Experimental and modeling study of the oxidation of toluene

Article abstract of DOI:10.1002/kin.20047

This paper describes an experimental and modeling study of the oxidation of toluene. The low-temperature oxidation was studied in a continuous flow stirred tank reactor with carbon-containing products analyzed by gas chromatography under the following experimental conditions: temperature from 873 to 923 K. 1 bar, fuel equivalence ratios from 0.45 to 0.91, concentrations of toluene from 1.4 to 1.7%, and residence times ranging from 2 to 13s corresponding to toluene conversion from 5 to 85%. The ignition delays of toluene-oxygen-argon mixtures with fuel equivalence ratios from 0.5 to 3 were measured behind reflected shock waves for temperatures from 1305 to 1795 K and at a pressure of 8.7 ± 0.7 bar. A detailed kinetic mechanism has been proposed to reproduce our experimental results, as well as some literature data obtained in other shock tubes and in a plug flow reactor. The main reaction paths have been determined by sensitivity and flux analyses. 2004 Wiley Periodicals, Inc.

Full text of DOI:10.1002/kin.20047

Experimental and Modeling
Study of the Oxidation
of Toluene
R. BOUNACEUR, I. DA COSTA, R. FOURNET, F. BILLAUD, F. BATTIN-LECLERC
D e´ partement de Chimie Physique des R e´ actions, UMR 7630 CNRS, INPL-ENSIC, 1, rue Grandville, BP 451,
5
4001 Nancy Cedex, France
Received 25 May 2004; accepted 12 August 2004
DOI 10.1002/kin.20047
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: This paper describes an experimental and modeling study of the oxidation of
toluene. The low-temperature oxidation was studied in a continuous flow stirred tank reactor
with carbon-containing products analyzed by gas chromatography under the following experi-
mental conditions: temperature from 873 to 923 K, 1 bar, fuel equivalence ratios from 0.45 to
0.91, concentrations of toluene from 1.4 to 1.7%, and residence times ranging from 2 to 13 s cor-
responding to toluene conversion from 5 to 85%. The ignition delays of toluene–oxygen–argon
mixtures with fuel equivalence ratios from 0.5 to 3 were measured behind reflected shock waves
for temperatures from 1305 to 1795 K and at a pressure of 8.7 ± 0.7 bar. A detailed kinetic mech-
anism has been proposed to reproduce our experimental results, as well as some literature data
obtained in other shock tubes and in a plug flow reactor. The main reaction paths have been
determined by sensitivity and flux analyses. ꢀC 2004 Wiley Periodicals, Inc. Int J Chem Kinet 37:
2
5–49, 2005
INTRODUCTION
Toluene is the most important aromatic compound
in gasolines (up to 10%), with a research octane number
of 120 [1]. Its slow oxidation has been studied experi-
mentally in a flow reactor by Brezinsky et al. [12,13], in
ajet-stirredreactorbyDagautetal. [14], andinasingle-
pulse shock tube by Sivaramakrishnan et al. [15]. The
autoignition of toluene has been studied in a shock tube
by Burcat et al. [16] and Pengloan et al. [14,17] and in
a rapid compression machine by Roubaud et al. [18].
Diffusion and laminar flames containing toluene have
been studied by Hamins and Seshadri [19] and by Davis
et al. [20], respectively. Several kinetic mechanisms
Aromatic compounds are present in significant
amounts in gasolines (∼35%) and diesel fuels (∼30%)
[1]. Nevertheless, detailed chemical kinetic models for
the combustion and the oxidation of aromatic com-
pounds are still scarce and are mainly restricted to
benzene [2–10]. We have recently published a paper
presenting an experimental and modeling study of the
oxidation of benzene [11] that we have used as a basis
for this work about the oxidation of toluene.
[2,13,14,21,22] have also been proposed to model these
experimental data.
It is worth noting that all the experiments in reactors
[12–14]havebeenperformedusingalowconcentration
of toluene (≤0.15%) and, consequently, with no notice-
able conversion for a temperature lower than 1000 K.
The two purposes of this paper then are:
Correspondence to: F. Battin-Leclerc; e-mail: frederique.battin-
leclerc@ensic.inpl-nancy.fr.
Contract grant sponsor: Institut Fran c¸ ais du P e´ trole.
Contract grant number: EVG1-CT-2002-00072 SAFEKINEX.
Contract grant sponsor: European Commission.
ꢀ
c 2004 Wiley Periodicals, Inc.

26
BOUNACEUR ET AL.
•
•
To present new experimental data for the oxida-
tion of toluene obtained both in a jet-stirred re-
actor at atmospheric pressure, at a concentration
of toluene from 1.4 to 1.7%, at temperatures from
The main carbon-containing products analyzed by
gas chromatography were carbon monoxide and diox-
ide, methane, C2 species (acetylene and ethylene
were not separated), allene, propene, 1,3-butadiene,
benzene, phenol, ethylbenzene, styrene, benzaldehyde,
benzyl alcohol, and bibenzyl. Chromatographs with
thermal conductivity detector and hydrogen as carrier
gas were used to analyze O2, CO, and CO2 with a Car-
bosphere packed column. Gas-phase light hydrocar-
bons (CH4, C2 species, C3H4, C3H6, and C4H6) were
analyzed on a 30% Squalane on Chromosorb P col-
umn with detection by flame ionization detection and
nitrogen as carrier gas. The analyses of products con-
taining more than five atoms of carbon were performed
by gas chromatography with flame ionization detection
by using a HP PONA capillary column and nitrogen as
carrier gas, which gave a good separation for the above-
mentioned products with a limit of detection around
8
73 to 923 K, and in a shock tube for tempera-
tures between 1305 and 1795 K at a pressure of
8
.7 ± 0.7 bar.
To propose a new mechanism able to model these
experimental results, and also to reproduce litera-
ture data obtained in a flow reactor at 1100 K [12]
and in different shock tubes [14–16]. Some of our
results obtained in a shock tube have already been
modeled by Pitz et al. [22], but the predicted ig-
nition delays were too long (at least a factor 3)
compared to the experimental delays, particularly
at the highest temperatures.
EXPERIMENTAL RESULTS OF THE
OXIDATION OF TOLUENE
−
5
10 in mole fraction [11]. For each sample, the carbon
balance was checked and an agreement was obtained
to within 5%. The formation of cyclopentadiene and
cresols was not detected.
A discussion of the experimental results presented here
is given later in the text with the comparisons with
simulations.
As this experimental study was performed using the
same two devices as in our study of the oxidation of
benzene, a jet-stirred reactor and a shock tube [11],
we shall not recall the experimental procedure here.
Toluene (99.8% pure) was provided by Fluka. Oxygen
Tables I and II present all the experimental results
obtained in a jet-stirred reactor. Figure 1 displays the
evolution of the conversion of toluene, of the formation
of the two major primary products, benzene and ben-
zaldehyde, and the production of carbon monoxide, vs.
residence time for an equivalence ratio of 0.9 and for a
temperature equal to 873 (Fig. 1a), 893 (Fig. 1b), and
923 K (Fig. 1c). This figure shows that the reaction is
autoaccelerated with an induction delay time that de-
creases with temperature. Figures 2 and 3 display the
evolution of the conversions of toluene and oxygen and
of the formations of all the observed products vs. res-
idence time at 893 K and for an equivalence ratio of
(
99.5% pure), argon, and helium (both 99.995% pure)
were supplied by Alphagaz-L’Air Liquide.
Jet-Stirred Reactor
As liquid toluene used for bubbling the flow of carrier
gas (helium) to obtain gas-phase toluene needs to be
maintained below room temperature (here at 289 K)
to avoid the condensation of hydrocarbon before it
enters the reactor, the maximum mole fraction was
0.45.
1
.7%, which was much lower than what was used for
Shock Tube
benzene (4.5%) [11].
As in our previous work [11], the temperature and the
pressure of the test gas behind the incident and the re-
flected shock waves were derived from the value of the
incident shock velocity by using ideal one-dimensional
shock equations. The onset of ignition was detected by
OH radical emission at 306 nm through a quartz win-
dow with a photomultiplier fitted with a monochroma-
tor at the end of the reaction part. The last pressure
transducer was located at the same place along the axis
of the tube as the quartz window. The ignition delay
time was defined as the time interval between the pres-
sure rise measured by the last pressure transducer due
to the arrival of the reflected shock wave and the rise
of the optical signal by the photomultiplier up to 10%
of its maximum value.
This study was performed under the following
experimental conditions:
•
•
•
temperature: 873, 893, and 923 K;
atmospheric pressure;
mixtures (helium: toluene: oxygen, in molar
percent) were 1.7:17:81.3 and 1.4:28:70.6, re-
spectively, which correspond to two different
equivalence ratios (ꢀ = 0.45 and 0.9) and lead
to conversion of toluene from 5 to 85%.
The equivalence ratio ꢀ is defined as ꢀ = 9 ×
%toluene/%oxygen) by using as a reference of the total
(
reaction
C7H8 + 9O2 = 7CO2 + 4H2O

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
27
Table I Experimental Results Obtained in a Jet-Stirred Reactor at 893 K and 1 atm
.4% of Toluene, ꢀ = 0.45
Residence Time (s)
1
Compound
2.5
4
6.5
9
11
13
Mole Fraction
2.64 × 10−
4
4
5
6
5
5.17 × 10
3.09 × 10
2.75 × 10
6.17 × 10
9.23 × 10
0.272
−4
−4
−5
−6
−6
6.37 × 10
−4
6.99 × 10
−4
−4
−5
−6
−6
6.99 × 10
2.14 × 10
2.92 × 10
3.89 × 10
4.99 × 10
0.206
−4
−4
−5
−6
−6
6.09 × 10
−4
−4
−5
−6
−6
CH4
−
−4
−5
−6
−5
C2H2 + C2H4
2.36 × 10
3.02 × 10
3.24 × 10
5.3 × 10
1.08 × 10
0.269
2.42 × 10
3.23 × 10
3.96 × 10
7.46 × 10
0.224
1.24 × 10
−
C3H4
1.31 × 10
2.44 × 10
−
C3H6
4.83 × 10
3.08 × 10
−
1
,3-C4H6
1.02 × 10
4.17 × 10
O2
CO
CO2
C7H8
C6H6
C8H10
C8H8
C6H5CHO
C6H5OH
C6H5CH2OH
C14H14
0.279
0.014
0.218
−
2
−2
−2
−2
0.0386
0.0046
5.89 × 10
6.16 × 10
5.94 × 10
5.94 × 10
0.00167
0.00603
0.0148
0.0144
0.0182
0.0207
0.00422
0.00319
0.00206
0.00224
0.0026
−
4
5
5
4
4
5
5
−4
−4
−5
−5
−4
−4
−5
−5
−4
−4
−4
3.53 × 10
5.87 × 10
5.51 × 10
4.43 × 10
2.91 × 10
5.6 × 10
5.9 × 10
6.79 × 10
4.51 × 10
2.38 × 10
5.6 × 10
4.97 × 10
5.01 × 10
3.49 × 10
2.45 × 10
5.93 × 10
–
5.33 × 10
4.08 × 10
−
−5
−5
−4
−4
−5
−5
−5
−5
−4
−4
−5
3.68 × 10
4.12 × 10
–
–
–
–
–
–
−
−5
2.62 × 10
3.1 × 10
−
−4
−4
2.28 × 10
1.19 × 10
−
2.3 × 10
5.07 × 10
−
3.05 × 10
5.77 × 10
5.8 × 10
2.69 × 10
3.22 × 10
–
–
7.97 × 10−
–
1.7% of Toluene, ꢀ = 0.9
Residence Time (s)
Compound
2.5
4
6
8
10
12
Mole Fraction
CH4
2.38 × 10−
5
6
6
7
6
1.24 × 10
2.35 × 10
4.89 × 10
6.32 × 10
3.52 × 10
0.169
−4
−5
−6
−6
−6
4.72 × 10
1.97 × 10
6.72 × 10
3.32 × 10
7.67 × 10
0.149
−4
−4
−6
−6
−6
7.59 × 10
3.11 × 10
5.97 × 10
4.31 × 10
1.21 × 10
0.131
−4
−4
−6
−5
−5
8.95 × 10
3.61 × 10
5.76 × 10
4.39 × 10
1.17 × 10
0.115
−4
−4
−6
−5
−5
1.03 × 10
4.13 × 10
4.94 × 10
4.5 × 10
9.67 × 10
0.112
−3
−4
−6
−5
−6
−
C2H2 + C2H4
1.51 × 10
−
C3H6
3.62 × 10
−
C3H4
7.93 × 10
−
1
,3-C4H6
3.95 × 10
O2
CO
CO2
C7H8
C6H6
C8H10
C8H8
C6H5CHO
C6H5OH
C6H5CH2OH
C14H14
0.167
−
−
2
4
−2
−3
−2
−3
−2
−3
–
–
–
–
1.91 × 10
5.87 × 10
3.03 × 10
3.6 × 10
3.89 × 10
5.51 × 10
6.1 × 10
6.31 × 10
0.0154
0.0122
0.00901
0.00726
0.00587
0.00801
0.00563
0.000887
−
5
4.42 × 10
0.000211
0.000608
0.00846
−6
−5
−5
−5
−5
−5
−4
−5
−5
−5
2.93 × 10
2.95 × 10
7.46 × 10
9.07 × 10
1.05 × 10
9.33 × 10
−5
–
2.12 × 10
5.2 × 10
4.81 × 10
4.89 × 10
6.7 × 10
−
5
1.28 × 10
0.000122
0.000205
0.000218
0.000255
0.000211
−6
−5
−5
−4
−4
−4
−4
5.27 × 10
4.64 × 10
2.23 × 10
2.9 × 10
4.02 × 10
7.8 × 10
4.1 × 10
4.3 × 10
−
5
–
–
–
–
–
3.54 × 10−
5
9.97 × 10
9.72 × 10
−5
−5
2.9 × 10
−5
2.28 × 10
−5
This study was performed under the following ex-
perimental conditions (after the reflected shock):
95.025, respectively, which correspond to four
different equivalence ratios (ꢀ = 0.5, 1, 1.5
and 3) and to two different concentrations of
toluene (1.25% (a graph summarizing the re-
sults for 1.25% has already been presented
[22]) and a new more dilute condition, 0.5%)
and lead to ignition delay times from 3 to
1560 ꢀs.
•
•
•
temperature (T ) from 1305 to 1795 K.
pressure from 8 to 9.5 bar.
mixtures (argon:toluene:oxygen, in molar per-
cent) were 1.25:22.5:76.25, 1.25:11.25:87.5,
1
.25:7.5:91.25, 1.25:3.75:95.0, and 0.5:4.475:

