Formation of C3H6 from the reaction C3H7 + O2 between 450 and 550 K

Article abstract of DOI:10.1021/jp981774t

The generation of C₃H₆ from the reaction C₃H₇ + O₂ (1) has been investigated as a function both of temperature (450-550 K) at constant density (5.5 × 10¹⁸ molecules/cm³) and of pressure (55-550 Torr) at 490 K. The experiments were carried out by UV irradiation of mixtures of C₃H₈, Cl₂, and O₂ to generate propyl radicals. C₃H₈, C₃H₆, and C₃H₇Cl were monitored by gas chromatographic analysis. The propylene yield is 0.7% at 450 K. Based on these measurements and previous data at 298 K, the propylene yield has an apparent activation energy which is less than 2.5 kcal mol^-1 below 450 K. Beginning near 450 K, the yield increases rapidly with an apparent activation energy of ～32 kcal mol^-1, similar to previous observations on the generation of C₂H₄ from the reaction C₂H₅ + O₂. At 490 K, the propylene yield from reaction 1 depends inversely on total pressure (YC₃H₆ ∝ P ^-0.6) between 55 and 550 Torr, while the overall value of k₁ has a much smaller pressure dependence (P^0.18). These observations show that above 450 K propylene is formed via reaction 1 through an excited propylperoxy adduct which can be stabilized by collision as was observed at 298 K.

Full text of DOI:10.1021/jp981774t

J. Phys. Chem. A 1998, 102, 5903-5906
5903
Formation of C3H6 from the Reaction C3H7 + O2 between 450 and 550 K
E. W. Kaiser
Research Laboratory, Mail Drop 3083/SRL, Ford Motor Company, Dearborn, Michigan 48121-2053
ReceiVed: April 7, 1998; In Final Form: May 27, 1998
The generation of C3H6 from the reaction C3H7 + O2 (1) has been investigated as a function both of temperature
(450-550 K) at constant density (5.5 × 1018 molecules/cm3) and of pressure (55-550 Torr) at 490 K. The
experiments were carried out by UV irradiation of mixtures of C3H8, Cl2, and O2 to generate propyl radicals.
C3H8, C3H6, and C3H7Cl were monitored by gas chromatographic analysis. The propylene yield is 0.7% at
450 K. Based on these measurements and previous data at 298 K, the propylene yield has an apparent activation
energy which is less than 2.5 kcal mol-1 below 450 K. Beginning near 450 K, the yield increases rapidly
with an apparent activation energy of ∼32 kcal mol-1, similar to previous observations on the generation of
C2H4 from the reaction C2H5 + O2. At 490 K, the propylene yield from reaction 1 depends inversely on total
pressure (YC H
P-0.6) between 55 and 550 Torr, while the overall value of k1 has a much smaller pressure
3
6
dependence (P0.18). These observations show that above 450 K propylene is formed via reaction 1 through
an excited propylperoxy adduct which can be stabilized by collision as was observed at 298 K.
Introduction
pressure decreases, the rate of (1b) decreases, increasing the
olefin yield even if formed via a pressure-independent abstrac-
tion channel.
The present publication extends this previous study of reaction
1 to include elevated temperatures (450-550 K) in order to
determine whether the propyl radical shows the onset of rapid
olefin generation in this temperature range as was observed for
the C2H5 + O2 reaction. No estimation of the C3H6 yield from
the individual, isomeric 1- and 2-propyl radicals is possible in
these experiments.
Detailed experiments1-5 and calculations6-9 have been carried
out on the reaction of ethyl radicals with O2 over the past decade.
This body of work has shown that at low to moderate
temperatures the reaction proceeds via an excited adduct which
can either isomerize and decompose to form C2H4 + HO2 or
be stabilized by collisions to form C2H5O2. Temperature-
dependent measurements5 have shown that the yield of ethylene
from this reaction has a very weak temperature dependence (Ea
e 1 kcal mol-1) at temperatures below that (450 K) at which
the O2 addition reaction begins to be reversible. Above this
temperature, the ethylene yield increases rapidly.
