Oxidation mechanisms for ethyl chloride and ethyl bromide under atmospheric conditions

Article abstract of DOI:10.1021/jp012932a

Partially chlorinated and/or brominated alkanes are present in the earth's atmosphere as the result of natural and anthropogenic activities, and their oxidation mechanisms under atmospheric conditions have been the subject of a number of recent studies. The Cl-atom initiated oxidation of ethyl chloride and ethyl bromide was studied as a function of temperature (220-298 K) in an environmental chamber equipped with an FTIR spectrometer. Products resulting from abstraction at the -CH₂Cl group included CH₃C(O)Cl, CO₂, CH₃C(O)OOH, CH₃C(O)OH, CH₂O, CO, HCOOH, and CH₃OH. The HCl-elimination reaction possessed an energy barrier of about 6 kcal/mole. As a result of this modest barrier, chemical activation played an important role in the chemistry of internally excited CH₃CHClO radicals generated from the exothermic reaction of the CH₃CHClO₂ radical with NO. BrCH₂CH₂ radicals, generated via Cl-atom abstraction at the -CH₃ group, reacted via Br-atom elimination to form ethylene or via addition of O₂ to form a peroxy radical. The generation of an alkene is likely to be general a occurrence in the tropospheric chemistry of alkyl bromides.

Full text of DOI:10.1021/jp012932a

312
J. Phys. Chem. A 2002, 106, 312-319
Oxidation Mechanisms for Ethyl Chloride and Ethyl Bromide under Atmospheric
Conditions
John J. Orlando* and Geoffrey S. Tyndall
Atmospheric Chemistry DiVision, National Center for Atmospheric Research, Boulder, Colorado 80305
ReceiVed: July 31, 2001; In Final Form: October 26, 2001
The Cl-atom initiated oxidation of ethyl chloride and ethyl bromide has been investigated as a function of
temperature (220-298 K) in an environmental chamber equipped with an FTIR spectrometer. In the absence
of NOx, reaction with O2 (CH3CHClO + O2 f CH3C(O)Cl + HO2, reaction 9) and decompositon via HCl-
elimination (CH3CHClO f CH3CO + HCl, reaction 8a) are shown to be competing fates of the CH3CHClO
radical generated in the ethyl chloride system, with k9/k8a ) 3.2 × 10-24 exp(2240/T) cm3 molecule-1. The
CH3CHClO radical is also shown to be subject to a chemical activation effect; when produced in the exothermic
reaction of CH3CHClO2 with NO, about 50% of the nascent CH3CHClO radicals decompose promptly via
HCl elimination, before collisional deactivation can occur. The reaction of Cl with ethyl bromide occurs
70-80% via abstraction from the -CH2Br group and 20-30% via abstraction from the -CH3 group. The
observation of ethene in this system indicates that the BrCH2CH2 radical (generated via abstraction from the
-CH3 group) undergoes Br-atom elimination (BrCH2CH2 f CH2dCH2 + Br, reaction 37) in competition
with its reaction with O2 (BrCH2CH2 + O2 f BrCH2CH2O2, reaction 34) At atmospheric pressure, this
competition is governed by the rate coefficient ratio, k37/k34 ) 1.4 × 1023 exp(-2800/T) molecule cm-3. The
chemistry of the CH3CHBrO radical (generated from abstraction at the -CH2Br site) is dominated by Br-
atom elimination; no evidence for its reaction with O2 could be found at any temperature.
Introduction
well as a suite of carbonyls, alcohols, and organic hydroper-
oxides:
Partially chlorinated and/or brominated alkanes are present
in the earth’s atmosphere as the result of both natural and
anthropogenic activities,1 and their oxidation mechanisms under
atmospheric conditions have been the subject of a number of
recent studies.2-20 Ethyl chloride and ethyl bromide are minor
components of the atmospheric halogen burden,1 arising mostly
from natural sources. In addition, these species are good model
compounds for larger partially halogenated organics found in
the atmosphere (e.g., chlorinated solvents, the HCFCs in use
as CFC replacement compounds, and the alkyl bromides being
proposed for use as industrial solvents).
CH3CHClO2 + RO2 f CH3CHClO + RO + O2 (4a)
CH3CHClO2 + RO2 f CH3C(O)Cl + ROH + O2
(4b)
CH3CHClO2 + RO2 f CH3CH(Cl)OH + RdO + O2 (4c)
CH3CHClO2 + HO2 f CH3CH(Cl)OOH + O2 (5)
ClCH2CH2O2 + RO2 f ClCH2CH2O + RO + O2
(6a)
The Cl-atom initiated oxidation of ethyl chloride has been
extensively investigated near 298 K by scientists at the Ford
Motor Company.2-4 Abstraction of an H-atom occurs at two
sites (with abstraction at the -CH2Cl group dominating), leading
to the formation of two peroxy radicals, CH3CHClO2 and
ClCH2CH2O2:
ClCH2CH2O2 + RO2 f ClCH2CHO + ROH + O2 (6b)
ClCH2CH2O2 + RO2 f ClCH2CH2OH + RdO + O2 (6c)
ClCH2CH2O2 + HO2 f ClCH2CH2OOH + O2 (7)
where RO2 represents any organic peroxy radical present in the
system.
Surprisingly large yields of HCl and CO2 observed in these
experiments2-4 were rationalized in terms of a novel HCl
elimination from the CH3CHClO radical, reaction 8a:
Cl + CH3CH2Cl f CH3CHCl + HCl (82%) (1a)
Cl + CH3CH2Cl f ClCH2CH2 + HCl (18%) (1b)
CH3CHCl + O2 + M f CH3CHClO2 + M
ClCH2CH2 + O2 + M f ClCH2CH2O2 + M
(2)
(3)
CH3CHClO + M f CH3CO + HCl + M
f CH3CHO + Cl + M
(8a)
(8b)
(9)
Subsequent chemistry of these peroxy species (in the absence
of the oxides of nitrogen) yields in part 1-chloroethoxy
(CH3CHClO) and 2-chloroethoxy (ClCH2CH2O) radicals, as
CH3CHClO + O2 f CH3C(O)Cl + HO2
Furthermore, the lack of a dependence of the product (HCl,
CH3C(O)Cl) yields on O2 partial pressure indicated that this
* Corresponding author.