28
BOUNACEUR ET AL.
Table II Experimental Results Obtained in a Jet-Stirred Reactor at 1 atm, for 1.7% of Toluene and ꢀ Equal to 0.9
Temperature 873 K
Residence Time (s)
Compound
3
4.4
6
7.5
8.7
11
13
Mole Fraction
1.14 × 10−
5
7
2.21 × 10
−5
−7
4.76 × 10
2.75 × 10
4.23 × 10
1.56 × 10
4.21 × 10
0.164
−5
−6
1.07 × 10
−4
3.6 × 10
1.06 × 10
3.53 × 10
2.13 × 10
6.99 × 10
–
−4
−4
−6
−5
−6
5.28 × 10
2.22 × 10
3.89 × 10
2.63 × 10
7.6 × 10
0.140
−4
−4
−6
−5
−6
7.15 × 10
2.66 × 10
4.06 × 10
3.75 × 10
8.22 × 10
–
−4
−4
−6
−5
−6
CH4
−
−5
C2H2 + C2H4
3.22 × 10
8.69 × 10
1.35 × 10
−
7
−6
C3H6
–
–
–
–
1.59 × 10
−7
−6
−6
−6
C3H4
2.7 × 10
4.56 × 10
−
6
1
,3-C4H6
–
0.166
–
5.05 × 10
O2
CO
CO2
C7H8
C6H6
C8H10
C8H8
C6H5CHO
C6H5OH
C6H5CH2OH
C14H14
0.167
–
–
0.139
−
4
−4
−3
−2
−3
–
–
4.05 × 10
5.21 × 10
4.95 × 10
7.73 × 10
1.61 × 10
−
4
–
–
–
3.47 × 10
0.0161
0.0156
0.0157
8.67 × 10
2.62 × 10
–
0.0137
1.69 × 10
3.51 × 10
2.2 × 10
1.05 × 10
–
0.0118
0.00993
0.00856
−5
−5
−5
−5
−5
−4
−5
−4
−5
−5
−4
−4
−5
−5
−4
−4
−4
2 × 10
4.18 × 10
4.26 × 10
5.74 × 10
3 × 10
4.48 × 10
5.9 × 10
5.19 × 10
1.81 × 10
3.13 × 10
–
6.65 × 10
6.39 × 10
5.46 × 10
2.32 × 10
4.54 × 10
–
−
5
−5
−5
−4
−4
–
–
–
–
−
5
−5
−5
−4
2.65 × 10
7.9 × 10
6.2 × 10
1.79 × 10
1.84 × 10
2.24 × 10
1.26 × 10
−
−
4
–
–
–
–
–
–
5
–
2.16 × 10−
5
1.16 × 10
8.19 × 10
−5
9.15 × 10
−5
−4
1.24 × 10
−4
1.21 × 10
−4
Temperature 923 K
Residence Time (s)
Compound
2.5
4
6
8
10
12
Mole Fraction
4.85 × 10−
4
4
5
5
5
9.55 × 10
5.22 × 10
1.34 × 10
5.25 × 10
2.1 × 10
–
−4
−4
−5
−5
−5
1.26 × 10
−3
1.58 × 10
−3
−4
−6
−5
−5
1.8 × 10
−3
1.76 × 10
7.12 × 10
3.76 × 10
4.19 × 10
1.41 × 10
0.111
−3
−4
−6
−5
−5
CH4
−
−4
−6
−5
−5
−4
C2H2 + C2H4
2.43 × 10
6.45 × 10
6.14 × 10
4.91 × 10
1.23 × 10
0.108
7.03 × 10
4.19 × 10
5.37 × 10
1.57 × 10
0.113
7.39 × 10
−
−6
C3H6
1.27 × 10
5.5 × 10
−
−5
C3H4
3.82 × 10
4.63 × 10
−
−5
1
,3-C4H6
1.92 × 10
1.21 × 10
O2
CO
CO2
C7H8
0.160
–
−
2
−2
−3
−2
−3
−2
−2
−2
−2
−2
−2
1.89 × 10
3.57 × 10
4.62 × 10
5.4 × 10
6.27 × 10
6.06 × 10
−
3
1.69 × 10
5.63 × 10
8.2 × 10
2.09 × 10
1.13 × 10
1.42 × 10
0.0095
0.00541
0.00401
0.00387
0.00326
0.00337
C6H6
C8H10
C8H8
C6H5CHO
C6H5OH
C6H5CH2OH
C14H14
0.000648
0.000903
0.000945
0.00113
0.00111
0.00108
−
4
−4
−4
−4
−4
−5
1.02 × 10
1.1 × 10
1.15 × 10
1.21 × 10
8.15 × 10
0.0238
1.01 × 10
7.96 × 10
−
5
−5
−5
−5
−5
−5
7.67 × 10
7.62 × 10
8.16 × 10
6.11 × 10
5.86 × 10
0.000331
4.31 × 10
0.0002687
0.00018
0.000265
0.000138
−
−
4
5
4
−4
−4
−4
−5
−4
−5
−4
6.18 × 10
4.54 × 10
1.14 × 10
5.67 × 10
5.57 × 10
1.06 × 10
4.91 × 10
6.14 × 10
–
5.36 × 10
7.05 × 10
4.67 × 10
–
−5
−4
−5
−4
−5
4.77 × 10
7.54 × 10
1.09 × 10−
–
Table III and Fig. 4 present the experimental data
thus obtained.
For the five mixtures, ignition delay times (τ)
decrease when T rises and varies exponentially vs.
but increase with the concentration of hydrocarbon for
a given equivalence ratio. It is worth noting that in
the case of benzene [11], ignition delay times decrease
when the concentration of hydrocarbon increases. The
lower part of Fig. 4 also presents the data obtained by
Burcat et al. [16] at ꢀ = 1, at 0.5% of toluene, and
at a pressure of 7 bar, which is close to the pressure
1
000/T . It is also shown that for a given T , igni-
tion delay times decrease with the equivalence ratio
of the mixture for a given concentration of toluene,

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
29
Figure 1 Oxidation of toluene in a jet-stirred reactor at ꢀ = 0.9. Comparison between experimental (symbols) and computed
(lines) mole fraction of species versus residence time at 873 K (a), 893 K (b), and 923 K (c).
•
(
8.4 bar) used in our experiments, and shows that both
A secondary mechanism including 134 reactions,
in which the reactants are the molecular prod-
ucts (apart from benzene) formed by the primary
mechanism and styrene.
The mechanism for the oxidation of benzene
which is presented in a previous paper [11] and
includes the C0 C6 reaction base, which is also
described in recent papers [24,25].
series of measurements are in good agreement.
DESCRIPTION OF THE
REACTION MECHANISM
•
This mechanism, which is available on request, has
been written in the CHEMKIN II [23] format and
includes three parts:
Benzene is an important primary product of the
oxidation of toluene. As previously presented [11],
its oxidation mechanism contains 135 reactions and
includes the reactions of benzene molecules and of
•
A primary mechanism containing 59 reactions, in
which only toluene and oxygen are considered as
molecular reactants,

30
BOUNACEUR ET AL.
Figure 2 Oxidation of toluene in a jet-stirred reactor at 893 K and at ꢀ = 0.45. Comparison between experimental (symbols)
and computed (lines) mole fraction of species versus residence time.
cyclohexadienyl, phenyl, phenylperoxy, phenoxy, hy-
droxyphenoxy, cyclopentadienyl, cyclopentadienoxy,
and hydroxycyclopentadienyl free radicals, as well
as the reactions of ortho-benzoquinone, phenol, cy-
clopentadiene, cyclopentadienone, and vinylketene
molecules, which are the primary products obtained.
The C0 C6 reaction base was constructed from
a review of the recent literature and is an extension
of our previous C0 C2 reaction base [26,24]. This
C0 C2 reaction base includes the reactions of radicals
or molecules including carbon, hydrogen, and oxygen
atoms and containing less than three carbon atoms. The
kinetic data used in this base were taken from the lit-
erature and are mainly those proposed by Baulch et al.
formation of benzene [24,25]. Pressure-dependent rate
constants follow the formalism proposed by Troe [29],
and efficiency coefficients have been included. This
reaction base was constructed in order to model ex-
perimental results obtained in a jet-stirred reactor for
methane and ethane [26]; profiles in laminar flames of
methane, acetylene, and 1,3-butadiene [24]; and shock
tube autoignition delays for acetylene, propyne, allene,
1,3-butadiene [24], 1-butyne, and 2-butyne [25].
Specific heats, heats of formation, and entropies
of the molecules or radicals considered have been
calculated using software THERGAS [30], based on
the group and bond-additivity methods proposed by
Benson [31], and stored as 14 polynomial coeffi-
cients, according to the CHEMKIN II formalism [23].
Table IV presents the names, the formulae, and the
heats of formation of the 25 aromatic species that are
not included in the mechanism for the oxidation of
benzene [11]. It must be kept in mind that the precision
obtained by using group additivity methods to estimate
[27] and Tsang and Hampson [28]. The C0 C6 reac-
tion base includes reactions involving C3H2··, ·C3H3,
C3H4 (allene and propyne), ·C3H5 (three isomers),
C3H6, C4H2, ·C4H3 (two isomers), C4H4, ·C4H5 (five
isomers), C4H6 (1,3-butadiene, 1,2-butadiene, methyl-
cyclopropene, 1-butyne, and 2-butyne), as well as the

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
31
Figure 3 Oxidation of toluene in a jet-stirred reactor at 893 K and at ꢀ = 0.45. Comparison between experimental (symbols)
and computed (lines) mole fraction of species versus residence time.
−1
heats of formation is around 2 kcal mol for molecules
and 4 kcal mol for radicals [31].
Primary Mechanism
−1
The primary and secondary mechanisms for the ox-
idation of toluene are presented in Table V, as well as
the references related to the kinetic data used. These
mechanisms have been constructed using as a basis
those proposed by Emdee et al. [2] and Lindstedt and
Maurice [21], but in order to be more comprehensive to
model experimental results around 900 K, e.g. termi-
nation steps of resonance-stabilized radicals have been
considered in more detail.
The primary mechanism includes the reactions of
toluene molecules and of benzyl, methylphenyl, per-
oxybenzyl, alcoxybenzyl, and cresoxy free radicals.
Unimolecular decompositions (1 and 2, numbers re-
ferring to Table V) of toluene molecules give either
benzyl radicals and ·H atoms or phenyl and methyl
radicals. Two rather different expressions for the rate
constant of the reaction involving the breaking of a
C C bond have been published. Pamidimukkala et al.

