2934 J. Phys. Chem. A, Vol. 105, No. 13, 2001
Letters
TABLE 1: Upper Limits for the Reactions of CIs with a
Number of Molecular Species
reactant H2O
SO2
butanone acetic acid
[S]max/molecule cm-3 5 × 1017
1 × 1016
2.5 × 1015 4 × 1015
k/cm3 molecule-1 s-1 e1 × 10-16 e4 × 10-15 e2 × 10-14 e1 × 10-14
the reactions of CIs with these types of species cannot inhibit
OH formation in the reaction of ozone with 2-methylbut-2-ene
under atmospheric conditions. At 295 K, the maximum con-
centration of H2O used in our experiments (ca. 20 000 ppm)
can be exceeded in the atmosphere by some 25%, but this small
increase cannot have a significant effect on the OH yield. At
higher temperatures, much higher concentrations are sometimes
experienced, e.g., at 313 K, concentrations of water vapor of
ca. 70 000 ppm are possible. However, the barrier to OH
formation from syn CIs has been calculated11 at approximately
60 kJ mol-1. Consequently, the rate of formation of OH from
the stabilized CI is expected to be about a factor of 4 faster at
313 K than it is at the temperature of the experiments described
here. The rate constant for the bimolecular reaction between
the CI and water may also be temperature dependent, but
because this reaction is believed to be an addition reaction, the
temperature dependence is likely to be weak; it is very difficult
to envisage how the bimolecular reaction could have a stronger
temperature dependence than the unimolecular decomposition.
At temperatures below 295 K, unimolecular decomposition is
less effective but concentrations of H2O are necessarily lower,
so that, competition from the bimolecular reaction can only be
significant if the reaction has a strong negative temperature
dependence. It therefore seems likely that, under atmospheric
conditions, the formation of OH from the reaction of ozone with
2-methylbut-2-ene is not inhibited by the presence of water
vapor.
It is important to know to what extent OH formation in the
reactions of ozone with other alkenes is inhibited by the presence
of CI scavengers. Although a small amount of OH is formed
from the decomposition of the simplest CI, CH2OO, most OH
is formed via the decomposition of more complex types, such
as syn-CH3CHOO, (CH3)2COO, and structurally related ana-
logues. The reaction of ozone with 2-methylbut-2-ene generates
both types of CI, and so it is reasonable to suppose that the
conclusions drawn about OH formation in this reaction can be
extended to the reactions of ozone with other alkenes. This
statement brings us to the major conclusion of this letter: the
currently accepted OH yields for the reactions of ozone with
alkenes are applicable to atmospheric conditions.
Figure 1. Plot of R vs kT[T]i/(kT[T]i + kA[A]i + kS[S]i) for the
ozonolysis of 2-methylbut-2-ene. 0, DMB, no CI scavenger; ×, DMB,
20 000 ppmv H2O; O, DMB, 100 ppmv SO2; b, DMB, 500 ppmv SO2;
4, DMB, 100 ppmv butanone; [, DMB, 150 ppmv acetic acid; 9,
TMB tracer experiments also included.
III. Experimental Section
The experimental apparatus employed for this study com-
prised a static reaction chamber with attached gas chromatograph
(GC) with flame ionization detection (FID). A mixture of
2-methylbut-2-ene and the relevant OH tracer and/or CI
scavenger (DMB, H2O, SO2, butanone, or acetic acid) was
prepared in a 50 L collapsible Teflon chamber using dry
synthetic air (BOC Gases, BTCA 74) as the diluent gas. Water
vapor was introduced into the mixture by passing the synthetic
air through three Dreschel bottles containing deionized water.
Typical initial hydrocarbon concentrations employed were 10
ppmv of the alkene and 10-50 ppmv of DMB. Typical initial
ozone mixing ratios ranged from ca. 0.5 to 8 ppmv. Experiments
were carried out by admitting a known concentration of ozone,
to a pressure of ca. 8 Torr, into a 0.5 L borosilicate glass reaction
chamber and adding a sample of the hydrocarbon mixture such
that a total pressure of 1 atm (760 ( 10 Torr) was effected.
Experiments were carried out at 295 ( 2 K. Ozone was
generated as a mixture in O2 by passing oxygen through a
Fischer ozone generator, its concentration being determined
spectrophotometrically by absorption at λ ) 254 nm. After the
mixture was left for sufficient time for the ozone to react, the
contents of the glass bulb were separated and detected by GC-
FID (Perkin-Elmer, model 8420). This procedure was typically
repeated for six different initial ozone concentrations during each
study. Chromatographic peak heights were related to concentra-
tions after calibration with pure standards. A 25 m × 0.53 mm
diameter Poraplot Q capillary column was used. A typical
temperature program employed held the column isothermally
at 200 °C for 10 min. Sulfur dioxide was supplied by Aldrich
(purity > 99.9%) and was used without further processing; all
other reagents employed were of analytical grade and underwent
a freeze-pump-thaw cycle before being used.
The measurements of Kroll et al. place limits on the lifetime
of the stabilized CI. Limits can also therefore be placed on the
rate constants for the bimolecular reactions of CIs with the
molecular species used in our experiments. The Kroll et al.
measurements were made at ca. 10 µs. If the lifetime for the
formation of OH from the stabilized CI were shorter than this
value, the OH yields would be expected to rise at higher
pressures, the opposite of their observations. The only reaction
for which the results indicate that this effect was observed is
for the reaction of ozone with 2,3-dimethylbut-2-ene, but as
the authors explain, at the higher pressures, the measurements
were made at reaction times longer than 10 µs. The scatter in
Figure 1 indicates that a reduction in OH yield of about 30%
IV. Results and Discussion
Figure 1 shows a plot of R vs kT[T]/(kT[T] + kA[A] + kS[S])
for all experiments carried out in this study. Also included are
the results of already-published experiments carried out using
TMB as the tracer.18 What the figure shows is that the measured
OH yield is unaffected by the presence of H2O, SO2, butanone,
and acetic acid, all of which are known to react with stabilized
CIs. The concentration of the latter three species (g100 ppm)
was far in excess of their concentrations (or the concentrations
of related compounds) in the atmosphere. It is clear, then, that
would be clearly detectable, and implies that kS[S] e 35 s-1
.
On this basis, upper limits for the bimolecular rate constants
for the reactions of the CI with the scavenger molecules used
in this study can be obtained and are listed in Table 1. Rate
constants for the reactions of CIs with the molecules studied
here have been quoted in the literature. However, the studies