Kinetics and products of chlorine atom initiated oxidation of HCF 2OCF2OCF2CF2OCF2H and HCF20(CF20)n(CF2CF20)mCF 2H

Article abstract of DOI:10.1002/kin.20349

Smog chamber/FTIR techniques were used to measure κ(Cl + HCF ₂OCF₂OCF₂-CF₂OCF₂H) = κ(Cl + HCF₂0(CF₂0)n(CF₂CF ₂0)mCF₂H) = (5.0 ± 1.4) × 10^-17 cm³ molecule^-1 s^-l in 700 Torr of N ₂/0₂ diluent at 296 ± 1 K. The Cl-initiated atmospheric oxidation of HCF₂OCF₂OCF₂CF ₂OCF₂H and the sample of HCF₂0(CF ₂0)n(CF₂CF₂0)MCF₂H used in this work gave COF₂ in molar yields of (476 ± 36)% and (859 ± 63)%, respectively, with no other observable carbon containing products (i.e., essentially complete conversion of both hydrofluoropolyethers into COF ₂). The results are discussed with respect to the atmospheric chemistry and environmental impact of hydrofluoropolyethers of the general formula HCF₂0(CF₂0)n(CF₂CF₂0) mCF₂H.

Full text of DOI:10.1002/kin.20349

Kinetics and Products of
Chlorine Atom Initiated
Oxidation of
HCF2OCF2OCF2CF2OCF2H
and HCF2O(CF2O)n-
(CF2CF2O)mCF2H
T. J. WALLINGTON,1 M. D. HURLEY,1 M. S. JAVADI,2 O. J. NIELSEN2
1System Analytics and Environmental Sciences Department, Ford Motor Company, Mail Drop RIC-2122, Dearborn,
MI 48121-2053
2Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark
Received 4 January 2008; revised 13 April 2008; accepted 15 April 2008
DOI 10.1002/kin.20349
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Smog chamber/FTIR techniques were used to measure k(Cl + HCF2OCF2OCF2-
CF2OCF2H) = k(Cl + HCF2O(CF2O)n(CF2CF2O)mCF2H) = (5.0 1.4) × 10−17 cm3 molecule−1
s−1 in 700 Torr of N2/O2 diluent at 296
1 K. The Cl-initiated atmospheric oxidation
of HCF2OCF2OCF2CF2OCF2H and the sample of HCF2O(CF2O)n(CF2CF2O)mCF2H used in
this work gave COF2 in molar yields of (476 36)% and (859 63)%, respectively, with
no other observable carbon containing products (i.e., essentially complete conversion of
both hydrofluoropolyethers into COF2). The results are discussed with respect to the atmo-
spheric chemistry and environmental impact of hydrofluoropolyethers of the general formula
HCF2O(CF2O)n(CF2CF2O)mCF2H. ꢀ 2008 Wiley Periodicals, Inc. Int J Chem Kinet 40: 819–825,
C
2008
with environmentally acceptable alternatives. Per-
fluoropolyethers (PFPEs) and hydrofluoropolyethers
(HFPEs) have been used as replacements for CFCs
as heat transfer fluids and refrigeration working fluids.
PFPEs and HFEs do not contain chlorine and there-
fore do not contribute to chlorine-based catalytic de-
struction of stratospheric ozone. PFPEs do not contain
hydrogen atoms and do not react in the lower atmo-
sphere. The atmospheric lifetimes of PFPEs are de-
termined by photolysis in the upper atmosphere and
are expected to be long, probably more than 800 years
INTRODUCTION
Recognition of the adverse environmental impact of
chlorofluorocarbon (CFC) release into the atmosphere
[1,2] has led to an international effort to replace CFCs
Correspondence to: T. J. Wallington; e-mail: twalling@
ford.com; O. J. Nielsen; e-mail: ojn@kiku.dk.
Contract grant sponsor: Danish Natural Science Research
Council.
c
ꢀ
2008 Wiley Periodicals, Inc.

820
WALLINGTON ET AL.
