I*(2P1/2) and Cl*(2P 1/2) production from chloroiodobenzenes in the ultraviolet

Article abstract of DOI:10.1021/jp049345k

The relative quantum yields of I*(²P_1/2) and Cl*(²P_1/2) production, φ(I) and φ*(Cl), respectively, have been measured at four different ultraviolet excitation wavelengths, e.g., 222, 236, 266, and 280 nm in the

Full text of DOI:10.1021/jp049345k

J. Phys. Chem. A 2004, 108, 7949-7953
7949
I*(2P1/2) and Cl*(2P1/2) Production from Chloroiodobenzenes in the Ultraviolet†
Dulal Senapati, Sandip Maity, and Puspendu K. Das*
Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore-560012, India
ReceiVed: February 13, 2004; In Final Form: May 31, 2004
The relative quantum yields of I*(2P1/2) and Cl*(2P1/2) production, φ*(I) and φ*(Cl), respectively, have been
measured at four different ultraviolet excitation wavelengths, e.g., 222, 236, 266, and 280 nm in the
photodissociation of o-, m-, and p-chloroiodobenzenes. The measured I* and Cl* quantum yields are, with
some exceptions, higher than those obtained, respectively, from iodobenzene and chlorobenzene, at the same
wavelengths. While at most wavelengths the major fraction of the iodine atoms is produced in the excited
state, the opposite is true for the chlorine atoms. Both direct and indirect dissociation pathways are involved
in the production of I* atoms whereas Cl* is produced only by indirect pathways since direct excitation of
the σ*(C-Cl) r n(Cl) transition is not possible at these wavelengths. The halogen atom in the ortho position
is found to be most effective in enhancing the yield of the other spin-orbit excited halogen atom in the
photolysis. While the nature of the initial transition, the extent of intersystem crossing in the excited states
and the exit channel effects need to be considered in interpreting the quantum yield results, some factors
seem to be more effective in influencing the final outcome. Induced dipole-induced dipole and quadrupole-
quadrupole interactions between the two halogen atoms (I and Cl in this case) seem to play an important role
in the exit channel dynamics. These electrostatic interactions facilitate the intersystem transfer in the excited
state and subsequent production of the spin-orbit excited halogen atoms.
1. Introduction
Reports on aryl dihalides and mixed halide systems are not
many although photodissociation of pentafluoroiodoiodobenzene
Like alkyl and aryl monohalides, aryl dihalides are potential
source of active halogen atoms and radicals in the upper
atmosphere and ocean water1 due to their strong absorption in
the UV region. In the upper atmosphere the halogen radical
reactions play an important role in depleting the ozone layer.2
Therefore, studying and understanding halogen radical produc-
tion in the gas phase, from the photolysis of various halogen
containing compounds, remain active areas of research.
The photodissociation dynamics of aryl halides in the
ultraviolet has been studied over the past 25 years starting with
the pioneering work by Bersohn and co-workers.3-5 Specifically,
the dissociation of iodobenezene6-10 and chlorobenzene11-15 in
the ultraviolet has been studied extensively. It has been
established that the mechanism of photodissociation of aryl
halides are qualitatively different from that of alkyl halides.
Unlike in alkyl iodides where the excitation in the 200-305
nm region is localized on the C-I bond (σ*(C-I) r n(I)
transition) and the dissociation is direct, the predominant part
of halobenzene dissociation is indirect and involves the π* r
π, σ* r π and π* r n transitions localized on the aromatic
has been studied in detail.8,10 In their initial time-of-flight (TOF)
dissociation studies Bersohn and co-workers3-5 investigated
many diiodoaromatics in the ultraviolet region. They concluded
that a small fraction of the available energy is disposed in
translation of the product fragments and the dissociation time
scale is on the order of 1 ps. Later Nishi and co-workers studied
the dissociation mechanism of o-, m-, and p-dichlorobenzene
at 193 nm using the TOF technique.19 They observed, in
agreement with Bersohn’s earlier work, that less than half of
the available energy goes into product translation. They also
observed three peaks in the TOF signals from the Cl atom and
assigned the peaks to three different dissociation mechanisms.
