Atmospheric chemistry of C2F5CHO: mechanism of the C2F5C(O)O2 + HO2 reaction

Article abstract of DOI:10.1016/j.cplett.2003.09.085

Smog chamber/FTIR techniques were used to study the gas-phase reaction of C2F5C(O)O2 with HO2 radicals in 100-700 Torr of air, or O2, diluted at 296 K. The reaction proceeds by two pathways leading to formation of C2F5C(O)OH and O3 in a yield of 24+/-4 percent and C2F5C(O)O radicals, OH radicals and O2 in a yield of 75+/-4 percent. The gas phase reaction of CnF(2n+1)C(O)O2 with HO2 radicals offers a potential explanation for at least part of the observed environmental burden of fluorinated carboxylic acids, CnF(2n+1)C(O)OH. As part of this work an upper limit for the rate constant of reaction of Cl atoms with C2F5C(O)OH at 296 K was determined; k(Cl + C2F5C(O)OH) < 1 x 10^-17 cm³ molecule^-1 s^-1.

Full text of DOI:10.1016/j.cplett.2003.09.085

Chemical Physics Letters 381 (2003) 14–21
www.elsevier.com/locate/cplett
Atmospheric chemistry of C2F5CHO:
mechanism of the C2F5C(O)O2 + HO2 reaction
b
b,
*
M.P. Sulbaek Andersen a,b, M.D. Hurley , T.J. Wallington
,
b
J.C. Ball , J.W. Martin , D.A. Ellis , S.A. Mabury
c
c
c
a
Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark
b
Ford Motor Company, P.O. Box 2053, Dearborn, MI 48121-2053, USA
Department of Chemistry, 80 St. George St. University of Toronto, Toronto, Ont., Canada M5S 3H6
c
Received 4 August 2003; in final form 18 September 2003
Published online: 14 October 2003
Abstract
Smog chamber/FTIR techniques were used to study the gas-phase reaction of C2F5C(O)O2 with HO2 radicals in
100–700 Torr of air, or O2, diluent at 296 K. The reaction proceeds by two pathways leading to formation of
C2F5C(O)OH and O3 in a yield of 24 ꢀ 4% and C2F5C(O)O radicals, OH radicals and O2 in a yield of 76 ꢀ 4%. The gas
phase reaction of CnF2n þ 1C(O)O2 with HO2 radicals offers a potential explanation for at least part of the observed
environmental burden of fluorinated carboxylic acids, CnF2n þ 1C(O)OH. As part of this work an upper limit for the rate
constant of reaction of Cl atoms with C2F5C(O)OH at 296 K was determined; k(Cl + C2F5C(O)OH) < 1 ꢁ 10ꢂ17 cm3
moleculeꢂ1 sꢂ1
.
Ó 2003 Published by Elsevier B.V.
1. Introduction
more carbon atoms [3,4]. The health effects asso-
ciated with long-term exposure to perfluorinated
acids are the subject of a present risk assessment
[5]. In animal studies and in vitro tests, perfluori-
nated acids are potent peroxisome proliferators [6],
inhibitors of gap-junction intercellular communi-
cation [7], and tumor promoters [8].
PFCAs are not generally used directly in con-
sumer or industrial materials, other than in aque-
ous film forming foams or as polymerization aids
in fluoropolymer manufacture [9]. The PFCAs
observed in the environment are presumably the
degradation products of precursor chemicals. The
identity of the precursors and mechanism by which
Long-chain perfluorinated carboxylic acids
(PFCAs, CnF2n þ 1C(O)OH, where n ¼ 6–12) have
been observed recently in fish [1] and mammals [2]
in a variety of locations around the world. Per-
fluorinated acids have no known significant biotic
or abiotic degradation loss mechanisms in the en-
vironment, and are bioaccumulative when the
perfluorinated chain reaches a length of six or
* Corresponding author. Fax: +1-3133227044.
E-mail address: twalling@ford.com (T.J. Wallington).