[3]. The long lifetimes of HFPEs combined with their
large instantaneous forcings result in these compounds
having high global warming potentials (GWPs). HF-
PEs contain one or more carbon–hydrogen bonds and
react with OH radicals in the troposphere. The reaction
with OH radicals reduces the atmospheric lifetime of
HFPEs compared to PFPEs and CFCs and thus reduces
their GWPs.
smog chamber/FTIR techniques were used to deter-
mine the kinetics and products of the chlorine atom ini-
tiated oxidation of HCF2OCF2OCF2CF2OCF2H and
HCF2O(CF2O)n(CF2CF2O)mCF2H. The results are re-
ported herein.
EXPERIMENTAL
HCF2O(CF2O)n(CF2CF2O)mCF2H + OH
→ HCF2O(CF2O)n(CF2CF2O)mCF2 + H2O (1a)
HCF2O(CF2O)n(CF2CF2O)mCF2H + OH
Experiments were performed in a 140-L Pyrex reactor
interfaced to a Mattson Sirus 100 FTIR spectrome-
ter [10]. The reactor was surrounded by 22 fluores-
cent blacklamps (GE F15T8-BL), which were used
to photochemically initiate the experiments. The loss
of HFPE and formation of products were monitored
by FTIR spectroscopy. IR spectra were derived from
32 coadded interferograms with a spectral resolu-
tion of 0.25 cm−1 and an analytical path length of
27.4 m.
→ CF2O(CF2O)n(CF2CF2O)mCF2H + H2O (1b)
The alkyl radicals produced in reaction (1) add O2
rapidly to give peroxy radicals,
HCF2O(CF2O)n(CF2CF2O)mCF2 + O2 + M
→ HCF2O(CF2O)n(CF2CF2O)mCF2OO + M (2a)
CF2O(CF2O)n(CF2CF2O)mCF2H + O2 + M
→ OOCF2O(CF2O)n(CF2CF2O)mCF2H + M (2b)
Cl atoms were generated by the photolysis of molec-
ular chlorine in air diluent,
Cl2 + hv → 2Cl
(7)
HCF2OCF2OCF2CF2OCF2H and HCF2O(CF2O)n-
(CF2CF2O)mCF2H were obtained from the Solvay
Chemical Company (Brussels, Belgium). The stated
purity of HCF2OCF2OCF2CF2OCF2H was more than
99%. HCF2O(CF2O)n(CF2CF2O)mCF2H was supplied
as a mixture with n values in the range 0–5 and m values
in the range 0–7, the mixture had an average molecular
weight of 572. A GC-MS analysis revealed the pres-
ence of nine major components of the HCF2O(CF2O)n-
(CF2CF2O)mCF2H mixture. The mass spectrum of
each of the+components had prominent+m/z peaks
at 51 (CHF2 ), 97 (C2F3O+), 100 (C2F4 ), and 213
(C4F7O+2 ), which are consistent with the general
formula of HCF2O(CF2O)n(CF2CF2O)mCF2H but do
not allow a specific identification of the components.
It is rather unusual in experimental studies such as the
present to work with a reactant, which is a mixture of
different compounds and this deserves some comment.
In the present work, we used a sample that was a mix-
ture of at least nine different compounds of the general
formula HCF2O(CF2O)n(CF2CF2O)mCF2H. It is well
known that CF2O and CF2CF2O units are un-
reactive toward Cl atoms (and OH radicals, important
in the atmosphere). The reactivity of HCF2O(CF2O)n-
(CF2CF2O)mCF2H is confined to the C H bonds on
either end of the molecule. The chemical environment
of the C H bonds is expected to be independent of
the number of CF2O and CF2CF2O units in the
molecule. For the purposes of the present study, we
The peroxy radicals will react with NO, NO2,
HO2, or CH3O2 [4–6], for HCF2O(CF2O)n-
(CF2CF2O)mCF2OO radicals, which we abbreviate to
RfOCF2OO:
RfOCF2OO + NO → RfOCF2O + NO2 (3a)
RfOCF2OO + NO → RfOCF2ONO2
(3b)
RfOCF2OO + NO2 → RfOCF2OONO2
(4)
RfOCF2OO + HO2 → RfOCF2OOH + O2 (5)
RfOCF2OO + CH3O2
→ RfOCF2O + CH3O + O2
RfOCF2OO + CH3O2
(6a)
→ RfOCF2OH + HCHO + O2
(6b)
The importance of these reactions is determined by the
relative abundances of NO, NO2, HO2, and CH3O2 and
the rates of reactions (3)–(5). Alkyl peroxynitrates are
thermally unstable and decompose rapidly to regener-
ate the peroxy radicals and NO2 [7,8]. The hydroper-
oxide is returned to the peroxy radicals via reaction
with OH or photolysis [9].
