Atmospheric chemistry of CF3CH2OCH3: Reaction with chlorine atoms and OH radicals, kinetics, degradation mechanism and global warming potential

Article abstract of DOI:10.1016/j.cplett.2011.12.047

FTIR smog chamber techniques were used to measure k(Cl + CF ₃CH₂OCH₃) = (2.28 ± 0.44) × 10^-11 and k(OH + CF₃CH₂OCH₃) = (4.9 ± 1.3) × 10^-13 cm³ molecule^-1 s^-1 in 700 Torr total pressure of air at 296 ± 2 K. The atmospheric lifetime of CF₃CH₂OCH₃ is estimated at 25 days. Reaction of Cl atoms with CF₃CH₂OCH ₃ proceeds 79 ± 4% at the CH₃ group and 22 ± 2% at the CH₂ group. Reaction with OH radicals proceeds 55 ± 5% at the CH₃ group yielding CF₃CH₂OCHO and 45 ± 5% at the CH₂ group yielding COF₂ and CH ₃OCHO as primary oxidation products. The infrared spectrum of CF ₃CH(O)OCH₃ was measured and a global warming potential GWP₁₀₀ = 8 was estimated. The atmospheric chemistry and environmental impact of CF₃CH₂OCH₃ is discussed in context of the use of hydrofluoroethers as CFC substitutes.

Full text of DOI:10.1016/j.cplett.2011.12.047

Chemical Physics Letters 524 (2012) 32–37
Contents lists available at SciVerse ScienceDirect
Chemical Physics Letters
journal homepage: www.elsevier.com/locate/cplett
Atmospheric chemistry of CF CH OCH : Reaction with chlorine atoms and OH
3
2
3
radicals, kinetics, degradation mechanism and global warming potential
a,
⇑
a
b
c
Freja From Østerstrøm , Ole John Nielsen , Mads P. Sulbaek Andersen , Timothy J. Wallington
a
Copenhagen Center for Atmospheric Research, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark
Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Mail Stop 183-901, Pasadena, CA 91109, USA
System Analytics and Environmental Sciences Department, Ford Motor Company, Mail Drop RIC-2122, Dearborn, MI 48121-2053, USA
b
c
a r t i c l e i n f o
a b s t r a c t
ꢁ
11
Article history:
FTIR smog chamber techniques were used to measure k(Cl + CF CH OCH ) = (2.28 ± 0.44) ꢀ 10
and
in 700 Torr total pressure of air at 296 ±
is estimated at 25 days. Reaction of Cl atoms with
group and 22 ± 2% at the ACH A group. Reaction with OH rad-
group yielding CF CH OCHO and 45 ± 5% at the ACH A group yielding
OCHO as primary oxidation products. The infrared spectrum of CF CH(O)OCH was mea-
sured and a global warming potential GWP100 = 8 was estimated. The atmospheric chemistry and envi-
3
2
3
Received 24 August 2011
In final form 19 December 2011
Available online 26 December 2011
ꢁ13
3
ꢁ1 ꢁ1
k(OH + CF
K. The atmospheric lifetime of CF
CF CH OCH proceeds 79 ± 4% at the ACH
icals proceeds 55 ± 5% at the ACH
COF and CH
3
CH
2
OCH
3
) = (4.9 ± 1.3) ꢀ 10
cm molecule
s
2
3
CH OCH
2
3
3
2
3
3
2
3
3
2
2
2
3
3
3
ronmental impact of CF
substitutes.
3
CH
2
OCH
3
is discussed in context of the use of hydrofluoroethers as CFC
Ó 2011 Elsevier B.V. All rights reserved.
1
. Introduction
Recognition of the detrimental environmental impact of the re-
kinetics and mechanism of reaction with chlorine atoms, (ii) kinet-
ics and mechanism of reaction with hydroxyl radicals, (iii) atmo-
spheric degradation products, and (iv) IR spectrum and global
warming potential. The results from these experiments compli-
ment and extend a recent study by Oyaro et al. [4] of the chemistry
of HFEs.
lease of chlorofluorocarbons (CFCs) and Halons into the atmo-
sphere [1,2] has led to an international effort to replace these
compounds with environmentally acceptable alternatives. Hydro-
fluoroethers (HFEs) are a class of compounds which have been
developed as potential CFC and Halon replacements in applications
such as the cleaning of electronic equipment, heat transfer, lubri-
cant deposition, and fire suppression. HFEs do not contain chlorine
and therefore do not contribute to chlorine based catalytic destruc-
tion of stratospheric ozone. A detailed knowledge of the atmo-
spheric chemistry of HFEs is needed to assess the environmental
impact of these compounds prior to their large scale industrial use.
