H atom attack on propene

Article abstract of DOI:10.1021/jp111476q

The reaction of propene (CH₃CH=CH₂) with hydrogen atoms has been investigated in a heated single-pulsed shock tube at temperatures between 902 and 1200 K and pressures of 1.5-3.4 bar. Stable products from H atom addition and H abstraction have been identified and quantified by gas chromatography/flame ionization/mass spectrometry. The reaction for the H addition channel involving methyl displacement from propene has been determined relative to methyl displacement from 1,3,5-trimethylbenzene (135TMB), leading to a reaction rate, k(H + propene) → H₂C=CH₂ + CH ₃) = 4.8×10¹³ exp(-2081/T) cm³/(mol s). The rate constant for the abstraction of the allylic hydrogen atom is determined to be k(H + propene → CH₂CH=CH₂ + H₂) = 6.4 × 10¹³ exp(-4168/T) cm³/(mol s). The reaction of H + propene has also been directly studied relative to the reaction of H + propyne, and the relationship is found to be log[k(H + propyne → acetylene + CH₃)/k(H + propene → ethylene + CH₃)] = (-0.461 ± 0.041)(1000/T) + (0.44 ± 0.04). The results showed that the rate constant for the methyl displacement reaction with propene is a factor of 1.05 ± 0.1 larger than that for propyne near 1000 K. The present results are compared with relevant earlier data on related compounds. This article not subject to U.S. Copyright. Published 2011 by the American Chemical Society.

Full text of DOI:10.1021/jp111476q

ARTICLE
pubs.acs.org/JPCA
H Atom Attack on Propene
Claudette M. Rosado-Reyes,* Jeffrey A. Manion, and Wing Tsang
National Institute of Standards and Technology, 100 Bureau Drive, Gaithersburg, Maryland 20899, United States
S Supporting Information
b
ABSTRACT: The reaction of propene (CH3CHdCH2) with hydrogen atoms
has been investigated in a heated single-pulsed shock tube at temperatures
between 902 and 1200 K and pressures of 1.5ꢀ3.4 bar. Stable products from H
atom addition and H abstraction have been identified and quantified by gas
chromatography/flame ionization/mass spectrometry. The reaction for the H
addition channel involving methyl displacement from propene has been
determined relative to methyl displacement from 1,3,5-trimethylbenzene
(135TMB), leading to a reaction rate, k(H þ propene) f H2CdCH2 þ
CH3) = 4.8 ꢁ 1013 exp(ꢀ2081/T) cm3/(mol s). The rate constant for the
abstraction of the allylic hydrogen atom is determined to be k(H þ propene f
CH2CHdCH2 þ H2) = 6.4 ꢁ 1013 exp(ꢀ4168/T) cm3/(mol s). The reaction
of H þ propene has also been directly studied relative to the reaction of H þ
propyne, and the relationship is found to be log[k(H þ propyne f acetylene þ
CH3)/k(H þ propene f ethylene þ CH3)] = (ꢀ0.461 ( 0.041)(1000/T) þ
(0.44 ( 0.04). The results showed that the rate constant for the methyl displacement reaction with propene is a factor of 1.05 ( 0.1
larger than that for propyne near 1000 K. The present results are compared with relevant earlier data on related compounds.
’ INTRODUCTION
chemical processes. For H-atom attack on propene the reaction
channels in Scheme 1 should be considered.
Liquid fuels are comprised of unsaturated and saturated—
normal, branched, and cyclic—hydrocarbons. An alkyl radical is
formed when a fuel molecule decomposes upon radical attack or
thermal excitation. The alkyl radical can breakdown via β-bond
scission to produce another alkyl radical and a 1-alkene molecule.
Propene is the simplest substituted of the 1-alkenes. As the fuel
breakdown progresses via unimolecular reactions and oxidation
steps, H atoms are produced. They are among the most
important intermediates in hydrocarbon combustion.
This paper is concerned with the reaction of H atoms with
propene, studied under shock-tube conditions. It is an extension
of earlier measurements of the mechanism and rate constants for
the reaction of H atoms with other unsaturated hydrocarbons.
