Rate coefficients and mechanisms of the reaction of Cl-atoms with a series of unsaturated hydrocarbons under atmospheric conditions

Article abstract of DOI:10.1002/kin.10135

The rate coefficients and mechanisms of the reaction of Cl-atoms with a series of unsaturated hydrocarbons under atmospheric conditions were studied. It was observed the product yielded from both trans-2-butene and 1-butene were found to be oxygen dependent. The results showed that the variation of the product yielded with O₂ in the case of 1-butene resulted from competitive reaction pathways for the two β-chlorobutoxy radicals involved in the oxidation.

Full text of DOI:10.1002/kin.10135

Rate Coefficients and
Mechanisms of the
Reaction of Cl-Atoms with
a Series of Unsaturated
Hydrocarbons Under
Atmospheric Conditions
JOHN J. ORLANDO,1 GEOFFREY S. TYNDALL,1 ERIC C. APEL,1 DANIEL D. RIEMER,2
SUZANNE E. PAULSON3
1Atmospheric Chemistry Division, National Center for Atmospheric Research, Boulder, Colorado 80307-3000
2RSMAS, University of Miami, Miami, Florida 33149
3Department of Atmospheric Sciences, University of California, Los Angeles, California 90095-1565
Received 16 December 2002; accepted 21 March 2003
DOI 10.1002/kin.10135
ABSTRACT: Rate coefficients and/or mechanistic information are provided for the reaction of Cl-
atoms with a number of unsaturated species, including isoprene, methacrolein (MACR), methyl
vinyl ketone (MVK), 1,3-butadiene, trans-2-butene, and 1-butene. The following Cl-atom rate
coefficients were obtained at 298 K near 1 atm total pressure: k(isoprene) = (4.3 0.6) ×
10−10 cm3 molecule−1 s−1 (independent of pressure from 6.2 to 760 Torr); k(MVK) = (2.2
0.3) × 10−10 cm3 molecule−1 s−1; k(MACR) = (2.4 0.3) × 10−10 cm3 molecule−1 s−1; k(trans-
2-butene) = (4.0 0.5) × 10−10 cm3 molecule−1
s
−1; k(1-butene) = (3.0 0.4) × 10−10 cm3
molecule−1 −1. Products observed in the Cl-atom-initiated oxidation of the unsaturated
s
species at 298 K in 1 atm air are as follows (with % molar yields in parentheses): CH2O (9.5
1.0%), HCOCl (5.1 0.7%), and 1-chloro-3-methyl-3-buten-2-one (CMBO, not quantified) from
isoprene; chloroacetaldehyde (75 8%), CO2 (58 5%), CH2O (47 7%), CH3OH (8%), HCOCl
(7 1%), and peracetic acid (6%) from MVK; CO (52 4%), chloroacetone (42 5%), CO2
(23 2%), CH2O (18 2%), and HCOCl (5%) from MACR; CH2O (7 1%), HCOCl (3%), acrolein
(≈3%), and 4-chlorocrotonaldehyde (CCA, not quantified) from 1,3-butadiene; CH3CHO (22
3%), CO2 (13 2%), 3-chloro-2-butanone (13 4%), CH2O (7.6 1.1%), and CH3OH (1.8
0.6%) from trans-2-butene; and chloroacetaldehyde (20 3%), CH2O (7 1%), CO2 (4 1%),
Correspondence to: John J. Orlando; e-mail: orlando@acd.ucar.
edu.
Contract grant sponsor: National Science Foundation.
2003 Wiley Periodicals, Inc.
c
ꢀ

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
335
and HCOCl (4 1%) from 1-butene. Product yields from both trans-2-butene and 1-butene were
found to be O2-dependent. In the case of trans-2-butene, the observed O2-dependence is the
result of a competition between unimolecular decomposition of the CH3CH(Cl) CH(O•) CH3
radical and its reaction with O2, with kdecomp/kO2 = (1.6 0.4) × 1019 molecule cm−3. The
activation energy for decomposition is estimated at 11.5 1.5 kcal mol−1. The variation of
the product yields with O2 in the case of 1-butene results from similar competitive reaction
pathways for the two β-chlorobutoxy radicals involved in the oxidation, ClCH2CH(O•)CH2CH3
•
and OCH2CHClCH2CH3. ꢀ 2003 Wiley Periodicals, Inc. Int J Chem Kinet 35: 334–353, 2003
C
mental understanding of the pathways involved in these
oxidations is desired. Cl-atom-initiated oxidation of
unsaturated species proceeds in large part via the for-
mation of ꢀ-chloroalkoxy radicals (e.g., ClCH2CH2O
from reaction of Cl with ethene [21]). While it has been
established that unimolecular decomposition of these
species (via C C bond cleavage) is not a favorable
process for smaller (C2 and C3) hydrocarbons [21–25],
studies of larger ꢀ-chloroalkoxy radicals are lacking.
In this work, measurements of the rate coeffi-
cients for, and/or end products of, the reactions of
Cl-atoms with isoprene, methyl vinyl ketone (MVK),
methacrolein (MACR), and 1,3-butadiene are reported
and compared with previous literature data. In addi-
tion, studies of the rate coefficient and mechanism of
the reaction of Cl with both trans-2-butene and 1-
butene are reported, and product data are interpreted
in terms of the behavior of the ꢀ-chloroalkoxy radicals
involved.
INTRODUCTION
The extent to which Cl-atoms impact the chemistry of
the troposphere has been debated over the years [e.g.,
1–11]. While it is now generally considered that Cl-
atoms play only a minor role on tropospheric oxida-
tion processes on a global scale [6,7,9], it is thought
that they may exert some influence in the boundary
layer, particularly in marine and coastal environments
[1–4,8,10]. For example, heterogeneous chemistry in-
volving sea-salt aerosols is thought to provide a source
of Cl-atom precursors, particularly in the presence of
NOx (e.g., in polluted coastal regions) [1,2,8]. The in-
fluence of heterogeneous sea-salt chemistry is particu-
larly evident in the Arctic springtime, where changes in
hydrocarbon mixing ratios are clearly the result of Cl-
atom (and, in the case of some hydrocarbons, Br-atom)
chemistry [e.g., 12,13]. It has also recently been pro-
posed [14,15] that Cl-atoms produced in the photolysis
of Cl2 emitted from industrial processes may enhance
hydrocarbon oxidation rates and ozone production in
urban environments, specifically Houston, Texas.
Finlayson-Pitts and coworkers [16–19] have pro-
posed the use of unique reaction products resulting
from the reaction of Cl with isoprene (a major bio-
genic emission, which is present in the marine bound-
ary layer) and 1,3-butadiene (a major anthropogenic
emission) as atmospheric markers for the occurrence
of Cl-atom chemistry. For example, using a combina-
tion of infrared and GC/MS detection methods, lab-
oratory experiments [17] have shown that 1-chloro-
3-methyl-3-buten-2-one (CMBO) is formed following
Cl-atom addition to isoprene, though its yield has yet
to be quantified [17]. Similarly, the formation of 4-
chlorocrotonaldehyde (CCA) in substantial yield fol-
lowing Cl-atom addition to 1,3-butadiene has been ob-
served in laboratory experiments [18,19].
EXPERIMENTAL
All experiments were conducted in a stainless steel en-
vironmental chamber, which has been described pre-
viously [26,27]. The chamber is constructed of stain-
less steel, is 2 m in length, and has a volume of 47 L.
The chamber is interfaced to a Bomem Fourier trans-
form spectrometer (operating in the infrared) via a set
of Hanst-type multipass optics that provided an op-
tical path length of 32.6 m. Photolysis of gas mix-
tures in the chamber was carried out using the filtered
(235–400 nm) output of a Xe-arc lamp. Temporal pro-
files of starting materials and reaction products were
monitored primarily using FTIR spectroscopy. Infrared
spectra were recorded over the range 800–3900 cm−1
,
with a spectral resolution of 1 cm−1. For some experi-
ments, products were also identified using a gas chro-
matography/mass spectrometer (GC/MS) system [28].
ForGC/MSanalysis, sampleswerewithdrawnfromthe
chamber via a 1/8ꢁꢁ PFA Teflon tube and collected on
a 1/16ꢁꢁ Silcosteel trap held at −145◦C. To avoid reac-
tion of Cl2 with organics in the cold trap prior to injec-
tion, a procedure similar to that described in Wang and
Recent field experiments [15,20] have indeed
demonstrated the presence of CMBO and CCA in
ambient air. Clearly, in order to interpret these field
data and to quantify levels of Cl-atoms in the tropo-
sphere, accurate rate coefficient and yield data for re-
action of Cl with these unsaturated species and their
by-products are needed. Furthermore, a more funda-