The database concerning the atmospheric chemistry of HFEs is
somewhat sparse and in some cases contradictory. For example,
2
. Experimental method
All experiments were performed in a 140-liter Pyrex reactor
interfaced to a Mattson Sirus 100 FTIR spectrometer [5]. The reac-
tor was surrounded by 22 fluorescent blacklamps (GE F40BLB)
which were used to photochemically initiate the experiments.
Chlorine atoms were generated by photolysis of Cl
Cl ! 2Cl
þ h
OH radicals were generated by UV irradiation of CH
air mixtures:
2
:
2
m
ð1Þ
ONO/NO/
3 2 2 3
in a recent study of CF CF CH OCH in our laboratory [3] we mea-
3
sured a rate constant for reaction with chlorine atoms which was
approximately 40% lower than the previous value reported in the
literature [4]. Furthermore, we observed different primary oxida-
tion products than those reported previously. Further studies on
HFEs seem warranted. To improve our understanding of the atmo-
spheric chemistry of HFEs in general, and CF
lar, the chlorine atom and OH radical initiated oxidation of
CF CH OCH was studied in the smog chamber facility at Ford Mo-
tor Company. Experiments were performed to determine the: (i)
CH
CH
3
ONO þ h
m
! CH
3
O þ NO
ð2Þ
ð3Þ
ð4Þ
3
O þ O
2
! HCHO þ HO
2
HO þ NO ! OH þ NO
2
2
3 2 3
CH OCH in particu-
Reactant and product concentrations were monitored using
in situ Fourier transform infrared spectroscopy. IR spectra were de-
rived from 32 co-added interferograms with a spectral resolution
3
2
3
ꢁ
1
of 0.25 cm
and an analytical path length of 26 m. Reference
spectra were acquired by expanding known volumes of reference
compounds into the chamber. Experiments were performed at
296 ± 1 K in 700 Torr of N /O diluent. All reagents except CH ONO
2 2 3
⇑
Corresponding author.
E-mail address: frejafrom@adslhome.dk (F.F. Østerstrøm).
0
009-2614/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2011.12.047

F.F. Østerstrøm et al. / Chemical Physics Letters 524 (2012) 32–37
33
were obtained from commercial sources at purities >99%.
CF CH OCH was supplied by Alfa Aesar. Ultra high purity nitro-
gen, oxygen, and synthetic air diluent gases were used as received.
CH ONO was prepared by the drop-wise addition of concentrated
SO to a saturated solution of NaNO in methanol and was de-
3
2
3
3
H
2
4
2
void of any detectable impurities using FTIR analysis.
In smog chamber experiments unwanted loss of reactants, ref-
erence compounds, and products via photolysis and heterogeneous
reactions need to be considered. The reactant and reference com-
pounds used in the present work (CF CH OCH , C H , and C H )
3 2 3 2 4 2 2
do not absorb at wavelengths above the Pyrex cut-off (>300 nm).
Neither these compounds, nor the expected oxidation products of
3 2 3
CF CH OCH , are expected to be photolyzed in our reactor. To test
for the presence of heterogeneous reactions product mixtures ob-
tained after UV irradiation were allowed to stand in the dark in
the chamber for 30 min. There was no observable (<2%) loss of
reactants or products, suggesting that heterogeneous reactions
are not a significant complication in the present experiments.
Analysis of the IR spectra was achieved through a process of spec-
tral stripping in which small fractions of a reference spectrum were
subtracted incrementally from the sample spectrum.
Figure 1. Loss of CF
symbols) in the presence of Cl atoms in 700 Torr total pressure of air, 296 ± 2 K.
