UV Photolysis of ClOOCl

Article abstract of DOI:10.1021/jp984410+

ClOOCl (ClO dimer) photolysis is believed to dominate the catalytic destruction of polar stratospheric ozone during springtime through the production of atomic chlorine. Decomposition by an alternate pathway to form ClO would not catalyze ozone loss. Molecular beam experiments have demonstrated that photoexcitation of ClOOCl at both 248 and 308 nm leads to dissociation via multiple dynamical pathways, producing ClO + ClO and 2Cl + O2. At 248 nm, both concerted and sequential dissociation to 2Cl + O2 were observed. The primary dissociation channels occurred within a rotational period at both excitation wavelengths. The relative Cl:ClO product yields are 0.88:0.12 and 0.90:0.10 at 248 and 308 nm, respectively. Lower limits on these ratios were determined. These results substantially confirm the importance of ClOOCl photolysis in catalyzing springtime polar ozone depletion.

Full text of DOI:10.1021/jp984410+

©
Copyright 1999 by the American Chemical Society
VOLUME 103, NUMBER 12, MARCH 25, 1999
LETTERS
UV Photolysis of ClOOCl
Teresa A. Moore and Mitchio Okumura*
Arthur Amos Noyes Laboratory of Chemical Physics, Mail Stop 127-72, California Institute of Technology,
Pasadena, California 91125
James W. Seale and Timothy K. Minton*
Department of Chemistry and Biochemistry, Montana State UniVersity, Bozeman, Montana 59717
ReceiVed: NoVember 11, 1998; In Final Form: February 2, 1999
ClOOCl (ClO dimer) photolysis is believed to dominate the catalytic destruction of polar stratospheric ozone
during springtime through the production of atomic chlorine. Decomposition by an alternate pathway to form
ClO would not catalyze ozone loss. Molecular beam experiments have demonstrated that photoexcitation of
ClOOCl at both 248 and 308 nm leads to dissociation via multiple dynamical pathways, producing ClO +
ClO and 2Cl + O
2
2
. At 248 nm, both concerted and sequential dissociation to 2Cl + O were observed. The
primary dissociation channels occurred within a rotational period at both excitation wavelengths. The relative
Cl:ClO product yields are 0.88:0.12 and 0.90:0.10 at 248 and 308 nm, respectively. Lower limits on these
ratios were determined. These results substantially confirm the importance of ClOOCl photolysis in catalyzing
springtime polar ozone depletion.
The large springtime depletion of stratospheric ozone over
the Poles is thought to be catalyzed primarily by the photolysis
of ClOOCl. Chlorine atoms released through ClOOCl dissocia-
form ClO products, then the consequent null cycle would not
deplete ozone. The branching ratio between Cl and ClO products
from ClOOCl photolysis is therefore critical in determining the
effectiveness of this mechanism for destroying ozone.
1
tion participate in ozone loss through the photochemical cycle:
ClOOCl has a strong, broad absorption centered at 245 nm.
Absorption in the 305-400 nm tail leads to photolysis in the
ClO + ClO + M f ClOOCl + M
ClOOCl + hν f ClOO + Cl
(1)
(2)
(3)
2
stratosphere during the polar spring. When ClOOCl is excited
in the UV (Ephot ) 265-500 kJ/mol), it can dissociate via
ClOO + M f Cl + O + M
3
2
several channels, including
2
[Cl + O f ClO + O ]
(4)
3
2
ClOOCl + hν f ClO + ClO ∆H ° ) 71.5 kJ/mol
(5)
(6)
0
net: 2O f 3O2
f ClOO + Cl
f 2Cl + O2
) 88.7 kJ/mol
3
) 108.4 kJ/mol (7)
) 336.8 kJ/mol (8)
In current models, this catalytic cycle accounts for 70-80% of
the ozone loss over Antarctica. However, if photolysis of
ClOOCl results in cleavage of the weakest (peroxy) bond to
f Cl + O + ClO
4
The NASA Data Evaluation Panel currently recommends a unit
quantum yield for reaction 6, based on high quantum yields for
*
To whom correspondence should be addressed.