Using polarized light to decompose signals from various
transitions, they concluded that the three different dissociation
channels are (1) a fast direct dissociation channel through the
repulsive σ* state, (2) an indirect predissociation channel via
the triplet excited state, and (3) another rather slow predisso-
ciation channel through the highly excited vibrational levels
close to the dissociation level of the ground state potential energy
surface. For the isomers of dichlorobenzene, they predicted that
the three channels are equally probable like in chlorobenzene,14
and no significant change was caused by the additional chlorine
atom. They also speculated that Cl* production would be a
minor process in the dissociation. A later report on o-
dichlorobenzene photodissociation at 266 nm by Zhu et al.20
corroborated the results of Nishi and co-workers at 193 nm.
1
1
ring. All the six transitions (1B2u(S1) r A1g(S0), B1u(S2) r
1A1g(S0), 1E1u(S3) r 1A1g(S0), 3B1u(T1) r 1A1g(S0), 3E1u(T2) r
1A1g(S0) and B2u(T3 ) r A1g(S0)) in benzene16 are also seen
in iodobenzene. However, when fluorine atom is substituted in
the place of iodine atom, the T2 r S0 band disappears and a
new singlet-singlet transition around 230 nm appears.17,18 In
general, substitution of hydrogen atoms by fluorine or other
halogen atoms alters (stabilizes) both the π and σ molecular
orbitals of benzene.
3
1
In this paper, we have investigated o-, m-, and p-chloroiodo-
benzenes along with iodobenzene and chlorobenzene at four
different ultraviolet wavelengths (222, 236, 266, 280 nm) and
probed the relative quantum yields of I* and Cl* production
directly. Two different resonance enhanced multiphoton ioniza-
tion (REMPI) schemes have been employed to detect iodine21,22
† Part of the special issue “Richard Bersohn Memorial Issue”.
* Corresponding author. Email: pkdas@ipc.iisc.ernet.in. Fax: 91-80-
23601552. Telephone: 91-80-22932662.
10.1021/jp049345k CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/09/2004

7950 J. Phys. Chem. A, Vol. 108, No. 39, 2004
Senapati et al.
and chlorine23,24 atom fragments in the ground (2P3/2) as well
as excited (2P1/2) states. The C-Cl bond dissociation energy is
97 kcal/mol in dichlorobenzene and the C-I bond dissociation
energy is 64 kcal/mol in iodobenzene. Assuming that the C-Cl
and C-I bond energies in chloroiodobenzenes will be similar
to those in dichlorobenzene and iodobenzene, respectively, all
four channels shown below
ClC6H4-I f ClC6H4 + I/I*
f IC6H4 + Cl/Cl*
are accessible at the photon energies employed in our experi-
ments. Further dissociation of C6H4X via the initial single photon
excitation is not energetically possible in the near-ultraviolet.
The basis of selecting these molecules is that, if there is any
effect of the chlorine (iodine) atom on the dissociation mech-
anism (dynamics) of the other carbon-halogen bond, it will be
more pronounced in these examples. The difference in electron
affinity and size of the chlorine atom is much larger than that
of the iodine atom and the interhalogen interaction is expected
to change in going from the para isomer to the ortho isomer.
The interatomic interaction between the two halogen atoms may
occur via induced dipole-induced dipole or quadrupole-
quadrupole interaction through space when they are situated
close to each other as in the ortho configuration, and this may
alter the exit channel dynamics of bond dissociation.