Prior to their large-scale industrial use, an assess-
ment of the atmospheric chemistry, and hence environ-
mental impact, of HFPEs is needed. To improve our
understanding of the atmospheric chemistry of HFPEs
International Journal of Chemical Kinetics DOI 10.1002/kin

KINETICS AND PRODUCTS OF CHLORINE ATOM INITIATED OXIDATION OF HFPEs
821
0.30
assume that we can treat the mixture as a simple com-
pound, and the validity of this assumption is discussed
in the discussion and atmospheric chemistry section.
The samples of HCF2O(CF2O)n(CF2CF2O)mCF2H
and HCF2OCF2OCF2CF2OCF2H were subjected
to freeze–pump–thaw cycling prior to use. All
0.25
0.20
0.15
0.10
0.05
0.00
CF3CF2H
t
experiments were performed at 296
1 K in 700
t
Torr total pressure of N2/O2 diluent. The decay of
the reactants and references were measured using
their characteristic absorptions in the infrared over
the following wavelength ranges (in cm−1): HCF2O-
(CF2O)n(CF2CF2O)mCF2H/HCF2OCF2OCF2CF2OC-
F2H 1015–1045 and 1350–1435; COF2 773–775;
CF2ClCH3 850–950; and CF3CF2H 710–730.
The relative rate method is a well-established tech-
nique for measuring the reactivity of Cl atoms with
organic compounds. Kinetic data were derived by mon-
itoring the loss of HFPE relative to one, or more, ref-
erence compounds. Providing that the HFPE and the
reference are lost only by the reaction with chlorine
atoms, and neither HFPE nor the reference compounds
are reformed in any process then it can be shown
that
CF2ClCH3
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
ln([Reference]t0 /[Reference]t)
Figure 1 Decay of HCF2O(CF2O)n(CF2CF2O)mCF2H
(circles) and HCF2OCF2OCF2CF2OCF2H (triangles) versus
CF3CF2H (open symbols) and CH3CF2Cl (filled symbols)
in the presence of Cl atoms in 700 Torr total pressure of air.
Cl + CF3CF2H → products
Cl + CH3CFlCl → products
(10)
(11)
ꢀ
ꢁ
ꢀ
ꢁ
[HFPE]t
kHFPE
[reference]t
0
0
ln
=
ln
[HFPE]t
kreference
[reference]t
Reaction mixtures consisted of 7.0–12.2 mTorr
HFPE, 433–497 mTorr Cl2, and either 2.6–4.2 mTorr
CF3CF2H or 3.2–6.1 mTorr CH3CF2Cl in 700 Torr
of air diluent. The observed loss of the HFPEs ver-
sus those of reference compounds is shown in Fig. 1.
As seen from Fig. 1, there was no discernible dif-
ference in the reactivity of the two HFPEs studied.
Linear least-squares analyses of the composite data
set in Fig. 1 give k8/k10 = k9/k10 = 0.205 0.019 and
k8/k11 = k9/k11 = 0.124 0.021, where the uncertain-
ties include uncertainties in the IR analysis and two
standard deviations from the regression analysis. Us-
where [HFPE]t , [HFPE]t , [reference]t , and
0
0
[reference]t are the concentrations of the HFPE and
reference at times t0 and t, and kHFPE and kreference are
the rate constants for reactions of the HFPE and the ref-
erence with Cl atoms. Plots of ln([HFPE]t /[HFPE]t )
0
versus ln([reference]t /[reference]t ) should be lin-
0
ear, pass through the origin and have a slope of
kHFPE/kreference. The uncertainties reported in this
paper are two standard deviations unless stated
otherwise.
ing literature values of k10 = (2.5
0.6) × 10−16
[12] and k11 = (3.90 0.52) × 10−16 [11] (based on
k(Cl + CH4) = 1.0 × 10−13) gives k8 = k9 = (5.1
1.3) × 10−17 and k8 = k9 = (4.8 1.0) × 10−17 cm3
molecule−1 s−1, respectively. We choose to cite a final
value, which is the average of the individual determi-
nations together with error limits that encompass the
extremes of the determinations, hence, k8 = k9 = (5.0
RESULTS
Kinetics of the Reaction of Cl Atoms
with HFPEs
Relative rate experiments were performed using the
FTIR system to investigate the kinetics of reactions (8)
and (9). The techniques used are described in detail
elsewhere [11]. The rates of reactions (8) and (9) were
measured relative to reactions (10) and (11):
1.4) × 10−17 cm3 molecule−1 s−1
.
