The role of 3Π0+ in the photodissociation of BrCl at 235 nm

Article abstract of DOI:10.1016/S0009-2614(00)00467-X

The photodissociation of BrCl has been investigated at 235 nm using a two-dimensional photofragment ion imaging technique. Three direct dissociation channels were observed with the quantum yields of Φ[Br(²P_1/2)+Cl(²P _3/2)]=0.58±0.05, Φ[Br(²P_3/2)+Cl(²P _1/2)]=0.16±0.05, and Φ[Br(²P_3/2)+Cl(²P _3/2)]=0.26±0.05, and angular distributions were represented by anisotropy parameters, β=1.98±0.05, -1.04±0.05, and 1.88±0.05, respectively. The results indicate that the 2341 ³Π₀⁺ state plays a key role in the dissociation dynamics after absorbing a photon. Based on the results of BrCl and newly found probe lines for Br atoms, the ionization probability ratios of Cl(²P_3/2 or ²P_1/2) to Br(²P_3/2 or ²P_1/2) were determined.

Full text of DOI:10.1016/S0009-2614(00)00467-X

26 May 2000
Ž
.
Chemical Physics Letters 322 2000 429–438
www.elsevier.nlrlocatercplett
The role of 3P0q in the photodissociation of BrCl at 235 nm
Moon-Soo Park, Young-Jae Jung, Sung-Hae Lee, Dong-Chan Kim,
Kyung-Hoon Jung)
(
)
Department of Chemistry and School of Molecular Science BK21 , Korea AdÕanced Institute of Science and Technology,
Taeduck Science Town, Taejon 305-701, South Korea
Received 3 September 1999; in final form 10 April 2000
Abstract
The photodissociation of BrCl has been investigated at 235 nm using a two-dimensional photofragment ion imaging
Ž2 Ž2
w
.
.x
technique. Three direct dissociation channels were observed with the quantum yields of F Br P1 qCl P3r2 s0.58"
r2
w
Ž2
.
Ž2
.x
w
Ž2
.
Ž2
.x
0.05, F Br P3 qCl P1r2 s0.16"0.05, and F Br P3r2 qCl P3r2 s0.26"0.05, and angular distributions were
r2
represented by anisotropy parameters, bs1.98"0.05, y1.04"0.05, and 1.88"0.05, respectively. The results indicate
that the 2341 3P0
state plays a key role in the dissociation dynamics after absorbing a photon. Based on the results of BrCl
q
and newly found probe lines for Br atoms, the ionization probability ratios of Cl P3 or 2P1r2 to Br P3r2 or 2P1r2 were
Ž2
.
Ž2
.
r2
determined. q 2000 Elsevier Science B.V. All rights reserved.
2
Ž
.
Ž .
denoted Br and
1. Introduction
ground state Br 4p P3r2
2
Ž
. Ž
.
Cl 3p P3r2 denoted Cl , and the first spin–orbit
SO excited state Br 4p P1r2 denoted Br
Cl 3p P1r2 denoted Cl . In the BrCl photolysis,
four one-photon dissociation channels are thermody-
namically possible in the ultraviolet UV wave-
length region.
2
)
Ž
.
Ž
. Ž
.
The photodissociation dynamics of homo- and
heterodiatomic halogens have long been investigated
due to their interesting properties such as the avoided
crossing and the predissociation of the potential
and
2
)
Ž
. Ž
.
Ž
.
w x
curves 1 . BrCl dissociates into bromine and chlo-
w
x
rine which have impacts on ozone depletion 2–4 .
BrCl, in addition, can be used as a good reference
compound to determine the formation ratio of
bromine and chlorine in the dissociation of both
bromine and chlorine containing molecules by corre-
lating the exact quantum yield of each dissociation
channel of BrCl with their dynamic behaviors.
BrCl BrqCl
BrqCl)
D Hs218.3 kJmoly1
D Hs228.3 kJmoly1
D Hs262.3 kJmoly1
D Hs272.3 kJmoly1
a
Ž .
b
Ž .
c
Ž .
d
Ž .
Br) qCl
Br) qCl)
1
Ž .
The halogen atom photofragments are generally
formed in two low-lying fine structural states, i.e. the
Ž
.