32
BOUNACEUR ET AL.
Table III Mixture Compositions, Shock Conditions, and Ignition Delays for Toluene
Composition
mole %)
Composition
(mole %)
(
−
−1
P1 (atm) V (m s ) P5 (atm) T5 (K) τ (s)
C7H8
O2
P1 (atm) V (m s 1) P5 (atm) T5 (K) τ (ꢀs)
C7H8
1.25
O2
1
.25
11.25
0.2631
.2237
0.1974
794
838
847
847
857
862
867
872
882
893
914
8.28
8.25
8.51
9.01
9.43
8.48
8.50
8.69
8.42
8.72
8.70
1305 1553
7.5
0.246
824
847
847
857
862
872
877
882
887
909
8.60
8.47
9.07
8.44
9.00
8.65
8.25
8.22
8.92
8.51
1418 179
1486 86
1486 111
1515 63
1530 54
1560 40
1576 25
1591 32
1608 19
0
1419
1456
1456
1474
1488
1513
1528
1558
1578
1656
90
59
66
50
57
17
18
13
4
0.2237
0.2395
0.2158
0.2263
0.2105
0.1974
0.1934
0.2066
0.1842
Φ = 1
Φ = 1.5
0
0
0
0
0
0
0
0
.2092
.2395
.2118
.2368
.2105
.2237
.1974
.1842
1675
5
3
1
.25
3.75
0.1987
0.1887
0.1802
862
893
909
8.28
8.68
8.88
1563 128
1660 37
1712 25
1
.25
22.5
0.2585
.246
0.2385
824
833
857
857
862
872
872
872
882
898
898
909
9.18
9.00
9.45
9.40
8.60
9.33
8.23
9.33
9.08
9.14
8.06
8.77
1311 203
1335 170
0
Φ = 3
0.5
Φ = 0.5
1398
1398
1411
1438
1438
1438
1465
1509
1509
1539
56
51
38
23
26
36
19
6
0.2368
0.2131
0.2239
0.1974
0.2237
0.2105
0.2010
0.1776
0.1868
4.475 0.2526
0.2329
843
862
888
909
8.84
8.68
7.97
8.52
1569 119
1633 38
Φ = 1.006
0.1974
0.1974
1720
1795
7
5
8
5
1 5 5
Note: P is the pressure of the mixture before the shock. V the speed of the incident wave. P and T are pressure and temperature behind
the reflected shock wave. τ is the ignition delay time.
1
3
−1
,
[
58] proposed k = 8.9 × 10 exp(−143750/T ) s
icals (5), and the addition of ·OH radicals produces
·H atoms and benzyl alcohol (6). We have considered
fewer addition reactions than Lindstedt and Maurice
[21], who proposed two possible additions with ·OH
radicals and three with ·O· atoms, but we have written
only the most probable channels for which a satisfac-
tory evaluation of the rate constant could be found. In
the case of the addition of ·OH radicals, the best data
available were proposed by Baulch et al. [39] for the
formation of an adduct between 200 and 300 K. We
have preferred to use the same value as that proposed
for the formation of phenol from benzene [11].
An important part of the reactions of toluene
molecules involves the formation of the resonance-
stabilized benzyl radicals by unimolecular (1) and
bimolecular (3) initiations and by metatheses with
H-abstraction (7–24). Because of the high stability of
benzyl radicals, metatheses are considered for a wide
range of radicals, especially resonance-stabilized rad-
icals, which are present in important concentration.
The radicals taken into account are then ·H, ·O·, ·OH,
·HO2, ·CH3, ·C2H3, ·C3H5 (allyl), ·nC4H5, ·iC4H5,
which was the value used by Lindstedt and Maurice
[
1
21], while Rao and Skinner [45] gave, k = 1.4 ×
1
6
−1
0 exp(−197600/T ) s at 4 bar, which is a pres-
sure close to the average value of the data that we
have used for the validations in shock tubes. We have
based our rate constant on this last expression, which
is more consistent with thermochemistry (the energy
−
1
of the broken C C bond is around 100 kcal mol ac-
cording to Benson [31]), but multiplying the A factor
by 7 to obtain a better agreement with the experiments
performed in a flow reactor at ꢀ = 1.4. Rao and Skin-
ner [45] have shown that these unimolecular initiations
display an important fall-off effect at 1600 K, for pres-
sures below 2 bar. That should be taken into account for
further modeling at higher temperature or lower pres-
sure. The initiations with oxygen molecules (3) have
also been considered, with a rate constant calculated as
recommended by Ingham et al. [46] for alkylbenzenes.
The addition of ·H atoms to toluene molecules leads
to the formation of benzene and methyl radicals (4), the
addition of ·O· atoms gives ·H atoms and cresoxy rad-

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
33
Table IV Names, formulae, and heats of formation for
aromatic species involved in Table V. The heats of
formation have been calculated by the software
−
1
THERGAS [30] at 298 K in kcal mol
ꢁ
Hf (298 K)
Molecules
Toluene (C6H4CH3)
Benzaldehyde (C6H5CHO)
Benzyl hydroperoxide (C6H5CH2OOH)
Cresols (HOC6H4CH3)
Benzyl alcohol (C6H5CH2OH)
Ethyl benzene (C6H5CH2CH3)
Styrene (C6H5CH CH2)
11.9
−6.0
−8.2
−30.1
−24.0
7.1
35.2
Bibenzyl (C6H5CH2CH2C6H5)
Stilbene (C6H5CH CHC6H5)
Hydroxy benzaldehyde (HOC6H4CHO)
34.1
58.1
−51.5
Radicals
Benzyl (·C6H5CH2)
49.4
73.1
26.9
27.3
5.3
26.0
6.9
9.3
−15.5
−16.7
−15.8
40.4
55.9
81.6
67.4
Methyl phenyl (·C6H4CH3)
Peroxy benzyl (·C6H5CH2O2)
Alcoxy benzyl (·C6H5CH2O)
Cresoxy (·OC6H4CH3)
Benzoyl (·C6H5CO)
Hydroxy benzyls (·HOC6H4CH2)
Hydroxy benzyls (·C6H5CHOH)
Hydroxy alcoxybenzyl (·HOC6H4CH2O)
Hydroxy benzoyl (·HOC6H4CO)
Hydroxy peroxybenzyl (·HOC H CH O )
Figure 4 Autoignition delays of toluene in a shock tube
from 7 to 9.5 bar. Semilog plot of experimental (symbols)
and computed (lines) ignition delays as a function of temper-
ature behind the reflected shock wave for different equiva-
lence ratios at 1.25% of toluene (a) and for different toluene
concentrations at ꢀ = 1 (b). The experimental results ob-
tained by Burcat et al. [16] at ꢀ = 1 and at 0.5% of toluene
is shown in (b), together with the related computed line.
6
4
2 2
C H (·C H CHCH )
8
9
6
5
3
C H -1 (·C H CH CH )
8
9
6
5
2
2
C H (·C H C CH )
8
7
6
5
2
C H (·C H CHCH C H )
1
4
13
6
5
2 6 5
zyl, respectively. The reactions with ·O· atoms, which
are important at high temperature, lead to phenyl rad-
icals and formaldehyde molecules (34) or to ·H atoms
and benzaldehyde molecules (35). For the rate con-
·
C5H5 (cyclopentadienyl), ·C6H5 (phenyl), ·C6H5O
(
(
(
phenoxy), ·C6H4CH3 (methyl phenyl), ·OC6H4CH3
cresoxy), C6H5CH2OO· (peroxybenzyl), C6H5CH2O·
alcoxybenzyl), and HOC6H4CH2· (hydroxybenzyl)
14
stant of this last channel, we use a value of 1 × 10
3
−1 −1
s , while Hippler et al. [38] proposed
cm mol
1
4
3
−1 −1
and Linstedt and Maurice [21]
radicals. In order to be as comprehensive as possible,
wehavealsoconsideredtheformationofmethylphenyl
radicals by abstraction of phenylic H-atoms (25–29).
These reactions were not considered by previous au-
thors [2,13,14,21,22].
4 × 10 cm mol
s
1
3
3
−1 −1
have used 3.5 × 10 cm mol
s .
The reactions of methyl phenyl radicals (40–47)
have been adapted from those proposed for phenyl radi-
cals [11] and includes reactions with oxygen molecules
to give either ·O· atoms and cresoxy radicals or methyl
radicals and benzoquinone molecules and termination
steps with other radicals.
Benzyl radicals can decompose by beta-scission as
proposed by Colket and Seery [35] (30,31), react with
oxygen molecules (32,33), or by termination reactions
with other radicals (34–39). Two reactions with oxy-
gen are possible, the formation of peroxybenzyl radi-
cals (32), which is important at low temperature, and
the production of ·O· atoms and alcoxybenzyl rad-
icals (33). Benzyl radicals can combine with ·OH,
Ellis et al. [59] have shown that the peroxybenzyl
radical isomerization and decomposition makes a sub-
stantial contribution to the formation of benzaldehyde
at 773 K. The kinetic data for this channel (48) have
been calculated according to the method proposed by
Warth et al. [48] for peroxy alkyl radicals. A-factor
has been derived from the change in the number of
·
CH3, ·HO2, and benzyl radical, to give benzyl alco-
hol, ethylbenzene, benzylhydroperoxide, and biben-