336
ORLANDO ET AL.
Finlayson-Pitts [18,19] was used to remove excess Cl2
upstream of the cold trap. For this purpose, a Gelman
GF/C 47-mm glass fiber filter coated with a solution of
1%(w/v)sodiumcarbonateina1%(v/v)glycerolin1:1
methanol–water solution was placed in a Teflon filter
holder and all samples were drawn through the filter
at ∼100 mL min−1 prior to trapping. Typically, 10–
25 mL at STP was trapped. The trap was then warmed
and the contents injected onto a chromatographic col-
umn (Agilent Technologies, HP-624, 20 m × 20 mm
i.d., 1.1-ꢁm film) installed in a custom oven enclosure.
The compounds were eluted into a mass spectrometer
(Agilent Technologies, 5973), which was run in full
scan mode. The column was held at 35◦C for 1 min
and then heated at 20◦C min−1 to 120◦C, and was held
there for a total run time of 8 min.
Rate coefficients for reactions of Cl with isoprene,
MVK, MACR, trans-2-butene, and 1-butene were de-
termined via FTIR spectroscopy, using standard rela-
tive rate procedures [e.g.,16,27,29–31], with propane,
propene, and ethene used as the reference compounds.
In most cases, mixtures of Cl2, the unsaturated species
under study, and one of the reference species were
photolysed in synthetic air at a total pressure of 700–
720 Torr. For isoprene, studies were also conducted
over a range of pressures (6–760 Torr) in either air
or N2. Detailed experimental conditions for specific
experiments are given in the Results section.
Product yield experiments involved the photolysis
of mixtures of Cl2 and the unsaturated species under
study in 700 Torr of O2/N2 diluent in the absence of
NOx (see Results section for further details). Typically,
four to eight irradiations of the chamber contents were
conducted (each of duration 15–60 s), and an IR spec-
trum was recorded following each irradiation (with
the photolysis lamp blocked). Most products (CH2O,
CO2, CH3OH, peracetic acid, 3-chloro-2-butanone,
CH3CHO, acrolein) were quantified via spectral strip-
ping routines, using standard spectra obtained in our
laboratory. Other species were quantified using in-
frared absorption cross-section data from the follow-
ing sources: HCOCl [33], chloroacetaldehyde [21],
chloroacetone∗, and CH3COCl [34].
Chemicals were obtained from the following
sources: isoprene, MVK, trans-2-butene, 1-butene,
1,3-butadiene, 3-chlorobutan-2-one (all Aldrich, 99%
or better); MACR (95%, Aldrich); propene (Mathe-
son, UHP); propane (Linde, instrument grade); ethene
(Linde, CP grade); Cl2 (Matheson, UHP); NO (Linde,
UHP). Gases were used as received. Liquids (isoprene,
MVK, MACR) were degassed by several freeze-pump-
thaw cycles before use. These minor components of
the mixture were added to the reaction chamber via
expansion from smaller calibrated volumes, while N2
(boil-off from a liquid N2 dewar) and O2 (UHP Grade,
U.S. Welding) were added directly to the chamber.
The rate of disappearance of the unsaturated species
under study (UNSAT) relative to the reference com-
pound (REF) was monitored by FTIR spectroscopy,
and rate coefficient ratios, kUNSAT/kREF, were deter-
mined as follows:
RESULTS AND DISCUSSION
Rate Coefficients
Isoprene. The rate coefficient for reaction of Cl with
isoprene (k1) was determined relative to k2 over a
range of pressures (6.2–760 Torr), from the photol-
ysis of mixtures of Cl2, (1.8–3.5) × 1015 molecule
cm−3; isoprene, (2.3–8.5) × 1014 molecule cm−3; and
propane, (7–14) × 1014 molecule cm−3 in either syn-
thetic air or N2 diluent.
kUNSAT
kREF
ln([UNSAT]o/[UNSAT])
ln([REF]o/[REF])
=
Absolute values for the rate coefficients for reaction of
Cl with isoprene, MVK, MACR, trans-2-butene, and
1-butene were obtained from currently recommended
[32] expressions for reaction of Cl with propane,
k = 1.43 × 10−10 cm3 molecule−1 s−1 (independent of
pressure), propene, k = 2.63 × 10−10 cm3 molecule−1
s−1 (at 700 Torr total pressure), and/or ethene, k =
0.99 × 10−10 cm3 molecule−1 s−1 (at 700 Torr total
pressure). Each rate coefficient ratio was obtained from
four to eight irradiations of at least two fills of the cham-
ber. The ratio of the concentrations of the species un-
der study and the reference compound was varied by at
least a factor of 2 between runs. No significant varia-
tion in the measured rate coefficient ratio was observed
in any case, suggesting that there were no effects of
secondary chemistry or of spectral interferences in the
measurements.
Cl + isoprene → products
Cl + propane → products
(1)
(2)
As previously noted by Ragains and Finlayson-Pitts
[16], a slow dark reaction occurs between isoprene and
Cl2. Under the conditions employed in our experiments
(i.e., relatively low [Cl2]), and based on a rate coeffi-
cient of 1 × 10−20 cm3 molecule−1 s−1 measured in our
laboratory, the effect of this dark reaction was minimal.
∗Wallington, T. J. personal communication.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
337
4.5 × 10−10 cm3 molecule−1 s−1 near 1 atm, although,
given the uncertainty inherent in the data, a mean value
of k1 = (4.0 0.5) × 10−10 cm3 molecule−1 s−1 (the
average of all data obtained between 0.25 and 760 Torr)
also provides an adequate description.
Observed isopreneconcentrationswerecorrected by no
more than 4% to account for its occurrence.
Representative relative rate data are plotted in Fig. 1,
and all measurements are summarized in Table I. No
evidence of any variation in the ratio k1/k2 with either
total pressure or diluent gas identity was noted; the non-
weighted mean of all data gives k1/k2 = 2.99 0.27.
Using k2 = 1.43 × 10−10 cm3 molecule−1 s−1 [32], a
value of k1 = (4.3 0.6) × 10−10 cm3 molecule−1 s−1
is obtained. In all cases in this paper, uncertainties in
measured rate coefficient ratios are 1ꢂ, precision only.
Uncertainties quoted for absolute rate coefficient data
include a (typically 10%) contribution for the uncer-
tainty in the reference rate coefficient data.
MVK. The rate coefficient for reaction of Cl with
MVK, k3, was determined relative to both propane (k2)
and propene (k4), from the photolysis of mixtures of Cl2
[(6–13) × 1015 molecule cm−3], MVK [(4–7) × 1014
molecule cm−3], and either propane [(1.0–1.4) × 1015
molecule cm−3] or propene [(3.5–5.6) × 1014 molecule
cm−3], in 700 Torr synthetic air. The measured ratios
k3/k2 = 1.50 0.12 and k3/k4 = 0.81 0.07 (see
Fig. 3), combined
Numerous measurements for k1 at 298 K have been
reported in the recent literature [16,29,31,35–38], over
a range of total pressures and using a variety of rel-
ative and absolute techniques. As shown in Fig. 2,
these data all fall within the range (2.6–5.6) × 10−10
Cl + MVK → products
Cl + propene → products
(3)
(4)
cm3 molecule−1 −1. The atmospheric pressure data
s
of Bierbach et al. [35] and Fantechi et al. [29] lie
slightly above the bulk of the measurements, though
the differences are not outside the combined uncertain-
ties of the various data sets. The lowest pressure data
of Ragains and Finlayson-Pitts [16] (below 0.25 Torr)
show a slight decrease in k1 with decreasing pressure,
though no fall-off is evident in the Bedjanian et al. [36]
data obtained at slightly higher pressure. The available
data suggest a slight increase in k1 with pressure, from
about 3.5 × 10−10 cm3 molecule−1 s−1 near 1 Torr to
with the literature values for k2 = 1.43 × 10−10
cm3 molecule−1 s−1 and k4 = 2.63 × 10−10 cm3
molecule−1 s−1 [32], yield values for k3 of (2.15
0.32) and (2.27 0.30) × 10−10 cm3 molecule−1 s−1
respectively. Averagingthesevaluesyieldsk3 = (2.2
,
0.3) × 10−10 cm3 molecule−1
s
−1. This value is in ex-
cellentagreementwithotherrecentlypublisheddatafor
this reaction obtained near atmospheric pressure, k3 =
(2.1 0.3) × 10−10 cm3 molecule−1 −1
[31], k3 =
s
(2.0 0.2) × 10−10 cm3 molecule−1 s−1 [30], and
Figure 1 Decays of isoprene relative to those of propane at 296 K at different total pressures and in the presence of air (filled
symbols) or N2 (open symbols): triangles, 6.2 Torr; circles, 260 Torr; squares, 760 Torr.