The bond dissociation energy, D0 Br–Cl , was de-
termined to be 218.3 kJ moly1 5 . The SO excita-
w x
tion energies of bromine and chlorine are 44 and 10
kJ moly1, respectively 6 . BrCl exhibits a continu-
)
w x
Corresponding author. Fax: q82-42-869-8123; e-mail:
ous absorption spectrum with three peaks at 225,
khjung@mail.kaist.ac.kr
0009-2614r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 00 00467-X

(
)
430
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
w
x
373, and 443 nm in the UV–visible region 7,8 . The
maxima at 373 and 443 nm have been assigned to
the transitions from the ground state X 1 Sq0q to the
C 1P1 and B 3P0q states, respectively, arising from
known, it can be deduced from the well-established
properties of other heterodiatomic halogen molecules,
w
x
ICl and IBr 1,16 . Since ICl has been studied in
detail by various spectroscopic and theoretical meth-
the electron configuration sg2 pu4 pg3su1 2431 corre-
lated with the dissociation channels as displayed in
Fig. 1. A, B and ion-pair states have been studied by
laser induced fluorescence LIF and emission spec-
troscopy 9–13 . The nonadiabatic dissociation and
ods 17–20 , its dissociation dynamics can help to
understand the electronic structure of BrCl. On the
contrary, there are some disagreements among the
reported results of ICl at 235 nm. The first disagree-
ment is in the existence of the IqCl channel. In the
REMPI–Doppler profile experiments, three dissocia-
Ž
.
w
x
Ž
.
w
x
predissociation on the B state were investigated
)
)
w
x
Ž
.
5,14 . The absorption at 225 nm was attributed to
tion channels IqCl, IqCl , and I qCl were
X 1 Sq0q. The photodissociation dynamics of
observed 18 , while only two channels IqCl and
q
)
Ž
.
w
x
Ž
D 0
)
.
Ž
BrCl in the UV region have been studied via the
ion-image technique coupled with the resonance en-
hanced multiphoton ionization REMPI at 235 and
260 nm 15 . In this previous study, however, the
I qCl were confirmed in the REMPI–TOF time
.
w x
of flight and the ion-image experiments 19 . The
Ž
.
second is related to the analysis of the contribution
3
w
x
of the 2341 P0q state. The UV absorption of ICl
3
photodissociation in the range of 235–540 nm, only
corresponds to transitions to the 2422 Sy0q, 2341
two channels BrqCl and Br qCl could be
observed at 235 nm due to the experimental limita-
tion.
Although the electronic states of BrCl correspond-
ing to the transitions in the UV region is not well
1P1 and P0q states from the ground state X 1 Sq0q
.
3
)
)
Ž
.
3
The 2341 P0q state adiabatically correlates to the
IqCl) dissociation channel with avoided crossing 1
3
3
y
Ž
.
AC1 between the 2422 S0q and the 2341 P0q
Ž
.
state, and avoided crossing 2 AC2 between the
2341 3P0q and the 2431 3P0q state as shown in
Fig. 1, the correlation diagram of BrCl, which is
similar to that of ICl. In the REMPI–Doppler study,
it has been suggested that the fraction of the 2341
3P0q state is 0.43. Meanwhile a small contribution
of this state to the absorption at 235 nm has been
assumed with no evidence of any parallel component
of the IqCl) channel where the parallel component
results from the Vs0q state and the perpendicular
component from the Vs1 state. In order to evaluate
the dissident results, it is very important to determine
the exact relative quantum yields for the dissociation
Ž . Ž .
channels, a – c and the nonadiabatic crossing prob-
ability at AC2.
In the case of the dissociation of BrCl, two Vs
3
3
y
0q states the 2422 S0q and the 2341 P0q states
Ž
.
which strongly contribute to the absorption at 235
nm have been already exhibited as the Br) qCl
channel. In the present study, we have suggested that
two Vs0q states correlate with the BrqCl)
andror the BrqCl channel irrespective of the nona-
diabatic transition at AC1 and AC2. We show the
3
exact dynamics of the 2341 P0q state of BrCl to
Fig. 1. Adiabatic correlation diagram of BrCl. The transitions
which are allowed only to the states with V s0 and 1 are
displayed. The vertical arrows represent the excitations by the 235
nm photons.
verify our suggestion at 235 nm and the possibility
of BrCl as a reference compound for the determina-
tion of the REMPI strengths for both bromine and

(
)
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
431
w
x
chlorine atoms. We have estimated the ionization
probabilities of Br, Br), Cl, and Cl) near 235 nm.
In order to obtain the angular and speed distributions
and to determine the branching ratio of each dissoci-
ation channel, the fragment ion-image technique was
employed. We report REMPI probe lines for bromine
atoms in the region of 232–236 nm. We have as-
signed the low and intermediate states of the 13
two-photon REMPI lines by investigating two-di-
mensional images and compared them with known
and at 233.69 for Br using the 2q1 -REMPI tech-
nique within the same laser pulse. The laser wave-
length was scanned over the range of ca. 4 cmy1 to
detect all velocity components of the fragments.