2. Experiment
The experimental setup has been described in detail else-
where.25 In brief, the photodissociation experiments were carried
out inside a stainless steel chamber, evacuated continuously by
a diffusion pump backed by a rotary pump to a base pressure
of 10-6 Torr. A constant sample pressure between 200 and 400
µTorr was maintained for I and I* detection and 50-800 µTorr
for Cl/Cl* detection. The pump and probe laser beams were
aligned perpendicular to each other at the center of the chamber
between two electrodes (2.5 × 2.5 cm2 and 0.5 mm thick
stainless steel plates) for the I atom detection. For the chlorine
atom detection, we aligned the pump and the probe lasers
parallel (counterpropagating) to each other. The delay between
the pump and probe lasers was maintained between 50 and 100
ns. The pump and probe laser wavelengths were generated by
various frequency doubling and mixing schemes described in a
previous paper.25 The probe beam was focused using a short
focal length (5-in. fused silica) lens and the pump beam was
contracted (0.5 cm dia) by a long focal length (typically 1 m
Figure 1. Absorption spectra of o-, m-, and p-C6H4ICl along with C6H5I
and C6H5Cl in cyclohexane.
TABLE 1: O*(I) at Various Photolysis Wavelengths from
Different Compoundsa
compounds
280 nm
266 nm
236 nm
222 nm
o-C6H4ICl 0.69 ( 0.02 0.78 ( 0.02 0.54 ( 0.01 0.73 ( 0.02
m-C6H4ICl 0.67 ( 0.005 0.77 ( 0.02 0.63 ( 0.01 0.72 ( 0.04
p-C6H4ICl 0.57 ( 0.03 0.65 ( 0.01 0.33 ( 0.01 0.65 ( 0.01
C6H5I
0.58 ( 0.01 0.51 ( 0.01 0.59 ( 0.01 0.62 ( 0.02
a The errors shown in the table are statistical errors obtained from,
at least, three measurements.
f.l.) lens The iodine and chlorine atoms in the ground (2P3/2
)
3. Results and Discussion
and excited (2P1/2) states produced in the photolysis were
detected by two different REMPI schemes.21,23 The anode was
maintained at +200 V and the voltage drop at the cathode across
a 1 MΩ/1 W resistor to the ground was taken as signal. The
output from the ion detector was collected through a homemade
electrometer, amplified (typically 25 times), averaged over 50
shots, and digitized in a digital storage oscilloscope. The pump
laser polarization was kept unchanged since, in our experiments,
iodine and chlorine atoms are detected which are spherically
symmetric and their recoil direction with the rotation of the
pump laser polarization should not influence the REMPI signal
intensity. All the compounds were purchased from Aldrich and
were fractionally distilled prior to each experiment. The
absorption spectra of o-, m-, and p-C6H4ICl along with C6H5-
Cl, and C6H5I in the ultraviolet were recorded in cyclohexane
in a BECKMAN DU-600 spectrometer (Figure 1).
The REMPI spectra for I, I* and Cl, Cl* were recorded by
scanning the probe laser across the (2+1) ionization lines of I
(304.67 nm), I* (304.02 nm),21 Cl (235.34 nm), and Cl* (235.21
nm).23 The relative quantum yield of spin-orbit excited halogen
atom, φ* ) N*/(N + N*), where N is the concentration of the
nascent spin-orbit state and * refers to the excited state, were
calculated from the REMPI signal intensities and are listed in
Table 1 and Table 2, for I* and Cl*, respectively. The measured
REMPI signal intensities S and S* may be expressed as
N* S* F
)
(1)
N
S F*
where F contains various factors such as ionization cross section
of the fragments, collection efficiency of the ions, and absorption
cross section of the molecule at the probe REMPI wavelengths.25

Photodissociation of Chloroiodobenzenes
J. Phys. Chem. A, Vol. 108, No. 39, 2004 7951
alkyl halides.29,30 All these three transitions lead to direct
