Kinetics of the reactions of chlorine atoms with a series of acetates

Article abstract of DOI:10.1016/j.cplett.2009.04.083

The kinetics of the reactions of Cl atoms with a series of alkyl acetates were investigated using relative and absolute rate methods in 2-700 Torr of N₂ at room temperature. Consistent results were obtained using the two methods. The data from

Full text of DOI:10.1016/j.cplett.2009.04.083

Chemical Physics Letters 474 (2009) 268–272
Contents lists available at ScienceDirect
Chemical Physics Letters
journal homepage: www.elsevier.com/locate/cplett
Kinetics of the reactions of chlorine atoms with a series of acetates
Jia-Hua Xing , Kenshi Takahashi , Michael D. Hurley , Timothy J. Wallington b
a
a,*
b
a
Pioneering Research Unit, Kyoto University, Gokasho, Uji 611-0011, Japan
Systems Analytics and Environmental Sciences Department, Ford Motor Company, Mail Drop SRL-3083, Dearborn, MI 48121, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The kinetics of the reactions of Cl atoms with a series of alkyl acetates were investigated using relative
Received 3 March 2009
In final form 30 April 2009
Available online 7 May 2009
2
and absolute rate methods in 2–700 Torr of N at room temperature. Consistent results were obtained
using the two methods. The data from the absolute rate measurements were more precise and were
ꢀ11
3
ꢀ1 ꢀ1
(
in units of 10
acetate, 1.97 ± 0.24; n-butyl acetate, 15.8 ± 0.94; i-butyl acetate, 9.97 ± 0.60; s-butyl acetate,
.01 ± 0.48; and t-butyl acetate, 1.64 ± 0.10. The reactivity of alkyl acetates is discussed in terms of struc-
ture–activity relationship.
cm molecule
s ): ethyl acetate, 1.76 ± 0.11; n-propyl acetate, 7.19 ± 0.11; i-propyl
8
Ó 2009 Elsevier B.V. All rights reserved.
1
. Introduction
These reactions have been studied previously by two groups [2,3]
using laser photolysis-resonance fluorescence (LP-RF) techniques.
Notario et al. [2] reported rate constants at 298 K for reactions
(1)–(7). Cuevas et al. [3] investigated the temperature dependence
of rate constants of reactions (1), (2), and (4) over the range 265–
383 K. The rate constants for reactions (1), (2), and (4) measured
at 298 K by Notario et al. [2] are systematically higher than those
by Cuevas et al. [3] by approximately 40–70%. Cuevas et al. [3] sug-
gested this discrepancy might be explained by errors in the calcula-
tion of acetate concentrations and losses of acetates to the walls of
the reaction slow flow system in either, or both, studies. Clearly,
further kinetic investigations of these reactions are required.
To resolve the discrepancies in the kinetic database for reac-
tions of chlorine atoms with acetates and facilitate an improved
understanding of the oxidation mechanisms of esters we con-
Interest is growing in the use of biofuels to reduce the current
dependence on conventional fuels derived from petroleum and to
reduce CO emissions and hence address climate change. However,
2
the combustion and atmospheric chemistry of biofuels is not as
well documented as that of petroleum-derived fuels. Accurate
information on the oxidation mechanisms of several classes of oxy-
genated fuels including alcohols, ethers, and esters are needed to
characterize their combustion in engines and oxidation in the
atmosphere. Biodiesel fuels typically contain methyl esters of
long-chain fatty acids derived from vegetable oils or animal fats.
Esters are formed in the atmospheric oxidation of ethers, for exam-
ple Orlando [1] has observed ethyl formate and ethyl acetate from
the oxidation of diethyl ether.
We report the results of a study of the kinetics of the reactions
of chlorine atoms with: ethyl acetate (CAS# 141-78-6), n-propyl
acetate (CAS# 109-60-4), i-propyl acetate (CAS# 108-21-4), n-bu-
tyl acetate (CAS# 123-86-4), i-butyl acetate (CAS# 110-19-0), s-bu-
tyl acetate (CAS# 105-46-4) and t-butyl acetate (CAS# 540-88-5).
1 7
ducted measurements of k –k at room temperature. Two different
experimental techniques were employed. Relative rate measure-
ments were made using smog chamber/FT-IR techniques. Absolute
rate measurements were made using pulsed-laser photolysis/la-
ser-induced fluorescence (PLP/VUV-LIF) techniques.