338
ORLANDO ET AL.
Table I Summary of Rate Coefficient Data Obtained in This Study
Absolute
Reference
Compound
Pressure
(Torr)
Measured Rate
Coefficient Ratio
Rate Coefficient
(10−10 cm3 molecule−1 s−1
)
Compound
Isoprene
Isoprene
Isoprene
Isoprene
Isoprene
Isoprene
MVK
MVK
MACR
MACR
Propane
Propane
Propane
Propane
Propane
Propane
Propane
Propene
Propane
Propene
Propene
Propene
Ethene
6.2 (N2)
760 (N2)
270 (air)
760 (air)
112 (air)
46 (air)
700 (air)
700 (air)
700 (air)
700 (air)
700 (air)
700 (air)
700 (air)
2.91
2.89
3.05
3.26
2.82
3.01
1.50
0.81
1.77
0.82
1.54
1.12
3.17
4.16
4.13
4.36
4.66
4.03
4.30
2.15
2.27
2.53
2.16
4.0
trans-2-Butene
1-Butene
1-Butene
2.9
3.1
All data were obtained at 296 K.
k3 = (2.0 0.3) × 10−10 cm3 molecule−1 s−1 [39]. A
flow-tube determination of k3 = (1.0 0.3) × 10−10
cm3 molecule−1 s−1 [40], made at a total pressure of
1.6 Torr, indicates a pressure dependence to the rate
coefficient.
from the photolysis of mixtures of Cl2 [(5–16) ×
1015 molecule cm−3], MACR [(2.2–7.0) × 1014 mole-
cule cm−3], and either propane [(1.0–1.5) × 1015
molecule cm−3] or propene [(2.0–3.5) ×1014 molecule
cm−3], in 700 Torr synthetic air. As shown in Fig. 4 and
in Table I, rate coefficient ratios k5/k2 = 1.77 0.15
and k5/k4 = 0.82 0.07 were determined. With k2 =
1.43 × 10−10 and k4 = 2.63 × 10−10 cm3 molecule−1
s−1 [32], values for k5 of (2.53 0.39) × 10−10
cm3 molecule−1 s−1 and (2.16 0.39) × 10−10 cm3
MACR. The rate coefficient for reaction of Cl with
MACR, k5, was determined relative to both k2 and k4,
Cl + MACR → products
(5)
Figure 2 Summary of rate coefficient measurements for the reaction of Cl with isoprene as a function of pressure (all data
at 296–300 K). Solid triangles, Ref. [16]; solid circles, Ref. [36]; open diamonds, Ref. [37]; open triangles, Ref. [38]; solid
diamonds, Ref. [40]; solid square, Ref. [29]; open square, Ref. [35]; open circles, this work.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
339
Figure 3 Decays of MVK versus those of propane (filled circles) or propene (open circles) in 700 Torr air at 296 K.
molecule−1 s−1 are derived. Averaging these data
Although all determinations agree within the quoted
yields k5 = (2.35 0.40) × 10−10 cm3 molecule−1
uncertainties, the disagreement between our value
and the atmospheric pressure value of Canosa-
Mas et al. [31] is surprising, given the similar-
ity in technique. A nonweighted average of the
four separate determinations of k5 yields an aver-
age value of k5 = (2.9 0.7) × 10−10 cm3 molecule−1
s
−1. This value is somewhat smaller than data pub-
lished by Canosa-Mas et al. [31,40] [k5 = (3.2
0.5) × 10−10 cm3 molecule−1 s−1 at atmospheric pres-
sure, and k5 = (3.3 0.6) × 10−10 cm3 molecule−1
s−1 at 1.6 Torr total pressure] and by Wang et al.
[39] [k5 = (2.9 0.8) × 10−10 cm3 molecule−1
s
−1].
s−1
.
Figure 4 Decays of MACR (circles) and trans-2-butene (squares) versus those of propane (open symbols) or propene (filled
symbols).

340
ORLANDO ET AL.
Trans-2-butene and 1-Butene. The rate coefficient
for reaction of Cl with trans-2-butene, k6, was made
relative to k4, from the photolysis of mixtures of Cl2
[≈1.2 × 1016 molecule cm−3], trans-2-butene [(1.1–
1.5) ×1015 molecule cm−3], and propene [(2.8–5.6) ×
1014 molecule cm−3], in 700 Torr synthetic air.
There are now a few measurements [41–44] avail-
able in the literature with which to compare our
k6 and k7 data. For 1-butene, our value of (3.0
0.4) × 10−10 cm3 molecule−1 s−1 is in reasonable
agreement with two very recent reports [(3.5 0.1) ×
10−10 cm3 molecule−1 s−1 [42] and (3.4 0.5) ×
10−10 cm3 molecule−1 s−1 [41]], though somewhat
higher than an earlier value reported by Stutz et al.
Cl + trans-2-butene → products
(6)
[43], (2.2 0.3) × 10−10 cm3 molecule−1
s
−1. Our
value for k6, (4.0 0.5) × 10−10 cm3 molecule−1 s−1
,
The rate coefficient ratio k6/k4, obtained from the data
shown in Fig. 4, is found to be 1.54 0.12, which
is slightly higher than two recent reports, (3.3 0.5) ×
10−10 cm3 molecule−1 s−1 [41] and (3.4 0.7) ×
10−10 cm3 molecule−1 s−1 [44], though all values agree
within the mutual uncertainties. Our values seem rea-
sonable on the basis of the structures of the compounds
involved. The finding of very similar values for k4 and
k7 is consistent with near identical reactivity of the ter-
minal alkene moieties in propene and 1-butene, while
the higher degree of substitution about the double bond
in trans-2-butene leads to a larger value for k6.
yields k6 = (4.0 0.5) × 10−10 cm3 molecule−1 s−1
.
For studies of the rate coefficient for reaction of Cl
with 1-butene, k7, both propene (k4) and ethene (k8)
were used as the reference compounds:
Cl + 1-butene → products
Cl + C2H4 → products
(7)
(8)
Experiments involved the photolysis of mixtures of
Cl2 [(3–5) ×1015 molecule cm−3], 1-butene [(7–12) ×
1014 molecule cm−3], and either propene or ethene
[(2.8–4.2) ×1014 molecule cm−3], in 700 Torr syn-
thetic air. Relative rate data are shown in Fig. 5, from
which the following rate coefficient ratios are obtained:
k7/k4 = 1.12 0.06 and k7/k8 = 3.17 0.20. Us-
ing k4 = 2.63 × 10−10 cm3 molecule−1 s−1 and k8 =
0.99 × 10−10 cm3 molecule−1 s−1 at 298 K and 700
Torr total pressure yields k7 = (2.9 0.3) × 10−10
cm3 molecule−1 s−1 and (3.1 0.4) × 10−10 cm3
Product Distributions
Isoprene. Major reaction pathways expected follow-
ing addition of Cl to isoprene in NOx -free air are shown
in Scheme 1 [16,17,29,37,45–47]. Reactions of per-
oxy radicals with HO2 (to form a variety of organic
hydroperoxides) are not shown, nor are the products
likely to arise from the molecular channels of the var-
ious RO2–RO2 reactions. Major products anticipated
from the chemistry include [16,17,29,37,45] CMBO;
four isomers of chloro-methyl-butenal (CMBA);
MVK and MACR with HCOCl as their coproduct;
formaldehyde with various coproducts including MVK
and chloromethyl vinyl ketone (ClMVK); and 2-
methylene-3-butenal and HCl, which arise via abstrac-
tion from the methyl group.
molecule−1 −1, respectively, from which an average
s
value of k7 = (3.0 0.4) × 10−10 cm3 molecule−1 s−1
can be reported.
Previous studies, based on observed HCl yields,
have shown that abstraction accounts for ≈15% of the
reaction [16,29,37]. Other previously observed prod-
ucts include HCOCl (8 5% [16], 4.7 3.3% [29]),
MVK (9 5%) [17], as well as CMBO and three iso-
mers of CMBA, which were identified by GC/MS [17].
Although not strictly quantitative, the GC/MS studies
pointed to the likelihood of substantial yields of the
CMBO/CMBA species.
Our studies involved the photolysis of mixtures of
Cl2, (1.0–3.7) ×1015 molecule cm−3, and isoprene,
(0.27–1.89) ×1015 molecule cm−3, in 700–740 Torr
synthetic air. The only products observable via IR spec-
troscopy were formaldehyde (9.5 1.0%) and HCOCl
(5.1 0.7%), as shown in Fig. 6. No MVK (<4%)
or MACR (<4%) was observed. While our HCOCl
yield agrees favorably with previous studies [16,29],
Figure 5 Decays of 1-butene versus those of ethene (filled
circles) and propene (open circles) in 700 Torr air at 296 K.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
341

