Quantum yields and energy partitioning in the ultraviolet photodissociation of 1,2 dibromo-tetrafluoroethane (Halon-2402)

Article abstract of DOI:10.1063/1.1313545

The photofragment translational spectroscopy with vacuum ultraviolet ionization was used to investigate the photodissociation of 1,2 dibromo-tetrafluoroethane (Halon-2402) at 193 nm. The state-selected translational spectroscopy with resonance-enhanced multiphoton ionization was used to investigate the photodissociation of the compound at 193, 233, and 266 nm. The ultraviolet photodissociation was further studied by determining the product branch ratios, angular distributions, and translational energy distributions. A model was proposed to describe the wavelength-dependent bromine quantum yield. The implications of these products in upper atmosphere were studied.

Full text of DOI:10.1063/1.1313545

Quantum yields and energy partitioning in the ultraviolet photodissociation of 1,2
dibromo-tetrafluoroethane (Halon-2402)
Peng Zou, W. Sean McGivern, Osman Sorkhabi, Arthur G. Suits, and Simon W. North
Citation: The Journal of Chemical Physics 113, 7149 (2000); doi: 10.1063/1.1313545
View online: http://dx.doi.org/10.1063/1.1313545
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/113/17?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Properties of the B+-H2 and B+-D2 complexes: A theoretical and spectroscopic study
J. Chem. Phys. 137, 124312 (2012); 10.1063/1.4754131
The electronic states of 1,2,3-triazole studied by vacuum ultraviolet photoabsorption and ultraviolet photoelectron
spectroscopy, and a comparison with ab initio configuration interaction methods
J. Chem. Phys. 134, 084309 (2011); 10.1063/1.3549812
Fourier transform microwave spectroscopy and molecular structure of the trans-1,2-difluoroethylene-hydrogen
fluoride complex
J. Chem. Phys. 131, 204302 (2009); 10.1063/1.3246841
Role of OH-stretch/torsion coupling and quantum yield effects in the first OH overtone spectrum of cis-cis
HOONO
J. Chem. Phys. 122, 104311 (2005); 10.1063/1.1859273
A statistical, ab initio, quantum mechanical study of the photolysis and final state distributions of singlet ketene
J. Chem. Phys. 119, 6003 (2003); 10.1063/1.1601603
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

JOURNAL OF CHEMICAL PHYSICS
VOLUME 113, NUMBER 17
1 NOVEMBER 2000
Quantum yields and energy partitioning in the ultraviolet photodissociation
of 1,2 dibromo-tetrafluoroethane „Halon-2402…
Peng Zou and W. Sean McGivern
Department of Chemistry, Texas A & M University, College Station, Texas 77842
Osman Sorkhabi and Arthur G. Suits
Chemical Sciences Division, Lawrence Berkeley Laboratory, University of California,
and Chemistry Department, University of California, Berkeley, CA 94720
Simon W. Northa)
Department of Chemistry, Texas A & M University, College Station, Texas 77842
͑
Received 11 July 2000; accepted 8 August 2000͒
The photodissociation of 1,2 dibromo-tetrafluoroethane ͑Halon-2402͒ has been investigated at 193
nm using photofragment translational spectroscopy with vacuum ultraviolet ionization and at 193,
233, and 266 nm using state-selected translational spectroscopy with resonance-enhanced
multiphoton ionization. The product branching ratios, angular distributions, and translational energy
distributions were measured at these wavelengths, providing insight into the ultraviolet
photodissociation dynamics of CF BrCF Br. The total bromine atom quantum yields were found to
2
2
be 1.9Ϯ0.1 at both 193 and 233 nm and 1.4Ϯ0.1 at 266 nm. The first C–Br bond dissociation
energy was determined to be 69.3 kcal/mol from ab initio calculations. The second C–Br bond
dissociation energy was determined to be 16Ϯ2 kcal/mol by modeling of the bromine quantum
yield. In addition, variational Rice–Ramsperger–Kassel–Marcus theory was used to calculate the
secondary dissociation rates for a range of dissociation energies above threshold. These results
suggest that CF CF Br photofragments with sufficient internal energies will undergo secondary
2
2
dissociation prior to collisional stabilization under atmospheric conditions. Based on the measured
translational energy distributions and product branching ratios, a model is proposed to describe the
wavelength-dependent bromine quantum yield and the implications of these results to atmospheric
chemistry are discussed. © 2000 American Institute of Physics. ͓S0021-9606͑00͒01941-3͔
I. INTRODUCTION
in the atmosphere is 25–40 years, it will make a significant
contribution to stratosphere ozone depletion for an extended
period of time.5
Although there are two C–Br bonds in this system, it has
been assumed that the total bromine atom quantum yield,
The role of anthropogenic chlorine and bromine contain-
ing molecules in stratospheric ozone depletion has been well
established. The importance of bromine and its chemistry in
1
⌽
, for this molecule follows the trend of other bro-
the atmosphere is receiving renewed interest based on pre-
dictions that bromine is almost 100 times more destructive to
stratospheric ozone than chlorine on an atom-for-atom
basis. One of the major sources of anthropogenic bromine
in the stratosphere are the halons. Recent measurements of
the halogenated species in the tropical tropopause found that
͓Br,Br*͔
6
momethanes and is unity. However, unlike the polybromi-
nated methane derivatives, the second C–Br bond in
CF BrCF Br is significantly weaker than a typical C–Br
bond, due to the energy associated with forming the carbon–
carbon double bond that largely offsets the energy necessary
to break the second C–Br bond. As a result there is sufficient
energy following UV absorption to ensure that both C–Br
bonds break, resulting in a ⌽͓
2
,3
2
2
7
1,2 dibromo-tetrafluoroethane, henceforth referred to as
CF BrCF Br, constitutes 0.134 of the total organic bromine
2
2
, of 2.0. The primary
Br,Br*͔
͑
21Ϯ1 pptv͒ detected in the tropical tropopause ͑between
C–Br bond cleavage results in either ground state bromine
4
3
0 °S and 30 °N latitudes͒. Since transport from the tropo-
sphere into the stratosphere occurs mainly in this region, the
concentration of CF BrCF Br in tropical tropapause provides
2
2
(
P3/2), denoted as Br, or excited state bromine ( P1/2), de-
noted as Br*, and an energized C F Br radical which may
2
4
2
2
contain sufficient energy to undergo secondary dissociation
BrCF CF Br→C F BrϩBr,Br*,
͑1͒
͑2͒
a good measure of the relative mixing ratio in the strato-
sphere. The title compound has found widespread use as a
fire suppressant and, although the production of CF BrCF Br
2
2
2 4
2
2
in development countries has been halted by the Copenhagen
Amendment to the Montreal Protocol, its release continues
from current inventories. Since the lifetime of this compound
C F Br→C F ϩBr.