0009-2614/$ - see front matter Ó 2003 Published by Elsevier B.V.
doi:10.1016/j.cplett.2003.09.085

M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
15
C2F5C(O)O2 with HO2. Results are reported
herein and discussed with respect to the atmo-
spheric chemistry and environmental impact of
fluorinated organic compounds. Chemical intui-
tion, combined with our recent experience with
long-chain fluorinated alcohols [20] and acids [21]
suggests that C2F5CHO will serve as a useful
model with which to understand the degradation
mechanism of longer chain perfluoroaldehydes.
they are converted into PFCAs are unknown at
the present time.
Fluorotelomer alcohols, CnF2n þ 1CH2CH2OH,
are used in a variety of industrial products (paints,
coatings, polymers, adhesives, waxes, polishes,
electronic materials, and caulks [10]). Global pro-
duction has been estimated at 5 ꢁ 106 kg yearꢂ1
.
As a result of the telomerization manufacturing
process the fluoroalkyl chains tend to have even
numbered carbon chain lengths (e.g. C4, C6; . . . ;
C16). Fluorotelomer alcohols (FTOHs) are named
according to the number of fluorinated carbons
within the chain as compared with the number of
hydrogenated carbon atoms, e.g. 4:2 FTOH ¼
C4F9CH2CH2OH. 6:2 FTOH, 8:2 FTOH, and
10:2 FTOH have been observed in the atmosphere
(at pg mꢂ3 concentrations [11,12]) and it has been
shown experimentally that fluorinated aldehydes
are formed during the atmospheric oxidation of
FTOHs. For example, C4F9CHO, C6F11CHO, and
C8F17CHO have been identified as products of the
simulated atmospheric oxidation of 4:2 FTOH, 6:2
FTOH, and 8:2 FTOH [13].
2. Experimental
The apparatus and experimental techniques
used in this work are described in detail elsewhere
[22]. Experiments were performed in a 140-liter
Pyrex reactor interfaced to a Mattson Sirus 100
FTIR spectrometer. The reactor was surrounded
by 22 fluorescent blacklamps (GE F15T8-BL)
which were used to photochemically initiate the
experiments. C2F5C(O)O2 and HO2 radicals were
generated by the UV irradiation of C2F5CHO/H2/
Cl2 mixtures in 100–700 Torr of air, or O2, diluent
at 296 ꢀ 2 K (M represents a third body, N2 or
O2).
It is well established that the atmospheric oxi-
dation of acetaldehyde, CH3CHO, proceeds via
the formation of acetyl peroxy radicals,
CH3C(O)O2. Furthermore, it is known that
CH3C(O)O2 react with HO2 radicals in the atmo-
sphere via two channels:
Cl2 þ hm ! 2Cl
ð2Þ
ð3Þ
ð4Þ
ð5Þ
ð6Þ
ð7Þ
Cl þ C2F5CHO ! C2F5CO þ HCl
C2F5CO þ O2 þ M ! C2F5CðOÞO2 þ M
Cl þ H2 ! H þ HCl
CH3CðOÞO2 þ HO2 ! CH3CðOÞOOH þ O2
ð1aÞ
H þ O2 þ M ! HO2 þ M
C2F5CðOÞO2 þ HO2 ! products
CH3CðOÞO2 þ HO2 ! CH3CðOÞOH þ O3
ð1bÞ
On the basis of a review of the literature data [14–
18], Tyndall et al. [19] recommend k1a=k1b ¼ 4 in
one atmosphere of air at 298 K (i.e., k1b=ðk1a þ
k1bÞ ¼ 20%). It is possible that a reaction channel
analogous to (1b) is important in the atmospheric
chemistry of fluorinated acyl peroxy radicals of the
general formula CnF2n þ 1C(O)O2. If so, this reac-
tion may be a source of the perfluorocarboxylic
acids observed in the environment. Unfortunately,
there are no available kinetic or mechanistic data
with which to make a determination of the likely
importance of this source. To improve this sit-
uation we have used smog chamber FTIR tech-
niques to study the mechanism of the reaction of
C2F5CðOÞO2 þ C2F5CðOÞO2
! C2F5CðOÞO þ C2F5CðOÞO þ O2
ð8Þ
Initial concentrations of the gas mixtures used
were 2.6–22.5 mTorr of C2F5CHO, 88–440 mTorr