3
CH
2
OCH
3
relative to C
2
H
2
(solid symbols) and C
2
H
4
(open
The relative rate method is a well established technique for
measuring the reactivity of Cl atoms and OH radicals with organic
compounds. Kinetic data were derived by monitoring the loss of
3 2 3
CF CH OCH relative to two or more reference compounds. The de-
3 2 3 x
3.2. Products of Cl + CF CH OCH in the presence of NO
cays of the reactant and reference are then plotted using the
expression:
To investigate the products of the chlorine atoms initiated oxi-
dation of CF CH OCH , experiments were performed using mix-
ꢀ½
ꢁ
ꢀ
ꢁ
3
2
3
CF
3
CH
2
OCH
OCH
CH
are the concentrations of CF
3
ꢂt
kreactant
½referenceꢂ
0
t
0
ln
¼ k
ln
ðIÞ
tures of 19.8–20.4 mTorr CF CH OCH , 117.5–146.9 mTorr Cl ,
3
2
3
2
½
CF
3
CH
2
3
ꢂt
reference
½referenceꢂ
t
2
6.4–36.0 mTorr NO and 10–600 Torr O
2
in a total pressure of
where [CF
3
2
3
OCH ]
to, [reactant]
t
, [reference]to, and [reference]
t
,
’
2
700 Torr made up with N diluent. Figure 2 shows spectra obtained
CH OCH
2
3
and reference at times ‘t
0
before and after a 30 s UV irradiation of a mixture of 20.4 mTorr
CF CH OCH , 118 mTorr Cl , and 26.4 mTorr NO in 700 Torr air.
The consumption of CF CH OCH was 28%, panel C shows the re-
sult of subtracting features attributable to CF CH OCH from panel
B. Comparison of panel C with reference spectra of CH OCHO and
COF in panels D and E shows the formation of these products.
The IR feature in panel C centered at 1799 cm is attributable to
the formation of ClNO. Subtracting features attributable to CH O-
CHO, COF , NO, NO , HONO, and ClNO from panel C gives the resid-
3
and ‘t’, and kreactant and kreference are the rate constants for reactions
of chlorine atoms or OH radicals with CF CH OCH and the refer-
3
2
3
2
3
2
3
3
2
3
ence. Quoted uncertainties include two standard deviations from
the linear regressions, uncertainties inherent in analysis of the IR
spectra (typically ±1% of the initial reactant concentration), and
uncertainties in the reference rate constants.
3
2
3
3
2
ꢁ
1
3
3
. Results and discussion
3 2 3
.1. Relative rate study of k(Cl + CF CH OCH )
2
2
ual shown in panel F.
3 2
Figure 3 shows plots of the formation of CH OCHO and COF
3
versus the fractional loss of CF
COF have been shifted vertically by 1 unit). The lines through
the data in Figure 3 are linear least squares fits which gives molar
yields of 21 ± 2% for COF and 22 ± 2% for CH OCHO. The simplest
explanation for the formation of CH OCHO and COF in indistin-
guishable yields is that these compounds are formed following
3 2 3
CH OCH (for clarity the data for
The rate of Reaction (5) was measured relative to Reactions (6)
and (7):
2
2
3
Cl þ CF
3
CH
2
OCH
3
! Products
ð5Þ
ð6Þ
ð7Þ
3
2
Cl þ C
Cl þ C
2
H
H
2
! Products
! Products
the decomposition of the CF
lowing reactions:
3
CH(O)OCH
3
alkoxy radical by the fol-
2
4
The initial mixtures consisted of 8.7–20.3 mTorr CF
8.2–111.7 mTorr Cl , and either 7.3 mTorr C or 4.4 mTorr
in a total pressure of 700 Torr air diluent. The loss of
OCH is plotted against the loss of the reference compounds
3 2 3
CH OCH ,
Cl þ CF
3
CH
2
OCH
þ O
3
! CF
! CF
þ NO ! CF
3
CHOCH
CHðOOÞOCH
CHðOÞOCH
OCHO
CðOÞOCH
3
þ HCl
ð5aÞ
ð8Þ
8
C
CF
2
2 2
H
2
H
4
CF
CF
CF
CF
CF
CF
CF
3
3
3
3
3
3
3
CHOCH
3
2
3
3
3
CH
2
3
CHðOOÞOCH
3
3
3
þ NO
2
þ O
2
ð9Þ
ð10Þ
ð11Þ
ð12Þ
ð13Þ
ð14Þ
in Figure 1.