34
BOUNACEUR ET AL.
Table V Primary and Secondary Mechanism for the Oxidation of Toluene
Reactions
A
n
Ea
References
No.
PRIMARY MECHANISM
Reactions of Toluene Molecules
Umimolecular inititiation
Toluene = Benzyl + H
1
1
5
7
1.0 × 10
1.0 × 10
0.0
0.0
88.9
97.0
HIPPLER90 [32]
Estimated
(1)
(2)
a
Toluene = C6H5 + CH3
Bimolecular inititiation
Toluene + O2 = Benzyl + HO2
Additions
Toluene + H = C6H6 + CH3
Toluene + O = OC6H4CH3 + H
Toluene + OH = HOC6H4CH3 + H
1
2
Estimatedb
2.1 × 10
0.0
38.6
(3)
1
1
1
3
3
3
5.8 × 10
1.7 × 10
0.0
0.0
0.0
8.1
3.6
10.6
BAULCH94 [27]
TAPPE89 [33]
(4)
(5)
(6)
c
1.3 × 10
Estimated
Metatheses
Metatheses with abstraction of a benzylic H-atom
Toluene + H = Benzyl + H2
1
1
9
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
4
1
1.2 × 10
6.3 × 10
5.2 × 10
4.0 × 10
1.6 × 10
4.0 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
7.9 × 10
1.6 × 10
7.9 × 10
1.6 × 10
4.0 × 10
1.6 × 10
1.6 × 10
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
8.4
0.0
0.87
14.0
11.1
8.0
15.1
15.1
15.1
15.1
11.1
12.0
15.1
12.0
15.1
14.0
11.1
15.1
HIPPLER90
HOEEMANN90[34]
BAULCH94
BAULCH94
COLKET94 [35]
COLKET94
(7)
(8)
(9)
Toluene + O = Benzyl + OH
Toluene + OH = Benzyl + H2O
1
2
2
2
2
2
2
2
3
1
3
1
1
1
1
Toluene + HO2 = Benzyl + H2O2
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
Toluene + CH3 = Benzyl + CH4
Toluene + C2H3 = Benzyl + C2H4
d
Toluene + C3H5 = Benzyl + C3H6
Estimated
Estimated
Estimated
Estimated
d
Toluene + C3H3 = Benzyl + pC3H4
Toluene + iC4H5 = Benzyl + 1,3-C4H6
Toluene + nC4H5 = Benzyl + 1,3-C4H6
Toluene + C5H5 = Benzyl + C5H6
d
d
d
Estimated
Toluene + C6H5 = Benzyl + C6H6
HECKMANN96 [36]
d
Toluene + C6H5O = Benzyl + C6H5OH
Toluene + C6H4CH3 = Benzyl + Toluene
Toluene + OC6H4CH3 = Benzyl + HOC6H4CH3
Toluene + C6H5CH2OO = Benzyl + C6H5CH2OOH
Toluene + C6H5CH2O = Benzyl + C6H5CH2OH
Toluene + HOC6H4CH2 = Benzyl + HOC6H4CH3
Metatheses with abstraction of a phenylic H-atom
Toluene + H = C6H4CH3 + H2
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
e
d
f
d
d
8
1
8
1
1
c
6.0 × 10
2.0 × 10
1.6 × 10
4.0 × 10
2.0 × 10
1.0
0.0
1.42
0.0
16.8
14.7
1.45
28.9
15.0
Estimated
Estimated
Estimated
Estimated
(25)
(26)
(27)
(28)
(29)
3
c
Toluene + O = C6H4CH3 + OH
c
Toluene + OH = C6H4CH3 + H2O
1
2
c
Toluene + HO2 = C6H4CH3 + H2O2
Toluene + CH3 = C6H4CH3 + CH4
c
0.0
Estimated
Reactions of Benzyl Radicals
Decompositions by beta-scission
Benzyl = C5H5 + C2H2
1
1
3
4
6.0 × 10
2.0 × 10
0.0
0.0
70.0
83.6
COLKET94
COLKET94
(30)
(31)
Benzyl = C3H3 + C4H4
Reactions with oxygen molecule
Benzyl + O2 = C6H5CH2OO
Benzyl + O2 = C6H5CH2O + O
Termination reactions
1
1
1
2
4.6 × 10
6.3 × 10
0.0
0.0
−0.38
FENTER94 [37]
LINSTEDT96 [21]
(32)
(33)
40.0
1
1
1
1
1
1
3
4
3
2
5
1
Benzyl + O = C6H5 + HCHO
Benzyl + O = C6H5CHO + H
Benzyl + OH = C6H5CH2OH
Benzyl + HO2 = C6H5CH2OOH
etC6H5 = Benzyl + CH3
3.5 × 10
1.0 × 10
2.0 × 10
5.0 × 10
6.1 × 10
2.5 × 10
0.0
0.0
0.0
0.0
0.0
0.4
0.0
0.0
0.0
0.0
75.1
0.0
EMDEE92 [2]
Estimated
HIPPLER90’ [38]
HIPPLER90’
BAULCH92 [39]
MULLER88 [40]
(34)
(35)
(36)
(37)
(38)
(39)
g
2
Benzyl = Bibenzyl
Reactions of Methyl Phenyl Radicals
Reactions with oxygen molecule
C6H4CH3 + O2 = OC6H4CH3 + O
1
3
Estimatedh
2.6 × 10
0.0
6.1
(40)
Continued

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
35
Table V Continued
Reactions
A
n
Ea
References
Estimatedh
No.
1
3
C6H4CH3 + O2 = OC6H4O + CH3
Termination reactions
3.0 × 10
0.0
9.0
(41)
1
1
1
1
6
1
4
3
4
3
Estimatedi
MILLER92 [41]
C6H4CH3 + H = Toluene
1.0 × 10
1.0 × 10
1.0 × 10
1.0 × 10
1.2 × 10
5.0 × 10
0.0
0.0
0.0
0.0
1.96
0.0
0.0
0.0
0.0
(42)
(43)
(44)
(45)
(46)
(47)
C6H4CH3 + H = Benzyl + H
C6H4CH3 + O = OC6H4CH3
C6H4CH3 + OH = HOC6H4CH3
C6H4CH3 + CH3 = Xylene
h
Estimated
Estimated
Estimated
Estimated
h
0.0
j
−3.7
2
h
C6H4CH3 + HO2 = OC6H4CH3 + OH
0.0
Reactions of Peroxybenzyl Radicals
Isomerisation-decomposition reactions
C6H5CH2OO = C6H5CHO + OH
Combination reactions
9
Estimatedk
Estimatedi
3.4 × 10
1.0 × 10
1.0
37.5
0.0
(48)
(49)
1
4
C6H5CH2OO + H = C6H5CH2OOH
0.0
Dispropotionnation reactions
1
1
1
1
0
0
l
C6H5CH2OO + HO2 = C6H5CH2OOH + O2
2.0 × 10
1.4 × 10
6.3 × 10
0.0
0.0
0.0
0.0
Estimated
(50)
(51)
(52)
l
2
C6H5CH2OO = C6H5CH2OH + C6H5CHO + O2
−0.72
Estimated
l
C6H5CH2OO + C6H5CH2OO = 2C6H5CH2O + O2
−0.72
Estimated
Reactions of Alcoxy Benzyl Radicals
Decomposition by beta-scission
C6H5CH2O = H + C6H5CHO
C6H5CH2O = C6H5 + HCHO
Reactions with oxygen
1
1
3
3
m
2.0 × 10
2.0 × 10
0.0
0.0
27.5
27.5
Estimated
Estimated
(53)
(54)
m
1
0
Estimatedn
C6H5CH2O + O2 = HO2 + C6H5CHO
6.0 × 10
0.0
1.6
(55)
Reactions of Cresoxy Radicals
CO elimination with rearrangement
OC6H4CH3 = H + C6H6 + CO
OC6H4CH3 = H + C4H6 + C2H2 + CO
OC6H4CH3 = C3H3 + C3H4 + CHO
Termination reactions
1
1
1
1
1
1
o
3.0 × 10
3.1 × 10
1.5 × 10
0.0
0.0
0.0
43.8
43.8
43.8
Estimated
Estimated
Estimated
(56)
(57)
(58)
o
o
1
4
Estimatedi
OC6H4CH3 + H = HOC6H4CH3
1.0 × 10
0.0
0.0
(59)
SECONDARY MECHANISM
Reactions of Benzaldehyde Molecules and Derived Radicals
1
1
1
1
1
1
1
5
1
3
3
2
2
2
C6H5CHO = C6H5CO + H
4.0 × 10
7.0 × 10
5.8 × 10
4.0 × 10
6.0 × 10
7.8 × 10
3.0 × 10
2.0 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
1.3 × 10
4.0 × 10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
83.7
39.5
8.1
3.2
1.8
0.0
11.0
1.5
GRELA86 [42]
(60)
(61)
(62)
(63)
(64)
(65)
(66)
(67)
(68)
(69)
(70)
(71)
(72)
(73)
(74)
(75)
(76)
(77)
(78)
b
C6H5CHO + O2 = C6H5CO + HO2
C6H5CHO + H = C6H6 + R5CHO
Estimated
p
Estimated
q
C6H5CHO + H = C6H5CO + H2
Estimated
C6H5CHO + O = C6H5CO + OH
BAULCH94
BAULCH94
C6H5CHO + OH = C6H5CO + H2O
C6H5CHO + HO2 = C6H5CO + H2O2
C6H5CHO + CH3 = C6H5CO + CH4
C6H5CHO + C2H5 = C6H5CO + C2H6
C6H5CHO + C3H5 = C6H5CO + C3H6
C6H5CHO + iC4H5 = C6H5CO + 1,3-C4H6
C6H5CHO + nC4H5 = C6H5CO + 1,3-C4H6
C6H5CHO + C5H5 = C6H5CO + C5H6
C6H5CHO + C6H5 = C6H5CO + C6H6
C6H5CHO + Benzyl = Toluene + C6H5CO
C6H5CHO + C6H5O = C6H5CO + C6H5OH
C6H5CHO + OC6H4CH3 = C6H5CO + HOC6H4CH3
C6H5CHO + HOC6H4CH2 = C6H5CO + HOC6H4CH3
C6H5CO = C6H5 + CO
q
Estimated
−
6
q
Estimated
1
1
1
1
1
1
1
1
1
1
1
2
q
7.5
Estimated
2
Estimatedr
Estimatedr
Estimatedr
Estimatedr
Estimatedr
11.5
11.5
7.5
11.5
11.5
11.5
11.5
11.5
11.5
29.5
2
2
1
1
1
1
1
1
4
r
Estimated
Estimatedr
r
Estimated
Estimatedr
SOLLY71 [43]
Reactions of Benzyl Hydropreroxide Molecules and Derived Radicals
1
6
Estimateds
C6H5CH2OOH = C6H5CH2O + OH
1.5 × 10
0.0
42.0
(79)
Continued