dissociation pathways and only the 3Q0 r n transition produces
spin-orbit excited halogen atom. All the other three states
including the ground state, that is, 3Q1, 1Q1, and n states correlate
asymptotically to the ground-state halogen atom product. Of
the six π* r π type aromatic transitions mentioned earlier, only
four transitions (S1 r S0, T3 r S0, T2 r S0, S2 r S0) are
accessible in the 210-300 nm region.10 On the basis of the
selection rules for one photon transition, only excitations of the
type S1/S2 r S0 will carry most of the transition strength in the
wavelengths used in this study. Both the singlet excited states,
S1 and S2, are bound states, and they cross the σ*(C-X) states
to dissociate into products. Therefore, excitation to the ring
aromatic states in halobenzenes leads to indirect dissociation
via the triplet σ* r n states. This intersystem crossing is induced
by spin-orbit interaction as well as a vibrational mode that helps
symmetry breaking at the crossing region. In another competitive
relaxation pathway, a fraction of the excited singlet molecules
may undergo internal conversion to highly excited vibrational
levels of the ground electronic state and predissociate. All these
TABLE 2: Relative O*(Cl) at Various Photolysis
Wavelengths from Different Compoundsa
compounds
280 nm
266 nm
235 nm
222 nm
o-C6H4ICl
0.16 ( 0.02 0.57 ( 0.02 0.28 ( 0.01 0.36 ( 0.02
m-C6H4ICl 0.21 ( 0.02 0.33 ( 0.02 0.23 ( 0.01 0.35 ( 0.01
p- C6H4ICl NDb
ND 0.25 ( 0.01 0.32 ( 0.02
C6H5Cl ND 0.39 ( 0.01 0.14 ( 0.01 0.24 ( 0.02
a The errors shown in the table are statistical errors obtained from,
at least, three measurements. b ND indicates undetectable Cl and Cl*
signals from the photodissociation
We evaluated the ratio f ) F/F* for I and I* using I2 photolysis
process26 at 304.67 and 304.2 nm in our apparatus, which should
yield equal amounts of I and I*. We found the value of f )
1.00 ( 0.05 which is close to the expected value and the value
of 0.90 reported by Eppink et al.22 We have taken f as 1.0 and
the quantum yield at various wavelengths were determined
directly from I and I* REMPI signals.
For the Cl* quantum yield determination from REMPI signal
intensities, the factor f has been revised several times in the
literature. Initially Kawasaki et al.27 used the value of f ) 1,
using two photon transitions, 4p2D3/2 r 3p2P3/2 (235.336 nm)
and 4p2D3/2 r 3p2P1/2 (237.808 nm) albeit different frequencies
from ours, to detect chlorine atoms by a (2 + 1) REMPI scheme.
This was later modified by Gordon and co-workers24 as 0.85 (
0.10 to find an agreement between statistical and measured Cl*
quantum yields in HCl. Since we use the same REMPI probe
wavelengths used by Gordon et al.,23 our natural choice for f
would have been the same. However, in a recent measurement
of this factor f, Regan et al.28 have recommended a value of
1.06 ( 0.17 for the REMPI signal calibration of Cl and Cl* at
the same REMPI probe wavelengths. The large error in the
calibration is due to propagation of errors as ascribed by the
authors. For our relative quantum yield measurement we have
chosen the value of f ) 1.06, and the error in the data reported
in the two tables is the statistical error obtained from three or
more separate measurements on different days.
1
three mechanisms have more or less equal probability (∼ /3) of
occurrence in dichlorobenzene as put forward earlier by Nishi
and co-workers.19 In the following sections, we discuss sepa-
rately the differences observed in the I* and Cl* quantum yields
in the o-, m-, and p-chloroiodobenzenes.