Linear least squares analyses of the data in Figure 1 gives: k
CHðOÞOCH
CHðOÞOCH
3
! CF
3
þ CH
! CF
3
5
/
3
þ O
2
3
3
þ HO
2
k
6
= 0.46 ± 0.07 and k
5
/k
7
= 0.24 ± 0.02. Using k
6
= (5.07 ± 0.34) ꢀ
ꢁ11
ꢁ11
1
0
0
[6] and k
7
= (9.29 ± 0.51) ꢀ 10
[5], gives k
cm molecule
5
= (2.33 ±
þ O
2
! CF
3
OO
ꢁ
11
ꢁ11
3
ꢁ1 ꢁ1
.39) ꢀ 10 , and (2.23 ± 0.22) ꢀ 10
which is the average of the indi-
vidual determinations together with uncertainties that encompass
the extremes of the individual determinations, hence, k = (2.28 ±
. This result is consistent with
s . We
OO þ RO
O þ NO ! COF
As seen from Figure 3, variation of [O
600 Torr had no discernable effect on the yields of COF
CHO indicating that Reaction (11) is not a competing fate for
2
! CF
3
O þ RO þ O
þ FNO
2
choose to quote a final value for k
5
2
5
2
] over the range 10–
or CH O-
ꢁ11
3
ꢁ1 ꢁ1
0
.44) ꢀ 10
the previous determinations of k
et al. [4] and
= (2.31 ± 0.10) ꢀ 10
Kambanis et al. [7].
cm molecule
s
2
3
ꢁ11
5
= (1.8 ± 0.9) ꢀ 10
by Oyaro
ꢁ
11
3
ꢁ1 ꢁ1
k
5
cm molecule
s
by
CF
3
CH(O)OCH
3
alkoxy radicals. The CF
3
radicals produced in Reac-
via Reactions (12)–(14). In
tion (10) will be converted into COF
2

3
4
F.F. Østerstrøm et al. / Chemical Physics Letters 524 (2012) 32–37
Figure 3. Formation of CF
squares) versus the fractional loss of CF
oxidation of CF CH OCH . For clarity the CF
displaced vertically by 1 unit. Experiments were conducted in the presence of
0 Torr (gray symbols), 140 Torr (open symbols) or 600 Torr O (solid symbols).
3
CH
2
OCHO (circles), CH
CH OCH in the chlorine-atom initiated
CH OCHO and COF data have been
3 2
OCHO (triangles) and COF
(
3
2
3
3
2
3
3
2
2
1
2
CF
3
CF
3
CF
3
CH
CH
CH
2
2
2
OCH
OCH
OCH
2
2
2
þ O
OO þ RO
O þ O ! CF
2
! CF
3
CH
2
OCH
2
OO
OCH
OCHO þ HO
ð15Þ
ð16Þ
ð17Þ
2
! CF
3
CH
2
2
O þ RO þ O
2
2
3
CH
2
2
Given the lack of any discernable CF
assumption that CF CH OCHO accounts for the balance of the loss
of CF CH OCH not accounted for by formation of COF and CH O-
CHO. Using this approach we calibrated the concentration of
CF CH OCHO responsible for the carbonyl absorption evident at
766 cm in panel F of Figure 2 and hence the yields for CF
CHO shown in Figure 3 were estimated (for clarity the data for
CF CH OCHO have been shifted vertically by 1 unit). The IR spec-
trum of CF CH OCHO is available from the authors upon request.
As seen from Figure 3, there was no discernable effect of varying
] over the range 10–600 Torr on the yield of CF CH OCHO. We
conclude that the atmospheric fate of CF CH OCH O radicals is
reaction with O to give the formate CF CH OCHO. Our determina-
tion of a CF CH OCHO yield of 79 ± 4% in the chlorine atom
initiated oxidation of CF CH OCH is consistent with the CF CH O-
3
CHO we proceed on the
3
2
3
2
3
2
3
3
2
ꢁ
1
1
3 2
CH O-
3
2
3
2
[
O
2
3
2
3
2
2
2
3
2
3
2
3
2
3
3
2
Figure 2. IR spectra of a CF
irradiation. Panel C is B – 0.72 ꢀ A. Panels D and E show reference spectra of
CH OCHO and COF , respectively. Panel F is the residual spectrum after subtraction
of IR features due to CF CH OCH , CH OCHO, COF , NO, NO , HONO and ClNO from
panel C (see text for details).