36
BOUNACEUR ET AL.
Table V Continued
Reactions
A
n
Ea
References
Estimatedt
No.
Reactions of Cresol Molecules and Derived Radicals
1
1
1
1
1
8
1
1
1
1
1
1
1
1
1
9
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
9
1
1
1
1
1
1
1
3
2
3
4
3
HOC6H4CH3 + O2 = OC6H4CH3 + HO2
1.0 × 10
2.1 × 10
5.8 × 10
1.2 × 10
1.3 × 10
1.4 × 10
1.0 × 10
1.8 × 10
4.9 × 10
4.9 × 10
4.9 × 10
4.9 × 10
4.9 × 10
1.2 × 10
6.3 × 10
5.2 × 10
4.0 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
7.9 × 10
1.6 × 10
7.9 × 10
1.6 × 10
0.0
0.0
0.0
0.0
0.0
38.9
39.7
8.1
12.4
2.9
(80)
(81)
(82)
(83)
(84)
(85)
(86)
(87)
(88)
(89)
(90)
(91)
(92)
(93)
(94)
(95)
(96)
b
HOC6H4CH3 + O2 = HOC6H4CH2 + HO2
HOC6H4CH3 + H = C6H5OH + CH3
Estimated
Estimated
Estimated
Estimated
p
t
HOC6H4CH3 + H = OC6H4CH3 + H2
t
HOC6H4CH3 + O = OC6H4CH3 + OH
t
HOC6H4CH3 + OH = OC6H4CH3 + H2O
HOC6H4CH3 + HO2 = OC6H4CH3 + H2O2
HOC6H4CH3 + CH3 = OC6H4CH3 + CH4
HOC6H4CH3 + C6H5 = OC6H4CH3 + C6H6
HOC6H4CH3 + C5H5 = OC6H4CH3 + C5H6
HOC6H4CH3 + C3H5 = OC6H4CH3 + C3H6
HOC6H4CH3 + iC4H5 = OC6H4CH3 + 1,3-C4H6
HOC6H4CH3 + C6H5O = OC6H4CH3 + C6H5OH
HOC6H4CH3 + H = HOC6H4CH2 + H2
1.4 −0.96 Estimated
2
1
2
1
1
1
1
4
1
t
t
t
t
t
t
t
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
10.0
7.7
4.4
9.4
9.4
9.4
9.4
8.4
0.0
0.87
14.0
11.1
15.1
15.1
15.1
11.1
15.1
12.0
15.1
12.0
15.1
14.0
11.1
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
u
u
i
HOC6H4CH3 + O = HOC6H4CH2 + OH
HOC6H4CH3 + OH = HOC6H4CH2 + H2O
HOC6H4CH3 + HO2 = HOC6H4CH2 + H2O2
HOC6H4CH3 + CH3 = HOC6H4CH2 + CH4
HOC6H4CH3 + C3H5 = HOC6H4CH2 + C3H6
HOC6H4CH3 + C3H3 = HOC6H4CH2 + pC3H4
HOC6H4CH3 + iC4H5 = HOC6H4CH2 + 1,3-C4H6
HOC6H4CH3 + nC4H5 = HOC6H4CH2 + 1,3-C4H6
HOC6H4CH3 + C5H5 = HOC6H4CH2 + C5H6
HOC6H4CH3 + C6H5 = HOC6H4CH2 + C6H6
HOC6H4CH3 + C6H5O = HOC6H4CH2 + C6H5OH
HOC6H4CH3 + C6H4CH3 = HOC6H4CH2 + Toluene
HOC6H4CH3 + OC6H4CH3 = HOC6H4CH2 + HOC6H4CH3
1
2
2
2
2
2
1
3
1
3
1
1
1
1
2
4
2
2
(97)
(98)
(99)
(100)
(101)
(102)
(103)
(104)
(105)
(106)
(107)
(108)
(109)
(110)
(111)
(112)
(113)
(114)
(115)
(116)
(117)
(118)
(119)
(120)
(121)
(122)
(123)
HOC6H4CH3 + C6H5CH2OO = HOC6H4CH2 + C6H5CH2OOH 4.0 × 10
HOC6H4CH3 + C6H5CH2O = HOC6H4CH2 + C6H5CH2OH
HOC6H4CH2 + O2 = HOC6H4CH2OO
HOC6H4CH2 + O2 = HOC6H4CH2O + O
HOC6H4CH2 + H = HOC6H4CH3
1.6 × 10
4.6 × 10
6.3 × 10
1.0 × 10
5.0 × 10
5.0 × 10
3.4 × 10
2.0 × 10
2.0 × 10
6.0 × 10
4.0 × 10
6.0 × 10
7.8 × 10
3.0 × 10
2.0 × 10
4.0 × 10
0.0 −0.38 Estimated
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.6
0.0
40.0
0.0
0.0
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
u
v
w
x
HOC6H4CH2 + HO2 = HOC6H4CH2OOH
HOC6H4CH2 + CH3 = C6H5OH + C2H4
HOC6H4CH2OO = HOC6H4CHO + OH
HOC6H4CH2O = H + HOC6H4CHO
0.0
37.5
27.5
27.5
1.6
3.2
1.8
0.0
11.0
1.5
3
3
0
3
2
2
2
x
HOC6H4CH2O = C6H4OH + HCHO
x
HOC6H4CH2O + O2 = OOH + HOC6H4CHO
HOC6H4CHO + H = HOC6H4CO + H2
HOC6H4CHO + O = HOC6H4CO + OH
HOC6H4CHO + OH = HOC6H4CO + H2O
HOC6H4CHO + HO2 = HOC6H4CO + H2O
HOC6H4CHO + CH3 = HOC6H4CO + CH4
HOC6H4CO = C6H5O + CO
y
y
y
y
−
1
6
y
4
Estimatedz
29.5
Reactions of Benzyl Alcohol Molecules and Derived Radicals
1
1
1
1
1
9
1
2
4
3
3
1
Estimatedb
EMDEE92
C6H5CH2OH + O2 = HO2 + C6H5CHOH
C6H5CH2OH + O2 = HO2 + C6H5CH2O
C6H5CH2OH + H = C6H6 + CH2OH
1.4 × 10
2.0 × 10
5.8 × 10
8.0 × 10
4.2 × 10
3.9 × 10
2.7 × 10
0.0
0.0
0.0
0.0
0.0
1.0
0.0
34.0
41.4
8.1
(124)
(125)
(126)
(127)
(128)
(129)
(130)
p
Estimated
aa
C6H5CH2OH + H = C6H5CHOH + H2
C6H5CH2OH + O = C6H5CHOH + OH
C6H5CH2OH + OH = C6H5CHOH + H2O
C6H5CH2OH + HO2 = C6H5CHOH + H2O2
6.4
Estimated
Estimated
Estimated
Estimated
aa
aa
aa
−2.0
−1.1
12.0
1
Continued

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
37
Table V Continued
Reactions
A
n
Ea
References
No.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
3
1
1
3
1
1
1
4
aa
C6H5CH2OH + CH3 = C6H5CHOH + CH4
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
5.2 × 10
1.1 × 10
1.1 × 10
5.2 × 10
1.1 × 10
1.1 × 10
1.1 × 10
2.0 × 10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
9.1
Estimated
(131)
(132)
(133)
(134)
(135)
(136)
(137)
(138)
(139)
(140)
(141)
(142)
(143)
(144)
aa
C6H5CH2OH + C3H3 = C6H5CHOH + pC3H4
C6H5CH2OH + C3H5 = C6H5CHOH + C3H6
C6H5CH2OH + iC4H5 = C6H5CHOH + 1,3-C4H6
C6H5CH2OH + nC4H5 = C6H5CHOH + 1,3-C4H6
C6H5CH2OH + C5H5 = C6H5CHOH + C5H6
C6H5CH2OH + C6H5 = C6H5CHOH + C6H6
C6H5CH2OH + C6H5O = C6H5CHOH + C6H5OH
C6H5CH2OH + Benzyl = C6H5CHOH + Toluene
C6H5CH2OH + C6H4CH3 = C6H5CHOH + C6H5CH2OH
C6H5CH2OH + OC6H4CH3 = C6H5CHOH + HOC6H4CH3
C6H5CH2OH + HOC6H4CH2 = C6H5CHOH + HOC6H4CH3
C6H5CH2OH + C6H5CH2O = C6H5CHOH + C6H5CH2OH
C6H5CHOH = C6H5CHO + H
13.1
13.1
13.1
13.1
9.1
10.0
13.1
13.1
10.0
13.1
13.1
9.1
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
bb
23.3
Estimated
Reactions of Ethylbenzene Molecules and Derived Radicals
1
1
1
1
1
9
1
1
1
1
1
1
1
1
1
1
1
7
1
6
1
4
2
3
3
3
cc
etC6H5 = C8H9 + H
4.3 × 10
1.4 × 10
1.2 × 10
5.8 × 10
8.0 × 10
3.9 × 10
1.2 × 10
2.7 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
1.1 × 10
2.8 × 10
5.1 × 10
2.7 × 10
6.0 × 10
0.3
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
0.0
2.0
0.0
4.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
83.6
34.0
49.0
8.1
Estimated
Estimated
Estimated
Estimated
(146)
(147)
(148)
(149)
(150)
(151)
(152)
(153)
(154)
(155)
(156)
(157)
(158)
(159)
(160)
(161)
(162)
(163)
(164)
(165)
(166)
(167)
(168)
(169)
b
etC6H5 + O2 = C8H9 + HO2
dd
etC6H5 + O2 = C8H9-1 + HO2
etC6H5 + H = C6H6 + C2H5
aa
aa
etC6H5 + H = C8H9 + H2
6.4
Estimated
aa
etC6H5 + O = C8H9 + OH
−1.1 Estimated
4
1
2
2
2
2
2
2
2
1
1
aa
aa
aa
aa
aa
aa
aa
aa
aa
aa
aa
ee
etC6H5 + OH = C8H9 + H2O
etC6H5 + HO2 = C8H9 + H2O2
etC6H5 + CH3 = C8H9 + CH4
etC6H5 + C3H5 = C8H9 + C3H6
etC6H5 + iC4H5 = C8H9 + 1,3-C4H6
etC6H5 + nC4H5 = C8H9 + 1,3-C4H6
etC6H + C5H5 = C8H9 + C5H6
etC6H5 + C6H5O = C8H9 + C6H5OH
etC6H5 + Benzyl = C8H9 + Toluene
etC6H5 + OC6H4CH3 = C8H9 + HOC6H4CH3
etC6H5 + HOC6H4CH2 = C8H9 + HOC6H4CH3
etC6H5 + H = C8H9-1 + H2
8.4
12.0
9.1
13.1
13.1
13.1
9.1
13.1
13.1
13.1
13.1
7.7
7.8
0.4
17.0
8.2
50.7
34.8
43.5
15.0
5.0
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
3
1
ee
etC6H5 + O = C8H9-1 + OH
ee
etC6H5 + OH = C8H9-1 + H2O
etC6H5 + HO2 = C8H9-1 + H2O2
etC6H5 + CH3 = C8H9-1 + CH4
C8H9 = H + Styrene
ee
ee
1
1
1
1
1
1
1
3
3
4
1
2
2
2
3.1 × 10
1.6 × 10
7.1 × 10
7.0 × 10
1.5 × 10
5.0 × 10
5.0 × 10
MULLER88
f f
C8H9-1 = H + Styrene
Estimated
C8H9-1 = C2H4 + C6H5
EBERT78 [44] (170)
gg
C8H9 + O2 = HO2 + Styrene
C8H9-1 + O2 = HO2 + Styrene
C8H9 + HO2 = OH + C6H5CHO
C8H9-1 + HO2 = OH + HCHO + Benzyl
Estimated
Estimated
Estimated
Estimated
(171)
(172)
(173)
(174)
gg
cc
cc
0.0
0.0
Reactions of Styrene Molecules and Derived Radicals
1
8
1
1
5
6
1
1
3
hh
hh
hh
hh
hh
hh
hh
ii
Styrene + O2 = HO2 + C6H5C2H2
Styrene + O = C6H5 + CH2CHO
Styrene + OH = Benzyl + HCHO
Styrene + OH = C6H5CHO + CH3
Styrene + H = C6H5C2H2 + H2
Styrene + OH = C6H5C2H2 + H2O
Styrene + CH3 = C8H7 + CH4
C8H7 + O2 = C6H5CO + HCHO
2.0 × 10
3.0 × 10
1.4 × 10
2.8 × 10
6.6 × 10
1.5 × 10
3.5 × 10
4.5 × 10
0.0
1.45
0.0
0.0
2.53
2.0
0.0
−1.39
57.9
0.9
0.0
0.0
9.2
0.4
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
Estimated
(175)
(176)
(177)
(178)
(179)
(180)
(181)
(182)
2
2
3
6
15.5
1.0
Continued