10.1021/jp012932a CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/13/2001

Oxidation of Ethyl Chloride and Ethyl Bromide
J. Phys. Chem. A, Vol. 106, No. 2, 2002 313
HCl elimination was the dominant (if not sole) fate of the
CH3CHClO radical over the range of conditions studied.
Similar studies5-9 on the CH2ClO radical (generated from
the Cl-atom initiated oxidation of methyl chloride) show that
HCl elimination does occur, though this is a minor process
compared with O2 reaction 11 under typical atmospheric
conditions:
ethyl bromide and the chemistry of the BrCH2CH2 radical are
also investigated.
Experimental Section
All experiments were carried out in a 2-m long, 47-L stainless
steel environmental chamber, described previously.6,7,16,17,22 The
cell temperature (which was varied from 220 to 298 K) was
controlled by flowing chilled ethanol (from a Neslab ULT-80DD
circulating bath) through a jacket surrounding the cell. The
chamber is interfaced to a Bomem DA 3.01 Fourier Transform
Infrared Spectrometer via a set of Hanst multipass optics, which
provided a total observational path length of 32.6 m. Spectra
were recorded over the range 800-3900 cm-1 at a resolution
of 1 cm-1, and were obtained from the co-addition of 200
interferograms (3.5 min acquisition time).
Experiments involved the photolysis of mixtures of Cl2 ((0.7-
10) × 1015 molecule cm-3) and either ethyl chloride ((0.8-8)
× 1015 molecule cm-3) or ethyl bromide ((1.7-7) × 1015
molecule cm-3), in mixtures of O2 (20-620 Torr) and N2 (100-
720 Torr) at total pressures of 700-760 Torr. NO ((4-7) ×
1014 molecule cm-3) was added to the reaction mixture in some
cases. The ratio of [O2]/[Cl2] was typically >2000, thus limiting
the occurrence of reactions of alkyl (e.g., CH3CHCl) or acyl
(e.g., CH3CO) radicals with Cl2 to <1%.4 Photolyses were
conducted with an unfiltered cw Xe-arc lamp, directed along
the length of the cell through a quartz window located at one
end. Reaction mixtures were photolyzed four to six times (each
photolysis period being 30-200 s), and an infrared spectrum
was recorded following each photolysis period. Under these
conditions, about 20% of the parent alkyl halide was consumed
over the course of an experiment.
Reactants (ethyl chloride and ethyl bromide) and products
(CO2, CO, CH3C(O)Cl, CH3C(O)Br, CH3CHO, CH3OH,
CH3C(O)OH, CH3C(O)OOH, CH2O, HC(O)OH and its dimer,
HC(O)Cl, BrCH2CHO, ClCH2CHO) were quantified by com-
parison with standard spectra recorded in our laboratory at the
same pressure and temperature as the experiment. For CO2,
which showed nonlinear absorbance vs concentration behavior
under the conditions employed, concentrations were obtained
by interpolation of absorbance versus [CO2] plots. Standard
samples of CH3C(O)Cl or CH3C(O)Br were generated in the
chamber from the brief photolysis of Cl2 or Br2 (≈20 mTorr)/
CH3CHO (≈20 mTorr)/N2 mixtures at atmospheric pressure.
Quantification of the CH3C(O)Cl and CH3C(O)Br was possible
with the assumption23 of unit conversion of CH3CHO to the
acetyl halide, e.g.,
CH2ClO + M f HCO + HCl + M
f CH2O + Cl + M
(10a)
(10b)
(11)
CH2ClO + O2 f HC(O)Cl + HO2
Furthermore, it was shown7 that the HCl elimination channel
(10a), as well as the C-Cl scission reaction 10b were enhanced
by a chemical activation effect in the presence of NO. That is,
conversion of CH2ClO2 to CH2ClO via exothermic reaction 12
results
CH2ClO2 + NO f CH2ClO* + NO2
(12)
in part in the formation of internally excited CH2ClO, whose
decomposition via reactions 10a and 10b is prompt compared
to its deactivation rate.
Theoretical studies of the CH3CHClO and CH2ClO reaction
systems have also been carried out.18-20 In accord with the
laboratory results, HCl elimination (8a) from CH3CHClO was
found to be the most rapid unimolecular pathway available,
occurring with a rate coefficient of about 106 s-1 at 1 atm, 298
K.18 Furthermore, the calculated barrier of only about 5 kcal/
mol for (8a) makes chemical activation likely in this system.
In an earlier study from the same group,19 Cl-atom elimination
was shown to be the lowest energy pathway for CH2ClO; the
HCl-elimination process (10a) was apparently not considered,
however. Semiempirical calculations show HCl elimination from
CH2ClO to be possible,20 though barriers to both reactions 10a
and 10b obtained by these methods are considerably higher than
has been shown by recent experiments.7
The oxidation of ethyl bromide has not yet been studied under
atmospheric conditions. However, some data on R-brominated
alkoxy radical chemistry are available.15-17,21 The chemistry of
the CH2BrO and CHBr2O radicals15-17 has been shown to be
dominated by Br-atom elimination over the temperature range
220-300 K, even in the presence of ≈1 atm of O2. The
CH3CHBrO radical has been studied indirectly by Bierbach et
al.21 in a study of the Br-atom-initiated oxidation of trans-2-
butene. While the major CH3CHBrO reaction pathway was via
Br-atom elimination reaction 13, small yields of CH3C(O)Br
indicated that some reaction with O2, reaction 14, may have
been occurring:
Cl2 + hν f Cl + Cl
(15)
(16)
(17)
Cl + CH3CHO f CH3CO + HCl
CH3CO + Cl2 f CH3C(O)Cl + Cl
CH3CHBrO + M f CH3CHO + Br + M
CH3CHBrO + O2 f CH3C(O)Br + HO2
(13)
(14)
Absorption cross sections for the two halides were found to be
1.3 × 10-18 cm2 molecule-1 (acetyl chloride, R-branch maxi-
mum at 1827 cm-1) and 1.8 × 10-18 cm2 molecule-1 (acetyl
bromide, R-branch maximum at 1831 cm-1) at 298 K; both
values increased by 20% at 220 K. These CH3C(O)Cl data are
in accord with previous determinations made in our laboratory.23
Reference spectra for BrCH2CHO (ClCH2CHO) were generated
from the reaction of Br (Cl) with ethene in the presence of O2,
and quantified using cross section data provided by Yarwood
et al.24
However, formation of CH3C(O)Br via this pathway is not
consistent with the anticipated rapid elimination of a Br-atom.