3 2 3 2
CH OCH /NO/Cl /air mixture before (A) and after (B)
CHO yields of ꢃ90% reported by Nolan et al. [9] and 84 ± 5% re-
ported by Oyaro et al. [4].
3
2
3
2
3
3
2
2
3.3. Products of OH + CF
3 2 3 x
CH OCH in the presence of NO
The products of the OH radical initiated oxidation of
contrast to Oyaro et al. [4] we do not find any evidence for the for-
mation of CF C(O)OCH and, as discussed in detail elsewhere [3],
we suggest that Oyaro et al. [4] incorrectly assigned the product
peaks in their GC–MS spectra.
CF
CF
3
CH
2
OCH
3
were investigated using mixtures of 58.3–61.6 mTorr
, 141.0–161.6 mTorr CH ONO, 18.7–22.0 mTorr NO
in a total pressure of 700 Torr made up with
diluent. The identified products were the same observed in the
Cl initiated oxidation; however in different yields: 45 ± 3% COF
4 ± 3% CH OCHO and 59 ± 5% CF CH OCHO (see Figure 4). There
are no previous OH radical initiated oxidation product studies to
compare with our results. From the COF yield we conclude that
18a/k18 ꢃ 0.45 ± 0.05, and hence, k18b/k18 ꢃ 0.55 ± 0.07.
3
3
CH
OCH
3
2
3
3
and 140–600 Torr O
2
N
2
Reaction of chlorine atoms at the ACH
will lead to the formation of CF CH OCH O alkoxy radicals which
will either react with O to give the formate CF CH OCHO, or will
decompose via OAC bond scission to give HCHO and the CF CH
alkoxy radical. Any CF CH O radicals formed will react with O
give CF CHO. The observed carbonyl product feature evident in pa-
nel F of Figure 2 (at 1766 cm ) does not match that of CF
3 3 2 3
group in CF CH OCH
2
,
3
2
2
4
3
3
2
2
3
2
3
2
O
2
3
2
2
to
k
3
ꢁ1
3
CHO at
787 cm [8]. We conclude that the residual spectrum in panel F
OH þ CF
OH þ CF
3
3
CH
CH
2
2
OCH
OCH
3
3
! CF
! CF
3
3
CHOCH
CH OCH
3
þ H
2
O
ð18aÞ
ð18bÞ
ꢁ1
1
2
2
2
þ H O
is attributable to the formate CF
ing reactions:
3
2
CH OCHO produced in the follow-
It is interesting to compare the branching ratios for reactions of
OH radicals and chlorine atoms with CF CH OCH . We find that
chlorine atoms are more discriminating. Interestingly, this finding
3
2
3
Cl þ CF
3
CH
2
OCH
3
3 2 2
! CF CH OCH
þ HCl
ð5bÞ

F.F. Østerstrøm et al. / Chemical Physics Letters 524 (2012) 32–37
35
Figure 5. Loss of CF
symbols) in the reaction with OH radicals in 700 Torr of air, 296 ± 2 K.
3
CH
2
OCH
3
relative to C
2
H
2
(open symbols) and C
2
H
4
(solid
Figure 4. Formation of CF
squares) as a function of decay of CF
Experiments were conducted in the presence of 140 Torr (open symbols) or
00 Torr O (solid symbols).
3
CH
2
OCHO (circles), CH
3 2
OCHO (triangles), and COF
(
3
CH OCH in the reaction with OH radicals.