38
BOUNACEUR ET AL.
Table V Continued
Reactions
A
n
Ea
References
No.
Reactions of Bibenzyl Molecules and Derived Radicals
1
1
1
1
9
1
1
1
1
1
1
4
2
4
1
Bibenzyl = C14H13 + H
1.0 × 10
2.8 × 10
1.6 × 10
8.4 × 10
7.0 × 10
5.4 × 10
2.2 × 10
2.2 × 10
2.2 × 10
2.2 × 10
7.1 × 10
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
83.0
35.0
6.4
−2.0
−1.1
12.0
9.1
9.1
9.1
9.1
30.0
HIPPLER90’
(183)
(184)
(185)
(186)
(187)
(188)
(189)
(190)
(191)
(192)
(193)
b
Bibenzyl + O2 = C14H13 + HO2
Bibenzyl + H = C14H13 + H2
Estimated
aa
Estimated
aa
Bibenzyl + O = C14H13 + OH
Bibenzyl + OH = C14H13 + H2O
Bibenzyl + HO2 = C14H13 + H2O2
Bibenzyl + CH3 = C14H13 + CH4
Bibenzyl + C3H5 = C14H13 + C3H6
Bibenzyl + C6H5O = C14H13 + C6H5OH
Bibenzyl + Benzyl = C14H13 + Toluene
C14H13 = Stilbene + H
Estimated
aa
Estimated
1
2
2
1
1
4
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
aa
Estimated
EBERT78
n
The rate constants are given at 1 atm (k = AT exp(−E
a
/RT )) in cc, mol, s, kcal units. Reference numbers are given in brackets when they
appear for the first time. This mechanism should be used together with our mechanism for the oxidation of benzene [11].
a
Rate constant taken equal to that proposed by Rao et al. [45] multiplied by 10.
The rate constant of this bimolecular initiation with oxygen molecule has been calculated as proposed by Ingham et al. [46]: A is taken
b
11
3
−1 −1
equal to n × 7 × 10 cm mol
s , where n is the number of benzylic hydrogen atoms abstractable, and the activation energy to the reaction
enthalpy.
c
Rate constant taken equal to that of the similar reaction in the case of benzene [11].
Rate constant taken equal to that of the H-abstraction with methyl radicals proposed by Colket et al. [35] with A divided by 10 for cyclic
radicals and with an activation energy 4 kcal mol higher for resonance stabilized radicals.
d
−1
e
Rate constant taken equal to that of the H-abstraction with phenyl radicals proposed by Heckmann et al. [36].
2
radicals proposed by Baulch et al. [27].
This rate constant is that proposed by Hippler et al. [38] with the A-factor divided by 4.
f
Rate constant taken equal to that of the H-abstraction with ·HO
g
h
i
Rate constant taken equal to that of the similar reaction in the case of phenyl radicals [11].
Rate constant taken equal to that of the recombination of ·H atoms with alkyl radicals as proposed by Allara et al. [47].
Rate constant taken equal to that of the similar reaction in the case of phenyl radicals by Rao et al. [45].
The kinetic data for this isomerization of a peroxy radical have been calculated state according to the method proposed by Warth et al. [48]
j
k
for peroxyalkyl radicals (see text).
l
Rate constant taken equal to that the disproportionation of peroxyalkyl radicals as proposed by Warth et al. [48].
For these beta-scissions involving the breaking of a C C or a C H bond, A-factor is an average value for beta-scissions [49] and activation
m
energies have been estimated from thermochemistry and to obtain the best results for simulations.
n
Rate constant taken equal to that of the similar reaction in the case of ethoxy radicals as proposed by Baulch et al. [39].
The global rate constant has been taken equal to that the decomposition of phenoxy radicals to give carbon monoxide and cyclopentadienyl
o
radicals, as proposed by Franck et al. [50].
p
Rate constant taken equal to that of the similar reaction in the case of toluene.
q
Rate constant estimated from the parameters proposed for acetaldehyde by Warnatz [51] (for ·H atoms), by Baulch et al. [26] (for ·HO
2
and
−1
·
CH
3
radicals), and by Holhein et al. [52] (for ·C
2 5
H radicals) with an activation energy 1 kcal mol lower due to the resonance stabilization of
the obtained radical.
r
Rate constant taken equal to that of the H-abstraction with ethyl radicals with A divided by 10 for cyclic radicals and with an activation
−
1
energy 4 kcal mol higher for resonance-stabilized radicals.
s
Rate constant taken equal to that the decomposition of hydroperoxide molecules as proposed by Sahetchian et al. [53].
Rate constant taken equal to that of the similar reaction in the case of phenol [11].
Rate constant taken equal to that of the similar reaction for benzyl radicals.
Rate constant taken equal to that of the combination calculated using software KINGAS [54] based on thermochemical kinetics methods
t
u
v
[
31]. Activation energies of recombination are taken to be zero and A factors are estimated from modified collision theory [48].
w
Rate constant taken equal to that of the similar reaction for benzyl peroxy radicals.
Rate constant taken equal to that of the similar reaction for benzyl alcoxy radicals.
Rate constant taken equal to that of the similar reaction for benzaldehyde.
x
y
z
Rate constant taken equal to that of the similar reaction for ·C
6
H
5
CO radicals.
aa
Rate constant taken equal to that of the similar reaction in the case of toluene with A multiplied by 2/3 (for benzyl alcohol and ethyl
−1
benzene) and by 4/3 (for bibenzyl) to take into account the number abstractable H atoms and with an activation energy 2 kcal mol lower due
OH substitution (for benzyl alcohol) or to the abstraction of secondary H atoms instead of primary ones (for ethyl benzene and bibenzyl).
bb
Rate constant taken equal to that of the similar reaction in the case of ·C
2 5
H O radicals proposed by Heicklen et al. [55].
cc
dd
ee
f f
Rate calculated using software KINGAS [54].
Rate constant taken equal to that the bimolecular initiation of alkanes as proposed by Warth et al. [48].
Rate constant taken equal to that the abstraction of three primary alkylic H atoms as proposed by Warth et al. [48].
Rate constant taken equal to that the beta-scission of an alkenyl radicals involving the abstraction of an allylic H-atoms [49].
gg
Rate constant taken equal to that the oxidation of allylic radicals (for ·C
6 5 3
H CHCH radicals) as proposed by Touchard et al. [56] or of
alkylic radicals (for ·C
6 5 2 2
H CH CH radicals) as proposed by Warth et al. [48].
hh
Rate constant taken equal to that of the similar reaction in the case of 1,3-butadiene [24] with A divided by 2 to take into account the
number of double bonds.
ii
Rate constant taken equal to that of the similar reaction in the case of ·C
2 3
H radicals proposed by Mebel et al. [57].

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
39
internal rotations as the reactant moves to the transi-
tion state. The activation energy has been set equal to
the sum of the activation energy for the abstraction of
an allylic secondary hydrogen atom by a peroxy radical
(83–92 and 93–108, respectively). For the formation of
HOC6H4CH2·radicals, therateparametersarededuced
from those used from toluene and for the formation of
cresoxy radicals from those used for phenol. The reac-
tions deriving from HOC6H4CH2· radicals (109–123)
are deduced from the major steps deriving from benzyl
radicals.
Apart from the additions of H-atoms, which lead
to ·CH2OH radicals and benzene (126) and the bi-
molecular initiation (125) leading to benzyl alcoxy
and ·HO2 radicals, the reactions of benzyl alcohol
molecules involve the formation of the resonance-
stabilized ·C6H5CHOH radicals by bimolecular ini-
tiation (124) and by metatheses with H-abstraction
(127–143). ·C6H5CHOH radicals can decompose to
hydrogen atoms and benzaldehyde (144).
In the case of ethylbenzene molecules, the addi-
tions of H-atoms lead to ·C2H5 radicals and benzene
(149). Unimolecular (146) and bimolecular initia-
tion (147) and metatheses with H-abstraction (150–
162) involve the formation of the resonance-stabilized
·C6H5CHCH3 radicals. We have also considered the
formation of ·C6H5CH2CH2 radicals by bimolecular
initiation (148) and by metatheses with H-abstraction
(163–167). Both radicals lead to the formation of
styrene, by decompositions involving the breaking of
a C H bond (168–169) or by reaction with oxygen
molecules (171,172) or combine with ·HO2 radicals
(173,174) to give hydroperoxide molecules, which
quickly decompose. ·C6H5CH2CH2 radicals can also
decompose to give ethylene and phenyl radicals (170).
In the case of styrene molecules, the reactions have
been derived from those proposed for 1,3-butadiene
−1
(
14.5 kcal mol [49]) and the strain energy of a four-
−1
membered cyclic transition state (23 kcal mol [31]).
Linstedt and Maurice [21] have also considered this re-
13
3
−1 −1
s and
action with an A-factor of 1 × 10 cm mol
−1
an activation energy of 28.7 kcal mol . We have also
considered the termination with H-atoms (49) and the
disproportionations of these radicals with ·HO2 radi-
cals and with themselves (50–52).
Alcoxy benzyl radicals can react with oxygen to
give benzaldehyde and ·HO2 radicals (55) or by beta-
scission. Two beta-scissions are possible, the first in-
volving the breaking of a C H bond (53) with an
−1
enthalpy of reaction of 15.9 kcal mol , the second
involving the breaking of a C C bond (54) with an
−1
enthalpy of reaction of 23.6 kcal mol . We have used
the same activation energy for both channels, since we
found the simulations to be much deteriorated if we
allowed the activation energies to be different.
Like phenoxy radicals [11], cresoxy radicals can
react by CO elimination (56–58) or combine with H-
atoms (59). CO elimination (56–58) would lead to the
formation of ·C5H4CH3 radicals, but Emdee et al. [2]
proposed a fast rearrangement of this species to give
·
H atoms and benzene (56). We propose here that beta-
scissions from this radical can also lead to the opening
of the ring and to the formation of small linear unsatu-
rated species (57,58).
Secondary Mechanism
[24] and include a bimolecular initiation (175), addi-
Apart from the additions of H-atoms, which lead to
tions of ·O· atoms (176) and ·OH radicals (177) to the
double bond, and metatheses with H-abstraction in-
volving the formation of ·C6H5CHCH2 radicals (178,
181), which can react with oxygen molecules to give
formaldehyde and benzoyl radicals (182).
·
CHO radicals and benzene (62), the reactions of
benzaldehyde molecules involve the formation of the
resonance-stabilized benzoyl radicals by unimolecular
(60) and bimolecular (61) initiations and by metatheses
with H-abstraction (63–77). Benzoyl radicals decom-
pose to produce phenyl radicals and carbon monoxide
In the case of bibenzyl molecules, we have only
considered the formation of the resonance-stabilized
·C6H5CHCH2C6H5 radicals by unimolecular (183) and
bimolecular (184) initiation and by metatheses with
H-abstraction (185–192). ·C6H5CHCH2C6H5 radicals
can decompose to hydrogen atoms and stilbene (193).
The formation of stilbene from the recombination of
benzylradicalshasbeenproposedbyHippleretal. [32].
(78).
Benzyl hydroperoxide molecules can easily decom-
pose by rupture of the O OH bond (79), and a rate con-
stant can be estimated by analogy with that proposed
for alkyl hydroperoxide molecules [53]. This reaction
leads to the formation of ·OH and alcoxy benzyl radi-
cals.
Apart from the additions of H-atoms, which lead
to methyl radicals and phenol (82), the reactions
of cresol molecules involve the formation of two
resonance-stabilized radicals, ·OC6H4CH3 (cresoxy)
and HOC6H4CH2·, by bimolecular initiations (80 and
COMPARISON BETWEEN COMPUTED
AND EXPERIMENTAL RESULTS
Simulations were performed using the CHEMKIN
II software [23]. We have tried to reproduce the
8
1, respectively) and by metatheses with H-abstraction

