion
‡
Present address: Department of Chemical Engineering, Ohio University,
1
79 Stocker Center, Athens, OH 45701.
§
1. Kinetic Study. The rate constant for the reaction of Cl
atom with CF3CH2CHF2 was determined from the decay rates
Deceased.
X
Abstract published in AdVance ACS Abstracts, March 15, 1997.
S1089-5639(96)03735-8 CCC: $14.00 © 1997 American Chemical Society

Reactions of CF3CH2CHF2 (HFC-245fa)
J. Phys. Chem. A, Vol. 101, No. 14, 1997 2649
-
1
Figure 2. Spectral data in the frequency region 500-1500 cm
obtained from the photolysis of a mixture containing CF CH CHF (0.5
Torr) and Cl (3.4 Torr) in 700 Torr of He: (A) before irradiation; (B)
after 25 s irradiation; (C) difference spectrum ) [(B) - (A)] × 3.
Figure 1. Plot of ln{[CF
CHClF] /[CF CHClF] }, in which the slope gives the rate constant ratio
k(CF CH CHF +Cl)/k(CF CHClF+Cl) ) 2.55 ( 0.30.
3 2 2 0 3 2 2 t 3
CH CHF ] /[CF CH CHF ] } against ln{[CF -
3
2
2
0
3
t
2
3
2
2
3
of CF3CH2CHF2 and a reference compound, CF3CHClF (HCFC-
24), in the UV photolysis of mixtures containing these two
1
compounds and Cl2 at 297 ( K, and using the integrated relative
rate equation:
Ln{[CF CH CHF ] /[CF CH CHF ] }
3
2
2 0
3
2
2 t
)
Ln{[CF CHClF] /[CF CHClF] }
3
0
3
t
k[CF CH CHF + Cl]
3
2
2
k[CF CHClF + Cl]
3
In the presence of O2, CF3O radicals may be produced in the
reaction system as an important intermediate product. These
CF3O radicals may initiate reactions in competition with the Cl
atoms and confound the rate constant measurement. To avoid
the possible side reactions of CF3CH2CHF2 and CF3CHClF with
8
CF3O radicals, the experiments were carried out in 700 Torr
of N2, using reactant mixtures containing CF3CH2CHF2 (0.4-
0
.5 Torr), CF3CHClF (0.4-0.5 Torr) and Cl2 (2-3 Torr). A
relative rate constant, k[CF3CH2CHF2 + Cl]/k[CF3CHClF +
Cl], was derived to be 2.55 ( 0.30 (2σ) by linear regression of
Ln{[CF3CH2CHF2]0/[CF3CH2CHF2]t} against Ln{[CF3CHClF]0/
[CF3CHClF]t}, as presented in Figure 1. Placed on an absolute
basis, the rate constant is k[CF3CH2CHF2 + Cl] ) (6.9 ( 0.8)
-
15
3
-1 -1
×
10
.7 × 10
uncertainty reported above refers only to experimental error.
. Reactions in He. The Cl atom abstracts an H atom from
(2σ) cm molecule
s , using k[CF3CHClF + Cl] )
-
15
3
-1 -1
9
Figure 3. Plot of the CF
conversion observed in 700 Torr of He. The slope gives the observed
yield of CF CH CClF to be 95% based on the conversion of CF CH
CHF
3 2 2 3 2 2
CH CClF yield against the CF CH CHF
2
cm molecule
s
as the reference. The
3
2
2
3
2
-
2
2
.
CF3CH2CHF2, leaving a radical site. In the absence of O2, this
radical site will be chlorinated by reaction with Cl2. To establish
the relative importance of central vs terminal H-atom abstraction
for Cl + CF3CH2CHF2 reactions, spectra were collected after
UV irradiation of mixtures containing CF3CH2CHF2 (0.4-0.7
Torr) and Cl2 (3-4 Torr) diluted in He to 700 Torr. The
absorbance spectra recorded in a typical run are illustrated in
Figure 2. Parts A and B of Figure 2 correspond to the spectra
recorded before and after 25 s photolysis of a mixture initially
containing CF3CH2CHF2 (0.5 Torr) and Cl2 (3.4 Torr) in 700
Torr of He; Figure 2C is the difference spectrum [2B -2A],
with the absorbance scale expanded by a factor of 3 for clarity.