Cl þ C
2
H
5
OCðOÞCH
3
! products
! products
k
1
ð1Þ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
ð6Þ
ð7Þ
Cl þ n-C
Cl þ ðCH
3
H
7
OCðOÞCH
3
k
2
2. Experimental
3
Þ CHOCðOÞCH
2
3
! products
k
3
2.1. Smog chamber/FT-IR techniques at Ford
Cl þ n-C
Cl þ ðCH
4
H
9
OCðOÞCH
3
! products
k
4
3
Þ CHCH
2
2
OCðOÞCH
ÞCHOCðOÞCH
Þ3COCðOÞCH
3
! products
! products
k
k
5
FT-IR/Smog chamber experiments were performed using the
Cl þ CH
3
ðC
3
2
H
5
3
6
smog chamber system at Ford Motor Company, consisting of a
40 L Pyrex reactor interfaced to a Mattson Sirus 100 spectrometer
4]. The reactor was surrounded by 22 fluorescent blacklamps (GE
1
[
Cl þ ðCH
3
! products
k
7
F15T8-BL). Chlorine atoms were produced by photolysis of molec-
ular chlorine.
*
Corresponding author. Fax: +81 774 38 4546.
E-mail addresses: kenshi.takahashi@kupru.iae.kyoto-u.ac.jp (K. Takahashi),
twalling@ford.com (T.J. Wallington).
Cl
2
þ h
m
! Cl þ Cl
ð8Þ
0
009-2614/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2009.04.083

J.-H. Xing et al. / Chemical Physics Letters 474 (2009) 268–272
269
Relative rate techniques were used to measure the rate coefficient
of the reaction of Cl atoms with acetates relative to a reference reac-
tion whose rate constant has been established previously. Kinetic
data were derived by monitoring the loss of the acetate relative to
the loss of the reference compounds, acetylene and ethylene of
reaction (9) and (10), respectively.
pumped by a single XeCl excimer laser (Lambda Physik, COMPex
201, FL3002 and Scanmate 2E). The fundamental output of the x1
laser at 425.12 nm was frequency-doubled in a BBO crystal to gen-
erate 212.56 nm light which is two-photon resonant with the 5p[1/
2] level of Kr. The wavelength of the x2 laser was ꢄ503 nm. Typical
0
pulse energies of the x1 and x2 lasers were 0.3 mJ and 4 mJ, respec-
tively. The laser beams were overlapped collinearly and focused by
an acromat lens (f = 200) into a cell containing Kr. The Kr cell and
Cl þ C
Cl þ C
2
2
H
H
2
! products
! products
ð9Þ
4
ð10Þ
2
the reaction chamber were separated by a thin MgF window. The
ꢀ1
VUV laser linewidth was estimated to be 0.8 cm full-width-at-
half-maximum (FWHM) with a Gaussian shape.
The decays of the acetate and the reference were then plotted using
the expression:
The VUV-LIF signal was detected by a solar-blind photomulti-
plier tube (EMR, 541J-08-17) through an interference bandpass fil-
ter (Acton, N135-1D). The PMT tube was mounted at right angles to
the propagation direction of the VUV probe beam and the 351 nm
photolysis beam. The 351 nm laser light and the vacuum UV laser
light crossed perpendicularly in the reaction cell. The pump and
probe lasers were operated at a repetition rate of 10 Hz. When
the 351 nm laser light was turned off, no discernible LIF signal
ꢀ½
ꢁ
ꢀ
ꢁ
acetateꢁ
k
½referenceꢁ
t
0
t
0
ln
¼ k ln
ð11Þ
½
acetateꢁ
ref
½referenceꢁ
t
t
where ½acetateꢁ , [acetate]
t
, ½referenceꢁ , and [reference]
t
are the
0
’ and ‘t’, k, and
t
0
t
0
concentrations of reactant and reference at times ‘t
ref are the rate coefficients for reactions of Cl atoms with the ace-
tate and the reference, respectively. Plots of lnð½acetateꢁ =
k
t
0
½
acetateꢁ Þ versus lnð½referenceꢁ =½referenceꢁ Þ should be linear, pass
t
t
0
t
was observed indicating that neither the probe,
photolyze Cl to any appreciable extent.
Reagents diluted with N were stored in 10-L glass-bulbs which
1 2
x , nor x lasers
through the origin and have a slope of k/kref
.
