9
792 J. Phys. Chem., Vol. 100, No. 23, 1996
Kaiser and Wallington
cases, the improvement upon reducing the value of Fcent. is of
the order of the experimental error. Changing the value of Fcent.
from 0.6 to 0.45 increases the calculated values of k1a(0) and
k1a(∞) only slightly more than the stated experimental errors.
The low-pressure rate constant for the addition of Cl to C3H6
is approximately 30 times faster than the low-pressure limiting
to form the C3H6Cl radical becomes insignificant. In this
pressure regime, the apparent rate constant for allyl formation
from the elimination channel reaches a constant value deter-
mined by the rate constants for HCl elimination (kE), elimination
of Cl from the activated complex to return to the reactants (kR),
and the high-pressure limiting rate constant (k1a(∞)):
-
29
6
-2 -1
rate constant for ethylene ()1.4 × 10
cm molecule
s
for Fcent. ) 0.6). This shows that a larger molecule such as
C3H6, containing more low-frequency vibrational modes, dis-
tributes the excess energy from the addition reaction more
efficiently within the molecule, resulting in a large increase in
the low-pressure addition rate constant.
kR
k ) k (P<10Torr) - k (P>200Torr) ) k (∞)/
+ 1
1b
1b
1a
[
]
kE
-
11
Substituting the measured values of k1b(P<10) ) 3.7 × 10
,
-11
-10
cm3
k1b(P>200) ) 2.3 × 10 , and k1a(∞) ) 2.7 × 10
molecule-
gives kR/kE ) 18.
k1a(∞) is close to the gas kinetic limit and indistinguishable
1
-1
-
10
3
-1 -1
s
from that of ethylene ()3.2 × 10
cm molecule
s
for
Fcent. ) 0.6) within the uncertainty of the determinations. Such
behavior is not surprising since there should be little steric
hinderance in the approach of the Cl atom to the II bond in
either molecule.
For pressures > 100 Torr, k1b ) (2.3 ( 0.3) × 10
molecule-
(circles in Figure 6). This value is very similar
to that for H atom abstraction from methyl groups in alkanes
Conclusions
The rate constant (k ) for the reaction of Cl atoms with C H
6
1
3
has been measured as a function of pressure in relative rate
experiments, using either FTIR or GC detection in two different
reactors. The results of the two experimental techniques yield
indistinguishable results, suggesting the absence of significant
systematic errors.
At high pressure (>100 Torr), the primary product observed
following UV irradiation of Cl2/C3H6/N2 mixtures is 1,2-
dichloropropane with a yield in excess of 85%, indicating that
addition of a Cl atom to the double bond (k1a) is the major
reaction channel. On the basis of the measured values of k1
and the 1,2-dichloropropane yield as a function of pressure, the
-11
cm3
1
-1
s
-11
3
-1 -1
(
3 × 10 cm molecule s ). If reaction channel 1b to form
the allyl radical is purely an abstraction process, no pressure
dependence would be expected. While k1b does not change
dramatically over the factor of 2000 pressure range studied, it
does increase by approximately 50% as the pressure decreases
-
11
from 100 to 10 Torr (from a value of 2.3 × 10
(200-700
-11
Torr) to 3.7 × 10
(<30 Torr). While this change is small,
it is observed in both GC and FTIR experiments and is greater
than the experimental uncertainty.
-10
3
-1 -1
high [k1a(∞) ) (2.7 ( 0.4) × 10
cm molecule s ] and
-28
6
-2
low pressure [k1a(0) ) (4.0 ( 0.4) × 10
cm molecule
We suggest that this observation may result from the existence
of two possible channels for allyl radical generation within
reaction 1b. The first channel, represented by the rate constant
measured at pressures >200 Torr, arises from direct hydrogen
abstraction at the methyl group in propylene. This rate constant
is independent of pressure, as expected for a metathesis reaction.
The second channel becomes apparent only at pressures <
-1
s ] limiting rate constants have been calculated assuming that
Fcent. ) 0.6. Because of the additional vibrational degrees of
freedom, the low-pressure limiting rate constant for propylene
is approximately 30 times larger than that for Cl addition to
C2H4.
At low pressure (<10 Torr), allyl chloride is the major product
yield 75-90%), indicating that as the addition reaction rate
(
1
00 Torr. We believe that this arises from an addition-
slows, removal of an H atom from the CH3 group in C3H6 is
the dominant channel. Measurement of the product yields over
the full pressure range studied (0.3-700 Torr) permits calcula-
tion of the pressure dependence of the rate constant for this
channel (k1b). Three different pressure regimes were identified
for k1b. Within the experimental uncertainties, there was no
discernable effect of total pressure on k1b over the ranges 0.3-
elimination process (reaction A) in which a Cl atom adds to
the double bond to form an activated complex (C3H6Cl*), from
which HCl can be eliminated.
k1a(∞)
kE
C3H6 + Cl
C3H6Cl*
C3H5 + HCl
(A)
kR
(M)
-
11
1
1
00 Torr (k1b ) 3.7 × 10 ) and 200-700 Torr (k1b ) 2.3 ×
-11
0
). In contrast, k1b increased by approximately 50% as the
C3H6Cl
total pressure was decreased from 100 to 10 Torr. We suggest
that this increase in the rate constant at lower pressure indicates
that the allyl radical may be formed by two processes: (1) a
pressure independent direct abstraction; (2) an addition-
elimination process which depends inversely on the total
pressure.
On the basis of a Lindemann analysis of the pressure
dependence of such a reaction, the overall rate constant (k) for
C3H5 formation by elimination from the activated complex is
d[C H ]
3
5
)
k[C H ][Cl]
3 6
dt
References and Notes
k + k [M]
R
M
where k ) k (∞)/
+ 1
1
a
(1) Aschmann, S. M.; Atkinson, R. Int. J. Chem. Kinet. 1995, 27, 613.
(2) Hooshiyar, P. A.; Niki, H. Int. J. Chem. Kinet. 1995, 27, 1197.
[
]
kE
(
3) Beichert, P.; Wingen, L.; Lee, J.; Vogt, R.; Ezell, M. J.; Ragains,
M.; Neavyn, R.; Finlayson-Pitts, B. J. J. Phys. Chem. 1995, 99, 13156.
4) Tyndall, G. S.; Orlando, J. J.; Wallington, T. J.; Dill, M. J.; Kaiser,
E. W. Int. J. Chem. Kinet., in press.
Formation of the allyl radical by this type of reaction will have
an inverse pressure dependence, as has been observed for the
addition-elimination reaction leading to the formation of
ethylene from C2H5 + O2.18 Near the high-pressure limit, the
activated complex is stabilized efficiently by third-body colli-
sions. In this pressure region, the rate of the HCl elimination
reaction is small and inversely dependent on total pressure. At
low pressures, the elimination reaction becomes increasingly
important. At sufficiently low pressure, the stabilization channel
(
(
(
5) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1996, 100, 4111.
6) Kaiser, E. W. Int. J. Chem. Kinet. 1992, 24, 179.
(7) Wallington, T. J.; Gierczak, C. A.; Ball, J. C.; Japar, S. M. Int. J.
Chem. Kinet. 1989, 21, 1077.
(
(
8) Atkinson, R.; Aschmann, S. C. Int. J. Chem. Kinet. 1985, 17, 33.
9) Wallington, T. J.; Skewes, L. M.; Siegl, W. O. J. Photochem.
Photobiol. 1988, 45, 167.
(10) Manning, R. G.; Kurylo, M. J. J. Phys. Chem. 1977, 81, 291.