1
0.1021/jp984410+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/04/1999

1
692 J. Phys. Chem. A, Vol. 103, No. 12, 1999
Letters
4
total Cl-atom production (ΦCl) observed in two experiments.
Cox and Hayman examined this process indirectly by steady-
beam velocity was typically 9.5 × 10 cm/s. Mass spectrometric
5
13
analysis confirmed that approximately 1.5 Torr of ClOOCl
state photolysis of Cl2O at 254 nm in a static cell. They inferred
the presence of ClOOCl from a kinetics model and deduced
ΦCl ) 3.8 ( 1.6 for ClOOCl photolysis. Molina et al. used
was formed. However, Cl2, Cl2O, and ClO were also present
with average abundances of 6, 0.4, and 0.06 relative to ClOOCl.
Although there was a strong signal at OClO , crossed-beam
6
+
resonance fluorescence to measure the Cl-atom yield directly
from ClOOCl photolysis at 308 nm in a flow tube and found
ΦCl ) 2.06 ( 0.24. They detected no ClO products with an
NO titration method, consistent with the observed Cl yield.
However, the calculation of ΦCl depended upon the ratio of
the ClOOCl absorption cross sections at 245 and 308 nm, for
which literature values3 vary from 13 to 22. If recalculated,
ΦCl could range from 1.2 to 2.06. Both groups reported no
evidence for the ClOO intermediate, and neither could distin-
guish between final product channels 2 and 7. These uncertain-
ties are significant enough to alter quantitatively model predic-
tions of polar stratospheric ozone loss.
nonreactive scattering measurements with Ne confirmed that
this signal arose from ionizer fragmentation of Cl2O2 or heavier
+
+
species. A weak Cl2O3 signal (<6% of Cl2O2 ) indicated the
presence of either small quantities of Cl2O3 or low relative
concentrations of clusters. These results corroborated model
predictions of insignificant OClO and Cl O concentrations.
2
3
,7
Light at 248 and 308 nm can photolyze the three major
impurities (Cl2, Cl2O, and ClO) to form Cl and ClO photofrag-
ments. We performed separate photolysis experiments on Cl2
and Cl2O at both wavelengths to obtain their translational energy
and angular distributions, which allowed us to model their
contributions to the observed photolysis signal. Photolysis of
any clusters would likely result in Cl or ClO photofragments
with low c.m. translational energies (ET).
The 245 nm absorption band arises from transitions to
2
antibonding σ* (ClO) orbitals, which led to suggestions that
ClOOCl photolysis would result only in Cl-O bond fission.
Examples of TOF spectra are shown in Figure 1 for 248 nm
and Figure 2 for 308 nm excitation. The signal-to-noise ratio at
However, a bond-selective model of photodissociation can be
8
too simplistic. In a study of ClONO2 photolysis, we showed
3
08 nm was less than that at 248 nm because the relative
that cleavage of the weaker ClO-NO2 bond competes ef-
fectively with Cl-O bond fission following excitation to a σ*
orbital localized on the ClO chromophore. On the basis of these
results, we raised the possibility that there might be a significant
yield of ClO products in ClOOCl photolysis.
7
ClOOCl absorption cross section is ∼18 times lower. Regard-
less of excitation wavelength, the largest signal was observed
+
at m/z ) 35 (Cl ) and a significant signal was still seen at m/z
)
+
+
51 (ClO ). No products were detected at m/z ) 67 (ClO2 ).
+
We also observed a fast O2 signal at 248 nm.