Experiment
The experimental apparatus consists of a cylindrical, 80 cm3,
Pyrex reactor, 20 cm in length, enclosed in a tube oven. The
reactor was irradiated by a single Sylvania F6T5 BLB fluores-
cent lamp, whose intensity peaks at 360 nm. The reactants,
containing C3H8, Cl2, O2, and N2, were premixed in a separate
flask, and the reactor was filled with the mixture to the desired
pressure. UV irradiation dissociates Cl2 to form Cl atoms. The
Cl atoms subsequently react with C3H8 to form 1- and 2-propyl
radicals which cannot be distinguished from one another in these
experiments:
Recent experiments have determined the propylene yield from
reaction 1
C3H7 + O2 ) products
(1)
as a function of total pressure at 298 K.10 The C3H6 yield
showed a strong inverse pressure dependence at ambient
temperature over a pressure range (P ) 0.5-100 Torr) within
which the addition reaction to form C3H7O2 is near its high-
pressure limit. As discussed in refs 5 and 10, this is a clear
indication of a channel that passes through an excited adduct
rather than an abstraction reaction:
Cl2 + hυ ) 2Cl
Cl + C3H8 ) C3H7 + HCl
C3H7 + O2 ) C3H7O2* ) C3H6 + HO2
C3H7O2* + M ) C3H7O2 + M
(1a)
(1b)
These reactions provide the source of propyl radicals for the
study of reaction 1.
Note that if reaction 1 is not near its high-pressure limit and
olefin formation is not its major channel, the presence of a
pressure-dependent olefin yield does not exclude the existence
of an abstraction channel. For example, if reaction 1b is at its
low-pressure limit and C3H6 is formed via abstraction (which
is pressure independent), then the C3H6 yield would show an
inverse dependence on pressure. This occurs because the olefin
yield depends on the rate of the abstraction reaction relative to
that of the addition process to form C3H7O2. Thus, as the
The mixture was irradiated for a predetermined time after
which the entire contents of the reactor were analyzed by gas
chromatography (GC) using flame-ionization detection; con-
sumption of C3H8 varied from 6% to 40%. There was no
discernible impact of degree of consumption on the measure-
ments. For the experiments in which propylene yields were
determined, the mixtures contained 10500 ppm of C3H8 and
18 000 ppm of Cl2, with the balance O2. Limited measurements
of the yields of 1- and 2-chloropropane were carried out with
S1089-5639(98)01774-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/29/1998

5904 J. Phys. Chem. A, Vol. 102, No. 29, 1998
Kaiser
TABLE 1: Yields (in ppm) of Chloropropanes at 490 Ka
P (Torr)
C3H8 consumed
1-C3H7Cl
2-C3H7Cl
k1/k2
550
61
550
275
275
61
1660
1840
2226
2045
3574
1883
95
160
155
155
260
185
155
270
285
295
470
325
0.32
0.19
0.23
0.20
0.22
0.16
a Initial concentrations: 10 600 ppm of C3H8; 18 200 ppm of Cl2;
318 000 ppm of O2; N2 balance.
Figure 1. Plot of C3H6 formed as a function of C3H8 consumed via
O2 reaction with propyl radicals at 512 ( 2 K and 290 Torr. Error
limits are shown as lines through the individual points. Open and closed
symbols were obtained on separate days. The slope is the “apparent”
yield uncorrected for the occurrence of reaction 2; see text. Initial
mixture: 10 900 ppm of C3H8; 17 900 ppm of Cl2; balance O2.
a mixture containing 10 500 ppm of C3H8, 18 200 ppm of Cl2,
and 318 000 ppm of O2 with N2 balance.