2
The loss of the reactant and reference compounds was moni-
tored by FTIR spectroscopy using an infrared path length of 27 m
and a resolution of 0.25 cm . Infrared spectra were derived from
2 co-added interferograms. Analysis of the IR spectra was
achieved through spectral stripping, in which small fractions of
the reference spectrum were subtracted incrementally from the
sample spectrum. The concentrations of reactant and reference
compounds were determined with a precision of ±1% of their initial
concentrations. Experiments were performed in 1–700 Torr of N
diluent at 296 ± 1 K. Alkyl acetates were obtained from Aldrich at
99% purity (except n-propyl acetate, 96%) and were subjected to
repeated freeze/pump/thaw cycling to remove volatile impurities
before use. Ethylene (Aldrich), acetylene (BOC), and Cl (Matheson,
2
were blackened to avoid photolysis of reactants by the room lights.
The gases used in the experiments had the following stated puri-
ties: Cl , >99% (Sumitomo Seika Co.); ethyl acetate, >99.5% (Wako
2
Pure Chemical Industries); n-propyl acetate, 98.0% (Tokyo Chemi-
cal Industry, TCI); i-propyl acetate, 99.0% (TCI); n-butyl acetate,
ꢀ1
3
>99.0% (Wako); i-butyl acetate, 99.0% (TCI); s-butyl acetate, >98.0
(
Wako); and t-butyl acetate, and N , >99.999% (Kyoto Tesisan).
2
2
To remove volatile impurities the alkyl acetates were subjected
to several freeze-pump-thaw cycles prior to use.
>
3
. Results
2
research grade) were obtained from commercial sources and were
trapped at the liquid nitrogen temperature for a cycle of freeze/
pump/thaw purification before use. Ultra-high-purity nitrogen
was used as the diluent gas.
3
.1. Relative rate measurements at Ford
Fig. 1 shows a plot of the loss of alkyl acetates versus those of
reference compounds following the UV irradiation of alkyl ace-
tate/reference/Cl mixtures in 1–700 Torr N diluent. Rate constant
ratios were derived from linear least-squares analysis of each set of
data and are summarized in Table 1 (uncertainties are two stan-
dard deviations from the linear regression analyses). The ratios in
Initial reagent concentrations were 7.4–9.9 mTorr acetate,
2
2
2
.9 mTorr ethylene or 1.5 mTorr acetylene, and 90 mTorr Cl
2
in
7
00 Torr N diluent. In smog chamber experiments it is important
2
to check for unwanted loss of reactants. In the beginning of each
set of experiments, the reaction mixtures were allowed to stand
in the dark in the chamber for 10 min. There was no observable
loss of reactants or products, suggesting that heterogeneous reac-
tions are not a significant complication in the present work. Unless
stated otherwise, quoted uncertainties are two standard deviations
from least-squares regressions.
ꢀ11
Table 1 were placed on an absolute basis using k
9
= 5.07 ꢂ 10
ꢀ11
3
ꢀ1 ꢀ1
and k10 = 9.29 ꢂ 10
cm molecule
s
[9] giving the rate con-
stants listed in Table 2 (quoted values are averages of the two
determinations with uncertainties which encompass the extremes
of the individual determinations). We estimate that uncertainties
9
in k and k10 could contribute an additional 10% uncertainty to
the values listed in Table 2.
2.2. PLP/VUV-LIF techniques at Kyoto University
The experimental setup is described in detail elsewhere [5,6].
3.2. VUV-LIF measurements at Kyoto University
PLP/VUV-LIF experiments were performed at 295 ± 2 K in N
ent. Reaction mixtures of 0.2–0.7 mTorr Cl , 1.7–28 mTorr acetate,
in 1.9–6.6 Torr of N diluent were flowed slowly through the reac-
tion cell. An excimer laser (Lambda Physik, COMPex102) was oper-
ated in a XeF mode to generate 351 nm light to photolyze Cl
Based on the Cl absorption cross section at 351 nm [7] and the
photolysis laser fluence, the initial concentration of Cl( P3/2) atoms
2
dilu-
2
In the PLP/VUV-LIF experiments, Cl /alkyl acetate/N2 gas mix-
tures were subjected to 351 nm pulsed laser irradiation to initiate
2
2
reactions (1)–(7). Cl has a broad absorption between 280 and
2
2
2
.
400 nm [10] and Cl( P ) atoms are produced following the photoab-
j
2
sorption. Matsumi et al. [11] measured the branching ratio of the
2
2
spin-orbit states of Cl( P ) produced from Cl photodissociation be-
j
2
11
ꢀ3
in the reaction chamber was estimated to be 4 ꢂ 10 atoms cm
All experiments were carried out under pseudo-first-order condi-
tions with [acetate] ꢃ [Cl]
.
tween 266 and 500 nm using resonance enhanced multi-photon
*
2
ionization spectroscopy. The reported value of [Cl ( P1/2)]/
2
0
.
