CH3O2 and C2H5O2
J. Phys. Chem., Vol. 100, No. 34, 1996 14381
The cross reactions of CH3C(O)CH2O2, t-C4H9O2, and
CCl3O2 with CH3O2 were derived from previous work and are
now briefly evaluated.
of magnitude as the rate constant for the CCl3O2 radical self-
reaction9 (k ) 4.0 × 10-12 cm3 molecule-1 s-1). The value of
the branching ratio Rc was optimized in simulations and found
to be equal to (0.5 ( 0.2) compared to the average of the self-
CH3C(O)CH2O2. The cross reaction of the acetonylperoxy
radical with CH3O2 was investigated by Bridier et al.6 at room
temperature using the flash photolysis of Cl2 in the presence of
acetone, methane, and oxygen. The chemistry in this system
is complicated by the formation of acetylperoxy radicals from
the decomposition of the acetoxy radical, CH3C(O)CH2O.
Taking account of all peroxy radicals interactions in the chemical
model, Bridier et al.6 proposed a value of (3.8 ( 0.4) × 10-12
cm3 molecule-1 s-1 for the cross-reaction rate constant. This
value is between the self-reaction rate constants for the
acetonylperoxy and the methylperoxy radicals (8.0 × 10-12 and
3.7 × 10-13 cm3 molecule-1 s-1, respectively, at 298 K).1,2,6
The value of the branching ratio Rc was determined to be (0.3
( 0.1) compared to the average of the self-reaction branching
ratios, i.e., Rc ) 0.54.
t-C4H9O2. The reaction of the tert-butylperoxy radical with
CH3O2 involves a radical bearing a tertiary central carbon atom
having a very slow self-reaction: k ) 3.0 × 10-17 cm3
molecule-1 s-1 at 298 K.1,2 The rate constant was studied by
two groups7,8 and was derived indirectly by modeling the
observed product yields of the t-C4H9O2 radical self-reaction
in which the methylperoxy radicals were produced from the
decomposition of the tert-butoxy radicals. The results obtained
by the two groups are in significant disagreement, Parkes
reporting a value of (1.0 ( 0.5) × 10-13 cm3 molecule-1 s-1 at
298 K,7 whereas Osbourne et al. propose a rate constant 2 orders
of magnitude lower8 (3.95 × 10-15 cm3 molecule-1 s-1). For
their part, Lightfoot et al.1 recommended 3.1 × 10-15 cm3
molecule-1 s-1, derived from the Arrhenius plot of the data of
Osbourne et al. Note that in our study of the reaction of neo-
C5H11O2 with CH3O2 in which the t-C4H9O2 and CH3O2 radicals
were formed in secondary reactions,1,2 the best fits of decay
traces at long reaction times were obtained using a value of
∼10-13 cm3 molecule-1 s-1 for the cross reaction of the two
latter radicals. The sensitivity to this value was, however, fairly
poor.
We have performed a few experiments, within the present
work, to measure the rate constant of the t-C4H9O2 + C2H5O2
cross reaction by laser-flash photolysis of tert-butyl chloride at
193 nm in the presence of ethane and oxygen (the investigation
of the cross reaction with CH3O2 could not be studied using
this method owing to the too large difference in reactivity of
Cl atoms with CH4 and t-C4H9Cl). Experimental results were
not very accurate, since the shape of decay traces could not be
simulated fully. Nevertheless, the measured values of the cross-
reaction rate constant, ranging from 1 × 10-13 to 2 × 10-13
cm3 molecule-1 s-1, were found again to be much larger than
the recommended value for the t-C4H9O2 + CH3O2 reaction,
whereas a similar value can reasonably be expected for the rate
constant of both reactions. We have ignored this last determina-
tion because of the low reliability of measurements, but it con-
firms the large uncertainty that exists for the rate constant of
the t-C4H9O2 + CH3O2 cross reaction. Nevertheless, we have
included in Table 5 the value recommended by Lightfoot et
al.1
CCl3O2. The reaction of the trichloromethylperoxy radical
with CH3O2 has been investigated recently in our laboratory9
using the flash photolysis of CCl4 in the presence of methane
and oxygen. The reaction mechanism is fairly well established,
and the measured value for the cross-reaction rate constant is
(6.6 ( 1.0) × 10-12 cm3 molecule-1 s-1 at 300 K. The value
of this rate constant is significantly higher than k31(CH2ClO2
+ CH3O2) measured in this work but is still of the same order
reaction branching ratios, i.e., Rc ) 0.66.
Analysis of Cross Reaction Kinetics. As for peroxy radical
self-reactions, the rate constants for cross reactions vary over a
very wide range: more than 4 orders of magnitude from the
rate constant of the t-C4H9O2 + CH3O2 reaction (k ) 3.1 ×
10-15 cm3 molecule-1 s-1) to that of the CH3C(O)O2 + CH3O2
reaction (k ≈ 10-11 cm3 molecule-1 s-1). To date, the
temperature dependence has been measured for the cross-
reaction rate constants of three radicals: CH2dCHCH2O2 (this
work), CH3C(O)O2,1,2 and t-C4H9O2.1,2 The values of E/R are
comparable to those observed for the corresponding self-
reactions, being positive for the slowest reactions (+1430 K
for the tert-butylperoxy radical reaction) and negative for most
other reactions (-430 K for the allylperoxy radical and -272
K for the acetylperoxy radical). The best recommendation that
can be made at the present time is to take for the cross reaction
a temperature dependence similar to that measured for the
peroxy radical self-reaction having the most comparable room-
temperature rate constant to the cross reaction.
It is observed that, in most cases, the cross reaction rate
constant is between the self-reaction rate constants of the two
reacting radicals and is often close to that of the fastest self-
reaction. Only in the cases of the reactions of CH3O2 with
CH2dCHCH2O2 and of CCl3O2 and C2H5O2 with CH2dC-
HCH2O2 is the cross reaction rate constant significantly larger
than that of either self-reaction. The following relationship
between cross and self-reactions of two radicals has been
proposed:3
k(RO + R′O ) ) 2 k(RO + RO ) × k(R′O + R′O ) (I)
x
2
2
2
2
2
2
and has been applied to the reactions of interest in this work.
The values calculated from expression I are included (in
parentheses) in Table 5. It is apparent that for most reactions,
the experimental and calculated rate constant do not differ by
more than a factor of 2 with the exception of the reactions of
the acetylperoxy radical, which seem to be always large, and
those of the tert-butylperoxy radical, which are small but
determined with poor accuracy, as discussed above. The other
exception is that of CCl3O2, which is larger than the predicted
value by a factor of almost 3 and was determined with fairly
good accuracy.9 We have no explanation for these particular
exceptions. Nevertheless, expression I may be a fairly good
approximation for estimating cross-reaction rate constants of
peroxy radicals, taking into account the large differences
observed from one reaction to another and the fairly large
uncertainties in the measurements as discussed above.
It can also be observed that rate constants of cross reactions
with CH3O2 are often close to the rate constants of RO2 self-
reactions. This is particularly true for the fastest reactions,
which are the most important in reaction systems. Thus, an
alternative recommendation might be to take for CH3O2 and
C2H5O2 cross reactions a rate constant close to that of the RO2
self-reaction. This would be a good approximation for fast
reactions and wrong for slow reactions, but the slow reactions
generally play a minor role in the reaction systems. The only
exception to the second rule of interest is the CH3O2 + C6H5-
CH2O2 reaction, which is found to be slower than predicted.
As already pointed out, this discrepancy may arise from large
experimental uncertainties resulting from the formation of
strongly absorbing benzaldehyde, which perturbs the flash
photolysis study of this system.