We report here a molecular beam experiment on the pho-
tolysis of ClOOCl at two wavelengths, 248 and 308 nm, which
employed photofragment translational spectroscopy as in our
ClONO2 studies. We have directly observed Cl, ClO, and O2
photofragments by mass spectrometric detection. The relative
yields of Cl and ClO products have been determined by
calibrating the relative detector sensitivity for ClO and Cl with
a Cl2O photolysis experiment.9
The ClO+ TOF distributions (Figure 1A) obtained in the
experiment at 248 nm consisted of overlapping signals from
ClO products of Cl2O and ClOOCl photodissociation. Although
+
the data looked similar to the ClO signal predicted for Cl2O
alone, the Cl2O photoproduct signal was somewhat faster than
that in the observed TOF spectrum, as shown in Figure 3. By
matching the predicted Cl2O product signal to the rising edge
of the observed signal, we found that Cl2O photolysis accounted
+
+
The experiments were performed on a crossed molecular
beams apparatus10 modified for laser photolysis. A molecular
beam of ClOOCl was crossed at right angles with the UV
radiation from a pulsed excimer laser focused at the interaction
region. The photolysis products were detected off the beam axis
by a mass spectrometer that is rotatable in the plane of the laser
and molecular beams. Product time-of-flight (TOF) distributions
were collected at various detector angles for each photofragment
mass-to-charge ratio (m/z) as a function of time. A forward
for ∼50% of the ClO signal. The remaining ClO signal gave
a bimodal peak (dashed line, Figure 1A) which we assigned to
ClO products from ClOOCl + hν f ClO + ClO with an
int
average total internal energy 〈E 〉 of 310 kJ/mol and anisotropy
+
parameter â ) 0.5. A slower ClO peak was also observed,
1
4
which we assigned to a minor channel: ClO(X) + ClO(A) f
3
15
ClO(X) + Cl + O( P). There were several contributions to
+
the Cl TOF distributions (Figure 1B) at 248 nm, but only a
small fraction of the signal could be attributed to photolysis of
8
16
convolution technique was used to determine the center-of-
impurities. Photolysis of ClO gave rise to a single peak (thin
+
mass (c.m.) translational energy and angular distributions for
each two-body dissociation channel. The angular distribution
is characterized by the anisotropy parameter â,11 which can
range from â ) -1 for a transition with the dipole perpendicular
to the recoil vector to â ) 2 for a pure parallel transition.
solid line, Figure 1B). We determined the Cl contribution from
photolysis of Cl O (short-dashed line, Figure 1B) from the
2
+
+
known Cl :ClO relative intensities for pure Cl O photolysis.
2
We also accounted for the signal at this m/z from the ClO
products from reaction 5, which underwent fragmentation to
+
17
+
Production of a beam of ClOOCl was key to the feasibility
of this experiment, which required partial pressures of ClOOCl
Cl in the ionizer. We modeled the remaining Cl signal as
Cl-atom products from ClOOCl photolysis. The main peak (160
µs, dot-dot-dashed line, Figure 1B) matched the expected TOF
distribution for Cl atoms recoiling with a total momentum equal
(pClOOCl ≈ 1-2 Torr) 1-2 orders of magnitude higher than had
been reported in the literature. We formed ClOOCl by recom-
bination of ClO (eq 1). This method is known to produce pri-
marily ClOOCl rather than other Cl2O2 isomers.1 After ex-
tensive photochemical modeling, we chose to generate high
transient concentrations of ClO radicals (pClO > 3 Torr) by 308-
nm photolysis of Cl2O in the high-pressure stagnation region.
Cl2O (10 Torr) was seeded in a buffer of 2% SF6 in He at a
total pressure of 190 Torr. The initial temperature was kept
below 220 K to prevent the ClO self-reaction. Modeling indi-
cated that most of the ClO reacted during the 300-500 µs resi-
dence time prior to expansion through a 100 µm nozzle. The
+
and opposite to that of the O2 fragments (105 µs, Figure 1C)
2
if we assumed that both Cl-O bonds broke nearly simulta-
neously. We therefore assigned this peak and the corresponding
+
O2 signal to the concerted dissociation to 2Cl + O2. The best
fits were obtained with an assumed dissociation geometry similar
to the ground-state geometry but with a smaller dihedral angle
(65° instead of 81°). The high translational energy release, on
average 72% of the available energy, is characteristic of a
1
8
concerted process. The angular distributions gave anisotropy
parameters âO ) -1.0 and âCl ) 0.1. For a primary two-body
2

Letters
J. Phys. Chem. A, Vol. 103, No. 12, 1999 1693
Figure 2. TOF distributions for products from ClOOCl photolysis at
3
08 nm. TOF spectra were collected at a detector angle of 40° for (A)
+
+
ClO and (B) Cl fragments. The contribution in B from the ClO
product channel arises from fragmentation to Cl in the ionizer.