A. I* Quantum Yield. From Table 1 it is clear that φ*(I)
for all the chloroiodobenzenes is more than 0.5 in all cases
except p-chloroiodobenzene at 236 nm. In other words, more
than 50% of the product iodine atoms are produced in their
spin-orbit excited state. This implies that, regardless of the
nature of the initially prepared excited state, the final state from
which the molecules dissociate correlates to the excited state
iodine atom. Only a minor fraction of the molecules dissociate
to produce the ground state iodine atom. Generally, we would
expect the relative I* yield to depend on (i) the electronic
character of the initial excited state(s), (ii) the extent of
intersystem crossing between the initially excited state and the
final state from which the molecule dissociate, and (iii) the exit
channel effects. With a few exceptions, the I* quantum yield
in chloroiodobenzenes is, in general, higher than that in
iodobenzene. While the exceptions point out the complex nature
of the dissociation, the data in Table 1 bring out the effect of
the chlorine atom on the production of I*(2P1/2) quantum state
in the dissociation. Two distinct effects are possibly exerted by
the chlorine atom to influence the dynamics of C-I bond
breaking in the excited state: (i) the reorganization of the π*
r π transition energy and (ii) the through space induced dipole-
induced dipole and quadrupole-quadrupole interactions with
the iodine atom (an exit channel effect). The first effect has
been realized from the absorption spectra of C6H5I and C6H5Cl
(Figure 1), and it is clear that there is a 8 nm red shift (from
257 to 265 nm) of the π* r π transition due to the presence of
chlorine atom in the ring. The shift indicates that the chlorine
atom stabilizes the π* antibonding orbital more than the π
bonding orbital. In the F atom substitution study on the I*
quantum yield in alkyl31as well as aryl iodides,8,10 the spin-
orbit interaction effects exerted by the F atom have been
recognized. In the direct dissociation mechanism, the additional
halogen atom (chlorine atom in the present context) increases
the singlet-triplet energy gap in the intersystem crossing region
and thus forces most of the molecules to dissociate from the
excited state which correlates to the I* product. The influence
is much more complex in the indirect dissociation mechanism
as discussed earlier by Kavita et al.10 in the case of pentaflu-
The main objectives of this study are to find out (i) whether
the photolysis of the mixed haloaryl compounds yield similar
amounts of the spin-orbit excited halogen atoms compared to
the corresponding monohalide compounds and (ii) how much
influence does a halogen atom exert on the yield and formation
of the other halogen atom in the excited state, specifically, when
they occupy neighboring positions.
From the previous reports on the dihalobenzene dissocia-
tion,3-5,19-20 it is clear that there are three pathways for
dissociation of these molecules in the ultraviolet. The initial
excitation takes the molecule to a one photon allowed state from
the ground state. Three such transitions are possible of which
two transitions are of σ* r n type centered on the two carbon-
halogen bonds and the third transition of π* r π type localized
on the aromatic core. The σ* r n transition originating from
the iodine 5p electrons of the C-I bond is weak and occurs
with a peak absorption at ∼260 nm with a fwhm of ∼50 nm.
The σ* r n transition originating from the chlorine 3p electrons
of the C-Cl bond is also weak but takes place at a much shorter
wavelength at 174 nm with a fwhm of ∼10 nm. The π* r π
transition of the aromatic ring electrons occur at ∼ 235 nm with
a fwhm of 20 nm. All these transitions are broad, and therefore,
at any given ultraviolet wavelength, it is possible to access all
or any combination of these transitions. Each of these transitions
can further be decomposed to many more pure transitions. For
example, underlying the σ* r n transition localized on the C-X
3
3
1
bond there are three transitions Q1 r n, Q0 r n, and Q1 r
n in increasing order of energy, as shown in the case of simple

7952 J. Phys. Chem. A, Vol. 108, No. 39, 2004
Senapati et al.
Figure 2. Pump laser power dependence of Cl/Cl* REMPI signal from o-C6H4ICl. The straight line is a linear least-squares fit through the
experimental points. The slope of the lines is mentioned inside the plots.
oroiodobenzene. Also the influence is stronger if the halogen
atom is attached to the carbon atom holding the C-I bond or
the number of halogen atom substitutions in the molecule is
high. In the example of chloroiodobenzene, only one halogen
atom is attached to the next nearest carbon atom. Then why is
there a significant variation in I* yield from iodobenzene to
o-chloroiodobenzene in many cases? The spin-orbit interaction
exerted by the lone Cl atom alone cannot account for such a
large change. If we look at the electron affinity of the chlorine
atom, it is 3.617 eV, which is large. This may in turn exhibit a
negative inductive effect (-I effect) on the carbon atoms in
the ortho and para positions and reduce the electron density on
those carbon atoms. On this basis alone, we would expect the
o- and p-chloroiodobenzene to produce more or less equal
amounts of I* atoms in the photolysis and m-chloroiodobenzene
which is not affected by the -I effect to exhbit similar quantum
yield to that of iodobenzene. However, the fact that the I*
quantum yield is much higher in the ortho compound than in
the para indicates that the -I effect exerted by the chlorine atom
cannot be the only factor affecting the dynamics at the exit
channel. If we look at the optimized geometry of the ortho,
meta, and para isomers using LanL2DZ basis set within the
Gaussian-98 set of programs, we find that the shortest distances
between the Cl and I atoms in the ground state are 3.485 Å for
the ortho, 5.833 Å for the meta, and 6.751 Å for the para
compounds. The distance in the o-chloroiodobenzene is very
much within the van der Waals distance of the two atoms (the
sum of the mean van der Waals radii of Cl and I atoms,32
respectively, is (1.98 + 1.75Å) or 3.73 Å), and, thus the
electronic interaction between the two atoms through space is
expected to be large. The distances are unlikely to change very
much in the excited states from which the molecule dissociates.