Results
Figure 1 shows the C3H6 mole fraction as a function of the
C3H8 consumed for five experiments on two separate days with
different percentages (6-43%) of C3H8 consumption at 512 K
and 290 Torr total pressure. To within the experimental error,
the yield is essentially independent of the percentage of C3H8
consumed. This indicates that any uncertainty associated with
secondary consumption of C3H6 is of the order of the measure-
ment error at this temperature, and no correction for secondary
consumption of C3H6 has been applied to the measured yields
throughout this publication. The Cl + C3H6 rate constant (k3)
has not been measured at 512 K. However, at 298 K11 and a
density of 5.5 × 1018 cm-3, k3 ) 1.8 × 10-10 cm3 molecule-1
s-1, which is slightly larger than that of Cl + C3H8 ()1.4 ×
10-10).12 Because k3 will decrease as the temperature increases
at constant density, the above value of k3 represents an upper
limit to its value at 512 K. Based on these rate constants at
298 K, this correction should be <20%, consistent with the data
in Figure 1 in which any correction is of the order of the data
scatter.
Figure 2. Rate constant ratio (k1/k2) presented as a function of total
pressure at 490 K (triangles, left axis). Filled and open symbols obtained
on different days. Initial mixture for k1/k2 experiments: 10 600 ppm
of C3H6; 18 200 ppm of Cl2; 318 000 ppm of O2, balance N2. Pressure
dependence of line ) P0.18. Yield of C3H6 plotted as a function of total
pressure at 490 K (circles, right axis). Initial mixtures for C3H6
measurements: A (open circles) 10 600 ppm of C3H6, 18 200 ppm of
Cl2, 318 000 ppm of O2, balance N2; B (filled circles) 10 400 ppm of
C3H6, 18 000 ppm of Cl2, balance O2. Pressure dependence of line )
P-0.6
.
of k1 must be known. The pressure dependence of k1 can be
determined directly from the ratio k1/k2 because reaction 2 will
have no pressure dependence. The ratio k1/k2 can be calculated
from the total yield of the chloropropanes (YRCl), from which
can be calculated the total yield of products from O2 reaction
(1 - YRCl) assuming that propyl radicals react only with O2 or
Cl2, using the known ratio of Cl2 to O2 in the initial mixture:
k1/k2 ) {[Cl2]/[O2]}{(1 - YRCl)/YRCl
}
However, the apparent 11% yield in Figure 1 should be
corrected for the formation of small yields of 1- and 2-chloro-
propane via reaction 2:
The measured ratios are presented in Table 1 and plotted in
Figure 2. There is a small increase (50 ( 15%) in the rate
constant ratio as the pressure increases by a factor of 10.
Because the rate constant for an abstraction reaction such as
(2) will not be pressure dependent, this increase must result
from a small increase in the rate constant of reaction 1, which
is near to but not quite at its high-pressure limit over the pressure
range studied at 490 K. The rate constant, k2, for isopropyl
radical reaction with Cl2 has been measured at 490 K ()4.1 ×
10-11 cm3 molecule-1 s-1).13 Assuming that the rate constant
for the n-propyl radical reaction with Cl2 is similar to this value,
k1 ≈ 1 × 10-11 cm3 molecule-1 s-1 at 550 Torr, a value which
is typical of O2 addition to alkyl radicals in the high-pressure
limit.
C3H7 + Cl2 ) C3H7Cl + Cl
(2)
since this channel does not involve reaction with oxygen. The
yields of the chloropropanes were measured at 490 K; results
from experiments at three pressures on two separate days are
presented in Table 1. These tabulated yields have been corrected
for a small amount of secondary consumption (e15%) of the
chloropropanes by Cl atoms using rate constants determined
previously at 298 K.12 The rate constants at 298 K for reaction
of Cl with 1- and 2-chloropropane are fast (4.8 × 10-11 and 20
× 10-11 cm3 molecule-1 s-1, respectively) and will not have a
large temperature dependence.
The measured rate constant ratio, k1/k2, plotted in Figure 2
as a function of pressure at 490 K, can be used to estimate the
correction to the apparent C3H6 yields caused by reaction 2.