342
ORLANDO ET AL.
Figure 6 Product data obtained in the reaction of Cl with isoprene in 700 Torr air at 296 K. Filled circles, CH2O; open circles,
HCOCl.
our upper limit for MVK lies on the low end of the
Nordmeyer et al. [17] value. Also, Fantechi et al. [29]
postulated that MACR was formed in their experi-
ments, but subsequently destroyed. Our experiments,
conducted under conditions of low isoprene conversion
(<10% in most cases) and hence negligible product
consumption, clearly exclude any appreciable MACR
formation in this system. The observation of formalde-
hyde (not previously reported, but apparent in the spec-
tra of Fantechi et al. [29]), coupled with the very low
yield of MVK, implies that either ClMVK is pro-
duced in the chemistry (see Scheme 1) or that the
CH2 C(CH3)CHClCH2O• radical is formed and sub-
sequently decomposes to generate formaldehyde and
other products. The most likely chemistry occurring
subsequent to this decomposition is as follows:
In conjunction with the FTIR product measure-
ments, studies were also conducted using the GC/MS
instrument. Though not quantitative at this time, these
measurements revealed the presence of CMBO and
three isomers of CMBA (identified via comparison
with [17]), as well as MVK and MACR. Relative inten-
sities seen in the chromatograms were similar to those
observed by Nordmeyer et al. [17].
Examination of Scheme 1 reveals that the expected
products from the reaction of Cl with isoprene can
be categorized as either (a) CMBO and isomers of
CMBA, obtained from reaction of O2 with the various
alkoxy radicals, or (b) four-carbon oxygenates and sim-
ple coproducts, such as HCOCl and CH2O, which arise
via alkoxy radical decomposition processes. The low
yields of the readily detectible HC(O)Cl and CH2O,
and of some of the key four-carbon species expected
as coproducts (MVK and MACR), point to the like-
lihood of high yields of CMBO and CMBA, as was
concluded previously by Nordmeyer et al. [17]. These
general conclusions are also supported by recent theo-
retical studies [45–47] which show that decomposition
reactions of the six possible alkoxy radicals are rela-
tively slow, and thus that high yields of CMBO (25–
30%) and CMBA (45–50% total yield of the four pos-
sible isomers) should be obtained. Note, however, that
one quantitative discrepancy exists between the theo-
retical predictions [45] and our observations; theoreti-
cally predicted yields of MVK and HCOCl, via decom-
CH2 C(CH3)CHClCH2O• (+O2)
•
→ CH2 C(CH3)CHClO2 + CH2O
•
CH2 C(CH3)CHClO2 + RO2
→ CH2 C(CH3)CHClO• + RO +O2
CH2 C(CH3)CHClO• (+O2)
•
→ CH2 C(CH3)C(O)O2 + HCl
•
CH2 C(CH3)C(O)O2 + RO2
→ CH2 C(CH3) + CO2 + RO + O2
CH2 C(CH3) + O2 →→ CH2O, CO, CO2
•
position of the ClCH2C(O )(CH3)CH CH2 radical,

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
343
are significantly higher (≈15%) than our observations
(≈5% or less). Experiments to quantify CMBO and
CMBA products of the Cl-initiated oxidation of iso-
prene, which would provide context for atmospheric
measurements of these species and would provide fur-
ther data for comparison with theory [45], are planned
in our laboratory.
Internal addition of Cl to MVK results in the pro-
duction of the CH3C(O)CH(Cl)CH2O radical, which
may undergo decomposition, isomerization, or reac-
tion with O2 (see Scheme 2 [29]). On the basis of the
chemistry of the 1-butoxy radical [43] and data on 1-
butene oxidation given below, we conclude that iso-
merization [k ≈ (1–2) × 105 s−1 [48]] should be the
dominant fate of this radical, resulting in the produc-
tion of CH2O, CO2, and HCOCl (see Scheme 2 [29]),
though decomposition will likely lead to a similar end-
product distribution. The very low observed HCOCl
yield is thus consistent with the occurrence of only a
small amount of internal addition.
MVK. Products of the Cl-initiated oxidation of MVK
were determined using FTIR spectroscopy via the pho-
tolysis of Cl2, (3.3–5.0) × 1015 molecule cm−3, and
MVK, (4.4–6.6) × 1014 molecule cm−3, in 720 Torr
synthetic air. Products observed (see Fig. 7) in-
cluded chloroacetaldehyde (ClCH2CHO, yield 75
8%), CO2 (58 5% yield), CH2O (47 7% yield), as
well as CH3OH (≈8% yield), HCOCl (7 1% yield),
and peracetic acid (≈6% yield). The observed prod-
ucts account for 70 9% of the reacted MVK; organic
hydroperoxides, arising from reaction of peroxy radi-
cals with HO2, likely account for a large fraction of the
remaining carbon.
The large yield of chloroacetaldehyde implies that
the majority (at least 75%) of the Cl-atom addition
to MVK occurs at the terminal carbon, and that the
decomposition of the resulting oxy radical is via
CH3CO elimination (see Scheme 2). The acetylperoxy
radical produced via this route will generate CO2,
CH2O, peracetic acid, acetic acid, and CH3OH.
The combined yields of these species implies an
acetylperoxy radical yield of 64 10%, similar to that
of the chloroacetaldehyde coproduct.
The only previous studies of the Cl-initiated oxida-
tion of MVK were brief investigations by Fantechi et
al. [29] and Canosa-Mas et al. [40]. Products identified
in [29] were chloroacetaldehyde, HCOCl, methylgly-
oxal, CH2O, HCl, CO, and CO2; no yields are reported.
Methylglyoxal was not observed in our study. Elim-
•
ination of CH2Cl from the CH3C( O)CH(O )CH2Cl
radical, as proposed by Fantechi et al. [29] to account
for the methylglyoxal formation, will not compete with
the more favorable CH3C(O) elimination, and thus the
production of methylglyoxal in this system seems un-
likely. The CO observed by Fantechi et al. [29], but
not seen in our work, may result from oxidation of the
primary CH2O product. Canosa-Mas et al. [40] report
the observation of chloroacetaldehyde and CH2O (in
agreement with our work, though no yield data were
given), and CO (yield ≈0.05). They also conclude that
externaladditionofCltotheMVKdoublebondleadsto
Figure 7 Product yield data obtained in the reaction of Cl with MVK in 700 Torr air at 296 K. Filled triangles, chloroacetalde-
hyde; filled circles, CO2; open circles, CH2O; open triangles, HCOCl.