2
4
2 4
An important issue from a dynamical perspective is whether
the distribution of available energy following the initial
a͒Electronic mail: north@mail.chem.tamu.edu
0021-9606/2000/113(17)/7149/9/$17.00
7149
© 2000 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.113.111.210 On: Wed, 24 Dec 2014 04:15:29
1

7150
J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Zou et al.
9
TABLE I. Anisotropy parameters and average total translational energies
for primary C–Br bond cleavage.
ionization have also been previously described. The nozzle
was heated to 100 °C to minimize clustering in the beam and
no evidence for clusters was observed.
Anisotropy parameter,
Total translational energy
␤
͑kcal/mol͒
B. State-selected ion time-of-flight experiments
193 nm
233 nm
266 nm
Br
Br*
Br
Br*
Br
Br*
0.67Ϯ0.10
1.20Ϯ0.10
1.26Ϯ0.10
1.39Ϯ0.10
1.10Ϯ0.10
1.28Ϯ0.10
26Ϯ1
22Ϯ1
21Ϯ1
18Ϯ1
19Ϯ1
16Ϯ1
The state-selected ion time-of-flight experiments were
conducted at Texas A & M University using a differentially
pumped molecular beam apparatus that has been previously
described.1
0,11
A
pulsed molecular beam of ϳ3%
CF BrCF Br in helium was collimated by a conical skimmer
2
2
and intersected by either one or two laser pulses.
The bromine atoms were probed using 2ϩ1 resonance-
enhanced multiphoton ionization ͑REMPI͒ transitions in the
wavelength region between 233 and 267 nm. The tunable
C–Br bond cleavage ͑see Table I͒ ensures that sufficient en-
ergy remains in the C F Br fragment for it to undergo sec-
12
2
4
ondary reaction and whether the breakage of bonds can be
characterized as stepwise or concerted. A more atmospheri-
cally relevant question is whether the loss of one or both Br
atoms changes the atmospheric impact of this molecule.
In the present article, we have investigated the photodis-
sociation of CF BrCF Br in a molecular beam at wave-
lengths ranging from 266 to 193 nm using two complemen-
tary experimental techniques. We have performed state-
selected ion time-of-flight mass spectrometry ͑TOF-MS͒
measurements to measure the Br and Br* relative branching
ratios, angular distributions, and translational energy distri-
butions. We have also performed photofragment translational
spectroscopy ͑PTS͒ experiments at 193 nm using tunable
photoionization detection to identify all of the dissociation
products. This technique provides a broad overview of the
dissociation and complements the state-specific measure-
laser pulses were generated by a dye laser that was pumped
by 355 nm light from the third harmonic of a Nd:YAD laser
operating at 10 Hz. Two laser experiments were performed
by overlapping the output of an excimer laser operating on
the ArF ͑193 nm͒ transition with the probe beam from the
dye laser. The excimer light was polarized with a pile-of-
plates polarizer and the reported anisotropy parameters were
corrected for the incomplete ͑85%͒ polarization. The time
delay between the pump and probe pulses was typically
2
2
0
–50 ns.
Ions were detected using a two-stage Wiley–McLaren
time-of-flight mass spectrometer in either ‘‘noncore
sampled’’ or ‘‘core-sampled mode,’’ using a set of dual
13–17
Chevron microchannel plates.
In noncore sampled
mode, the signal was sent to a 400 MHz digital oscilloscope,
which was interfaced to a Pentium PC. Subtracting off-
resonance traces from on-resonance signals effectively re-
moved the minor pump-oil related background contribution
to the TOF spectra. A fast photodiode was used to trigger the
data acquisition. In core-sampled mode, a copper plate with a
ments. We have unambiguously determined that ⌽͓
Br,Br*]
is
near 2.0 at both 193 and 233 nm, and 1.4 at 266 nm. A
model, based on the experimental distributions, provides a
reasonable description of the wavelength-dependent Br
quantum yields for this molecule. In addition, we present ab
initio results on the C–Br bond dissociation energies in both
C F Br and CF BrCF Br and compare them to the experi-
3
.0-mm-diam pinhole was placed approximately 29 cm from
the interaction region, allowing only a ‘‘core’’ of the ion
packet to strike the detector. The core-sampled signal was
averaged using a multichannel scaler with a 5 ns bin size. In
order to minimize cluster contributions, the pulsed valve
nozzle was typically held at 70 °C, and an early part of the
molecular beam pulse was probed. No significant change in
the TOF profiles was observed as a function of the delay
time between the laser and the molecular beam pulse, indi-
cating that cluster formation was negligible.
2
4
2
2
mental bond energies. The atmospheric significance of the
results will also be discussed.
II. EXPERIMENT
A. Photofragment translational spectroscopy
experiments
These experiments were carried out at the Chemical Dy-
namics Beamline at the Advanced Light Source at Lawrence
Berkeley National Laboratory. A complete description of the
CF BrCF Br 99% was obtained from Aldrich Chemical
Co. and used without further purification.
2
2
8
apparatus can be found elsewhere. Briefly, a continuous mo-
III. RESULTS AND ANALYSIS
lecular beam of Ͻ1% CF BrCF Br seeded in helium was
2
2
A. Photofragment translational spectroscopy
experiments at 193 nm
collimated with conical skimmers and intersected at 90° with
the output of a excimer laser operating on the ArF transition
͑
193 nm͒. There was no change in the shape of any of the
Center-of-mass translational energy distributions,
2
TOF spectra over a laser fluence range of 60–500 mJ/cm ,
providing strong evidence that all of the observed signals are
the result of single photon absorption. Neutral photodissocia-
tion products traveled 15.2 cm where they were ionized by
tunable vacuum ultraviolet ͑VUV͒ undulator radiation, mass
selected, and counted as a function of time. The characteris-
tics of the VUV undulator radiation used for product photo-
P(E ), were obtained from the TOF spectra using the for-
T
1
8
ward convolution ͑FC͒ technique. For all of the PTS TOF
spectra presented, the circles represent the data and the lines
represent the forward convolution fit. TOF spectra were ob-
tained for masses C F ͑m/z 100͒ and Br ͑m/z 79, 81͒. No
signal was observed at any other masses including C F Br
ϩ
4
ϩ
2
ϩ
2
4
͑m/z 179, 181͒ using a VUV ionization energy of 15 eV.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Photodissociation of Halon-2402
7151
ϩ
FIG. 1. Unpolarized TOF spectra for m/z 79 (Br ) photoproducts at scat-
tering angles of 10° and 15° and a photoionization energy of 15.0 eV arising
from CF BrCF Br photodissociation at 193 nm. The circles are the experi-
2
2
mental data, and the solid lines are the forward convolution fits using the
FIG. 2. The upper panel shows the primary P(E ) used to fit the primary Br
T
P(E ) distributions in Fig. 2.