40
BOUNACEUR ET AL.
experimental data described in this paper (jet-stirred
reactor and shock tube), but to extend the validity of
the proposed mechanism, we have also attempted to
model results of the literature obtained in other shock
tubes in a wider pressure range (from 2.4 to above 600
bar) [14–16] and in a flow tube [12,13]. The mecha-
nism presented here before has been used for all the
different performed simulations.
the important part of the curve for the determination of
ignition delays. Both experimentally and theoretically,
the ignition delay is determined at 10% of the maxi-
mum OH peak. Modeled results reproduce the system-
atic trends of the measurements, including variations
in ignition delay times with temperature, equivalence
ratio, and concentration of toluene. While the global
agreement is better than that obtained with the mecha-
nism of Pitz et al. [22], it deteriorates when temperature
increases, for the most concentrated mixtures, 1.25%
of toluene and ꢀ equal to 0.5 or to 1. In these cases,
computed times can be up to three times longer than
the experimental values. This could be due to the fact
that the dilution of the mixtures (76.25% of Ar for the
most concentrated mixture) is low enough for non-ideal
processes starting to occur. That could explain why the
agreement is correct in stoichiometric mixtures at 0.5
of toluene and not at 1.25% of toluene as shown in
Fig. 4b.
Ignition delay times behind reflected shock waves
have been also measured by Burcat et al. [16] for pres-
sures between 2.4 and 7 bar and by Pengloan et al.
[14,17] at 1.1 bar. Figures 4b and 5a show that our
mechanism allows us to reproduce the results obtained
by Burcat et al. [16] very accurately, whatever the
pressure. The agreement in the case of the results of
Pengloan et al. [14,17] deteriorates slightly, but is still
acceptable.
Sivaramakrishnan et al. [15] have measured the con-
sumption of the organic reactant and the formation of
carbon monoxide, benzene, and acetylene during the
oxidation of toluene behind reflected shock wave in a
single pulse shock tube at a pressure in excess of 600
bar and in a temperature range from 1250 to 1450 K.
Figure 6 shows that our mechanism is able to correctly
reproduce these results, apart from the formation of
carbon monoxide and acetylene at the highest temper-
atures. As Sivaramakrishnan et al. [15] consider that
the residence time in the reaction zone was between
1.5 ± 0.1 ms, we have used a value of 1.5 ms in our
simulations.
Perfectly Stirred Reactor
Figures 1 to 3 display comparisons between the ex-
perimental (symbols) and computed (lines) results for
some of the experiments performed in a jet-stirred re-
actor. Figures 1 and 2 show that our mechanism leads to
an acceptable agreement for the conversion of toluene
and the formation of the two major primary prod-
ucts, benzene and benzaldehyde, and for that of car-
bon monoxide, under all the conditions studied, apart
from that the modeled increase in the concentration of
benzene is steeper than the experimental one for an
equivalence ratio of 0.9 and the formation of carbon
monoxide is underestimated by a factor 2 at 893 K and
for an equivalence ratio of 0.45. As an example of the
agreement obtained, Figs. 2 and 3 present the simulated
profiles of oxygen and products at 893 K and for an
equivalence ratio of 0.45; the same kind of agreement
can be obtained in other conditions. Conversion of oxy-
gen and the formations of minor primary products, phe-
nol, benzyl alcohol, and ethylbenzene are also correctly
reproduced. Bibenzyl is systematically overestimated,
irrespective of residence time, which is probably in-
dicative of a condensation problem for this compound
before it enters the trap. As regards the nonprimary
products, the formation of styrene, allene, carbon diox-
ide, and methane are satisfactorily reproduced, while
the production of C2 compounds is underpredicted at
short residence times and correctly modeled at long res-
idence times. At 893 K, at an equivalence ratio of 0.45
and a residence time of 4 s, C2 compounds contain 52%
acetylene and 48% ethylene. The profiles of propene
and butadiene are seriously underpredicted whatever
the residence time.
Flow Reactor
Brezinsky et al. [12] have studied the oxidation of
toluene in a flow reactor at 1180 K, at atmospheric
pressure, with nitrogen as bath gas, for a concentra-
tion of toluene of 1400 ppm and for two equivalence
ratios, 0.63 and 1.4. Figures 7 and 8 display compar-
isons between the experimental (symbols) and com-
puted (lines) results at ꢀ equal to 0.63 and 1.4, re-
spectively, and show that a globally correct agreement
can be observed in both cases. The model can satis-
factorily reproduce the consumption of toluene and the
Shock Tubes
Computed results for our shock tube appear in Fig. 4.
It is worth noting that the experimental OH emission
at 306 nm is related to the electronically exited OH
population and is not directly proportional to OH radi-
calconcentration. Nevertheless, previous work [24] has
shown a correct agreement between the shapes of the
profiles of experimental emission and calculated OH
concentration during the rise of the signal, which is

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
41
formation of benzene, benzaldehyde, ethyl benzene,
styrene, cresols ethylene, methane, and carbon monox-
ide. The production of phenol is well simulated at ꢀ
equal to 1.4, while it is strongly overestimated for the
longest residence times at ꢀ equal to 0.63. The for-
mation of acetylene is correctly simulated at ꢀ equal
to 0.63, while it is overestimated by a factor around 2
at ꢀ equal to 1.4. The production of benzyl alcohol is
overestimated by a factor around 2 in both cases.
Our model leads to a global agreement similar to
that obtained by Lindstedt and Maurice [21]; but it is
worth noting that unlike these authors, we have not
found it necessary to introduce a time shift, just as in
our previous modeling of the oxidation of benzene [11]
in the same apparatus.
ANALYSIS OF THE MECHANISM
AND DISCUSSION
The fluxes and sensitivity analyses presented hereafter
have been performed with the mechanism described
previously and the accuracy of conclusions derived
from them can only reflect the accuracy of the proposed
reaction channels and of the used thermodynamic and
kinetic data.
Figures 9a and 10 present sensitivity and fluxes anal-
ysis computed in a jet-stirred reactor at 893 K, Figs. 9b
and 11 present sensitivity and fluxes analysis computed
in a flow reactor at 1180 K, and Figs. 9c and 12 present
sensitivity and fluxes analysis computed in a shock
tube. As already shown in previous studies [11,24,25],
Fig. 9c demonstrates the determinant role for ignition
delays of the branching reaction between ·H atoms and
oxygen molecules.
Figure 5 Autoignition delays of toluene in a shock tube at
.1 and 2.4 bar. Semilog plot of experimental (symbols) and
computed (lines) ignition delays as a function of temperature
behind the reflected shock wave for the experimental results
obtained by Burcat et al. [16] at 2.4 bar for different mixtures
1
(a) and those of Pengloan et al. [14,17] at 1.1 bar, at 0.4% of
toluene and different equivalence ratios (b).
Main Reaction Pathways
of Toluene Molecules
In all cases, toluene reacts by four main channels,
metatheses by H· and ·O· atoms, OH·, methyl, HO2·,
phenoxy, and cresoxy radicals to form benzyl radicals
(7–11, 21, 23 (reaction numbers refer to Table V)),
metatheses by OH· radicals to form methylphenyl rad-
icals (27), addition of ·O· atoms to give cresoxy radi-
cals and H· atoms (5), and addition of H· atoms to form
benzene and methyl radicals (4). In a perfectly stirred
reactor at 893 K, metatheses giving benzyl radicals ac-
count for 61% of the toluene consumption, while 18%
is due to the addition of ·O· atoms, 14% to metatheses
giving methylphenyl radicals, and 7% to the addition
of H· atoms. This last channel is favored when equiva-
lence ratio or temperature increases, while the ratio be-
tween the three other channels does not depend much
on temperature or equivalence ratio. In a shock tube
Figure 6 Oxidation of toluene in a shock tube at 613 bar and
ꢀ
= 5. Comparison between experimental (symbols) and
computed (lines) concentration of species versus tempera-
ture at an estimated residence time of 1.5 ms and an initial
concentration of toluene of 14 ppm [15].

42
BOUNACEUR ET AL.
Figure 7 Oxidation of toluene in a flow reactor at 1180 K and at ꢀ = 0.63 [12]. Comparison between experimental (symbols)
and computed (lines) mole fraction of species versus residence time.
at 1800 K, 30% of toluene also decomposes to phenyl
and methyl radicals (2).
6% of the consumption of toluene. The rate constant
that we use for this reaction allows us considerably
to improve our simulations in these conditions, both
for the conversion of toluene and for the formation of
products deriving from methyl radicals, methane, and
ethylbenzene.
At 893 K, the main ways of production of propene
and 1,3-butadiene, which are strongly underestimated
in our simulations, are the metatheses of allyl and
i-C4H5 radicals with toluene. The formation of
these radicals derives from the decomposition of
C5 cyclic compounds, such as cyclopentadienyl
radicals or cyclopentadienone, which are still not well
understood.
Figure 9 shows that, in all cases, the metatheses
leading to the formation of the resonance-stabilized
benzylradicalshaveastronginhibitingeffect, whilethe
metathesis involving to the production of methylphenyl
radicals has a promoting effect, as these radicals mainly
react through branching step with oxygen molecules to
give ·O· atoms and cresoxy radicals. The important
sensitivity coefficient of these reactions shows that it is
important to take them into account in the mechanism.
In a flow tube at ꢀ = 1.4, the decomposition of
benzylradicalstophenylandmethylradicalshasanim-
portant promoting effect, while it contributes for only

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
43
Figure 8 Oxidation of toluene in a flow reactor at 1190 K and at ꢀ = 1.4 [12]. Comparison between experimental (symbols)
and computed (lines) mole fraction of species versus residence time.
Reactions Deriving from the Formation
of Benzyl Radicals
radicals (79), which decompose to produce benzalde-
hyde and H· atoms (53) or formaldehyde and phenyl
radicals (54). As it enables the very stable benzyl
radicals to be transformed into the reactive H· atoms,
In all cases, benzyl radicals react mainly by termi-
nation steps with ·O· atoms and with OH·, methyl,
and HO2· radicals and with themselves (34–39). In
a perfectly stirred reactor, the major recombination
is that with HO2· radicals producing benzylhydroper-
oxide molecules, which rapidly yield alcoxybenzyl
the recombination of benzyl and HO · radicals is
2
the reaction with the strongest accelerating effect at
893 K.
The reactions of phenyl and derived radicals are
not presented in great detail in Figs. 10–12, because

44
BOUNACEUR ET AL.
Figure 9 Sensitivity analyses related to (a) the conversion of toluene in a jet-stirred reactor, at 893 K and a residence time of
s, (b) the mole fraction of carbon monoxide in a flow reactor [12] at 1180 K and a residence time of 0.02 s, and (c) the mole
4
fraction of hydroxyl radicals in a shock tube at ꢀ = 1, 8.7 bar and 0.5% of toluene and a residence time corresponding to 50%
conversion of toluene (for the clarity of the figure, only reactions, with an absolute value of the sensitivity coefficient above a
given threshold depending on the studied reactor, are shown).

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
45
Figure 10 Oxidation of toluene in a jet-stirred reactor at 893 K. Flux analysis at ꢀ = 0.5 and a residence time of 4 s.

46
BOUNACEUR ET AL.
Figure 11 Oxidation of toluene in flow reactor at 1180 K. Flux analysis at ꢀ = 0.63 and a residence time of 0.02 s.

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
47
Figure 12 Oxidation of toluene in a shock tube at 1800 K. Flux analysis at ꢀ = 1, 8.7 bar and 0.5% of toluene and a residence
time corresponding to 50% conversion of toluene.
they are similar to what was observed during our pre-
vious study of the oxidation of benzene [11]. They react
mainly with oxygen molecules to give benzoquinone,
which yields cyclopentadienone by CO elimination or
phenoxy radicals, which lead to phenol at 893 K and
to cyclopentadienyl radicals by CO elimination at 1180
and 1800 K. As the production of benzoquinone occurs
mainlythroughtheformationofperoxyphenylradicals,
it decreases with temperature. At 1180 K, cyclopenta-
dienyl radicals react mainly with oxygen molecules to
produceC4H4Oandcarbonmonoxide, whileat1800K,
it yields 1,3-butadiene and carbon monoxide.
From 1180 K, the major recombination is that with
·O· atoms yielding benzaldehyde and H· atoms (35).
Figure 9 shows that the rate constant of this reaction is a
sensitive parameter under the experimental conditions
of the flow reactor. The recombination giving phenyl
radicals and formaldehyde is less important (36).
Benzaldehyde reacts mainly by metatheses with OH·
radicals to give benzoyl radicals (65), which decom-
pose to produce phenyl radicals and carbon monox-
ide (78).
The recombination of benzyl radicals giving biben-
zyl is important at 893 K, but decreases with