In this paper, the Cl-atom-initiated oxidation of ethyl chloride
and ethyl bromide is explored over a range of temperature (220-
298 K) and O2 partial pressures (20-650 Torr), both in the
absence and presence of NOx. Product yield data are interpreted
in terms of the fate of the R-halogenated alkoxy radicals,
CH3CHClO and CH3CHBrO, for conditions representative of
the troposphere. The mechanism for the Cl-atom reaction with

314 J. Phys. Chem. A, Vol. 106, No. 2, 2002
Orlando and Tyndall
TABLE 1: Measured Product Yields (%) as a Function of
Temperature from the Photolysis of Cl2/CH3CH2Cl/O2/N2
Mixtures at a Total Pressure of 710 Torra
Test experiments were conducted to ensure that no unac-
counted-for loss of products or starting materials was occurring.
Loss of ethyl chloride and ethyl bromide by photolysis (i.e., in
the absence of Cl2) was found to be negligibly slow. In some
cases, reacted mixtures were left sitting in the dark for periods
of up to an hour. No gain or loss of any reaction product or
starting material was noted.
Chemicals used in these experiments were obtained from the
following sources: Cl2 (Matheson, UHP); ethyl chloride, ethyl
bromide, acetaldehyde (all Aldrich, > 99+ %); O2 (U.S.
Welding); N2, (boil-off from liquid N2, U.S. Welding); NO
(Linde, UHP). Gases were used as received, while liquids were
subjected to several freeze-pump-thaw cycles before use.
temperature (K)
298
273
254
237
220
O2 pressure (Torr) 20 550 20 620 20 620 20 500 20 620
CO2
66 52 57 46 51 32 58 24 50 12
12 19 12 22 14 34 20 43 26 48
CH3C(O)Cl
CH3OH
5
4
4
2
4
2
4
2
4
5
3
12
2
1
3
2
9
CH2Ob
32 28 29 24 23 10 11
7
17 19 15 14 13 12 11 11
peracetic acid
ClCH2CHOb
total C balance
7
5
6
5
6
5
9
87 87 77 78 71 74 73 74 70 67
a Uncertainties in the measured yield data are as follows: for CO2,
(5%; for CH3C(O)Cl, CH3OH, CH2O, and peracetic acid, (2%; for
ClCH2CHO, (4%; data are shown for the extremes of the range of O2
partial pressures investigated. b The yield of CH2O includes those of
its byproducts, CO and HCOOH (monomer + dimer). See text for
details. c The yield of ClCH2CHO includes that of its byproduct,
HC(O)Cl.
Results and Discussion
1. Ethyl Chloride. Products observed at all temperatures
following the photolysis of Cl2/CH3CH2Cl mixtures in O2/N2
diluent were CO2, CH3C(O)Cl, CO, CH2O, CH3OH, HC(O)-
OH (and its dimer at low temperature), CH3C(O)OOH (peracetic
acid), HC(O)Cl, and ClCH2CHO. Yield data for selected
experiments are shown in Table 1. The observed products
accounted for about 70-85% of the reacted ethyl chloride, and
can be explained from a consideration of reactions 1-8 above,
as well as chemistry of the acetyl radical generated in reaction
8a,
and CH2O (30%, if its byproducts CO and HCOOH are
included). Again, although Shi et al. observed no dependence
of these yields on [O2], their yields are quite similar to ours.
Since ClCH2CHO and its byproduct HC(O)Cl are the major
products obtained following reaction 1b, the total yield of these
species provides a lower limit to the branching ratio k1b/k1. The
sum of the yields of these two species was found to be 18 (
4% at 298 K, and to decrease to 9 ( 4% at 220 K. Our value
of k1b/k1 at 298 K, though strictly a lower limit, is in agreement
with those determined by Shi et al.2 and Niedzielski et al.,25
who both obtained values of k1b/k1 of 0.18 ( 0.02 using a
competitive chlorination technique.
As alluded to earlier, products resulting from abstraction at
the -CH2Cl group included CH3C(O)Cl, CO2, CH3C(O)OOH,
CH3C(O)OH, CH2O, CO, HCOOH, and CH3OH. No evidence
was found for the presence of CH3CHO under any conditions.
Based on an upper limit of 4% to its yield, we can conclude
that Cl-atom elimination, reaction 8b, accounts for no more than
10% of the chemistry of the CH3CHClO radical.
CH3CO + O2 + M f CH3C(O)O2 + M
(18)
2CH3C(O)O2 (+ 2O2) f 2CH3O2 + 2CO2 + O2 (19)
CH3C(O)O2 + HO2 f CH3C(O)OOH + O2 (20a)
f CH3C(O)OH + O3
(20b)
CH3C(O)O2 + CH3O2 (+ O2) f
CH3O + CH3O2 + CO2 + O2 (21)
At a given temperature, the CH3C(O)Cl yield was found to
increase with increasing O2 partial pressure (note that the yields
in Figure 1 have been corrected for a very small, e1.5%,
occurrence of the reaction of CH3CO with Cl2 at low [O2]). On
the other hand, the total yield of all products originating from
the acetylperoxy radical chemistry (CO2, CO, HC(O)OH, CH3C-
(O)OOH, CH2O, CH3OH) was found to decrease with increasing
O2 partial pressure. These observations reflect the competition
between reactions 8a and 9, which can be expressed as follows:
CH3C(O)O2 + CH3O2 f CH3C(O)OH + CH2O + O2 (22)
2CH3O2 f CH3OH + CH2O + O2
f 2CH3O + O2
(23a)
(23b)
(24)
CH3O2 + HO2 f CH3OOH + O2
CH3O + O2 f CH2O + HO2
(25)
Y(CH3COCl) )
Y(CH3CHClO) * {(k9/k8a) [O2]/(1+ (k9/k8a) [O2])} + C (A)
CH2O + HO2 T HOCH2OO f f HCOOH (26)
2HO2 f H2O2 + O2 (27)
Y(CO2corr) ) Y(CH3CHClO) * {1/(1+ (k9/k8a) [O2])}
(B)
and secondary reactions of ClCH2CHO, which lead to HC(O)-
Cl. As with other peroxides in our chamber system, H2O2, CH3-
OOH, and CH3CH(Cl)OOH were not detected and are presum-
ably lost rapidly at the cell walls. Before considering in more
detail the implications of the observed yields and their depen-
dence on temperature and O2 partial pressure, it is worthwhile
to compare our room-temperature yields with previous2,4 work.