344
ORLANDO ET AL.
Scheme 2 Major reactions occurring in the reaction of Cl-atoms with MVK in NOx -free air.
the major observed products (chloroacetaldehyde and
CH2O).
formation of ≈2 CH2O, 0.4 CO, and 1.6 CO2 and this
appears to account for about 11% of the reaction (based
on the observed yield of CH2O and the yield of CO
in excess of that of chloroacetone). Acetyl chloride,
CH3COCl, a likely product of internal addition, was not
observed (yield <2%). The other possible end product
of internal addition is HC(O)CCl(CH3)CHO. Although
no strong, unidentified carbonyl absorption features
were present in the product spectra, this species could
have a fairly low vapor pressure and thus be rapidly
removed from the gas phase. However, on the basis of
the behavior of other ꢀ-chlorinated alkoxy radicals (see
discussion below), it is likely that decomposition will
be a major fate of the HC(O) CCl(CH3) CH2O rad-
ical, and thus large yields of HC(O) CCl(CH3)CHO
are not likely. Nonetheless, with the information avail-
able at present, it is not possible to exclude a significant
occurrence of Cl-atom addition to the internal carbon
in MACR.
MACR. FTIR spectroscopy was used to determine the
endproductsoftheCl-initiatedoxidationofMACR, via
the photolysis of mixtures of MACR, (2.1–7) × 1014
molecule cm−3, and Cl2, (2.1–5.6) × 1015 molecule
cm−3, in 710 Torr synthetic air. Products observed
(see Fig. 8) included CO (52 4% yield), chloroace-
tone (42 5% yield), CO2 (23 2% yield), CH2O
(18 2% yield), and HCOCl (≈5% yield), which ac-
count for 55 7% of the reacted carbon. As with any
study conducted in a NOx -free environment, organic
hydroperoxides likely contribute much of the remain-
ing carbon.
Three major reaction pathways are available for
the Cl/MACR reaction, addition to either end of the
double bond, and abstraction of the aldehydic hydro-
gen (see Scheme 3). Addition to the terminal carbon
should yield chloroacetone and CO in equal yields,
thus indicating that at least 42% of the reaction pro-
ceeds via this pathway (undetected hydroperoxide,
ClCH2C(CH3)(OOH)CHO, could also result from ter-
minal addition). Abstraction, as is the case in the re-
action of OH with MACR [49,50], should lead to the
A brief investigation of the mechanism of the reac-
tion of Cl with MACR was carried out by Fantechi
et al. [29]. They report the formation of methyl-
glyoxal, HCOCl, formic acid, HCl, CO, CO2, and
possibly CH2O, though no quantitative information
is given. From an examination of their Fig. 4, it

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
345
Figure 8 Product data obtained in the reaction of Cl with MACR in 700 Torr air at 296 K. Open circles, CO; filled circles,
chloroacetone; filled triangles, CO2; open triangles, CH2O.
Scheme 3 Major reactions occurring in the reaction of Cl-atoms with MACR in NOx -free air.

346
ORLANDO ET AL.
appears that chloroacetone (1743, 1365, 1225 cm−1
)
with O2; and CH2O, CO, and CO2, following de-
composition of CH2 CH CH(Cl)CH2O . Studies per-
•
has been misidentified as methylglyoxal (1731, 1366,
1229 cm−1). Additional chloroacetone absorption fea-
tures at 1428, 810, and 725 cm−1 are also evident
in their product spectrum, as are absorptions due to
CH2O.
More recently, Canosa-Mas et al. [40] identi-
fied chloroacetone (41 4% molar yield), CH2O
(23 2%), CO (75 6%), and HCl (18 2%) as prod-
ucts of the Cl-initiated oxidation of MACR. Although
their yields of chloroacetone and CH2O are similar to
ours, their CO yield is significantly higher. Our con-
clusions regarding the mechanism are similar to those
drawn by Canosa-Mas et al. [40]. Their HCl yield indi-
cates an 18% channel to abstraction, similar to our less
direct conclusion derived on the basis of the yields of
CO, CO2, and CH2O.
formed by Wang and Finlayson-Pitts [18,19], using a
combination of FTIR, GC/MS, and atmospheric pres-
sure ionization–MS, have identified CCA (29% yield),
ClMVK (19%), acrolein (6%), CH2O, and possibly 2-
chloro-3-butenal as products of the Cl-initiated oxi-
dation of 1,3-butadiene in NOx -free air. A number of
chlorinated hydroxybutenes, the result of the occur-
rence of the “molecular channels” of various RO2/RO2
reactions, were also identified.
Inourstudies, bothFTIRandGCanalyseswereused
to investigate the products of the the reaction of Cl with
1,3-butadiene. Experiments consisted of the photolysis
of mixtures of Cl2 [(1.1–3.1) × 1015 molecule cm−3
]
and 1,3-butadiene [≈2.8 × 1015 molecule cm−3]. Al-
though yields were quite small (see Fig. 9), CH2O
(with a yield of 7 1%), HCOCl (3 1%), and acrolein
(≈3%, not shown) were clearly observed in the in-
frared. Numerous other product absorption features
were observed, including bands centered near 978,
1067, 1107, 1275, 1340, 1414, 1721, 2736, 2815, and
2970 cm−1. Although these absorption features could
not be identified, they appear very similar in position
and intensity to the peaks shown in Fig. 2 of Wang and
Finlayson-Pitts [19], many of which were identified as
belonging to CCA and ClMVK [19]. However, with-
out authentic reference spectra for these species, we
are not able to quantify their presence in our product
spectra.
1,3-Butadiene. Major reactions expected to occur fol-
lowing the addition of Cl to 1,3-butadiene are shown
in Scheme 4 [18,19]. Three alkoxy radicals are likely
•
to be produced: ClCH2CH(O )CH CH2 from external
•
addition, ClCH2CH CHCH2O from external addi-
tion followed by allylization; and CH2 CH CH(Cl)
•
CH2O from internal Cl addition. Likely end products
•
include ClMVK, from reaction of ClCH2CH(O )CH
CH2 with O2; HC(O)Cl and acrolein, following de-
•
composition of ClCH2CH(O )CH CH2; CCA, from
•
reaction of ClCH2CH CHCH2O with O2; 2-chloro-
•
3-butenal, from reaction of CH2 CH CH(Cl)CH2O
Scheme 4 Major reactions occurring in the reaction of Cl-atoms with 1,3-butadiene in NOx -free air.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
347
Figure 9 Product data obtained in the reaction of Cl with 1,3-butadiene in 700 Torr air at 296 K. Filled circles, CH2O; open
circles, HCOCl.
The similar yields of HCOCl and acrolein ob-
served in our study are consistent with their simul-
taneous formation from the decomposition of the
ClCH2 CH(O )CH CH2 radical, as described above
and as is shown in Scheme 4:
GC/MS analysis of the Cl-initiated oxidation of 1,3-
butadiene indicated the presence (in decreasing order
of abundance) of CCA, ClMVK, and acrolein though,
as was the case with isoprene, these data were not
quantitative.
•
•
ClCH2 CH(O )CH CH2
Trans-2-butene and 1-Butene. Product studies for
the reaction of Cl-atom with trans-2-butene were car-
ried out at 1 atm pressure, over a range of O2 partial
pressures, from the photolysis of mixtures of Cl2 [(2.0–
8.4) × 1015 molecule cm−3] and trans-2-butene [(0.6–
1.2) × 1015 molecule cm−3] in O2 (25–600 Torr) and
N2 (100–700 Torr). Products observed were 3-chloro-
2-butanone, CH3CHO, CH2O, CH3COCl, CH3OH,
and CO2, which are all expected following addition
of Cl to trans-2-butene (see Scheme 5). Abstraction
has been shown to be a minor, but measurable, channel
in the reaction of Cl with isoprene, ethene, and propene
[16,29,37,51,52]. In the case of trans-2-butene, ab-
straction from the CH3 group would generate either
crotonaldehyde or MVK, neither of which are observed
in our experiments (MVK yield <4%; crotonaldehyde
yield <5%). The finding of less than 10% abstraction
in the Cl/trans-2-butene reaction is in disagreement
with a recently published structure–reactivity scheme
for Cl/alkene reactions [41], which predicts ≈15–20%
abstraction for this reaction.
→ ClCH2 + CH2 CH CHO
ClCH2 + O2 →→ HCOCl
The observed CH2O may result from the decomposi-
tion of the CH2 CH CHCl CH2O radical:
CH2 CH CH(Cl)CH2O
→ CH2 CH CHCl + CH2O
Subsequent chemistry of the CH2 CHCHCl likely
generates the vinyl radical, a further source of CH2O.
CH2 CH CHCl + O2 → CH2 CH CHClO2
CH2 CH CHClO2 + RO2
→ CH2 CH CHClO + RO + O2
CH2 CH CHClO
→ HC(O)Cl + CH2 CH →→ CO, CO2, CH2O
As shown in Fig. 10, reaction product yields
were found to be O2-dependent, with the 3-chloro-
2-butanone yield increasing with increasing O2 par-
tial pressure, and the yields of the other product
Thus, the overall occurrence of internal addition fol-
lowed by CH2 CHCH(Cl)CH2O decomposition only
need be about 4% to account for the observed CH2O.