+
Figure 1. Time-of-flight (TOF) distributions for products from ClOOCl
photolysis at 248 nm. TOF spectra were collected at a detector angle
+
+
+
2
of 30° for (A) ClO , (B) Cl , and (C) O
fragments. The contribution
in B from the ClO + ClO product channel arises from fragmentation
+
to Cl in the ionizer, and the contribution from the ClO + ClO(A)
channel arises from predissociation of the ClO(A).
dissociation, the â parameters should be identical. The fact that
they are different and of opposite sign is strong evidence that
the dissociation is concerted.19 After subtracting this main peak,
both a faster and slower signal remained. We assigned the faster
peak to primary Cl atoms from the Cl + ClOO channel (long-
dashed line, 125 µs, Figure 1B). The angular distribution gave
Figure 3. ClO+ TOF distribution of products photolyzed at 248 nm
and collected at a detector angle of 30°. The solid line denotes the fit
+
if the ClO products arise solely from Cl
2
O photolysis.
+
an anisotropy parameter of â ) 0.6. No ClOO signal was
+
observed, indicating that all ClOO products dissociated. This
result is not surprising, because the internal energy of the ClOO
fragment (〈Eint〉 ) 276 kJ/mol) was far in excess of its
dissociation energy (D0 ≈ 20 kJ/mol). We modeled the
secondary dissociation20 of ClOO assuming formation of
ground-state Cl + O2 products to obtain the contribution to the
slow shoulder of the main Cl peak (long-dashed line, 185 µs,
Figure 1B). The observed spectra could not be fit using RRKM
+
theory. Finally, the slowest peak in the Cl distribution (dotted
lines, Figure 1A) was modeled as a unimolecular decay21 via a
1
Cl + ClOO* channel to Cl and electronically excited O2 ( ∆)
products.

1
694 J. Phys. Chem. A, Vol. 103, No. 12, 1999
TABLE 1: Relative Yields for the ClOOCl Photolysis Channels Observed at 248 and 308 nm
48 nm
Letters
2
308 nm
relative yield
relative yield
â
â
Dynamical Channels
i. Cl + ClOO f Cl + Cl + O
2
0.55
0.18
0.6
â 2
0.5
1.5
ii. Cl + O
2
+ Cl (concerted)
O
) -1.0
â
Cl ) 0.1
iii. ClO + ClO
0.09
0.14
0.04
0.5
0.0
0.0
0.1
0.4
2.0
0.0
2
iv. Cl + ClOO* f Cl + Cl + O *
v. ClO + ClO* f ClO + Cl + O
Relative Product Yieldsa
2
Cl + O
ClO + ClO
branching ratio Cl:ClO:O
2
0.88 ( 0.07 (0.89)
0.12 ( 0.07 (0.11)
1.0:0.15:0.02
0.90 ( 0.1 (0.83)
0.10 ( 0.1 (0.17)
1.0:0.11:0.0
†
(1.0:0.13:0.00)
(1.0:0.20:0.0)
a
Yields and ratios in parentheses are calculated assuming that products with low center-of-mass translational energies (c.m. E
T
) are assigned to
cluster photodissociation rather than channels iv and v.
We observed contributions to the photolysis signal at 308
nm from ClO, Cl2, Cl2O, and ClOOCl as shown in Figure 2.
We could not unambiguously determine the contributions from
products are formed by cluster dissociation; however, these are
minor products whose assignments have little effect on the Cl:
ClO yields. If we attribute the low c.m. ET products to clusters
rather than channels iv and v (Table 1), the resulting product
yields and branching ratios (given in parentheses in Table 1)
are unaffected within the experimental uncertainty.