Shi et al.2 and Kaiser and Wallington4 both reported yields of
≈15% CH3C(O)Cl from the Cl-atom initiated oxidation of ethyl
chloride, similar to our yields which range from 12% at low
[O2] to 19% at high [O2]. However, Shi et al. observed no
dependence of the CH3C(O)Cl yield on [O2], in contrast to our
data. Shi et al. also report yields for CO2 (53%), CH3OH (7%),
corr
where Y(X) reflects the yield of species X, CO2 reflects the
yield of CH3CO radical as described below, and C is used to
quantify any O2-independent formation of CH3COCl, for
example via reaction 4b. Inclusion of the term Y(CH3CHClO),
the fraction of the oxidized ethyl chloride that proceeds via CH3-
CHClO formation, serves to normalize the measured product
yields to the amount of CH3CHClO actually generated.
While the occurrence of reaction 9 can be tracked uniquely
through the O2-dependence of the CH3C(O)Cl yield, reaction
8a leads to the generation of either peracetic acid, acetic acid
(not observed, but minor compared to peracetic acid based on
measurements of k20a/k20b ) 2.5 ( 0.5),26-30 or CO2 and a

Oxidation of Ethyl Chloride and Ethyl Bromide
J. Phys. Chem. A, Vol. 106, No. 2, 2002 315
Figure 3. Rate coefficient data for the HCl elimination from CH3-
CHClO, reaction 8a, plotted in Arrhenius form.
Figure 1. Yield of CH3C(O)Cl as a function of the O2 partial pressure
from the photolysis of Cl2/CH3CH2Cl/O2/N2 mixtures at a total pressure
of 710 ( 10 Torr: solid circles, 298 K; open triangles, 273 K; solid
squares, 254 K; open diamonds, 237 K; solid triangles, 220 K. Lines
are fits of equation (A) to the data, see text for details.
as follows. First, all CH3C(O)Cl and CO2corr yield data at a given
temperature were fit simultaneously, with Y(CH3CHClO), k9/
k8a, and C as fit parameters. Because the CH3C(O)Cl yield data
were deemed to be the most accurate, a second fit was performed
using only the CH3C(O)Cl data, but with the value of Y(CH3-
CHClO) (which is best determined by the CO2 data at low [O2])
fixed to the value obtained in the first fit. A couple of trends
with temperature are evident in the data. First, the value of C
was found to increase with decreasing temperature (from 12%
at 298 K to 22% at 220 K), probably reflecting an increase in
the branching ratio to molecular products in the peroxy radical
self-reactions (4) and also the increase in the branching ratio
k1a/k1 with decreasing temperature. As alluded to above, our
yield of CH3C(O)Cl at 298 K and low O2 is similar to previous
reports,2,4 though no O2 dependence to the yield was noted in
these previous studies.
More importantly, the rate coefficient ratio k9/k8a was found
to increase with decreasing temperature. Fits of the temperature-
dependent k9/k8a data to an Arrhenius expression yielded k9/k8a
) 3.2 × 10-24 exp(2240/T) cm3 molecule-1. If one estimates
the value of k9 to be 6 × 10-14 exp(-550/T) cm3 molecule-1
s-1, on the basis of data for the rate coefficient for the reaction
of ethoxy radicals with O2,31 an estimated expression for k8a
can be obtained, k8a ) 1.9 × 1010 exp(-2790/T) s-1, for 1 atm
total pressure (see Figure 3). Thus, the three-center HCl
elimination is characterized by a relatively low A-factor
(compared to values of 1013-1014 typical of simple alkoxy
radical bond cleavage reactions), and a relatively low activation
barrier (about 6 kcal/mol). The agreement between our Arrhe-
nius expression for k8a and that obtained in the theoretical
analysis of Hou et al.,18 k8a ) 6.5 × 1010 exp(-2869/T) s-1, is
quite remarkable given the uncertainties involved in the two
studies. The theoretical studies also show that the reaction is in
the pressure fall-off region at atmospheric pressure, reaching
its high-pressure limit near 10 atm with an A-factor of order
corr
Figure 2. Yield of CO2 as a function of the O2 partial pressure
from the photolysis of Cl2/CH3CH2Cl/O2/N2 mixtures at a total pressure
of 710 ( 10 Torr: solid circles, 298 K; open circles, 273 K; solid
triangles, 254 K; open triangles, 237 K; solid squares, 220 K. Lines
are fits of equation (A) to the data, see text for details.
CH3O2 radical (which eventually generates CH2O, HC(O)OH,
CH3OH, CO, or CH3OOH). The most accurate method for
determining the extent of occurrence of reaction 8a was from
the sum of the CO2 and peracetic acid yields (with the peracetic
acid yield multiplied by 1.4 to account for the undetected acetic
acid), a quantity referred to as CO2corr. Plots of CO2 versus
corr
O2 as a function of temperature are shown in Figure 2. Although
the data exhibit considerable scatter, they do show the expected
1012 s-1
.
corr
trend of decreasing CO2 with increasing O2 and decreasing
The low activation barrier to reaction 8a makes it quite likely
that this decomposition will be prone to chemical activa-
tion;7,18,30,32 that is, decomposition yields will be larger when
the CH3CHClO radical is generated from the exothermic
reaction of CH3CHClO2 with NO, than following its generation
from the near-thermoneutral peroxy radical self-reaction. Thus,
temperature. The sum of the CO, CH2O, HC(O)OH, and CH3-
OH yields were typically 20-40% less than those of CO2,
probably reflecting the (undetected) formation of CH3OOH.