The identified products after irradiation of the mixtures were
CF3CH2CClF2 and HCl. The observed yield of CF3CH2CClF2
is plotted as a function of the reactant conversion in Figure 3.
The slope gives the percentage yield of CF3CH2CClF2 to be
95%. The observation of CF3CH2CClF2 as the predominant
product is consistent with the occurrence of reaction 1a, H-atom

2650 J. Phys. Chem. A, Vol. 101, No. 14, 1997
Chen et al.
-
1
-
1
Figure 5. Spectral data in the frequency region 1500-2500 cm
obtained from the photolysis of a mixture containing CF CH CHF (0.6
Torr) and Cl (0.5 Torr) in 700 Torr of air: (A) before irradiation; (B)
Figure 4. Spectral data in the frequency region 500-2500 cm
obtained from the photolysis of a mixture containing CF CH CHF (3
mTorr) and Cl (0.5 Torr) in 700 Torr of air: (A) before irradiation;
B) after 3 min irradiation; (C) difference spectrum ) [(B) - (A)].
3
2
2
3
2
2
2
2
after 3 min irradiation; (C) difference spectrum ) [(B) - (A)] × 1.5.
(
abstraction from the terminal carbon producing CF3CH2CF2
radical, followed by reaction with Cl2 molecule, reaction 2.
experimental conditions, no H-containing products, except
inorganic HCl and HF, were identified. Because of the slow
-
15
3
reaction rate of CF3CH2CHF2 + Cl (k ) 6.9 × 10
cm
CF CH CHF + Cl f CF CH CF + HCl
(1a)
(1b)
(2)
-1 -1
3
2
2
3
2
2
molecule s ), it is quite possible that some active H-
containing intermediate products were formed in the system and
then scavenged by the faster secondary reactions with Cl atoms.
For example, CF3CHO is a possible product and its reaction
CF CH CHF + Cl f CF CHCHF + HCl
3
2
2
3
2
CF CH CF + Cl f CF CH CClF + Cl
-12
3
-1 -1 11
3
2
2
2
3
2
2
rate with Cl atoms (k ) 2.7 × 10
cm molecule s ) is
more than 400 times faster than that of CF3CH2CHF2 with Cl.
To limit the possible secondary reactions of the intermediate
products with Cl atoms, the experiments were also carried out
with high reactant concentrations and at low reactant conver-
sions. That is, the initial concentrations of CF3CH2CHF2 and
Cl2 were both 500-1000 mTorr, and the conversions of
CF3CH2CHF2 were a few percent. Parts A and B of Figure 5
correspond to the spectra recorded before and after 3 min
irradiation of a mixture initially containing CF3CH2CHF2 (600
mTorr) and Cl2 (500 mTorr) in 700 Torr of air; Figure 5C is
the difference spectrum [5B - 5A]. As shown in Figure 5, no
CF3CH2CClF2 was identified above the detection limit (0.02
mTorr). Thus, in 700 Torr of air, the Cl2 reaction with the
initially formed CF3CH2CF2 radicals, reaction 2, was negligible,
while O2 addition, reaction 4, is the predominant channel.
It is to be noted that CF3CHClCHF2 would be formed by a
central H-atom abstraction producing CF3CHCHF2 radical,
reaction 1b, followed by reaction 3.
CF CHCHF + Cl f CF CHClCHF + Cl
(3)
3
2
2
3
2
After subtracting CF3CH2CClF2 and HCl from Figure 2C,
the residual spectrum showed IR peaks at 1281, 1217, and 1182
-
1
cm , which are in the C-F stretching frequency range. If all
the residual spectra are assigned to CF3CHClCHF2, a lower limit
for the yield of CF3CH2CClF2 and a higher limit for the yield
of CF3CHClCHF2 are estimated to be 95% and 5%, respectively.
In other words, at least 95% of the CF3CH2CHF2 + Cl reactions
proceed via terminal H-atom abstraction. Reactions via central
H-atom abstraction, if they occur, account for no more than
5
% of the total.