T
loss from CF BrCF Br. The internal energy of the C F Br radical, formed in
2
2
2 4
coincidence with ground state bromine, is indicated on the top axis. The
lower panel shows the secondary P(E ) distribution used to fit the second-
ary Br and C2F4 fragments shown in Figs. 1 and 3.
T
ϩ
TOF spectra for m/z 79 (Br ) at laboratory angles of 10° and
15° are shown in Fig. 1. Two contributions to the TOF spec-
tra were observed and are shown using dashed lines in the
figure. The narrow feature at early arrival times results from
primary C–Br bond cleavage. The second contribution at
longer times is due to secondary Br fragments that arise from
ity on the detector axis, v , the angle between the laser elec-
z
2
2,23
tric field, ⑀ˆ , and the detector axis, zˆ , is given by
ϱ
g͑v͒
v
z
the spontaneous dissociation of C F Br. The relative inten-
f v ,␹ϭ
͵
ͫ
1ϩ␤P ͑cos ␹͒P2ͩ ͪ
ͬ
dv,
͑3͒
2
4
ͩ
z
2
2
vz
͉ ͉
v
v
sity of the fast to slow components in the TOF indicates that
the primary to secondary Br yield is 1.0:0.97 suggesting an
overall Br quantum yield of 1.98Ϯ0.09. The primary Br TOF
where ␤ is the spatial anisotropy parameter, P (x) is the
second Legendre polynomial, and g(v) is the center-of-mass
2
spectrum has been fit using the P(E ) distribution shown in
speed distribution. In the noncore sampled fitting, the depen-
T
the top panel of Fig. 2. The P(E ) distribution is nearly
dence of f(v ,␹) on ␤ was eliminated by co-adding normal-
T
z
Gaussian in shape with a maximum at 26Ϯ1 kcal/mol and a
full width at half maximum ͑FWHM͒ of 10 kcal/mol.
ized profiles with ␹ϭ90° and ␹ϭ0° with a 2:1 weighting. A
trial speed distribution, g(v), was adjusted to obtain a best
fit to the co-added angular profile. A one-parameter fit to the
individual horizontal and vertical profiles was then used to
obtain ␤. For core-sampled experiments, the presence of the
coring aperture discriminates against fragments with large
ϩ
Figure 3 shows C2F4 ͑m/z 100͒ TOF spectra at a labo-
ratory angle of 20° and a photoionization energy of 15 eV.
Only one contribution to the m/z 100 spectra is present and is
due to the formation of C F from the secondary fragmenta-
2
4
tion of C F Br. The secondary fragments in both the m/z 79
2
4
and m/z 100 TOF spectra were fit using the P(E ) distribu-
T
tion shown in the lower panel of Fig. 2. The distribution is
peaked near zero with an average energy of 5 kcal/mol. A
secondary angular distribution with forward–backward sym-
metry, consistent with an intermediate lifetime longer than a
rotational period, was used in the forward convolution
1
9,20
fitting.
B. State-selected ion time-of-flight experiments
The FC technique was used to fit the experimental TOF
profiles, and details can be found elsewhere.1 We find this
approach preferable to a direct inversion of the TOF profiles,
which ignores instrumental broadening factors and is valid
only for high velocities. Briefly, the photofragment velocity
distribution in terms of the projection of the fragment veloc-
0,21
ϩ
4
FIG. 3. Unpolarized TOF spectra for m/z 100 (C F ) at a scattering angle of
2
2
0° photoionization energy of 15.0 eV arising from CF BrCF Br photodis-
sociation at 193 nm. The circles are the experimental data, the solid line is
the forward convolution fit using the P(E ) distributions shown in Fig. 2.
2 2
T
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.113.111.210 On: Wed, 24 Dec 2014 04:15:29
1

7152
J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Zou et al.
FIG. 5. Translational energy distributions used to fit the Br fragment TOF
profile at 233 nm. The primary and secondary components are indicated on
the figure.
isotropy parameter of 1.40Ϯ0.10 was obtained, suggesting
that the transition dipole moment is approximately parallel to
the C–Br bond.
The Br TOF profiles, shown in the upper panel of Fig. 4,
are substantially different and cannot be adequately fit with a
single contribution. The fast feature has an anisotropy pa-
rameter of 1.20Ϯ0.10 and an average translational energy of
FIG. 4. Core-sampled time-of-flight spectra for Br and Br* arising from
CF BrCF Br photodissociation near 234 nm. The filled circles are the ex-
2
2
perimental data and the solid line is a forward convolution fit that includes
7
9
both isotopes of bromine ͑for clarity the Br peaks are indicated with ar-
rows͒. The dashed lines show the primary and secondary contributions to
the TOF spectrum.
2
1Ϯ1 kcal/mol with a FWHM of ϳ10 kcal/mol. We attribute
this feature to primary dissociation to give ground state Br.
The observed spatial anisotropy suggests that the initial
C–Br bond breaks rapidly ͑Ͻ1 ps͒ and with unit efficiency
following absorption of the photon. The slower feature ex-
hibits little or no velocity anisotropy and is not observed in
the Br* TOF profiles. Linear momentum conservation with
the primary Br fragments provides an average velocity for
velocity components perpendicular to the flight axis, thereby
increasing the velocity resolution. The difference between
core-sampled and noncore sampled fitting has been described
1
7
max
previously. A speed-dependent vz array was calculated at
each set of ion-optic voltage settings and used in the subse-
quent forward convolution fitting. By interactively adjusting
a trial speed distribution g(v) and anisotropy parameter, ␤,
we can obtain best fits to both vertical ͑␹ϭ0°͒ and horizontal
the C F Br counter fragment of approximately 500 m/s. The
2
4
secondary Br fragments will acquire an additional recoil
from the spontaneous dissociation of the C F Br fragment.
2
4
However, the average velocity for the secondary Br in the
laboratory frame is symmetric about the C F Br velocity,
͑
␹ϭ90°͒ data simultaneously.
2 4
implying that the C F Br fragments are relatively long lived.
