Kinetics and products of gas-phase reactions of ozone with methyl methacrylate, methyl acrylate, and ethyl acrylate

Article abstract of DOI:10.1021/jp104451v

The kinetics and products of the gas-phase reactions of ozone with methyl methacrylate, methyl acrylate, and ethyl acrylate have been investigated at 760 Torr total pressure of air and 294 ± 2 K. The rate coefficients obtained (in cm³ molecule^-1 s^-1 units) were as follows: k(methyl methacrylate) = (6.7 ± 0.9) × 10^-18, k(methyl acrylate) = (0.95 ± 0.07) × 10^-18, and k(ethyl acrylate) = (1.3 ± 0.1) × 10^-18. In addition to formaldehyde being observed as a product of the three reactions, the other major reaction products were methyl pyruvate from reaction of ozone with methyl methacrylate, methyl glyoxylate from reaction of ozone with methyl acrylate, and ethyl glyoxylate from reaction of ozone with ethyl acrylate. Possible reaction mechanisms leading to the observed products are presented and discussed.

Full text of DOI:10.1021/jp104451v

8
376
J. Phys. Chem. A 2010, 114, 8376–8383
Kinetics and Products of Gas-Phase Reactions of Ozone with Methyl Methacrylate, Methyl
Acrylate, and Ethyl Acrylate
F. Bernard, G. Eyglunent, V. Da e¨ le, and A. Mellouki*
ICARE-CNRS, 1C AVenue de la Recherche Scientifique, 45071 Orl e´ ans, Cedex 02, France
ReceiVed: May 16, 2010; ReVised Manuscript ReceiVed: July 1, 2010
The kinetics and products of the gas-phase reactions of ozone with methyl methacrylate, methyl acrylate, and
ethyl acrylate have been investigated at 760 Torr total pressure of air and 294 ( 2 K. The rate coefficients
3
-1 -1
-18
obtained (in cm molecule
s
units) were as follows: k(methyl methacrylate) ) (6.7 ( 0.9) × 10 , k(methyl
-18
-18
acrylate) ) (0.95 ( 0.07) × 10 , and k(ethyl acrylate) ) (1.3 ( 0.1) × 10 . In addition to formaldehyde
being observed as a product of the three reactions, the other major reaction products were methyl pyruvate
from reaction of ozone with methyl methacrylate, methyl glyoxylate from reaction of ozone with methyl
acrylate, and ethyl glyoxylate from reaction of ozone with ethyl acrylate. Possible reaction mechanisms leading
to the observed products are presented and discussed.
1
. Introduction
Acrylate esters are volatile liquids, soluble in most organic
Polymers and copolymers of methyl methacrylate are also used
in undissolved surface coatings, adhesives, etc. Methyl acrylate
can be used as a copolymer in the process of polymerization of
polyanionic cellulose (PAC) polymers to reduce the glass-
transition temperature of PAC polymers. It is also used in
making vitamin B1. Ethyl acrylate is used in the production of
polymers including resins, plastics, etc.
In this work, the reaction rate coefficients of ozone with the
three acrylates have been measured using both absolute and
relative rate methods
solvents but only slightly soluble in water. Containing a double
bond and a functional carboxylic group, they are highly reactive
chemicals and, therefore, used mainly as intermediates in the
production of other materials. For example, individual molecules
of acrylic acid or esters, monomers, readily combine with
themselves or other monomers to form long chains of repeating
units to constitute polymers which have chemical and physical
properties different from the constituent monomers. They are
used in formulating paints and dispersions for paints, inks, and
adhesives. They are also used in making cleaning products,
antioxidant agents, surfactants for making aqueous resins, and
dispersions for textiles and papers. Part of these compounds
may escape into the atmosphere, where they can play a role in
the atmospheric chemistry and air quality at local and regional
scales.
CH dCH(CH )C(O)OCH + O f products k1
2
3
3
3
CH dCHC(O)OCH + O f products k2
2
3
3
CH
dCHC(O)OCH
CH
+ O
f products
k
3
2
2
3
3
Acrylate esters contain a vinyl group, which makes them more
reactive than simple esters in the gas phase toward the main
atmospheric oxidants such as hydroxyl radicals (OH), nitrate
The mechanisms of these reactions have been also investigated
using in situ FT-IR spectroscopy, which enabled us to determine
the product formation yields. The results obtained are compared
radicals (NO
3 3
), ozone (O ), and chlorine atoms (Cl). Although
numerous studies were conducted on the atmospheric degrada-
with previous studies on reactions of ozone with unsaturated
1
-3
tion of saturated esters
reactions with OH radicals and Cl atoms,
and mechanistic investigations have been performed on the
and unsaturated esters initiated by
15,21
esters. To date, only Grosjean and Grosjean
reported reaction
4
-7
8-12
fewer kinetic
product formation yields from reaction of methyl acrylate with
14
16
ozone while Grosjean et al. and Grosjean and Grosjean
1
3-17
unsaturated esters through reaction with ozone
and NO
3
reported the single kinetic studies of O
late and methyl acrylate, respectively.