. Reactions in Air. To determine the subsequent reaction
CF CH CF + O (+M) f CF CH CF OO (+M) (4)
3
3
2
2
2
3
2
2
pathways for CF3CH2CF2 radical in the presence of O2,
experiments were carried out in 700 Torr of air. Parts A and B
of Figure 4 correspond to the spectra recorded before and after
The identified carbon-containing products were CF2O (1944,
-
1
1243, and 774 cm ), CO2, CF3CHO (2864, 1787, and 706
-
1
12,13
8,15
-1 14
3
min irradiation of a mixture initially containing CF3CH2CHF2
cm ),
CF3C(O)OH (3584 and 1827 cm ), CF3OH (3664
-
1
(3 mTorr) and Cl2 (500 mTorr) in 700 Torr of air; Figure 4C is
cm ),
and a trace amount of CO. The product yields are
the difference spectrum [4B - 4A]. As shown in Figure 4, the
plotted as a function of the conversion of CF3CH2CHF2 in Figure
6. Since the initial concentration of CF3CH2CHF2 was very
high, all the IR peaks of fundamentals were saturated, and those
of overtones and combinations were too weak to be sensitive
enough to a few percent change of the reactant. Thus, the
reactant conversions in Figure 6 were estimated from the
observed yields of total products. The concentrations of
identified carbon-containing products were CO2, CF2O, and CF3-
-
1 10
OOOCF3 (1290, 1252, and 1169 cm ).
As reported by
1
0
Wallington et al., CF3OOOCF3 may be formed from the
reaction of CF3O and CF3OO radicals. However, CF3OOOCF3
is unlikely to form in the troposphere since the concentrations
of the CF3O and CF3OO would be very low. Under the

Reactions of CF3CH2CHF2 (HFC-245fa)
J. Phys. Chem. A, Vol. 101, No. 14, 1997 2651
Figure 6. Plots of product yields against conversion of CF
Data were obtained from experiments carried out by the photolysis of
3
CH
2
CHF
2
.
Figure 7. Yields of CF
Symbols O, 3, and 2 express yields of CF
experiments, corrected for the secondary reaction with Cl atoms and
derived from the sum of observed percentage yields of CF CHO, CO
and CF C(O)OH, respectively.
3
CHO against conversion of CF
3 2 2
CH CHF .
3
CHO observed from
mixtures initially containing CF
in 700 Torr of air.
3 2 2 2
CH CHF and Cl (500-1000 mTorr)
3
2
,
3
-
1
CF3OH were calculated from the IR absorbance σ (3664 cm )
-
3
-1
-1 8
minor importance and reaction 8 is the main degradation channel
)
2.4 × 10 mTorr M .
-
1
for CF CH OO radicals.
A small IR peak was observed at 3628 cm , which could
3
2
be the O-H stretching of a ROOH type molecule produced by
an ROO + HOO f ROOH + O2 reaction. By comparison
with the IR absorbance of the O-H stretching of CH3OOH and
C2H5OOH, acquired previously in our laboratory, the yield of
this ROOH type product is estimated to be less than 2%.
As shown in Figure 6, the yield of observed CF3CHO (55-
2
CF CH OO f CF CH OH + CF CHO + O (10)
3
2
3
2
3
2
Unimolecular dissociation producing CF3 radicals and CH2O
was expected to be a degradation channel for CF3CH2O radicals.
However, no CH2O was observed; our detection limit for CH2O
is about 0.01 mTorr. The unimolecular dissociation, reaction
2
5%) decreased with increasing CF3CH2CHF2 conversion,
1
1, was thus negligible.
indicating that loss of CF3CHO to secondary reactions is
important. The main secondary reaction of CF3CHO was
expected to be with Cl atoms, so the observed yield of CF3-
CHO could be corrected by the method suggested by Atkinson
et al.,16 using the rate constants k(CF3CH2CHF2 + Cl) and
k(CF3CHO + Cl) noted above. The yields of CF3CHO before
and after correcting for the secondary reaction with Cl atoms
are plotted against the conversion of CF3CH2CHF2 in Figure 7.
The plot after correction is linear, with a slope of 0.97 ( 0.05,
which strongly suggests that CF3CHO was a main primary
product and that degradation of CF3CH2CF2OO radicals pro-
duced a stoichometric amount of CF3CHO. This result is
consistent with reactions 5-9 following reaction 4.