2
4
In addition, a forward–backward symmetric angular distri-
bution of fragments was required to fit the secondary contri-
bution to the 193 nm PTS data, and a faster C F Br disso-
1
. Photodissociation at 233 nm
2
4
Representative one-color core-sampled vertical ͑⑀ˆ
ʈ
zˆ ͒ and
ciation at longer wavelengths with lower internal energy
would not be expected. Rice–Ramsperger–Kassel–Marcus
͑RRKM͒ calculations on the C F Br dissociation ͑vide infra͒
horizontal ͑ ⑀ˆ Ќ zˆ ͒ TOF profiles of Br ͑upper panel͒ and Br*
͑lower panel͒ for dissociation wavelengths near 233 nm are
2
4
shown in Fig. 4. The Br* TOF profile ͑lower panel͒ consists
of a single component at large velocities and exhibits signifi-
cant velocity anisotropy. The FC fit of the Br* TOF profiles
also indicate that the C F Br fragment lives longer than the
2 4
rotational period. Rotational averaging is expected to dimin-
ish the lab-frame anisotropy parameter, as observed in the
TOF spectra. We therefore attribute the slow component of
the TOF spectra to secondary Br atoms arising from the
spontaneous dissociation of C F Br radicals.
gives a near Gaussian P(E ) distribution peaked at 18Ϯ1
T
kcal/mol with a FWHM ϳ10 kcal/mol, which is consistent
with a direct dissociation on a repulsive potential energy
surface. Therefore, we attribute the Br* products to the pri-
mary C–Br bond fission. Based on the measured intensity
change between vertical and horizontal geometries, an an-
2
4
The velocity distributions derived from fitting the TOF
spectra are shown in Fig. 5. The relative intensities of pri-
mary and secondary contributions to the Br TOF spectra are
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Photodissociation of Halon-2402
7153
TABLE II. Br/Br* total and primary branching ratio.
Total Br/Br*
Primary Br/Br*
193 nm
233 nm
266 nm
8.3Ϯ0.2
3.3Ϯ0.2
2.1Ϯ0.2
3.8Ϯ0.2
1.3Ϯ0.2
1.1Ϯ0.1
1
.0 and 1.6, respectively. This ratio and the measured total
Br/Br* branching ratio are necessary to provide a direct mea-
sure of the bromine atom quantum yield, ⌽͓
.
Br,Br*͔
The branching ratio of Br and Br* was determined by
scanning the probe laser over the Br/Br* Doppler profiles
and correcting the ratio of integrated areas of the Br and Br*
REMPI signals by the detection efficiency of Br and Br*
REMPI transitions. The photodissociation of CH Br was
3
used to calibrate the Br/Br* detection efficiency, and the
relative Br/Br* detection efficiencies were previously deter-
1
7
mined to be 0.46:1 for these states. The Br/Br* branching
ratio measured in this manner includes Br from both the
primary and secondary steps and, therefore, overestimates
the primary Br/Br* branching ratio. Assuming that the first
step in the dissociation is exclusively the loss of Br or Br*,
the primary Br/Br* ratio may be derived by correcting for
the relative contributions of primary to secondary Br. The
FIG. 6. Noncore sampled time-of-flight spectra for Br and Br* arising from
CF BrCF Br photodissociation near 266 nm. The filled circles are the ex-
perimental data and the solid lines are forward convolution fits. The angle
between the electric field vector of the dissociation laser and the flight axis,
2
2
␹, is indicated.
total bromine atom quantum yield, ⌽͓
, at 233 nm is
Br,Br*͔
1.93Ϯ0.09, implying that nearly all the C F Br radicals un-
2 4
Adjusting this value to obtain the measured Br quantum
yield we find that the secondary bond dissociation energy is
dergo secondary dissociation regardless of whether they arise
from the C F BrϩBr or C F BrϩBr* channel. The Br/Br*
2
4
2 4
1
6Ϯ2 kcal/mol with the error bound reflecting uncertainties
total branching ratios and primary branching ratios for a
number of photolysis wavelengths are shown in Table II.
in the Br quantum yield.
3. Photodissociation at 193 nm
2. Photodissociation at 266 nm
The FC fitting of the two-color REMPI data using 193
One-color noncore sampled TOF spectra are shown in
nm photolysis gives average translational energies of 26Ϯ1
and 22Ϯ1 kcal/mol for the primary Br and Br* dissociation
channels, respectively. These values result in a state-
averaged translational energy of 25Ϯ1 kcal/mol, which is in
excellent agreement with the PTS results. The anisotropy
parameters of the primary Br and Br* are 0.5Ϯ0.1 and 0.9
Ϯ0.1, respectively, indicating a stronger perpendicular con-
tribution than observed at longer wavelengths. The primary
Br/Br* branching ratio was found to be 3.8Ϯ0.2, and the
ratio of primary to secondary Br is 1.2Ϯ0.1. From these
Fig. 6 for the photodissociation of C BrCF Br at 266 nm.
2
2
The Br TOF profiles ͑left panels͒ show a bimodal velocity
distribution similar to that observed at 233 nm. The primary
Br and Br* channels give anisotropy parameters of ␤ϭ1.10
Ϯ0.10 and ␤ϭ1.28Ϯ0.10, indicating that both arise from a
transition dipole moment of predominately parallel character.
The average translational energies are 19Ϯ1 and 16Ϯ1 for
the primary Br and Br* channels, respectively. Performing a
similar analysis to that described above for 233 nm, we at-
tribute the fast and slow Br contributions to primary and
secondary dissociation, respectively. Based on the FC fits,
we find that the primary to secondary Br ratio is 1.0:0.9 and
the primary Br/Br* branching ratio is 1.1Ϯ0.1, resulting in
ratios, the total bromine atom quantum yield, ⌽͓
, was
Br,Br*͔
found to be 1.9Ϯ0.1. The translational energy distribution
associated with the secondary dissociation has an average
value of 5Ϯ1 kcal/mol with FWHM of ϳ10 kcal/mol, which
is consistent with the results derived from the PTS data. The
secondary angular distribution was found to be nearly isotro-
pic, consistent with a C F Br lifetime in excess of its rota-
⌽͓
ϭ1.4. This value is lower than at 233 nm, implying
Br,Br*͔
that, because there is less available energy at 266 nm, not all
C F Br fragments secondarily dissociate. The fraction of
2
4
2
4
C F Br fragments that undergo secondary C–Br bond disso-
2
4
tional period.
ciation will depend on both the distribution of the internal
energy following the primary dissociation step and the sec-
C. Ab initio and rate calculations
ondary C–Br bond strength. Using the primary P(E ) dis-
T
tributions we can estimate the secondary C–Br bond disso-
ciation energy by comparing the predicted Br quantum yield
to the experimental measurement. The secondary C–Br bond
energy provides a dividing line between stable and unstable
C F Br radicals arising from the first step in the dissociation.
In order to provide additional insight into the mechanism
for the photodissociation of CF BrCF Br, we have per-
2
2
formed ab initio calculations to determine the primary and
secondary C–Br bond strengths of CF BrCF Br. All ab ini-
2
2
2
4
tio calculations were performed using Gaussian 98 on an
2
4
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

7154
J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Zou et al.