The ion fragment cloud was projected onto a
two-dimensional position sensitive detector plate,
Ž
.
consists of a microchannel plate MCP rphosphor
screen Galileo, FM2040 and a charge coupled de-
Ž
.
Ž
.
Ž
.
vice CCD camera Photometric, CH250 . The im-
age was summed over at least 10 000 shots and the
background was removed by subtracting the refer-
ence image collected at an off-resonant wavelength
under the same conditions. The TOF mass spectra
photomultiplier tube
Hamamatu, 1P21 instead of the CCD camera. The
2q1 -REMPI spectrum of bromine atom was ob-
w
x
literature values 21 . Among these lines, a pair of
REMPI probe lines at 233.69 nm for Br and 234.03
nm for Br) were chosen to be good probe lines
because the wavelengths are close and the ionization
signal intensities are strong.
were acquired using
a
Ž
w
.
x
tained by placing a photomultiplier tube at the back
of phosphor screen and was scanned between 232
and 236 nm. The ion signals from the phosphor
screen were monitored using a boxcar integrator.
During the scan, the power of the laser was main-
tained within "10%.
2. Experiment
The photofragment ion-image system used in this
study has been described elsewhere in detail 22 .
w
x
Ž
.
BrCl 20 Torr seeded in ca. 1.4 atm He was injected
into the reaction zone through a skimmer and a 1
mm pinhole using a pulsed molecular beam valve
3. Result and analysis
Ž
.
General Valve Series 9 operated at 10 Hz. BrCl
Ž
Ž
was synthesized by mixing 200 Torr Cl2 gas MG
IND. purity )99.9% with 10 Torr Br2 vapor Al-
drich, purity )99.5% in a mixing chamber and was
3.1. Assignment of 13 REMPI lines for bromine
atoms in the region of 232–235 nm
.
.
kept overnight to reach equilibrium. Because the
absorption cross-section of Cl2 is smaller than those
CCl3Br is used as a source of Br and Cl atoms
owing to its large absorption cross-section and high
of Br2 and BrCl at 235 nm Cl2: 0.04=10y20, Br2:
Ž
2
0.9=10y20, and BrCl: 6=10y20 cm , excess Cl2
product quantum yields at 235 nm 22 . Thirteen
REMPI lines of Br were observed in the region of
232–236 nm as shown in Fig. 2. Since the power
dependence of REMPI lines was found to be
quadratic, the observed REMPI lines seem to be
.
w x
Ž
in the equilibrium gas Br2 qCl2 2BrCl: Keq s
.
6.4 does not interfere with the detection of
w x
photofragments from the dissociation of BrCl 8 .
Ž
.
Ž
CCl3Br 10 Torrr1.4 atm He and CF3Br 20
Torrr1.4 atm He were injected into the reaction
.
w
x
produced by the 2q1 photon process. The wave-
zone in the same manner as BrCl.
lengths of REMPI lines in Table 1 were calibrated
Ž
A linearly polarized UV laser light typically 50
against the well-established REMPI line of Cl at
2
mJrpulse , generated through doubling the output of
a Nd:YAG 355 nm pumped dye laser, was focused
perpendicularly onto the ionization zone of the
235.336 nm 4p D3r2 3p 2 P3r2 23 . The low and
intermediate states of 10 lines among the observed
lines were assigned, and they coincide with the
.
Ž
. w x
Ž
.
Ž
.
w x
molecular beam with a lens fl s150 mm and
paralleled onto an image plane. CCl3Br, CF3Br, and
BrCl were photolyzed by the UV laser light, and the
fragments were then selectively ionized at 235.20 nm
for Cl), at 235.34 nm for Cl, at 234.03 nm for Br),
literature 21 . Five lines originated from the first
2
)
Ž
.
excited state of bromine Br , 4p P1r2 , and the
2
Ž
.
other five lines from the ground state Br, 4p P3r2
However, it is difficult to assign three lines numbers
2, 11 and 13 in Table 1 because of the large
.
Ž
.

(
)
432
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
Fig. 2. Two photon atomic transitions of bromine atom: CCl3Br was used as a precursor.
difference between the obtained values and the re-
ported values. Among these assigned lines, we se-
lected a pair of lines as reference lines, 233.69 nm
for Br and 234.03 nm for Br), to identify the origins
of the three lines.
We have investigated the angular distributions of
bromine atoms to identify which of the three unas-
signed lines is from Br or Br), using the two-dimen-
sional imaging technique. The angular distribution of
bromine atoms is given by
Table 1
Two photon atomic transitions of bromine in the region of 232–236
nm
No.
Wavelength
Literature
State
b
Ž
.
