Thermal decomposition of 2-bromopropene, and 2-chloropropene

Article abstract of DOI:10.1039/b212670b

2-Chloropropene and 2-bromopropene have been decomposed in single pulse shock tube experiments. The only products under all conditions are propyne and allene. The high pressure rate expressions are k(2-BrC₃H₅ propyne/allene + HBr) =

Full text of DOI:10.1039/b212670b

View Online / Journal Homepage / Table of Contents for this issue
Thermal decomposition of 2-bromopropene, and 2-chloropropene
Karin Roy, Iftikhar A. Awan, Jeffrey A. Manion and Wing Tsang*
National Institute of Standards and Technology, Gaithersburg, MD 20899, USA
Received 3rd January 2003, Accepted 21st February 2003
First published as an Advance Article on the web 24th March 2003
2-Chloropropene and 2-bromopropene have been decomposed in single pulse shock tube experiments. The
only products under all conditions are propyne and allene. The high pressure rate expressions are
kð2-BrC3H5 ) propyne=allene þ HBrÞ ¼ 1014:9expðꢀ32 830=RTÞ sꢀ1
kð2-ClC3H5 ) propyne=allene þ HClÞ ¼ 1014:8expðꢀ34 200=RTÞ sꢀ1
in the temperature range 1100 to 1250 K and pressures of 150 to 800 kPa. The propyne to allene ratios are 1.8
and 1.6 for the brominated and chlorinated compounds respectively with minimal temperature dependence.
Results are compared with those for the alkyl compounds and ab-initio calculations on 2-chloropropene.
Differences in energy transfer efficiencies for 2-chloropropene and 4-methylcyclohexene decompositions are
explored.
stability of the chlorides in comparison to the bromides and
Introduction
iodides and the increase in rate constants with methyl substitu-
tion. These effects all arise from changes in the activation ener-
gies. The A-factor appears to be dependent only on the number
of hydrogen atoms available. The activation energies are found
to be about 0.28 of the ion dissociation energy (DH ¼
(RX ) R+ + Xꢀ).2 The general subject has been reviewed1
and there is a consensus that the transition state must be
semi-ionic in character. Nevertheless, the rate data results
maybe dependent in part on some contributions from the
homolytic bond energies.
The only experimental thermal results on hydrogen halide
elimination from unsaturated compounds are those for HCl
and HBr elimination from vinyl chloride and bromide. Clearly
rate constants are much smaller than those of the saturated
species. This makes experiments more difficult since there is
the added possibility of contributions from other channels.
The smaller rate constants are mostly due to large increases
in the activation energy, although the A-factors also appear
to be larger. Note that for all these species there is also the pos-
sibility that hydrogen halide elimination will be through a
three centered transition state (1,1 elimination).7 This means
that the rate constants given above may be the maximum pos-
sible if one wishes to consider only contributions from 1,2
elimination. The data on alkyl halide decomposition are all
from single pulse shock tube experiments. The data from the
vinyl halides are from a variety of methods. Interestingly,
two of these determinations lead to exactly the same rate
expressions. This is probably accidental. The review by
Maccoll1 showed that results from static and flow systems on
the alkyl halides lead to much larger spreads of rate para-
meters. Also included in Table 1 are results from recent
ab initio calculations.6
Dehydrohalogenation reactions have long been of interest to
kineticists.1 This is largely due to the analogies that exist
between gas phase and solution phase reaction rate constants
in polar media.1 It is therefore always of interest to expand
the database of relevant reactions. Furthermore, advances in
theory promise to offer new insights into the details of the reac-
tions. Table 1 summarizes some of the existing data2–6 on the
elimination of hydrogen halide from organic halides. We have
divided the results into alkyl halide and vinylic halide cate-
gories. Most of the results involve alkyl halides where the
hydrogen halide is ejected by 1,2 elimination. Note the greater
Table 1 Summary of data on hydrogen halide elimination from alkyl
and vinyl halide compounds. The alkyl halide results are from single