These reactions are of particular importance with respect to the
mechanisms associated with the formation of the initial aromatic
ring and subsequent soot formation. H-atom reaction with larger
unsaturated hydrocarbons leads to resonance stabilized radicals,
such as allyl, benzyl, and propargyl radicals. They are intermedi-
ates in the cyclization steps that initiate the production of
aromatic rings, as precursors of polycyclic aromatic hydrocar-
bons (PAHs) and soot. Under fuel-rich conditions, the concen-
trations of these radicals can build up in combustion systems.
The reactivity of H-atoms is such that their attack on any
hydrocarbon of even moderate complexity leads to multiple reaction
channels. In the case of unsaturated hydrocarbons (always
present in high temperature rich systems) there are contributions
from abstraction as well as addition channels. Their relative
importance dictates the reaction pathways for subsequent
Reaction 1 involves terminal H-addition to propene to
produce an isopropyl radical. Previous work of the unimolecular
decomposition of iso-C3H7 has shown that at high temperatures
relevant to combustion, reaction 1 is reversible, favoring back
decomposition into propene and H (log Kꢀ1 (1000 K) = ꢀ15.63).1
This channel is invisible to our experiments. Reaction 2 consists
of the addition of H atom to the central carbon atom with
subsequent displacement of the methyl moiety to produce
ethylene. This is actually a chemically activated decomposition.
The thermochemistry favors the methyl displacement process
overwhelmingly. Thus the measured rate constant refers to the
addition process.
H-atom abstraction reactions can take place in three different
ways, abstraction of allylic H (reaction 3), vinylic H (reaction 4),
and methylvinylic H (reaction 5). Reaction 3 is a source of allyl
radicals. A possible sink for allyl radicals in our experiments is
their recombination reaction to form 1,5-hexadiene and their
reaction with methyl radicals to form 1-butene. Reactions 4 and 5
produce 1- and 2-propenyl radicals. 1-Propenyl radicals decom-
pose into acetylene and methyl radical, while the decomposition
of 2-propenyl radicals favors the propyne þ H channel. These
channels are thermodynamically less favorable than the abstrac-
tion of an allylic H-atom.
Received: December 2, 2010
Revised:
February 18, 2011
Published: March 15, 2011
This article not subject to U.S. Copyright.
Published 2011 by the American Chemical Society
2727
dx.doi.org/10.1021/jp111476q J. Phys. Chem. A 2011, 115, 2727–2734
|

The Journal of Physical Chemistry A
ARTICLE
Scheme 1
In a recent study we have examined the analogous reaction
mechanism for H-atom reaction with propyne.2 This furnishes an
interesting basis for comparison, involving the effect of the degree
unsaturation on nonterminal addition and the consequence for
abstraction in going from a propargyl-H to an allylic-H.
residence times, leading to the possibility of studying primary
chemical processes, assuring the absence of surface reactions, and
minimizing contributions from chain reactions and secondary
chemistry. (Certain commercial materials and equipment are
identified in this paper in order to specify adequately the
experimental procedure. In no case does such identification
imply recommendation of endorsement by the National Institute
of Standard and Technology, nor does it imply that the material
or equipment is necessarily the best available for the purpose.)
Table 2 contains a listing of the mixtures used in these studies
along with the temperature and pressure conditions achieved
with each. Note particularly the large difference in concentration
between the radical source and the target molecules. This
guarantees that radicals will not attack radical source or reaction
products. Two sources are used to generate H atoms. One of
them is hexamethylethane (HME, (CH3)3CꢀC(CH3)3). Its
thermal decomposition takes place via, HME f 2 t-butyl f2
i-C4H8 þ 2H. The reaction is monitored by the disappearance of
HME and the formation of i-butene. The i-butene concentration
is a measure of the number of H atoms released into the system.