Ž
.
nm
value
Lowera
Upperb
1
2
3
4
5
6
7
8
9
10
11
12
13
232.72
232.96
233.05
233.11
233.21
233.69
234.03
234.34
234.56
234.85
235.22
235.86
235.98
232.72
232.94
233.06
233.12
233.21
233.70
234.04
234.35
234.57
234.86
235.32
235.66
235.87
Br
6p4 D3r2
7p4 P5r2
6p4 D5r2
6p4 P1r2
6p4 D7r2
6p4 P3r2
6p2S1r2
6p2 D5r2
6p4S3r2
6p2 D3r2
Br)
Br
1
4
P u s
Ž .
1qbP cos u
2 Ž
2
Ž .
.
p Ž
.
Br
Br
Br
Ž .
where P u is the ion intensity at u angle between
the laser polarization axis and the recoil velocity
vector, and the b value indicates the anisotropy
Br)
Br)
Br)
Br)
Br)
Br)
Br)
Ž
.
Ž
.
parameter y1(b(2 , and P2 cos u is the sec-
ond-order Legendre polynomial. If the b values of
Br at 233.69 nm and Br) at 234.03 nm are quite
different, we can measure the origins of the unas-
signed lines by comparing their b values with the b
values of reference lines. We have used CF3Br as a
precursor for Br and Br) and observed a large
5p2 F5r2
X 2
5p
P
1r2
X 2
5p
P
3r2
a
Br and Br) refer to the 4p2 P3r2 and 4p2 P1r2
.
b
w
x
Ref. 21 .

(
)
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
433
)
Ž
.
Ž
.
Ž
.
and l2. N Br rN Br
l2
difference of b values, Br bs0.66 at 233.69 nm
denotes the ratio of Br
l1
and Br) bs1.83 at 234.03 nm 22 . The b
values of bromine images acquired at three lines
were 1.69"0.10, 1.65"0.10, and 1.69"0.10, re-
spectively. These results show that the three lines
originate from Br). According to the selection rules
and Br) formed at photolysis wavelengths, l1 and
Ž
.
w x
)
Ž
.
Ž
.
l2. F Br rF Br
indicates the ratio of prod-
l1
l2
uct quantum yields of two different state bromine
atoms, and slR – BrrslR – Br the absorption cross-sec-
tions at two di1fferent resonance wavelengths. Using
2
Ž
Ž .
for the two photon transition of a heavy atom Dls
Eq. 4 we can obtain the product quantum yields
.
0,"2, D Js0,"1,"2 , the three lines have been
from the ion signals, when the ionization probabili-
ties of Br and Br) are known.
assigned in Table 1.
When these resonance wavelengths of REMPI
probe lines, l1 and l2 , are very close, the absorption
cross-sections of alkyl bromide at l1 and l2 can be
approximated to be equal in a practical sense, espe-
3.2. Photodissociation of BrCl at 235 nm
Since the REMPI technique is highly selective
and sensitive, Br and Br) can be ionized selectively
at two different resonance wavelengths. The product
quantum yields of these two different energy state
atoms at a given photolysis wavelength can then be
expressed by .
Ž .
cially in the case of one-color experiments. Eq. 4
Ž .
can be approximated by Eq. 5 . Further, in order to
obtain precise quantum yields, the ionization proba-
bilities of Br at l1 and Br) at l2 have to be large in
order to obtain a high signal-to-noise ratio. The good
REMPI probe lines satisfying the conditions of
closeness and strong ionization probabilities are use-
ful for determining the branching ratio of X and X).
Among newly found lines, a pair of good REMPI
probe lines at 233.69 nm for Br and at 234.03 nm for
S Br
f Br
N Br
l1
Ž
.
Ž
.
Ž
.
l1
s
3
Ž .
S Br)
f Br)
N Br)
Ž
.
Ž
.
Ž
.
l2 l2
R – Br
f Br
F Br
s
s
Ž
.
Ž
.
l1
l1
l1
s
f
4
Ž .
f Br)
F Br)
R – Br
l2
Ž
.
Ž
.
l2
l2
Br) were chosen.
f Br
F Br
Ž
.
Ž
.
l1
l1
)
Ž
.
Ž
.
f Br y233.69 nmrf Br
was estimated to
5
Ž .
y234.03 nm
f Br)
F Br)
Ž
.
Ž
.
l2
l2
be 0.42"0.02 from the Br2 photolysis under the
)
Ž
.
Ž
Ž
.
.
w x
where S Br rS Br is the ion signal ratio of Br and
same experimental condition 24 . In this study, the
)
)
)
Ž
.
Ž
.
Ž .
Br , f Br rf Br
the ionization probability ra-
l2
ratio of N Br rN Br was found to be 1.38"0.12.
l1
tio of Br and Br) at their resonance wavelengths, l1
From the ratio, the relative quantum yields, F Br
)
Ž
.