3
with methyl methacry-
radicals.1
8-20
In this work, we report kinetic and mechanistic studies on
ozone reactions with three acrylates: methyl methacrylate
2
. Experimental Section
Experiments were carried out in a 7300 L Teflon chamber in
(
MMA, CH
CH dCHC(O)OCH
O)OCH CH ). Methyl methacrylate is the monomer to make
2
dC(CH
3
)C(O)OCH
3
), methyl acrylate (MA,
2
3
), and ethyl acrylate (EA, CH
2
dCHC-
the dark at 294 ( 2 K and 760 Torr total pressure of purified
air (<5% of relative humidity). Rapid mixing of reactants was
ensured within 1-2 min using a set of two fans made of Teflon
fitted into the chamber. After each experiment, the chamber was
flushed with purified air for at least 12 h. Reactants were
introduced into the chamber either via the liquid phase by
streaming purified air or as a gas using a calibrated cylinder
with a known volume (0.9 L) equipped with two pressure
sensors (0-10 and 0-100 Torr, MKS Baratron). The temper-
ature and relative humidity data were recorded by a combined
(
2
3
polymethyl methacrylate acrylic plastics (PMMA), used as a
shatterproof replacement for glass (this application consumes
approximately 80% of the MMA). Methyl methacrylate is also
used for production of the copolymer methyl methacrylate-
butadiene-styrene (MBS), used as a modifier for PVC. In
addition, MMA is a key ingredient in a number of products.
*
To whom correspondence should be addressed. E-mail: mellouki@
cnrs-orleans.fr.
1
0.1021/jp104451v  2010 American Chemical Society
Published on Web 07/22/2010

Kinetics and Products of Gas-Phase Reactions of Ozone
J. Phys. Chem. A, Vol. 114, No. 32, 2010 8377
sensor (T870, Series T800, Dostmann Electronic). Ozone
concentrations were monitored by a Horiba (APOA-360)
ultraviolet ozone analyzer. Temporal resolution of the ozone
monitor enables obtaining a measurement every 5 s. The other
reactants were monitored by in situ Fourier transform infrared
absorption spectroscopy (Nicolet 5700 Magna FT-IR spectrom-
eter) coupled to a white-type mirror system resulting in an
optical path of about 129 m.
Experiments were conducted in the presence of an excess
amount of cyclohexane to avoid secondary reactions with OH
radicals (which could be formed during ozonolysis of unsatur-
ated compounds). The experiments have been repeated to enable
us to derive averaged ratios for kesters/kref. The quoted uncertain-
ties on the obtained rate coefficients originate from the
uncertainties associated to the value of the slope (one standard
deviation, 1σ).
Rate coefficients for the studied reactions were determined
using both absolute and relative methods. In the absolute method
studies, the decays of ozone were measured in the presence of
excess concentrations of the esters. For each reaction investi-
gated, one run was conducted in the presence of cyclohexane
to scavenge OH radicals, which could be produced during
ozonolysis. The initial ester concentrations in the chamber were
In addition to the kinetic studies, a series of experiments was
performed to investigate the mechanism of the ozonolysis of
the three esters. These runs were conducted in the presence of
cyclohexane (as OH scavenger) at 760 Torr of purified air and
293 ( 1 K. Initial concentrations of reactants were in the range
13
-3
(1.25-5.01) × 10 molecules cm . Organic compounds were
continuously monitored by FT-IR spectroscopy and ozone by
the Horiba analyzer. FT-IR spectra were recorded every 4-5
1
3
-3
in the range of (1.36-12.2) × 10 molecules cm for methyl
1
3
-3
-1
methacrylate and (2.8-14.5) × 10 molecules cm for methyl
min by coadding 110-120 scans with a resolution 1 cm . In
acrylate and ethyl acrylate, while the initial ozone concentrations
all experiments, esters were first introduced in the chamber in
order to assess their loss rate, in the presence of cyclohexane.
The temporal concentration profiles of organic compounds were
monitored for 1 h before adding ozone to initiate the reaction.
Unsaturated esters were monitored using absorption features
12
-3
were in the range of (1.55-13.2) × 10 molecules cm . Under
pseudo-first-order conditions, the rate of disappearance of O
followed a simple exponential rate law
3
-
1
-
k′t
over the following wavenumber ranges (cm ): methyl meth-
acrylate, 1262-1112; methyl acrylate, 1252-1156; ethyl acry-
late, 1348-1262. Gas-phase products of the investigated
reactions were identified and tentatively assigned by FT-IR
spectroscopy using reference spectra when possible and their
formation yields obtained by plotting the concentration of
formed products versus the concentration of consumed ester.
Three runs were performed for each unsaturated ester-ozone-
cyclohexane-air mixture system. The resulting formation yields
of gas-phase products arising from each experiment were then
averaged, leading to the final gas-phase product yields. The
quoted error on the yield values originates from one standard
deviation (1σ) on the averaged yield.
[
O ] ) [O ] e
3 t 3 0
where k′ ) k
rate coefficients for the reaction of O
be determined and k′ the first-order rate coefficient for O
removal in the absence of ester, which was due to dilution and
loss of ozone on the wall of the chamber. Further, k can be
versus [ester]
0
i
[ester] + k′
0
with k
i
(i ) 1-3) representing the
3
with the three esters to
0
3
i
obtained from the slope of the plots of k′ - k′
0
considering the following equation
k′-k′ ) k [ester]
0
0
i
Chemicals. Unsaturated esters (methyl methacrylate, methyl
acrylate, and ethyl acrylate) and methyl pyruvate were provided
by Fluka with a stated purity of g99.0% and 97% (GC),
respectively. Ethyl glyoxylate (50% in toluene) was provided
by Alfa Aesar. Cyclohexane was obtained from Sigma Aldrich
with a purity of g99.9%. Propene and ethene were provided
by Air Liquide.