Products of HFPE Oxidation
The atmospheric degradation mechanism was stud-
ied using the UV irradiation of HCF2O(CF2O)n-
(CF2CF2O)mCF2H/Cl2/O2/N2 mixtures. Reaction
mixtures consisted of 0.79–3.89 mTorr HCF2O-
Cl + HCF2O(CF2O)n(CF2CF2O)mCF2H → products
(8)
Cl + HCF2OCF2OCF2CF2OCF2H → products
(9)
International Journal of Chemical Kinetics DOI 10.1002/kin

822
WALLINGTON ET AL.
25
20
15
10
5
0.4
A: before UV
0.2
0.0
0.4
0.2
B: after UV
C: B - 0.76* A
D: COF2
0.0
0.2
0.1
0.0
0.6
0.4
0.2
0.0
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Δ [HCF2O(CF2O)n(CF2CF2O)mCF2H] (mTorr)
800
1000
1200
1400
1600
1800
2000
Figure
3
Formation of COF2 versus the loss of
Wavenumber (cm )
HCF2O(CF2O)n(CF2CF2O)mCF2H
(triangles)
and
Figure 2 IR spectra of a mixture consisting of 3.67 m
Torr HCF2OCF2OCF2CF2OCF2H and 4.11 Torr Cl2 in 700
Torr total pressure of N2/O2 diluent before (A) and after (B)
23 min UV irradiation. Subtraction of features attributed to
HCF2OCF2OCF2CF2OCF2H from panel B gives the prod-
uct spectra shown in panel C. A reference spectrum of COF2
is shown in panel D. The small structured features at 1300–
1800 cm−1 in panels B and C are attributable to the presence
of small amounts of H2O in the IR beam.
HCF2OCF2OCF2CF2OCF2H (circles) observed following
UV irradiation of mixtures consisting of 1.23–3.89 m
Torr HCF2O(CF2O)n(CF2CF2O)mCF2H and 4.44–4.75
Torr Cl2 in 700 Torr total pressure of N2/O2 diluent at
296
1 K and mixtures consisting of 1.94–7.46 mTorr
HCF2OCF2OCF2CF2OCF2H and 4.06–4.24 Torr Cl2 in
700 Torr total pressure of N2/O2 diluent.
HCF2O(CF2O)n(CF2CF2O)mCF2H and (476 36)%
for the oxidation of HCF2OCF2OCF2CF2OCF2H, re-
spectively. Quoted uncertainties include two standard
deviations from the regression analysis and 5% uncer-
tainties in the calibration of COF2 and HFPE spec-
tra. Within the experimental uncertainties, the ob-
served formation of COF2 accounts for the entire
loss of the HCF2OCF2OCF2CF2OCF2H. Given the
absence of any other product features, we also con-
clude that COF2 accounts for the entire loss of the
HCF2O(CF2O)n(CF2CF2O)mCF2H.
(CF2O)n(CF2CF2O)mCF2H or 3.7–7.5 mTorr HCF2-
OCF2OCF2CF2OCF2H, 4.1–5.1 Torr Cl2, and 71–84
Torr of O2 in 700 Torr of N2 diluent. Figure 2 shows
IR spectra at 800–2000 cm−1 obtained before (panel
A) and after (panel B) subjecting a mixture containing
3.7 mTorr HCF2OCF2OCF2CF2OCF2H and 4.1 Torr
Cl2 in 700 Torr of N2/O2 to UV irradiation for
23 min. Subtraction of the IR features attributed to
HCF2OCF2OCF2CF2OCF2H from panel B gives
the product spectrum in panel C. Comparison of
the product features in panel C with the reference
spectrum in panel D shows the formation of COF2.
Figure 3 shows a plot of the observed forma-
tion of COF2 versus the loss of HCF2O(CF2O)n-
(CF2CF2O)mCF2H and HCF2OCF2OCF2CF2OCF2H.