As discussed in the Introduction, to verify that the propylene
is being formed via an excited adduct, the pressure dependence

Formation of C3H6 from C3H7 + O2
J. Phys. Chem. A, Vol. 102, No. 29, 1998 5905
adduct rather than by direct H atom abstraction from propyl
radicals. Because reaction 1b is at its high-pressure limit for
pressures greater than 1 Torr at ambient temperature, if
abstraction were the source of the olefin formation at 298 K,
no pressure dependence would be present. However, a strong
negative pressure dependence was observed, consistent with a
competition between reactions 1a and 1b at 298 K.
To verify that this is also the case at elevated temperature,
measurements of the propylene yield were made as functions
of pressure (55-550 Torr) at 490 K; the results are presented
in Figure 2. The C3H6 yields in Figure 2 were obtained at two
initial O2 concentrations; the yield is independent of O2 as
expected. The yield of propylene is strongly pressure dependent,
decreasing by a factor of 4 over this pressure range. This
represents a P-0.6 pressure dependence, similar to the P-0.68
observed at 298 K. As discussed above, the rate constant ratio
k1/k2 increases by approximately 50 ( 15%, indicating that k1
increases by 50% over this pressure range (P0.18). These two
sets of results show that propylene cannot be formed primarily
by abstraction at 490 K because the pressure dependence of
reaction 1 is much smaller than that observed for propylene
generation. Therefore, it is likely that, over the entire temper-
ature range studied, propylene arises from the decomposition
of an excited propylperoxy radical (reaction 1a). This observa-
tion is consistent with conclusions drawn for the ethyl radical
at elevated temperature.6
Figure 3. Plot of percentage yield of C3H6 as a function of 1/T at a
constant density [(5.5 ( 0.5) × 1018 molecules cm-3]. Initial mixture:
10 500 ppm of C3H6; 18 000 ppm of Cl2; balance O2.
This correction is small (6-9% depending on total pressure)
for the ratio [O2]/[Cl2] ) 54 used in the C3H6 determinations
and has been applied to all C3H6 yields presented in the
following discussion. The ratio k1/k2 is assumed constant over
the experimental temperature range.
Figure 3 presents the yields of C3H6 (corrected for C3H7Cl
formation) determined over the temperature range 450-550 K
in the current experiments at a total gas density of (5.5 ( 0.5)
× 1018 molecules cm-3 of O2. Also included in Figure 3 is the
C3H6 yield at this density measured at 298 K in the earlier
experiments.10 The data show a temperature dependence similar
to that observed for the ethylene yield from C2H5 + O2. For
temperatures between 298 and 450 K, the rate of increase in
propylene yield is small. No data are available for intermediate
temperatures in this range in contrast to the ethyl radical data
because the C3H6 yield is too small to be measured by the GC
in the presence of a large C3H8 signal.
Conclusions
Mixtures of C3H8, O2, and Cl2 have been irradiated with 350
nm UV light to form C3H7 radicals, which subsequently react
with O2. The C3H6 yields from the C3H7 + O2 reaction were
measured at pressures between 55 and 550 Torr and tempera-
tures from 450 to 550 K. The C3H6 yield depends inversely
(P-0.6) on the total pressure at 490 K, a temperature at which
reaction 1 shows a much smaller pressure dependence (P0.18).