Additional measurements of the rate coefficients were conducted
using the well-established relative rate method. Kinetic data were
derived by monitoring the loss of unsaturated esters relative to
one or more reference compounds. The loss rates of esters and
reference compounds in the absence of ozone, k
d
(esters) and
-1
k
d
(ref), (in s ), respectively, were measured before each run,
and the values obtained have been considered in the treatment
of the data. Provided that the unsaturated esters and references
were lost only by reaction with ozone and by dilution and neither
ester nor the reference compound was formed, it can be shown
that
3. Results and Discussion
3.1. Rate Coefficient Measurements. Absolute Study. Fig-
ure 1 displays an example of the typical loss of ozone versus
reaction time in the absence and presence of different concentra-
tions of the ester (methyl methacrylate), and Figure 2 shows
ln([esters] /[esters] ) - k (esters)t )
the plots of (k′ - k′
0
) versus the ester concentrations for different
were derived from the least-squares fit of
0
t
d
esters. Values of k
i
kesters/k (ln([ref] /[ref] ) - k (ref)t)
ref
0
t
d
the straight lines. The measured k′
absence of added reactants was k′
0
ozone decay rates in the
-
4
-1
0
) (0.17-0.25) × 10 s .
0 t 0
where [esters] , [esters] , [ref] , and [ref]
t
are the concentrations
Loss rates of ozone from dilution and adsorption on Teflon film
were measured before addition of the ester. The contribution
of this later loss was found to represent 3-19% of the total
decay of ozone for the experiments with methyl methacrylate
and 9-40% for those with methyl acrylate and ethyl acrylate.
The experimental conditions and measured values of the rate
coefficients are listed in Table 1. As shown in Table 1 and Figure
2, the presence of an excess amount of cyclohexane led to the
same rate coefficient values, which indicate that secondary
reactions, involving, for example, OH radicals, were not
significant in our experimental conditions. This was expected
for possible OH reactions because the latter would preferentially
of the esters and reference at times t
are the rate coefficients for reactions of esters and the reference
with O . Plots of (ln([esters] /[esters] ) - k (esters)t) versus
ln([ref] /[ref] ) - k (ref)t) should be linear, pass through the
origin, and have a slope of kesters/kref
0
and t and kesters and kref
3
0
t
d
(
0
t
d
.
Ethene and propene were used as the reference compounds.
Reactants and reference compounds were monitored by FT-IR
spectroscopy using absorption features over the following
-1
wavenumber ranges (cm ): methyl methacrylate, 1250-1120;
methyl acrylate, 1100-1049 and 1252-1156; ethyl acrylate,
1
332-1246; ethene, 1055-873; propene, 968-867.

8
378 J. Phys. Chem. A, Vol. 114, No. 32, 2010
Bernard et al.
reactions of O
3
with ethene and propene. Prior to addition of
3
O , the loss rates of reactants were measured in the presence of
cyclohexane (in order to estimate the dilution and adsorption
rates on the Teflon film). The initial concentrations of reactants
1
3
-3
were in the range of (2.51-8.01) × 10 molecules cm , while
1
3
those of ozone were in the range of (3.26-10.0) × 10
-
3
molecules cm . The duration of the experiments was varied
from 30 to 120 min, depending on the reactivity of the studied
d d
esters. k (esters) and k (ref) were found to be similar in our
-
5
-5
experimental conditions, ranging from 1.1 × 10 to 1.3 × 10
-
1
s . The loss of the unsaturated esters through reaction with
ozone represents about 87-92% for methyl methacrylate,
2
6-82% for methyl acrylate, and 40-78% for ethyl acrylate
of the total loss of these species. Examples of the obtained plots
of (ln([esters] /[esters] ) s k (esters)t) versus (ln([ref] /[ref] ) s
(ref)t) are displayed in Figures 3-5. Linear squares analysis
0
t
d
0
t
k
d
of the data gives the results shown in Table 2. The measured
rate coefficient ratios were placed on an absolute scale using
-
17
3
-1 -1
k(propene + O
3
) ) 1.0 × 10 cm molecule
s
and k(ethene
-
18
3
-1 -1 22
+
O
3
) ) 1.6 × 10
cm molecule s , leading to the
3
-1 -1
Figure 1. Example of plots of ozone consumption versus reaction time
in the absence and presence of esters (example of methyl methacrylate).
following rate coefficient values (in cm molecule s ), taken
as an average of all values
CH dCH(CH )C(O)OCH + O f products
2
3
3
3
-18
-18
-18
k1 ) (6.5 ( 0.7) × 10
CH dCHC(O)OCH + O f products
2
3
3
k2 ) (0.94 ( 0.06) × 10
CH dCHC(O)OCH CH + O f products
2
2
3
3
k3 ) (1.3 ( 0.1) × 10
As shown above, the rate coefficient values obtained using
absolute and relative methods are in excellent agreement. In
addition to these measurements, experiments were performed
to determine the decay of ethyl acrylate (EA) against methyl
acrylate (MA) and methyl methacrylate (MMA). These experi-
ments resulted in the rate coefficients ratios of k(EA + O
k(MA + O ) ) 1.25 ( 0.14 and k(MMA + O )/k(EA + O ) )
.62 ( 0.13, which are in good agreement with the values of
.38 ( 0.19 and 5.0 ( 0.9 obtained from the ratios of the
measured rate coefficients for these esters with ethene and
propene as reference compounds.
3
)/
3
3
3
Figure 2. Plots of absolute rate data for the ozone reaction with methyl
methacrylate, methyl acrylate, and ethyl acrylate. Experiments con-
ducted without cyclohexane are represented with filled symbols, while
those conducted in the presence of an excess of cyclohexane are shown
with open symbols.