TABLE III. RRKM calculations of the unimolecular dissociation rates of
C F Br using 16 kcal/mol C–Br bond energy.
2
4
(
͑
E*ϪE0)
kcal/mol͒
Unimolecular reaction rate (sϪ1
)
1
2
3
4
.0
.0
.0
.0
2.5ϫ108
8
8.3ϫ10
1.8ϫ109
8
3.3ϫ10
1
0.0
3.5ϫ1010
assess, making the choice of a truncation point challenging.
Although the value derived from the PTS studies is close to
the calculated value of 19.3 kcal/mol, it is significantly
higher than our experimental result of 16Ϯ2 kcal/mol and
outside the mutual error bounds. Using a secondary C–Br
bond energy of 22.3 kcal/mol and our measured primary
state-selected translational energy distributions would predict
a quantum yield of 1.13 that is inconsistent with our value of
FIG. 7. Schematic energy diagram of the dissociation of CF BrCF Br. The
energies are the results of ab initio calculations as described in the text.
2
2
1
.4, which we consider reliable. Both our value and the value
of Nathanson et al. are higher than 13.6 kcal/mol reported by
Wu and Rogers32 based on the heat of reaction for the bro-
mination of C F and an initial C–Br bond energy of 69.3
Origin 2000 Supercomputer and an SGI Power Challenge
workstation. Geometries and frequencies for the ground state
species were calculated at the MP2/6-311ϩG* level, and vi-
brational frequencies were left unscaled for the zero-point
corrections and RRKM calculations. The ab initio methodol-
ogy used to calculate the bond energies has been successful
in calculating C–Br bond energies in haloalkanes with mod-
est computational expense.25 This methodology corrects
MP2/cc-pVtz energies for correlation to the coupled cluster
level including single and double excitations and perturba-
tive treatment of triple excitations ͑CCSD͑T͒͒ and basis-set
effects to the cc-pV5z level using correction factors derived
from monohalogenated compounds. The primary C–Br bond
energy is 69.3 kcal/mol, and the second C–Br bond dissocia-
tion energy is 19.3 kcal/mol from these calculations. Since
the ab initio methodology yields good agreement with ex-
periment for closed-shell compounds, we believe the value
for the primary bond energy of CF BrCF Br to be accurate,
2
4
kcal/mol.
The unimolecular rates for the secondary Br elimination
channel were calculated using variational RRKM theory33 by
minimizing the sum of states along the reaction coordinate
‡
dN ͑E,J,R͒
ϭ0.
͑4͒
dR
For consistency all frequencies and energies used in the
RRKM calculations, including reactant and product species,
were obtained from ab initio theory. A direct count algo-
rithm was used to determine the sum of states using the
harmonic oscillator approximation for both sets of calcula-
tions. The frequencies of the conserved modes were assumed
to change smoothly with the C–Br distance, R, according to
͑i͒͑R͒ϭv ͑R ͒exp͑ϪaR͒
͑i͒
v
r
e
2
2
͑i͒
ϩvp ͑R ͓͒1Ϫexp͑ϪaR͔͒,
͑5͒
ϱ
though the value for the second bond energy appears to be
overestimated. A schematic diagram of the sequential loss of
(
i)
where v (R ) is the ith frequency of the reactant molecule
r
e
(
i)
bromine atoms is shown in Fig. 7.
at the equilibrium C–Br bond length, R , v (R ) is the
p
e ϱ
Based on the available heat of formation data,2
6–29
the
corresponding product frequency, and a is a constant taken
4,35
energy required for the loss of two Br atoms from
to be 1.2.3
The transitional modes were assumed to decay
CF BrCF Br is 82.9 kcal/mol. Combining the ab initio result
exponentially to the C F rotational constant with the same
2
2
2 4
for primary C–Br bond ͑69.3 kcal/mol͒ with relevant heat of
formation information, the secondary C–Br bond energy is
value of a. The sum of a Morse function and a centrifugal
barrier was used to model the potential along the C–Br
stretching coordinate. The centrifugal barrier was derived us-
ing the average rotational energy of the C F Br correspond-
13.6 kcal/mol. Krajnovich, Butler, and Lee reported an upper
limit for the C F Br second C–Br bond energy of 19Ϯ3
2
4
2 4
kcal/mol from their PTS data analysis from the photodisso-
ing to Jϭ75 derived from the soft fragment impulsive
model following primary loss of bromine. Moments of in-
3
0
36
ciation C F BrI. This result was subsequently refined by
2
4
3
1
Nathanson et al. to 22.3Ϯ2.5 kcal/mol based on a similar
analysis to Krajnovich and co-workers which relied on a
nonabrupt truncation of the C F Br TOF spectrum. In those
ertia at each fixed geometry were obtained by changing only
the C–Br bond length. One rocking mode and one wagging
mode were treated as transitional modes. Variational RRKM
calculations were performed at a series of energies above the
threshold energy for Br loss from C F Br, and the rates are
2
4
studies the Br fragments were not detected and there was an
overlap between I and I* components at energies where the
coincident C F Br was unstable. Rotational metastability of
2
4
shown in Table III as a function of the energy above thresh-
old. For the RRKM calculations, we have adopted an
asymptotic energy of 16 kcal/mol for the secondary C–Br
2
4
the C F Br should be less important in the dissociation of
2
4
CF BrCF Br than in C F IBr but its effects are difficult to
2
2
2 4
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.113.111.210 On: Wed, 24 Dec 2014 04:15:29
1

J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Photodissociation of Halon-2402
7155
bond energy. Rate constants for Br loss increase from 2.5
suggesting that C F Br fragments derived from the C F Br
2
4
2 4
8
10 Ϫ1
ϫ10 to 3.5ϫ10 s for energies above threshold between
and 10 kcal/mol. In all cases, the lifetime of the C F Br
*
ϩBr channel do not have sufficient internal energy for sec-
ondary dissociation.
1
2
4
fragment under collisionless conditions was less than the
temporal width of the laser pulses, ensuring that all second-
ary products were observed in the TOF-MS experiments.