As seen from Fig. 3, the formation of COF2 scaled
linearly with the HFPE loss over the range studied (up
to 71% consumption of HCF2O(CF2O)n(CF2CF2O)m-
CF2H and up to 40% consumption of HCF2OCF2-
OCF2CF2OCF2H). The linearity of the formation
of COF2 shown in Fig. 3 suggests the absence
of significant loss of COF2 via secondary reac-
tions in the chamber. No other carbon-containing
products were discernible in the IR spectra. The
lines through the data in Fig. 3 give molar yields
DISCUSSION AND ATMOSPHERIC
CHEMISTRY
The reaction of Cl atoms with HCF2O(CF2O)n-
(CF2CF2O)mCF2H occurs via hydrogen abstraction
from the terminal CF2H groups:
Cl + HCF2O(CF2O)n(CF2CF2O)mCF2H
→ CF2O(CF2O)n(CF2CF2O)mCF2H + HCl
(12a)
Cl + HCF2O(CF2O)n(CF2CF2O)mCF2H
→ HCF2O(CF2O)n(CF2CF2O)mCF2 + HCl
(12b)
of COF2 of (859
63)% for the oxidation of
International Journal of Chemical Kinetics DOI 10.1002/kin

KINETICS AND PRODUCTS OF CHLORINE ATOM INITIATED OXIDATION OF HFPEs
823
The radicals formed in reaction (12) will add O2 to
give peroxy radicals (reaction (2)), which will undergo
self- or cross-reactions to produce the corresponding
alkoxy radicals. The alkoxy radicals are expected to
decompose via the elimination of a COF2 molecule
and initiate a sequence of reactions in which the radi-
cal “unzips” by shedding successive COF2 molecules.
In contrast to the situation in the experimental cham-
ber, the concentrations of fluorinated peroxy radicals
in the atmosphere will be extremely small, and the
self- and cross-reactions described above will not be
of atmospheric significance. In the atmosphere, the
fate of the fluorinated peroxy radicals will be reaction
with NO, NO2, HO2, or CH3O2 radicals (reactions
(3)–(6)). Reaction with NO gives the corresponding
alkoxy radical and NO2 as major products with the
fluorinated organic nitrate as a minor product. Reac-
tion with NO2 gives a thermally unstable peroxyni-
trate whose fate is decomposition to reform NO2 and
the peroxy radical. Reaction with HO2 radicals gives
a hydroperoxide, which will be returned back to the
fluorinated peroxy radical pool via reaction with OH
radicals. Reaction with CH3O2 radicals is expected to
proceed via two channels ((6a) and (6b)) leading to
the formation of alkoxy radicals (6a) and a fluorinated
alcohol and formaldehyde (6b). The atmospheric fate
of the fluorinated alcohol is expected to be heteroge-
neous elimination of HF followed by hydrolysis of
the acid fluoride to give a carboxylic acid, for ex-
ample, from HCF2OCF2OCF2CF2OCF2H one might
expect formation of HCF2OCF2OCF2CF2OC(O)OH
and HOC(O)OCF2OCF2CF2OCF2H. Long-chain per-
fluorocarboxylic acids, such as perfluorooctanoic acid
(C7F15C(O)OH), have no natural sources, are bioaccu-
mulative, are found in fauna in remote locations, and
are the subject of considerable current research inter-
est. The possibility that small amounts of fluorinated
acidic compounds may be formed during the atmo-
spheric oxidation of HFPEs is interesting and merits
further study but is beyond the scope of the present
work.
Tuazon used a relative rate method similar to that used
in the present work and reported k13/k10 = 0.208
0.006, k14/k10 = 0.188 0.004, and k15/k10 = 0.149
0.003. We can compare these results to the deter-
mination of k8/k10 = k9/k10 = 0.205
0.019 in the
present work. The reactivities of HCF2OCF2OCF2H
and HCF2OCF2CF2OCF2H toward chlorine atoms re-
ported by Tuazon are indistinguishable, within the ex-
perimental uncertainties, from those for the HFPEs
studied in the present work. For reasons which are
unclear, the reactivity of HCF2OCF2OCF2CF2OCF2H
reported by Tuazon is approximately 25% below those
of the other HFPEs.