5
1
The rate coefficients recommended in this work are the
averaged values obtained using both absolute and relative
methods with error limits that encompass the extremes of σ error
of the individual determinations
react with esters (in excess) and will not have any effect on the
3
-1 -1
decay rate of ozone. The rate coefficients (in cm molecule s )
obtained for reactions of O
3
with the three esters at 294 ( 2 K
and 760 Torr of air are as follows
CH dCH(CH )C(O)OCH + O f products
2
3
3
3
k(O + methyl methacrylate) )
3
-18
-18
-18
k1 ) (6.8 ( 0.4) × 10
-18
cm molecule s-1
3
-1
(6.7 ( 0.9) × 10
k(O + methyl acrylate) )
CH dCHC(O)OCH + O f products
3
2
3
3
0.95 ( 0.07) × 10 cm molecule s-1
-18
3
-1
(
k2 ) (0.95 ( 0.06) × 10
k(O + ethyl acrylate) )
3
CH dCHC(O)OCH CH + O f products
2
2
3
3
1.3 ( 0.1) × 10 cm molecule s-1
-18
3
-1
(
k3 ) (1.28 ( 0.06) × 10
RelatiWe Rate Study. The kinetics of the reactions of ozone
with the three acrylates were measured relative to those of the
The only other measurements reported in the literature so far
on the title reactions are from Grosjean et al.14 and Grosjean

Kinetics and Products of Gas-Phase Reactions of Ozone
J. Phys. Chem. A, Vol. 114, No. 32, 2010 8379
TABLE 1: Reaction of O with Methyl Methacrylate, Methyl Acrylate, and Ethyl Acrylate: Summary of the Experimental
3
Conditions and Results from the Absolute Study at 294 ( 2 K and 760 Torr of Air
[
ester]
[O
3
]
0
(k′ - k′ ( 1σ)
0
(1013 molecules cm-3)
(1012 molecules cm-3)
(10
-4 -1
s )
ester
methyl methacrylate,
1.36 ( 0.04
1.55
3.05
2.50
4.22
13.2
2.69
0.94 ( 0.01
2.04 ( 0.01
3.63 ( 0.02
4.68 ( 0.02
5.30 ( 0.02
8.32 ( 0.02
a
CH
2
dCH(CH
3
)C(O)OCH
3
3.0 ( 0.2
d
5
7
8
.4 ( 0.2
.1 ( 0.3
.0 ( 0.4
1
2.2 ( 0.4
-
18
k
1
) (6.8 ( 0.4) × 10
methyl acrylate,
CH dCHC(O)OCH
2.8 ( 0.1
5.3 ( 0.2
2.47
3.12
2.85
3.03
3.09
2.93
0.29 ( 0.01
0.51 ( 0.01
0.57 ( 0.01
1.04 ( 0.01
1.25 ( 0.01
1.40 ( 0.01
b
d
2
3
5
.8 ( 0.2
10.4 ( 0.4
13.1 ( 0.5
14.5 ( 0.6
-
18
k
k
2
) (0.95 ( 0.06) × 10
ethyl acrylate,
CH dCHC(O)OCH
2.80 ( 0.07
1.65
1.97
2.75
3.86
2.58
2.31
0.27 ( 0.01
0.59 ( 0.01
0.96 ( 0.01
1.19 ( 0.01
1.59 ( 0.01
1.79 ( 0.01
CH c
5.4 ( 0.2
d
2
2
3
7
.7 ( 0.3
10.2 ( 0.2
13.2 ( 0.3
14.5 ( 0.3
-
18
3
) (1.28 ( 0.06) × 10
a
-4 -1 b
For methyl methacrylate experiments, ozone decay rate was k′
0
) (0.23-0.25) × 10
s . For methyl acrylate experiments, ozone decay
-
4
-1
c
-4 -1
d
rate was k′
0
) (0.17-0.19) × 10
s
.
For ethyl acrylate experiments, ozone decay rate was k′
0
) (0.18-0.19) × 10
s . Experiments
1
5
-3
conducted in the presence of cyclohexane (∼7.5 × 10 molecules cm ).
Figure 3. Plots of relative kinetic data from the ozonolysis of MMA
using ethene and propene as organic references in the presence of
cyclohexane (∼(2.7-4.7) × 10 molecules cm ).
Figure 4. Plots of relative kinetic data from the ozonolysis of EA
1
5
-3
using MMA and MA as organic references in the presence of
1
5
-3
cyclohexane (∼4.7 × 10 molecules cm ).
and Grosjean16 using the absolute method and Munshi et al.
13
using the second-order kinetic condition. Grosjean et al.14
measured a rate coefficient of ozone reaction with methyl
consumption of ozone than acrylates, which may indicate the
occurrence of secondary reactions in their system, explaining
the larger rate coefficient values they obtained.
-18
3
-1 -1
methacrylate of k ) (7.5 ( 0.9) × 10
which differs only by 15% from the one measured in the present
work. For the rate coefficient of the reaction of O with methyl
acrylate, Grosjean and Grosjean obtained k ) (1.05 ( 0.15)
cm molecule s ,
3
As expected, the rate coefficient value for reaction of O with
3
methyl methacrylate is higher than those with methyl acrylate
and ethyl acrylate (by a factor of ∼7 and ∼5, respectively).
This might be due to the presence of the electron-donating
methyl group on the carbon-carbon double bond, which
increases the electronic density and therefore the reactivity
toward ozone. Table 3 summarizes the rate coefficients values
1
6
-18
3
-1 -1
×
10 cm molecule s . This value is less than 10% higher
than that obtained in this work. Munshi et al. reported much
higher rate coefficient values for reactions of ozone with methyl
1
3
-18
acrylate (k ) (2.9 ( 0.2) × 10 ) and ethyl acrylate (k ) (5.7
-18
(
0.6) × 10 ), respectively, a factor three and five higher
available for reactions of O
tion of the data presented in this table shows that k(O
CH dCHC(O)OCH ) < k(O + CH dCHC(O)OCH CH ) <
3
with unsaturated esters. Examina-
than our determinations. It has to be mentioned that in their
work Munshi et al., using a flow tube, observed a larger
3
+
2
3
3
2
2
3

8
380 J. Phys. Chem. A, Vol. 114, No. 32, 2010
Bernard et al.