The extent of the reaction of HME decomposition is also used to
determine the temperature of the experiment by means of its rate
The available literature data on the reaction of H atoms with
propene and other relevant unsaturated hydrocarbons are com-
piled in Table 1. Arrhenius parameters have been measured for
the overall reaction at temperatures of 298ꢀ445 K, resulting in
the rate expression of 1.32 ꢁ 1013 exp(ꢀ785/T) cm3/(mol s),3
3
and at temperatures of 177ꢀ473 K leading to a rate expression of
6.13 ꢁ 1012 exp(ꢀ609/T) cm3/(mol s).4 These are mostly due
3
to terminal addition. Little work has been done at room
temperature.5 Estimates on the ratio of rate constants of non-
terminal versus terminal H atom addition have been made at
room temperature.6 The rate constant available for abstraction is
an estimate. Particularly important is the absence of direct
measurements of the individual rate constants for hydrogen
atom attack on propene. A direct measure of methyl displace-
ment and allyl formation may be of particular importance in
simulation studies due to the difference in reactivity of these
two radicals. The present study is carried out with this aim in
mind. It will also lead to a better basis of estimates of related
reaction processes.
expression k(HME f 2 t-butyl) = 1016.40 exp(ꢀ34400/T) sꢀ1 8
.
To cover lower temperature regions, experiments have been
carried out using t-butylperoxide (tBPO, (CH3)3CO-OC-
(CH3)3) in an excess of H2 (≈20%) to generate H atom,
(CH3)3CO-OC(CH3)3 f 2(CH3)3CO f 2CH3C(O)CH3 þ
2CH3, 2CH3 þ 2H2 f 2CH4 þ 2H. In this system, chlorocy-
clopentane is used as a temperature standard. The thermal
decomposition of chlorocyclopentane into cyclopentene and
hydrogen chloride is characterized by the rate expression
4.47 ꢁ 1013 exp(ꢀ24570/T) sꢀ1.9 Experimental pressures are
calculated from the temperature and mixture composition,
’ EXPERIMENTAL METHODS
Experiments were carried out in a heated single pulsed shock
tube. Details of the apparatus and general procedures have been
previously published.7 For the present purposes, it can be
regarded as a pulse reactor with heating time, derived from
recorded pressure traces, of 500 μs. The unique features of the
instrument are the generation of high temperatures and short
2728
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
Table 1. Summary of Relevant Data on the Reaction of Hydrogen Atoms with Alkenes and Alkynes
reaction
rate constant, cm3/(mol s)
T (K)
P (bar)
reference
3
Addition
1.33 ꢁ 1013 exp(ꢀ785/T)
6.13 ꢁ 1012 exp(ꢀ609/T)
5.3 ꢁ 1011
H þ propene f products
298ꢀ445
177ꢀ473
298
0.067
Harris et al. (1982)3
Kurylo et al. (1971)4
Munk et al. (1986)5
H þ propene f i-C3H7
H þ propene f n-C3H7
0.067
1.01 ꢁ 105
¥
133ꢀ1200
¥
1.32 ꢁ 1013 exp(ꢀ785/T)
5.70 ꢁ 109 T1.16 exp(ꢀ440/T)
9.33 ꢁ 1011
300ꢀ2500
177ꢀ910
298
Tsang, W. (1991, 1992)16,17,a
Seakins et al. (1993)1
Kerr et al. (1972)12,a
Tsang, W. (1991, 1992)16,17a
Kerr et al. (1972)12,a
this work
Lightfoot et al. (1987)19
Kerr et al. (1972)12,a
Kerr et al. (1972)12,a
Kyogoku et al. (1983)20
Kyogoku et al. (1983)20
Rosado-Reyes et al. (2010)2
1.32 ꢁ 1013 exp(ꢀ1641/T)
5.50 ꢁ 1010
500ꢀ2500
298
¥
¥
H þ propene f n-C3H7 f CH3 þ H2CdCH2
H þ H2CdCH2 f •C2H5
4.8 ꢁ 1013 exp(ꢀ2081/T)
2.64 ꢁ 1013 exp(ꢀ1087/T)
8.13 ꢁ 1011
922ꢀ1080
285ꢀ604
298
1.7ꢀ3.4
0.06ꢀ0.8
¥
H þ CH3CH2CHdCH2 f i-butyl
H þ CH3CH2CHdCH2 f n-butyl
5.01 ꢁ 1010
298
¥
H þ cis-CH3CHdCHCH3 f sec-C4H9
H þ trans-CH3CHdCHCH3 f sec-C4H9
H þ CH3CtCH f 1-propenyl f C2H2 þ CH3
1.61 ꢁ 1013 exp(ꢀ965/T)
2.39 ꢁ 1013 exp(ꢀ1060/T)
6.26 ꢁ 1013 exp(ꢀ2267/T)
200ꢀ500
200ꢀ500
889ꢀ1068
0.799
0.799
1.6ꢀ7.6
Abstraction
6.4 ꢁ 1013 exp(ꢀ4168/T)
H þ propene f products þ H2
H þ propene f allyl þ H2
H þ propene f 2-propenyl þ H2
H þ propene f 1-propenyl þ H2
a Data Evaluation.