Fits of the CH3C(O)Cl and CO2 data to the expressions
above are shown in Figures 1 and 2. The fitting procedure was
corr

316 J. Phys. Chem. A, Vol. 106, No. 2, 2002
Orlando and Tyndall
some ethyl chloride experiments were carried out in the presence
of NOx. Major carbon-containing products observed in the
photolysis of Cl2/C2H5Cl/NO/O2/N2 mixtures were CO2, CH3C-
(O)Cl, PAN, CH2O, and CO. As was the case in the absence of
NOx, CH3CHO was not found in these experiments (yield <4%).
Under these conditions, the chemistry of the CH3CHClO2 radical
is dominated by its reaction with NO,
quantification, plots of percent yield of these species versus ethyl
bromide consumption were constructed, and actual yields were
obtained via extrapolation of these plots to zero ethyl bromide
consumption. Yields of CH3CHO and BrCH2CHO were difficult
to quantify because the spectra of these two species are strongly
overlapped, and because the apparent yields were small and
decreased with increasing reaction time, again likely due to rapid
reaction with Br.
By analogy to reaction of Cl with ethyl chloride2 and ethyl
fluoride,37 abstraction at the -CH3 group (33b) is likely to be
minor compared to reaction at the -CH2Br group (33a):
CH3CHClO2 + NO f CH3CHClO + NO2
(28)
and all CH3C(O)Cl must originate from this reaction, followed
by reaction 9. Other major products listed above are obtained
pursuant to reactions 8a and 18, with the sum of the CO2 and
PAN yields providing a measure of the occurrence of (8a):
Cl + CH3CH2Br f CH3CHBr + HCl
Cl + CH3CH2Br f BrCH2CH2 + HCl
(33a)
(33b)
CH3C(O)O2 + NO (+O2) f CH3O2 + CO2 + NO2 (29)
CH3C(O)O2 + NO2 + M f CH3C(O)O2NO2 + M (30)
Products anticipated following the occurrence of the minor
channel (33b) include BrCH2CHO, BrCH2CH2OH, and BrCH2-
CH2OOH. Of these species, only BrCH2CHO could be identified
in the product spectra.
CH3O2 + NO f CH3O + NO2
CH3O + O2 f CH2O + HO2
HO2 + NO f OH + NO2
(31)
(25)
(32)
BrCH2CH2 + O2 + M f BrCH2CH2O2 + M (34)
BrCH2CH2O2 + RO2 f BrCH2CH2O + RO + O2 (35a)
BrCH2CH2O2 + RO2 f BrCH2CHO + ROH + O2 (35b)
Surprisingly, the sum of the yields of CO2, PAN, and CH3C-
(O)Cl was only 65 ( 7%, lower than the 80% yield expected
on the basis of the branching ratio k1a/k1. This discrepancy was
also noted by Shi et al.2 The absence of CH3CHO in these
experiments indicates that reaction 8b does not play a major
role in the chemistry of excited CH3CHClO radicals, and cannot
account for the missing carbon. Note also that OH is generated
when NO is present; this should have little effect on the observed
chemistry, however, since the branching ratio for attack of OH
on the two ends of ethyl chloride should be about equal to that
for Cl-atom attack.33
BrCH2CH2O2 + RO2 f BrCH2CH2OH + RdO + O2
(35c)
BrCH2CH2O2 + HO2 f BrCH2CH2OOH + O2 (36)
Generation of C2H4 also likely occurs following reaction 33b,
as the result of the decomposition of the BrCH2CH2 radical,
BrCH2CH2 + M f Br + C2H4 + M
(37)
To explore the extent of the effects of chemical activation,
experiments were conducted in the presence of NOx under
conditions most likely to favor reaction of the CH3CHClO
radical with O2, reaction 9, relative to its decomposition reaction
8a (i.e., under conditions of low temperature, 220 K, and high
O2 partial pressures, 300 Torr). In the NOx-free experiments
discussed above, the ratio k9[O2]/k8a was about unity for these
conditions. However, in the presence of NOx, the (CO2 + PAN)
yield exceeded that of CH3C(O)Cl by about a factor of 3,
implying that about half of the nascent CH3CHClO radicals
decompose promptly via (8a), before collisional deactivation
occurs. Chemical activation was also predicted by the theoretical
study of Hou et al.; they calculate a lifetime of ≈10-100 ps
for CH3CHClO radicals containing 8 kcal/mol of internal energy,
about the average amount expected to be imparted in the
exothermic reaction 28.32
2. Ethyl Bromide. Products observed in the photolysis of
Cl2/C2H5Br/O2/N2 mixtures at all temperatures were CO, CO2,
CH2O, CH3C(O)Br, CH3CHO, C2H4, BrCH2CHO, and CH3C-
(O)OOH. On average, these species accounted for 80 ( 10%
of the reacted carbon. The apparent yields of CH2O and C2H4
were observed to roll off with time. This roll-off is too rapid to
be explained by reaction of these species with Cl, and is very
likely due to their rapid reaction with Br-atoms.34-36 As will
be discussed in more detail below, the Cl-atom initiated
oxidation of ethyl bromide generates substantial yields of Br
(via decomposition of BrCH2CH2 and CH3CHBrO radicals);
because Br is not reactive with ethyl bromide, its main fate is
thus reaction with CH2O, CH3CHO, and C2H4 products. For
which occurs in competition with its reaction with O2, reaction
34. This reaction sequence is identical to that seen in the reaction
of Br with ethene (and other alkenes),34,36,38,39 in which the
effective rate coefficient is dependent on O2, reflecting a
competition between Br/ethene adduct decomposition and its
reaction with O2.