TABLE 3: Rate Constants of Unsaturated Esters by
Reaction with Ozone in the Gas Phase
k (cm molecule s-1
3
-1
)
ref
unsaturated esters
-
-
18
methyl methacrylate,
(7.5 ( 0.9) × 10
(6.7 ( 0.9) × 10
14
this work
18
CH
2
dC(CH
3
)C(O)OCH
3
-
18
methyl acrylate,
CH dCHC(O)OCH
(1.05 ( 0.15) × 10
16
13
this work
-
18
2
3
(2.9 ( 0.2) × 10
-
18
(
0.95 ( 0.07) × 10
-
18
18
ethyl acrylate,
CH dCHC(O)OCH
(5.7 ( 0.6) × 10
13
this work
-
2
2
CH
3
(1.3 ( 0.1) × 10
-
18
methyl crotonate,
CH CHdCHC(O)OCH
(4.4 ( 0.3) × 10
14
17
17
3
3
-
-
18
ethyl methacrylate,
2 3
CH dC(CH )C(O)OCH CH
(7.68 ( 0.88) × 10
2
3
18
n-butyl acrylate,
2 3 3
CH dCHC(O)O(CH ) CH
(2.40 ( 0.29) × 10
2
Figure 5. Plots of relative kinetic data from the ozonolysis of EA and
MA using ethene as organic reference in the presence of cyclohexane
pyruvate (CH
+ methyl methacrylate, methyl glyoxylate (HC(O)-
C(O)OCH ) was observed as a product from O + methyl
acrylate, and ethyl glyoxylate (HC(O)C(O)OCH CH ) was
observed as a product from O + ethyl acrylate.
Figure 6 shows IR spectra of the gas mixture in the chamber
at different stages of the reaction. Figure 6 A shows the IR
spectrum of methyl methacrylate. Figure 6B is a mixture of
3 3
C(O)C(O)OCH ) was observed as a product from
1
5
-3
(
∼5.5 × 10 molecules cm ).
O
3
3
3
k(O
the length of the R′ group in RC(O)OR′ may have an effect on
the reactivity even though O is expected to add mainly to the
3
+ CH
2
dCHC(O)O(CH
2
)
3
CH
3
), which could indicate that
2
3
3
3
>
CdC<. According to the limited existing data, k(O
3
+
+
+
CH
2
dCHC(O)OCH
3
) is a factor 1.3 lower than k(O
CH ) and a factor 2.4 lower than k(O
CH
ity needs to be confirmed. On the other hand, the presence of
the methyl group on >CdC< enhances the reactivity (k(O
CH dCHC(O)OCH ) < k(O + CH CHdCHC(O)OCH ) < k(O
CH dC(CH )C(O)OCH )).
.2. Gas-Phase Reaction Products. A set of three experi-
3
CH
CH
2
dCHC(O)OCH
2
3
3
3
methyl methacrylate and O at the start of the experiment. Figure
2
dCHC(O)O(CH
)
2 3
3
). However, this trend in the reactiv-
6C shows the IR spectrum after 4 h of the reaction. Figure 6D
and Figure 6E display, respectively, IR spectra of methyl
pyruvate and formaldehyde (main reaction products). Figure 6F
shows the residual spectrum obtained after subtraction of the
identified products. Figure 7 shows a plot of the observed
formaldehyde and methyl pyruvate formation versus methyl
methacrylate loss (in the presence of cyclohexane). As seen from
Figure 7, formation of formaldehyde and methyl pyruvate scales
linearly with the loss of methyl methacrylate (up to 63%
consumption of methyl methacrylate). This was also the case
for product formation for reactions of ozone with the other
esters. The linearity of the formation of HCHO and the other
3
+
3
2
3
3
3
3
+
2
3
3
3
ments has been conducted for each unsaturated ester in the
presence of an excess amount of cyclohexane and one experi-
ment in its absence. Reaction mixtures consisted of (1.25-4.99)
1
3
-3
13
×
10 molecules cm of the esters and (1.0-2.5) × 10
-3
molecules cm of ozone. The concentration of cyclohexane,
1
5
-3
when present, was typically 7.5 × 10 molecules cm . The
gas-phase reaction products were monitored and tentatively
assigned by FT-IR spectroscopy using reference spectra when
available. Cyclohexanone, a product of the OH-initiated oxida-
tion of cyclohexane, was not observed in our experimental
conditions, which indicates that its formation was below the
detection limit of our instrumentation or/and the OH formation
yield was negligible. While formic acid (HCOOH) and carbon
monoxide (CO) were observed as reaction products with minor
yields (<5%), formaldehyde (HCHO) was one of the main
reaction products in the three reactions investigated. Methyl
carbonyl compounds (methyl pyruvate (CH
methyl glyoxylate (HC(O)C(O)OCH ), and ethyl glyoxylate
(HC(O)C(O)OCH CH )) versus consumption of the acrylates
3 3
C(O)C(O)OCH ),
3
2
3
suggests the absence of a significant loss or formation of the
products observed via secondary reactions in the chamber.
Tables 4, 5, and 6 summarize the measured formation yields of
the oxidation products detected in the ozonolysis of methyl
methacrylate, methyl acrylate, and ethyl acrylate in the presence
and absence of OH scavenger. Formic acid and carbon monoxide
are formed in the three systems with low yields (less than 5%).