The secondary C–Br bond dissociation is correlated ex-
clusively to ground state Br products, which appear as a slow
component in the Br TOF profiles and are absent in the Br
TOF spectra. Analysis of the C F product PTS data gives an
*
2
4
average total translational energy of 5 kcal/mol, strongly in-
dicating that the C F molecules and slow Br fragment are
2
4
IV. DISCUSSION
derived from the same secondary dissociation channel. The
present results may be compared to a prior distribution for
the loss of an atom from a polyatomic molecule, in which
A. Product angular distributions and translational
energy distributions
0
.5
sϩ0.5
The positive anisotropy for the primary bromine frag-
ment observed at all wavelengths under investigation indi-
cates that the dissociation occurs on a repulsive potential
prior to parent rotation and arises from a transition dipole
moment predominately parallel to the C–Br bond. There are
P(fT)ϭfT (1ϪfT)
, where fT is the fraction of avail-
able energy partitioned into translation and s is the number
of oscillators in the polyatomic product.40 For 193 nm disso-
ciation, the available energy is 40 kcal/mol, assuming that
the average translational energy of the C2F4Br fragment is 26
kcal/mol and the bond energies are 69.3 and 16 kcal/mol for
the first and second steps, respectively. The average transla-
tional energy from the prior distribution is therefore 4 kcal/
mol, in agreement with the PTS results. The agreement be-
tween the average values of the prior distribution and the
observed P(ET) distribution implies that little or no barrier
exists for the secondary dissociation step. Secondary transla-
tional energy distributions at longer wavelengths are qualita-
tively consistent with the prior distribution, but the uncer-
tainty of the primary translational energy distribution
precludes a more detailed analysis.
three electronic excited states that contribute to the first UV
3
absorption band of haloalkane systems and are labeled Q ,
1
3
1
37
3
Q , and Q . Among these states, the Q state diabati-
0
1
0
3
cally correlates to excited state halogen atoms, while the Q
1
1
and Q states correlate to ground state products. The tran-
1
sition is predominantly parallel at 266 nm, becomes increas-
ingly parallel at 233 nm, and becomes more perpendicular at
193 nm. This behavior is very similar to other bromoalkanes
and is the result of overlapping contributions from different
parallel and perpendicular electronic states.1
0,17,38
Based on
magnetic circular dichroism experiments, Gedanken and
3
9
Rowe found that due to the smaller spin-orbit coupling of
The present results can be compared to the time-resolved
UV study of C2F4I2 reported by Zewail and co-workers.4
The authors observed a single prompt rise ͑р1 ps͒ when
1,42
bromine relative to iodine, there was a decrease in the rela-
3
tive contribution of the parallel Q state relative to the per-
0
1
probing the I
*
fragments but a biexponential appearance
pendicular Q state in methyl bromide at short wavelengths
1
when compared to methyl iodide. The present results are
curve for ground state I atoms. The fast signal was consistent
with direct bond dissociation, identical to the origin of the I
*
fragments. The slower component ͑30–150 ps͒ was due to
spontaneous secondary bond cleavage. Based on the energet-
ics of the CF2BrCF2Br system compared to C2F4I2 we would
expect to observe similar stepwise behavior.
1
consistent with the significant contribution of the Q state
1
3
near 193 nm but the overall dominance of the Q state.
0
From the translational energy distributions of the state-
selected fragments, the internal energy distribution of the
C F Br radicals for the Br and Br* channels may be calcu-
2
4
lated precisely using the conservation equation
C F Br
int
0
0
Br
elec
2
2
B. Model for the wavelength dependent bromine
quantum yield
E
ϭhvϪD ͑C–Br͒ϪEtransϪE
,
͑6͒
0
0
where D (C–Br) is the primary C–Br bond strength, E
is
As shown in Table II, the total quantum yield of bromine
trans
the total translational energy for the primary dissociation
atoms, ⌽͓
, depends strongly on the dissociation wave-
Br,Br*͔
Br
step, and E
is the electronic energy of the Br fragment
length. Due to the small secondary C–Br bond energy, pho-
todissociation at 193 and 233 nm results in internal energies
of C F Br from both the Br and Br* channels that are suffi-
elec
relative to the ground spin-orbit state. Since only C F Br
2
4
radicals with internal energies larger than the secondary
C–Br bond energy will spontaneously dissociate to C F and
2
4
cient to ensure complete secondary dissociation. At 266 nm,
however, only C F Br radicals formed in coincidence with
2
4
Br, the observed total quantum yield, ⌽͓
, provides a
Br,Br*͔
2 4
measure of the secondary C–Br bond strength. A stronger
secondary C–Br bond will result in fewer secondary Br
products since fewer C F Br radicals contain sufficient inter-
ground state Br with low translational energies will undergo
secondary dissociation. In order to quantitatively evaluate
the bromine atom quantum yield over the 193–266 nm
wavelength region, functional forms of both the wavelength
dependence of the Br/Br* branching ratios and the depen-
dence of the translational energies on the photolysis energy
must be determined.
2
4
nal energy to break the second C–Br bond. This is confirmed
by our observation of a wavelength-dependent quantum
yield. At 193 and 233 nm, the total bromine atom quantum
yields are close to 2.0, indicating that almost all the C F Br
2
4
fragments from both the C F BrϩBr and C F BrϩBr*
channels will experience secondary dissociation. At 266 nm,
however, the bromine atom quantum yield is reduced to 1.4
The forward convolution results show that the primary
translational energy distributions are well described by a
near symmetric functional form43 with a FWHM of ϳ13
2
4
2 4
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

7156
J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Zou et al.
second C–Br bond resulting in a bromine atom quantum
yield near 2. The atmospheric pressure in the lower strato-
sphere is 10–50 Torr, and temperature in this region is typi-
4
4
cally ϳ220 K. The collision frequency under these condi-
tions ranges from 2ϫ108 to 1ϫ10 s , suggesting that
9 Ϫ1
nascent C F Br radicals with internal energies near their dis-
2
4
sociation threshold may be partially stabilized by collisions.
The solar flux in the upper troposphere is concentrated in the
wavelength region Ͼ290 nm, where the photolysis energy is
insufficient to break both C–Br bonds and the bromine atom
quantum yield will be unity. In the lower stratosphere where
the solar flux has increasingly significant contributions from
shorter wavelengths, the internal energies of C F Br radicals
2
4
are much higher than the dissociation threshold. The unimo-
lecular dissociation rates for these high-energy fragments are
much faster than the collisional frequency, yielding a pho-
tolysis quantum yield of 2. The resulting C F molecules will
2
4
be removed primarily by reaction with OH radicals. Acer-
boni et al.45 determined the rate constant of this reaction at
Ϫ11
Ϫ3
Ϫ1 Ϫ1
2
98 K to be 1.13Ϯ0.33ϫ10
cm mol
s
implying
that low stratospheric lifetime of C F is approximately one
day.
FIG. 8. Wavlength-dependent quantum yield of bromine using the model
described in the text and three different values of the secondary C–Br bond
energy ͑A͒ 13.6 kcal/mol; ͑B͒ 16.0 kcal/mol, and ͑C͒ 22.3 kcal/mol. The
closed and open circles represent the ion TOF data and PTS data, respec-
tively.