pulse shock tube studies. The vinyl halides are from all available results
1,2-Alkyl halide eliminations1 (experiments) [rate constants at
1000] sꢀ1
k(CH3CH2Cl ! CH2 CH2 + HCl) ¼ 1013.3exp(ꢀ28 330/T ) [9.9]
=
k(CH3CH2Br ! CH2 CH2 + HBr) ¼ 1013.3exp(ꢀ26 900/T ) [41.5]
=
k((CH3)2CHCl ! CH3CH CH2 + HCl) ¼ 1013.8exp(ꢀ25 600/T ) [457]
=
k((CH3)2CHBr ! CH3CH CH2 + HBr) ¼ 1013.7exp(ꢀ24 000/T )
=
[1887]
k((CH3)2CHI ! CH3CH CH2 + HI) ¼ 1013.8exp(ꢀ22 600/T ) [9185]
=
k((CH3)3CCl ! (CH3)2C CH2 + HCl) ¼ 1013.9exp(ꢀ22 400/T )
=
[14 960]
k((CH3)3CBr ! (CH3)2C CH2 + HBr) ¼ 1014.0exp(ꢀ20 800/T )
=
[92 610]
k((CH3)3CI ! (CH3)2C CH2 + HI) ¼ 1013.9exp(ꢀ19 070/T ) [417 921]
=
Vinylic halide eliminations (experiments)
k(C2H3Cl ¼ C2H2 + HCl) ¼ 1014exp(ꢀ34 880/T )3 [0.071]
k(C2H3Cl ¼ C2H2 + HCl) ¼ 1014exp(ꢀ34 880/T )4 [0.071]
k(C2H3Br ¼ C2H2 + HBr) ¼ 1015.2exp(ꢀ32 960/T )5 [7.3]
Recently, Mueller et al.8 described experiments where a
beam of 2-chloropropene was crossed with laser light at
193.3 nm. From the translational energy distribution they
inferred that molecular elimination of HCl from the ground
state molecule, resulted in the formation of both propyne
and allene with the former predominating. This is contrary
to the strengths of the C–H bond being broken, which differs
Vinylic halide eliminations (theory)
k(2-ClC3H5 ¼ p-C3H4 + HCl) ¼ 1015.15exp(35 090/T )6 [0.81]
k(2-ClC3H5 ¼ a-C3H4 + HCl) ¼ 1015.03exp(35 300/T )6 [0.51]
1806
Phys. Chem. Chem. Phys., 2003, 5, 1806–1810
This journal is # The Owner Societies 2003
DOI: 10.1039/b212670b

View Online
tal procedures can be found in earlier publications.2 The
by about 90 kJ molꢀ1.2 We were sufficiently intrigued by their
observations in demonstrating that the course of reaction is
completely uninfluenced by homolytic bond cleavage consid-
erations that we have now carried out single pulse shock
studies on the thermal decomposition of 2-bromo and chloro-
propenes where the halides are adjacent to the double bonds.
Nevertheless, Maccoll1 has pointed out that the rate constant
for elimination of an allylic hydrogen is only marginally
increased in comparison to a normal secondary hydrogen.
For that case the 50 kJ molꢀ1 (a factor of 150 in rate constants
if directly reflected in the activation energy at 1100 K) differ-
ence has minimal effects on rate constants for elimination.
Single pulse shock tube studies provide an ideal means of
studying the thermal decomposition of polyatomic molecules.2
This is due to a combination of short reaction times, the cap-
ability of working at very low concentrations, and the use of
chemical inhibitors. Thus it is possible to eliminate surface
and radical induced contributions to the decomposition pro-
cess. The only possible molecular destruction mechanism is
that of unimolecular decomposition. These considerations
can be strikingly demonstrated when one recalls that at typical
reaction conditions of 200 kPa and 500 ms residence time there
are in fact only a few million collisions. Indeed when one
works at dilute concentrations, of the order of a few hundred
parts per million, and in the presence of large quantities of a
radical scavenger, it is easy to show that induced decomposi-
tion is impossible. By comparison, it is well known that in clas-
sical static or flow systems special measures must be taken to
eliminate surface contributions to the decomposition process.
Single pulse shock tube experiments under the conditions
described here can therefore unambiguously establish in a
quantitative manner the thermal cracking pattern for most
polyatomic organic molecules.
This ability to isolate individual unimolecular reactions for
study means that one can simultaneously study several reac-
tions at the same time. If the rate expression for decomposition
for one of these reactions is well established, then this reaction
can serve as an internal thermometer. This circumvents the
main problem in obtaining truly quantitative results from
shock tube studies, the uncertainty in the reaction temperature
that is determined from the shock velocity. This is the basis of
the comparative rate method and extremely accurate rate
expressions can be obtained. Finally by varying inhibitor to
reactant ratios it is possible to demonstrate unambiguously
the correctness of the proposed mechanism.