2
3
6
2
of the individual determinations and propagating an additional
1
1% uncertainty associated with conversion of COF
2
formation
cm mole-
ꢁ
13
3
into CF
cule
tions of
3
CH
2
OCH
3
loss gives
k
18 = (4.9 ± 1.3) ꢀ 10
runs counter to the typical behavior of chlorine atoms which typ-
ically tend to be less discriminating than OH radicals in their reac-
tions with organic compounds. This somewhat unusual behavior
ꢁ
1 ꢁ1
s . Our result is consistent with the previous determina-
ꢁ
13
k
18 = (5.7 ± 0.8) ꢀ 10
by Oyaro et al. [4], (6.24 ±
ꢁ
13
ꢁ13
3
0
.67) ꢀ 10
by Zhang et al. [12] and (6.4 ± 0.5) ꢀ 10
cm mol-
may reflect the substantial inductive effect of the CF
which will strengthen the CAH bonds in the ACH A group. The
HACl bond strength is 432 kJ mol while the HAOH bond strength
3
A group
ꢁ1
ꢁ1
ecule
s
reported by Nolan et al. [9].
2
ꢁ1
ꢁ
1
is 497 kJ mol and the rate of hydrogen atom abstraction reac-
tions slows down more rapidly for chlorine atom than OH radical
reactants as the strength of the CAH bond approaches
3
3 2 3
.5. Atmospheric lifetime of CF CH OCH
Organic compounds are removed from the atmosphere via pho-
tolysis, wet deposition, dry deposition, and reaction with OH radi-
cals, NO radicals, Cl atoms, and O . For saturated compounds such
as CF CH OCH reaction with NO radicals and O are too slow to
be of importance. The concentration of chlorine atoms in the tropo-
sphere is believed to be greatest in the marine boundary layer with
ꢁ1
4
32 kJ mol .
3
3
3
3 2 3
.4. Kinetics of OH + CF CH OCH in 700 Torr of air
3
2
3
3
3
The rate of Reaction (18) was measured relative to Reactions
19) and (20):
4
ꢁ3
a mean value of approximately 1.8 ꢀ 10 cm [13]; approximately
(
two orders of magnitude lower than the OH radical concentration.
CF
3
CH
2
OCH
3
þ OH ! Products
ð18Þ
ð19Þ
ð20Þ
3 2 3
Given the rate constant ratio kCl/kOH = 50, loss of CF CH OCH via
reaction with chlorine atoms may compete with loss via reaction
with OH radicals in the marine boundary layer. However,
C
C
2
H
H
4
þ OH ! Products
þ OH ! Products
2
2
3 2 3
CF CH OCH is not expected to be emitted in substantial amounts
into the marine boundary layer and it seems likely that reaction
with chlorine atoms will be a minor overall atmospheric fate for
Experiments were performed using mixtures of of 15.1–
8.2 mTorr CF CH OCH , 119–157 mTorr CH ONO, 14.7 mTorr NO
and 2.2–2.4 mTorr C or 3.2 mTorr C , respectively in a total
of 700 Torr air diluent. The loss of CF OCH was modest (1–
0%) and difficult to measure directly. To improve the precision
of the measurements the loss of CF CH OCH was monitored indi-
rectly by observing the formation of COF and using the product
yield described in Section 3.3 to calculate the loss of CF CH OCH
The loss of CF CH OCH is plotted against the reference com-
pounds in Figure 5.
1
3
2
3
3
2
H
2
2 4
H
3
CH
2
3
1
3
2
3
2
3
2
3
.
3
2
3
The lines through the data in Figure 5 are linear least squares
fits which give k18/k19 = 0.053 ± 0.005 and k18/k20 = 0.63 ± 0.05.
Quoted uncertainties are two standard deviations from the linear
regressions together with our estimate of the uncertainty associ-
ated with the product determination. The relative rate data can
be placed upon an absolute basis using literature values of
ꢁ
12
ꢁ13
k
[
1
19 = (8.52 ± 1.28) ꢀ 10
11] to give
18 = (4.5 ± 0.8) ꢀ 10
cm molecule
[10] and
k
20 = (8.45 ± 0.85) ꢀ 10
ꢁ13
k
and
(5.3 ± 0.4) ꢀ
ꢁ13
3
ꢁ1 ꢁ1
0
s
. Taking an average of the individual
3 2 3
Figure 6. IR spectrum of CF CH OCH in 700 Torr air diluent at 296 ± 2 K. The insert
determinations with error limits which encompass the extremes
shows the calibration plot.

3
6
F.F. Østerstrøm et al. / Chemical Physics Letters 524 (2012) 32–37
CF
14] and photolysis of CF
troposphere. Fluorinated molecules such as CF
hydrophobic and wet deposition is unlikely to be of importance.