This observation shows that the propylene is formed through
an excited propylperoxy radical at this temperature as has been
observed previously at 298 K.10
Beginning near 450 K, the C3H6 yield increases sharply,
reaching 35% at 550 K. Again, this observation is similar to
that for C2H5 + O2. In the ethyl radical case, data obtained in
the intermediate temperature range showed that the apparent
activation energy for ethylene generation was e1 kcal/mol. This
value is consistent with the data for reaction 1 presented in
Figure 3, for which any activation energy must be less than 2.5
kcal/mol as shown by the line connecting the 298 and 450 K
data. At temperatures above 450 K, the propylene yield
increases with an apparent initial activation energy of ap-
proximately 32 kcal/mol. As demonstrated by chemical kinetic
model calculations for the C2H5 + O2 reaction in ref 5, this
rapid increase in C3H6 likely occurs because the reaction forming
C3H7O2 becomes reversible and can return to reactants (C3H7
+ O2) above this temperature on the time scale of the
experiments. Because of this reversibility, multiple passes
through the excited adduct occur, each pass having a small
probability of forming C3H6. Therefore, the C3H6 yield
increases above that observed at lower temperature for a single
pass. The apparent initial activation energy of approximately
32 kcal/mol is also of the order of the heat of reaction for
isopropyl radicals with O2 to form C3H7O2 ()37.8 kcal/mol).14
This is consistent with the suggestion that the reversibility of
reaction 1 plays an important role in the rapid increase of the
propylene yield for temperatures above 450 K.
The temperature dependence of the propylene yield is similar
to that observed for ethylene formed in the C2H5 + O2 reaction,
which also is generated through an excited peroxy intermediate.5
Between 298 and 450 K, the propylene yield increases slowly
with an apparent activation energy <2.5 kcal mol-1. Beginning
near 450 K, the C3H6 yield increases sharply with an apparent
activation energy of approximately 32 kcal. This behavior is
ascribed to the onset of reversibility in reaction 1, allowing
repeated opportunity for C3H6 formation with each passage
through the excited peroxy intermediate. These results are
similar to the temperature dependence observed for C2H5 + O2,
demonstrating that the influence of excited peroxy radicals in
low to intermediate temperature oxidation is not restricted to
ethyl radicals.
Note Added in Proof. Because of temperature nonunifor-
mity along the cylinder axis (see ref 5), the average absolute
reactor temperature is estimated to be accurate to (2%; relative
temperatures are uncertain to (0.5%. At 520 K, the C3H6 yield
in the absence of O2 is <10% of that observed with O2 present,
verifying that the C3H6 is formed by reaction with O2 as required
by reaction 1.
References and Notes
As discussed in ref 10, the formation of propylene from
reaction 1 at 298 K must occur via an excited propylperoxy
(1) Slagle, I. R.; Feng, Q.; Gutman, D. J. Phys. Chem. 1984, 88, 3648.
(2) Plumb, I. C.; Ryan, K. R. Int. J. Chem. Kinet. 1981, 13, 1011.

5906 J. Phys. Chem. A, Vol. 102, No. 29, 1998
Kaiser
(3) Kaiser, E. W.; Lorkovic, I. M.; Wallington, T. J. J. Phys. Chem.
1990, 94, 3352.
(4) McAdam, K. G.; Walker, R. W. J. Chem. Soc., Faraday Trans. 2
1987, 1509.
(5) Kaiser, E. W. J. Phys. Chem. 1995, 99, 707.
(6) Wagner, A. F.; Slagle, I. R.; Sarzynski, D.; Gutman, D. J. Phys.
Chem. 1990, 94, 1853.
(9) Bozzelli, J. W.; Dean, A. M. J. Phys. Chem. 1990, 94, 3133.
(10) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1996, 100, 18770.
(11) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1996, 100, 9788.
(12) Tyndall, G. S.; Orlando, J. J.; Wallington, T. J.; Dill, M.; Kaiser,
E. W. Int. J. Chem. Kinet. 1997, 29, 43.
(7) Quelch, G. E.; Gallo, M. M.; Schaefer, H. F. J. Am. Chem. Soc.
1992, 114, 8239.
(8) Quelch, G. E.; Gallo, M. M.; Shen, M. S.; Xie, Y.; Schaefer III,
H. F.; Moncrieff, D. J. Am. Chem. Soc. 1994, 116, 4953.
(13) Timonen, R. S.; Gutman, D. J. Phys. Chem. 1986, 90, 2987.
(14) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr, J.
A.; Troe, J. J. Phys. Chem. Ref. Data 1992, 21, 1125.