[Cl( P3/2)] = 0.016 ± 0.001 at 355 nm is expected to be close to that
2
Cl( P3/2
)
atoms were detected using LIF at 134.72 nm
at the 351 nm photolysis wavelength used in the present work. The
4
2
5 2
* 2
(
3p 4s P3/2–3p
erated by two-photon resonance four-wave difference frequency
mixing (2 ) in 40 Torr of Kr [8], using two dye lasers
P
3/2 transition). Tunable VUV radiation was gen-
LIF intensity of Cl ( P1/2) at 135.17 nm corresponding to the
4
2
5 2
Cl(3p 4s P1/2–3p P1/2) transition was negligibly small compared
2
x
1
ꢀ
x
2
with that of Cl( P3/2) at 134.72 nm. We conclude that physical

2
70
J.-H. Xing et al. / Chemical Physics Letters 474 (2009) 268–272
Table 2
Rate coefficients for reactions of Cl atoms with
a series of acetates at room
temperatures.
Ester
k(Cl + acetate)a
Techniqueb
Ref.
Ethyl acetate
2.01 ± 0.25
PLP-RF
PLP-RF
RR
Notario et al. [2]
Cuevas et al. [3]
This work
1
1
1
.37 ± 0.20
.76 ± 0.26
.76 ± 0.11
PLP/VUV-LIF
This work
n-Propyl acetate
7.76 ± 0.92
PLP-RF
PLP-RF
RR
Notario et al. [2]
Cuevas et al. [3]
This work
4.60 ± 0.41
7.10 ± 2.00
7.19 ± 0.47
PLP/VUV-LIF
This work
i-Propyl acetate
n-Butyl acetate
2.68 ± 0.34
PLP-RF
RR
PLP/VUV-LIF
Notario et al. [2]
This work
This work
2
1
.49 ± 0.47
.97 ± 0.24
17.4 ± 2.0
PLP-RF
PLP-RF
RR
Notario et al. [2]
Cuevas et al. [3]
This work
12.3 ± 3.4
13.7 ± 1.8
15.8 ± 0.94
PLP/VUV-LIF
This work
i-Butyl acetate
s-Butyl acetate
t-Butyl acetate
10.8 ± 1.3
PLP-RF
RR
PLP/VUV-LIF
Notario et al. [2]
This work
This work
1
9
0.4 ± 1.4
.97 ± 0.60
9.1 ± 1.0
.92 ± 1.07
.01 ± 0.48
PLP-RF
RR
PLP/VUV-LIF
Notario et al. [2]
This work
This work
8
8
2.42 ± 0.28
PLP-RF
RR
PLP/VUV-LIF
Notario et al. [2]
This work
This work
1
1
.49 ± 0.13
.64 ± 0.10
a
ꢀ11
cm3 molecule
ꢀ1 ꢀ1
In units of 10
s
, uncertainties are 2
r
statistical errors.
b
Experimental techniques: RR, relative rate; PLP/VUV-LIF, pulsed laser photolysis
coupled with vacuum ultraviolet laser-induced fluorescence spectroscopy; PLP-RF,
pulsed laser photolysis coupled with resonance fluorescence detection.
Fig. 1. Loss of acetates versus ethylene (a) and acetylene (b) in the presence of Cl
atoms in 700 Torr N at 296 K: ethyl acetate (filled circles), n-propyl acetate (filled
2
squares), i-propyl acetate (open squares), n-butyl acetate (filled diamonds), s-butyl
acetate (open diamonds), i-butyl acetate (triangles), and t-butyl acetate (inverted
triangles). The lines are linear least-squares fits.
4
laser wavelength was fixed at the center of resonance line Cl(3p 4s
2
5 2
P3/2–3p
P
3/2) at 134.72 nm. The delay time was scanned to cover
the whole time domain of the decaying LIF signal, namely t = ꢀ30
to 300 s with step t = 5 s. At each step, the LIF signal was aver-
l
D
l
Table 1
Rate constant ratios obtained in the present relative rate measurements in
aged over 10 laser shots. The LIF signal exhibits an initial jump at
t = 0 reflecting the instantaneous production of Cl atoms from the
7
2
00 Torr N diluent at 296 K.
Referencea
2
pulsed laser photolysis of Cl . At times >1 ls there was a slow de-
Reactant
cay of the LIF signal which we attributed to chemical loss of Cl
atoms.