TABLE 2: Reaction of O
3
with Methyl Methacrylate, Methyl Acrylate, and Ethyl Acrylate: Summary of the Experimental
Conditions and Results from the Relative Rate Study at 294 ( 2 K and 760 Torr of Air
(
k ( 1σ),
-1
esters
references
propene
N° of runs
(k/kref ( 1σ)
(cm molecule s-1
)
3
-
-
18
methyl methacrylate,
2
1
0.65 ( 0.07
5.62 ( 0.13
(6.5 ( 0.7) × 10
(7.2 ( 0.2) × 10
18
CH
2
dCH(CH
3
)C(O)OCH
3
ethyl acrylate
-
18
methyl acrylate,
CH dCHC(O)OCH
ethene
3
0.59 ( 0.04
(0.94 ( 0.06) × 10
2
3
2
-18
ethyl acrylate,
CH dCHC(O)OCH
ethene
methyl acrylate
5
2
0.83 ( 0.06
1.25 ( 0.14
(1.3 ( 0.1) × 10
-18
2
CH
3
(1.2 ( 0.1) × 10

Kinetics and Products of Gas-Phase Reactions of Ozone
J. Phys. Chem. A, Vol. 114, No. 32, 2010 8381
TABLE 4: Reaction of Methyl Methacrylate with O
3
:
Gas-Phase Product Yields in the Absence and Presence of
Scavenger (cyclohexane)
scavenger
cyclohexane
(7.5 × 10 )
15
(
molecules cm-3)
-
number of runs
methyl pyruvate
formaldehyde
carbon monoxide
formic acid
3
1
0.55 ( 0.02
0.69 ( 0.01
0.05 ( 0.02
0.02 ( 0.02
0.59 ( 0.02
0.552 ( 0.003
0.69 ( 0.02
0.02 ( 0.01
0.030 ( 0.009
0.58 ( 0.01
carbon balance
TABLE 5: Reaction of Methyl Acrylate with O
3
: Gas-Phase
Product Yields in the Absence and Presence of Scavenger
(
cyclohexane)
scavenger
cyclohexane
(7.5 × 10 )
15
(
molecules cm-3)
-
number of runs
methyl glyoxylate
formaldehyde
carbon monoxide
formic acid
3
1
0.30 ( 0.06
0.43 ( 0.01
<0.01
0.32 ( 0.01
0.565 ( 0.005
<0.01
<0.01
<0.01
-
-
1 a
1 a
1
1
113 cm
214 cm
1.00 ( 0.13
1.00 ( 0.02
0.33 ( 0.05
1.04 ( 0.01
0.99 ( 0.01
0.38 ( 0.01
carbon balance
a
Yield of unidentified FT-IR bands have been normalized to
unity.
3
TABLE 6: Reaction of Ethyl Acrylate with O : Gas-Phase
Product Yields in the Absence and Presence of Scavenger
(
cyclohexane and water)
Figure 6. Methyl methacrylate + O
Reference IR spectra of methyl methacrylate (A). IR spectra of a
mixture of methacrylate and O at the start of the experiment (B). IR
spectra of a mixture of methacrylate and O after 4 h of reaction (C).
3
(in the presence of cyclohexane).
cyclohexane
1
5
scavenger
cyclohexane
(7.5 × 10 ), water
3
molecules cm-3) (7.5 × 10 )
15
-
(2.0 × 10 )
14
(
3
Reference IR spectra of methyl pyruvate (D) and formaldehyde (E).
IR residual spectra after subtraction of reactants (methyl methacrylate
and ozone) and identified products (major products, methyl pyruvate
and formaldehyde; minor products, carbon monoxide and formic acid)
number of runs
3
1
1
ethyl glyoxylate 0.53 ( 0.08 0.56 ( 0.01
formaldehyde 0.45 ( 0.07 0.23 ( 0.01
carbon monoxide 0.03 ( 0.01 0.02 ( 0.01
0.71 ( 0.02
0.60 ( 0.04
0.23 ( 0.01
(
F).
formic acid
0.015
0.022 ( 0.003
-
-
1 a
1
1
113 cm
211 cm
1.00 ( 0.16 0.94 ( 0.01
1.00 ( 0.12 0.75 ( 0.02
0.52 ( 0.08 0.492 ( 0.009
0.19 ( 0.01
0.73 ( 0.03
1 a
carbon balance
a
Yield of unidentified FT-IR bands have been normalized to
unity.
in the presence and absence of cyclohexane. For the reaction
of ozone with methyl acrylate, formaldehyde and methyl
glyoxylate were the main reaction products with formation yields
of ∼43% and ∼30%, respectively. The run conducted in the
absence of cyclohexane led to the same formation yield of
methyl glyoxylate but a slightly higher yield for formaldehyde.
Methyl glyoxylate was quantified using a reference spectrum
obtained from the Euphore facilities. Ethyl glyoxylate (∼53%)
and formaldehyde (∼45%) were observed to be the main
products of the ozonolysis of ethyl acrylate.
To the best of our knowledge, among the studied esters, only
the gas-phase products from the ozonolysis of methyl acrylate
have been reported so far.15 Grosjean and Grosjean reported
the following formation yields: 0.54 ( 0.04 for formaldehyde
and 0.43 ( 0.01 for methyl glyoxylate. Nevertheless, methyl
glyoxylate was quantified in their study using ethyl glyoxylate
as a surrogate based on DnPH derivatization. Moreover, in the
studies of Grosjean and Grosjean, experiments were conducted
in a system with a relative humidity equivalent to 50%, which
could impact the fate of Criegee biradical intermediate and
potentially the formation yields of oxygenated products such
as formaldehyde, formic acid, and carbonyl compounds.