This competition was explored in the present case by
examining the ethene yield as a function of O2 at various
temperatures, ranging from 226 to 298 K. As discussed earlier,
C2H4 yields were taken as the intercepts of plots of ethene yield
(molar %) vs ethyl bromide conversion. Ethene yield data,
defined in this manner, are shown as a function of O2 partial
pressure and temperature in Figure 4. The data show the
expected trendss the ethene yield decreases with increasing
O2 (i.e., with increasing rate of reaction 34), and with decreasing
temperature (as the thermal stability of the BrCH2CH2 adduct
increases). From a consideration of the reaction mechanism, it
can be seen that the ethene yield is described by the following
expression:
YC H ) YBrCH CH {1/(k34[O2]/k37 + 1)}
(C)
2
4
2
2
where YC H is the observed C2H4 yield, and YBrCH CH is the
2
4
2
2
yield of BrCH2CH2, that is, the branching ratio k33b/k33. Ethene
yield data at the three highest temperatures (298, 273, and 250
K) were fit to this expression, with k37/k34 and YBrCH CH as fit
2
2
parameters. Retrieved values for k33b/k33 and k37/k34 are shown
in Table 2. The fits (shown as solid lines in Figure 4) show a

Oxidation of Ethyl Chloride and Ethyl Bromide
J. Phys. Chem. A, Vol. 106, No. 2, 2002 317
Figure 4. Yield of ethene (molar %) as a function of the O2 partial
pressure from the photolysis of Cl2/CH3CH2Br/O2/N2 mixtures at a total
pressure of 710 ( 10 Torr: solid circles, 298 K; open circles, 273 K;
solid triangles, 250 K; open triangles, 226 K. Solid lines are fits of
equation (C) to the data, see text for details.
Figure 5. Rate coeffient data obtained for the decomposition of BrCH2-
CH2 radical, reaction 37, plotted in Arrhenius form.
For completeness, a couple of ethene yield determinations
were made at 228 K. Ethene yields were found to be quite small,
5% at 20 Torr O2 partial pressure and 2% at 500 Torr. The
“fit” curve for these data shown in Figure 4 was obtained by
extrapolation of the temperature-dependent expression for k37/
k34 to 228 K, with an assumed YBrCH CH of 19% (also obtained
TABLE 2: Data for k33b/k33, k37/k34, and k37, Obtained from
an Analysis of Ethene Yield Data from the Cl-Atom-Initiated
Oxidation of Ethyl Bromide (see text for details)a
temp (K)
k33b/k33
k37/k34 (molecule cm-3
)
k37 (s-1
)
2
2
from an extrapolation of the higher temperature data). The two
measured ethene yields are consistent with these extrapolated
fit parameters.
298
273
250
226
0.29
0.26
0.22
0.19
1.2 × 1019
4.9 × 1018
2.0 × 1018
6.0 × 1017
9.1 × 107
3.7 × 107
1.5 × 107
4.5 × 106
The production of an alkene from the atmospheric oxidation
of alkyl bromides will not be unique to the case of ethyl
bromide. In fact, a similar reaction to reaction 37 has been noted
in the case of the bromacetyl radical, BrCH2CO, where Br-
atom elimination to form ketene occurs in competition with
C-C bond cleavage and addition of O2 in 1 atm air at 298 K.40
In addition, propene has recently been observed in the atmo-
spheric oxidation of n-propyl bromide,41 a compound currently
being considered for use as an industrial solvent. The elimination
of Br-atom from a â-brominated alkyl radical immediately
following the initial OH attack on the parent alkyl bromide will
serve to decrease the likelihood of the formation of long-lived
brominated organics (e.g., bromoacetone from n-propyl bro-
mide), and thus lessen the likelihood of Br transport to the
stratosphere via these intermediate species. However, it should
also be noted that data from Bedjanian et al.38,39 imply a stronger
Br/alkene bond with increasing alkene size, which will serve
to decrease the yield of alkene from the correspondingly larger
bromoalkanes. More information on the kinetics and energetics
of the Br/alkene adducts (from studies of Br + alkene kinetics
and alkyl bromide oxidation) is clearly desirable.
a Data at 226 K are obtained via extrapolation of higher temperature
values.
slight decrease in branching ratio k33b/k33 with decreasing
temperature, from about 29% at 298 K to 23% at 250 K, and
imply a slightly higher fraction of reaction occurring at the
-CH3 group than in the case of ethyl chloride (18% at room
temperature2). As expected, retrieved values of k37/k34 decreased
with decreasing temperature, reflecting the greater thermal
stability of the BrCH2CH2 radical (i.e., a decrease in k37 with
temperature, see Table 1). The k37/k34 data are described by the
following Arrhenius expression, k37/k34 ) 1.4 × 1023 exp(-
2800/T) molecule cm-3. With a reasonable assumption about
the magnitude of k34 (k ) 7.5 × 10-12 cm3 molecule-1 s-1
,
independent of temperature over the range considered here, by
analogy to the rate coefficient for reaction of C2H5 with O2 at
1 atm37), an Arrhenius expression for the decomposition reaction
37 can be estimated, k37 ) 1 × 1012 exp(-2800/T), see Table
2 and Figure 5.
These data are in reasonable agreement with k37 data obtained
from studies of the reaction of Br with ethene,34,36,38 the present
data set being about a factor of 1.5-2 lower. While this level
of agreement is within the combined uncertainties of the two
data sets, it is possible that larger k37 values are obtained in the
Br/ethene studies due to internal energy in the adduct that may
allow for a more rapid decomposition before stabilization can
occur. In any event, it is quite clear from studies of both Br +
C2H4 and of the Cl-atom-initiated oxidation of ethyl bromide
that the bond strength of the Br/C2H4 adduct is of order 6 kcal/
mol, and that this adduct has a lifetime of order 10 ns at 298
K, 1 atm.
On the basis of our measured ethene yields at low O2 partial
pressure, it is apparent that about 70% of the Cl-atom reaction
with ethyl bromide occurs via pathway 33a, with this fraction
increasing to about 80% near 220 K:
Cl + CH3CH2Br f CH3CHBr + HCl
(33a)
In a NOx-free environment, products expected following path-
way 33a include CH3C(O)Br, CH3CH(Br)OOH, CH3CHO, and
the products obtained from CH3CO chemistry (CH2O, CH3OH,
CH3C(O)OOH, CH3C(O)OH, and CO). All these species were

318 J. Phys. Chem. A, Vol. 106, No. 2, 2002
Orlando and Tyndall
indeed observed, except CH3CH(Br)OOH, which is likely to
decompose in our chamber system.
rate. On the basis of our CH3C(O)Br yield data, we estimate
that the rate of decomposition exceeds the rate for reaction with
O2 by a factor of 10, even at 226 K and in the presence of 500
Torr. Thus, with a likely value for k42 exceeding 1 × 10-14
CH3CHBr + O2 + M f CH3CHBrO2 + M
(38)
cm3 molecule-1 s-1 8,9,31
the rate coefficient for the decomposi-
,
CH3CHBrO2 + RO2 f CH3CHBrO + RO + O2
(39a)
tion reaction must be at least 2 × 106 s-1 at 226 K.