C
2
H
2
2 4
C H
Ethyl acetate
0.336 ± 0.040
1.46 ± 0.33
0.499 ± 0.085
2.72 ± 0.32
1.98 ± 0.16
1.76 ± 0.21
0.295 ± 0.024
0.196 ± 0.016
0.732 ± 0.073
0.264 ± 0.024
1.46 ± 0.15
1.17 ± 0.10
1.05 ± 0.13
The concentration of Cl atoms at a delay time t, [Cl]
t
, is de-
n-Propyl acetate
i-Propyl acetate
n-Butyl acetate
i-Butyl acetate
s-Butyl acetate
t-Butyl acetate
scribed by the equation:
0
½
Clꢁ ¼ ½Clꢁ expðꢀk tÞ
ð12Þ
ð13Þ
t
0
0
k ¼ k½acetateꢁ þ k
d
0.159 ± 0.013
0
where k is the pseudo-first-order rate coefficient, k is the rate con-
a
) = 5.07 ꢂ 10ꢀ11 and
Placed on an absolute basis using
k
9
(C
2
H
2
k
10
) = 9.29 ꢂ 10ꢀ cm molecule
11
3
ꢀ1 ꢀ1
d
stant for the reaction of Cl with the alkyl acetates, and k is the dif-
(
2
C H
4
s
[9].
fusive loss rate of Cl atoms.
2
The decay of the Cl( P3/2) LIF signal shown in Fig. 2 contains
information on the kinetics of the reaction of Cl atoms with the
n-propyl acetate. The curve through the data in Fig. 2 is a least-
*
2
quenching and/or chemical reaction of Cl ( P1/2) does not interfere
with our kinetic measurements.
The Cl atoms produced from photodissociation of Cl
2
at 351 nm
have a nascent kinetic energy of 12.1 kcal molecule . To thermal-
ize the translationally hot Cl atoms 2–7 Torr of N diluent was
squares fit of a first order decay to the data at delay times >1 ls.
As seen from Fig. 2, the first order fit provides a good description
of the observed decay of Cl atoms in the system. The loss of Cl
ꢀ1
2
added to the reaction mixtures. Doppler profiles of the Cl atoms
as a function of delay time were recorded by scanning the VUV laser
wavelength to ensure thermalization of the translational energy of
Cl atoms was achieved before their reaction with the alkyl acetates.
We observed that the translational energy distribution of Cl atoms
atoms followed pseudo-first-order kinetics at delay times >1 ls
in all experiments. Pseudo-first-order rate constants obtained from
fits such as that shown in Fig. 2 are plotted versus the reactant con-
centration in Fig. 3. The straight lines through the data in Fig. 3 are
linear least-squares fits which have slopes equal to the bimolecular
rate coefficients for reactions (1)–(7). The results are listed in Table
2. The listed values are the average of the results obtained in sev-
eral experimental sets. Quoted uncertainties include two standard
deviations from the least-squares fit analysis and systematic
uncertainties associated with the reactant concentration.
was thermalized by collisions with N
2
within 1
ls. Kinetic data
were acquired by fitting the Cl atom decay traces at times >1
l
s.
2
Fig. 2 shows a typical trace of the Cl( P3/2) LIF signal at
1
34.72 nm obtained in an experiment using a mixture of 0.2 mTorr
2 2
Cl and 5.0 mTorr n-propyl acetate in 1.9 Torr N diluent. The probe

J.-H. Xing et al. / Chemical Physics Letters 474 (2009) 268–272
271
we are able to decouple the time scale of chlorine atom loss from
that of the regeneration of chlorine atoms from reactions involv-
ing alkyl radical products of reactions (1)–(7) with molecular
chlorine.