)
Ž .
Ž .
Fig. 3. Raw ion images of the chlorine fragments from the BrCl photolyses at 235 nm. a and b represent the Cl and Cl images. The
polarization vector of the laser is vertical.

(
)
434
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
Table 2
The relative quantum yields for three channels
Channel
b
Speed of chlorine fragments
F
xI
xH
Br) qCl
BrqCl)
BrqCl
1.98"0.05
y1.04"0.05
1.88"0.05
3101 m sy1
3306 m sy1
3368 m sy1
0.58
0.16
0.26
1
0
0.96
0
1
0.04
Ž
.
and F Br , were determined to be 0.58"0.02 and
0.42 " 0.02, respectively. We have measured
the parallel transition while the Cl) image in Fig. 3b
shows the characteristics of the perpendicular transi-
tion. The projected velocity distributions of the chlo-
rine fragments show maxima at approximately 3370
and at 3100 m sy1 as shown in Fig. 3a, and at 3300
m sy1 as shown in Fig. 3b. The theoretical recoil
speeds of chlorine fragments for the dissociation
)
)
Ž
Ž
.
Ž
.
Ž
.
Ž .
N Br rN Br rather than N Cl rN Cl because
)
.
Ž
.
f Cl rf Cl at 235 nm is not available due to the
lack of a proper reference compound.
Two-dimensional images of chlorine fragments
produced from the BrCl photolysis at 235 nm are
displayed in Fig. 3. These raw images are two-di-
mensional projections of the three-dimensional speed
and angular distributions of the chlorine fragments.
The Cl image in Fig. 3a shows the characteristics of
Ž . Ž .
channels a – c can be calculated based on the
energy and momentum conservation. The internal
energy of parent molecule is assumed to be zero
under the supersonic molecular beam condition. The
Ž
.
Fig. 4. The speed distribution of Cl fragments at 235 nm Fig. 3a is fitted in terms of three Gaussian functions. Two Gaussian functions
corresponding to dissociation channels, BrqCl and Br) qCl, are displayed in dashed and dot lines, respectively. The straight line indicates
another Gaussian function corresponding to background noise. The open triangle denotes the sum of the three Gaussian functions.

(
)
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
435
Ž . Ž .
calculated speeds for dissociation channels a – c at
tion axis and the recoil velocity of the bromine
fragment. The b values extracted from the Cl) and
Cl images at 235 nm photolysis are listed in Table 2.
235 nm are listed in Table 2. When the bromine
fragments were probed, the translational energy dif-
Ž .
Ž .
ference between dissociation channels a and b is
only 10 kJ moly1, which is too small to resolve these
channels. Since the translational energy difference
4. Discussion
Ž .
Ž .
between dissociation channels a and c is 44 kJ
moly1, which is enough to resolve them when the
chlorine fragments were probed, the images were
acquired to probe chlorine fragments rather than
bromine fragments.
3
4.1. The role of P0 q in the photodissociation of
BrCl
In the photolysis of BrCl at 235 nm, three dissoci-
ation channels Br qCl, BrqCl , and BrqCl
were observed. On the other hand, in the previous
In order to reduce the noise effect on the transfor-
mation, the raw images were smoothed through the
Gaussian filter with a 5=5 window and a typical
standard deviation of 2 in a pixel unit. Afterwards,
they were used to reconstruct three-dimensional ve-
locity distributions by performing the inverse Abel
transformation 25,26 . By integrating the recon-
structed three dimensional speed distribution over all
angles at each speed, the speed distribution for the Cl
fragments has been obtained as shown in Fig. 4. The
)
)
Ž
.
study focused in the visible region, only two chan-
)
)
Ž
.
w x
nels Br qCl and BrqCl were observed 15 . A
discrepancy similar to that found in the BrCl results
w
x
was also reported in ICl photodissociation 18,19 .
The discrepancy in ICl was attributed to be the
misinterpretation of the Doppler spectrum of the Cl
w
x
3
fragments, and it was concluded that the 2341 P0q
X 1 Sq0q transition makes a minor contribution to
the ICl absorption at 235 nm based on the absence of
the IqCl channel. The BrqCl channel, therefore,
is very important as an evidence of the contribution
of the 2341 3P0q state in the photodissociation
dynamics in the UV region.
)
)
Ž
. Ž Ž
.
Ž
branching ratio, N Br qCl r N BrqCl qN Br
..
qCl , was obtained from the speed distribution as
shown in Fig. 4. Each speed distribution has been
fitted into three Gaussian functions, which represent
the BrqCl formation channel, the Br) qCl chan-
nel, and broad background noise. N Br) q
Ž
3
The 2341 P0q state correlates to the BrqCl)
)
. Ž Ž
.