CF CH O (+M) f CF + CH O (+M)
(11)
3
2
3
2
Atkinson and Carter summarized the thermodynamic and
kinetic properties of a series of alkoxy radicals and introduced
an empirical method for assessing the relative importance of
17
reactions 9 and 11 under atmospheric conditions. This method
utilizes the differences in the enthalpies of reactions, ∆(∆H) )
∆H11 - ∆H9), where ∆H11 and ∆H9 refer to enthalpies of
(
reactions 11 and 9, respectively. For ∆(∆H) < 40 kcal/mol,
the unimolecular dissociation (11) dominates over O2 reaction,
reaction 9, while the opposite is true for ∆(∆H) > 43 kcal/
mol, at room temperature and atmospheric pressure. ∆(∆H) is
calculated to be 47.2 kcal/mol for CF3CH2O radical. In the
calculation, the value for the enthalpy of formation of CF3CHO
was estimated to be -187.5 kcal/mol by using the empirical
2
CF CH CF OO f 2CF CH CF O + O
(5)
3
2
2
3
2
2
2
CF CH CF O (+M) f CF CH + CF O (+M) (6)
18
3
2
2
3
2
2
AM1 method, while those for other products of reactions 9
1
9
and 11 were found from the literature.
This estimation
CF CH + O (+M) f CF CH OO (+M)
(7)
(8)
(9)
3
2
2
3
2
supports that the O2 reaction with the CF3CH2O radicals
producing CF3CHO and HOO radicals, reaction 9, dominates
over the unimolecular dissociation, reaction 11, and is consistent
with the previous report by Nielsen et al. that >99.3% of the
CF3CH2O radicals react with O2 to produce CF3CHO.20
Another possible decomposition channel for CF3CH2O radical
2
CF CH OO f 2CF CH O + O
3 2 3 2 2
CF CH O + O f CF CHO + HOO
3
2
2
3
2
1
The self-reactions of CF3CH2OO radicals may proceed via
two possible channels, reactions 8 and 10. The absence of IR
peaks assignable to CF3CH2OH suggests that reaction 10 is of
is elimination of a hydrogen atom, reaction 12. Because the
HOO radical is about 50 kcal/mol more stable than the H atom,19
reaction 9 must be thermodynamically favored over reaction

2652 J. Phys. Chem. A, Vol. 101, No. 14, 1997
Chen et al.
1
2. Thus, CF3CH2O radical mainly degrades via reaction 9
be negligible compared to reaction 16. Using a global averaged
9
-3 24
under atmospheric conditions.
value for HO2 concentration of 10 molecule cm
and
assuming that the rate constant for reaction 17 was gas kinetic,
the rate of unimolecular decomposition, reaction 16, would be
at 5 orders of magnitude faster than the formation of trifluo-
roacetic acid, reaction 17. Therefore, the trifluoroacetic acid
formed in these experiments resulted from reactions subsequent
to the secondary reaction of Cl atoms with CF3CHO, reaction
CF CH O (+M) f CF CHO + H (+M)
(12)
3
2
3
Based on the above-discussed reaction mechanism, the yield
of CF2O should be equal to that of CF3CHO. Actually, the
observed percentage yield of CF2O was about 30-60% higher
than that of CF3CHO and increased with the reactant conversion.
Thus, part of the observed CF2O was probably produced by
reactions subsequent to the CF3CHO + Cl reaction. The
reaction of CF3CHO with Cl has been studied by several
groups1 and is expected to produce CF3CO radicals, reaction
13, and would not be an expected product from the atmospheric
degradation of 1,1,3,3-pentafluoropropane, HFC-245fa.
In this study, the CF3CH2O radicals were produced by the
self-reaction of two CF3CH2OO radicals, reaction 15, which is
a nearly thermoneutral process. Recent work by Wallington
et. al. indicated that when a similar alkoxy radical, CF3CFHO,
was generated by a more exothermic reaction between CF3-
CFHOO and NO (reaction 23), a “hot” CF3CFHO radical was
produced, and compared to the thermalized radical (reaction
1,22
1
3.