In fact, both the halogen atoms have large quadrupole moments,
and they can interact strongly through quadrupole-quadrupole
interaction apart from the induced dipole-induced dipole (both
the C-Cl and C-I bonds have dipole moments) interaction.
All these exit channel interactions perhaps give greater I*
production through the direct dissociation pathway. This inter-
action will also help the indirect pathway for I* production by
mixing the π* states with the dissociative state(s) by providing
higher density of states in the interaction region. Since the
separation distance between the two halogen atoms follow the
order o- < m- < p-, the I* quantum yield at a given wavelength
is expected to follow the reverse order; that is o- > m- > p-.
What is observed is more like o- > m- g p- since the difference
in the quantum yields between meta and para isomers are
marginally different. However, the reason for the deviation of
the quantum yield data at 236 nm for p-chloroiodobenzene (φ*-
(I) ) 0.33) from the general trend observed in this study is not
at all obvious.
B. Cl* Quantum Yield. It is natural to think that the
contribution from the σ* r n transition localized on C-Cl
chromophore which peaks at ∼170-180 nm will be insignificant
in the photodissociation at all the dissociation wavelengths
employed in this study, and consequently no Cl/Cl* will be
produced in the near-ultraviolet dissociation of C6H4ICl. How-
ever, the most exciting observation in this study is that all of
the compounds produce Cl* fragments in a measurable quantity
at most of the wavelengths of investigation. This is not surprising
energetically since it is possible to produce both Cl and Cl*
atoms in a single photon excitation process at the wavelengths
chosen here. In fact, the threshold for Cl/Cl* production in
wavelength is ∼295 nm, and this has been verified by trying to
dissociate chloroiodobenzene at ∼304 nm. No Cl/Cl* was
detected at ∼304 nm. A laser power dependence study show
that the production of Cl/Cl* is unlikely to be originated via a
two-photon process (Figure 2). We rule out, first, the dissociation
of the C-I bond followed by a further dissociation of the
chloroaryl radical, that is, a secondary dissociation pathway for
the production of Cl/Cl*. The energy of a single photon in the
wavelength range 222-280 nm is sufficient to break the C-Cl
bond and produce either Cl or Cl* atoms. Table 2 lists the Cl*
quantum yield, φ*(Cl), for all the compounds (C6H5Cl, and
isomers of C6H4ICl) at various photolysis wavelengths. For
chlorine atom detection, the pump wavelength of 236 nm was
not used; instead, photons from the same probe laser at ∼235
nm were used for dissociation as well as detection. The relative
quantum yields in Table 2 are less than 0.5 with the exception
of o-C6H4ICl. In most cases, more than 50% of the product
chlorine atoms are produced in the spin-orbit ground state. This
is in sharp contrast to what has been observed for the I*
production as discussed above. The quantum yield of Cl*
production in chloroiodobenzenes is higher than that of chlo-
robenzene at all wavelengths except for m-C6H4ICl at 266 nm.
With this exception, the measured Cl* quantum yields reveal
the effect of the neighboring iodine atom on the dynamics of
the C-Cl bond rupture in much the same way as does the
chlorine atom influence the I* yield.