The high pressure thermal rate expression contains all kine-
tically useful experimental information on the nature of the
transition state for a particular reaction. It is difficult to deter-
mine accurately since it is dependent on a slope measurement
and correct results can be obfuscated by the mechanistic pro-
blems in static experiments. Single pulse shock tube experi-
ments of the type described here have proved to be the one
of the few consistent sources of such information. A possible
source of error is that at sufficiently high temperatures, unim-
olecular reactions may no longer be at the high pressure limit.
There are some indications that this may be a problem in the
present work and will be discussed subsequently. The direct
formation of molecules in thermal decomposition is of particu-
lar interest since unlike straightforward bond breaking, pre-
dictive capabilities are not satisfactory. We are therefore
particularly interested in comparing the present results with
the theoretical calculations of Parsons et al.6 on the transition
state structure and hence the high pressure rate expression for
2-chloropropene.
heated aspect of the shock tube enables us to study a much lar-
ger range of compounds than is possible with a room tempera-
ture instrument as well as use larger organic molecules with
lower vapor pressures as inhibitors. When working at low con-
centrations, it also guarantees that adsorption on the walls will
not deplete reactants or products. The experimental tempera-
tures ranged from 1000 K–1250 K and reaction pressures
between 1.5 bar to 12 bar pressure. The heating times were
in the 500 ms range.
Product analysis utilized a dual column Hewlett Packard
6890 GC9 equipped with flame ionization detection (FID).
Except for hydrocarbons smaller than C4, most organics were
well-separated on a Restek 30 m ꢁ 0.053 mm (530 mm) id Rtx-
624 fused silica column (a crossbonded 6% cyanopropylphe-
nyl/94% dimethyl polysiloxane). Hydrocarbons up to about
C7 and C1 and C2 species containing up to three chlorines
were separable on a 1 m ꢁ 1 mm id Hayesep S followed by a
1 m ꢁ 1 mm id Hayesep Q column (Restek). We have found
this combination generally gives good separation of isomeric
hydrocarbons as well as those differing only in degree of un-
saturation. The GC was operated in the temperature pro-
grammed mode with constant carrier gas flow and gave good
separation of all species of interest.
The chemical inhibitors used in these studies were a number
of the methylbenzenes. They remove reactive radicals via the
reactions
ꢂ
ꢂ
R þ C6H3ðCH3Þn ) RH þ C6H3ðCH3Þnꢀ1CH2
ꢂ
) CH3 þ C6H4ðCH3Þnꢀ1
where n is less than 3. Given the overwhelming amount of the
methylbenzenes, the consequence of these reactions is to sub-
stitute a relatively unreactive benzyl radical for any more reac-
tive radicals. In the present environment, the benzyl radical
can only react with other radicals in the system through fast
combination reactions. In particular, it cannot attack the com-
pounds whose decomposition characteristics we are attempting
to determine. Thus the benzyl radicals are themselves extre-
mely effective inhibitors of chain processes. The reactions
being studied do not produce radicals. The presence of the
inhibitor assures that any radicals from minor channel impuri-
ties will not alter the mechanism by inducing pathways to
decomposition. As it turns out the reactions studied do not
involve radicals. Thus inhibitors are probably not necessary.
Nevertheless their presence guarantees that radical induced
decomposition cannot occur. Furthermore note that for bro-
mopropene decomposition reactions were carried out with
wide variations in inhibitor to reactant concentrations. The
failure to observe any systematic deviations provides the stron-
gest validation for the postulated mechanism.