2
2
ClO or Cl2O photolysis based on the TOF data at either m/z.
+
The Cl TOF spectrum was dominated by dissociation of Cl2
(long-dashed line, Figure 2B). We therefore estimated the Cl2O
(
short-dashed line) and ClO (dot-dot-dashed lines) contribu-
The observation of multiple dissociation pathways, including
peroxy-bond fission, suggests that a simple bond-selective
picture is inadequate. More striking is the occurrence at 248
nm of both concerted and sequential cleavage of the two Cl-O
bonds. Furthermore, we conclude from the anisotropies of the
tions relative to the Cl2 peak (long-dashed line) based on
abundances determined by mass spectrometer measurements.
The Cl2O contribution to the ClO signal was then obtained
from known Cl :ClO relative intensities. This analysis was
consistent for different source conditions under which the
relative abundances varied by as much as a factor of 4.
Significant ClO+ signal remained (∼30-60%), which we
assigned to ClOOCl + hν f ClO + ClO (dotted line, Figure
2
3
+
+
+
two 2Cl + O channels that different electronic states are
2
initially excited. Several excited states have been predicted, but
more detailed calculations of the excited-state surfaces are
needed to explain the observed dynamics.
The ClO yield in Table 1 is based on our best estimate of
2
2
)
A). The ClO products were vibrationally excited (〈total Eint〉
201 kJ/mol). The angular distribution indicated a parallel
the signal from Cl O photolysis, but we can place a strict upper
+
transition (â ) 2.0).
limit on the ClO yield if we assume that all of the ClO signal
+
Cl signal also remained after accounting for contributions
is from ClOOCl f ClO + ClO. In this case, the relative ClO:
from photolysis of impurities and fragmentation in the ionizer
of ClO products. We assigned the fast shoulder at 135 µs (dotted
line, Figure 2B) to primary Cl from the ClOOCl f Cl + ClOO
channel and the slower broader signal to Cl atoms from the
secondary process ClOO f Cl + O2 (dotted line, Figure 2B).
Cl branching ratio can be no greater than 0.19:0.81 at 248 nm
2
5
and 0.31:0.69 at 308 nm. Hence, ClO + ClO is at most a
minor product channel from ClOOCl photolysis at 248 and 308
+
nm. Alternatively, we can assign all of the ClO signal to Cl2O
photolysis, but the poor fits indicate that this assignment is
unlikely.
+
We again interpreted the absence of ClO2 product signal as
an indication that the ClOO counterfragment, formed with 〈Eint〉
Photolysis of other isomers of Cl O could conceivably
2
2
)
184 kJ/mol, spontaneously dissociated. The angular distribu-
contribute to the observed signal; however, ClOOCl is the most
tion for the primary Cl gave â ) 1.5. High background at m/z
stable isomer, with chloryl chloride (ClClO ) and chlorine
2
+
26
)
32 (O2 ) prevented us from identifying a concerted 2Cl +
chlorite (ClOClO) less stable by 4 and 40 kJ/mol, respectively.
O2 channel. A slow signal was observed at m/z ) 35 but not at
Furthermore, we produce ClOOCl at high pressures by the three-
+
12
m/z ) 51. This slow Cl signal, centered at 330 µs, was much
body recombination reaction 1, which is known to generate
2
7
larger in proportion to the fast Cl + ClOO signal than it was at
predominately the ClOOCl isomer. Although we cannot
entirely rule out the presence of other isomers, the source
chemistry favors ClOOCl as the dominant isomer present in
our experiment.
2
48 nm. We again modeled this signal as a statistical (RRKM)
dissociation to Cl + ClOO* f Cl + O2*.