of Cl2, and 223–4511 mTorr of H2 in 700 or 100 Torr
of air or oxygen diluent (ultra high purity). Reac-
tions (7) and (8) are competing loss mechanisms for
C2F5C(O)O2 in the system. To study the mechanism
of reaction (7) a series of experiments was per-
formed in which the initial concentration ratio
[H2]0/[C2F5CHO]0 was varied from 0 to 1545. The

16
M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
rate constant ratio k3/k5 is 1.97 ꢁ 10ꢂ12/1.73 ꢁ
10ꢂ14 ¼ 114 [23,28]. Thus, the initial rate of pro-
duction of HO2 radicals was 0–13.6 times that of
C2F5C(O)O2 radicals. As the [H2]0/[C2F5CHO]0
ratio increases, more HO2 radicals are generated in
the system for a given level of C2F5C(O)O2 and the
relative importance of reaction (7) as a loss mech-
anism for C2F5C(O)O2 increases. The increased
importance of reaction (7) should be manifested by
a change in the products observed in the system. For
sufficiently high values of [H2]0/[C2F5CHO]0 essen-
tially all of the C2F5C(O)O2 radicals should be lost
via reaction (7). At this point, further increases in
[H2]0/[C2F5CHO]0 will have no discernable effect
on the observed products which can then be as-
cribed to reaction (7). This general approach has
been used in previous studies of the reactions of
HO2 radicals with CH2FO2 [24], CH2ClO2 [25],
CHCl2O2 [26], CCl3O2 [26], CH3OCH2O2 [27], and
CH3C(O)O2 [18].
(<2%) loss of reactants or products. Samples of
C2F5CHO and C2F5C(O)OH were subjected to
UV irradiation in 700 Torr of air diluent for 15 min
without any observable loss (<1%) suggesting that
photolysis of C2F5CHO and C2F5C(O)OH is not a
significant complication in the present work.
3. Results
Prior to the study of the mechanism of reaction
(7) an experiment was performed to investigate the
rate of reaction of Cl atoms with C2F5C(O)OH
using the relative rate technique. A mixture con-
taining 3 mTorr C2F5C(O)OH, 5 mTorr CF2ClH
and 1.1 Torr of Cl2 in 700 Torr of N2 was intro-
duced into the chamber. Irradiation of the gas
mixture for 3 min led to 89% consumption of
CF2ClH but no discernable loss (<1%) of
C2F5C(O)OH. Using k(Cl + CF2ClH) ¼ 1.7 ꢁ
10ꢂ15 [23] we derive an upper limit of k(Cl +
Reactant and product concentrations were
monitored using in situ Fourier transform infrared
spectroscopy. IR spectra were derived from 32
coadded interferograms with a spectral resolution
of 0.25 cmꢂ1 and an analytical path length of
27.1 m. Reference spectra were acquired by ex-
panding calibrated volumes of reference materials
into the chamber. The vapor above liquid
C2F5C(O)OH contains monomer and dimer. A
C2F5C(O)OH) < 1 ꢁ 10ꢂ17 cm3 moleculeꢂ1 sꢂ1
.
This result can be compared to our previous de-
termination of k(Cl + CF3C(O)OH) < 1 ꢁ 10ꢂ17
cm3 moleculeꢂ1 sꢂ1 [29]. There is little, or no, re-
action of Cl atoms with perfluorocarboxylic acids
under ambient conditions.
Fig. 1 shows typical spectra acquired before (A)
and after (B) a 35 s irradiation of a mixture of
18.8 mTorr of C2F5CHO, 2.84 Torr of H2, and
218 mTorr of Cl2 in 700 Torr of air diluent. The
loss of C2F5CHO was 8.5 mTorr. Comparison
with reference spectra of C2F5C(O)OH, COF2,
and CF3OH shown in panels C, D and E indicates
the formation of these species. The sharp peaks of
C2F5C(O)OH, COF2, and CF3OH are located at
3577, 1944, and 3664 cmꢂ1, respectively. Features
attributable to the peracid C2F5C(O)OOH were
sought but not found. On the basis of the strength
of the carbonyl absorption feature of peracetic
acid, CH3C(O)OOH, we estimate an upper limit of
6% for the formation of C2F5C(O)OOH.