Ž
..
Cl r N BrqCl qN Br qCl was calculated by
integrating over the fitted functions and found to be
0.69"0.02 as shown in Fig. 4. The relative quantum
yield can be calculated from the following relation-
ships and are listed in Table 2.
channel adiabatically via AC1 and AC2 as shown in
Fig. 1. Since this channel is also formed from the
2341 1P1 state, it may exhibit both parallel and
perpendicular components. In order to estimate the
relative contributions of these components to the
dissociation channels, classical relationships were
used:
F Br) qCl sF Br) s0.58"0.02
6
Ž .
Ž
.
Ž
.
N Br) qCl
Ž
.
N Br) qCl qN BrqCl
bsxI bI qxH bH
xI qxHs1
9
Ž .
Ž
.
Ž
.
10
F Br) qCl
Ž
.
Ž
.
s
F Br) qCl qF BrqCl
.
bI and bH represent the anisotropy parameters for
pure parallel and perpendicular transitions. The xI
and xH values, the fractions for each dissociation
channel, are listed in Table 2.
Ž
.
Ž
s0.69"0.02
7
Ž .
F Br) qCl qF BrqCl) qF BrqCl s1 .
Ž
.
Ž
.
Ž
.
The absorption cross-section corresponding to the
8
Ž .
transition, 2341 P0q X 1 Sq0q, should be responsi-
3
Ž .
The angular distribution, P u , has been obtained
by integrating the reconstructed three-dimensional
speed distribution over a proper range of speed at
each angle. u is the angle between the laser polariza-
ble for the parallel component of the BrqCl) chan-
nel andror the existence of the BrqCl channel with
bs2 via nonadiabatic transition at AC2. The anisot-
ropy parameter for the BrqCl) channel, bs

(
)
436
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
y1.04"0.05, indicates that the Vs0q state the
3P0q state does not correlate with the BrqCl
)
Ž
.
3
2341 3P0q andror the 2422 Sy0q state does not
channel adiabatically but with the BrqCl channel
via nonadiabatic transition at AC2. We suggest that
Pr2 , the nonadiabatic transition probability at AC2,
is nearly 1 based on the above statements.
.
contribute to the formation of the BrqCl) channel.
It has been concluded that the channel only arises
from the 2341 P1 state. This conclusion coincides
1
with the previous studies of BrCl and ICl systems, in
which it is suggested that the contribution of the
2341 3P0q state to the absorption at 235 nm was
neglected due to the absence of the parallel compo-
It is possible to determine the range of the contri-
bution of the 2342 3P0q using the relationships with
the nonadiabatic transition probability at AC1, Pr1:
Sy0
P0
P1
P1
x2422 3 qqx2341 3 qqx2341 1 qx2341 3 s1 11
Ž
.
nent for the BrqCl) channel 15,18,19 . On the
contrary, the discovery of the BrqCl channel with
w
x
F Br) qCl sx
1yPr qx
2422 3 Sy0
1
2341 3P0
qPr1
Ž
.
q Ž
.
Ž
.
the quantum yield F BrqCl s0.26 in the present
study seems not to be in good agreement with the
suggestion. The BrqCl channel can be produced by
several pathways as shown in Fig. 1. The pathways
12
Ž
.
where xi indicates the fraction of the i state. The
absorption at 235 nm is assumed to be only due to
3
3
the transition to the 2422 Sy0q, the 2341 P0q, the
3
y
Ž .
with Vs0 are 1 the 2422 S0q state via nonadia-
1P1 and the P1 states from the X 1 Sq0q state. The
3
Ž .
batic transitions at AC1 and at AC2, 2 the 2341
3P0q state via nonadiabatic transition at AC2, and
values of the x2341 1 qx2341 3
are found to be
P1
P1
3
3
Sy0
q
0.17 as listed in Table 3. X and Y represent x2422 3
Ž . Ž .
Ž .
the 2431 P1, and 6 the A P1 states as candidates
3 the B P0q, while there are 4 the 2341 P1, 5
1
3
and x2341 3 q, respectively. Y and Pr1 can be rear-
Ž .
with Vs1. Among these states, 3 B P0q, 5 the
P0
3
ranged into the simple expression:
Ž .
Ž .
1
3
Ž .
2431 P1, and 6 A P1 states are excluded since
their vertical energies are too low to absorb the
photon at 235 nm. According to bs1.88"0.05 for
BrqCl, it can be suggested that the formation of
this channel results mainly from the Vs0q states
Yy0.25
2Yy0.83
0.83Pr1 y0.25
2 Pr1 y1
Pr1 s
,
Ys
.
13
Ž
.