The volatility of CF CH OCH will render dry deposition an unlikely
removal mechanism. The major removal process for CF CH OCH
in the atmosphere is reaction with OH radicals. The value of
k(OH + CF CH OCH ) derived in the present work can be used to
provide an estimate of the atmospheric lifetime of CF CH OCH
Using a global weighted-average OH concentration of 1.0 ꢀ
3
CH
2
OCH
3
. Ethers do not absorb at UV wavelengths >200 nm
CH OCH will not be important in the
CH OCH are
CF
of 0.21 W m ppb
The GWP for time horizon t is defined as:
3 2 3 3 2 3
CH OCH we calculate a radiative efficiency for CF CH OCH
ꢁ
2
ꢁ1
[
3
2
3
.
0
3
2
3
R
0
t
0
F
x
expðꢁt=
FCO2 RðtÞdt
s
x
Þdt
3
2
3
0
GWPðxðt ÞÞ ¼
R
0
ðIIÞ
t
3
2
3
0
where FCO2 is the radiative forcing of CO
2
, R(t) is the response func-
, F is
is the atmospheric life-
time of compound x. The denominator in Eq. (II) is the absolute glo-
bal warming potential (AGWP) for CO which has been evaluated as
3
2
3
tion that describes the decay of an instantaneous pulse of CO
the radiative forcing of compound x, and
2
x
3
2
3
.
s
x
6
ꢁ3
1
0 molecules cm
[15] leads to an estimated lifetime of
CF CH OCH with respect to reaction with OH radicals of 25 days.
3
2
3
2
ꢁ1
ꢁ
2
0
.192, 0.676, and 2.223 W m ppm for 20, 100 and 500 year time
horizons, respectively [18]. Using the radiative efficiency deter-
mined in this work, and assuming a lifetime for CF CH OCH of
5 days, we estimate GWPs over 20, 100 and 500 year time horizon
for CF CH OCH of 29, 8 and 2, respectively.
The approximate nature of this lifetime estimate should be
stressed; the average daily concentration of OH radicals in the
atmosphere varies significantly with both location and season
3
2
3
2
[
16]. The value above is an estimate of the global average lifetime;
local lifetimes could be significantly different.
3
2
3
4
. Implications for atmospheric chemistry
3 2 3
3.6. IR spectra and global warming potential of CF CH OCH
We present here a large body of self-consistent kinetic and
mechanistic data concerning the atmospheric chemistry of
CF CH OCH which extends and refines the information available
in the literature. We provide the first determination of the GWP
of CF CH OCH and the product distribution following OH radical
The IR spectrum of CF
3 2 3
CH OCH recorded in 700 Torr of air dil-
uent at 296 K is shown in Figure 6. The integrated absorption cross
3
2
3
ꢁ1
ꢁ16
ꢁ1
section (600–1600 cm ) is (1.55 ± 0.08) ꢀ 10
cm molecule .
Using the method of Pinnock et al. [17] and the IR spectrum of
3
2
3
3 2 3
Figure 7. Degradation scheme of CF CH OCH . Boxes indicate the observed products.

F.F. Østerstrøm et al. / Chemical Physics Letters 524 (2012) 32–37
37
initiated oxidation (the major atmospheric fate). We show that a
previous report that CF C(O)OCH is formed is incorrect. With re-
gard to the environmental impact of CF CH OCH we can make
the following statements. First, CF CH OCH does not contain any
chlorine and will not contribute to stratospheric ozone depletion
via the well established chlorine based chemistry. As with all
hydrofluorocarbons (HFCs) and hydrofluoroethers (HFEs), the
References
3
3
[
[
[
1] M.J. Molina, F.S. Rowland, Nature 249 (1974) 810.
2] J.D. Farman, B.G. Gardiner, J.D. Shanklin, Nature 315 (1985) 207.
3] D.L. Thomsen, V.F. Andersen, O.J. Nielsen, T.J. Wallington, Phys. Chem. Chem.
Phys. 13 (2011) 2758.
3
2
3
3
2
3
[
[
4] N. Oyaro, S.R. Sellevåg, C.J. Nielsen, J. Phys. Chem. A 109 (2005) 337.