2
value of Keq ¼ ½Dimerꢃ=½Monomerꢃ ¼ 0:30 ꢀ 0:03
Torrꢂ1 [21] was used to correct for C2F5C(O)OH
dimer (5–26% partial pressure) in the calibrated
volumes. Analysis of the IR spectra was achieved
through a process of spectral stripping in which
small fractions of the reference spectrum were sub-
tracted incrementally from the sample spectrum.
C2F5CHO was synthesized as described else-
where [28] and purified by vacuum distillation. All
other reactants were obtained from commercial
sources at purities >99%. Ultrahigh purity air
(THC < 0.1 ppm) and O2 (>99.994%) diluent gases
were used as received. In smog chamber experi-
ments unwanted loss of reactants and products via
photolysis and heterogeneous reactions has to be
considered. Control experiments were performed
in which product mixtures obtained after UV ir-
radiation were allowed to stand in the dark in the
chamber for 30 min. There was no observable
Fig. 2 shows a plot of the observed formation of
C2F5C(O)OH versus the loss of C2F5CHO for a
series of experiments with [H2]0/[C2F5CHO]0
ranging from 18.3 to 1034.9 in 100–700 Torr of air,
or oxygen, diluent. The linearity of the plots in
Fig. 2 suggest that loss of C2F5C(O)OH is not

M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
17
C2F5C(O)OH and its low reactivity towards Cl
atoms. The lines through the data in Fig. 2 are
linear fits (unweighted, forced through origin) to
the data which give molar yields of C2F5C(O)OH
in the various experiments.
0.4
0.3
0.2
0.1
0.0
A: Before UV
0.4
0.3
0.2
0.1
0.0
0.10
0.05
0.00
The observed yields of C2F5C(O)OH and COF2
are plotted versus [H2]0/[C2F5CHO]0 in Fig. 3.
CF3OH was measured simultaneously with COF2
and its yield followed the same pattern as that of
COF2. As seen from a comparison of the open,
gray, and filled symbols in Fig. 3 there was no
discernable effect of total pressure over the range
100–700 Torr or the use of air, or O2, diluent on
the product yields. For experiments conducted in
the absence of H2 there was evidence in the
product spectra for formation of the trioxides
CF3O3CF3 and C2F5O3C2F5 formed by addition
of CF3O radicals to CF3O2 and C2F5O2 radicals.
In the presence of H2 there was no evidence for the
formation of trioxides in the system and it seems
reasonable to assume that the fate of CF3O radi-
cals is reaction with H2 to give CF3OH.
B: After UV
C: C2F5C(O)OH
E: CF3OH
D: COF2
1700
1800
1900
2000
3500
3600
3700
Fig. 1. IR spectra obtained before (A) and after (B) 55 s of
irradiation of a mixture of 18.8 mTorr C2F5C(O)H, 218 mTorr
Cl2 and 2.8 Torr H2 in 700 Torr of air. The consumption of
C2F5C(O)H was 63%. Panels C, D and E are reference spectra
of C2F5C(O)OH, COF2, and CF3OH, respectively.
As seen from Fig. 3 the yield of C2F5C(O)OH
increases as the concentration ratio [H2]0/
[C2F5CHO]0 is increased from 0 to 600, presum-
ably reflecting the increased importance of reac-
tion (1a) as a loss of C2F5C(O)O2 radicals.
1.5
1.0
0.5
0.0
90
COF2
80
70
20
C2F5COOH
0
2
4
6
8
10 12 14 16 18 20
10
0
Fig. 2. Formation of C2F5COOH versus loss of C2F5CHO for
experiments employing various [H2]0/[C2F5CHO]0 ratios: 18.3
(diamonds, 700 Torr O2), 89.7 (inverted triangles, 700 Torr air),
121.5 (circles, 700 Torr air), 234.0 (triangles, 100 Torr air), and
1034.9 (squares, 700 Torr air).