3
Two ranges of the contribution of the 2341 P0q
Ž .
state Y and the nonadiabatic transition probability
the 2422 S0q andror the 2341 3P0q state and the
Pr1 are calculated from Eq. 13 under the condi-
3
y
Ž
.
Ž
.
Ž .
3
Ž .
minor portion originates from 4 the 2341 P1 state.
tion of 0(Y(0.83 and 0(Pr1 (1. One set is
0.58(Y(0.83 and 0.7(Pr1 (1. The other is 0(
Y(0.25 and 0(Pr1 (0.3. We have excluded the
0(Y(0.25 and 0(Pr1 (0.3 region for two rea-
3
However, the exact contribution of the 2341 P0q
state to the formation the BrqCl channel cannot be
measured because the product from the two Vs0q
3
3
y
Ž
.
w x
states the 2422 S0q andror the 2341 P0q state
sons. In the ICl study 19 , the high value of the
Ž
.
were not distinguished in our system. The two results
nonadiabatic transition probability Pr1 at AC1 has
been determined. According to the ICl energy level,
the 2422 3 Sy0q state is located at a 4000 cmy1 higher
energy than the 2341 3P0q state, which shows a
maximum at 235 nm in the Frank–Condon region.
Ž
in this study the strong perpendicular characteristics
for BrqCl) with bsy1.04"0.05 and the exis-
.
tence of BrqCl with bs1.88"0.05 indicate that
3
y
the Vs0q state s the 2422 S0q andror the 2341
Ž . Ž
Table 3
The electronic excited states relevant to the channels
Channel
F
Assignment
Br) qCl
I
0.58
0
0.16
0.25
0.01
2422 3 Sy0 , 2341 3P0
2422 3 Sy0
q
q
q
3
BrqCl)
I
2341 3P0 , 2422 Sy0
2341 3P0
q
q
q
H
I
H
2341 1P1
3
BrqCl
2341 3P0
2431 3P0 , 2422 Sy0
2341 3P0
2431 3P0
q
q
q
q
q
2341 3P1

(
)
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
437
Table 4
Ž
.
Ratios of 2q1 -REMPI probabilities of bromine and chlorine atoms
)
)
)
)
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
f Cl rf Br
f Cl rf Br
f Cl rf Br
f Cl rf Br
1.24"0.30
3.35"0.84
1.46"0.42
3.94"0.97
) Br, Br), Cl and Cl) were ionized at 233.69, 234.03, 235.336 and 235.204 nm.
)
)
Ž
.
Ž
.
Ž
.
Because of the similarity of ICl and BrCl except for
the energy levels of BrCl which are slightly higher
than that of ICl, it is expected that the absorption
corresponding to the transition to the 2341 3P0q
state at 235 nm is strong. Consequently, 0.58(
f Br
were obtained easily using f Br rf Br
)
Ž
.
Ž
.
and f Cl rf Cl in Table 4.
f Cl
f Br)
S Cl
Ž
F Br) qCl
Ž
.
.
Ž .
s
.
S Br) F Br) qCl qF BrqCl
Ž
.
Ž . Ž . Ž .
x2341 3 q(0.83 and 0.7(Pr1 (1 have been cho-
15
Ž
.
P0
Ž .
sen. The value of 0.58 which resulted from Eq. 6
The newly found good REMPI probe lines are
very useful for determining the ratio of quantum
yields of Br and Cl producing channels, and thus for
investigating the photodissociation reactions of envi-
ronmentally harmful Br and Cl containing com-
signifies the lower bound of the contribution of the
2341 P0q state.
3
In the photodissociation dynamics of BrCl at 235
nm, the quantum yields for the dissociation channels
and the angular distribution are reported. The 2341
3P0q state has been found to contribute to the
formation of the BrqCl channel. It is suggested that
the lower bound of the fraction of the state is 0.58,
and the nonadiabatic transition probabilities at AC1
and AC2 with 0.7(Pr1 (1 and Pr2 s1 are ob-
tained.
Ž
.
pounds such as polychloro bromo dibenzo-p-
Ž
.
Ž
.
dioxin PCDD and CF2 BrCl halon 1211 .
Acknowledgements
The authors gratefully acknowledge the Korea
Research Foundation for its support of this research
by the Korea–Germany Joint Project, 1998–2001.
)
(
)
(
)
4.2. Determination of f Cl rf Br
Ž
.
The ion signal intensity, S Br , is proportional to
the ionization probability f Br and the sum of the
Ž
.
quantum yields of the channels producing Br:
References
S Br Af Br F qF
. Ž
.
b
14
Ž
.
Ž
.
Ž
.
a
w x
Ž
.
1
w x
2
M.C. Heaven, Chem. Soc. Rev. 15 1986 405.
)
Ž
.