Tuazon reported the formation of COF2 in a
near unit yield per carbon atom from the chlo-
rine atom initiated oxidation of HCF2OCF2OCF2H,
HCF2OCF2CF2OCF2H, and HCF2OCF2OCF2CF2-
OCF2H in 740 Torr of air at 298 K. Cavalli et al.
[13] conducted four experiments of the chlorine atom
initiated oxidation of HCF2OCF2OCF2CF2OCF2H in
740 Torr of air and reported the formation of COF2
in a molar yield of 300%–485%. The large range
reported by Cavalli et al. presumably reflects dif-
ficulties in measuring the small amounts of COF2
formed (0.16–0.74 mTorr) and HFPE lost (0.03–0.24
mTorr) in their experiments; approximately an order
of magnitude smaller than in the present work (see
Fig. 3). The measurement of a molar COF2 yields
of 476
36% in the chlorine atom oxidation of
HCF2OCF2OCF2CF2OCF2H in the present work is
consistent with expectations based on the previous
work by Tuazon [6] and Cavalli et al. [13].
The atmospheric oxidation of HFPEs is initiated by
reaction with OH radicals. The kinetic database for
reactions of HFPEs with OH radicals is sparse. The
reported kinetic data for reactions of Cl atoms and OH
radicals with HFPEs can be compared to expectations
based upon the correlation of the rates of abstraction
of hydrogen atoms from hydrofluorocarbons and hy-
drofluorochlorocarbons reported by Sulbaek Andersen
et al. [14]. The circles in Fig. 4 (adapted from Fig. 9
in Sulbaek Andersen et al. [14]) are the data used by
Sulbaek Andersen et al. [14] to derive the relation-
The results from the present work can be com-
pared with those from the study by Tuazon [6]
of the kinetics and products of the chlorine atom
initiated oxidation of the hydrofluoropolyethers
HCF2OCF2OCF2H, HCF2OCF2CF2OCF2H, and
HCF2OCF2OCF2CF2OCF2H.
ship log (k(OH)) = (0.412
0.049) × log(k(Cl)) –
(8.16 0.72), which is shown as the solid line. The
dotted lines show variation from the solid line by a
factor of 2. It can be seen from Fig. 4 that the reactivity
of most saturated hydrofluorocarbons and hydrofluo-
rochlorocarbons toward OH radicals can be predicted
to within a factor of 2 from their reactivity toward
chlorine atoms. The stars in Fig. 4 show the literature
data for polyfluorinated HFEs and HFPEs (see Table I).
From the very limited data, it appears that the trend of
reactivity of HFEs/HFPEs toward OH radicals and Cl
Cl + HCF2OCF2OCF2H → products
Cl + HCF2OCF2CF2OCF2H → products
(13)
(14)
Cl + HCF2OCF2OCF2CF2OCF2H → products
(15)
International Journal of Chemical Kinetics DOI 10.1002/kin

824
WALLINGTON ET AL.
Table I Kinetic Data for HFEs and HFPEs
Compound
k(OH)
Reference
k(Cl)
Reference
HCF2OCF2H
2.4 × 10−15
2.4 × 10−15
4.6 × 10−15
4.9 × 10−16
1.2 × 10−14
1.1 × 10−14
1.2 × 10−14
[15]
[13]
[13]
[15]
[15]
[15]
[15]
5.73 × 10−14
5.0 × 10−17
5.0 × 10−17
2.2 × 10−17
1.36 × 10−14
1.1 × 10−13
1.3 × 10−14a
[16]
[6]
This work
[17]
HCF2OCF2OCF2H
HCF2OCF2OCF2CF2OCF2H
CF3OCF2H
CH3OCF3
CH3OC2F5
[18]
[19]
[20]
[21]
CHF2OCH2CF3
[22]
[23]
CH3OCF2CHF2
2.2 × 10−14
[15]
2.49 × 10−13
a Average from available studies.
straction, it is difficult to understand why the rates of
these reactions are not better correlated. The scatter
of the HFE/HFPE data in Fig. 4 suggests that per-
haps there are differences in the mechanism of the
OH and Cl reactions, or perhaps that there are er-
rors in the kinetic database. Given the similar chem-
ical environment of the C H bond in HCF2OCF2H,
HCF2OCF2OCF2H, and HCF2OCF2CF2OCF2H, it is
not surprising that these compounds react with OH
radicals at similar rates, but it is puzzling that the reac-
tivity of HCF2OCF2H toward chlorine atoms is a fac-
tor of 1000 times greater than for HCF2OCF2OCF2H
and HCF2OCF2CF2OCF2H. Further experimental and
computational studies are needed to improve our lim-
ited understanding of the kinetics of the reactions of
Cl atoms and OH radicals with HFPEs and hence the
atmospheric lifetimes of these species.