2 4
Despite the insufficient photolysis energy in the lower
troposphere to break both C–Br bonds, surviving C F Br
2
4
radicals will rapidly react with O , and the resulting peroxy
2
radical C F BrO will react primarily with NO to produce
2
4
2
C F BrO and NO . The alkoxy radical, C F BrO, will likely
2
4
2
2 4
kcal/mol at all three wavelengths. We adopt this functional
undergo C–C bond fission, giving CF Br and CF O as prod-
2
2
form as an estimate of the P(E ) distributions at other wave-
T
ucts. It has been reported that 85% of C Cl O radicals un-
2
5
lengths. Wavelength-dependent average translational ener-
gies were assumed to be well described by a linear function
of the available energy. For dissociation on a repulsive po-
tential, this assumption is the basis for the impulsive models,
which are widely used to estimate the fraction of available
energy partitioned into product translation. The proportion-
ality factors for both Br and Br* primary channels were de-
rived from the values reported in Table I. The branching
ratios of the primary Br/Br* fragments were also assumed to
depend linearly on wavelength and the proportionality con-
stants were determined from the results shown in Table II.
Figure 8 shows the simulation using this model for a
secondary C–Br bond dissociation energy of 16 kcal/mol.
dergo thermal C–Cl bond fission and 15% undergo C–C
bond fission. However, since the C–F bond is significantly
stronger than the C–Cl bond, little C–F bond cleavage chan-
4
6
nel would be expected for C F BrO radical dissociation.
2
4
Exclusive C–C bond breakage has been previously reported
4
7
for C F O radicals. The CF Br will undergo subsequent
2
5
2
4
4
oxidation to yield CF O and Br atom. The final CF O prod-
2
2
uct will be effectively taken up by raindrops or aerosols. As
a result, it may be expected that both Br atoms present in the
CF BrCF Br will be released as active bromine upon absorp-
2
2
tion of a single UV photon, irrespective of the photolysis
energy.
The total Br quantum yield, ⌽͓
, is found to decrease
Br,Br*͔
V. CONCLUSIONS
from 2.0 for wavelengths shorter than ϳ230 nm to 1.0 for
wavelengths longer than ϳ320 nm. The resulting expression
can be used to accurately predict the wavelength Br quantum
yield of this compound in the UV
In the present study, we have examined the photodisso-
ciation of CF BrCF Br ͑Halon 2402͒ using both photofrag-
2
2
ment translational spectroscopy at 193 nm and state-selected
translational spectroscopy at 193, 233, and 266 nm. At each
wavelength, translational energy distributions, angular distri-
butions, and branching ratios were determined. Forward con-
volution fitting of the observed Br spectra showed two com-
ponents, corresponding to a fast primary loss of bromine
followed by a slower spontaneous dissociation of the C2F4Br
fragment. Based on the measured Br/Br* branching ratios
and the primary to secondary branching ratios, we have de-
termined the total bromine quantum yield at the three wave-
lengths. The results suggest that state-selected ion TOF is a
viable method for studying three-body dissociation. Using
these results, a model for the wavelength-dependent quantum
yield has been developed that provides an estimate for the
1
⌽͓
ϭ2.0Ϫ
.
͑7͒
Br,Br*͔
͓
1ϩexp͑Ϫ͑␭Ϫ263͒/10.5͔͒
Shown for comparison in Fig. 8 are the quantum yield pre-
dictions using the previously recommended values of 13.6
and 22.3 kcal/mol for the secondary C–Br bond energy.
C. Atmospheric fate of CF BrCF Br
2
2
The ultraviolet photodissociation of CF BrCF Br will
2
2
initially result in C F Br fragments and Br/Br* atoms. For
2
4
dissociation at wavelengths shorter than 230 nm, almost all
C F Br radicals have sufficient internal energy to break the
2
4
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Wed, 24 Dec 2014 04:15:29

J. Chem. Phys., Vol. 113, No. 17, 1 November 2000
Photodissociation of Halon-2402
7157
secondary C–Br bond dissociation energy. Though condi-
tions in the upper troposphere will allow some of the C F Br
fragments to be collisionally stabilized, subsequent oxidation
Chem. 92, 5 ͑1988͒; R. Ogorzalek Loo, H.-P. Haerri, G. E. Hall, and P. L.
Houston, J. Chem. Phys. 90, 4222 ͑1989͒.
J. A. Syage, J. Chem. Phys. 105, 1007 ͑1996͒.
P. Zou, W. S. MeGivern, and S. W. North, Phys. Chem. Chem. Phys. 2,
3785 ͑2000͒.
A. M. Wodtke, Ph.D. thesis, University of California, Berkeley, 1986; X.
Zhao, Ph.D. thesis, University of California, Berkeley, 1989.
S. W. North, D. A. Blank, J. D. Gezelter, C. A. Longfellow, and Y. T.
Lee, J. Chem. Phys. 102, 4447 ͑1995͒; S. W. North, D. A. Blank, and Y.
T. Lee, Chem. Phys. Lett. 224, 38 ͑1994͒; P. M. Kroger and S. J. Riley, J.
Chem. Phys. 67, 4483 ͑1977͒.
2
4
1
1
6
7
will yield a second bromine and two CF O fragments. As a
2
1
1
8
9
result, the photodissociation of CF BrCF Br may be ex-
2
2
pected to yield two Br atoms and two CF O fragments, re-
2
gardless of the conditions in the upper atmosphere.
ACKNOWLEDGMENTS
2
0
E. J. Hinsta, X. Zhao, and Y. T. Lee, J. Chem. Phys. 92, 2280 ͑1990͒.
K. Bergman, R. T. Carter, G. E. Hall, and J. R. Huber, J. Chem. Phys. 109,
Work by S.W.N., P.Z., and W.S.M. was partially sup-
ported by a Research Enhancement from Texas A&M Uni-
versity. The work by A.G.S. and O.S. was supported by the
Director, Office of Energy Research, Office of Basic Energy
Science, Chemical Sciences Division of the U. S. Depart-
ment of Energy under Contract No. DE-AC03-76SF00098.
The experiments were conducted at the Advanced Light
Source, Lawrence Berkeley National Laboratory which is
supported by the same source. Hardware and software sup-
port for the ab initio calculations was given by the Texas
A&M University Supercomputing Facility and Texas A&M
University Laboratory for Molecular Simulations under Na-
tional Science Foundation Grant No. CHE-9528196.
21
474 ͑1998͒.
2
2
2
2
3
4
R. N. Zare, Mol. Photochem. 4, 1 ͑1972͒.
R. N. Dixon, J. Chem. Phys. 85, 1866 ͑1986͒.
GAUSSIAN 98, revision A.6, M. J. Frisch, G. W. Trucks, H. B. Schlegel
et al., Gaussian, Inc., Pittsburgh, Pennsylvania, 1998.