348
ORLANDO ET AL.
Scheme 5 Major reactions occurring in the reaction of Cl-atoms with trans-2-butene in NOx -free air.
species decreasing. These observations can be ratio-
nalized (see Scheme 5) in terms of competing fates of
the CH3CHCl CH(O)CH3 radical, decomposition (to
yield CH3CHO, CH2O, CH3COCl, CH3OH, and CO2)
and reaction with O2 (to yield 3-chloro-2-butanone).
Note that the O2-dependence of the CH3COCl yield
is weaker than for the other species; its formation is
simultaneously anticorrelated with [O2] [through the
competition between reactions (9) and (10)]
were considered, since these are direct products of
the CH3CH(Cl) CH(O ) CH3 chemistry. The O2-
dependence of the yields of these species can be ex-
pressed as follows:
•
Y{CH3CH(Cl)C( O)CH3}
•
= Y(RO ) × [k10[O2]/k9/{k10[O2]/k9 + 1}] + A
(I)
ꢁ
•
Y{CH3CHO} = Y(RO )/{k10[O2]/k9 + 1} + A
CH3CH(Cl) CH(O) CH3
(II)
→ CH3CHO + CH3CHCl
(9)
CH3CH(Cl) CH(O) CH3 + O2
→ CH3 CH(Cl)C( O)CH3 + HO2 (10)
•
where Y(RO ) is the yield of the CH3CH(Cl)
CH(O ) CH3 radical. The constants A and Aꢁ ac-
count for the formation of these species from other
O2-independent sources (for example, CH3CH(Cl)-
C( O)CH3 may result from the molecular channels
•
and positively correlated with [O2], through the chem-
istry of the CH3CHClO radical [53] (see Scheme 5).
For quantitative analysis of the competition be-
tween decomposition (9) and reaction with O2
(10), yields of 3-chloro-2-butanone and CH3CHO
•
of the reactions of CH3CH(Cl) CH(OO ) CH3 with
other peroxy radicals present in the system)ꢁ. A simulta-
•
neous four-parameter [Y(RO ), k9/k10, A, A ] fit (shown

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
349
Figure 10 Product yield data in the reaction of Cl with trans-2-butene, as a function of O2 partial pressure at 296 K. Open
circles, acetaldehyde; open triangles, CO2; filled circles, chlorobutanone; filled squares, CH2O; filled triangles, CH3COCl; open
squares, CH3OH. Solid lines denote results of fits of Eqs. (I) and (II) to the yields of 2-chloro-butanone, and acetaldehyde; see
text for details.
as the solid curves in Fig. 10) of the data yields the
following results: Y(RO ) = 0. 32 0.04, A = 0.07
four identifiable products (chloroacetaldehyde, CH2O,
CO2, and HCOCl) were observed to decrease with in-
creasing O2 partial pressure. This observation points
to the occurrence of competing pathways in the chem-
istry of the two ꢀ-chlorinated alkoxy radicals derived
from Cl addition to 1-butene, CH3CH2CH(O)CH2Cl,
and CH3CH2CH(Cl)CH2O.
As shown in Scheme 6, the CH3CH2CH(O)CH2Cl
radical (derived from external Cl addition) has three
possible reaction pathways, reaction with O2 to gen-
erate CH3CH2C( O)CH2Cl, decomposition to gener-
ate ClCH2CHO, or decomposition to generate HCOCl.
Furthermore, there are no other likely pathways to the
production of either ClCH2CHO or HCOCl, thus indi-
cating that the O2-dependence of the product yield data
forthesespeciescanbeuseddirectlyintheexamination
of the chemistry of the CH3CH2CH(O)CH2Cl radical.
The O2-dependence of the HCOCl and ClCH2CHO
yields can thus be expressed as follows:
•
0.02, Aꢁ = 0.00 0.02, and k9/k10 = (1.6 0.4) ×
1019 molecule cm−3. With some reasonable assump-
tions, the measured k9/k10 ratio can be used to ob-
tain approximate kinetic and thermodynamic param-
eters for the decomposition reaction (9). Reactions
of alkoxy radicals with O2 are typically of order
1 × 10−14 cm3 molecule−1 s−1 [48]. With this assump-
tion, and an assumed A-factor for k9 at atmospheric
pressure of 2 × 1013 s−1 [54], an activation barrier
for k7 of 11.5 1.5 kcal mol−1 can be estimated, thus
yielding k9 ≈ 2 × 1013 exp(−5750/T ) s−1. The chem-
istry of the 3-chloro-2-butoxy radical is thus very sim-
ilar to the structurally similar 3-bromo-2-butoxy radi-
cal [55], which was shown to also undergo both reac-
tion with O2 and unimolecular decomposition at room
temperature.
The chemistry occurring following reaction of Cl
with 1-butene was also examined via FTIR spec-
troscopy, from the photolysis of mixtures of Cl2
[(3–4) × 1015 molecule cm−3] and 1-butene [≈7 ×
1014 molecule cm−3] at 700 Torr total pressure (O2
partial pressure 25–600 Torr; balance N2). In all ex-
periments, ClCH2CHO (chloroacetaldehyde), HCOCl,
CH2O, and CO2 were observed. Also present in the
product spectra was a carbonyl absorption band cen-
tered near 1744 cm−1, that could not be assigned to any
specific species. As shown in Fig. 11, the yields of all
•
Y(HCOCl) = Y(RO ) × {kd1/(kd1 + kd2 + kO2)}
(III)
•
Y(ClCH2CHO) = Y(RO ) × {kd2/(kd1 + kd2 + kO2)}
(IV)
•
where Y(RO ) is the formation yield of the
•
CH3CH2CH(O )CH2Cl radical, kO2 is the rate