0
200 400 600 800 1000 1200 1400 1600 1800
Fig. 3. Yield of C2F5COOH following Cl atom initiated oxi-
dation of C2F5CHO in the presence of HO2 radicals versus
[H2]0/[C2F5CHO]0 in 700 Torr of air (filled), 700 Torr of O2
(gray), and 100 Torr of air (open) diluent.
significant during these experiments. This is
consistent with expectations given the the
absence of photolytic or heterogeneous losses of

18
M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
Concurrently, the yield of COF2 decreases. For
concentration ratios greater than 600, there is no
observable dependence of product yields on [H2]0/
[C2F5CHO]0 suggesting that, under these condi-
tions, essentially all the C2F5C(O)O2 radicals are
consumed by reaction with HO2 radicals. Aver-
aging all the product data obtained using [H2]0/
[C2F5CHO]0 of 600 or more, gives the following
product yields; C2F5C(O)OH, 24 ꢀ 4%; COF2,
75 ꢀ 5%; and CF3OH 70 ꢀ 5%. Quoted errors are
two standard deviations from the linear least
squares regressions. Together, C2F5COOH, COF2,
and CF3OH account for 97 ꢀ 6% of the loss of
C2F5C(O)O2.
Loss of C2F5CHO via reaction with HO2 radi-
cals is a potential complication in the present
work. It is known that HO2 radicals add to HCHO
and CH3CHO to give an a-hydroxy peroxy radical
[18]. In the case of C2F5CHO: the addition of HO2
would give C2F5CH(OH)O2 radicals.
action (9) experiments were performed using the
UV irradiation of C2F5CHO/CH3Cl/Cl2/H2 and
C2F5CHO/C2H5Cl/Cl2/H2 in air diluent and the
loss of C2F5CHO was measured relative to CH3Cl
and C2H5Cl. The results are compared to previous
results obtained in relative rate experiments con-
ducted in N2 diluent in the absence of H2 in Fig. 4.
As seen from Fig. 4 the rate of loss of C2F5CHO
relative to those of the reference compounds
(CH3Cl and C2H5Cl) in the presence of substantial
concentrations of H2 ([H2]0/[C2F5CHO]0 ¼ 600)
was indistinguishable from that in N2 diluent in
the absence of H2. There is no observable effect of
the presence of HO2 radicals on the rate of loss of
C2F5CHO in the system. We conclude that reac-
tion of C2F5CHO with HO2 radicals is not a sig-
nificant complication in the present work.
4. Discussion
HO2 þ C2F5CHO $ C2F5CHðOHÞO2
ð9Þ
It is expected that the reaction of HO2 with
C2F5C(O)O2 radicals proceeds via the formation
of a short lived tetraoxide intermediate. As illus-
trated in Fig. 5, the tetraoxide can decompose via
cyclic transition states to give either the peracid
and O2, or C2F5C(O)OH and O3. Decomposition
can also occur via C2F5C(O)O–OOOH or C2F5C
(O)OOO–OH bond scission to give C2F5C(O)O
radicals, O2, and OH.
The a-hydroxy peroxy can then either decom-
pose to reform the aldehyde and HO2 radical,
or react with HO2 or other peroxy radicals in
the system. To test for loss of C2F5CHO via re-
C2F5CðOÞO2 þ HO2 ! C2F5CðOÞOOH þ O2
ð7aÞ
1.5
CH3Cl
C2F5CðOÞO2 þ HO2 ! C2F5CðOÞOH þ O3 ð7bÞ
1.0
C2F5CðOÞO2 þ HO2 ! C2F5CðOÞO þ O2 þ OH
ð7cÞ
C2H5Cl
0.5
The observed formation of C2F5C(O)OH in a
molar yield of 24% shows that channel (7b) plays a
role in the reaction. Evidence for the formation of
ozone (the co-product of C2F5C(O)OH in channel
(7b)) was sought in the IR spectra, but not found.