48
BOUNACEUR ET AL.
temperature, while the recombination of benzyl rad-
icals with methyl radicals to form ethylbenzyl is opti-
mum at 1180 K and that with OH· radicals to produce
benzyl alcohol is favored at 1800 K. Ethylbenzene re-
acts mainly by metatheses with H· atoms and OH· rad-
icals to form the resonance-stabilized ·C6H5CHCH3
radicals (150, 152), which decompose to give styrene
and H· atoms (168). At 1180 K, ·C6H5CH2CH2 radi-
cals are formed (163, 164) and lead to the production
of styrene (169) and of ethylene and phenyl radicals
radicals (58). The importance of these reactions, which
are responsible for most of the formation of acetylene
and allene at 893 K, increases with temperature.
CONCLUSION
This paper presents new experimental results for the
oxidation of toluene in a jet-stirred reactor at tempera-
tures from 873 to 923 K and in a shock tube between
(170).
1305 and 1795 K and proposes a detailed mechanism
At 893 K, 2% of benzyl radicals react with oxygen
able to reproduce these data, but also previously pub-
lished experiments in others shock tubes [14–17] and
in a flow reactor [12,13].
The reactions of importance in this mechanism have
been determined by using flux and sensitivity analyses.
Amongst them, the reactions of cresols, styrene, and
related radicals are yet very uncertain and should be
subjectofstudiesforspecialistsofelementaryreactions
to progress toward a better modeling of the oxidation
of aromatic compounds.
molecules to give peroxybenzyl radicals (32), which
isomerize and decompose to give benzaldehyde and H·
atoms (48). At 1800 K, 13% of benzyl radicals decom-
pose to give cyclopentadienyl radicals and acetylene
(30).
Reactions Deriving from the Formation
of Cresoxy Radicals
At 893 K, resonance-stabilized cresoxy radicals react
mainly by disproportionation with ·HO2 radicals (80)
or by H-abstraction with phenol to form cresols (92).
The reaction of cresoxy and ·HO2 radicals is the re-
action with the strongest inhibiting effect at 893 K.
According to our simulations, cresols should have been
important primary products under our experimental
conditions; the fact that we have not observed them
could be due to condensation problems for these com-
pounds before the entrance of the trap, as in the case of
bibenzyl. It is also worth noting that the rate constants
of the reactions involved in the formation and the con-
sumption of cresols are mostly estimated and that, at
this low temperature, there are large uncertainties con-
cerning the reactions forming or consuming ·O· atoms.
At higher temperatures, cresols are mainly formed
by recombination between H· atoms and cresoxy
radicals
Despite the uncertainties in the chemistry of some
species, themechanismproposedhereconstitutesa first
basis in the further development of models for the ox-
idation of other aromatic compounds, such as xylenes
or methylnaphthalenes. It could also be included in a
wider model to reproduce the oxidation and autoigni-
tion of surrogate fuel mixtures (e.g. n-heptane–toluene
mixtures), but that should be done carefully at low tem-
perature as this model has not been validated below
893 K.
BIBLIOGRAPHY
1
. Guibet, J. C. In Fuels & Engines; Publications de
l’Institut Fran c¸ ais du P e´ trole, Editions Technip, 1999;
Vols. 1, 2.
2. Emdee, J. L; Brezinsky, K.; Glassman, I. J Phys Chem
The abstractions of the hydroxylic H-atom by ·O·
1
992, 96, 2151–2161.
(
84) atoms and OH· (85) and ·HO2, at 893 K, radi-
3. Lindstedt, R. P.; Skevis, G. Combust Flame 1994, 99,
cals (86) to give back cresoxy radical are important
reactions of cresols accounting for 75% of their con-
sumption at 893 K. This is due to the fact that cresoxy
radicals are resonance stabilized. Cresols can also re-
act by abstraction of the benzylic H-atoms by H· (93)
and ·O· (94) atoms and OH· (95) radicals to form hy-
droxybenzyl radicals, which can recombine with ·HO2
radicals (112) and ultimately lead to the production
of hydroxybenzaldehyde or carbon monoxide and cy-
clopentadienone.
551–561.
4. Zhang, H. Y.; Mc Kinnon, J. T. Combust Sci Technol
1995, 107, 261.
5
6
7
8
9
. Tan, Y.; Frank, P. Symp Int Combust Proc 1996, 26,
6
77–684.
. Shandross, R. A.; Longwell, J. P.; Howard, J. B. Proc
Combust Inst 1996, 26, 711–719.
. Richter, H.; Grieco, W. J.; Howard, J. Combust Flame
1
999, 119, 1–22.
. Alzueta, M. U.; Glarborg, P.; Dam-Johansen, K. Int J
Chem Kin 2000, 32, 498–522.
. Ristori, A.; Dagaut, P.; El Bakali, A.; Pengloan, G.;
Cathonnet, M. Combust Sci Technol 2001, 167, 223–
256.
Cresoxy radicals can also react by CO elimination
to give benzene and H· atoms (56), acetylene, viny-
lacetylene, and H· atoms (57) or allene and propargyl

EXPERIMENTAL AND MODELING STUDY OF THE OXIDATION OF TOLUENE
49
1
0. Dupont, L.; El Bakali, A.; Pauwels, J. F.; Da Costa,
I.; Meunier, P.; Richter, H. Combust Flame 2003 135,
35. Colket, M. B.; Seery, D. J. Proc Combust Inst 1994, 25,
883–891.
1
71–183.
36. Heckmann, E.; Hippler, H.; Troe, J. Proc Combust Inst
1996, 26, 543–550.
1
1
1
1
1
1. Da Costa, I.; Fournet, R.; Billaud, F.; Battin-Leclerc, F.
Int J Chem Kin 2003, 35, 503–524.
2. Brezinsky, K.; Litzinger, T. A.; Glassman, I. Int J Chem
Kin 1984, 16, 1053–1074.
3. Klotz, S. D.; Brezinsky, K.; Glassman, I. Proc Combust
Inst 1998, 27, 337–344.
4. Dagaut, P.; Pengloan, G.; Ristori, A. Phys Chem Chem
Phys 2002, 4, 1846–1854.
5. Sivaramakrishnan, R.; Tranter, R. S.; Brezinsky, K.;
Durgam, S.; Vasudevan, H. In Proceedings of the 3rd
Joint Meeting of the U.S. Sections of the Combustion
Institute, 2003.
37. Fenter, F. F.; Nozi e` re, B.; Caralp, B.; Lesclaux, R. Int J
Chem Kin 1994, 26, 171–189.
38. Hippler, H.; Reihs, C.; Troe, J. Proc Combust Inst 1990,
23, 37–43.
39. Baulch, D. L.; Cobos, C. J.; Cox, R. A.; Esser, C.;
Franck, P.; Just, Th.; Kerr, J. A.; Pilling, M. J.; Troe,
J.; Walker, R. W.; Warnatz J. J Phys Ref Data 1992, 21,
411–429.
40. M u¨ ller-Markgraf, W.; Troe, J. J Phys Chem 1988, 92,
4899.
41. Miller, R.; Melius, C. F. Combust Flame 1992, 91, 21–
39.
1
6. Burcat, A.; Snyder, C.; Brabbs, T. NASA TM-87312,
1
986.
42. Grela, M. A.; Collussi, A. J. J Phys Chem 1986, 90,
434–437.
43. Solly, R. K.; Benson, S. W. J Am Chem Soc 1971, 23,
2127–2131.
1
1
7. Pengloan, G. Ph.D. Thesis, Universit e´ d’Orl e´ ans, 2001.
8. Roubaud, A.; Minetti, R.; Sochet, L. R. Combust
Flame 2000, 123, 535–541.
1
2
2
2
9. Hamins, A.; Seshadri, K. Combust Flame 1987, 68,
44. Ebert, K. H.; Ederer, H. J.; Schmidt, P. S. In ACS Sym-
posium Series 65, 1978.
45. Rao, S. V.; Skinner, G. B. J Phys Chem 1989, 93, 1864–
1869.
46. Ingham, T.; Walker, R. W.; Woolford, R. E. Proc Com-
bust Inst 1994, 25, 767–774.
47. Allara, D. L.; Shaw, R. J Phys Chem Ref Data 1980, 9,
523–559.
2
95–307.
0. Davis, S. G.; Wang, H.; Brezinsky, K.; Law, C. K. Proc
Combust Inst 1996, 26, 1025–1033.
1. Lindstedt, R. P.; Maurice, L. Q. Combust Sci Technol
1
996, 120, 119–167.
2. Pitz, W. J.; Seiser, R.; Bozzelli, J. W.; I. Da Costa, I.;
Fournet, R.; Billaud, F.; Battin-Leclerc; Seshadri, K.;
Westbrook, C. K. In Proceedings of the 2nd Joint Meet-
ing of the U.S. Sections of the Combustion Institute,
48. Warth, V.; Stef, N.; Glaude, P. A; Battin-Leclerc, F.;
Scacchi, G.; C oˆ me, G. M. Combust Flame 1998, 114,
81–102.
2
001.
2
2
2
2
2
3. Kee, R. J.; Rupley, F. M.; Miller, J. A. In Sandia
Laboratories Report, S 89-8009B, 1993.
4. Fournet, R.; Baug e´ , J. C.; Battin-Leclerc, F. Int J Chem
Kin 1999, 31, 361–379.
5. Belmekki, N.; Glaude, P. A.; I. Da Costa, I.; Fournet, R.;
Battin-Leclerc, F. Int J Chem Kin 2002, 34, 172–183.
6. Barb e´ , P.; Battin-Leclerc, F.; C oˆ me, G. M. J Chim Phys
49. Heyberger, B.; Belmekki, N.; Conraud, V.; Glaude, P. A;
Fournet, R.; Battin-Leclerc, F. Int J Chem Kin 2002, 34,
666–677.
50. Frank, P.; Herzler, J.; Just, Th.; Wahl, C. Proc Combust
Inst 1994, 25, 833.
51. Warnatz, J. In Combustion Chemistry; Gardiner, W. C.,
Jr. (Ed.); Springer-Verlag: Berlin 1984.
52. Hohlein, G.; Freeman, G. R. J Am Chem Soc 1970, 92,
6118–6125.
53. Sahetchian, K. A.; Rigny, R.; Tardieu De Maleissye, J.;
Batt, L.; Anwar Khan, M.; Mathews, S. Proc Combust
Inst 1992, 24, 637–643.
1
995, 92, 1666–1692.
7. Baulch, D. L.; Cobos, C. J.; Cox, R. A.; Franck, P.;
Hayman, G. D.; Just, Th.; Kerr, J. A.; Murrells, T.
P.; Pilling, M. J.; Troe, J.; Walker, R. W.; Warnatz, J.
Combust Flame 1994, 98, 59–79.
2
8. Tsang, W.; Hampson, R. F. J Phys Chem Ref Data 1986,
54. Bloch-Michel, V. Ph.D. Thesis, Institut National
Polytechnique de Lorraine, Nancy, 1995.
55. Heicklen, J. Adv Photochem 1988, 14, 177.
56. Touchard, S.; Fournet, R.; Glaude, P. A.; Warth, V.;
Battin-Leclerc, F.; Ribaucour, M.; Minetti, R., Proc
Combust Inst in press.
57. Mebel, A. M.; Diau, E. W. G.; Lin, M. C.; Morokuma,
K. J Am Chem Soc 1996, 118, 9759–9771.
58. Pamidimukkala, K. M.; Kern, R. D.; Patel, M. R.;
Wei, H. C.; Kiefer, J. H. J Phys Chem 1987, 91,
2148–2154.
1
5, 1087–1279.
2
3
9. Troe, J. Ber Bunsen-Ges Phys Chem 1974, 78, 478–485.
0. Muller, C.; Michel, V.; Scacchi, G.; C oˆ me, G. M. J Chim
Phys 1995, 92, 1154–1177.
3
3
3
3
1. Benson, S. W. In Thermochemical Kinetics, 2nd ed.;
Wiley: New York, 1976.
2. Hippler, H.; Reihs, C.; Troe, J. Z Phys Chem 1990, 167,
1
–16.
3. Tappe, M.; Schliephake, V.; Wagner, H. Gg. Z Phys
Chem 1989, 162, 129–145.
4. Hoffmann, A.; Klatt, M.; Wagner, H. Gg. Z Phys Chem
59. Ellis, C.; Scott, M.; Walker, R. W. Combust Flame
2003, 132, 291–304.
1
990, 168, 1–12.