The reaction temperature was determined from the reverse
Diels–Alder reaction of methylcyclohexene and in a few cases
cyclohexene. These reactions involve the direct formation of
stable molecules and therefore cannot contribute reactive radi-
cals to the system. The rate expressions are2
kðC6H10 ) C2H4 þ 1,3-C4H6Þ
¼ 1015:15expðꢀ33 500 K=TÞ sꢀ1
kðCH3C6H9 ) C3H6 þ 1,3-C4H6Þ
¼ 1015:3expðꢀ33 400 K=TÞ sꢀ1
where the uncertainties in the A-factors are estimated to be a
factor of 1.3 and the activation energy 3 kJ molꢀ1. The tem-
perature was determined through a measurement of the extent
of reaction of the internal standard. Thus if the rate expression
for the decomposition of one of the compounds is known to be
Experimental
The experimental studies are carried out in a heated single
pulse shock tube. Details of the shock tube and the experimen-
kuni ¼ AexpðꢀE=RTÞ
Phys. Chem. Chem. Phys., 2003, 5, 1806–1810
1807

View Online
where A is the pre-exponential factor, E the activation energy,
R the gas constant and T the reaction temperature, then from
the extent of decomposition of the standard reaction, the tem-
perature can be determined or
fln½1=tꢃlnðC0=CÞ ꢀ AgR=E ¼ 1=T
where t is the heating time and has been found to be about 500
ms for the present range of conditions. Since the molecules are
suffering through the same time as well as temperature and
pressure history, there is cancellation of errors. We have earlier
given a detailed analysis of the uncertainties from these types
of experiments.2 Indeed, if the activation energies are the same
then there is exact cancellation. Since it is difficult to carry out
experiments with widely different activation energies, the dis-
crepancies that can arise are small. Unimolecular decomposi-
tion rate constants are calculated in the standard manner
kunið2-halopropeneÞ
¼ ð1=tÞ½lnð1-ðpropyne þ alleneÞ=halopropeneÞꢃ
The propyne and allene are lumped together since, as will be
shown below, the ratio of the two products is invariant over
the conditions studied. kuni(2-halopropene) is thus the total
rate constant for product formation. Rate expressions for the
individual channels can be determined by multiplying the rate
expression by the constant factors a/1 + a and 1/1 + a for pro-
pyne and allene respectively where a is the ratio of propyne to
allene.
The 2-halopropenes, cyclohexene and 4-methylcyclohexene
were obtained from Aldrich Chemicals and were used without
further purification. Gas chromatographic analysis indicated
impurities were below 0.1%. The argon is Ultra-Pure carrier
grade from Air Products.9
Fig. 1 A: Arrhenius plot for the elimination of HBr from 2-bromo-
propene. The rate constants are for the sum of the two products. B:
Ratio of propyne and allene from 2-bromopropene decomposition.
(1) Filled circles are experiments at 2 bar. Closed squares are experi-
ments at 8 bar. Mixture composition: 200 ppm 2-bromopropene and
200 ppm 4-methylcyclohexene in 1% 1,3,5-trimethylbenzene. (2) Open
circles are for experiments at 1.6 bar. Mixture composition: 100 ppm
2-bromopropene and 100 ppm cyclohexene and 0.5% meta-xylene.
(3) Open triangles (up) are for experiments at 1.6 bar. Mixture compo-
sition: 100 ppm 2-bromopropene, 100 ppm 4-methylcyclohexene and
0.5% meta-xylene. (4) Open triangles (down) are for experiments at
2.5 bar. Mixture composition: 80 ppm 2-bromopropene, 80 ppm 4-
methylcyclohexene and 5% meta-xylene. (5) Open diamonds are for
experiments at 6 bar. 50 ppm 2-bromopropene, 50 ppm 4-methylcyclo-
hexene and 3% meta-xylene. The least squares fit of the data for the
propyne to allene ratio is 1.92 ꢄ 0.15exp(ꢀ177 ꢄ 140/RT ).
Results
For the two secondary halides studied here, the primary pro-
ducts are propyne and allene. Propyne is the predominant pro-
duct. The propyne to allene ratios are virtually constant over
the entire range of experimental conditions. The values are
1.8 for the bromide and 1.6 for the chloride. On a per hydrogen
basis the values are 2.7 and 2.4 respectively. However, if one
assumes that the hydrogens in the methyl groups are equiva-
lent (free rotor) and that 1,2-dehydrohalogenation can only
occur from the cis configuration then the ratio would be 5.4
and 3.6. This constant ratio is maintained to the highest extent
of decomposition or approximately 0.80 mole fraction for the
bromide. All of these experiments are carried out under condi-
tions where the isomerization reaction propyne , allene do
not make any contributions. We have verified this by separate
studies for this reaction.