CF CHO + Cl f CF CO + HCl
(13)
3
3
The unimolecular dissociation of CF3CO radicals leading to CO
and CF3 radicals was confirmed to be a minor channel by the
result that less than 2% yield of CO was identified under the
experimental conditions. Wallington et al.22 studied the deg-
radation mechanism of CF3CO radicals in the presence of 27
mTorr of O2 and reported that 84% of the CF3CO radicals
reacted with O2 forming CF3C(O)OO radicals, reaction 14, and
27), the extent of unimolecular decomposition (reaction 24) was
increased over the competing hydrogen abstraction reaction
2
5
(reaction 25).
CF CHFOO + NO f CF CHFO* + NO
2
(23)
3
3
CF CHFO* + (M) f CF + CH(O)F + (M) (24)
3
3
1
6% decomposed to give CO and CF3 radicals. Because of
the presence of 140 Torr of O2, reaction 14 is the most likely
fate of CF3CO radicals at our experimental conditions, and thus
at atmospheric conditions.
CF CHFO + HOO f CF C(O)F + O
2
(25)
(26)
(27)
3
3
CF CHFO* (+M) f CF CHFO + (M)
3
3
CF CO + O (+M) f CF C(O)OO + (+M) (14)
3
2
3
CF CHFO (+M) f CF + CH(O)F + (M)
3
3
The identified product yields of CO2 (40-60%), CF3C(O)OH
5-8%), CF3OH (10-30%) changed with the reactant conver-
(
Bevilacqua et al.26 and Pultau et al.27 have also studied the
reaction of NO with fluoroperoxy radicals. Comparing the
reactivity of the fluoroalkoxy radicals produced by the reaction
of NO with the precursor fluoroperoxy radicals, the results
indicated that the CF3CHFO radical did undergo further carbon-
carbon bond cleavage which is not inconsistent with the results
sion. They and part of the CF2O were formed via a series of
reactions which follow reaction 14.
2
CF C(O)OO f 2CF C(O)O + O
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
3
3
2
CF C(O)O (+M) f CF + CO (+M)
3
3
2
25
of Wallington et al., whereas the CF3CH2O did not undergo
further decomposition. Therefore, based on results from these
two studies, it is unlikely that there would have been any
significant unimolecular decomposition of CF3CH2O, reaction
1, even if the CF3CH2O radicals were produced in this study
by the reaction of NO with the peroxy radical, CF3CH2OO.
CF C(O)O + HOO f CF C(O)OH + O
3
3
2
CF + O (+M) f CF OO (+M)
3
2
3
1
2
CF OO f 2CF O + O
3 3 2
Conclusion
CF O + HOO f CF OH + O
2
3
3
Our investigation of Cl atom initiated oxidation of CF3CH2-
CHF2 (HFC-245fa) in 700 Torr of air suggests that terminal H
abstraction producing CF3CH2CF2 radicals will be the predomi-
nant channel during the tropospheric oxidation of CF3CH2CHF2
by OH radicals. At atmospheric conditions, self-reactions
(reactions 5, 8, 15, and 19) of the peroxy radicals produced via
the subsequent reactions of CF3CH2CF2 radicals are negligible.
CF O + RH f CF OH + R
3
3
CF OH + wall f CF O + HF + wall
3
2
Scollard et al. studied the degradation mechanisms of a series
CX3C(O)O radicals (X ) Cl and F) and reported that the radicals
mainly dissociate to give CO2 and CX3 radicals.11 In the present
experiments, the observed yield of CO2 was about 8 times larger
than that of CF3C(O)OH. Thus, the unimolecular dissociation
of CF3C(O)O radicals leading to CO2 and CF3 radicals, reaction
9
By analogy to the behavior of other peroxy radicals, it is
expected that the rate constants of NO with the peroxy radicals
-12
3
-1
involved in this study are larger than 5 × 10 cm molecule
-
1
s . With an estimated background tropospheric NO concentra-
9
-3 28,29
1
1
6, dominates over the reaction with HOO radicals, reaction
7. The sum of observed percentage yields of CF3CHO, CO2,
tion of about 1 × 10 molecule cm ,
the lifetimes of these
peroxy radicals with respect to NO reactions are estimated to
be 2-3 min. Thus, at atmospheric conditions, reactions with
NO producing the corresponding alkoxy radicals are likely to
be the main degradation channels for these peroxy radicals. The
present study has shown that the fate of CF3CH2CF2O radicals
is unimolecular dissociation leading to CF2O and CF3CH2
radicals. The observation of CF3CHO as a main primary
product and the absence of CH2O suggest that the atmospheric
and CF3C(O)OH is very similar to the yield of CF3CHO after
correcting for the secondary reaction with Cl atoms, as shown
in Figure 7. This result is consistent with the above reaction
mechanism. Since the lifetime for the unimolecular decomposi-
-6
tion for CF3C(O)O is 8.3 × 10 s as determined by Wallington
2
3
et al. at 760 Torr, under atmospheric conditions, the rate for
reaction 17 forming CF3C(O)OH (trifluoroacetic acid) would

Reactions of CF3CH2CHF2 (HFC-245fa)
J. Phys. Chem. A, Vol. 101, No. 14, 1997 2653
fate of CF3CH2O radicals is to react with O2 producing CF3-
CHO and HOO radicals, while the unimolecular dissociation
leading to CF3 and CH2O is negligible. The atmospheric
degradation of CF3CHO will produce CO2 and CF3 radicals as
(10) Wallington, T. J.; Sehested, J.; Dearth, M. A.; Hurley, M. D. J.
