Reaction of Cl- with CH3Br
J. Phys. Chem. A, Vol. 101, No. 30, 1997 5545
shown in reaction sequence 1 will be assumed to be dissociation
to the reaction products Br- and CH3Cl. From the rate constants
for this process also provided by Seeley et al., this step can be
assumed to occur very quickly (the half-life of the complex in
nitrogen buffer gas at 125 °C will be much less than 1 µs18).
Therefore, the ion complex will not be detected in the present
IMS-based experiments and the dissociation step will not affect
the overall observed rate constant for this pathway.
For the sake of this discussion, several simplifications of the
detailed reaction dynamics represented by reaction sequence 1
will be made and the possible implications of these assumptions
should be pointed out. One of these is that the rate at which
the initially formed set of entrance-channel ion complexes
undergo backdissociation will be described here by a single rate
constant kr. It is acknowledged, however, that a more rigorous
treatment of the backdissociation process would involve a sum
of several exponential decay constants representing all of the
important mode-specific species initially formed in the ion-
molecule collision. In the present study, it is possible that only
a subset of the entrance channel complexes that have relatively
long lifetimes (smallest kr’s) are being affected by increased
buffer gas pressure. Therefore, the simplified model developed
here can be expected to provide only a semiquantitative
description of the actual events and has validity only to the
extent that the relatively long-lived fraction of the collision-
formed complexes constitutes a significantly large portion of
the total population.
Another point of simplification in reaction sequence 1 is that
the passage of Cl-(CH3Br)* over the transition state barrier is
shown to lead directly to products. Wang et al.14 have shown,
however, that the exit-channel complex, Br-(CH3Cl)*, that is
first formed upon passage over the transition state is likely to
recross the transition-state barrier to reform Cl-(CH3Br)*. In
addition, Wang et al. predict that several such recrossings of
the central barrier will then occur prior to the dissociation of
one of the two complexes to either reactants or to products.
Therefore, the rate constant kp shown in reaction sequence 1
should be viewed to represent the passage over the central barrier
only of those entrance-channel complexes that do not recross
the barrier. In addition, it is recognized that our use here of a
single value for kp is also a simplification of a much more
complex reality in which a set of mode-specific rate constants
would actually be required in a rigorous treatment of the forward
motion of the excited entrance-channel complexes along the
reaction coordinate.
methane buffer gases is not known. If it is first assumed that
kq will be equal to the rate constant for collisions between the
ion complex and the buffer gas molecules, Langevin theory19
predicts kq ) 1.00 × 10-9 cm3 s-1 for methane buffer gas and
kq ) 6.5 × 10-10 cm3 s-1 for nitrogen buffer gas. The best fit
to the experimental data obtained at 640 Torr was then obtained
when the lifetime of the entrance-channel complex was assumed
to be about 0.7 ps (i.e., kr ) 1.5 × 1012 s-1). These predictions
are shown as the solid (methane) and dashed (nitrogen) lines
in Figure 1.
If the efficiency for collisional quenching of the entrance-
channel ion complex is less than unity, the value of kr ) 1.5 ×
1012 s-1 would have been overestimated in the above treatment.
For example, if the quenching efficiency of one of the buffer
gases is only 10%, then the best fit of kr to the data in Figure
1 would be about kr ) 1.5 × 1011 s-1 (corresponding to a
complex lifetime of 7 ps). For this reaction system, it seems
reasonable to speculate that the efficiency of quenching by
methane buffer gas would be at least 10%.20 Therefore, it seems
reasonable to conclude that the 1/kr lifetime suggested by the
present set of measurements and the above simple model for
the reaction dynamics operative at 640 Torr is somewhere in
the low picosecond range.
If all of the rate constants for the Cl-/CH3Br reaction system
measured over the entire pressure range between 300 and 1100
Torr (Figure 1) are considered, significant differences between
the values measured in both buffer gases and the predictions of
the simple model are apparent. Both sets of experimental values
form lines of lower slopes than those of the predictions and
extrapolate to low-pressure limits that lie significantly above
the experimentally determined value of kLP ) 2.2 × 10-11 cm
s-1. Although the cause of this discrepancy is not presently
known, it could result from limitations of the simplified model
described above. For example, if the simplified model was
altered only slightly in a manner that allowed just 1% of the
initially formed complexes in reaction sequence 1 to have
distinctly longer lifetimes (i.e., 1/kr > 50 ps) against backdis-
sociation, an initial increase of about 1 × 10-11 cm3 s-1 in the
rate constant would then be expected for pressures less than
300 Torr. This change would account for the observed
extrapolations to higher-than-expected low-pressure limits for
the rate constants. Within this modified view, the slopes of
the measurements shown in Figure 1 over the pressure range
300-1100 Torr would then lead to deductions for kr that are
about 5 times greater than those derived above from the
simplified model (these modified kr values would apply only
to the set of collision complexes that are being affected by the
pressure change from 300 to 1100 Torr and not to the set that
are affected by lower pressures). To more completely charac-
terize the dynamics of this and other reaction systems, experi-
mental methods that are operative over all pressure ranges of
interest are clearly needed. For this purpose, an IMS-based
apparatus that will operate over the pressure range from about
10 to 300 Torr is presently being constructed in our laboratory.
Within this simplified view of reaction sequence 1 and with
the reasonable assumption that kr . (kp + kq[B]) under our
experimental conditions, the observed rate constant, kobs, would
be given by
kobs ) kckp/kr + kckq[B]/kr
(2)
Since the rate constant observed for this reaction under buffer
gas conditions of relatively low pressure will be given by kLP
) kckp/kr, eq 3 can be written as
Conclusions
kobs ) kLP + kckq[B]/kr
(3)
The experiments performed here indicate that the reaction
of chloride ion with methyl bromide is not moved onto its high-
pressure limit of kinetic behavior by use of buffer gas pressures
near 1 atm. The pressure dependence of the rate constants
observed in this pressure range is qualitatively consistent with
predictions by Wang et al.14 of very short lifetimes for the
reaction complexes of this reaction and with the prediction by
Seeley et al.18 that this reaction should proceed with high
efficiency in its high-pressure limit.
Also shown in Figure 1 are predictions of kobs based on eq 3
where the following values of kLP, kc, kq, and kr were selected.
From prior measurements12 of this reaction by PHPMS in 3
Torr buffer gas at 125 °C, kLP was determined to be equal to
2.2 × 10-11 cm3 s-1. By ADO theory,19 kc is predicted to be
equal to 1.42 × 10-9 cm3 s-1 at 125 °C. Accurate predictions
of kq are more difficult to obtain because the efficiency of
collisional quenching of the excited complexes by nitrogen and