The atmospheric oxidation of HCF2O(CF2O)n-
(CF2CF2O)mCF2H and HCF2OCF2OCF2CF2OCF2H
gives carbonyl fluoride, COF2, as the major prod-
uct. COF2 is removed from the troposphere via con-
tact with water surfaces and hydrolysis to HF and
CO2 in rain/aerosol/cloud/sea water with a lifetime
of approximately 5–10 days [4]. HCF2O(CF2O)n-
(CF2CF2O)mCF2H and HCF2OCF2OCF2CF2OCF2H
do not contain any chlorine and will not con-
tribute to stratospheric ozone depletion via the well-
established chlorine-based chemistry. As with all hy-
drofluorocarbons (HFCs), hydrofluoroethers (HFEs),
and hydrofluoropolyethers (HFPEs), the ozone de-
pletion of HCF2O(CF2O)n(CF2CF2O)mCF2H and
HCF2OCF2OCF2CF2OCF2H is, for all practical pur-
poses, zero.
CH3OCF2CHF2
CHF2OCH2CF3
HCF2OCF2OCF2CF2OCF2H
CF3OCH3
C2F5OCH3
k
HCF2OCF2H
HCF2OCF2OCF2H
CF3OC(CF3)2H
CF3OCF2CF2H
CF3CFHCF2OCF2H
CF3CFHCF2OCF3
CF3OCF2H
log (k(Cl))
Figure 4 log(k(OH)) versus log(k(Cl)) for saturated hy-
drofluoroethers and hydrofluoropolyethers (stars), and hy-
drofluorocarbons and hydrofluorochlorocarbons (circles).
The solid line is the expression reported by Sulbaek Andersen
et al. [14], the dotted lines show variation by a factor of 2
from this expression. Data for hydrofluorocarbons, hydroflu-
orochlorocarbons, CF3OC(CF3)2H, CF3CFHCF2OCF3,
CF3CFHCF2OCF2H, and CF3OCF2CF2H were taken from
Sulbaek Andersen et al. [14]. All other data were taken from
Table I.
atoms is broadly similar to that for HFCs and HCFCs.
However, there is substantial and somewhat surpris-
ing scatter of the HFE/HFPE data in Fig. 4. For ex-
ample, CF3OCF2H and HCF2OCF2CF2OCF2H react
with Cl atoms at similar rates but react with OH rad-
icals at rates that differ by approximately an order of
magnitude. HCF2OCF2OCF2H and HCF2OCF2H re-
act with OH radicals at rates that are indistinguishable,
but HCF2OCF2OCF2H reacts with Cl atoms approx-
imately 1000 times slower than does HCF2OCF2H.
Given that the mechanism of reaction of both Cl atoms
and OH radicals with HFPEs is hydrogen atom ab-
Finally we return to discuss the validity of
the assumption introduced in the experimental sec-
tion that in the kinetic and mechanistic analysis,
we can treat the HCF2O(CF2O)n(CF2CF2O)mCF2H
mixture as a single compound. In addition to
International Journal of Chemical Kinetics DOI 10.1002/kin

KINETICS AND PRODUCTS OF CHLORINE ATOM INITIATED OXIDATION OF HFPEs
825
chemical intuition, there are three pieces of infor-
mation that support the assumption that the chem-
ical environment of the C H bonds is indepen-
dent of the number of CF2O and CF2CF2O
units in HCF2O(CF2O)n(CF2CF2O)mCF2H: First,
the fact that the reactivities of Cl atoms toward
HCF2OCF2OCF2H and HCF2OCF2CF2OCF2H mea-
sured by Tuazon et al. [6] are indistinguishable from
that for HCF2OCF2OCF2CF2OCF2Hmeasuredherein;
second, the linearity of the relative rate plots in Fig. 1,
which show that the fractional loss of the different
components during the course of the experiment are
indistinguishable; and third, the linearity of the COF2
formation versus HCF2O(CF2O)n(CF2CF2O)mCF2H
loss, shown in Fig. 3, which also indicates that the
different components react at rates that are indistin-
guishable.