Cl þ RH ! HCl þ R
R þ Cl ! RCl þ Cl
ð14Þ
ð15Þ
2
As shown in Fig. 3, the pseudo-first-order loss rates for chlorine
atoms with respect to reactions (1)–(7) in our work were of the or-
4
5 ꢀ1
der of 10 –10 s . For the experimental conditions used in the
present work with [Cl ] = 0.2–0.7 mTorr and assuming that k15 is
2
ꢀ11
3
ꢀ1 ꢀ1
of the order of 10
cm molecule
s
[12] the pseudo-first-or-
der rate for chlorine atom formation in reaction (15) is of the order
ꢀ
1
of 10–100 s . Loss of chlorine atoms via reactions (1)–(7) in the
present work was measured in the absence of interference via
regeneration of chlorine atoms via reaction (15). In the experi-
ments conducted by Notario et al. [2] and Cuevas et al. [3] the
pseudo-first-order loss rates for chlorine atoms with respect to
Fig. 2. Typical VUV-LIF Cl(2P3/2) decay observed using a mixture of 0.2 mTorr Cl
and 5.0 mTorr n-propyl acetate in 1.9 Torr of N diluent. The initial jump in the
2
2
2
profile reflects the photolytic formation of Cl atoms from Cl . The solid curve is a
least-squares first order decay fit.
3
4 ꢀ1
reactions (1)–(7) were of the order of 10 –10 s while the rate
2
3 ꢀ1
of regeneration via reaction (14) was of the order of 10 –10 s
.
The time scale for loss and regeneration of chlorine atoms was
not fully separated and as a result biexponential decays of chlorine
atoms were observed following the flash photolysis of Cl
2
/acetate/
He mixtures.
To convert alkyl radicals into alkyl peroxy radicals (RO
2
) and re-
duce regeneration of Cl atoms via reaction (15), Notario et al. [2]
1
6
ꢀ3
and Cuevas et al. [3] added up to 1 ꢂ 10 molecule cm of O
the reaction mixtures. As discussed elsewhere [13], the addition
of O brings at least two complications. Firstly, it leads to an over-
estimation of the rate constants as a result of the rapid reaction of
Cl toward RO . This overestimation will only be significant when
14[RH] is relatively small. n-Propyl, n-butyl, i-butyl, and s-butyl
acetate react rather rapidly with chlorine atoms and so additional
loss via Cl + RO reaction should not be a major complication. Sec-
ondly, O absorbs at 135 nm (Schumann-Runge continuum) and
quenches fluorescence from excited chlorine atoms. Notario et al.
2] reported a significant decrease in detection sensitivity on addi-
2
to
2
2
k
2
2
[
3
Fig. 3. Pseudo-first-order plots for Cl(2P3/2) + acetates at 295 ± 2 K: ethyl acetate
filled circles), n-propyl acetate (filled squares), i-propyl acetate (open squares), n-
butyl acetate (filled diamonds), s-butyl acetate (open diamonds), i-butyl acetate
triangles), and t-butyl acetate (inverted triangles). The lines through the data are
linear least-squares fits.
tion of O
O( D), and electronically excited O
acetates to give OH radicals, alkyl, and hence alkyl peroxy radicals.
The precise impact of the presence of O( P), O( D), and electroni-
2
. Absorption by O
2
might lead to the formation of O( P),
1
3
ꢀ
1
(
ðB
R
Þ. O( D) atoms react with
2
g
(
3
1
3
ꢀ
cally excited O
2
ðB
R
Þ on the measurements by Notario et al. [2]
g
and Cuevas et al. [3] is difficult to assess. The simplest explanation
of the discrepancy between the results from the present work (rel-
ative and absolute) and the previous studies by Notario et al. [2]
and Cuevas et al. [3] is that unwanted secondary chemistry lead
to errors in the previous work.
4
. Discussion
The results from the present work are compared to the litera-
ture data in Table 2. As seen from Table 2, consistent results were
obtained using our two different experimental techniques. Where
comparison is possible, Notario et al. [2] reported rate constants
The reactivity of chlorine atoms towards the –OCH
methyl acetate is approximately 13 times less than that towards
the –CH group in ethane [14,15]. Interestingly, the reactivity of
ethyl acetate measured in the present work, (1.76 ± 0.11) ꢂ 10
is also significantly lower than that of the –CH group in ethane
[15]) showing that the deactiva-
3
group in
3
ꢀ11
3
0–40% higher than those of Cuevas et al. [3]. Our results lie be-
tween the previous two studies. Given the similarity in approaches,
it is surprising that the two previous PLP-RF studies gave inconsis-
tent results. Cuevas et al. [3] suggested that the inconsistency
might be attributable to errors in the reactant concentrations in
one, or both, studies possibly caused by losses to the walls of the
reaction system. We have used two very different experimental
systems and approaches. The consistency of results from the two
systems suggests that the present work is free from complications
caused by uncertainties in reactant concentrations.