Most of the experimental work deals with the decomposition
of 2-bromopropene. Conditions were varied to establish unam-
biguously that the reaction is truly hydrogen bromide elimina-
tion. The results are summarized in Fig. 1 and include studies
over a six-fold range of the concentration of the bromide.
Changes in inhibitor to reactant ratio range from 50 to 1000
to one and pressures from 180 kPa to 1200 kPa. As can be seen
from Fig. 1, the results show no significant trends with changes
in these variables and it can be concluded that true unimolecu-
lar rate expressions are being determined. The best fit of these
results leads to the rate expression.
constants and expression leads to possible errors of 20% in the
rate constants and factors of 2 in the A-factor and 6 kJ molꢀ1
in the activation energy. Note that this includes the uncertainty
in the standard reaction.
One problem with these results is the possibility of a pressure
dependence. Results at 2 bar and 8 bar from the same mixture
are in fact different by a factor of 1.10. By working with the
same mixture, errors from differences in mixture preparation
are eliminated. The overall dependence is of the order of p0.1
and in terms of the precision of the measurement is probably
no more than factors of 2 or 3 larger then the scatter of the
results. Since it is difficult to carry out experiments that encom-
pass greater ranges of pressure the data has been treated as if it
were at the high pressure limit. If this is not the case then the
limiting rate parameters should be somewhat larger. Fig. 1B
contains data pertinent to the propyne/allene ratio. Note the
invariance with temperature.
Results on the decomposition of 2-chloropropene can be
found in Fig. 2. In this case there does not appear to be any
pressure dependence. Note that the reaction pressures are var-
ied by the same amount as in the case of 2-bromopropene.
Also included in Fig. 2 are the results from the 2-bromopro-
pene study. It can be seen that the rate constants for the chloro
compound are about a factor of 5 smaller than those for the
bromo compound. The rate expression is
kð2-BrC3H5 ) propyne=allene þ HBrÞ
¼ 1014:92ꢄ0:16expðꢀ32 830 ꢄ 400 K=TÞ sꢀ1
kð2-ClC3H5 ) propyne=allene þ HClÞ
¼ 1014:8ꢄ:2expðꢀ34 190 ꢄ 500 K=TÞ sꢀ1
The stated uncertainties reflect the precision of the measure-
ments. In absolute terms, the uncertainties in the standard rate
1808
Phys. Chem. Chem. Phys., 2003, 5, 1806–1810

View Online
Table 2 Calculated k/k for 4-methylcyclohexene decomposition as
1
functions of pressure, temperature and step size down
Step size
down ¼ 250 cmꢀ1
Step size
down ¼ 500 cmꢀ1
Temp (K)/
pressure (bar)
1100
2
4
8
2
4
8
0.94
0.90
0.82
0.97
0.93
0.88
0.98
0.95
0.92
0.98
0.95
0.92
0.99
0.97
0.95
0.99
0.99
0.97
1160
1200
on the high pressure rate expression. For example for the
500 cmꢀ1 step down case the deviations from the high pressure
rate constants would lead to a lowering of the activation
energy by 7 kJ molꢀ1
.
However, if the deviation from the high pressure value of the
target reaction is the same, then cancellation of the fall-off
effects means that the use of the high pressure rate expression
for the standard will lead to high pressure rate expression for
the target. This would appear to be the case here. Nevertheless
the halopropenes are much smaller molecules than the 4-
methylcyclohexene used as a standard. Since the rate expres-
sions are not that different it is difficult to reconcile similar
pressure dependence unless both reactions are well into the
high pressure region. The data in Table 2 suggest that this is
not the case. Results for step sizes down of 500 cmꢀ1 and
1000 cmꢀ1 and the rate parameters from the ab-initio calcula-
tions of Parsons et al. can be found in Table 3. The conclusion
is that with the ab-initio rate expression a pressure dependence
of the rate constants should have been observed unless the
step-size down parameter is larger than that used here. It can
be seen that the agreement in the degree of fall-off can only
be attained when the step-size down parameter for 2-chloro-
propene decomposition is considerably higher than that for
4-methylcyclohexene or alternatively if they are the same the
rate expressions must be in the 1014exp(ꢀ32 000/T ) range.
This is beyond the range of errors from the present experi-
ments. The larger step size down parameter necessary to fit
results on chlorinated compounds is in accord with the obser-
vations of Lim and Michael.14 There is therefore a clear differ-
ence in the energy transfer characteristics of hydrocarbons and
chlorinated compounds.