The reason for the production of less Cl* at longer wave-
lengths can be speculated. Since the σ*(C-Cl) r n(Cl)
transition centered on the C-Cl bond is much higher in energy,
and therefore, the direct dissociation pathway via the C-Cl bond

Photodissociation of Chloroiodobenzenes
J. Phys. Chem. A, Vol. 108, No. 39, 2004 7953
two indirect) for dissociation are available, for the C-Cl bond
scission only two indirect paths are possible. The absence of
the direct excitation pathway in the latter results in a low yield
of Cl*. Among the three isomers of chloroiodobenzene, the ortho
compound produces maximum amounts of Cl* and I*, indicating
stronger coupling between various photodissociation pathways
in the exit channel. The coupling is, perhaps, aided by through
space induced dipole-induced dipole and/or quadrupole-quad-
rupole interactions between the two halogen atoms. A quantita-
tive dynamics calculation using accurate potential energy
surfaces is necessary for developing a more quantitative
understanding of the ultraviolet dissociation of mixeddihalo
benzenes.
Acknowledgment. We thank Prof. A. G. Menon for helping
us with the design of the REMPI detector. The research
described here has been funded generously by the Department
of Atomic Energy, Government of India.
References and Notes
(1) Vogt, R.; Sander, R.; Glasow, R. V.; Crutzen, P. J. J. Atmos. Chem.
1999, 32, 375.
(2) Dubtsov, S. N.; Skubnevskaya, G. I.; Kutzenogii, K. P. J. Aerosol
Sci. 1991, 23, 181.
(3) Dzvonik, M.; Yang, S.; Bersohn, R. J. Chem. Phys. 1974, 61, 4408.
(4) Kawasaki, M.; Lee, S. J.; Bersohn, R. J. Chem. Phys. 1977, 66,
2647.
Figure 3. Schematic potential energy diagram describing the mech-
anism of the C-Cl bond dissociation. IVR is intramolecular vibrational
relaxation, and IC, internal conversion. Under Cs geometry, 9A′, 10A′,
and 11A′ are the (σ*(C-Cl) r n(Cl)) transitions.
excitation is not available for the Cl/Cl* production. However,
the two indirect pathways as applicable to the C-I bond
dissociation are accessible. The Cl* atoms are then produced
by a curve crossing mechanism from the initially excited S1/S2
r S0 state (s) via the dissociative state(s). Observation of a
low φ*(Cl) indicates that this curve crossing must be hindered.
Since the initially prepared excited state is low in energy
compared to the σ*(C-Cl) r n(Cl) states, the intersystem curve
crossing must be taking place at a much larger C-Cl bond
distance allowing the molecules to spend a long time in the
exit channel and dissociate by forming Cl atoms in the ground
spin-orbit state. This has been shown in a schematic potential
energy diagram in Figure 3. In this diagram, a molecule excited
to the S2 state first undergoes intramolecular vibrational
relaxation (IVR) to lower vibrational energy levels of the excited
state. A fraction of the molecules then crosses over to the
dissociative states to produce both Cl* and Cl products. A small
fraction of the molecules can also undergo fast internal
conversion to the high vibrational levels of the S0 state which
will eventually dissociate and produce ground-state Cl atom.
Also in this case, the ortho effect on φ*(Cl) is much more
prominent. The variation of φ*(Cl) as a function of wavelength
has to do with the extent to which the various initial excitation
carry the total transition strength and how the final repulsive
states are placed with respect to the initial excited states. In
other words, the variation reflects the actual dynamics of the
dissociation from the excited-state potential energy surfaces.
(5) Freedman, A.; Yang, S. C.; Kawasaki, M.; Bersohn, R. J. Chem.
Phys. 1980, 72, 1028.
(6) Freitas, J. E.; Hwang, H. J.; El-Sayed, M. A. J. Phys. Chem. 1993,
97, 12481.
(7) Cheng. P. Y.; Zhong, D.; Zewail, A. H. Chem. Phys. Lett. 1995,
237, 399.
(8) Griffiths, J. A.; Jung, K. W.; El-Sayed, M. A. J. Phys. Chem. 1996,
100, 7989.