3
ꢀ11
3
ꢀ1 ꢀ1
(2.95 ꢂ 10
cm molecule
s
tion by the RC(O)O– extends to the b-position. Notario et al. [2] and
Cuevas et al. [3] also concluded that the RC(O)O– group has a deac-
tivating effect at the b-position. Cuevas et al. [3] calculated substi-
2
tuent factors to estimate the reactivity of –CH – groups a and b to
the ester functionality using the structure activity relationship
(SAR) developed by Aschmann and Atkinson [16]. Our measured
rate constants differ significantly from those reported by Cuevas
et al. [3] and we have recalculated the substituent factors. In the
SAR method [16], calculation of overall H-atom abstraction rate
The intensity of the pulsed laser excitation source used in the
present LIF experiments is substantially greater than that of the
resonance lamp used in the RF experiments. With greater sensi-
3 2
constants for alkanes is based on estimation of –CH , –CH –, and
tivity we are able to employ lower [Cl]
concentrations. The practical benefit of greater sensitivity is that
o
and higher alkyl acetate
>CH– group rate constants. These group rate constants
depend on the identity of the alkyl substituents around these

2
72
J.-H. Xing et al. / Chemical Physics Letters 474 (2009) 268–272
Table 3
5
. Summary
Comparison of the rate constants for reactions of Cl + acetates between the
experimental values and the calculated values based on the structure–activity
relationship.
The rate constants of the reactions of seven alkyl acetates with
chlorine atoms were investigated using PLP/VUV-LIF techniques at
Compounda
k
exp
b
k
calc
c
2
–7 Torr of N
2
diluent and smog chamber/FTIR techniques at
Ethyl acetate
1.76
2.23
7.15
14.7
10.2
1.7
2.2
6.1
12.0
7.9
7
00 Torr N diluent. The results from the absolute and the relative
2
i-Propyl acetate
n-Propyl acetate
n-Butyl acetate
i-Butyl acetate
s-Butyl acetate
methods were in good agreement. Our results are 5–50% lower
than reported by Notario et al. [2] and 15–50% higher than re-
ported by Cuevas et al. [3]. Discrepancies in the previous kinetic
database for reaction of chlorine atoms with acetates probably re-
flect complications stemming from the fact that the time scales for
chlorine atom loss and regeneration were not sufficiently decou-
pled in the previous work. In the present work the experimental
conditions were chosen to ensure adequate decoupling of the time
scales for chlorine atom loss and regeneration. We show that the
deactivating influence of the ester group extends to the b but not
8.46
6.1
a
Note that ethyl and i-propyl acetates are used for derivation of the substituent
factors.
b
ꢀ11
cm3 molecule
ꢀ1 ꢀ1
In unit of 10
s . Values are the average of the rate constants
derived in relative and absolute measurement in this study.
c
ꢀ11
3
ꢀ1 ꢀ1
In unit of 10
F(RC(O)
O–) = 0.066, F(RC(O)OCH
see text).
cm molecule
s
.
Calculation using substituent factors
2
–) = 0.36, and F(RC(O)OCH–) = 0.27 derived in this study
(
the c position and provide SAR parameters to estimate the reactiv-
ity of chlorine atoms with acetates.
groups with k(CH
k(X–CH–Y(–Z)) = ktertF(X)F(Y)F(Z), where kprim, ksec, and ktert are
the rate constants per –CH , –CH –, and >CH– group for a standard
substituent, X, Y, and Z are the substituent groups, and F(X), F(Y),
3
–X) = kprimF(X), k(X–CH
2
–Y) = ksecF(X)F(Y),
Acknowledgements
This study was supported by Program for Improvement of Re-
search Environment for Young Researchers from Special Coordina-
tion Funds for Promoting Science and Technology (SCF)
commissioned by the Ministry of Education, Culture, Sport, Science
and Technology (MEXT) of Japan (K.T.). This work was also sup-
ported by Grant-in-Aids from the MEXT for Scientific Research on
Innovative Areas ‘Impacts of Aerosols in East Asia on Plants and Hu-
man Health’.
3
2
and F(Z) are the substituent factors. For the Cl reactions, kprim
=
ꢀ
11
ꢀ11
ꢀ11
3
.32 ꢂ 10 , ksec = 8.34 ꢂ 10 , ktert = 6.09 ꢂ 10
cm molecule s–1), F(–CH
) = 1.00 and F(–CH –) = F(>CH–) =
2
(in units of
3
–1
3
F(>C<) = 0.79 have been proposed [16]. It has been shown by Chris-
tensen et al. [17] that reaction of Cl atoms with the acetyl group
CH
3
C(O)O in methyl acetate proceeds with a rate constant of
ꢀ13
3
ꢀ1 ꢀ1
<1.3 ꢂ 10
cm molecule
s . We will assume that reaction at
the acetate group makes a negligible contribution to the reactivity
of the acetates studied in the present work. The substituent factor
References
[
[
[
1] J.J. Orlando, Phys. Chem. Chem. Phys. 9 (2007) 4189.