Fig. 2 A: Arrhenius plot for elimination of HCl from 2-chloropro-
pene. The rate constants are for the sum of the two products. B: Ratio
of propyne to allene from 2-chloropropene decomposition. Filled
squares are for experiments at 8 bar. Filled circles are for experiments
at 2 bar. Mixture composition: 200 ppm 2-chloropropene and 200 ppm
4-methylcyclohexene in 1% 1,3,5-trimethylbenzene. Dotted lines are
results derived from the transition state properties from Parsons
et al. Dashed line are results for 2-bromopropene. The least squares
fit of the data for the propyne to allene ratio is 1.27 ꢄ
1.4exp(262 ꢄ 300/T ) with a scatter of 4%.
Data pertaining to the propyne and allene ratios can be found
in Fig. 2B. As with the bromo compound results are indepen-
dent of pressure and temperature.
Discussion
Table 4 contains results of calculations using the rate expres-
sion that was obtained in the present study and a step size
down of 1000 cmꢀ1. It can be seen that under such conditions
the pressure dependence will be below detectable limits. We
have also carried out calculations with the reaction threshold
reduced to that for 2-bromopropene. Due to the lower thresh-
old we are slightly more into the fall-off region. This will be
further accentuated by the somewhat looser structure of the
bromo compound. In the present work the rate expression at
the highest pressures of 800 kPa (from an Arrhenius fit of
the 800 kPa results) is in fact very close to that given here.
The consequence is that the present experiments are probably
sufficiently near the high pressure limit so that the experimen-
tal rate expressions reported here are valid for such situations.
The rate constants derived here are in quite good agreement
with the calculated results of Parsons et al.6 based on ab initio
calculations. This is a very impressive demonstration of the
capability of such calculations. Nevertheless it is possible that
the degree of agreement may partly be due to the compensa-
tion of errors in the A-factor and activation energies. Unlike
the experimental situation where rate constants are the directly
measured quantities, the calculated rate parameters are in
some sense independent of each other. This will be discussed
below.
An important issue is the degree that the rate expressions
given here represent high pressure values. The high tempera-
tures utilized in effecting decomposition will always require
examination of energy transfer effects.10 For this purpose it
will be necessary to consider contributions from the decompo-
sition of the standard as well as the target reaction. Table 2
summarizes results of fall-off calculations11 for 4-methylcyclo-
hexene decomposition for step sizes down of 250 cmꢀ1 and 500
cmꢀ1 in the temperature range covered by these experiments.
This range of step-sizes down is characteristic of hydrocarbon
systems in this temperature range.12 It is clear that experiments
are near the high pressure limit. The transition state is based
on that derived by Huang et al. for cyclohexene.13 Thus the
rate constants for 4-methylcyclohexene are high pressure
values and derived temperatures are correct. Nevertheless
these small differences are sufficient to have significant effects
Table 3 Calculated k/k for 2-chloropropene decomposition as
1
functions of pressure, temperature and step size down. Results are
based on the rate expression derived from the results of Parsons et al.
Step size
down ¼ 500 cmꢀ1
Step size
down ¼ 1000 cmꢀ1
Temp (K)/
pressure (bar)
1100
2
4
8
2
4
8
0.70
0.61
0.54
0.78
0.71
0.63
0.85
0.79
0.72
0.82
0.75
0.68
0.88
0.82
0.76
0.92
0.88
0.83
1160
1220
Phys. Chem. Chem. Phys., 2003, 5, 1806–1810
1809

View Online
Table 4 Calculated k/k for 2-chloropropene decomposition as functions of pressure and temperature for a step size down of 1000 cm based on
1
the rate expression derived from present work and that for 2-bromopropene decomposition
2-Chloropropene
2-Bromopropene
1014.8exp(ꢀ34 190/T ) sꢀ1
1014.92exp(ꢀ32 830/T ) sꢀ1
Temp (K)/pressure (bar)
2
4
8
2
4
8
1100
1160
1220
0.90
0.86
0.81
0.94
0.91
0.87
0.96
0.94
0.91
0.87
0.82
0.75
0.92
0.88
0.83
0.95
0.91
0.89
The rate constants for the bromo- and chloropropenes
are each about two orders of magnitude smaller than those for
isopropyl bromide and chloride. This suggests that the effect of
halogen substitution in 1,2 elimination is not significantly
affected by the double bond. It would thus appear that the
rate expression for 2-iodopropene could be estimated by
the data in Table 1. The effect of methyl substitution is less
certain. Extrapolation of the published result on Shilov and
Sbirova5 for vinyl bromide decomposition leads to rate con-
stants very similar to those for the 2-bromopropene from the
present work. This is contrary to the situation for the chloro
compounds. Vinyl chloride is much more stable than 2-
chloropropene and the difference is in fact similar to that
observed for the analogous alkyl chlorides. We have preli-
minary data that suggest that for 1-bromopropene, rate con-
stants are in fact smaller than for 2-bromopropene. There is
a question as to the contribution of a three centered transi-
tion state to these rate expressions. We hope to carry out
some studies on vinyl bromide in the near future.