350
ORLANDO ET AL.
Figure 11 Product yield data in the reaction of Cl with 1-butene, as a function of O2 partial pressure at 296 K. Filled circles,
chloroacetaldehyde; open circles, HC(O)Cl; open triangles, CH2O; filled triangles, CO2.
Scheme 6 Major reactions occurring in the reaction of Cl-atoms with 1-butene in NOx -free air.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
351
Table II Summary of Available Thermodynamic and
Kinetic Data for the Unimolecular Decomposition of
β-Chlorinated Alkoxy Radicals and Their Unsubstituted
Counterparts (Where Available)
coefficient for its reaction with O2, and kd1 and kd2 are
the rate coefficients for its unimolecular decomposi-
tion to HCOCl and ClCH2CHO, respectively. A least-
squares fit of the HCOCl and ClCH2CHO yield data
versus O2, with kO2 fixed at 10−14 cm3 molecule−1 s−1
Species
ꢀHr
Ea
Reference
•
[43] and with Y(RO ), kd1, and kd2 as fit parameters,
3-Chloro-2-butoxy
2-Butoxya
1.5
4.9
11.5
12.2
11.5
12.2
12.5
This work
[54]
This work
[54]
This work
[54]
yielded the following results (shown as solid lines
in Fig. 11): Y(RO ) = 0.38, kd1 = 1.1 × 104 s−1, and
•
1-Chloro-2-butoxya
2-Butoxya
kd2 = 8.5 × 104 s−1. With the assumption that the A-
factors for these decomposition reactions are about
2 × 1013 s−1 at atmospheric pressure [54], activation
barriers for the two decomposition pathways of
≈12.5 kcal mol−1 (for the HCOCl formation channel)
and 11.5 kcal mol−1 (for the ClCH2CHO formation
channel) are obtained.
4.9
6.4
1-Chloro-2-butoxyb
2-Butoxyb
14.5
2-Chloro-1-butoxy
1-Butoxy
1-Chloro-2-propoxy
2-Propoxy
2-Chloro-1-ethoxy
Ethoxy
≈12
This work
[54]
[22]
[54,56]
[22]
10.6
2.6
7.8
10.2
13.1
14.9
13.4
15.2,14.9
17
Less quantitative information can be obtained re-
16.8,17.6
[54,57]
•
garding the chemistry of the CH3CH2CHClCH2O rad-
a Decomposition via rupture of the C2 C3 bond.
b Decomposition via rupture of the C1 C2 bond.
ical, which is produced from internal Cl addition to
1-butene. Threepathwaysareplausible(seeScheme6):
decomposition to give CH2O (and also CO2), iso-
merization to yield an array of multifunctional com-
pounds (including HOCH2CH2CH(Cl)CHO), and re-
action with O2 to yield CH3CH2CHClCHO. Clearly,
some decomposition is occurring given that both CH2O
and CO2 are observed, and some reaction with O2 must
also be occurring, since the CH2O and CO2 yields
decrease with increasing O2. However, without the
ability to quantify the yield of any products derived
from the isomerization channel, a quantitative analy-
sis cannot be carried out. Nonetheless, the fact that
the decrease in the yields of CH2O and CO2 with
O2 partial pressure are similar to those for HCOCl
and ClCH2CHO, the barrier to decomposition (and
probably isomerization as well) are likely close to
endothermicity of the decomposition (ꢀHd) by the fol-
lowing relationship:
Ea = (2.4IP − 8.1) + 0.36ꢀHd
where IP refers to the ionization potential of the (com-
mon) leaving group in eV, and Ea and ꢀHd are in
kcal mol−1. Consideration of the data in Table II re-
veals that, in this context, ꢀ-chlorinated alkyl radicals
are slightly poorer leaving groups than CH3 radicals
(higher activation energy for a given endothermicity),
despite the fact that the ionization potential for CH2Cl
(8.9 eV [59]) is somewhat lower than that of the CH3
radical (9.8 eV [59]).
Thus, to a first approximation, the atmospheric
chemistry of the ꢀ-chlorinated alkoxy radicals is sim-
ilar to that of nonsubstituted (i.e., alkane-derived)
species [48,54] of similar size and structure. For ex-
ample, smaller and less branched species (e.g., ethoxy,
propoxy, and their ꢀ-chlorinated counterparts) pre-
dominantly react with O2. On the other hand, decom-
position plays a role in the atmospheric chemistry
of larger, more branched species (such as 2-butoxy,
branchedC4 andlargerradicals, andtheirꢀ-chlorinated
counterparts). Decomposition plays a slightly larger
role in the atmospheric chemistry of some mid-size
ꢀ-chlorinated species in comparison with their unsub-
stituted counterparts (for example, 1-chloro-2-butoxy
and 2-chloro-1-butoxy).
12 kcal mol−1
.
Summary of β-Chloroalkoxy Radical
Chemistry
The data presented here for a number of ꢀ-chlorinated
alkoxy radicals, combined with recent theoretical stud-
ies [22,45], allow for some general conclusions to be
drawn. The available thermodynamic and kinetic data
([22] and this work) for the unimolecular decomposi-
tion of some of the simpler ꢀ-chlorinated alkoxy rad-
icals are collected in Table II. The decompositions of
the ꢀ-chlorinated species are characterized by some-
what lower ꢀH values (3–5 kcal mol−1) than for the
corresponding unsubstituted radical decompositions
[54,56,57], which leads to only a modest reduction in
thecorrespondingactivationenergies(0–2kcal mol−1).
It has been shown [48,58] that the activation energy
(Ea) for a series of alkoxy radical decompositions
involving a common leaving group is related to the
From the standpoint of the formation of unique at-
mospheric tracers of Cl-atom chemistry, largest tracer
yields are expected from the oxidation of the smaller
alkenes (e.g., nearly 100% yield of chloroacetalde-
hyde from ethene, and chloroacetone from propene are