2] A. Notario, G. Le Bras, A. Mellouki, J. Phys. Chem. A 102 (1998) 3112.
3] C.A. Cuevas, A. Notario, E. Martínez, J. Albaladejo, Atmos. Environ. 39 (2005)
5091.
3
F(RC(O)O–) was calculated from the rate constant k(Cl + CH -
ꢀ12
3
–1 –1
C(O)OCH
k(Cl + CH
at the b position were calculated to be: F(RC(O)OCH
3
) = 2.2 ꢂ 10
C(O)OCH )/kprim = 0.066. Substituent factors for attack
1
–) = (k –
cm molecule
s
[17]: F(RC(O)O–) =
3
3
[
[
4] T.J. Wallington, S.M. Japar, J. Atmos. Chem. 9 (1989) 399.
5] K. Hitsuda, K. Takahashi, Y. Matsumi, T.J. Wallington, J. Phys. Chem. A 105
2
k
ðk
sec ꢂ F(RC(O)O–) ꢂ F(–CH
3
))/kprim = 0.36, and FðRCðOÞ OCH—Þ ¼
(
2001) 5131.
ꢅ
ꢅ
2
3
—k FðRCðOÞO—Þ Fð—CH
tert
3
Þ Þ=2kprim ¼ 0:27. The substituent fac-
[6] K. Takahashi, E. Iwasaki, T. Nakayama, Y. Matsumi, T.J. Wallington, Int. J. Chem.
Kinet. 39 (2007) 328.
tor for attack at the
c
position was calculated to be:
[
[
7] R. Atkinson et al., Atmos. Chem. Phys. 7 (2007) 981.
8] R. Hilbig, R. Wallenstein, IEEE J. Quant. Electron. QE-19 (1983) 1759.
F(RC(O)OCH
C(O)OCH
–) ꢂ F(–CH
uncertainties F(RC(O)OCH
indicating that the deactivating effect does not extend to the
sition. Using the values of F(RC(O)O–), F(RC(O)OCH –), and
2
CH
2
–) = (k
))/kprim = 1.12. Within the experimental
CH –) is indistinguishable from unity
po-
2
–ksec ꢂ F(RC(O)O–) ꢂ F(–CH
2
–)–ksec ꢂ F(R-
2
3
[9] T.J. Wallington, J.M. Andino, I.M. Lorkovic, E.W. Kaiser, G. Marston, J. Phys.
Chem. A 94 (1990) 3644.
2
2
[
10] H. Okabe, Photochemistry of Small Molecules, John Wiley and Sons, New York,
c
1978.
2
[11] Y. Matsumi, K. Tonokura, M. Kawasaki, J. Chem. Phys. 97 (1992) 1065.
[12] A.J. Eskola, V.A. Lozovsky, R.S. Timonen, Int. J. Chem. Kinet. 39 (2007) 614.
F(RC(O)OCH–) derived above, the estimated rate constants
according to SAR for Cl + n-propyl, n-butyl, i-butyl, and s-butyl ace-
[
13] T. Ide, E. Iwasaki, Y. Matsumi, J.-H. Xing, K. Takahashi, T.J. Wallington, Chem.
Phys. Lett. 467 (2008) 70.
ꢀ
11
ꢀ11
ꢀ11
tates were 6.1 ꢂ 10
,
12.0 ꢂ 10
,
7.9 ꢂ 10
,
and 6.1 ꢂ
[
14] L.K. Christensen, J.C. Ball, T.J. Wallington, J. Phys. Chem. A 104 (2000) 345.
ꢀ
11
cm molecule s–1, respectively. As shown in Table 3, the
3
–1
1
0
[15] R. Atkinson et al., Atmos. Chem. Phys. 6 (2006) 3625.
[
[
16] S.M. Aschmann, R. Atkinson, Int. J. Chem. Kinet. 27 (1995) 613.
17] L.K. Christensen, J.C. Ball, T.J. Wallington, J. Phys. Chem. A 104 (2000) 345.
SAR estimates provide a reasonable description of the experimen-
tally observed reactivity trends.