Photochem. Photobiol. A: Chem. 1993, 70, 5.
(11) Scollard, D. J.; Treacy, J. J.; Sidebotton, H. W.; Balestra-Garcia,
C.; Laverdet, G.; LeBras, G.; MacLeod, H.; Teton, S. J. Phys. Chem. 1993,
97, 4683.
the main products.
(12) Francisco, J. S.; Williams, I. M. Mol. Phys. 1992, 76, 1433.
(13) Berney, C. V. Spectrochim. Acta. 1969, 25A, 793.
The atmospheric fate of CF3 radicals has been well studied.5,30
(14) Redington, R. L.; Lin, K. C. Spectrochim. Acta. 1971, 27A, 2445.
(15) Sehested, J.; Wallington, T. J. EnViron. Sci. Technol. 1993, 27, 1.
(16) Atkinson, R.; Aschmann, S. M.; Carter, W. P. L.; Winer, A. M.;
The main terminating processes will be reaction with NO and
CH4 which produce, respectively, CF2O and CF3OH.
most probable fate for both CF2O and CF3OH would be uptake
by cloud, rain, or ocean water to yield CO2 and HF.
Consequently, no long-term decomposition products such as
trifluoroacetic acid (TFA) are expected to result from the
atmospheric degradation of CF3CH2CHF2(HFC-245fa).
8,31,32
The
Pitts, J. N. Jr. J. Phys. Chem. 1982, 86, 4563.
(17) Atkinson, R.; Carter, W. J. Atmos. Chem. 1991, 13, 195.
3
3
(
18) Dewar, M. J. S.; Zoebisch, E. G.; Heuly, E. F. J. Am. Chem. Soc.
985, 107, 3902.
19) Data from the following publications were used for the calcula-
tions: (a) Shum, L. G. S.; Benson, S. W. J. Phys. Chem., 1983, 87, 3479.
b) Cox, J. D., Wagman, D. D., Medvedev, V. A., Eds. CODATA. Key
Values for Thermodynamic; Hemisphere Publishing Corp.: New York, 1989.
c) Handbook of Chemistry and Physics, 75th ed.; Lide, D. R., Ed.; CRC
Press: Boca Raton, FL, 1995.
20) Nielsen, O. J.; Gamborg, E.; Sehested, J.; Wallington, T. J.; Hurley,
M. D.; J. Phys. Chem. 1994, 98, 9518.
21) Francisco, J. S.; Li, Z.; Bradley, A.; Knight, A. E. W. Chem. Phys.
Lett. 1993, 214, 77.
22) Wallington, T. J.; Sehested, J.; Nielsen, O. J. Chem. Phys. Lett.
994, 226, 563.
23) Wallington, T. J.; Hurley, M. D; Nielsen, O. J.; Sehested, J.; J.
Phys. Chem. 1994, 98, 5686.
24) Atkinson, R. In “Scientific Assessment of Stratospheric Ozone:
1
(
(
Acknowledgment. We thank AlliedSignal and British Gas/
Consumers Gas for financial support. The late H. Niki was
the holder of the British Gas/Consumers Gas/AES Industrial
Research Chair in Atmospheric Chemistry. We also thank Dr.