1061ꢀ1200
500ꢀ2500
500ꢀ2500
500ꢀ2500
1.7ꢀ3.4
¥
¥
¥
this work
1.7 ꢁ 105 T2.5 exp(ꢀ1254/T)
7.8 ꢁ 105 T2.5 exp(ꢀ6182/T)
3.9 ꢁ 105 T2.5 exp(ꢀ2929/T)
Tsang, W. (1991, 1992)16,17,a
Tsang, W. (1991, 1992)16,17,a
Tsang, W. (1991, 1992)16,17,a
Table 2. List of Mixtures Used
mixture
componentsa
T (K)
P (bar)
A
B
C
4500 ppm propene, 4500 ppm 135TMB, 100 ppm HME, 75 ppm CCP
4066 ppm propene, 3996 ppm propyne, 100 ppm CCP, 50 ppm tBPO, 23% H2
3900 ppm propene, 195 ppm HME
922ꢀ1080
902ꢀ1036
1061ꢀ1200
1.6ꢀ2.3
1.5ꢀ1.9
2.4ꢀ3.4
a Key: CCP = chlorocyclopentane; 135TMB = 1,3,5-trimethylbenzene; tBPO = t-butylperoxide; HME = hexamethylethane (Note: 1% = 10000 ppm).
assuming ideal gas properties using the standard shock relations,
and no boundary layer or vibrational relaxation effects.
capillary) column. Once eluted, the detection and quantification
of the components were made by flame ionization and mass
spectrometry (Agilent Technologies 5973 inert mass selective
detector) detection. The GC oven temperature was programmed
from ꢀ60 to 180 ꢀC.
The rate constants for the reaction of H atoms were determined
here in comparison with that of H-atom attack on 1,3,5-trimethyl-
benzene (135TMB, 1,3,5-(CH3)3C6H3) to form m-xylene. The
process is, H þ 1,3,5-(CH3)3C6H3 f 1,3-(CH3)2C6H4 þ CH3.
The rate expression for this reaction has been previously
determined, k(950 to 1100 K) = 6.7 ꢁ 1013 exp(ꢀ3255/
Retention times and FID response factors of species under
analysis were determined from samples of pure compounds. Peak
areas were determined using HP ChemStation software and
converted to molar quantities based on the experimentally
determined response factors.
T) cm3/(mol s).10 An additional role of 135TMB is that of a
3
radical scavenger to minimize chain processes.
All hydrocarbon samples were from Aldrich Chemicals and
were used without purification. Only allene was found as a signi-
ficant impurity in propyne samples, which was subtracted from
the products formed from the course of the reaction. Reaction
mixtures were allowed to mix for at least 24 h prior to analysis.
Gas chromatography (Hewlett-Packard 6890N GC), with two
chromatographic columns, was used for the product analysis of
the reaction mixtures. Light organics (C1ꢀC4) were separated
on a 50 m, 0.53 mm i.d., 6.0 μm RT-Alumina PLOT column. An
advantage from using this column was the successful signal
separation of i-butene and 1-butene. This was difficult to do in
the past and was the reason that H þ propene had not yet been
studied. Heavier species were separated using a J&W Scientific
DB-1 30 m, 0.53 mm i.d., 5% phenylmethylsiloxane (fused silica
’ RESULTS
Product Distribution. The important products that were
observed in the after-shock propene mixtures are ethylene,
1-butene, and i-butene. i-Butene is the result of the decom-
position of HME, while the rest of the products come from
hydrogen atom attack on propene. Ethylene is the primary
product from H addition on propene. When allyl radical,
produced from the H allylic abstraction, recombines with
methyl radicals, 1-butene is obtained. Other products observed
are methane, ethane, acetylene, propyne, and 1,5-hexadiene.