5] T.J. Wallington, C.A. Gierczak, J.C. Ball, S.M. Japar, Int. J. Chem. Kinet. 21 (1989)
1077.
ozone depletion potential of CF
purposes zero [19,20]. Second, the atmospheric lifetime of
CF CH OCH is approximately 25 days, and consequently this com-
pound has a negligible GWP (see Section 3.6). Third, the atmo-
spheric oxidation of CF CH OCH gives CF CH OCHO, COF and
CH OCHO, see Figure 7. The atmospheric fate of COF is uptake into
rain, cloud, and ocean water, followed by hydrolysis to produce
HC(O)OH, CO , and HF. At the levels anticipated in the environ-
3
CH
2
OCH
3
is for all practical
[6] T.J. Wallington, J.M. Andino, I.M. Lorkovic, E.W. Kaiser, G. Marston, J. Phys.
Chem. 94 (1990) 3644.
[
7] K.G. Kambanis, Y.G. Lazarou, P. Papagiannakopoulos, J. Phys. Chem. A 102
1998) 8620.
8] Y. Hashikawa et al., J. Fluorine Chem. 125 (2004) 1925.
3
2
3
(
[
[9] S. Nolan, N. O’Sullivan, J. Wenger, H. Sidebottom, J. Treacy, in: R. Djahjhs (Ed.),
Proceedings of EUROTRAC Symposium ‘98: Transport and Chemical
Transformation in the Troposphere, Garmisch-Partenkirchen, Germany,
March 23-27, 1998, WIT Press, Ashurst, Southampton, UK, 1999, pp. 1–120.
[10] J.G. Calvert, R. Atkinson, J.A. Kerr, S. Madronich, G.K. Moortgat, T.J. Wallington,
G. Yarwood, The Mechanisms of Atmospheric Oxidation of the Alkenes, Oxford
University Press, Oxford, 2000.
3
2
3
3
2
2
3
2
2
3 2 3
ment, the atmospheric oxidation products of CF CH OCH are not
of concern.
[
11] M. Sørensen, E.W. Kaiser, M.D. Hurley, T.J. Wallington, O.J. Nielsen, Int. J. Chem.
Kinet. 35 (2003) 191.
[
[
12] Z. Zhang, R.D. Saini, M.J. Kurylo, R.E. Huie, J. Phys. Chem. 96 (1992) 9301.
13] W. Allan, H. Struthers, D.C. Lowe, S.E. Mikaloff Fletcher, J. Geophys. Res. 115
(2010). doi: 10.129/2009JD012924.
Acknowledgements
[
[
[
14] J.G. Calvert, J.N. Pitts Jr., Photochemistry, John Wiley, New York, 1966.
15] R.G. Prinn et al., Science 292 (2001) 1882.
16] G. Klecka et al., Evaluation of Persistence and Long-Range Transport of Organic
Chemicals in the Environment, SETAC Press, Pensacola, 2000.
17] S. Pinnock, M.D. Hurley, K.P. Shine, T.J. Wallington, T.J. Smyth, J. Geophys. Res.
The authors thank the Danish Natural Science Research Council,
the Villum Kann Rasmussen Foundation and EUROCHAMP2, for
financial support. We thank S.P. Sander (JPL) for helpful discus-
sions. This work was performed partly at the Jet Propulsion Labo-
ratory, California Institute of Technology, under a contract with
the National Aeronautics and Space Administration. M.P.S.A. is sup-
ported by an appointment to the NASA Postdoctoral Program,
administered by Oak Ridge Associated Universities through a con-
tract with NASA.
[
[
100 (1995) 23227.
18] World Meteorological Organization (WMO). Scientific Assessment of Ozone
Depletion: 2006, Global Ozone, Research and Monitoring Project – Report 50,
Geneva, Switzerland, 2007.
19] A.R. Ravishankara, A.A. Turnipseed, N.R. Jensen, S. Barone, M. Mills, C.J.
Howard, S. Solomon, Science 263 (1994) 71.
[
[20] T.J. Wallington, W.F. Schneider, J. Sehested, O.J. Nielsen, J. Chem. Soc., Faraday
Discuss. 100 (1995) 55.