Table 1 gives the relative yields of the primary dissociation
24
Our results support the previous reports5,6 that the stronger
Cl-O bond is preferentially broken and that Cl atoms are the
dominant products from UV photolysis of ClOOCl. The absolute
quantum yield determined by Molina et al. at 308 nm depends
on the ClOOCl absorption cross section. The most recent
channels for both wavelengths. Although we observe products
from multiple dissociation pathways, ultimately there are two
main product channels: 2Cl + O2 and ClO + ClO. Their relative
yields may be interpreted as absolute quantum yields, because
excitation into strong, broad UV absorption bands typically
results in dissociation with unit quantum efficiency. As seen
by the Cl:ClO product ratios (Table 1), chlorine atoms are the
dominant products (> 85%) upon photolysis at both wave-
lengths. A small O-atom yield at 248 nm is inferred from ClO-
7
absorption cross-section measurement results in a recalculated
yield for the 2Cl + O channel (7) at 308 nm of Φ ) 0.88, in
2
7
good agreement with our results (Φ ) 0.9). Furthermore, the
7
high ClO vibrational excitation that we observe could explain
(
A) predissociation from the possible ClO(X) + ClO(A) channel.
Although we assign the products with low c.m. ET to ClOOCl
their inability to detect ClO products. Recent experiments show
that their detection method, titration with NO, is less efficient
than assumed when ClO is vibrationally excited.
2
8
photolysis channels, we cannot exclude the possibility that these

Letters
We conclude from our results at 308 nm that in the
J. Phys. Chem. A, Vol. 103, No. 12, 1999 1695
(11) The anisotropy parameter â is given by â ) 2P2(cos ø), where P2
is the second Legendre polynomial and ø is the angle between the electronic
transition moment and the center-of-mass recoil axis of the fragments.
Busch, G. E.; Wilson, K. R. J. Chem. Phys. 1972, 56, 3626.
(12) Schwell, M.; Jochims, H.-W.; Wassermann, B.; Rockland, U.;
Flesch, R.; R u¨ hl, E. J. Phys. Chem. 1996, 100, 10700.
stratosphere
ClOOCl + hν f 2Cl + O2
Φ ) 0.9 ( 0.1
7
(7)
(5)
f ClO + ClO Φ ) 0.1 ( 0.1
5
(13) The nominal electron energy was lowered to 30 eV in order to
reduce fragmentation.
3
(
14) ClO(A) spontaneously predissociates to produce Cl and O( P) atoms.
with an absolute upper limit of Φ5 ) 0.31 for the ClO product
channel. The intermediate ClOO (reaction 6) is formed with
Eint . D0 and likely dissociates spontaneously under strato-
spheric conditions. The high Cl yield we observed implies that
the catalytic cycle involving photolysis of ClOOCl is an
important mechanism for ozone loss under perturbed polar
conditions. Further modeling studies are necessary to determine
whether the small ClO yield observed would require additional
ozone loss mechanisms to balance the ozone budget, especially
in the Arctic vortex where modeling underestimates ozone loss
rates.
(
15) The translational energy distribution was obtained by a fit to an
RRKM mechanism with â ) 0.
(16) Davis, H. F.; Lee, Y. T. J. Phys. Chem. 1996, 100, 30. To fit our
ClO photolysis signal (the thin solid line at 240 µs in Figure 1B) at 248
nm, a value of â ) 2.0 was needed.
(
17) The fragmentation of ClO from Cl2O photolysis was determined
based on the pure Cl2O photolysis experiments, which had three ClO
components with differing average internal energies. The fragmentation of
ClO from ClOOCl photolysis was determined based on the average internal
energy of the ClO and knowledge of the Cl2O fragmentation pattern.
(
(
18) Maul, C.; Gericke, K.-H. Int. ReV. Phys. Chem. 1997, 16, 1.
19) Baum, G.; Felder, P.; Huber, R. J. J. Chem. Phys. 1993, 98, 1999.
Wannenmacher, E. A. J.; Felder, P.; Huber, J. R. J. Chem. Phys. 1991, 95,
86.
9
(
20) At m/z ) 32, the signal-to-noise ratio was very low due to high O2
Acknowledgment. This work was supported by NASA/
UARP Grant Nos. NAGW-3893 and NAG5-3911; calculations
were performed with support of the JPL Supercomputing
Project. We are grateful to Christine M. Nelson, Donna J.