Ž
.
In order to determine f Cl rf Br , we have
L.A. Barrie, J.W. Bottenheim, R.C. Schnell, P.J. Crutzen,
measured the ion intensities of Cl and Br) pho-
tolyzed from the BrCl precursor and the REMPI
probe lines of bromine and chlorine in the region of
233.69–235.34 nm. In TOF mass spectrometry, a
long ion flight time decreases the ion peak intensity
because of the scattering in the flight region. Con-
trolling the repelling electrode voltage, the flight
time of the Cl ion could be adjusted to the same as
that of the bromine. The ion peak intensity ratio of
Ž
.
R.A. Rasmussen, Nature 334 1988 138.
w x
3
Ž
.
J.G. Anderson, D.W. Toohey, W.H. Brne, Science 251 1991
39.
E. Tschuikow-Roux, T. Yano, J. Niedzielski, J. Chem. Phys.
w x
4
Ž
.
82 1985 65.
w x
Ž .
5
J. Tellinghuisen, J. Chem. Phys. 89 1988 6150.
w x
Ž
C.E. Moore, Atomic Energy Levels, vol. 2 NSRDS-NBS35
6
.
1971 .
w x
7
D. Maric, J.P. Burrows, G.K. Moortgat, J. Photochem. Pho-
Ž
.
tobiol. A: Chem. 83 1994 179.
w x
8
S. Hubinger, J.B. Nee, J. Photochem. Photobiol. A: Chem.
)
)
Ž
.
Ž
.
Cl and Br , S Cl rS Br , was 1.80"0.44. The
Ž
.
86 1995 1.
)
Ž
.
Ž
.
Ž .
f Cl rf Br calculated by Eq. 15 was 1.24"0.30.
w x
9
J.J. Wright, W.S. Spates, S.J. Davis, J. Chem. Phys. 66
)
)
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Similarly, f Cl rf Br , f Cl rf Br and f Cl r
1977 1566.

(
)
438
M.-S. Park et al.rChemical Physics Letters 322 2000 429–438
w
w
x
Ž
.
Ž
.
10 J.A. Coxon, J. Mol. Spectrosc. 50 1974 142.
Yabushita, S. Fujimura, K. Saito, J. Chem. Phys. 99 1993
3461.
x
11 M.A.A. Clyne, I.C. Zai, J. Chem. Soc. Faraday Trans. 2
Ž
.
w
w
w
w
w
x
1974 1221.
19 L.J. Rogers, M.N.R. Ashfold, Y. Matsumi, M. Kawasaki, B.
w
w
w
w
x
Ž .
Whitaker, Chem. Phys. Lett. 258 1996 159.
12 A. Hopkirk, D. Shaw, R.J. Donovan, K.P. Lawley, A.J.
Ž
.
x
Yencha, J. Phys. Chem. 93 1989 7338.
20 K.-W. Jung, T.S. Ahmadi, M.A. El-Sayed, J. Phys. Chem. A
x
Ž .
101 1997 6572.
13 S.W. Brown, C.J. Dowd Jr., J. Tellinghuisen, J. Mol. Spec-
Ž
.
x
trosc. 132 1988 178.
21 C.E. Moore, Atomic Energy Levels, vol. 2, NBS Circ. 35,
x
Ž
.
14 J. Cao, H.P. Loock, C.X.W. Qian, Can. J. Chem. 72 1994
Natl. Bur. Stand., Washington, DC, 1971.
x
758.
22 Y.-J. Jung, M.S. Park, Y.S. Kim, K.-H. Jung, J. Chem. Phys.
x
Ž .
111 1999 4005.
15 M.J. Cooper, P.J. Jackson, L.J. Rogers, A.J. Orr-Ewing,
Ž
.
x
23 S. Arepalli, N. Presser, D. Robie, R.J. Gordon, Chem. Phys.
M.N.R. Ashfold, B.J. Whitaker, J. Chem. Phys. 109 1998
Ž .
Lett. 118 1985 88.
4367.
w
x
Ž
.
w
x
16 Y.S. Kim, Y.-J. Jung, K.-H. Jung, J. Chem. Phys. 107 1997
24 Y.-J. Jung, Y.J. Jee, K.-H. Jung, submitted.
w
x
3805.
25 S.M. Candel, IEEE Trans. Acoust. Speech Signal Proc. 29
w
w
x
x
Ž
.
Ž
.
17 C.-K. Ni, G.W. Flynn, Chem. Phys. Lett. 210 1993 333.
18 K. Tonokura, Y. Matsumi, M. Kawasaki, H.L. Kim, S.
1981 963.
w
x
Ž
.
26 R.N. Zare, D.R. Herschbach, Proc. IEEE 51 1963 173.