Methane originates from methyl radicals produced from dis-
placement reactions, while ethane results from methyl radicals’
2729
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
Figure 1. Rate plot for methyl displacement by atomic hydrogen from propene relative to that of 1,3,5-trimethylbenzene.
recombination. Acetylene, propyne, and 1,5-hexadiene are
minor products and will not be treated in the present analysis.
1,5-Hexadiene is produced from the allyl þ allyl recombina-
tion, while acetylene and propyne can be formed from a variety
of processes. When 135TMB is added to the mixtures m-xylene is
found. When tBPO is added to the system, acetone, the large organic
decomposition product, is formed. In the presence of chlorocyclo-
pentane, cyclopentene is produced. The relative amounts of methane
and ethane that are formed are dependent on the nature of the
mixture studied. Another product observed in smaller concen-
trations is 1,3-dimethyl-5-ethylbenzene, resulting from methyl
recombination reactions with dimethylbenzyl, where the latter
species is formed by H abstraction in the H þ 135TMB reaction.
Nonterminal Addition and Methyl Displacement. A direct
measure of the relative rate of methyl displacement from propene
and 135TMB is obtained by comparing the yields of m-xylene
and ethylene, normalized for initial concentration of the target
molecules or
of methyl from 135TMB, for which kHþ135TMB = 6.7 ꢁ 1013exp-
(ꢀ3255/T) cm3/(mol s), we obtain
3
kðH þ propene f ethylene þ CH3Þ
¼ 4:8 ꢁ 1013 expð ꢀ 2081=TÞcm3=ðmol sÞ; ð922ꢀ1080 KÞ
3
The methyl displacement channels during the reactions of H
þ propene and H þ propyne are compared to each other in a
comparative relative rate experiment. The results are shown in
Figure 2, where the yields of acetylene are compared to those of
ethylene. For the reaction of methyl displacement by H atoms,
the rate constant for the reaction with propene is a factor of 1.05
( 0.1 larger than that for propyne, as determined from fitting the
experimental results
kðH þ propyne f acetylene þ CH3Þ
log
kðH þ propene f ethylene þ CH3Þ
1000
¼ ðꢀ0:461 ( 0:041Þ
þ ðꢀ0:44 ( 0:04Þ; ð902ꢀ1036 KÞ
kH þ propene=kH þ 135TMB
T
¼ ½ethyleneꢂ½135TMBꢂi=½m-xyleneꢂ½propeneꢂi
Abstraction of H Atoms. Information about the H atom
abstraction channels was obtained on the basis of being able to
count H atoms in experiments conducted without added in-
hibitor (mixture C). To recall, a direct measure of the H atoms
released in our experiments is the molar concentration of i-butene.
Assuming that all H atoms produced react with propene via
reactions 1ꢀ5(Scheme 1), the ethylene formed represents dis-
placement of methyl by H atoms and the difference between the
concentrations of i-butene and ethylene represents the H-atoms
that have disappeared from the system via abstraction reactions.
This leads to
Figure 1 shows an Arrhenius plot of kHþpropene/kHþ135TMB
,
the relative rate constant, for displacement of methyl radical from
propene and 1,3,5-trimethylbenzene. The following relative rate
expression was obtained
kðH þ propyne f ethylene þ CH3Þ
log
¼
kðH þ 135TMB f m-xylene þ CH3Þ
1
ð509:58 ( 68Þ ꢀ ð0:144 ( 0:067Þ; ð922ꢀ1080 KÞ
T
The uncertainties listed above are (σ. Experimental uncertain-
ties in comparative-rate single-pulse shock tube studies, mostly
systematic in nature, have been previously discussed.11 On the
basis of the experimental technique, we estimate that the relative
rate constants should be accurate to within (10%, while the esti-
mated uncertainty in the relative activation energies is approximately
5 kJ/mol. Relative to the standard reaction of displacement
ꢀ
ꢁ
½ethyleneꢂ
½i-buteneꢂ
kðabstractionÞ ¼ 1 ꢀ
kðdisplacementÞ
An Arrhenius plot of the rate constant determined on this basis
is shown in Figure 3; it corresponds to a composite value for
abstraction of hydrogen from propene by H atoms via reactions
2730
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
Figure 2. Temperature dependence of rate constant for methyl displacement by atomic hydrogen from propyne relative to that of propene.