Garton, and Thomas S. Schindler for experimental assistance.
background. The slower, broader signal from the secondary dissociation of
ClOO was not detectable.
(
21) An RRKM calculation was used to model this dissociation pro-
cess.
22) At 308 nm, ClO photodissociated to Cl( P3/2,1/2) + O( P) and was
observed at 195 µs in Figure 2B.
23) We estimated the relative abundances of ClO, Cl2, and Cl2O from
2
3
(
(
References and Notes
the beam mass spectrometric measurements, their relative electron impact
ionization cross sections, and their ionizer fragmentation patterns (observed
in Cl2 and Cl2O photolysis experiments and estimated for ClO). The
expected contributions to the Cl photolysis signal were then determined
from the relative absorption cross sections at 308 nm.
(
(
(
1) Molina, L. T.; Molina, M. J. J. Phys. Chem. 1987, 91, 433.
2) Stanton, J. F.; Bartlett, R. J. J. Chem. Phys. 1993, 98, 9335.
3) ∆Hf°(0K)ClOOCl ) 31.2 kcal/mol from Nickolaisen, S. L.; Friedl,
+
R. R.; Sander, S. P. J. Phys. Chem. 1994, 98, 155.
(24) The relative yield for the concerted channel was determined from
(4) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
a primary c.m. flux distribution that was obtained by modeling the Cl
products as arising from a primary two-body channel.
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling, EValuation Number 12; JPL Publication 97-4, JPL, 1997.
(25) Some ClO photofragments could go undetected if they absorbed a
second photon and dissociated, but the photolysis laser fluence was kept
low enough to minimize this process. Under these conditions, little ClO
photodissociation was observed in the Cl2O photolysis experiments.
(
(
5) Cox, R. A.; Hayman, G. D. Nature 1988, 332, 796.
6) Molina, M. J.; Colussi, A. J.; Molina, L. T.; Schindler, R. N.; Tso,
T.-L. Chem. Phys. Lett. 1990, 173, 310.
(
26) Lee, T. J.; Rohlfing, C. M.; Rice, J. E. J. Chem. Phys. 1992, 97,
(
(
7) Huder, K. J.; DeMore, W. B. J. Phys. Chem. 1995, 99, 3905.
8) Minton, T. K.; Nelson, C. M.; Moore, T. A.; Okumura, M. Science
6
593. Stanton, J. F.; Rittby, C. M. L.; Bartlett, R. J.; Toohey, D. W. J.
Phys. Chem. 1991, 95, 2107. McGrath, M. P.; Clemitshaw, K. C.; Rowland,
F. S.; Hehre, W. H. Geophys. Res. Lett. 1988, 15, 883. Jensen, F.;
Oddershede, J. J. Phys. Chem. 1990, 94, 2235.
1
992, 258, 1342. Moore, T. A.; Okumura, M.; Tagawa, M.; Minton, T. K.
Faraday Discuss. Chem Soc 1995, 100, 295. Nelson, C. M.; Moore, T. A.;
Okumura, M.; Minton, T. K. Chem. Phys. 1996, 207, 287.
(
9) Nelson, C. M.; Moore, T. A.; Okumura, M.; Minton, T. K. J. Chem.
(27) While the ClOClO isomer could form from ClO clustering in the
nozzle expansion, the ClO concentration in the stagnation region was so
low (∼30 mTorr) that this process can be neglected.
Phys. 1994, 100, 8055. Moore, T. A.; Okumura, M.; Minton, T. K. J. Chem.
Phys. 1997, 107, 3337.
10) Lee, Y. T.; McDonald, J. D.; LeBreton, P. R.; Herschbach, D. R.
ReV. Sci. Instrum. 1969, 40, 1402.
(
(28) Tyndall, G. S.; Kegley-Owen, C. S.; Orlando, J. J.; Calvert, J. G.
J. Chem. Soc., Faraday Trans. 1997, 93, 2675.