M. Van Der Puy from the Buffalo Research Laboratory of
AlliedSignal for synthesizing CF3CH2CC1F2, and Drs. O. Melo
and G. Harris for critical review of the manuscript.
(
(
(
(
1
(
References and Notes
(
(1) “Scientific Assessment of Stratospheric Ozone”. World Meteoro-
1989”, World Meteorological Organization Global Ozone Research and
Monitoring Project; Report No. 20, Volume II, Appendix: AFEAS Report,
Geneva, Switzerland, 1989; p 167.
logical Organization Global Ozone Research and Monitoring Project, Report
No. 20; 1989, Vol. I.
(
2) Edney, E. O.; Gay, B. W.; Driscoll, D. J. J. Atmos. Chem. 1991,
2, 105.
3) Wallington, T. J.; Hurley, M. D.; Ball, J. C.; Kaiser, E. W. EnViron.
Sci. Technol. 1992, 26, 1318.
4) Nielsen, O. J.; Ellerman, T.; Sehested, J.; Bartkiewicz, E.; Wall-
ington, T. J.; Hurley, M. D. Int. J. Chem. Kinet. 1992, 24, 1009.
5) Wallington, T. J.; Schneider, W. F.; Worsnop, D. R.; Nielsen, O.
J.; Sehested, J.; Debruyn, W. J.; Shorter, J. A. EnViron. Sci. Technol. 1994,
8, 320 (and references therein).
6) (a) “Montreal Protocol on Substances That Deplete the Ozone
(
25) Wallington, T. J.; Hurley, M. D; Fracheboud, J. M.; Orlando, J. J.;
Tyndall G. S.; Sehested, J.; Møgelberg, T. E.; Nielsen, O. J.; J. Phys. Chem.
996, 100, 18116.
26) Bevilacqua, T. J., Hanson, D. R.; Jensen, N. R.; Howard, C. J. In
Kinetics and Mechanisms for the Reactions of Halogenated Organic
1
(
1
(
(
“
Compounds in the Troposphere”, STEP-HALOCIDE/AFEAS Workshop,
(
Dublin, Ireland, March 23-25, 1993; p 163.
(27) Pultau, V.; Peeters, J. In “Kinetics and Mechanisms for the
2
Reactions of Halogenated Organic Compounds in the Troposphere”, STEP-
HALOCIDE/AFEAS Workshop, Dublin, Ireland, March 23-25, 1993; p
4.
(
Layer”. UNEP, 1994 Report of the Flexible and Rigid Foams Technical
Options Committee; Chapter III: Polyurethane Foams. (b) Knopeck, G.
M.; Parker, R. C.; Richard, R. G.; Shankland, I. R. Proc. SPI/C 48th Annu.
Conf. 1993, 465. (c) Knopeck, G. M., Parker, R. C.; Richard, R. G.;
Shankland, I. R. Proc. 35th Annu. Polyurethane Tech./Marketing Conf.,
Oct. 9-12, 1994, 115.
3
(28) Logan, J. A. J. Geophys. Res. 1983, 88, 10785.
(29) Bollinger, M. J.; Hahn, C. J.; Parrish, D. D.; Murphy, P. C.;
Albrittom, D. L.; Fehsenfeld, F. C. J. Geophys. Res. 1984, 89, 9623.
(
(
30) Niki, H.; Chen J.; Young, V. Res. Chem. Intermed. 1994, 20, 277.
31) Barone, S. B.; Turnipseed, A. A.; Ravishankara, A. R. J. Phys.
(7) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. J. Mol.
Chem. 1994, 98, 4594.
(32) Turnipseed, A. A.; Barone, S. B.; Ravishankara, A. R. J. Phys.
Chem. 1994, 98, 4604.
Struct. 1980, 59, 1.
(
(
8) Chen, J.; Zhu, T.; Niki, H. Geophys. Res. Lett. 1992, 19, 2215.
9) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. “Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling”, Evaluation No. 11, NASA and JPL, December, 15, 1994.
(33) “Scientific Assessment of Ozone Depletion”; World Meteorological
Organization Global Research and Monitoring Project, Report No. 37, 1994;
Chapter 12: Atmospheric Degradation of Halocarbon Substitutes; p 12.14.