Figure 3. Arrhenius plot of the rate constant for the abstraction of H atoms from propene by H atoms.
3, 4, and 5. From the previously determined rate expression for
nonterminal addition (above), the temperature-dependent rate
expression for hydrogen abstraction is determined to be
nonterminal addition to abstraction (Figure 4). An uncertainty of
2ꢀ3% for each concentration measurement is estimated. The data
show that H addition takes place ≈81% of the time during the H þ
propene reaction, while ≈19% corresponds to H abstraction.
kðH abstractionÞ ¼ 6:4 ꢁ 1013 expð ꢀ 4168=TÞcm3=ðmol sÞ;
3
ð1061ꢀ1200 KÞ
’ DISCUSSION
The possible contribution of propene pyrolysis to the H atom
reservoir in the experiments was checked by shocking propene in
the absence of hexamethylethane, in a mixture of 0.45% propene
and 0.0075% CCP. Ethylene was observed as a product, but its
yield was less than 0.01% of the initial propene concentration at
the highest temperature.
Of particular importance is the vast difference in concen-
trations between the radical source and the target molecule.
This guarantees that the radicals that are generated, in
particular the H-atoms, will only attack the target molecule.
The two radical products from H-atom attack are methyl and
allyl. Here again their concentration is much smaller than that
of the target molecule. It is also important to notice the high
reactivity of H-atoms toward propene and the large rates of
The yields of ethylene are plotted against those of i-butene
to obtain information about the branching ratio of H atom
2731
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
Figure 4. Branching ratio information; ethylene yield vs i-butene yield.
Figure 5. Comparison of relevant data pertaining to H atom addition reactions. (*Data evaluation.)
combination of the radicals that are formed. The consequence
is that H-atom attack on propene is the only reaction that can
occur. Note that the radicals formed cannot add to the
unsaturated compounds since under the high temperature
conditions the reactions are reversed. This is validated,
through carrying out reactions with a variety of substrates
and differences in substrate to target molecule concentrations,
by the invariance of the results. The general procedure has
been employed and similarly validated for studies on similar
systems (2, 10, 13, 14). Since all measurements are based on
internal standards, small discrepancies in concentrations are
significant in terms of uncertainty in the results.
The rate constant for nonterminal H addition on propene has
been determined relative to the analogous reaction on H þ
135TMB. The rate constant for the reaction of H þ propene f
ethylene þ CH3 is two times faster than the H þ 135TMB f m-
xylene þ CH3 reaction. On a per-methyl basis the results show
that H atoms displace methyl radicals from propene 6 times faster
2732
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
Figure 6. Comparison of reported rate constant expression of H-atom abstraction by atomic hydrogen on a per H-atom basis.
Table 3. Summary of Data Pertaining to Relevant H Abstraction Reactions
reaction
k, cm3/(mol s)
k (1000 K)
k per H atom (1000 K)
reference
3
H þ 1,3,5-TMB (9H)
H þ toluene (3H)
H þ isobutene (6H)
H þ propyne (3H)
H þ propane (6H)
H þ propene (3H)
3.7 ꢁ 1014 exp(ꢀ4342/T)
1.2 ꢁ 1014 exp(ꢀ4138/T)
1.7 ꢁ 1014 exp(ꢀ4024/T)
1.2 ꢁ 1014 exp(ꢀ4940/T)
2.5 ꢁ 1012(T/298 K)2.54 exp(ꢀ3400/T)
6.4 ꢁ 1013 exp(ꢀ4168/T)
4.81 ꢁ 1012
1.91 ꢁ 1012
3.04 ꢁ 1012
8.59 ꢁ 1011
1.81 ꢁ 1012
9.91 ꢁ 1011
5.35 ꢁ 1011
6.38 ꢁ 1011
5.07 ꢁ 1011
2.86 ꢁ 1011
3.01 ꢁ 1011
3.30 ꢁ 1011
Tsang et al. (1990)10
Robaugh et al. (1986)13
Tsang et al. (1988)14
Rosado-Reyes et al. (2010)2
Tsang (1988)15
this work
than from 135TMB. The ratio of available sites is however 6 or 3
to 1 in favor of the aromatic compound. It is clear that on a
numerical basis H-atom addition, at this temperature, to the
olefin may be as much as 12 or 6 times larger per site. The large
difference is attributed to the barrier arising from the breaking
of the resonance energy of the aromatic ring into a cyclohex-
adiene structure, indicative of the greater stability of aromatic
structures.
for nonterminal H atom addition to propene, when extrapolated
to room temperature, differs by ≈240% to the recommended
value of 5.5 ꢁ 1010 cm3/(mol s) of Kerr et al.12 It is interesting
3
that the differences between terminal and nonterminal addition
decreases as the temperature is increased. Thus differences in rate
constants are largely due to those in the activation energy.