The absence of observable amounts of O3 can be
rationalized by the fact that Cl atoms react with O3
at a rate which is 1.2 ꢁ 10ꢂ11 [23]/2.0 ꢁ 10ꢂ12 [28] ¼
6 times faster than the reaction with C2F5CHO.
0.0
0.0
0.5
1.0
1.5
2.0
2.5
Fig. 4. Loss of C2F5CHO versus the reference compounds
CH3Cl (triangles) and C2H5Cl (circles) following UV irradia-
tion of either C2F5CHO/Cl2/reference/N2 (open symbols) or
C2F5CHO/Cl2/reference/H2/air mixtures (filled symbols). See
text for details.

M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
19
+
HO2
CF3
O
O
O
CF3
CF2
O
O
H
O
CF2
a
b
c
H
O
O
O
H
O
O
O
O
O
CF3
CF2
O
O
H
O
O
CF3
CF3
CF2
O
O
CF2
O
Fig. 5. Mechanistic pathways for the C2F5C(O)O2 + HO2 reaction.
C2F5O þ M ! CF3 þ COF2 þ M
CF3 þ O2 þ M ! CF3O2 þ M
CF3O2 þ RO2 ! CF3O þ RO þ O2
CF3O þ H2 ! CF3OH þ H
ð13Þ
ð14Þ
ð15Þ
ð16Þ
For the initial concentrations (6 mTorr for ex-
periments with [H2]0/[C2F5CHO]0 > 600) and
consumptions of C2F5CHO employed in the
present work (10–85%) loss of O3 via reaction with
Cl atoms will keep the concentration of this
product below the detection limit (0.5 mTorr).
It is interesting to note that while peracetic acid
(CH3C(O)OOH) is a major product (approxi-
mately 80% yield) in the reaction of CH3C(O)O2
with HO2 [19], there was no evidence for the for-
mation of the analogous peracid in the reaction of
C2F5C(O)O2 with HO2 radicals. Peracetic acid is
not subject to heterogeneous or photolytic loss in
the chamber and reacts slowly with Cl atoms. We
expect C2F5C(O)OOH to be similarly unreactive.
If C2F5C(O)OOH was formed in the chamber we
would expect to observe it. From the absence of
any IR features attributable to C2F5C(O)OOH
product and from the fact that we can account for
100% of the loss of C2F5CHO by other products
we conclude that channel (7a) is of minor impor-
tance (<10%).
Theobservation of COF2 andCF3OHinyieldswhich
are indistinguishable from each other and which ac-
count for the balance of reaction products provides a
strong indication that channel (7c) accounts for the
balance of reaction not accounted for by channel
(7b). The simplest explanation of the observed
products is that reaction (7) proceeds 24ꢀ 4% via
channel (7b) and 76ꢀ 4% via channel (7c).
It is interesting that channel (7a) does not play
any discernable role in the reaction. Hydroperox-
ides are the dominant, if not exclusive, products of
reaction of non-substituted alkyl peroxy radicals
(e.g., CH3O2, C2H5O2, C6H11O2) with HO2. For
substituted alkyl peroxy radicals (e.g., CH2FO2,
CH3OCH2O2) [24,27] additional reaction channels
become important. Reaction of CCl3O2 with HO2
radicals proceeds exclusively via a channel leading
to CCl3O + O2 + OH [26]. The behavior of reaction
(7) follows the broad pattern of other reactions of
peroxy radicals with HO2 radicals. Computational
studies are needed to elucidate the factors re-
sponsible for the changes in reaction mechanisms
of the different peroxy radicals.
C2F5C(O)O radicals formed in channel (7c) are
expected to decompose and initiate a series of re-
actions giving COF2 and CF3OH.
C2F5CðOÞO ! C2F5 þ CO2
C2F5 þ O2 ! C2F5O2
ð10Þ
ð11Þ
ð12Þ
C2F5O2 þ RO2 ! C2F5O þ RO þ O2

20
M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
5. Implications for atmospheric chemistry
the overall reaction at lower temperatures this
needs to be investigated experimentally.