W. S. McGivern, A. Derecskei-Kovacs, S. W. North, and J. S. Francisco,
J. Phys. Chem. A 104, 436 ͑2000͒.
M. W. Chase, Jr., NIST-JANAF Thermochemical Tables, 4th ed. J. Phys.
Chem. Ref. Data, Monogr. 9, 1 ͑1998͒.
H. V. Wartenberg and J. Z. Schiefer, Anorg. Chem. 278, 326 ͑1955͒.
V. P. Kolesov, I. D. Zenkov, and S. M. Skuratov, Russ. J. Phys. Chem.
͑Engl. Transl.͒ 36, 45 ͑1962͒; C. A. Neugebauer and J. L. Margrave, J.
Phys. Chem. 60, 1318 ͑1956͒.
25
2
6
2
7
28
2
3
9
0
F. W. Kirbride and F. G. Davidson, Nature ͑London͒ 174, 174 ͑1954͒.
D. Krajnovich, L. J. Butler, and Y. T. Lee, J. Chem. Phys. 81, 3031
͑
1984͒.
1
3
1
M. J. Molina and F. S. Rowland, Nature ͑London͒ 248, 810 ͑1974͒.
S. Solomon, M. Mills, L. E. Heidt, W. H. Pollock, and A. F. Tuck, J.
Geophys. Res., ͓Space Phys.͔ 97, 25 ͑1992͒.
R. R. Garcia and S. Solomon, J. Geophys. Res., ͓Space Phys.͔ 99, 12937
G. M. Nathanson, T. K. Minton, S. F. Shane, and Y. T. Lee, J. Chem.
Phys. 90, 6157 ͑1989͒.
2
32
33
E.-C. Wu and A. S. Rogers, J. Am. Chem. Soc. 98, 6112 ͑1976͒.
K. A. Holbrook, M. J. Pilling, and S. H. Robertson, Unimolecular Reac-
tions, 2nd ed. ͑Wiley Interscience, New York, 1996͒; W. Forst, Theory of
Unimolecular Reactions ͑Academic, New York, 1973͒.
W. L. Hase, Chem. Phys. Lett. 139, 389 ͑1987͒.
A. S. Sudbo, P. A. Schultz, E. R. Grant, Y. R. Shen, and Y. T. Lee, J.
Chem. Phys. 70, 912 ͑1979͒.
G. E. Busch and K. R. Wilson, J. Chem. Phys. 56, 3639 ͑1972͒.
R. S. Mulliken, Phys. Rev. 61, 277 ͑1942͒.
T. Gougousi, P. C. Samartzis, and T. N. Kitsopolous, J. Chem. Phys. 108,
5
A. Gedanken nad M. D. Rowe, Chem. Phys. Lett. 34, 39 ͑1974͒.
J. T. Muckerman, J. Phys. Chem. 93, 179 ͑1989͒.
J. L. Knee, L. R. Khundkar, and A. H. Zewail, J. Chem. Phys. 83, 1996
3
͑1994͒.
4
5
6
S. M. Schauffler, L. E. Heidt, W. H. Pollack, T. M. Gilpin, J. F. Vedder,
S. Solomon, R. A. Leub, and E. L. Atlas, Geophys. Res. Lett. 20, 2567
34
35
͑1993͒.
J. B. Burkholder, R. R. Wilson, T. Gierczak, E. Talukdar, S. A. McKeen,
J. J. Orlando, G. L. Vaghjiani, and A. R. Ravishankara, J. Geophys. Res.,
36
37
38
͓
Space Phys.͔ 96, 5025 ͑1991͒.
L. T. Molina, M. J. Molina, and F. S. Rowland, J. Phys. Chem. 86, 2672
1982͒.
͑
742 ͑1998͒.
7
8
S. W. Benson, J. Chem. Educ. 42, 502 ͑1973͒.
X. Yang, D. A. Blank, J. Lin, P. A. Heimann, A. M. Wodtke, A. Suits, and
Y. T. Lee, Rev. Sci. Instrum. 68, 3317 ͑1997͒.
M. Koike, P. A. Heimann, A. H. Kung, T. Namioka, R. DiGennaro, B.
Gee, and N. Yu, Nucl. Instrum. Methods Phys. Res. A 347, 282 ͑1994͒; P.
A. Heimann, M. Koike, C. W. Hsu, M. Evans, C. Y. Ng, D. Blank, X. M.
Yang, C. Flaim, A. G. Suits, and Y. T. Lee, Proc. SPIE 9, 2865 ͑1996͒.
W. S. McGivern, R. Li, P. Zou, and S. W. North, J. Chem. Phys. 111,
39
40
41
9
͑
1985͒.
4
4
2
3
L. R. Khundkar and A. H. Zewail, J. Chem. Phys. 92, 231 ͑1989͒.
Based on translational energy distributions of the form P(E)
a
b
1
1
1
0
1
2
ϭEmax͓f (1Ϫf ) ͔. We find that values of 10 for both parameters a and b,
E
E
respectively, reproduce the experimental distributions.
G. P. Brasseur, J. J. Orlando, and G. S. Tyndal, Atmospheric Chemistry
and Global Change ͑Oxford University Press, New York, 1999͒.
G. Acerboni, N. R. Jensen, B. Rindone, and J. Hjorth, Chem. Phys. Lett.
309, 364 ͑1999͒.
T. E. Mogelberg, J. Sehested, O. J. Nielsen, and T. J. Wallington, J. Phys.
Chem. 99, 16932 ͑1995͒.
J. Sehested, T. Ellermann, O. J. Nielsen, T. J. Wallington, and M. D.
Hurley, Int. J. Chem. Kinet. 25, 701 ͑1993͒.
5
771 ͑1999͒.
W. S. McGivern, R. Li, P. Zou, T. Nguyen, and S. W. North, Chem. Phys.
49, 237 ͑1999͒.
S. Arepalli, N. N. Presser, D. Robie, and R. J. Gordon, Chem. Phys. Lett.
17, 64 ͑1985͒.
44
45
46
2
1
1
1
3
4
W. C. Wiley and I. H. McLaren, Rev. Sci. Instrum. 26, 1150 ͑1955͒.
H. J. Hwang, J. Griffiths, and M. A. El-Sayed, Int. J. Mass Spectrom. Ion
Processes 131, 265 ͑1994͒.
47
1
5
R. Ogorzalek Loo, G. E. Hall, H.-P. Haerri, and P. L. Houston, J. Phys.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.113.111.210 On: Wed, 24 Dec 2014 04:15:29
1