Our overall rate constant for H abstraction, H þ propene f
products þ H2, is compared with other relevant abstraction rate
constants2,10,13ꢀ15 on a per H-atom basis in Figure 6. The data
are summarized in Table 3. Our determination of the H abstrac-
tion rate expression for the reaction of H þ propene of 4.9 ꢁ
1013 exp(ꢀ4033/T) is comparable to that of propyne within 15%
and to that of propane within 10% at 1000 K. Even more
surprising is the small difference in rate constants between
propane and the two unsaturated compounds. This is contra-
dictory for example to the lower resonance energy of propargyl
compared to that of allyl of about 12 kJ/mol. The resonance
energy is not manifested in the rate constant during abstraction,
in contrast to the breaking of a resonance stabilized bond. One
does note the apparently larger rate parameters for propane. This
larger activation energy is compensated by a larger A-factor,
which is expected. This recommended rate expression for H þ
propane is however based on an extrapolation from low tem-
perature measurements.
The relative rate studies between propene and propyne
showed that, for displacement of methyl radical, the reaction of
H þ propene is a factor of 1.05 faster than the reaction of H þ
propyne (Figure 2). This relative rate ratio is compared to the
same ratio taken from the individual rate expressions deter-
minations for both reactions, k(C3H6 þ H, this work) = 4.8 ꢁ
1013 exp(ꢀ2081/T) cm3/(mol s), and (C3H4 þ H, ref 2) =
3
6.3 ꢁ 1013 exp(ꢀ2267/T) cm3/(mol s). The comparison of
3
the individual determinations shows the contrary; that the
reaction of H þ propyne is faster by a factor of 1.08. The
discrepancy is approximately 13%. This is slightly larger than
the 5ꢀ10% discrepancy that we would have expected from our
analysis of the uncertainties. It may however be a more
realistic value.
Table 1 and Figure 5 provide comparisons of relevant rate
constants for H atom addition to the terminal and nonterminal
positions of C2 to C4 species. The existing data on low tempera-
ture terminal addition appear to be well established. Our experi-
mental rate constant determination of 4.8 ꢁ 1013 exp(ꢀ2081/T)
It is interesting that the ratio for methyl and allyl formation
from this study is a factor of 4 higher from the recommended
values by Tsang,16,17 estimated on the basis of existing data at
2733
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734

The Journal of Physical Chemistry A
ARTICLE
lower temperatures. This is due to errors (not unreasonable) of
factors of 2 in opposite directions and emphasizes the impor-
tance of direct determinations. The experimentally determined
ratio of rate expressions is
(5) Munk, J.; Pagsberg, P.; Ratajczak, E.; Sillesen, A. Chem. Phys. Lett.
1986, 132, 417.
(6) Falconer, W. E.; Sunder, W. A. Int. J. Chem. Kinet. 1971, 3, 395.
(7) Awan, I. A. M., W.S.; Tsang, W.; Manion, J. A. J. Phys. Chem. A
2010, 114, 7832.
(8) Tsang, W. J. Chem. Phys. 1966, 44, 4283.
kðH þ propene f n-propyl f CH3 þ C2H4Þ
kðH þ propene f H2 þ allylÞ
(9) Awan, I. A. T., W.; Manion, J. A. Standard Reactions for
Comparative Rate Studies: Experiments and Computations on the
Dehydrochlorination Reactions of 2-Chloropropane, Chlorocyclopen-
tane, and Chlorocyclohexane. Int. J. Chem. Kinet., in press.
(10) Tsang, W.; Cui, J. P.; Walker, J. A. In 17th International
Symposium on Shock Tubes and Waves; American Institute of Physics:
New York, 1990; Vol. 208, p 63.