15
Figure 7. Formation of formaldehyde and methyl pyruvate versus loss
of methyl methacrylate in the presence of ozone and an excess of
1
5
-3
cyclohexane (∼7.5 × 10 molecules cm ).
15
The main products of the reaction of ozone with methyl
methacrylate are formaldehyde (∼55%) and methyl pyruvate
(
∼69%). Calibrated reference spectra of methyl pyruvate were
obtained in our experimental conditions from which calibration
curves were derived. The observed yields obtained were similar

8
382 J. Phys. Chem. A, Vol. 114, No. 32, 2010
Bernard et al.
work using an IR spectrum of pyruvic acid taken in our
laboratory, which was compared to the residual spectra after
subtraction of the known compounds during ozonolysis of
MMA. This enabled us to conclude that the unidentified FT-IR
-1
band near 1775 cm observed in this work does not correspond
to pyruvic acid (Figure 6F).
Regarding the energy-rich Criegee intermediate [CH
which is formed in the three reactions studied here, it can be
either stabilized or decompose to form HCOOH, CO, CO , H O,
, and OH radical.
It should be noted that the experiments conducted in the
2
OO]*,
2
2
H
2
presence and absence of cyclohexane led essentially to the same
product yield formation. On the other hand, the ratio ∆[esters]/
∆
3
[O ] was close to unity. These two observations indicate that
OH radicals were not significantly formed through ozonolysis
of methyl methacrylate. Grosjean and Grosjean15 reported that
ozonolysis of methyl acrylate leads to less than 1% of the OH
radicals formation yield.
Figure 8. Comparison of residual spectra after subtraction of all known
products: methyl methacrylate + O (A), methyl acrylate + O (B),
ethyl acrylate + O (C), and ethyl acrylate + O in the presence of an
excess amount of water (D).
In order to assess the role of the Criegee intermediates in
formation of the unidentified products, one run was conducted
on the ozonolysis of ethyl acrylate in the presence of water
vapor, which is known to scavenge the Criegee intermediate
biradicals. The run was performed in the presence of cyclo-
3
3
3
3
As shown in Figure 8, after subtraction of the identified
reaction products, the residual spectrum obtained from ozo-
nolysis of methyl acrylate and ethyl acrylate showed unidentified
hexane. In these conditions, the biradical [CH
2
OO]* reacts with
H
2
O to form hydroxymethyl hydroperoxide (HMHP), which
23
decomposes, leading to formic acid
-1
IR features at 1113 cm for both esters and 1214 and 1211
-1
cm for methyl acrylate and ethyl acrylate, respectively.
Assessments of IR bands have been carried out by subtraction
of all known oxidation products. The residual spectrum was
then used to integrate unknown FT-IR bands from the global
spectra. Absorbance time profiles of unknown IR bands near
[
CH OO]* + H O f HOCH OOH f HCOOH + H O
2
2
2
2
Figure 6 compares the residual IR spectra of MMA + O
reaction (Figure 6A), the residual IR spectra of MA + O
reaction (Figure 6B), the residual IR spectra of EA + O
Figure 6C), and the residual IR spectra EA + O in presence
of water (Figure 6D). From ethyl acrylate experiments, the IR
3
113, 1211, and 1214 cm- were derived, and their yields were
1
3
1
3
reaction
normalized to unity for the experiments conducted in the
presence of cyclohexane and compared to those without
cyclohexane and in the presence of water vapor (Tables 5 and
(
3
-
1
bands near 1113 and 1211 cm observed in the absence of
O disappear in its presence (Figure 6D). In the presence of
6
).
The reaction of ozone with unsaturated compounds proceeds
H
2
-
1
an excess of H
2
O, the 1113 cm IR band decreased 80%
by electrophilic addition on the carbon-carbon double bond.
This leads to the formation of a primary ozonide, which
decomposes in energy-rich Criegee biradical and a correspond-
ing carbonyl as follows for the studied unsaturated esters
compared to dry conditions. The same trend is observed for
-
1
the 1211 cm IR band, where it disappeared completely under
humid conditions. In the same experiment, the yields of ethyl
glyoxylate increased from 53% to 71% and from 45% to 60%
for formaldehyde compared to dry experiments. These observa-
tions indicate that the Criegee intermediates may have contrib-
uted to formation of the unidentified products corresponding to
the observed bands. However, we are not able to draw definite
conclusions from this single experiment. Further investigations
are required in order to understand the fate of the Criegee
intermediates formed through the ozonolyzis of such unsaturated
oxygenated VOCs (e.g., in the presence of scavengers such as
HCOOH and HCHO).
H CdCR C(O)OR + O
2
1
2
3
f [CH OO]* + R C(O)C(O)OR (A)
2
1
2
f HCHO + [R COOC(O)OR ]*
(B)
1
2
with R
1
) R
2
) -CH
) -CH
2 3
) -CH CH to ethyl acrylate. As shown by the experimental
3
corresponding to methyl methacrylate,
R
R
1
) H and R
2
3
to methyl acrylate, and R ) H and
1
2
4
. Atmospheric Implications
data, formaldehyde is a primary product formed in the ozo-
nolysis of the three studied esters at similar formation yield for
methyl and ethyl acrylate, ∼43% and ∼45%, respectively, and
at higher yield for methyl methacrylate (∼9%). This indicates
that the two decomposition pathways of the primary ozonide
formed in the reactions of ozone with methyl acrylate and ethyl
acrylate are of the same importance while channel B will
dominate decomposition of the biradical from ozonolysis of
methyl methacrylate. To date, the fate of the Criegee intermedi-
In the troposphere, unsaturated esters react with the major
atmospheric oxidants O , OH, and NO radicals and Cl atoms.