352
ORLANDO ET AL.
expected [21–25] in the presence of NOx ), while lesser
yields are obtained from larger and more branched
species (e.g., the yield of 1-chloro-2-butanone from 1-
butene is probably about 30% in the presence of NOx ).
Reasonably large tracer yields also result from the oxi-
dation of the dienes (1,3-butadiene and isoprene) [16–
19,45], since decomposition of the chlorinated oxy rad-
icals formed from these species are not favorable pro-
cesses [45], often involving the formation vinyl-type
radicals.
Silva, P. J.; Jimenez, J. L.; Canagaratna, M. R.; Neece,
J. D.; Mullins, C. B.; Allen, D. T. Atmos Environ 2003,
37, 1393–1400.
16. Ragains, M. L.; Finlayson-Pitts, B. J. J Phys Chem A
1997, 101, 1509–1517.
17. Nordmeyer, T.; Wang, W.; Ragains, M. L.; Finlayson-
Pitts, B. J.; Spicer, C. W.; Plastridge, R. A. Geophys Res
Lett 1997, 24, 1615–1618.
18. Wang, W.; Finlayson-Pitts, B. J. Geophys Res Lett 2000,
27, 947–950.
19. Wang, W.; Finlayson-Pitts, B. J. J Geophys Res 2001,
106, 4939–4958.
20. Riemer, D. D.; Apel, E. C.; Orlando, J. J.; Tanaka, P. L.;
Allen, D.; Neece, J.; manuscript in preparation.
21. Orlando, J. J.; Tyndall, G. S.; Bilde, M.; Ferronato, C.;
Wallington, T. J.; Vereecken, L.; Peeters, J. J Phys Chem
A 1998, 102, 8116–8123.
The authors are indebted to Alan Fried and Dan Marsh of
NCAR for their careful reading of the manuscript, and to two
anonymous reviewers for helpful suggestions.
22. Kukui, A.; Le Bras, G. Phys Chem Chem Phys 2001, 3,
175–178.
BIBLIOGRAPHY
23. Yarwood, G.; Peng, N.; Niki, H. Int J Chem Kinet 1992,
24, 369–383.
24. Wallington, T. J.; Andino, J. M.; Japar, S. M. Chem Phys
Lett 1990, 165, 189–194.
25. Kleindienst, T. E.; Shepson, P. B.; Nero, C. M. Int J Chem
Kinet 1989, 21, 863–884.
1. Keene, W. C.; Pszenny, A. A. P.; Jacob, D. J.; Duce,
R. A.; Galloway, J. N.; Schultz-Tokos, J. J.; Sievering,
H.; Boatman, J. F. Global Biogeochem Cycles 1990, 4,
407–430.
2. Finlayson-Pitts, B. J. Res Chem Intermed 1993, 19, 235–
249.
26. Shetter, R. E.; Davidson, J. A.; Cantrell, C. A.; Calvert,
J. G. Rev Sci Instrum 1987, 58, 1427–1428.
27. Orlando, J. J. Int J Chem Kinet 1999, 31, 515–524.
28. Apel, E. C.; Riemer, D. D.; Hills, A.; Baugh, W.;
Orlando, J.; Faloona, I.; Tan, D.; Brune, W.; Lamb, B.;
Westberg, H.; Carroll, M. A.; Thornberry, T.; Geron,
C. J Geophys Res 2002, 107(D3), DOI: 10.1029/
2000JD000225.
29. Fantechi, G.; Jensen, N. R.; Saastad, O.; Hjorth, J.;
Peeters, J. J Atmos Chem 1998, 31, 247–267.
30. Finlayson-Pitts, B. J.; Keoshian, C. J.; Buehler, B.; Ezell,
A. A. Int J Chem Kinet 1999, 31, 491–499.
31. Canosa-Mas, C. E.; Hutton-Squire, H. R.; King, M. D.;
Stewart, D. J.; Thompson, K. C.; Wayne, R. P. J Atmos
Chem 1999, 34, 163–170.
32. Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R.
F.; Kerr, J. A.; Troe, J. J Phys Chem Ref Data 1997, 26,
1326–1499.
33. Wallington, T. J.; Orlando, J. J.; Tyndall, G. S. J Phys
Chem 1995, 99, 9437–9442.
34. Tyndall, G. S.; Orlando, J. J.; Wallington, T. J.; Hurley,
M. J. Int J Chem Kinet 1997, 29, 655–663.
35. Bierbach, A.; Barnes, I.; Becker, K. H. Int J Chem Kinet
1996, 28, 565–577.
36. Bedjanian, Y.; Laverdet, G.; Le Bras, G. J Phys Chem A
1998, 102, 953–959.
3. Graedel, T. E.; Keene, W. C. Global Biogeochem Cycles
1995, 9, 47–77.
4. Singh, H. B.; Gregory, G. L.; Anderson, B.; Browell, E.;
Sachse, G. W.; Davis, D. D.; Crawford, J.; Bradshaw, J.
D.; Talbot, R.; Blake, D. R.; Thornton, D.; Newell, R.;
Merrill, J. J Geophys Res 1996, 101, 1907–1917.
5. Wingenter, O. W.; Kubo, M. K.; Blake, N. J.; Smith, T.
W., Jr.; Blake, D. R.; Rowland, F. S. J Geophys Res 1996,
101, 4331–4340.
6. Singh, H. B.; Thankur, A. N.; Chem, Y. E.; Kanakidou,
M. Geophys Res Lett 1996, 23, 1529–1532.
7. Rudolph, J.; Koppmann, R.; Plass-Du¨lmer, Ch. Atmos
Environ 1996, 30, 1887–1894.
8. Spicer, C. W.; Chapman, E. G.; Finlayson-Pitts, B. J.;
Plastridge, R. A.; Hubbe, J. M.; Fast, J. D.; Berkowitz,
C. M. Nature 1998, 394, 353–356.
9. Rudolph, J.; Ramacher, B.; Plass-Du¨lmer, Ch.; Mu¨ller,
K.-P.; Koppmann, R. Tellus 1997, 49B, 592–601.
10. Oum, K. W.; Lakin, M. J.; DeHaan, D. O.; Brauers, T.;
Finlayson-Pitts, B. J. Science 1998, 279, 74–77.
11. Wingenter, O. W.; Blake, D. R.; Blake, N. J.; Sive, B. C.;
Rowland, F. S. J Geophys Res 1999, 104, 21819–21828.
12. Jobson, B. T.; Niki, H.; Yokouchi, Y.; Bottenheim, J.;
Hopper, F.; Leaitch, R. J Geophys Res 1994, 99, 25355–
25368.
13. Ramacher, B.; Rudolph, J.; Koppmann, R. Tellus 1997,
49B, 607–623.
37. Suh, I.; Zhang, R. J Phys Chem A 2000, 104, 6590–
6596.
14. Tanaka, P. L.; Oldfield, S.; Neece, J. D.; Mullins, C.
B.; Allen, D. T. Environ Sci Technol 2000, 34, 4470–
4473.
15. Tanaka, P. L.; Riemer, D. D.; Chang, S.; Yarwood, G.;
McDonald-Buller, E. C.; Apel, E. C.; Orlando, J. J.;
38. Notario, A.; Le Bras, G.; Mellouki, A. Chem Phys Lett
1997, 281, 421–425.
39. Wang, W.; Ezell, M. J.; Ezell, A. A.; Soskin, G.;
Finlayson-Pitts, B. J. Phys Chem Chem Phys 2002, 4,
1824–1831.

RATE COEFFICIENTS AND MECHANISMS OF THE REACTION OF Cl-ATOMS
353
40. Canosa-Mas, C. E.; Cotter, E. S. N.; Duffy, J.;
Res Lett 1999, 26, 2191–2195.
Thompson, K. C.; Wayne, R. P. Phys Chem Chem Phys
2001, 3, 3075–3084.
51. Kaiser, E. W.; Wallington, T. J. J Phys Chem 1996, 100,
4111–4119.
41. Ezell, M. J.; Wang, W.; Ezell, A. A.; Soskin, G.;
Finlayson-Pitts, B. J. Phys Chem Chem Phys 2002, 4,
5813–5820.
42. Coquet, S.; Ariya, P. A. Int J Chem Kinet 2000, 32, 478–
484.
43. Stutz, J.; Ezell, M. J.; Ezell, A. A.; Finlayson-Pitts, B. J.
J Phys Chem A 1998, 102, 8510–8519.
44. Ullerstam, M.; Ljungstro¨m, E.; Langer, S. Phys Chem
Chem Phys 2001, 3, 986–992.
52. Kaiser, E. W.; Wallington, T. J. J Phys Chem 1996, 100,
9788–9793.
53. Shi, J.; Wallington, T. J.; Kaiser, E. W. J Phys Chem
1993, 97, 6184–6192.
54. Somnitz, H.; Zellner, R. Phys Chem Chem Phys 2000,
2, 1907–1918.
55. Bierbach, A.; Barnes, I.; Becker, K. H. Tellus 1997, 49B,
566–582.
56. Devolder, P.; Fittschen, Ch.; Frenzel, A.; Hippler, H.;
Poskrebyshev, G.; Striebel, F.; Viskolcz, B. Phys Chem
Chem Phys 1999, 1, 675–681.
45. Zhang, D.; Zhang, R.; Allen, D. T. J Chem Phys 2003,
118, 1794–1801.
46. Lei, W.; Zhang, R. J. Chem Phys 2000, 113, 153–157.
47. Lei, W.; Zhang, R.; Molina, L. T.; Molina, M. J. J Phys
Chem 2002, 106, 6415–6420.
48. Atkinson, R. Int J Chem Kinet 1997, 29, 99–111.
49. Tuazon, E. C.; Atkinson, R. Int J Chem Kinet 1990, 22,
1221–1236.
57. Caralp, F.; Devolder, P.; Fittschen, Ch.; Gomez, N.;
Hippler, H.; Mereau, R.; Rayez, M. T.; Striebel, F.;
Viskolcz, B. Phys Chem Chem Phys 1999, 1, 2935–
2944.
58. Choo, K. Y.; Benson, S. W. Int J Chem Kinet 1981, 13,
833–844.
50. Orlando, J. J.; Tyndall, G. S.; Paulson, S. E. Geophys
59. NIST WebBook, Available at http://webbook.nist.gov.