It is clear that the correlations developed for alkyl halide
decomposition are not applicable upon the introduction of
the double bond adjacent to the halide. The activation energy
is larger than that calculated on the basis of the correlation
between ion dissociation energy and the activation energy
developed for the alkyl halides. The A-factor is about an order
of magnitude larger and is indicative of a looser transition
state. It is independent of the number of available H-atoms.
Parsons et al. have interpreted the propensity for propyne for-
mation in terms of the locking of the motion of the H-atoms in
the internal rotor required for allene formation. These results
must however be crucially dependent on the nature of the
cation in the semi-ionic transition state. It is unfortunate that
the calculations of Parsons et al. did not examine this aspect of
the problem. In addition, a particularly interesting results
would be similar calculations on the isopropyl chloride and a
comparison of the semi-ionic transition state structures for
the two cases.
unimolecular dehydrohalogenation. High pressure rate expres-
sions have been obtained. The results confirm the observation
of Muller et al. regarding the predominance of propyne to
allene in 2-chloropropene decomposition. The rate expressions
and constants are in fair agreement with the ab-initio calcula-
tions of Parsons et al. The effect of energy transfer on the
relative rate constants for 4-methylcyclohexene and the
2-halopropenes have been considered. It is concluded that
the latter must be more easily activated.
References
1
2
A. Maccoll, Chem. Rev., 1969, 69, 33.
W. Tsang, Comparative Rate Single Pulse Shock Tube Studies on
the Thermal Stability of Polyatomic Molecules, in Shock Waves
in Chemistry, ed. A. Lifshitz, Marcel Dekker Inc., New York,
1981, p. 59.
3
J. A. Manion and R. Louw, J. Chem. Soc., Perkin Trans. 2, 1988,
1547.
F. Zabel, Int. J. Chem. Kinet., 1977, 9, 6551.
A. E. Shilov and R. D. Sbirova, Kinet. Catal., 1964, 5, 32–39.
B. F. Parsons, L. J. Butler and B. Ruscic, Mol. Phys., 2002,
1001, 865.
4
5
6
7
8
J. F. Riehl and K. Morokuma, J. Chem. Phys., 1994, 100, 8976.
J. A. Mueller, B. F. Parson, L. J. Butler, F. Qi, O. Sorkhabi and
A. G. Suits, J. Chem. Phys., 2001, 114, 4505.
9
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 or
endorsement by the National Institute of Standards and Technol-
ogy, nor does it imply that the material or equipment is necessarily
the best available for the purpose.
10 K. A. Holbrook, M. J. Pilling and S. H. Robertson, Unimolecular
Reactions, John Wiley, 1996.
11 V. Bedanov, W. Tsang and M. R. Zachariah, J. Phys. Chem.,
1995, 99, 11 452.
12 W. Tsang and J. H. Kiefer, Unimolecular Reactions over Extended
Pressure and Temperature Ranges, in Dynamics and Kinetics
of Small Radicals, ed. K. Liu and A. Wagner, World Scientific
Company, Singapore, 1995, p. 59.
Summary
13 C. H. Huang, L. C. Tsai and W. P. Hu, J. Phys. Chem., 2001, 105,
9045.
14 K. P. Lim and J. V. Michael, J. Chem. Phys., 1993, 98, 3919–3928.
2-Bromo- and 2-chloropropene have been decomposed in sin-
gle pulse shock tube experiments. The mechanism involves
1810
Phys. Chem. Chem. Phys., 2003, 5, 1806–1810