3 3
Using a typical concentration of ozone of 50 ppb and the
reaction rate coefficients measured in this work, the tropospheric
lifetimes of the three unsaturated esters with respect to reaction
with ozone are estimated to be ∼10 days for methyl acrylate,
∼7 days for ethyl acrylate, and around 33 h for methyl
methacrylate. These lifetimes can be compared with those
obtained against reactions with the other major atmospheric
oxidants. The available kinetic data shows that the reaction with
OH radicals is the dominant degradation process of unsaturated
ate [CH
et al.21 speculated that [CH
carboxylic acid (CH C(O)C(O)OH). This was checked in our
3
COOC(O)OCH
3
]* has not been elucidated. Grosjean
3
OC(O)CHOO]* may lead to the
3

Kinetics and Products of Gas-Phase Reactions of Ozone
J. Phys. Chem. A, Vol. 114, No. 32, 2010 8383
esters in the atmosphere with lifetimes of a few hours.24
However, the reaction with ozone can constitute an important
sink for these compounds in some areas (such as the highly
polluted urban areas where these esters might be present) where
the average concentration of ozone could be much higher than
(6) Blanco, M. B.; Teruel, M. A. J. Phys. Org. Chem. 2008, 21, 397.
(
7) Blanco, M. B.; Bejan, I.; Barnes, I.; Wiesen, P.; Teruel, M. A. J.
Phys. Chem. A 2009, 113, 5958.
8) Blanco, M. B.; Bejan, I.; Barnes, I.; Wiesen, P.; Teruel, M. A.
(
Atmos. EnViron. 2009, 43, 5996.
(9) Blanco, M. B.; Bejan, I.; Barnes, I.; Wiesen, P.; Teruel, M. A.
EnViron. Sci. Pollut. Res. 2009, 16, 641.
5
0 ppb used to estimate the lifetimes here). The short atmo-
(
10) Teruel, M. A.; Achad, M.; Blanco, M. B. Chem. Phys. Lett. 2009,
spheric lifetimes indicate that these compounds will be oxidized
near their emission sources. The reaction of ozone with
unsaturated esters leads to formaldehyde and other highly
oxygen-containing compounds such as methyl pyruvate, methyl
glyoxylate, and ethyl glyoxylate, which are expected to be
4
79, 25.
(11) Blanco, M. B.; Barnes, I.; Teruel, M. A. Chem. Phys. Lett. 2010,
88, 135.
12) Mart ´ı n Porrero, M. P.; Gallego-Iniesta Garc ´ı a, M. P.; Espinosa Ruiz,
J. L.; Tapia Valle, A.; Caba n˜ as Gal a´ n, B.; Salgado Mu n˜ oz, M. S. EnViron.
Sci. Pollut. Res. 2010, 17, 539.
4
(
removed from the atmosphere through partitioning in the
(13) Munshi, H. B.; Rama Rao, K. V. S.; Mahadeva Iyer, R. Atmos.
EnViron. 1989, 23, 1971.
aqueous phase.25
(
14) Grosjean, D.; Grosjean, E.; II, E. L. W. Int. J. Chem. Kinet. 1993,
2
5, 783.
Acknowledgment. The research leading to these results
received funding from the European Community’s Seventh
Framework Programme under grant agreement no. 228335
(
15) Grosjean, E.; Grosjean, D. J. Atmos. Chem. 1997, 27, 271.
(16) Grosjean, E.; Grosjean, D. Int. J. Chem. Kinet. 1998, 30, 21.
(
(
17) Gai, Y.; Ge, M.; Wang, W. Chem. Phys. Lett. 2009, 473, 57.
18) Canosa-Mas, C. E.; Carr, S.; King, M. D.; Shallcross, D. E.;
(
EUROCHAM 2 Project) and the French German programme
Thompson, K. C.; Wayne, R. P. Phys. Chem. Chem. Phys. 1999, 1, 4195.
19) Canosa-Mas, C. E.; Flugge, M. L.; King, M. D.; Wayne, R. P. Phys.
Chem. Chem. Phys. 2005, 7, 643.
INSU-DFG.
(
(
(
20) Wang, K.; Ge, M.; Wang, W. Atmos. EnViron. 2010, 44, 1847.
References and Notes
21) Grosjean, E.; Grosjean, D. J. Atmos. Chem. 1999, 32, 205.
(
1) Le Clav e´ , S.; Le Bras, G.; Mellouki, A. J. Phys. Chem. A 1997,
(22) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.; Hampson,
R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J. Atmos. Chem.
Phys. 2006, 6, 3625.
(23) Neeb, P.; Sauer, F.; Horie, O.; Moortgat, G. K. Atmos. EnViron.
1997, 31, 1417.
1
1
2
01, 9137.
(
2) Cavalli, F.; Barnes, I.; Becker, K. H. J. Phys. Chem. A 2000, 104,
1310.
(
3) Wallington, T. J.; Hurley, M. D.; Haryanto, A. Chem. Phys. Lett.
006, 432, 57.
(
24) Teruel, M. A.; Lane, S. I.; Mellouki, A.; Solignac, G.; Le Bras, G.
Atmos. EnViron. 2006, 40, 3764.
25) Gelencs e´ r, A.; Varga, Z. Atmos. Chem. Phys. 2005, 5, 2823.
(
4) Saunders, S. M.; Baulch, D. L.; Cooke, K. M.; Pilling, M. J.;
Smurthwaite, P. I. Int. J. Chem. Kinet. 1994, 26, 113.
5) Blanco, M. B.; Taccone, R. A.; lane, S. I.; Teruel, M. A. Chem.
Phys. Lett. 2006, 429, 389.
(
(
JP104451V