The goal of the present work was to investigate
the possibility that reaction of CnF2n þ 1C(O)O2
with HO2 radicals is an atmospheric source of
perfluorocarboxylic acids (CnF2n þ 1C(O)OH). We
show here that in 100–700 Torr of air diluent at
296 K the reaction of C2F5C(O)O2 with HO2
radicals gives C2F5C(O)OH in a yield of 24 ꢀ 4%.
Chemical intuition and the available data for long-
chain fluorinated alcohols [20] and acids [21] sug-
gest that C2F5C(O)O2 is a good model with which
to understand the atmospheric chemistry of larger
CnF2n þ 1C(O)O2 radicals. It seems likely that the
reaction of CnF2n þ 1C(O)O2 with HO2 radicals will
give CnF2n þ 1C(O)OH in a yield which is compa-
rable to that measured in the present work for
C2F5C(O)O2.
In conclusion, the gas phase reaction of
CnF2n þ 1C(O)O2 with HO2 radicals offers a quali-
tative explanation for the presence of fluorinated
carboxylic acids, CnF2n þ 1C(O)OH, observed in the
environment. Further work is required to quantify
the significance of this source of CnF2n þ 1C(O)OH.
Acknowledgements
We thank Ole John Nielsen (Copenhagen Uni-
versity) for helpful discussions. MPSA thanks the
Danish Research Council (FUR) for a research
stipend. This research was funded, in part, by an
NSERC Strategic Grant.
Translation of the results from the present work
into an estimation of the fraction of fluorinated
aldehydes in the atmosphere which are converted
into fluorinated acids is complicated by three
factors. First, photolysis and reaction with OH
radicals are competing atmospheric fates of
CnF2n þ 1CHO. Reaction with OH leads to the
formation of CnF2n þ 1C(O)O2 radicals in 100%
yield. The rate and products of photolysis of
CnF2n þ 1CHO are unclear and so it is difficult to
establish what fraction of CnF2n þ 1CHO is con-
verted into CnF2n þ 1C(O)O2. Second, reaction with
NO and HO2 radicals are competing atmospheric
fates for CnF2n þ 1C(O)O2 radicals. Oxidation of
C2F5CHO in the presence of excess NO does not
lead to the formation of C2F5C(O)OH [28]. In the
absence of kinetic data for reactions with NO and
HO2 it is difficult to estimate the fraction of
CnF2n þ 1C(O)O2 which react with HO2 radicals.
Third, mechanistic data for reaction (7) are needed
at temperatures in the atmospherically relevant
range 220–300 K. In the analogous reaction of
CH3C(O)O2 radicals the channel giving acid in-
creases from 25% to 50% of the overall reaction
over the temperature range 298–220 K. Decreasing
temperature increases the tetraoxide lifetime which
provides greater opportunity for the complex
molecular rearrangement of the tetraoxide inter-
mediate inherent in channel (7b). While it seems
likely that channel (7b) will play a greater role in
References
[1] C.A. Moody, J.W. Martin, W.C. Kwan, D.C.G. Muir,
S.A. Mabury, Environ. Sci. Technol. 36 (2002) 545.
[2] J.W. Martin, S.A. Mabury, M. Smithwick, C.D. Sonne-
Hansen, M. Gamberg, B. Braune, D.C.G. Muir, Canadian
Arctic Contaminants Assessment Symposium, Northern
Contaminants Program, Ottawa, March 4–7, 2003.
[3] J.W. Martin, S.A. Mabury, K.R. Solomon, D.C.G. Muir,
Environ. Toxicol. Chem. 22 (2003) 189.
[4] J.W. Martin, S.A. Mabury, K.R. Solomon, D.C.G. Muir,
Environ. Toxicol. Chem. 22 (2003) 196.
[5] U.S. Environmental Protection Agency. Preliminary risk as-
sessment of the developmental toxicity associated with exposure
to perfluorooctanoic acid and its salts, Office of Pollution
Prevention and Toxics, Risk Assessment Division, 2003.
[6] J. Berthiaume, K.B. Wallace, Toxicol. Lett. 129 (2002) 23.
[7] B.L. Upham, N.D. Deocampo, B. Wurl, J.E. Trosko, Int.