9. World Meteorological Organization. Scientific Assess-
ment of Stratospheric Ozone: 2006; WMO: Geneva,
Switzerland, 2007.
10. Wallington, T. J.; Japar, S. M. J Atmos Chem 1989, 9,
399.
11. Wallington, T. J.; Hurley, M. D. Chem Phys Lett 1992,
189, 437.
12. Sehested, J.; Ellermann, T.; Nielsen, O. J.; Wallington,
T. J.; Hurley, M. D. Int J Chem Kinet 1993, 25, 701.
13. Cavalli, F.; Glasius, M.; Hjorth, J.; Rindone, B.; Jensen,
N. R. Atmos Environ 1998, 32, 3767.
14. Sulbaek Andersen, M. P.; Nielsen, O. J.; Wallington,
T. J.; Hurley, M. D. J Phys Chem A 2005, 109, 3926.
15. Sander, S. P.; Friedl, R. R.; Golden, D. M.; Kurylo,
M. J.; Moortgat, G. K.; Keller-Rudek, H.; Wine, P. H.;
Ravishankara, A. R.; Kolb, C. E.; Molina, M. J.;
Finlayson-Pitts, B. J.; Huie, R. E.; Orkin, V. L. Chemical
kinetics and photochemical data for use in atmospheric
studies: Evaluation Number 15, Jet Propulsion Labora-
tory, National Aeronautics and Space Administration,
Pasadena, CA, 2006.
BIBLIOGRAPHY
16. Kambanis, K. G.; Lazarou, Y. G.; Papaglannakopoulos,
P. J Phys Chem A 1998, 102, 8620.
17. Inoue, Y.; Kawasaki, M.; Wallington, T. J.; Hurley, M. D.
Chem Phys Lett 2001, 343, 296.
1. Molina, M. J.; Rowland, F. S. Nature 1974, 249, 810.
2. Farman, J. D.; Gardiner, B. G.; Shanklin, J. D. Nature
1985, 315, 207.
18. Christensen, L. K.; Wallington, T. J.; Guschin, A.;
Hurley, M. D. J Phys Chem A 1999, 103, 4202.
19. Nohara, K.; Toma, M.; Kutsuna, S.; Takeuchi, K.;
Ibusuki, T. Environ Sci Technol 2001, 35, 114.
20. Oyaro, N.; Sellevag, S. R.; Nielsen, C. J. J Phys Chem
A 2005, 109, 337.
21. Wallington, T. J.; Hurley, M. D.; Fedotov, V.; Morrell,
C.; Hancock, G. J Phys Chem A 2002, 106, 8391.
22. Hickson, K. M.; Smith, I. W. M. Int J Chem Kinet 2001,
33, 165.
3. Young, C. J.; Hurley, M. D.; Wallington, T. J.; Mabury,
S. A. Environ Sci Technol 2006, 40, 2242.
4. Wallington, T. J.; Schneider, W. F.; Worsnop, D. R.;
Nielsen, O. J.; Sehested, J.; Debruyn, W. J.; Shorter,
J. A. Environ Sci Technol 1994, 28, 320.
5. Tyndall, G. S.; Cox, R. A.; Grainer, C.; Lesclaux, R.;
Moortgat, G. K.; Pilling, M. J.; Ravishankara, A. R.;
Wallington, T. J. J. Geophys Res 2001, 106, 12157.
6. Tuazon, E. C. Environ Sci Technol 1997, 31, 1817.
7. Zabel, F.; Reimer, A.; Becker, K. H.; Fink, E. H. J Phys
Chem 1989, 93, 5500.
23. Papadimitriou, V. C.; Kambanis, K. G.; Lazarou, Y.
G.; Papagiannakopoulos, P. J Phys Chem A 2004, 108,
2666.
8. Ko¨ppenkastrop, D.; Zabel, F. Int J Chem Kinet 1991,
23, 1.
International Journal of Chemical Kinetics DOI 10.1002/kin