Some room-temperature ethyl bromide oxidation experiments
were also carried out in the presence of NO. In these experi-
ments, conversion of the peroxy radical CH3CHBrO2 to the oxy
radical CH3CHBrO is via the exothermic reaction 43:
CH3CHBrO2 + RO2 f CH3C(O)Br + ROH + O2 (39b)
CH3CHBrO2 + HO2 f CH3CH(Br)OOH + O2 (40a)
CH3CHBrO2 + NO f CH3CHBrO + NO2
(43)
f CH3C(O)Br + H2O + O2 (40b)
CH3CHBrO + M f CH3CO + HBr + M (41a)
f CH3CHO + Br + M (41b)
and complications due to peroxy radical self- and cross-reactions
are eliminated. In addition, the presence of NO and NO2 serves
to lower the steady-state Br-atom concentration (through the
formation of BrNO and BrNO2, and the subsequent reactions
of these species with Br15,16), allowing the concentrations of
species reactive toward Br-atoms to build up to higher levels
than was the case in the NOx-free experiments described above.
In the presence of NOx, CH3CHO is clearly observed as a
reaction product following reaction of Cl with ethyl bromide,
with an initial yield of order 60% (recall that only 70% of the
reaction occurs via abstraction at the -CH2Br group). Thus, in
the case of internally excited CH3CHBrO radicals generated
from reaction 43, Br-atom elimination dominates. Finally, we
note that no CH3C(O)Br was observed in these experiments,
again confirming the lack of importance of reaction 42 in the
chemistry of CH3CHBrO radicals.
CH3CHBrO + O2 f CH3C(O)Br + HO2
(42)
CH3C(O)Br formation can be via two pathwayssvia molecular
channels of the reaction of the peroxy radical CH3CHBrO2 with
other peroxy radicals (which would be independent of O2 partial
pressure) and via reaction of the CH3CHBrO radical with O2,
which may occur in competition with the decomposition of this
alkoxy species. CH3C(O)Br yields were found to be 5.5 ( 1.0%,
independent of O2 (varied from 20 to 500 Torr) and temperature
(varied from 226 to 298 K). The small yield of CH3C(O)Br
indicates that the radical-propagating channels of the CH3-
CHBrO2 self- and cross-reactions, (39a), predominate as has
been observed for other R-halogenated peroxy radicals.2,42
Furthermore, there is no evidence for the occurrence of reaction
42, even at temperatures as low as 226 K, and the decomposition
reactions (either 41a or 41b) must dominate. The conclusions
drawn here regarding the chemistry of the CH3CHBrO radical
are similar to those reached by Bierbach et al.20 in their study
of the Br-atom initiated oxidation of trans-2-butene. However,
these workers interpreted their observation of a small CH3C-
(O)Br yield as evidence for the occurrence of reaction 42. Our
demonstration of the independence of the CH3C(O)Br yield on
the O2 partial pressure clearly shows that this species is obtained
from chemistry of CH3CHBrO2 radicals, and not from the alkoxy
species.
The relative rate of occurrence of the two decomposition
reactions 41a and 41b is difficult to assess under the conditions
of our experiments, as the rapid reaction of Br with CH3CHO
(k ) 3.2 × 10-12 cm3 molecule-1 s-1 at 298 K)35 kept observed
CH3CHO levels very low (apparent yield ≈ 1% of the ethyl
bromide consumption), and essentially rendered the two de-
composition reactions indistinguishable. However, there is some
evidence pointing to the occurrence of reaction 41b. First, the
very fact that the apparent CH3CHO yield was observed to fall
off rapidly with ethyl bromide conversion is evidence for the
production of substantial yields of Br atoms. Furthermore, on
the basis of thermodynamic arguments, it is likely that the simple
Br-atom elimination reaction 41b is predominant. Using data
from the NASA JPL evaluation,37 the Br-atom elimination is
more exothermic than HBr elimination by about 2 kcal/mol.
This is in contrast to the situation for CH3CHClO, in which the
HCl elimination channel is about 15 kcal/mol more exothermic
than Cl-atom elimination.18 The trans-2-butene experiments of
Bierbach et al.21 also point toward the predominance of Br-
atom elimination; they observed CH3CHO, but only very minor
yields of the expected end-products of CH3CO production (i.e.,
CH2O and CH3OH). Regardless of the nature of the decomposi-
tion reaction 41, some conclusions can be drawn regarding its
Conclusion
The oxidation mechanisms of both ethyl chloride and ethyl
bromide have been studied under tropospherically relevant
conditions. Ethyl chloride oxidation proceeds mainly via the
CH3CHClO radical, whose chemistry involves decomposition
via HCl-elimination or reaction with O2. The HCl-elimination
reaction was found to possess an energy barrier of about 6 kcal/
mol. As a result of this modest barrier, chemical activation plays
an important role in the chemistry of internally excited CH3-
CHClO radicals generated from the exothermic reaction of the
CH3CHClO2 radical with NO.
Reaction of Cl-atoms with ethyl bromide occurs mostly (70-
80%) at the -CH2Br group. The CH3CHBrO radical thus
obtained reacts exclusively via Br-atom elimination under
tropospheric conditions. BrCH2CH2 radicals, generated via Cl-
atom abstraction at the -CH3 group, react via Br-atom elimina-
tion to form ethene or via addition of O2 to form a peroxy
radical. Kinetic parameters obtained for the Br-atom elimination
reaction are broadly consistent with those obtained in previous
studies of the reaction of Br with ethene. The generation of an
alkene is likely to be a general occurrence in the tropospheric
chemistry of alkyl bromides.
Acknowledgment. The National Center for Atmospheric
Research is operated by the University Corporation for Atmo-
spheric Research, under the sponsorship of the National Science
Foundation. This work was funded in part by the NASA Upper
Atmosphere Research Program. Thanks are due to Chris Cantrell
and Didier Voisin of NCAR for their comments on the
manuscript.