(9) Unny, S.; Du, Y.; Zhu, L.; Truhins, K.; Gordon, R. J.; Sugita, A.;
Kawasaki, M.; Matsumi, Y.; Delmdahl, R.; Parker, D. H.; Berces, A. J.
Phys. Chem. A 2001, 105, 2270.
(10) Kavita, K.; Das, P. K. J. Chem. Phys. 2002, 117, 2038.
(11) Ichimura, T.; Mori, Y.; Shinohara, H.; Nishi, N. Chem. Phys. Lett.
1985, 122, 51.
(12) Ichimura, T.; Mori, Y.; Shinohara, H.; Nishi, N. Chem. Phys. Lett.
1985, 122, 55.
(13) Satyapal, S.; Tasaki, S.; Bersohn, R. Chem. Phys. Lett. 1993, 203,
349.
(14) Ichimura, T.; Mori, Y.; Shinohara, H.; Nishi, N. Chem. Phys. 1994,
189, 117.
(15) Wang, G.; Zhu, R.; Zhang, H.; Han, K.-L.; He, G.-Z.; Lou, N.-Q.
Chem. Phys. Lett. 1998, 288, 429.
(16) Doering, J. P. J. Chem. Phys. 1977, 67, 4065.
(17) Frueholz, R. P.; Flicker, W. M.; Mosher, O. A.; Kupperman, A. J.
Chem. Phys. 1979, 70, 3057.
(18) Trudell, B. C.; Price, S. J. W. Can. J. Chem. 1979, 57, 2256.
(19) Ichimura, T.; Mori, Y.; Shinohara, H.; Nishi, N. J. Chem. Phys.
1997, 107, 835.
(20) Zhu, R.-S.; Zhang, H.; Wang, G.-J.; Gu, X.-B.; Han, K.-L.; He,
G.-Z.; Lou, N.-Q. Chem. Phys. 1999, 248, 285.
(21) Gedanken, A.; Robin, M. B.; Yafet, Y. J. Chem. Phys. 1982, 76,
4798.
(22) Eppink, A. T. J. B.; Parker, D. H. J. Chem. Phys. 1998, 109, 4758.
(23) Arepalli, S.; Presser, N.; Robie, D.; Gordon, R. J. Chem. Phys.
Lett. 1985, 118, 88.
(24) Liyanage, R.; Yang, Y.; Hashimoto, S.; Gordon, R. J.; Field, R.
W. J. Chem. Phys. 1995, 103, 6811.
4. Concluding Remarks
(25) Senapati, D.; Kavita, K.; Das, P. K. J. Phys. Chem. A 2002, 106,
8479.
From I* and Cl* quantum yield measurements in the
photolysis of o-, m-, and p-chloroiodobenzenes in the ultraviolet,
we have shown, in this paper, that both chlorine and iodine
atoms are produced in their spin-orbit excited as well as ground
states. The dynamics of spin-orbit excited halogen atom
production is very complex and depends on (i) the electronic
nature of the initially prepared state, (ii) the intersystem crossing,
and (iii) the exit channel dynamics. While, in the case of the
C-I bond rupture, three competitive pathways (one direct and
(26) Hwang, H. J.; El-Sayed, M. A. J. Phys. Chem. 1991, 95, 8044.
(27) Matsumi, Y.; Tonokura, K.; Kawasaki, M.; Inoue, G.; Satyapal,
S.; Bersohn, R. J. Chem. Phys. 1991, 94, 2669.
(28) Regan, P. M.; Langford, S. R.; Ascenzi, D.; Cook, P. A.; Orr-
Ewing, A. J.; Ashfold, M. N. R. Phys. Chem. Chem. Phys. 1999, 1, 3247.
(29) Gedanken, A.; Rowe, M. D. Chem. Phys. Lett. 1975, 34, 39.
(30) Phillips, D. L. Prog. React. Kinet. Mech. 1999, 24, 223 and
references therein.
(31) Kavita, K.; Das, P. K. J. Chem. Phys. 2000, 112, 8426.
(32) Bondi, A. J. Phys. Chem. 1964, 68, 441.