¼ 0:75 expð2087=TÞ
This can be compared with the similar ratio of rate expressions
for toluene or
(11) Tsang, W. In Shock Waves in Chemistry, 1st ed.; Lifshitz, A., Ed.;
Marcel Dekker, Inc.: New York, 1981; p 59.
kðH þ toluene f methylcyclohexadienyl f CH3 þ benzeneÞ
kðH þ toluene f H2 þ benzylÞ
(12) Kerr, J. A.; Parsonage, M. J. Evaluated Kinetic Data on Gas Phase
Addition Reactions: Reactions of Atoms and Radicals with Alkenes, Alkynes,
and Aromatic Compounds; Butterworth & Co., Ltd.: London, 1972.
(13) Robaugh, D.; Tsang, W. J. Phys. Chem. 1986, 90, 4159.
(14) Tsang, W., Walker, James A. In Twenty-Second Symposium
(International) on Combustion; The Combustion Institute: 1988; p 1015.
(15) Tsang, W. J. Phys. Chem. Ref. Data 1988, 17, 887.
(16) Tsang, W. J. Phys. Chem. Ref. Data 1991, 20, 221.
(17) Tsang, W. Ind. Eng. Chem. Res. 1992, 31, 3.
¼ 0:100 expð1560=TÞ
Comparisons of these ratios demonstrate that the branching ratio
favors benzyl as opposed to allyl formation. This is indicative of
the greater inhibitory powers of toluene as compared to propene
through more efficient scavenging of H atoms.
During the experiments, yields of acetylene and propyne were
observed in small traces and did not provide means to reliably
determine the source of these compounds and experimental rate
expression for the abstraction of the vinylic and methylvinylic
hydrogens in propene. The acetylene yields in our experiments
led to a H-vinylic abstraction rate constant that is larger by a
factor of 6 (per H atom basis) than that estimated by Knyazev
et al.18 for the analogous reaction of ethylene þ H. A possible
mechanism involves H atom reaction with allyl radical leading to
H þ allyl f propene* f CH3 þ C2H3 f HCtCH þ H via a
chemically activated process. Due to the complication of the
sequence of reactions involving allyl radical, it is not possible to
quantitatively examine this possibility in our system.
ꢀ
(18) Knyazev, V. D.; Bencsura, A.; Stoliarov, S. I.; Slagle, I. R. J. Phys.
Chem. 1996, 100, 11346.
(19) Lightfoot, P. D. P.; Pilling, M. J. J. Phys. Chem. 1987, 91, 3373.
(20) Kyogoku, T. W., T.; Tsunashima, S.; Sato, S. Bull. Chem. Soc.
Jpn. 1983, 56, 19.
’ SUMMARY AND CONCLUSION
This study presents kinetic and mechanistic information on
the H-atom attack on propene. We are deriving the general rules
covering abstraction and displacement processes of reactions of
H atoms in high-temperature combustion systems. These studies
demonstrate little effect of resonance stabilization on the rate
constant for removing an H atom by an H atom. We have also
found the effect of aromatic versus olefinic and acetylenic
structures on the addition of H atoms thus should provide a
basis for the estimation of hydrogen atom’s reactivity to other
unsaturated systems.
’ ASSOCIATED CONTENT
S
Supporting Information. Product data and conditions of
b
individual experiments (Table S1). This material is available free
of charge via the Internet at http://pubs.acs.org.
’ REFERENCES
(1) Seakins, P. W.; Robertson, S. H.; Pilling, M. J.; Slagle, I. R.;
Gmurczyk, G. W.; Bencsura, A.; Gutman, D.; Tsang, W. J. Phys. Chem.
1993, 97, 4450.
(2) Rosado-Reyes, C. M.; Manion, J. A.; Tsang, W. J. Phys. Chem. A
2010, 114, 5710.
(3) Harris, G. W.; Pitts, J. N., Jr. J. Chem. Phys. 1982, 77, 3994.
(4) Kurylo, M. J.; Peterson, N. C.; Braun, W. J. Chem. Phys. 1971,
54, 4662.
2734
dx.doi.org/10.1021/jp111476q |J. Phys. Chem. A 2011, 115, 2727–2734