References and Notes
(1) Atmospheric Chemistry and Global Change; Brasseur, G. P.,
Orlando, J. J., Tyndall, G. S., Eds.; Oxford University Press: New York,
1999.

Oxidation of Ethyl Chloride and Ethyl Bromide
J. Phys. Chem. A, Vol. 106, No. 2, 2002 319
(2) Shi, J.; Wallington, T. J.; Kaiser, E. W. J. Phys. Chem. 1993, 97,
6184.
(3) Maricq, M. M.; Shi, J.; Szente, J. J.; Rimai, L.; Kaiser, E. W. J.
Phys. Chem. 1993, 97, 9686.
(25) Niedzielski, J.; Tschuikow-Roux, E.; Yano, T. Int. J. Chem. Kinet.
1984, 16, 621.
(26) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. J. Phys.
Chem. 1985, 89, 588.
(4) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1995, 99, 8669.
(5) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1994, 98, 5679.
(6) Wallington, T. J.; Orlando, J. J.; Tyndall, G. S. J. Phys. Chem.
1995, 99, 9437.
(7) Bilde, M.; Orlando, J. J.; Tyndall, G. S.; Wallington, T. J.; Hurley,
M. D.; Kaiser, E. W. J. Phys. Chem. A 1999, 103, 3963.
(8) Wu, F.; Carr, R. W. Chem. Phys. Lett. 1999, 305, 44.
(9) Wu, F.; Carr, R. W. J. Phys. Chem. A 2001, 105, 1423.
(10) Bilde, M.; Sehested, J.; Nielsen, O. J.; Wallington, T. J. J. Phys.
Chem. A 1997, 101, 5477.
(11) Wallington, T. J.; Bilde, M.; Møgelberg, T. E.; Sehested, J.; Nielsen,
O. J. J. Phys. Chem. 1996, 100, 5751.
(12) Platz, J.; Nielsen, O. J.; Sehested, J.; Wallington, T. J. J. Phys.
Chem. 1995, 99, 6570.
(27) Moortgat, G. K.; Veyret, B.; Lesclaux, R. Chem. Phys. Lett. 1989,
160, 443.
(28) Horie, O.; Moortgat, G. K. J. Chem Soc., Faraday Trans. 1992,
88, 3305.
(29) Crawford, M. A.; Wallington, T. J.; Szente, J. J.; Maricq, M. M.;
Francisco, J. S. J. Phys. Chem. A 1999, 103, 365.
(30) Orlando, J. J.; Tyndall, G. S.; Vereecken, L.; Peeters, J. J. Phys.
Chem. A 2000, 104, 11578.
(31) Atkinson, R. Int. J. Chem. Kinet. 1997, 29, 99.
(32) Orlando, J. J.; Tyndall, G. S.; Bilde, M.; Ferronato, C.; Wallington,
T. J.; Vereecken, L.; Peeters, J. J. Phys. Chem. A 1998, 102, 8116.
(33) Kwok, E. S. C.; Atkinson, R. Atmos. EnViron. 1995, 29, 1685.
(34) Barnes, I.; Bastian, V.; Becker, K. H.; Overath, R.; Tong, Z. Int. J.
Chem. Kinet. 1989, 21, 499.
(13) Wallington, T. J.; Hurley, M. D.; Schneider, W. F. Chem. Phys.
Lett. 1996, 251, 164.
(14) Villenave, E.; Lesclaux, R. Chem. Phys. Lett. 1995, 236, 376.
(15) Chen, J.; Catoire, V.; Niki, H. Chem. Phys. Lett. 1995, 245, 519.
(16) Orlando, J. J.; Tyndall, G. S.; Wallington, T. J.; Dill, M. Int. J.
Chem. Kinet. 1996, 28, 433.
(17) Orlando, J. J.; Tyndall, G. S.; Wallington, T. J. J. Phys. Chem.
1996, 100, 7026.
(18) Hou, H.; Wang, B.; Gu, Y. J. Phys. Chem. A 2000, 104, 1570.
(19) Wang, B.; Hou, H.; Gu, Y. J. Phys. Chem. A 1999, 103, 2060.
(20) Catoire, V.; Lesclaux, R.; Lightfoot, P. D.; Rayez, M. T. J. Phys.
Chem. 1994, 98, 2889.
(35) Ramacher, B.; Orlando, J. J.; Tyndall, G. S. Int. J. Chem. Kinet.
2000, 32, 460.
(36) Ramacher, B.; Orlando, J. J.; Tyndall, G. S. Int. J. Chem. Kinet.
2001, 33, 198.
(37) DeMore, W. B.; Golden, D. M.; Hampson, R. F.; Howard, C. J.;
Kolb, C. E.; Kurylo, M. J.; Molina, M. J.; Ravishankara, A. R.; Sander, S.
P. Chemical kinetics and photochemical data for use in stratospheric
modeling; evaluation number 12. NASA JPL Publ. No. 97-4, 1997.
(38) Bedjanian, Y.; Poulet, G.; Le Bras, G. J. Phys. Chem. A 1999, 103,
4026.
(39) Bedjanian, Y.; Poulet, G.; Le Bras, G. J. Phys. Chem. A 1998, 102,
5867.
(21) Bierbach, A.; Barnes, I.; Becker, K. H. Tellus 1997, 49B, 566.
(22) Shetter, R. E.; Davidson, J. A.; Cantrell, C. A.; Calvert, J. G. ReV.
Sci. Instrum. 1987, 58, 1427.
(23) Tyndall, G. S.; Orlando, J. J.; Wallington, T. J.; Hurley, M. D. Int.
J. Chem. Kinet. 1997, 29, 655.
(24) Yarwood, G.; Peng, N.; Niki, H. Int. J. Chem. Kinet. 1992, 24,
369.
(40) Chen, J.; Young, V.; Catoire, V.; Niki, H. J. Phys. Chem. 1996,
100, 6580.
(41) Gilles, M. K.; Gierczak, T.; Burkholder, J. B.; Marshall, P.;
Ravishankara, A. R. J Phys. Chem. A, in preparation.
(42) Lesclaux, R. Combination of peroxy radicals in the gas phase. In
The Chemistry of Free Radicals: Peroxyl Radicals; Alfassi, Z. B., Ed.;
John Wiley and Sons Ltd: New York, 1997.