J. Cancer 78 (1998) 491.
[8] L.B. Biegel, M.E. Hurtt, S.R. Frame, J. OÕConnor, J.C.
Cook, Toxicol. Sci. 60 (2001) 44.
[9] Determination of low levels of fluoropolymer polymeriza-
tion aids – a guidance document, The Society of the
Plastics Industry, SPI Literature Catalogue # BZ-102, New
York, 2003.
[10] E. Kissa, in: M.J. Schick, F.M. Fowkes (Eds.), Fluorinated
Surfactants, Synthesis, Properties, and Applications, Mar-
cel Dekker, New York, 1994.
[11] J.W Martin, D.C.G. Muir, W.C. Kwan, C.A. Moody,
K.R. Solomon, S.A. Mabury, Anal. Chem. 74 (2002) 584.
[12] N.L. Stock, F.K. Lau, J.W. Martin, D.C.G. Muir, S.A.
Mabury, Environ. Sci. Technol. (2003), submitted.
[13] D.A. Ellis, J.W. Martin, S.A. Mabury, M.P. Sulbaek
Andersen, M.D. Hurley, T.J. Wallington, Environ. Sci.
Technol. (2003), submitted.

M.P. Sulbaek Andersen et al. / Chemical Physics Letters 381 (2003) 14–21
21
[14] W.R. Stockwell, J.B. Milford, G. Dongfen, Y.-J. Yang, J.
Atmos. Environ. 29 (1995) 1591.
[22] T.J. Wallington, S.M. Japar, J. Atmos. Chem. 9 (1989) 399.
[23] W.B. DeMore, S.P. Sander, D.M. Golden, R.F. Hampson,
M.J. Kurylo, C.J. Howard, A.R. Ravishankara, C.E.
Kolb, M.J. Molina, 1997, Jet Propulsion Laboratory
Publication 97-4, Pasadena, CA.
[15] H. Niki, P.D. Maker, C.M. Savage, L.P. Breitenbach, J.
Phys. Chem. 89 (1985) 588.
[16] G.K. Moortgat, B. Veyret, R. Lesclaux, Chem. Phys. Lett.
160 (1989) 443.
[24] T.J. Wallington, M.D. Hurley, W.F. Schneider, J.
Sehested, O.J. Nielsen, Chem. Phys. Lett. 218 (1994)
34.
[17] O. Horie, G.K. Moortgat, J. Chem. Soc., Faraday Trans.
88 (1992) 3305.
[18] M. Crawford, T.J. Wallington, J.J. Szente, M.M. Maricq,
J.S. Francisco, J. Phys. Chem. 103 (1999) 365.
[19] G.S. Tyndall, R.A. Cox, C. Granier, R. Lesclaux, G.K.
Moortgat, M.J. Pilling, A.R. Ravishankara, T.J. Walling-
ton, J. Geophys. Res. 106 (2001) 12157.
[25] T.J. Wallington, M.D. Hurley, W.F. Schneider, Chem.
Phys. Lett. 251 (1996) 164.
[26] V. Catoire, R. Lesclaux, W.F. Schneider, T.J. Wallington,
J. Phys. Chem. 100 (1996) 14356.
[27] T.J. Wallington, M.D. Hurley, J.C. Ball, M.E. Jenkin,
Chem. Phys. Lett. 211 (1993) 41.
[20] D.A. Ellis, J.W. Martin, S.A. Mabury, M.D. Hurley, M.P.
Sulbaek Andersen, T.J. Wallington, Environ. Sci. Technol.
37 (2003) 3816.
[28] M.P. Sulbaek Andersen, M.D. Hurley, T.J. Wallington,
J.C. Ball, J.W. Martin, D.A. Ellis, S.A. Mabury, O.J.
Nielsen, Chem. Phys. Lett. 379 (2003) 28.
[21] M.D. Hurley, M.P. Sulbaek Andersen, T.J. Wallington,
D.A. Ellis, J.W. Martin, S.A. Mabury, J. Phys. Chem. A.
(2003), submitted.
[29] T.J. Wallington, M.D. Hurley, Int. J. Chem. Kinet. 27
(1995) 189.